M A TE RI A L S C H A RAC TE RI ZA T ION 5 9 ( 2 00 8 ) 1 4 8 1–1 4 9 0

Diffusion bonding of commercially pure titanium to low carbon

steel using a silver interlayer

Evren Atasoy a , Nizamettin Kahraman b,⁎

a
Tokat Adocim Cement Factory, 60100, Tokat, Turkey
b
Karabuk University, Technical Education Faculty, 78050, Karabuk, Turkey

AR TIC LE D ATA ABSTR ACT

Article history: Titanium and low carbon steel plates were joined through diffusion bonding using a silver
Received 3 September 2007 interlayer at various temperatures for various diffusion times. In order to determine the
Received in revised form strength of the resulting joints, tensile-shear tests and hardness tests were applied.
11 January 2008 Additionally, optical, scanning electron microscopy examinations and energy dispersive
Accepted 11 January 2008 spectrometry elemental analyses were carried out to determine the interface properties of
the joint. The work showed that the highest interface strength was obtained for the
Keywords: specimens joined at 850 °C for 90 min. It was seen from the hardness results that the highest
Diffusion welding/bonding hardness value was obtained for the interlayer material and the hardness values on the both
Titanium sides of the interlayer decreased gradually as the distance from the joint increased. In
Interlayer energy dispersive spectrometry analyses, it was seen that the amount of silver in the
Microstructure interlayer decreased markedly depending on the temperature rise. In addition, increasing
diffusion time also caused some slight decrease in the amount of silver.
© 2008 Elsevier Inc. All rights reserved.

1. Introduction metal of choice for dental implants due to its biocompatibility

and superior physical and chemical properties, high surface
Titanium and its alloys have been considered as one of the energy, medium elastic modulus and chemical inertness [13].
best engineering materials for use in industrial applications With the increased use of Ti, the bonding of Ti and its alloys
[1,2]. They are known with their corrosion resistance due to a has become more important [14]. Diffusion bonding has en-
stable, protective and strongly adherent oxide film layer [3]. abled the fabrication of complex titanium structures [15].
Titanium and its alloys have high specific strength and good Welding of titanium and its alloys is difficult as titanium is
erosion resistance; thus, they have been widely used in aero- extremely chemically reactive at high temperatures. During
space and chemical industries [4]. Titanium and its alloys welding, titanium alloys pick up oxygen and nitrogen from the
have been used in a number of applications in industry due to atmosphere easily [1,16]. For this reason, diffusion bonding is
their excellent corrosion resistant [5], low thermal conductiv-
Table 1 – Chemical composition of pure Ti sheet (wt.%)
ity [6], low specific gravity [7], low density which makes them
attractive for aerospace applications [8] and relatively high ASTM Transformation Alloying elements (wt.%)
melting temperature [9]. standard temperatures (°C)
Commercially pure titanium is widely used as dental im- Alpha (α) Beta (β) N C H Fe O
plant material because of its appropriate mechanical pro-
Grade 2 913 890 0.03 0.10 0.015 0.30 0.25
perties and excellent biocompatibility [10–12]. Titanium is the

⁎ Corresponding author.
E-mail address: nizamettinkahraman@gmail.com (N. Kahraman).

1044-5803/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.matchar.2008.01.015
1482 M A TE RI A L S C H A RAC TE RI ZA T ION 5 9 ( 2 00 8 ) 1 4 8 1–1 4 9 0

Table 2 – Chemical composition of low carbon steel (wt.%)

Element C Si Mn P S Cr Ni Mo Cu Al Ti Fe

Weight (%) 0.0986 0.022 0.0307 0.015 0.0183 0.0363 0.0233 0.006 0.026 0.0586 0.001 Balance

a recommended joining method [17–19]. Diffusion bonding is a applied. The transverse microhardness measurements across
solid-state bonding method [20]. Diffusion bonding provides a the silver interlayer, the two interfaces (titanium-silver and
novel joining operation for similar and dissimilar materials steel silver) and the base materials were carried out. One of the
without gross microscopic distortion and with minimum di- microhardness measurements was taken in the centre of the
mensional tolerance [21,22] and almost no phase transforma- interlayer and the others on both sides of the bonded
tion or microstructural change occurs during the process [23]. specimens by the distances of 25, 40, 40, and 200 µm away
Diffusion bonding can minimise structural in-homogeneities from the silver interlayer. Three measurements were carried
due to negligible effect of temperature gradient, which results out from each point and the results were averaged. From the
in considerable cost and weight savings in comparison to tra- diffusion bonded assemblies a transverse section was taken
ditional joining technique and provides a near net shape form- and then prepared by usual techniques (grinding and polish-
ing process [24]. ing) for metallographic observation. The Ti side was etched
The present study considers the shear strengths of diffu- with a mixture of 73% H2O–16% HF–6% HNO3–5% HCl while the
sion bonding of commercially pure titanium and low carbon low carbon steel side was etched with 2% nitric acid + 98%
steel using silver as interlayer at the temperatures of 700, 750, methanol. A NIKON Epiphot 200 optical microscopy was used
800, and 850 °C. Four different bonding durations (30, 60, 90 for the microscopic examination of the etched specimens.
and 120 min) were used in the study. This investigation also Scanning electron microscopy (SEM) examinations were car-
examines the influence of bonding temperature and diffusion ried out on these specimens using a JEOL 6060 LV SEM device
time on the interface microstructure and tensile-shear proper- and the diffusion zones were analysed by energy dispersive
ties of the bonded assemblies. spectrometry (EDS).

2. Experimental Procedure 3. Results and Discussions

In this study, commercially pure titanium (ASTM Grade 2) was The results of diffusion bonding operations showed that bon-
bonded to low carbon steel by diffusion bonding method. The ding was effected for almost all the conditions. On the other
bonding was carried out using a silver interlayer, which was hand, bonding at 700 °C and for 30 and 60 min of diffusion times
supplied in sheets of 10 × 10 × 0.08 mm. The hardness of the could not be achieved. The failure of bonding at 700 °C for 30 and
original materials were measured as 122, 132, 268 HV for low 60 min of diffusion times can be attributed to both the low
carbon steel, titanium and silver, respectively. Chemical temperature and the insufficient diffusion times. When diffu-
analyses of all the materials are given in Tables 1, 2, and 3. sion time was increased to 90 min or bonding temperature was
Specimens were prepared in the dimensions of 40 × increased to 750 °C without any increment in diffusion time,
10 × 1.5 mm and all the specimens' surfaces were subjected bonding could be effected. It is well known that adequate heat,
to mirror-like polishing using the conventional metallo- diffusion time, and pressure are required for atoms to diffuse in
graphic polishing steps. And then, they were cleaned by this bonding method. Diffusion time is a dependent operation
alcohol prior to the welding. Samples were then placed in the parameter and is interrelated with temperature, pressure, and
chamber of the diffusion welding equipment (Fig. 1). The the type of bonding. Time needed for bonding different
samples were oriented as shown in Fig. 2. In order to eliminate
the oxidation problem, Ar gas was introduced into the test
chamber during welding. Temperature was then increased up
to the test temperature in steps of 20 °C/min and the samples
were kept there for predetermined diffusion times under
3 MPa uniaxial load and then cooled down to the room tem-
perature. Welding parameters for these experiments are given
in Table 4.
To determine the mechanical properties of the bonding,
tensile-shear test was carried out on the bonded specimens
(Autograph-Shimadzu) with the speed of 1 mm s− 1. Hardness
specimens were mounted in bakelite. Microhardness was
measured using a Shimadzu HMV unit and 200 g load was

Table 3 – Chemical composition of silver (wt.%)

Element Ag Cu Zn Fe Ti

Weight (%) 51,043 36,730 11,810 0,294 0,124

Fig. 1 – Experimental setup for diffusion bonding.
 M A TE RI A L S C H A RAC TE RI ZA T ION 5 9 ( 2 00 8 ) 1 4 8 1–1 4 9 0 1483

Table 4 – Diffusion bonding parameters used in this study

Material Temperature Welding time
(°C) (min)

Ti–Ag–Fe 1 700 30
Ti–Ag–Fe 2 700 60
Ti–Ag–Fe 3 700 90
Ti–Ag–Fe 4 700 120
Ti–Ag–Fe 5 750 30
Ti–Ag–Fe 6 750 60
Ti–Ag–Fe 7 750 90
Ti–Ag–Fe 8 750 120
Ti–Ag–Fe 9 800 30 Fig. 2 – Dimensions of the diffusion bonded specimens (mm).
Ti–Ag–Fe 10 800 60
Ti–Ag–Fe 11 800 90 materials or material couples can vary from several seconds to
Ti–Ag–Fe 12 800 120
several hours. Longer diffusion time is not only disadvanta-
Ti–Ag–Fe 13 850 30
Ti–Ag–Fe 14 850 60
geous for bonding economy, but also it leads to some undesir-
Ti–Ag–Fe 15 850 90 able effects which deteriorate the mechanical properties of
Ti–Ag–Fe 16 850 120 bonding such as; porosity formation, alteration of the composi-
tion and formation of brittle intermetallic compounds [25].

Fig. 3 – Test graphs of tensile-shear tests for the specimens bonded at (a) 700 °C, (b) 750 °C, (c) 800°C and (d) 850 °C for 120 min.
1484 M A TE RI A L S C H A RAC TE RI ZA T ION 5 9 ( 2 00 8 ) 1 4 8 1–1 4 9 0

When the tensile curves in Fig. 4 are examined, it is seen

that per cent elongation increases with increasing tempera-
ture. It is 0.4% for 700 °C and 8% for 850 °C. Therefore, it can be
said that per cent elongations of the bonded specimens in-
crease with increasing temperature.
The influence of temperature and diffusion time on the
tensile-shear behaviour is seen from these test results. At
700 °C temperature and under 3 MPa pressure, the shear
strength is found to be 1415.7 N for 90 min of diffusion time.
When the diffusion time is increased to 120 min at the same
temperature and pressure, the tensile-shear strength
decreases to 1077.9 N. Similarly, the tensile-shear strength of
specimens increases steadily with increasing diffusion time
Fig. 4 – Results of tensile-shear test. up to 90 min at 750, 800 and 850 °C temperatures and under
3 MPa holding pressure. However, the diffusion times of more
than 90 min results in decrease in strength, Fig. 4. The results
3.1. Tensile-shear Test of the experimental studies show that tensile-shear strength
increases with increasing temperature. However, it can be
Fig. 3 gives the tensile-shear curves of the diffusion welded seen clearly from the Fig. 4 that increase in strength slows
specimens bonded at various temperatures (700, 750, 800, and down depending on the increasing temperature.
850 °C) for 120 min of diffusion time while Fig. 4 gives the According to the tensile-shear test results, the highest tensile-
tensile-shear test results. The specimens subjected to tensile- shear strength was found to be 3222.8 N for the specimen bonded
shear tests showed that fracture mainly took place at the low at 850 °C for 90 min. The lowest strength was found to be 1077.9 N
carbon steel–silver interface of the bonded specimens. In for the specimen bonded at 700 °C for 120 min. It can be concluded
addition, it was seen that there was no significant reduction in from Fig. 4 that tensile-shear strength increases with increasing
the cross-sectional area of the test specimens after the frac- temperature. However, as diffusion time increases, the bond
ture and therefore there was no significant changes in the part strength increases until a maximum value is reached beyond
geometry. which it decreases at all the temperatures.

Fig. 5 – Hardness results taken from the specimens bonded at (a) 700, (b) 750, (c) 800 and (d) 850 °C.
 M A TE RI A L S C H A RAC TE RI ZA T ION 5 9 ( 2 00 8 ) 1 4 8 1–1 4 9 0 1485

Fig. 6 – Macrostructures of the specimens obtained at 750 and 850 °C for various diffusion times.

In their study, Peng et al. [4] stated that strength increased interlayer as distance from the joint increases. In addition, it
with increasing diffusion time up to a certain point and then it was determined that hardness values taken from the speci-
decreased due to intermetallic formation in diffusion welding. mens (titanium, low carbon steel, and silver as interlayer)
Ghosh and Chatterjee [26] bonded pure titanium and 304 bonded at 700, 750, and 800 °C are lower than the values taken
stainless steel couples at 950 °C for diffusion times of 30, 60, 90 from the original materials. However, the hardness value of
and 120 min and determined that 30 min of diffusion time led the silver interlayer of the specimen bonded at 850 °C is sig-
to the highest strength. They also determined that strength nificantly higher (about 450 HV). The decrease in hardness
decreased sharply with increasing diffusion time. In their values after bonding can be attributed to the decrease in
work, this sharp decrease was attributed to the growth of number of dislocations during bonding, which were relatively
intermetallic compounds and formation of voids. In a similar abundant in the original material. Another reason for this
study, Liming et al. [27] bonded different metallic materials reduction in hardness may be grain growth which occurs in the
at various diffusion temperatures and reported that shear both materials due to high temperature and long diffusion time.
strength increased by diffusion temperature. When the hardness values measured on the interlayer of
the bonded specimens are compared to each other, it can be
3.2. Hardness Test seen that hardness increases with increasing temperature,
however, there is no evidence of change in hardness depend-
Fig. 5 shows the hardness test results obtained from the ing on the diffusion time. Increasing hardness with increasing
diffusion bonded joints. When the curves in Fig. 5 are gene- temperature can be attributed to the formation of interme-
rally examined, it is seen that the highest hardness value tallic compounds at high temperatures.
among the all specimens belongs to the interlayer material Kolukisa [28], Orhan et al. [19], and Yildirim and Keleste-
and hardness decreases gradually at the both sides of the mur, [29] bonded various materials by diffusion bonding at

Fig. 7 – Microstructures of the original materials.

1486 M A TE RI A L S C H A RAC TE RI ZA T ION 5 9 ( 2 00 8 ) 1 4 8 1–1 4 9 0

Fig. 8 – Microstructures of the specimens bonded at 700 °C for diffusion times of 90 and 120 min.

different temperatures. They reported that the highest hard- (Rene 80) using a nickel alloy interlayer at different environ-
ness values were obtained from interfaces and the hardness ments. They reported in their study that the highest hardness
decreased as the distance from the interface increased. was observed at the interface and that the hardness decreased
Similarly Ekrami et al. [30] bonded nickel based super alloys as the distance from the interface increased.

Fig. 9 – Microstructures of the specimens bonded at 850 °C for diffusion times of (a) 30, (b) 60, (c) 90 and (d) 120 min.
 M A TE RI A L S C H A RAC TE RI ZA T ION 5 9 ( 2 00 8 ) 1 4 8 1–1 4 9 0 1487

Fig. 10 – EDS and EDS line element analysis taken from the interlayer of the specimen bonded at 700 °C for 120 min.

Fig. 11 – EDS and EDS line element analysis taken from the interlayer of the specimen bonded at 850 °C for 120 min.
1488 M A TE RI A L S C H A RAC TE RI ZA T ION 5 9 ( 2 00 8 ) 1 4 8 1–1 4 9 0

creeping deformation, c—diffusion and recrystallization and

grain boundary immigration), the reduction in the bonding
region area with increasing temperature and diffusion time is
regarded as quite normal. That is because, the migration of
atoms from the interlayer to the both sides increases with
increasing temperature and diffusion time under load. This, in
turn, decreases the thickness of the interlayer. Similar findings
were also reported in Ref. [25].
When the bonded specimens were examined by naked eye,
it was seen that metal flow from the interface took place and
this, in turn, caused the interlayer to get smaller. This reduc-
tion was obvious especially for the specimens bonded at high
temperatures and for longer diffusion times.

3.4. Microstructure

Fig. 7 shows the microstructure of the low carbon steel and the
titanium before they were diffusion welded. Both materials
originally consist of equiaxed grains and the grains are almost
similar in size.
The microstructures of the joints bonded at 700 °C are
similar to those of the original materials and their grain sizes
are also the same as those of the original materials, Fig. 8.
However, grain growth occurs due to the temperature rise in
the microstructures of the joints bonded at 850 °C, Fig. 9.
Microstructural analysis indicates that the change in grain
morphology is more obvious for the titanium with the in-
creasing diffusion temperature and time than for the steel.
Equiaxed and more homogeneous grains are seen in these
specimens due to absence of HAZ (heat affected zone), which
Fig. 12 – (a) Weight percent of silver in the interlayer versus exist in the fusion welding methods.
temperature and time (b) weight percent of silver and copper Grain growth in bonded materials can be attributed to
versus temperature for 120 min. recrystallization and to the enveloping of small grains by
bigger ones. The tendency for grain growth after recrystalliza-
tion is related to grain boundary energy. In order to obtain a
3.3. Macroscopic Examination lower level of energy, total grain boundary per unit volume
needs decreasing and this, in turn, requires the growth of
The effects of temperature and diffusion time on bonding area grains. However, it is well known that grain growth is not
were examined in detail using an optical microscope. The desirable [31,32].
photos of the bonding areas were taken before the etching, Fig. 6.
From the optical macroscopy, the bonding region decreased in 3.5. SEM and EDS Examinations
area with increasing temperature and diffusion time for all the
specimens. If it is considered that diffusion bonding is generally The SEM and EDS analyses were performed on the silver
effected at four stages (a—plastic deformation under load, b— interlayer of the specimens diffusion bonded at 700 and 850 °C
for diffusion time of 120 min. The SEM and EDS analysis

Fig. 13 – Weight percent variations of silver and copper versus Fig. 14 – Weight percent variations of copper and silver at
the diffusion time at 850 °C. different temperatures for 120 min.
 M A TE RI A L S C H A RAC TE RI ZA T ION 5 9 ( 2 00 8 ) 1 4 8 1–1 4 9 0 1489

between 1.830% and 3.825% after the bonding. Such a significant

loss of the silver in the interlayer at high temperatures can be
attributed to partial melting of the interlayer during diffusion
bonding and to flow of this melted metal from the interlayer.
Fig. 14 shows the weight percent variations of copper and
silver both in the interlayer and in areas next to the interlayer
for the specimens bonded at 700, 750, 800 and 850 °C and for
120 min. The weight percent of silver and copper are
maximum at the interlayer and their diffusion rates decrease
as the distance from the interlayer increases. In addition, the
weight percent of copper in the interlayer is higher than that
of silver and generally the weight percent of both elements in
Fig. 15 – Weight percent variations of copper and silver at the interlayer decrease with increasing temperature.
850 °C for different diffusion times. Fig. 15 shows the weight percent variations of copper and
silver at the interlayer and at the regions next to the interlayer
for the specimens bonded at 850 °C for 30, 60, 90 and 120 min.
results taken from the interlayer (the area denoted by 3 in the In Fig. 15, the curves with high values represent copper quantity
photos) are given in Figs. 10 and 11. The weight percent of while the curves with low values represent silver quantity.
silver in the interlayer significantly decreases with increasing Similarly, the atom concentration diffusing to the both metals
temperature from 66.2 to 3.825% while the weight percent of next to the interlayer from the interlayer continued in a de-
titanium and iron significantly increase from 3.708 to 52.243% creasing manner and it was observed that the interlayer material
and from 3.295 to 8.982%, respectively, Figs. 10 and 11. This can diffused to the titanium rather than the low carbon steel.
also be seen clearly from the element analysis scanning When all the analyses results are evaluated generally, both the
results. These figures clearly show that the diffusion of titanium amount of silver in the interlayer and the silver diffused are
in silver is higher than that of iron in silver. This can generally be considered to be less than the amount of copper although the
explained by the lower atomic weight (atomic mass) of titanium original interlayer material contains the highest amount of silver.
compared to that of iron. Atomic weights of the elements In addition, it was also determined that the amount of silver
analysed were found to be 47.9 g/mol for titanium, 55.85 g/mol decreased with increasing diffusion time and increasing
for iron, 63.54 g/mol for copper and 107.87 g/mol silver. temperature.
Fig. 12 gives the temperature dependent variation of When the temperature and diffusion times are taken into
the silver and copper in the interlayer. It can be seen from consideration, it is seen that diffusion depends mainly on
Fig. 12(a) that the weight percent of silver in the interlayer temperature and diffusion increases with diffusion time.
significantly decreases with increasing temperature. How- However, this increase slows down as the diffusion time
ever, increasing diffusion time slightly decreases weight per- increase. This finding is in agreement with Fick's Second Law,
cent of silver. By visual examination of the bonded specimens, a partial differential equation that describes the rate at which
some metal flow was seen from the bond interface to the atoms are redistributed in a material by diffusion [31,32].
outside at high temperatures and for long diffusion times.
Therefore, it can be said that the metal flowed from the
interface to the outside is mainly silver. Fig. 12(b) indicates 4. Conclusions
that the weight percent of silver in the interlayer continuously
decrease with increasing temperature for the specimens • For the diffusion bonding of titanium and steel using a silver
bonded for 120 min. For the same specimens, the weight interlayer, the maximum tensile-shear strength was
percent of copper increases with increasing temperature up to obtained for the specimen bonded at 850 °C for 90 min.
800 °C and then decreases significantly with further increase • The tensile-shear strength of the bonded specimens was
in temperature. found to increase with increasing temperature until a
Fig. 13 gives the variation of the weight percent of silver and maximum value is reached beyond which it decreased.
copper in the interlayer as a function of diffusion time at 850 °C. It • The interlayer had the highest hardness and the hardness
is seen that the weight percent of copper in the interlayer is values on the both sides of the interlayer decreased
almost as much as that of copper in the original material (the gradually as distance from the joint increased.
weight percent of copper in the original material is 36.730%). From • The macro- and microstructural studies revealed that the
the EDS analyses, it is seen that the weight percent of the cross-section of the interlayer decreased with increasing
elements are between; 54.539% and 50.722% for titanium, 8.982% temperature and diffusion time.
and 7.846% for iron, 37.454% and 34.940% for copper and 1.830% • The microstructural studies also showed that grain growth
and 3.825% for silver. Thus, it can be concluded that the weight occurred in both steel and titanium in the bonded specimens and
percent of the elements in the interlayer does not change that it increased with increasing temperature and diffusion time.
significantly with increasing diffusion time. • The EDS analysis revealed that the weight percent of silver in
Contrary to the weight percent of copper, the weight percent of the interlayer markedly decreased with increasing tempera-
silver in the interlayer was found to be significantly lower than that ture. However, it slightly decreased with an increase in the
of the original material for all diffusion times. While the weight diffusion time. In addition, the amount of diffusion decreased
percent of silver in the original material was 51.043%, it decreased to as the distance from the interlayer increased.
1490 M A TE RI A L S C H A RAC TE RI ZA T ION 5 9 ( 2 00 8 ) 1 4 8 1–1 4 9 0

REFERENCES [16] Sun Z, Annergren I, Pan D, Mai TA. Effect of laser surface
remelting on the corrosion behavior of commercially pure
titanium sheet. Mat Sci Eng A 2003;345(1–2):293–300.
[1] Kahraman N, Gulenc B, Findik F. Corrosion and [17] He P, Feng JC, Zhang BG, Qian YY. A new technology for
mechanical–microstructural aspects of dissimilar joints of diffusion bonding intermetallic TiAl to steel with composite
Ti6Al4V and Al plates. Int J Impact Eng 2007;34:1423–32. barrier layers. Mater Charact 2003;50(1):87–92.
[2] Barreda JL, Santamaría F, Azpiroz X, Irisarri AM, Varona JM. [18] Kurt B, Orhan N, Evin E, Çalik A. Diffusion bonding between
Electron beam welded high thickness Ti6Al4V plates using Ti–6Al–4V alloy and ferritic stainless steel. Mater Lett
filler metal of similar and different composition to the base 2007;61(8–9):1747–50.
plate. Vacuum 2001;62(2–3):143–50. [19] Orhan N, Khan TI, Eroğlu M. Diffusion bonding of a microduplex
[3] Lee HS, Yoon JH, Yi YM. Oxidation behavior of titanium stainless steel to Ti–Al–4V. Scripta Mater 2001;45(4):441–6.
alloy under diffusion bonding. Thermochım Acta [20] Özdemir N, Aksoy M, Orhan N. Effect of graphite shape in
2007;455(1–2):105–8. vacuum-free diffusion bonding of nodular cast iron with gray
[4] Peng H, Jiuhai Z, Ronglin Z, Xiaoqiang L. Diffusion bonding cast iron. J Mater Process Tech 2003;141(2):228–33.
technology of a titanium alloy to a stainless steel web with an [21] Kundu S, Chatterjee S. Interfacial microstructure and
Ni interlayer. Mater Charact 1999;43(5):287–92. mechanical properties of diffusion-bonded titanium–stainless
[5] Lee HK, Han HS, Son KJ, Hong SB. Optimization of Nd:YAG steel joints using a nickel interlayer. Mat Sci Eng A
laser welding parameters for sealing small titanium tube 2006;425(1–2):107–13.
ends. Mat Sci Eng A 2006;415(1–2):149–55. [22] Ghosh M, Bhanumurthy K, Kale GB, Krishnan J, Chatterjee S.
[6] Katou M, Oh J, Miyamoto Y, Matsuura K, Kudoh M. Freeform Diffusion bonding of titanium to 304 stainless steel. J Nucl
fabrication of titanium metal and intermetallic alloys by Mater 2003;322(2–3):235–41.
three-dimensional micro welding. Mater Design [23] Arik H, Aydin M, Kurt A, Turker M. Weldability of Al4C3–Al
2007;28(7):2093–8. composites via diffusion welding technique. Mater Design
[7] Liu L, Du X, Zhu M, Chen G. Research on the microstructure 2005;26:555–60.
and properties of weld repairs in TA15 titanium alloy. [24] Ghosh M, Chatterjee S. Diffusion bonded transition joints of
Mat Sci Eng A 2007;445–446:691–6. titanium to stainless steel with improved properties. Mat Sci
[8] Saresh N, Pillai MG, Mathew J. Investigations into the effects Eng A 2003;358(1–2):152–8.
of electron beam welding on thick Ti–6Al–4V titanium alloy. [25] Mahoney MW, Bampton CC. Fundamentals of diffusion
J Mater Process Tech. In Press. Corrected Proof. Available bonding. Welding, Brazing and Soldering. ASM handbook
online 19 April 2007. USA: ASM International Materials Park; 2000. p. 156–9.
[9] Lima MSF, Folio F, Mischler S. Microstructure and surface [26] Ghosh M, Chatterjee S. Effect of interface microstructure on
properties of laser-remelted titanium nitride coatings on the bond strength of the diffusion welded joints between
titanium. Surf Coat Tech 2005;199(1):83–91. titanium and stainless steel. Mater Charact 2005;54:327–37.
[10] Aparicio C, Gil FJ, Fonseca C, Barbosa M, Planell JA. Corrosion [27] Liming L, Meili Z, Longxiu P, Lin W. Studying of micro-bonding
behaviour of commercially pure titanium shot blasted with in diffusion welding joint for composite. Mat Sci Eng A
different materials and sizes of shot particles for dental 2001;315:103–7.
implant applications. Biomaterials 2003;24(2):263–73. [28] Kolukısa S. The effect of the welding temperature on the
[11] Huang HH. Effects of fluoride concentration and elastic weldability in diffusion welding of martensitic (AISI 420)
tensile strain on the corrosion resistance of commercially stainless steel with ductile (spheroidal graphite-nodular) cast
pure titanium. Biomaterials 2002;23(1):59–63. iron. J Mater Process Tech 2007;86:33–6.
[12] Kahraman N. The influence of welding parameters on the [29] Yıldırım S, Kelestemur MH. A study on the solid-state welding
joint strength of resistance spot-welded titanium sheets. of boron-doped Ni3Al–AISI 304 stainless steel couple. Mater
Mater Design 2007;28:420–7. Lett 2005;59:1134–7.
[13] Miguel A, Puig I. Custom-made laser-welded titanium implant [30] Ekrami A, Moeinifar S, Kokabi AH. Effect of Transient liquid
prosthetic abutment. J Prosthet Dent 2005;94(4):401–3. phase diffusion bonding on microstructure and properties of
[14] Lee WB, Lee CY, Chang WS, Yeon YM, Jung SB. Microstructural a nickel base superalloy Rene 80. Mat Sci Eng A 2007;56:93–8.
investigation of friction stir welded pure titanium. Mater Lett [31] Smith WF. Principles of materials science and engineering.
2005;59(26):3315–8. second Edition. New York: McGraw-Hill publishing Company;
[15] Tuppen SJ, Bache MR, Voice WE. A fatigue assessment of 1990. p. 290–6.
dissimilar titanium alloy diffusion bonds. Int J Fatıgue [32] Askeland RD. The science and engineering of materials. Third
2005;27(6): 651–8. Edition. Boston: PWS Publishing Company; 1994. p. 110–27.