Materials

& Design
Materials and Design 27 (2006) 719726
www.elsevier.com/locate/matdes

The joint properties of brass plates by friction stir welding

Cemal Meran

Mechanical Engineering Department, Engineering Faculty, Pamukkale University, 20020 Camlik, Denizli, Turkey
Received 20 January 2005; accepted 5 May 2005
Available online 6 July 2005

Abstract
Brass materials widely used as engineering materials in industry because of their high strength, high corrosion resistance, high
electrical and thermal conductivity. They are easily shaped and they possess nice appearance. However, it is dicult to fusion welding of brasses. The main problem of these alloys in fusion welding is the evaporation of the zinc during the welding process. After the
welding, the weld metal becomes porous. Moreover, since the amount of the zinc in the alloy is reduced due to evaporation, the brass
material loses its physical and chemical properties which it normally possesses. Not being solved these problems good enough in
fusion welding of brass materials, it had been directed investigators to apply new methods. It seems that friction stir welding which
is one of the new method developed nowadays, will solve these problems. So, it needs experimental investigations in this subject.
Even though there are so many researches in the literature about friction stir welding in aluminum and its alloys, researches about
copper and its alloys especially brass are limited. In this research, it was pointed out friction stir welding capability especially brass
plates which are 3 mm in thickness. For this reason, experiment conditions were determined and friction stir welding procedures
were applied to brass plates in dierent rotation and welding speeds. Obtained welded joints subjected about physical virtual,
mechanical tests and microstructure investigations and the results had been evaluated.
 2005 Elsevier Ltd. All rights reserved.
Keywords: Welded joint; Brass; Friction stir welding; Rotation speed; Welding speed

1. Introduction
Friction stir welding is a relatively new technique
developed by The Welding Institute (TWI) for the joining of aluminum alloys [1]. Friction stir welding (FSW)
is a relatively new joining process that is presently
attracting considerable interest. The process is solidstate in nature and relies on the localized forging of
the weld zone to produce the joint. In this welding process, a rotating welding tool is driven into the material at
the interface of, for example, two adjoining plates, and
then translated along the interface. Friction stir welding
oers ease of handling, precise external process control

Tel.: +90 258 212 5532; fax: +90 258 212 5538.
E-mail address: cmeran@pamukkale.edu.tr.

0261-3069/$ - see front matter  2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2005.05.006

and high levels of repeatability, thus creating very

homogeneous welds. No special preparation of the sample is required and little waste or pollution is created
during the welding process. Furthermore, its applicability to aluminum alloys, in particular dissimilar alloys or
those considered unweldable by conventional welding
techniques, such as tungsten inert gas (TIG) welding,
make it as an attractive method for the transportation
sector. The friction stir process involves the translation
of a rotating cylindrical tool along the interface between
two plates. Friction heats the material which is then
essentially extruded around the tool before being forged
by the large down pressure. The weld is formed by the
deformation of the material at temperatures below the
melting temperature. The simultaneous rotational and
translational motion of the welding tool during the
welding process creates a characteristic asymmetry

720

C. Meran / Materials and Design 27 (2006) 719726

between the adjoining sides. On one side, where the tool

rotation is with the direction of the translation of the
welding tool one speaks of the advancing side, whereas
on the other side, the two motions, rotation and translation counteract and one speaks of the retreating side [2].
Although the work piece does heat up during friction
stir welding, the temperature does not reach the point of
melting. The friction stir welded joint, then, does not
have the dendritic structure typical of a fusionweld
joint, with its attendant serious degradation of mechanical properties.
Basically, a non-consumable tool with a specially designed rotating probe is entered into the abutting edges
of the sheets or plates to be welded. Once entered, the
rotating tool produces frictional heat and plastic deformation in the workpiece. The tool, or the part, is then
translated along the joint to complete the joining
process.
In evaluating friction stir welding, critical issues (beyond a sound joint) include microstructure control and
localized mechanical property variations. A serious
problem with fusion welding, even when a sound weld
can be made, is the complete alteration of microstructure and the attendant loss of mechanical properties.
Being a solid-state process, friction stir welding has the
potential to avoid signicant changes in microstructure
and mechanical properties [3].
Since friction stir welding (FSW) was invented for the
joining of Cu alloys which are not easily joined with
conventional fusion welding method, FSW has been applied to the joining of Cu alloys [49], Al alloys [1018],
Mg alloys [19], Ti alloys [20] and Fe alloys [21,22] over
the past 10 years. These materials were successfully
joined and also showed higher mechanical properties.
Although a few report have [49] been made on the
weldability of FSW for copper alloys, only few attempts
have been made thus far concerning the application of
the FSW process for CuZn alloys and observing in detail of the metallurgical and mechanical properties of the
welds.
Welding of brass is usually dicult by conventional
fusion welding processes because the brass has the high
thermal diusivity, which is about 10100 times higher
than in many steels and nickel alloys. Brass materials,
as known, are copper (Cu) and zinc (Zn) alloys. The
welding temperature, for example in the TIG welding,
reaches almost 4200 C in the electrode (cathode) and
nearly 3200 C in the work piece (anode) during welding. These temperatures exceed both the melting and
boiling temperatures of both copper (1083/2590 C)

and zinc (419/907 C). In addition, the melting temperature of zinc oxide occurs as a result of zinc evaporation
is 1970 C. For this reason, during welding it is possible
for zinc and copper to evaporate from the work piece.
Zinc with the lower boiling temperature evaporates
more than copper [23].
To overcome prevent zinc and copper evaporation,
the FSW which is one of the solid-state welding techniques is applied to the joining of brass. In this present
study, we explored the friction stir welding of brass
plates which 3-mm thickness. This study especially characterized and compared of microstructures and mechanical properties, and the correspondence between the
mechanical properties and microstructures changes in
a wide range of the welding conditions for 70%Cu
30%Zn alloy.
This is clear and environment protective process because of not having negative subjects such as arc formation could be caused industrial accident, radiation,
releasing of toxic gas or harmful laser beam for human
being eyes. Since there is no arc formation which can
cause industrial accidents, radiation, toxic gas releasing
and harmful laser beam for human eyes, it is a clean and
environment friendly process. Besides, in this method
cracks and porosite formation do not occur, because it
is a solid-state welding method with low total heat input.
In this method, the lost of the strength in welding zone is
lower level than all welding methods of Cu-alloys because of low heat input.
There are three factors that determine joint characteristics obtained by friction stir welding used denite pin
surface geometry. These are pins rotation speed, pins
welding speed and pins depth of sinking. Pins rotation
speed and pins welding speed can be controlled easily.
But, pins depth of sinking is so critical factor and dicult to be controlled. The depth of sinking must be constant along welding process. But, that is not possible to
provide especially in joint process of long plates unless
surfaces are smooth. Therefore, preparation of surface
before welding is so critical and must be paid attention
to this case.

2. Experimental procedure
Friction stir welds were produced in CuZn30 alloy (materials number CW505L) plates of 3-mm
thickness. Tables 1 and 2 show the chemical composition and mechanical properties of this material,
respectively.

Table 1
Chemical composition of the brass (CuZn30 (CW508L) (EN 1652))
%Sn

%Pb

%P

%Mn

%Fe

%Ni

%Si

%Mg

%Al

%S

%Zn

%Cu

0.0010

0.0110

0.00213

0.00050

0.0146

0.00196

0.00251

0.00050

0.00050

0.00100

29.23

70.73

C. Meran / Materials and Design 27 (2006) 719726

721

Table 2
Mechanical properties of base materials (CuZn30)
Material

Tensile strength Rm (MPa)

Yield strength Rp0.2 (MPa)

Elongation (%)

E (kN/mm2)

CuZn30 (CW508L)

360

215

67

110

The stirrer was made by special hot working steel,

X32CrMoV12-28 (EN ISO 4957, material number
1.2365) during FSW of brass, as shown in Fig. 2. Length
of stirrer tip had been chosen 0.1 mm lower than thickness of material to obtain better stirring eect.
The experiments were performed in Laboratory of
University of FH-Regensburg-Germany. The welds
were made on one side using two combinations of tools
rotation speed and welding speed as shown in Table 3
where each test result was obtained from the average
of three tests. The tilt angle for all welds was maintained
at 90 as shown in Fig. 1. Stirrer tip had been rotated in
clockwise in all welding. Besides, for all specimens M5
right screw had been threaded to stirrer tip except No.
4 specimen on which M5 left screw had been threaded.
Considering previous works the welding process was
carried out by rotating the stirrer at 2050 rpm and by
moving the plates at various speeds under a constant friction force. By using the predetermined welding parameters, three dierent samples were welded for mechanical
tests and metallographic examinations. Tension and
hardness measurements were made as mechanical tests.
Scanning electron microscopy (SEM) examinations were
carried out on the weld and base materials.

Fig. 1. Friction stir welding setup of brass materials.

3. Results and discussion

Fig. 2. Geometry and dimensions of stirrer.

To carry out FSW process, two brass plates, 3 mm in

thickness, 120 mm length and 60 mm in width, were
place on a at steel plate. These two brass plates were
then clamped with a vice so that they would not separate
during welding process, Fig. 1. During the experiments
it had been used a milling machine which has high rotation speed to rotate stirrer tip. Besides, specimen to be
welded had been cut by laser in desired dimensions.
Thus, stirrer tip had been prevented to contact on dierent points surface of metal that will be welded.

Unchanged yellowish colour of brass and smooth formation of joint are both important parameters in welding of brass materials. For this reason, welded joints had
been inspected according to ANSI/AWS B1.11-88
(guide for the visual inspection of welds).
First defect detected on welded joints is lower weld
surface quality. This defect occurs because of unstable
pins action caused by improper contact to surface and
improper sank of sinker to plate during weld line. Second defect is unstirred zone near bottom because of
insucient sinking depth of sinker tip. Third defect is
porosite formation look like tunnel. It was determined

Table 3
Friction stir welding parameters for the square groove butt welding
Sample

Rotation speed (rpm)

Welding speed (mm/min)

Screw properties

Remark

1
2
3
4
5
6
7

2050
2050
2050
2050
2050
2050
2050

20
40
56
56
80
112
140

M5 1.5
M5 1.5
M5 1.5
M5 1.5
M5 1.5
M5 1.5
M5 1.5

Metal edges stripped from surface were detected

Lesser edges stripped from surface
Rough welding surface
Less penetration
A good welding bead
A good welding bead
Less penetration

right-hand thread
right-hand thread
right-hand thread
left-hand thread
right-hand thread
right-hand thread
right-hand thread

722

C. Meran / Materials and Design 27 (2006) 719726

Fig. 3. The appearance of upper surface and root side of welding beads of brass plates produced with friction stir welding.

that screw direction and smoothness of prole have

important role.
Main reasons of welding defects in friction stir welding are unsuitable sinker tip design and process parameters. Better sinker tip design, better control of dipping
process and constant depth of sinking during welding
can provide perfect weld [10]. It can be obtained perfect
weld by eliminating this weld defects with a good sinker
tip design, good controlling of being dipped the pin process that will be welded and keep the depth of sink constantly by adjusting during welding [10].
Metal edges stripped from surface were detected because of overheating in the welding speed of 20 mm/

min. Also it has been determined that faster welding

speed creates better results in both top and root joints.
It must be provided recommended heat input to welding zone for good joint. The welding speed of 140 mm/

Fig. 4. Dimension of tensile test specimen.

C. Meran / Materials and Design 27 (2006) 719726

723

It seems that weld joints mechanical properties improve and its appearance gets better if welding speed is
being raised in 2050 rpm. Weld joints breakages occur
either in stirred welding zone or heat aected zone. It
has been determined that welded joints tensile properties
are satisfactory even though not comparable with base
materials. The best weld joint appearance has been obtained in the welding speed of 112 mm/min and
2050 rpm rotation speed. However, it has not occurred
expected joint with higher welding speed.

min has not provided sucient contact in weld joint because of over speed of welding (Fig. 3).
To determine mechanical properties of obtained weld
joints, it has been performed tensile tests according to
EN 895 by preparing three test specimens. Their details
are given in Fig. 4. Average tensile strength and yield
strength are given in Fig. 5 matched with strength properties of base metal.

Vickers hardness, HV0, 1N

200

Fig. 5. Tensile test results of CuZn30 at dierent welding speed with

the rotation speed 2050 rpm, where Rm is tensile strength and Rp0.2 is
0.2%. Yield strength of welded joint: *, fractured in stirred zone; **,
fractured in heat aected zone.

160
120
80

Half of
stirred
zone

40
0
0

10

Distance from center of stired zone,mm

Fig. 8. Harness proles in cross-section of the welds at 2050 rpm and
112 mm/min.

Fig. 6. Optical microstructures in cross-section of the welds at 2050 rpm112 mm/min.

Fig. 7. Eect of thread way on FSW of brass.

724

C. Meran / Materials and Design 27 (2006) 719726

Fig. 9. SEM microstructures and chemical composition of: (a) stirred welding zone; (b) heat aected zone; (c) base metal (welding conditions:
2050 rpm and 112 mm/min).

Fig. 10. The chemical analysis results of the welding bead by FSW, 2050 rpm and 112 mm/min.

C. Meran / Materials and Design 27 (2006) 719726

During welding of brass materials, three dierent

microstructures occurred. Base material, heat aected
zone (HAZ) and stirred welding, Fig. 6.
Experiments were performed in the welding speed of
56 mm/min using both left-hand thread and right-hand
thread screw as seemed Fig. 7. Results showed that
while stirrer tip rotating in clockwise, right-hand thread
created good weld joints, however, left-hand thread did
not create good joints because of insucient stirring of
metal. So, it must be paid attention to suitable both stirrer tip direction and rotation direction during welding.
If there is a defect on surface of base plate (like holes,
etc.), softened metal yields to these holes during welding.
So surface of bottom plates must at and clean.
Results of micro Vickers hardness measurement on
best obtained weld joints under condition of 2050 rpm
and 112 mm/min can be seen in Fig. 8. As expected
hardness had been obtained in the best stirred welding
zone and decreased to same level with base metal.
SEM images of welding zone that obtained by stirred,
heat aected zone and base metal zone are given in Fig. 9.
Besides, chemical analysis was performed on three dierent points of each zone. Measurements showed that these
three zones chemical composition stays unchanged, only
microscopic structures appearance changes.
As seen from the results of chemical analysis of the
internal structure shown in Fig 10, chemical compositions of root side (regions 1 and 2) and work piece which
is not aected by the welding (region 3) are same. The
chemical analyses results of the upper surface of the
welding bead (regions 4 and 5) show that there is not
much dierence in the quantity of copper and zinc. This
result shows that evaporation of zinc and copper obtained in fusion welding, is being prevented by friction
stir welding.

4. Conclusion
To clarify the feasibility of FSW process to CuZn30,
the characteristic of structures, mechanical properties
and microstructures of friction stir welds were investigated in dierent welding test conditions. The main conclusions are summarized as follows:
1. Especially in welding of at brass plates, friction stir
welding gives good results for some points according
to fusion welding.
2. Evaporation of zinc and copper which makes welding
more dicult disappears in friction stir welding
because of not reaching to melting point of metal
during welding.
3. The formation of pore in microstructure of welding
zone disappears. However, some pores have been
observed in zones where stirring had not been done
well.

725

4. Since there was no evaporation, colour of brass material did not change.
5. It is not recommended additional metal that used
when needed and shielding gas which has to be used
in fusion welding.
6. The mechanical properties of obtained weld joints
can reach to base metal strength level if suitable welding parameters are determined. Fractures usually
occur either in heat aected zone or in weld joint,
however, fractures had occurred on base material
under conditions of 2050 rpm and 112 mm/min.
7. Geometry of stirrer tip screw must be compatible
with stirrer tip direction of rotation. If there is
right-hand thread screw, stirrer tip should be rotated
clockwise, if there is left-hand thread screw stirrer tip
should be rotated unclockwise. Otherwise the metal
that softened like paste can not be stirred.
8. Surface of base plate should be at. Root joint of
softened weld metal is occurred straight in this
way.

Acknowledgements
The author thank Mr. Volkan Kovan for his assistance, I also emphasize my deepest appreciation to Prof.
Dr. Joahim Hammer for valuable help and his quidence,
and Cagdas Makine Kalip San. Denizli.

References
[1] Thomas WM, et al. Friction Stir Butt Welding, International
Patent Appl. No. PCT/GB92/02203 and GB Patent Appl. No.
9125978.8, Dec. 1991, U.S. Patent No. 5460,317.
[2] Peel M, Steuwer A, Preuss M, Withers PJ. Microstructure,
mechanical properties and residual stresses as a function of
welding speed in aluminum AA5083 friction stir welds. Acta
Mater 2003;51:4791801.
[3] Rhodes CG, Mahoney MW, Bingel WH. Microstructural changes
determining joint strength in friction stir welding of aluminum
alloys. Scripta Mater 1997;36(1):1569.
[4] Park HS, Kimura T, Murakami T, Nagano Y, Nakata K, Ushio
M. Microstructures and mechanical properties of friction stir
welds of 60% Cu40% Zn copper alloy. Mater Sci Eng A
2004;371:1609.
[5] Murr LE. Dynamic recrystallization in friction stir welding of
aluminum alloy 1100. J Mater Sci Lett 1997;16:180113.
[6] Andersson CG, Andrews RE. Fabrication of containment canisters for nuclear waste by friction stir welding. In: Proceedings of
the 1st international symposium on friction stir welding, Thousand Oaks, CA, USA; 1416 June 1999.
[7] Andersson CG, Andrews RE, Dance BGI, Russel MJ, Olden EJ,
Sanderson RM. Comparison of copper canister fabrication by the
electron beam and friction stir processes. In: Proceedings of the
2nd international symposium on friction stir welding, Gothenburg, Sweden; 2729 June 2000.
[8] Okamoto K, Doi M, Hirano S, Aota K, Okamura H, Aono Y,
Ping TC. Fabrication of backing plate of copper alloy by friction

726

[9]

[10]

[11]

[12]

[13]
[14]
[15]

C. Meran / Materials and Design 27 (2006) 719726

stir welding. In: Proceedings of the 3rd international symposium
on friction stir welding, Kobe, Japan; 2728 September
2001.
Murr LE, Liu G, McClure JC. A TEM study of precipitation and
related microstructures in friction stir welded 6061 aluminum. J
Mater Sci 1998;33:124351.
Cam G. Al Alasimlari Icin Gelistirilen Yeni Kaynak Yontemleri,
Kaynak Teknolojileri III. Ulusal Kongresi, Istanbul; 2001. p.
26777.
Dawes CJ. Friction stir welding, The Welding Institute, Abington
Hall, Cambridge; 1999 [Adapted for TALAT by Roy Woodward,
Birmingham and Christian Leroy, EAA].
Lee WB, Yeon YM, Jung SB. The improvement of mechanical
properties of friction stir welded A356 Al alloy, materials. Sci Eng
A 2003;355:60712.
Johnson R, Kale S. Friction stir welding. Mater World
1999;7(12):7513.
Kalle S, Nicholas D. Causing a stir in the future. Weld Join
1998;February:1821.
Kaluc E, Bozduman B. Surtunen Elemanlarla Birlestirme. Makina Magazin, Say:27, 1998.

[16] Kulekci MK. Mechanical properties of friction stir welded joints

of AlCu4SiMg aluminum alloy. Metal Mater Kovove Materialy
2003;41(2):208.
[17] Mahoney MW. Science friction. Weld Join 1997;January/
February:712.
[18] Sohonida T, Kondoh Y. Friction stir welding of aluminum plate.
Journal JFWA 1995;2:1247.
[19] Lee WB, Jung SB. The joint properties of copper by friction stir
welding. Mater Lett 2004;58:10416.
[20] Hori H et al. Joint strength of thick sheet welded by friction stir
welding. In: Proceedings of the 3rd international symposium on
friction stir welding, Kobe, Japan; 2728 September 2001.
[21] Konkol PJ, Mathers JA, Pickens JP. Friction stir welding of
HSLA-65 steel for shipbuilding. In: Proceedings of the 3rd
international symposium on friction stir welding, Kobe, Japan;
2728 September 2001.
[22] Tomas WM, Threadgil PL, Nicholas ED. The feasibility of
friction stir welding steel. Sci Technol Weld Join 1999;4:365772.
[23] Meran C. Weldability of brass materials and determination of
available welding parameters, Ph.D. Thesis, Pamukkale University, Denizli, Turkey; 2001 [in Turkish].