Materials Science and Engineering A 529 (2011) 275– 281

Contents lists available at SciVerse ScienceDirect

Materials Science and Engineering A

jo ur n al hom epage: www.elsev ier .com/ locate /msea

High-strength, lead-free machinable �–� duplex phase brass

Cu–40Zn–Cr–Fe–Sn–Bi alloys

Haruhiko Atsumia,∗, Hisashi Imaib, Shufeng Lib, Katsuyoshi Kondohb,Yoshiharu Kousakac, Akimichi Kojimac

a Graduate School of Engineering, Osaka University, 565-0871, 2-1, Yamadaoka, Suita, Osaka, Japanb Joining and Welding Research Institute, Osaka University, 567-0047, 11-1, Mihogaoka, Ibaraki, Osaka, Japanc San-etsu Metals Co. Ltd., 939-1315, 1892, Ohta, Tonami, Osaka, Japan

a r t i c l e i n f o

Article history:

Received 8 February 2011

Received in revised form 7 September 2011

Accepted 9 September 2011

Available online 22 September 2011

Keywords:

Brass

Lead-free

Bismuth

Machinability

Mechanical properties

High-strength

a b s t r a c t

High-strength, lead-free machinable �–� duplex phase brass (Cu–40Zn) with 0.3 mass% chromium,

0.2 mass% iron, 0.6 mass% tin, and 1–3 mass% bismuth (Cu–40Zn–Cr–Fe–Sn–Bi) were prepared using a

casting process, and their microstructures, mechanical properties, and machinability were investigated.

Cast Cu–40Zn–Cr–Fe–Sn–Bi exhibited �–� duplex phase structures dispersed with Cr–Fe intermetal-

lic compounds (IMCs) and spherical Bi particles that existed in the �-phase. The Bi particle size in this

phase was smaller than that of irregularly shaped Bi particles in or around the �-phase; thus, cast spec-

imens with large area fractions of the �-phase had more finely dispersed Bi particles. Furthermore, the

additional Bi did not react with the added solid solution strengthening elements of Cr, Fe, and Sn. Con-

versely, the extruded Cu–40Zn–Cr–Fe–Sn–Bi consisted of fine, uniform �–� duplex phases dispersed with

fine, discrete Cr–Fe IMCs, and the Bi particles were also slightly elongated along the extrusion direction.

Analysis of back-scattered SEM images determined that the number of Bi particles in the wrought alloy

matrix was 1500–3000/mm2 in the transverse cross-section of the extrusion direction. The average yield

strength (YS) and average ultimate tensile strength (UTS) of the extruded Cu–40Zn–Cr–Fe–Sn–Bi alloy

were 288 MPa and 601 MPa, respectively. Based on the similar tensile properties of this alloy to those

of Cu–40Zn–Cr–Fe–Sn, the main strengthening mechanism in the former alloy was due to solid solution

strengthening with elemental additives and the increased area present as the hard �-phase. Furthermore,

this extruded alloy exhibited an increase of 29% YS and 40% UTS compared to traditional machinable brass

Cu-40Zn with 3.2 mass% lead (Cu–40Zn–Pb). The machinability of the extruded Cu–40Zn–Cr–Fe–Sn–Bi

was also 25% lower than that of the Cu–40Zn–Pb alloy.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Brass alloys are used for pipes, valves, and fittings in systems that

transport water and other aqueous fluids because of their excel-

lent corrosion resistance. Pb at levels of 2–5 mass% is also added

to traditional brasses to improve their machinability [1–5]. How-

ever, the addition of Pb to these materials is a significant, severe

hazard to the environment and human health [6,7]. Thus, mate-

rial designs must consider safety according to the regulations of

the Restriction of Hazardous Substance (RoHS) and Waste Electrical

and Electronic Equipment (WEEE) Directives [8–10]. As alternatives

to Pb, Bi, silicon (Si), and graphite (C) particles have been consid-

ered to improve the machinability of brass [11–15]. Bi has similar

properties to Pb, including its melting point, solid solubility in

∗ Corresponding author. Fax: +81 6879 8669.

E-mail address: atsumi-h@jwri.osaka-u.ac.jp (H. Atsumi).

copper, and density. Furthermore, Bi does not pose a danger to the

environment or human health.

Methods for solid solution strengthening and increasing the area

fractions of the �-phase have been applied to brass materials to

improve their mechanical properties. Analysis of a Cu–Zn binary

phase diagram suggests that brass with 35–48 mass% Zn has suf-

ficient �–� duplex phase structures to yield suitable strength and

hot forginability. For example, commercial high-strength brass has

been used for marine propellers and bridge bearings due to its supe-

rior mechanical properties and corrosion resistance. This material

is alloyed with aluminum (Al), manganese (Mn), and Fe for solid

solution strengthening. Moreover, its matrix is strengthened by

an increase in the area fraction of the hard �-phase [16,17]. How-

ever, when larger amounts of elemental materials are added to the

alloy for strengthening, coarser and more brittle IMCs are produced

in the matrix. In particular, these IMCs drastically decrease the

machinability of brass materials. A trade-off in the balance between

hardness and machinability of these alloys has been reported in

0921-5093/$ – see front matter © 2011 Elsevier B.V. All rights reserved.doi:10.1016/j.msea.2011.09.029

276 H. Atsumi et al. / Materials Science and Engineering A 529 (2011) 275– 281

Table 1Chemical compositions of brass cast ingots used in this study (mass%).

Alloys Sn Pb Zn Ni Fe Al Cr Bi Cu

CAST1 0.59 −0.005 40.86 −0.004 0.22 −0.005 0.34 – Bal.

CAST2 0.595 −0.005 40.81 −0.004 0.229 −0.005 0.256 0.994 Bal.

CAST3 0.6 −0.005 40.64 −0.004 0.23 −0.005 0.26 2.02 Bal.

CAST4 0.578 −0.005 40.83 −0.004 0.219 −0.005 0.22 2.85 Bal.

the literature [18,19]. In the present study, new high-strength, Pb-

free machinable �–� duplex brass alloys with dispersed Bi particles

were produced using a casting and extrusion process. The effects

of Bi dispersoids on the mechanical properties and machinability

of the extruded specimens were investigated.

2. Materials and methods

Brass alloys were prepared using a casting process. Cu–40Zn

with additions of 0.3 mass% Cr, 0.2 mass% Fe, and 0.6 mass%

Sn (Cu–40Zn–Cr–Fe–Sn, CAST1) was prepared as a refer-

ence material, and Cu–40Zn–Cr–Fe–Sn with 1–3 mass% Bi

(Cu–40Zn–Cr–Fe–Sn–Bi, CAST2–4) were alloyed. Cu–Cr and Cu–Fe

alloys were melted at over 1673 K, and pure Zn, Sn, and Bi were

added to the molten alloys at 1273 K, followed by casting of the

alloys. The chemical compositions of the cast ingots are shown in

Table 1. The addition of small amounts of elemental Cr, Fe, and Sn

strengthened the �–� phase brass as described in previous reports

[20,21]. The cast ingots (60 mm in diameter) were machined into

billets with a diameter of 41 mm. The billets were extruded using a

hydraulic direct press machine (SHP-200-450, Shibayamakikai Co.)

with a press capacity of 2000 kN. The extrusion speed was 3 mm/s.

Before extrusion, the billets were preheated at 923 K for 180 s in an

atmosphere of Ar gas in a muffle furnace (KDF S-70, Denken Co.).

The final diameter of the ingots after hot extrusion was 7 mm. The

extruded CAST1–4 specimens are denoted as EXT1–4 in this study.

Microstructural observations and analyses of the cast and

extruded cast specimens were conducted using an optical

microscope (OM, BX-51P, Olympus), and scanning electron

microscope–electron dispersive spectroscopy (SEM–EDS, JSM-

6500F, Jeol; EX-64175JMU, Jeol), and the phase compositions

were identified using X-ray diffraction (XRD, XRD-6100, Shi-

madzu). Back-scattered electron images (BSIs) acquired with SEM

were used to evaluate the morphologies of Bi particles dis-

persed in the extruded specimens. The number of Bi particles was

measured using image analyses of the BSIs. For electron back-

scattered diffraction (EBSD) analyses, specimens with dimensions

of 10 mm × 7 mm × 3 mm were cut parallel to the extrusion direc-

tion from the extruded specimens. The specimens were ground

with #1000 and #4000 SiC abrasive papers and polished with

a 0.05 �m Al2O3 polishing suspension. They were then electro-

chemically polished in 30% HNO3 in methanol at 243 K and 30 V

for 3 s. EBSD texture measurements were conducted with an SEM

5mm Extruded brass alloy

Rotating speed900 rpm

Applied force

Drill

98 N

Fig. 1. Schematic illustration of experimental set-up of drilling test in this study.

equipped with a DigiView IV Detector (EDAX-TSL) and OIM Data

Collection 5.31 software (TSL Solutions K.K.).

For evaluations of machinability, drilling tests were performed

under dry conditions using a TiN-coated high-speed steel drill tool

(EX-SUS-GDS, OSG) with a 4.5 mm diameter. During drilling, the

drill rotation speed was 900 rpm, and the applied load was 98 N. The

experimental set-up is shown in Fig. 1. The drilling time to make a

hole with a 5 mm depth was measured; after repeating this drilling

test 10 times, the average drilling speed was used as a machinabil-

ity parameter for the extruded specimens. Mechanical properties

of the specimens were evaluated using tensile and hardness tests.

The extruded specimens were machined into tensile test specimens

with a 3 mm diameter in accordance with ICS 59.100.01. Tensile

tests were performed on a universal testing machine (Autograph

AG-X 50 kN, Shimadzu) with a strain rate of 5 × 10−4/s. Hardness

tests were performed using a Vickers microhardness tester (HMV-

2T, Shimadzu) with a testing load of 245.2 mN (0.025 kgf) for 15 s

at room temperature.

3. Results and discussion

3.1. Microstructure of cast ingots

Optical microstructures of the cast alloys are shown in Fig. 2.

The CAST1 specimen without Bi had �–� duplex phase structures

and coarse, dispersed IMCs with diameters of 10–30 �m. CAST2–4

specimens had similar morphologies in terms of �–� phase struc-

tures and IMCs to those of the CAST1 specimen. The Zn content was

maintained at 40 mass% in the cast ingots, as shown in Table 1. As

the amount of added Bi increased and the ratio of Cu to Zn decreased

in the CAST2–4 specimens, an increase in the area fraction of the

Fig. 2. Optical microstructures of cast ingots without Bi (a), with 0.99 mass% Bi (b), with 2.02 mass% Bi (c), and with 2.85 mass% Bi (d) in the transverse cross-section of the

casting direction.

H. Atsumi et al. / Materials Science and Engineering A 529 (2011) 275– 281 277

25 27 29 31 33 35 37 39 41 43 45

α-phase β-phase Cr1.36Fe0.52 Bi resin

Inte

nsity

/a.u

.

Diffraction angle, 2θ / º

(b) CAST2(0.99mass%Bi)

(c) CAST3(2.02mass%Bi)

(d) CAST4(2.85mass%Bi)

(a) CAST1(without Bi)

(012) (104) (110) (110) (111)

Fig. 3. X-ray diffraction patterns of cast ingots without Bi (a), with 0.99 mass% Bi (b), with 2.02 mass% Bi (c), and with 2.85 mass% Bi (d).

�-phase was observed. Bi in the �-phase formed spherical par-

ticles; however, irregularly shaped Bi particles also formed in or

around the �-phase. The difference in surface tension between Bi

particles and the matrix decreased as the elemental Zn content in

the matrix increased, and then spherical Bi particles were formed

in the �-phase with higher elemental Zn concentration compared

to the �-phase [22,23]. Many spherical Bi particles were observed

in the CAST4 specimen, which had a large area fraction of the �-

phase. Furthermore, the average size of the spherical Bi particles

was smaller than that of the irregularly shaped Bi particles. The

number of Bi particles also increased in proportion to the Bi content

in the CAST2–4 specimens.

X-ray diffraction analysis results for the cast alloys are shown

in Fig. 3. CAST1-4 specimens exhibited large peaks due to the �-

phase at 42.3◦ with a (1 1 1) crystal plane and to the �-phase

at 43.5◦ with a (1 1 0) crystal plane. A small peak at 44.5◦ cor-

responding to Cr1.36Fe0.52, confirmed with a standard ICDD PDF

card, was detected in each specimen in addition to the main peaks

attributed to the �- and �-phase structures. CAST2–4 specimens

exhibited a diffraction angle similar to that of Cr–Fe IMCs, which is

also comparable to that of the CAST1 specimen. The crystal plane

of the Cr–Fe IMCs is unknown. The �-phase volume fractions in

CAST1-4 specimens were 68%, 63%, 72%, and 76%, respectively, as

calculated from the �-phase (1 1 1) and the �-phase (1 1 0) peak

intensity ratios in the XRD profiles. The IMC volume fraction in each

specimen was 0.2–0.4%, which is quite small in proportion to the

�–� phase matrix. XRD peaks attributed to Bi were found at 27.3◦

(0 1 2), 38.1◦ (1 0 4), and 39.7◦ (1 1 0) in the CAST2–4 specimens,

respectively. Thus, the Bi peak intensity increased in proportion to

the amount of added Bi. These XRD analysis results corresponded

to the OM observations shown in Fig. 2. In addition, a peak at

around 26◦ indicated the presence of a resin for metallographic

observation. SEM–EDS point analysis results for the �-phase, �-

phase, and Cr–Fe IMC components in the cast ingots are shown

in Table 2. Similar levels of elemental solutes, including 0.1 mass%

Cr, 0.2 mass% Fe, and 0.2 mass% Sn (�-phase) and 0.8 mass% Sn (�-

phase), were detected in each specimen. Because the Cr–Fe IMC

compositions were nearly identical in the cast alloys, the elemental

Bi did not reacted with the solid solution strengthening elements

of Cr, Fe and Sn.

The Vickers microhardness values of the �- and �-phases in the

CAST2–4 specimens were similar to that of the CAST1 specimen

(Table 2). The average hardnesses of the �- and �-phases in the

CAST1-4 specimens were 110 Hv and 168 Hv, respectively, which

are 10% and 17% higher than those for Cu–40Zn alloy (�-phase:

100 Hv; �-phase: 144 Hv).

3.2. Microstructure and mechanical properties of extruded alloys

Optical microstructures of the extruded specimens (EXT1–4) are

shown in Fig. 4. The area fractions of the �-phase increased in pro-

portion to the added Bi concentration, similar to the results for the

cast alloys. The �-phase volume fractions in the EXT1–4 specimens

were 80%, 84%, 91%, and 97%, respectively, as calculated from the

�–� phase peak intensity ratios of the XRD profiles of the extruded

specimens. In addition, each extruded specimens exhibited an �-

phase with finer grains than in the cast alloys, which suggested

that the cast alloys possessed larger area fractions of the �-phase

due to the extrusion conditions of 923 K and the 180 s holding

time as compared to room temperature processing of the extruded

alloys. This result is expected on the basis of the Cu-Zn binary phase

diagram, and is due to the nucleation of the �-phase from the �-

phase during cooling after hot extrusion [24]. Discrete Cr–Fe IMCs

also formed along the extrusion direction and were mechanically

Table 2SEM–EDS analysis results of �-phase, �-phase and Cr–Fe IMC on brass cast ingots without Bi (a), with 0.99 mass% Bi (b), with 2.02 mass% Bi (c), and with 2.85 mass% Bi (d).

Alloys Elements (mass%) Vickers micro

hardness/Hv0.025

Cu Zn Cr Fe Sn Bi

CAST1

� 63.81 35.74 0.10 0.23 0.13 – 110(±4)

� 56.52 42.53 0.06 0.13 0.76 – 170(±4)

IMC 3.22 2.44 77.33 16.91 0.10 – –

CAST2

� 62.51 36.91 0.10 0.18 0.21 0.10 108(±5)

� 55.81 42.90 0.05 0.09 0.99 0.17 165(±7)

IMC 4.33 3.33 77.04 15.02 0.22 0.06 –

CAST3

� 63.10 36.13 0.10 0.20 0.22 0.25 111(±6)

� 56.24 42.61 0.07 0.23 0.79 0.06 168(±6)

IMC 3.21 2.34 81.47 12.57 0.24 0.16 –

CAST4

� 62.68 36.55 0.06 0.15 0.31 0.26 112(±5)

� 55.79 43.15 0.07 0.09 0.76 0.15 169(±5)

IMC 2.97 2.81 78.12 15.98 0.10 0.02 –

278 H. Atsumi et al. / Materials Science and Engineering A 529 (2011) 275– 281

Fig. 4. Optical microstructures of extruded cast specimens without Bi (a), with 0.99 mass% Bi (b), with 2.02 mass% Bi (c), and with 2.85 mass% Bi (d) in the longitudinal

cross-section of the extrusion direction.

broken by hot extrusion, forming fine particles. Bi particles were

also slightly elongated along the extrusion direction. Bi possessed

a similar electronic structure to that of Pb, remaining mostly

insoluble in the brass matrix and contributing to machinability per-

formance, as previously discussed in the introduction. However, the

elongated Bi particles also caused hot shortness at the surface of the

extruded brass alloy. Previous studies have reported that alloy ele-

ments including Bi and Pb, which have low melting points, promote

hot shortness [25]. In this study, hot shortness was not observed in

the extruded brass alloys. When brass alloys are extruded into elab-

orate shapes, such as hexagonal rods, hot-shortness cracks can be

reduced by controlling the extrusion temperature and the pressing

speed. A detailed Bi morphology will be discussed later.

SEM–EDS and XRD analysis results of the extruded specimens

remained almost unchanged when compared to those of the cast

specimens. The distributions of the �–� duplex phase structures

in the EXT1–4 specimens on the transverse cross-section of the

extrusion direction, as discerned by EBSD, are shown in Fig. 5. The

area fractions of the �-phase in the EXT1–4 specimens were 60%,

57%, 62%, and 80%, respectively. The increase in the fractional area

of the �-phase in these specimens was due to the increase in the

ratio of elemental Zn to Cu in each alloy. The average grain sizes of

the �- and �-phase structures are also shown in the right bottom

corner of each image. The EXT4 specimen had a large area frac-

tion of the �-phase and contained coarser grains in the �–� phases

than the EXT1–3 specimens. The grain size of extruded �–� brass is

determined by the phase transformation from �-phase to �-phase.

Extruded �–� brass had fine structures due to the nucleation of fine

�-phase during cooling after hot extrusion, as mentioned above.

Thus, because the EXT4 specimen had the highest concentration of

elemental Zn, this specimen had the coarsest structures without

considerable phase transformation from �-phase to �-phase, and

200

300

400

500

600

700

0 1 2 320

25

30

35

40

YS UTS Elongation

Elo

ngat

ion,

ε (%

)

Tens

ile p

rope

rties

/MPa

Bi content (mass%)

Fig. 6. Tensile properties of extruded cast specimens without Bi, with 0.99 mass%

Bi, with 2.02 mass% Bi, and with 2.85 mass% Bi.

the average �-phase grain size of EXT4 was 5.52 �m, which is 60%

larger than that of the EXT1 specimen (3.47 �m).

The dependence of the tensile properties of the extruded alloys

on their Bi content is shown in Fig. 6. The extruded high-strength

brass alloy Cu–40Zn–Cr–Fe–Sn (EXT1) had a YS of 285 MPa and an

UTS of 582 MPa, which are values that are 17% and 25% higher than

those of the extruded Cu–40Zn (YS: 244 MPa and UTS: 467 MPa),

respectively. The EXT1 specimen was strengthened by the solid

solution additives Cr, Fe, and Sn and the increased area frac-

tion of the hard �-phase. YS and UTS values for the extruded

Cu–40Zn–Cr–Fe–Sn–Bi (EXT2–4) alloys had average values of

288 MPa and 601 MPa, respectively, due to the same strengthening

Fig. 5. Grain distribution of �–� duplex phase structures on EXT1 (a), EXT2 (b), EXT3 (c), and EXT4 (d) by EBSD analysis in the transverse cross-section of the extrusion

direction.

H. Atsumi et al. / Materials Science and Engineering A 529 (2011) 275– 281 279

Fig. 7. SEM observation on fractured surface of tensile test specimens of EXT2 (0.99 mass% Bi) (a) and EXT4 (2.85 mass% Bi) (b), and Bi mapping of EXT2 (c) and EXT4 (d)

acquired by SEM–EDS analysis.

mechanisms; theses values are comparable to those of the EXT1

specimen, although only EXT2–4 specimens contained Bi particles.

The increasing amounts of soft and brittle Bi particles in the latter

specimens caused the deterioration of their mechanical proper-

ties. However, because the EXT2–4 specimens had larger �-phase

fractions than the EXT1 specimen, they demonstrated high tensile

strength similar to that of the EXT1 specimen. The YS and UTS of the

extruded brass specimens were influenced by a balance between

the grain size and the area fraction of the �-phase.

Elongation of the EXT2–4 specimens decreased as the elemental

Bi content increased. Notably, the elongation of the EXT4 specimen

(2.85 mass% Bi) was 22%, the lowest value observed for the extruded

specimens. The deterioration in elongation behavior was caused by

the increasing amounts of brittle Bi particles in the specimens. SEM

images and SEM–EDS maps for Bi content in the fractured surface

of the tensile specimens of EXT2 and EXT4 are shown in Fig. 7.

Fine, uniform dimples were observed on the fractured surface of

the EXT2 specimen due to the presence of fine �–� duplex phase

structures (Fig. 5(b)). The EXT4 specimen had more cleavage modes

in its fractured surface than the EXT2 specimen. Because the EXT4

specimen had a larger area fraction of the �-phase (Fig. 5(d)), which

has lower ductility at room temperature than the �-phase. Fur-

thermore, coarser particles with brittle surface in the cracks were

also observed on the fractured surface of the EXT4 specimen when

compared to the EXT2 specimen. These brittle, fractured surface

areas were identified as Bi particles by SEM–EDS analysis. Because

Bi particles are soft but brittle, the degree of elongation observed for

extruded specimens also decreased due to the presence of coarse

Bi particles.

3.3. Relationship between the morphology of bismuth particles

and the machinability of extruded specimens

BSIs of the EXT2–4 specimens in the transverse cross-sections of

the extrusion direction are shown in Fig. 8. The BSI shows the bina-

rized image that segregates Bi particles from the �–� phase matrix.

Because elemental Bi is heavier than elemental Cu, Zn, Cr, Fe and Sn,

white areas indicate the presence of Bi particles in the matrix. The

number of Bi particles in a selected analysed area was measured by

BSI image analyses. The relationships between the Bi particle den-

sity and the elemental Bi content are shown in Fig. 9. The Bi particle

numbers of extruded cast Cu–40Zn–X Bi alloys (X = 1.0, 2.33, 2.53,

2.6, 2.95, and 5.45 mass%) are also shown in Fig. 9 as a reference.

The Bi particle density of Cu–40Zn–X Bi specimens increased in

proportion to the amount of elemental Bi added. Although smaller

amounts of elemental Bi were added to the EXT2–4 specimens than

to the extruded Cu–40Zn–X Bi alloys, the former specimens exhib-

ited a better dispersion of Bi particles than the extruded Cu–40Zn–X

Bi alloys. Bi morphology in an extruded specimen depends signif-

icantly on that of the cast alloy. Cast Cu–40Zn–Cr–Fe–Sn–Bi alloys

had a large area fraction of the �-phase than cast Cu–40Zn–X

Bi alloy. As previously discussed for the cast alloys, because

Fig. 8. Back scattered electron images of EXT2 (0.99 mass% Bi) (a), EXT3 (2.02 mass% Bi) (b), and EXT4 (2.85 mass% Bi) (c) by SEM in the transverse cross-section of the

extrusion direction.

280 H. Atsumi et al. / Materials Science and Engineering A 529 (2011) 275– 281

Table 3The microstractural, mechanical, and machinability characteristics of the extruded specimens in the present study.

Alloys Bi content �-Phase content Grain size (�m) Bi particle density Drilling speed UTS Elong. Hardness

mass% % � � /mm2 mm/s MPa % Hv

EXT1 0 80 1.96 3.47 – – 582 37 154 ± 4

EXT2 0.994 84 2.51 4.05 940 0.14 597 29 157 ± 8

EXT3 2.02 91 2.83 3.87 1834 0.34 596 27 155 ± 7

EXT4 2.85 97 3.98 5.52 3221 0.28 622 22 159 ± 7

spherical Bi particles are formed in the �-phase domains,

these in �-rich brass become finer than those in �-rich

brass.

The dependence of the drilling speed on the number of Bi parti-

cles in the matrix is shown in Fig. 10. The drilling speed on extruded

specimens increased in proportion to the number of Bi particles in

the matrix. A good correlation was observed between machinability

of the specimen and the Bi particle density in the matrix. However,

the EXT4 specimen (with 2.85 mass% Bi) did not have sufficient

machinability when compared with the EXT2 and EXT3 specimens.

Although Bi was dispersed well in the EXT4 matrix, this specimen

had coarser grains than the other specimens. A previous study has

reported that alloys with coarse grains had poor machinability, due

to sliding wear and friction resistance between the cutting tools

and material surface [26]. Table 3 summarizes the characteristics of

the extruded alloys, including the microstructural, mechanical, and

machinability properties determined in the present study. Fig. 11

0

500

1000

1500

2000

2500

3000

3500

0 2 4 6

Cu-40Zn-Cr-Fe-Sn-BiCu-40Zn-Bi

Bi content (mass%)

Num

ber o

f Bi p

artic

les

(/mm

2 )

Fig. 9. Relationship between the number density of Bi particles in matrix and

amount of additions of Bi on extruded cast brass materials.

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0 1000 2000 3000 4000

Cu-40Zn-Cr-Fe-Sn-BiCu-40Zn-Bi

Number of Bi particles (/mm2)

Dril

ling

spee

d (m

m/s

)

EXT2(0.99mass% Bi)

EXT3(2.02mass% Bi)

EXT4(2.85mass% Bi)

Fig. 10. Dependence of drilling speed on number of Bi particles in matrix on

extruded cast brass materials.

300

350

400

450

500

550

600

650

0 0.1 0.2 0.3 0.4 0.5

Cu-40Zn-Cr-Fe-Sn-BiCu-40Zn-BiCu-40Zn-Pb

Tens

ile s

treng

th :

UTS

/MP

aMachinability : drilling speed (mm/s)

EXT2(0.99mass% Bi) EXT3(2.02mass% Bi)

EXT4(2.85mass% Bi)

Fig. 11. Relationships between tensile strength and machinability of extruded cast

brass alloys.

shows the relationship between tensile strength (UTS) and machin-

ability (drilling speed) of the extruded cast brass materials. Data

for the extruded Cu–40Zn–X Bi alloys and a traditional machin-

able brass Cu–40Zn alloy with 3.2 mass% Pb (Cu–40Zn–Pb) are also

included for reference. When the specimens contained coarse Bi

dispersoids, the UTS of extruded Cu–40Zn–X Bi alloys decreased

as the machinability increased, which is consistent with a trade-

off balanced between hardness and machinability. The extruded

Cu–40Zn–Cr–Fe–Sn–Bi alloys had superior tensile properties and

excellent machinability because this alloy had better Bi particle

dispersion than the extruded Cu–40Zn–X Bi alloys. In addition,

the UTS of extruded Cu–40Zn–Cr–Fe–Sn–Bi alloys were 40% higher

than that of conventional extruded Cu–40Zn–Pb alloy. However,

because the drilling speed for extruded Cu–40Zn–Cr–Fe–Sn–Bi

alloys maintained 75% of the machinability of the extruded

Cu–40Zn–Pb alloy, the extruded Cu–40Zn–Cr–Fe–Sn–Bi alloys

deviate from the traditional trade-off balanced between hardness

and machinability observed for the conventional machinable brass

materials.

4. Conclusion

In this study, the microstructures, mechanical properties, and

machinability of high-strength, Pb-free machinable �–� duplex

phase brass Cu–40Zn–Cr–Fe–Sn–Bi alloys were investigated. The

extruded Cu–40Zn–Cr–Fe–Sn–Bi alloys consisted of �–� phases

containing the fine, uniform Cr–Fe IMCs and Bi particles. Aver-

age YS and UTS of the extruded Cu–40Zn–Cr–Fe–Sn–Bi alloys

were 288 MPa and 601 MPa, respectively. Theses alloys exhibited

YS and UTS that were 29% and 40% higher than conventional

machinable brass Cu–40Zn–Pb alloy. The machinability of the

extruded Cu–40Zn–Cr–Fe–Sn–Bi alloys also maintained 75% of

the machinability of the Cu–40Zn–Pb alloy; thus, the extruded

Cu–40Zn–Cr–Fe–Sn–Bi alloys deviated from the traditionally

observed trade-off between hardness and machinability in conven-

tional machinable brass materials.

H. Atsumi et al. / Materials Science and Engineering A 529 (2011) 275– 281 281

Acknowledgement

Nippon Atomized Metal Powders Corporation is acknowledged

for their help in preparing the brass alloy cast ingots.

References

[1] G. Pantazopoulos, J. Mater. Eng. Perform. 11 (2002) 402–407.[2] Japan Capper and Brass Association, Base and Industrial Technology of Copper

and Copper Alloys, 1994 (in Japanese).[3] P. García, S. Rivera, M. Palacios, J. Belzunce, Eng. Fail. Anal. 17 (2010) 771–776.[4] E.W. Thiele Jr, K.J.A. Kundig, D.W. Murphy, G. Soloway, B. Duffin, CDA’s Univer-

sal Machinability Index. SAE Technical Paper Series, No. 900365, InternationalCongress and Exposition, Detroit, MI, 1990.

[5] J.X. Wu, M.R. Ji, M. Galeotti, A.M. Giusti, G. Rovida, Surf. Interface Anal. 22 (1994)323–326.

[6] X. Chen, A. Hu, M. Li, D. Mao, J. Alloys Compd. 460 (2008) 478–484.[7] S. Kuyucak, M. Sahoo, Can. Metall. Q. 35 (1) (1996) 1–15.[8] B.D. Mellow, D.J. Thomas, M.J. Joyce, P. Kolkowski, N.J. Roberts, S.D. Monk, Nucl.

Instrum. Methods Phys. Res. A 577 (3) (2007) 690–695.[9] GE plastics, RoHS and WEEE compliant flame retardants developed for electrical

connectors, Reinf. Plastics 51 (8) (2007) 12.

[10] C. Mans, S. Hanning, C. Simons, A. Wegner, M. Jan�en, M. Kreyenschmidt, Spec-trochim. Acta Part B 62 (2007) 116–122.

[11] L.V. Whiting, P.D. Newcombe, M. Sahoo, Trans. Am. Foundrymen’s Soc. 103(1995) 683–691.

[12] A.L. Fontaine, V.J. Keast, Mater. Charact. 57 (2006) 424–429.[13] V. Vazquez, A.J. Hernandez, A. Mascarenas, P. Zambrano, M.A.L.H. Rodriguez,

Eng. Fail. Anal. 17 (2010) 1285–1289.[14] T. Andoh, T. Atsumi, Y. Yoshikawa, J. JCBRA 40 (2001) 253–256.[15] H. Imai, Y. Kosaka, A. Kojima, S. Li, K. Kondoh, J. Umeda, H. Atsumi, Mater. Trans.

51 (2010) 855–859.[16] H. Mindivan, H. C imenoglu, E.S. Kayali, Wear 254 (2003) 532–537.[17] M. Sundberg, R. Sundberg, S. Hogmark, R. Otterbergc, B. Lehtinenc, S.E. Hgrn-

strgm, S.E. Karlsson, Wear 115 (1987) 151–165.[18] C. Vilarinho, J.P. Davim, D. Soares, F. Castro, J. Barbosa, J. Mater. Process. Technol.

170 (2005) 441–447.[19] L. Jin, R. Sandström, Mater. Des. 15 (1994) 339–346.[20] H. Atsumi, H. Imai, S. Li, Y. Kousaka, A. Kojima, K. Kondoh, Mater. Sci. Forum

654–656 (2010) 2552–2555.[21] S. Li, H. Imai, H. Atsumi, K. Kondoh, J. Alloys Compd. 493 (2010) 128–133.[22] T. Andoh, T. Atsumi, Y. Yoshikawa, J. JRICu 42 (2003) 17–20.[23] J.T. Plewes, D.N. Loiacono, Adv. Mater. Processes (USA) 140 (1991) 23–27.[24] C. Mapelli, R. Venturini, Scr. Mater. 54 (2006) 1169–1173.[25] G. Pantazopoulos, Practical Fail. Anal. 3 (4) (2003) 14–22.[26] T.E. Raghy, P. Blau, M.W. Barsoum, Wear 238 (2000) 125–130.