Correlations Between Cooling Rate, Dendritic Spacing and Hardness in a Brass Alloy for a Variety of Cooling Conditions

Abstract

As-cast Cu-Zn components can be manufactured under a wide range of solidification conditions, reflected in varying cooling rates. While industrial processes are well-established, the relationships between grain size (GS), dendritic spacing (DS), and cooling rates in Cu-Zn alloys remain barely explored. Such knowledge becomes fundamental since these microstructural features significantly influence mechanical properties such as hardness. Under this context, the present study investigates the solidification behavior of a Cu-30 wt.% Zn alloy through a combination of SEM-EDS, optical microscopy, XRD, and CALPHAD (Calculation of Phase Diagrams) computation. Systematic measurements of average GS and DS have been conducted on samples solidified under unidirectional solidification (US) conditions (slow) and centrifugal casting (CC) in Cu molds (rapid cooling). The ratios between GS and DS have been analyzed and correlated with hardness variations as a function of the solidification rates. Additionally, SEM-EDS and CALPHAD analyses elucidate the formation of phases and Zn segregation patterns under varying cooling conditions. Appropriate approaches to compute cooling rates have been used so that ranges from 0.60 to 0.95 K/s, and from 15 to 47 K/s have been determined for US and CC samples respectively. These findings provide valuable insights into the microstructural evolution and mechanical property optimization of cast brass alloys.

Keywords:
Cu-Zn; Brass; microstructure; grain size; dendritic spacing; solidification

1. Introduction

Copper and its alloys are among the most important commercial metals, valued for their exceptional properties. These include excellent electrical and thermal conductivities, outstanding corrosion resistance, ease of fabrication, and suitable strength. Copper and copper alloys are generally nonmagnetic, making them suitable for specialized applications. They can be readily soldered and brazed, and many alloys in this family are compatible with various welding and foundry techniques¹.

Copper and its alloys are typically categorized into six main families: pure coppers, dilute-copper (or high-copper) alloys, brasses, bronzes, copper nickels, and nickel silvers. Pure copper, while offering superior conductivity, is challenging to cast due to issues such as surface cracking, porosity, and internal cavities. To address these limitations, small amounts of alloying elements like beryllium, silicon, nickel, tin, zinc, and chromium have to be added to enhance its casting characteristics. Larger additions of alloying elements are used to further improve mechanical and physical properties². Due to their suitable corrosion resistance, Cu-Zn alloys (brasses) are widely used in pipes, valves, and fittings for various systems. Among them, α-β phase brass offers an optimal balance of superior mechanical performance and durability, making it highly suitable for both technological and service applications³.

To optimize the performance of as-cast manufactured brass components based on processing variables, understanding and tailoring a comprehensive set of properties is essential. An effective approach to influence the component's performance is by controlling the microstructural morphology through adjustments in processing conditions, i.e., controlling the dendritic spacing, the phase morphologies and segregation. Several studies have highlighted the strong correlation between the dendritic spacing and the resulting properties in metals⁴-⁶.

The dendritic array is characterized by primary and secondary spacing arrangements and the segregated products greatly affect the mechanical properties and homogenization kinetics of solidified alloys. Therefore, to control the properties of casting materials, it is important to understand the solidification parameters that affect the growth of dendritic spacings during solidification. For several cast alloys, dendrite arm spacing decreases with increasing cooling rates. For cooling rates between 1 and 10 K/s, the primary dendrite arm spacing (PDAS) ranges from approximately 50 to 200 µm, while the secondary dendrite arm spacing (SDAS) is about 10 to 50 µm. At cooling rates of 10 to 100 K/s, the PDAS decreases to 20 to 100 µm, and the SDAS ranges from 5 to 20 µm. When cooling rates exceed 100 K/s, the PDAS further reduces to 5 to 50 µm, with SDAS becoming less than 5 µm¹. This general knowledge has been proved mainly for Sn-based⁷,⁸ and Al-based⁹,¹⁰ alloys. Although these relationships exist for various alloy systems, for Cu brass and Fe-based alloys, such information is almost nonexistent, not only regarding dendritic spacing (DS) but also grain size (GS). Data on both parameters remain essential for benefiting the control and planning of casting based on cooling rates.

While direct studies on rapidly solidified Cu-Zn alloys at cooling rates exceeding 10 K/s are scarce, the general principles observed in other alloy systems can be extrapolated. Implementing high cooling rates in the casting of Cu-Zn alloys is expected to produce a finer dendritic structure, which can enhance hardness and strength. Although specific experimental data for Cu-Zn alloys at cooling rates above 10 K/s are limited, general metallurgical principles and studies on similar alloys suggest that such high cooling rates would lead to a refined microstructure with reduced dendrite arm spacing. Further experimental research focused on Cu-Zn alloys under these conditions would provide more definitive insights. Cu-Zn alloys are cast in many types of processes such as sand, investment, permanent mold, centrifugal, and die castings¹. These processes produce castings with different cooling rates, producing non-equilibrium conditions. However, there is limited knowledge about the effects of cooling rates under various cooling conditions across these processes for brasses.

One of the key aspects for controlling and designing Cu-brass components is understanding the dependencies and correlations between grain size (GS) and dendritic spacing (DS) for this alloy, which is already established, for instance, for Al-2017¹¹, Al-6201¹², and Al-319¹³ alloys. Silva et al.¹¹ established that GS/DS ratios were approximately 32 and 15 for the 2017 and 2017-NbB alloys, respectively. This GS/DS ratio exceeded 20 over the whole range of cooling rates in the 6201 alloy¹². Gomes et al.¹³ demonstrated that for conventional dendritic structures in the 319 alloy, the grain size differs greatly from the primary dendritic spacing (PDS), but for refined (with Al 5 wt.%Ti–1wt.%B inoculant) structures, the relationship is direct, in fact PDS = 0.9997 GS, or simple PDS = GS. Regarding the GS/DS ratio, a value of approximately 9 was obtained for the 319 alloy. Therefore, GS and DS are considered key microstructural features and must be explored in Cu-brass alloys.

To clarify the contribution of the present investigation, the significance of GS and DS microstructural parameters lies in their influence on the control of mechanical properties. Both parameters, while reflecting coarsening, directly impact the hardening mechanisms in brasses. According to Fan et al.¹⁴, the stress–strain behavior of 70/30 brass with varying grain sizes, evaluated through standard room-temperature tensile tests, demonstrates that fine-grained specimens exhibit higher flow stress than coarse-grained specimens, in accordance with the Hall–Petch effect. Specifically, the fine-grained brass shows a yield strength (YS) of 111 GPa and an ultimate tensile strength (UTS) of 503 MPa, whereas the coarse-grained brass presents a YS of 78 GPa and a UTS of 376 MPa. Moreover, Iqbal et al.¹⁵ demonstrated that Vickers hardness exhibited a decreasing trend with increasing pouring temperature in a 70/30 brass alloy. The increase in pouring temperature may extend the solidification time, reduce the cooling rate and promote dendritic growth, which in turn contributes to the reduction in hardness.

The Cu-30Zn alloy is a common brass, used in industries such as automotive, marine, construction, and instrumentation. Understanding its solidification helps improve casting and machining processes. The aim of this research is to investigate the coarsening associated with the DS and GS in the Cu-30 wt.% Zn alloy under slow and rapid cooling regimes. Moreover, segregation of Zn has been assessed for a range of cooling rates. Solidification insights are gathered so that the comprehension of microstructure and hardness on this alloy could be advanced. The impact of cooling rate and microstructure coarsening is explored on the alloy hardness, while the ratio between GS and DS could be assessed and verified under quite different solidification conditions.

2. Materials and Methods

The Cu-30 wt.% Zn alloy was fabricated using a conventional vacuum induction melting (VIM) furnace (Inductotherm, VIP Power-Trak model). Commercially pure Zn (99.997 wt.%) and Cu (99.92 wt.%) were melted in a graphite crucible, followed by a 10-minute homogenization of the molten bath before solidification. The composition of Zn element is as follows (wt. %): Al 0.0001%, Cu 0.0002%, Sn 0.0001%, Cd 0.0006%, Zn 99.9967%, Pb 0.0026%, Fe 0.0001%. The Cu composition is Sb 0.0004%, Pb 0.0005%, Fe 0.001%, Ni 0.001%, Sn 0.0005%, S 0.0015%, Ag 0.0025%. The very low ppm levels of these elements are unlikely to interfere with solidification or act as grain refiners. To compute the alloy's solidification path, the Thermo-Calc software was employed, offering phase diagram calculations using both the Lever Rule and Scheil models¹⁶. The Lever Rule assumes complete equilibrium, while the Scheil model represents the opposite extreme, assuming no diffusion in the solid phase. This dual-model approach facilitated a comprehensive analysis of the alloy solidification.

The alloy was solidified using centrifugal casting in a stepped Cu mold to achieve rapid solidification. The process utilized a Linn High Therm GmbH Titancast 700-VAC system. Rods with diameters of 5, 7, 9, and 11 mm were produced in duplicate to provide sufficient material for microstructural and microhardness analysis. The alloy was melted in situ, with precise temperature control during heating. Once the target melt temperature was reached, the heating system was deactivated, and the mold was rotated, causing the molten alloy to fill the stepped cavities and naturally solidify.

For the directional solidification experiments, the molten metal was poured into a stainless-steel mold, and once the melt temperature exceeded the liquidus temperature by 5%, the electric heaters were turned off. At this point, water flow at the base of the container was initiated to start the cooling system. This setup ensured that cooling commenced as soon as the melt reached the desired temperature. The water-cooled base triggered the onset of solidification and maintained the system until the casting was fully formed. The mold, manufactured with the AISI 304 stainless steel, was cylindrical and longitudinally split, with a height of 160 mm, an internal diameter of 60 mm, and a wall thickness of 5 mm. The carbon-steel base plate liquid-contact surface was polished to a 1200 mesh finish.

For metallographic analysis, the samples were polished and etched with a solution of FeCl₃, HCl, and H₂O (5 g: 50 ml: 100 ml) for 10 seconds. The following positions (P) have been examined: 1.2 mm, 10.2 mm, 22.2 mm, 52.8 mm, 61.8 mm, and 73.8 mm. Optical microscopy was used to examine the microstructures, focusing on the dendrite spacing. The grain size (GS) was calculated using the Heyn linear intercept method¹⁷ after images were taken using a stereomicroscope. The primary and secondary dendritic spacing (DS) was measured through the triangle and intercept techniques¹⁸. A minimum of 40 measurements for each selected position have been carried out, obtaining mean and standard deviations. What it is referred to as SDAS may, in fact, also represent higher order spacings, such as tertiary arms. However, this does not invalidate the analysis of the dendritic network coarsening, nor any of the analyses presented here.

The dendritic morphology has been further analyzed with a Philips XL-30 FEG SEM equipped with an energy-dispersive X-ray spectroscopy (EDS) system, which provided phase compositions and insights into the effects of cooling rates on the final liquid composition. Thermodynamic calculations were performed using the CALPHAD method with the TCSLD3 database. Vickers microhardness tests were performed on the transverse sections of the rods using a 200 gf load and a dwell time of 15 seconds. The tests ensured that the indentation encompassed all microconstituents, offering a representative measurement of the overall microstructure. Hardness values were determined as an average of at least 20 measurements per sample.

X-ray diffraction (XRD) was performed using a Bruker D8 Advance ECO system to identify the phases present. Scans were conducted over a 2θ range of 5° to 90°, with a step size of 0.02° and a count time of 0.25 seconds per step, using Cu Kα radiation (λ = 1.5406 Å). This analysis provided comprehensive phase identification to support microstructural and hardness evaluations.

3. Results and Discussion

3.1. Solidification analysis and microstructures

The optical microscopy images in Figure 1 show the cross-sectional view, revealing a more cellular-type morphology with a refined array at position 22.2 mm and a more dendritic structure associated with the coarser microstructure at position 61.8 mm.

Figure 1
Optical micrographs showing examples of the cellular/dendritic arrays observed in the US Cu-30Zn samples at the positions (a) 22.2 mm and (b) 61.8 mm from the metal/mold interface.

The coarsening of the dendritic array is quite noticeable when comparing positions 22.2 mm and 61.8 mm from the water-cooled bottom, as shown in Figure 1, the microstructure at 61.8 mm is observed to be slightly coarser. Sections closer to the cooled base of the casting, such as 22.2 mm, tend to solidify more quickly since the thermal resistance of the solidified layers originating from the base of the casting is lower compared to more distant sections or positions, such as 61.8 mm¹⁹.

Both the α-phase and the β-phase can be observed forming the microstructure in Figure 1, which is expected to be a single a-phase (FCC) in equilibrium. This reveals that Zn must have segregated into the liquid ahead the solid-liquid interface in which the β-phase (BCC - darker regions in the images in Figure 1) solidified as the last portion of the liquid. Similar trends were also demonstrated by Korojy and Fredriksson²⁰. Additionally, some bright spots of the β’ phase appear to have precipitated after solidification, probably from the β-phase, as previously demonstrated by Zhou²¹.

Figure 2a shows the phase evolution of the Cu–Zn alloy as a function of temperature in equilibrium. At temperatures above approximately 950 °C, the alloy is fully liquid. As the temperature decreases between about 950 °C and 925 °C, the liquid phase progressively transforms into a solid FCC_L12 (α) phase. Below approximately 925 °C, the alloy is completely solid, predicting a consistency entirely of the α phase. In contrast, the Scheil diagram in Figure 2b illustrates the non-equilibrium solidification, which initiates at approximately 955 °C with the formation of a primary α-phase from the liquid. As the solid fraction increases, the composition of the remaining liquid becomes progressively enriched in Zn. At approximately 900 °C, a secondary BCC_B2 (β−phase) precipitates, leading to a two-phase solid structure (α + β) upon completion of solidification near 870 °C. As such, the solidification of the β-phase as observed in the microstructures in Figure 1 under non-equilibrium conditions may be confirmed by observing the Scheil solidification path in Figure 2b.

Figure 2
(a) Equilibrium and (b) Scheil solidification diagrams for the Cu-30 wt.% Zn alloy.

The equiaxed (randomly oriented) grains that form as the different sections of the alloy solidify are shown in Figure 3. Images corresponding to six distinct positions, (i.e., cooling rates) can be examined. Thus, a comprehensive analysis of this parameter can be conducted with a focus on its coarsening evolution along the casting length. The average GS values also proved to be significant, ranging from 4,067 μm to 6,020 μm. The GS tends to increase with decreasing the cooling rate, exhibiting a certain sensitivity to it. An increase of approximately 50% in the GS is observed when comparing the base region (1.2 mm) to the top (73.8 mm) of the casting. As there are no specific studies on the evolution of grain size for brass alloys, a comparison with Al alloys is the most feasible. An increase of approximately 100% has been reported for directional solidification experiments of the Al-2017 and Al-6201 alloys¹¹,¹² with parameterization and a furnace similar to those used in the present analysis. It, thus, appears that the brass alloy analyzed here is less sensitive to variations in solidification conditions in terms of their impact on the final GS.

Figure 3
Optical micrographs showing the grain morphology observed in the unidirectionally solidified (US) Cu-30Zn samples. P: Position from the casting bottom. Grain sizes (in micrometers) for the six different samples: 4068.0 μm (1.2 mm), 3886.0 μm (10.2 mm), 4107.0 μm (22.2 mm), 6020.0 μm (52.8 mm), 6013.0 μm (61.8 mm), 5881.0 μm (73.8 mm).

3.2. Dendritic microstructure scale

To predict the cooling rates associated with the primary dendritic spacing (PDS) measured for the US brass alloy, mathematical modeling results provided by Brito et al.²² have been adopted. This work involved the study of the solidification of Zn-Cu alloys and was deemed the most appropriate for establishing the cooling rate values. The mathematical model determined the following relationship for cell spacing: CS=55 (Ṫ)^-0.55 (CS: μm and Ṫ: K/s). Since the reported model focused on cellular growth, it has been corrected by a factor of 4 to be representative for the dendritic growth, as demonstrated by Rosa et al.²³. The spacing between the primary trunks is typically much greater, as they have lateral branches that push the trunks further apart during solidification, which is not the case for the cells. This adjustment by a factor of 4 reflects a reasonably realistic ratio of the spacing difference between cellular and primary dendritic structures, resulting in the appropriate correlation for this study, being: PDS=220 (Ṫ)^-0.55 (PDS: μm and Ṫ: K/s).

While a highly cellular and dendritic pattern with few lateral branches was observed, only PDS data could be measured. Figure 4 shows the variation of primary dendritic spacing as a function of position along the solidified casting and cooling rate. The averages and experimental deviations are shown. Using the correlation allowed the estimation of the cooling rates for the US processing of the brass alloy, ranging from approximately 0.95 K/s to 0.60 K/s for PDS between 225 μm and 282 μm, as can be seen in Figure 4.

Figure 4
Primary dendritic spacing as a function of the (a) position, and (b) cooling rate in the US Cu-30Zn alloy.

Figure 5 shows the dendritic network formed during solidification, characterized by tree-like structures, while Figure 6 shows the larger, regular grain structures, where atoms are organized in distinct crystallographic orientations.

Figure 5
Optical micrographs showing the dendritic arrays observed in the rapidly solidified (RS) Cu-30Zn samples at the following section diameters: (a) 5 mm, (b) 7 mm, (c) 9 mm, (d) 11 mm.

Figure 6
Optical micrographs showing the grain morphology observed in the RS Cu-30Zn samples at the following section diameters: (a) 5mm (GS=82.2 μm), (b) 7mm (GS=117.6 μm), (c) 9mm (GS=125.8 μm), (d) 11mm (GS=162.6 μm).

The presence of primary dendrites has been characterized as non-aligned ones as can be seen in Figure 5 for the brass alloy being produced through CC. Therefore, due to this misalignment, in conjunction with the high number of observed randomly developed secondary branches, only SDAS was measured in these samples. A prevalence of a fully dendritic and significantly refined microstructure was observed compared to the microstructure related to the US processed samples in Figure 1.

The variation in the diameter of the CC sample resulted in coarsening with an increase in diameter, due to the larger volume of liquid, and the consequently thinner copper mold wall. This translates into a reduction in the solidification rate as the diameter increases. The rapid solidification of the single-phase brass alloy also showed some β-phase formation in the interdendritic regions, as can be observed in Figure 5.

As expected, the grain size decreased significantly with rapid cooling, as observed in Figure 6 compared to Figure 3. For the slower solidified samples shown in Figure 3, the grain size varied between 4,000 and 6,000 µm. In contrast, the rapidly solidified samples exhibited a grain size range from 82 mm to 163 mm, representing an approximate reduction of 40 times. It is noteworthy that grain size remains dependent on the cooling rate even under rapid cooling conditions, such as those achieved via CC.

A variation of SDAS from 11.0 to 17.0 mm was observed for the brass alloy due to the decrease in the cooling rate, as can be seen in Figure 7a and Figure 7c, that shows the SDAS shifts as a function of each of the different parameters. It has been demonstrated that the primary factor influencing the SDAS is the local solidification time²⁴-²⁶, expressed as t_SL​=ΔT/(G⋅V)​, where ΔT represents the temperature difference between the liquidus and solidus, and G and V are temperature gradient and solidification velocity, respectively. Typically, the relationship between spacing and time is described by an equation of the form: SDAS=K₁. t_SL ^a, where “a” and K₁ are parameters associated to a particular alloy system. The experimental results can be normally compared with predictions from two models; the theoretical model due to Feurer²⁵ and the empirical model due to Grugel²⁶.

Figure 7
Secondary dendritic spacing as a function of the (a) diameter, (b) local solidification time (t_SL) and (c) cooling rate in the rapidly solidified centrifuged Cu-30Zn alloy.

Ares et al.²⁴ determined the experimental relationship between SDAS and time, and found SDAS=9.2. t_SL ^1/3 for a Cu-36%Zn alloy. This correlation has been applied to the present results, obtaining the t_SL data in Figure 7b.

Considering that the cooling rate: Ṫ = G. V = ΔT/t_SL, one can determine the cooling rate since DT and t_SL are known. ΔT can be extracted from the Scheil diagram in Figure 2, being approximately 93 K. Therefore, the SDAS x cooling rate profile could be plotted in Figure 7c. As expected, much higher cooling rates have been calculated for CC samples (Figure 7) as compared to those in Figure 4.

3.3. Phase composition, grain/dendritic spacing correlations and hardness

Although the Scheil equation can predict the accumulation of Zn and the formation of an interdendritic phase in a brass alloy that is typically single-phase, experimental validation and the impact of cooling rate are important complements to this investigation. Figure 8 shows results of SEM-EDS points obtained from the dendritic and interdendritic regions related to the Zn content and shows the corresponding EDS signal count spectra. It can be observed, therefore, from the Zn content vs. cooling rate graph in Figure 8, a certain stability at higher levels as a function of the cooling rate. Furthermore, as predicted by the Scheil model, the interdendritic region indeed forms the β phase, showing a higher Zn content.

Figure 8
SEM-EDS data (Zn content) at the dendritic and interdendritic portions in the rapidly solidified Cu-30Zn alloy. It includes an image with one example of the SEM-EDS image points, and two typical registered EDS count spectra at the dendritic matrix (# 1), and interdendritic region (# 4).

According to Hong et al.²⁷, in Cu-Zn alloys, the β phase emerges when the zinc content exceeds approximately 35% wt.%. This phase is characterized by a body-centered cubic (BCC) crystal structure and is typically harder and stronger than the α phase. However, the β-phase composition values measured by the EDS were lower. It is important to highlight that EDS is a semi-quantitative technique, and the measurements are limited by the interaction volume of the electron beam, which extends beneath the sample surface. This interaction results in signal mixing from adjacent phases and subsurface regions, thereby reducing the spatial resolution and the accuracy of the compositional analysis.

The differences in EDS measurements between the α and β phases remained between 4 and 5 wt.%, which is quite reasonable considering the difference between alloy nominal composition and the 35% predicted by Hong et al.²⁷ for the β-phase.

Two XRD spectra show the spectra obtained for the Cu-Zn alloy in the centrifugal cast condition can be seen in Figure 9. These results identified only the α phase, as the β phase fraction is relatively limited to be counted. In other words, there is a significant predominance of the α phase under CC conditions.

Figure 9
X-ray diffraction pattern for the CC Cu-30%Zn alloy. Two spectra registered at two different locations are presented to confirm the repeatability of the duplicate experiments.

An interesting relationship in castings is the GS/DAS ratio. Ratios between GS and DAS can be seen for the two types of processing in Figure 10. It is observed that this ratio is significantly higher in the case of the US alloy, as seen in Figure 10. Grains grow substantially under this condition because the diffusion time is much longer compared to the rapid solidification in centrifugal casting with a copper mold. In directional solidification, the unidirectional heat flow promotes the alignment of dendrites along specific crystallographic orientations, leading to a pronounced growth texture. This structure alignment can result in multiple dendrite arms coalescing into larger grains, thereby increasing GS to DAS ratio.

Figure 10
Grain size / DAS ratios considering both (a) US, and (b) CC Cu-Zn samples.

It is fundamental to monitor this ratio, which is around ~ 20 for cooling rates below 1 K/s and between 7 and 9 for rates above 14 K/s. Considering that both microstructural parameters can affect mechanical properties, the dimensional relationship between them is of great importance for casting design of brasses.

Figure 11 shows the data on the average and standard deviations of microhardness as a function of DAS. Although microhardness was not significantly altered by the solidification condition in both the US and CC cases, there is a certain tendency for an increase in hardness with increasing cooling rate and a reduction in interdendritic spacing, as shown in Figure 11. As expected, the microhardness of the CC samples was slightly higher (~15% higher), varying from 66 HV to 69 HV.

Figure 11
Microhardness Vickers variation with the dendritic spacing in the (a) slowly and (b) rapidly solidified Cu-30Zn alloy.

Cooling rate decreases as solidification progresses, influencing the final as-cast microstructure, which consists of α-phase dendrites with β-phase in the interdendritic interstices. The hard β-phase reinforces the ductile α-phase, enhancing hardness. This effect is particularly pronounced in samples with finer phases, such as those generated through CC, where microstructural refinement further strengthens the alloy. Indeed, the dendritic boundaries can act as barriers to dislocation movement, contributing to increased hardness.

As dislocations move through the α-phase matrix, they accumulate at dendritic and phase boundaries, increasing resistance to deformation through indentation. The harder β-phase at interdendritic regions further restricts dislocation motion, strengthening the alloy through a combination of boundary strengthening and second-phase hardening²⁸.

4. Conclusions

The following conclusions can be drawn from the present investigation:

• Non-equilibrium solidification of the Cu-Zn alloy reveals that Zn likely segregated into the liquid ahead of the solid-liquid interface, where the β-phase (BCC) solidified as the final portion of the liquid. As a result, a two-phase structure was formed during solidification: α-phase (FCC) and the β-phase. The microstructural analysis of the Cu-30Zn samples revealed a moderate coarsening of the dendritic array as the solidification front progressed. The observed increase in grain size with decreasing cooling rate aligns with prior studies on Al alloys, although the brass alloy studied here appears to exhibit lower sensitivity to variations in solidification conditions (i.e., variations in cooling rate).

• The XRD data confirmed the predominance of the α-phase, suggesting that the β-phase fraction remained limited. These findings highlight the importance of experimental validation in conjunction with thermodynamic simulations to better understand phase transformations in the Brass alloys.

• CC significantly reduced the grain size, with values decreasing from thousands of micrometers in the US samples to below 200 µm in the rapidly solidified ones. This reduction of approximately 40 times underscores the effectiveness of rapid solidification in refining microstructure. These results emphasize the importance of thermal management in controlling microstructural features during solidification.

• The grain size to dendritic spacing (GS/DAS) ratio exhibited a strong dependence on the solidification conditions, being significantly higher in unidirectional solidification due to prolonged diffusion times. For cooling rates below 1 K/s, this ratio was approximately 20, whereas for rates above 14 K/s, it decreased to values between 7 and 9. Due to the direct influence of these parameters on the mechanical properties, systematically monitoring their interrelationship is essential for optimizing the casting design of brass alloys.

5. Acknowledgments

The authors are grateful to CNPq – National Council for Scientific and Technological Development, Brazil and to FAPESP – São Paulo Research Foundation, Brazil (grant number #2023/06107-3) for their financial support. This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001.

• Data Availability
  Data presented in this study are available on request from the corresponding author.

6. References

• ¹ Fredriksson H, Åkerlind U. Solidification and crystallization processing in metals and alloys. New York: John Wiley & Sons; 2012. http://doi.org/10.1002/9781119975540
  » http://doi.org/10.1002/9781119975540
• ² Silva TEF, Amaral RL, Paiva RM, Reis ARL, Jesus AMP. On the influence of lead in the hot workability of brass alloys. Mater Proc. 2022;8(1):15. http://doi.org/10.3390/materproc2022008015
  » http://doi.org/10.3390/materproc2022008015
• ³ Atsumi H, Imai H, Li SF, Kousaka Y, Kojima A, Kondoh K. Microstructure and mechanical properties of high strength brass alloy with some elements. Mater Sci Forum. 2010;654-656:2552-5. http://doi.org/10.4028/www.scientific.net/MSF.654-656.2552
  » http://doi.org/10.4028/www.scientific.net/MSF.654-656.2552
• ⁴ Leal JRS, Saldanha FE, Ganju E, Spinelli JE, Gouveia GL. Effect of cooling rate on AlFe primary and eutectic phase growth evolution in an Al-2Fe-1Mn alloy. J Alloys Compd. 2025;1010:177870. http://doi.org/10.1016/j.jallcom.2024.177870
  » http://doi.org/10.1016/j.jallcom.2024.177870
• ⁵ Carvalho CC, Sobral BS, Sousa RB, Paixão JL, Dantas SLA, Spinelli JE, et al. Effects of different zinc content on solidification, microstructure, and mechanical properties in tin–bismuth alloy. Adv Eng Mater. 2024;26(21):2401074. http://doi.org/10.1002/adem.202401074
  » http://doi.org/10.1002/adem.202401074
• ⁶ Dias JMS, Bogno AA, Spinelli JE, Oliveira R, Cheung N, Garcia A, et al. Microstructure and hardness of an Al–8 wt%si–2.5 wt%bi alloy subjected to solidification cooling rates from 0.1 to 800 K s⁻¹ Adv Eng Mater. 2023;25(3):2201060. http://doi.org/10.1002/adem.202201060
  » http://doi.org/10.1002/adem.202201060
• ⁷ Rocha OL, Siqueira CA, Garcia A. Heat flow parameters affecting dendrite spacings during unsteady-state solidification of Sn-Pb and Al-Cu alloys. Metall Mater Trans, A Phys Metall Mater Sci. 2003;34(4):995-1006. http://doi.org/10.1007/s11661-003-0229-3
  » http://doi.org/10.1007/s11661-003-0229-3
• ⁸ Tong X, Beckermann C, Karma A. Velocity and shape selection of dendritic crystals in a forced flow. Phys Rev E Stat Phys Plasmas Fluids Relat Interdiscip Topics. 2000;61(1):R49-52. http://doi.org/10.1103/PhysRevE.61.R49
  » http://doi.org/10.1103/PhysRevE.61.R49
• ⁹ Ceschini L, Morri A, Morri A, Gamberini A, Messieri S. Correlation between ultimate tensile strength and solidification microstructure for the sand cast A357 aluminium alloy. Mater Des. 2009;30(10):4525-31. http://doi.org/10.1016/j.matdes.2009.05.012
  » http://doi.org/10.1016/j.matdes.2009.05.012
• ¹⁰ Reyes RV, Bello TS, Kakitani R, Costa TA, Garcia A, Cheung N, et al. Tensile properties and related microstructural aspects of hypereutectic Al-Si alloys directionally solidified under different melt superheats and transient heat flow conditions. Mater Sci Eng A. 2017;685(1):235-43. http://doi.org/10.1016/j.msea.2016.12.096
  » http://doi.org/10.1016/j.msea.2016.12.096
• ¹¹ Silva AS, Sousa SMA, Gouveia GL, Garcia A, Spinelli JE. The influence of NbB inoculation on dendritic spacing and grain size of an Aluminum 2017 alloy at different cooling rates. Int J Adv Manuf Technol. 2023;125(11-12):5681-96. http://doi.org/10.1007/s00170-023-11104-x
  » http://doi.org/10.1007/s00170-023-11104-x
• ¹² Sousa SMA, Gouveia GL, Spinelli JE. Evaluating grain size, dendritic scale, and tensile properties of a NbB-inoculated 6201 alloy using solidification rate. Mater Sci Eng A. 2022;835:142680. http://doi.org/10.1016/j.msea.2022.142680
  » http://doi.org/10.1016/j.msea.2022.142680
• ¹³ Gomes IP, Barros A, Brollo GL, Silva CAP, Cheung N, Zoqui EJ. The effects of cooling rate on microstructure formation during solidification of 319 alloy. Int J Met Cast. 2024;18(3):2079-91. http://doi.org/10.1007/s40962-023-01145-z
  » http://doi.org/10.1007/s40962-023-01145-z
• ¹⁴ Fan R, Magargee J, Hu P, Cao J. Influence of grain size and grain boundaries on the thermal and mechanical behavior of 70/30 brass under electrically-assisted deformation. Mater Sci Eng A. 2013;574:218-25. http://doi.org/10.1016/j.msea.2013.02.066.
• ¹⁵ Iqbal, Ali N. Husin H, Akhyar, Khairil, Farhan A. Differences in pour temperature affect hardness properties of CuZn brass alloy through metal casting. IOP Conf Series Mater Sci Eng. 2021;1082(1):012001. http://doi.org/10.1088/1757-899X/1082/1/012001
  » http://doi.org/10.1088/1757-899X/1082/1/012001
• ¹⁶ Andersson JO, Helander T, Höglund L, Shi P, Sundman B. Thermo-calc & DICTRA, computational tools for materials science. Calphad. 2002;26(2):273-312. http://doi.org/10.1016/S0364-5916(02)00037-8
  » http://doi.org/10.1016/S0364-5916(02)00037-8
• ¹⁷ ASTM: American Society for Testing and Material. ASTM E112-13: standard test methods for determining average grain size. West Conshohocken: ASTM International; 2015.
• ¹⁸ Gündüz M, Çadirli E. Directional solidification of aluminium-copper alloys. Mater Sci Eng A. 2002;327(2):167-85. http://doi.org/10.1016/S0921-5093(01)01649-5
  » http://doi.org/10.1016/S0921-5093(01)01649-5
• ¹⁹ Kurz W, Fisher JD. Fundamentals of solidification. Bäch: Trans Tech Publications; 1992.
• ²⁰ Korojy B, Fredriksson H. On solidification and shrinkage of brass alloys. Int J Cast Met Res. 2009;22(1–4):183-6. http://doi.org/10.1179/136404609X367623
  » http://doi.org/10.1179/136404609X367623
• ²¹ Zhou P. An in situ kinetic investigation of the selective dissolution mechanism of Cu alloys [thesis]. Paris: Université Pierre et Marie Curie; 2017.
• ²² Brito C, Siqueira CA, Spinelli JE, Garcia A. Cellular growth during the transient directional solidification of Zn-Rich Zn-Cu monophasic and peritectic alloys. J Phys Chem Solids. 2012;73(9):1173-81. http://doi.org/10.1016/j.jpcs.2012.05.014
  » http://doi.org/10.1016/j.jpcs.2012.05.014
• ²³ Rosa DM, Spinelli JE, Ferreira IL, Garcia A. Cellular/dendritic transition and microstructure evolution during transient directional solidification of Pb-Sb alloys. Metall Mater Trans, A Phys Metall Mater Sci. 2008;39(9):2161-74. http://doi.org/10.1007/s11661-008-9542-1
  » http://doi.org/10.1007/s11661-008-9542-1
• ²⁴ Ares AE, Caram R, Schvezov CE. The effect of solidification parameters on dendrite spacing in unidirectional solidification. In: TMS Annual Meeting & Exhibition; 2004 Mar 14-18; Charlotte. Proceedings. Pennsylvania: TMS; 2004. p. 751-65.
• ²⁵ Feurer U. Influence of alloy composition and solidification conditions on dendrite arm spacing, feeding and hot tearing properties of aluminium alloys. In: The Symposium on Quality Control of Engineering Alloys; 1977 Mar 3-4; Delft. Proceedings. Delft: Technical University of Delft; 1977. p. 131-45.
• ²⁶ Grugel RN. Secondary and tertiary dendrite arm spacing relationships in directionally solidified Al-Si alloys. J Mater Sci. 1993;28(3):677-83. http://doi.org/10.1007/BF01151244
  » http://doi.org/10.1007/BF01151244
• ²⁷ Hong HL, Wang Q, Dong C, Liaw PK. Understanding the Cu-Zn brass alloys using a short-range-order cluster model: significance of specific compositions of industrial alloys. Sci Rep. 2014;4(1):7065. http://doi.org/10.1038/srep07065
  » http://doi.org/10.1038/srep07065
• ²⁸ Zhang P, An X, Zhang Z, Wu S, Li S, Zhang Z, et al. Optimizing strength and ductility of Cu-Zn alloys through severe plastic deformation. Scr Mater. 2012;67(11):871-4. http://doi.org/10.1016/j.scriptamat.2012.07.040
  » http://doi.org/10.1016/j.scriptamat.2012.07.040

Edited by

• Associate Editor:
  Aloisio Klein.

• Editor-in-Chief:
  Luiz Antonio Pessan.

Data availability

Data presented in this study are available on request from the corresponding author.

Publication Dates

• Publication in this collection
  18 Aug 2025
• Date of issue
  2025

History

• Received
  03 Feb 2025
• Reviewed
  27 Apr 2025
• Accepted
  29 June 2025