Thermodynamic Assessment of the Fe-B System in the Ssol5 and User Databases

Mateusz Szymanski et al. Int. Journal of Engineering Research and Application www.ijera.com
ISSN : 2248-9622, Vol. 7, Issue 1, ( Part -1) December 2017, pp.59-62
www.ijera.com 59 | P a g e
Thermodynamic Assessment of the Fe-B System in the Ssol5 and
User Databases
Mateusz Szymanski*, Viera Homolová**, Marcin Leonowicz*
*(Faculty of Materials Science and Engineering, Warsaw University of Technology, 141 Woloska Street, 02-507
Warsaw, Poland)
** (Institute of Materials Research, Slovak Academy of Sciences, Watsonova 47, Košice 040 01 Slovakia)
ABSTRACT
Thermodynamic assessment of Fe-B system, including phase diagram, Gibbs energy, enthalpy, heat capacity
and activity, was performed in the ThermoCalc software ver. 4.1 (Sweden). Two databases were used: the
commercial SSOL5 database for solid solutions (substitutional approach) and the USER made database based
on work of T. Van Rompaey et al. (intersticial approach). Results obtained were compared with experimental
data gathered from work of M. Van Ende et al. In low boron regime the curve of the Fe-B phase diagram is
represented more reliable in the USER database. However, temperatures of the phase transformations are
calculated with more accuracy in the SSOL5 database. For boron content higher than 0.3 mole fraction phase
transformation temperatures are better assessed in the USER database, except for melting point of the Fe2B
phase. Gibbs energy, enthalpy and heat capacity of the FeB and the Fe2B are difficult to evaluate because
experimental data are spread and inaccurate. Activities of iron and boron in liquid Fe-B alloy, calculated at
selected temperatures, are almost identical for both databases. paramount importance.
Keywords: thermocalc
I. INTRODUCTION
Iron and Boron are important constituents
of steels and magnetic materials. There is a need to
simulate properties and behavior of these products.
The milestone papers allowing for calculation of the
Nd-Fe-B system were provided by Hallemans et al.
in 1994 [1] and 1995 [2]. Not so long ago, a
significant article was published by Van Ende et al.
who provided a critical review of all theoretical and
experimental data available for the Nd-Fe-B system
and proposed improved parameters for calculations
[3]. Very recently Zhou et al. investigated
thermodynamic description of the Nd-Fe-B system
including metastable phases Fe3B, Fe17Nd2B and
Fe23Nd2B3 which have effect on magnetic properties
of dual-nanoscale-phase nanocomposite magnets [4].
However, obtained phase diagrams cannot be easily
compared because one must suspend stable phases
from calculations, e.g. Fe2B phase, in order to
display metastable one, e.g. Fe3B phase. In order to
predict ternary system one has to properly assess
binary one first [5]. Therefore, this work is focused
on the Fe-B system which is calculated using
different approaches. In the SSOL5 database iron
and boron are recognized as substitutional solutions,
while in the USER database boron is modelled as an
interstitial element in the FCC and BCC solid
solutions.
II. EXPERIMENTAL
Calculations were performed using
ThermoCalc ver. 4.1 software equipped with two
databases: commercial SSOL5 for solid solutions
(substitutional approach) based on Hallemans et al.
[1,2] and USER made database with data from work
of T. Van Rompaey [6]. Conditions used in
calculations (reference states, temperatures etc.)
were based on those used in Van Ende’s et al. [3]
thus allowing for easy comparison. Experimental
data, gathered by various authors, were read from
Van Ende’s et al. paper using DataThief III software
developed by B. Tummers [7].
III. RESULTS AND DISCUSSION
Boron-rich part of calculated Fe-B phase
diagram is presented together with experimental data
in Fig. 1. For boron content up to 0.3 mole fraction
transformation temperatures are too high in USER
database (e.g. difference in eutectic transformation at
0.16 B is higher by more than 20℃ between the
databases). In contrary, transformation temperatures
for boron content more than 0.3 mole fraction are
reduced in the USER database (e.g. melting point of
FeB phase is almost 30℃ lower comparing with the
SSOL5) what fits better experimental data. Melting
point of the FeB phase, calculated at 1651.1℃
byVan Ende et al., is reduced in both databases:
1632.6℃ in the SSOL5 and 1602.8℃ in the USER.
Unfortunately, experimental data for the FeB
compound varies significantly and one has to choose
among the results.
RESEARCH ARTICLE OPEN ACCESS

Fig. 1 Phase diagrams of the Fe-B system calculated
in the SSOL5 (solid line) and the USER (dotted line)
database with experimental data (triangles) obtained
by various authors [3].
An iron-rich corner of calculated the Fe-B
phase diagram is presented together with
experimental data in Fig. 2. A single phase region of
the FCC phase is clearly visible only in the USER.
Indeed, the FCC phase appears in the SSOL5 but has
extremely narrow existence’s range, making it
practically invisible. This difference, including
experimental results, should be considered in a favor
of the USER database.
Fig. 2 Fe-rich corner of the Fe-B binary phase
diagrams calculated in the SSOL5 (solid line) and
the USER (dotted line) database with experimental
data (triangles) obtained by various authors [3].
Despite the FCC phase exists in the SSOL5
database, it is virtually invisible in the left side of the
diagram.
The heat capacity of FeB and Fe2B versus
temperature is presented together with experimental
data in Fig. 3. Simulation for FeB calculated in the
SSOL5 should be slightly elevated (around +5
J/mol-K) at temperatures below 1300 K and slightly
reduced at higher temperatures in order to cover the
experimental data. These adjustments are actually
what the USER database provides. Peaks denote
Curie temperature. Moreover, neither of the database
fits experimental points measured above 1900 K.
This part of curves should be elevated around +15
J/mol-K.
The heat capacity of Fe2B calculated in the
SSOL5 covers experimental data well. Simulation
performed in the USER lies too high and should be
reduced. There are no experimental data for Fe2B at
temperatures above 1500 K so this part of curves
cannot be verified.
Fig. 3 Heat capacity of FeB and Fe2B calculated in
the SSOL5 and the USER database using
hfr(phase).t formula at standard reference states
(SER). Experimental data obtained by various
authors (squares for FeB and asterisks for Fe2B) [3].
Enthalpy of formation of FeB and Fe2B
calculated in the SSOL5 and the USER database is
presented together with experimental data in Fig. 4.
First of all, one should notice experimental points
are spread and uncertain, hindering evaluation.
Indeed, simulations obtained in the SSOL5 are
closer to experimental data, both databases should be
improved by elevating enthalpy values at
temperatures above 1500 K for both FeB and Fe2B.
Experimental points measured between 1000-1300K
do not fit any calculation at all including database
improved by Van Ende.
Fig. 4 Enthalpy of formation of FeB and Fe2B
calculated in the USER and the SSOL5 database
with experimental data (squares for FeB and
triangles for Fe2B) obtained by various authors [3].
The reference states are set as recommended: iron in
BCC and boron in Bbeta at default temperature.

The Gibbs energy of FeB and Fe2B
calculated in the USER and the SSOL5 database is
presented together with experimental data in Fig. 5.
Experimental data are spread, inaccurate and there
are no measuring points below 1000 K. Calculations
provided in both databases cover experimental data
at temperatures between 1400-1700 K for Fe2B but
not for FeB. At temperatures between 1000-1300 K
simulations for both databases should be reduced by
around -10 kJ/mol in order to fit experimental points
for both phases. Although. calculated curves are
generally lower for the USER database value’s
reduction should be greater.
Fig. 5 Gibbs free energy of FeB and Fe2B calculated
in the SSOL5 and the USER database with
experimental data (squares for FeB and triangles for
Fe2B) obtained by various authors [3]. Reference
states are: BCC for iron and BETA_RHOMBO_B
for boron.
The activity of Fe and B in liquid Fe-B
alloy calculated at selected temperatures is presented
with experimental data in Fig. 6-8 for the SSOL5
and in Fig. 9-11 for the USER database,
respectively. One could see the activities of Fe and B
are calculated almost identically in the two
databases. Activity of Fe calculated in the USER
database for around 0.5 boron content is slightly
elevated at lower temperature (1673 K) and slightly
reduced at higher temperatures (1823 K and 1873
K). The iron’s activity curve at 1923 K seems to be
the same for both databases. On the other hand,
activity of B calculated in the USER database is
slightly elevated at higher temperatures (1773 K and
1823 K) and slightly reduced at lower temperature
(1673 K).
Fig. 6 Activity of Fe in liquid Fe-B alloy calculated
at selected temperatures in the SSOL5 database with
experimental data [3].
Fig. 7 Activity of Fe in liquid Fe-B alloy calculated
at selected temperatures in the USER database with
Fig. 8 Activity of B in liquid Fe-B alloy calculated
at selected temperatures in the SSOL5 database with

Fig. 9 Activity of B in liquid Fe-B alloy calculated
at selected temperatures in the USER database with
Fig. 10 Activity of B at low boron content in liquid
Fe-B alloy calculated at selected temperatures in the
SSOL5 database with experimental data [3].
Fig. 11 Activity of B at low boron content in liquid
Fe-B alloy calculated at selected temperatures in the
USER database with experimental data [3].
IV. CONCLUSION
The assessment of the Fe-B binary system,
including calculation of phase diagrams and
determining thermodynamic parameters, was carried
out in the Thermo-Calc software ver. 4.1. Two
databases were used: the commercial database for
solid solutions SSOL5 (substitutional approach) and
the USER database (intersticial approach). The
assessments were compared each other and to
experimental data collected by Van Ende et al. Each
of the database has some advantages and
disadvantages, depending on the property calculated.
Some differences are visible only in specific
composition ranges. In addition, the experimental
data is spread and inaccurate what hinders the
evaluation. The following conclusions, referring to
the Thermo-Calc calculations, could be drawn:
I. Phase diagram optimization might be in
contrary to assessment of thermodynamic properties.
One must improve both the shape of the curve (e.g.
transformation temperatures) and the
thermodynamic functions (Gibbs free energy,
enthalpy of formation and heat capacity).
II. Instead of creating new databases it
might be beneficial to modify current databases by
amending selected parameters.
III. Reference states should be established
in order to allow for unmistakable comparison
between models and databases.
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the
founding of research fellowship provided by the
KMM-VIN community. The present work was
supported by Slovak Grant Agency (VEGA) under
the project No.2/0153/15.
REFERENCES
[1]. B. Hallemans, P. Wollants, J.R. Roos,
Thermodynamic re-assessment and
calculation of the Fe-B phase diagram, Z
Metallkde, 85, 1994, 676–682.
[2]. B. Hallemans, P. Wollants, J.R. Roos,
Thermodynamic assessment of the Fe-Nd-B
phase diagram, Journal of Phase
Equilibria, 16(2), 137–149.
[3]. M.A. Van Ende, I.H. Jung, Critical
thermodynamic evaluation and
optimization of the Fe-B, Fe-Nd, B-Nd and
Nd-Fe-B systems, Journal of Alloys and
Compounds, 548, 2013, 133–154.
[4]. G.J. Zhou, Y. Luo, Y. Zhou,
Thermodynamic Reassessment of the Nd-
Fe-B Ternary System, Journal of Electronic
Materials, 45(1), 2015, 418–425.
[5]. A. Kroupa, Modelling of phase diagrams
and thermodynamic properties using
Calphad method - Development of
thermodynamic databases, Computational
Materials Science, 66, 2013, 3–13.
[6]. V.T. Rompaey, K.C. Hari Kumar, P.
Wollants, Thermodynamic optimization of
the B-Fe system, Journal of Alloys and
Compounds, 8388, 2002, 173–181.
[7]. B. Tummers, DataThief III, 2006.
http://datathief.org/.

Thermodynamic Assessment of the Fe-B System in the Ssol5 and User Databases

More Related Content

Viewers also liked (6)

Similar to Thermodynamic Assessment of the Fe-B System in the Ssol5 and User Databases (8)

Recently uploaded (20)

Thermodynamic Assessment of the Fe-B System in the Ssol5 and User Databases