Wetting of Graphite and Platinum Substrate by Oxide System with Graded B₂O₃ Content

Abstract

This work focuses on wetting two types of substrates (a platinum substrate and a polished graphite substrate) by molten polycomponent oxide system CaO–MgO–SiO₂–Al₂O₃–B₂O₃ to test the level of interaction at high temperatures. The tested systems were subjected to high-temperature wetting tests in the temperature range from liquidus temperature to 1550 °C using the sessile drop method. A total of four oxide systems were tested with graded boron oxide contents ranging from 0 to 30 wt%. The experiments were conducted in a CLASIC high-temperature resistance observation furnace and an inert atmosphere of high-purity argon. Droplet silhouettes were obtained with a CANON EOS 550D high-resolution camera during heat treatment, with reactive and non-reactive wetting occurring depending on the substrate type. The dependence of the average wetting angles on temperature and time was evaluated, and it was found that boron oxide decreased the average wetting angles of molten oxide droplets. The analyses were accompanied by the SEM/EDX analysis of the substrate and FTIR analysis of the droplets after high-temperature experiments. The phase composition of the oxide systems was evaluated by XRD analysis.

1. Introduction

Boroaluminosilicate glasses, multicomponent boron oxide-containing oxide glasses, have a wide range of applications, from lithium battery-sealing materials and liquid crystal display substrates to nuclear waste disposal [1,2]. However, they are also crucial in the glass fiber industry, with E-type glass fibers exhibiting excellent dielectric properties due to their B₂O₃ content at the expense of significantly reduced mechanical properties [1,2,3,4]. It is worth noting that in its pure form, borate glass is not ordinarily usable for any application because it has meagre chemical resistance and a high affinity to water. Therefore, it is used in combination with other oxides, such as Al₂O₃ or SiO₂, which leads to improved chemical durability but at the cost of higher processing temperatures [1]. Low processing temperatures are critical in applications such as sealing and the passivation of electronic devices [5]. Thus, it is logical that boron oxide is added as a fluxing agent, as is the case with fluxes where they significantly reduce the melting temperature of the flux by replacing CaF₂ and improving the desulphurization capacity of the flux, whereby the sulfur content of the metal can be reduced to thousandths of a percent [6,7].

Wettability is the ability of a liquid to spread on a solid substrate and is measured by the angle of wetting between the tangent drawn at the triple point and the substrate. There are two types of wetting: non-reactive and reactive. The former does not involve a reaction or absorption between the spreading liquid and the substrate, while the latter does. It should be noted that Young’s equation provides the equilibrium contact angle, and its formulation is valid under the assumption of spreading a non-reactive liquid on an inert, smooth, homogeneous solid [8,9,10]. Platinum and graphite are commonly used as measuring systems [11,12,13,14,15]; however, substrates manufactured from these materials often interact with the sample. Carbon tends to reduce oxides. Platinum can form alloys with metals such as iron, lead, zinc, tin, and antimony, while the melting point of these alloys is lower than that of platinum. Under reducing conditions, non-metallic elements can also react with platinum, particularly arsenic, phosphorus, boron, bismuth, silicon, and sulfur.

A reliable and reproducible contact angle measurement is essential for assessing the wetting behavior of the system since understanding high-temperature wetting behavior is essential for improving industrial processes involving a liquid phase and the quality of the final product. Over the years, various methods have been developed to evaluate the wettability of solids by liquids [16,17]. Among these techniques is the viral sessile drop method. Wetting is influenced by many factors that relate to both the properties of the liquid and the properties of the substrate and the measurement conditions. A common cause of contact angle hysteresis is the physical properties of the substrate surface, such as its roughness, presence of cracks, microcavities, coatings, etc., with large values of roughness resulting in significant scatter in the measured angles [18]. The type of gaseous atmosphere also plays an equally important role, with oxidizing conditions capable of significantly distorting the results of metal contact angle measurements [19]. What is also of note are the physical and chemical interactions of the molten droplet with the substrate, such as the dissolution of the substrate forming craters under the droplet, the formation of interfacial reaction products requiring structural characterization of the phase interface after the experiment, and the formation of carbon whiskers on the droplet surface when tested under a high vacuum [8,9,19].

Over the last few decades, many authors and research groups have studied the wettability of refractory materials, such as platinum and graphite, by oxide systems. Using the sessile-drop method, the wettability of platinum and other metal substrates and graphite by slag was measured. The slag wets Pt relatively well, while increasing the temperature promotes wettability. Parry and Ostrovski studied the wetting of solid metals (iron, nickel, and platinum) with molten CaO–SiO₂–Al₂O₃ [13] and MnO–SiO₂ [20] slag in a graphite furnace with a reducing atmosphere containing trace oxygen. In the case of the first oxide system, silica reduction, silicon dissolution (which did not occur for the platinum substrate), and oxygen adsorption substantially changed the phase interface, making the conditions dynamic. The platinum substrate was wetted the most, and the wetting angles decreased with increasing temperatures. Oxygen desorption led to the development of gas bubbles, as observed in experiments with platinum substrates. In contrast, the wetting of the platinum substrate was the lowest for the second oxide system. Modifying the metal–oxide interface due to oxide reduction and oxygen adsorption created conditions for dynamic wetting behavior and affected the interface properties. The reaction at the phase interface aided wetting, and the dissolution of the reduced manganese occurred for all three substrates. In paper [14], the difference in the thermal expansion coefficient of platinum and slag caused a facile separation of these systems during cooling. On the other hand, the slag did not wet the graphite. However, the increase in this wettability at high temperatures was due to a chemical reaction that also caused the drop to foam [14]. Obviously, graphite as a substrate can only be used in a reducing atmosphere and may react with the oxide system [14,21]. The wetting of platinum by the MnO–SiO₂ and CaO–AI₂O₃–SiO₂ oxide system was analyzed in [22], where oxide reduction and the dissolution of Mn into the substrate occurred, and the reaction at the phase interface promoted wetting, resulting in dynamic wetting behavior. In the case of the graphite substrate, the absence or minimal wetting by molten oxide mixtures is due to the weak van der Waals forces underlying interfacial bonding [23]. However, for some oxide systems, the contact angle decreases with time to values below 90° due to interfacial reaction [24,25,26]. This has been addressed in several works on the wettability of carbonaceous materials by CaO–SiO₂–Al₂O₃–MgO–FeO_x slags in the 200 K range from 1673 K [24,25,26,27,28], and it has been found that wettability can be improved by increasing the iron oxide content and temperature. In the wettability test measurements of the CaO–SiO₂–Al₂O₃–MgO system and carbon substrate [29], it was found that the reduction of SiO₂ in the slag and the formation of SiC on the substrate surface are the dominant factors improving the wettability. Furthermore, the reduction of MgO proceeded preferentially and inhibited the spreading of molten slag. The wettability of graphite by the CaO–SiO₂–Al₂O₃–FeO–MgO oxide system was investigated with a focus on interfacial phenomena using the sessile drop method, and it was found that the wettability was mainly influenced by the initial magnesium and iron oxide content, and that iron reduction occurred due to the penetration of the oxide system into the graphite [30]. Duchesne studied the temperature dependence of wetting angles, density, and surface tension of slag on graphite, molybdenum, and alumina substrates. Although alumina and molybdenum provided lower contact angles, they interacted less with the slag and were stable under oxidizing conditions [31]. Using the dispensed drop method, Liu et al. investigated the wettability of graphite by TiO₂–CaO–SiO₂–Al₂O₃–MgO slag in an argon atmosphere and found that the TiO₂ content significantly aggravated the wettability in contrast to the CaO/SiO₂ ratio. Interfacial reaction and reduction of the corresponding metal oxides also occurred during the high-temperature test [32].

The aim of this paper was to describe the interaction at the interface between the CaO–MgO–SiO₂–Al₂O₃–B₂O₃ oxide system and graphite and platinum substrates using a wettability test. The phase interface after high-temperature tests was characterized by SEM and EDX analyses. The phase composition of the oxide system was analyzed by XRD. Attention has been paid to the effect of B₂O₃ on wettability due to the frequent use of B₂O₃ as a component that reduces the liquidus temperature. Our literature survey revealed that no work has been published to date on wetting platinum and graphite substrates with a molten oxide system similar in composition to E-glasses and with a graded boron oxide content. This provides this work with practical significance because a material with low interaction at high temperatures can be used as a measuring system, e.g., crucible, mold, or substrate, to measure critical thermophysical properties such as surface tension, density, and viscosity of this oxide system. Furthermore, the description of the phase behavior aids in understanding the interfacial processes for the widely used graphite substrate and the minimally reactive platinum substrate. This work can also contribute to optimizing ironmaking processes when contact between molten slag and coke occurs, and it can also increase the understanding of glass container manufacturing processes when glass is stuck to metal molds.

2. Materials and Methods

2.1. Preparation of the Samples

The oxide system based on CaO–MgO–SiO₂–Al₂O₃–B₂O₃ glass with varying boron oxide content between 0–30 wt% (samples 1–4), including two types of substrates (graphite plate with dimensions 30 × 30 × 5 mm and platinum plate with diameter 35 mm and thickness 1 mm), was subjected to a high-temperature wettability test. Oxide systems were prepared from pure chemicals in powder form of analytical purity with a minimum content of 96.5% of the substance, the total weight being 10 g. Calcium oxide, silicon dioxide, and aluminium oxide were purchased from Lach:ner (Lach-Ner, s.r.o., Neratovice, Czech Republic), boron oxide was purchased from Alfa Aesar (Alfa Aesar GmbH, Karlsruhe, Germany), and magnesium oxide was purchased from Mach chemikálie (Mach chemikálie, s.r.o., Ostrava–Hrušov, Czech Republic). Based on the predetermined composition in wt%, while keeping the basicity constant at 0.4, the weights of the individual samples were determined (

Table 1. Chemical composition of oxide samples in wt%.

Sample	B2O3	SiO2	CaO	MgO	Al2O3
1	0.0	64.6	15.9	10.0	9.5
2	5.0	61.1	14.4	10.0	9.5
3	15.0	53.9	11.6	10.0	9.5
4	30.0	43.2	7.3	10.0	9.5

). Pure oxides were mixed after weighing, ground in a Retsch PM 100 laboratory mill (Retsch GmbH, Haan, Germany), and remixed to ensure homogenization of the system. Approximately 0.7 g of the sample was then weighed and pressed into tablets with a diameter of 14 mm (Figure 1). Immediately before the experiment, the substrate surface was cleared of impurities by acetone.

2.2. Determination of Liquidus Temperatures

The liquidus temperature of the studied oxide systems (Samples 1–4) was investigated by an Anton Paar FRS 1600 high-temperature rheometer (Anton Paar GmbH, Graz, Austria) monitoring the position of the spindle acting vertically on the sample surface as a function of the temperature during heating. A detailed description of the device can be found in the article [33]. During heating, the value of the normal force imposed by the spindle on the sample surface was also monitored. The liquidus temperature was determined when the spindle position reached the measuring position and the normal force dropped to 0 N. The initial normal force applied by the spindle to the sample surface was 3 N. Dilatation of the samples also occurred during heating. However, in the case of the samples studied, the volume dilatation was negligible. Heating to 1550 °C was conducted in two steps (from room temperature to 900 °C at a heating rate of 30 °C·min⁻¹ and then to 1550 °C at a rate of 3 °C·min⁻¹). The determined liquidus temperatures can be seen in

Table 2. Liquidus temperatures for Samples 1–4.

Sample	Liquidus Temperature (°C)
1	1510
2	1482
3	1456
4	1332

The relative measurement error is 0.5%.

.

2.3. High-Temperature Wettability Test

Experimental determination of wetting angles at the oxide melt/substrate interface was conducted using the sessile drop method in a CLASIC high-temperature observation resistance furnace (CLASIC CZ, s.r.o., Řevnice, Czech Republic). The apparatus has already been described in a previous paper [34]. The temperature range was determined from the liquidus temperature, obtained by measuring the normal force on an Anton Paar FRS 1600 viscometer (see Section 2.2), up to 1550 °C. The oxide system pressed into a tablet was placed in a furnace on a polished graphite or platinum substrate. The furnace was then hermetically sealed, evacuated to approximately 1 Pa, and flushed with high-purity Argon (6N). The heating rate was set to 5 °C/min, which is an appropriate rate to ensure sample heating due to the furnace arrangement and sample size. During temperature load, the temperature was measured with a Pt—13% Rh/Pt thermocouple located near the sample. During the high-temperature wettability test, the droplet silhouette of the sample was captured by a Canon EOS550D high-resolution camera. The wetting angles were evaluated using the ADSA (Axisymmetric Drop Shape Analysis) method based on fitting the drop profiles to a Laplacian curve using a nonlinear regression procedure [35].

2.4. SEM, EDX, FTIR, and XRD Methods

After performing the high-temperature wettability test, the interaction between the oxide system and the respective substrate was studied by SEM, EDX, and XRD analyses. The substrate surface was analyzed by a Quanta–650 field emission gun (FEG) electron microscope (Thermo Fisher Scientific, Waltham, Massachusetts, USA), equipped with an energy dispersive detector (EDS, EDAX Elect Plus) was employed. The microscope operated under these conditions: voltage 20 kV, current 8–10 nA, beam diameter 4 μm, and high vacuum. Samples were scanned without metal coating.

FTIR spectra were measured by ATR technique with a diamond crystal using Thermo Scientific Nicolet 6700 FT–IR. The parameters were set as follows: number of scans 32, spectral resolution 4 cm⁻¹, spectral range 400–4000 cm⁻¹. The spectra were adjusted using ATR and baseline correction.

The phase composition of oxide samples after interaction with graphite or platinum substrate was measured by a Bruker AXS D8 Advance X-ray diffractometer equipped with a LynxEye position-sensitive silicon strip detector under the following conditions: CuKα/Ni-filtered radiation, voltage 40 kV, current 40 mA, step mode with a step of 0.014° 2θ, total time 25 s per step, and angular extent 5–80° 2θ. The Bruker AXS Diffrac and Bruker EVA software (version 4.2.) processed the data. The PDF–2 database (International Centre for Diffraction Data) was used for phase identification.

3. Results and Discussion

3.1. Determination of Wetting Angles

The effect of boron oxide content ranging from 0 to 30 wt% on the wettability of platinum and graphite substrates was investigated by a high-temperature wettability test from the liquidus temperature of the system to 1550 °C. Figure 2 shows the dependence of the average wetting angle on temperature, with the oxide system/graphite substrate couple showing a non-monotonic decrease in contact angle with temperature for all samples. The non-wetting behavior where the contact wetting angle was greater than 90 degrees occurred up to 1530 °C for Samples 1 (0 wt% B₂O₃) and 2 (5 wt% B₂O₃), 1524 °C for Sample 3, and 1490 °C (15 wt% B₂O₃) for Sample 4 (30 wt% B₂O₃). From these temperatures onwards, there was a significant decrease in contact angle and a transition from non-wetting to wetting behavior. The most significant decrease in contact angles was observed for Sample 4 with the maximum addition of boron oxide (30 wt%), with contact angle values reaching only 80 degrees in the final measurement phase (from 1490 °C to 1550 °C). On the other hand, in the case of wetting on the platinum substrate, the wetting angles were less than 90 degrees over the entire measured temperature interval, and the wetting angles varied with temperature, when they decreased, and with boron oxide content. The higher the boron oxide content, the smaller the contact angle.

Since, for the platinum substrate, the wetting angle decreased linearly with temperature, it can be approximated by a linear function (Equation (1)), which was adapted from [36].

$$\theta ={\theta }_{ref}+\frac{\partial \theta }{\partial T}\left(T-T_{ref}\right)$$

(1)

where ${\theta }_{ref}$ is the wetting angle at the reference temperature, i.e., the liquidus temperature ($T_{ref}$) obtained from Figure 2B, and $\frac{\partial \theta }{\partial T}$ is the corresponding temperature coefficient. The results (parameters ${\theta }_{ref}$ and $\frac{\partial \theta }{\partial T}$) are listed in

Table 3. Results of the linear fit of the average wetting angle dependence on temperature.

Sample	${\mathit{T}}_{\mathit{r}\mathit{e}\mathit{f}}$ (°C)	${\mathit{\theta }}_{\mathit{r}\mathit{e}\mathit{f}}$ (deg.)	$\frac{\partial \mathit{\theta }}{\partial \mathit{T}}$ (deg·°C−1)	$\mathsf{\Delta }\mathit{T}$ (°C)
1	1510	72	−11.60 × 10−3	1510–1550
2	1482	70	−29.33 × 10−3	1482–1550
3	1456	65	−27.56 × 10−3	1456–1550
4	1332	65	−42.72 × 10−3	1332–1550

for all oxide systems.

The table shows that as the boron oxide content of the samples increased, the slopes of the temperature dependence decreased, and the decrease in wetting angle with temperature was more abrupt.

The images of selected oxide systems (Sample 1—0 wt% B₂O₃, Sample 4—30 wt% B₂O₃) on graphite and platinum substrates can be observed in Figure 3A–L. The change in droplet shape and wetting angles can be observed at three different temperatures, namely for Sample 1 at the liquidus temperature of 1510 °C (Figure 3A—graphite, Figure 3B—platinum), at 1530 °C (Figure 3C—graphite, Figure 3D—platinum), and at the maximum temperature of 1550 °C (Figure 3E—graphite, Figure 3F—platinum). For Sample 4, this occurs at the liquidus temperature of 1332 °C (Figure 3G—graphite, Figure 3H—platinum), 1390 °C (Figure 3I—graphite, Figure 3J—platinum), and at the maximum temperature of 1550 °C (Figure 3K—graphite, Figure 3L—platinum).

All solidified droplets for both graphite and platinum substrates did not peel off the surface, indicating good adhesion of these systems during the cooling process to ambient temperature. The adhesion was more intense in the case of interaction with the graphite substrate, and the cause could be the reactive wetting and the formation of chemical products at the phase interface. On the other hand, the lower adhesion on the platinum substrate supports the assumption of non-reactive wetting, which can also be inferred from the lower scatter of wetting angles compared to the graphite substrate [37]. For the graphite substrate, the variation of the contact angle is probably due to the gas evolution in the reaction process. The reaction proceeds slowly at lower temperatures, and the gas has time to escape. However, at high temperatures, the gas evolution rate exceeds the escape rate, and gas can accumulate in the droplet volume, causing shape changes and variations in wetting angles (Figure 3C,I). Moreover, platinum is a metal not expected to interact chemically with an oxide in a neutral atmosphere [38]. Due to the assumption of non-reactive wetting on the platinum substrate, it can be argued that the spreading rate was controlled by the viscous flow of the liquid [39]. Reactive and non-reactive wetting is more thoroughly discussed in the following section.

3.2. Oxide System/Graphite (Platinum) Substrate Interaction

All samples studied, including oxide system/graphite (platinum) substrates, were subjected to SEM, EDX, and XRD analyses after high-temperature wettability tests. Since, as described in Section 3.1, there was a significant interaction between the respective oxide system and graphite substrate, the cross-sections of a particular couple were investigated through SEM and EDX analyses due to the non-separability of the two interacting systems. In Figure 4, SEM images of the cross-sections of two selected couples can be observed (Sample 1—0 wt% B₂O₃/graphite substrate, Sample 4—30 wt% B₂O₃/graphite substrate). From the images, the reciprocal interaction can be observed where the oxide system penetrates the surface layer of the graphite substrate. As the boron oxide content increases, a more pronounced interaction (penetration of the oxide system particles to a greater depth) is observed, probably due to the longer contact time between the liquid phase of the oxide system and the solid surface of the substrate. The liquidus temperatures of the samples decrease with the addition of boron oxide (see Section 2.2).

Silicon, magnesium, and aluminum oxides, as well as possibly reduced metals in the surface layer of the graphite substrate, were determined by EDX analysis to a depth of about 770 µm for Sample 4 with 30 wt% B₂O₃. The results of the EDX analysis at the marked points in Figure 4 and Figure 5 are shown in

Table 4. Results of semi-quantitative EDX microanalyses of the interaction between the oxide system and graphite and platinum substrates.

Point	Caption	C	O	Mg	Al	Si	Ca
				(wt%)
1	Oxides	31.9	40.3	3.5	3.2	19.0	2.1
2	Oxides	36.6	35.0	2.5	2.6	21.6	1.7
3	Silicates, SiO2, SiC, reduced metals	51.3	28.2	3.8	3.0	11.4	2.3
4	Silicates, SiO2, SiC, reduced metals	48.2	24.2	3.5	3.0	18.8	2.3
5	Silicates, SiO2, SiC, reduced metals	52.2	16.7	2.3	2.0	25.3	1.5
6	SiO2, SiC	72.6	4.5	―	―	22.9	―
7	SiO2, SiC	66.7	4.2	―	―	29.1	―
8	Oxides	8.7	39.2	7.4	7.8	27.3	9.6
9	Silicates/SiO2/Al2O3, reduced metals	11.4	44.4	5.7	5.3	28.6	4.6
10	Silicates/SiO2/Al2O3, reduced metals	25.8	32.9	10.3	8.4	15.6	7.0
11	SiO2, SiC, Si	64.1	2.7	―	―	33.2	―
12	SiO2, SiC, Si	80.1	4.2	―	―	15.7	―
13	Silicates/SiO2/SiC	41.5	19.6	1.5	5.2	27.7	4.5
14	Al2O3, oxides, SiO2, SiC	28.6	32.0	9.1	19.7	9.7	0.9
15	Al2O3 > silicates, SiO2, SiC	44.5	24.3	―	19.4	6.4	5.4
16	Al2O3 > silicates, SiO2, SiC	39.8	23.7	―	22.0	8.5	6.0
17	SiO2, SiC	42.5	5.6	―	―	51.9	―
18	SiO2, SiC > Al2O3	59.4	4.7	―	0.7	35.2	―

In the EDX point analysis, the influence of the surroundings must be considered for small particles and thin layers.

.

The surfaces of the oxide system/graphite substrate couples were also investigated by SEM/EDX analyses (Figure 5). Figure 5A shows Sample 1 (0 wt% B₂O₃) with graphite substrate. In Figure 5B, the sample with a maximum addition of B₂O₃ (30 wt%), including graphite substrate, can be observed. A comparison of these figures confirms the increased adhesion followed by significant interaction as the B₂O₃ content of the samples increases. Silica, alumina, silicon carbide, silicates, and, most probably, reduced metals (even in a negligible amount) were identified on the graphite substrate’s surface in the droplet’s proximity (Figure 5C–F and Table 4). In the proximity of the droplet, the surface was coated due to the spreading of molten oxide. However, with increasing distance, the gas phase was sputtered and condensed, and it reacted with the carbon substrate. Interestingly, in the case of Sample 1, these phases were uniformly distributed and consisted mainly of silica and silicon carbide. However, in the case of Sample 4, they formed zones on the surface with a predominance of a particular phase. Almost all phases were represented in the proximity of the droplet, but with increasing distance, there was a prevalence of alumina, which was subsequently replaced by the onset of silica and silicon carbide.

The reduction of silica by carbon in powder mixtures occurs considerably under vacuum or in a flow-through argon atmosphere at temperatures close to 1400 °C [25]. The reaction consists of two reaction steps (Equations (2) and (3)):

$$\mathrm{S}\mathrm{i}{\mathrm{O}}_2+\mathrm{C}=\mathrm{S}\mathrm{i}\mathrm{O}+\mathrm{C}\mathrm{O}$$

(2)

$$\mathrm{S}\mathrm{i}\mathrm{O}+2\mathrm{C}=\mathrm{S}\mathrm{i}\mathrm{C}+\mathrm{C}\mathrm{O}$$

(3)

The overall reaction is then,

$$\mathrm{S}\mathrm{i}{\mathrm{O}}_2+3\mathrm{C}=\mathrm{S}\mathrm{i}\mathrm{C}+2\mathrm{C}\mathrm{O}$$

(4)

SiO is a component of the emerging gas phase and can form silicon carbide after condensation on the surface of the graphite substrate. We assume that SiO evaporated and was not absorbed by the oxide system because this system was acidic, with a CaO-to-SiO₂ ratio of about 0.2. Other gas phase components may have been carbon dioxide and carbon monoxide, formed by reducing the corresponding alkaline earth metal oxides with carbon to form carbides and reduced metals, and the formation of metallic silicon cannot be ruled out, either. However, EDX analysis on a graphite substrate cannot distinguish carbides from pure metals. We do not assume that alumina reduction would occur as this requires higher temperatures than those achieved in the experiment. In addition, in the case of boroaluminosilicate glasses, the formation of alkali earth metal silicates also occurs at elevated temperatures.

The adhesion was not as prominent in the case of the oxide system/platinum substrate couple. Since these systems, which were in contact during the high-temperature wettability test, were easily separated after the test, no interface cross-sections were performed, and only the platinum surfaces under and in the close surroundings of the droplet were observed by SEM/EDX analysis. The results of these analyses confirmed non-reactive wetting for all samples. The platinum surface under the droplet, and in its proximity, was not affected by the oxide system. Only recrystallization of platinum and accentuation of grain boundaries occurred (Figure 6B,C).

Furthermore, the droplets containing 0% B₂O₃ and 30% B₂O₃ were subjected to FTIR analysis after a high-temperature wettability test on graphite (Figure 7A) and platinum substrate (Figure 7B). The wavenumber regions for the structural units from the relevant references are listed in

Table 5. FTIR bands of different structural units of Samples 1, and 4 droplets tested on graphite and platinum substrates.

0 wt% B2O3 on Graphite	30 wt% B2O3 on Graphite	0 wt% B2O3 on Pt	30 wt% B2O3 on Pt	IR Assignment	Ref.
Wavenumber (cm−1)
―	1370	―	1371	B–O stretching vibration of varied borate groups in [BO3] units	[40]
―	1044	―	1057	[SiO4]-tetrahedral stretching and [BO4]-tetrahedral stretching	[40]
928	914	938	919	Si–O bonds occurring in Si–O–Si (νas S–O–Si) and Si–O–Al bridges (νas S–O–Al)	[41]
799/779	796	795/780	797/781	The motions of Si atoms against tetrahedral oxygen cage	[42]
―	672	―	674	B–O–B bending vibration	[40]

.

FTIR spectra correspond to the amorphous/glassy phase, which indicates the broad bands. Graphite and platinum are not IR active, so only the glass oxide system can be observed. In the sample with the addition of boron, a change in structure is clearly visible. The samples without B₂O₃ are very similar; only a slight shift to the higher wavenumber is observed in the band at 928/938 cm⁻¹ in the case of platinum substrate. This band originates from the asymmetrical stretching vibration of Si–O bonds occurring in Si–O–Si (ν_as S–O–Si) and Si–O–Al bridges (ν_as S–O–Al) [41]. This band can also be found in spectra with B₂O₃, where its shift could be associated with the influence of boron on the initial structure. The spectrum without B₂O₃ also shows a doublet ~800 cm⁻¹, which is related to the motions of Si atoms against the tetrahedral oxygen cage [42], and its small intensity may indicate the decomposition of the quartz [41]. This doublet is also visible in the spectra with 30 wt% of B₂O₃, and it is more pronounced in the spectrum with Pt substrate. Other bands visible in spectra with B₂O₃ are associated with the presence of boron in the structure. Vibrations of B–O in the [BO₃] unit are at 1370/1371 cm⁻¹ for the sample on graphite/Pt. A slight shift is observed in the case of the [BO₄] unit vibration at 1044/1057 cm⁻¹ for the sample on graphite/Pt. A broader shape is found for the sample on the graphite, and the band’s intensity at ~917 cm⁻¹ is higher when compared to the intensity of the band of ~1050 cm⁻¹, which could be explained by the more glassy/amorphous phase in the sample.

The results of the XRD analyses were similar for all samples analyzed. In Figure 8, diffraction patterns of selected samples of oxide systems (sample 1— 0 wt% B₂O₃ and sample 4—30 wt% B₂O₃) that were in contact with graphite or platinum substrate can be observed. The samples exhibit a strongly amorphous character with a small fraction of the crystalline phase, which, in the case of samples in contact with graphite, was not greater than 5%. However, in the case of contact with platinum, the fraction of the crystalline phase was determined to be, at most, 3%. In addition, the presence of the crystalline phase was also confirmed by FTIR analysis. In both cases, the crystalline phase was identified as quartz, corresponding to the vibration of Si–O bonds on the recorded FTIR spectra. The presence of carbon was also detected in the samples of the oxide system in contact with graphite. The maximum B₂O₃ content of Sample 4 was almost twice that of the sample without B₂O₃, which can also be explained by the firm adhesion of Sample 4 to the graphite substrate and, thus, the contamination of the oxide system by graphite during the separation of the two systems.

4. Conclusions

The results of the particular study dealing with the interaction of the CaO–MgO–SiO₂–Al₂O₃–B₂O₃ glass oxide system with graded boron oxide content (0–30 wt%) and graphite or platinum substrate during the high-temperature wettability test up to 1550 °C can be summarized as follows:

• The contact angle values decreased with increasing temperature for the oxide system/graphite substrate system, and the effect of the boric oxide content on the contact angle values was negligible. The wetting of the graphite substrate by the investigated oxide system can be considered reactive. The strong interaction of the two contacting phases was accompanied by the formation of reaction products at the phase interface, and the interaction intensity increased with increasing boron oxide content.
• In the case of the oxide system/platinum substrate system, it can be stated that the contact angles decreased with increasing temperature, as well as with increasing boron oxide content. The wettability of these systems can be considered non-reactive based on the supporting SEM/EDX analyses.
• FTIR analysis showed that oxide system samples, after high-temperature wettability tests on both graphite and platinum substrates, contain a minimum crystalline phase. It was further confirmed that adding boron oxide changed the structure of the sample.
• The results of XRD analysis confirmed the highly amorphous nature of all the investigated oxide system samples, with a small fraction of the crystalline phase identified as quartz. The oxide system samples that wetted the graphite substrate were slightly contaminated with graphite due to the high adhesion.

The results of this work complement the results of earlier studies on wetting and interfacial interaction of molten oxide systems and graphite or platinum substrate, which find their application in metallurgical, ceramic, and glass industries. Wetting data are essential for process control and optimization of industrial operations with oxide systems.