Influence of Impurities on the Hot Shortness of Brass Alloys ^†

Abstract

Brass alloys are critical for numerous modern applications. However, a significant knowledge gap exists regarding the impact of impurities on their processability, particularly regarding hot cracking susceptibility. This issue is exacerbated by increasing recycling rates, leading to a higher concentration of impurities in the alloy pool. This study employed computational thermodynamics to assess the impact of common impurities in brasses. Based on the results, the combined impact of impurities and alloying content significantly shifted the α and β solvus to the left, the α solvus indicatively is shifted by approximately 10% at low temperatures, and introduced multiple solvus and liquidus lines, crucial for the hot formability of the alloy. These preliminary results provide insight on the need for process optimization, considering the possibly deleterious effects of some elements to minimize the risk of hot cracking, a step towards enhancing the reliability and sustainability of these alloys.

1. Introduction

Brass alloys, composed primarily of copper (Cu) and zinc (Zn), are widely utilized in engineering and manufacturing due to their good combination of chemical, mechanical, and physical properties [1,2,3]. Brass alloys are classified into three main types, alpha (α), beta (β), and alpha-beta (α+β) brasses, based on the main microstructural components, usually given as a function of their Zn content. Specifically, alpha and beta brasses are single-phase alloys with a face-centered cubic (FCC) and body-centered cubic (BCC) crystal structure, respectively [2,3]. Alpha-beta brasses consist of a dual-phase (α+β) microstructure of both FCC and BCC crystal structure [2,3].

Apart from the two main components, brasses also contain both other alloying elements and impurities. The presence of impurities such as P, As, and Bi can influence/alter/affect their properties and formability window. A critical issue that usually arises is the hot shortness, or hot cracking, which is a phenomenon during which the material becomes brittle and prone to cracking at elevated temperatures. This is driven by the presence of low-melting-point impurities or phases that segregate at grain boundaries. When segregated, they weaken the interatomic bonds, causing dislocation emission and/or decohesion [4]. Hot cracks are defined by the literature as inter-granular or inter-dendritic defects, meaning that they form between grains/dendrites; they are attributed to a combination of metallurgical effects and structural loads [5] and is considered to be one of the most severe defects [6]. This type of failure can be caused either by liquid metal embrittlement (LME) and/or solid metal-induced embrittlement (SMIE). Thus, hot cracks can be classified as follows [5]: (i) solidification cracks, (ii) liquation cracks, or (iii) ductility dip cracks (DDC) that form in the solid state after the manufacturing process has ended.

These phenomena are further enhanced by recent trends and challenges dictating the need to increase recycling rates of copper, for the successful implementation of the Green Transition, while retaining or enhancing properties like machinability and formability [2]. However, there is a limited availability of literature data regarding Cu-alloys, partly because the execution of extensive physical experiments is a resource intensive process, both in terms of costs and time requirements. Currently, the published work on this issue is rather limited. Some of the available studies have focused on Phosphorus-P, that improves corrosion resistance, and machinability by forming (Mn,Fe)₂P and restricting γ-phase (Cu₅Zn₈) due to the formation of the eutectic Cu₃P. However, it is referred that even an addition of 50 ppm (0.005%) can prevent hot cracking [7,8,9,10,11]. Bismuth-Bi, due to its low melting point and large atomic size, causes segregation and embrittlement. Specifically, the morphology of the segregation varies according to Bi content; >1.2 wt% forms bulk Bi phase whereas <0.8 wt% forms a thin film at grain boundaries [12,13,14,15].

The scope of this work is to fill this literature gap via computational engineering methods. This could allow for process optimization based on thermodynamical predictions of the microstructural components and behavior, minimizing the risk of hot cracking and ultimately enhancing the reliability and sustainability of these essential alloys.

2. Methodology

Computational thermodynamic simulations were chosen to evaluate the impact of alloying and impurity content in the phase diagram of a brass alloy. The experiments were conducted according to the CalPhaD methodology via ThermoCalc [16], utilizing TCCU6 and MOBCU5 databases.

Firstly, the binary Cu-Zn diagram was calculated and was used as a reference point. Then, the pseudo-binary diagram Cu-Zn-Pb was calculated in order to highlight the impact of Pb, a crucial element for enhanced machinability of traditional brasses. This was followed by calculations of pseudo-binary systems with (i) basic alloying elements (those that have a defined quantity limit in the relevant standard, i.e., Copper (Cu), Zinc (Zn), Lead (Pb), Iron (Fe), Tin (Sn), Nickel (Ni), and Aluminum (Al) and (ii) common impurities (those that count cumulatively towards the impurity limit, i.e., Phosphorus (P), Chromium (Cr), Bismuth (Bi), Cobalt (Co), Silver (Ag), Silicon (Si), Manganese (Mn), Cadmium (Cd), Selenium (Se), Arsenic (As), excluding Sulfur (S), Tellurium (Te), and Antimony (Sb) due to limitations of the current software database). The composition of elements in all calculated diagrams was selected in accordance with international standards, namely the BS EN 12164:2016 [17] for CW614 alloy. The impurity composition that was used in the calculations can be found in

Table 1. Composition according to EN12164:2016 standard and utilized in calculations.

Element	EN12164:2016	Utilized
As	-	1.0
Sn	0.3	0.30
P		0.2
S		-
Te		-
Se		0.02
Sb		-
Cr		0.2
Ni	0.3	0.20
Fe	0.3	0.30
Al	0.05	0.05
Bi		0.02
Pb	2.5–3.5	3.2
Zn	REM	REM
Co		0.2
Ag		0.2
Si		0.3
Mn	-	1.0
Cd		0.2
Cu	57.0–59.0	58.2

.

3. Results and Discussion

Figure 1 (binary) and Figure 2 (pseudo-binary) show the Cu-Zn phase diagram before and after the addition of Pb, respectively. In Figure 2, three different compositions of Pb are portrayed. Namely, these are the minimum (2.5% wt.) and maximum (3.5% wt.) and their medium (3% wt.), as mentioned in the standard. It can be seen that Pb content does not significantly alter the α and β solvus, yet it introduces the low temperature Pb solidus. It is well known that the existence of this low temperature solidus, in combination with the low solubility it has in Cu, is tied to machinability enhancement, since it promotes Pb segregation at grain boundaries leading to chip breakage.

On the other hand, Figure 3 and Figure 4 show the Cu-Zn phase diagram with the addition of basic alloying elements and maximum impurity content, respectively. The alloying content (Figure 3) shifts the diagram lowering the solidus by approximately 50 °C and creates additional solvus lines, mostly for BCC transformations and intermetallic compounds. Maximum impurity content (Figure 4) demonstrates a notably higher level of complexity, and its utilization requires a further extensive study of: (i) the impurities content, both individually in pseudo binary diagrams and (ii) possible para-equilibrium eutectic phase formation due to segregation effects. Moreover, the nature of the various phases present should also be addressed. Specifically, the BCC_B2 #1 phase refers to a Ni-Al or Fe-Ni rich (80%) phase, with the presence of Mn and Co (about 8% each) while the BCC_B2 #2 is the typical β-phase found in brasses consisting of Cu-Zn. Regarding the FCC_L₁₂ phases, the first one which forms at low temperatures is the solidified fraction of the segregated Pb, while the second one refers to the α-phase of brasses. Similarly, the low temperature solidus lines correspond to a liquid mixture of mainly Pb-As with some Cu additions while the first one corresponds to the totality of the system.

4. Conclusions

This research indicates that alloying elements and impurities have a significant effect on the system solidus and liquidus lines, as well as on the α-β solvus. Specifically, the alloying elements’ content shifts the initial diagram, lowering the solidus (by approximately 100 °C) and introducing additional solvus lines, mostly for intermetallic phases and elements with low solubility to the copper matrix. Moreover, the impurity content was also found to have a significant impact on the system’s behavior, shifting the α and β solvus (indicatively, α solvus exhibits a leftward shift of approximately 10% in low temperatures, accompanied by a change in its slope) and introducing liquidus lines of the low melting elements, possibly increasing the alloys’ hot cracking susceptibility.