Analysis of microstructural effects on mechanical properties of copper alloys 2017 Journal of Science Advanced Materials and Devices

Analysis of microstructural effects on mechanical properties of copper alloys 2017 Journal of Science Advanced Materials...

Trang 1

Original Article

Analysis of microstructural effects on mechanical properties of copper

alloys

a Graduate School of Natural Science and Technology, Okayama University, 3-1-1 Tsushimanaka, Kita-ku, Okayama, 700-8530, Japan

b Department of Materials Science and Engineering, Ehime University, 3 Bunkyo-cho, Matsuyama, Ehime, 790-8577, Japan

a r t i c l e i n f o

Article history:

Received 16 November 2016

Received in revised form

17 December 2016

Accepted 22 December 2016

Available online 30 December 2016

Keywords:

Bronze

Brass

Strength

Ductility

Microstructural characteristics

a b s t r a c t

With the aim of obtaining copper alloys with favorable mechanical properties (high strength and high ductility) for various engineering applications, the microstructural characteristics of two conventional copper alloys d an aluminum bronze (AlBC; CueAl9.3eFe3.8eNi2eMn0.8) and a brass (HB: CueAl4eZn25 eFe3eMn3.8) d and a recently developed aluminum bronze (CADZ: CueAl10.5eFe3.1eNi3.5eMn1.1eSn3.7), were controlled by subjecting the alloys to two different processes (rolling and casting) under various conditions For the rolling process, the rolling rate and temperature were varied, whereas for the casting process, the solidification rate was varied Microstructural characteristics, as examined by electron backscatter diffraction analysis, were found to differ among the alloys Complicated microstructures formed in CADZ led to high hardness and high tensile strength (sUTS), but low ductility (εf) For CADZ, casting at a high solidification rate allowed an increase in ductility to be obtained as a result of fine-grained structure and low internal stress In contrast, high ductility (with a fracture strain of more than 30%) was found for both cast AlBC and cast HB; moreover, both of these alloys possessed high tensile strength when produced by warm rolling at 473 K For CADZ, on the other hand, no clear effect of rolling on tensile strength could be found, owing to the many microcracks caused by its brittleness The results of this study indicate that copper alloys with excellent mechanical properties can be produced This is especially the case for the conventional alloys, with a high tensile strengthsUTS¼ 900 MPa and a high fracture strainεf¼ 10% being obtained for warm-rolled brass

© 2016 The Authors Publishing services by Elsevier B.V on behalf of Vietnam National University, Hanoi This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)

1 Introduction

Copper alloys, including bronzes, are currently employed in a

wide range of engineering applications because of their high

ductility, high corrosion resistance, non-magnetic properties,

excellent machinability, and high hardness[1] Copper is used for

electric wiring and in heat exchangers, pumps, tubing, and several

other products, while aluminum bronze and high-strength brass

are found in marine applications, for example in propellers and

propeller shafts[2] Furthermore, shiny brass is widely employed

for coins and for musical instruments However, in spite of their

excellent material characteristics, there is still scope for technical

improvements to increase the strength and ductility of these alloys

To achieve improvements in mechanical strength, several copper

alloys with high dislocation density andﬁne microstructure, con-taining solid solutions, have been proposed The mechanical strength of ultraﬁne-grained or nanocrystalline CueAl alloys, pre-pared by equal-channel angular pressing (ECAP), has been inves-tigated, and the strength and uniform elongation of these alloys have been simultaneously improved by lowering the stacking fault energy[3] The hardness of even nanocrystalline copper with grain size as small as 10 nm still follows the HallePetch relation[4]

A variety of methods have been used to make high-strength copper alloys Maki et al.[5]attempted to create a higher-strength CueMg alloy through a solid-solution hardening effect, in which supersaturation with Mg increases the strength compared with that

of a representative solid-solution CueSn alloy[5] A high tensile strength of 600 MPa was reported by Sarma et al.[6], who produced

a CueAl alloy with ultraﬁne-grained microstructure and very ﬁne annealing twins by cryorolling and annealing at 523 K for 15 min The higher strength of this CueAl alloy was interpreted in terms of the enhanced solid-solution strengthening effect of Al, which is about 1.7 times higher than the corresponding effect in CueZn alloys

* Corresponding author Fax: þ81 86 251 8025.

E-mail address: mitsuhiro.okayasu@utoronto.ca (M Okayasu).

Peer review under responsibility of Vietnam National University, Hanoi.

Contents lists available atScienceDirect Journal of Science: Advanced Materials and Devices

j o u r n a l h o m e p a g e : w w w e l s e v i e r c o m / l o c a t e / j s a m d

http://dx.doi.org/10.1016/j.jsamd.2016.12.003

Journal of Science: Advanced Materials and Devices 2 (2017) 128e139

Trang 2

[6] In recent years, CueZn30eAl0.8alloys exhibiting nanostructure

have been fabricated by cryomilling of brass powders and

subse-quent spark plasma sintering [7] Such alloys have a high

compressive yield strength of 950 MPa, which is much higher than

the values of 200e400 MPa found in commercially available alloys This increase in mechanical strength has been attributed to pre-cipitation hardening and grain boundary strengthening [7] The effect of grain size on yield stress was examined in polycrystalline copper and CueAl alloys at 77 and 293 K, and the yield stress was found to satisfy the HallePetch relation in both materials[8] The

inﬂuence of hydrogen on the mechanical properties of aluminum bronze was investigated, and it was found that neither tensile nor fatigue properties were affected[9] After low-temperature thermal treatment, strained CueAl alloys exhibited high mechanical strength, which is caused by increases both in the degree of order and in the electron-to-atom (e/a) ratio[10] The effects of micro-structural characteristics on the mechanical strength of CueNi26eZn17alloy were investigated, and it was found that solid-solution strengthening of the alloy was affected by the interaction of

Ni and Zn atoms with screw dislocations and by the effective interaction caused by the modulus mismatch[11] In order to un-derstand the material properties of copper alloys, it is important to investigate their microstructural characteristics, including texture The textures of copper alloys after rolling and recrystallization were analyzed by electron backscatter diffraction analysis (EBSD) [12] The evaluation of grain boundaries in copper bicrystals during one-Fig 1 Maximum rolling rates for copper alloys at different temperatures.

M Okayasu et al / Journal of Science: Advanced Materials and Devices 2 (2017) 128e139 129

Trang 3

pass ECAP was systematically investigated by several methods,

including EBSD[13]

The above literature survey shows that there are various

ap-proaches that can be adopted to improve the mechanical properties

of copper alloys, including grain reﬁnement, solid solutions, and

high dislocation density In many practical applications, it is

desirable to reduce the weight of components and structures made

from such alloys by enhancing their mechanical properties Thus, in

the present work, an attempt is made to create copper alloys with

favorable tensile properties (high strength and ductility) via

microstructural modiﬁcation using forging and casting processes

under various conditions To analyze the mechanical strength and

ductility of these alloys, their microstructural characteristics are

investigated by EBSD

2 Experimental

2.1 Sample preparations

Two commercial copper alloys, namely, an aluminum bronze

(AlBC: CueAl9.3eFe3.8eNi2eMn0.8) and a brass (HB:

CueAl4eZn25eFe3eMn3.8), were studied, as well as a newly

Mn1.1eSn3.7) It should be pointed out that CADZ was developed on the basis of a CueAl10.5alloy in Dozen-Kogyo Co Ltd The material characteristics of CADZ were originally developed by described in detail elsewhere[14]

The test samples of the alloys were produced by casting and forging (rolling) In the casting process, two different cooling rates, and thus solidification speeds, were adopted At the low cooling rate (slow cooling, SC: 20 K/s), the melts were solidified slowly in a furnace In this case, the solidification process was carried out under an argon gas atmosphere to prevent oxidation

At the high cooling rate (rapid cooling, RC: 150 K/s), the melts were solidiﬁed rapidly in a copper mold The solidiﬁcation speeds for both the rapid and slow cooling processes were measured directly using a thermocouple In the rolling process, the alloys were forged at different deformation rates, using a 10-ton twin-rolling machine (Yoshida Kinen Co., Ltd.) with high-strength

diameter 200 mm) Samples of thickness 10 mm were forged under severe deformation at different temperatures: 293 K (cold rolling, CF), 493 K (warm rolling, WF), and 1073 K (hot rolling, HF)

Fig 2 (continued)

Trang 4

2.2 Material properties

Tensile tests were conducted at room temperature using a

hy-draulic servo-controlled testing machine with 50 kN capacity

Rectangular dumbbell-shaped specimens were employed with

di-mensions 3 mm 20 mm 2 mm The loading speed was set at

1 mm/min untilﬁnal failure The tensile properties (ultimate tensile

strengthsUTSand fracture strainεf) were evaluated via tensile stress

versus tensile strain curves, which were monitored by a data

acquisition system in conjunction with a computer through a

standard load cell and strain gauge Hardness measurements were

made using a micro-Vickers tester at 2.94 N for 15 s In this test, a

diamond indenter was loaded manually at about 0.3 N/s to the

sample surface, which had been polished to a mirrorﬁnish

The microstructural and lattice characteristics of the alloys were

investigated by EBSD using aﬁeld emission scanning electron

mi-croscope (SEM; JEOL JSM-7000F), with an acceleration voltage of

15 kV, a beam current of 5 nA, and a step size of 20mm The samples

were sectioned to less than 10 mm thick, and the sample faces for

the observation were polished to a mirrorﬁnish in a vibropolisher,

using colloidal silica for no longer than 2 h

3 Results and discussion 3.1 Microstructural characteristics

Fig 1shows the maximum possible rolling rates for the alloys With the cold-rolling process, the maximum rolling rate of CADZ is about 9%, which is about 50% and 33% lower than those for HB and AlBC, respectively With the warm-rolling process, the rolling rate is still as low as 12% for CADZ, although severe deformation of more than 75% is obtained for HB and AlBC after warm rolling With the hot-rolling process, a high rolling rate of more than 75% is obtained for the three alloys

Fig 2shows optical micrographs of the three alloys, made using both the casting and the rolling processes Essentially, the three alloys consist of matrix and eutectic structures The main eutectic phases of the CADZ sample are found to be (Fe, Ni)3Cu, CueNieSn and CueAl[14], as indicated by the arrows in cast CADZ The AlBC sample is essentially formed from eutectic Fe-, CueAleNi-, and CueAl-based phases, while for the HB sample, eutectic Fe-, CueZneAl-, and CueZn-based phases are observed The grain size clearly varies for all the cast alloys, where the higher the cooling Fig 2 (continued)

Trang 5

rate, the smaller the grain size For the rolled samples, no clear

changes in grain size can be detected, especially for CADZ This

could be due to the low rolling rates for CF- and WF-CADZ On the

other hand, grain growth (or recrystallization) occurs for HF-CADZ

and for HF-HB For CF- and WF-AlBC, slightly strained

microstruc-tural formations can be seen To understand these microstrucmicrostruc-tural

characteristics in detail, an EBSD analysis was carried out

Fig 3displays the inverse poleﬁgure (IPF) and misorientation

(MO) angle maps of the cast alloys, obtained by EBSD As can be

seen, complicated microstructures with high MO angles are formed

almost throughout both the cast and rolled CADZ samples In

contrast, high MO angles are found mainly in the eutectic phases of

AlBC, whereas high MO angles are widely distributed in RC- and

CF-AlBC Similar trends are observed in the corresponding HB samples The MO angles for the CADZ samples are overall higher than those for AlBC and HB The higher MO angles for the CF samples are considered to be due to increased dislocation density, while the low

MO angles for the HF samples result from a reduction in internal stress due to the high-temperature processing In addition, it is notable that deformation twins can be clearly detected in the rolled AIBC samples, but not in the others This can be attributed to the different extents of stacking fault energy (SFE): the lower the SFE, the weaker the deformation twins In previous work, it has been reported that the SFE decreases with an increasing proportion of Al

in the alloy composition: for example, the SFE of CueAl2.3alloy is about 6 times higher than that of CueAl alloy[15] Since it has Fig 3 (aec) Inverse pole ﬁgure (IPF) and misorientation angle maps of (a) CADZ, (b) AlBC, and (c) HB (d) IPF map for CADZ with and without Fe element.

Trang 6

Fig 3 (continued)

Trang 7

Fig 3 (continued)

Trang 8

been reported that the SFE of CueZn24alloy is about 4 times higher

than that of CueAl8alloy[16], our AIBC should have a much lower

SFE compared with HB, leading to deformation twinning in the

former but not the latter

The grain size of the cast alloys was measured directly, and the

results are summarized inTable 1 It should be pointed outﬁrst that

for the Cu-based phases, measurements were made of straight

di-agonal lines on each grain, and the grain size was determined as the

mean value of more than 50 measurement data Since grain for-mation is not clearly seen for CADZ, image analysis was conducted

on the cast CADZ, with the Fe element being removed from the IPF maps; seeFig 3(d) FromTable 1, it can be seen that the grain size varies, depending on the sample and the casting speed The dif-ferences in microstructural characteristics lead to difdif-ferences in mechanical properties The average grain size of the alloys made by rapid cooling is less than 38 mm The grain size increases with

Fig 3 (continued)

Table 1

Grain sizes of the copper alloys CADZ, AlBC, and HB (SD: standard deviation).

Trang 9

decreasing cooling rate: for example, for HB, a large grain size of

680mm is obtained, which is more than 10 times greater than that

for AIBC

3.2 Mechanical properties

Figs 4 and 5show Vickers hardness data for the three alloys

made by rolling and casting processes under different conditions

For the cast samples shown in Fig 4(a), a high hardness is

ob-tained overall for CADZ: for example, the value of about 2.5 GPa

for AS-CADZ is about 15% and 70% higher than those for AS-HB

and AS-AlBC, respectively An improvement in hardness is

obvious for all the alloys with a higher solidiﬁcation rate (the RC

samples): for example, for RC-CADZ, the hardness is as high as

3.3 GPa, which is more than 1.4 times that for SC-CADZ On the

other hand, the lowest hardness of about 1.4 GPa is obtained for

the SC-AlBC samples These differences in hardness are due to a

number of reasons, including the fact that different grain sizes

lead to different grain boundary strengths Fig 4(b) shows the

relationship between grain size and hardness for the alloys

Although there are only a few data points, clear correlations can be

seen, and the HallePetch relation appears to be satisﬁed A similar

HallePetch relation is also obtained for nanocrystalline copper

(10 nm)[4]

For the rolled samples shown inFig 5, the hardness value

in-creases with increasing rolling rate and decreasing rolling

temperature These trends are presumably due to the differences in dislocation density, deformation twinning, and internal stress arising during the rolling process, as indicated by the distributions

of MO angles seen inFig 3 In particular, a high hardness is obtained for the cold-rolling process, owing to dislocation tangling, despite the low rolling rate On the other hand, the low hardness of the samples made by hot rolling is a consequence of their recrystalli-zation and grain growth, as described previously It should also be pointed out that the deformation characteristics of AIBC and HB can vary depending on the SFE, as mentioned above In general, it ap-pears that deformation twining occurs for the alloys with lower SFE, namely, the AIBC samples This deformation occurs when dislocation is dominated by the rolling process, i.e., work hardening occurs[17]

Fig 6shows representative tensile stress versus tensile strain curves for the three alloys made by rolling and by casting, while

Fig 7 summarizes their tensile properties in terms of ultimate tensile strength versus fracture strain It should be pointed out that more than three specimens were employed here to obtain the tensile properties From the stressestrain curves, it can be seen that high ductility is obtained for the cast samples, with the fracture strain for AIBC being higher than that for HB and CADZ The reason for this is the presence of deformation twinning in AIBC, as mentioned above Huang et al.[18]reported that the deformation twins in coarse-grained Cu occurred mainly in shear bands and at their intersections, as a result of the very high local stress caused by Fig 5 Vickers hardness of copper alloys made by rolling: (a) CADZ; (b) AlBC; (c) HB.

Trang 10

severe plastic deformation On the other hand, a high tensile

strength is obtained overall for the rolled samples compared with

the cast ones In particular, higher tensile strengthssUTSare

ob-tained overall for AlBC and HB made at a high rolling rate and a

rapid cooling rate The highestsUTSvalues (>900 MPa) are obtained

for WF-AlBC, WF-HB, and RC-CADZ On the other hand, lowsUTS

values are found for HF-HB and HF-AlBC, even when high rolling

rates were applied The data plots of tensile properties are relatively

scattered for CADZ, which may be due to the low sample quality

Fig 8shows an SEM image of the HF-CADZ sample after rolling but before the tensile test As can be seen, several microcracks have been generated along the grain boundaries, as indicated by the dashed lines Such microcracks could lead to a deterioration in mechanical properties It should be noted that no clear microcracks were detected in the other rolled alloys, because of their high ductility

For the cast samples inFig 7, higher tensile strengths are ob-tained for the alloys made at a high solidiﬁcation rate (the RC

Fig 6 Stressestrain curves for copper alloys made by rolling and by casting: (a) CADZ; (b) AlBC; (c) HB.

Định dạng
Số trang	12
Dung lượng	8,83 MB