metals
Article
Joining Ti6Al4V to Alumina by Diffusion Bonding Using
Titanium Interlayers
Marcionilo Silva, Jr. 1,2,3,4 , Ana S. Ramos 4






Citation: Silva, M., Jr.; Ramos, A.S.;
Simões, S. Joining Ti6Al4V to
Alumina by Diffusion Bonding Using
Titanium Interlayers. Metals 2021, 11,
1728. https://doi.org/10.3390/
met11111728
Academic Editor: Daolun Chen
Received: 28 September 2021
Accepted: 25 October 2021
Published: 29 October 2021
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Metals 2021, 11, 1728. https://doi.org/10.3390/met11111728
and Sónia Simões 2,3, *
1 Department of Mechanical Engineering, Federal University of Amazonas, General Rodrigo Octavio Jordão
Ramos ST, Manaus 69067-005, Brazil; marcionilo@ufam.edu.br
2 DEMM, Department of Metallurgical and Materials Engineering, University of Porto, R. Dr. Roberto Frias,
4200-465 Porto, Portugal
3 LAETA/INEGI, Institute of Science and Innovation in Mechanical and Industrial Engineering, R. Dr. Roberto
Frias, 4200-465 Porto, Portugal
4 Department of Mechanical Engineering, University of Coimbra, CEMMPRE, R. Luís Reis Santos,
3030-788 Coimbra, Portugal; sofia.ramos@dem.uc.pt
* Correspondence: ssimoes@fe.up.pt; Tel.: +351-225081424
Abstract: This work aims to investigate the joining of Ti6Al4V alloy to alumina by diffusion bonding
using titanium interlayers: thin films (1 µm) and commercial titanium foils (5 µm). The Ti thin films
were deposited by magnetron sputtering onto alumina. The joints were processed at 900, 950, and
1000 ◦ C, dwell time of 10 and 60 min, under contact pressure. Experiments without interlayer were
performed for comparison purposes. Microstructural characterization of the interfaces was conducted
by optical microscopy (OM), scanning electron microscopy (SEM) with energy dispersive X-ray
spectroscopy (EDS), and electron backscatter diffraction (EBSD). The mechanical characterization
of the joints was performed by nanoindentation to obtain hardness and reduced Young’s modulus
distribution maps and shear strength tests. Joints processed without interlayer have only been
achieved at 1000 ◦ C. Conversely, joints processed using Ti thin films as interlayer showed promising
results at temperatures of 950 ◦ C for 60 min and 1000 ◦ C for 10 and 60 min, under low pressure.
The Ti adhesion to the alumina is a critical aspect of the diffusion bonding process and the joints
produced with Ti freestanding foils were unsuccessful. The nanoindentation results revealed that the
interfaces show hardness and reduced Young modulus, which reflect the observed microstructure.
The average shear strength values are similar for all joints tested (52 ± 14 MPa for the joint processed
without interlayer and 49 ± 25 MPa for the joint processed with interlayer), which confirms that the
use of the Ti thin film improves the diffusion bonding of the Ti6Al4V alloy to alumina, enabling a
decrease in the joining temperature and time.
Keywords: diffusion bonding; thin film; titanium; Al2 O3 ; sputtering
1. Introduction
Ti6Al4V is the most used commercial titanium alloy due to its excellent performance
and attractive properties, such as high-temperature specific strength, low density, excellent
creep, and corrosion resistance [1]. Combining this titanium alloy with advanced ceramics,
e.g., alumina (Al2 O3 ), which has high thermal stability, excellent wear and corrosion resis-
tance, and chemical inertness [2], can be interesting for applications in the medical, aircraft,
nuclear and micro-electro-mechanical systems (MEMS) sectors [1,3–6]. However, obtain-
ing sound joints with good mechanical properties between titanium alloys and ceramic
materials has been challenging due to their different properties, e.g., thermal conductiv-
ity, coefficient of thermal expansion (CTE), and chemical properties. This mismatch of
properties induces the formation of residual stresses at the joint’s interface during cooling.
Researchers have worked to improve knowledge about the mechanisms that affect the
joining of metals to ceramics [7–14]. The most reported technologies in the literature with
https://www.mdpi.com/journal/metals
Metals 2021, 11, 1728 2 of 17

regards to joining dissimilar materials are brazing and solid-state diffusion bonding [15–17].
Brazing [18–28] is an attractive process that consists of adding a filler metal between the
base materials. Ag-based commercial alloys can be used successfully as brazing alloys since
they improve the reaction between metal and ceramic and enhance wetting [29]. However,
this brazing alloy produces an interface with a lower service temperature than the titanium
alloy. Furthermore, heterogeneous interfaces at microstructure and chemical composition
level are formed, and post-processing heat treatments are necessary [30]. Diffusion bond-
ing can obtain sound joints between dissimilar materials, mainly when interlayers are
applied to join the faying diffusion surfaces. In this case, cracks due to residual stress can
be prevented [10,17,31–38]. Active elements have been used as interlayers, e.g., titanium,
niobium, and zirconium [15,16]. Titanium is the most used interlayer; for instance, it reacts
with Al2 O3 by diffusion bonding resulting in the formation of TiO, TiAl, and Ti3 Al [8]. The
use of interlayers in diffusion bonding can be one option to reduce the bonding tempera-
ture, pressure and time and make the process more industrially attractive [39]. This may
be even more important in metal-ceramic joining, when the joint is subjected to thermal
cycling or thermal shock, large stress concentrations may be introduced in parts of the
ceramic.
Researchers have been working to improve the joining between dissimilar materials
by diffusion bonding. Barrena et al. [40] investigated the joining of Ti6Al4V to Al2 O3 by
diffusion bonding using as interlayer an Ag-Cu alloy (72 wt.% Ag, 28 wt.% Cu) with a
thickness of 60 µm produced by electron beam evaporation. The interface’s microstructure
was composed of different Ti-Cu phases and (Ag), and the shear tests showed a maximum
of 140 MPa obtained at a temperature of 750 ◦ C and a dwell time of 30 min. Yu et al. [10]
joined Ti6Al4V to Ti2 SnC using Cu foil (50 µm) as an interlayer in Ar atmosphere performed
at a temperature of 750 ◦ C during 60 min; a relatively low temperature was allowed due to
the destabilization of Ti2 SnC that occurs below 700 ◦ C. The interface’s microstructure is
formed mainly by Ti3 Cu4 and TiCu4 , and the maximum value obtained at the shear tests
was around 86 MPa [10]. Kliauga and Ferrante [41] joined Al2 O3 to AISI304 stainless steel
using a commercial Ti sheet with a thickness of 500 µm by diffusion bonding processed
at 800 ◦ C during 180 min and at 900 and 1000 ◦ C during 120 min under a pressure of
15 MPa in vacuum. The authors investigated the interfacial reaction products of Ti/Al2 O3 .
The microstructure characterization of the surface fracture showed that the composition
depends on the temperature and dwell time. The joints produced at a temperature of
800 ◦ C for 180 min exhibit the Ti3 Al phase at the interface close to the titanium base
material, while TiAl is observed on the ceramic side. However, for joints obtained at 900
and 1000 ◦ C during 120 min, the interface is constituted only by Ti3 Al. In similar work,
Travessa and Ferrante [42] diffusion bonded the same base materials and interlayer at
700 ◦ C for 120 min, and 800 ◦ C for 15 and 120 min. The experiments were unsuccessful,
and no reaction could be seen at the interface. Rocha et al. [43] joined pure Ti to Al2 O3 by
diffusion bonding at 800 ◦ C for 90 min under a pressure of 5 MPa. The characterization
of the Ti/Al2 O3 interface identified the presence of TiAl adjacent to alumina and Ti3 Al
adjacent to pure Ti. In addition, the authors applied galvanic corrosion tests to study the
chemical degradation of the interfaces, and the results showed that the corrosion resistance
was impaired by the TiAl phase. During the joining of Ti to Al2 O3 , TiAl intermetallic phase
always comes up in zones adjacent to alumina, and the TiAl/Ti3 Al ratio can increase with
the thickness of the Ti interlayer [44].
Although some works referred to low temperature bonding (750–800 ◦ C), the devel-
opment of new approaches should be investigated to reduce the formation of (Ag) at the
interface, as is the case when Ag-based alloys are used as interlayers. The presence of (Ag)
is detrimental to the service temperature and should be avoided.
The objective of the present work consists in the study of the feasibility of joining
Ti6Al4V to Al2 O3 by diffusion bonding using titanium as interlayer material. Different
titanium interlayers were investigated: Ti thin films deposited onto Al2 O3 base material
by magnetron sputtering and freestanding Ti thin foils. The joints processed using Ti foils
Metals 2021, 11, 1728 3 of 17

have a lower processing cost and are more straightforward than those using thin films
obtained by magnetron sputtering. However, the lack of adhesion between the Ti foils and
ceramics might make it difficult to get sound joints. The Ti interlayer was selected to allow
an interface with a chemical composition similar to one of the base materials, avoiding the
formation of phases that will impair the service temperature and with characteristics that
promote the diffusion, enabling a decrease in the diffusion bonding processing conditions.
Joining without interlayer was also conducted using the same parameters to evaluate the
potential of these interlayers. The microstructural characterization of the joints’ interface
was carried out by optical microscopy (OM), scanning electron microscopy (SEM), energy
dispersive X-ray spectroscopy (EDS), and electron backscatter diffraction (EBSD), while
the mechanical characterization was performed by nanoindentation tests across the joints’
interface and shear strength tests.

2. Materials and Methods

2.1. Base Materials
Ti6Al4V alloy and polycrystalline Al2 O3 (purity of 99%) were purchased from Good-
fellow in rods with 7 and 6 mm diameters, respectively. They were cut 5 mm in length,
ground, and polished down to 1 µm diamond suspension and 0.03 µm silica using standard
metallographic procedure, then cleaned with deionized water, acetone, and ethanol in an
ultrasonic bath and dried with heat blow air. The results of the polishing were assessed by
optical microscopy (OM) (DM4000, Leica Microsystems, Wetzlar, Germany) and average
roughness (Ra) of the surfaces was measured by profilometry (Perthometer SP4, with laser
probe (Mahr Perthometer SP4, Göttingem, Germany)).

2.2. Titanium Interlayer

The titanium thin films were deposited onto the polished surfaces of alumina
(substrate) by direct current magnetron sputtering using a Ti (99.99% pure) target
(150 mm × 150 mm × 6 mm thick). After achieving a base pressure below 5 × 10−4 Pa in
the sputtering chamber, Ar was introduced (P ~ 1.5 × 10−1 Pa). The substrate materials
were cleaned by heating followed by Ar+ (current of 20 A) etching using an ion gun.
To avoid residual impurities from the substrates, assuring a good adhesion between the
substrate and the Ti film, the total etching time employed was increased to 12–10 min
higher than the usual conditions. The deposition starts once the cleaning of the substrate
is concluded, after introducing more argon into the sputtering chamber (4.0 × 10−1 Pa
deposition pressure). The power density applied to the Ti target was 6.70 × 10−2 W·mm−2 .
The Ti films were produced using a substrate rotation speed of ~23 rpm and a deposition
time of 20 min to achieve a thickness of ~1.0 µm. The titanium foil was purchased from
Goodfellow with a purity of 99.6%, dimensions of 25 mm × 25 mm, and thickness of
5 µm. For the joining experiments, the foil was cut into rectangular parts with sizes of
7 mm × 7 mm.

2.3. Adhesion
The adhesion strength between the Ti thin films and Al2 O3 was measured by a pull-off
test using an apparatus as referred to in [45,46]; however, the substrate has a surface area
three times higher than in the literature. The test consists of gluing the film deposited onto
the alumina substrate to a rigid rod, following the curing time of the glue. Then, the set
was fixed by the grips of a tensile test machine. The tensile tests were carried out under
environmental conditions using a load cell of 500 N and a loading speed of 10 µm/min.
The adhesion strength was estimated for three specimens to obtain the average value.

2.4. Diffusion Bonding

Ti6Al4V and Al2 O3 joining was performed in a tubular horizontal furnace (Termolab
Electrical Furnace, Agueda, Portugal) under a vacuum level of 10−2 Pa. Figure 1 displays
the device used to assure contact pressure of the faying surfaces. The dissimilar base
Metals 2021, 11, x FOR PEER REVIEW

Metals 2021, 11, 1728 4 of 17

the device used to assure contact pressure of the faying surfaces. The d
materials were joined with and without Ti interlayer, and diffusion bondi
materials were 950,
out at 900, joinedand
with 1000
and without Ti interlayer,
°C, during 10 andand 60
diffusion
min. bonding was carried
The heating and coolin
out at 900, 950, and 1000 ◦ C, during 10 and 60 min. The heating and cooling rates were 5
and
and 3 ◦3C/
°C/
min,min, respectively.
respectively.

Figure
Figure 1. 1. Device
Device usedused to apply
to apply the pressure
the contact contactatpressure
the fayingat the faying
surfaces duringsurfaces during the
the diffusion
bonding experiments.
ing experiments.
2.5. Microstructural Characterization
2.5.Optical
Microstructural Characterization
microscopy (OM) was used to overview the soundness of the joints’ inter-
face. A DM 4000 M optical microscope equipped with a Leica DFC 420 camera (Leica
Optical
Microsystems, microscopy
Wetzlar, Germany)(OM) was
was used to used
achieveto overview
this objective. the soundness of the jo
A For
DMthe4000 M optical
microstructural and microscope equippedcross-sections
chemical characterization, with a Leica DFC 420
of the joints
were prepared using Wetzlar,
Microsystems, standard metallographic
Germany) techniques.
was usedAfterward,
to achieve the joints’ interfaces
this objective.
were analyzed by scanning electron microscopy (SEM) (FEI Quanta 400FEG ESEM/EDAX
GenesisFor
X4Mthe
(FEI microstructural andUSA))
Company, Hillsboro, OR, chemical
operatingcharacterization, cross-section
at an accelerating voltage of
15were prepared
keV, coupled withusing
energy standard metallographic
dispersive X-ray techniques.
spectroscopy (EDS) Afterward,
(Oxford Instrument, Ox- the jo
fordshire, UK) by theby
were analyzed standardless
scanning quantification method. Electron
electron microscopy backscatter
(SEM) (FEI diffraction
Quanta 400FEG
(EBSD) analyses were conducted using an acceleration voltage of 15 keV to obtain Kikuchi
Genesis
patterns X4M
using (FEI TSL-EDAX
a detector Company, Hillsboro,
EBSD Unit (EDAX OR,
Inc.USA))
(Ametek),operating
Mahwah, NJ, at USA),
an accelera
15 keV,
allowing thecoupled with energy
phases in localized zones ofdispersive X-ray
the interfaces to spectroscopy
be identified. (EDS)
The indexation of (Oxfo
the patterns was made
Oxfordshire, UK) by ICDD PDF2standardless
by the (2006) database. quantification method. Electro

diffraction
2.6. (EBSD) analyses
Mechanical Characterization were conducted using an acceleration voltag
obtain Kikuchi
Hardness patterns
and reduced using
Young’s a detector
modulus TSL-EDAX
were evaluated across the EBSD
joints byUnit (EDAX
nanoin-
dentation. The experiments were carried out in a computer-controlled
Mahwah, NJ, USA), allowing the phases in localized zones of the in apparatus equipped
with a Berkovich diamond indenter (NanoTest, Micro Materials Limited, Wrexham, UK).
identified. The indexation of the patterns was made by ICDD PDF2 (2006)
A maximum load of 5 mN was selected, and 8 × 12 matrices were defined, starting on
the Ti6Al4V side, crossing the joints’ interface, and moving towards the alumina base
2.6. Mechanical
material. Characterization
The distances between rows and columns were 5 and 3 µm, respectively. Load-
ing/unloading was carried out in 30 s, with 30 s at maximum load and at 10% of maximum
Hardness
load during unloadingand reduced
for thermal Young’sHardness
drift correction. modulus were Young’s
and reduced evaluatedmodu-across
nanoindentation.
lus were determined by theTheOliver
experiments were[47].
and Pharr method carried out in a computer-contro
Mechanical characterization was also performed by shear tests; three specimens were
equipped with a Berkovich diamond indenter (NanoTest, Micro Materials
tested at room temperature at a rate of 1 mm/min for each joint. Figure 2 shows the shear
ham, UK).
strength A maximum
apparatus load of 5 and
used in this investigation mNa was selected,
schematic andThe
illustration. 8 ×fracture
12 matrices
starting
surface of theon thesubmitted
joints Ti6Al4V side,
to shear crossing
strength theevaluated
tests was joints’byinterface, and movin
digital microscope
DVM6 (Leica Microsystems, Wetzlar, Germany).
alumina base material. The distances between rows and columns were
respectively. Loading/unloading was carried out in 30 s, with 30 s at maxi
at 10% of maximum load during unloading for thermal drift correction.
reduced Young’s modulus were determined by the Oliver and Pharr meth
Mechanical characterization was also performed by shear tests; three s
tested at room temperature at a rate of 1 mm/min for each joint. Figure 2 s
strength apparatus used in this investigation and a schematic illustratio
surface of the joints submitted to shear strength tests was evaluated by dig
DVM6 (Leica Microsystems, Wetzlar, Germany).
Metals2021,
Metals 2021,11,
11,1728
x FOR PEER REVIEW 55ofof17
17
Metals 2021, 11, x FOR PEER REVIEW 5 of 17

Al2O3
Al2O3

Ti6Al4V
Ti6Al4V

Figure 2. Shear strength test device and respective schematic illustration.

Figure2.2.Shear
Figure Shearstrength
strengthtest
testdevice
deviceand
andrespective
respectiveschematic
schematicillustration.
illustration.
3. Results and Discussion
3.3.Results
ResultsandandDiscussion
Discussion
3.1.Ti
3.1. TiThin
ThinFilms
FilmsDeposited
Deposited ontoAl Al2O3
3.1. Ti Thin Films Depositedonto onto Al2 2OO33
The adhesion
The adhesion of the Ti
Ti films
films to Al22O
toAl O33 3plays
playsaa vital role in in
the joints processed by dif-
The adhesion of of the
the Ti films to Al 2O plays avital
vital role
role in thethe joints
joints processed
processed by
by dif-
fusion
diffusionbonding,
bonding, andand is affected
is affected by the
by surface
the surface topography
topography of the
of Al
the2O3 substrate. Profilom-
Al O substrate. Pro-
fusion bonding, and is affected by the surface topography of the Al2O3 substrate. Profilom- 2 3
etry measured
filometry measured the substrate’s
the substrate’s surface roughness,
surface and the
roughness, and results showed
the results Ra values
showed of 0.10
Ra values
etry measured the substrate’s surface roughness, and the results showed Ra values of 0.10
µm.
of Figure
0.10 3
µm. Figureshows the
3 shows cross-section of
the cross-section the titanium thin
of the titanium film
thinas deposited onto the sur-
µm. Figure 3 shows the cross-section of the titanium thin film as film as deposited
deposited onto theontosur-
facesurface
the of Al2Oof3. Al
The Ti
2 OTi
thin
The film
3 . thin Ti is highlighted
thin film is highlighted in lightin grey,
lightand its and
grey, total thickness is closeisto
face of Al2O 3. The film is highlighted in light grey, and its totalits total thickness
thickness is close to
1 µm,toas1 desired.
close The adhesion
µm, as desired. pull-off test
The adhesion shows testan average an adhesion strength of 8 MPa,
1 µm, as desired. The adhesion pull-off pull-off
test shows anshows
average average adhesion
adhesion strength strength
of 8 MPa,
slightly
of below
8 MPa,below
slightlythe 10 MPa referred to in the literature [48]. However, the thin films
the were
slightly thebelow
10 MPa thereferred
10 MPatoreferred to in the literature
in the literature [48]. However, [48]. However,
the thin films thin
were
not
films detached
were not from the
detached alumina
from substrate
the alumina at the end
substrate of
atthe
the experiments;
end of the only the
experiments; interface
only
not detached from the alumina substrate at the end of the experiments; only the interface
glue/film
the was
interface broken, was
glue/film which suggests
broken, which that the adhesion
suggests that thestrength
adhesion is strength
sufficient. is Therefore,
sufficient.
glue/film was broken, which suggests that the adhesion strength is sufficient. Therefore,
this value this
Therefore, seems enough
value seems forenough
the envisaged application.
for the envisaged application.
this value seems enough for the envisaged application.

Figure3.
3. SEM image
imageof
ofthe
thecross-section ofof
as-deposited Ti thin filmfilm
withwith
totaltotal
thickness closeclose
to 1 µm.
Figure 3. SEM
Figure SEM image of the cross-section
cross-section as-deposited
of as-deposited Ti thin
Ti thin film with thickness
total thickness to
close to 1 µm.
1 µm.
Metals 2021, 11, 1728 6 of 17

Metals 2021, 11, x FOR PEER REVIEW 6o

3.2. Diffusion Bonding without Interlayer

The joining3.2. Diffusion
between Bondingand
Ti6Al4V without
Al2 OInterlayer
3 without interlayer was only successful by
diffusion bonding atThe 1000 ◦ C with dwellTi6Al4V
times ofand 10 Aland
joining between 2O360 min with
without contact
interlayer waspressure.
only successful
Figure 4 shows diffusion
the cross-section
bonding at of 1000
the joints
°C withinterface.
dwell times Someof 10defects
and 60are observed
min in pressu
with contact
zones adjacent to alumina.
Figure 4 shows Thisthecan be explained
cross-section of theby theinterface.
joints following factors:
Some defectsthearelack of
observed in zon
adjacent to
base materials surfaces’ alumina.
contact, theThis can be explained
formation of brittleby the following
phases, or the factors: the lackof
development of base ma
rials surfaces’
residual stress, which promotes contact,
cracksthe formation
during of brittle
the cooling phases,
of the joints.orThe
the same
development
defects of resid
stress, which promotes cracks during the cooling of
were observed in joints obtained for both dwell times, 10 and 60 min. Figure 4b shows the joints. The same defects were
the interface in high magnification, and EDS analyzed zones highlighted in red. Figure 5the interf
served in joints obtained for both dwell times, 10 and 60 min. Figure 4b shows
shows the SEM in high of
image magnification,
the interfaceand EDS analyzed
produced for 10 zones
min and highlighted
the Ti, Al,in red.
and Figure
O EDS5 shows
SEM image of the interface produced for 10 min
elemental distribution mapping. Despite the short bonding time, the EDS results showand the Ti, Al, and O EDS elemen
distribution mapping. Despite the short bonding time,
the occurrence diffusion of Al from Al2 O3 and Ti and Al from Ti6Al4V to the interface. the EDS results show the occ
rence diffusion of Al from Al 2O3 and Ti and Al from Ti6Al4V to the interface. Zone Z1
Zone Z1 in the vicinity of Ti6Al4V has contents of 78.9% Ti, 19.6% Al, and 1.5% V (at. %),
combining the EDS the vicinity
results of withTi6Al4V hasphase
ternary contents of 78.9%
diagram Ti, 19.6%
[49], suggestsAl, and 1.5% phase
the α-Ti V (at. %),
is combin
the EDS results with ternary phase diagram [49], suggests the α-Ti phase is predomina
predominant. Zone Z2 comprises 73.3% Ti, 25.3% Al, and 1.4% V (at. %), and besides the
Zone Z2 comprises 73.3% Ti, 25.3% Al, and 1.4% V (at. %), and besides the α-Ti phase,
α-Ti phase, the α2 -Ti3 Al phase may have also been formed. Some round and elongated
α2-Ti3Al phase may have also been formed. Some round and elongated particles n
particles near Al2 O3 (Z3) have a chemical composition of 61.9% Ti, 36.6% Al, and 1.5% V
Al2O3 (Z3) have a chemical composition of 61.9% Ti, 36.6% Al, and 1.5% V and can
and can be identified as γ-TiAl and α2 -Ti3 Al phases.
identified as γ-TiAl and α2-Ti3Al phases.

(a)

(b)
Figure 4. SEM4.images
Figure of the joints
SEM images without
of the joints interlayer processed processed
without interlayer at (a) 1000 °C
at for
(a) 60 min◦ C
1000 and
for(b)
601000
min°C for(b)
and 10 min.
1000 ◦ C for 10 min.
 Metals 2021, 11, 1728 7 of 17
Metals 2021, 11, x FOR PEER REVIEW 7 of 17

(a) (b)

(c) (d)
Figure 5. (a)
Figure SEM
5. (a) image
SEM imageand EDS
and EDSelemental maps
elemental maps of of
(b)(b)
Ti,Ti,
(c)(c)
Al,Al,
and (d)(d)
and O of thethe
O of interface obtained
interface obtained
without interlayer
without at at
interlayer 1000 °C◦and
1000 dwell
C and time
dwell of of
time 1010min.
min.

3.3.
3.3. Diffusion
Diffusion Bonding
Bonding Using
Using TiTi Thin
Thin Films
Films Deposited
Deposited onto
onto AlAl
2O23O3

TheThe joining
joining ofof dissimilar
dissimilar materials
materials using
using interlayers
interlayers onon faying
faying surfaces
surfaces ofof thebase
the base
materials is an approach to reduce the residual stresses in the diffusion
materials is an approach to reduce the residual stresses in the diffusion bonding process bonding process
andimprove
and improve thethe contact
contactbetween
betweenthethe faying surfaces
faying [8,29,50–54].
surfaces Joining
[8,29,50–54]. between
Joining Ti6Al4V
between
and Al2and
Ti6Al4V O3 using
Al2O3Tiusing
thin films deposited
Ti thin onto Al2onto
films deposited O3 was Al2performed at temperatures
O3 was performed of 900,
at tempera-
950,ofand ◦ C and dwell times of 10 and 60 min. Unsuccessful joints were obtained
tures 900,1000
950, and 1000 °C and dwell times of 10 and 60 minutes. Unsuccessful joints
at 900 ◦ C, for both bonding times, and joints processed at 950 ◦ C for 10 min. The base
were obtained at 900 °C, for both bonding times, and joints processed at 950 °C for 10 min.
materials
The were separated
base materials after cooling,
were separated proving that
after cooling, the diffusion
proving bonding conditions
that the diffusion bonding con- were
insufficient to promote the bonding of the dissimilar materials.
ditions were insufficient to promote the bonding of the dissimilar materials. SuccessfulSuccessful joints were
obtained
joints were at 950 ◦ C at
obtained for950
60 °C
minforand
60 1000
min and
◦ C for 10 and 60 min with Ti thin film interlayer.
1000 °C for 10 and 60 min with Ti thin film
Figure 6 shows
interlayer. Figure SEM
6 showsimages
SEMofimages
the interfaces of the joints
of the interfaces processed
of the under these
joints processed conditions.
under these
conditions. Table 1 shows the EDS results of the zones marked on the SEM images of6.Fig-
Table 1 shows the EDS results of the zones marked on the SEM images of Figure The
identification of the possible phase(s) was performed by combining
ure 6. The identification of the possible phase(s) was performed by combining the EDS the EDS results with
the Ti-Al-V
results with the phase diagram
Ti-Al-V phase[49].
diagram [49].
Figure 6a shows the joint processed at 950 ◦ C and 60 min that exhibits an interface
Figure 6a shows the joint processed at 950 °C and 60 min that exhibits an interface
with a total thickness of ~15 µm, and the bond line is not visible. The interface can be
with a total thickness of ∼15 µm, and the bond line is not visible. The interface can be
divided into two distinct layers. The layer close to the Ti6Al4V (Z2) exhibits a chemical
divided into two distinct layers. The layer close to the Ti6Al4V (Z2) exhibits a chemical
composition similar to the Ti base material (Z1), although slightly higher V content. Ti6Al4V
composition similar to the Ti base material (Z1), although slightly higher V content.
is constituted by α-Ti phase (Z1) matrix with a β-Ti phase (Z4) rich in V at the grain
Ti6Al4V is constituted by α-Ti phase (Z1) matrix with a β-Ti phase (Z4) rich in V at the
boundaries. Combining the EDS results with the Ti-Al-V phase diagram, a layer of α-Ti
grain boundaries. Combining the EDS results with the Ti-Al-V phase diagram, a layer of
phase can be identified. The layer adjacent to Al2 O3 (Z3) has twice the Al content (22.5 at%)
α-Ti phase can be identified. The layer adjacent to Al2O3 (Z3) has twice the Al content (22.5
than that found in the layer adjacent to Ti6Al4V (10.2 at.% Al). The EDS results combined
at%) than that found in the layer adjacent to Ti6Al4V (10.2 at.% Al). The EDS results com-
with the phase diagram suggest that this layer is mainly composed of the α2 -Ti3 Al phase.
bined with the phase diagram suggest that this layer is mainly composed of the α2-Ti3Al
Close to Al2 O3 , darker particles were observed that could be identified as γ-TiAl (Z5).
phase. Close to Al2O3, darker particles were observed that could be identified as γ-TiAl
Figure 6b shows the joint interface obtained at 1000 ◦ C using a bonding time of 10 min.
(Z5).
The interface has a thickness of ~13 µm, and two different layers are also observed. Despite
theFigure 6b showstime,
short bonding the joint interface temperature
the bonding obtained at 1000 °C using
promoted theaformation
bonding timeof a of 10
sound
min. The interface has a thickness of ∼13 µm, and two different layers are also observed.
Metals 2021, 11, x FOR PEER REVIEW 8 of 17
Metals 2021, 11, 1728 8 of 17

Despite the short bonding time, the bonding temperature promoted the formation of a
sound interface
interface becausebecause the diffusion
the diffusion coefficient
coefficient exponentially
exponentially increases
increases with temperature
with temperature [55].
[55]. Identical
Identical to the interface
to the interface produced at 950 ◦ C,
produced at the
950layer
°C, the layertoadjacent
adjacent Ti6Al4Vto(Z3) Ti6Al4V (Z3) is
is composed
composed
of α-Ti withofaα-Ti with aof
thickness thickness
~5 µm. of ∼5 layer
The µm. The layer adjacent
adjacent to Al2 O3to Al2Owith
(Z4), 3 (Z4), with a thick-
a thickness of
~8
ness
µm, ∼8 µm,
of has contents of 75.0 at.%
has contents Ti at.%
of 75.0 and Ti25.0
andat.%
25.0Alat.%
suggesting that the
Al suggesting thatTi3the
Al Ti
phase was
3Al phase

formed
was formedduring the diffusion
during bonding
the diffusion bondingprocess. In-between
process. In-between thethe
AlAl
2 O23Oand
3 andthis
thisinterface
interface
layer,
layer,ititisispossible
possibleto
toobserve
observeaathin
thinlayer
layer(~1 µm)composed
(∼1µm) composedof ofalternated
alternateddifferent
differentgrey
grey
columnar
columnarsmall smallgrains.
grains.However,
However,it itwaswasnot
notpossible
possible to to
perform
performEDSEDS analyses
analyses owing to its
owing to
reduced
its reducedthickness.
thickness.

(a) (b)

(c)
Figure6.6.SEM
Figure SEMimages
imagesofofthe
thejoints
joints with
with Ti Ti thin
thin film
film interlayer
interlayer processed
processed at 950
at (a) ◦ C for
(a) 950 °C 60
formin,
60 min, ◦ C for
(b) 1000
(b) 1000 °C10for 10 min,
min, and
and (c) 1000
◦ °C for
(c) 1000 C for 60 min.60 min.

Increasing the bonding time at 1000 ◦ C to 60 min promotes the change in the thickness
Increasing the bonding time at 1000 °C to 60 min promotes the change in the thick-
of the layers that compose the interface. This interface can be observed in Figure 6c. The
ness of the layers that compose the interface. This interface can be observed in Figure 6c.
increase in the bonding time results in the decrease in the α-Ti phase and the growth of the
The increase in the bonding time results in the decrease in the α-Ti phase and the growth
Ti3 Al layer with TiAl particles close to Al2 O3 (Z4 in Figure 6c).
of the Ti3Al layer with TiAl particles close to Al2O3 (Z4 in Figure 6c).
The confirmation of the phases at the interfaces was conducted by EBSD, which
The confirmation of the phases at the interfaces was conducted by EBSD, which al-
allows Kikuchi patterns of tiny zones to be obtained due to the reduced interaction volume.
lows Kikuchi patterns of tiny zones to be obtained due to the reduced interaction volume.
Figure 7 shows EBSD Kikuchi patterns of the joint processed at 1000 ◦ C for 60 min. This
Figure 7 shows EBSD Kikuchi patterns of the joint processed at 1000 °C for 60 min. This
technique was of paramount importance to confirm the presence of the γ-TiAl intermetallic
phase close to the Al2 O3 base material.
 technique was of paramount importance to confirm the presence of the γ-TiAl intermetal-
lic phase close to the Al2O3 base material.
Metals 2021, 11, 1728 9 of 17
Table 1. Chemical composition obtained by EDS of the zones in Figure 6.

Element (at.%)
Conditions Zone Possible Phases
Tiby EDS of
Table 1. Chemical composition obtained Althe zonesVin Figure 6.
1 88.5 10.3 1.2
Element (at.%)
α-Ti
Conditions Zone2 88.3 10.3 1.5 Possible
α-Ti Phases
Ti Al V
950 °C/60 min 3 76.5 22.5 1.0 α2-Ti3Al
1 88.5 10.3 1.2 α-Ti
24 80.3
88.3 6.7 10.3 13.0 1.5 β-Ti
α-Ti
950 ◦ C/60 min 35 65.1
76.5 33.722.5 1.2 1.0 α2-Ti3Alα2+-Tiγ-TiAl
3 Al
4 80.3 6.7 13.0 β-Ti
1 86.8 11.6 1.6 α-Ti
5 65.1 33.7 1.2 α2 -Ti3 Al + γ-TiAl
2 73.3 19.7 7.0 β-Ti
1000 °C/10 min 1 86.8 11.6 1.6 α-Ti
◦ 23 86.0
73.3 11.919.7 2.1 7.0 α-Ti
β-Ti
1000 C/10 min
34 75.0
86.0 25.011.9 * 2.1 α2-Tiα-Ti
3Al
4 75.0 25.0 * α2 -Ti3 Al
1 88.3 11.7 * α-Ti
12 88.3
77.1 6.9 11.7 16.0 * α-Ti
β-Ti
1000 °C/60
◦ min
1000 C/60 min
2 77.1 6.9 16.0 β-Ti
33 74.5
74.5 25.525.5 * * α2-Ti 3Al
α2 -Ti3 Al
44 57.6
57.6 42.442.4 * * α2-Ti Al3 Al
α23-Ti + γ-TiAl
+ γ-TiAl
*—not
*—not detected.
detected.

(b)

2
2
1

(a) (c)
Figure 7. (a)Figure
SEM 7.images
(a) SEM of images
the interface produced
of the interface with Tiwith
produced thinTifilm
thinprocessed at 1000
film processed at 1000 ◦ C 60
°C for formin, (b)(b)
60 min, EBSD Kikuchi
EBSD Kikuchipatterns of
the grain marked
patternsasof1 the
in (a) indexed
grain markedasasγ-TiAl,
1 in (a)and (c) EBSD
indexed Kikuchi
as γ-TiAl, and patterns
(c) EBSD of the grain
Kikuchi marked
patterns of theasgrain
2 in marked
(a) indexed
as 2 in α2-Ti3Al.
as (a)
indexed as α2 -Ti3 Al.
The nanoindentation experiments were performed across the joints interface and
adjacentThe
basenanoindentation experiments were performed across the joints interface and
materials. Hardness and reduced Young’s modulus (Er) maps were obtained
adjacent base materials. Hardness and reduced Young’s modulus (Er ) maps were obtained
to understand their distribution across the joints interface (Figure 8). The reduced Young’s
to understand their distribution across the joints interface (Figure 8). The reduced Young’s
modulus
modulusmapmapis is
only
onlyshown
shownfor
for the joint interface
the joint interfaceprocessed
processed at 1000
at 1000 °C 60
◦ C for formin.
60 min.
The The
different hardness values allow the base materials to be identified, as well as the interface.
different hardness values allow the base materials to be identified, as well as the interface.
As As
expected, the
expected, theAlAl
2O
2O3 3displays
displays the highesthardness
the highest hardnessvalue,
value, around
around 30–50
30–50 GPa.GPa.
The The
interface hardness is similar to that of titanium alloy but increases slightly near the Al2 O3
Metals 2021, 11, x FOR PEER REVIEW 10 of 17

Metals 2021, 11, 1728 10 of 17

interface hardness is similar to that of titanium alloy but increases slightly near the Al2O3
base material. The increase in temperature and time of diffusion bonding promotes an
base material. The increase in temperature and time of diffusion bonding promotes an
increase in the hardness of the interface that is in accordance with the microstructural
increase in the hardness of the interface that is in accordance with the microstructural
characterization. The increase in the diffusion bonding temperature or time induces the
characterization. The increase in the diffusion bonding temperature or time induces
increase in the thickness of the
the increase in the thickness α2-Ti
of the Al3 Al
α23-Ti andand
α2-Ti 3Al + γ-TiAl layers, which can be associ-
α2 -Ti 3 Al + γ-TiAl layers, which can be
ated with the
associated increase
with in theinhardness.
the increase the hardness.

(a) (b)

(c) (d)
Figure 8. Figure
(a–c) Hardness maps ofmaps
8. (a–c) Hardness the joints
of theprocessed at 950at°C
joints processed for
950 ◦ C60 min,
for 10001000
60 min, °C for
◦ C 10
for min, and
10 min, 1000
and °C◦ C
1000 forfor6060min,
min,respectively,
and (d) reduced Young’s
respectively, modulus
and (d) reducedmap across
Young’s the joint
modulus mapprocessed
across the at 1000
joint °C for 60
processed min.◦ C for 60 min.
at 1000

Regarding the
Regarding the reduced
reducedYoung’s
Young’s modulus,
modulus,the the
mapmap
(Figure 8d) allows
(Figure the different
8d) allows the different
regions to be distinguished from the lower values (Ti6Al4V) towards,
regions to be distinguished from the lower values (Ti6Al4V) towards, the higher the higher values values
(Al
(Al 2O2O 3 ),passing
3),
passing by
by the
the two
twointerface
interfacelayers. TheThe
layers. thicker layerlayer
thicker corresponding
correspondingto α2 -Tito
3 Al
α2-Ti3Al
has a slightly higher modulus than the Ti alloy base material, while the thinner layer
has a slightly higher modulus than the Ti alloy base material, while the thinner layer (α2-
(α2 -Ti3 Al + γ-TiAl) has an even higher modulus (Figure 8d). The Al2 O3 base material has
Tithe
3Al + γ-TiAl) has an even higher modulus (Figure 8d). The Al2O3 base material has the
highest reduced Young’s modulus.
highest The reduced Young´s
shear strength ofmodulus.
the joints was evaluated by shear tests. Table 2 displays the
Theofshear
values strength
the shear strengthofofthe
thejoints was evaluated
joints processed ◦ C shear
at 1000 by tests.with
for 60 min, Tableand 2without
displays the
values of the shear strength of the joints processed at 1000 °C for 60 min, with and without
interlayers, and at 950 °C for 60 min with an interlayer (Ti thin film). At 950 °C, only two
values are presented because, for the third sample, during the test, the sample was frac-
tured but fitted in the support, and the computer continued to record the force—this test
Metals 2021, 11, 1728 11 of 17

interlayers, and at 950 ◦ C for 60 min with an interlayer (Ti thin film). At 950 ◦ C, only
two values are presented because, for the third sample, during the test, the sample was
fractured but fitted in the support, and the computer continued to record the force—this
test being considered invalid. The average shear strength values are very similar for the
samples processed at 1000 ◦ C for 60 min (52 ± 14 MPa for the joint processed without
interlayer and 49 ± 25 MPa for the joint processed with interlayer). In fact, the use of the Ti
interlayer allows the decrease in the bonding conditions without impairing the mechanical
properties. This can be explained since the microstructure of the interface is very similar,
and the interface is composed of the same phases. The Ti thin films improve the adhesion
to the ceramic base material and, due to the fine microstructure, enhance the diffusion
across the interface during the joining process. These results confirm that these interlayers
improve the bonding process of these dissimilar materials.
Figure 9 presents the fracture surface of the samples with higher and lower shear
strength values. It can be observed that when the fracture occurs at the ceramic base
material, the shear strength is higher. The fracture occurs at the Al2 O3 and then propagates
to the interface (Figure 9a). On the contrary, when the fracture occurs at the interface
and propagates through the interface, the shear strength value is lower, (Figure 9b). This
happen when the joint presents some defects at the interface, such as pores in the layer
close to the Al2 O3 , that promote the nucleation of the fissure at the interface.

Table 2. Shear strength values of the joints produced without interlayer at 1000 ◦ C for 60 min and
with Ti thin film at 1000 and 950 ◦ C for 60 min.

Joint Processing Conditions Interlayers Shear Strength (MPa)

66
1000 ◦ C for 60 min without 54
38
77
1000 ◦ C for 60 min Ti thin film 41
27
43 1
950 ◦ C for 60 min Ti thin film
30
1 —only two values are present since one test was considered invalid.

Based on the reaction between the Ti thin film and the base materials observed at
different diffusion bonding temperatures and times, a common mechanism can be pointed
out to explain the formation of the interface microstructure during the joining of Al2 O3
to Ti6Al4V alloy. Figure 10 shows a schematic illustration of a possible mechanism for
the formation of the interfaces, as well as the microstructure of the interface obtained at
different temperatures and times. During the bonding process, due to the temperature
increase, the Ti of the film starts to diffuse from the film towards the base materials, while
the Ti and Al of the Ti alloy and the Al of the ceramic diffuse towards the interface. The
Ti film at the interface starts to grow significantly, and this is also due to the fine grain
microstructure of the film because it is produced by magnetron sputtering, which enhances
diffusion. The Al diffusing from the Al2 O3 will combine with the Ti at the interface and
form a layer of α2 -Ti3 Al. With increasing diffusion due to higher temperature and time,
the α2 -Ti3 Al layer thickness will increase. At high temperatures and long times, diffusion
will result in a thicker α2 -Ti3 Al layer and the formation of alternating α2 -Ti3 Al and γ-
TiAl grains along the Al2 O3 base material. During cooling, different microstructures will
form, which are represented in Figure 10d–f. The thickness of the layers that compose the
interfaces depend on the temperature and time of diffusion bonding, since this process is
ensured by the diffusion of the elements that react and form the different reaction layers.
Metals 2021, 11, 1728 12 of 17
Metals 2021, 11, x FOR PEER REVIEW 12 of 17

Higher shear strength values Lower shear strength values

Interface

Interface

Al2O3

(a) (b)
Interface
Interface

Al2O3
(c) (d)

(e) (f)

(g) (h)
Figure 9.
Figure 9. Fracture
Fracture surface
surface of
of the
the joints
joints with
with (a,c,e,g)
(a,c,e,g) higher
higher and
and (b,d,f,h)
(b,d,f,h) lower
lower shear
shear strength
strength values:
values:
(a) and (b) low magnification of the Al2O3 samples, (c) and (d) high magnification of the regions
(a,b) low magnification of the Al2 O3 samples, (c,d) high magnification of the regions marked in (a,b),
marked in (a) and (b), respectively, (e,f,g,h) 3D representation of regions (c) and (d).
respectively, (e–h) 3D representation of regions (c,d).
 Metals 2021, 11, 1728 13 of 17
Metals 2021, 11, x FOR PEER REVIEW 13 of 17

(a)

(b)

(c)

(d) (e) (f)

Figure 10. Microstructure evolution during the diffusion bonding of Ti6Al4V to Al2O3 using Ti thin films: (a) initial microstructure,
Figure 10. Microstructure evolution during the diffusion bonding of Ti6Al4V to Al2 O3 using Ti thin films: (a) initial
(b) sequence of the formation and growth of the α2-Ti3Al layer due to Ti and Al diffusion, (c) growth of α-Ti, formation of α2-Ti3Al
microstructure, (b) sequence of the formation and growth of the α2 -Ti3 Al layer due to Ti and Al diffusion, (c) growth of α-Ti,
layer and formation of α2-Ti3Al and γ-TiAl grains close to Al2O3 base material, (d) microstructure of the interface formed at 950 °C
formation
for 60 min, of α2 -Ti3 Al layer
(e) microstructure of theand formation
interface of αat2 -Ti
formed 3 Al°C
1000 and 10 mingrains
forγ-TiAl close
and (f) to Al2 O3 base
microstructure of material, (d) microstructure
the interface formed at 1000 °C
of the interface formed at 950 ◦ C for 60 min, (e) microstructure of the interface formed at 1000 ◦ C for 10 min and (f)
for 60 min.
microstructure of the interface formed at 1000 ◦ C for 60 min.
Metals 2021, 11, 1728 14 of 17
Metals 2021, 11, x FOR PEER REVIEW 14 of 17

3.4. Diffusion Bonding

3.4. Diffusion Bonding withwith Freestanding
Freestanding Ti Ti Thin
Thin Foils
Foils
Diffusion
Diffusion bonding
bonding experiments
experiments of of Ti6Al4V
Ti6Al4V to to Al O33 using
Al22O using55µm freestandingTi
µmfreestanding Ti foils
foils
were ◦ C for 10 and 60 min. The joining was unsucessful under these
were performed
performed at at 950
950 °C for 10 and 60 min. The joining was unsucessful under these
bonding ◦ C for
bonding conditions. Figure 11
conditions. Figure 11 shows
shows OM OM images
images of of the
the interface
interface produced
produced at at 950
950 °C for
60
60 min. Based on
min. Based on these
these images,
images, the the lack
lack of
of joint
joint between
between the the Ti
Ti foil
foil and
and the
the AlAl22O base
O33 base
material
material is is clear.
clear. InIn addition,
addition, between
between the theTi
Ti foil
foil and
and the
the Ti6Al4V
Ti6Al4V alloy,
alloy, there
there areare also
also
unbounded areas that can be observed. These results suggest that
unbounded areas that can be observed.These results suggest that the bonding conditions the bonding conditions
are
are not
notsufficient
sufficienttoto promote
promote thethe
joining between
joining betweendissimilar materials,
dissimilar revealing
materials, that having
revealing that
the Al2 Othe
having 3 base
Al2O material coated with Ti thin films is a more promising approach. The films
3 base material coated with Ti thin films is a more promising approach.
deposited
The films on Al2 O3 areoncrucial
deposited Al2O3toare guarantee
crucial the
to adhesion
guaranteebetween the Ti interlayer
the adhesion between and the the
Ti
ceramic
interlayer and the ceramic material. This adhesion is essential for the success of state
material. This adhesion is essential for the success of the ceramic/metal solid the
diffusion bonding
ceramic/metal solid process. Furthermore,
state diffusion bondingtheprocess.
fine microstructure
Furthermore, ofthe
thefine
thinmicrostructure
film deposited
by sputtering seems to be a parameter to take into account. The bond
of the thin film deposited by sputtering seems to be a parameter to take into account. between the TiThefoil
and the Ti6Al4V alloy also fails, showing that this interlayer does not present
bond between the Ti foil and the Ti6Al4V alloy also fails, showing that this interlayer does the requisites
to
notbepresent
appliedthe inrequisites
the diffusionto bebonding
appliedprocess under study.
in the diffusion bonding process under study.

Figure 11.
Figure 11. OM
OM images
images of
of the
the joint
joint processed
processed at
at 950
950 °C
◦ C for
for 60
60 min
min using
using aa Ti
Ti freestanding
freestanding thin
thin foil
foil as
as interlayer.
interlayer.

4. Conclusions
4. Conclusions
Diffusion bonding of Ti6Al4V to Al2O3 using titanium as interlayer was investigated
Diffusion bonding of Ti6Al4V to Al2 O3 using titanium as interlayer was investigated
in this work. Two different interlayers were considered to evaluate the effect of the adhe-
in this work. Two different interlayers were considered to evaluate the effect of the adhesion
sion during
during the diffusion
the diffusion bonding
bonding process.
process. Sputtered
Sputtered Ti thinTifilms
thin (1
films
µm)(1deposited
µm) deposited
onto Al onto
2 O3
Al 2O3 base material and commercial freestanding titanium foils (5 µm) were used for this
base material and commercial freestanding titanium foils (5 µm) were used for this purpose.
purpose.interlayers,
Without Without interlayers,
the diffusion thebonding
diffusionexperiments
bonding experiments were onlyatachieved
were only achieved 1000 ◦ C forat
1000 °C for 10 and 60 min. However, some defects are observed
10 and 60 min. However, some defects are observed close to the ceramic base material. close to the ceramic base
material.
Using Usingwith
Ti films Ti films
~1 µm with ∼1 µm thickness,
thickness, the contactthe contactthe
between between the base was
base materials materials was
improved,
improved, allowing sound joints to be obtained at 950 °C for 60
allowing sound joints to be obtained at 950 ◦ C for 60 min. The joining interface wasmin. The joining interface
was characterized
characterized by α-Ti,
by α-Ti, α2 -Tiα3 2Al
-Tiand
3Al and γ-TiAl phases. The increase in the joining temper-
γ-TiAl phases. The increase in the joining temperature
and time promote the increase in theinthickness
ature and time promote the increase the thickness
of theoflayers
the layers
at theatinterface.
the interface. An in-
An intense
tense diffusion is observed, and the fine microstructure of the thin
diffusion is observed, and the fine microstructure of the thin film seems to improve the film seems to improve
the diffusion
diffusion bonding
bonding process.
process. TheThe use use
of Tioffreestanding
Ti freestandingfilmsfilms
waswas unsuccessful,
unsuccessful, proving
proving that
that the adhesion to the ceramic material is a crucial aspect for this approach.
the adhesion to the ceramic material is a crucial aspect for this approach. The hardness The hardness
maps confirm
maps confirm thethe microstructural
microstructural results,
results, revealing
revealing that
that the
the interface
interface exhibited
exhibitedaahardness
hardness
similar to
similar to the
the Ti6Al4V,
Ti6Al4V, butbut it it increases
increases with
with the
the increase
increase in in temperature
temperature and and time
time ofof the
the
diffusion bonding process. The average shear strength values are
diffusion bonding process. The average shear strength values are similar for all joints similar for all joints
tested, confirming
tested, confirming that that the
the useuse ofof the
the Ti
Ti thin
thin film
film improves
improves the the diffusion
diffusion bonding
bonding of of the
the
Ti6Al4V alloy to alumina, promoting a decrease in the temperature and time required to
achieve sound ceramic/metal joining.
Metals 2021, 11, 1728 15 of 17

Ti6Al4V alloy to alumina, promoting a decrease in the temperature and time required to
achieve sound ceramic/metal joining.

Author Contributions: Conceptualization, M.S.J.; methodology, S.S. and A.S.R.; validation, S.S. and
A.S.R.; investigation, M.S.J., S.S. and A.S.R.; writing—original draft preparation M.S.J.; writing—
review and editing, S.S. and A.S.R.; Supervision, S.S. and A.S.R.; funding acquisition, S.S. All authors
have read and agreed to the published version of the manuscript.
Funding: This work was financially supported by: Project PTDC/CTM-CTM/31579/2017—POCI-
01-0145-FEDER-031579—funded by FEDER funds through COMPETE2020—Programa Operacional
Competitividade e Internacionalização (POCI) and by national funds (PIDDAC) through FCT/MCTES.
This research was also supported by FEDER funds through the program COMPETE—Programa
Operacional Factores de Competitividade, and by national funds through FCT—Fundação para a
Ciência e a Tecnologia, under the project UIDB/EMS/00285/2020.
Data Availability Statement: Data can be available upon request from the authors.
Acknowledgments: The authors are grateful to CEMUP-Centre of Materials of the University of
Porto for expert assistance with SEM.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Titanium and Titanium Alloys; Leyens, C.; Peters, M. (Eds.) Wiley: Hoboken, NJ, USA, 2003; ISBN 9783527305346.
2. Carter, C.B.; Norton, M.G. Ceramic Materials; Springer: New York, NY, USA, 2007; ISBN 978-0-387-46270-7.
3. Simões, S. Recent Progress in the Joining of Titanium Alloys to Ceramics. Metals 2018, 8, 876. [CrossRef]
4. Li, H.; Ma, Y.; Xu, B.; Bridges, D.; Zhang, L.; Feng, Z.; Hu, A. Laser welding of Ti6Al4V assisted with nanostructured Ni/Al
reactive multilayer films. Mater. Des. 2019, 181, 108097. [CrossRef]
5. Kaneko, A.; Katayama, T.; Morishita, S. Micro Fabrication of Au Thin-Film by Transfer-Printing Using Atomic Diffusion Bonding.
Int. J. Autom. Technol. 2019, 13, 810–816. [CrossRef]
6. Cazajus, V.; Seguy, S.; Welemane, H.; Karama, M. Residual Stresses in a Ceramic-Metal Composite. Appl. Mech. Mater. 2012, 146,
185–196. [CrossRef]
7. Shi, G.; Zhang, L.; Wang, Z. Modelling the Elements Reaction–Diffusion Behavior on Interface of Ti/Al2 O3 Composite Prepared
by Hot Pressing Sintering. Metals 2020, 10, 259. [CrossRef]
8. Misra, A.K. Reaction of Ti and Ti-Al alloys with alumina. Met. Mater. Trans. A 1991, 22, 715–721. [CrossRef]
9. Lu, Y.-C.; Sass, S.; Bai, Q.; Kohlstedt, D.; Gerberich, W. The influence of interfacial reactions on the fracture toughness of Ti-Al2 O3
interfaces. Acta Met. Mater. 1995, 43, 31–41. [CrossRef]
10. Yu, W.; Zhao, H.; Huang, Z.; Chen, X.; Aman, Y.; Li, S.; Zhai, H.; Guo, Z.; Xiong, S. Microstructure evolution and bonding
mechanism of Ti2 SnC-Ti6Al4V joint by using Cu pure foil interlayer. Mater. Charact. 2017, 127, 53–59. [CrossRef]
11. Loh, N.L.; Wu, Y.L. Diffusion bonding of ceramic to metals. Mater. Manuf. Proc. 1993, 8, 159–181. [CrossRef]
12. Nicholas, M.G.; Crispin, R.M. Diffusion bonding stainless steel to alumina using aluminium interlayers. J. Mater. Sci. 1982, 17,
3347–3360. [CrossRef]
13. Yin, X.; Li, M.; Zhou, Y. Microstructure and mechanical strength of diffusion-bonded Ti3 SiC2 /Ni joints. J. Mater. Res. 2006, 21,
2415–2421. [CrossRef]
14. Hattali, M.; Valette, S.; Ropital, F.; Mesrati, N.; Tréheux, D. Interfacial behavior on Al2 O3 /HAYNES® 214™ joints fabricated by
solid state bonding technique with Ni or Cu–Ni–Cu interlayers. J. Eur. Ceram. Soc. 2012, 32, 2253–2265. [CrossRef]
15. Uday, M.B.; Ahmad-Fauzi, M.N.; Noor, A.M.; Rajoo, S. Current Issues and Problems in the Joining of Ceramic to Metal. In Joining
Technologies; Ishak, M., Ed.; InTech: Rijeka, Croatia, 2016; Chapter 8.
16. Elssner, G.; Petzow, G. Metal/ceramic joining. ISIJ Int. 1990, 30, 1011–1032. [CrossRef]
17. Zhang, Y.; Feng, D.; He, Z.-Y.; Chen, X.-C. Progress in Joining Ceramics to Metals. J. Iron Steel Res. Int. 2006, 13, 1–5. [CrossRef]
18. Yang, Z.; Lin, J.; Wang, Y.; Wang, D. Characterization of microstructure and mechanical properties of Al2 O3 /TiAl joints vacuum-
brazed with Ag-Cu-Ti + W composite filler. Vacuum 2017, 143, 294–302. [CrossRef]
19. Cao, J.; Zheng, Z.; Wu, L.; Qi, J.; Wang, Z.; Feng, J. Processing, microstructure and mechanical properties of vacuum-brazed
Al2 O3 /Ti6Al4V joints. Mater. Sci. Eng. A 2012, 535, 62–67. [CrossRef]
20. Niu, G.; Wang, D.; Yang, Z.; Wang, Y. Microstructure and mechanical properties of Al2 O3 /TiAl joints brazed with B powders
reinforced Ag-Cu-Ti based composite fillers. Ceram. Int. 2017, 43, 439–450. [CrossRef]
21. Xue, H.; Wei, X.; Guo, W.; Zhang, X. Bonding mechanism study of active Ti element and α-Al2 O3 by using first-principle
calculation. J. Alloys Compd. 2020, 820, 153070. [CrossRef]
22. Qiu, Q.; Wang, Y.; Yang, Z.; Wang, D. Microstructure and mechanical properties of Al2 O3 ceramic and Ti6Al4V alloy joint brazed
with inactive Ag–Cu and Ag–Cu + B. J. Eur. Ceram. Soc. 2016, 36, 2067–2074. [CrossRef]
Metals 2021, 11, 1728 16 of 17

23. Wang, Z.M.; Yang, J.H.; Qi, F.Q.; Hou, A.L.; Yang, Z.W.; Wang, Y.; Wang, D.P. Interfacial characterization and mechanical properties
of reactive air brazed ZrO2 ceramic joints with Ag-CuO-Al2 TiO5 composite filler metal. Ceram. Int. 2021, 47, 29128–29138.
[CrossRef]
24. Xin, C.; Yan, J.; Li, N.; Liu, W.; Du, J.; Cao, Y.; Shi, H. Microstructural evolution during the brazing of Al2 O3 ceramic to kovar
alloy by sputtering Ti/Mo films on the ceramic surface. Ceram. Int. 2016, 42, 12586–12593. [CrossRef]
25. Dai, X.; Cao, J.; Liu, J.; Wang, D.; Feng, J. Interfacial reaction behavior and mechanical characterization of ZrO2 /TC4 joint brazed
by Ag–Cu filler metal. Mater. Sci. Eng. A 2015, 646, 182–189. [CrossRef]
26. Barbier, F.; Peytour, C.; Revcolevschi, A. Microstructural Study of the Brazed Joint between Alumina and Ti-6AI-4V Alloy. J. Am.
Ceram. Soc. 1990, 73, 1582–1586. [CrossRef]
27. Emadinia, O.; Guedes, A.; Tavares, C.; Simões, S. Joining Alumina to Titanium Alloys Using Ag-Cu Sputter-Coated Ti Brazing
Filler. Materials 2020, 13, 4802. [CrossRef]
28. Peytour, C.; Barbier, F.; Revcolevschi, A. Characterization of ceramic/TA6V titanium alloy brazed joints. J. Mater. Res. 1990, 5,
127–135. [CrossRef]
29. Mishra, S.; Sharma, A.; Jung, D.H.; Jung, J.P. Recent Advances in Active Metal Brazing of Ceramics and Process. Met. Mater. Int.
2019, 26, 1087–1098. [CrossRef]
30. Akselsen, O.M. Diffusion bonding of ceramics. J. Mater. Sci. 1992, 27, 569–579. [CrossRef]
31. Simões, S.; Viana, F.; Ramos, A.S.; Vieira, M.T. Microstructural Characterization of Dissimilar Titanium Alloys Joints Using Ni/Al
Nanolayers. Metals 2018, 8, 715. [CrossRef]
32. Simões, S.; Viana, F.; Ramos, A.S.; Vieira, M. Diffusion Bonding of TiAl to Ti6Al4V Using Nanolayers. J. Mater. Eng. Perform. 2018,
27, 5064–5068. [CrossRef]
33. Simões, S.; Viana, F.; Koçak, M.; Ramos, A.S.; Vieira, M. Diffusion bonding of TiAl using reactive Ni/Al nanolayers and Ti and Ni
foils. Mater. Chem. Phys. 2011, 128, 202–207. [CrossRef]
34. Aucott, L.; Bamber, R.; Lunev, A.; Darby, T.; Maquet, P.; Gimbert, N.; Pak, S.; Walsh, M.; Udintsev, V.; Eaton, G.; et al. Solid-State
Diffusion Bonding of Glass-Metal for the International Thermonuclear Experimental Reactor (ITER) Diagnostic Windows. In
TMS 2020 149th Annual Meeting & Exhibition Supplemental Proceedings; Springer International Publishing: Cham, Switzerland,
2020; pp. 2085–2094.
35. Cavaleiro, A.; Ramos, A.; Fernandes, F.B.; Schell, N.; Vieira, M. Follow-up structural evolution of Ni/Ti reactive nano and
microlayers during diffusion bonding of NiTi to Ti6Al4V in a synchrotron beamline. J. Mater. Process. Technol. 2020, 275, 116354.
[CrossRef]
36. Weiqi, Y.; Jincheng, L.; Run, A.; Lili, X. High shear strength and ductile ZrC–SiC/austenitic stainless steel joints bonded with
Ti/Ni foam interlayer. Ceram. Int. 2020, 46, 3036–3042. [CrossRef]
37. Niu, J.B.; Wang, Y.; Yang, Z.W.; Wang, D.P. Microstructure and Mechanical Properties of Titanium–Zirconium–Molybdenum and
Ti2 AlNb Joint Diffusion Bonded with and without a Ni Interlayer. Adv. Eng. Mater. 2019, 21, 1900713. [CrossRef]
38. Wang, Q.; Chen, G.-Q.; Wang, K.; Fu, X.-S.; Zhou, W.-L. Microstructural evolution and growth kinetics of interfacial compounds
in TiAl/Ti3 SiC2 diffusion bonding joints. Mater. Sci. Eng. A 2019, 756, 149–155. [CrossRef]
39. Zhang, Y.; Chen, Y.; Yu, D.; Sun, D.; Li, H. A review paper on effect of the welding process of ceramics and metals. J. Mater. Res.
Technol. 2020, 9, 16214–16236. [CrossRef]
40. Barrena, M.; Matesanz, L.; de Salazar, J.G. Al2 O3 /Ti6Al4V diffusion bonding joints using Ag–Cu interlayer. Mater. Charact. 2009,
60, 1263–1267. [CrossRef]
41. Kliauga, A.; Ferrante, M. Interface compounds formed during the diffusion bonding of Al2 O3 to Ti. J. Mater. Sci. 2000, 35,
4243–4249. [CrossRef]
42. Travessa, D.; Ferrante, M. The Al2 O3 -t itanium adhesion in the view of the diffusion bonding process. J. Mater. Sci. 2002, 37,
4385–4390. [CrossRef]
43. Rocha, L.; Ariza, E.; Costa, A.M.; Oliveira, F.J.; Silva, R.F. Electrochemical behavior of Ti/Al2 O3 interfaces produced by diffusion
bonding. Mater. Res. 2003, 6, 439–444. [CrossRef]
44. Heikinheimo, L.S.K.; De With, G. Interface Structure and Fracture Energy of Al2 O3 -Ti Joints: Dissertation; VTT Technical Research
Centre of Finland: Espoo, Finland, 1995.
45. Lim, J.D.; Lee, P.M.; Chen, Z. Understanding the bonding mechanisms of directly sputtered copper thin film on an alumina
substrate. Thin Solid Film. 2017, 634, 6–14. [CrossRef]
46. Silva, M.; Ramos, A.; Vieira, M.; Simões, S. Diffusion Bonding of Ti6Al4V to Al2 O3 Using Ni/Ti Reactive Multilayers. Metals 2021,
11, 655. [CrossRef]
47. Oliver, W.; Pharr, G. An improved technique for determining hardness and elastic modulus using load and displacement sensing
indentation experiments. J. Mater. Res. 1992, 7, 1564–1583. [CrossRef]
48. Lim, J.D.; Lee, P.M.; Rhee, D.M.W.; Leong, K.C.; Chen, Z. Effect of surface treatment on adhesion strength between magnetron
sputtered copper thin films and alumina substrate. Appl. Surf. Sci. 2015, 355, 509–515. [CrossRef]
49. Hayes, F.H. The Al-Ti-V (aluminum-titanium-vanadium) system. J. Phase Equilibria Diffus. 1995, 16, 163–176. [CrossRef]
50. Zhu, S.; Włosiński, W. Joining of AlN ceramic to metals using sputtered Al or Ti film. J. Mater. Process. Technol. 2001, 109, 277–282.
[CrossRef]
Metals 2021, 11, 1728 17 of 17

51. Valenza, F.; Artini, C.; Passerone, A.; Cirillo, P.; Muolo, M.L. Joining of ZrB2 Ceramics to Ti6Al4V by Ni-Based Interlayers. J. Mater.
Eng. Perform. 2014, 23, 1555–1560. [CrossRef]
52. Zhang, X.; Lu, G. Properties and Interface Structures of Ni and Ni-Ti Alloy Toughened A12 O3 Ceramic Composites. J. Eur. Ceram.
Soc. 1994, 15, 225–232. [CrossRef]
53. Lu, Y.; Zhu, M.; Zhang, Q.; Hu, T.; Wang, J.; Zheng, K. Microstructure evolution and bonding strength of the Al2 O3 /Al2 O3
interface brazed via Ni-Ti intermetallic phases. J. Eur. Ceram. Soc. 2020, 40, 1496–1504. [CrossRef]
54. Yi, J.; Zhang, Y.; Wang, X.; Dong, C.; Hu, H. Characterization of Al/Ti Nano Multilayer as a Jointing Material at the Interface
between Cu and Al2 O3 . Mater. Trans. 2016, 57, 1494–1497. [CrossRef]
55. Gietzelt, T.; Toth, V.; Huell, A. Diffusion Bonding: Influence of Process Parameters and Material Microstructure. In Joining
Technologies; Intech Open Publishers: London, UK, 2016; pp. 195–216. [CrossRef]