Journal of Manufacturing Processes 95 (2023) 14–26

Contents lists available at ScienceDirect

Journal of Manufacturing Processes

journal homepage: www.elsevier.com/locate/manpro

Investigation on the thermodynamic stability of phases evolved in Al-based

hybrid metal matrix composite fabricated using in-situ stir casting route
Sudhir Ranjan, Pradeep Kumar Jha *
Department of Mechanical and Industrial Engineering, Indian Institute of Technology, Roorkee 247667, India

A R T I C L E I N F O A B S T R A C T

Keywords: The objective of this study is to investigate the thermodynamic stability of different phases evolved during
Thermal stability fabrication of Al-12wt%Si-(TiB2 + Al2O3) hybrid composite using in-situ stir casting technique. The hybrid
Hybrid composites composite was fabricated using an approach that involved exothermic reactions between K2TiF6 and H3BO3 salts
Metal matrix composites
with the aluminium melt. Microstructural characterization and thermodynamic analysis of the fabricated com­
AMMC
TiB2
posite was successfully carried out to confirm and investigate the evolution of phases in the composite respec­
tively. The investigation of the reaction mechanism in the (Aluminium-Titanium-Boron) ternary and
(Aluminium-Titanium-Boron-Silicon) quaternary systems was conducted by utilizing the Wilson model and
Extended Miedema model to evaluate the excess Gibbs free energy. TiB2 was found out to be the most stable
phase in the study followed by AlB2. Silicon being used as alloying element was found to largely hinder the
formation of Al3Ti and to some extent also for AlB2. Incorporating silicon caused the excess Gibbs free energy for
Al3Ti to become positive after reaching a temperature of 1070 K, which effectively prevented the formation of
Al3Ti at higher temperatures. The formation of TiB2 was seen to be unaffected by the addition of silicon in the
aluminium melt. The findings from this investigation may be helpful in the development of processing param­
eters to optimize the microstructure and properties of the composite and can also provide valuable information
about specific temperatures at which reactions occur. This can simplify the experimental process and may result
in saving of time and resources.

properties [4]. Aluminium is often alloyed with other elements to

1. Introduction enhance its mechanical, physical, or chemical properties. Aluminium
silicon alloys are a versatile and reliable choice for many industrial
The focus of modern industries is on the design of new materials, applications, more specifically in automobile sectors [5]. They are
which has resulted in the advancement of composite materials. Com­ known for their excellent thermal conductivity and low coefficient of
posite materials have the unique ability to be tailored to different ap­ thermal expansion, making them ideal for applications where heat
plications in various fields, allowing for the molding of properties dissipation is important, such as in engine components and heat sinks.
according to the requirements [1]. This has also led to the development The addition of silicon in aluminium creates a eutectic mixture, which
of hybrid metal matrix composite, where two or more reinforcements enhances the strength and hardness of the alloy [6]. A near eutectic
are used to fabricate a composite, resulting in a material that is superior Al–Si alloy has a silicon content that is very close to the eutectic point
to a single reinforcement composite in terms of properties [2]. (≈12.6wt%Si), where the alloy has the lowest melting point and
Aluminium metal matrix composites (AMMCs) have become increas­ maximum fluidity during casting. This makes it easier to cast complex
ingly popular among researchers in the field of composite materials shapes and reduce defects in the casting process, resulting in a higher
owing to their high strength-to-weight ratio, which makes them an quality product with good machinability, and weldability [7].
excellent choice for industries like automobile, aerospace, and marine. Various reinforcements can be added to aluminium to fabricate
They are also preferred over other materials due to their excellent aluminium MMCs, but TiB2-reinforced aluminium metal matrix com­
strength-to-cost ratio, which addresses the issue of high-cost materials in posites are at a higher end when it comes to high-temperature engi­
these industries [3]. AMMCs possess high stiffness, hardness, strength, neering applications. TiB2 is an extremely hard ceramic particle that is
and exhibit better structural, thermal, wear, and transportation very stable at high temperatures and possesses a high corrosion

* Corresponding author.
E-mail addresses: [email protected] (S. Ranjan), [email protected] (P.K. Jha).

https://doi.org/10.1016/j.jmapro.2023.03.084
Received 18 January 2023; Received in revised form 25 March 2023; Accepted 30 March 2023
Available online 11 April 2023
1526-6125/© 2023 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

wear behavior of Al-Si-Mg/TiB2 MMC fabricated by in-situ stir casting

Nomenclature with the salt reaction mechanism. The wear rate was observed to
decrease in the AMMCs while the hardness increased as compared to
MMC metal matrix composite base alloy due to the formation of TiB2 and the ageing process. The
AMMC aluminium metal matrix composite development of oxidative layers on the worn surface also contributes to
SEM scanning electron microscopy the reduction in wear at the elevated ageing time [20]. Process opti­
XRD X-ray diffraction mization can be done to acquire the best optimal parameters for the
DSC Differential Scanning Calorimetry composite fabrication. Sheelwant et al. applied ANN-GA-based para­
Gibbs free energy metric optimization to the processing of Al-TiB2 MMC by optimizing the
◦
ΔG
ΔG excess free energy parameters of ultrasonic casting and predicting the porosity. The process
ΔH enthalpy change for the reaction parameters finalized for the optimization were the depth of the horn,
ΔS entropy change for the reaction melt temperature, and ultrasonication time. The ANN results were seen
K equilibrium constant for the reaction to match the experimental findings, and the grain boundaries were also
investigated to be fine by ultrasonic stirring resulting in uniform
dispersion of TiB2 [21]. The difficult task of achieving uniform particle
distribution in the matrix can be solved by a variety of methods. Gupta
resistance, high melting point, high modulus, excellent electrical and et al. fabricated Al 6061-Al3Ti composite using in-situ stir casting
thermal conductivity [8]. TiB2-reinforced AMMCs can be easily fabri­ coupled with ultrasonic stirring. The salt reaction of Al6063 was with
cated by in-situ reactions with less cost resulting in superior properties K2TiF6, and the morphology, mechanical properties with strengthening
of the reinforced composite [5]. Al2O3 in aluminium matrix improves mechanism were studied. The Al3Ti particulates were observed in
hardness, oxidation resistance, interfacial bonding and reduces the co­ blocky morphology, and their position was within the grains, enhancing
efficient of thermal expansion. The presence of Al2O3 in aluminium the fabricated composite's mechanical properties. Al3Ti also acts as a
composite can create a barrier that is more wear-resistant than the base nucleating agent, which caused the composite's grain size to decrease.
alloy, which can resist the plastic deformation of the matrix material [22]. Sheelwant et al. developed a novel hybrid ultrasonic casting sys­
[9,10]. tem to fabricate TiB2 reinforced aluminium metal matrix composite and
Fabrication of aluminium metal matrix composites always remains observed that higher ultrasonic power led to a more refined micro­
challenging in terms of uniform dispersion and better properties than structure and improved mechanical properties. However, excessively
the monolithic counterpart [11]. Liquid processing route, powder met­ high ultrasonic power could cause agglomeration of the TiB2 particles
allurgy, friction stir processing, etc., are various diverse fabrication and negatively affect the mechanical properties of the AMMC [23]. The
routes to manufacture a composite [12]. The stir casting route is one of insertion of a master alloy into the melt of aluminium can also produce
the liquid processing techniques which is the most preferred one due to in-situ reactions. Ramesh et al. fabricated Al6061-TiB2 MMC with the
low cost as compared to other processing techniques, and it is also one of stir casting process by introducing Al-10%Ti and Al-3%B master alloy in
the oldest preferred processing techniques [13]. AMMCs can be fabri­ the aluminium melt. The in-situ reaction was carried out at 810 ◦ C, and
cated by two techniques, i.e., in-situ or ex-situ. In the ex-situ technique contents of Al3Ti and TiB2 were observed in the fabricated composite.
externally, synthesized reinforcements are added into the matrix during An increment in mechanical properties with better ductility was also
the development of the composite, whereas the in-situ approach syn­ investigated [4]. Singh et al. focuses on the characterization of the 3-
thesizes the reinforcement through chemical reactions within the metal dimensional morphology of in-situ grown TiB2 particulates in molten
matrix (metallic melt in stir casting) itself [14]. The in-situ developed aluminium by employing a kinetic-geometric model. The composite was
MMCs have more thermodynamical stability, which results in a fine and fabricated using the in-situ halide salt method, and the investigation
uniform dispersal of reinforcements within the matrix with strong revealed that the TiB2 particles possessed a complex three-dimensional
interfacial bonding between the reinforced particles and the matrix with morphology, specifically a hexagonal pyramid morphology. Addition­
better properties [15]. The reinforcement in the fabrication of in-situ ally, some particles exhibited a dendritic structure, while others had a
composites is developed as a product of precipitation or interfacial in­ faceted shape [24].
teractions inside the melt during the fabrication of the composite [16]. The thermodynamic analysis of the precipitates developed during
One of the best ways to fabricate metal matrix composites is through the fabrication of the composite is a necessary task to achieve the best
in-situ salt-aided reactions, and various studies have also been done on thermodynamic and mechanical properties in the salt assisted method.
this process. Chen et al. investigated mechanical and microstructural Using the stir casting technique and an in-situ salt reaction between the
properties in Al-TiB2 composite. The composite was fabricated through salts of K2TiF6 and KBF4, Mozammil et al. fabricated an Al-4.5%Cu-TiB2
salt reactions between K2TiF6 and KBF4, which results in better distri­ MMC. The mechanical and microstructural properties were investigated
bution of TiB2 particulates and enhanced mechanical properties [17]. and found to be enhanced compared to the base alloy. The formation of
Ma et al. examined the in-situ formation of Al-7wt%Si/Al3Ti aluminium TiB2 particulates was also studied through thermodynamic analysis of its
matrix composite through the stir casting technique and studied the precipitation, and it was investigated that at lower temperatures, the
microstructural and mechanical properties of the composite. K2TiF6 salt appearance of this particulate is relatively easy due to more negative
was introduced in the Al-7wt%Si melt to fabricate the composite with an Gibbs free energy [1]. Fan et al. studied the thermodynamic influence of
in-situ salt reaction. The mechanical properties and thermal coefficient various alloying elements on the formation of TiB2 particles in AMMC
were observed to be enhanced in the manufactured composite through fabricated through flux-assisted reaction. It was observed that adding
reduced grain size due to the introduction of Al3Ti [18]. Chen et al. different alloying elements promotes the development of TiB2 and Al3Ti
presented a study on the microstructure evolution, and hot deformation phases [25]. Xue et al. focuses on investigating the effect of ScZr micro-
behavior of a composite material composed of TiB2 and TiAl3 reinforced alloying on the precipitation behavior and mechanical properties of
2024 aluminium alloy fabricated through an in-situ salt synthesis TiB2/Al-4.5Cu composite and observed that the introduction of Zr and
method. The addition of the TiAl3 and TiB2 reinforcements improved the Sc significantly enhanced the distribution of TiB2 particles and facili­
deformation resistance of the composite material, resulting in a more tated more efficient grain refinement, resulting in a fine and equiaxed
homogeneous deformation behavior and playing a significant role in α-Al grain structure. This, in conjunction with the ordered structure of
promoting dynamic recrystallization and inhibiting grain growth [19]. θ′ -Al2Cu precipitates and Al3 (Sc, Zr) dispersoids, contributed to the
In earlier work, the tribological characteristics of the MMCs created superior ductility of the composite [26]. Moldovan et al. calculated the
during these reactions were also examined. Rathod et al. analyzed the thermodynamic data for TiB2 formation in the aluminothermic reaction

15
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

between K2TiF6 and KBF4 salt in aluminium melt between 750 and Table 1
950 ◦ C, and found that the formation of TiB2 particulates is more stable Weight of salts added.
than Al3Ti in the aluminothermic reaction [27]. Usurelu et al. success­ Material Reinforcement TiB2 (%) Al2O3 (%) K2TiF6 H3BO3
fully fabricated AA6063-TiB2 metal matrix composite by reaction be­ (%) [TiB2 + (approx.) (approx.) (g) (g)
tween K2TiF6 and KBF4 salts in aluminium melt. The main finding is the Al2O3]
increase in formation of TiB2 due to the cracks and voids formed on the Al-12%Si 0 0 0 0 0
external surface of AlTi3, which results in an increment of the dissolution Al-12%Si- 12.5 5 7.5 176.90 90.06
of Al3Ti particulates in the melt [28]. Butu et al. concluded that the time 12.5 %
(TiB2 +
needed to form TiB2 increases proportionally with the increment in the Al2O3)
percentage amount of TiB2. The formation of TiB2 particulates also in­
creases with the cracks over the surface of Al3Ti particulates and with
the heat of exothermic reaction [29]. holding to remove the slag and maintaining the melt purity easily. The
In recent years, many researchers have focused on the in-situ pro­ temperature for adding two salts and the holding time was decided
cessed aluminium metal matrix composites including exothermic re­ based on the thermodynamic analysis discussed in this article. The melt
actions between the various aluminium alloy melts and salt mixtures. was taken out at 1053 K temperature from the furnace and after
Most of the researchers are focused on the theoretical mechanism of the removing the slag from the top of the melt, the molten material was
process and there is a need to focus over the thermodynamics behind the poured into a preheated steel mold to create a plate of size 160 × 160 ×
reactions. With the introduction of each alloying element with the 7 mm3 for further investigation. The reactions involved are discussed in
aluminium the formation of reinforcement in the composite gets the next section of the article.
affected. Component activity, one of the essential factors to determine Samples of different sizes were taken out from the fabricated hybrid
the thermodynamic equilibrium of the reaction directly gets affected by composite plate for microstructural analysis with a Zeiss Sigma 300
the alloying element. For the detailed thermodynamic study, there is a Field Emission Scanning Electron Microscope (FESEM) with Oxford
need to study various ternary or quaternary alloy systems. The data detectors. Accelerating voltage (electron high tension, EHT) of 15 kV
available for these studies are less extensive and obtaining experimental and a variable working distance in a secondary electron mode were
measurements of activity coefficients is a time-consuming process. So, used. Energy-Dispersive Spectroscopy (EDS) was also done on FESEM to
there is a need to study these systems and calculate the reaction activity identify the elements present in the sample. The sample surface was
coefficients with the available methods. cleaned and polished with various grit-size emery papers, followed by
The present article focuses on the fabrication of an Al-12wt%Si/ etching with Keller's' reagent for microstructural study in SEM. Fresh
TiB2/Al2O3 hybrid AMMC using the in-situ stir casting method. In Keller's' reagent was prepared by mixing 2.5 ml nitric acid (HNO3), 1.5
addition, a detailed thermodynamic analysis has been performed to ml hydrochloric acid (HCl), and 1 ml hydrofluoric acid (HF) in 85 ml of
better understand the reactions involved and phases evolved during the distilled water. Separate samples were also deep etched with hydroflu­
fabrication of the composite. Unlike previous studies that used K2TiF6 oric acid (HF acid) to get a better view of particles in the SEM images.
and KBF4 salts to create TiB2 in-situ composite, present work employs a Bruker D8 ADVANCE, X-Ray diffractometer was used to study the phase
different salt combination, namely K2TiF6 and H3BO3. Investigations compositions of the fabricated composite with copper-Kα radiation and
have been carried out on the Al-Ti-B ternary and Al-Ti-B-Si quaternary a scanning rate of 0.5. Quantification of phases and XRD Patterns were
systems to gain insights into the formation of phases in the composite, as analyzed by Xpert Highscore software. Differential Scanning Calorim­
well as to examine the effect of adding silicon as an alloying element. etry (DSC) was conducted on Exstar SII TG/DTA 6300 with an air at­
mosphere (200 ml/min) using a sample weight of 10.04 mg and a
2. Experimentation and methodology reference weight of 10 mg. The reference material used was alumina and
the samples were heated at a rate of 10◦ per minute.
2.1. Material selection
3. Reaction mechanism
In the present study, Al-12wt%Si master alloy was selected as a
matrix material in the form of ingots, which were cut into various sizes The fabrication of the hybrid composite is approached by flux-
for further melting process in a muffle furnace with a stir casting setup. assisted synthesis (FAS), a well-developed technique for reactions at
Two pure salts of K2TiF6 and H3BO3 were selected to be added into the higher temperatures. The salts K2TiF6 and H3BO3 are thermodynami­
melt to produce hybrid composite. C2Cl6 was also taken in a small cally stable in the temperature range of 1000–1060 K. Comparing the
amount to be added at the end of the reaction for better removal of slag stability of both salts with one another, it is observed that K2TiF6 is more
and to maintain the purity of the melt. stable than H3BO3. Both salts are added into aluminium melt, subse­
quently one after another. K2TiF6 salt gets dehydrated at 648 K in the
2.2. Sample preparation aluminium melt, followed by an exothermic reaction at 1001 K (727 ◦ C),
which suggests the starting of the formation of Al3Ti precipitate in the
The Al-12wt%Si/12.5% (TiB2 + Al2O3) hybrid composite was melt. When K2TiF6 is incorporated into the melt, the formation of the
fabricated in a muffle furnace (assembled by Lab Scientific Instruments, Al3Ti intermetallic compound starts according to Eq. (1). H3BO3 salt gets
Delhi, India, capacity: up to 1473 K), starting with the 1 kg Al-12wt%Si dehydrated above 443 K (170 ◦ C) to form HBO2 and again gets dehy­
alloy and maintaining the melt at 1053 K (780 ◦ C) temperature for half drated to lose its remaining water content at higher temperatures to
hour. K2TiF6 salt was preheated in an oven at 573 K for removal of any form B2O3. With the introduction of H3BO3 in the aluminium melt, a
type of moisture content. After maintaining the melt at 1053 K, K2TiF6 second reaction (Eq. (2)) takes place with the formation of AlB2 and
salt wrapped in aluminium foil was added. A mechanical stirrer was Al2O3 compounds. The last reaction occurs in the melt after the pro­
used to stir the melt continuously for 2 min, and then the melt was kept duction of both the components Al3Ti and AlB2, forming TiB2 particu­
at this temperature for the next 10 min. H3BO3 salt wrapped in lates in the melt in accordance with Eq. (3). The combined reaction for
aluminium foil was then introduced into the melt at the same temper­ the whole process is shown in Eq. (4).
ature with a continuous stirring for 2 min. The weights of the two salts
3K2 TiF6 + 13Al→3Al3 Ti + K3 AlF6 + 3KAlF4 (1)
added to the melt are mentioned in Table 1. The melt was then main­
tained at 1053 K for 1.5 h while being stirred for 2–3 min every half- H3 BO3 + 3Al→Al2 O3 + AlB2 + 3H2 O (2)
hour. C2Cl6 (0.5 % of 1000 g) was added to the melt after 1.5 h of

16
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

Al3 Ti + AlB2 →TiB2 + 4Al (3) Table 3

Thermodynamic data for Eq. (3).
3K2 TiF6 + 10Al + 6 H3 BO3 →K3 AlF6 + 3KAlF4 + 9H2 O + 3 Al2 O3 + 3TiB2 AlB2 + Al3Ti → TiB2 + 4Al
(4) T ΔH ΔS ΔG K Log(K)

The thermodynamic data for above the reactions Tables 2 and 3 in K kJ J/K kJ
the temperature range (1000− 1300 K) specifies the possibility of the 1000.000 63.931 64.194 − 0.263 1.032E+000 0.014
occurrence of these reactions in the working range with a negative value 1050.000 64.188 64.446 − 3.479 1.490E+000 0.173
of Gibbs free energy (ΔGT). The thermodynamic data for the reactions is 1100.000 64.378 64.623 − 6.706 2.082E+000 0.318
◦

1150.000 64.501 64.732 − 9.941 2.828E+000 0.452

calculated by HSC Chemistry 6 software.
1200.000 64.556 64.779 − 13.179 3.747E+000 0.574
The individual reactions of K2TiF6 and H3BO3 with aluminium melt 1250.000 64.544 64.769 − 16.418 4.854E+000 0.686
can form Al3Ti and AlB2, respectively, as seen in the previous section of 1300.000 64.463 64.706 − 19.655 6.163E+000 0.790
this article. Therefore, the combined mechanism of the salt-metal re­
actions can be further bifurcated as follows.
Formula FM Conc. Amount Amount Volume
i. First, K2TiF6 will get decomposed in the aluminium melt to form TiF4 g/mol wt-% mol g l or ml
gas and KF liquid, as expressed in Eq. (5)
AlB2 48.602 27.389 1.000 48.602 15.236
K2 TiF6 →2KF (l) + TiF4 (g) (5) Al3Ti 128.845 72.611 1.000 128.845 0.000
TiB2 69.520 39.178 1.000 69.520 15.872
Al 26.982 60.822 4.000 107.926 39.973
ii. The titanium ions are released and get diffused into the aluminium
melt with the aluminothermic reaction of KF (l) and TiF4 (g), which
occurs at the flux-liquid interface, as in Eq. (6) B2 O3 (l) + [Al]→[B]Al + Al2 O3 (10)
TiF4 (g) + [Al]→[Ti]Al + AlF3 (l) (6)
vi. According to Eq. (11), the saturation stage between the liquid
iii. When the saturation is reached between liquid aluminium and aluminium and boron will pump the formation of the interme­
solute Ti, it will get separated as an intermetallic compound, as tallic compound, AlB2.
expressed in Eq. (7) 2[B]Al + [Al]→AlB2 (11)
[Ti]Al + 3[Al]→Al3 Ti (7)
vii. Finally, both boron and titanium in the aluminium melt will form
iv. With the addition of H3BO3 in aluminium melt, the salt will get the second desired reinforcement at saturation inside the melt,
dehydrated and form B2O3 according to Eqs. (8) and (9) TiB2 (Eq. (12)).

H3 BO3 →HBO2 (l) + H2 O (8) [Ti]Al + 2[B]Al →TiB2 (12)

HBO2 →B2 O3 (l) + H2 O (9) 4. Results and discussion

The amount of the added salts, alloying elements, and the tempera­
v. The aluminothermic reaction between B2O3 and liquid aluminium
ture has played a vital role in the microstructural and thermal evolution
will now release the boron ions into the aluminium melt and diffuse
of the hybrid composite. Further, the results of the performed charac­
them into it along with the creation of Al2O3, a desirable reinforce­
terization and thermal analysis of the reactions on the fabricated hybrid
ment (Eq. (10)).
AMMC have been discussed in the section below.

Table 2 4.1. Microstructure and XRD patterns

Thermodynamic data for Eq. (2).
2H3BO3 + 3Al → Al2O3 + AlB2 + 3H2O The experiment aims to achieve suitable amount of TiB2 and Al2O3
T ΔH ΔS ΔG K Log(K)
reinforcements in the Al–Si matrix. The formation of these particles can
be observed in the SEM images below (Fig. 1). In FESEM, a beam of high-
K kJ J/K kJ
energy electrons is focused on the surface of the sample, triggering the
1000.000 − 521.229 − 14.470 − 506.758 2.969E+026 26.473 emission of secondary electrons and other signals, such as backscattered
1050.000 516.164 − 9.528 506.159 1.521E+025 25.182
− −
electrons and X-rays. These signals are then detected and analyzed to
1100.000 − 510.975 − 4.701 − 505.804 1.049E+024 24.021
1150.000 − 505.666 0.018 − 505.687 9.353E+022 22.971 produce images and other data about the sample. The confirmation for
1200.000 − 500.241 4.636 − 505.804 1.044E+022 22.019 the formation of both the reinforcement can be expressed through the
1250.000 − 494.702 9.157 − 506.149 1.421E+021 21.153 SEM image, the color mapping of the samples taken out from the
1300.000 − 489.053 13.589 − 506.718 2.301E+020 20.362 fabricated hybrid composite (Figs. 2 and 3). The color mappings can be
used to estimate the approximate locations of particles, and Energy-
dispersive spectroscopy can subsequently be conducted to confirm the
Formula FM Conc. Amount Amount Volume presence of specific elements. Energy-dispersive spectroscopy (EDS), a
g/mol wt-% mol g l or ml technique to analyze the elemental composition of solid materials was
H3BO3 61.832 60.439 2.000 123.664 86.177 performed to approximate the elements present in the fabricated hybrid
Al 26.982 39.561 3.000 80.945 29.979 composite. Point analysis EDS was performed at various locations in the
Al2O3 101.961 49.832 1.000 101.961 25.715 sample, which measures the elemental composition of the spot selected
AlB2 48.602 23.753 1.000 48.602 15.236 on the particles seen in the SEM image. Fig. 2 was magnified to perform
H2O 18.015 26.414 3.000 54.046 58.937
point analysis EDS, and the results are presented in Fig. 4, which

17
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

Fig. 1. SEM image of the fabricated sample with particles formation.

Fig. 2. Color mapping for the hybrid composite showing TiB2 particles. (For interpretation of the references to color in this figure legend, the reader is referred to the
web version of this article.)

confirms the presence of elements and their compositions. The color respectively, and a Cu-Kα radiation (λ = 1.54 Å) was used. The absence
mapping in Fig. 2 and the point analysis EDS both confirmed the exis­ of peaks corresponding to other intermetallic compounds in the XRD
tence of TiB2 particles. Similarly, Fig. 3 was subjected to point analysis patterns indicated the completion of the reactions involved in the
EDS, and the resulting data confirmed the formation of Al2O3 particles, experiment. The XRD data clearly showed peaks corresponding to TiB2
whose locations were previously estimated using color mapping in and Al2O3 phases. However, these peaks were relatively smaller in in­
Fig. 5. The weight percentage determined in EDS analysis is only an tensity than the peaks for aluminium and silicon due to their higher
approximate value of the element's composition because the efficiency volume fraction in the samples.
of X-ray detection can vary with different elements, leading to different
levels of sensitivity and accuracy. Additionally, the depth of X-ray
4.2. Thermodynamics of the reactions
penetration into the sample can also influence the measured concen­
tration, especially in samples that are not homogeneous. The SEM image
The formation of TiB2 does not take place through the interfacial
of the deep-etched sample clearly demonstrates the hexagonal structure
reactions but preferably by the precipitations from the aluminium melt.
of TiB2 particles (Fig. 6). The XRD pattern of the hybrid aluminium
As a result, during the fabrication of the aluminium composite, TiB2 is
metal matrix composite, shown in Fig. 7, clearly indicates the formation
mostly generated by precipitation in the aluminium melt when the
of two reinforcements in the matrix. The X-Ray diffractometer works on
concentration of titanium and boron reaches saturation [25,30]. The
the principle that X-rays are diffracted by atoms in a crystal lattice,
stability of the precipitates formed can be discovered through the excess
generating a diffraction pattern that can be used to determine the crystal
free energy (ΔG) calculations [31]. It is vital to calculate the excess free
structure of the material. The phase compositions of the composites
energy of reactions represented by Eqs. (7), (11), and (12) to form TiB2
were determined by analyzing the X-ray diffraction (XRD) pattern. The
particulates. ΔG calculations include the reactants in a liquid state to
step size and scanning speed were maintained at 0.02◦ and 0.5 s/step,
form solid-state products. Excess free energy is not only a function of

18
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

Fig. 3. Color mapping for the hybrid composite showing Al2O3 particles. (For interpretation of the references to color in this figure legend, the reader is referred to
the web version of this article.)

Fig. 4. EDS analysis indicating the presence of TiB2 particles.

temperature but also depends on the concentration of reactants and 4.2.1. Calculation of the activity
products or their molar weight during the reactions. The activity for the solid products Al3Ti, AlB2, and TiB2 can be
Expressions for excess free energy (ΔG) of the Eqs. (7), (11), and (12) approximated to unity in the Al melt. The value of the excess free energy
can be expressed as follows: of the three compounds in the liquid melt depends on the activity of the
components in the melt. Once the activity value is calculated, the excess
(13)
◦
ΔGAl3 Ti = ΔGAl3 Ti − 3RTlnαAl − RTlnαTi
free energy of Eqs. (13)− (15) can be determined. The activity directly
gets affected by the introduction of any alloying element in the melt,
(14)
◦
ΔGAlB2 = ΔGAlB2 − RTlnαAl − 2RTlnαB which further affects the excess free energy of the reactions involved
[32,33]. The calculations of the activity of the component can quickly
(15)
◦
ΔGTiB2 = ΔGTiB2 − 2RTlnαB − RTlnαTi determine the specific temperature and conditions for the occurrence of
the reactions. The procedure for the calculation of component activity is
where ΔG denotes the Gibbs free energy required to form the com­
◦
discussed below.
pound. Its value for different compounds can be calculated using
Table 4. αi is the activity for different components (i) in the melt. Procedure:

19
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

Fig. 5. EDS analysis indicating the presence of Al2O3 particles.

Fig. 6. Hexagonal geometry of TiB2 (deep-etched sample with HF acid).

The fabricated composite has silicon (12 wt%) as the alloying ( ) ⎡ ( )⎤

element, which affects the formation of Al3Ti, AlB2, and TiB2 in the ∑ ∑ xj 1 − Ai/j
composite melt. The addition of silicon changes the activity of Al, Ti, and lnγi = − ln 1 − xj Aj/i +1− ⎣ ∑ ⎦ (17)
j j
1− k xk Ak/j
B in the composite melt. The Al-Ti-B system is a ternary system, and the
introduction of silicon makes it a quaternary system of Al-Ti-B-Si. where, xi is the molar fraction of the component i in the melt. Ai/j and
The activity of any component (i) can be determined as, Aj/i are the adjustable parameters. The activities obtained through bi­
nary systems are used to get the activities of higher systems.
αi = xi ⋅γi (16)
The above Eq. (17) for binary (i-j), ternary (i-j-k) and quaternary
xi: molar concentration of component i, and γi: activity coefficient. system (i-j-k-l), expressed as:
The calculation of the activity coefficient γi in multiple components
can be calculated using the method suggested by Wilson [34] expressed Binary system:
as:

20
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

density difference contributes favorably [35,36]. When the Miedema

model is extended to ternary alloys, only the binary interaction between
atoms is taken into account, leaving out the ternary atomic interactions
between the elements [37]. An extended Miedema model [38] was
modified by Fan et al. [25] to determine the activity coefficients using
the following theoretical equations.
The heat of formation, ΔQij for a binary liquid alloy system (i-j) can
be stated through Miedema model as,
[ ( )] [ ( )]
xi 1 + ui xj ρi − ρj xj 1 + uj xi ρj − ρi
ΔQij = tij [ ( ) ] 2/3 [ ( ) ] 2/3 (23)
xi 1 + ui xj ρi − ρj Vi + xj 1 + uj xi ρj − ρi Vj

where, tij can be obtained as,

[ {( 1 ) ( 1 ) } ( )2 ( )]
2/3 2/3
2pVi Vj qp n3ws − n3ws j − ρi − ρj − b pr
tij = i
(24)
( 11 ) + ( 11 )
n3ws n3ws
Fig. 7. XRD pattern for the hybrid composite. i j

where, V is the molar volume, ρ is the electron density. Other symbols, u,

Table 4 b, r, p, and q are the factors related to material properties. The units used
Standard Gibbs free energy calculations. for these are as follows:
Enthalpy in kJ/mol, electron density in volts, V in cm3, nws in density
Reactions ΔG◦ J/mol (T, Kelvin)
units.
3Al (l) + Ti (l) → Al3Ti (s) − 335,925 + 168.326 T
Al (l) + 2B (l) → AlB2 (s) − 237,812 + 162.962 T
Ti (s) + B (s) → TiB2 (s) − 284,510 + 20.5 T
• The value of p is 12.3, 10.6, and 14.1 for alloys of a transition metal
Ti (s) → Ti (l) 15,480–7.95 T with non-transition metal, two non-transition metals, and two tran­
B (s) → B (l) 50,200–21.84 T sition metals.
• The value of qp equals to 9.4V2((d.u.)2/3)
( ) • b is 0.73, 1.0, and 0 for liquid alloys, solid alloys, and other alloys.
( ) xi xj 1 − Ai/j • Other parameters are listed in the Table 4
lnγi = 1 − ln 1 − xj Aj/i − − (18)
1 − xj Aj/i 1 − xi Ai/j
In an i-j binary system, the excess molar Gibbs free energy (ΔGEi ) of
the component i, can be exhibited as,
Ternary system:
( ) ΔGEi = RTlnγi (25)
( ) xi xj 1 − Ai/j
lnγi =1 − ln 1 − xj Aj/i − xk Ak/i − − − The excess molar Gibbs free energy of the component and the entire
( )
1 − xj Aj/i − xk Ak/i 1 − xi Ai/j − xk Ak/j excess Gibbs free energy, GEij are related as,
xk 1 − Ai/k
− ∂GEij
1 − xi Ai/k − xj Aj/k ΔGEi = GEij + (1 − xi ) (26)
∂xi
(19)
GEij = ΔQij − TSEij (27)
Quaternary system:
( ) xi where, SEij is the excess entropy of mixing. The approximate value of SEij
lnγi =1 − ln 1 − xj Aj/i − xk Ak/i − xl Al/i −
1 − xj Aj/i − xk Ak/i − xl Al/i can be expressed as,
( ) ( )
( ) 1 1
xj 1 − Ai/j xk 1 − Ai/k SEij = 0.1 × ΔQij − (28)
− − (20) Tmi Tmj
1 − xi Ai/j − xk Ak/j − xl Al/j 1 − xi Ai/k − xj Aj/k − xl Al/k
( )
−
xl 1 − Ai/l where, T is the temperature of the aluminium melt and Tmi and Tmj are
1 − xi Ai/l − xj Aj/l − xk Ak/l the melting temperature of components i and j in the liquid melt.
The values for other parameters have been listed in Table 5.
The calculation for the parameters of binary system Ai/j and Aj/i can
Using Eqs. (23)–(28), the activity coefficients can be summarized as,
be performed based on binary infinitely dilute activity coefficients by
considering lnγxi i→0 and lnγxj j→0. These two parameters can be calculated
as,
( )
lnγxi i →0 = − ln 1 − Aj/i + Ai/j (21)
( )
(22)
x →0
lnγj j = − ln 1 − Ai/j + Aj/i
Table 5
The experimental data can be used to determine the values of lnγxi i→0 Values of other parameters [39].
and lnγxj j→0. Calculation of these values is temperature dependent which Component V2/3 (cm3) ρ (V) n1/3
ws ((d.u.)
1/3
) u r/p Tm (K)
needs extensive experimental data. Alternately, these parameters can be
Al 4.6 4.2 1.39 0.07 1.9 933
calculated through some theoretical model. Miedema's model stated that Ti 4.8 3.65 1.47 0.04 1.0 1943
in a binary alloy system, the electronegativity difference between the B 2.8 4.75 1.55 0.04 1.9 2365
two parts contributes negatively to the heat of mixing, while the electron Si 4.2 4.70 1.50 0.04 2.1 1685

21
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

[ { ( )
αij ΔQij 1 1 ui ρi − ρj Table 7
lnγi = 1 +(1− x) − − [ ( )] Calculated activity values for Al, Ti, and B.
RT xi 1 − xi 1 + ui (1 − xi ) ρi − ρj
( ) Temp, K Al Ti B
uj ρj − ρi
+[ ( )] 1000 0.984100 0.000004 0.456570
1 + uj xi ρj − ρi
1100 0.970100 0.000011 0.407676
2[ ( )] 2[ ( )]}] 1200 0.956400 0.000026 0.373724
V3i 1 + ui (1 − 2xi ) ρi − ρj + V3j − 1 + uj (1 − 2xi ) ρj − ρi
− 2[ ( )] 2[ ( )] 1300 0.947300 0.000053 0.336143
xi V3i 1 + ui (1 − xi ) ρi − ρj + (1 − xi )V3j 1 + uj xi ρj − ρi
(29)
Table 8
The value for the term αij can be calculated from the equation below,
Calculated excess free energy.
[ ( )− 1 ]
αij = 1 − 0.1T (Tmi )− 1 + Tmj (30) Temp ΔGAl3Ti ΔGAlB2 ΔGTiB2
K kJ/mol kJ/mol kJ/mol
Hence, the activity coefficient of the component i in an infinite so­ 1000 − 64.5320 − 61.6865 − 148.3123
lution j can be calculated from Eq. (23) as 1100 − 45.74148 − 41.8721 − 141.3634
{ ( )} 1200 − 27.2757 − 22.1834 − 134.9559
αij tij 1 + ui ρi − ρj 1300 − 9.0657 − 1.8212 128.0239
(31)
−
lnγi = 2/3
RTVj

Thus, from Eqs. (17), (21), (22), and (31), the activity coefficients
can be easily calculated in binary or higher multi-alloy systems. The
thermodynamic data predicted from these equations are applicable for
multiple liquid alloys according to the physical parameters and are in­
dependent of the experimental data.

4.2.2. Calculated data for multialloy systems

The excess free energies, ΔGAl3Ti, ΔGAlB2, and ΔGTiB2can now be
calculated based on the procedure discussed above. Thermodynamic
study of the formation of TiB2 in the fabricated composite is carried out
based on ternary Al-Ti-B and quaternary Al-Ti-B-Si melt system. The
effects of silicon, the alloying element on TiB2 formation and the com­
parison with ternary system is investigated through the calculations. The
values for the parameter Ai/j evaluated by the Wilson and extended
Miedema equation considering the binary systems are listed in the
Table 6. The temperature range for the calculation is kept at 1000–1300
K. The parameter computations for ternary systems will also employ the
parameters mentioned in Table 6.

4.2.3. Al-Ti-B ternary system

In the fabrication of the hybrid in-situ composite, considering the Fig. 8. Ternary Al-Ti-B phase diagram at logarithmic scale (Al-corner) [40].
melt as a ternary system of Al-Ti-B for the formation of TiB2 particulates,
the calculated values of αijk parameters are listed in Table 7. The values
are calculated through Eqs. (16) and (19) and the data from Table 5 (i =
Al, j = Ti, k = B). The molar fraction is defined through the solubility of
titanium and boron in liquid aluminium. The solubility of Ti and B is
very low in aluminium melt.
The values calculated for activities of the various components in the
aluminium melt are imported in Eqs. (13)–(15) to get the free excess
energies. The excess free energies are represented in tabular form in
Table 8.
Jones and Pearson [40] presented a new non-conventional practical
phase diagram for Al-Ti-B ternary system with an aluminium corner at
1000 K temperature (Fig. 8). In the phase diagram, it can be noted that
the concentration should be in the shaded region for the formation of
TiB2 in the liquid aluminium in the shaded region to avoid the formation
of Al3Ti along with liquid aluminium and TiB2. The value of excess free
energy for the equations can be determined as a function of the con­
centration of titanium in liquid aluminium at a temperature of 1000 K.

Table 6
Calculated values of Wilson parameters.
T, K AAl/Ti ATi/Al AAl/B AB/Al ATi/B AB/Ti Fig. 9. Excess free energy curves for Ternary (Al-Ti-B) System.
1000 − 11.289 − 11.029 − 1.737 0.730 − 20.563 − 11.644
1100 − 9.974 − 9.738 − 1.551 0.685 − 18.745 − 10.515
1200 − 8.837 − 8.623 − 1.414 0.655 − 16.882 − 9.360
1300 − 7.884 − 7.668 − 1.264 0.612 − 15.314 − 8.390

22
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

When the reaction starts the concentration of titanium and boron Table 10
released into the liquid aluminium is extremely low due to which the Activities (αi) for aluminium, titanium, and boron.
formation of Al3Ti and AlB2 is easier than the formation of TiB2 owing to Temp, K Al Ti B
less excess free energy of Al3Ti and AlB2 as compared to TiB2. With
1000 0.984100 0.000004 0.456570
passage of time, the concentration of titanium and boron is enhanced in 1100 0.970100 0.000011 0.407676
the melt, due to which the excess free energy decreases for all three 1200 0.956400 0.000026 0.373724
reactions (Eqs. (13)–(15)). TiB2 formation becomes easier when the 1300 0.947300 0.000053 0.336143
excess free energy ΔGTiB2 becomes more negative than ΔGAl3Ti and
ΔGAlB2, as in Table 8. The free energies of these phases for a temperature
range of 1000–1300 K are also represented in graphical form in Fig. 9, Table 11
justifying that the excess free energy of TiB2 phase is negative in whole Excess free energies in quaternary system.
temperature range. Hence, it can be concluded that the TiB2 phase can Temp, K ΔGAl3Ti ΔGAlB2 ΔGTiB2
be easily formed with the increment in titanium concentration, or in K kJ/mol kJ/mol kJ/mol
other words, it can be stated that TiB2 is more thermodynamically stable 1000 − 14.7149 − 26.93792 − 70.9869
in aluminothermic reaction than the other two phases at this 1100 6.6886 − 6.2475 − 60.6782
temperature. 1200 27.3381 14.3578 − 51.2140
TiB2 particles may form directly at the junction of flux and liquid 1300 45.3601 34.4928 − 44.9382

aluminium. The TiB2 phase produced in this way is more easily rejected
to the inter-dendritic region during the hybrid composite's solidification
process. The development of TiB2 phase in a temperature range of
1000–1300 K is not true for other liquid melts because aluminium acts as
a medium for the precipitation of TiB2 at this temperature range.

4.2.4. Al-Ti-B-Si quaternary system

In the previous section, the composite melt is considered a ternary
system (Al-Ti-B), but the matrix alloy for the composite fabrication is Al-
12%Si, which also includes silicon as the alloying element. Introducing
an alloying element in the liquid aluminium will affect the concentration
of components, resulting in a change in activities and excess free en­
ergies of the three phases [25]. The calculations for activities in the
quaternary system can be done using Eq. (20). The activity values for the
binary systems Ti–Si, B–Si, and Al–Si can be calculated through the
Wilson and extended Miedema model discussed previously in Section
4.2. The estimated values in temperature range 1000–1300 K are
expressed in a tabular form in Table 9.
Placing the values of activity coefficients from Table 9, the activities
for Al, Ti, and B in a temperature range of 1000–1300 K are calculated
using Eq. (20), which are listed in the Table 10.
The activities are used to calculate the excess free energies of Al3Ti,
AlB2, and TiB2 in the quaternary system. The excess free energy for these Fig. 10. Excess free energy curves for quaternary (Al-Ti-B-Si) system.
reactions in a temperature range of 1000–1300 K are calculated using
the values input in Eqs. (13)–(15), and they are listed in Table 11.
will become positive approximately beyond 1070 K (796 ◦ C), obstruct­
The plot of excess free energies for the formation of the three com­
ing the formation of Al3Ti particles. This is because the aluminium melt
pounds, Al3Ti, AlB2, and TiB2 are represented in the graphical form in
now contains 12 wt% Si, which has a direct impact on how Ti and Al
Fig. 10. The difference between the curves of ternary and quaternary
behave in the melt. Without silicon (as alloying element) ΔGAl3Ti at
systems shows the effect of adding alloying element, silicon in the liquid
1053 K is approximated to − 54.59 kJ/mol, which is significantly lower
aluminium. When 12 wt% silicon is added into the melt, it directly af­
than the quaternary counterpart. At 1300 K (1027 ◦ C) the value of excess
fects the excess free energies of formation of Al3Ti, AlB2, and TiB2. While
free energy for Al3Ti remains negative for the ternary system. This in­
in the ternary system, ΔG for all three phases is negative in the whole
dicates that in the temperature range of 1000–1300 K, Al3Ti production
temperature range (1000− 1300 K), in the quaternary system, it is not
is simpler in the absence of silicon.
negative in the entire temperature range. The formation of Al3Ti and
ΔGAlB2is approximated to be − 15.956 kJ/mol at 1053 K (Fig. 10),
AlB2 gets affected due to the addition of silicon resulting in the positive
suggesting that the formation of AlB2 is easier than Al3Ti. The same
excess free energies above a specific temperature.
value in the ternary is approximated to − 51.21 kJ/mol. For the ternary
ΔGAl3Ti is approximated to − 3.32 kJ/mol at 1053 K (780 ◦ C) from
system, ΔGAlB2 values are negative in the whole temperature range of
Fig. 10; due to which the temperature for the addition of K2TiF6 salt is
1000–1300 K, while for the quaternary system, it becomes positive
selected as 1053 K. As we go above 1053 K the ease of the formation of
approximately beyond 1132 K (859 ◦ C). The addition of silicon affects
Al3Ti will become low. The excess free energy of the formation of Al3Ti
the activities of aluminium and boron in the aluminium melt resulting in
the changes in excess free energy of AlB2.
Table 9 ΔGTiB2 is approximated to − 65.453 kJ/mol at 1053 K (Fig. 10) in the
Calculated Wilson parameters for quaternary system. quaternary system. In the ternary system, the value of ΔGTiB2 is
approximated as − 144.69 kJ/mol. So, it can be interpreted from these
T, K AAl/Si ASi/Al ATi/Si ASi/Ti AB/Si ASi/B
excess free energy values that the addition of alloying element, silicon
1000 − 5.5903 0.9906 − 20.5815 − 19.3852 0.3623 − 0.2962 has almost negligible effect on the formation of TiB2 particles, as both
1100 4.9358 0.9836 18.2823 17.2019 0.3544 − 0.2995
the values of ternary and quaternary ΔG are highly negative. For the
− − −
1200 − 4.4243 0.9748 − 16.3766 − 15.3928 0.3475 − 0.3017
1300 − 4.012 0.9646 − 14.7730 − 13.8705 0.3411 0.3030 whole temperature range 1000–1300 K the ΔGTiB2 value remains

23
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

negative in both ternary and quaternary systems. At 1300 K the excess

free energy is approximated to − 128.076 kJ/mol and − 43.814 kJ/mol,
respectively, for ternary and quaternary systems.

4.2.5. Considering very low solubility of titanium and boron in aluminium

melt
The solubility of Ti and B is very low in aluminium melt. So,
considering the molar fraction of both Ti (xTi → 0) and B (xB → 0) as zero
in aluminium melt, the activity coefficients of the titanium (γTi) and
boron (γB) in a quaternary system can be calculated using the Eq. (20) as,
( ) ( )
( ) xAl 1− ATi/Al xSi 1− ATi/Si
lnγi xTiTi →0,xB →0 = 1− ln 1− xAl AAl/Ti − xSi ASi/Ti − −
1− xSi ASi/Al 1− xAl AAl/Si
(32)
( ) ( ) ( )
xAl 1− AB/Al xSi 1− AB/Si
lnγi xBTi →0,xB →0 = 1− ln 1− xAl AAl/ − xSi ASi/B − −
B 1− xSi ASi/Al 1− xAl AAl/Si
(33)
The calculated free excess energies are represented in Table 12 and
graphical form in Fig. 11.
Fig. 11. Excess free energy curves comparison for TiB2 formation.
The comparison between the calculated excess free energy for qua­
ternary system and (ΔGTiB2) by considering Ti (xTi → 0) and B (xB → 0) is
also represented in Fig. 11. decided by performing many iterations for the experiment.
It can be observed from the graph that for both the excess free energy In the ternary system, Al-Ti-B four compounds are formed namely,
curves, calculated ΔGTiB2 are in the negative region with slight variation. Al3Ti, AlB2, AlB12, and TiB2. AlB12 is an unstable compound with lesser
Therefore, to simplify the calculations for excess free energies only for Gibbs free energy than AlB2, so the formation of AlB12 is easier than
TiB2 in a quaternary system, it is better to consider the solubilities of AlB2. AlB12 compound reacts with Al3Ti in liquid aluminium to form
titanium and boron as zero. The large equations will be converted into TiB2 particulates at a high holding time, further confirming the impor­
two simple equations, reducing the calculation time and work involved. tance of holding time.
AlB12 + Al3 Ti + Al→TiB2 + 5AlB2 (34)
4.3. DSC analysis
The reaction has a very high negative Gibbs free energy Table 13,
confirming the reaction in the temperature range of 1000–1300 K. This
In DSC, a sample and a reference material are heated or cooled under
reaction will enhance the number of TiB2 reinforced particles in the
controlled conditions and the heat flow is measured as a function of
melt. It will result in the distribution of more TiB2 particulates in
temperature. As the temperature changes, the material undergoes
aluminium matrix.
physical or chemical changes, which result in a change in the heat flow.
These changes in heat flow are recorded as a DSC curve, which provides
5. Conclusions
information about the thermal properties of the material. The Differ­
ential Scanning Calorimetry is performed on the samples of fabricated
The study investigates the reaction mechanism of aluminothermic
hybrid composite in the air (200 ml/min) with a heating rate of 10◦ per
reactions to ensure in-situ formation of reinforcement particles in a
minute and alumina as reference material. The temperature range is
hybrid metal matrix composite.
kept between room temperature and 1123 K (850 ◦ C) for one sample and
1523 K (1250 ◦ C) for another. The results of which are reflected in
1) Al-based MMC was successfully fabricated using the stir casting
Fig. 12, where it can be observed that during the heating cycle of the first
technique with the addition of K2TiF6 and H3BO3 salts. To
sample, aluminium begins to melt at a temperature of about 848 K.
confirm the successful formation of TiB2 and Al2O3 as reinforce­
There are no traces of any other reaction, so in the next sample the
ment particles in the Al-12%Si matrix, microstructural charac­
temperature range is enhanced to 1523 K (1250 ◦ C). In the heating cycle
terization was carried out. Additionally, XRD analysis was used to
of the second sample, a reaction is observed at 1518 K (1245 ◦ C).
investigate the phase composition of the hybrid composite, which
Although the Al-12wt%Si-(TiB2 + Al2O3) is manufactured at a tem­
further confirmed the presence of these reinforcement particles.
perature of roughly 1053 K, the DSC study indicates that the reaction to
2) Operating parameters such as temperature of salt addition into
fabricate the composite occurs at 1518 K. This reflects the importance of
the melt, holding time, and stirring time were optimized based on
the holding time, which is 1.5 h in our case. To fabricate the hybrid
thermodynamic calculations and experimental findings.
composite, it is necessary to hold the melt after introducing both salts
3) Gibbs free energy (ΔG◦ ) of the reactions involved was found to be
into the melt, so all the reactions can be completed to form desired re­
negative in the operating temperature range of 1000–1300 K.
inforcements. If no holding time is provided in the experiment, then all
4) Excess free energies were calculated for the formation of Al3Ti,
the reactions will complete only at 1523 K. The holding time of 1.5 h is
AlB2, and TiB2 particles in the aluminium melt. Both ternary (Al-
Ti-B) and quaternary (Al-Ti-B-Si) systems were studied to find the
Table 12 temperature range for the formation of reinforcement particles.
Excess free energy comparison.
5) The addition of silicon affects the formation of Al3Ti, and its
Temperature ΔGTiB2 (considering low solubility) ΔGTiB2 (original) addition hinders the formation of Al3Ti above 1070 K. Hence in
K kJ/mol kJ/mol
the present study, K2TiF6 salt was added in the melt at 1053 K
1000 − 74.7273 − 70.9869 (ΔGAl3Ti = − 3.32 kJ/mol).
1100 − 65.8971 − 60.6782
1200 − 57.6709 − 51.2140
1300 − 49.6807 − 44.9382

24
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

Fig. 12. DSC analysis of the fabricated hybrid composite up to (i) 1123 K (850 ◦ C) and (ii) 1523 K (1250 ◦ C).

composite to obtain desired properties such as high strength,

Table 13
wear resistance etc.
Thermodynamic data for Eq. (34).
AlB12 + Al3Ti + Al → TiB2 + 5AlB2
6. Future scope
T ΔH ΔS ΔG K Log(K)

K kJ J/K kJ The current research is limited to thermodynamic analysis of Al-

12wt%Si-(TiB2 + Al2O3) composite. However, the technique used in this
1000.000 − 638.380 − 59.805 − 578.575 1.676E+030 30.224
1050.000 − 638.416 − 59.840 − 575.584 4.326E+028 28.636 study can be applied to examine the influence of additional alloying
1100.000 − 638.377 − 59.805 − 572.592 1.557E+027 27.192 elements on the formation of various types of reinforcements in a salt
1150.000 − 638.264 − 59.704 − 569.604 7.488E+025 25.874 assisted fabrication of a composite. This future research could provide
1200.000 − 638.076 − 59.545 − 566.623 4.640E+024 24.666 insight into the specific temperatures at which salts should be added and
1250.000 − 637.814 − 59.331 − 563.651 3.594E+023 23.556
1300.000 − 637.479 − 59.068 − 560.690 3.394E+022 22.531
reactions occur, which would help to streamline the experimental pro­
cess and save time for researchers.

Formula FM Conc. Amount Amount Volume Declaration of competing interest

g/mol wt-% mol g l or ml

AlB12 156.702 50.140 1.000 156.702 61.452

The authors declare that they have no known competing financial
Al3Ti 128.845 41.227 1.000 128.845 0.000 interests or personal relationships that could have appeared to influence
Al 26.982 8.633 1.000 26.982 9.993 the work reported in this article.
TiB2 69.520 22.244 1.000 69.520 15.872
AlB2 48.602 77.756 5.000 243.008 76.178
Acknowledgments

6) The formation of AlB2 particles was slightly affected by the This research did not receive any specific grant from funding
addition of silicon, excess free energy for its formation remains agencies in the public, commercial, or not-for-profit sectors.
negative till 1132 K.
7) No effect of silicon on the formation of TiB2 was observed, as its References
excess free energy remains negative in the whole temperature
range of 1000–1300 K. [1] Mozammil S, Karloopia J, Verma R, Jha PK. Effect of varying TiB2 reinforcement
and its ageing behaviour on tensile and hardness properties of in-situ Al-4.5%Cu-
8) When the concentrations of titanium and boron in aluminium xTiB2 composite. J Alloys Compd 2019;793:454–66. https://doi.org/10.1016/j.
melt reaches saturation, TiB2 particles are formed through pre­ jallcom.2019.04.137.
cipitation. The solubility of titanium and boron is negligible in [2] Ranjan S, Karloopia J, Jha PK. Recent advances in aluminium-based hybrid metal
matrix composites: a review. Miner Met Mater Ser 2022:53–81. https://doi.org/
liquid aluminium. Hence while calculating ΔGTiB2, the molar 10.1007/978-3-030-92567-3_4.
fraction of titanium and boron can be considered as zero for [3] Sujith SV, Mahapatra MM, Mulik RS. An investigation into fabrication and
making the equations simple and less time-consuming. characterization of direct reaction synthesized Al-7079-TiC in situ metal matrix
composites. Arch Civ Mech Eng 2019;19:63–78. https://doi.org/10.1016/j.
9) Holding duration should be optimized based on DSC analysis to
acme.2018.09.002.
ensure completion of all the reactions involved. [4] Ramesh CS, Pramod S, Keshavamurthy R. A study on microstructure and
10) The thermodynamic analysis of a metal matrix composite can mechanical properties of Al 6061-TiB2 in-situ composites. Mater Sci Eng A 2011;
provide insights into the stability and formation of different 528:4125–32. https://doi.org/10.1016/j.msea.2011.02.024.
[5] Karloopia J, Mozammil S, Jha PK. Influence of in situ titanium diboride particulate
phases in the composite. Phase evolution analysis can help in reinforcement on mechanical properties of aluminum-silicon based metal matrix
understanding the kinetics of phase transformation during the composites. Jom 2020;72:2927–36. https://doi.org/10.1007/s11837-020-04245-
processing and manufacturing of metal matrix composites. This x.
[6] FC Robles-hernandez. Al-Si alloys. n.d.
information can be used to develop processing parameters that [7] Asensio-Lozano J, Suarez-Peña B. Effect of the addition of refiners and/or modifiers
can optimize the microstructure and properties of the composite, on the microstructure of die cast Al-12Si alloys. Scr Mater 2006;54:943–7. https://
and can be used to design and optimize the composition of the doi.org/10.1016/j.scriptamat.2005.10.067.
[8] Ünsal H, Grasso S, Kovalčíková A, Hanzel O, Tatarková M, Dlouhý I, et al. In-situ
graphene platelets formation and its suppression during reactive spark plasma

25
S. Ranjan and P.K. Jha Journal of Manufacturing Processes 95 (2023) 14–26

sintering of boron carbide/titanium diboride composites. J Eur Ceram Soc 2021;41: [24] Singh C, Prywer J, Yadav D. On the three-dimensional morphology of in situ grown
6281–9. https://doi.org/10.1016/j.jeurceramsoc.2021.06.053. TiB2 particles in molten aluminum. J Mater Eng Perform 2023. https://doi.org/
[9] Ravikumar M, Reddappa HN, Suresh R, Reddy MS, Babu ER, Reddy NC, et al. 10.1007/s11665-022-07779-0.
Evaluation of corrosion properties of Al2O3 and SiC reinforced aluminium metal [25] Fan T, Yang G, Zhang D. Thermodynamic effect of alloying addition on in-situ
matrix composites using Taguchi’s techniques. J Sci Res 2021;65:253–9. https:// reinforced TiB2/Al composites. Metall Mater Trans A Phys Metall Mater Sci 2005;
doi.org/10.37398/jsr.2021.650132. 36 A:225–33. https://doi.org/10.1007/s11661-005-0154-8.
[10] Rosenberger MR, Forlerer E, Schvezov CE. Wear behavior of AA1060 reinforced [26] Xue Y, Lou Z, Hao Q, Li X, Yu W, Zhang H, et al. Insight into the precipitation
with alumina under different loads. Wear 2009;266:356–9. https://doi.org/ behavior and mechanical properties of Sc-Zr micro-alloying TiB2/Al-4.5Cu
10.1016/j.wear.2008.06.007. composites. J Alloys Compd 2022;929:167209. https://doi.org/10.1016/j.
[11] Mozammil S, Verma R, Karloopia J, Jha PK. Investigation and measurement of jallcom.2022.167209.
porosity in Al + 4.5Cu/6wt%TiB2 in situ composite: optimization and statistical [27] Moldovan P, Butu M, Popescu G, Buzatu M, Usurelu E, Soare V, et al.
modelling. J Mater Res Technol 2020;9:8041–57. https://doi.org/10.1016/j. Thermodynamics of interactions in Al-K2TiF6-KBF 4 system. Rev Chim 2010;61:
jmrt.2020.05.045. 828–32.
[12] Zhou MY, Ren LB, Fan LL, Zhang YWX, Lu TH, Quan GF, et al. Progress in research [28] Uşurelu EM, Moldovan P, Buţu M, Ciucǎ I, Drǎguţ V. On the mechanism and
on hybrid metal matrix composites. J Alloys Compd 2020:838. https://doi.org/ thermodynamics of the precipitation of TiB2 particles in 6063 matrix aluminum
10.1016/j.jallcom.2020.155274. alloy. UPB Sci Bull Ser B Chem Mater Sci 2011;73:205–16.
[13] Ramanathan A, Krishnan PK, Muraliraja R. A review on the production of metal [29] Butu M, Moldovan P, Usurelu E, Mitrica D. The mechanism and kinetics of TiB2
matrix composites through stir casting – furnace design, properties, challenges, and nanoparticles and microparticles formation via aluminothermic reaction. Rev Chim
research opportunities. J Manuf Process 2019;42:213–45. https://doi.org/ 2011;62:723–6.
10.1016/j.jmapro.2019.04.017. [30] Liu Z, Rakita M, Xu W, Wang X, Han Q. Ultrasound assisted salts-metal reaction for
[14] Wang M, Chen D, Chen Z, Wu Y, Wang F, Ma N, et al. Mechanical properties of in- synthesizing TiB2 particles at low temperature. Chem Eng J 2015;263:317–24.
situ TiB2/A356 composites. Mater Sci Eng A 2014;590:246–54. https://doi.org/ https://doi.org/10.1016/j.cej.2014.11.043.
10.1016/j.msea.2013.10.021. [31] Yue NL, Lu L, Lai MO. Application of thermodynamic calculation in the in-situ
[15] Liu Z, Cheng N, Zheng Q, Wu J, Han Q, Huang Z, et al. Processing and tensile process of Al/TiB2. Compos Struct 1999;47:691–4. https://doi.org/10.1016/
properties of A356 composites containing in situ small-sized Al3Ti particulates. S0263-8223(00)00037-4.
Mater Sci Eng A 2018;710:392–9. https://doi.org/10.1016/j.msea.2017.11.005. [32] Wu Z, Troparevsky MC, Gao YF, Morris JR, Stocks GM, Bei H. Phase stability,
[16] Fjellstedt J, Jarfors AEW. On the precipitation of TiB2 in aluminum melts from the physical properties and strengthening mechanisms of concentrated solid solution
reaction with KBF4 and K2TiF6. Mater Sci Eng A 2005;413–414:527–32. https:// alloys. Curr Opin Solid State Mater Sci 2017;21:267–84. https://doi.org/10.1016/
doi.org/10.1016/j.msea.2005.09.054. j.cossms.2017.07.001.
[17] Chen F, Chen Z, Mao F, Wang T, Cao Z. TiB2 reinforced aluminum based in situ [33] Avner SH. Introduction to Physical Metallury: Second Edition. 1974.
composites fabricated by stir casting. Mater Sci Eng A 2015;625:357–68. https:// [34] Vapor-Liquid Wilson GM, Equilibrium XI. A new expression for the excess free
doi.org/10.1016/j.msea.2014.12.033. energy of mixing. J Am Chem Soc 1964;86:127–30. https://doi.org/10.1021/
[18] Ma S, Wang Y, Wang X. The in-situ formation of Al3Ti reinforcing particulates in ja01056a002.
an Al-7wt%Si alloy and their effects on mechanical properties. J Alloys Compd [35] Miedema AR. The electronegativity parameter for transition metals: heat of
2019;792:365–74. https://doi.org/10.1016/j.jallcom.2019.04.064. formation and charge transfer in alloys. J Less-Common Met 1973;32:117–36.
[19] Chen G, Geng H, Ji X, Xu P, Li X, Zhang H. Investigation of the hot deformation https://doi.org/10.1016/0022-5088(73)90078-7.
behavior and microstructure evolution of TiB2 +TiAl3/2024Al composite. J Alloys [36] Miedema AR, de Châtel PF, de Boer FR. Cohesion in alloys - fundamentals of a
Compd 2023:933. https://doi.org/10.1016/j.jallcom.2022.167765. semi-empirical model. Phys B+C 1980;100:1–28. https://doi.org/10.1016/0378-
[20] Rathod N, Menghani J. Dry sliding wear behavior and its relation to microstructure 4363(80)90054-6.
of artificially aged Al-Si-Mg/TiB2 in situ composites. Metall Mater Eng 2022;28: [37] Zhang B, Jesser WA. XX Formation energy of ternary alloy systems calculated by
269–90. https://doi.org/10.30544/824. Miedema model.pdf 2002;315:123–32.
[21] Sheelwant A, Jadhav PM, Narala SKR. ANN-GA based parametric optimization of [38] Gonçalves AP, Almeida M. Extended miedema model: predicting the formation
Al-TiB2 metal matrix composite material processing technique. Mater Today enthalpies of intermetallic phases with more than two elements. Phys B Condens
Commun 2021;27:102444. https://doi.org/10.1016/j.mtcomm.2021.102444. Matter 1996;228:289–94. https://doi.org/10.1016/S0921-4526(96)00480-2.
[22] Gupta R, Chaudhari GP, Daniel BSS. Strengthening mechanisms in ultrasonically [39] Lupis CHP, Elliott JF. Generalized interaction coefficientsPart I: Definitions. Acta
processed aluminium matrix composite with in-situ Al3Ti by salt addition. Compos Metall 1966;14:529–38. https://doi.org/10.1016/0001-6160(66)90320-8.
Part B Eng 2018;140:27–34. https://doi.org/10.1016/j.compositesb.2017.12.005. [40] Jones GP, Pearson J. In: Factors Affecting the Grain-Refinement of Aluminum
[23] Sheelwant A, Dutta S, Sonti KSM, Narala SKR. Processing and performance Using Titanium and Boron Additives A) THE PRACTICE OF GRAIN-REFINEMENT
assessment of particulate TiB2 reinforced aluminum MMC developed via a novel IN ALUMINUM ALLOYS A1) Reasons for Grain-Refinement; 1976. p. 7.
hybrid ultrasonic casting system. Mater Manuf Process 2022;37:186–96. https://
doi.org/10.1080/10426914.2021.1960996.

26