Solar Energy Materials and Solar Cells 203 (2019) 110171

Contents lists available at ScienceDirect

Solar Energy Materials and Solar Cells

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

Nanoparticles as a high-temperature anticorrosion additive to molten nitrate T

salts for concentrated solar power
Udayashankar Nithiyananthama,b, Yaroslav Grosua,∗, Argyrios Anagnostopoulosc,
Enrique Carbó-Argibayd, Oleksandr Bondarchukd, Luis González-Fernándeza, Abdelali Zakia,
Josu Mirena Igartuab, María Elena Navarroc, Yulong Dingc, Abdessamad Faika,∗∗
a
CIC Energigune, Albert Einstein 48, Miñano, Álava, 01510, Spain
b
Applied Physics II Department, Faculty of Science and Technology, University of the Basque Country (UPV/EHU), PO Box 644, Bilbao, 48080, Spain
c
BCES Birmingham Centre of Energy Storage, University of Birmingham, United Kingdom
d
International Iberian Nanotechnology Laboratory, Av. Mestre José Veiga, s/n, 4715-330, Braga, Portugal

ARTICLE INFO ABSTRACT

Keywords: Hot corrosion is profoundly detrimental to construction elements in Concentrated Solar Power (CSP) plants
Corrosion affecting their lifetime, running costs and safety. In this work we have studied the anticorrosion effect of TiO2
Molten salt nanoparticles additives on carbon steel using XRD, XPS with depth profiling, EDX and FIB/SEM techniques. The
Thermal energy storage results revealed that the addition of 1 wt% of TiO2 nanoparticles to molten binary nitrate salt reduces the
Concentrated solar power
corrosion rate of carbon steel more than twice and stabilizes the corrosion scale at 390 °C. The anticorrosion
effect of TiO2 nanoadditive was attributed to the formation of iron-titanium mixed oxide on the carbon steel
surface. It was confirmed by XRD and TGA techniques that addition of TiO2 nanoparticles does not alter the
stability of the salt. In view of presented results, the feasibility of molten salts based nanofluids in the CSPs can
be reconsidered in terms of improved compatibility with construction materials.

1. Introduction
Reaching necessary dispatchability is a major challenge for renew-
able technologies, which is typically addressed through efficient energy
storage. Concentrated solar power (CSP) technology [1] possesses
several important advantages in this sense, as it allows for reliable, low-
cost and relatively simple thermal energy storage (TES) system [2,3]
instead of more expensive and complex chemical or mechanical storage
systems [4]. There is a great deal of effort from scientific and en-
gineering community for further improvement of TES technologies at
CSP plants by using two-tank or single-tank thermocline configurations
[5], where different storage materials and heat transfer fluids (HTF),
like molten salts, thermal oils, air, phase change materials, etc., can be
used [6]. Corrosion issues [7], particularly with most common storage
material NaNO3–KNO3 molten salt [8], are considered important not
only in terms of safety and power plant reliability, but also in terms of
the electricity cost reduction [9,10].
With these regards, a rapidly growing number of works dedicated to
the corrosion of molten salts are evident in the literature [11–47].
Nitrate salts are often the object of investigation due to the high level of
maturity of this storage material. HitecXL salt (48 wt% Ca(NO3)2–7wt%
NaNO3–45 wt% KNO3) was studied in terms of corrosion with AISI 304,
430 stainless steels, low-Cr alloy steel T22 [11] and with carbon steel
A516Gr70, stainless steels AISI 304, AISI 316 [24]. The increase in
corrosion rates due to the chlorides in molten nitrate salts was de-
monstrated for Hitec salt (53 wt% KNO3–40 wt% NaNO2–7wt% NaNO3)
[12], Solar salt (60 wt% NaNO3 – 40 wt% KNO3) [13] and
KNO3–NaNO2–NaNO3–KCl salt [14]. On the contrary, the positive effect
of chromium in steels on the compatibility between different molten
nitrate salts and construction materials has been reported [16,21]. The
corrosion rates generally increase with temperature as was demon-
strated for stainless steels SS-304 and SS-316 at 570 °C and for carbon
steel A36 at 316 °C with several nitrate salts mixtures [17]. The same
negative effect of temperature stands for Solar salt, which was tested at
550 °C [20], 650 °C [21], 680 °C [25] and for Hitec salt tested at 450,
600 and 680 °C [26]. It should be noted that corrosion rate is often non-
linearly depends on temperature, for example, SS-316 steel was de-
monstrated to be resistant at 450 °C, while being severely corroded at

∗
Corresponding author.
∗∗
Corresponding author.
E-mail addresses: ygrosu@cicenergigune.com (Y. Grosu), afaik@cicenergigune.com (A. Faik).

https://doi.org/10.1016/j.solmat.2019.110171
Received 17 June 2019; Received in revised form 26 August 2019; Accepted 8 September 2019
Available online 13 September 2019
0927-0248/ © 2019 Elsevier B.V. All rights reserved.
U. Nithiyanantham, et al.
600 and 680 °C by Hitec salt [26].
It is highly important to carry out the corrosion tests under the
conditions close to those of real CSP plants. Ruiz-Cabañas et al. in-
vestigated the corrosion of carbon steel (CS) A516.Gr70 at the pilot
plant scale by continuously or partially exposing it to solar salt at 390 °C
[15]. From the analysis, the continuously exposed CS shows 5.46 μm/
year and partially exposed CS shows 3.57 μm/year corrosion rates [15].
Wang et al. studied compatibility of X80 carbon steel, 304, 316L
stainless steels, and 600, 825 nickel alloys with Solar salt and HITEC
salt under static and dynamic conditions [18]. The authors demon-
strated the corrosion resistance behaviour of the stainless steel and
nickel based alloy at 450 °C in the following order
304 < 316L < 600 < 825. They also found that the stainless steel
has lower corrosion resistance compared to nickel based alloy [18].
Encinas-Sanchez et al. presented the electrochemical impedance spec-
troscopy method for monitoring corrosion process at the CSP plants
[19]. The corrosion test was performed by using P91 steel at 580 °C for
the time period of 1000h and the observed corrosion rate of ∼300 μm/
year. They also found the maximum mass gain of 0.59 mg/cm2 after
72 h corrosion test [19]. Interestingly the mass gain reduced with fur-
ther exposure time due to the detachments of the corrosion layer from
the P91 steel [19].
Walczak et al. described the corrosion problems relevant to TES
systems using different combinations of construction materials and
molten salts [10]. A recent review on the corrosion aspects of molten
nitrate salts is available in Ref. [33]. Corrosion issues for molten salts
other than nitrates are also actively discussed. In particular, carbonate
[23,41] and chloride [34–42] salts in view of their potential to be used
at the third generation CSP plants at temperatures above 600 °C. A re-
view on sensible storage materials at temperatures above 600 °C, in-
cluding corrosion aspects, can be found in Ref. [48].
Several anticorrosion methods were proposed to cope with com-
patibility issue in the CSP plants. In particular, Al slurry coating reduces
corrosion rates in nitrate salts at 580 °C and in carbonate salts at 650 °C
[29]. Recently we proposed a spry-graphitization coating for carbon
steel against corrosion attack of molten nitrate salts under cycling
conditions up to 500 °C [22,30]. Encinas-Sanchez et al. tested
ZrO2–Y2O3 coating for protecting P91 against molten nitrate salt at
500 °C, demonstrating environmental benefits of such combination
compared to SS-304 alloy [31]. Porcayo-Calderon et al. compared
Ni20Cr coating to SS-304 in molten ZnCl2–KCl salt at 350, 400 and
450 °C [43]. Gomez-Vidal et al. used pre-oxidation for alumina forming
alloys to improve the corrosion resistance under isothermal [45] and
cycling [35] conditions. MCrAlX coatings were proposed against cor-
rosion of molten chlorides salts [46]. Several strategies, namely, the
addition of Mg and pre-oxidation, were suggested by Ding et al. to re-
duce the corrosion attack of chloride salts at 700 °C [47].
On the other hand, enhancement of thermophysical properties of
molten salts due to nanoparticles doping (Al2O3, Graphite, SiO2, TiO2,
etc.) are actively discussed in the literature. In particular, the en-
hancement of thermophysical properties for Al2O3 and SiO2 based na-
nofluids are well-known and studied by several researchers [48,49].
Furthermore, the in-situ preparation of the TiO2 based nanofluids shows
7.5% enhancement of its specific heat capacity [50]. However, there is
no information available regarding the corrosion effect of TiO2 nano-
particles doped molten nitrate salts.
With these regards, corrosion aspects of such molten salts based
nanofluids become the point of interest [27,28]. At the moment the
understanding of the influence of the nanoparticles on the corrosivity of
molten salt is very limited as well as the number of corresponding ar-
ticles [21,28,32,51,52]. Adding 1 wt% of silica nanoparticles into
42.7%Li2CO3–K2CO3 salt was reported to reduce the corrosion rates at
520 °C based on mass variation [51,52]. Fernandez et al. detected the
incorporation of alumina and silica nanoparticles into the corrosion
scale, which reduced the corrosion rates of SS-347 with Solar salt at
565 °C [28]. Earlier incorporation of Al2O3 and SiO2 nanoparticles into
2
Solar Energy Materials and Solar Cells 203 (2019) 110171
the corrosion scale was also reported for HitecXL salt explored with
carbon steel at 310 °C [27]. Most recently, our research group in-
vestigated the corrosion effect of Al2O3 and SiO2 based molten salts
nanofluids at 310 and 390 °C with CS [32]. Both Al2O3 and SiO2 doped
molten salts demonstrated around 3 times lower corrosion layer
thickness compared to pure molten salt at 390 °C [32]. However, the
corrosivity was found to be higher when nanoparticles are added to the
salt at 310°C due to numerous microbubbles of air entrapped in the
interparticle porosity, which increased the rates of oxidation [27,32].
From this limited number of works it is clear that nanoparticles have
pronounced effect on the corrosivity of molten salts, however, the
mechanism behind this effect is not yet understood. Understanding this
mechanism has tremendous potential to increase the feasibility of
molten salts based nanofluids at CSP plants, bringing the benefits not
only in terms of thermophysical properties enhancement, but also in
terms of cost reduction of the construction materials.
The aim of this work is to define the dominant phenomena affecting
the corrosivity of molten salts based nanofluids. For this purpose, a set
of techniques was applied to carbon steel tested with molten binary
nitrate salt doped with TiO2 nanoparticles. The corrosion test condi-
tions as well as materials were chosen to be representative for the real
CSP application. In particular, 390 °C is the maximum temperature for
the cold tank in commercial parabolic through system. While carbon
steel is a typical material for such tank. The corrosion rates were found
to decrease considerably when nanoparticles were added, which was
confirmed by three independent techniques (mass gain, cross-section
analysis and XPS depth profiling). It was identified that the formation of
mixed iron-titanium oxide is responsible for the stabilization of the
corrosion layer and reduction of the corrosion rates.
2. Materials and techniques
2.1. Material
The eutectic salt was prepared by mixing pre-dried NaNO3 and
KNO3 in 51:49 wt% ratio. Next, the mixture was melted at 360 °C for 6 h
and grinded using agate mortar to obtain a homogeneous fine powder
of the eutectic salt. Both the NaNO3 and KNO3 were received from SQM
with a purity of 99% and the impurities present in the salts are given
in our previous work [27]. The TiO2 nanoparticles with a particle size
of 21 nm were purchased from Sigma Aldrich. The nanofluids were
prepared by using 99 wt% of the eutectic salt with 1 wt% of nano-
particle by physical shaking method. The physical shaking process was
carried out through SPEX Sample Prep 9000-series High-Energy shaker
mill for the time period of 15 min. This method has been demonstrated
to be suitable for molten salts based nanofluids preparation [27].
A516 Gr70 carbon steel (CS) used for the corrosion tests has the
following chemical composition: 98.68% Fe, 0.97% Mn, 0.31% C and
0.04% P (weight percentage).
2.2. Corrosion test protocol
For the corrosion tests, pieces of CS A516 Gr70 having a size of
around 3 × 12 × 14 mm were used. These pieces were cleaned with
ultra-sonication for 15 min using acetone followed by ethanol and fi-
nally with DI water. It is worth to mention here that, similar to previous
work [24,27,30], the samples were used under the conditions close to
practical applications without applying any polishing or other external
treatment of the CS surface, which may enormously affect the evolution
of the corrosion rate. The mass of the CS was controlled before and after
the corrosion tests. The corrosion tests were conducted in alumina
crucible by immersing CS pieces in 5 g of eutectic salt or TiO2 based
nanofluid. The corrosion tests were conducted under air at 390 °C for
the periods of 250, 500, 1000 and 1500 h under static immersion
condition. After the corrosion tests the samples were carefully washed
with hot water.
U. Nithiyanantham, et al. Solar Energy Materials and Solar Cells 203 (2019) 110171

Fig. 1. SEM images of the main steps of cross-section view preparation using FIB technique: a) carbon steel; b) Pt protective layer of ∼500 nm deposited on carbon
steel; c) staircase-shaped cavity excavated using focused Ga ion beam; d) 52° tilted view of the cross-section.

2.3. Methods

2.3.1. Scanning electron microscopy - SEM

FEI Helios NanoLab 450S DualBeam – Focused ion beam (FIB) with
FEG SEM was used to record micrographs of the surfaces of carbon steel
samples after the corrosion tests. Also, the FIB technique was deployed
to observe cross-sectional view of the corrosion layer. Fig. 1 shows the
steps of cross-section preparation procedure. First, Pt protective layer
(∼1 μm) was deposited above the spot of interest on the carbon steel
surface (Fig. 1a) via electron beam (3 keV) stimulated decomposition of
a Pt-containing gaseous precursor (Fig. 1b). Next, a staircase-shaped
cavity was excavated using a focused Ga ion beam (30 keV, 9.6 nA)
(Fig. 1c). The dimensions of this cavity were around 15 μm wide, 30 μm
long and 20 μm deep at the cross-sectional plane. Finally, polishing of
the cross-section surface was performed by low-current (30 keV,
100 pA) focused Ga ion beam followed by 52° tilt of the sample for the
cross-section analysis (Fig. 1d). Using a low-voltage (3 keV) and low-
current polishing combined with Pt protective layer, allowed to pre-
serve the corrosion layer in its initial form and to gain the details of its
Fig. 2. Raman spectrum of TiO2–nanoparticles. Insert: SEM micrograph.
topology usually inaccessible when mechanical polishing is applied.
Additionally, Quanta 200 FEG SEM was used in high vacuum mode

Fig. 3. SEM micrographs and EDX-analysis of the surface of carbon steel after 1500h corrosion test at 390 °C with nanoparticles-free nitrate salt (a, c, d) and nitrate
salt doped with 1 wt% of TiO2 nanoparticles (b, e, f).

3
U. Nithiyanantham, et al. Solar Energy Materials and Solar Cells 203 (2019) 110171

Fig. 4. SEM micrographs and EDX-mapping of the surface of carbon steel after 1500h corrosion test at 390 °C with nitrate salt doped with 1 wt% of TiO2 nano-
particles.

Fig. 6. XRD diffractograms of carbon steel before and after the 1500h corrosion
tests at 390 °C with nitrate salt and nitrate salt doped with 1 wt% of TiO2 na-
noparticles.
Fig. 5. XPS spectra of the surfaces of carbon steel (CS) after the 1500h corrosion
tests with nitrate salt and nitrate salt doped with 1 wt% of TiO2 nanoparticles
(np). Table 1
Quantitative XPS analysis (at%) of carbon steel after 1500h corrosion tests at
390 °C with nitrate salt (CS) and nitrate salt doped with nanoparticles
with electron beam energies 10 kV, 20 kV and 30 kV with BSED and
(CS + TiO2).
ETD detectors. Energy-Dispersive X-Ray Spectroscopy (EDX) was per-
formed for elemental mapping. Ti2p Fe2p O1s C1s Mg

For observing cross-section in SEM, the samples were put in a resin CS + TiO2 5 26 47 22 –
holder made of Aka-Resin and Aka-Cure catalyst (1 mL + 0,135 mL CS – 32 66 1 1
proportion respectively) to maintain the corrosion layer intact during
the cutting and polishing processes.
2.3.3. Thermogravimetric analysis – TGA
For the thermogravimetric analysis NETZSCH STA 449 F3 Jupiter
2.3.2. X-ray diffraction - XRD was used with a constant airflow of 60 mL/min in the temperature
Bruker D8 Discover X-ray diffractometer was used with a LYNXEYE- range from 40 °C to 700 °C with the heating rate of 10°C / min.
XE detector using CuKα1 radiation (λ = 1.5418 Å) and Bragg-Brenato
θ:2θ geometry. The data collection was carried out at room tempera- 2.3.4. X-ray photoelectron spectroscopy characterization – XPS
ture, between 10° and 80° with a step of 0,02° and dwell time of 1,03s The XPS measurements were performed in an ultra-high vacuum
per step. The EVA program was used to determine the phase compo- (UHV) system ESCALAB250Xi (Thermo Fisher Scientific). The base
sition of the material. pressure in the system was below 5x10−10 mbar. XPS spectra were

4
U. Nithiyanantham, et al.
Figure 7. a) mass gain and b) corrosion layer thickness for carbon steel tested with molten nitrate salt and nitrate salt doped with 1 wt% of TiO2 nanoparticles at
390 °C.
Fig. 8. Cross-section of carbon steel after 1500h corrosion test at 390 °C with a)
nitrate salt and b) nitrate salt doped with 1 wt% of TiO2 nanoparticles.
acquired with a hemispherical analyzer and a monochromated X-ray
source (Al Kα radiation, hν=1486.6 eV) operated at 15 keV and power
200 W. The XPS spectra were recorded with pass energies 20 eV and
200 eV for high resolution and survey spectra, respectively. The spec-
trometer was calibrated by setting the Au 4f7/2 level to 84.0 eV mea-
sured on a gold foil.
XPS spectra were peak-fitted using Avantage data processing soft-
ware. For peak fitting, the Shirley-type background subtraction was
used. Quantification was done using elemental sensitivity factors pro-
vided by the Avantage library.
The XPS system features a dual mode Ar ion source MAGCIS
(Thermo Fisher Scientific) which can operate in monoatomic and
cluster modes. The samples’ surfaces before XPS characterization were
cleaned by using Ar cluster ions of 1000 atoms. For Ar cluster cleaning
an accelerating voltage of 6 keV was used. For depth profiling the
MAGCIS was operated in the monoatomic mode with beam energy
4 keV and beam current 3 μA.
2.3.5. Raman spectroscopy
Raman spectra were recorded with Alpha300R spectrometer from
WITec using 532 nm laser operated at 9x10−4 W. No radiation induced
effects were observed during the measurements. The accumulations of
100 cycles were used for the measurements to achieve a sufficient
signal-to-noise ratio.
5
Solar Energy Materials and Solar Cells 203 (2019) 110171
2.3.6. Contact angle measurements
A KRUSS HT-2 Contact Angle was used to perform the contact angle
(CA) measurements. All the CA measurements were conducted under
ambient atmosphere, with a 2 °C/min heating rate. The presented CA
values are the average of 3 measurements. CA values dispersion is
subject to several factors, among which surface roughness and drop
dimension are the most pertinent. To minimize their effect, a sample
preparation protocol was followed. The salt eutectic powder was
weighed, then milled and afterwards compressed at 40 MPa for 3 min in
a 4 mm diameter round compression dye, to form same size pellets of
0.1g. The measurements were conducted on a carbon steel substrate
(20 mm × 20 mm x 3 mm), which roughness was measured before and
after each measurement using an EPS Interferometer. The surface
roughness is found to have a constant value of S = 0.43 ± 0.20 μm.
3. Results
SEM imaging of TiO2 nanoparticles (Fig. 2, insert) confirmed the
average size of 21 nm specified by the supplier. Raman spectroscopy is
the technique of choice for TiO2 polymorphs characterization. Fig. 2
shows a Raman spectrum from the TiO2 nanoparticles. The main peaks
centred at 148 cm −1, ∼400 cm −1, ∼520 cm −1 and ∼640 cm −1 are
close to the peaks’ positions in anatase, thus concluding that the TiO2
nanoparticles used in this study are mainly anatase polymorph.
Fig. 3 presents SEM images of the carbon steel surface after corro-
sion test with pure nitrate salt (Fig. 3a) and with nanoparticles doped
nitrate salt (Fig. 3b).
A pronounced separation of the corrosion layer (peel-off) in the case
of pure nitrate salt is clearly seen (Fig. 3a), while homogeneous cor-
rosion appears when TiO2 nanoparticles were added (Fig. 3b). The
high-magnification micrographs reveal the difference in the corrosion
layer morphology (Fig. 3c–d versus Fig. 3e–f). Specifically, one can see
that the nanoparticles are merging with the typical flakes of iron oxide
(Fig. 3f) - the nanoparticles no longer possess the well-defined crystal
shape as compared to the pristine nanoparticles (Fig. 2, insert). Such
observation is in favour of the chemical reaction between the nano-
particles and the corrosion layer. Another argument to support the idea
about possible chemical reactions between TiO2 and carbon steel comes
from the EDX-analysis. Tables in Fig. 3 list elemental concentrations for
the corresponded surfaces. In case of nanoparticles-free nitrate salt
apart from iron and oxygen, magnesium was detected. Magnesium is a
common impurity in nitrate salts and is often detected in the corrosion
layers [11,15,24,27,30]. When salt without nanoparticles is used, Mg
reacts with carbon steel during the corrosion test and incorporates into
U. Nithiyanantham, et al.
Fig. 9. XPS depth profiling for carbon steel after 1500h corrosion test at 390 °C
with a) nitrate salt and b) nitrate salt doped with 1 wt% of TiO2 nanoparticles.
Red lines present depth profile of the iron oxide component (FeOx) of the Fe2p
XPS spectral line. FeOx component intensity represents contribution from
several oxidation states of iron excluding metallic iron. (For interpretation of
the references to colour in this figure legend, the reader is referred to the Web
version of this article.)
Fig. 10. Schematic representation of corrosion layer of carbon steel after cor-
rosion test at 390 °C with molten a) nitrate salt and b) nitrate salt doped with
TiO2 nanoparticles based on XPS depth profiling and XRD analysis.
the corrosion layer. This was demonstrated using XPS (Fig. 5) and EDX
(Fig. 3) techniques. While when TiO2 is present in the salt, TiO2 reacts
with the carbon steel instead of magnesium. This was demonstrated by
EDX (Fig. 3), XPS (Fig. 5) and XRD (Fig. 6). The reactions of TiO2 with
carbon steel is preferable to the reaction of magnesium with carbon
6
Solar Energy Materials and Solar Cells 203 (2019) 110171
steel. That is why Mg is not found in the corrosion layer after the
corrosion tests with TiO2 doped salt. The similar mechanism was found
in our previous work with Al2O3 and SiO2 nanoparticles [32]. On the
other hand, for the nitrate salt doped with TiO2 nanoparticles titanium
was detected instead of magnesium, which indicates the predominance
of the TiO2-iron reaction over the magnesium-iron reaction.
TiO2 was found all over the surface of the carbon steel. In some
places, titanium forms large agglomerates ∼5 μm in size, as it is evi-
denced from Fig. 4.
XPS spectra shown in Fig. 5 confirm the above mentioned ob-
servations, namely, the presence of titanium and the absence of mag-
nesium in the corrosion layer of carbon steel for the case of nitrate salt
doped with TiO2 nanoparticles (green line in Fig. 5) while the presence
of magnesium was detected in the case of nanoparticle-free nitrate salt
(red line in Fig. 5).
XRD diffractograms presented in Fig. 6 reveal the evolution of the
chemical phase composition of carbon steel after the corrosion tests
with the nanoparticle-free nitrate salt and with the TiO2 nanoparticles
doped nitrate salt. In particular, the surface of carbon steel after 1500h
corrosion tests with nanoparticles-free nitrate salt contains two phases,
namely, iron (Fe) and hematite (Fe2O3). On the other hand, the cor-
rosion products after the tests with nitrate salt doped with TiO2 are
formed by magnetite (Fe3O4), which is a lower degree of oxidation
compared to hematite, and iron-titanium mixed oxide (Fe1.7Ti0.3O3).
The main mechanism for the reduced corrosion rates in the molten salt
doped with TiO2 was found to be the formation of mixed iron-titanium
oxide phase (Fe1.7Ti0.3O3) on the CS surfaces (Fig. 6). This mixed oxide
layer was found to be more stable as compared to iron oxide layer,
which prevented the peel-off of the corrosion scale (Fig. 3) and con-
sequently reduced the rate of further oxidation. The formed phase
suggests that upon oxidation of carbon steel in molten nitrate salt some
Fe3+ iron atoms are substituted by titanium in hematite phase. For-
mation of such mixed oxide can be an explanation of the stabilization
effect, which prevents corrosion layer from peeling off (Fig. 3). Quan-
titative XPS analysis of the carbon steel surface elemental composition
after corrosion tests (Table 1) gives Ti:Fe ratio ∼0.20 which is in accord
with the XRD derived result: Ti:Fe = 0.18.
In order to quantify the effect of iron-titanium mixed oxide forma-
tion in the corrosion layer, the rate of oxidation was determined using
three independent techniques. First, the mass gain evolution was ex-
amined, which was found to be more than twice lower for nitrate salt
doped with nanoparticles than that for the nanoparticles-free salt after
1500h corrosion tests – Fig. 7a. Most probably, the lower mass gain and
corrosion layer thickness at 500h were obtained for the pure molten salt
due to the corrosion layer peel-off (Fig. 3). Apparently, such peel-off
becomes predominant at 500h. This question is directly related with the
kinetics of the corrosion layer formation, which requires more study
and is planned in the future. Next, a similar trend was observed for the
corrosion layer evolution (Fig. 7b), which was determined from the
statistical examination of the cross-sections of carbon steel (Fig. 8). For
example, it was demonstrated by Wang et al. that the corrosion rate of
CS is decreasing after certain period of time while being continuously
exposed to the molten salt [18]. This was related with the reduction of
oxygen and water vapor compared to their initial state and also for-
mation of a new mixed oxide phases [18]. Finally, the obtained results
were confirmed by the XPS depth profiling technique, which revealed
the enhanced thickness of the oxidation layer for nanoparticles-free
nitrate salt compared to the salt doped with TiO2 nanoparticles – Fig. 9.
In particular, the average corrosion layer after 1500h tests for pure
molten salt is 11.6 μm and for TiO2 doped molten salt is 5.2 μm. These
results reveal that the addition of 1 wt% of TiO2 nanoparticles to molten
binary nitrate salt reduced the corrosion rate of carbon steel more than
twice at 390 °C and had a stabilizing effect on the corrosion scale.
Additional important observation can be extracted from XPS depth
profiling experiments. In the case of nanoparticles-free nitrate salt
magnesium incorporation into the corrosion layer of carbon steel can be
U. Nithiyanantham, et al. Solar Energy Materials and Solar Cells 203 (2019) 110171

Fig. 11. SEM images of the cross-section view of carbon steel after 1500h corrosion test at 390 °C with a) nitrate salt and b) nitrate salt doped with 1 wt% of TiO2
nanoparticles. Obtained with focused ion beam technique.

Fig. 13. XRD pattern of binary nitrate salt doped with 1 wt% TiO2 nanoparticles
after 1500h corrosion test with carbon steel at 390 °C.

Fig. 12. Contact angles of a) molten nitrate salt, b) molten binary nitrate
salt + 1 wt% of TiO2 NPs and c) their temperature dependence.

tracked all along the oxide layer (Fig. 9a, green line). This suggests that
Mg forms mixed oxide with iron upon oxidation in molten nitrate salt.
We were not able to determine the Mg containing phase by means of
XRD due to its very low content as can be seen from the EDX (Fig. 3a)
and from the XPS (Figs. 5 and 9a) data. On the contrary, titanium
containing phase (Fe1.7Ti0.3O3) was found to be located mostly in the
topmost surface of the corrosion layer (Fig. 9b). In particular, the
concentration of Ti drops exponentially after sputtering away ∼1 μm of Fig. 14. Thermogravimetric curves of as prepared nitrate salt, nitrate salt +
material and continues to decrease with depth in the corrosion scale. 1 wt% TiO2 nanoparticles and nitrate salt + 1 wt% TiO2 nanoparticles after
1500h corrosion test at 390 °C.
Such observation allows suggesting that after the formation of the
mixed iron-titanium oxide further oxidation of carbon steel takes place
by the diffusion of oxygen through the Fe1.7Ti0.3O3 phase. Such diffu- Based on the XPS depth profiling and XRD analysis presented above,
sion is apparently slower as compared to the diffusion through the while taking into account EDX mapping, one can schematically re-
mixed iron-magnesium oxide, which results in the lower corrosion rates present the corrosion scale of carbon steel after the corrosion tests with
in the case of nitrate salt doped with nanoparticles compared to na- nanoparticles-free molten nitrate salt and nitrate salt doped with TiO2
noparticles-free salt. nanoparticles (Fig. 10).

7
U. Nithiyanantham, et al.
Finally, focused ion beam technique was used to obtain textural
properties of the corrosion layers' cross-section. Due to the low-voltage
ion polishing the cross-section of the corrosion scale reveals the textural
features (like cracks, voids, etc), which are typically lost during the
mechanical polishing applied in conventional cross-section preparation.
Fig. 11 clearly shows that corrosion layer of the carbon steel after the
corrosion tests with molten nitrate salt poses more cracks (Fig. 11a) as
compared to the case of molten salt doped with TiO2 nanoparticles
(Fig. 11b).
All the described above observations provide strong evidence of
anticorrosion effect of TiO2 nanoparticles doping into the molten ni-
trate salt. This effect we assign to formation of mixed iron-titanium
oxides. It is known from literature that the doping by other nano-
particles like silica [21,28,32,51,52] or alumina [27,28,32] also results
in their incorporation into the corrosion layer. In particular, the in-
corporation of Al2O3 [28] and SiO2 [51,52] nanoparticles was reported
to decrease the corrosivity of a molten salt. While in other work SiO2
nanoparticles incorporation had a negligible effect on the corrosion
rates [28]. In the case when corrosion tests were performed at lower
temperature (310 °C) micro-bubbling of the nanofluids were found to be
a dominating effect which led to increased corrosion rates, even
though, the incorporation of Al2O3 and SiO2 nanoparticles was also
reported [27,32]. In this work, we have found that the mechanism of
TiO2 nanoparticles incorporation involves the formation of mixed oxide
with iron. Such results may also shed some light on the possible me-
chanism of SiO2 and Al2O3 nanoparticles doping. In particular, in work
[28] iron-silicon mixed oxide was reported as a corrosion product after
corrosion test of Solar salt with SS 347. However, this Fe2SiO4 phase did
not have a predominant role for the corrosion rates and the overall
corrosivity of the salt did not change despite of nanoparticles doping
[28].
It is generally known that the nanoparticle doping can change the
wetting properties of a fluid [53], which pertains to corrosion phe-
nomena. With these regards, high-temperature contact angle measure-
ments were performed for molten salts and salt doped with nano-
particles. It was found that in the investigated temperature range,
nanoparticles doping did not change the wetting properties of molten
nitrate salt (Fig. 12). Therefore, obtained reduction of corrosivity can
be attributed to the mixed oxide formation, rather than to the wetting
properties of the nanofluid.
It should be noted that both salts and nanoparticles were examined
in terms of possible degradation after the corrosion tests by XRD
(Fig. 13) and no evidence of such modifications were detected. It was
also verified that nanoparticles doping does not affect the decomposi-
tion temperature (and hence the operational temperature range) of the
salt (Fig. 14).
4. Conclusions
In this work corrosivity of molten salts based nanofluids has been
explored experimentally by means of XRD, XPS, EDX, FIB/SEM tech-
niques. In particular, the effect of TiO2 nanoparticles on the corrosion
rates of eutectic NaNO3–KNO3 molten salt was investigated for the first
time. The following conclusions can be formulated:
• Addition of 1 wt% of TiO 2 nanoparticles to molten binary nitrate salt
reduced corrosion rate of carbon steel tested at 390 °C more than
twice, which was demonstrated by mass gain, cross-section analysis
and XPS depth profiling.
• The presence of TiO2 nanoparticles in the salt resulted in a homo-
geneous oxidation of the surface of carbon steel at 390 °C, which is
in drastic contrast to the case of nanoparticles-free salt, where
pronounced corrosion scale separation and peel-off were evident
from SEM experiments.
•
Formation of mixed iron-titanium oxide (Fe1.7Ti0.3O3) was demon-
strated by XRD analysis and suggested to be responsible for the
8
Solar Energy Materials and Solar Cells 203 (2019) 110171
corrosion rates reduction and corrosion scale stabilization. By means
of XRD, EDX and XPS it was shown that growth of surface iron-
titanium oxide is competitive to formation of iron-magnesium oxide
- a chemical phase formed due to the presence of Mg impurity in the
nitrate salt.
• By means of XPS depth profiling, it was demonstrated that such
mixed oxide is located on the surface of corrosion scale on top of
magnetite phase, hindering the diffusion of molten salt and further
oxidation of carbon steel.
• The investigated molten salt based nanofluid was found to have si-
milar decomposition temperature and surface properties (contact
angle on the carbon steel) as nanoparticles-free nitrate salt.
Obtained results provide important insights into the corrosion me-
chanisms of molten salts based nanofluids and allow to reconsider their
feasibility for CSP applications due to the improved compatibility.
Further work should be focused on the optimization of the concentra-
tion of TiO2 nanoparticles, as well as on the feasibility analysis of such
nanofluids.
Acknowledgements
UN gratefully acknowledges the Nanouptake COST Action CA15119
for the funding of a short-term scientific mission (STSM-Grant No
42643), between CIC Energigune and University of Birmingham Center
for Energy storage. YG gratefully acknowledges the Nanouptake COST
Action CA15119 for the funding of a short-term scientific mission
(STSM-Grant No 42475), between CIC Energigune and International
Iberian Nanotechnology Laboratory. The authors express their sincere
thanks to Yagmur Polat, Leticia Martinez and Cristina Luengo for their
technical support.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.solmat.2019.110171.
References
[1] E. Zarza-Moya, A Comprehensive Guide to Solar Energy Systems, first ed., Academic
Press, 2018.
[2] H. Zhang, J. Baeyens, G. Cáceres, J. Degrève, Y. Lv, Thermal energy storage: recent
developments and practical aspects, Prog. Energy Combust. Sci. 53 (2016) 1–40
https://doi.org/10.1016/j.pecs.2015.10.003.
[3] J. Lilliestam, T. Barradi, N. Caldés, M. Gomez, S. Hanger, J. Kern,
N. Komendantova, M. Mehos, W.M. Hong, Z. Wang, A. Patt, Policies to keep and
expand the option of concentrating solar power for dispatchable renewable elec-
tricity, Energy Policy 116 (2018) 193–197 https://doi.org/10.1016/j.enpol.2018.
02.014.
[4] S. Kuravi, J. Trahan, D.Y. Goswami, M.M. Rahman, E.K. Stefanakos, Thermal energy
storage technologies and systems for concentrating solar power plants, Prog. Energy
Combust. 39 (2013) 285–319 https://doi.org/10.1016/j.pecs.2013.02.001.
[5] B. chen Zhao, M. song Cheng, C. Liu, Z. min Dai, System-level performance opti-
mization of molten-salt packed-bed thermal energy storage for concentrating solar
power, Appl. Energy 226 (2018) 225–239 https://doi.org/10.1016/j.apenergy.
2018.05.081.
[6] T. Esence, A. Bruch, J. Fourmigué, B. Stutz, A versatile one-dimensional numerical
model for packed-bed heat storage systems, Renew. Energy 133 (2019) 190–204
https://doi.org/10.1016/j.renene.2018.10.012.
[7] G. Peiró, C. Prieto, J. Gasia, A. Jové, L. Miró, L.F. Cabeza, Two-tank molten salts
thermal energy storage system for solar power plants at pilot plant scale: lessons
learnt and recommendations for its design, start-up and operation, Renew. Energy
121 (2018) 236–248 https://doi.org/10.1016/j.renene.2018.01.026.
[8] X. Li, E. Xu, S. Song, X. Wang, G. Yuan, Dynamic simulation of two-tank indirect
thermal energy storage system with molten salt, Renew. Energy 113 (2017)
1311–1319 https://doi.org/10.1016/j.renene.2017.06.024.
[9] A. Gottschalk, U. Ramamoorthi, Parametric simulation and economic estimation of
thermal energy storage in solar power tower, Mater. Today: Proc. 5 (2018)
1571–1577 https://doi.org/10.1016/j.matpr.2017.11.248.
[10] M. Walczak, F. Pineda, Á.G. Fernández, C. Mata-Torres, R.A. Escobar, Materials
corrosion for thermal energy storage systems in concentrated solar power plants,
Renew. Sustain. Energy Rev. 86 (2018) 22–44 https://doi.org/10.1016/j.rser.2018.
01.010.
[11] A.G. Fernández, H. Galleguillos, E. Fuentealba, F.J. Pérez, Corrosion of stainless
U. Nithiyanantham, et al.
steels and low-Cr steel in molten Ca(NO3)2–NaNO3–KNO3 eutectic salt for direct
energy storage in CSP plants, Sol. Energy Mater. Sol. Cells 141 (2015) 7–13 https://
doi.org/10.1016/j.solmat.2015.05.004.
[12] K. Federsel, J. Wortmann, M. Ladenberger, High-temperature and corrosion beha-
vior of nitrate nitrite molten salt mixtures regarding their application in con-
centrating solar power plants, Energy Procedia 69 (2015) 618–625 https://doi.org/
10.1016/j.egypro.2015.03.071.
[13] A.S. Dorcheh, R.N. Durham, M.C. Galetz, High temperature corrosion in molten
solar salt: the role of chloride impurities, Mater. Corros. (2017) 1–9 https://doi.org/
10.1002/maco.201609300.
[14] W. Wang, B. Guan, X. Li, J. Lu, J. Ding, Corrosion behavior and mechanism of
austenitic stainless steels in a new quaternary molten salt for concentrating solar
power, Sol. Energy Mater. Sol. Cells 194 (2019) 36–46 https://doi.org/10.1016/j.
solmat.2019.01.024.
[15] F.J. Ruiz-cabañas, C. Prieto, R. Osuna, V. Madina, A.I. Fernández, L.F. Cabeza,
Corrosion testing device for in-situ corrosion characterization in operational molten
salts storage tanks: a516 Gr70 carbon steel performance under molten salts ex-
posure, Sol. Energy Mater. Sol. Cells 157 (2016) 383–392 https://doi.org/10.1016/
j.solmat.2016.06.005.
[16] W. Cheng, D. Chen, C. Wang, High-temperature corrosion of Cr-Mo steel in molten
LiNO3-NaNO3-KNO3 eutectic salt for thermal energy storage, Sol. Energy Mater. Sol.
Cells 132 (2015) 563–569 https://doi.org/10.1016/j.solmat.2014.10.007.
[17] S. Goods, R. Bradshaw, Corrosion of stainless steels and carbon steel by molten
mixtures of commercial nitrate salts, J. Mater. Eng. Perform. 13 (2004) 78–87
https://doi.org/10.1361/10599490417542.
[18] J. Wang, Y. Jiang, Y. Ni, A. Wu, J. Li, Investigation on static and dynamic corrosion
behaviors of thermal energy transfer and storage system materials by molten salts in
concentrating solar power plants, Mater. Corros. 70 (2019) 102–109 https://doi.
org/10.1002/maco.201810362.
[19] V. Encinas-Sánchez, M.T. de Miguel, M.I. Lasanta, G. García-Martín, F.J. Pérez,
Electrochemical impedance spectroscopy (EIS): an efficient technique for mon-
itoring corrosion processes in molten salt environments in CSP applications, Sol.
Energy Mater. Sol. Cells 191 (2019) 157–163 https://doi.org/10.1016/j.solmat.
2018.11.007.
[20] A. Gomes, M. Navas, N. Uranga, T. Paiva, I. Figueira, T.C. Diamantino, High-tem-
perature corrosion performance of austenitic stainless steels type AISI 316L and
AISI 321H, in molten solar salt, Sol. Energy 177 (2019) 408–419 https://doi.org/
10.1016/j.solener.2018.11.019.
[21] J.W. Slusser, J.B. Titcomb, M.T. Heffelfinger, B.R. Dunbobbin, Corrosion in molten
nitrate-nitrite salts, J. Met. 37 (1985) 24–27 https://doi.org/10.1007/BF03259692.
[22] J. Piquot, U. Nithiyanantham, Y. Grosu, A. Faik, Spray-graphitization as a protec-
tion method against corrosion by molten nitrate salts and molten salts based na-
nofluids for thermal energy storage applications, Sol. Energy Mater. Sol. Cells 200
(2019) 110024 https://doi.org/10.1016/j.solmat.2019.110024.
[23] M.T. de Miguel, V. Encinas-Sánchez, M.I. Lasanta, G. García-Martín, F.J. Pérez,
Corrosion resistance of HR3C to a carbonate molten salt for energy storage appli-
cations in CSP plants, Sol. Energy Mater. Sol. Cells 157 (2016) 966–972 https://doi.
org/10.1016/j.solmat.2016.08.014.
[24] Y. Grosu, O. Bondarchuk, A. Faik, The effect of humidity, impurities and initial state
on the corrosion of carbon and stainless steels in molten HitecXL salt for CSP ap-
plication, Sol. Energy Mater. Sol. Cells 174 (2018) 34–41 https://doi.org/10.1016/
j.solmat.2017.08.026.
[25] A.M. Kruizenga, D.D. Gill, M.E. LaFord, G. McConohy, Corrosion of High
Temperature Alloys in Solar Salt at 400, 500, and 680oC, SANDIA REPORT, SAND
2013-8256 California, (2013).
[26] M. Zhu, S. Zeng, H. Zhang, J. Li, B. Cao, Electrochemical study on the corrosion
behaviors of 316 SS in HITEC molten salt at different temperatures, Sol. Energy
Mater. Sol. Cells 186 (2018) 200–207 https://doi.org/10.1016/j.solmat.2018.06.
044.
[27] Y. Grosu, N. Udayashankar, O. Bondarchuk, L. González-Fernández, A. Faik,
Unexpected effect of nanoparticles doping on the corrosivity of molten nitrate salt
for thermal energy storage, Sol. Energy Mater. Sol. Cells 178 (2018) 91–97 https://
doi.org/10.1016/j.solmat.2018.01.002.
[28] A.G. Fernández, B. Muñoz-Sánchez, j. Nieto-Maestre, A. García-Romero, High
temperature corrosion behavior on molten nitrate salt-based nanofluids for CSP
plants, Renew. Energy 130 (2019) 902–909 https://doi.org/10.1016/j.renene.
2018.07.018.
[29] A. Agüero, P. Audigié, S. Rodríguez, V. Encinas-Sánched, Protective coatings for
high temperature molten salt heat storage systems in solar concentration power
plants, AIP Conf. Proc. 2033 (2017) 090001, , https://doi.org/10.1063/1.5067095.
[30] Y. Grosu, U. Nithiyanantham, A. Zaki, A. Faik, A simple method for the inhibition of
the corrosion of carbon steel by molten nitrate salt for thermal storage in con-
centrating solar power applications, npj Mater. Degrad. 2 (2018) 34 https://doi.
org/10.1038/s41529-018-0055-0.
[31] V. Encinas-Sánchez, E. Batuecas, A. Macías-García, C. Mayo, R. Díaz, F.J. Pérez,
Corrosion resistance of protective coatings against molten nitrate salts for thermal
energy storage and their environmental impact in CSP technology, Sol. Energy 176
(2018) 688–697 https://doi.org/10.1016/j.solener.2018.10.083.
[32] U. Nithiyanantham, Y. Grosu, L. González-Fernández, A. Zaki, J.M. Igartua, A. Faik,
Solar Energy Materials and Solar Cells 203 (2019) 110171
Corrosion aspects of molten nitrate salt-based nanofluids for thermal energy storage
applications, Sol. Energy 189 (2019) 219–227 https://doi.org/10.1016/j.solener.
2019.07.050.
[33] Á.G. Fernández, L.F. Cabeza, Molten salt corrosion mechanisms of nitrate based
thermal energy storage materials for concentrated solar power plants: a review, Sol.
Energy Mater. Sol. Cells 194 (2019) 160–165 https://doi.org/10.1016/j.solmat.
2019.02.012.
[34] G. Mohan, M. Venkataraman, J. Gomez-Vidal, J. Coventry, Thermo-economic
analysis of high-temperature sensible thermal storage with different ternary eu-
tectic alkali and alkaline earth metal chlorides, Sol. Energy 176 (2018) 350–357
https://doi.org/10.1016/j.solener.2018.10.008.
[35] J.C. Gomez-Vidal, A.G. Fernandez, R. Tirawat, C. Turchi, W. Huddleston, Corrosion
resistance of alumina forming alloys against molten chlorides for energy produc-
tion. II: electrochemical impedance spectroscopy under thermal cycling conditions,
Sol. Energy Mater. Sol. Cells 166 (2017) 234–245 https://doi.org/10.1016/j.
solmat.2017.03.025.
[36] J.C. Gomez-Vidal, R. Tirawat, Corrosion of alloys in a chloride molten salt (NaCl-
LiCl) for solar thermal technologies, Sol. Energy Mater. Sol. Cells 157 (2016)
234–244 https://doi.org/10.1016/j.solmat.2016.05.052.
[37] B. Liu, X. Wei, W. Wang, J. Lu, J. Ding, Corrosion behavior of Ni-based alloys in
molten NaCl-CaCl2-MgCl2 eutectic salt for concentrating solar power, Sol. Energy
Mater. Sol. Cells 170 (2017) 77–86 https://doi.org/10.1016/j.solmat.2017.05.050.
[38] W. Ding, A. Bonk, J. Gussone, T. Bauer, Electrochemical measurement of corrosive
impurities in molten chlorides for thermal energy storage, J. Energy Storage 15
(2018) 408–414 https://doi.org/10.1016/j.est.2017.12.007.
[39] K. Vignarooban, P. Pugazhendhi, C. Tucker, D. Gervasio, A.M. Kannan, Corrosion
resistance of Hastelloys in molten metal-chloride heat-transfer fluids for con-
centrating solar power applications, Sol. Energy 103 (2014) 62–69 https://doi.org/
10.1016/j.solener.2014.02.002.
[40] K. Vignarooban, X. Xu, K. Wang, E.E. Molina, P. Li, D. Gervasio, A.M. Kannan,
Vapor pressure and corrosivity of ternary metal-chloride molten-salt based heat
transfer fluids for use in concentrating solar power systems, Appl. Energy 159
(2015) 206–213 https://doi.org/10.1016/j.apenergy.2015.08.131.
[41] M. Sarvghad, T.A. Steinberg, G. Will, Corrosion of stainless steel 316 in eutectic
molten salts for thermal energy storage, Sol. Energy 172 (2018) 198–203 https://
doi.org/10.1016/j.solener.2018.03.053.
[42] H. Sun, J. Wang, Z. Li, P. Zhang, X. Su, Corrosion behavior of 316SS and Ni-based
alloys in a ternary NaCl-KCl-MgCl2 molten salt, Sol. Energy 171 (2018) 320–329
https://doi.org/10.1016/j.solener.2018.06.094.
[43] J. Porcayo-Calderon, O. Sotelo-Mazon, V.M. Salinas-Bravo, C.D. Arrieta-Gonzalez,
J.J. Ramos-Hernandez, C. Cuevas-Arteaga, Electrochemical performance of Ni20Cr
coatings applied by combustion powder spry in ZnCl2-KCl molten salts, Int. J.
Electrochem. Sci. 7 (2012) 1134–1148.
[44] H. Cho, J.W. Van Zee, S. Shimpalee, B.A. Tavakoli, J.W. Weidner, B.L. Garcia-Diaz,
M.J. Martinez-Rodriguez, L. Olson, J. Gray, Dimensionless analysis for predicting
Fe-Ni-Cr alloy corrosion in molten salt systems for concentrated solar power sys-
tems, Corrosion 72 (2016) 742–760 https://doi.org/10.5006/1865.
[45] J.C. Gomez-Vidal, A.G. Fernandez, R. Tirawat, C. Turchi, W. Huddleston, Corrosion
resistance of alumina-forming alloys against molten chlorides for energy produc-
tion. I: pre-oxidation treatment and isothermal corrosion tests, Sol. Energy Mater.
Sol. Cells 166 (2017) 222–233 https://doi.org/10.1016/j.solmat.2017.02.019.
[46] J.C. Gomez-Vidal, Corrosion resistance of MCrAlX coatings in a molten chloride for
thermal storage in concentrating solar power applications, npj Mater. Degrad. 7
(2017) 1 https://doi.org/10.1038/s41529-017-0012-3.
[47] W. Ding, H. Shi, A. Jianu, Y. Xiu, A. Bonk, A. Weisenburger, T. Bauer, Molten
chloride salts for next generation concentrated solar power plants: mitigation
strategies against corrosion of structural materials, Sol. Energy Mater. Sol. Cells 193
(2019) 298–313 https://doi.org/10.1016/j.solmat.2018.12.020.
[48] G. Mohan, M.B. Venkataraman, J. Coventry, Sensible energy storage options for
concentrating solar power plants operating above 600°C, Renew. Sustain. Energy
Rev. 107 (2019) 319–337 https://doi.org/10.1016/j.rser.2019.01.062.
[49] M. Arthur, A. Karim, An investigation into the thermophysical and rheological
properties of nanofluids for solar thermal applications, Renew. Sustain. Energy Rev.
55 (2016) 739–755 https://doi.org/10.1016/j.rser.2015.10.065.
[50] M. Lasfargues, A. Bell, Y. Ding, In situ production of titanium dioxide nanoparticles
in molten salt phase for thermal energy storage and heat transfer fluid applications,
J. Nanoparticle Res. 18 (2016) 150 https://doi.org/10.1007/s11051-016-3460-8.
[51] M. Schuller, F. Little, D. Malik, M. Betts, Q. Shao, J. Luo, W. Zhong, S. Shankar,
A. Padmanaban, Molten Salt-Carbon Nanotube Thermal Energy Storage for
Concentrating Solar Power Systems, United States: N. P. (2012), https://doi.org/10.
2172/1036948 Web https://www.osti.gov/servlets/purl/1036948.
[52] A. Padmanaban, The Effect of Silica Nanoparticles on Corrosion of Steel by Molten
Carbonate Eutectics, Master thesis (2011).
[53] M. Hernaiz, V. Alonso, P. Estellé, Z. Wu, B. Sundén, L. Doretti, S. Mancin,
N. Çobanoğlu, Z.H. Karadeniz, N. Garmendia, M. Lasheras-Zubiate, L. Hernández
López, R. Mondragón, R. Martínez-Cuenca, S. Barison, A. Kujawska, A. Turgut,
A. Amigo, G. Huminic, A. Huminic, M.R. Kalus, K.G. Schroth, M.H. Buschmann, The
contact angle of nanofluids as thermophysical property, J. Colloid Interface Sci. 547
(2019) 393–406 https://doi.org/10.1016/j.jcis.2019.04.007.