metals

Article
Dissolution of Grain Boundary Carbides by the Effect
of Solution Annealing Heat Treatment and Aging
Treatment on Heat-Resistant Cast Steel HK30
Francisco J. G. Silva 1, * , Jorge Santos 1,2 and Ronny Gouveia 1
1 Department of Mechanical Engineering, ISEP-School of Engineering, Polytechnic of Porto,
4249-015 Porto, Portugal; jorge.santos@ju.se (J.S.); ronnygouveia@gmail.com (R.G.)
2 School of Engineering, Jönköping University, SE-55111 Jönköping, Sweden
* Correspondence: fgs@isep.ipp.pt; Tel.: +351-228-340-500

Received: 29 May 2017; Accepted: 26 June 2017; Published: 5 July 2017

Abstract: Decreasing the weight of heavy-duty vehicles is an ongoing concern. However, the need to
deal with high temperatures in components such as manifolds imposes, by itself, some restrictions
regarding material selection, being further limited when other required properties (e.g., functional,
manufacturing or cost requirements) are taken into account. Cast austenitic stainless steels may
represent a good choice in this context but the existence of concentrated chromium carbides
can generate undesirable results. A good combination of heat treatments can be applied
to cast heat-resistant austenitic stainless steels, in an effort to achieve the dispersion of fine
carbides, consequently improving their microstructure, mechanical properties and creep resistance.
In this work, an austenitic stainless steel usually used in high temperature applications was
characterized and subjected to solution annealing and aging heat treatments. The material analyzed
was the austenitic cast stainless steel HK30 and the goals of the work were to evaluate the effects of
solution annealing heat treatments on the dissolution of grain boundary chromium carbides and the
effects of aging treatments on creep resistance. The results show that the elimination of grain boundary
chromium carbides is possible by applying a solution annealing heat treatment. Additionally, the
precipitation of fine dispersed carbides is obtained after the aging treatment with an increase
of hardness and, consequently, an expected improvement of creep resistance. Thus, the novelty
presented by this work consists of selecting the best heat treatment combination in order to promote
dispersion of carbides, thus avoiding further crack nucleation phenomena when parts are cyclically
subjected to load and unload; this work also found the most adequate mechanical properties and
achieved corrosion resistance regarding the application in heavy-duty vehicle components subjected
to mechanical and thermal fatigue. By discovering methods of improving the properties of cast
materials, large savings can be made both in terms of production costs as well as in the overall weight
of the components.

Keywords: cast stainless steel; solution annealing; aging heat treatment; intergranular corrosion;
grain boundary carbides

1. Introduction
The increasing demand in the automotive industry to reduce CO2 emissions is the driving force
behind improving fuel economy and enhancing combustion efficiency in the heavy-duty vehicle
industry [1]. This will result in a change in composition of the emitted exhaust gases and an even
higher exhaust temperature, bringing higher demands for both corrosion resistance and heat-resistance
materials used for exhaust components. Heat-resistant cast stainless steels have been developed to
meet the increasing demands in several applications [2,3], including exhaust components.

Metals 2017, 7, 251; doi:10.3390/met7070251 www.mdpi.com/journal/metals

Metals 2017, 7, 251 2 of 12

Austenitic cast stainless steels are usually found in the most demanding conditions where
ferritic alloys show limitations [2–6]. Besides chromium, they present high amounts of nickel in
their composition to stabilize the austenite phase [7–9].
The heat-resistant cast stainless steel HK30 (ASTM A351) is a 25Cr/20Ni alloy commonly found in
exhaust manifolds of high power gasoline engines, where the gas temperatures may reach 1000 ◦ C [7–9].
The microstructure is composed of austenite with massive carbides embedded in the matrix. When the
cooling rate is slow, precipitation of carbides at the grain boundaries may occur, depleting the matrix
of chromium and making the steel more sensitive to intergranular corrosion. Furthermore, the ductility
is reduced and the material becomes more brittle. This precipitation phenomenon occurs when
materials are subjected to working conditions that present large thermal amplitudes, ranging from
room temperature to elevated temperatures, and then are allowed to cool slowly enough to permit
steel sensitization [10]. Welding processes may also lead to a similar outcome when slow cooling is
also present.
The literature refers to the susceptibility of stainless steels to intergranular corrosion when
chromium carbides precipitate at the grain boundaries, during high temperature exposure [10,11].
These chromium carbides, type M7C3 and M23C6 (M = Fe, Cr) [11,12], will deplete the surrounding
matrix of chromium and make the alloy more sensitive to intergranular corrosion. This is easy to
understand since these chromium carbides present a chromium content between 24% and 60% [13,14].
These carbide networks usually appear when the alloy is cooled very slowly through high temperature
ranges, in which the carbon-rich austenite rejects the carbon (maximum carbon solubility of 2.06% at
1148 ◦ C in the austenite) as grain boundary networks instead of dispersed particles [7,8,10].
Solution annealing heat treatment is usually applied in austenitic cast stainless steels to prevent
alloy sensitization. The treatment consists of a holding time at high temperatures in the range of
1040 to 1205 ◦ C in order to dissolve the carbides back to the matrix, followed by severe cooling to
prevent carbides re-precipitation. This is possible due to the increase of carbon solubility in austenite—
a consequence of the elevated temperatures. Another important consideration is that carbon carbides
will always be present to some extent in the microstructure of alloys with carbon contents greater than
0.20%, such as HK30 alloy, regardless of the solution annealing treatment applied [15,16].
Usually, the solution annealing heat treatment is followed by an aging treatment in order to
promote a fine dispersed precipitation of carbides, improving creep resistance [10,17]. Aging at 760 ◦ C
promotes the precipitation of fine, uniformly dispersed carbides resulting in a higher strength level and
can be kept up to temperatures of around 980 ◦ C, at which point secondary carbides start to agglomerate
and spheroidize, decreasing the creep strength [10]. Another study [18] regarding 23-8-N nitronic steel,
an austenitic stainless steel containing nitrogen in its matrix in order to improve its mechanical strength,
toughness, ductility, erosion and corrosion resistance, showed that a solution treatment at 1220 ◦ C for
150 min ensures the best carbide dissolution in the austenite phase, also showing a good combination
of high toughness and elevated mechanical strength, improving significantly erosion resistance.
The purpose of this work is to study the possibility of eliminating the grain boundary carbides
that may appear during material processing, by applying solution annealing heat treatments and the
effects of aging treatments on the microstructure and hardness of the heat-resistant cast steel HK30.

2. Materials and Methods

2.1. Materials Characterization

Cross-section samples were removed from several exhaust manifolds (seen in Figure 1) produced
in an automated sand casting molding process using HK30 cast steel. Five samples were extracted
regarding each heat treatment condition and considering three different areas, as depicted in
Figure 1a, attending as well to different cooling conditions in the application considered in this
work. The chemical composition of HK30 is given in Table 1. The samples presented a rectangular
shape, 20 mm in length and around 6 mm in thickness.
Metals 2017, 7, 251 3 of 12

work. The2017,
Metals chemical
7, 251 composition of HK30 is given in Table 1. The samples presented a rectangular
3 of 12
shape, 20 mm in length and around 6 mm in thickness.

(a) (b)
Figure 1. HK30
Figure cast cast
1. HK30 stainless steel steel
stainless turboturbo
manifold; (a) location
manifold; of sample
(a) location removal;
of sample (b) 3D
removal; (b) CAD
3D CAD
(Computer-aided
(Computer-aided design)
design) drawing
drawing of turbo
of turbo manifold.
manifold.

Table 1. Chemical
Table composition
1. Chemical of austenitic
composition stainless
of austenitic steelsteel
stainless HK30 (wt %).
HK30 (wt %).

Composition Cr Si Mn Cr Ni Mo Cu
Composition Cr Si Mn Cr Ni Mo Cu
HK30 0.34 1.40 1.11 24.37 19.13 0.38 0.37
HK30 0.34 1.40 1.11 24.37 19.13 0.38 0.37
Material characterization was conducted through microstructure analyzes and hardness
measurements.
Material Allcharacterization
the samples werewas cut, conducted
mounted, polished
throughand etched. The etch
microstructure reagentand
analyzes usedhardness
was
Kallings Etsmedel etchant
measurements. All the (LabService,
samples were Stockholm, Sweden)
cut, mounted, (5 g and
polished of copper
etched.chloride,
The etch100 mL ofused
reagent
hydrochloric acid, 100 mL of ethanol, 100 mL of water).
was Kallings Etsmedel etchant (LabService, Stockholm, Sweden) (5 g of copper chloride, 100 mL of
A CARL ZEISS
hydrochloric acid,AX10
100 mL (Carl Zeiss Microscopy,
of ethanol, Jena, Germany) optical microscope with an
100 mL of water).
incorporated
A CARLAXIOCAMZEISS MRC
AX10digital
(Carl camera (Carl Zeiss Microscopy,
Zeiss Microscopy, Jena, Germany) Jena,optical
Germany) was usedwith
microscope for an
microstructure
incorporated AXIOCAM MRC digital camera (Carl Zeiss Microscopy, Jena, Germany) was aused
observation. Micro and macro hardness measurements were performed with
Matsuzawa MxT30 (Matsuzawa,
for microstructure observation.Akita, Japan)
Micro and and a KB
macro 3000 BVZ
hardness (ball diameter
measurements of 2.5
were mm withwith
performed a a
nominal test load of 187.5 kg, Hochdorf, Assenheim, Germany) testing machines,
Matsuzawa MxT30 (Matsuzawa, Akita, Japan) and a KB 3000 BVZ (ball diameter of 2.5 mm with respectively. The
results presented
a nominal areload
test the of
average
187.5 of
kg,five measurements.
Hochdorf, Assenheim, Germany) testing machines, respectively.
Some microstructural analyzes, carried out by a scanning electron microscope (SEM) and X-ray
The results presented are the average of five measurements.
microanalyses
Some microstructural analyzes, carried out FEG-SEM/EDS
(EDS) were performed using a Zeiss microscope
by a scanning electron Sigma VP (SEM)
(Carl and
Zeiss
X-ray
Microscopy, Jena, Germany) provided with a Gemini field emission column
microanalyses (EDS) were performed using a FEG-SEM/EDS Zeiss Sigma VP (Carl Zeiss Microscopy, and element
microanalysis (Carl provided
Jena, Germany) Zeiss Microscopy, Jena,field
with a Gemini Germany)
emissionwas carried
column andout with an
element Energy Dispersive
microanalysis (Carl Zeiss
Spectrometer, EDS INCA Oxford Instruments (Oxford Instruments, Abingdon, UK).
Microscopy, Jena, Germany) was carried out with an Energy Dispersive Spectrometer, EDS INCA
Oxford Instruments (Oxford Instruments, Abingdon, UK).
2.2.Heat Treatments
2.2. Heat Treatments
A series of heat treatments was conducted, changing three variables: solution annealing
holding time, quenching
A series medium (after
of heat treatments the solutionchanging
was conducted, treatment holding
three time) solution
variables: and the aging holding
annealing holding
time.time,
The quenching
heat treatment effectiveness was evaluated through microstructural analysis by optical
medium (after the solution treatment holding time) and the aging holding time.
and The
scanning electron microscopes
heat treatment andwas
effectiveness hardness measurements.
evaluated Table 2 shows the
through microstructural chosen
analysis byvariables
optical and
and scanning
range of respective values.
electron microscopes and hardness measurements. Table 2 shows the chosen variables and
range of respective values.
Table 2. Variables and experimental parameters.
Table 2. Variables and experimental parameters.
Variables − +
ST—Solution annealing temperature holding time 1.5 h 3h
Variables mean
QM—Quenching −
Oven (O.C.) Water (W.Q.)+
AT—Aging
ST—Solution annealing holding time holding time
temperature 0 h1.5 h 24 h 3 h
QM—Quenching mean Oven (O.C.) Water (W.Q.)
AT—Aging holding time 0h 24 h
Metals 2017, 7, 251 4 of 12
Metals 2017, 7, 251 4 of 12

Table 33 shows
Table shows the
theexperimental
experimentalmatrix obtained
matrix afterafter
obtained decomposing the variables.
decomposing the variables.

Table3.3.Matrix
Table Matrixof
ofexperiments.
experiments.

Sample Designation ST QM AT
Sample Designation ST QM AT
ST/1.5 + O.C. − − −
ST/1.5 + O.C.
ST/3 + O.C. − + −− − −
ST/3 + O.C.
ST/1.5 + W.Q. + − +− − −
ST/1.5 + W.Q.
ST/3 + W.Q. − + ++ − −
ST/3 + W.Q.
ST/1.5 + O.C. + AT + − −+ + −
ST/1.5 + O.C. + AT − − +
ST/3 + O.C. + AT + − +
ST/3 + O.C. + AT + − +
ST/1.5 + ST/1.5
W.Q. ++AT
W.Q. + AT − − ++ + +
ST/3++AT
ST/3 + W.Q. W.Q. + AT + + ++ + +

Fixed temperatures were selected for solution annealing and aging heat treatments, 1200 and
Fixed temperatures were selected for solution annealing and aging heat treatments,
760 °C, respectively. The solution heat treatment temperature (1200 °C) applied is almost the
1200 and 760 ◦ C, respectively. The solution heat treatment temperature (1200 ◦ C) applied is almost the
maximum temperature referred to by the literature [10] in order to ensure the effective dissolution of
maximum temperature referred to by the literature [10] in order to ensure the effective dissolution of
grain boundary carbides in the matrix. At 760 °C [10], the precipitation of fine, uniformly dispersed
grain boundary carbides in the matrix. At 760 ◦ C [10], the precipitation of fine, uniformly dispersed
carbides is promoted, improving the creep resistance. The cooling applied after the aging holding
carbides is promoted, improving the creep resistance. The cooling applied after the aging holding time
time was still performed by air. Figure 2 shows schematically the different thermal cycles employed.
was still performed by air. Figure 2 shows schematically the different thermal cycles employed.

Figure2.2. Thermal
Figure Thermalcycles
cyclesapplied
appliedin
inheat
heattreatments.
treatments.

3.
3. Results
Results and
and Discussion
Discussion

3.1.
3.1. As-Cast
As-Cast Microstructure
Microstructure
The
The HK30 isis aa 25Cr/20Ni
HK30 25Cr/20Ni alloy,
alloy, so
so itit isis an
an iron-chromium-nickel
iron-chromium-nickel alloy; alloy; aa microstructure
microstructure isis
expected
expected that
that is
is composed
composed of of an
an austenitic
austeniticmatrix
matrixcontaining
containingrelatively
relativelylarge
largecarbides
carbidesininthe
the form
form of
of
either unconnected islands or as networks [10]. Figure 3a presents the as-cast microstructure of
either unconnected islands or as networks [10]. Figure 3a presents the as-cast microstructure of
HK30
HK30samples
samplescomposed
composed of aoffully austenitic
a fully matrixmatrix
austenitic and carbides, as individual
and carbides, islands and
as individual networks
islands and
at grain boundaries. As can be seen in Figure 3b, there are unconnected carbide
networks at grain boundaries. As can be seen in Figure 3b, there are unconnected carbide islands islands dispersed
into each grain
dispersed into in the grain
each as-castinsamples. As samples.
the as-cast referred toAspreviously,
referred tothepreviously,
carbide networks formed
the carbide at the
networks
grain
formedboundaries
at the graincanboundaries
promote material intergranular
can promote materialcorrosion and embrittlement.
intergranular For these reasons,
corrosion and embrittlement. For
heat treatments were designed in order to eliminate the carbide networks
these reasons, heat treatments were designed in order to eliminate the carbide networks and consequently improve
and
ductility, creepimprove
consequently and corrosion resistance
ductility, at high
creep and temperatures.
corrosion resistance at high temperatures.
Metals 2017, 7, 251 5 of 12
Metals 2017,
Metals 2017, 7,
7, 251
251 55 of
of 12
12

Grain boundary
carbides

Austenite Unconnected
carbides
(a)
(a) (b)
(b)
Figure 3.
Figure 3.
Figure (a) Microstructure
3. (a)
(a) Microstructure obtained
Microstructure obtained in
obtained in as-cast
in as-cast state
as-cast state composed
state composed of
composed of aaa fully
of fully austenitic
fully austenitic matrix,
austenitic matrix, grain
matrix, grain
grain
boundary chromium-rich
boundary chromium-rich
boundary carbides
chromium-rich carbides and massive
carbides and
and massive carbides
massive carbidesin the
carbides in matrix;
in the (b)
the matrix; SEM
matrix; (b) image
(b) SEMof grain
SEM image boundary
image ofof grain
grain
carbides
boundaryand
boundary unconnected
carbides
carbides and carbide islands
and unconnected
unconnected in the
carbide
carbide as-cast
islands
islands sample.
in the
in the as-cast sample.
as-cast sample.

3.2. Solution
3.2. Solution Annealing
Annealing Followed
Followed by
by Water Quenching
Water Quenching
Figure 444shows
Figure showsmicrostructures
shows microstructuresobtained
microstructures obtained
obtained from
from
from samples
samples
samples after
after after solution
solution
solution annealing
annealing
annealing treatment.
treatment.
treatment. The
The
The quenching
quenching
quenching meanmean
mean for for samples
for both
both both samples
samples was was water.
was water.
water. Figure Figure
Figure 4a represents
4a represents
4a represents the microstructure
the microstructure
the microstructure of a
of aa sample
of sample
sample ◦ C for 1.5 h. It is possible to distinguish some differences compared
solutionsolution
solution annealed
annealed annealed
at 1200 at
at 1200 °C1200
°C for 1.5
for 1.5 h. It
h. It is
is possible
possible toto distinguish
distinguish some some differences
differences compared
compared with with
with
the the as-cast
the as-cast
as-cast sample.
sample. sample.
When When
When comparing
comparing comparing
Figure 3a
Figure Figure
with3a
3a with with4a,
Figure
Figure Figure
4a, 4a, a decrease
aa decrease
decrease in grain
in in grain boundary
grain boundary
boundary carbides
carbides
carbides
is clear. is
is clear. The
Theclear. The carbides
carbides
carbides were were partially
were partially
partially dissolved
dissolved dissolved
back into
back backthe
into intomatrix,
the the matrix,
matrix, becoming
becoming
becoming thinner
thinner thinner
and
and
and disconnected. For the sample solution annealed at 1200 ◦ C for 3 h (Figure 4b), there are far
disconnected. For
disconnected. For the
the sample
sample solution
solution annealed
annealed at at 1200
1200 °C°C for
for 33 hh (Figure
(Figure 4b),
4b), there
there are
are far
far less
less grain
grain
less grain carbides
boundary
boundary boundary
carbides carbideswith
compared
compared compared
with the sample
the withsolution
sample the sample
solution solution
annealed
annealed annealed
for 1.5
for 1.5 h, which
h, which for 1.5expected,
was
was h, whichsince
expected, was
since
expected, since
the dissolution
the dissolution of the dissolution
of carbides
carbides is is alsoof carbides
also aa time is
time dependent also a time
dependent process. dependent
process. A A sufficientprocess.
sufficient amount A
amount of sufficient
of time
time is amount
is necessary
necessary of
time
to is necessary
to accomplish
accomplish the to
the accomplish
maximum
maximum the maximum
carbon
carbon dissolutioncarbon
dissolution in dissolution
in austenite
austenite in austenite
[19]. From
[19]. From the X-ray
the [19].
X-ray From the X-ray
microanalysis,
microanalysis, the
the
microanalysis,
chromium content
chromium the chromium
content near carbides
near carbidescontent
wasnear
was higher
higher carbides
(26.3%)
(26.3%)was higher
than
than in the
in (26.3%)
the middle
middle than
of in
of the
thethe middle
matrix
matrix of thefor
(24.5%)
(24.5%) matrix
for the
the
(24.5%) for
sample ST/1.5
sample the
ST/1.5 h. sample
h. This
This canST/1.5
can be h. This
be explained
explained by can be
by the explained
the dissolution by
dissolution of the dissolution
of grain
grain boundary of grain
boundary carbides boundary
carbides during carbides
during solution
solution
during solution
annealing,
annealing, annealing,
increasing
increasing increasingcontents
the chromium
the chromium the chromium
contents tocontents
close to
close carbides.
carbides. close to carbides.

(a)
(a) (b)
(b)
Figure 4. (a) Microstructure obtained after solution annealing at 1200 ◦C
for 1.5
1.5 hh followed by water
Figure 4.
Figure 4. (a)
(a) Microstructure
Microstructure obtained
obtained after
after solution
solution annealing
annealing at at 1200
1200 °C
°C for
for 1.5 h followed
followed by
by water
water
quenching; (b) Sample solution annealed at 1200 ◦ C for 3 h followed by water quenching.
quenching; (b)
quenching; (b) Sample
Sample solution
solution annealed
annealed at
at 1200
1200 °C
°C for
for 33 hh followed
followed byby water
water quenching.
quenching.
Metals 2017, 7, 251 6 of 12
Metals 2017, 7, 251 6 of 12

For both samples

samples in Figure 4, very small small carbides
carbides are
are present near massive
massive andand grain
grain boundary
boundary
carbides. These
Thesesmall
smallcarbides
carbides or or decomposed
decomposed carbides
carbides maymay be theberesult
the result of the dissolution
of the dissolution of lamellarof
lamellar
massive andmassive
grainand grain boundary
boundary carbides
carbides during the during
solutionthe solution since
annealing, annealing,
they aresince
not they are
visible in not
the
visible
as-cast in the as-cast
state. state. The
The literature literature
states that atstates that at high temperatures,
high temperatures, such as thosesuchused asinthose
this used in this
experiment,
experiment,
carbides can carbides can be
be dissolved dissolved or spheroidized
or spheroidized [19]. Increasing [19].
theIncreasing the solution
solution annealing annealing
holding time,
holding time, carbides
these massive these massive
becomecarbides becomedue
bigger, possibly bigger,
to thepossibly due toofthe
agglomeration agglomeration
decomposed carbidesof
decomposed
from previouscarbides
massivefrom previous
carbides. massive between
A difference carbides.massive
A difference between
carbide sizes ismassive
visible, carbide sizes in
being bigger is
visible, being
the sample biggerannealed
solution in the sample
for 3 solution
h (Figureannealed
4b) than for
1.5 3h h(Figure
(Figure4a).
4b) than 1.5 h (Figure 4a).
From
From these experiments,
experiments, we we can
can conclude
conclude that
that during
during solution
solution annealing
annealing heatheat treatment,
treatment,
carbide dissolution
dissolution takes
takesplace
placeuntil
untilthe
themaximum
maximumcarbon
carbon solubility
solubility in in austenite
austenite is reached.
is reached. WhenWhenthe
the solubility
solubility limitlimit of carbon
of carbon is reached,
is reached, carbide
carbide dissolution
dissolution stops stops and carbide
and carbide agglomeration
agglomeration starts,
starts, due to
due to thetemperature
the high high temperature
[15]. [15].

3.3. Solution
3.3. Solution Annealing
Annealing Followed
Followed by
by Oven
Oven Cooling
Cooling
Microstructures obtained
obtained for
for sample
sample solutions
solutions annealed
annealed atat 1200 ◦ C, followed by oven cooling,
1200 °C,
Microstructures followed by oven cooling,
are presented in Figure 5, for 1.5 and 3 h holding time. When comparing sample
are presented in Figure 5, for 1.5 and 3 h holding time. When comparing sample solutions solutions annealed
annealed
for 1.5
for 1.5 hh (Figures
(Figures 4a4a and
and 5a),
5a), larger
larger massive
massive carbides
carbides and
and thicker
thicker grain
grain boundary
boundary carbides
carbides are
are visible
visible
in the
in the oven
oven cooled
cooled sample
sample than
than in
in the
the water
water quenched
quenched sample.
sample. The
The sample
sample solution
solution annealed
annealed for
for 33 h
h
followed by oven cooling presents a greater quantity of grain boundary carbides
followed by oven cooling presents a greater quantity of grain boundary carbides than when than when compared
with the water ◦ C/h) may be sufficient
compared with quenched
the water sample.
quenched The slow cooling
sample. The slowin cooling
an ovenin(around
an oven 120
(around 120 °C/h) may be
to keep the samples at high temperatures long enough for the carbide agglomeration
sufficient to keep the samples at high temperatures long enough for the carbide agglomeration process to take
process to take place. Also, a higher carbide concentration in the microstructure (increase ofratio)
place. Also, a higher carbide concentration in the microstructure (increase of the carbide/matrix the
is visible. This can
carbide/matrix also
ratio) be explained
is visible. by the
This can alsoslow cooling, which
be explained by theenables furtherwhich
slow cooling, carbide precipitation,
enables further
due to the
carbide decreasing due
precipitation, of carbon
to thesolubility
decreasing inof
austenite with temperature.
carbon solubility in austenite with temperature.

(a) (b)
Figure 5. (a) Microstructure
5. (a) Microstructure obtained
obtained after
after solution
solution annealing
annealing atat 1200 ◦ C for 1.5 h followed by oven
1200 °C
Figure for 1.5 h followed by oven
cooling; (b)
(b) Sample
Sample solution
solution annealed
annealed at
at 1200 ◦
1200 °CC for
for 33 hh followed
followed by
by oven
oven cooling.
cooling.
cooling;

The carbide precipitation

precipitation (due
(due to
to the
the carbon
carbon excess
excess inin the
the austenite)
austenite) at
at high
hightemperature
temperature ranges,
ranges,
takes place
placeasasgrain
grainboundary
boundary networks,
networks, instead
instead of asofdispersed
as dispersed particles
particles in the in the matrix
matrix [19].
[19]. This This
carbide
carbide precipitation
precipitation tendencytendency
in grain in grain boundaries
boundaries can be seencan by
be the
seen by the
more more continuous
continuous carbide
carbide networks
networks
presentedpresented when comparing
when comparing with bothwith both
water water quenched
quenched samples samples (Figure
(Figure 4). The 4). The conclusion
conclusion drawn
drawn
at the at
endtheofend of these
these trialstrials is that
is that slowslow cooling
cooling after
after solutionannealing
solution annealingheat
heat treatment
treatment promotes
the
the agglomeration
agglomeration of of carbides
carbides andand precipitation
precipitation at
at grain
grain boundaries
boundaries of of previous
previous massive
massive carbides.
carbides.
So, the dissolution effect of carbides in austenite after solution annealing loses its effect, if further
cooling is slow enough to admit carbide precipitation and agglomeration at high temperatures.
effect of the solution annealing in the aging treatment.
Figure 6 presents the samples’ microstructures after aging. Figure 6a represents the 1.5 h
solution annealed sample and Figure 6b the 3 h solution annealed sample, both having been
quenched in water and aged for 24 h at 760 °C.
MetalsThe
2017, microstructures
7, 251 presented in Figure 6 show a big difference when compared with 7 ofthe
12
samples that have been just solution annealed for both quenching media. Fine dispersed carbides in
the matrix are visible as we can see at higher magnifications in Figure 6c,d. Carbon further from
So, theboundaries
grain dissolution remains
effect of incarbides in austenite
supersaturated after solution
austenite, annealing
with longer loses its
holding effect,
times andif further
greater
cooling is slow enough to admit carbide precipitation and agglomeration at high temperatures.
supersaturation being needed for the nucleation and growth of these precipitates to occur [20]. Thus,
the previous samples (without aging) did not remain at high temperatures long enough to admit
3.4. Aging Heat Treatment—Sample Solutions Annealed Followed by Water Quenching
precipitation of these fine dispersed carbides. The aged samples both present fine dispersed carbides
in theTheaustenite
aging treatments
matrix. were applied
However, asafter solution
referred annealing heat
to previously, fortreatments, to compare
solution annealed andthe water
effect
of the solution annealing in the aging treatment.
quenched samples, the 1.5 h solution annealed sample presents disconnected thin grain boundary
Figure
carbides, 6 presents
whereas thethe
3 hsamples’
solutionmicrostructures
annealed sample after aging.no
shows Figure 6a grain
visible represents the 1.5carbides
boundary h solutionin
annealed
Figure 6. sample and Figure
As for solution 6b the 3samples
annealed h solution annealed
after water sample, boththe
quenching, having
higherbeen
thequenched in water
holding time the
and aged
larger thefor 760 ◦ C. are.
24 h atcarbides
massive

(a) (b)

(c) (d)
◦C
Figure 6. (a)
Figure 6. (a) Microstructure
Microstructure obtained
obtained after
after solution
solution annealing
annealing atat 1200
1200 °C for
for 1.5
1.5 h
h followed
followed by
by water
water
quenching and aged at 760 ◦ C for 24 h; (b) Sample solution annealed at 1200 ◦ C for 3 h followed by
quenching and aged at 760 °C for 24 h; (b) Sample solution annealed at 1200 °C for 3 h followed by
water ◦ for 24 h. (c) SEM micrograph for the sample solution annealed for
water quenching
quenchingandandaged
agedatat760
760 C°C for 24 h. (c) SEM micrograph for the sample solution annealed
1.5
for 1.5 h followed by water and
h followed by water quenching aged forand
quenching 24 h; aged
(d) Thefor
same24 sample
h; (d)at The
highersame
magnification.
sample at
higher magnification.
The microstructures presented in Figure 6 show a big difference when compared with the samples
that have been just solution annealed for both quenching media. Fine dispersed carbides in the matrix
are visible as we can see at higher magnifications in Figure 6c,d. Carbon further from grain boundaries
remains in supersaturated austenite, with longer holding times and greater supersaturation being
needed for the nucleation and growth of these precipitates to occur [20]. Thus, the previous samples
(without aging) did not remain at high temperatures long enough to admit precipitation of these fine
dispersed carbides. The aged samples both present fine dispersed carbides in the austenite matrix.
However, as referred to previously, for solution annealed and water quenched samples, the 1.5 h
solution annealed sample presents disconnected thin grain boundary carbides, whereas the 3 h solution
Metals 2017, 7, 251 8 of 12

Metals 2017, 7, 251 8 of 12

3.5. Aging Heat Treatment—Sample Solutions Annealed Followed by Oven Cooling
The effect of samples oven cooled after the solution annealing on aging treatment was
annealed sample shows no visible grain boundary carbides in Figure 6. As for solution annealed
also evaluated.
samples
From after water
Figure 7, quenching,
one thing is theclear:
higherthethe holding time
precipitation ofthe
finelarger the massive
dispersed carbidescarbides are. less
was much
intense when
3.5. Aging Heatcompared with the water
Treatment—Sample Solutionsquenched
Annealedsamples.
FollowedActually, the sample solution annealed
by Oven Cooling
for 1.5 h and aged does not present fine dispersed carbides in the matrix. The reason for the 3 h
The annealed
solution effect of sample
samplespresenting
oven cooled less after the solution
fine dispersed annealing
carbide on aging
precipitation is, treatment
as mentionedwas
also evaluated.
before, the precipitation of carbon far from grain boundaries needs, besides a much longer period of
time,From Figure
greater 7, one thing isDuring
supersaturation. clear: thetheprecipitation
slow cooling of fine
in thedispersed carbides was of
oven, precipitation much less intense
carbides takes
when compared with the water quenched samples. Actually, the sample solution
place, depleting the matrix in carbon. When the aging treatment is applied, the austenite is less annealed for 1.5 h and
aged does not present
supersaturated fine decreasing
in carbon, dispersed carbides in the matrix.
the possibility The reasonoffor
of precipitation thedispersed
fine 3 h solution annealed
carbides far
sample presenting less
from the grain boundaries. fine dispersed carbide precipitation is, as mentioned before, the precipitation of
carbon far from grain boundaries needs, besides a much longer period of time, greater
The lower the holding time at the solution annealed temperature, the less carbides are dissolved supersaturation.
During
and the slow cooling
consequently in the
there will beoven, precipitation ofaustenite.
less supersaturated carbides takes place,for
Therefore, depleting
the sample the matrix in
solution
carbon. When the aging treatment is applied, the austenite is less supersaturated
annealed for 1.5 h (Figure 7a), no fine dispersed carbides are visible in the matrix while for the in carbon, decreasing
the possibility
sample of precipitation
with a higher of fine
holding time (3 dispersed
h), some finecarbides far from
dispersed the grain
carbides boundaries.
are visible (Figure 7b).

(a) (b)
Figure 7. (a) Microstructure
7. (a) Microstructure obtained
obtained after
after solution
solution annealing
annealing atat 1200 ◦C
1200 °C for
for 1.5
1.5 hh followed
followed by
by oven
oven
Figure
cooling and aging; (b) Sample solution annealed at 1200 ◦ C for 3 h followed by oven cooling.
cooling and aging; (b) Sample solution annealed at 1200 °C for 3 h followed by oven cooling.

3.6. Hardness
The lowerMeasurements
the holding time at the solution annealed temperature, the less carbides are dissolved
and consequently there will be less supersaturated austenite. Therefore, for the sample solution
Figure 8 presents the macro-hardness obtained by each heat treatment. Figure 9 shows the
annealed for 1.5 h (Figure 7a), no fine dispersed carbides are visible in the matrix while for the sample
micro-hardness measured on the matrix of each heat-treated sample. Note that the micro-hardness
with a higher holding time (3 h), some fine dispersed carbides are visible (Figure 7b).
measurements for the cast stainless steel were conducted at three different areas of the samples’
surface, namely
3.6. Hardness inside, in the middle and outside. These three places on the samples were analyzed
Measurements
in order to test the different cooling rates that may be subjected, despite the relatively low thickness
of theFigure
samples8 presents
(~6 mm).the
Themacro-hardness obtained by
inside surface represents theeach heat
surface intreatment.
contact withFigure 9 shows
the core the
sand, the
micro-hardness
outside surface ismeasured
in contacton thethe
with matrix of each
mould sandheat-treated
and the coresample. Note that
of the sample is inthe
the micro-hardness
middle.
measurements for the cast stainless steel were conducted at three different areas of the samples’ surface,
namely inside, in the middle and outside. These three places on the samples were analyzed in order
to test the different cooling rates that may be subjected, despite the relatively low thickness of the
samples (~6 mm). The inside surface represents the surface in contact with the core sand, the outside
surface is in contact with the mould sand and the core of the sample is in the middle.
Metals 2017, 7, 251 9 of 12

Metals 2017, 7, 251 9 of 12

Figure 8. Macro-hardness obtained for different heat treatments.

Figure 8. Macro-hardness obtained for different heat treatments.
Returning to Figure Figure 8,8.there
Macro-hardness
is one main obtained for different
conclusion drawnheat treatments.
from the results: sample ST/1.5 +
W.Q. + AT clearly presents the highest hardness of all tested HK30 samples.
Returning to
Returning to Figure
Figure8, 8,therethere is one
is the
one mainmain conclusion
conclusion drawndrawn from from
the the results:
results: sample
From Figure 9, it is visible that highest hardness values are generally foundsample
on theST/1.5
outside +
ST/1.5+ +
W.Q. AT W.Q. + AT
clearly clearly the
presents presents
highest thehardness
highest hardness
of all testedof all tested
HK30 HK30 samples.
samples.
surfaces when compared to the inside and middle surface measurements. This may be due to
From Figure
From Figure rates9,
9, it
it is
isinvisible
visible that
that the
the highest hardness values are generally found on the
the outside
different cooling each edge. Thehighest
insidehardness
surface invalues
contact arewith
generally
the corefoundsand,on and outside
middle
surfaces
surfaces when compared to the inside and middle surface measurements. This may be due to different
surface suffered lower cooling rates leading to lower hardness while the higher cooling ratedue
when compared to the inside and middle surface measurements. This may be to
of the
cooling
differentrates in each
coolingresulted edge. The
rates inineach inside
edge. surface in
The inside contact with
surface the core
in contact sand, and
with the middle
corethe surface
sand, suffered
outside surface higher hardness. The inside surface is in contact with core and
sandmiddle
which
lower cooling
surface suffered rates leading
lower cooling to lower
rates hardness
leading to while
lower the higherwhile
hardness cooling
the rate
higherof the outside
cooling ratesurface
of the
is surrounded by metal walls. During solidification, these metal walls release heat through the
resultedsurface
outside in higher hardness.
resulted in The inside
higher hardness. surface
The isinside
in contact
surface with
is the
in core sand
contact with which
the is surrounded
core sand which
mould sand to the outside and to the core sand. The inside surface will have a slower decrease in
by surrounded
is metal walls. by During
metal solidification,
walls. these metal wallsthese release heatwalls
through the mould sand to the
temperature because the heat canDuring
only besolidification,
released through the metal release
cast component’s heat walls,
metal through the
which
outside
mould and
sand to
to the
the core sand.
outside The
and to inside
the surface
core sand. will
The have
insidea slower
surface decrease
will havein temperature
a slower because
decrease in
consequently only occurs when there is a temperature decrease in the surrounding mould sand. For
the heat can because
temperature only be released
the heat through
can only thereleased
be cast component’s
through themetal
cast walls, which metal
component’s consequently
walls, only
which
this reason, the inside surface during cooling remains at higher temperatures for a longer time than
occurs when there
consequently onlyisoccurs
a temperature
when there decrease in the surrounding
is a temperature decreasemould in thesand. For this reason,
surrounding mould the sand.inside
For
the outside surface.
surface during
this reason, the cooling
inside remains
surface at
duringhigher temperatures
cooling remains for
at a longer
higher time than
temperatures the
for outside
a longer surface.
time than
For the as-cast HK30 sample in Figure 9, a higher error range in micro-hardness is visible at the
the For thesurface.
outside as-cast HK30 sample in Figure 9, a higher error range in micro-hardness is visible at the
outside surface, indicating some matrix heterogeneity. Actually, the highest error ranges were
outside surface,
For on
thethe indicating
as-cast HK30 some matrix
sample heterogeneity. Actually, the inhighest error ranges is were verified
verified outside surfaces forinthe
Figure 9, a higher
different error range
heat treatments. micro-hardness visible at the
on the outside surfaces for the different heat treatments.
outside surface, indicating some matrix heterogeneity. Actually, the highest error ranges were
verified on the outside surfaces for the different heat treatments.

Figure 9. Micro-hardness measured on the matrix obtained for each heat treatment.
treatment.

Figure 9. Micro-hardness measured on the matrix obtained for each heat treatment.
Metals 2017, 7, 251 10 of 12
Metals 2017, 7, 251 10 of 12

From X-ray
From X-ray microanalysis
microanalysis seen
seen in
in Figure
Figure 10,
10, the
the matrix
matrix presented
presented less
less chromium
chromium content
content in
in the
the
middle (23.0%)
middle (23.0%) than
than on
on the
the outside
outside surface
surface (24.1%).
(24.1%). This
This also
also shows that there
shows that there may
may be
be less
less carbon
carbon
content in the middle than on the inside surface.
content in the middle than on the inside surface.

(a)

(b)
Figure 10. (a)
Figure 10. (a) Element
Element weight
weight percentage
percentage results
results given
given by
by an
an X-ray
X-ray microanalysis
microanalysis performed
performed on
on the
the
middle and outer surface of an as-cast sample; (b) Element atomic percentage results given
middle and outer surface of an as-cast sample; (b) Element atomic percentage results given by by an X-ray
an
microanalysis performed
X-ray microanalysis on the on
performed middle and outer
the middle andsurface of an as-cast
outer surface sample.
of an as-cast sample.

The variation of hardness between the different zones is lower for the samples samples that have been
solution annealed, which decreased the effect of having different cooling rates, experimented during
the casting process. The The increase
increaseof ofhardness
hardnesswas wasalready
alreadyexpected
expectedbefore
beforethethetrial,
trial,since
sinceapplying
applyinga
asolution
solutionannealing
annealingheat heattreatment
treatmentfollowed
followedby byaging
agingheat
heattreatment
treatment results
results inin the
the highest creep
resistance [10].
resistance [10]. This
Thisisisdue
duetotothe
thefine
fine dispersed
dispersed carbide
carbide precipitation
precipitation which
which consequently
consequently leadsleads to
to the
the increasing
increasing hardness.
hardness. Analyzing
Analyzing the microstructure
the microstructure and SEM andmicrographs,
SEM micrographs, a dispersion
a dispersion of fine
of fine carbides
carbides
on on the
the matrix matrix as
is visible, is visible,
expected. as This
expected. This precipitation
precipitation may be themayreasonbe for
thethe
reason
higher forhardness
the higherin
hardness in the 1.5 h solution annealed sample followed by water
the 1.5 h solution annealed sample followed by water quenching and aged for 24 h. quenching and aged for 24 h.
Comparing the
Comparing themicro-hardness
micro-hardness obtained
obtained for for the middle
the middle section,
section, this sample
this sample presentspresents the
the highest
highest hardness
hardness compared compared with all
with all HK30 HK30 for
samples samples
the samefor zone.
the same zone.
In real In real applications,
applications, the aging
the aging treatment
treatment
does doestonot
not need need totobepromote
be applied appliedthe to precipitation
promote the of precipitation
fine dispersedof fine dispersed
carbides as, once carbides as,
in service,
oncecomponent
the in service, will
the component
achieve the will achieve the
temperatures temperatures
needed to promote needed to promote this precipitation.
this precipitation.
sample ST/3
The sample ST/3 ++W.Q.
W.Q. ++ATAT(Figure
(Figure 9)9)presents
presents lower
lower average
average micro-hardness
micro-hardness than the sample
ST/1.5 ++W.Q.
ST/1.5 W.Q. ++AT;AT; the
the only
only difference
difference between
between these samples is the length of the solution annealing
time. Comparing the microstructures of both (Figure 6), the only visible difference is the presence of
grain boundary carbides on the ST/1.5 + W.Q. + AT sample, which may not be the reason for this
difference. Both samples present fine dispersed carbides in the matrix, although a longer stage at the
Metals 2017, 7, 251 11 of 12

grain boundary carbides on the ST/1.5 + W.Q. + AT sample, which may not be the reason for this
difference. Both samples present fine dispersed carbides in the matrix, although a longer stage at the
solution annealed temperature may have resulted in the growth of austenite grain boundaries and
consequently a decrease in hardness for the sample ST/3 + W.Q. + AT.
The samples ST/1.5 + O.C. + AT and ST/3 + O.C. + AT presented lower hardness compared with
the sample ST/1.5 + W.Q. + AT. This can be explained by the slow cooling rate verified in the oven,
causing precipitation and agglomeration of carbides at high temperatures during the slow cooling,
reducing the carbon available in the matrix. Consequently, the carbon available for the formation of fine
dispersed carbides (and to increase the hardness) during aging treatment is lower, thus less carbides
will precipitate. From the microstructures in Figure 7, lower precipitation is visible compared with
the samples quenched in water (Figure 6), in agreement with previous findings. Actually, the sample
ST/1.5 + O.C. + AT does not present any fine dispersed carbides (Figure 7a), as previously mentioned.
Despite the microstructure of the ST/1.5 + O.C. + AT sample not showing visible dispersion of fine
carbides, higher hardness than the sample ST/1.5 + O.C. (Figure 8) is present. The cause of this
difference may be fine precipitation that occurred during aging near the massive carbides, resulting in
a darker zone of precipitates in Figure 7, when compared with Figure 5. This further precipitation may
be responsible for the hardness increase.
Comparing both samples that were oven cooled and aged, we can conclude that the hardness is
similar (Figure 8) due to the higher error range of the sample (ST/1.5 + O.C.).

4. Conclusions
Solution annealing and aging treatments were applied to the heat-resistant austenitic cast stainless
steel HK30 and their effects were evaluated. The conclusions are the following:

• The formation of grain boundary carbides during alloy processing is possible. This can increase
the risk of intergranular corrosion and crack initiation.
• The elimination of the grain boundary carbides can be achieved through the application of
a solution annealing heat treatment.
• Aging treatment results in the precipitation of fine dispersed carbides and increase of hardness.
• Slow cooling at high temperature ranges promotes the precipitation of grain boundary carbides.
• The aging treatment will occur naturally in service at high temperatures.

Thus, the main novelty achieved in this work was to determine the best heat treatment
combination and corresponding parameters in order to optimize the mechanical properties, fatigue
resistance and corrosion behavior of the HK30 cast steel. With a proper combination of heat treatments,
it is possible to achieve the best dispersing effect of grain boundary carbides, combining very good
hardness values with an elevated toughness. Indeed, the first heat treatment steps can improve the
hardness but allow for remaining concentrations of grain boundary carbides; these conditions allow
crack nucleation when subjected to cyclic loads. The dispersion of carbides eliminates this concern,
conferring very good global properties to the components, allowing HK30 cast steel’s application
in heavy-duty vehicle components. For the example presented in this work (an exhaust manifold),
the aging treatment will be carried out in service, allowing for additional cost savings.

Acknowledgments: This work was conducted at Scania CV AB in the scope of the Master’s Degree in
Mechanical Engineering-Materials and Manufacturing Technologies at ISEP-School of Engineering, Polytechnic of
Porto. The author is thankful to Scania for the thesis project opportunity. The author would like to acknowledge
Mimmi Härdeman, Henrik Sieurin and Baohua Zhu from Scania for valuable support and discussions.
Author Contributions: Jorge Santos conceived and designed the experiments, as well as performed the
experiments; Jorge Santos, Francisco J. G. Silva and Ronny Gouveia analyzed the data; Jorge Santos and
Francisco J. G. Silva wrote the paper. Ronny Gouveia revised the manuscript.
Conflicts of Interest: The authors declare no conflict of interest.
Metals 2017, 7, 251 12 of 12

References
1. Schöggl, J.P.; Baumgartner, R.J.; Hofer, D. Improving sustainability performance in early phases of product
design: A checklist for sustainable product development tested in the automotive industry. J. Clean. Prod.
2017, 3, 1602–1617. [CrossRef]
2. Park, D.B.; Huh, M.Y.; Jung, W.S.; Suh, J.Y.; Shim, J.H.; Lee, S.C. Effect of vanadium addition on the creep
resistance of 18Cr9Ni3CuNbN austenitic stainless heat resistant steel. J. Alloys Compd. 2013, 574, 532–538.
[CrossRef]
3. Koa, S.J.; Kim, Y.J. High temperature fatigue behaviors of a cast ferritic stainless steel. Mater. Sci. Eng. A
2012, 534, 7–12. [CrossRef]
4. Kim, Y.J.; Jang, H.; Oh, Y.J. High temperature low cycle fatigue properties of a HF30-type cast austenitic
stainless steel. Mater. Sci. Eng. A 2009, 526, 244–249. [CrossRef]
5. Shi, S.; Lippold, J.C. Microstructure evolution during service exposure of two cast, heat-resisting stainless
steels—HP-Nb modified and 20–32Nb. Mater. Charact. 2008, 59, 1029–1040. [CrossRef]
6. Nunes, F.C.; Dille, J.; Delplancke, J.L.; De Almeida, L.H. Yttrium addition to heat-resistant cast stainless steel.
Scr. Mater. 2006, 54, 1553–1556. [CrossRef]
7. Ekström, M.; Jonsson, S. High-temperature mechanical- and fatigue properties of cast alloys intended for
use in exhaust manifolds. Mater. Sci. Eng. A 2014, 616, 78–87. [CrossRef]
8. Ekström, M. Development of a Ferritic Ductile Cast Iron for Increased Life in Exhaust Applications.
Bachelor’s Thesis, KTH Industrial Engineering Management, Stockholm, Sweden, 7 May 2013.
9. Li, D.; Sloss, C. Comment on High-temperature mechanical and fatigue properties of cast alloys intended for
use in exhaust manifolds by Ekström et al. (Mater. Sci. Eng. A 616 (2014) 78). Mater. Sci. Eng. A 2015, 624,
90–91. [CrossRef]
10. Davis, J. Heat-Resistant Materials; ASM International: Geauga County, OH, USA, 1997; pp. 247–322,
ISBN 978-0-87170-596-9.
11. Sustaita-Torres, I.A.; Haro-Rodríguez, S.; Guerrero-Mata, M.P.; Garza, M.; Valdés, E.; Deschaux-Beaumed, F.;
Colás, R. Aging of a cast 35Cr–45Ni heat resistant alloy. Mater. Chem. Phys. 2012, 133, 1018–1023. [CrossRef]
12. Laird, R.G. Abrasion-Resistant Cast Iron Handbook; ASM International: Geauga County, OH, USA, 2000;
pp. 125–286, ISBN 978-0-87433-224-7.
13. Santos, H. Aços Inoxidáveis Austeníticos; Lesson #14; Faculdade de Engenharia da Universidade do Porto:
Porto, Portugal, 2007. (In Portuguese)
14. Guimarães, A.A.; Mei, P.R. Precipitation of carbides and sigma phase in AISI type 446 stainless steel under
working conditions. J. Mater. Process. Technol. 2004, 155–156, 1681–1689. [CrossRef]
15. Sourmail, T. Precipitation in creep resistant austenitic stainless steels. Mater. Sci. Technol. 2001, 1, 1–14.
[CrossRef]
16. Ha, V.T.; Jung, W.S. Effects of heat treatment processes on microstructure and creep properties of a high
nitrogen 15Cr–15Ni austenitic heat resistant stainless steel. Mater. Sci. Eng. A 2011, 528, 7115–7123. [CrossRef]
17. Wang, J.Z.; Liu, Z.D.; Bao, H.S.; Cheng, S.C. Evolution of precipitates of S31042 heat-resistant steel during
700 ◦ C aging. J. Iron Steel Res. Int. 2013, 20, 113–121. [CrossRef]
18. Kumar, A.; Sharma, A.; Goel, S.K. Effect of heat-treatment on microstructure, mechanical properties and
erosion resistance of cast 23-8-N nitronic steel. Mater. Sci. Eng. A 2015, 637, 56–62. [CrossRef]
19. Davis, J.R. Stainless Steels; ASM International: Geauga County, OH, USA, 1994; pp. 459–474,
ISBN 978-0-87170-503-7.
20. McGuire, M.F. Stainless Steels for Design Engineers; ASM International: Geauga County, OH, USA, 2008; p. 77,
ISBN 978-0-87170-717-8.

© 2017 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 (http://creativecommons.org/licenses/by/4.0/).