High-Temperature Interaction of Molten Gray Cast Iron with Al2O3-ZrO2-SiO2 Ceramic

High-temperature interaction between molten conventional gray cast iron (wt.%: 4.1 C; 1 Si; 0.20 Mn; 0.03 S; 0.04 P, the rest—Fe) and multiphase Al2O3-ZrO2-SiO2 ceramic substrate (wt.%: 51.5 Al2O3, 33.8 ZrO2, 13.3 SiO2, ~ 1 NaO2, 0.15 Fe2O3 + TiO2 + CaO) was examined at a temperature of 1450 °C for 15 min under inert flowing gas atmosphere (Ar, 850-900 hPa) using a sessile drop method coupled with contact heating procedure. Melting, wetting behavior and solidification of gray cast iron sample were continuously recorded with a high-speed high-resolution CCD camera. Under conditions of this study, selected gray cast iron alloy does not wet the Al2O3-ZrO2-SiO2 ceramic forming the final contact angle with average value of θf ~ 135°. The wetting kinetic curve was not smooth and its shape suggests the effect of substrate chemical heterogeneity due to multiphase composition of the substrate material. After the sessile drop test, the solidified drop was easily detached from the ceramic substrate along the drop-side interface showing a lack of bonding between the drop and the substrate. Structural characterization of the drop/substrate couple evidenced the nucleation and growth of graphite at the drop surface and at the drop/substrate interface taken place during solidification and resulting into the formation of discontinuous interfacial graphite layer.


Introduction
High-temperature interaction between metal oxides and molten alloys takes place in several metallurgical processes such as metal casting (melt/crucible, melt/mold and melt/filter interactions) (Ref 1, 2), production of cast metal matrix composites (MMCs) containing oxide reinforcements ( Ref 3,4), reinforcing castings with ceramic inserts (Ref 5) and joining oxide-based ceramics (Ref 6). In these liquid-assisted processes, wettability plays a key role in the manufacture of highquality and defect-free products. Wetting phenomenon occurs when two media (liquid and solid) are brought together and it symbolizes the ability of a liquid to maintain contact with a solid surface.
Usually, wetting properties of materials are characterized by the value of contact angle (h) established at the triple point, where three phases meet together, i.e., a liquid (L), a solid (S) and a gas (G) ( Ref 7). The contact angle, as a quantitative measure of the wetting in a S/L system, can be expressed by the YoungÕs equation, which describes the balance between interfacial tensions r SG , r SL and r LG at the tree-phase S/L/G contact: Moreover, the contact angle characterizes also the S/L bonding (Ref 8) since it is related to the work of adhesion (W a ) via the Young-Dupré equation: In metal casting processes, non-wetting conditions (h > 90°) resulting in week melt/crucible, melt/mold and melt/filter interactions are required. On the contrary, in all liquid-assisted processes used for joining dissimilar materials (e.g., production of metal-ceramic composites by powder metallurgy, casting processes and additive manufacturing, joining ceramics, reinforcing of metal castings with ceramic inserts) good wetting (h ( 90°) is a prerequisite property for ensuring desired metal-ceramic bonding and high-quality final products. Therefore, the investigation of the wetting phenomena at elevated temperatures constitutes one of the most important scientific aspects of high-temperature liquid-assisted materials processing stimulated by the demands of modern metallurgy and foundry industry ( . In the last 50 years, extensive development has been done in the field of nonferrous MMCs, and nowadays, ferrous matrix composites have attracted many researchers due to technology developments, particularly those of iron foundry base (e.g., . Recently, attention is given to reinforcing cast iron alloys using casting process through incorporation of ceramic particles ( Ref 3,4) ceramics, as candidate materials for reinforcements, is of great practical importance. However, for oxides, there are only a few reports on wetting and interfacial phenomena with cast iron alloys ( . The aim of this study is to clarify the wetting behavior and reactivity of molten conventional gray cast iron in contact with Al 2 O 3 -ZrO 2 -SiO 2 ceramic in terms of the successful manufacture of cast iron-based MMCs. Alumina-rich multiphase Al 2 O 3 -ZrO 2 -SiO 2 refractory material has been selected as a potential ceramic reinforcement since it has beneficial combination of properties, i.e., improved strength, toughness, hardness and wear resistance over alumina alone combined with a lower cost than zirconia. Directly before wettability test, the gray cast iron sample was cut from a bulk casting and cleaned mechanically and ultrasonically in isopropanol. The AZS substrate of 15 9 15 mm size was cut from the AZS block (in wt.%: 51.5 Al 2 O 3 , 33.8 ZrO 2 , 13.3 SiO 2 , $ 1 NaO 2 , 0.15 Fe 2 O 3 + TiO 2 + CaO) produced by the Institute of Metal Science, Equipment and Technologies with Hydro-aerodynamics Centre (Bulgaria). The ceramic substrate was polished and also cleaned ultrasonically in isopropanol directly before testing.

Experimental
The sessile drop test was performed in protective atmosphere (Ar, 99.999%) using contact heating procedure, i.e., the cast iron sample was loaded into the vacuum chamber at room temperature by placing it on the ceramic substrate. After closing chamber and pumping up to a pressure of p = 2 9 10 À6 mbar, the metal/ceramic couple was contact heated with a rate of 15°C/min up to a temperature of 500°C, followed by introduction of flowing gas (Ar, 850-900 hPa). Next, the testing couple was heated with the same rate up to the experimental temperature of 1450°C, followed by isothermal heating at that temperature for 15 min before start of cooling process with a rate of 20°C/min. Throughout the entire test cycle, the images of the metal/substrate couple were recorded by a high-speed highresolution CCD camera MC1310 with a rate of 10 fps (frame per second) and while cooling, the recording rate was 1 fps. The collected images were processed by ASTRA2 software developed by CNR-IENI, Italy (Ref 14,15) for the determination of contact angles by an automatic image analysis. The contact angle values of the left (h l ) and the right (h r ) sides of the drop were determined from the collected high-resolution images of the sessile drop test with systematic uncertainty of than ± 2°. The same images were also used for generating a real-time movie of the high-temperature test (see Supplementary material).
After wettability test, the solidified sessile drop couple was investigated by scanning electron microscopy (SEM) using TM3000 HITACHI device equipped with energy-dispersive Xray spectroscopy (EDS) analyzer Quantax 70. Structural observations were performed both on the drop and the substrate surfaces as well as on the cross-sectioned drop/substrate couple. Figure 1 shows the images of the metal/substrate couple during different stages of the sessile drop test, including that at room temperature (Fig. 1a), during melting gray cast iron sample (Fig. 1b), wetting behavior during drop formation ( Fig. 1c) and wettability test (Fig. 1d,e) as well as upon drop solidification (Fig. 1f).

Results
The results of wettability test are presented in Fig. 2 showing the change of the left and the right contact angle values with time. Immediately after melting gray cast iron The next stage was accompanied with decrease in Dh value and the formation of perfectly symmetrical drop with corresponding Dh = 0°in about 5th minute at 1450°C. The first average contact angle h = (h l + h r )/2 formed at the test temperature (1450°C, t = 0 s) was h 0 $ 130°. During isothermal heating at that temperature for 15 min, both Dh value and the average contact angle value were slightly increased to the final value of h f $ 135°. Thus, one may conclude that under testing conditions of this study, the selected gray cast iron alloy does not wet the AZS substrate showing the contact angle h ) 90°.
Another very important fact was noted after cooling the couple to room temperature, i.e., the solidified drop was found separated from the substrate. This observation represents experimental evidence of weak bonding established during high-temperature test between the cast iron drop and the AZS substrate. Thus, one may conclude that weak bonding is in a good agreement with experimental results of non-wetting behavior of examined system. However, the detailed analysis of the real-time behavior of the couple during high-temperature test (see a movie in Supplementary material and extracted images in Fig. 3) supported our predictions that for examined couple, the most critical moment for establishing bonding between the drop and the substrate was the solidification of gray cast iron drop accompanied with increase of its volume. This positive shrinkage effect is demonstrated by the phenomenon of squeezing smaller ''daughter'' droplets from the ''mother'' drop in the final stage of its solidification well distinguished in Fig. 1f and Fig. 3a and b. Therefore, significant CTE (coefficients of thermal expansion) mismatch of the alloy (volume expansion) and the ceramic substrate (volume shrinkage) at the moment of the drop solidification and also during subsequent cooling of the solidified drop/substrate couple (dissimilar CTE values for the ceramic and the alloy) contributes to the debonding of two materials.
SEM observations of the couple from the top view ( Fig. 4ad) show that the surface of the drop is covered with large black graphite blots while occasionally between them there are small daughter droplets of white color growing from white region of the mother drop. This white region as well as small daughter droplets correspond to lower melting temperature part of the drop as they were squeezed from solidifying drop at the last moment of solidification due to expansion of solidifying gray cast iron. The substrate surface around the drop does not indicate any important structural changes (Fig. 4a) thus suggesting that the dark ring in the vicinity of the drop, well distinguished in Fig. 3a, is related with evaporation and deposition from the drop.
SEM/EDS analysis of cross-sectioned couple revealed no changes in the substrate structure after high-temperature test since the phase composition remains the same, composed of three binary oxides, e.g., Al 2 O 3 , m-ZrO 2 and t-ZrO 2 phases (Fig. 5a, b). Under law magnification (Fig. 5a), the microstructure of the drop looks homogeneous while under higher magnifications (Fig. 5b-d), less graphite precipitates are in the drop in the vicinity of its surface (Fig. 5c) and near the drop/substrate interface (Fig. 5d). This effect might be related with experimentally evidenced nucleation and growth of graphite at the drop surface (already mentioned above) and at the drop/substrate interface (see explanation below), both The detailed analysis of interfacial and near-interfacial regions (Fig. 6) exposed the presence of dark discontinuous layer up to about 35 lm thickness of which EDS analysis showed the appearance of carbon only. It is believed that this layer was formed during cooling through nucleation and growth of graphite phase at the interface, i.e., similar to that at the open surface of the drop. SEM observations showed that the debonding of the drop/substrate couple during cooling took place along the drop-side interface while a good bonding was established between the substrate and the graphite layer at the substrate-side interface. Nevertheless, based on the current observations, it is difficult to identify which phase of the substrate plays a role of a nucleation site for graphite because of discontinuous character of the graphite layer with its fragments locally detached from the substrate, most probably, during sample preparation for structural characterization (cutting and polishing). It should be highlighted that under magnifications used in this study for SEM observations (up to 5000 9), we did not distinguish any interfacial reaction products.
Local chemical analysis in different regions of the drop proved that the alloy matrix, particularly near the interface, contains about 3.3-3.9 Al, 0.3-0.6 Zr and 0.3-0.6 Na (at.%). No separate particles of new phases were identified in the solidified drop. These observations put forward a slight selective dissolution of the substrate constituents (mainly Al 2 O 3 and ZrO 2 ) in molten drop and suggest that no chemical reactions take place between selected gray cast iron alloy and AZS ceramic.

Discussion
Contrary to pure iron, only a few publications are available in the literature for information on contact angles and interface structures formed between molten cast iron alloys on oxidebased refractories. For Al 2 O 3 , SiO 2 and Al 2 O 3 -SiO 2 substrates Based on the results of this study, we may conclude that under applied testing conditions, weak interaction takes place between selected cast iron alloy and Al 2 O 3 -ZrO 2 -SiO 2 ceramic accompanying with slight dissolution of the substrate in molted drop. The formation of discontinuous graphite layer during solidification of molten drop contributes to a weak bonding between the drop and the substrate.
The above observations suggest that for the selected couple of materials, the improvement of their physico-chemical and thermomechanical compatibilities are needed in order to produce high-performance metal matrix composites by liquidassisted processes using gray cast iron (as a matrix) and Al 2 O 3 -ZrO 2 -SiO 2 ceramic (as a reinforcing phase or insert in the casting components). On the other hand, both non-wetting and non-reactive character of examined couple advises the applicability of Al 2 O 3 -ZrO 2 -SiO 2 refractory in foundry practice of the selected molten cast iron, e.g., as a material for porous ceramic filters used to remove non-metallic inclusions from molten metal and to improve the quality of final castings.

Summary
Under testing conditions in the sessile drop test of this study, the selected conventional gray cast iron alloy does not wet alumina-based Al 2 O 3 -ZrO 2 -SiO 2 ceramic and after melting and holding at T = 1450°C for 15 min, it forms a high contact angle of $ 135°. After the sessile drop test, the solidified drop was easily detached from the ceramic substrate showing a lack of bonding between them.
Structural characterization of the drop/substrate couple by SEM + EDS analysis evidenced the nucleation and growth of graphite at the drop surface and at the drop/substrate interface resulting in the formation of discontinuous interfacial graphite layer well bonded with the substrate. Non-wetting behavior accompanied with a lack of any interfacial reaction products and slight selective dissolution of the substrate constituents in molten drop suggests a weak interaction between the selected alloy and Al 2 O 3 -ZrO 2 -SiO 2 ceramic. Moreover, significant CTE mismatch between the alloy and the ceramic, particularly during alloy solidification, contributes to debonding of the drop/substrate couple that takes place along the drop-side interface.
These findings suggest that for liquid-assisted processes of metal matrix composites using gray cast iron (as a matrix) and Al 2 O 3 -ZrO 2 -SiO 2 ceramic (as a reinforcing phase or insert in the casting components), the improvement of physico-chemical and thermomechanical compatibilities between the selected couple of materials are needed through improvement of wetting properties and optimal design of structure and chemistry of interfaces.

Open Access
This article is licensed under a Creative Commons Attribution 4.0 International License, 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 licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence 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. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.