Study on Machinability of Lead-Free α + β Brass 62.5Cu–1Si–Zn Alloy

Abstract

Free-cutting brass contains 3% lead to obtain excellent machinability, but because lead is harmful to the human body, lead regulations are being tightened in Europe, starting with the 2010 and 2014 lead control laws for drinking water-related equipment in the United States, and including RoHS regulations, ELV regulations, and the 4MS Positive List. Being affected by these increasingly strict lead regulations in Europe and the United States, demand for brass with significantly reduced lead content is on the rise.

In order to meet the needs of the times, the authors have developed the world’s first lead-free, free-cutting α + β brass “62.5Cu–1Si–Zn alloy” with a lead content of less than 0.1%.

As a result of investigating the machinability of this alloy by comparing it with CW510L, it was demonstrated that the quality of machined products made of the 62.5Cu–1Si–Zn alloy is good due to reduced cutting resistance realized by shear type chips that can be generated by the action of Si solid-solubilized in the β-phase, the presence of fine P compounds, and the effect of fine α-crystal grains.

 

This Paper was Originally Published in Japanese in J. Japan Institute of Copper 61 (2022) 213–217.

Fig. 4 Appearance, cross-sectional microstructure and thickness of chips in turning, and HV hardness of center of chip cross section.

1. Preface

Free-cutting brass contains 3% lead to obtain excellent machinability, but because lead is harmful to the human body, lead regulations are being tightened in Europe, starting with the 2010 and 2014 lead control laws for drinking water-related equipment in the United States, and including RoHS regulations,¹⁾ ELV regulations,²⁾ and the 4MS³⁾ Positive List. Being affected by these increasingly strict lead regulations in Europe and the United States, demand for brass with significantly reduced lead content is on the rise. Among such brasses, EN standard alloy CW510L has been distributed as a brass containing lead up to 0.2%. However, they point out that this alloy has problems related to cutting resistance and cutting quality.⁴⁾

In order to meet the needs of the times stated above, the authors have developed the world’s first lead-free, free-cutting alpha-beta brass “62.5Cu–1Si–Zn alloy” with a lead content limited to less than 0.1%.⁵⁾ This paper compares the newly developed alloy with CW510L and substantiates the superiority of the alpha-beta brass in machinability.

2. Experiment Method

Drawn ϕ25-mm bars of 62.5Cu–1Si–Zn and C3604 made by mass production facility were prepared. For comparison, a commercially available ϕ35-mm CW510L bar was also prepared. Chemical components of the test specimens are listed in Table 1.

Table 1 Chemical composition of specimens.

For each test specimen, tensile test, hardness measurement, observation of microstructure, and machinability evaluation were conducted. For the tensile test, the test specimens were processed according to the specification of Test Piece No. 10 of JIS Z2241, a test method of the Japan Industrial Standard, then subjected to a tensile test in one trial. To examine the hardness, Vickers hardness (hereinafter, “HV hardness”) test was conducted in accordance with JIS Z2244. The hardness was measured at three different locations for each of the test specimens, and the average of the measurements was regarded as the HV hardness of the material. For the observation of microstructure, metallographic micrographs were taken with a 500X metallographic microscope. Machinability was evaluated based on the results of a turning test using a lathe and a drilling test in which holes having a depth three times the diameter were made. Table 2 and Fig. 1 respectively show conditions of the cutting tests and illustrations that depict how the tests were conducted. In the testing, cutting resistance (three component forces) and cutting power were measured with the dynamometers attached to the cutting machines. Machinability index was calculated according to the following formula: cutting resistance (cutting power when drilled) of C3604/cutting resistance (cutting power when drilled) of the test specimen × 100. Regarding the turning test specimens, pictures of the appearance as well as micrographs of cross-sectional microstructure after the test were taken at the location circled with a dashed line in Fig. 1(a). Further, appearance and cross-sectional microstructure of the chips generated during turning and drilling were observed, their thickness was measured, and HV hardness was measured at the center of a cross section indicated with a painted circle in Fig. 2. Regarding surfaces of the test specimens after turning, surface roughness was measured, and SEM images were taken.

Table 2 Cutting test condition.

Fig. 1

Schematic diagram of cutting test.

Fig. 2

Chip thickness measurement method and HV hardness measurement position (left), and surface roughness measurement position (right).

3. Test Results and Observation

3.1 Turning

Table 3 shows the results of the tensile test conducted on the test specimens. Compared with CW510L, the 62.5Cu–1Si–Zn alloy exhibited equivalent elongation of about 25%, but its tensile strength and 0.2% proof stress were approximately 10% higher, about 560 MPa and about 380 MPa respectively.

Table 3 Mechanical property of specimens.

Figure 3 shows changes in cutting resistance (principal cutting force) over the duration of the turning test. Table 4 shows the results of the cutting tests. The cutting resistance of the 62.5Cu–1Si–Zn alloy was constant from the beginning to the ending of the test duration, being approximately half the level of CW510L, and the machinability index was 89, about twice the level of CW510L.

Fig. 3

Time variation of principal cutting force during turning.

Table 4 Cutting test results.

Figure 4 shows the appearance, the cross-sectional microstructure, and the thickness of the chips generated during the turning test, as well as the HV hardness of the center portions of their cross sections. Chips of the 62.5Cu–1Si–Zn alloy, like those of C3604, were finely broken, and their cross sections had a shear-type shape. On the other hand, chips of CW510L were continuous and spiral, and their cross section had a flow-type shape. The thickness of the chips of C3604 and that of the 62.5Cu–1Si–Zn alloy were approximately the same being 120 µm whereas that of CW510L was 270 µm, more than twice the thickness of the chips of C3604 or the 62.5Cu–1Si–Zn alloy. That is, the ratio between C3604 or the 62.5Cu–1Si–Zn alloy and CW510L in chip thickness was nearly reciprocal to that in machinability index. Incidentally, it is said that chip thickness and cutting resistance have a relationship that can be expressed with the following Formula (1):⁶⁾   

\begin{align} \text{Cutting resistance} &= \text{C}\times \text{shear yield stress}\\ &\quad \times \text{chip thickness} \end{align} (1)

• C: a constant determined by the cutting conditions (rake angle, cutting depth, and feed)

When Value “C” is calculated based on the resultant force of cutting resistance in accordance with the above Formula (1) assuming that the shear yield stress is approximately the same as the tensile strength, the value of the 62.5Cu–1Si–Zn alloy and that of CW510L both became about 1.9. Since the above proportionality equation is basically valid for both alloys, we believe that the reason the 62.5Cu–1Si–Zn alloy has better machinability index is because its chips are thinner.

Fig. 4

Appearance, cross-sectional microstructure and thickness of chips in turning, and HV hardness of center of chip cross section.

3.2 Drilling

Figure 5 shows the change in cutting power over the duration of the drilling test. The cutting power required for drilling the 62.5Cu–1Si–Zn alloy was about 2/3 of that required by CW510L in the early to middle stages of drilling. On the other hand, the machinability index of the 62.5Cu–1Si–Zn alloy was 74, about 1.5 times that of CW510L. C3604 was the alloy that required the lowest cutting power in the early stage, but the required cutting power slightly increased toward the late stage of the drilling. In contrast, the cutting power required by the 62.5Cu–1Si–Zn alloy was stable from the early to late stages of the drilling, and it was almost the same as that of C3604 at the end of the drilling. On the other hand, the cutting power required by CW510L increased significantly in the late stage of the drilling from 1.5 times the cutting power required by the 62.5Cu–1Si–Zn alloy in the early stage to twice in the latter half of the drilling. This increase in the required cutting power in the latter half of the drilling is presumably related to clogging of chips.

Fig. 5

Time variation of principal cutting power during drilling.

Figure 6 shows the appearance, the cross-sectional microstructure, and the thickness of the chips generated by the middle stage of the drilling. Chips of the 62.5Cu–1Si–Zn alloy generated by drilling, like those generated by turning, were finely broken similar to those of C3604, and the cross-sectional shape of the chips was shear-type like that of C3604. Chips of CW510L were relatively broken, but their cross-sectional shape was flow-type. The thickness of the chips of C3604 and that of the chips the 62.5Cu–1Si–Zn alloy were almost the same, 150 µm, while that of the chips of CW510L was about 1.7 times larger. If Value “C” in Formula (1) is calculated applying the cutting power indicated in Table 4 as the cutting resistance, Value “C” of the 62.5Cu–1Si–Zn alloy and that of CW510L are both 5.1, revealing that Formula (1) is also applicable to drilling operation. From this fact, we figure that the reason the 62.5Cu–1Si–Zn alloy exhibits better machinability index also in drilling is because the alloy generates thinner chips.

Fig. 6

Appearance, cross-sectional microstructure and thickness of chips in drilling.

3.3 Cutting quality

First, the maximum difference between the thickest and thinnest portions of the wall of the pipes made of the test materials by drilling was compared to see the variation in wall thickness. It was 40 µm in the pipes made of the 62.5Cu–1Si–Zn alloy or C3604, whereas 100 µm in the CW510L-made pipe. The drill path in this CW510L-made pipe was significantly meandering, suggesting that problems with cutting quality may arise when drilling a CW510L material.

Next, Fig. 7 shows the appearance and the cross-sectional microstructure of the test specimens after the turning test. The edge of the specimen made of the 62.5Cu–1Si–Zn alloy is cut sharp and clean at the location circled with a dotted line in Fig. 1(a) whereas generation of burrs was observed in the specimen made of CW510. In general, burrs are more likely be formed when turned at a small rake angle and a low cutting speed. Even though the cutting conditions applied this time were relatively burr-prone, none were observed on the specimen of the 62.5Cu–1Si–Zn alloy. If burrs are formed on a machined product like the specimen of CW510L, modification of cutting conditions or adding an extra step to remove burrs may become necessary, which creates a problem in mass production.

Fig. 7

Appearance and cross-sectional microstructure of the edge after turning.

Figure 8 shows schematic diagrams (illustrations) that indicate the locations where burrs are formed when a workpiece of the 62.5Cu–1Si–Zn alloy or CW510L is turned. Burrs are formed on an end face of a workpiece, but we think that in the case shear-type chips are generated like when the 62.5Cu–1Si–Zn alloy is turned, they are less likely be formed because 62.5Cu–1Si–Zn alloy chips are thin and finely brokens including those formed on an end face. In contrast, flow-type chips such as those of CW510L are thick and continuous including those formed on an end face. That is why we presume that thick and continuous chips are more likely to remain attached to the workpiece in the form of burr.

Fig. 8

Schematic diagram during burr generation.

Figure 9 shows the surface roughness, Rz, of the test specimens after the turning test. All of the specimens exhibited surface roughness close to the theoretical surface roughness when the feed rate was 0.1 mm/rev or more. However, when the feed rate was less than 0.1 mm/rev, the surface roughness of C3604 exceeded the theoretical surface roughness. On the other hand, the surface roughness of the 62.5Cu–1Si–Zn alloy and CW510L remained close to the theoretical surface roughness regardless of the feed rate. When compared at a feed rate of 0.05 mm/rev, the Rz values of these two alloys were roughly 1/3 that of C3604.

Fig. 9

Surface roughness of specimens after turning.

The SEM images of the turned surfaces in Fig. 10 show that fine burrs that originate from lead scatter over the surface of C3604, and the regularity in the linear cutting marks on the surface is disarranged. In contrast, very limited fine burrs are observed on the surfaces of the specimens of the 62.5Cu–1Si–Zn alloy and CW510L, and regularity in the cutting marks is maintained. Based on these facts, it is presumed that the surface roughness of a C3604 material increases when turned at a low feed rate due to the fine burrs that originate from lead.

Fig. 10

SEM image of the surface after turning (a) 62.5Cu–1Si–Zn alloy (b) CW510L (c) C3604.

3.4 Microstructure and cutting mechanism

Figure 11 shows the microstructures of the 62.5Cu–1Si–Zn alloy and CW510L. Both of them consist of alpha-beta phases, with the β phase accounting for roughly 50% of the entire constituent phases. The 62.5Cu–1Si–Zn alloy contains 1% Si which is solid-solubilized in both of the phases. Further, compared with CW510L, this alloy has finer α-phase grains, and fine precipitates mainly comprising P are observed in the β phase. It is presumed that α-phase grains of the 62.5Cu–1Si–Zn alloy are refined due to an influence from Si, an element having an effect of lowering stacking fault energy, contained in the alloy, and control of the alloy’s microstructure.⁷⁾

Fig. 11

Microstructure of specimens (a) 62.5Cu–1Si–Zn alloy (b) CW510L.

Figure 12 shows the relationship between the cold working rate and the HV hardness of the test specimens when they were cold compressed. Based on the measurements of the HV hardness taken at the center of cross sections of the chips generated during the turning test shown in Fig. 4, the cold working rate at the center of a cross section of a chip of each test specimen can be estimated from Fig. 12 to be about 15% for the 62.5Cu–1Si–Zn alloy as the alloy’s HV hardness is 185, higher than 50% for CW510L as its HV hardness is 190, and about 30% for C3604 as its HV hardness is 150. That is, the cold working rate at the center of a chip was lowest with the 62.5Cu–1Si–Zn alloy, followed by C3604, and it was the highest with CW510L. That suggests that at the center of a chip of the 62.5Cu–1Si–Zn alloy, shear fracture advanced with little plastic deformation whereas in the case of C3604 chip, stress locally concentrated on lead particles after it was subjected to moderately large plastic deformation, which induced shear fracture to advance. In contrast, CW510L does not have many stress concentration sources that cause shear fracture. Therefore, not only the center portion but also the entire chip was subjected to large plastic deformation, resulting in generation of flow-type chips. Consequently, chips of CW510L are thick and unbroken, causing its cutting resistance to increase. Unlike CW510L, the 62.5Cu–1Si–Zn alloy has origins of shear fracture in the matrix itself, which is more influential than the large-scale plastic deformation. As a result, thin, shear-type chips are formed. In the 62.5Cu–1Si–Zn alloy, β phase in which Si, an element having an effect of reducing the stacking fault energy, is solid-solubilized, fine P-containing compounds uniformly dispersed in the β phase, and refined α-phase grains are present. As these factors make stress concentration sources, thin, shear-type chips are formed during machining with little influence from plastic deformation on the center portion of the chips. Thus, we presume that the 62.5Cu–1Si–Zn alloy provides low cutting resistance, and its chips are easily broken without relying on Pb contained at a concentration as high as 3%. It is further presumed that thin, shear-type chips are preventing formation of burrs and restraining drill path from meandering. Also, formation of fine burrs on the machined surface is inhibited because the alloy is free of lead, resulting in more ideal surface roughness. That the 62.5Cu–1Si–Zn alloy has excellent machinability and provides good cutting quality has been substantiated by the test results and the theories provided above.

Fig. 12

Relationship between cold working ratio and HV hardness.

4. Conclusion

1. (1)    Machinability indices of the 62.5Cu–1Si–Zn alloy was 89 in turning, nearly twice the figure of CW510L, and 74 in drilling, about 1.5 times the figure of CW510L.
2. (2)    Chips of the 62.5Cu–1Si–Zn alloy generated in turning were finely broken and shear-type whereas those of CW510L were thick, flow-type, and continuous in a spiral shape.
3. (3)    In drilling the 62.5Cu–1Si–Zn alloy, finely broken shear-type chips were generated, and required cutting power was stable through to the latter half of the drilling process when the depth of holes was three times the diameter.
4. (4)    Cutting quality of the 62.5Cu–1Si–Z alloy was excellent based on the facts that burrs were less likely be formed and surface roughness was favorable in turning, and drill path was quite straight in drilling.
5. (5)    We consider that the 62.5Cu–1Si–Zn alloy provides good cutting quality because when the alloy is machined, thin and shear-type chips are formed, fragmentation of such chips is promoted, and cutting resistance is reduced due to the action of Si solid-solubilized in the β phase, presence of fine P-containing compounds, and the effect of refined α-phase grains.

Patent

The developed alloy stated above is patented in Japan under the registration numbers of 6796355 and 6795872.

REFERENCES

• 1)  Oeko-Institut e.V.: Study to assess requests for a renewal of nine (-9-) exemptions 6(a), 6(a)-I, 6(b), 6(b)-I, 6(b)-II, 6(c), 7(a), 7(c)-I, and 7 (c)-II of Annex III of Directive 2011/65/EU (2021).
• 2)  Fraunhofer-Institute for Reliability and Microintegration (IZM), the United Nations Institute for Training and Research (UNITAR), and Bio Innovation Service (BioIS): 11th adaptation to scientific and technical progress of exemptions, 2(c)(i), 3 and 5(b) of Annex II to Directive 2000/53/EC (ELV) (2021).
• 3)  4MSI Joint Management Committee: Acceptance of Metallic Materials Used for Products in Contact with Drinking Water Part B – Positive list of compositions for metallic materials, 16th revision, (2021).
• 4)  RoHS Umbrella Industry Project (“the Umbrella Project”): Exemption Request Form - Exemption #6(c), (2020) p. 27.
• 5)   K.  Oishi,  H.  Goto and  K.  Suzaki: Materia 60 (2021) 110–112. doi:10.2320/materia.60.110
• 6)  K. Nakayama: The Metal Cutting in Its Principle, (Corona Publishing Co., Ltd., Tokyo, 1978) p. 94.
• 7)   K.  Oishi,  I.  Sasaki and  J.  Otani: Mater. Lett. 57 (2003) 2280–2286. doi:10.1016/S0167-577X(02)01211-9