**Study of Semi-Solid Magnesium Alloys (With RE Elements) as a Non-Newtonian Fluid Described by Rheological Models**

#### **Marta Sl ˛ ´ ezak ID**

Department of Ferrous Metallurgy, Faculty of Metals and Industrial Computer Science, AGH University of Science and Technology, Al. Mickiewicza 30, 30-059 Kraków, Poland; mslezak@agh.edu.pl; Tel.: +48-12-617-26-02

Received: 28 December 2017; Accepted: 23 March 2018; Published: 28 March 2018

**Abstract:** This paper includes the results of high-temperature rheological experiments on semi-solid magnesium alloys and the verification of different models describing the rheological behaviour of semi-solid magnesium alloys. Such information is key from the point of view of designing alloy forming processes in their semi-solid states. Magnesium alloys are a very attractive material, due to their light weight and good plastic properties; on the other hand, this material is very reactive in a liquid (semi-solid) state, which is challenging from a testing and forming perspective. Formulating/finding models for an accurate description of the rheological behaviour of semi-solid magnesium alloys seems to be key from the standpoint of developing and optimising forming processes for semi-solid magnesium alloys.

**Keywords:** rheological model; semi-solid state; Mg alloys; high-temperature rheology; rheological properties

#### **1. Introduction**

Magnesium alloys are currently growing in importance as materials for parts used in the automotive industry. In addition, materials made of magnesium alloys have been accepted by the European Civil Aviation Conference and NASA as materials for the production of parts which are not prone to corrosion. Magnesium, with its specific gravity of 1.8 g/cm3, is the lightest structural material. It is over four times lighter than steel and 1.5 times lighter than aluminium; at the same time it maintains very good mechanical properties, including ductility, which can be modified by the addition of appropriate alloying elements. Magnesium ranks eighth as the most frequently-occurring element in the lithosphere. It is produced from seawater, brines or magnesium rock and, therefore, its resources are enormous. Moreover, it is 100% recyclable. At present, annual magnesium output is estimated at about 500,000 tonnes p.a.

On the other hand magnesium is a very reactive material in a liquid (semi-solid) state, which makes it challenging from a laboratory and industrial perspective.

Viscosity is a property of liquid metals which plays a key role in many effects occurring in high-temperature conditions. It is a very important parameter when controlling manufacturing processes in which liquid metal is present: casting, forming [1–5], also in the semi-solid state. Data from measurements taken at high-temperatures are necessary for engineering new processes, and for the optimisation of those that already exist [6,7]. Many mathematical models that can assist in describing the thermodynamics, kinetics, fluid flow, and heat exchange have been created in recent years [8–10]. Obtaining the correct measurement data has been the basis for the creation of accurate models. The above-mentioned models may be helpful in modelling/optimising processes with the participation of the liquid phase. Mathematical modelling and the control of molten metal processing operations require knowledge of the thermophysical properties of liquid metals. The accuracy of the measurements of these properties is the basic precondition for the development of processes in materials engineering.

The issue of the influence of rheological parameters on semi-solid metal forming processes (SSM) has been considered in the subject literature. The beginning of semi-solid metal forming (SSM—so-called thixotropic forming) goes back to 1970 [11]. At the moment, it is believed that knowledge of the rheological properties in the semi-solid metal alloy forming processes plays a key role in process engineering [12–14]. Current semi-solid metal forming processes have been applied primarily in light metals processing [15,16]. Viscosity is the main rheological parameter considered in the SSM processes [17–21]; it is an indicator defining the capability of the metal to fill a mould, and it determines the force required to deform a material.

Many authors have taken up the subject of analysing the value of the dynamic viscosity coefficient of magnesium alloys [22–28]; however, this data did not concern systems in which rare earth elements had been added. Additionally, authors usually make rheological investigations of semi-solid slurries of alloys [22–28]. At the same time, the most frequently tested magnesium alloy—AZ91—was analysed in a slightly different way: by analysing rest time and subjecting the system to the impact of forces [22,23], instead of gradually changing the shear rate. This paper supplements the research on the rheological characteristics of magnesium alloys containing rare earth metals and include the results of rheological analysis by using models which are often mentioned in papers about aluminum alloys [16–21].

This paper contains the selected results of rheological tests of semi-solid magnesium alloys of the Mg-Zn-Al, Mg-Zn-RE groups: three chemical compositions with applied shear rates from 10 to 150 s<sup>−</sup>1. The results of rheological tests conducted on magnesium alloys have been used to verify rheological models by Herschel-Bulkley, Ostwald, Carreau, and Bingham, which are most often used in the subject literature to describe the rheological behaviour of semi-solid metallic systems (aluminium, magnesium alloys). These models may be used for modelling alloy-forming processes in semi-solid states, and for computing individual rheological parameters (dynamic viscosity coefficient, shear stress, etc.) without the need to conduct expensive, complicated, and time-consuming tests.

The research materials presented in this paper form part of the tests and analyses performed, which, due to the complexity of the topic, constitute a cycle of studies concerning broadly-understood rheological analyses of liquid and semi-solid magnesium alloys (with various shares of the solid and liquid phases)**.**

#### **2. Materials and Methods**

The rheological tests were carried out with a high-temperature rheometer [29–34] designed by the Anton Paar company (Anton Paar GmbH, Graz, Austria). The FRS1600 rheometer is a very precise instrument, equipped with an air bearing, one of the few instruments of this type that enables measurements to be performed at high temperatures, testing a very broad range of liquids, characterised by both high and low viscosity values (thanks to the measurement range of torque from 0.05 mNm to 200 mNm). Basically, it consists of the head of a rheometer and a furnace which enables a temperature in the range 673–1805 K to be obtained. There is also the possibility of providing measurements at room temperature. One of the main advantages is an operating system based on pneumatic servomotors used to manipulate the crucibles and the rotating rods inside the furnace. Control of the furnace is also possible using the rheometer software (which is called Rheoplus), which allows experiments with changes of temperature to be programmed. It is not only possible to study the rheological properties of materials, in the liquid state in this type of rheometer, but also in the semi-solid state.

The measurement is performed in a fixed crucible into which a sample of the material tested is placed, then a rotating spindle is immersed within the material being tested. The crucible is then placed inside a ceramic shroud, being a component of a heating furnace. The furnace, which is comprised of four electrically-heated SiC-type heating elements, enables the maximum temperature of 1793 K to be obtained within the sample. The whole device is shielded from the outside with an insulating material. The temperature inside the furnace is controlled by a change in feeder power in the measurement and control system. The heating rate, along with maintaining the temperature at a constant level, are set in the control panel of the Rheoplus software of the rheometer. Rotary movements of the spindle are controlled by a motorised measurement head—the spindle being suspended on a ceramic tube placed in an air bearing. The head is cooled with water and air in order to ensure a low temperature.

In this rheometer the torque values are measured by the head and then the software calculates the values of shear stress, viscosity, etc. [34]. The instrument features torque accuracy of 0.001 mNm. The parameters of the geometry of the measurement system used are implemented in the Rheoplus software before starting the experiments. This method of measurement, with adequate equations for viscosity calculation, is fully described in [34].

The measurements were provided in a Searle-type system [18,20,29–34]. Concentric cylinder systems are described by the standards ISO 3219 and DIN 53019. Bobs with perforated surfaces, with diameters of 16 mm, and cups with smooth inner surfaces and an internal diameter of 30 mm were used for the tests. Materials were selected for the tools that prevented the tool surface reacting with the sample tested. The measurement system was made of low-carbon steel.

The rheological tests were conducted for three magnesium alloys with different chemical compositions. Table 1 presents the chemical compositions of the magnesium alloys tested.

For each of the aforesaid grades (Table 1), the values of the liquidus temperatures and temperatures of the solid phase content of 50% (Table 2) were determined with DSC (Differential Scanning Calorimetry) analysis (Figure 1).


**Table 1.** Magnesium alloys tested.

**Table 2.** The values of the liquidus temperature and the temperature of the solid phase content of 50% for the magnesium alloys analysed according to the DSC analysis.

**Figure 1.** Graphs of DSC analysis of the magnesium systems analysed: AZ91, E21 and WE43B.

The determination of a fraction of the liquid phase as a function of temperature was calculated using data collected from the differential scanning calorimeter. It was assumed that a fraction of a liquid phase is proportional to the absorbed/released energy during the transformation (melting/solidification). The estimation of the liquid phase fraction changes was carried out by application of a partial peak area integration. Liquid fraction at a given temperature is determined by calculating the ratio of the area corresponding to the partial heat of melting over the total peak area. The former area is limited by the solidus temperature and the temperature range between solidus and liquidus lines (semi-solid range), the latter area is limited by solidus to liquidus temperature. It is expressed in volume percentage.

The amount of liquid phase was also determined as 50% on the basis of the content of eutectic phase in the sample after cooling from semi-solid temperature range. The discrepancy in the volume of liquid phase results from different heating rates, as well as an inhomogeneous chemical composition of the feedstock.

The samples were cylindrically shaped: 40 mm height, 25 mm in a diameter.

The rheological tests were conducted from the liquidus temperature to the temperature of a solid phase share of 50% (cooling rate was 2 K/min) for each alloy tested. The magnesium alloy tests were carried out in conditions of variable rheological parameters: shear rate values varied from 10 to 150 s<sup>−</sup>1, and the objective of the tests was to find the influence of the aforementioned variables on the value of shear stress and, thus, to attempt to determine the rheological nature of the magnesium alloys tested.

The scheme of measurements (for each Mg alloy) was as follows:


The findings are presented in the form of flow curves.

In the case of research on very reactive magnesium alloys, conducted on a wide range of solid phase shares (the upper range of the nominal torque operation of the measuring head), it was decided to carry out the measurements at the maximum shear rate of 150 s<sup>−</sup>1. However, the description of data by rheological models was also assessed by paying special attention to the approximation of the shear rate to higher values, referring to thixoforming process conditions.

#### **3. Results**

Figure 2 presents a graph of changes in the shear stress value of AZ91 for variable shear rate values. The values of shear stress of alloy AZ91, for a solid phase share of 50%, grow non-linearly as the shear rate grows, which shows non-Newtonian rheological behaviour of the body tested. The shear stress values grow from about 10 Pa to about 50 Pa, as the shear rate values grow from 10 to 150 s<sup>−</sup>1.

The next figure (Figure 3) presents a graph of changes in the shear stress value of alloy E21 for variable shear rate values.

The values of shear stress of alloy E21, for 50% solid phase share, grow non-linearly as the shear rate grows, in a similar manner to alloy AZ91. However, for alloy E21, the shear stress values grow from about 3 to about 20 Pa.

Figure 4 presents a graph of changes in the tangential stress value of alloy WE43B for variable shear rate values.

*Metals* **2018**, *8*, 222

The values of the shear stress of alloy WE43B, for a share of 50% solid phase, grow non-linearly as the shear rate increases. However, the values of shear stress obtained for alloy WE43B were the lowest, from about 4 to about 10 Pa.

**Figure 2.** Flow curve of alloy AZ91, at a solid phase share of 50%.

**Figure 3.** Flow curve of alloy E21 for 50% of the solid phase.

**Figure 4.** Flow curve of alloy WE43B for 50% of the solid phase.

Analysis of the results obtained allowed us to establish that the highest values of shear stress were obtained for alloy AZ91 (50 Pa), while the shear stress values for alloys E21 and WE43B achieved a maximum of 10–20 Pa. It should be borne in mind that alloys E21 and WE43B contain rare earth elements (yttrium, neodymium, gadolinium), which may influence changes in the shear stress values. However, rheological tests of alloys with various contents of the above-mentioned rare earth metals would need to be performed to verify this.

Alloy E21 shows some deviation during the end of the test, it is probably the effect of shearing particles (the high amount of solid fraction). The author provided the wider rheological and microstructure measurements of these Mg alloy and did not observe the deviations of E21 alloy behaviour in comparison to the WE43B alloy (both contain RE elements).

In addition, all of the alloys tested showed a tendency to non-linear increase of the shear stress value as the shear rate increased. This may indicate a tendency towards shear-thinning (a decline in the value of the dynamic viscosity coefficient resulting from the forces applied). This is one of the key examples of non-Newtonian behaviour, characteristic for metals in a semi-solid state.

#### *3.1. Data Description with Rheological Models*

The rheological behaviour of a material is described by the relationships between stresses, strains, shear rates, and the time during which the material has been subjected to such strains. Such relationships are called rheological equations of the state of the material, or rheological equations, for short. The main task of rheology is to formulate models for describing the behaviours of bodies that have been subjected to an impact force.

In the subject literature, many attempts have been made to describe the flow curve with an appropriate rheological mathematical model [32–37]. The foregoing models are necessary for the analytical solution of problems related to non-Newtonian fluid flow [38–40].

Rheological models constitute a group of equations, which, apart from dynamic viscosity, also take into account other rheological parameters—shear rate, shear time, etc. The simplest mathematical rheological model, which describes a non-Newtonian fluid flow curve within a range of intermediate shear rates, is the so-called Ostwald-de Waele power law model in the form [33]:

$$
\pi = k(\dot{\gamma})^{\pi},
\tag{1}
$$

where *<sup>k</sup>* is the empirically-determined constant (Pa·sn), *<sup>n</sup>* is the empirically-determined index exponent (-), *<sup>τ</sup>* is the shear stress (Pa), and . *γ* is the shear rate (s<sup>−</sup>1).

The power law model created by Ostwald and DeWaele is the simplest mathematical rheological model of a generalised Newtonian fluid, containing only two constants that need to be determined.

To better describe experimental data, numerous authors have proposed to use mathematical rheological models with more complex structures. This study attempted to approximate the results obtained with four selected rheological models, which are most often used in the subject literature to compute (approximate and describe) the values of the shear stress of aluminum and magnesium alloys intended to be formed in a semi-solid state [24,28,41,42].

The following models were used to describe the data:


$$
\pi = \pi\_{\rm HB} + k(\dot{\gamma})^n \tag{2}
$$

• Carreau (Equation (3)):

$$\eta = \frac{\eta\_0 - \eta\_{\infty}}{1 + \left(c \cdot \dot{\gamma}\right)^{2p}} + \eta\_{\infty} \tag{3}$$

Form of the Carreau model which is used in the Rheoplus calculations:

$$\tau = \frac{\tau\_0 - \tau\_\infty}{1 + \left(\boldsymbol{c} \cdot \dot{\boldsymbol{\gamma}}\right)^{2p}} + \tau\_\infty \tag{3a}$$

#### • Bingham (Equation (4)):

$$
\tau = \tau\_\mathcal{B} + k(\dot{\gamma}),
\tag{4}
$$

where *<sup>τ</sup>* is the shear stress (Pa), *<sup>τ</sup>HB* is the Herschel-Bulkley shear stress (Pa). . *γ* is the shear rate (s<sup>−</sup>1), *k* is the Bingham constant (Pa·s), *n* is the empirically-determined index exponent (-), *c* is the Carreau constant (s), *p* is the Carreau exponent (-), *η* is the dynamic viscosity coefficient (Pa·s), *η*<sup>0</sup> is the dynamic viscosity coefficient for shear rates approaching 0 (Pa·s), *η*<sup>∞</sup> is the dynamic viscosity coefficient for shear rates approaching ∞ (Pa·s), *τ*<sup>B</sup> is the Bingham shear stress (Pa), *τ*<sup>0</sup> is the shear stress for shear rates approaching 0 (Pa), and *τ*<sup>∞</sup> *is the* shear stress for shear rates approaching ∞ (Pa).

#### *3.2. Rheoplus Calculations*

Rheoplus V3.40 (Anton Paar GmbH, Ostfildern, Germany) is the integrated software for Anton Paar rheometers. By using Rheoplus it is possible to control instruments during measurement and analyse measurement data after testing. The different rheological models which were used to describe and fit the results obtained were implemented in the software.

Using Rheoplus software for each of the alloys tested over a range with variable shear rate values from 10 to 150 s−1, an approximation of the results obtained was attempted with the four selected rheological models. The results obtained were presented in the form of a graph (with the calculated correlation coefficient *R*2—the degree to which the model matched the actual data) with the actual flow curve obtained by measurements and the flow curves obtained from each model marked on the graph. The curves were presented in the shear stress *<sup>τ</sup>*—shear rate . *γ* system (flow curve), as such relationships occur in three of the rheological equations presented. However, the Carreau equation is usually only defined for the dynamic viscosity coefficient *η*, so to enable models to be compared, the Carreau equation was also presented as shear stress versus shear rate.

Figure 5 presents the graphs of flow curves for alloy AZ91: actual and for three models.

**Figure 5.** Flow curves for alloy AZ91: actual and computed from the Hershel-Bulkley (*R*<sup>2</sup> = 0.99873), Ostwald (*R*<sup>2</sup> = 0.87918), Carreau (*R*<sup>2</sup> = 0.99875), and Bingham (*R*<sup>2</sup> = 0.91265) models.

As we can observe, on graphs (Figure 5) the model and actual flow curves largely overlap, with the greatest deviations being seen for the flow curve described by the Ostwald model, which is reflected in the lowest value of the correlation coefficient *R*<sup>2</sup> out of all those computed.

Figure 6 presents the graphs of the flow curves for alloy E21: actual results and for three models. The graphs (Figure 6) display the large overlap between the model and actual flow curves.

The greatest deviations can be seen for the flow curves described by the Ostwald and Bingham models (the lowest values of coefficients *R*2).

Figure 7 presents graphs of flow curves for alloy WE43B: actual and for three models.

**Figure 6.** Flow curves for alloy E21: actual and computed from the Hershel-Bulkley (*R*<sup>2</sup> = 0.9861), Ostwald (*R*<sup>2</sup> = 0.89346), Carreau (*R*<sup>2</sup> = 0.98654), and Bingham (*R*<sup>2</sup> = 0.90697) models.

**Figure 7.** Flow curves for alloy WE43B: actual and computed from the Hershel-Bulkley (*R*<sup>2</sup> = 0.97892), Ostwald (*R*<sup>2</sup> = 0.97174), Carreau (*R*<sup>2</sup> = 0.93052), and Bingham (*R*<sup>2</sup> = 0.97731) models.

On the basis of the analysis of graphs on Figure 7, one may find that for flow curves describing the rheological behaviour of alloy WE43B, all four models represent the measurement results rather well; correlation coefficients over 0.93 were obtained for all four models.

As the Rheoplus software only enabled the quality of models to be assessed by analysing the correlation coefficient *R*2, and the models were non-linear, the models were verified in Wolfram Mathematica. This analysis enabled the quality of the data description (mean prediction bands) to be more explicitly assessed by the selected rheological models, and allowed us to determine which non-linear model works best for the description of data from the measurements of semi-solid magnesium alloys with a 50% solid phase share.

#### *3.3. Wolfram Mathematica Calculations*

Wolfram Mathematica 11 (developed by The Wolfram Centre, Long Hanborough, United Kingdom) is a mathematical symbolic computation program used in many scientific, engineering, mathematical, and computing fields. The data from measurements where analysed in the Mathematica software implemented four different rheological models (Equations (1)–(4)). The results are presented as graphs and Equations (5)–(16), with calculated factors.

Below (Tables 3–5, Equations (5)–(16)), mathematical formulae of rheological models (calculated with Wolfram Mathematica software) are presented for each of the test alloys. In each of the cases the shear stress *τ* is given in Pa.

**Table 3.** Mathematical formulas of rheological models for the AZ91 alloy.


**Table 4.** Mathematical formulas of rheological models for the E21 alloy.


**Table 5.** Mathematical formulas of rheological models for the WE43B alloy.


On the basis of the analysis of the above graphs and equations, we can conclude that the Herschel-Bulkley and Carreau models describe the results obtained well and, in addition, the mean prediction bands are relatively narrow for both cases. The mean prediction bands are the confidence bands for mean predictions and give functions of the predictor variables. This confirms the good quality of the description of the measurement data for these models. However, the Carreau model is much more complicated mathematically; therefore, the H-B model seems to be more appropriate for describing the rheological data obtained for the aforementioned magnesium alloys. Furthermore, the Carreau model is a function for calculations of flow behaviour including zero-shear and infinite-shear viscosity, thus, it dedicated for systems in which a wide range of shear rates are measured.

As a result of the analysis of the above models, it was found that the Herschel-Bulkley model best described the rheological behaviour of semi-solid magnesium alloys with a significant solid phase share in the alloy tested. For this model, the shear stress value was approximated for shear rates of 200 s<sup>−</sup>1. The results obtained are presented in Figure 8.

As a result of the analysis of the approximation of the flow curves described with the Herschel-Bulkley model, one may find that this model is suitable for alloys AZ91, WE43B (large coverage with measurement data, small calculation uncertainty). This model describes data for alloy E21 as being slightly worse and, moreover, there is greater model uncertainty at higher values of the shear rate. However, this is likely to be related to the deviations of recorded measurement points that were not observed for the other two alloys.

**Figure 8.** Approximated flow curves, along with mean prediction bands (functions of the predictor variables) calculated for the Herschel-Bulkley models, for each of the tested alloys: AZ91, E21 and WE43B (black continuous line—flow curve, grey dotted line—mean prediction bands).

#### **4. Discussion**

Flow curves obtained from measurements were compared with curves resulting from the use of four different rheological models: Herschel-Bulkley, Ostwald, Carreau, and Bingham. This allowed us to determine that all models provided a good level of accuracy of description, however, the measurement data were most accurately described by the models of Carreau and Herschel-Bulkley. Due to a simpler mathematical form, the model that is the most recommended for the description of

data from rheological measurements of semi-solid magnesium alloys (with 50% solid phase share in the alloy) is the Herschel-Bulkley model.

The approximation conducted for higher values of the shear rate, for the Herschel-Bulkley model, showed that the model predicted shear stress values well, in particular for alloys AZ91 and WE43B (narrow mean prediction bands), and that it might be used to calculate shear stress values for Mg alloys under higher values of shear rate (according to the thixoforming process). The model performs slightly worse for alloy E21, but this is likely to be a result of the greater span of measurement points obtained.

#### **5. Conclusions**


**Acknowledgments:** Research financed through statutory funds of AGH University of Science and Technology in Krakow, No. 11.11.110.502. Special thanks to Bogusz Kania for his help in the execution of the computing.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2018 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Effect of Segregation and Surface Condition on Corrosion of Rheo-HPDC Al–Si Alloys**

#### **Maryam Eslami <sup>1</sup> ID , Mostafa Payandeh 2, Flavio Deflorian 1, Anders E. W. Jarfors <sup>2</sup> ID and Caterina Zanella 2,\* ID**


Received: 22 February 2018; Accepted: 23 March 2018; Published: 24 March 2018

**Abstract:** Corrosion properties of two Al–Si alloys processed by Rheo-high pressure die cast (HPDC) method were examined using polarization and electrochemical impedance spectroscopy (EIS) techniques on as-cast and ground surfaces. The effects of the silicon content, transverse and longitudinal macrosegregation on the corrosion resistance of the alloys were determined. Microstructural studies revealed that samples from different positions contain different fractions of solid and liquid parts of the initial slurry. Electrochemical behavior of as-cast, ground surface, and bulk material was shown to be different due to the presence of a segregated skin layer and surface quality.

**Keywords:** Al–Si alloys; rheocasting; HPDC; electrochemical evaluation

#### **1. Introduction**

High pressure die casting (HPDC) is one of the most used manufacturing process for light alloy components [1,2], due to its high productivity, capability to cast complex geometry, dimensional accuracy, reduced need for finishing operations, and producing component with fine grain microstructure and good mechanical properties [3–5].

Coupling semisolid metal (SSM) casting to HPDC (SSM-HPDC) is a promising technology to produce high quality components with sound microstructure. Higher viscosity of semisolid material in this process reduces air entrapment and consequent porosity in the component. Such a technology introduces a new opportunity to enhance the castability of a component, which is impossible to achieve by traditional manufacturing methods [2,3,6,7].

There are two kinds of semisolid processes: "rheocasting" and "thixocasting". In 1976, Flemings et al. [8] introduced rheocasting as an alternative manufacturing process to die casting and even forging, and suggested it can be used to prepare high quality parts. Rheocasting, the method which is used in the current research, involves shearing force during a first solidification to produce a slurry. The slurry is then transferred into a mold and solidifies with non-dendritic microstructure [9]. Rheocasting can be coupled with high pressure die casting and Rheo-HPDC parts have a globular microstructure and usually show low porosity. This leads to heat treatability and high performance [2,6]. Proper materials for Rheo-HPDC process are limited to those which have good castability with HPDC process and also have low sensitivity of the solid fraction to variations of temperature [10]. Hypoeutectic Al–Si alloys in the range of 5–8% silicon content are suitable choices for this process [11].

While Rheo-HPDC widens the composition range of castable alloys and allows for casting of thin sections, it is a process which increases the inhomogeneity of the microstructure at the macroscale in the final component, in comparison to the conversional HPDC process [5]. This phenomenon arises from the fact that the primary α-Al phase solidifies at higher temperature during slurry preparation, and is characterized with low solubility of alloying elements, therefore, these elements will be higher in the remnant liquid phase [11]. In addition, macrosegregation is formed during the filling process: solid and liquid fractions tend to separate in the gating system (longitudinal macrosegregation) and from the surface to the core of the component (transverse macrosegregation), increasing the microstructure inhomogeneity of the final component. This leads to variations of properties in different locations of the component either in microscopic or macroscopic scale [12–14]. The liquid will preferentially fill the furthest parts of the mold, or the thinner sections, while the solid fraction will concentrate in the core. This is due to the higher viscosity of the slurry compared to the liquid molten metal during casting [15]. As a consequence, higher yield strength and ultimate tensile strength are developed near the vent, where more liquid fraction and refined grains solidify compared to the region near the gate [11].

Park et al. [16] showed that the tensile elongation of thixoformed 357-T5 semi solid aluminum alloy is strongly affected by the α1-Al volume fraction which changes in different locations of the component.

Masuku et al. [17–19] investigated the corrosion behavior of the surface layer of SSM-HPDC 7075-T6 and 2024-T6 alloys in sodium chloride solution. They observed a surface liquid segregated layer (mainly formed by the eutectic) in all the alloys, with higher amounts of alloying elements (such as copper). They did not report any difference between the pitting potential, however, they mentioned that pitting morphology is affected by the amount and distribution of the intermetallic particles and therefore differences are expected between the SSM-HPDC and wrought alloys [18]. How longitudinal macrosegregation affects the corrosion resistance of semisolid aluminum alloys and the behavior of the as-cast surface is still not studied.

Limitation of scientific and technical knowledge makes it essential to evaluate properties and behavior of aluminum alloys produced by means of Rheo-HPDC process under different operational circumstances. Corrosion resistance is a critical property for Al alloys, especially in outdoor applications, and therefore, is an interesting subject either for researchers or industries.

Many authors have investigated corrosion resistance of Al–Mg–Si (6xxx series) [20–28] and Al–Si alloys [29–35]. Regarding Al–Mg–Si alloys, pitting and intergranular corrosions (IGC) have been reported as localized corrosion features [23,27]. IGC susceptibility is especially influenced by the amount of copper, iron, and Mg/Si ratio in the alloy composition [23,24,27,36,37]. The localized corrosion occurs in the presence of phases such as β-phase (Mg2Si), silicon (in alloys with excess Si) and copper-containing phases such as Q-phase (Al4Cu2Mg8Si7) (in alloys with Cu) [38]. Iron-rich intermetallics in Al–Mg–Si alloys are nobler compared to the matrix [21]. Nobler intermetallic particles (IMs) in grain boundaries form a microgalvanic couple with the adjacent precipitate free zones (PFZ), and result in IGC [37]. In chloride containing solutions, Mg2Si undergoes Mg dealloying, before turning to an active cathodic site [21,22].

Al–Si alloys generally suffer from the localized corrosion (pitting) in the Al–Si eutectic, due to impurities, such as Fe [39]. Generally, corrosion behavior of these alloys depends on the amount and morphology of iron-rich IMs, such as β-AlFeSi and π-AlFeSiMg [34]. Both Fe and Si are cathodic with respect to the aluminum. Therefore, together they can form a microgalvanic couple, resulting in localized corrosion [34]. Silicon also increases the corrosion potential [23,27].

Previous corrosion studies of semisolid-cast Al-Si alloys such as those performed by Yu et al. [40], Park et al. [31], Tahamtan et al. [30,35] and Arrabal et al. [34], have mostly emphasized on the pitting corrosion in the eutectic regions of A356 and A357 alloys.

It is shown that semisolid-cast process, such as thixoforming, can effectively modify the Si morphology in 357 alloy, resulting in a higher corrosion resistance [40]. The acicular eutectic silicon phase in permanent mold cast 357 alloy has more contact area with the aluminum matrix in comparison to that of the globular eutectic silicon in the thixoformed alloy, and this can encourage the galvanic corrosion [40].

Rheocast process can increase the concentration of silicon in α-Al particles in A356 aluminum alloy. This leads to smaller potential differences between this phase and the eutectic silicon phase and β-AlFeSi IM particles, which result in higher resistance to pitting corrosion [34]. Although both of the eutectic silicon phase and iron-rich IMs contribute in the localized corrosion, some authors consider the contribution of IM particles to be more important [34,41].

The present study focused on the Al low Si alloys produced by Rheo-HPDC process and on the effect of the microstructure inhomogeneity on the corrosion resistance. In addition, the influence of silicon content and surface condition on the microstructure characteristics and the corrosion behavior is examined.

#### **2. Materials and Methods**

A telecom cavity filter (Figure 1) was rheocast by a 400 ton HPDC machine equipped with an automated RheoMetalTM slurry generator. The RheoMetalTM (Stockholm, Sweden) process uses an Enthalpy Exchange Material (EEM) as slurry generator [42].

**Figure 1.** The experimental cavity filter used for Rheo-high pressure die cast (HPDC) process.

The alloys were prepared, and their composition were adjusted by adding pure silicon to a primary alloy in a resistance furnace. The chemical compositions of the alloys were measured by optical emission spectroscopy (OES) (SPECTRO Analytical Instruments GmbH, Kleve, Germany) and are presented in Table 1.


**Table 1.** Measured composition (wt %) of alloys.

In the casting process, the temperature of the fixed half of the die was maintained at 230–250 ◦C, while the temperature of the moving half was set to 280–320 ◦C. Shots of about 5 kg were held at 675 ◦C in the ladle where 5% of EEM was added under stirring at 900 rpm. The final slurry temperature was 610 ◦C and had 40% of solid fraction. The die was filled in two stages with piston speed at 0.23 and 5.2 m/s, respectively, and the shot time was 31 ms. The solid fraction in the slurry was estimated by image analyses of a quenched sample of slurry after light etch using a 5% NaOH.

Specimens for corrosion tests were taken from the thin walls (thickness ≈ 1.5 mm), in as-cast or ground condition, or from the thick bottom plate (thickness ≈ 4 mm), underneath the component only in ground condition.

To investigate longitudinal segregation, both thin wall and thick bottom plate samples were taken from different locations: near the feeding gate (G) or near the die vent (V). The detailed samples designation is presented in Table 2.


**Table 2.** Designation of samples.

R = Rheo-HPDC, G = Samples from near the gate, V = Samples from near the vent, P = Thin wall samples in ground surface condition, B = Samples from the thick bottom plate ground to the half of thickness (Bulk samples).

Polarization tests and electrochemical impedance spectroscopy (EIS) were performed with a 3-electrode cell and a potentiostat (Parstat 2273, Ametek, Berwyn, PA, USA). The aluminum alloy with an exposure area of 1 cm<sup>2</sup> was connected as working electrode, platinum as a counter and silver/silver chloride (Ag/AgCl·3M·KCl) as reference electrodes. Due to the working environment of telecom components, diluted Harrison solution (0.5 g/L NaCl and 3.5 g/L (NH4)2SO4) was used to simulate the electrochemical behavior of the alloys exposed to an acid rain [43,44].

Regarding the polarization test, the sweep rate was 0.166 mV/s, and the delay time before each the test was 600 s, to let the open circuit potential (OCP) reaches its stable value. For each sample, mainly anodic branch was collected (as cathodic branch did not exhibit any significant information), starting from OCP. Anodic polarization was stopped after the maximum current density of 9 × <sup>10</sup>−<sup>4</sup> A/cm2 was reached. In order to highlight the effect of chloride ions, all the polarization experiments were also repeated using a solution of 3.5 g/L (NH4)2SO4.

EIS measurements were collected for 24 h of immersion at the room temperature in the diluted Harrison solution from 100 kHz to 10 MHz with 36 points and 10 mV of amplitude of the sinusoidal potential.

EIS measurements were conducted on as-cast and ground surfaces of thin wall samples, and also on ground surface of the thick bottom plate samples, to investigate the effect of transverse macrosegregation and as-cast condition.

To distinguish the results of these different experiments, the letter P is added to the name of the samples of thin walls, which were ground before the electrochemical test. These samples were wet ground by silicon carbide abrasive papers from P600 to P4000 to the extent that the skin layer was completely removed. In the case of bulk samples (of the thick bottom plate), the surfaces were ground until the middle of each sample was reached. In this way, the electrochemical behavior of the bulk of each alloy can be investigated. Repeatability of the results was tested by conducting each experiment on at least three specimens. ZsimpWinTM software (3.5, Echem software, Ann Arbor, MI, USA, 2013) was used to fit the EIS spectra. After the corrosion tests, corroded surfaces were examined using SEM/EDS (JSM-IT300) (Jeol, Akishima, Japan).

For metallographic analyses, samples were wet ground, followed by polishing using diamond paste (3 μm and 1 μm). NaOH solution (10 wt %) was used to clean the surface and to reveal the constituents of the microstructure. The microstructure of the surfaces and the bulk of components were studied using a light optical microscope (LOM) (Zeiss, Oberkochen, Germany) and scanning electron microscopy (SEM) (Jeol, Akishima, Japan). Energy/wave-dispersive X-ray spectroscopy (EDS/WDS) (EDAX, Mahwah, NJ, USA) was used to measure the composition of the different phases.

#### **3. Results and Discussion**

#### *3.1. Microstructural Features*

Generally, the microstructure of rheocast low silicon content aluminum alloys exhibit the presence of α-Al phase together with Al–Si eutectic mixture and some intermetallic particles [45]. Based on the presence of Fe in the two alloys (Table 1), and since its solubility in Al is very low [46], the presence of Fe-rich intermetallic particles is expected. The sequence of phase formation (aluminum phase and eutectic reaction) was calculated using ThermoCalcTM (2015b, Solna, Sweden, 2015) software [47,48]. The results are shown in Figure 2. As it is predicted by the thermodynamic model, the needle shape β-AlFeSi intermetallic particle was the most favored intermetallic phase for precipitation, and it is formed before the eutectic silicon.

**Figure 2.** Sequence of formation of different phases in the (**a**) Alloy 2.5 and (**b**) Alloy 4.5.

The microstructural features of the polished surfaces of thin walls from different positions in the cavity, with different percentage of silicon, are illustrated in LOM images in Figure 3. From microstructure images in Figure 3, two different range sizes of α-Al phase can be observed: a coarse globular α-Al phase (α1-Al) and a finer α-Al phase (α2-Al).

The formation of α1-Al and α2-Al is related to the multi-stage solidification in the semisolid metal process. α1-Al grains are nucleated in the ladle, under shear forces, due to stirring at a higher temperature due to contact with the EEM and form the slurry, while α2-Al grains are mostly formed in the solidification stage inside the die, at a higher cooling rate [49].

Regarding the effect of the position, the microstructure near the gate (Figure 3a,c) consists of a higher amount of α1-Al particles compared to the region near to the vent (Figure 3b,d). However, this difference is not significant in alloy 2.5.

This microstructure heterogeneity is attributed to the distribution among the die of the liquid and solid fraction of the slurry during the injection stage: the liquid part squeezes out and leaves the solid fraction behind near to the gate. Easton et al. [50] show this behavior in a SSM-HPDC component and defined this type of separation of the slurry as sponge effect.

α2-Al particles in thin wall samples (of both alloys) are finer near the vent in comparison to the region near the gate, due to the higher undercooling in this region of the cavity. This effect is more evident for alloy 4.5. It seems that by increasing the amount of silicon, aluminum phase is refined by undercooling, which can be due to more nucleation of the eutectic silicon [51].

**Figure 3.** Light optical microscope (LOM) images of sample (**a**) 2.5 RGP; (**b**) 2.5 RVP; (**c**) 4.5 RGP and (**d**) 4.5 RVP.

LOM images of sample 2.5 RGB and 4.5 RGB from the thick plate are presented in Figure 4. In comparison to the microstructure of thin walls, the bulk microstructure contains more α1-Al particles in both of the alloys, which is expected. In fact, due to the high viscosity of semisolid slurry, thin walls are mostly filled by liquid, while the relatively thicker parts are filled by the solid fraction of the slurry [15].

**Figure 4.** LOM image of sample (**a**) 2.5 RGB and (**b**) 4.5 RGB.

LOM image of cross-section of the thick plate of Alloy 2.5 in the region near the gate is reported in Figure 5. On the surface, there is a segregation of the liquid fraction to the surface and α1-Al solid fraction, which tend to aggregate in the center of samples. This phenomenon is considered as transverse macrosegregation. Three different phenomena lead to transverse macrosegregation: skin effect [12], sponge effect [13,50] and shearing band during melt flow that lead to porosity or

eutectic-rich segregation band [52]. Study of transverse macrosegregation in different positions of the same Rheo-HPDC component performed by Payandeh et al. [53] showed that the thickness of the surface segregation layer increases by increasing the liquid faction. Govender et al. [54] also reported the existence of a surface layer consisting of mainly liquid or eutectic phase in SSM-HPDC A356 Alloy.

**Figure 5.** LOM images of cross-sectional view of sample 2.5 RGB.

Figure 6 compares the as-cast and polished surfaces of sample 2.5RG/P of the thin wall section. It is noticeable how the as-cast surface is enriched with eutectic phase and intermetallic particles. Moreover, localized defects, such as porosities and/or voids among the grains and inclusions, are visible.

**Figure 6.** Scanning electron microscopy (SEM) image of sample (**a**) 2.5 RG (as-cast condition) and (**b**) 2.5 RGP (polished surface).

SEM image of sample 4.5 RGP at higher magnification in Figure 7a depicts the microstructure of the eutectic and the intermetallic particles. The Si particles have a flake shape and form a continuous network [55].

**Figure 7.** SEM image (**a**) and spot EDS analysis (**b**) of polished surface microstructure of the sample 4.5 RGP.

According to the EDS results in Figure 7b, in this figure, intermetallic particles are rich in iron, and as is expected in hypoeutectic alloys, the intermetallic particles are β-AlFeSi. These intermetallic particles usually have a needle shape [34], and a platelet morphology in 3D tomographic volume [56].

Figure 8 represents the concentration of silicon at the center of α1-Al and α2-Al grains in alloys 2.5 and 4.5. α2-Al particles, nucleated from the liquid portion of the slurry [57], have higher amounts of silicon, which leads to smaller potential differences between them and the silicon eutectic phase and/or iron-rich intermetallic particles, and results in less severe corrosion [34].

**Figure 8.** Concentration of Si in α<sup>1</sup> and α2-Al phases in alloy 2.5 and 4.5.

#### *3.2. Corrosion Studies*

#### 3.2.1. Potentiodynamic Polarization Curves

Potentiodynamic polarization curves of as-cast surface of the thin wall samples (2.5 RG, 2.5 RV, 4.5 RG, and 4.5 RV) in the diluted Harrison solution and in the solution of 3.5 g/L (NH4)2SO4 are reported in Figure 9.

**Figure 9.** Potentiodynamic polarization curves in (**a**) the solution of 3.5 g/L (NH4)2SO4 and (**b**) the diluted Harrison solution.

In both of the solutions, values of corrosion potential are similar for different samples regardless of the position and the silicon content. Regarding the effect of the position, samples from near the gate (2.5 RG and 4.5 RG) possess higher pitting potentials, in the solution of 3.5 g/L (NH4)2SO4, compared to the samples taken from near the vent (2.5 RV and 4.5 RV). In the diluted Harrison solution, all of the samples (except for 4.5 RV) show pitting from the OCP, which is due to the chloride ions and the higher amount of iron-rich intermetallic particles, as well as the defective condition of the as-cast surfaces.

Considering the bulk microstructure, the results of potentiodynamic polarization test in the diluted Harrison solution and in the solution of 3.5 g/L (NH4)2SO4, of samples of 2.5 RGB and 4.5 RGB, are presented in Figure 10. In both of the solutions, the corrosion potentials of the two samples shows no significant difference.

**Figure 10.** Potentiodynamic polarization curves in (**a**) the solution of 3.5 g/L (NH4)2SO4 and (**b**) the diluted Harrison solution.

The stability of the passive oxide layer is higher for the bulk samples compared to the as-cast surface, especially in the chloride containing solution. This is due to the surface condition, since the bulk samples have higher α1-Al fraction, less eutectic fraction, and are ground while the thin wall samples are tested in as-cast condition.

SEM images of the corroded surfaces of samples 2.5 RG and 4.5 RG after the polarization test in the diluted Harrison solution are presented in Figure 11. For both samples, corrosion is localized, and a galvanic couple between the eutectic silicon phase and/or iron-rich intermetallic particles and aluminum matrix can be observed.

**Figure 11.** Corroded surfaces of sample (**a**) 2.5 RG and (**b**) 4.5 RG after polarization test in the diluted Harrison solution.

#### 3.2.2. Electrochemical Impedance Spectroscopy

EIS spectra were obtained during 24 h of immersion in the diluted Harrison solution. The Bode plots of EIS spectra of thin wall samples in as-cast condition are reported in Figure 12. The figure also compares the effect of different positions in the cavity and the silicon content in the alloy.

**Figure 12.** Bode plots of EIS spectra of samples 2.5 RG, 4.5 RG, 2.5 RV, and 4.5 RV after (**a**) 1 h; (**b**) 6 h; (**c**) 12 h; and (**d**) 24 h of immersion in the diluted Harrison solution.

For all of the samples, the impedance values at low frequencies decrease with the immersion time, which indicates the progressive corrosion process on the surface. In addition, the phase angle peak depresses through the immersion time, suggesting that the pitting corrosion activity is increasing [58].

Regarding the effect of position, samples taken from near the vent almost always, during the 24 h of immersion, show slightly higher impedance values at low frequencies, compared to the samples taken near the gate, except for 24 h, when all the samples show the same values. As it was discussed before, according to LOM images (Figure 3), the samples, which are taken from near the vent, possess a higher fraction of α2-Al particles, and they also have a finer microstructure compared to the samples which are taken from near the gate.

Concerning the effect of silicon, no noticeable difference is detectable. However, all the previous research has indicated the positive effect of silicon on the pitting resistance of aluminum alloys [29,59,60]. Nevertheless, it is worth mentioning that these studies are focused on silicon content higher than 6%, which is higher than the percentage of silicon in both of our alloys.

To remove the effect of as-cast surface quality, the thin wall samples were ground using SiC abrasive papers to the extent that the skin layer was totally removed. The samples were then monitored by EIS during 24 h of immersion in the same solution. Bode presentation of EIS spectra of these samples are presented in Figure 13. According to these spectra, the ground thin wall samples show one order of magnitude higher impedance values at low frequencies compared to the same samples in as-cast condition. In addition, the impendence values increased during 24 h of immersion for all of these samples. This can be related to the presence of a protective oxide, which is more stable to pitting. This protective oxide later is provided by better finishing quality on the ground surfaces. By grinding, the skin layer is removed, and a surface containing more α1-Al particles and fewer intermetallic particles (Figure 6) is exposed to the corrosive solution.

**Figure 13.** Bode plots of EIS spectra of samples 2.5 RGP, 4.5 RGP, 2.5 RVP, and 4.5 RVP after (**a**) 1 h; (**b**) 6 h; (**c**) 12 h and (**d**) 24 h of immersion in the diluted Harrison solution.

The difference between the corrosion performances of samples from different positions is negligible in this condition. Impedance tends slightly to increase after 6 h, and no pits are developed after 24 h. This proves that the poor behavior of as-cast surface is due to the poor surface quality and the higher amounts of intermetallic particles.

EIS spectra of the samples taken from the thicker plate, which are ground to the middle to expose the bulk microstructure, are reported in Figure 14. These results indicate the growth of a protective oxide layer on the surface.

The equivalent circuits used to describe the electrochemical responses of the samples 2.5 RG(P/B), 4.5 RG(P/B), 2.5 RVP, and 4.5 RVP are shown in Figure 15.

The electrochemical behavior of aluminum surface is affected by the presence of the passive oxide layer and the interface between the intermetallic particles and the aluminum. Although all these constituents are present, their time constants strongly overlap, or one dominates. Therefore, only one peak in the phase diagram can be observed for all of them.

**Figure 14.** Bode plots of EIS spectra of samples 2.5 RGB, 4.5 RGB after (**a**) 1 h; (**b**) 6 h; (**c**) 12 h; and (**d**) 24 h of immersion in the diluted Harrison solution.

**Figure 15.** Equivalent circuits for the studied alloy after (**a**) a short time and (**b**) a longer time of immersion in the diluted Harrison solution.

The oxide layer is represented by a parallel circuit containing a resistor and a capacitor representing, respectively, the oxide ionic conduction and its dielectric properties [61].

The circuit in Figure 15a with one time constant has been used for the first 6 h of immersion, for almost all the samples, while the circuit in Figure 15b has been used for the later hours of immersion, when a second time constant was visible in the EIS spectrum. In the circuits depicted in Figure 15a,b, *REL* represents the resistance of the electrolyte. *QDL* (or *CDL*) and *RCT* stand for capacitive behavior of the electrical double layer at the interface between the surface and the solution, and for the resistance against the charge transfer (or polarization resistance), respectively. The second time constant at longer immersion time shows the oxidation of aluminum, and can represent the oxide layer (for ground surface of thin wall and thick plate samples). It can also stand for the corrosion products formed due to the localized corrosion attack in the case of thin wall samples in as-cast condition. *RCP* and *CCP* stand for the resistance and capacitive behavior of the oxide layer (or the corrosion products).

The fitting results of the two resistances (*RCT* and *RCP*) for thin wall samples, ground and in as-cast condition, and ground thick plate samples, are presented in Figures 16–18, respectively.

**Figure 16.** Fitted EIS parameters of as-cast thin wall samples: (**a**) *RCT* and (**b**) *RCP*.

**Figure 17.** Fitted EIS parameters of ground thin wall samples: (**a**) *RCT* and (**b**) *RCP*.

**Figure 18.** Fitted EIS parameters of ground thick plate (bulk) samples: (**a**) *RCT* and (**b**) *RCP*.

Values of *RCT* of thin wall samples in as-cast condition (Figure 16a) decrease with the immersion time for all the samples, which indicates an increase of the activity and a higher corrosion rate in the localized attack [62]. It should be noted that this decrease is immediate for the samples taken from near the gate, while for the samples taken from near the vent, it starts after 5–6 h of immersion. These values are higher for the samples near the vent, compared to the samples near the gate. Regarding the effect of silicon content, samples with lower amounts of silicon show slightly higher *RCT* values at the very first hours of immersion. The values of *RCP* (Figure 16b), that for these samples represent the resistance of corrosion products in the pits, are very low, and slightly decrease for all the samples due to the corrosion attack.

Regarding the ground thin wall samples, *RCT* values are generally one order of magnitude higher compared to the thin wall samples in as-cast condition. The values increase during immersion time for all the samples (Figure 17a). The values of *RCT* for different samples, with regard to the position, and with different amounts of silicon, are close to each other. In the case of ground samples, values of *RCP* represent the presence of the passive oxide layer. They remain almost constant for all of the samples, and show a similar trend.

In the case of ground thick plate (bulk) samples, reported in Figure 18, the behavior of 2.5 RGB and 4.5 RGB samples are very similar. After some initial fluctuation, values of *RCT* remain almost constant for the two samples from 12 to 24 h of immersion, but never reach the high resistance showed by the ground thin wall samples. *RCP* values increase from the first hour of immersion for sample 4.5 RGB.

SEM micrographs and map analysis of some selected corroded surfaces are shown in Figure 19. As clearly visible in Figure 19, corrosion is mainly localized in all the samples, and takes place especially in the eutectic region, at the interface between the silicon and aluminum and at the interface of aluminum grains and iron-rich intermetallic particles. These results are in accordance with results of other researchers about corrosion of Al–Si alloys in similar solutions [34,35].

**Figure 19.** (**a**) SEM micrograph and elemental map analysis of corroded surface of sample 2.5 RG; (**b**) Al; (**c**) Si; (**d**) Fe; and (**e**) O.

#### **4. Conclusions**

In this paper, the corrosion behavior of the alloy in different parts of the component geometry were compared, and the microstructure macrosegregation due to Rheo-HPDC process was evaluated in terms of corrosion properties for two aluminum alloys containing 2.5 wt % and 4.5 wt % silicon. These alloys are prone to localized corrosion in the eutectic region at the interface between iron-rich intermetallic particles and the aluminum. Therefore, segregation of these phases influences the electrochemical behavior of the component. It was shown that the samples taken from different positions and different parts with various thicknesses differ in the amount of α1-Al and α2-Al particles. Samples also show a transverse macrosegregation. α1-Al particles which are the solid fraction of the slurry, tend to segregate at the center of the samples, while the surface is richer in α2-Al. Longitudinal segregation induces higher fraction of α1-Al particles in the area nearer to the higher fraction of α2-Al and the eutectic phase near the vent.

Both kinds of these segregations were shown to have influence on the corrosion behavior. A big change in the corrosion resistance was shown by grinding the samples. This is due to the as-cast surface morphology and composition, and to higher concentrations of intermetallic particles on the surface. In as-cast surface conditions, samples with higher amounts of α2-Al and finer microstructure (near the vent) show slightly higher corrosion resistance.

Differently to most of the previous papers, the as-cast surface was tested. It was shown how the surface segregation and the increase of intermetallic particles on the very surface influence the pitting resistance of the component. Both ground thin wall and thick plate samples possess better corrosion resistance compared to the thin wall samples in as-cast condition. This improvement is due to the better surface condition. Nevertheless, relatively thinner samples show a higher corrosion resistance compared to the thicker samples. Surfaces of the rheocast component present a purely liquid microstructure due to the segregation, which makes them more resistant to corrosion when ground and therefore without surface defects.

**Acknowledgments:** This research work was partially supported by the KK-foundation (CompCast, Project No. 20100280) which is gratefully acknowledged. The authors would like to thank COMPtech AB, Sweden for the production of the component and technical support.

**Author Contributions:** Maryam Eslami and Caterina Zanella conceived and designed the experiments; Maryam Eslami performed the experiments; Maryam Eslami, Mostafa Payandeh, Flavio Deflorian, Anders E. W. Jarfors and Caterina Zanella analyzed the data; Flavio Deflorian contributed reagents/materials/ analysis tools; Maryam Eslami, Mostafa Payandeh and Caterina Zanella wrote the paper.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Tribological Behavior of Nano-Sized SiCp/7075 Composite Parts Formed by Semisolid Processing**

**Jufu Jiang 1,\*, Guanfei Xiao 1, Ying Wang <sup>2</sup> and Yingze Liu <sup>1</sup>**


Received: 26 December 2017; Accepted: 17 February 2018; Published: 25 February 2018

**Abstract:** The tribological behavior of the rheoformed and thixoformed nano-sized SiCp/7075 composite parts is investigated. The semisolid stirring temperature has a little influence on the friction coefficient and wear resistance of the rheoformed composite parts. As for the thixoformed composite parts, the average value of the steady-state coefficient of friction increases firstly and then decreases with increasing reheating temperature. Higher wear resistance is achieved at a reheating temperature of 580 ◦C. The average value of the steady-state friction coefficient of the rheoformed composite parts varies from 0.37 to 0.45 upon applied loads of from 20 to 50 N. Weight loss increases slightly upon an increase of applied load from 20 to 40 N. An applied load of 50 N leads to a significant increase of the weight loss. The wear rate decreases firstly and then increases with increasing applied load. As for the thixoformed composite part, the average value of the steady-state friction coefficient and the weight loss decreased with an increasing applied load. However, the wear rate decreases firstly with increasing applied load and then increases. As for the rheoformed composite part, the average value of the steady-state friction coefficient decreases firstly and then increases a little with increasing sliding velocity. Weight loss and wear rate show a first increase and a followed decrease with increasing sliding velocity. As for the thixoformed composite part, the average value of the steady-state friction coefficient shows a decrease with increasing sliding velocity. Weight loss and wear rate exhibit, at first, an increase, and then a decrease with increasing sliding velocity. The average friction coefficient varies from 0.4 to 0.44 with increasing volume fraction of SiC. Weight loss and wear rate decrease with increasing volume fraction of SiC. An increase in dislocation density around the nano-sized SiC particles and the mismatch of the coefficient of thermal expansion (CTE) between 7075 matrix and nano-sized SiC particles during solidification improve the wear resistance of the composite. The dominant wear mechanisms of the rheoformed and thixoformed composite parts are adhesive wear, abrasive wear and delamination wear.

**Keywords:** nano-sized SiC particle; wear rate; friction coefficient; rheoformed; thixoformed

#### **1. Introduction**

Particle reinforced aluminum matrix composite (PRAMC) has received much attention because of their improved specific strength and modulus, good wear resistance, and modified thermal properties [1–4]. Wear behavior is an important evaluation parameter of the PRAMC. A large amount of research has been focused on tribological behavior (or wear behavior) of the PRAMC. Abdollahi et al. [5] investigated dry sliding tribological behavior of Al2024-5 wt. % B4C nanocomposite fabricated by mechanical milling and hot extrusion and found that mechanical milling and adding B4C increased the wear resistance of the nanocomposite. Kumar et al. [6] reported dry sliding wear behavior of stir cast AA6061-T6/AlNp composite and developed a regression model predicting

wear rate. The results showed that wear rate of cast AA6061/AlNp composite decreased with an increase in the mass fraction of AlN particles, and the regression model could predict wear rate at a 95% confidence level. SiCp/Al-Cu matrix composites were produced by the direct squeeze casting (i.e., liquid aluminum melt is infiltrated into a preform of SiC particles under pressure) method, and their dry sliding properties were examined [7]. It was concluded that the friction coefficient decreased with increasing applied load and sliding velocity. Abrasive wear properties of SiC reinforced aluminum matrix composite produced by compocasting (i.e., casting of a stirred mixture of liquid aluminum and SiC particles) were studied [8]. The results revealed that the matrix hardness had a strong influence on the dry sliding wear behavior of the composite, and the lowest wear rate occurred in the composite with the lowest matrix hardness. Tribological behavior of composites with different aluminum matrix fabricated by squeeze cast was reported [9]. The results revealed that the matrix alloy had no remarkable influence on the tribological performance of the composites at low test loads less than 3 N.

Semisolid processing (SSP) has been widely used in automotive and 3C fields since it was developed by M. C. Flemings and his coworker [10–12]. Two typical technical routes such as rheoforming and thixoforming are included in SSP [13–17]. Rheoforming involves direct forming of semisolid slurries with spheroidal solid grains and liquid phase. In thixoforming process, semisolid billet obtained via solidifying semisolid slurries undergoes reheating and forming. SSP has shown an apparent advantage in dispersing the ceramic reinforcements of composite [18,19]. It was illustrated that SSP was suitable for fabricating PRAMC [20,21]. Therefore, some research has focused on wear properties of the PRAMC fabricated by SSP. Mazahery and Shabani [22] investigated the wear behavior of the sol-gel coated B4C particle reinforced A356 matrix composites and concluded that the wear rate of the composites reinforced with coated B4C was less than that of the matrix alloy and decreased with increasing volume fraction of B4C particles. The research results of wear behavior of the rheocasted SiCp/Al metal matrix composites (MMC) showed that the wear rate of the 11% SiC MMC was higher than that of the 50% SiC MMC [23]. The A356/Al2O3 metal matrix composites were fabricated by conventional stirring and semisolid processing [24]. It was concluded that the volume loss of the composites fabricated by semisolid processing was lower than that of the composites fabricated by conventional casting. The friction and wear of the aluminum alloy reinforced by TiO2 particles fabricated by semisolid stirring was mentioned by Sarajan [25]. The results showed that accumulated volume loss was significantly higher when wear debris was removed by camel brush during dry sliding wear.

The aluminum matrix composite (AMC), reinforced with nano-sized ceramics, has received much attention due to high temperature creep resistance and fatigue life [26–29]. However, research on the wear behavior of the AMC reinforced with nano-sized ceramic particles was little reported. Therefore, the present study aims to investigate the wear behavior of nano-sized SiCp/7075 composite parts formed by SSP and find the influence laws of the process parameters such as the volume fraction of SiC particles, and the applied load and sliding velocity on the wear behavior of the nanocomposite parts.

#### **2. Materials and Methods**

#### *2.1. Fabrication of the Rheoformed and Thixoformed Nanocomposite Parts*

Wrought 7075 aluminum alloys are used as matrix material of the composite. Its chemical composition content contained 6.0 wt. % Zn, 2.3 wt. % Mg, 1.56 wt. % Cu, 0.26 Si wt. %, 0.27 wt. % Mn, 0.17 wt. % Cr, 0.03 wt. % Ti and balance of Al. Nano-sized SiC particles with an normalized average size of 80 nm supplied by Xuzhou Jiechuang New Materials Co. Ltd of China (Xuzhou, China) are used as reinforcement of the composite [30]. The solidus of 546 ◦C and liquidus of 637 ◦C temperatures were achieved by using the Differential Thermal Analyzer (DTA) (Mettler Toledo, Zurich, Switerland) [30]. DTA data were converted into DSC (differential scanning calorimetry) data and then gave a curve of solid fraction vs temperature by integrating the DSC data. The semisolid slurry of

nano-sized SiCp/7075 (this denotes 7075 aluminum matrix composite reinforced by nano-sized SiC particles) composite was fabricated by ultrasonic-assisted semisolid stirring (UASS) method [30,31]. 7075 aluminum alloy was melt and held for 10 min at 650 ◦C. The nano-sized SiC particles were added into the melt. The melt with added nano-sized SiC particles was treated for 10 min via a ultrasonic device with a 2 kw power at a frequency of 20 kHz. Then the 7075 melt was cooled down while stirring to the required semisolid temperatures and stirred isothermally for the required times in order to obtain semisolid slurries. The required temperatures and times were shown in Table 1. Some of the semisolid slurries were directly used as rheoforming the cylinder parts of the composite. The other semisolid slurries were rheoformed into a cylindrical semisolid billet with a diameter of 70 mm and height of 58 mm. These cylindrical semisolid billets were reheated to given semisolid temperatures, soaked and then used as thixoforming cylinder parts (Table 1).


**Table 1.** Experimental procedures of dry sliding wear tests of nano-sized SiCp/7075 composite parts fabricated by rheoforming and thixoforming.

In the rheoforming process, a die was firstly preheated to 300 ◦C and then the semisolid slurries were carried into the die cavity. The upper die moved downwards and kept closed with lower die. The semisolid slurries were filled into the die cavity and rheoformed into the final composite cylinder part under a pressure of 398 MPa. In the thixoforming process, the reheated semisolid billet was carried into the die cavity with a preheated termperature of 400 ◦C. Then the upper die moved down and kept closed with the lower die. The reheated semisolid billet was filled into the die cavity and thixoformed into the final composite cylinder part under a pressure of 398 MPa. For the rheoformed cylinder parts, the process parameters included the semisolid stirring temperatures of 615, 620, 625 and 628 ◦C (corresponding to solid fractions of 0.46, 0.35, 0.30, 0.23 [30]), the stirring time of 20 min, the ultrasonic treatment time of 10 min and the volume fractions of 0, 0.5, 1.0, 1.5 and 2%. For the thixoformed cylindrical parts, the process parameters involved the semisolid stirring of 620 ◦C, the stirring time of 20 min, the ultrasonic treatment time of 10 min, the soaking time of 20 min, the volume fractions of 1.5% and reheating temperatures of 580, 590, 600 and 610 ◦C. The micrograph of the rheoformed and thixoformed parts were presented in Figure 1. As shown in Figure 1, the solid grains of the rheoformed composite cylinder part exhibited smaller deformation along flowing direction as compared to the thixoformed composite part. The TEM (transmission electron microscope) images of the rheoformed and thixoformed parts exhibited a uniform distribution of nano-sized SiC particles in the 7075 alloy matrix. It was attribited to double effect of the acoustic and cavitation created by ultrasonic wave and

controllable viscosity of semisolid slurry [31,32]. In addition, it can be noted that needlelike second phase η-MgZn2 existed in the microstructure of the rheoformed and thixoformed composite parts.

**Figure 1.** Micrographs of the rheoformed and thixoformed nanocomposite parts. (**a**) metallograph of the rheoformed at a stirring temperature of 620 ◦C (**b**) metallograph of the thixoformed at a remelting temperature of 590 ◦C. (**c**) TEM image of the rheoformed part (**d**) TEM image of the thixoformed part.

#### *2.2. Dry Sliding Wear Tests of the Rheoformed and Thixoformed Parts*

The dry sliding wear tests were carried out on a pin-on-disc wear-testing apparatus. The disc was made from GCr15 steel (Chinese National Standard GB/T18254-2016), which exhibited good quenching degree and high hardness (HRC 62) due to containing a large amount of chromium element [33]. Its chemical composition content contained 0.95–1.05 wt. % C, 0.25–0.45 wt. % Mn, 0.15–0.35 wt. % Si, 1.4–1.65 wt. % Cr, less than 0.02 wt. % S, less than 0.025 wt. % P, less than 0.1 wt. % Mo, less than 0.25 wt. % Ni, less than 0.25 wt. % Cu and balance of iron. After the rheoformed and thixoformed cylindrical parts were formed, they were machined into the samples with dimensions of ˟<sup>9</sup> <sup>×</sup> 20 mm for the dry sliding wear tests. The diameter of the disc was 100 mm. The radius of the contact track of the dry slide samples used for their experiments was 27 mm. The *Ra* values of the sample and disc were 3.2 μm. Seven dry sliding wear tests were performed in these experiments, as shown in Table 1.The sample has the same section area (i.e., same contact area between pin sample and wear disc) so that applying force has the same effect of applying force per unit area. All the samples were performed under a wear distance of 1000 m. All dry sliding tests were performed under unidirectional sliding. 2.7 g/cm<sup>3</sup> was used as the density value of the 7075 aluminum alloy samples in order to calculate the wear volume. The density of the composite sample was determined by Archimedes drainage in order to calculate the wear volume of the composite sample. The density values of the composite sample with 0.5, 1.0, 1.5 and 2% SiC were 2.712, 2.716, 2.721 and 2.726 g/cm3.

At the end of each wear test, the surface of the disc was washed by alcohol. The samples were weighed carefully and the weight loss was recorded and used as calculating wear rate of the composite parts. In order to investigate the surface morphology of worn samples and the wear mechanism, the worn surface was examined by scanning electron microscopy (SEM) (FEI, Hillsboro, OR, USA) with an energy dispersive X-ray spectrometer (EDS) (FEI, Hillsboro, OR, USA).

#### **3. Results and Discussion**

#### *3.1. Influence of Semisolid Stirring Temperature on Tribological Behavior of Rheoformed Composite Parts*

Figure 2 shows friction coefficient, weight loss, and wear rate of the rheoformed composite parts at different semisolid stirring temperatures. As indicated in Figure 2a–d, friction coefficient of the rheoformed composite parts at different semisolid stirring temperatures firstly increased significantly and then kept a fluctuation with increasing wear time. After the running-in period, the average value of the steady-state coefficient of friction was calculated as can be seen in Figure 2e. It is noticed that average value of friction coefficient varies from 0.42 to 0.46. It illustrates that semisolid stirring temperature has a little influence on friction of the rheoformed composite parts. As shown in Figure 2f, the weight loss varies from 2.84 × <sup>10</sup>−<sup>3</sup> g to 3.15 × <sup>10</sup>−<sup>3</sup> g. The wear rate was determined as defined by Equation (1):

$$K = \frac{W}{F\_{\text{N}} \cdot S} \tag{1}$$

in which *W* is the wear volume (mm3 ), *F*<sup>N</sup> is the applied load (N), and *S* is the sliding distance (m), as reported by Zhang and Wang [34]. Wear rate is determined according to the data of weight loss (Figure 2g).

The wear rate of the rheoformed composite part at different semisolid stirring temperatures varies from 3.48 × <sup>10</sup>−<sup>5</sup> to 3.86 × <sup>10</sup>−<sup>5</sup> mm3/m·N. Stirring temperature affects the solid fraction of semisolid slurries and further determines the deformation degree of solid grains rheoforming process. However, the deformation degree of solid grain during the rheoforming is obviously lower than that of thixoformed parts. Hence, the results of Figure 2f,g illustrate that semisolid stirring temperature has a little effect on the wear resistance of the rheoformed composite parts.

Figure 3 gives secondary electron (SE) images of the rheoformed composite parts at different semisolid stirring temperatures. As indicated in Figure 3, the delamination and shallow grooves are found in the microstructure of worn surface. Cracks are propagated in both transverse and longitudinal directions due to higher shear force on the sliding surfaces. It led to the loss of material from the worn surface in the form of flakes, as reported by Kumar [6]. In addition, some wear debris was found in the surface microstrucutre of the sample when semisolid stirring temperature was 628 ◦C (Figure 3d). It illustrates that abrasive wear also plays a role in the wear process of the composite.

**Figure 2.** *Cont*.

**Figure 2.** Friction coefficient, weight loss and wear rate of the rheoformed composite parts at different semisolid stirring temperatures: (**a**) friction coefficient at 615 ◦C, (**b**) friction coefficient at 620 ◦C, (**c**) friction coefficient at 625 ◦C, (**d**) friction coefficient at 628 ◦C, (**e**) average friction coefficient, (**f**) weight loss and (**g**) wear rate.

Energy dispersive X-ray (EDX) analysis revealed that some Fe and Cr elements occurred in the microstructure of the sample's surface (Figure 4). It illustrates also that wear debris of the disc made from GCr15 steel was retained on the surface of the composite sample, indicating the occurrence of abrasive wear.

Present oxygen element is due to the fact that wear test was done in an air environment, as reported by Sarajan [25]. Hence, the dominant wear mechanisms of the rheoformed composite parts at different semisolid stirring temperatures involve adhesive wear, abrasive wear and delamination wear.

**Figure 3.** SEM images showing the worn surface morphology of the rheoformed composite parts at different semisolid stirring temperatures: (**a**) 615 ◦C, (**b**) 620 ◦C, (**c**) 625 ◦C and (**d**) 628 ◦C.

**Figure 4.** Energy dispersive X-ray (EDX) analysis of the rheoformed composite parts at a semisolid stirring temperature of 620 ◦C: (**a**) SEM micrograph and (**b**) distribution of elements.

#### *3.2. Influence of Reheating Temperature on Tribological Behavior of Thixoformed Composite parts*

Figure 5 exhibits friction coefficient, weight loss and wear rate of the thixoformed composite parts at different reheating temperatures. The curves of friction coefficient vs time reveal that friction coefficient increases firstly and then keeps a fluctuation with increasing time (Figure 5a–d). The average value of friction coefficient increases firstly and then decreases with increasing reheating temperature. When reheating temperature increases from 580 to 600 ◦C, the average friction coefficient increases from 0.41 to 0.54. As can be seen in Figure 5e, the average value of the steady state coefficient of friction decreases to from 0.54 to 0.50 upon a further increase from 600 to 610 ◦C. Weight loss and wear rate of the thixoformed composite part at reheating temperature of 580 ◦C are lower than those of the thixoformed composite parts at reheating temperatures of 590, 600 and 610 ◦C (Figure 5f,g). It indicates highest wear resistance was achieved in the thixoformed composite part at the reheating temperature of 580 ◦C.

**Figure 5.** Friction coefficient, weight loss and wear rate of the thixoformed composite parts at different reheating temperatures: (**a**) friction coefficient at 580 ◦C, (**b**) friction coefficient at 590 ◦C, (**c**) friction coefficient at 600 ◦C, (**d**) friction coefficient at 610 ◦C, (**e**) average friction coefficient, (**f**) weight loss and (**g**) wear rate.

Plastic deformation of solid grains occurs severely because of low liquid phase fraction. It led to occurrence of more dislocations in solid grains. These dislocations can improve wear resistance of the thixoformed composite parts at 580 ◦C. When reheating temperature is elevated to a reheating temperature above 590 ◦C, the plastic deformation of solid grains decreases due to more liquid phase, leading to decreased dislocations. Therefore, the weight loss and wear rate change slightly when reheating temperatures are 590, 600 and 610 ◦C. In addition, it can be noticed that the weight loss and wear rate of the thixoformed composite parts at 590, 600 and 610 ◦C are close to those of the rheoformed composite parts (Figures 2 and 5).

Delamination, shallow grooves, craters and wear debris were found in the microstructure of worn surface (Figure 6). It illustrates that wear mechanisms of the thixoformed composite parts belong to delamination wear, abrasive wear and adhesive wear.

**Figure 6.** Worn surface morphology of the thixoformed composite parts at different reheating temperatures: (**a**) 580 ◦C, (**b**) 590 ◦C, (**c**) 600 ◦C and (**d**) 610 ◦C.

#### *3.3. Influence of Applied Load on Tribological Behavior of the Rheoformed and Thixoformed Composite Parts*

Friction coefficient, weight loss and wear rate of the rheoformed composite parts presented in Figure 7. Friction coefficient exhibits firstly a significant increase and then a fluctuation with increasing time (Figure 7a–c). Before 5min, friction coefficient shows firstly a significant increase and then fluctuation with increasing time. After 5 min, the friction coefficient exhibits a fluctuation again. The friction coefficient shows a significant increase and fluctuation with increasing time before 5 min. However, it is noted that friction coefficient increase slightly with increasing time after 5 min (Figure 7d). The average friction coefficient varies from 0.37 to 0.45 upon applied loads of from 20 to 50 N. As for applied load of 50 N, the average friction coefficient exhibited the lowest value of 0.37 (Figure 7e). The results achieved by Onat [7] revealed that friction coefficient decreased with increasing applied load. This result is almost consistent with the results achieved by Onat [7], except for the 40 N applied load. This work shows a decrease of friction coefficient upon an increase of applied load from 20 to 30 N. However, average friction coefficient increases with an increase of applied load from 30 to 40 N. Average friction coefficient exhibited a decrease with an increase of applied load from 40 to 50 N again. This may be due to different selected range of applied load. For research of Onat [7], the selected range of applied load is a range from 5 to 15 N. However, this selected range of applied load is from 20 to 50 N. Weight loss increases a little upon an increase of applied load from 20 to 40 N. However, when applied load reached to 50 N, the weight loss increased significantly (Figure 7f).

**Figure 7.** Friction coefficient, weight loss and wear rate of the rheoformed composite parts at different applied load: (**a**) friction coefficient at 20 N, (**b**) friction coefficient at 30 N, (**c**) friction coefficient at 40 N, (**d**) friction coefficient at 50 N, (**e**) average friction coefficient, (**f**) weight loss and (**g**) wear rate.

Wear rate exhibited a first decrease and then a slight increase with increasing applied load (Figure 7g). The surface microstructure is characterized by delamination, wear debris, and shallow grooves (Figure 8). It indicates that wear mechanisms depend on delamination wear, abrasive wear and adhesive wear. Especially, a large area of delamination was noted in the microstructure, indicating a dominant delamination wear.

**Figure 8.** Worn surface morphology of the rheoformed composite parts at different applied load: (**a**) 20 N, (**b**) 30 N, (**c**) 40 N and (**d**) 50 N.

Friction coefficient, weight loss and wear rate of the thixoformed composite part are present in Figure 9. A similar law to the rheoformed composite parts that friction coefficient increases significantly and keeps fluctuation is found in the curves of friction coefficient vs time (Figure 9a–d). The average friction coefficient decreased with increasing applied load (Figure 9e). This result is in agreement with the results achieved by Onat [7] and Zhang and Wang [34]. Weight loss increases with increasing applied load as shown in Figure 9f. It is an agreement with the results obtained by Nartarajan et al. [35].

However, the wear rate does not keep the similar law to weight loss with an increase of applied load. As to wear rate, it firstly keeps a decrease and then increases with increasing applied load. This change in trend is in agreement with the rheoformed composite parts as shown in Figure 9g. Delamination is a dominant characteristic of the surface morphology of the worn sample (Figure 10). It illustrates that wear mechanism of the thixoformed composite under different applied load belongs to delamination wear. In addition, it can be noticed that the delamination area firstly decreases and then significantly increases with increasing load. It is consistent with the change law of the wear rate. Especially at an applied load of 50 N, the delamination area increases significantly, indicating a large weight loss. This is the main reason for increase of the wear rate.

**Figure 9.** Friction coefficient, weight loss and wear rate of the thixoformed composite parts at different applied load: (**a**) friction coefficient at 20 N, (**b**) friction coefficient at 30 N, (**c**) friction coefficient at 40 N, (**d**) friction coefficient at 50 N, (**e**) average friction coefficient, (**f**) weight loss and (**g**) wear rate.

**Figure 10.** Worn surface morphology of the thixoformed composite parts at different applied load: (**a**) 20 N, (**b**) 30 N, (**c**) 40 N and (**d**) 50 N.

#### *3.4. Influence of Sliding Velocity on Tribological Behavior of the Rheoformed and Thixoformed Composite Parts*

Friction coefficient, weight loss, and wear rate of the rheoformed composite parts at different sliding velocities are displayed in Figure 11. When sliding velocity is 0.8 m/s, friction coefficient shows a significant increase and a followed fluctuation with increasing time (Figure 11a), which is similar to those of above mentioned rheoformed and thixoformed composite parts. However, an increase of from 0.8 to 1.2 m/s led to an obvious change. As shown in Figure 11b, three stages such as significant increase, significant decrease, and fluctuation were found in the curve of friction coefficient vs time.

A peak value of friction coefficient presented in the curve. The curve of friction coefficient vs time at a sliding velocity of 1.6 m/s exhibits a different change from those at 1.2 m/s (Figure 11c). Friction coefficient shows a significant increase and then enters a fluctuation stage with increasing time. A little difference is its fluctuation extent is lower as compared with the curve at 0.8 m/s. Average friction coefficient firstly decreases and then increases a little with increasing sliding velocity (Figure 11d). Weight loss shows a first increase and then a followed decrease with increasing sliding velocity (Figure 11e). Wear rate exhibits a similar law to weight loss upon an increase of sliding velocity (Figure 11f). Delamination and shallow grooves were also found in the worn surface microstructure, indicating an adhesive wear and delamination wear (Figure 12).

**Figure 11.** Friction coefficient, weight loss and wear rate of the rheoformed composite parts at different sliding velocity: (**a**) friction coefficient at 0.8 m/s, (**b**) friction coefficient at 1.2 m/s, (**c**) friction coefficient at 1.6 m/s, (**d**) average friction coefficient, (**e**) weight loss and (**f**) wear rate.

**Figure 12.** *Cont*.

**Figure 12.** Worn surface morphology of the rheoformed composite parts at different sliding velocities: (**a**) 0.8 m/s, (**b**) 1.2 m/s and (**c**) 1.6 m/s.

Figure 13 depicts friction coefficient, weight loss, and wear rate of the thixoformed composite parts at different sliding velocities. A law of a significant increase and a fluctuation with increasing time was shown. A little difference among them is the fluctuation extents at 1.2 and 1.6 m/s are lower than that at 0.8 m/s. Average friction coefficient shows a decrease with increasing sliding velocity. This work shows a disagreement with the results obtained by Nartarajan et al. [35]. It may be due to effect of the different material sample and disc. Weight loss and wear rate exhibit a first increase and a decrease with increasing sliding velocity. Delamination and wear debris were found the worn surface microstructure (Figure 14), indicating delamination wear and adhesive wear play a role in the wear process of the thixoformed composite parts at different sliding velocities.

**Figure 13.** *Cont*.

**Figure 13.** Friction coefficient, weight loss and wear rate of the thixoformed composite parts at different sliding velocities: (**a**) friction coefficient at 0.8 m/s, (**b**) friction coefficient at 1.2 m/s, (**c**) friction coefficient at 1.6 m/s, (**d**) average friction coefficient, (**e**) weight loss and (**f**) wear rate.

*3.5. Influence of Volume Fraction of SiC Particle on Tribological Behavior of the Rheoformed and Thixoformed Composite Parts*

Figure 15 represents friction coefficient, weight loss, and wear rate of the rheoformed composite parts with different volume fraction of SiC particles. It is noted that friction coefficient also undergoes a significant increase and a fluctuation with increasing time (Figure 15a–e). It illustrates that the friction coefficient of the matrix 7075 aluminum alloy has a similar change law with those of the composite. The average friction coefficient varies from 0.4 to 0.44, and no obvious law is found in the curve of average friction coefficient with increasing volume fraction of SiC particles (Figure 15f). However, weight loss and wear rate decrease with increasing volume fraction of SiC particles (Figure 15g,h). The weight loss and wear rate of the matrix 7075 aluminum alloy are 3.78 × <sup>10</sup>−<sup>3</sup> <sup>g</sup> and 4.63 × <sup>10</sup>−<sup>5</sup> cm3/m·N, respectively. Upon an addition of 2% nano-sized SiC particles, the weight loss and wear rate reach 2.16 × <sup>10</sup>−<sup>3</sup> g and 2.64 × <sup>10</sup>−<sup>5</sup> mm3/m·N, respectively. The decrease extents are 42.8 and 43.0%, respectively. It illustrates that addition of nano-sized SiC particles can improve significantly the wear resistance of the composite parts. This work is consistent with the results presented in the A356/Al2O3 metal matrix composites by Alhawari et al. [24] and the stir cast AA6061-T6/AlNp composite, as reported by Kumar et al. [6]. The addition of nano-sized SiC particles with low coefficient of thermal expansion (CTE) into the alloy matrix with higher coefficient of thermal expansion led to mismatch of CTE of the composite, as mentioned by Kumar et al. [6] and Zhong et al. [36].

**Figure 15.** *Cont*.

**Figure 15.** Friction coefficient, weight loss and wear rate of the rheoformed composite parts with different volume fraction of SiC particles: (**a**) 0, (**b**) 0.5%, (**c**) 1.0%, (**d**) 1.5%, (**e**) 2.0%, (**f**) average friction coefficient, (**g**) weight loss and (**h**) wear rate.

This contributes to the improved strength and wear resistance of the composite. An increase in the volume fraction of nano-sized SiC particles leads to an increase in dislocation density around the nano-sized SiC particles during solidification, as reported by Gutam and Srivastant [37]. The interaction between nano-sized SiC particles and dislocations also improves the wear resistance of the composite, as reported by Kumar et al. [6]. Delamination and shallow grooves were found in the worn surface microstructure (Figure 16). This shows that wear mechanisms depend mainly on delamination wear and adhesive wear.

**Figure 16.** *Cont*.

**Figure 16.** Worn surface morphology of the rheoformed composite parts with different volume fraction of SiC particles: (**a**) 0, (**b**) 0.5%, (**c**) 1.0%, (**d**) 1.5% and (**e**) 2.0%.

#### **4. Conclusions**

(1) The curves of the friction coefficient vs time exhibit a significant increase and a followed fluctuation with increasing time. As for the rheoformed composite part at the different semisolid stirring temperature, the average value of the steady-state friction coefficient varies from 0.42 to 0.46. This semisolid stirring temperature has a little influence on the friction and wear resistance of the rheoformed composite parts. As for the thixoformed composite parts, the average value of the steady-state friction coefficient increases firstly and then decreases with an increase in the reheating temperature. When the reheating temperature increases from 580 to 600 ◦C, the average value of steady-state friction coefficient increases from 0.41 to 0.54. The average value of the steady-state friction coefficient decreases to from 0.54 to 0.50 upon a further increase from 600 to 610 ◦C. The best wear resistance was achieved in the thixoformed composite part at a reheating temperature of 580 ◦C. The dominant wear mechanisms of the rheoformed and thixoformed composite parts involve adhesive wear, abrasive wear and delamination wear.

(2) As for the rheoformed composite part, the average value of the steady-state friction coefficient varies from 0.37 to 0.45 upon the applied loads of from 20 to 50 N. The weight loss increases slightly upon an increase of applied load from 20 to 40 N. However, it was noted that the weight loss increase significantly when the applied load reached 50 N. The wear rate decreases obviously and then slightly increases with the increasing applied load. Delamination, wear debris, and shallow grooves indicate wear mechanisms depend on delamination wear, abrasive wear, and adhesive wear. As for the thixoformed composite part, the average value of the steady-state friction coefficient decreased with the increasing applied load. Weight loss decreased with the increasing applied load. However, the wear rate firstly decreases with the increasing applied load and then increases.

(3) As for the rheoformed composite part, the average value of the steady-state friction coefficient firstly decreases and then increases a little with the increasing sliding velocity. Weight loss and wear rate show, at first, an increase and followed by a decrease with regard to the increasing sliding velocity. As for the thixoformed composite part, the average value of the steady-state friction coefficient shows a decrease with increasing sliding velocity. Weight loss and wear rate exhibit, at first, an increase and a then decrease with the increasing sliding velocity. Adhesive wear and delamination wear are the dominant wear mechanisms of the rheoformed and thixoformed composite parts at different sliding velocities.

(4) The average value of the steady-state friction coefficient varies from 0.4 to 0.44 and no obvious law is found in the curve of the friction coefficient with the increasing volume fraction of the SiC particles. However, weight loss and wear rate decrease with the increasing volume fraction of the SiC particles. The wear resistance of the composite parts was improved significantly due to the addition of the nano-sized SiC particles. An increase in the dislocation density around the nano-sized SiC particles and the mismatch of the coefficient of thermal expansion (CTE) between the 7075 matrix and the nano-sized SiC particles during solidification improved the wear resistance of the composite.

**Acknowledgments:** This work is supported by the Natural Science Foundation of China (NSFC) under Grant No.51375112.

**Author Contributions:** Jufu Jiang designed most experiments, analyzed the results and wrote this manuscript. Guanfei Xiao and Yingze Liu performed most experiments. Ying Wang helped analyze the experimental data and gave some constructive suggestions about how to write this manuscript.

**Conflicts of Interest:** The authors declare no conflicts of interest.

#### **References**


© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **An Experimental Evaluation of Electron Beam Welded Thixoformed 7075 Aluminum Alloy Plate Material**

#### **Ava Azadi Chegeni and Platon Kapranos \***

Department of Materials Science & Engineering, University of Sheffield; Sir Robert Hadfield Building, Mappin Street, South Yorkshire, Sheffield S1 3JD, UK; avaazadi.922@gmail.com

**\*** Correspondence: p.kapranos@sheffield.ac.uk; Tel.: +44-114-22-25509

Received: 8 November 2017; Accepted: 13 December 2017; Published: 15 December 2017

**Abstract:** Two plates of thixoformed 7075 aluminum alloy were joined using Electron Beam Welding (EBW). A post-welding-heat treatment (PWHT) was performed within the semi-solid temperature range of this alloy at three temperatures, 610, 617 and 628 ◦C, for 3 min. The microstructural evolution and mechanical properties of EB welded plates, as well as the heat-treated specimens, were investigated in the Base Metal (BM), Heat Affected Zone (HAZ), and Fusion Zone (FZ), using optical microscopy, Scanning Electron Microscopy (SEM), EDX (Energy Dispersive X-ray Analysis), and Vickers hardness test. Results indicated that after EBW, the grain size substantially decreased from 67 μm in both BM and HAZ to 7 μm in the FZ, and a hardness increment was observed in the FZ as compared to the BM and HAZ. Furthermore, the PWHT led to grain coarsening throughout the material, along with a further increase in hardness in the FZ.

**Keywords:** 7075 aluminum alloy; thixoforming; post-welding-heat treatment; electron beam welding (EBW)

#### **1. Introduction**

7075 wrought aluminium alloys are used for a wide variety of applications in aerospace and automotive industries due to the outstanding characteristics that they possess, such as high-strength-to-weight ratio, ductility, toughness, low density, and resistance to fatigue [1–4]. Promising fabrication techniques are required to produce high quality and integrity parts for such applications. Hence, semi-solid metal processing as a single step manufacturing method providing good quality near net shape products has been widely employed to aluminium alloys due to the advantages that this technology offers over the conventional casting techniques [5–8].

Weldability of the materials is another important factor in aerospace and automotive industries. However, although Al alloys have in the past been considered as difficult-to-weld materials through conventional arc welding techniques, improvements have removed these difficulties and quite few studies have focused on other technologies that offer improvements of weld performance, such as high-power density fusion joining, namely, laser beam welding (LBW) and electron beam welding [1,9, 10], and, of course, Friction Welding.

Electron-beam welding (EBW) is a fusion welding process, in which a beam of high-velocity electrons is applied to two materials to be joined. The workpieces melt and flow together as the kinetic energy of the electrons is transformed into heat upon impact. EBW is often performed under vacuum conditions to prevent dissipation of the electron beam. Electron beam welding provides high-quality welded joints for a wide range of thicknesses, and can be operated with high welding speeds [11,12]. Using EBW generates low distortions in the Fusion Zone (FZ), together with a narrow Heat Affected Zone (HAZ), and low residual stresses in comparison with conventional welding techniques [12,13]. To take advantage of these features, many studies centred around the investigation of the EBW on

different aluminium alloys. Cam et al. [14] investigated the effects of EBW on mechanical properties and microstructural characterisation of 5005, 2024, and 6061 aluminium alloys, and concluded that a defect free weld line was observed in these alloys. Kocak et al. [12] also evaluated the impacts of EBW on aluminium 7020 alloy, and reported that a loss of hardness in FZ was observed due to the loss of strengthening phases. Currently, there is no reported data published on EBW of thixoformed aluminium 7075 alloy and the aim of this study is to investigate the effect of EBW on the microstructure and mechanical properties of two thixoformed aluminium plates of Al 7075 alloy. These results have been obtained through optical microscopy, Scanning Electron Microscopy (SEM), EDX (Energy Dispersive X-ray Analysis) on the electron beam welded plates, along with hardness measurements. In addition, a post welding heat treatment of the weld zone was conducted at the semi-solid temperature range in order to investigate any microstructural and property changes in the weld material, as well as the parent material. The choice of the Semi-solid range was to replicate the conditions experienced during the thixoforming process that takes place within this temperature range at approximately 50% liquid content.

#### **2. Materials and Methods**

Two thixoformed plates of 7075 wrought aluminium alloy were used as starting materials. The chemical composition of the alloy is presented in Table 1 [1].


**Table 1.** Chemical composition of wrought 7075 Al alloy (wt %) [1].

The plates were welded using electron beam welding with a speed of 1000 mm/min at TWI Ltd. Cambridge (Great Abington, Cambridge, UK). The accelerating voltage and beam current that were used were 130 kV and 21 mA, respectively. The microstructure of the welded plates was investigated using standard optical metallographic methods. Samples that were cut along the length of the weld line and from the parent material were ground with standard SiC grinding paper and polished with 6 and 1 μm monocrystalline diamond suspension and 0.05 μm silica suspension; the specimens were subsequently etched using sodium hydroxide (10 g NaOH diluted with 100 mL water). The microstructures of the Base Metal (BM), HAZ, and FZ were evaluated using a Nikon Eclipse LV150 optical microscope (Nikon, Tokyo, Japan) and TM3030Plus Tabletop Scanning Electron Microscope (Hitachi, Tokyo, Japan). The chemical analysis of the phases was performed using EDX (Energy Dispersive X-ray Analysis, Hitachi, Tokyo, Japan). Image J software (An open platform for scientific image analysis, https://imagej.net/Welcome) was used to measure the average grain size from optical images using the linear intercept method, the shape factor of the solid grains, calculated based on their perimeter and area, as well as the liquid fraction content of the specimens. The Vickers hardness measurements were conducted using a Zwick hardness tester (ZHU250CL, Ulm, Germany) with 10 kgf applied force for 10 s across different locations on the plates, as shown in Figure 1, and the average values were reported. To investigate the influence of post welding heat treatment on the microstructure and mechanical properties of the welded plates, specimens were heat treated by being kept for 3 min at 610, 617, and 628 ◦C, respectively (i.e., within the semi-solid temperature range of the alloy) and fast cooled in water.

**Figure 1.** Illustration of the welded plates and the areas of hardness tests.

#### **3. Results and Discussion**

#### *3.1. Microstructure of the Base Material*

Micrographs of the as-received plate of 7075 thixoformed aluminum alloy are presented in Figure 2; the images are taken from different regions of two plates. When considering Figure 2 from the top left to the bottom right, the microstructure of the alloy is uniform throughout the thixoformed plates, and consists of globular, fine, non-dendritic grains in a solid matrix consisting of the last liquid to solidify.

**Figure 2.** Optical micrographs showing the thixoformed base metal (BM) of 7075 Al Electron Beam (EB) welded plates (**a**–**d**) with typical near spheroidal microstructure at different magnifications.

Scanning Electron Microscopy (SEM) micrographs of the BM are shown in Figure 3a,b. It can be observed that the liquid phase specified with a square was formed at the grain boundaries during the thixoforming process. In addition, the last liquid to solidify from the semi-solid state around the grains is illustrated in Figure 3c. There are two ways by which the liquid can be entrapped inside the grains. First, by segregation of the alloying elements inside the solid grains, leading to the formation of fine liquid droplets during the partial re-melting, and, secondly, when grains are combined to reduce the solid-liquid interfacial energy during the heating stage of the thixoforming process, giving rise to the creation of relatively large liquid droplets contained within the sub-grains. Hence, the base metal consists of alpha-Al, eutectic liquid phase and occasional trapped liquid pools within sub-grains [5].

**Figure 3.** (**a**) Scanning Electron Microscopy image of Al 7075 alloy showing the last liquid phase to solidify at grain boundaries; (**b**) a higher magnification SEM image of the liquid phase; (**c**) optical microscopy image showing the last liquid solidified around the grains.

#### *3.2. Microstructural Evolution after Welding*

Figure 4 illustrates the micrographs of the EB welded plates. As can be seen, the microstructure contains three regions, namely, Heat-Affected Zone (HAZ), Fusion Zone (FZ), and Base Metal (BM), as presented in Figure 4a. Generally, high heat input and preheating are two factors that increase the width of the HAZ in precipitation hardenable aluminum alloys. However, it can be observed that the HAZ is relatively narrow due to the low heat input during the electron beam welding process [9,14,15]. In addition, it can be said that although aluminum 7075 alloy is prone to cracking, the EBW process did not pose any significant problems, and only few numbers of pores were formed in the FZ, as shown in Figure 4b. Possible reasons for the formation of these pores could be the high specific energy density and evaporation of metal that is associated with EBW and thermophysical features of aluminum,

including its low melting point, high thermal conductivity, and surface oxide films, with high melting points [9,14,16]. Figure 4c,d show the weld area at different magnifications. As can be observed, the fusion zone consists of a fine grain microstructure with a significant reduction in the average grain size, as compared to the BM and HAZ. As the total heat input into the material during the EBW is lower than that of other fusion welding techniques due to the higher power density of the EBW process, a finer microstructure in the FZ can be typically obtained using EBW. The difference between the grain size in FZ and HAZ is presented in Figure 4e.

**Figure 4.** (**a**) Optical micrographs of Al 7075 EB welded plates, (**b**–**d**) show details of the Fusion Zone (FZ) and (**e**) detail of the plate weld boundary.

Figure 5a shows graphically the grain size as a function of the position from the weld zone. It can be seen that the mean grain size in the BM is around 67 μm, and this number is approximately the same in the HAZ, which shows a consistency of the microstructure in the thixoformed plate. However, the grain size in the FZ is significantly reduced to around 7 μm, which is considerably smaller than that of either the BM or HAZ. There are several reasons for the reduction of the grain size after welding. First, the presence of the alloying elements that precipitated out at the grain boundaries impedes the severe growth of the grains. In addition, the fast cooling of the joint after the welding process can also prevent grain growth in the FZ [17]. Furthermore, due to the high welding rates of the EB process that are caused by the high melting speeds of the focused heat source, the time that is required for the welding to be accomplished is also reduced so that the grains do not have enough time to grow during the EBW [12]. Figure 5b demonstrates a graph of the shape factor against the distance from the weld area. The shape factor represents the circularity of the grains, which has a maximum of one for a totally spherical grain and zero for a grain with a complex shape. As can be seen from the graph, more spherical grains can be observed in the BM when compared to the HAZ and FZ, since the material was

exposed to the high temperatures during the EBW, leading to the deviation from the spherical grains that are present in the BM. Measurements were taken across various positions in the different zones, and there is clearly a deviation between the spheroidicity of the different grains, as expected as the FZ has undergone melting that destroyed the original non-dendritic, near spheroidal microstructure of the base material.

**Figure 5.** (**a**) Graph of grain size as a function of position, (**b**) Graph of shape factor against position.

Results of the EDX point analysis conducted for the last liquid to solidify phase at the boundaries and grains in the BM, as well as the grains in the FZ are shown in Figure 6 and Table 2. From the graphs, it can be said that the liquid phase at the boundaries mainly consisted of aluminum, magnesium, copper, and zinc, implying the presence of the alloying elements in these regions. In addition, it can be observed that aluminum was by far the highest constituent of the grains in the base metal, whilst the percentage of the other elements was low within the grains, which confirms that the alloying elements precipitated out at the boundaries. Moreover, aluminum is the main element in the FZ, with almost the same content as that of the BM. However, EDX point analysis reveals that the matrix phase in FZ contains less Zn than the BM matrix, whereas the Mg and Cu contents of the FZ are higher than that of the BM. Cam et al. [9] suggested that the heat input during the fusion welding processes may give rise to the evaporation of the solute atoms with low melting points in the fusion zone, hence, the lower amount of zinc in this region can be attributed to the evaporation mechanism due to the high amounts of heat applied during welding. The loss of Mg/Zn/Cu was also reported by other researchers [12,14]. Furthermore, rapid cooling after the EBW process leads to the presence of super saturated amount of Mg and Cu in the FZ. It is worth adding that the EDX data for the liquid phase at the boundaries in the FZ could not be measured due to the low amount of the last liquid to solidify phase that remained at the boundaries after EBW.

**Figure 6.** Results of Energy Dispersive X-ray Analysis (EDX) point analysis for: (**a**) grains in BM, (**b**) eutectic phase at boundaries, (**c**) grains in FZ.


**Table 2.** Weight percent of alloying elements at different points of microstructure. Reported numbers are the average of at least three attempts.

The amount of the last liquid to solidify phase on quenching was calculated using image J software, and a plot of liquid fraction versus distance from the weld centre is indicated in Figure 7. The graph represents a downward trend from the BM to FZ, which suggests that the fraction of eutectic, which is around 35% in the thixoformed plate, is higher than this amount in both HAZ and FZ, which is 22% and 16%, respectively. The higher percentage of the eutectic phase in the BM can be related to the nature of the semi solid metal processing technique, which typically contains between 30–50% liquid. On the other hand, the lower amount of the eutectic phase in the FZ can be attributed to the complete melting in the FZ during the EBW, followed by the fast cooling of this region. It is worth pointing out that the actual amount of the eutectic phase in the BM is higher than the above mentioned number, as can be seen from the micrographs, since some of the liquid is entrapped within the grains and the software could not measure it during the calculations. We will use the terms eutectic solid and last liquid to solidify interchangeably as the terminology used in the Semi-solid forming usually refers to the eutectic solid as liquid fraction.

**Figure 7.** Liquid fraction against distance from the centre of the weld for Al 7075 alloy EB welded plate.

#### *3.3. Post Weld Heat Treatment*

Micrographs of EB welded specimens following heat treatment at 610, 617, and 628 ◦C are presented in Figure 8. According to Figure 8a, the microstructure of the base metal and the heat affected zone at different temperatures consisted of fine equiaxed solid grains that are uniformly distributed throughout the material; which is similar to that of the welded material before the heat treatment. The differences between the grain size of the FZ and HAZ can be observed from Figure 8b. Micrographs of the fusion zone for heat-treated samples, at the three temperatures, are displayed in Figure 8c, as compared to the microstructure of the joint before the post weld process. The grain structure in the FZ has coarsened after heat treatment at three temperatures. In addition, there is a noticeable change in the morphology in this region, in that grains are more spherical in the FZ when

compared to the as-received grain structure. This can be explained due to the post weld heat treatment in the semi-solid temperature range to which the material was subjected.

**Figure 8.** Optical microscopy micrographs of Al 7075 alloy EB welded plates after heat treatment at three temperatures: 610, 617 and 628 ◦C. (**a**) as received plate material, (**b**) HAZ and plate boundary and (**c**) FZ weld area. (Magnifications top from left to right in μm: 200, 500, 500, 500); (magnifications middle from left to right in μm: 100, 200, 100, 100); (magnifications bottom from left to right in μm: 20, 50, 50, 50).

A graph representing the grain size of the BM, HAZ, and FZ for the heat-treated materials at 610, 617, and 628 ◦C is shown in Figure 9a. From the graph, it can be seen that the mean grain size in the BM, HAZ, and FZ follows a similar trend, with a slight reduction in the grain diameter at 617 ◦C, followed by a rise in the average grain size at 628 ◦C. It is worth adding that grains started to grow in all three regions of specimens, including BM, HAZ, and FZ when compared to the grain structure before the heat treatment. This is due to the exposure of samples to high temperatures during the heat treatment and holding at these temperatures for three minutes, resulting in the grain coarsening. Moreover, the average grain size in the BM is slightly higher than that of the HAZ, but the mean grain diameter in the FZ is by far lower than those in both HAZ and BM, as shown in the micrographs of Figure 8b. Figure 9b shows the shape factor measurement for the heat-treated specimens. As can be seen, there is a significant change in the shape factor of particles in the FZ when compared to before heat treatment. During the heat treatment, solid grains in the FZ became more spherical as they were heated up to the thixoforming temperature range and held for three minutes at these temperatures. The shape factor of grains in the HAZ at the three temperatures is almost the same. In addition, the circularity of the particles in the BM has experienced an upward trend with temperature increase.

**Figure 9.** (**a**) Graph of grain size against temperature, (**b**) Plot of shape factor as a function of the three temperatures used, 610, 617, and 628 ◦C for Al 7075 alloy EB welded plates after heat treatment.

#### *3.4. Vickers Hardness Test*

Results of the Vickers hardness test performed on the EB welded plates of the thixoformed 7075 aluminium alloy are summarised in Figure 10. As can be observed, the graph shows a relatively symmetric shape, and the hardness number in the BM of two plates is roughly the same, since these regions belong to the parent material with similar properties and fabrication method, either side of the weld. In addition, a rise in hardness can be observed from both plates toward the FZ. As already mentioned, evaporation of zinc was observed after welding in the FZ, however, this amount was not considerable (0.307 wt % Zn depletion based on numerical values in Table 2) and can be eliminated, so it did not cause hardness reduction in this region. Moreover, based on the results of the EDX point analysis presented before, the weight percent of Mg and Cu increased in the FZ as compared to the BM as a result of rapid cooling that occurred after the welding process. Therefore, unlike the results provided by several researchers about some alloys of aluminium, including 5005 and 2024 series showing a hardness reduction in the fusion zone due to the loss of strengthening elements, such as Cu and Mg in this area [14,18], herein the hardness increment can be attributed to the surplus amount of these strengthening elements in the FZ, supported by researchers that have reported the hardness increase that is obtained in the FZ by using appropriate filler wire during the welding process to compensate for the evaporation of elements [9].

**Figure 10.** Graph of hardness versus position from the weld centre for Al 7075 alloy EB welded plates.

Figure 11 illustrates Vickers hardness over distance from the weld centre after heat treatment at 610, 617 and 628 ◦C. The overall behaviour of the three graphs is similar, demonstrating a peak in the FZ, and as the distance from the weld centre increases, a drop in the hardness can be observed, implying property improvement in the FZ. In comparison with the graph of the EB welded plate before heat treatment in the FZ, a rise in hardness can be seen after heat treatment at the three temperatures, with hardness increasing from 144 HV before to 156, 184, and 203 HV after heat treatment at 628, 610, and 617 ◦C, respectively, showing a maximum at 617 ◦C. After heat treatment, grain growth has occurred in the FZ of the material, and, hence, a drop in hardness was expected due to grain coarsening. However, the hardness increase can be attributed to the refined grain size, as well as chemical changes in the FZ after heat treatment as different diffusion coefficients of the critical elements in the semi-solid state locally can explain the movement of atoms from the Al-lattice at these high temperatures. In addition, the hardness number at the edges of both heat-treated plates is approximately 125 HV, which is roughly the same as that in the parent material before heat treatment. It can be concluded that although heat treatment improved the hardness in the FZ, it did not change the properties of the base metal.

**Figure 11.** Graph of hardness versus position from the weld centre for Al 7075 alloy EB welded plates after heat treatment at three temperatures: 610, 617 and 628 ◦C.

Work done by various researchers on Laser hardening, [19] has shown that the most relevant parameter in the hardened layer depth is the scanning speed, followed by the hardened track width. Another group of researchers have given a good account of modelling such processes by modelling the Added Layers by Coaxial Laser Cladding [20].

#### **4. Conclusions**

Electron beam welding was employed on thixoformed plates of 7075 aluminium alloy, along with a post welding heat treatment in semi-solid temperature range in order to evaluate the resultant microstructures and mechanical properties in the weld area. The following conclusions were inferred from the work:


**Acknowledgments:** Technical staff at the University of Sheffield for their support in various analytical techniques used in this work.

**Author Contributions:** For research articles with several authors, a short paragraph specifying their individual contributions must be provided. The following statements should be used: Ava Azadi Chegeni and Platon Kapranos conceived and designed the experiments; Avi Azadi Chegeni performed the experiments; Avi Azadi Chegeni and Platon Kapranos analyzed the data; Avi Azadi Chegeni and Platon Kapranos wrote the paper.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Metals* Editorial Office E-mail: metals@mdpi.com www.mdpi.com/journal/metals

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18