materials
Article
Resistance to High-Temperature Oxidation of Ti-Al-Nb Alloys
Joanna Małecka






Citation: Małecka, J. Resistance to
High-Temperature Oxidation of
Ti-Al-Nb Alloys. Materials 2022, 15,
2137. https://doi.org/10.3390/
ma15062137
Academic Editor:
Donatella Giuranno
Received: 28 January 2022
Accepted: 12 March 2022
Published: 14 March 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the author.
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/).
Materials 2022, 15, 2137. https://doi.org/10.3390/ma15062137
Faculty of Mechanical Engineering, Opole University of Technology, 45-271 Opole, Poland; j.malecka@po.edu.pl;
Tel.: +48-77-4498466
Abstract: The research presented in this paper concerns the assessment of the resistance to high-
temperature oxidation behaviour of a Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy and the explanation of
the role of niobium during oxidation processes. The basic problem concerned the evaluation of the
resistance of the studied alloy to cyclic oxidation in an air atmosphere, with particular attention to
the influence of temperature, surface roughness and cooling rate from heating temperature to room
temperature. The issue analysed was the effect of niobium addition on the corrosion kinetics as a
high-melting element causing improved oxidation resistance, contributing to the reduction in the
oxidation rate.
Keywords: oxidation; intermetallic; Ti-Al alloys; high temperature
1. Introduction
The alloys whose structure consists of intermetallic phases of the Ti-Al system are
attractive materials due to their properties which allow using them in many areas of activity,
including commercially in aerospace, aircraft and automobile applications. These alloys,
thanks to their properties, are used even at temperatures above 600 ◦ C, where classical
titanium alloys are not suitable for use due to low creep and oxidation resistance [1,2].
Recent research works [3–11] largely concern improving the oxidation resistance of Ti-Al
intermetallic alloys.
This issue is currently the main problem causing application limitations of the dis-
cussed alloys, especially for components intended for long-term operation at cyclically
varying temperatures. Increasing the resistance to oxidation is possible, among other
things, by modifying the chemical composition with suitable alloying elements, which is
the subject of this work.
The oxidation of intermetallic Ti-Al alloys is significantly dependent on their chemical
composition. A slowly growing, continuous and compact scale can exhibit protective
capabilities, thus providing resistance to high-temperature oxidation of materials. Such
properties are exhibited by alloys that are characterised by the ability to form an oxide
layer based on oxides of Al2 O3 , among others [12]. The oxidation kinetics of Ti-Al inter-
metallic alloys depend mainly on the temperature, but also on the cooling rate [13], surface
preparation [14], the degree of contamination and the type of microstructure [15], and the
presence of nitrogen [16,17], water vapor [18,19] or exhaust gas [20]. Alloys containing
more than 65% Al have the ability to form a favourable protective Al2 O3 layer. At lower
aluminium concentrations, the activity of this element is less than that of titanium, while
TiO oxide is more stable thermodynamically. Practically, this oxide never forms, and TiO2
oxide is formed instead. Therefore, alloys based on Ti3 Al and TiAl intermetallic phases
in which the Al content is lower (42–48% Al) undergo faster oxidation. Therefore, these
alloys should contain alloying additives that improve their oxidation resistance [21]. It is
generally believed that small amounts of Nb significantly reduce the kinetics of oxidation in
air, leading to the formation of a continuous layer of nitride at the oxide–metal interface that
controls downstream oxidation, and similar results were observed during the oxidation of
the Ti-47Al-2Cr-2Nb and Ti-45Al-2Cr-8Nb alloys [22]. Nb seems to be especially important
https://www.mdpi.com/journal/materials
Materials 2022, 15, 2137 2 of 19

because it is concentrated in the substrate near the phase boundary and limits formation
of the solid solution α-Ti(Al,N,O). Nb, being a β-forming element in relation to titanium,
promotes the formation of a poorly soluble phase β-Ti(Al,N,O) of nitrogen and oxygen. It
limits the diffusion of interstitial elements to the core, and as a consequence, it slows down
the increase in oxidation products.
This paper assumes that the increase in oxidation resistance of the discussed alloys
can be achieved by the introduction of Nb as an alloying element, which can reduce the
rate of oxidation of intermetallics. Optimal Nb content is in the range of 4–10 at.%, so this
research was concentrated on an alloy containing 7% Nb. The main goal of this work was
to determine the range of oxidation resistance of an alloy containing 7% Nb. On the basis
of an analysis of corrosion products and diffusion processes in the metallic substrate and
oxide layer, the mechanism of corrosion at high temperatures was determined. The next
stage of the work consisted of determining the influence of increased niobium content on
the course of oxidation, with the main focus on the influence of niobium on the rate of the
process. The most common alloy studied in the literature for resistance to high-temperature
oxidation is the alloy containing 2% Nb [22,23]. In this work, the literature data on the
alloy Ti-48Al-2Cr-2Nb [24] were treated as a comparison and reference for the investigated
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy. This paper presents the results of reference alloys at
temperatures below 875 ◦ C only for informative results, not for comparison.
The most common oxidation process is isothermal oxidation; however, for practical
reasons, the introduction of cyclic temperature changes is a process designed to approximate
the test conditions to the actual working conditions of the material. Components of
structural equipment may work not only in high-temperature oxidation conditions but
also in cyclic operation conditions, requiring cyclic start-up and shut-down. During cyclic
oxidation, the scale is usually detached from the substrate. This is the primary reason for
the poorer resistance of the material under these conditions. The main factors of cyclic
oxidation determining the oxidation rate and the thickness of the oxide layer formed are
heating temperature and cycle frequency. It is worth noting that the compactness of the
oxidation product layer formed on multicomponent Ti-Al alloys is also determined by
other factors important for methodological reasons. Apart from the chemical composition
of the alloys, the atmosphere and the temperature at which they operate, the cooling rate
and the surface condition of the alloys prior to operation may play important roles [2,13].
Therefore, the present study considered not only the possibility of improving the oxidation
resistance by the effect of Nb addition, but also the sensitivity of the alloys to particular
factors of cyclic oxidation.

2. Material and Test Methods

A two-phase multi-component Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy was selected as
the test material. The material in the form of a cylinder with a diameter of about 69 mm
was purchased from Flowserve Corporation Titanium and Reactive Metals Foundry (RMI
Titanium Company Inc., Pittsburgh, PA, USA). This alloy was prepared by the arc melting
process. In order to remove the internal defects and obtain a homogenised sample, the
castings were processed by the HIP (hot isostatic pressing) method. The surfaces were
abraded on abrasive paper of 800 grade or by abrasive blasting using alundum of F40 grain
and finally cleaned with acetone. Specimens with dimensions of ≈20 mm × 15 mm × 2 mm
and surface roughness Ra = 0.06 µm (after grinding) and Ra = 5.8 µm (after abrasive
blasting) were prepared for testing. The surface roughness was measured using a TOPO-
01P profilometer (Digital Surf Companies, Besançon, France) with a diamond stylus radius
of 2 µm.
Cyclic oxidation experiments were realised at temperatures of 875 ◦ C, 900 ◦ C, 925 ◦ C,
950 C and 975 ◦ C for 60 cycles in atmospheric air. The samples were heated up in the
◦

desired atmosphere with the tube furnace (high-accuracy temperature sensors: ±1 ◦ C)

(Czylok, Jastrzebie-Zdroj, Poland), then annealed in a given temperature, and subsequently
cooled down to the room temperature. Each traction cycle consisted of heating the samples
Materials 2022, 15, x FOR PEER REVIEW 3 of 19

(Czylok, Jastrzebie-Zdroj, Poland), then annealed in a given temperature, and subse-

Materials 2022, 15, 2137 quently cooled down to the room temperature. Each traction cycle consisted of heating 3 of 19
the samples to a specific temperature, annealing at this temperature for 1 h with subse-
quent cooling, and holding at room temperature for 15 min. The gravimetric method was
selected
to to measure
a specific the reaction
temperature, annealingspeed. The temperature
at this control of weight
for 1 change was performed
h with subsequent on a
cooling,
RADWAG precision scale with accuracy of 10 −4 g. Trials were repeated three times and
and holding at room temperature for 15 min. The gravimetric method was selected to
the presented
measure test results
the reaction were
speed. Theaveraged.
control ofDuring
weightcyclic
change oxidation in air, inon
was performed addition
a RADWAG to the
temperature, the variable factors were
− 4 the surface condition as determined
precision scale with accuracy of 10 g. Trials were repeated three times and the presented by the Ra and
the value of the cooling rate from the heating temperature to room
test results were averaged. During cyclic oxidation in air, in addition to the temperature, temperature (18 °C).
Thevariable
the microstructure was then
factors were observed
the surface using aasHITACHI
condition determined S-4200
by thescanning
Ra and the electron
valuemicro-
of the
scope (Hitachi High-Tech in Europe GmbH, Krefeld, Germany)
cooling rate from the heating temperature to room temperature (18 C). The microstructureoperating
◦ at 20 kV. Sec-
ondary electron emission (SE) and backscattered electron emission
was then observed using a HITACHI S-4200 scanning electron microscope (Hitachi High- (BSE) were used in the
tests. Microanalyses of chemical composition were performed with
Tech in Europe GmbH, Krefeld, Germany) operating at 20 kV. Secondary electron emission the EDS method using
the Therm
(SE) NORAN attachment
and backscattered with System
electron emission (BSE)Six werecoupled
used inwith HITACHI
the tests. microscope.
Microanalyses of
Phase composition
chemical composition tests were
were performed
performed withbythe
X-rayEDSdiffraction
method usingusingthe theTherm
JDX-7SNORANdiffrac-
tometer (JEOL
attachment withCompanies,
System Six Tokyo,
coupledJapan) from JEOLmicroscope.
with HITACHI using a copperPhase X-ray tube (λCuKα 
composition tests
=  1.5406
were Å). Theby
performed recording was carried
X-ray diffraction outthe
using by JDX-7S
a step method with a step
diffractometer (JEOL ofCompanies,
0.05° and a
count of
Tokyo, 3 s infrom
Japan) the range
JEOL of 20° atocopper
using 90° 2θ.X-rayPhasetubeidentification
(λCuKα =was carried
1.5406 out recording
Å). The using the
was carried out by a step method with a step of 0.05 and a count of 3 s in the range of 20◦
ICDD database. ◦

to 90◦ 2θ. Phase identification was carried out using the ICDD database.
3. Test Results
3. Test Results
3.1. Microstructure Analysis
3.1. Microstructure Analysis
The structure of the tested Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy and the XRD microa-
nalysisTheisstructure
shown inofFigure
the tested Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
1. According to the microstructurealloy and the XRD
classification microanal-
of Ti-Al alloys,
ysis is shown in Figure 1. According to the microstructure classification
the tested alloy can be described as duplex. It reveals the clear contrasts between of Ti-Al alloys, the
the
tested alloy can be described as duplex. It reveals the clear contrasts between the phases.
phases. The TiAl and Ti3Al phase form a plate system and the TiAl phase is present locally
The TiAl and Ti3 Al phase form a plate system and the TiAl phase is present locally in the
in the granular form. Moreover, an irregular concentration of Ti3Al phase is found on the
granular form. Moreover, an irregular concentration of Ti3 Al phase is found on the border
border of the plate grain system. The distance between Ti3Al plates in a plate mixture Ti3Al
of the plate grain system. The distance between Ti3 Al plates in a plate mixture Ti3 Al + TiAl
+ TiAl is from ca. 0.5 μm to ca. 1.5 μm. The thickness of TiAl phase plates is 4 to 10 times
is from ca. 0.5 µm to ca. 1.5 µm. The thickness of TiAl phase plates is 4 to 10 times higher
higher than Ti3Al phase.
than Ti3 Al phase.
(a) (b)

Figure 1. The
Figure The(a)
(a)structure
structureand (b)(b)
and XRD microanalysis
XRD results
microanalysis of the
results of Ti–46Al–7Nb–0.7Cr–0.1Si–
the Ti–46Al–7Nb–0.7Cr–0.1Si–
0.2Ni alloy.
0.2Ni alloy.

3.2.
3.2. Determination
Determination of of Kinetics
Kinetics of
of Cyclic
Cyclic Oxidation
Oxidation
◦ to 925 ◦ C) over the range of the 60 per-
Oxidation
Oxidation at lower temperatures(from
at lower temperatures (from875875C°C to 925 °C) over the range of the 60
formed
performedcycles onlyonly
cycles results in mass
results gain (Figure
in mass 2). Oxidation
gain (Figure at 950 ◦at
2). Oxidation C and
950 975
◦ C induces
°C and 975 °C
mass gain only in the first cycles. ◦ C, this occurs after about 30 oxidation cycles
induces mass gain only in the firstFor 950 For
cycles. 950 °C, this occurs after about 30 oxidation
and forand ◦ C, after 14 oxidation cycles. At 975 ◦ C, after 30 oxidation cycles, the sample
975 for
cycles 975 °C, after 14 oxidation cycles. At 975 °C, after 30 oxidation cycles, the
reaches
sample reachesclose
a mass a massto close
the initial
to theone, and
initial with
one, and further cycles, cycles,
with further only mass
onlyloss
masstakes
lossplace.
takes
place.
Materials 2022, 15, x FOR PEER REVIEW 4 of 19

Materials
Materials 2022,
2022, 15,
15, x2137
FOR PEER REVIEW 4 4of
of 19
19

Figure 2. Mass change of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni per unit area with respect to exposed tem-
perature. Masschange
changeofofTi-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
Figure 2.
Figure 2. Mass perper
unitunit
areaarea
withwith respect
respect to exposed
to exposed tem-
temperature.
perature.
For the Ti-48Al-2Cr-2Nb alloy treated as a reference alloy in this work, the course of
For
cyclicFor the Ti-48Al-2Cr-2Nb
oxidation in the temperature alloy treated as a°C to 875 °C
reference alloy in thisinwork, the course of
the Ti-48Al-2Cr-2Nb alloyrange
treated of as
800a reference
◦ C to 875
is shown
alloy
◦ C in
Figure
this work, the3. course
The tem-of
cyclic oxidation
peratures of 800 °C,
cyclic oxidation in the
in the
825 °C
temperature
and
temperature 850 °C
range
showed of 800
relatively small is shown
mass gains inthroughout
range◦ of 800 °C to 875 °C is shown in Figure 3. The tem-
Figure 3. The the
◦ C, 825 ◦alloy
of°C,
temperatures
test. At 875 of 800reference C andachieved
850 C showed relatively1.2small mass gains throughout
800◦the a mass gain of mg/cm 2 after only aboutthe
30
peratures °C, 825 °C and 850 °C showed relatively small mass gains throughout
2 after
the test. Atcycles
oxidation 875 (Figure
C, the reference
3), while alloy
the achieved
tested alloy a with
mass7gainat.% of
Nb 1.2content
mg/cm had a only
mass about
gain of
test. At 875 °C, the reference alloy achieved a mass gain of 1.2 mg/cm2 after only about 30
30 oxidation
about 0.8 mg/cm cycles (Figure
and about 3), while
1 mg/cm the tested
afteralloyalloy
60 cycles with 7 at.% Nb content had a mass gain
withat this temperature (Figure 2). gain
At 900
2 2
oxidation cycles (Figure2 and 3), while the tested 2 after 60 7 at.% Nb content had a mass of
of about
°C and 0.8°C,
mg/cm about 1 mg/cm cycles at this temperature (Figure 2). At
about ◦ 0.8 mg/cm◦ and about 1 mg/cm after 60 cycles at this temperature (Figure 2). Atsuit-
925 the2 products formed of 2 the Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy showed 900
900 adhesion
able C and 925 the C, the products formed ofthe thetest.
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy showed
°C and 925 °C,to substrate
the products throughout
formed On the other hand, for alloy
of the Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni the reference
showedalloy
suit-
suitable
at 900 °C adhesion
and 925 to
°C,the substrate
weight loss throughout
can be observedthe test.
as On
early the
as other
10 and hand,
four for the reference
oxidation cycles,
able adhesion to the substrate throughout the test. On the other hand, for the reference alloy
alloy at 900 ◦(Figure
respectively C and 925The ◦ C, weight loss can be observed as early as 10 and four oxidation
at 900 °C and 925 °C,3). weightresults
loss can show that the alloy
be observed withas7%
as early 10niobium
and fourcontent
oxidation is charac-
cycles,
cycles,
terised respectively
by higher (Figure
resistance 3).
to The results
high-temperature show that the
oxidation alloy
than with
the 7%
alloy niobium
containing content
2% ni-is
respectively (Figure 3). The results show that the alloy with 7% niobium content is charac-
characterised
obium, by which by higher
manifests resistance
itself to high-temperature oxidation than the alloy containing
terised
2% niobium, higher
which resistance
manifests toin a reduction in the
high-temperature
itself in a reduction
oxidation
oxidation rate.
than
in the oxidation the rate.
alloy containing 2% ni-
obium, which manifests itself in a reduction in the oxidation rate.

Figure 3.
Figure Mass change
3. Mass change of
of Ti-48Al-2Cr-2Nb
Ti-48Al-2Cr-2Nb per
per unit
unit area
area with
with respect
respect to
to exposed
exposed temperature.
temperature.
FigureStudies
3. Masson
change of Ti-48Al-2Cr-2Nb
the effect of high coolingper unit(water)
rates area with
andrespect to exposed
very low rates (1 ◦ C/min)
coolingtemperature.
Studies on the effect of high cooling rates (water) and very low cooling rates (1
are presented in [2] and [13], respectively. Although at low cooling rates, less spalling of
°C/min) are presented
Studies on be in [2]
theexpected
effect and
ofdue
high [13], respectively.
cooling rates (water)Althoughvery at lowlowcooling rates, less
the scale would to its deformation by creepand cooling
at high temperature, rates
the (1
exper-
spalling
°C/min) of the scale
are presented
imental results would
indicatein be expected
[2]spalling
that and [13], due to its
respectively.
increases. deformation
Although
The reason by creep
at low
for this at high
is notcooling tempera-
rates, less
clear, although it
ture, theof
spalling experimental
is speculated thethat
scale results
thewould indicate
be expected
micro-crack formed that
at spalling
due to
high increases.
its cooling
deformation The
rates, by reason
creep
oriented at for this
high is not
tempera-
perpendicular to
clear,
ture, although
the
the alloy it increases
surface, is speculated
experimental results that the micro-crack
indicate
the resistance that spallingformed
to spalling at high
asincreases. cooling
Thethe
it decreases reason rates,
stresses oriented
forinthis
the is not
oxide
perpendicular
clear,
layer. although to the
it islow
In contrast, alloy surface,
speculated
coolingthat increases
ratesthe the
micro-crack
facilitate resistance
lateralformed
growth to spalling
at strain as
high cooling it decreases
rates,increases
[25], which the
oriented
perpendicular
the stresses. to the alloy surface, increases the resistance to spalling as it decreases the
Materials 2022, 15, x FOR PEER REVIEW 5 of 19
Materials 2022, 15, x FOR PEER REVIEW 5 of 19

Materials 2022, 15, 2137 stresses in the oxide layer. In contrast, low cooling rates facilitate lateral growth strain5[25], of 19
which increases
stresses the layer.
in the oxide stresses.In contrast, low cooling rates facilitate lateral growth strain [25],
which The study of
increases thethe effect of cooling rate was carried out for Ti-46Al-7Nb-0.7Cr-0.1Si-
stresses.
0.2NiThealloy cyclically
study of theoxidised 900 °Crate
effect ofatcooling 925 °C
and was (Figure
carried out4).for
For cycles of heating at 900
Ti-46Al-7Nb-0.7Cr-0.1Si-
The study of the effect of cooling rate was carried out for Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
°C with
and°C °C (Figure
0.2Ni subsequent
alloy cyclically cooling
oxidised in compressed ◦ C air, higher mass4).gains were found than after
alloy cyclically oxidised at 900 ◦atC 900 925and 925
(Figure Forofcycles
4). For cycles of heating
heating at 900 ◦ Catwith
900
°C
cooling
with in still air. Forcooling
subsequent the temperature
in compressed °C,higher
of 925air, differences
mass between
gains cooling
were methods
found than are
after
subsequent cooling in compressed air, higher mass gains were found than after cooling in
practically
cooling in non-existent.
still air. For the temperature◦ of 925 °C, differences between cooling
still air. For the temperature of 925 C, differences between cooling methods are practically methods are
practically
non-existent. non-existent.

Figure 4. Mass change of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni per unit area with respect to exposed tem-
Figure 4.
perature
Figure Masschange
Mass change
4.(continuous of of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
curves—cooling perper
unitunit
areaarea
in still air, dot curves—cooling
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni in with respect
compressed
with respect to exposed
to air).
exposed tem-
temperature
perature (continuous
(continuous curves—cooling
curves—cooling in still
in still air, air,
dot dot curves—cooling
curves—cooling in compressed
in compressed air).air).
The higher surface roughness (Ra = 5.8 μm) contributes to a slight increase in mass
Thethe
gain The
for higher surface roughness
temperature
higher surface roughness (Ra == 5.8
of 900 °C (Figure
(Ra 5.8 µm) is
5). This
μm) contributes to
due to the to
contributes aa slight
slight
specific increase
surface inwhich
area,in
increase mass
mass
gain for the temperature of 900 ◦ C (Figure 5). This is due to the specific surface area, which
is notfor
gain included in the macroscopically
the temperature of 900 °C (Figure determined
5). This is total surface
due to area. However,
the specific at higher
surface area, which
is not included
temperatures,
is in the
not includeddespite macroscopically determined
this specific surface
in the macroscopically total
area, the total
determined surface
situation area.
changes.
surface However, at higher
area. However, at higher
temperatures, despite
temperatures, despite this
this specific
specific surface
surface area,
area, the
the situation
situation changes.
changes.

Figure 5.
Figure Masschange
5. Mass changeof of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni perper
unitunit
areaarea
withwith respect
respect to exposed
to exposed tem-
temperature
perature
Figure (two
5. (two surfaces:
Masssurfaces:
change of Ra = 0.06 µm—continuous
RaTi-46Al-7Nb-0.7Cr-0.1Si-0.2Nicurves
= 0.06 μm—continuous curvesper
and and Ra
Ra area
unit = 5.8
= 5.8with µm—dot
μm—dot curves).
curves).
respect to exposed tem-
perature (two surfaces: Ra = 0.06 μm—continuous curves and Ra = 5.8 μm—dot curves).
At 925 ◦ C, larger mass gains are observed for samples with low surface roughness,
925 °C,
At larger mass gains are observed for samples with low surface roughness,
and at ◦
and At975
at 975 C,
°C,from
925°C, from “smooth”
larger“smooth” surfaces,
mass gains thethe
surfaces, scale
are observed easily
scale chips
foreasily off
chips
samples from
offlow
with thesurface
from metallic substrate,
the metallic sub-
roughness,
while
strate,
and it adheres
975 °C,
at while well
it adheresto surfaces with high
well to surfaces
from “smooth” surfaces,withroughness.
the high
scale roughness.
easily chips off from the metallic sub-
strate, while it adheres well to surfaces with high roughness.
3.3. Effect of Cyclic Oxidation on the Microstructure and Chemical Composition of the Scale
The diffusion processes become more active with increasing temperature and oxidation
time. This was observed at 950 ◦ C and 975 ◦ C (Figure 2). As indicated on the basis of the
performed investigations, analysing the images of the surfaces and cross-sections, the
formed layer is always characterised by the order of sublayers. An increase in process
 3.3. Effect of Cyclic Oxidation on the Microstructure and Chemical Composition of the Scale
3.3. Effect of Cyclic Oxidation
The diffusion processeson the Microstructure
become more activeand
withChemical Composition
increasing of theand
temperature Scale
oxida-
The diffusion
tion time. This wasprocesses at 950 °C
observedbecome more 975 °Cwith
andactive increasing
(Figure temperature
2). As indicated andbasis
on the oxida-of
Materials 2022, 15, 2137 tion time. This was observed at 950 °C and 975 °C (Figure 2). As indicated
the performed investigations, analysing the images of the surfaces and cross-sections, the on the basis
6 of of
19
the performed
formed layer isinvestigations, analysing
always characterised bythe
theimages
order of
ofthe surfacesAn
sublayers. andincrease
cross-sections, the
in process
formed layer however,
temperature, is always results
characterised by the
in greater order of of
thicknesses sublayers.
the oxideAn increase
layer in process
formed. On the
temperature, however, results in greater thicknesses of the oxide layer
temperature, however, results in greater thicknesses of the oxide layer formed. On(875–
oxidised alloy, there are columnar crystallites (Figure 6). At the lower formed.
temperatures On the
the
925 °C),
oxidised alloy,
they there
are veryare columnar
small, and crystallites
therefore, (Figure 6).
observations At the
were lower
made
oxidised alloy, there are columnar crystallites (Figure 6). At the lower temperatures temperatures
at magnifications(875–
of
925 °C),
(875–925 they
◦
4000×. TheC),oxideare very
they layers small,
are very and
formed, therefore,
which
small, and observations
adhere observations
therefore, were made
well to the metallic
were made at magnifications
substrate throughout
at magnifications of
4000×.
the4000
of test×The oxide
in. The
this layers
temperature
oxide formed,
layers range,which
formed, whichadhere
undergo well
spalling
adhere totothe
under
well metallic
dynamic
the substrate
metallicbending
substrate throughout
loads already
throughout
the test
after
the in this
oxidation
test in at 875 °C (Figure
this temperature
temperature range,7).
range, undergo spalling
undergo spalling under
under dynamic
dynamic bending
bending loads
loads already
already
after oxidation
after oxidation at 875 ◦°C
at 875 (Figure 7).
C (Figure 7).

(a) (b)
(a) (b)
Figure 6. Surface of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy exposed to air at 875  °C: (a) SEM micro-
Figure 6.6.Surface
graph—surface
Figure Surfaceprepared
ofofTi-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
by grinding [26]; (b) XRD alloy exposed
microanalysis
alloy exposed to air 875at◦ 875
toresult—small
atair C: (a) °C:
SEM(a)micrograph—
SEMcrystal-
columnar micro-
graph—surface
lites of prepared
surface TiO2. prepared
by grindingby [26];
grinding [26];microanalysis
(b) XRD (b) XRD microanalysis result—small
result—small columnar of
columnar crystallites crystal-
TiO2 .
lites of TiO2.

Figure 7.
Figure 7. SEM
SEM micrograph
micrograph of
of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy
alloy exposed
exposed to
to air
air at
at 875
875 ◦°C—surface
C—surface
Figure 7.
prepared SEM
by micrograph
grinding of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
(fragmentary chipping of products).
prepared by grinding (fragmentary chipping of products). alloy exposed to air at 875  °C—surface
prepared by grinding (fragmentary chipping of products).
The cross-sections and
The andmetallographic
metallographicspecimens
specimens ofofthethescale
scale alsoalso
show that that
show the sub-
the
The
layers
sublayersdocross-sections
not
do fully coverand
not fully themetallographic
cover entire surface
the entire specimens
(Figure
surface of 8).
8). EDS
(Figure the scale
analysis also showperformed
performed
EDS analysis that
in the sub-
location in
layers
1 do
according not tofully cover
Figure 8 (athe entire
clear surface
dominance (Figure
of titanium8). EDScan analysis
be seen)
location 1 according to Figure 8 (a clear dominance of titanium can be seen) reveals that the performed
reveals in
that location
the outer
1sublayer
according
outer sublayer to Figure
is rutile 8 (Figure
(Figure
is rutile (a 9a).
clearLocally,
dominance
9a). of
another
Locally, titanium
sublayer
another can beis
is in
sublayer seen)
direct reveals
in directcontact that
withthe
contact theouter
with oxi-
the
sublayer
oxidant. is
dant. Between rutile
Between (Figure
thethe
rutile
rutile9a).
andandLocally,
the
thenext another
nextsublayer, sublayer
sublayer,therethereare is in direct
arenano-voids
nano-voids(incontact with
(in BSE the
BSE as black). oxi-
dant. Between
Immediately
Immediately the rutile
below
below the and the next
the columnar
columnar rutilesublayer,
rutile crystallites
crystallitesthere arenext
is the
is the nano-voids
next sublayer,
sublayer, (in BSE as
which
which black).
contrasts
contrasts
Immediately
heterogeneously
heterogeneously belowgrey
greytheincolumnar
in the
the BSE. rutile crystallites
BSE. Microanalysis
Microanalysis of is sublayer
of this
this the next sublayer,
sublayer (Figure
(Figure 9b) which
9b) shows
shows contrasts
that
that Al
Al
heterogeneously
dominates there, grey
dominates there, within
with the
aa much
much BSE. Microanalysis
smaller
smaller participationof this
participation ofsublayer
of Ti. Thus,(Figure
Ti. Thus, itit can
can be 9b)
be shows that
concluded
concluded Al
that
that
dominates
there there, with aofmuch
thereisisaapredominance
predominance smalleroxides
aluminium
of aluminium participation
in this
oxides oflayer.
layer.
in this Ti.The
Thus,
thirdit
The can oxide
oxide
third besublayer
concluded that
extends
sublayer ex-
there
up is
to the
tends a predominance
up metallic substrate
to the metallic of aluminium
and it has
substrate andthe oxides in
greatest
it has this layer.
thickness.
the greatest The
thickness. third oxide sublayer ex-
tends up to the metallic substrate and it has the greatest thickness.
Materials
Materials2022,
Materials 2022,15,
2022, 15,xx2137
15, FOR
FOR PEER
PEERREVIEW
REVIEW 77 7of
ofof19
19
19

(a)
(a) (b)
(b)

Figure
Figure8.
Figure 8.8.SEM-BSE
SEM-BSE
SEM-BSE images
images of
of Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
images Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni alloy
alloy after
of Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni after cyclic
alloy after oxidation
cyclic oxidation at
at (a)
(a) 875
cyclic oxidation 875
at
°C
°C [26];
[26]; (b)
◦(b) 925
925 °C
°C (surface
(surface
◦ prepared
prepared by
by grinding).
grinding).
(a) 875 C [26]; (b) 925 C (surface prepared by grinding).

(a)
(a) (b)
(b)

(c)
(c) (d)
(d)

Figure 9. EDS analysis: (a) in point 1, (b) in point 2, (c) in point 3 and (d) in point 4 marked in
Figure9.
Figure
Figure 8b.
9. EDS
EDSanalysis:
analysis:(a)
(a)in
inpoint
point1,
1,(b)
(b)in
inpoint
point2,
2,(c)
(c)in
inpoint
point33and
and(d)
(d)in
inpoint
point44marked
markedininFigure
Figure
8b.
8b.
Ti oxides are dominant in this sublayer but Al oxides are present, too (Figure 9c).
The Ti
effect
Ti oxides
oxidesof are
thedominant
are oxidationin
dominant ofthis
in thesublayer
this Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
sublayer but
but Al
Al oxides
oxides are present,alloy
are present, too carried
too (Figure
(Figure out
9c).
9c). The
Thein
the temperature range 875–925 ◦ C is not only the formation ofcarried
reaction products but
effect
effect of
of the
the oxidation
oxidation of
of the
the Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni alloy
alloy carried out
out in
in the
the tem-
tem-
also the
perature course
perature range of
range 875–925diffusion
875–925 °C °C is processes
is not
not only in
only the the metallic
the formation
formation of substrate.
of reaction There
reaction products is a
products but bright
but also band
also thethe
of discontinuous
course
course of diffusiongroupings
of diffusion processes
processes in ofthe
in elements
the metallicwith
metallic large There
substrate.
substrate. atomic
There isnumbers
is aa bright (Figure
bright band
band of 8b, point
of discontin-
discontin-
#4). groupings
uous
uous This is an of
groupings area
of rich inwith
elements
elements elements
with large with
large atomic
atomichigh atomic(Figure
numbers
numbers numbers
(Figure 8b, such
point as
8b, point #4).Nb,
#4). This
This Niis
isand
an
an
Cr
area (Figure 9d). Such observations regarding the structure of oxidation products were
area rich
rich in
in elements
elements with
with high
high atomic
atomic numbers
numbers suchsuch asas Nb,
Nb, Ni
Ni and
and Cr
Cr (Figure
(Figure 9d).9d). Such
Such
confirmed by quantitative
observations microanalysis studies (results arewere
summarised in Table 1).
observations regarding
regarding the the structure
structure of of oxidation
oxidation products
products were confirmed
confirmed by by quantita-
quantita-
tive
tive microanalysis
microanalysis studies
studies (results
(results areare summarised
summarised in in Table
Table 1).
1).
Materials 2022, 15, x FOR PEER REVIEW 8 of 19

Materials 2022, 15, 2137 8 of 19

Table 1. EDS analysis in points marked in Figure 8 (mean values and standard deviations).

Element Ti Al Nb Cr Ni Si
Table 1. EDS analysis in points marked in Figure 8 (mean values and standard deviations).
Chemical composition from %mass 75.9 ± 2.16 23.1 ± 0.67 - - - 1.0 ± 0.12
the location #1 %at.
Element 64.0
Ti ± 1.69 34.6Al± 0.85 Nb- Cr- Ni- 1.4Si
± 0.16
Chemical
Chemical composition
composition fromfrom
the %mass
%mass 75.9 ±±
43.3 1.01
2.16 50.0
23.1 ± ±0.67
1.13 6.7- ± 0.7 -- -- 1.0 ±-0.12
location
the #1 #2
location %at.
%at. 64.0 ±
32.0 ±1.69
0.92 34.6
65.5± ±0.85
2.03 2.5 -± 0.28 -- -- 1.4 ±-0.16
Chemical composition fromfrom
the %mass 43.3 ± 1.01 50.0 ± 1.13 6.7 ± 0.7 - - -
Chemical composition %mass 73.1 ± 2.61 17.1 ± 0.53 9.8 ± 0.62 - - -
location #2 %at. 32.0 ± 0.92 65.5 ± 2.03 2.5 ± 0.28 - - -
the location #3 %at. 67.4 ± 1.72 27.0 ± 0.84 5.6 ± 0.76 - - -
Chemical composition from the %mass 73.1 ± 2.61 17.1 ± 0.53 9.8 ± 0.62 - - -
Chemical composition
location #3 from %mass
%at. 67.4 ± ±
44.5 1.12
1.72 27.0 ± ±0.84
25.2 0.64 5.6 ± ±0.76
26.1 0.69 1.3 -± 0.13 2.9-± 0.33 - -
the location #4
Chemical composition from the %at.
%mass 44.5 ±±
42.9 0.97 25.2
1.12 41.5± ±0.64 12.7±±0.69
0.86 26.1 0.6±±0.13
0.42 1.3 2.3±±0.33
0.04 2.9 0.26 --
location #4 %at. 42.9 ± 0.97 41.5 ± 0.86 12.7 ± 0.42 0.6 ± 0.04 2.3 ± 0.26 -
Investigations of the tested alloys with surface prepared by grinding after oxidation
at 950 Investigations
°C and 975 °Cofwere the tested alloystowith
impossible surface
carry prepared
out without by grinding
violating after oxidation
the compactness and
at 950 ◦ C and 975 ◦ C were impossible to carry out without violating the compactness and
cohesion of the layer. Already after about 30 h of oxidation at 950 °C and after 14 h of
cohesion
oxidationof at the
975 layer.
°C, theAlready
productsafter onlyabout
locally30retained
h of oxidation at 950 ◦ CThe
their continuity. andobservation
after 14 h ofof
oxidation at 975 ◦ C, the products only locally retained their continuity. The observation of
the external surface of the products indicates that the formed columnar rutile crystallites
the
are external
much bigger surface of the
in size thanproducts
after theindicates
oxidation that the formed
at 875 °C to 925columnar rutile
°C (Figures 6 andcrystallites
10). The
are much bigger in size than after the oxidation at 875 ◦ C to 925 ◦ C (Figures 6 and 10). The
formation of such products, as shown in Figure 10, requires explanation. These crystallites
formation
are formedofnot such products,
only as shown
by out-core in Figure
diffusion 10, requires
of titanium but explanation.
also by surfaceThese crystallites
diffusion. The
are formed not only by out-core diffusion of titanium but also by surface diffusion. The
interpretation can be based on the model presented by Jungling and Rapp [27]. Studies by
interpretation can be based on the model presented by Jungling and Rapp [27]. Studies
Rapp and co-workers [28,29] using hot-stage environmental SEM and video camera re-
by Rapp and co-workers [28,29] using hot-stage environmental SEM and video camera
cordings have shown that at lower temperatures, surface diffusion is the fastest process
recordings have shown that at lower temperatures, surface diffusion is the fastest process
that leads to whisker formation. A slight share of lattice diffusion causes their extension,
that leads to whisker formation. A slight share of lattice diffusion causes their extension,
i.e., growth in the perpendicular direction. As temperature increases, the participation of
i.e., growth in the perpendicular direction. As temperature increases, the participation of
lattice diffusion becomes significant, so oxide growth occurs rapidly over the entire grain
lattice diffusion becomes significant, so oxide growth occurs rapidly over the entire grain
surface, while in the vicinity of the dislocation, surface stresses and energy are minimised
surface, while in the vicinity of the dislocation, surface stresses and energy are minimised
by cavity growth in depth. The observed crystallites therefore form with the interaction
by cavity growth in depth. The observed crystallites therefore form with the interaction
of both types of diffusion. This proves that there is an out-core transfer of titanium ions
of both types of diffusion. This proves that there is an out-core transfer of titanium ions
through the product layer of both Al2O3-rich and TiO2-rich products. At the same time, it
through the product layer of both Al2 O3 -rich and TiO2 -rich products. At the same time,
proves
it proves that
that a sublayer
a sublayer of of
AlAl2O3 sufficiently homogeneous to be capable of inhibiting the
2 O3 sufficiently homogeneous to be capable of inhibiting
diffusion processes and, therefore,
the diffusion processes and, therefore, of markedly
of markedlyincreasing the resistance
increasing to oxidation,
the resistance was
to oxidation,
not formed. The outer surfaces of samples oxidised at 950 °C and
was not formed. The outer surfaces of samples oxidised at 950 C and 975 C for 14 and◦ 975 °C for◦ 14 and 30 h,
when cooled to ambient temperature, show fragmentary spalling of
30 h, when cooled to ambient temperature, show fragmentary spalling of the essential part the essential part of
thethe
of oxidation
oxidation products.
products. As the
As holding
the holdingtimetime
at room temperature
at room increased,
temperature stresses
increased, accu-
stresses
mulated and further “delayed” spalling occurred until the essential
accumulated and further “delayed” spalling occurred until the essential part of the layer part of the layer was
completely lost (Figures 11 and
was completely lost (Figures 11 and 12). 12).

Figure 10. SEM micrograph of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni

Figure Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Nialloy alloyexposed
exposedto
toair
airat 950 ◦°C—surface
at 950 C—surface
prepared by
prepared by grinding
grinding (bigger
(biggercolumnar
columnarcrystallites
crystallitesof
ofTiO
TiO22).).
Materials 2022,15,
Materials2022, 15,2137
x FOR PEER REVIEW 99 of 19
19

Figure 11.
Figure 11. SEM micrograph
micrograph of
of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Nialloy
alloyexposed
exposedto
toair
airat 900◦°C—surface
at900 C—surface
prepared by grinding (fragmentary chipping of products) [30].
prepared by grinding (fragmentary chipping of products) [30].

Figure12.
Figure 12. Surface
Surfaceof
ofTi-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy
alloy after
after spallation
spallation of the
of the basic
basic layer
layer of oxida-
of oxidation
tion products (prepared by grinding). The structures of products in fields A, B and C
products (prepared by grinding). The structures of products in fields A, B and C are shown in are shown in
Figure 13.
Figure 13.

Thesurface
The surfaceof ofthe
theoxidised
oxidisedalloy
alloyafter
afterspallation
spallationof ofthe
the basic
basic layer
layer of
of oxidation
oxidation products
products
is shown in Figure 13. The structure of these products is extremely fine-grained.
is shown in Figure 13. The structure of these products is extremely fine-grained. The process The process
of such fragmentary spalling occurs not only in one cycle but also in
of such fragmentary spalling occurs not only in one cycle but also in many previous cycles. many previous cycles.
In the micro-areas
In micro-areas in inFigure
Figure12,12,differences
differencesinin composition
composition become
become apparent.
apparent.Thus, the sur-
Thus, the
face fragments
surface fragments in which
in which point A was
point A waschosen
chosen according
according to to
Figure
Figure 1212areare
clearly enriched
clearly enriched in
titanium
in titanium (about 76.776.7
(about at.%)at.%)
and contain aluminium
and contain (Figure(Figure
aluminium 14a, Table
14a,2),Table
while2),thewhile
composi-
the
composition of the that
tion of the sublayer sublayer that remained
remained after the in
after the spallation, spallation,
which pointsin which
B andpoints
C wereBchosen,
and C
besides
were titanium
chosen, andtitanium
besides aluminium, also contains
and aluminium, alsonickel,
contains chromium and niobium
nickel, chromium (Figure
and niobium
(Figure 14b,c).
14b,c). This This structure
structure of the
of the part part
of the of the sublayer
sublayer remaining remaining on the metallic
on the metallic substratesubstrate
after the
after
basicthe basic
layer layer of oxidation
of oxidation productsproducts
has spalled hasoff
spalled off indicates
indicates that they that they are multiphase
are multiphase and mul-
and multilayer
tilayer but compact.but compact. The morphology
The morphology and layering andarelayering are nofrom
no different different
those from those
obtained at
obtained at lower temperatures
lower temperatures (Figure 15).
(Figure 15). However, However,
it can it canthat
be observed be observed
micropores that micropores
develop more
develop
actively.more actively.
This is related This
to theisincrease
related toin the increase in temperature.
temperature.

Table2.2.EDS
Table EDSanalysis
analysisresults
resultsin
inpoints
pointsmarked
markedin
inFigure
Figure12
12(mean
(meanvalues
valuesand
andstandard
standarddeviations).
deviations).

Element Element
Ti Ti Al Al Nb Nb Cr Cr NiNi
Chemical composition from the %mass 85.4
%mass± 3.01 14.6 ±
85.4 ± 3.01 0.4314.6 ± 0.43 - - - - - -
Chemical location
composition
#A from the location
%at. #A 76.7 ± 2.18 23.3 ± 0.71 - - -
%at. 76.7 ± 2.18 23.3 ± 0.71 - - -
Chemical composition from the %mass 61.7 ± 1.72 10.5 ± 0.73 27.8 ± 0.91 - -
%at. #B %mass
65.1 ± 1.79 61.7 19.7
± 1.72
± 0.4310.5 ± 15.2 27.8 ± 0.91 -
0.73± 0.42 - - -
Chemical location #B
composition from the location
Chemical composition from the %mass %at.
61.5 ± 1.45 65.1 22.3
± 1.79
± 0.5819.7 ± 0.43 15.2
14.0 ± 0.40 ± 0.42
0.9 ± 0.05 - -
1.3 ± 0.15
location #C %at. 55.8
%mass± 1.12 61.5 35.9 ± 1.0122.3 ± 0.58
± 1.45 6.5 ± 0.06 0.8 ± 0.48
14.0 ± 0.40 ± 0.13
0.9 ± 0.051.01.3 ± 0.15
Chemical composition from the location #C
%at. 55.8 ± 1.12 35.9 ± 1.01 6.5 ± 0.06 0.8 ± 0.48 1.0 ± 0.13
Materials
Materials2022,
2022,15,
15,xxFOR
FORPEER
PEERREVIEW
REVIEW 10
10ofof1919
Materials2022,
Materials 2022,15,
15,2137
x FOR PEER REVIEW 1010ofof19
19

(a)
(a) (b)
(b) (c)
(c)
(a) (b) (c)

Figure
Figure13.
13.Fragment ofofthe surface ininfields
fields(a)
(a)A,
A,(b)
(b)BB
Band
and(c)
(c)CC
Cmarked
markedininFigure
Figure12.
Figure 13. Fragment
Figure 13. Fragment of the
Fragment the surface
surface in fields (a) A, (b) and (c) marked in Figure 12.
12.

(a)
(a) (b)
(b) (c)
(c)
(a) (b) (c)

Figure 14. EDS analysis: (a) in area 1,1,(b)

(b)inin
inarea
area222and
and(c)
(c)in
inarea
area333marked
markedin
inFigure
Figure12.
Figure
Figure 14. EDS analysis:
14. EDS analysis: (a) in area 1, (b) area and (c) in area marked in Figure
Figure 12.
12.
12.

(a)
(a) (b)
(b)
(b)
Figure
Figure15.
Figure 15.SEM-BSE
15. SEM-BSE
SEM-BSE
SEM-BSE images
imagesofofTi–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
images Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Nialloy
alloyafter
of Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
alloy aftercyclic
alloy
after after oxidation
cyclic
cyclic cyclic
oxidation
oxidation at
at(a)
at (a)925
oxidation
(a) at
925
925
°C;
°C;
(a) (b)
°C;925
(b)950
◦
(b) 950
950 °C (surface
◦
°C (surface
°C
C; (b) prepared
prepared
950 C (surface by grinding)
by grinding)
prepared [26].
[26]. [26].
by grinding)

The
Thebanded
The banded
banded contrast
The banded
contrastbetween
contrast
between the
theproduct
between the layer
product layerand
product the
themetallic
andlayer and the
metallic
metallic substrate
metallicisisissubstrate
substrate
substrate an
anincentive
an to
is an
incentive
incentive tocheck
to incen-
check
check
whether
tive
whether
whether the
thechemical
to check
the chemical
chemical composition
whethercomposition ofofthe
the chemical different
differentsublayers
thecompositionsublayers
sublayers was
of the
was
was affected
affectedby
different
affected bythethetemperature
sublayers
by the increase.
was affected
temperature
temperature by
increase.
increase.
ToTothis
To
the this
thisend,
end,microanalyses
end, microanalyses
microanalyses
temperature were
increase. Toperformed
were performed
this atatmultiple
multiplepoints
end, microanalyses points
pointsin
inthe
in
were the
the layer
layer(Figure
layer
performed (Figureat 16,
(Figure 16,Table
16, Table3).
Table
multiple 3).
3).
points in
the layer (Figure 16, Table 3).
Table
Table3.3.
Table 3.EDS
EDSanalysis
EDS analysisresults
resultsininpoints
pointsmarked
markedin
inFigure
Figure15
15(mean
15 (meanvalues
(mean valuesand
values andstandard
and standarddeviations).
standard deviations).
deviations).
Table 3. EDS analysis
Elementresults in points
Element Ti marked inAl
Ti Figure
Al 15 (mean
Nbvalues andCr
Nb standard
Cr deviations).
Ni
Ni
Element Ti Al Nb Cr Ni
Element %mass
%massTi 92.8
92.8 ±±2.96
2.96Al7.2 0.48 Nb ---
7.2 ±±0.48 Cr --- Ni---
Chemical
Chemicalcomposition
Chemical compositionfrom
composition fromthe
from thelocation
the location
location#1
#1
Chemical composition from the %mass %at. 87.9 ± 3.01
87.9 ± 3.01
92.8 ± 2.96
%at. 12.1 ± 1.01
12.1 ± 1.01 -
7.2 ± 0.48 --- -
- -- ----
location #1 %at. 87.9 ± 3.01
%mass
%mass 39.1 12.1 ± 55.2
39.1±±1.21
1.21 55.2±±1.23
1.01 1.23 -5.7 5.7±±±0.7
5.7 0.7
0.7 - --- ---
Chemical
Chemical
Chemicalcomposition
Chemical composition
from
compositionfrom
from
from the
thelocation
the location
location
%mass#2
#2 39.1 ± 1.21 55.2 ± 70.0
1.23 ± 1.76 5.7 ± 0.7 - -
composition the %at.
%at. 27.9 ± 1.03
27.9 ± 1.03 70.0 ± 1.76 2.1 2.1 ± 0.27
2.1 ±± 0.27
0.27 -- ----
location #2 %at. 27.9 ± 1.03 70.0 ± 1.76 2.1 ± 0.27 - -
%mass
%mass 61.3
61.3±±1.71
1.71 24.9
24.9±±0.74
0.74 13.8
13.8 ±±±0.56
0.56 - ---
Chemical
Chemicalcomposition
Chemical composition from
compositionfrom the %mass
thelocation
the
from location#3 61.3 ± 1.71 24.9 ± 0.74 13.8 ±13.8
0.56 0.56 - -- -
Chemical composition from the location #3 %at. 54.4
54.4±±±1.56
location #3 %at. %at.
54.4 ± 1.56
%at. 54.4 39.3 ± 39.3
1.56
1.56 39.3±±±0.83
39.3
0.83 0.83
6.3 ±
0.83 6.3
6.3
6.3±±±0.23
0.23 0.23
0.23 - --- ----
Chemical composition from the %mass %mass
%mass
%mass 29.6
29.6±±±0.98
29.6 ± 0.98
29.6 27.4 ± 27.4
0.98
0.98 27.4±±±0.85
0.85
27.4 37.237.2
0.85
0.85 ±37.2
37.2 ±±±0.81
0.81 0.81 2.32.3
0.81 ± ±0.28
2.3
2.3 ±±0.28
0.28 3.5
0.28 3.5 ±±±0.33
3.5± 0.33
0.33
0.33
Chemical
Chemicalcomposition
Chemical composition
composition from
fromthe
thelocation
%at.#4
location #4
location #4 from the location #4 28.9
%at.
%at. ± 0.99
28.9 ± 47.5
0.99
28.9 ±± 0.99 ± 1.13
47.5 ± 1.1318.7 ±
18.70.53
± 0.53 2.1
2.1± 0.29
± 0.29 2.8
2.8 ±
± 0.21
0.21
%at. 28.9 0.99 47.5
47.5 ±± 1.13
1.13 18.718.7 ±± 0.53
0.53 2.1 2.1 ±± 0.29
0.29 2.8
2.8±±0.21 0.21
Chemical composition from the %mass 49.3 ± 1.15 32.3 ± 0.89 17.8 ± 0.54 0.6 ± 0.05 -
%mass
%mass
%mass 49.3
49.3± ±1.15
1.15
49.3 ± 1.15 32.3
32.3±± 0.89
0.89
32.3 ± 0.89 17.8
17.8 ± ± 0.54
0.54 0.6
0.6 ±
± 0.05
0.05 - -
Chemical
Chemicalcomposition
Chemical location #5 from
composition
composition fromthe
from thelocation
the %at.#5
location
location #5
#5 42.3 ± 0.98 49.3 ± 1.02 7.9 ±17.8
0.61± 0.54 0.50.6 ± 0.05
± 0.02 --
%mas %at.
%at.
%at. 42.3
42.3±±±0.98
42.3
45.6 ± 1.03 33.2 ± 49.3
0.98
0.98 49.3±±±1.02
49.3
0.76 1.02
1.02 7.9
20.3 ±7.9
7.9±±±0.61
0.44 0.61 0.90.5
0.61 ± ±0.05
0.5
0.5 ±±0.02
0.02
0.02 ----
Chemical composition from the
location #6 %at. %mas
%mas 45.6
45.6±±±1.03
39.4 ± 0.78
%mas 45.6 50.8 ± 33.2
1.03
1.03 33.2±±±0.76
33.2
1.02 0.76
9.0 ±
0.76 20.3
20.3 ±±±0.44
0.32
20.3 0.44 0.70.9
0.44 ± ±0.03
0.9
0.9 ±±0.05
0.05
0.05 ---
Chemical
Chemicalcomposition
Chemical compositionfrom
composition fromthe
from thelocation
the location#6
location #6
#6 %at.
%at.
%at. 39.4
39.4± ±0.78
0.78
39.4 ± 0.78 50.8
50.8±± 1.02
1.02 9.0
9.0 ±± 0.32
0.32
50.8 ± 1.02 9.0 ± 0.32 0.7 ± 0.03 0.7
0.7 ±
± 0.03
0.03 ---
 Materials 2022, 15, x FOR PEER REVIEW 11 of 19
als 2022, 15,Materials
x FOR PEER
2022, REVIEW
15, 2137 11 of 19 11 of 19

(a) (b) (c)

(a) (b) (c)

(d) (e) (f)

(d) (e) (f)

Figure 16. EDS analysis: (a) in point 1, (b) in point 2, (c) in point 3, (d) in point 4, (e) in point 5 and
Figure 16. EDS Figure
(f) EDS
16.(a)
analysis:
in point 6 in analysis:
point
marked 1, (a)ininpoint
in (b)
Figure point
15. 2,1,(c)
(b)ininpoint
point3,2,(d)
(c)ininpoint
point4,3,(e)
(d)ininpoint
point54,and
(e) in point 5 and
(f) in point 6 marked in Figure
(f) in point 15. in Figure 15.
6 marked
Analysis of columnar crystals (Figure 16a) indicates that these are rutile crystals. The
Analysis next Analysis
of columnar of columnar
crystals
dark-contrasting (Figure
oxide crystals (Figure
16a) indicates
sublayer 16a)
contains indicates
thatthethese
mostareAl,that
rutile these
which are rutile
crystals.
may The crystals.
suggest that AlThe2O3
next
next dark-contrasting dark-contrasting
oxide sublayer oxide sublayer
contains the contains
most Al, the
which mostmay
is the dominant phase (Figure 16b). However, the presence of titanium (which indicates Al, which
suggest may
that suggest
Al 2 O 3 that Al 2 O3
is the dominantis the
phasedominant
(Figure phase
16b). (Figure
However, 16b).
the However,
presence the
of presence
titanium
that it is bonded in TiO2) manifested, among others, by bright, very fine nanoparticles of
(whichtitanium
indicates(which indicates
that
that it is bonded
means it is
in TiObonded
that 2)Al in TiO )among
manifested,
2O3 is not2 the
manifested, phaseamong
onlyothers, others,
by bright,
component ofby
very bright,
fine
this very The
nanoparticles
sublayer. fine third
nanoparticles
sublayer
means that Almeans O is thatthe
not Al2only
O3 isphase
not the only phaseofcomponent
component this sublayer.of this
extending down to the metallic substrate contrasts much brighter, although locally, the
2 3 The sublayer.
third The third sublayer
sublayer
extending down extending down tosubstrate
to theismetallic the metallic substrate
contrasts much contrasts
brighter, much brighter, although
although locally, the
contrast deeper. Microanalysis in point 3 (according to Figurelocally, the in
15) selected this sub-
contrast
contrast is deeper. is deeper.
Microanalysis Microanalysis
in point 3 in point
(according 3 (according
to Figure to
15) Figure
selected 15)in selected
this sub- in this sublayer
layer may indicate the presence of titanium oxide, aluminium and alloying element oxides
may the
layer may indicate indicate
presencethe of presence
titanium ofoxide,
titanium oxide, aluminium and alloying element oxides
(Figure 16c). Similarly, a bright bandaluminium
enriched with andNb,alloying
Cr and element
Ni wasoxides
observed at a lower
(Figure
(Figure 16c). Similarly, 16c). Similarly,
a bright band a bright
enriched band enriched
with Nb, of Crthe with Nb,
andchemical Cr
Ni was observedand Ni was observed at a lower
at a lower
temperature (Figure 16d). Microanalysis composition in different areas of
temperature
temperature (Figure (Figure 16d). Microanalysis
16d). substrate
Microanalysis of the ofcomposition
the chemicalin composition in different areas of
the metallic is shown inchemical
Figure 16e,f. different areas of
the
the metallic substratemetallic substrate is shown in Figure 16e,f.
As isin shown
the study in Figure 16e,f.
of the structure and composition of the products obtained when oxi-
As in the dation
study Asofinthe the study ofand
structure thecomposition
structure and compositionobtained
of◦the of the products obtained when
was carried out at 875 °C◦to 925 °C, the products
cross-sections of the when
scale,oxi-the formation of
oxidation
dation was carried was carried out at
°C to 925 additional 875 C to 925
°C, the cross-sections C, the cross-sections of the scale, of the formation
whichout didatnot875introduce interactionsofonthe thescale,
layer, the formation
were observed and analysed
of
which did notunder which
introduce did not introduce
additional additional
interactions interactions on the layer, were observed and analysed
the studied conditions (975 on◦°C).
the layer,
The were observed
cross-sectional areasandof analysed
the oxide scale formed
under
under the studied the studied
conditions conditions
(975 °C). The (975 C).
cross-sectional The cross-sectional
areas of the areas of
oxidechipping the
scale formedoxide scale formed
on the
on the alloy after
alloy after grinding
grindingshow showaathin, thin,uniform
uniformlayer layer(due(duetoto chipping during
during cooling)
cooling)of
on the alloy after
about grinding
6 μm show a thin,
thickness, with uniform
a running layer (dueband
bright to chipping
enriched during
in Nb cooling)
(Figure of Stress build-
17).
of about 6 µm thickness, with a running bright band enriched in Nb (Figure 17). Stress
about 6 μm thickness,
up in thewith a running to bright band enriched in Nb (Figure 17). Stress build-
build-up inscale leads leads
the scale thetoinitiation
the initiationof microcracks,
of microcracks, which promotes
which promotes the formation
the formationof a
up in the scalemultilayer
leads to the initiation
scale. The of
extensionmicrocracks,
of which
processing promotes
time causes the
the formation
formation of successive
of a sub-
of a multilayer scale. The extension of processing time causes the formation of successive
multilayer scale. The which
layers, extension of processing
is shown by the time causes
photos taken the formation
during of successive
microanalyses sub-
of cross-sections of the
sublayers, which is shown by the photos taken during microanalyses of cross-sections of
layers, which oxide
is shown by the photos taken during microanalyses ofhere,
cross-sections ofbetween
the
the oxide layers formed. As the oxidation time increases here, the contact betweenscale
layers formed. As the oxidation time increases the contact the the
oxide layers formed.
and the As the oxidation time increases here,andthe thecontact between of thethescale
scale andmetallic
the metallicsubstrate
substrateis locally broken
is locally broken and construction
the construction second
of the secondlayer be-
layer
and the metallic substrate is locally broken and the construction of the second layer be-
gins. The
begins. The sites
sitesofoffragmentary
fragmentaryspalling spallingofofthe thescale
scaleafter
afteroxidation
oxidation at 975 °C
at 975 ◦ C are shown in
are shown in
gins. The sitesFigures
of fragmentary
18 and 19.spalling of the scale after oxidation at 975 °C are shown in
Figures 18 and 19.
Figures 18 and 19.

Figure 17.
Figure 17. SEM-BSE
SEM-BSE images
images of
of Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Nialloy
alloyafter
aftercyclic
cyclicoxidation
oxidationat 975◦°C
at975 C
Figure 17. SEM-BSE images
(surface of Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
prepared by grinding). alloy after cyclic oxidation at 975 °C
(surface prepared by grinding).
(surface prepared by grinding).
Materials
Materials 2022,
2022, 15,
15, x2137
FOR PEER REVIEW 1212ofof 19
19

(a) (b)

Figure
Figure 18.
18.The
Thecross-sectional
cross-sectionalareas ofof
areas thethe
scale formed
scale on on
formed the the
alloy oxidised
alloy at °C
at 975
oxidised 975(surface pre-
◦ C (surface
pared by grinding): (a) high-magnification, (b) low-magnification.
prepared by grinding): (a) high-magnification, (b) low-magnification.

(a) (b)

Figure
Figure 19.
19. SEM-BSE
SEM-BSE images
images of
ofTi–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni alloy
alloy after
after cyclic
cyclic oxidation
oxidation at
at 975
975 °C
◦C
(surface prepared by grinding): (a) low-magnification, (b) high-magnification.
(surface prepared by grinding): (a) low-magnification, (b) high-magnification.

The
The way
way of surface preparation,
of surface preparation,especially
especiallyhigher
higherroughness,
roughness,definitely
definitely hashas a benefi-
a beneficial
cial
effect on the kinetics of cyclic oxidation, causing an increase in the cohesion of theof
effect on the kinetics of cyclic oxidation, causing an increase in the cohesion the
oxide
oxide
with thewith the metallic
metallic substrate.
substrate. This This is because
is because the diffusion
the diffusion processes
processes occurring
occurring at theat the
flat
flat product–oxidation
product–oxidation phasephase interface
interface cause cause coalescence
coalescence of vacancies,
of vacancies, formation
formation of micro-
of micro-voids
voids andmerging
and their their merging into bands
into bands (Figure(Figure
19). The 19). The structure
structure of the product
of the product layer onlayer on the
the surface
surface with Ra = 5.8 μm roughness is different. This is because the
with Ra = 5.8 µm roughness is different. This is because the layer has an uneven thickness, layer has an uneven
thickness, and the processes
and the diffusion diffusion processes in the direction
in the direction of decreasing
of decreasing concentrations
concentrations do not
do not produce
produce voids20a).
voids (Figure (Figure
The20a).
voidsThe do voids
not formdo not form probably
probably because the because
outer the outerofsurface
surface the oxideof
the oxide
is not is nottoparallel
parallel to theofsurface
the surface the phaseof the phase interface,
interface, and on the and on the microscopic
microscopic scale, the scale,
shear
the shear
stress stress
at each at each
point point
of the phase of boundary
the phase liesboundary lies inplane,
in a different a different
whichplane, which
does not does
promote
stress
not accumulation.
promote However, it should
stress accumulation. However, be mentioned
it should that the chemical
be mentioned thatcomposition
the chemical of
the oxide sublayers
composition formed
of the oxide does not
sublayers change.
formed doesAnot cross-section of the resulting
change. A cross-section of theproducts
result-
of the
ing alloy whose
products surface
of the alloy was surface
whose prepared was using abrasive
prepared usingblasting is shown
abrasive blasting inisFigure
shown20b.in
There are
Figure 20b.visible
There irregularities produced produced
are visible irregularities by the sample
by thesurface
samplepreparation by blasting.
surface preparation by
On the surface
blasting. On theofsurface
the alloy
of prepared
the alloy by abrasive
prepared byblasting,
abrasiveno verticalno
blasting, columnar
verticalrutile
columnarTiO2
crystallites
rutile are formed.are
TiO22 crystallites This is due to
formed. lower
This stresses,
is due which
to lower prevent
stresses, the rutile
which fromthe
prevent growing
rutile
vertically.
from growing Thevertically.
lower stresses present
The lower here present
stresses result inhere
lessresult
rutileingrowth compared
less rutile growthto the
com-
stresses present on the ground surface.
pared to the stresses present on the ground surface.
Materials 2022,
Materials 2022, 15, xx FOR
FOR PEER REVIEW 13 of
of 19
Materials 2022, 15,
15, 2137 PEER REVIEW 13
13 of 19
19

(a)
(a) (b)
(b)

Figure 20.
Figure 20. SEM-BSE images
images of Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni alloy afterafter cyclic oxidation
oxidation at 975
975 ◦°C
Figure 20. SEM-BSE
SEM-BSE images of
of Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni alloy
alloy after cyclic
cyclic oxidation at
at 975 °CC
(surface
(surface prepared
prepared by
by abrasive
abrasive blasting):
blasting): (a)
(a) high-magnification,
high-magnification, (b)
(b) low-magnification.
low-magnification.
(surface prepared by abrasive blasting): (a) high-magnification, (b) low-magnification.
XRD
XRD phase
XRD phase analysis
phaseanalysis results
analysisresults forfor
for
results Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni after
after oxidation
oxidation
after at 875
at
oxidation 875
at
°C,
°C, 900
◦
875 900 °C
C, 900 and
°C and
◦ 950 °C
950950
C and are
°C are
◦ C arepresented
presented
presentedin Figures
in in
Figures
Figures 21–23.
21–23. The
21–23.The nucleating
Thenucleating phase
nucleatingphase components
phasecomponents
components
are
are rutile
are rutile TiO and
TiO222and
rutileTiO Al
andAlAl2O3. Apart from these, there are also nitrides TiN, Ti2AlN and a me-
2O 3. 3Apart from these, there are are
alsoalso
nitrides TiN,TiN,
Ti2AlN and aand
me-a
2O . Apart from these, there nitrides Ti2 AlN
tallic
tallic substrate
substrate
metallic phase
phase
substrate (γ-TiAl
(γ-TiAl
phase phase).
phase).
(γ-TiAl To obtain
To obtain
phase). information
information
To obtain about the
about the
information phase
phase
about at the
theatphasemetallic
the metallic
at the
substrate,
substrate, the scale
the scalethe
metallic substrate, wasscale
was removed
removed from the
from
was removed thefrom
samples
samples (Figure 21).
(Figure
the samples 21).
(Figure 21).

Figure 21. XRD pattern showing the predominant phases formed during oxidation of Ti–46Al–7Nb–
Figure 21. XRD
XRD pattern
pattern
0.7Cr–0.1Si–0.2Ni
Figure 21. showing the
at 875 ◦ Cthe
alloyshowing predominant
(scale phases
was removed
predominant formed
from
phases during oxidation
the samples).
formed during oxidation of
of Ti–46Al–7Nb–
Ti–46Al–7Nb–
0.7Cr–0.1Si–0.2Ni alloy at 875 °C (scale was removed from the samples).
0.7Cr–0.1Si–0.2Ni alloy at 875 °C (scale was removed from the samples).
Materials 2022, 15, x FOR PEER REVIEW 14 of 19
Materials 2022, 15, x FOR PEER REVIEW 14 of 19
Materials 2022, 15, 2137 14 of 19

Figure 22. XRD pattern showing the predominant phases formed during oxidation of Ti–46Al–7Nb–
Figure22.
Figure XRDpattern
22.XRD patternshowing
showing the predominant phases formed during oxidation of Ti–46Al–7Nb–
0.7Cr–0.1Si–0.2Ni alloy at 900 °C.the predominant phases formed during oxidation of Ti–46Al–7Nb–
0.7Cr–0.1Si–0.2Ni alloy at ◦ C.
900 °C.
0.7Cr–0.1Si–0.2Ni alloy at 900

Figure23.
Figure XRDpattern
23.XRD patternshowing
showingthethepredominant
predominantphases
phasesformed
formedduring
duringoxidation
oxidationofofTi–46Al–7Nb–
Ti–46Al–7Nb–
Figure 23. XRD pattern
0.7Cr–0.1Si–0.2Ni alloyshowing
at 950 ◦ C.
the predominant phases formed during oxidation of Ti–46Al–7Nb–
0.7Cr–0.1Si–0.2Ni alloy at 950 °C.
0.7Cr–0.1Si–0.2Ni alloy at 950 °C.
4. Discussion
4. Discussion
In order to determine the range of resistance of the Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy
4. Discussion
In order to determine the range of resistance of the Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni al-
to oxidation
orderin toair, tests were carried outresistance
on cyclic of oxidation of the alloy in the temperature
loy toInoxidation determine
◦ C toin air,◦teststhe range
were of
carried out on cyclic theoxidation
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
of the alloy in the tem- al-
range
loy to of 875
oxidation in 975
air, C.
testsIt was
were found that
carried with
out on traditional
cyclic surface
oxidation preparation
of the alloy bythe
in grinding
tem-
perature range of 875 °C to 975 °C. It was found that with traditional surface preparation
to a low range
perature Ra value875 (Ra°C ≈ 0.06 µm °C.in the study), visible scale spalling occurs 950 ◦ C and
atpreparation
by grinding to aoflow Ra valueto 975 (Ra ≈It0.06
was μmfoundin thethat with
study), traditional
visible scale surface
spalling occurs at
above.
by This is
grinding to because
a low Rathe diffusion
value (Ra ≈ processes
0.06 μm occurring
in the at the
study), flat product–oxidation
visible scale spalling phase
occurs at
950 °C and above. This is because the diffusion processes occurring at the flat product–
interface
950 °C andlead to
above. the formation
This is becauseof microcavities
the diffusion and merging
processes into
occurring bands.
at theAnflat increase
product– in
oxidation
surface phase interface
roughness up to lead
Ra ≈to6 the
µm formation
increases of
the microcavities
mass gain and merging
during oxidation into at bands.
900 ◦C
oxidation
An increase phase interface lead toup thetoformation
Ra ≈ 6 μm of microcavities andgain
merging into bands.
(this is due into surface
a largerroughness
surface development, whichincreases the mass
is not included in theduring oxidation
macroscopically
An increase in
900 °C (thistotal
atdetermined surface
is due roughness
to a larger up to Ra ≈ 6 μm increases the mass gain during oxidation
surface area ofsurface development,
the oxidised sample), which
while atishigher
not included in the macro-
temperatures (925 ◦ C,
at 900 °C
scopically (this is
determined due to a
total larger surface
surface area development,
of the oxidised which is
sample), not included
while at in
higher the macro-
temper-
975 ◦ C), it strongly reduces the processes of scale spalling. This is because higher surface
scopically
atures (925determined
°C, 975 °C),the total surface area ofthe
it strongly theprocesses
oxidised sample), while at This
higher temper-
roughness increases cohesionreduces
of the metallic of scale
substrate withspalling.
the oxide and is because
causes a
atures
higher (925 °C,
surface 975 °C), it strongly
roughness increases reduces the processes of scale spalling. This is because
reduction in compressive stresses inthe thecohesion
layer. This of is
the metallic
because thesubstrate
layer haswith the
a non-uniformoxide
higher
and surface roughness increases the cohesion of the metallic substrate with the oxide
thickness and the diffusion processes that take place do not produce voids. Studies onhas
causes a reduction in compressive stresses in the layer. This is because the layer the
and causes a reduction in compressive stresses in the layer. This is◦ because the layer has
effect of high cooling rates (water) and very low cooling rates (1 C/min) are presented
Materials 2022, 15, x FOR PEER REVIEW 15 of 19
Materials 2022, 15, x FOR PEER REVIEW 15 of 19

Materials 2022, 15, 2137

a non-uniform thickness and the diffusion processes that take place do not produce15 of 19
voids.
a non-uniform thickness and the diffusion processes that take place do not produce voids.
Studies on the effect of high cooling rates (water) and very low cooling rates (1 °C/min)
Studies on the effect of high cooling rates (water) and very low cooling rates (1 °C/min)
are presented in [2] and [13], respectively. At low cooling rates, less spalling of the scale
are presented in [2] and [13], respectively. At low cooling rates, less spalling of the scale
in [2] and
would [13], respectively.
be expected due to its At low cooling
deformation rates, at
by creep less
high spalling of the scale
temperatures, would
but the be
experi-
would be expected due to its deformation by creep at high temperatures, but the experi-
expected due to indicate
mental results its deformation by creep
that spalling at high temperatures,
increases. The reason for but theisexperimental
this results
not clear, although
mental results indicate that spalling increases. The reason for this is not clear, although
indicate that spalling
the micro-crack formed increases.
at high The reason
cooling ratesfororiented
this is not clear, although
perpendicular the alloy
to the micro-crack
surface
the micro-crack formed at high cooling rates oriented perpendicular to the alloy surface
formed
may increase the resistance to spalling. On the other hand, low cooling rates facilitatethe
at high cooling rates oriented perpendicular to the alloy surface may increase lat-
may increase the resistance to spalling. On the other hand, low cooling rates facilitate lat-
resistance
eral growth to strain,
spalling. On the
which other hand,
increases low cooling
compressive rates
stresses in facilitate
the oxidelateral growth
layer [25]. Thestrain,
study
eral growth
which strain,
increases which increases compressive
compressive stresses in[25].
the oxide layer of [25].
theThe study
of the effect of the cooling ratestresses in the oxide
on the oxidation layer
of the The study
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni effect of
alloy
of
thethe effect
cooling of the cooling rate on the oxidation of the Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy
proved thatrate on the oxidation
increasing the cooling ofrate
the does
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
not affect the acceleration alloy proved that
or deceleration of
proved that
increasing theincreasing
cooling thedoes
rate cooling
not rate
affect does not affect theoracceleration
theobservation
acceleration deceleration or deceleration of
the oxidation rate. This is confirmed by the of Yoshihara andofKim the [2],
oxidation
as two
the oxidation
rate. This rate. Thisbyisthe confirmed by the observation of Kim
Yoshihara and ofKim the[2],
sixas two
of the six is confirmed
alloys they tested (oxidisedobservation of Yoshihara
cyclically and
at 875 °C), namely,[2],the
as two
Ti-45Al-7.5Nb-0.3B-alloys
of the
they six alloys they tested (oxidised cyclically
◦ at 875 °C), namely, the Ti-45Al-7.5Nb-0.3B-
0.1Ytested (oxidised
alloy plus 0.2% each cyclically
(W, Hf, at Zr,
875C C), andnamely,
O) and the Ti-45Al-7.5Nb-0.3B-0.1Y alloy,
the Ti-47Al-2Cr-0.8Nb-0.2O alloy plus
were
0.1Y alloy
0.2% each plusHf,
(W, 0.2% Zr,each
C and (W, O) Hf, Zr,the
and C and O) and the Ti-47Al-2Cr-0.8Nb-0.2O
Ti-47Al-2Cr-0.8Nb-0.2O alloy, were not alloy, were
sensitive to
not sensitive to changes in the cooling rate. However, it should be noted that in their work
not sensitive
changes in the tocooling
changes in the
rate. coolingitrate.
However, should However,
be noted itthat
shouldin be noted
their work that
[2], in their
200 work
oxidation
[2], 200 oxidation cycles were performed at the analysed temperature, while only 60 cycles
[2], 200were
cycles oxidation
performed cycles were performed at the analysed temperature, while only 60 cycles
were performed in the at the analysed
present work. temperature, while only 60 cycles were performed
were performed
in theThe
present in the present work.
nextwork.
stage of the research was the analysis of corrosion products and diffusion
The next
The next stage
stage of of the
the research
research waswas the the analysis
analysis of of corrosion
corrosion products
products and and diffusion
diffusion
processes occurring in the oxide layer and in the metallic substrate. The analysis of the
processes occurring in the oxide layer and in the metallic
processes occurring in the oxide layer and in the metallic substrate. The analysis of substrate. The analysis of the
the
obtained results allows us to conclude that the oxidation of the tested alloy results in the
obtained results allows us to conclude that the oxidation of
obtained results allows us to conclude that the oxidation of the tested alloy results in thethe tested alloy results in the
formation of scale and transformations in the metallic substrate. They are caused by the
formation of scale
formation scale and and transformations
transformations in the the metallic
metallic substrate.
substrate. They are caused by the the
out-core diffusion of alloying elements and the formation of phases and solid solutions as
out-core diffusion of alloying
out-core alloyingelements
elementsand andthe theformation
formationofofphases phases andandsolid
solid solutions
solutionsas
a result of the in-core diffusion of nitrogen and oxygen. Based on the investigations per-
a result
as a resultof the
of the in-core
in-corediffusion
diffusion of nitrogen
of nitrogen andand oxygen.
oxygen. Based on the
Based on investigations
the investigations per-
formed, the most general conclusion is that the oxidation products consist of three sub-
formed, the the
performed, most most general
general conclusion
conclusion is that the oxidation
is that the oxidation products consist
products of three
consist sub-
of three
layers, and a Nb-rich band is formed in the metallic substrate (Figures 24 and 25).
layers, and a Nb-rich band is formed in the metallic substrate
sublayers, and a Nb-rich band is formed in the metallic substrate (Figures 24 and 25). (Figures 24 and 25).

oxidation
productsofofoxidation
TiO outer layer
TiO22 outer layer

Al O + (TiO ) sublayer rich in Al O

Al22O33 + (TiO22) sublayer rich in Al22O33
products

TiO + Al O basic sublayer

TiO22+ Al22O33 basic sublayer

Nb (Cr, Ni) Nb rich region

Nb (Cr, Ni) Nb rich region
substrate
substrate

TiAlNb (γ+α ) metallic substrate

TiAlNb (γ+α22) metallic substrate

Figure 24.General
General structureof
of scaleafter
after oxidationof
of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Nialloy.
alloy.
Figure 24. General structure
Figure 24. structure of scale
scale after oxidation
oxidation of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy.

Figure 25. SEM-BSE images of Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni alloy after cyclic oxidation.

Figure 25.
Figure 25. SEM-BSE images
images of
of Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Ni
Ti–46Al–7Nb–0.7Cr–0.1Si–0.2Nialloy
alloyafter
aftercyclic
cyclicoxidation.
oxidation.
Materials 2022, 15, 2137 16 of 19

During the oxidation of the investigated alloy, it is not possible for a continuous
protective layer of Al2 O3 to form because the outer oxide layer always consists of TiO2 .
The reason for this is the main alloying element, titanium, which does not provide an
opportunity for a protective layer to form on the basis of aluminium oxides. A characteristic
feature of titanium is that it can have both cation and anion sublattice defected, depending
on the oxygen partial pressure.
The formation of the outer rutile layer causes a change in the balance of the Ti and
Al quantitative fraction and, as a result of the slower diffusion of this element, a mixed
Al2 O3 -rich sublayer is formed under the TiO2 layer. The out-core diffusing titanium cations
mostly migrate towards the surface. The resulting Al2 O3 layer is a heterogeneous layer and
it is not compact. It also includes, although in smaller amounts, rutile TiO2 . The presence of
rutile determines the further course of the oxidation process. Rutile has a defective anionic
sublattice where the oxygen in-core diffusion takes place. Oxygen diffuses toward the
product–metallic substrate interface, causing an increase in the thickness of the oxide layer
due to the formation of titanium and aluminium oxides in this layer. The occurrence of
defects in the rutile depends on the partial pressure of oxygen. At higher temperatures,
the in-core diffusion of oxygen (through the doubly and singly ionised oxygen vacancies
of the rutile) and the out-core diffusion of the titanium, aluminium and alloying elements
occur. This produces a third mixed sublayer extending from the Al2 O3 -rich sublayer to the
metallic substrate.
In the region of the metallic substrate of the investigated alloy, diffusion processes
also take place in both oxidising atmospheres used and, therefore, a brightly contrasting,
interrupted, niobium-rich micro-band forms in the metallic substrate in BSE. Niobium does
not diffuse in-core simultaneously with Al and Ti, but remains in the metallic substrate,
which relatively increases its concentration. Unfortunately, it was not possible to determine
by X-ray methods in which compounds this element occurs in this band. A similar problem
was observed in other works. Kim and Yoshihara [2] and Roy [31] did not identify Nb-
containing oxides in the resulting products, despite the quite diverse content of this element
in the alloys studied (2–15 at.%).
Through the product layer, nitrogen and oxygen diffuse into the metallic substrate.
Oxygen is the main component of the formed products, while nitrogen does not form
nitrides in it. It migrates toward the metallic substrate and combines mainly in Ti2 AlN and
TiN, and dissolves together with oxygen in the αTi(Al) solid solution. The possibility of
forming such solid solutions is due to the equilibrium systems of Ti-O and Ti-N [32].
The influence of Nb on increasing the oxidation resistance of the investigated alloy
is undeniable. Performed studies on the oxidation of Ti-46Al-7Nb alloys in relation to
literature data [24] on the oxidation of Ti-48Al-2Cr-2Nb alloys show that the alloy with
7 at.% Nb content is characterised by higher resistance to high-temperature oxidation than
the alloy containing 2 at.% Nb, which is manifested by a reduction in the oxidation rate and
smaller mass gains during oxidation tests. To clarify, the effects of niobium on the structure
and layer of oxidation products were analysed. In the analysed alloy, the products include
niobium and other elements of the metallic substrate diversely distributed in the different
sublayers. In the columnar sublayer of rutile crystallites, there are no elements such as Nb,
Cr or Ni. In the dark band of the Al2 O3 -rich sublayer, Nb was observed in a much lower
amount than in the next TiO2 -rich sublayer. In the sublayer with balanced amounts of TiO2
and Al2 O3 , the Nb content is comparable to its local concentration in the metallic substrate.
A definite increase in niobium content and occurrence of Cr and Ni was observed in the
bright band, which is formed during oxidation, independently of the process temperature,
oxidising atmosphere and other factors studied in this work (surface roughness, cooling
rate). It was found that such a distribution of niobium in a narrow micro-band inhibits
the diffusion processes through the sublayer closest to the metallic substrate. Why does
this happen? Both allotropic varieties of titanium form terminal solid solutions with O and
N [32]. The solubility of oxygen in α-Ti is about 34%, while that of nitrogen is about 23 at.%
On the other hand, about 4% of oxygen and 2 at.% nitrogen can be dissolved in the β-phase.
Materials 2022, 15, 2137 17 of 19

With the increase in aluminium content in these alloys, α2 -TiAl and γ-TiAl phases are
formed. In the γ-TiAl phase, oxygen dissolves in a very small amount, while nitrogen does
not dissolve at all. The α2 -TiAl phase is characterised by much higher solubility of both
elements. The bidirectional diffusion that occurs during the oxidation of these alloys makes
rutile present both at the product–metal interface and at the product–oxidant interface. Due
to the high chemical activity of aluminium, in nanometre-sized regions, quite a significant
part of Al will undergo selective oxidation and there will be a decrease in Al concentration
in the metallic part of the alloy. This leads to a local disappearance of the γ-TiAl phase
and an increase in the proportion of the α2 -TiAl phase, and the substrate is enriched in
nitrogen and oxygen. Al, O and N are alpha-forming elements and exhibit favourable
solubility in α-TiAl. Nb is a beta-forming element with respect to titanium (it extends
the occurrence of the β-Ti phase) and therefore, by forming the β-Ti phase, it limits the
solubility of nitrogen and oxygen [33–36]. Nb in the alloy, concentrated in the substrate near
the phase interface, limits the possibility of α-Ti(Al,N,O) solid solution formation and, as a
beta-forming element relative to titanium, promotes the formation of a β-Ti(Al,N,O) phase
with low O and N solubility. This limits the process of in-core interstitial element diffusion,
which slows down the growth of oxidation layers. This is evidenced by the smaller mass
increase in the tested alloy with respect to the Ti-48Al-2Cr-2Nb alloy during oxidation.
In addition to the mentioned beta-forming role of Nb in alloys and the associated
lower solubility of oxygen and nitrogen [33–36], the effect of niobium on increasing heat
resistance is also related to other factors [35,37–39]:
- Niobium inhibits film growth kinetics by decreasing titanium activity;
- By reducing the activity of titanium, it increases the activity of aluminium, and
supresses the growth of rutile TiO2 ;
- Nb ions replace Ti ions, thus leading to a reduction in oxygen vacancies that reduce
oxygen diffusion; and
- Niobium hinders the mass transfer of TiO2 .
As shown in this work, better corrosion resistance is provided by higher surface
roughness and increased Nb content in the material, which, in the range of 7 at.%, reduces
the oxidation rate compared to the reference alloy containing 2 at.% Nb. Taking into account
the presented test results, it can be concluded that the addition of 7 at.% Nb causes changes
in the structure of the metallic substrate and contributes to a significant improvement in
resistance to high-temperature oxidation.

5. Conclusions
1. The addition of 7 at.% Nb causes a decrease in the oxidation rate, which is manifested
by a 30% lower mass gain and better adhesion of the product layer compared to an
alloy containing 2 at.% Nb.
2. Multiphase scale is formed, which is always characterised by a sequence of sublayers:
- Outer rutile sublayer;
- Sublayer rich in Al2 O3 and containing TiO2 in smaller amounts; and
- Sublayer containing Al2 O3 , TiO2 , and oxides of Nb, Cr and Ni.
1. In the oxidation tests, changes in the metallic substrate were observed, consisting of
Al diffusion and the dominance at the oxidation temperature of the β-Ti(Nb,Al,N,O)
phase with low amounts of dissolved oxygen and nitrogen.
2. Nb forming a discontinuous band in the metallic substrate inhibits the diffusion
processes taking place.
3. The increase in surface roughness contributes to a decrease in the oxidation rate and
inhibits oxidation product spallation processes due to a decrease in compressive stress
and an increase in cohesion between the oxide and the metallic substrate.

Funding: This research received no external funding.

Institutional Review Board Statement: Not applicable.
Materials 2022, 15, 2137 18 of 19

Informed Consent Statement: Not applicable.

Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Conflicts of Interest: The author declares that there is no conflict of interest.

References
1. Clemens, H.; Kestler, H. Processing and Applications of Intermetallics γ-TiAl—Based Alloys. Adv. Eng. Mater. 2000, 2, 551–570.
[CrossRef]
2. Yoshihara, M.; Kim, Y.W. Oxidation behavior of gamma alloys designed for high temperature applications. Intermetallics 2005, 13,
952–958. [CrossRef]
3. Bauer, P.; Laska, N.; Swadźba, R. Increasing the oxidation resistance of γ-TiAl by applying a magnetron sputtered aluminum and
silicon based coating. Intermetallics 2021, 122, 107177. [CrossRef]
4. Donchev, A.; Mengis, L.; Couret, A.; Mayer, S.; Clemens, H.; Galetz, M. Effects of tungsten alloying and fluorination on the
oxidation behavior of intermetallic titanium aluminides for aerospace applications. Intermetallics 2021, 139, 107270. [CrossRef]
5. Dai, J.; Zhu, J.; Chen, C.H.; Weng, F. High temperature oxidation behavior and research status of modifications on improving
high temperature oxidation resistance of titanium alloys and titanium aluminides: A review. J. Alloys Compd. 2016, 685, 784–798.
[CrossRef]
6. Ostrovskaya, O.; Badini, C.; Baudana, G.; Padovano, E.; Biamino, S. Thermogravimetric investigation on oxidation kinetics of
complex Ti-Al. Intermetallics 2018, 93, 244–250. [CrossRef]
7. Park, S.Y.; Seo, D.Y.; Kim, S.W.; Kim, S.E.; Hong, J.K.; Lee, D.B. High temperature oxidation of Ti-46Al-6Nb-0.5W-0.5Cr-0.3Si-0.1C
alloy. Intermetallics 2016, 74, 8–14. [CrossRef]
8. Laska, N.; Braun, R.; Knittel, S. Oxidation behavior of protective Ti-Al-Cr based coatings applied on the γ-TiAl alloys Ti-48-2-2
and TNM-B1. Surf. Coat. Technol. 2018, 349, 347–356. [CrossRef]
9. Durda, E.; Przybylski, K. Oxidation behavior of Ti-46Al-8Nb alloy with boron and carbon addition under thermal cycling
conditions. Intermetallics 2016, 71, 51–56. [CrossRef]
10. Mitoraj, M.; Godlewska, E.M. Oxidation of Ti-46Al-8Ta in air at 700 ◦ C and 800 ◦ C under thermal cycling conditions. Intermetallics
2013, 34, 112–121. [CrossRef]
11. Narayana, P.L.; Kim, J.H.; Yun, D.W. High temperature isothermal oxidation behavior of electron beam melted multi-phase γ-TiAl
alloy. Intermetallics 2022, 141, 107424. [CrossRef]
12. Smialek, J.L.; Barrett, C.A.; Scheaffer, C. Design for oxidation resistance. In Materials Selection and Design, ASM Handbook; Dieter,
G.E., Ed.; ASM International: Materials Park, OH, USA, 1997; pp. 589–602.
13. Lowell, C.E.; Deadmore, D.L. The role of thermal shock in cyclic oxidation. Oxid. Met. 1980, 14, 325–336. [CrossRef]
14. Rakowski, J.M.; Pettit, I.S.; Meier, G.H. The effect of surface preparation on the oxidation behaviour of gamma Ti-Al-base
intermetallic alloys. Scr. Mater. 1996, 35, 1417–1422. [CrossRef]
15. Pérez, P.; Jiménez, J.A.; Frommeyer, G.; Adeva, P. The influence of the alloy microstructure on the oxidation behavior of
Ti-46Al-1Cr-0.2Si alloy. Oxid. Met. 2000, 53, 99–124. [CrossRef]
16. Quadakkers, W.J.; Schaaf, P.; Zheng, N.; Gil, A.; Wallura, E. Beneficial and detrimental effects on nitrogen on the oxidation
behaviour of TiAl-based intermetalics. Mater. Corros. 1997, 48, 28–34. [CrossRef]
17. Rakowski, J.M.; Pettit, F.S.; Meier, G.M. The effect of nitrogen on the oxidation of γ-TiAl. Scr. Metall. Mater. 1995, 33, 997–1003.
[CrossRef]
18. Zeller, A.; Dettenwanger, F.; Schütze, M. Influence of water vapor on the oxidation behaviour of titanium aluminides. Intermetallics
2002, 10, 59–72. [CrossRef]
19. Brady, M.P.; Smialek, J.L.; Humphrey, D.L.; Smith, J. The role of Cr in promoting protective alumina scale formation by γ-based
Ti-Al-Cr alloys. Part II: Oxidation behavior in air. Acta Mater. 1997, 45, 2371–2382. [CrossRef]
20. Li, X.Y.; Taniguchi, S. Correlation of high temperature oxidation with tensile properties for Ti–48Al–2Cr–2Nb and Ti–48Al–2Cr–2Fe
alloys. Intermetallics 2005, 13, 683–693. [CrossRef]
21. Małecka, J.; Grzesik, W.; Hernas, A. An investigation on oxidation wear mechanisms of Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni
intermetallic-based alloys. Corossion Sci. 2010, 52, 263–272. [CrossRef]
22. Terner, M.; Biamino, S.; Pelissero, F.; Sabbadini, S.; Fino, P.; Pavese, M.; Badini, C. Oxidation behavior of Ti-47Al-2Cr-2Nb and
Ti-45Al-2Cr-8Nb produced by Electron Beam Melting. In Proceedings of the Conference: EuroPM, Volume 2: Full Density and
Alternative Consolidation, Gothenburg, Sweden, 15–18 September 2013.
23. Jang, Y.-D.; Lee, D.B.; Seo, D.Y. High temperature oxidation of thermomechanically treated Ti-47Al−2Cr−2Nb. Met. Mater. Int.
2002, 8, 503–506. [CrossRef]
24. Król, S. Effect of tantalum on the oxidation resistance of Ti3 Al + TiAl titanium intermetallic alloy. In Proceedings of the 14th
International Metallurgical & Materials Conference, Brno, Czech Republic, 24–25 May 2005.
25. Rhines, F.N.; Wolf, J.S. The role of oxide microstructure and growth stresses in the high-temperature scaling of nickel. Metall.
Mater. Trans. 1970, 62, 1701–1718. [CrossRef]
Materials 2022, 15, 2137 19 of 19

26. Małecka, J. Transformation and precipitation processes in a metal substrate of oxidized TiAl-based alloys. Oxid. Met. 2019, 91,
365–380. [CrossRef]
27. Jungling, T.; Rapp, R.A. High temperature oxidation of iron at 1200 ◦ C in hot-stage environmental scanning electron microscope.
Metall. Mater. Trans. 1982, 134A, 929–935. [CrossRef]
28. Reynaud, G.M.; Clark, W.A.T.; Rapp, R.A. In situ observation of copper oxidation at high temperatures. Metall. Mater. Trans. 1984,
15A, 573–586. [CrossRef]
29. Rapp, R.A. The high temperature oxidation of metals forming cation-diffusing scales. Metall. Mater. Trans. 1984, 15A, 765–782.
[CrossRef]
30. Małecka, J. Oxidation resistance of AlCrN coating deposited on Ti-46Al-7Nb-0.7Cr-0.1Si-0.2Ni alloy. Arch. Metall. Mater. 2014, 59,
621–627. [CrossRef]
31. Roy, T.K.; Balasubramaniam, R.; Ghosh, A. Effect of nitrogen on the oxidation behavior of Ti3 Al-based intermetallic alloys. Metall.
Mater. Trans. 1996, 27A, 4003–4010. [CrossRef]
32. ASM Handbook. Vol. 3: Alloy Phase Diagrams, Metal Treatment, Structure and Joining Collection. In Section: Binary Alloy Phase
Diagrams—Standard Content; ASM International: Materials Park, OH, USA, 1998.
33. Taniguchi, S.; Shibata, T. Influence of additional elements on the oxidation behaviour of TiAl. Intermetallics 1996, 4, 85–93.
[CrossRef]
34. Haanappel, V.A.C.; Clemens, H.; Stroosnijder, M.F. The high temperature oxidation behaviour of high and low alloyed TiAl-based
intermetallics. Intermetallics 2002, 10, 293–305. [CrossRef]
35. Varma, S.K.; Chan, A.; Mahapatra, B.N. Static and cyclic oxidation of Ti–44Al and Ti–44Al–xNb alloys. Oxid. Met. 2001, 55,
423–435. [CrossRef]
36. Fergus, J.W. Review of the effect of alloy composition on the growth rates of scales formed during oxidation of gamma titanium
aluminide alloys. Mater. Sci. Eng. 2002, 338A, 108–125. [CrossRef]
37. Jung, H.G.; Kim, K.Y. Effect of ternary elements on the oxidation behavior of aluminized TiAl alloys. Oxid. Met. 2002, 58, 197–216.
[CrossRef]
38. Becker, S.; Rahmel, A.; Schorr, M. Mechanism of isothermal oxidation of the intel-metallic TiAl and of TiAl alloys. Oxid. Met. 1992,
38, 425–464. [CrossRef]
39. Jiang, H.; Hirohasi, M.; Lu, Y.; Imanari, H. Effect of Nb on the high temperature oxidation of Ti-(0–50at.%)Al. Scr. Mater. 2002, 46,
639–643. [CrossRef]