B. [ULER et al.: MODIFICATION OF NON-METALLIC INCLUSIONS WITH RARE-EARTH METALS ...
441–447
MODIFICATION OF NON-METALLIC INCLUSIONS WITH
RARE-EARTH METALS IN 50CrMoV13-1 STEEL
MODIFIKACIJA NEKOVINSKIH VKLJU^KOV S KOVINAMI
REDKIH ZEMELJ V JEKLU 50CrMoV13-1
Bla` [uler
1
, Jaka Burja
2
, Jo`e Medved
3
1
SIJ Metal Ravne d.o.o., Koro{ka cesta 14, 2390 Ravne na Koro{kem, Slovenia
2
Institute of Metals and Technology, IMT, Lepi pot 11, 1000 Ljubljana, Slovenia
3
Faculty of Natural Sciences and Engineering, Department of Materials and Metallurgy, A{ker~eva cesta 12, 1000 Ljubljana, Slovenia
Prejem rokopisa – received: 2018-12-13; sprejem za objavo – accepted for publication: 2019-05-06
doi: 10.17222/mit.2018.271
The potential use of the rare-earth metals in clean steel production was investigated. Reactions between the non-metallic
inclusions and the rare-earth metals were investigated in a cold-work tool-steel grade (50CrMoV13-1). A rare-earth metal alloy
misch metal (Ce, La, Nd, Pr) was used for the inclusion modification. Aluminium oxide non-metallic inclusions that have a
negative effect on the mechanical properties usually occur in the investigated steel. The experimental melts were made in a
vacuum induction furnace with different rare-earth metal additions (50, 150, 340, 950 and 2900) ppm. The laboratory cast ingots
were hot forged and analysed with light and electron microscopy. It was found that misch metal additions influence the size,
composition and type of the non-metallic inclusions in the steel. Most importantly, very small inclusions are formed at lower
rare-earth metal additions (up to 340 ppm), but at high additions large non-metallic inclusions are formed (2900 ppm).
Keywords: non-metallic inclusions, inclusion modification, rare earth metals, clean steel
V {tudiji smo raziskali mo`nost uporabe redkih zemelj za izdelavo jekel z nizko vsebnostjo nekovinskih vklju~kov. Reakcije
med redkimi zemljami in nekovinskimi vklju~ki so bile raziskane v orodnem jeklu za delo v hladnem 50CrMoV13-1. Za
dodatek redkih zemelj (Ce, La, Nd, Pr) smo uporabili me{anico redkih zemelj t.i. misch metal. Jeklo, ki smo ga uporabili v
{tudiji, je zaradi svoje kemi~ne sestave nagnjeno k tvorbi aluminatnih nekovinskih vklju~kov, kateri pa negativno vplivajo na
lastnosti jekla. Eksperimentalni del je potekal na vakuumski indukcijski pe~i, kjer smo v jeklo dodajali razli~ne koli~ine redkih
zemelj (50, 150, 340, 950 in 2900) ppm. Ulite ingote smo prekovali v palice, slednje smo analizirali s pomo~jo svetlobnega in
elektronskega mikroskopa. Ugotovljeno je bilo, da dodatek redkih zemelj vpliva na velikost, kemi~no sestavo in tipe
nekovinskih vklju~kov v jeklu. Zelo majhni nekovinski vklju~ki so nastali pri ni`jih vsebnostih redkih zemelj, za razliko od
velikih nekovinskih vklju~kov, ki so nastali v jeklu z visoko vsebnostjo redkih zemelj.
Klju~ne beside: nekovinski vklju~ki, modifikacija vklju~kov, kovine redkih zemelj, ~istost jekla
1 INTRODUCTION
Cold-work tool steels, as the name suggests, are used
to form or cut materials at room temperature. They have
great wear resistance, high hardness, good impact
toughness and they are also heat softening resistant. The
50CrMoV13-1 steel grade can be air quenched and is
mainly used for cold-work applications. Typical applica-
tions include chisels, punches, river sets, driver bolts, as
well as tools for hot punching and shearing. Typically,
the 50CrMoV13-1 is air quenched from 950 °C. The
hardness after quenching reaches up to 61 HRC. The
tempering is usually carried out in the temperature range
from 150 °C to 400 °C, depending on the hardness
requirements.
The content and type of the non-metallic inclusions
can influence the mechanical properties of the steel and
lifetime of the steel components;
1–3
this is especially true
for high-hardness steels (such as cold-work tool steels).
The term "clean steel" refers to a low level of non-me-
tallic inclusions. The requirements vary with the steel
grade and its end use. Typically, oxides are the most
common non-metallic inclusions in low-sulphur steels.
Therefore, a low total oxygen content is required for a
low-oxide, non-metallic inclusion content.
4,5
Oxide
non-metallic inclusions are typicaly composed from
aluminium, calcium, magnesium, manganese and silicon
oxides.
The aluminium oxide (Al
2
O
3
) non-metallic inclusions
are the result of Al deoxidation; they are solid at
steelmaking temperatures and cause problems during
steel casting, like nozzle clogging during continuous
casting.
6
Brittle Al
2
O
3
non-metallic inclusions break into
fragmented stringers during deformation, leaving sharp,
hard particles, that deteriorate the mechanical and fatigue
properties of steel.
4
However, Mn and Si de-oxidation
cause deformable, less-harmful inclusions, but the result
is unfortunately, a higher total oxygen content, as they
have worse de-oxidation properties. In order to obtain
the desired low oxygen content in the steel melt, certain
Al levels need to be guaranteed.
7,8
The harmful effect of
the aluminium oxide inclusions can be reduced by
modification. The calcium (Ca) treatment is the most
Materiali in tehnologije / Materials and technology 53 (2019) 3, 441–447 441
UDK 620.1:669.1:669.85/.86 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 53(3)441(2019)
*Corresponding author e-mail:
blaz.suler@metalravne.com

widely used technique to modify the aluminium oxide
non-metallic inclusions. The Ca treatment can modify
the high-melting-temperature inclusions into low-tempe-
rature melting inclusions that are present as liquid in the
steel melt. The liquid inclusions can agglomerate more
easily and grow, this eases the floatation and are also
deformable with a round shape.
3,4,9–11
Alternatively, other
elements such as rare-earth metals or zirconium can be
used to modify the shape, number and size of the
non-metallic inclusions.
12–14
In this research we used
rare-earth metals (misch metal alloy) with the aim to
modify the non-metallic inclusions and thus achieve
"clean steel" production.
2 MATERIALS AND EXPERIMENTAL
PROCEDURE
Bars of tool steel 50CrMoV13-1 for hot-work and
cold-work applications and misch metal, with the che-
mical composition given in Tables 1 and 2, were used in
the experimental work. The bars were melted in a
vacuum induction melting furnace (VIM) with 20 kg
capacity. The weight of materials used in each batch was
around 14.5 kg. Six different melts were made with
different mass fractions of rare-earth metals (REMs).
Before the melting the vacuum chamber of the induction
furnace was first vacuumed to5×10
–2
mbar, purged with
Ar, vacuumed again to5×10
–2
mbar, and then Ar was
flooded to 300 mbar. The melting was carried out in a
zirconia crucible. The REMs were added after the steel
was completely melted through a special pre-vacuumed
charger, that is a part of the VIM furnace.
Table 2: Chemical composition misch metal in mass fractions (w/%)
Ce La Nd Pr Oth. REM
45-55 20-25 15-20 5-8 Bal.
The first melt consisted only of the 50CrMoV13-1
steel with no REM addition. The second melt was made
with an addition of 0.1 g of REM per 1 kg of steel, the
third with 0.5 g per kg, the fourth with 1 g per kg, the
fifth with 2 g per kg, and finally the sixth melt with 5 g
per kg of steel. After the REM addition the melt was
homogenized for 5 minutes before casting into an ingot
under an argon protective atmosphere. The cast ingots’
dimensions were roughly 80 mm
3
×80mm
3
× 300 mm
3
.
The as-cast ingots were heated to 1175 °C for 1 h, after
that they were hot forged into 30 mm round bars. The
bars were air cooled to room temperature and annealed
a t6 0 0° Cf o r2h( t oachieve low hardness for easier
machining). The specimens for chemical analysis and
metallographic investigations were cut out of the bars in
the longitudinal direction. The C and S contents were
analysed with a LECO CS800, while the other elements
were analysed with a Coupled Plasma-Optical Emission
Spectrometer (ICP-OES Agilent 720). The microscopic
examinations of the non-metallic inclusions based on the
standard DIN50602 were made on an optical microscope
Zeiss Axio Imager 1m. The electron microscopy was
carried out with a scanning electron microscope Jeol
JSM6510 with a X-ACT EDS using the INCA automatic
non-metallic inclusion analyser. The automatic SEM
non-metallic inclusion analysis was made on each steel
sample (1-6). The analytical program automatically
found, measured and made an EDS analysis for each
inclusion ona9m m
2
area. Only inclusions with a
diameter of 0.98 μm and larger were analysed.
3 RESULTS AND DISCUSSION
3.1 Melting
The results of the chemical analysis for the experi-
mental batches are given in Table 3. The amount of
REM is continuously increasing in the chemical analysis
from sample 1 to 6 (higher amounts of REM added to the
melt). The REM yield was around 50 %. The relatively
low yield is attributed to the high vapour pressure of the
REM at steelmaking temperatures (1600 °C), also a thin
oxide film formed on the melt surface after the REM
addition. The film formation was observed during the
REM alloying.
Table 3: Results of the rare-earth metal chemical analysis in the steel
melt
Sample w/% Ce w/% La
Total ppm of
REM
1 0.0 0.0 0
2 0.004 0.001 50
3 0.010 0.005 150
4 0.027 0.007 340
5 0.070 0.025 950
6 0.200 0.090 2900
3.2 Optical microscopy
The DIN50602 standard method K was used for the
non-metallic inclusions’ evaluation under an optical
microscope (OM). The results of the non-metallic inclu-
sion examinations are given in Table 4. The inclusions in
sample 1 are classified as the AO (aluminate oxide) type
in the DIN50602 standard (fragmented stringers). The
inclusions in sample 6 are classified as AO and OG
(oxide globular) type.
Table 5 clearly shows that the cleanest steel samples
are 2, 3, 4 and 5. The first sample has no REM additions
B. [ULER et al.: MODIFICATION OF NON-METALLIC INCLUSIONS WITH RARE-EARTH METALS ...
442 Materiali in tehnologije / Materials and technology 53 (2019) 3, 441–447
Table 1: Chemical composition 50CrMoV13-1 in mass fractions (w/%)
C Si Mn P S Cr Ni Cu Mo V Ti Nb Al Fe
0.48 0.25 0.54 0.018 0.005 3.30 0.18 0.09 1.60 0.220 0.0020 0.0050 0.014 Bal.

and contains small inclusions in the K1 and K0 size
category, according to DIN50602 standard. The K1 and
K0 size categories are expected for VIM refining.
Samples 2 to 5 have inclusions that are too small to fall
into any category. However, sample 6 has an increased
amount of non-metallic inclusions that are large enough
to fit into the K4 category. Figure 1 shows an OM
example of non-metallic inclusions in sample 1 and
sample 3.
Table 4: Analysis results of non-metallic inclusion – DIN50602 –
method K
REM (ppm) K0 K1 K2 K3 K4
094000
5 000000
1 5 000000
3 4 000000
9 5 000000
2900 468 468 468 368 221
3.3 Electron microscopy
The automatic SEM non-metallic inclusion analysis
gave us the size distribution, the amount and the chemi-
cal compositions of the non-metallic inclusions. The
total surface area of the non-metallic inclusions is pre-
sented in Figure 2. The area of the non-metallic inclu-
sions in the unmodified sample is 1492 μm
2
. When
REMs are added the non-metallic inclusion area is
gradually reduced and reaches a minimum of 376 μm
2
at
150 ppm REM addition. After further increasing the
REM content, the non-metallic inclusion area rises, at
950 ppm of REM the area is 1643 μm
2
. The non-metallic
inclusion area at 2900 ppm of REM reaches a staggering
10,884 μm
2
. The number and distribution of the non-me-
tallic inclusions for all samples are shown in Figure 3.
The vast majority of inclusions in all the samples are
smaller than 2 μm (Figure 3). The REM additions
further increase the fraction of small inclusions. There
are still some inclusions above 4 μm when 50 ppm
REMs are added. But there are no inclusions above 4 μm
when 150 ppm to 950 ppm REMs are added. As men-
tioned above, further additions (2900 ppm) result in an
increase of both the number and size of the non-metallic
inclusions.
The addition of REMs changed the inclusion surface
area compared to the unmodified sample, 50 ppm REM
reduced it by 24 %, 150 ppm REM by 75 %, 340 ppm by
27 %, but 950 and 2900 ppm of REM increased the area
by 10 % and 728 %, respectively.
The chemical composition and the type of the non-
metallic inclusions also changes with the REM additions.
The REM additions from 50 ppm to 340 ppm increase
the number of oxysulphide inclusions (Figure 4). When
950 ppm of REM are added, the number of oxides in-
creases. This can be explained by the limited concentra-
tion of sulphur (50 ppm, see Table 1). The high oxide
number is attributed to the lower oxidation states of the
REMs as they form different oxides.
14
Another diffe-
rence between the REM modified and unmodified inclu-
B. [ULER et al.: MODIFICATION OF NON-METALLIC INCLUSIONS WITH RARE-EARTH METALS ...
Materiali in tehnologije / Materials and technology 53 (2019) 3, 441–447 443
Figure 3: Number and distribution of inclusions for each sample
Figure 1: Optical micrographs of non-metallic inclusions in a) sample
1 – no REM and b) sample 3 – 150 μg/g REM
Figure 2: Area of non-metallic inclusion for all samples

sions is that the oxysulphides in sample 1 (0 μg/g REM)
are in fact complex (multi-phase) inclusions, as is com-
mon for steels,
15,16
while the REMs form oxysulphide
compounds.
14,17
The REM non-metallic inclusions have
high atomic numbers Z, so they appear bright in the
backscattered electron images. The type of REM inclu-
sions (oxide/sulphide/oxysulphide) is difficult to deter-
mine from the electron microscope images, as they also
change little in morphology. Examples of non-metallic
inclusions, (a) aluminium oxide in sample 1, (b) REM
oxide in sample 3, (c) REM sulphide in sample 4 and (d)
REM oxysulphide in sample 3 with the associated EDS
analyses are shown in Figure 5.
The average compositions of the oxide non-metallic
inclusions are shown in Figure 6a. Sample no. 1 con-
tains mainly aluminium oxides with some Si, Mg and
Ca. Silicon oxide is the least stable of the analysed
oxides and is a common de-oxidation agent. Our samples
B. [ULER et al.: MODIFICATION OF NON-METALLIC INCLUSIONS WITH RARE-EARTH METALS ...
444 Materiali in tehnologije / Materials and technology 53 (2019) 3, 441–447
Figure 5: a) aluminium oxide in sample 1 (0 ppm REM), b) rare-earth metal oxide in sample 3 (150 ppm REM), c) rare-earth metal sulphide in
sample 4 (340 ppm REM) and d) rare-earth metal oxysulphide in sample 3 (150 ppm REM), with EDS spectra in w/%
Figure 4: Type of non-metallic inclusions in the samples

did not contain silicon-based inclusions, only SiO
2
dis-
solved in the alumina. MgO·Al
2
O
3
spinel-type inclusions
are commonly found in Al-killed steels, and are formed
through refractory/slag/melt interactions.
15
In our
particular case the spinel-type inclusions were found as
part of the fractured alumina stringers (the low magne-
sium content is shown in Figure 6a). Ca is also common
and is either the result of Ca modification or slag/melt
interaction. There were not any Ca-modified inclusions
found in our case. The composition of the oxide inclu-
sions is significantly changed after the REMs addition.
The REM oxides become dominant, as a result of the
thermodynamic stability.
The additions of REM cause the formation of oxysul-
phide inclusions, their number increases with the REM
content increase. However, the maximum increase of
REM oxysulphides occurred at 340 ppm REMs addition
(sample 4), a further increase of the REMs content re-
sulted in a sharp decline in oxysulphide inclusion
numbers (Figure 6b). The REMs addition also resulted
in the formation of REM sulphides, replacing the domi-
nant MnS. The addition of 150 ppm REM completely
prevented the formation of the MnS inclusions, as seen
in Figure 6c. Further REM additions resulted in an
increase of REM content in the non-metallic inclusions.
The ratio between REM/S increased, due to the forma-
tion of different type of REM sulphides.
3.4 Thermodynamic calculations
Table 5: Enthalpy and entropy values for Ce oxidation reactions
12,18
Reaction ÄH (J) ÄS (J)
2[Al]+3[O]=Al
2
O
3
(s) –1205090 –387.72
2[Ce]+2[O]+[S]=Ce
2
O
2
S(s) –1493010 –438.9
[Ce]+[S]=CeS(s) –422783 –120.58
20[Ce]+2[O]=CeO 2(s) –854274.7 –249.06
2[Ce]+3[O]=Ce
2
O
3
(s) –1431090 –360.06
2[Ce]+3[S]=Ce
2
S
3
(s) –1074584 –328.24
As cerium is the dominant element in misch metal
(Table 2), the REM additions can be described as an
increase in Ce. Therefore, Ce was used for the thermo-
dynamic calculations. The values for the reaction
enthalpies and entropies are given in Table 5. The
element interaction coefficients in a diluted iron melt at
1600 °C are given in Table 6. The interaction
coefficients show that S and O have a major impact on
lowering the Ce activity. The calculated Gibbs free
energy values for the reaction in Table 5 at (50, 150 and
340) ppm of Ce at (1, 10 and 100) ppm of oxygen and at
1873 K are given in Table 7. The thermodynamic
calculations in Table 7 for the (50, 150 and 340) ppm of
Ce are presented, because the experimental results are
the most promising in terms of inclusion content.
Table 6: First-order interaction coefficients#,inliquid iron at 1873 K
Ce S O Al
Ce 0,0039
14
–40
14
–106
19
/
S –9,1
14
–0.046
14
–0.27
20
0.035
20
O –64
14
–0.133
20
–0.17
14
–3.9
20
Al / 0.03
20
–6.6
20
0.045
20
The thermodynamic calculations show that Ce has a
high affinity for oxygen and preferably reacts with
oxygen. Ce sulphide formation is only enabled at low
oxygen contents (10 ppm). On the other hand, Ce
oxysulphides formation is possible even at high oxygen
contents (100 ppm). This means the REM addition will
primarily react with oxides in the melt (SiO
2
,Al
2
O
3
) and
replace them with REM-type oxides.
21
At low REM
contents (50 ppm) the oxygen consumed the majority of
the REM, so only REM oxides and oxysulphides are
formed. But, when the REM content was increased to
150 ppm, there was enough REM left over to enable the
formation of REM sulphides (as shown in Figure 6c).
Table 7 also shows that the REM content itself is not as
crucial to the reaction as is the oxygen activity/content.
This means that the oxygen contents will dictate the
REM non-metallic inclusion type formation. In
comparison to the Ce oxidation, the aluminium oxide
formation requires relatively high oxygen contents (100
ppm), even at low Ce concentrations (50 ppm). The
experimental results presented in Figure 6a confirm that
aluminium oxides are the first to be reduced by the REM
additions.
B. [ULER et al.: MODIFICATION OF NON-METALLIC INCLUSIONS WITH RARE-EARTH METALS ...
Materiali in tehnologije / Materials and technology 53 (2019) 3, 441–447 445
Figure 6: Chemical composition of: a) oxide non-metallic inclusions, b) oxysulphide non-metallic inclusions, c) sulphide non-metallic inclusions

4 CONCLUSIONS
Rare-earth metal (REM) additions cause changes in
the non-metallic inclusion chemistry, size and distribu-
tion. Three types of REM-based non-metallic inclusions
are formed: oxides, sulphides and oxysulphides. The
modification and refining of the non-metallic inclusions
confirm that REMs show potential in "clean steel" pro-
duction.
The maximum non-metallic inclusion area-surface
reduction, compared to the unmodified sample, was 75
% at 150 ppm REM.
Lower additions of REMs (50 ppm) will modify
oxides and oxysulphides, but not sulphides. Higher REM
(150 ppm) will result in the elimination of all non-
REM-based inclusions (Al
2
O
3
, MnS, SiO
2
, MgO·Al
2
O
3
).
REM additions produced smaller/finer non-metallic
inclusions. The number of detectable inclusions (<0.98
μm) increased, while the total non-metallic inclusion
area decreased, the minimum was reached at 150 ppm
REM.
Higher REM additions resulted in more and larger
non-metallic inclusions, worsening the steel’s
cleanliness. Very high REM additions (2900 ppm REM)
greatly increased the number and size of the non-metallic
inclusions.
5 REFERENCES
1
R. Arreola-Herrera, A. Cruz-Ramírez, J. E. Rivera-Salinas, J. A.
Romero-Serrano, R. G. Sánchez-Alvarado, The effect of non-metal-
lic inclusions on the mechanical properties of 32 CDV 13 steel and
their mechanical stress analysis by numerical simulation, Theor.
Appl. Fract. Mech., 94 (2018), 134–146, doi:10.1016/j.tafmec.2018.
01.013
2
M. Jiang, X. Wang, B. Chen, W. Wang, Laboratory study on evo-
lution mechanisms of non-metallic inclusions in high strength
alloyed steel refined by high basicity slag, ISIJ Int., 50 (2010),
95–104, doi:10.2355/isijinternational.50.95
3
L. Zhang, B. G. Thomas, State of the art in the control of inclusions
during steel ingot casting, Metall. Mater. Trans. B, 37 (2006),
733–761, doi:10.1007/s11663-006-0057-0
4
H. Y. Tang, Y. Wang, T. Wu, J. S. Li, S. F. Yang, Characteristics
analysis of inclusion of 60Si2Mn-Cr spring steel via experiments and
thermodynamic calculations, Ironmak. Steelmak, 44 (2017)5 ,
377–388, doi:10.1080/03019233.2016.1212563
5
K. Steneholm, N. A. I. Andersson, A. Tilliander, P. G. Jönsson, The
role of process control on the steel cleanliness, Ironmak, Steelmak., 0
(2016) 1–11, doi:10.1080/03019233.2016.1245917
6
F. Tehovnik, J. Burja, B. Arh, M. Knap, Submerged entry nozzle
clogging during continuous casting of al-killed steel, Metalurgija, 54
(2015)
7
Z. Deng, M. Zhu, Deoxidation mechanism of al-killed steel during
industrial refining process, ISIJ Int., 54 (2014), 1498–1506,
doi:10.2355/isijinternational.54.1498
8
M.-K. Paek, J.-M. Jang, Y.-B. Kang, J.-J. Pak, Aluminum deoxi-
dation equilibria in liquid iron: Part I: Experimental, Metall. Mater.
Trans. B, 46 (2015) 4, 1826–1836, doi:10.1007/s11663-015-0368-0
B. [ULER et al.: MODIFICATION OF NON-METALLIC INCLUSIONS WITH RARE-EARTH METALS ...
446 Materiali in tehnologije / Materials and technology 53 (2019) 3, 441–447
Table 7: Calculated Gibbs free energy for reaction products (J), at 1873 K, 50 ppm S, 200 ppm Al
50 ppm Ce
O( w/%) 2/3[Al]+[O]=1/3Al
2
O
3
[Ce]+[O]+1/2[S]=1/2Ce
2
O
2
S [Ce]+[S]=CeS
0.0001 44963.84 –76555.1 –22842.2
0.001 9272.361 –108965 –19414.5
0.01 –25092.5 –110515 14862.93
O( w/%) 1/2[Ce]+[O]=1/2CeO
2
2/3[Ce]+[O]=2/3Ce
2
O
3
2/3[Ce]+[S]=1/3Ce
2
S
3
0.0001 8804.683 –34579.8 –9173.47
0.001 –25319.2 –68133.9 –6885.41
0.01 –44008.1 –81124.4 15995.21
150 ppm Ce
O( w/%) 2/3[Al]+[O]=1/3Al
2
O
3
[Ce]+[O]+1/2[S]=1/2Ce
2
O
2
S [Ce]+[S]=CeS
0.0001 67900.73 –69088.4 –36681.8
0.001 32209.25 –101498 –33254
0.01 –2155.58 –103049 1023.372
O( w/%) 1/2[Ce]+[O]=1/2CeO
2 2/3[Ce]+[O]=2/3Ce 2O 3 2/3[Ce]+[S]=1/3Ce 2S 3
0.0001 23191.12 –23043.5 –17312.7
0.001 –10932.8 –56597.6 –15024.7
0.01 –29621.7 –69588.1 7855.954
340 ppm Ce
O( w/%) 2/3[Al]+[O]=1/3Al 2O 3 [Ce]+[O]+1/2[S]=1/2Ce 2O 2S [Ce]+[S]=CeS
0.0001 111480.8 –35149.4 –43224.6
0.001 75789.33 –67559.4 –39796.8
0.01 41424.5 –69109.6 –5519.44
O( w/%) 1/2[Ce]+[O]=1/2CeO
2 2/3[Ce]+[O]=2/3Ce 2O 3 2/3[Ce]+[S]=1/3Ce 2S 3
0.0001 60401.53 12043.67 –19609.1
0.001 26277.64 –21510.4 –17321
0.01 7588.741 –34500.9 5559.595

9
R. Dekkers, B. Blanpain, P. Wollants, F. Haers, C. Vercruyssen, B.
Gommers, Non-metallic inclusions in aluminium killed steels,
Ironmak. Steelmak., 29 (2002), 437–444, doi:10.1179/
030192302225004584
10
Y. Ren, L. Zhang, S. Li, Transient evolution of inclusions during cal-
cium modification in linepipe steels, ISIJ Int., 54 (2014), 2772–2779,
doi:10.2355/isijinternational.54.2772
11
Y. Wang, S. Sridhar, M. Valdez, Formation of CaS on Al2O3-CaO
inclusions during solidification of steels, Metall. Mater. Trans. B, 33
(2002), 625–632, doi:10.1007/s11663-002-0042-1
12
Y. Li, C. Liu, T. Zhang, M. Jiang, C. Peng, Liquid inclusions in
heat-resistant steel containing rare earth elements, Metall. Mater.
Trans. B, 48 (2017), 956–965, doi:10.1007/s11663-016-0873-9
13
M. Kole`nik, J. Burja, B. S. Bati~, A. Nagode, J. Medved, De-oxi-
dation of Pk942 steel with Ti and Zr, Mater. Tehnol., 51 (2017),
1031–1036, doi:10.17222/mit.2017.082
14
F. Pan, J. Zhang, H.-L. Chen, Y.-H. Su, Y.-H. Su, W.-S. Hwang,
Thermodynamic calculation among cerium, oxygen, and sulfur in
liquid iron, Sci. Rep., 6 (2016) 35843, doi:10.1038/srep35843
15
J. H. Shin, J. H. Park, Formation mechanism of oxide-sulfide com-
plex inclusions in high-sulfur-containing steel melts, Metall. Mater.
Trans. B, 49 (2018), 311–324, doi:10.1007/s11663-017-1152-0
16
L. Krajnc, P. Mrvar, J. Medved, Thermodynamic characterization of
multiphase non-metallic inclusions in re-sulphurised steel grades,
Mater. Tehnol., 48, (2014), 923–929
17
K. Zeli~, J. Burja, P. J. McGuiness, M. Godec, Effect of rare earth
elements on the morphology of eutectic carbides in AISI D2 tool
steels: Experimental and modelling approaches, Scientific Reports,
8( 2018), 9233, doi:10.1038/s41598-018-27658-w
18
Y. Li, C. Liu, C. Li, M. Jiang, A coupled thermodynamic model for
prediction of inclusions precipitation during solidification of
heat-resistant steel containing cerium, J. Iron Steel Res. Int., 22
(2015), 457–463, doi:10.1016/S1006-706X(15)30027-3
19
Q. Han, C. Xiang, Y. Dong, S. Yang, D. Chen, Equilibria between
the rare earth elements, oxygen and sulfur, in molten iron, Metall.
Mater. Trans. B, 19 (1988) 3, 409–418
20
G. K. Sigworth, J. F. Elliot, The thermodynamics of liquid dilute iron
alloys, Met. Sci. B, (1974), 298–310
21
S. K. Kwon, J. S. Park, J. H. Park, Influence of refractory-steel inter-
facial reaction on the formation behavior of inclusions in
Ce-containing stainless steel, ISIJ Int., 55 (2014), 2589-2596,
doi:10.2355/isijinternational.ISIJINT-2015-125
B. [ULER et al.: MODIFICATION OF NON-METALLIC INCLUSIONS WITH RARE-EARTH METALS ...
Materiali in tehnologije / Materials and technology 53 (2019) 3, 441–447 447