nach oben

Erschienen in:

Open Access 2024 | OriginalPaper | Buchkapitel

2. In Situ Observation of Collision Between Exogenous and Endogenous Inclusions on Steel Melts for Active Steel Filtration

verfasst von : Christina Schröder, Xingwen Wei, Undine Fischer, Gert Schmidt, Olena Volkova, Christos G. Aneziris

Erschienen in: Multifunctional Ceramic Filter Systems for Metal Melt Filtration

Verlag: Springer International Publishing

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config

KI-gestützte Suche

Patentsuche

Aus

Abstract

Investigations at the high-temperature confocal laser scanning microscope (HT-CLSM) allow observing the interaction of non-metallic particles in terms of potential filter material with endogenous particles of molten steel in the high-temperature range. The respective particle velocities are determined from the particle movements and conclusions are drawn about the attractive forces of the particles. The interactions of exogenous Al₂O₃ particles, MgO and MgAl₂O₄ spinel particles, and CA₆ calcium aluminate particles with endogenous constituents of molten steel X15CrNiSi25-20 are analyzed in the present work. Accompanying experiments were performed in a heating microscope on the interaction between steel and MgO and steel and CA₆. Scanning electron microscopy SEM/EDX/EBSD studies reveal not only the interactions of the non-metallic inclusions with each other, but also reactions of the molten steel with the exogenous particles that affect the agglomeration behavior. While exogenous and endogenous Al₂O₃ particles exhibit high attractive forces and almost no react with the molten steel, a liquid reaction layer forms around the magnesia particles, which leads to a reduction of the attractive forces. After dissolution of the reaction layer, the attractive forces increase. Spinel particles are surrounded by a strong meniscus in the observed steel melt. Endogenous particles moving toward the spinel do not adhere to the particle. Reactions were also observed when CA₆ particles came into contact with molten steel. In this process, the calcium aluminate is depleted of calcium. Only loose connections of the exogenous Ca-depleted CA₆ with endogenous Al₂O₃ particles have been detected.

2.1 Introduction

Non-metallic oxide inclusions are usually undesirable in steel because, mainly by forming clusters, they negatively affect the mechanical properties, especially the strength of the steel, induce cracking under stress and promote their growth. They can be formed both by reoxidation and by the decrease in oxygen solubility during cooling. Carbon-bonded refractory material have a high thermal shock and creep resistance and are used in both the tundish and the casting mold to reduce inclusions from molten steel.

Janiszewski and Kudlinski [1] provided the theoretical basis for the application of active filters. Oxide coatings with the same chemical composition as the inclusions to be filtered in the melt serve as active filter material and open up the possibility of eliminating semiliquid or liquid inclusions as well. The capillary forces between the particles play a decisive role in the attraction of the inclusions. The principle of attraction between colloidal particles has been extensively studied by Kralchevsky and Nagayama [2, 3] and Danov et al. [4]. In addition to the flotation force caused by their own weight, the particles are affected by the Archimedes force resulting from gravity and capillary forces due to wettability. Thereby, the flotation and immersion forces can be attractive or repulsive. The meniscus slope angles Ψ₁ and Ψ₂ of the particles in the melt are the decisive quantities for a movement of the particles towards or away from each other (Fig. 2.1). The capillary force is attractive when the product (sin Ψ1·sin Ψ2) is greater than zero and repulsive when the product (sin Ψ1·sin Ψ2) is less than zero. Mainly three factors influence the interaction between the particles: the particle density, the contact angle and the surface tension [5]. The higher the density the stronger gets the capillary attraction. But in reality the particle form clusters, so that an apparent density of particles have to be considered. The capillary force also increases with increasing contact angle of the oxide particles. The influence is big with increasing angles from 100 to ca. 130° but smaller at increasing angle from 130 to 170°. With increasing surface tension of the steel melt, the capillary forces of the particle decrease slightly, also with the presence of sulfur as surfactant element.

To study the clogging tendency against endogenous and exogenous non-metallic inclusions, carbon-bonded nozzles with and without alumina active coatings were tested in a steel casting simulator by Aneziris et al. [6]. ‘Denser’ as well ‘coral-like’ clogging areas were found in both cases, whereas the coated nozzles show a stronger clogging. This opens the possible use of the material for filters. The impact of the wetting properties and possible reactions between filter material and melt on the filtration efficiency of solid inclusions are not clear yet. High-temperature investigations at the HT-CLSM provide the opportunity to observe the interaction of endogenous particles with exogenous particles in situ in the melt. Yin et al. [7] were the first to observe in situ the collision and agglomeration behaviour of non-metallic particles in the steel melt of low carbon Al-killed steels. For the case, that one particle, guest particle, move to the other, host particle, they calculated the attractive force F as the product of the mass m₁ of the guest particle and its acceleration a_i at each time i.

$$F_{i} = m_{1} \cdot a_{i}$$

(2.1)

In the case that both guest and host particle are moved simultaneously, a revision factor was introduced with

$$m_{2} /(m_{1} + m_{2} )$$

(2.2)

where m₂ is the mass of the larger host particle. The mass of the particles is estimated by its size and the material density.

Yin et al. [7] found that the strength of the capillary attraction forces increases for inclusions and inclusion pairs as follows: It is zero for a liquid/liquid pair < liquid/semiliquid pair < liquid/solid pair < semiliquid/solid pair < solid/solid pair. The solid/solid pair has the strongest attraction force if only the effect of particle morphology is considered. For alumina particle smaller than 3 µm, a long-range strong attraction force over 10^–16 N was determined in a distance over 10 µm between the particles or their agglomerates. The acting distance is affected by the inclusion size and shape. Mu et al. [8] determined a wide-ranging acting distance of over 150 µm for Al₂O₃ clusters with equivalent radius greater 40 µm and attractive capillary force greater 2·10^–13 N. They show, that model calculations using equivalent radius R_k with

$${{\text{R}}}_{{\text{k}}}=\sqrt{{{\text{A}}}_{{\text{k}}}/\uppi }$$

(2.3)

where A_k is the specific area of the particle k, leads to a greater confirmation between calculated attractive force and experimental determined attractive force for inclusion clusters than using the effective radius R_k,eff with

$${\text{R}}_{{{\text{k}},{\text{eff}}}} { } = \frac{{{\text{P}}_{{\text{k}}} }}{{2{\uppi }}} = {\text{ R}}_{{\text{k}}} /\sqrt {{\text{CF}}_{{\text{k}}} }$$

(2.4)

where P_k is the perimeter of the particle k, and CF_k is the particle circularity.

For MgO inclusion pairs and 93%Al₂O₃·7%MgO inclusions pairs Kimura et al. [9] found similar attractive forces at a maximum acting distance of 21–22 µm. They are with 5·10^–18–5·10^–16 N approximately a power of ten lower as for Al₂O₃ inclusion pairs. The attractive forces of solid Al₂O₃·SiO₂ pairs is also weaker and their acting length shorter than between alumina particles [7]. In contrast to the alumina particles, their shapes were smoother and their clusters were denser.

In a continuing paper by Yin et al. [10] the collision and agglomeration of particles in the systems CaO-Al₂O₃ and CaO-Al₂O₃-SiO₂ were investigated. The attractive forces between semi-fluid particles of the CaO-Al₂O₃ system are 10^–14-10^–15 N with a range of about 42 µm. If one particle of the pair has an irregular shape (solid particle), the attractive force increases to 8·10^–13–7·10^–14 N and the range rises slightly. Larger attractive forces than alumina particles exhibit were determined for solid CaO-80 wt.% Al₂O₃- and for solid (CaO·Al₂O₃)-95 wt.% SiO₂ pairs. Both particle pairs have a denser structure as the alumina clusters and so a greater difference of liquid steel surface height between inside and outside of the pair. No attraction was found between totally liquid inclusion particles. SiO₂ lowers the liquidus temperature in CaO-Al₂O₃-SiO₂ inclusions. The attractive forces between solid CaO-Al₂O₃-SiO₂ particles decrease to <10^–15 N. No capillary attractive forces were observed between liquid CaO-Al₂O₃-SiO₂ inclusions. A summary on the in situ agglomeration observations of non-metallic inclusions at the interface of liquid steel/argon gas and liquid steel/slag is given by Mu et al. [11].

Uemura et al. [12] investigated the filtering mechanism of non-metallic inclusions in steel by ceramic loop filters. According to their research, the filter efficiency depends on the initial oxygen content in addition to the fineness of the filters. Equal filter efficiencies were determined for the CaO·6Al₂O₃, short CA₆, and Al₂O₃ filter materials. Storti et al. [13] studied carbon-bonded filters functionalized with coatings based on calcium aluminates CA₂ and CA₆ in contact with molten steel at 1650 °C. The coating aims to react with the molten steel and generate a liquid layer, that act as an active collector for endogenous inclusions. In addition, the boiling Ca and forming CO move the melt and promote the clustering of fine alumina so that it can rise to the surface of the melt pool. They describe the formation of a thin secondary layer on the with CA₆-coated carbon-bonded alumina filter material after the immersion test. On the other hand, they show in thermodynamic considerations that the Ca activity is low with 10^–5 to 10^–4. However, the analyses of the solid steel showed excellent filtration performances of both compositions.

The aim of the presented investigations within subproject A01 of the CRC 920 was to observe and analyze the interactions of endogenous particles with exogenous non-metallic alumina, magnesia, MgAl₂O₄ spinel und CA₆ particles as a possible filter material. For this purpose, various steels were tested for their suitability for high-temperature investigations in the HT-CLSM with respect to a homogeneous melt pool over the experimental period.

2.2 Experimental Details

2.2.1 Preparation of the Steel Samples

For the investigations at the HT-CLSM, steel platelets with a thickness of approx. 1 mm were cut from hot-rolled round bars with a 10 mm diameter. Previous removal of the rolling surface of the round bars by turning reduces the introduction of oxides during the melting test. The sample platelets were metallographically ground and polished so that they have a thickness of approx. 600 µm in the test. Immediately before being placed in the furnace, the steel platelets were polished with 1 µm grit diamond emulsion to minimize the oxide layer on the specimen surface.

For the experiments in the heating microscope, plane-parallel steel cylinders with a diameter of 6 mm and a height of 10 mm were manufactured. In order to reduce the oxide layer on the specimens, the steel specimens were etched in dilute hydrochloric acid directly before the experiment in the heating microscope.

2.2.2 Exogenous Non-metallic Test Material

To investigate the interaction of exogenous non-metallic materials with the melt and its endogenous non-metallic precipitates, individual 50–80 µm particles of the material to be investigated were preferably positioned in the centre of the steel lamina before the start of the melting test. The characteristic properties of the non-metallic test material are presented in Table 2.1.

Table 2.1

Specifics of the non-metallic exogenous materials

Ceramic	Graining	Specific properties	Fabricator
Al₂O₃	Tabular alumina; T60/64	Particle size 50–70 μm, Specific surface 1.13 m²/g	Almatis, Inc., Leetsdale, PA
MgO	Fused magnesia	Particle size 63–80 μm; include 0.35 wt.% CaO	Qinghua refractories, Dashigao, China
MgAl₂O₄	Magnesium aluminate spinel; AR 78	Particle size 60–80 µm >74.0 wt.% Al₂O₃, ca. 22.5 wt.% MgO, ca. 0.24 wt.% CaO, <0.15 wt.% SiO₂, <0.32 wt.% Na₂O, <0.25 wt.% Fe₂O₃; bulk spec. density 3.3 g cm⁻³, apparent porosity <2.6 vol.%, water absorption <0.8 vol.%	Almatis GmbH, Germany
CaO·6Al₂O_3; (CA₆)	Bonite	Particle size 0–0.020 mm, d₅₀ = 6.0 µm >99 wt.% CA₆	Kerneos, France
CaO·6Al₂O_3; (CA₆)	Bonite	Particle size 500–1000 µm ca. 91 wt.% Al₂O₃, ca. 7.6 wt.% CaO, ca. 0.09 wt.% Fe₂O₃, ca. 0.9 wt.% SiO₂; bulk spec. density 3.0 g cm⁻³, apparent porosity 9.8 vol.%, water absorption 3.4 vol.%	Almatis GmbH, Germany

MgO and CA₆ tablets were prepared for wetting behaviour investigation in the heating microscope. For MgO tablets electro fused magnesia with 97.8 wt.% MgO and 1 wt.% CaO was mixed with 2 wt.% Mg-ligninsulfonate Duriment BV P as temporary pressing additive and 2 wt.% deionized water. The mass was pressed in cylindric samples with 20 mm diameter and 2.5 mm height and a pressure of 25 MPa. After drying at 120 °C for 4 h the tablets were sintered at 1600 °C/1 h under atmosphere. The CA₆-tablets have a diameter of 45 mm and are 7 mm high. They were pressed with the binder dextrin at 120 MPa, dried at 120 °C and sintered at 1600 °C for 5 h. The surface roughness after polishing was 1.57 µm with an open porosity of 17%.

For calculations of the phase fractions during heating and melting of the steel alloys the software ThermoCalc with the thermodynamic database of steels/Fe-alloys Version 8.1 were used. The software FactSage 7.2 with the databases FTmisc 6.3, FToxid 6.3, and FactPS 6.3 were utilised to calculate possible interactions between the refractory material and the oxygen from the passive layer of steel or others oxygen sources.

2.2.3 Experimental Equipment and Experimental Conditions

The HT-CLSM 1LM21H/SVF17SP (Lasertec, Japan) makes it possible to digitally record interactions of molten steel with non-metallic particles in situ in the high temperature range. The principle sketch of the furnace operation is described in an earlier article by Aneziris et al. [14]. The temperature regime of the experiments can be pre-programmed with the aid of the device's own control software. Figure 2.2 shows an example of the temperature control during the tests. 5 working steps have been programmed: In the first stage, heating is performed at 200 K/min to 600 °C to preserve the durability of the halogen lamp. This is followed by an increased heating rate of 400 K/min to avoid precipitate formation during heating up to 1300 °C with a hold time of 30 s to eliminate overheating due to the high heating rate. In steps 3 and 4 the heating rates are reduced first to 120 K/min up to a temperature of 1420 °C with a holding time of 30 s and to 30 K/min up to a temperature of 1515 °C and a holding time of 6 min.

Step 4 allows observation of partial melting and the interaction between the melt and the non-metallic particle. The heating temperature and the holding time can be manually increased or extended during the test. In program step 5, the sample cools down very quickly by switching off the heating source due to the low molten sample volume. A HeNe laser scans the sample surface for in situ observation at a frame rate of up to 30 frames per second. Particles down to a size of 0.5 µm can be detected in the image. To analyse the particle trajectories of the HT-CLSM videos, image sequences were selected for automatic motion analysis using WINanalyse software (Mikromak Services, Germany).

The assembly of a heating microscope is shown schematically by Wei et al. [15]. The heating rates in this work were 20 K/min up to 400 °C and 40 K/min above 400 °C up to the test temperature of 1450 °C. Due to its surface tension, the liquid steel forms a drop (sessile drop method). To determine the surface tension of the steel droplet and its contact angle with the ceramic surface, a calculation software is used based on the findings of Rotenberg et al. [16] and described in the work of Chebykin et al. [17].

After the HT-CLSM and heating microscope investigations the micro-structural phase evaluations of the steel discs were carried out at room temperature (RT) at a scanning electron microscopy (SEM) Ultra 55 (Zeiss, Germany) using integrated energy dispersive X-ray spectrometer (EDX) from Edax/Ametek and electron back scattering diffractometer (EBSD). EDAX OIM software was used for data processing. The untreated sample surfaces, as well as vertically cut, grinded and polished samples were investigated.

2.3 Results

2.3.1 Melting Behaviour of Different Steels in HT-CLSM

A prerequisite for observing the interaction of non-metallic particles in the melt is the generation of a melt pool which is maintained as homogeneously as possible over a certain experimental period. Table 2.2 summarizes the chemical composition of three investigated steels. The steels have graded carbon contents.

Table 2.2

Chemical composition of the studied steels [wt.%], rest Fe

Steel	C	Si	Mn	P	S	Cr	Mo	Ni	Al	Cu	Ti
42CrMo4	0.438	0.26	0.83	0.016	0.031	1.09	0.18	0.21	0.013	0.27	−
X20Cr13	0.216	0.40	0.56	0.283	0.004	12.98	0.15	0.78	0.004	0.06	0.002
X15CrNiSi25-20	0.118	2.38	1.03	0.030	0.027	25.10	0.07	18.80	0.020	0.07	0.020

42CrMo4 is a low-alloyed steel for quenching and tempering used as a base alloy in several CRC 920 projects. According to ThermoCalc calculations Fig. 2.3a, ferritic-austenitic solidification starts at a liquidus temperature T_L of 1487 °C. The primarily formed δ-ferrite is peritectically transformed into austenite in a narrow temperature interval of approx. 5 K. Its solidus temperature T_S is 1406 °C. After solidification, the steel has a completely austenitic structure. Below 750 °C, pearlite formation from the austenitic high-temperature phase begins. Melted (1) and partially resolidified (2) areas are seen on the surface of 42CrMo4 at 1556 °C during HT-CLSM investigation in Fig. 2.3b.

Despite the temperature increase, a part of the melt solidifies again. A strong movement and/or doubling of the grain boundaries is visible, which may indicate the transformation of austenite to δ-ferrite. This observation leads to the following conclusion: Due to the low carbon solubility in δ-ferrite, carbon preferentially diffuses into the melt, reacts with the residual oxygen on the steel surface and is lost. The carbon depletion leads to an increase in the melting temperature. An indirect evidence of this assumption are black deposits on the furnace wall of the HT-CLSM. A uniform melt pool could not be formed without melting the peripheral areas of the steel sample and led to droplet formation due to the surface tension of the steel.

X15CrNiSi25-20 steel is a high-alloy stainless steel that undergoes primary ferritic-austenitic solidification (Fig. 2.4). Compared to 42CrMo4 and X20Cr13, its carbon content is lowered, reducing the tendency for carbon loss during melting. The higher proportions of the alloying elements chromium, nickel and especially silicon lead to the reduction of the liquidus and solidus temperatures to about 1378 and 1300 °C, respectively. 40 ppm oxygen was analysed in the steel.

Due to the reaction of solved aluminium in the steel with oxygen, which is present by reduction of the chromium oxide layer, endogenous particles of approx. 1 µm in size are formed on the surface of the melt. These move freely in the melt, interact with other endogenous particles and form conglomerates. Due to the stability of the melt over an investigation period of >300 s and the possibility of observing the interaction of endogenous particles with endogenous and exogenous particles, the further melting experiments were carried out with steel platelets of the alloy X15CrNiSi25-20.

2.3.2 Structural Changes in the Steel X15CrNiSi25-20 During Heating

In the initial state of the steel, (Mn,Cr,Al,Fe)-sulphides with about 36 at.% Mn, 8 at.% Cr, 3 at.% Fe and 2–7 at% Al and (Cr,Fe,Ni)-carbides are detectable. No sigma phase was observed.

During heating, the reflectivity of the steel plates decreases briefly at about 850 °C and about 1240 °C. The change in reflectance behaviour may be due to new grain formation during a phase transition. It can also be caused by the formation of non-metallic precipitates on the steel surface due to changes in the solubility of the alloying elements during a phase transition. According to ThermoCalc calculations (Fig. 2.4), phase transitions occur in both temperature ranges. Tables 2.3 and 2.4 show the calculated equilibrium phases with their volume fractions and the chemical composition of the phases. The bcc phases are rich in chromium, nickel and silicon and take up lower carbon, manganese and phosphorus contents and slightly higher sulphur and titanium contents compared to the fcc phase. (Cr,Ni)₃-phosphide, which appears at 894 °C under equilibrium calculation, is dissolved at 1130 °C in the matrix. Carbon and manganese have a lower solubility in the ferritic high temperature phase and can be enriched at the phase interfaces and the steel surface during the phase transition from fcc to bcc. The Ti and C rich fcc#2 phase takes up sulphur at 1130 °C and forms Ti₄C₂S₂ under equilibrium conditions. All phase transitions can influence the reflectance ratios of the steel surface. In order to investigate the phase formations in the steel during the melting and solidification process, steel samples without exogenous inclusions were examined in the HT-CLSM.

Table 2.3

Calculated phases at formation of bcc at 894.5 °C and the phase compositions; rest Fe

Phases at 894.5 °C	Phase fraction	C	Si	Mn	P	S	Cr	Mo	Ni	Al	Cu	Ti
	Vol.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%
Fcc#1	95.0	1E-2	2.4	2.0	1E-2	5E-9	22.5	0.15	18.0	9E-3	0.17	4E-3
Sigma	3.1	–	3.0	0.4	–	–	44.0	0.14	1.9	5E-7	–	7E-5
M23C6	1.7	5.5	–	0.6	–	–	77.0	5.3	0.9	–	–	–
M3P	0.1	–	–	–	15.9	–	40.4	–	24.2	–	1E-5	–
MnS	2E-2	–	–	63.1	–	36.9	3E-7	–	–	–	3E-8	–
Fcc#2	4E-3	18.4	5E-8	1E-3	5E-11	5E-11	3.1	0.2	7E-4	5E-11	2E-7	78.3
bcc	0.0	2E-3	2.6	1.03	9E-4	6E-9	38.9	0.2	6.3	7E-3	2E-2	7E-3

Table 2.4

Calculated phases at transition temperature fcc to bcc at 1127 °C and its compositions; rest Fe

Phases at 1127 °C	Phase fraction	C	Si	Mn	P	S	Cr	Mo	Ni	Al	Cu	Ti
	Vol.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%	wt.%
Fcc#1	99.6	8E-2	2.4	1.9	3E-2	4E-7	23.9	0.23	17.2	8E-3	0.16	4E-3
M23C6	0.3	5.5	–	0.9	–	–	72.8	1.9	1.9	–	–	–
MnS	2E-2	–	–	63.0	–	36.9	8E-6	–	–	–	6E-7	–
Ti4C2S2	4E-3	8.6	–	–	–	22.9	–	–	–	–	–	68.5
Bcc	0.0	2E-3	2.7	1.2	2E-3	6E-7	33.8	0.3	9.2	8E-3	4E-2	8E-3

Figure 2.5 shows a steel disc after the melting test. The sample has a metallic sheen without increased formation of visible oxide layers. The not yet melted peripheral area of the sample and the dendritically solidified structures from the periphery to the interior can be clearly seen. In the centre, the residual melt is fine-grained crystalline. In the border area between molten and non-molten material (Fig. 2.5c, d), a ((Mn,Cr,Fe)-S)-rich phase (1) has solidified at the grain boundaries. Fine, non-metallic aluminium oxides have also been deposited preferentially at the grain boundaries (2).

In order to investigate the cause of the radiation intensity loss at approx. 1300 °C, a steel sample was heated to 1300 °C and cooled in 10 cycles at 400 K/min. The steel did not melt during this process. The surface of the steel plate lost its metallic sheen. Drop-like precipitations were formed. In addition, a cluster of excretions was observed (Fig. 2.6). In contrast to the melting test, predominantly (Si,Cr)-rich and occasionally (Cr,Si,Mn)-rich oxidic phases were detected under the given conditions. EDX phase analyses of the spots marked in Fig. 2.6 (1–2.4) are summarised in Table 2.5.

Table 2.5

EDX analyses of the marked oxides on the surface of X15CrNiSi25-20 steel disc after heating treatment at 1300 °C (Fig. 2.6), fraction in at.%

Spot	Si	Cr	Mn	Al	Ni	Ti	O
1	14.07	28.07	12.01	0.84	0.54	0.47	42.39
2.1	27.25	12.66	4.61	–	1.51	–	49.14
2.2	23.05	19.77	8.01	0.53	0.88	0.43	44.33
2.3	18.01	15.21	1.14	0.30	8.47	0.19	31.34
2.4	30.04	7.04	2.53	0.56	1.29	0.18	54.03

On the one hand, the high-purity argon inert gas (99.999 wt.%) with a residual oxygen content of approx. 2 ppm serves as an oxygen source for oxide formation. Secondly, steel with chromium contents ≥11 wt.% forms a chromium oxide passive layer on the surface, which can serve as a further source of oxygen. According to the literature, steels with a chromium oxide coating are scale resistant up to approx. 850 °C. Aluminium oxide coated steels have a higher scaling resistance up to approx. 1000 °C [18]. The passive layers soften at higher temperatures and contract into drops due to their surface tension and reduced adhesion to the polished steel matrix. According to the Richardson-Ellingham diagram, the oxides of the metals Mn, Si, Ti, and Al have lower Gibbs free energies than chromium oxide [19]. In this context, the Gibbs free energy for the oxidation of the metals decreases in the listed order. As the above metals are dissolved in the steel, they can reduce chromium oxide.

Above about 1400 °C, chromium oxide can also be reduced by carbon to form carbon monoxide. The steel matrix absorbs the reduced chromium. In this context, it is interesting to note that a broadening of the grain boundaries was also observed in the case of X15CrNiSi25-20 immediately before the matrix melted. Since, according to the phase quantity diagram Fig. 2.4, ferrite is formed in the high-temperature range and, in addition, chromium stabilizes the bcc ferritic phase, it can be assumed that the manifestation of the grain boundaries is caused by the formation of δ-ferrite. The austenite begins to melt due to its higher carbon content, while the δ-ferrite remains in the solid state at the former grain boundaries.

2.3.3 Interaction of Exogenous Alumina Particles with Molten Steel and Its Endogenous Particles

In contrast to the investigations of Yin et al. [7], who observed particle agglomeration on the completely melted sample surface, in the present experiments the occurrence of local surface flow of the melt could be almost prevented by partial melting within still existing grain boundaries. The rapid collision and agglomeration tendency and growth of alumina clusters due to strong attractive forces over a long range on the surface of the molten steel are confirmed. Figure 2.7 shows an exogenous alumina particle of about 50–70 µm in size on a partially melted steel disc with endogenous particles or particle clusters.

Endogenous alumina particles move over a range of >50 µm to the exogenous host particle. The attractive length is limited by the grain boundaries. The red lines indicate exemplary paths of two endogenous particles. Attractive forces of 10^–15–10^–16 N were determined, which are slightly larger than those obtained by Yin et al. [7] and correspond to the forces of alumina clusters by Mu et al. [8] for particle pairs with nearly the same particle sizes ratio. The influence of the viscosity of the steel melt on the mobility of the non-metallic particles has not yet been investigated. It is to be expected that a lower viscosity with increasing temperature also leads to an increase in particle mobility.

The own investigations show that the attractive forces between the 1–6 µm sized endogenous alumina particles and the exogenous alumina particle are somewhat larger than between the endogenous particles themselves and that the attractive forces increase with increasing particle or cluster size of guest particle (Fig. 2.8) [14]. With increasing examination time at °C the grain boundaries dissolve. At 1515 °C also liquid, round inclusions are seen at the steel surface. The liquid inclusions attach to both solid and liquid particles and likewise form their own loose clusters. They contribute to a densification of the solid inclusion clusters. Titanium oxide was frequently detected. During cooling of the sample, two distinguished phases based on α-Al₂O₃ and Ti₂O₃ were formed in the neighborhood of the alumina particles. No formation of Al₂TiO₅ could be detected.

2.3.4 Interaction of Exogenous Magnesia Particles with Molten Steel and Its Endogenous Particles

A different behavior was observed for magnesia host particles [20]. Figure 2.9a shows an exogenous MgO particle on a steel disc close to the steel melting temperature. First melt has formed in the zones marked with arrows. As soon as the exogenous particle comes into contact with the molten steel, a meniscus with smooth borders forms around the magnesia (Fig. 2.9b). Inside the inner meniscus and the magnesia particle a lighter area was formed and the edges of the MgO particle are visible. In this area moving particles were observed, showing that a liquid phase was built between exogenous particle and steel melt, which increase the perimeter of the magnesia up to 57%. At this stage, no endogenous alumina remains attached to the magnesia. They flow past the exogenous particle at the liquid–liquid phase boundary between the magnesia layer and the melt and move away. Red traces mark the path of individual endogenous particles in the image. With increasing temperature and test duration, the layer becomes unstable and partially dissolves (Fig. 2.9c, d). Same endogenous inclusions that reach the surface of the free MgO adhere only weakly. Others continue to move past the particle.

The maximum real forces determined for endogenous inclusions are summarized in the diagram Fig. 2.10 for the different stages of MgO wetting. For all particle pairs, the attraction force increases with increasing particle size. At about 1420 °C, before contact of the MgO particles with the melt, the endogenous particles exhibit the same attractive forces among themselves as in the experiments with exogenous Al₂O₃ (Fig. 2.10a). After formation of the liquid layer around the exogenous MgO, due to the interaction of the particle with the melt, the attractive forces of the endogenous particles among themselves decrease by about one decimal power (Fig. 2.10b). The attractive forces of the endogenous inclusions towards the exogenous MgO are slightly lower than the forces between the endogenous inclusions (Fig. 2.10c). After partial dissolution of the liquid layer, an increase in the attractive forces was recorded for the interaction of endogenous inclusions to the exogenous particle to values slightly above those of the endogenous particles among themselves at the beginning of melt formation (Fig. 2.10d). SEM/EDX images show that the endogenous particles are enriched in magnesium [20]. Titanium oxide was detected in contact with the Al₂O₃-MgO cluster and at an MgO host particle. The cross-sectioned MgO particle has a layer of two phases on its surface, a (Si,Cr,Mg) rich phase and a magnesia phase enriched with (Ti,Cr and/or Al). The rounded contours of these phases reveal that they were in a liquid or semi-liquid state. It can be assumed that the Ti and Cr enriched phase precipitated from the supersaturated Si rich phase during cooling.

Wetting experiments in the heating microscope also confirm the reaction of the MgO with the molten steel to form a reaction layer [20]. Park et al. [21] describe the dissolution behaviour of MgO inclusions in CaO-Al₂O₃-SiO₂ slags at 1550 °C with the formation of ring-like Al₂O₃·MgO spinel structures when Al₂O₃ rich slags are used and the formation of circularly arranged Ca₂SiO₄ inclusions in (Si,Ca)-rich slags. In both cases, the MgO was surrounded by a liquid phase. In agreement with the observations of Yin et al. [10], this liquid layer prevents the agglomeration of endogenous particles with the exogenous MgO.

2.3.5 Interaction of Exogenous MgO·Al₂O₃ Spinel Particles with Molten Steel and Its Endogenous Particles

As reported in the previous section, a cluster of Al₂O₃ and MgO inclusions is formed during the reaction of exogenous MgO with the molten steel and its endogenous particles. The interaction of exogenous MgO·Al₂O₃ spinel particles with the melt was the focus of further investigations.

Figure 2.11a shows an exogenous spinel particle (1) on the steel surface with first molten steel (2) during the melting experiment in the HT-CLSM at 1439 °C. The grain boundaries have thickened in the area where austenite has transformed to δ-Ferrit (3). A strong meniscus forms around the particle once the exogenous spinel particle is contacted by the melt (Fig. 2.11b). Endogenous particles move to the exogenous spinel, but slide along the meniscus and move away again (1, 2). The exogenous particle achieves a higher wettability with increasing temperature and test time (Fig. 2.11c). At the same time, the sharp-edged particle contours of spinel become visible again. Endogenous particles or clusters (3, 4) move to these areas of the exogenous particle, adhere and form clusters. Based on the roundness of the endogenous particles, it can be assumed that a large number of the inclusions are in semiliquid or liquid state (Fig. 2.11d).

It is remarkable that, unlike in experiments with MgO particles, the exogenous spinel particle sinks deeply into the steel matrix, as shown by a cross-section on the SEM in the article by Aneziris et al. [22]. The spinel interacts with the molten steel. Manganese accumulated in the spinel lattice. Alumina was detected at all interfaces towards the steel matrix, especially also under the spinel inclusion. Magnesium aluminate spinel is known as refractory oxide for its high stability [23]. It can absorb non-stoichiometrically many impurities from liquid slag/steel systems. The manufacturer Almatis GmbH gives upper limits for CaO, SiO₂, Na₂O und Fe₂O₃ impurities for the used Spinel AR78 (Table 2.1). However, EBSD examinations of the spinel prior to use did not detect any significant impurities. Another indication of a strong interaction of the steel with the spinel is that the molten steel overflowed some spinel particles and covered them completely. Individual spinel particles were still detected approx. 10 µm below the steel surface. The formation of a liquid layer around the spinel particle was not observed.

Approximately equal real maximum forces of 5·10^–15 N were calculated for the endogenous particles at about 1487 °C when moving along the meniscus, almost independently of their size. These forces are not primarily based on attractive forces, but appear to be influenced by convection and the Marangoni effect [22]. The Marangoni effect results from a gradient in surface tension caused by diffusion processes of different materials or temperature differences. It is a possible reason for the observed cyclic acceleration of the endogenous particles on their way to the host particle. The simplified calculation of the attractive force according to Eq. (2.1) is not valid in this case. With the dissolution of meniscus on the exogenous spinel particle at about 1504 °C, maximum real forces of 1·10^–17 N to 4·10^–15 N were determined for the endogenous particles, Fig. 2.12. As already observed for Al₂O₃ and MgO host particles, the maximum real force of endogenous particles increases with increasing endogenous particle size. While endogenous particle pairs with a size less than about 3 µm attract each other more strongly, the forces between endogenous Al₂O₃ and exogenous spinel particles are larger from a particle size of about 3 µm.

2.3.6 Interaction of Exogenous CaO·6Al₂O₃-Particles with Molten Steel and Its Endogenous Particles

Two different particle classifications were available for the investigations of exogenous CA₆ particles with the steel melt X15CrNiSi25-20. The powders with an average particle size d₅₀ of approx. 6 µm form conglomerates due to adhesion forces. Individual particles could not be separated.

Different phases of the melting of a steel plate with a CA₆ conglomerate can be seen in Fig. 2.13. First liquid areas formed on the steel surface already at 1345 °C, at temperatures approx. 90–120 K lower than in the previous investigations. When the melt comes into contact with the conglomerates, an irregular pulsating movement of the melt is observed, which is attributed to bubble formation. It was also observed that the melt resolidifies in the immediate surroundings of the conglomerate and new grain boundaries form around the conglomerate, Fig. 2.13b. This newly formed δ-ferrite has a higher melting temperature and remains on the conglomerate, even if the steel is already in the molten state in the surrounding area. The observations indicate that oxygen from the conglomerate reacts with the carbon supersaturated steel melt. Upon contact of the melt with the CA₆ conglomerate, µm-sized particles move towards it, but it is not possible to identify whether they are endogenous Al₂O₃ particles or CA₆ particles of the same species.

Further information on the interaction of CA₆ particles with the steel melt is provided by investigations on approx. 500 µm large particles. Figure 2.14 shows part of an exogenous CA₆ particle. At approx. 1485 °C, the first molten areas can be seen (Fig. 2.14a). Even though the temperature is increased further, the melt on the particle solidifies (Fig. 2.14b), similar to the observations with powder conglomerates. A holding time of approx. 30 s at 1515 °C led to renewed melting of the steel associated with an intensive accumulation of inclusions and clusters on the exogenous particle. Figure 2.14c shows individual paths from endogenous particles to CA₆ and an attached cluster. The maximum observed distance of the attractive forces is approx. 230 µm. After about 40 s, clusters of endogenous particles formed around the CA₆ (Fig. 2.14d). It can be seen that the endogenous particles are more densely bonded to each other than to the exogenous particle.

The maximum calculated attractive forces for endogenous particles and clusters are shown in Fig. 2.15. The attractive forces are several orders of magnitude larger than those of the previously studied exogenous materials. A reason for this is the much larger particle size of CA₆.

A section through a CA₆ particle after the melting test shows that the exogenous particle is not immersed (Fig. 2.16). It lies on a relatively flat solidified steel surface.

These observations support the thesis that the buoyancy forces acting on the CA₆ are large due to its low density and approximately 10% porosity. Despite the visible dendritic solidification of the steel under the particle, indicating that large portions of the steel were in the molten state, the particle only partially contacts the steel surface. Voids were formed between the particle and the steel and non-metallic particles have deposited on both the steel plate and the exogenous particle. The CA₆ particle exhibits intense cracks (Fig. 2.16b). Both Al₂O₃ and steel particles were detected on the contact surface between exogenous particle and steel. Detailed images of a CA₆ particle after the melting test show a porous body with a brighter oxidic phase enriched in Mn, Si and Cr from the steel [Fig. 2.17a, b (4, 5, 7)].

The pores of CA₆ particle are partially filled with an (Al,Si,Ca)-rich slag (3, 8). The Ca contents of the slag are greater than that in CA₆. The peripheral regions of CA₆ are depleted in Ca (Fig. 2.17c) and enriched in Mn (Fig. 2.17d). This is illustrated by the higher values of the Al/Ca-at.% ratio compared to the ratio in CA₆, which has the value of 12. Also pure alumina was analysed (6). The composition of the phases and their Al/Ca ratios are summarised in Table 2.6.

Table 2.6

EDX-analysis of areas in Fig. 2.17a, b (fractions in at.%) and Al/Ca ratio

Spot	Al	Ca	Si	Mn	Cr	Al/Ca
1	35.0	2.8	0.3	0.06	0.4	12.6
2	22.6	2.3	0.4	0.06	–	9.7
3	25.7	4.8	7.4	1.7	0.7	5.3
4	22.9	0.8	3.5	5.2	2.8	26.9
5	27.8	1.5	1.2	4.1	1.8	18.6
6	40.0	–	–	–	–	–
7	21.6	0.6	2.7	7.3	0.8	36.0
8	10.6	2.8	13.2	6.1	0.2	3.8

Melting tests of cylindrical steel samples using the sessile drop method on sintered CA₆ tablets in the heating microscope confirm the formation of an oxide phase (2) on the liquid steel drop (Fig. 2.18). The oxide layers influence the wetting angle and led to divergent results between 115 and 140° at a test temperature of 1450 °C for three tests. Monaghan et al. [24] found wetting angles of 130° at 1450 °C and <110 °C at 1550 °C for iron with a dissolved carbon content of 5 wt.% on CA₆ plates. Wei et al. [25] determined a contact angle of 134° for electrolytic iron on a CA₆ tablet at 1600 °C. Analog to Yin´s conclusions, the high wetting angles between 100° and 130 lead to high attractive forces also [5].

After the test, individual ceramic particles adhere to the ceramic-steel interface, whereby it cannot be clearly assigned whether the particles were newly formed from the steel matrix or detached from the CA₆ tablet. The liquid (Al,Si,Ca)-rich phase was also observed in the pores of the CA₆ tablets to a depth of approx. 300 µm (Fig. 2.19). While the slag at a depth of approx. 160 µm has a Ca content of 5 at.%, the Ca content at a depth of approx. 90 µm is <1 at.%. Additionally, in contact with the liquid phase, the CA₆ is depleted in Ca and enriched in Cr. The CA₆ particles that were in contact with the slag have smoother boundaries than the unaffected CA₆ (Fig. 2.19b, c). Small dendrites grow from their boundaries to the slag (Fig. 2.19d).

It is possible, that the liquid oxide from the slag solidified at the former CA6 grain boundaries. After immersion tests, Storti [13] already found whiskers on the CA6-coated surface of carbon-bonded alumina filter material. Jiang et al. [21] report a steel/slag reaction in which solid oxidic particles transferred into lower melting point inclusions in presence of only 0.0002 wt.% dissolved Ca. Raviraj et al. [26] observed the reduction of a Si-Ca-Al rich solid slag with the formation of cavities with embedded inclusions using low and high Al steels. In this case, SiO is reduced to Al₂O₃, unfortunately without taking into account the high CaO content originally present in the slag.

In the binary Al₂O₃-CaO diagram (Fig. 2.20), CA and CA₆ are in equilibrium with CA₂ at about 1300 °C. As the CaO content increases, the melting point of the phases drops from 1766 to 1370 °C. The liquid silicon oxide, which is formed by reducing the Cr₂O₃ passive layer on the steel surface and is also present in minute quantities in the exogenous CA₆ particle, can contribute to a further reduction in the melting temperature. FactSage calculations of the steel with the addition of 10^–2 g each of Cr₂O₃, Al₂S₃, FeS and MnS detected as sulphides on the steel surface, 40 ppm oxygen and 1 g CA₆ related to 100 g steel lead to the formation of 0.36 g liquid slag with 69 wt.% Al₂O₃, 23 wt.% CaO, 4 wt.% SiO₂, 2 wt.% Ti₂O₃ and approx. 1 wt.% (TiO₂ and CrO). CA₆ no longer exists as an equilibrium phase. Calculations without Cr₂O₃ and sulphides with 40 and 400 ppm oxygen at 1515 °C lead to a small reduction of CA₆ under the formation of slag from Al₂O₃, SiO₂, Ti₂O₃, TiO₂, CaO, MnO, CrO and FeO.

2.4 Conclusion

The trend lines of the maximum real forces of endogenous Al₂O₃ particles interacting with exogenous Al₂O₃, MgO und MgAl₂O₄ particles are summarized in Fig. 2.21. Accordingly, endogenous Al₂O₃ particles smaller than about 2.5 µm in diameter exhibit the largest forces to the conspecific material. Clusters with a particle diameter greater than 4 µm based on its area are more strongly accelerated to magnesia. The endogenous alumina particles with size of 2.5 µm and larger were most strongly attracted to the exogenous spinel particle. However, the endogenous Al₂O₃ particles formed into clusters detached from the spinel in the progress of the experiment (Fig. 2.11d).

In contrast to the investigations presented above, the initial material of CA₆ had a very small particle size of approx. 6 µm or a much larger particle size of approx. 500 µm. A direct comparison of the attractive forces to the other refractory materials was therefore not possible. However, it could be shown that, analogous to the observations of Storti et al. [13], a liquid phase forms on the surface of the CA₆ particle in contact with the melt. An (Al,Si,Ca)-rich liquid phase was also detected in the pores of the coarsely ground CA₆ refractory material. Since this phase has more Ca in deeper areas than in the peripheral areas, it can be assumed that the dissolved Ca evaporates. The cracks in the CA₆ particle and the cavities under the particle after cooling of the sample could be due to vapour phase formation. On the other hand, the formation of an (Al,Mn,Cr)-rich oxide is observed at the particle surface and inside the particle in direct contact with the liquid phase. The deposition of alumina, manganese and chromium oxide from the molten steel can help to increase the purity of the steel.

2.5 Summary

Refractory materials are used in metallurgy to protect ladles and crucibles due to their high temperature resistance. They are also used as filter materials to remove/reduce non-metallic inclusions and contribute to increasing the degree of purity of steels. In the present work, the interactions of Al₂O₃, MgO, MgAl₂O₄ and CA₆ refractories with molten steel X15CrNiSi25-20 were investigated. The steel X15CrNiSi25-20 is an aluminium-killed stainless steel for quenching and tempering. Its carbon content is low, which minimises the tendency for carbon loss during the melting process. Since the ferritic high-temperature phase forms only to a small extent, a relatively stable melt pool is possible over a sufficiently long time interval to observe the interaction of endogenous and exogenous particles during initial contact with the melt. Endogenous aluminium oxides form on steel surface during melting. In this process, the dissolving chromium oxide passivation layer serves as oxygen source.

Exogenous alumina refractory does not react with the steel melt. The particle remains unchanged in its chemical composition. Already at the first contact of the molten steel with the exogenous alumina, the endogenous alumina particles move towards the larger exogenous particle and adhere to it. In contrast, chemical reactions were observed in the interaction of exogenous MgAl₂O₄ spinel, MgO and CA₆ particles with the molten steel. They lead to the formation of a strong meniscus in the case of the spinel and to the formation of a liquid phase on the MgO and CA₆ particles, respectively. Contact of the endogenous inclusions with the exogenous refractory at the surface of the molten steel was observed only after degradation of the meniscus or dissolution of the liquid interface in the HT-CLSM. CA₆ has a special feature. In the porous CA₆ particle, a liquid (Al,Si,Ca)-rich phase forms in the voids. In addition, the formation of (Al,Mn,Cr)-rich oxides was observed. The CA₆ particle is depleted of Ca. The emission of Ca can contribute to greater bath movement in the molten steel and lead to new slag being able to penetrate the porous particle.

The maximum forces determined between the endogenous alumina particles and the tested refractories depend on the size of the endogenous inclusions. As the endogenous particle size increases, the maximum real forces acting on the particles increase. Endogenous particles smaller than about 2.3 µm were most strongly attracted to conspecific exogenous particles. The maximum real forces were largest for larger endogenous alumina particles interacting with exogenous spinel. At the same time, it was observed that the adhesion of endogenous particles to the spinel was low. A direct comparison to the attractive forces using CA₆ as host particle was not possible due to the different particle classification.

Acknowledgements

The authors would like to thank Ms. Grahl for metallographic sample preparation and Mr. Fischer for technical assistance with equipment maintenance. The studies were carried out with financial support from the German Research Foundation (DFG) within the framework of the Collaborative Research Center CRC 920 “Multi-Functional Filters for Metal Melt Filtration–A Contribution toward Zero Defect Materials” (Project-ID 169148856) subproject A01.

Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

Vorheriges Kapitel Carbon-Bonded Filter Materials and Filter Structures with Active and Reactive Functional Pores for Steel Melt Filtration

Nächstes Kapitel Ceramic Filter Materials and Filter Structures with Active and Reactive Functional Pores for the Aluminum Melt Filtration

K. Janiszewski, Z. Kudliński, Removal of liquid non metallic inclusion from molten steel using method filtration. Metal 1–9 (2006). Retrieved from http://metal2014.tanger.cz/files/proceedings/metal_06/indexe.htm

P.A. Kralchevsky, V.N. Paunov, N.D. Denkov, I.B. Ivanov, K. Nagayama, J. Colloid Interface Sci. 155(2), 420–437 (1993). https://doi.org/10.1006/jcis.1993.1056CrossRef

P.A. Kralchevsky, K. Nagayama, Langmuir 10(1), 23–36 (1994). https://doi.org/10.1021/la00013a004CrossRef

K.D. Danov, B. Pouligny, P.A. Kralchevsky, Langmuir 17(21), 6599–6609 (2001). https://doi.org/10.1021/la0107300CrossRef

H. Yin, H. Shibata, T. Emi, J. KIM, Capillary attraction of solid particles at gas/steel melt interface. HTC 97, 380–388 (1997)

C.G. Aneziris, S. Dudczig, J. Hubálková, M. Emmel, G. Schmidt, Ceram. Int. 39(3), 2835–2843 (2013). https://doi.org/10.1016/j.ceramint.2012.09.055CrossRef

H. Yin, H. Shibata, T. Emi, M. Suzuki, In-situ” observation of collision, agglomeration and cluster formation of alumina inclusion particles on steel melts. ISIJ Int. 37(10), 936–945 (1997)CrossRef

W. Mu, N. Dogan, K.S. Coley, Metall. Mater. Trans. B 48(5), 2379–2388 (2017). https://doi.org/10.1007/s11663-017-1027-4CrossRef

S. Kimura, K. Nakajima, S. Mizoguchi, Metall. Mater. Trans. B 32(1), 79–85 (2001). https://doi.org/10.1007/s11663-001-0010-1CrossRef

10.

H. Yin, H. Shibata, T. Emi, M. Suzuki, ISIJ Int. 37(10), 946–955 (1997). https://doi.org/10.2355/isijinternational.37.946CrossRef

11.

W. Mu, N. Dogan, K.S. Coley, JOM 70(7), 1199–1209 (2018). https://doi.org/10.1007/s11837-018-2893-1CrossRef

12.

K. Uemura, M. Takahashi, S. Koyama, M. Nitta, ISIJ Int. 32(1), 150–156 (1992). https://doi.org/10.2355/isijinternational.32.150CrossRef

13.

E. Storti, M. Farhani, C. G. Aneziris, C. Wöhrmeyer, C. Parr, Steel Res. Int. 88(11), 1700247 (2017). https://doi.org/10.1002/srin.201700247

14.

C.G. Aneziris, C. Schroeder, M. Emmel, G. Schmidt, H.P. Heller, H. Berek, Metall. Mater. Trans. B 44(4), 954–968 (2013). https://doi.org/10.1007/s11663-013-9828-6CrossRef

15.

X. Wei et al., J. Eur. Ceram. Soc. 42(11), 1676–4685 (2022). https://doi.org/10.1016/j.jeurceramsoc.2022.04.058CrossRef

16.

Y. Rotenberg, L. Boruvka, A.W. Neumann, J. Colloid Interface Sci. 93(1), 169–183 (1983). https://doi.org/10.1016/0021-9797(83)90396-XCrossRef

17.

D. Chebykin, T. Dubberstein, H.-P. Heller, O. Fabrichnaya, O. Volkova, Ceram. Int. 48(3), 3771–3778 (2022). https://doi.org/10.1016/j.ceramint.2021.10.160CrossRef

18.

W. Weißbach, M. Dahms, C. Jaroschek, in Werkstoffe und ihre Anwendungen: Metalle, Kunststoffe und mehr, ed. by W. Weißbach, M. Dahms, C. Jaroschek (Springer Fachmedien, Wiesbaden, 2018), pp. 463–487. https://doi.org/10.1007/978-3-658-19892-3_12

19.

M. Hasegawa, in Treatise on Process Metallurgy, ed. by S. Seetharaman, (Elsevier, Boston, 2014), pp. 507–516. https://doi.org/10.1016/B978-0-08-096986-2.00032-1

20.

C. Schröder, U. Fischer, A. Schmidt, G. Schmidt, O. Volkova, C. G. Aneziris, Adv. Eng. Mater. 19(9), 1700146. https://doi.org/10.1002/adem.201700146

21.

J.-H. Park, I.-H. Jung, H.-G. Lee, ISIJ Int. 46(11), 1626–1634 (2006). https://doi.org/10.2355/isijinternational.46.1626CrossRef

22.

C.G. Aneziris et al., Adv. Eng. Mater. 15(12), 1168–1176 (2013). https://doi.org/10.1002/adem.201300155CrossRef

23.

I. Ganesh et al., Ceram. Int. 28(3), 245–253 (2002). https://doi.org/10.1016/S0272-8842(01)00086-4CrossRef

24.

B. Monaghan, M. Chapman, S. Nightingale, The wetting of liquid iron carbon on aluminate minerals formed during coke dissolution in iron. Faculty of Engineering—Papers (Archive) (2009). https://ro.uow.edu.au/engpapers/498

25.

X. Wei et al., J. Eur. Ceram. Soc. 42(5), 2535–2544 (2022). https://doi.org/10.1016/j.jeurceramsoc.2022.01.011CrossRef

26.

A. Raviraj et al., Metall. Mater. Trans. B 52(2), 1154–1163 (2021). https://doi.org/10.1007/s11663-021-02091-zCrossRef

Titel: In Situ Observation of Collision Between Exogenous and Endogenous Inclusions on Steel Melts for Active Steel Filtration
verfasst von: Christina Schröder
Xingwen Wei
Undine Fischer
Gert Schmidt
Olena Volkova
Christos G. Aneziris
Verlag: Springer International Publishing
Buch: Multifunctional Ceramic Filter Systems for Metal Melt Filtration
Print ISBN: 978-3-031-40929-5

Electronic ISBN: 978-3-031-40930-1

Copyright-Jahr: 2024
DOI: https://doi.org/10.1007/978-3-031-40930-1_2

Premium Partner

Marktübersichten

Die im Laufe eines Jahres in der „adhäsion“ veröffentlichten Marktübersichten helfen Anwendern verschiedenster Branchen, sich einen gezielten Überblick über Lieferantenangebote zu verschaffen.

Zur Marktübersicht