A proposal to the AISI

By

Alan W. Cramb
Professor
Department of Materials Science and Engineering
Carnegie Mellon University
Pittsburgh
Pa 15217

Quantifying the Thermal Behavior of Slags



Executive Summary



Successful operation of a continuous caster is based upon control of heat transfer in the mold. The mold slag is a key component in the success of continuous casting; however, the phenomena that occur in the gap between the shell and the mold are largely unknown as until recently there have been no techniques which have allowed visualization and quantification of the solidification behavior of liquid slags. This has lead to slag design being an empirical science or art.

Recently a new experimental technique was developed at Carnegie Mellon University that allowed the solidification behavior of a slag to be observed and quantified under conditions that simulate the thermal conditions that occur in steelmaking environments. This technique allows ladle, tundish and mold slags to be characterized under extreme conditions including those found between the mold wall and the growing shell of a continuous caster. Initial studies have shown that the solidification behavior of a slag is a function of its cooling rate and its environment.

It is proposed that a program be initiated to quantify and describe the phenomena that occur during the solidification of a slag in a steel mill environment. This will allow slag design to become an engineering science rather than an empirical exercise.

There will be two outcomes of this project:

&Mac183; The further development of a tool which will have broad use in the quantification of slag melting and solidification behavior
&Mac183; The development of a set of meaningful design criteria for slag application in steel mill environments.

This project has been developed by discussion with the member companies of the Center for Iron and Steelmaking at Carnegie Mellon University and the necessary industrial cost share for the project has already been approved by the board of the CISR. No additional steel industry funds are requested in the project. Industrial co-sponsors include the following AISI producer member companies: USS, Bethlehem Steel, AK Steel, LTV, National Steel, Ispat-Inland, Cleveland Cliffs, The Timken Company and North Star Steel and the following associate member companies: Air Liquide-America, Air Products, Foseco, Inc. and Praxair.






Introduction

Successful operation of a continuous caster is based upon control of heat transfer in the mold. The mold slag is a key component in the success of continuous casting; however, the phenomena that occur in the gap between the shell and the mold are largely unknown. Until recently there have been no techniques which have allowed visualization and quantification of the solidification behavior of liquid slags. This has lead to slag design being an empirical science or art. In general, there is not a detailed fundamental understanding of the phenomena that occurs during the solidification of a slag. This is a general problem in steelmaking and occurs in the ladle, tundish and mold of a caster, wherever slags are used. For example, slag crusting, which limits the application of basic fluxes to tundishes and occurs commonly in the ladle is a direct reaction of the liquid slag to a thermal gradient.

To overcome the limitations of differential thermal analysis (DTA) and direct casting experimentation in the measurement and understanding of the solidification phenomena of slags, the double and single hot thermocouple techniques (DHTT and SHTT) for the direct observation and measurement of mold slag crystallization were developed at Carnegie Mellon University. These methods enable the solidification and melting process of transparent slags to be observed "in situ" under conditions where the temperature or temperature gradient can be measured and controlled. The SHTT and DHTT allow a sample to be subjected to rapid cooling rates or to be held under isothermal conditions. The DHTT allows large temperature gradients to be developed between the two thermocouples and allows a simulation of the transient conditions, which can occur in the infiltrated slag film that occurs between the mold and the solidifying shell in the mold of a continuous caster. By these techniques both isothermal and non-isothermal phenomena can be studied.

A number of slags are optically transparent or translucent at steelmaking temperatures while the crystalline phase which precipitates upon cooling is opaque and can be clearly observed using optical microscopy. Thus the DHTT can be connected to an image capturing and analysis system that allows the onset and growth of the opaque crystals which precipitate from the slags to be documented. The development and application of these techniques to slag crystallization will be discussed in this proposal.


Technical Issues

Mold slags play an important role in the control of the continuous casting process and an understanding of mold slag properties and solidification behavior is the basis for optimization of continuous caster operation1-19). The precipitation of crystalline solids from mold slags during continuous casting is well documented and the use of casting powders with a high crystallization tendency ) has been reported to be beneficial in the casting of peritectic steel grades where the formation of a significant crystal fraction in the mold slag layer between the mold and the strand reduces mold heat flux and helps alleviate problems with longitudinal cracking2-7). A high crystallization tendency has, however, been shown to be deleterious to caster operation under high-speed casting conditions where increased friction between the mold and the strand and a higher incidence of sticking type breakouts has been documented8-15). For example, friction measurements carried out by Shibata et al8) in a slab caster showed that powders with a higher crystallization temperature gave larger friction forces. Hering et al9) noted that friction in the continuous casting mold depends on the viscosity of mold slag and the type of crystalline phases present.
The influence of crystalline phases on heat transfer in the mold has been the subject of a number of investigations2-6, 16-18). In general, it is commonly accepted that a larger proportion of crystalline phases (or a casting powder with higher crystallization temperature) in a liquid slag film decreases the overall heat transfer rate through the slag layer. This has been observed both in laboratory experiments3) and in plant trials2-3). This reduction in the heat transfer has been attributed to the formation of pores in the crystalline layer4) and to the scattering of radiation by the crystals, which reduces the radiative portion of the overall heat flux6. Jenkins7) has modeled heat transfer across the slag film present between the shell and mold of a continuos caster and has shown that the relative position of the crystalline layer in the slag film can also modify the total rate of heat transfer that is found in the mold of a continuous caster. Thus, it is necessary to understand the exact thermal conditions for precipitation and growth of a crystalline phase from a slag to understand the heat flux encountered in the mold of a continuous caster. It is also necessary to understand the crystal morphology and the temporal development of the fraction of solid to understand lubrication.
Quantification of mold slag solidification phenomena is normally conducted using differential thermal analysis (DTA) or slag casting in a chill mold; however, these techniques are limited as there is no direct method of observation of the solidification phenomena and only effects which liberate significant quantities of heat can be measured. The DTA is one of the most popular techniques for the study of slag crystallization. In this case, the slag sample is melted in a furnace and then cooled, together with a reference substance, and the temperature difference between both the sample and the reference material is recorded. When a reaction that involves heat generation or heat absorption takes place in the sample, a temperature difference between the sample and the reference can be detected. During sample cooling in a DTA, the highest temperature associated with an exothermic peak is usually called the "Crystallization Temperature" of the casting powder. In general, a higher "Crystallization Temperature" is thought to be related to a higher fraction of crystalline phase in the mold slag layer between the mold and the strand. However, a unique criterion to establish the cooling rate of the sample has not been established in these studies and only a few studies have considered the effect of the cooling rate on crystallization behavior16-18). Furthermore, the term "Crystallization Temperature" is sometimes confused with the so called "Solidification Temperature" of the slag which is determined from viscosity tests where a rapid rate of viscosity increase with decreasing temperature is incorrectly used as an indication of the onset of crystallization within the slag19).
Another technique that is frequently employed to determine the crystallization behavior of a mold slag consists of melting the mold slag sample and elevating its temperature to 1550°C before quenching in a metallic mold9, 11,12,16,18). After casting, the sample can be inspected at room temperature by different techniques. For example, the proportion of crystalline and glassy phases can be evaluated using microscopy and X-ray diffraction can be performed to determine the composition of the crystalline phases present. Obviously, these methods are useful to define a "crystallinity index" which compares, qualitatively, the tendency of different slags to give glassy or crystalline phases; however, these indexes do not provide a quantitative measurement that can characterize the slag behavior in the mold.
To properly characterize the precipitation of a second solid phase from a liquid, it is necessary to define the thermal field, the phase diagram and the nucleation and growth behavior of the solid. Slags are liquid oxides and easy glass formers under high cooling rates. It is thus necessary to be able to describe the conditions under which glass formation is possible, the conditions for the initiation of solidification, the crystal morphology, chemistry and growth rate and the time evolution of the fraction of solid to properly define the solidification behavior of such a material. It is well known from classical nucleation theory that the onset of crystallization in slags must be a function of cooling rate and, that to determine the solidification behavior of a liquid slag one must construct either isothermal time temperature transformation diagrams (TTT curves) or continuous cooling transformation diagrams (CCT curves). In addition, the growth rate, morphology and solidified fraction of the slag under varying cooling rates are important in the determination of the effect of crystallization of the slag on heat transfer and rheology. Thus a technique, whereby the thermal field can be determined as the solidification process is observed, was developed. This technique which combines the hot thermocouple technique with video observation and image analysis allows crystal growth rates, morphologies and solidified fractions to be determined under defined thermal conditions.
The concept of the single hot thermocouple technique dates back to the end of nineteenth century and the hot thermocouple method itself was successfully developed by Ordway 20) and Welch, et.al. 21) in the 1950’s. The progress of electronic development has made the hot thermocouple method easier and more reliable and has lead to renewed interest in the technique. Yanagase and Morinaga reported new developments of the hot thermocouple method and its application22-24) in the 1970’s and in 1980. Ohta et al24) applied the hot thermocouple method to the measurement of liquidus temperatures, clarification of the existence of two-phase regions in slags and also to the understanding of slag reactions. More recently (1993) Asayama, et al25) has discussed glass formation in silicate slags using the hot thermocouple method.
Ishii and Kashiwaya first developed the double hot thermocouple method and applied it to a microgravity experiment to determine the microstructural change of a superconducting oxide in 199227-29). The technique was also used by Murayama, et al26) to study Marangoni flow in silicate slags under microgravity and by Kuranaga et al27) to measure the ultimate length of a silicate slag film to clarify the mechanism of the separation of a bubble from a liquid slag surface.
Recently30-34) Kashiwaya developed the double hot thermocouple method for use with slags. A number of mold slags are optically transparent or translucent at steelmaking temperatures while the crystalline phase which precipitates upon cooling is opaque and can be clearly observed using optical microscopy. Thus the SHTT and DHTT are connected to an image capturing and analysis system that allows the onset and growth of the opaque crystals which precipitate from the slags to be documented. An example of the results from the technique is given in Figure 1 where a TTT curve for an actual mold slag was developed. This, to the authors’ knowledge, was the first complete TTT curve developed for an operating mold slag. X-ray analysis of the phases can be used to determine the actual solidification path.




Figure 1: TTT Curve for an Actual Mold Flux

In addition to studying actual fluxes the technique is most suitable to the study of the effect of chemistry variations on the solidification behavior of slags. An example is given in Figure 2 where the effect of sodium oxide additions to the solidification behavior of slags is shown.







Figure 2: Effect of Sodium Oxide on TTT Curves

Details of the solidification process can also be quantified and the evolution of the volume fraction of crystals as a function of time can be developed using this technique. An example is shown in Figure 3.



Figure 3: Effect of Sodium Oxide Additions to the Solidification
Behavior of a Slag.


In addition to isothermal experimentation, non-isothermal experiments are possible and cooling rates in excess of 100 C per second can be achieved. This is more than is necessary to quantify the conditions that occur with steelmaking environments. This is an area of extreme interest. The output of computer simulations of the thermal conditions within a mold can be used as an input to the experimental apparatus and, for the first time, the influence of the exact thermal conditions within the mold on the solidification of a slag can be observed and quantified. An example of thermal gradients from a simulation is given in Figure 4 and a schematic of the result is given in Figure 5.


Figure 4. Temperature profile of continuous cooling experiment with thermal gradient.




Figure 5. Schematic representation of the layer in the sample.

In addition to solidification, melting phenomena can also be observed. An example is given in Figure 6, where gas generation on melting is evident. This gas generation is related to atmospheric contamination by water vapor. In Figure 7 an example of melting under a defined gradient is shown that indicates that during melting from one side fluid flow is developed which leads to extreme shear stresses leading fragmentation of the crystalline array.




Figure 7: Melting behavior of a slag



Figure 8: Details of Remelting of a Slag

Thus the DHTT allows full quantification of the phenomena that occur during the melting and solidification of transparent slags.


Direction of Research

The development of the technique allows the following information to be developed:

1. The mode of solidification (equiaxed or columnar dendritic) of a slag as a function of cooling rate
2. Measurement of the chemistry of the precipitating crystal phase
3. Measurement of the segregation phenomena that occur during solidification
4. Determination of the response of a liquid slags to a thermal gradient
5. Determination of the environment (humidity) on slag solidification phenomena
6. Observation of the solidification phenomena that occur within the mold of a continuous caster.
7. Development of TTT and CCT curves as a function of slag chemistry.

This research will be aimed at developing a systematic understanding of the solidification phenomena of slags under conditions that are found in steelmaking environments.

Benefits of the Program

This program will allow rational design criteria to be developed for use in the applications of slags in ladles, tundishes and molds. Currently, this specialized information is completely missing and many industrial problems from slag crusting to the operation of a continuous caster are dependent upon knowledge of the response of a slag to its thermal history. In addition, to operational issues, the understanding of the solidification behavior of slags is a key in the understanding of the response of inclusional material to subsequent thermo-mechanical processing.

This program will allow the operation of continuous caster to be better understood and lead to increased productivity and decreased defect formation in continuous cast product. This information will also be a key to improving the castability of difficult to cast grades on all types of casters. Thus this work will aid in reducing the overall energy consumption of the steel industry by aiding in increased productivity and decreased defect ratio, both key components of successful hot charging.

Environmentally the project is also very important. As the environmental restrictions on steelmaking become more stringent, it will be necessary to develop fluxes that do not contain fluorides. At this point of time it is not possible to produce a fluoride free casting flux. This work will allow the criteria to be developed that will allow the successful development of such a flux.



Goal

Develop a detailed understanding of the solidification behavior of a slag to allow slag design to become an engineering science

Objectives

1. Develop a systematic understanding of the effect of cooling rate on slag solidification
2. Develop a systematic understanding on the effect of slag chemistry changes on slag solidification behavior.
3. Develop an method to characterize slag melting
4. Develop an understanding of the role of the environment on slag solidification and melting
5. Develop the ability to understand slag solidification under the conditions that occur in a continuous caster
6. Develop an ability to predict the solidification behavior of slags
7. Develop the criteria for optimization of slags in steelmaking environments where they are under thermal gradients

Project Plan


The sponsoring companies of the CISR will monitor the project and detailed project planning will occur at the biannual CISR meeting that will also serve as the official meetings of the project oversight committee. The plan for the project follows the objectives outlined above:

1. Defined slags of known chemistry will be quantified using the DHTT. It is planned to begin with CaO-Al2O3; CaO-SiO2; CaO-Al2O3-MgO as these are the route slags for ladle and tundish applications and also the emulsified inclusions which occur in steel products. TTT and CCT curves will be developed and precipitated phases identified.
2. After the above slags are quantified, the effect of Na2O, K2O, TiO2 and CaF2 additions to CaO-SiO2will be determined, as this is the beginning of mold slag design. TTT and CCT curves will be developed and precipitated phases identified
3. The effect of heating rate on slag melting phenomena will be determined for the slags outlined in 1 and 2.
4. The effect of atmospheric water vapor on solidification and melting behavior of the above slags will then be determined.
5. Run experimental trials using data on the thermal conditions from actual operating molds to determine the response of mold slags to the details of the thermal environment of a continuous caster mold.
6. Using nucleation theory and knowledge of the kinetic factors governing crystal growth a model will be developed to allow prediction of the solidification behavior of the slags.
7. Gather industrial data from operating casters and correlate operating results to the results of the experiments and determine the criteria for successful slag operation.



Schedule and Deliverables

The project is budgeted to last three years. The project start date is August 1, 1999. There will be monthly updates posted on a web page for the project and official quarterly and annual reports to the DOE. In addition there will be biannual project meetings with the projects advisory board held in conjunction with the CISR meetings. A Gantt chart outlining project timing is attached.

Project Budget

The total project budget is broken down in the official university budget attached to this proposal. Details are given in Table 1. The cost share of 30% will be supplied by the industrial co-sponsors. The following AISI producer member companies will be part of this project: USS, Bethlehem Steel, AK Steel, LTV, National Steel, Ispat-Inland, Cleveland Cliffs, The Timken Company and North Star Steel and the following associate member companies: Air Liquide-America, Air Products, Foseco, Inc. and Praxair.

Table 1: Budget Summary

Personnel 70,349 72,921 75,535
Departmental Expenses 13,200 13,200 13,200
Overhead 35,298 36,278 37,266
Total Project Cost 118,847 122,399 126,001
Cost Share 35, 654 (30%) 36720 (30%) 37800 (30%)
Sponsor Cost 83,193 85680 88201




References

1. M. Wolf : ISS Process Technology Conference, 13,(1995), 99
2. M. Emi: ISS Steelmaking Proceedings, (1991), 623.
3. P. Riboud and M. Larrecq: ISS Steelmaking Proceedings, (1979), 78.
4. H. Kyoden, T. Doihara, O. Nomura: ISS Steelmaking Proceedings, (1986), 153.
5. K. Mills, M. Susa, V. Ludlow: ISS Process Technology Conference, 13, (1995), 157.
6. A. Yamauchi, K. Sorimachi, T. Sakuraya, T. Fuji: ISIJ International, 33, (1993), 140.
7. M. Jenkins: ISS Steelmaking Proceedings, (1995), 315.
8. H. Shibata, J. Cho, T. Emi, M. Suzuki: Molten Slags, Fluxes and Salts ‘97 Conference, (1997), 771
9. L. Hering, H. Heller, H. Fenzke: Stahl und Eisen, 112(1992), 61.
10. H. Nakato, T. Sakuraya, T. Nozaki, T. Emi, H. Nishikawa: ISS Steelmaking Proceedings, (1986), 137.
11. H. Lindenberg, J. Loh.: ISS Steelmaking Proceedings, (1986), 161.
12. K. Sorimachi, M. Kuga, H. Nishikawa, K. Marumoto, H. Nakato: Transactions ISIJ, 21(1982), B-435.
13. K. Sorimachi, M. Kuga, M. Saigusa, T. Sakuraya: Fachberichte Hutttenpraxis Metall., 20(1982), 244.
14. S. Chang, I. Lee, M. Kim, S. Yang, J. Choi, J. Park: Conf. on Continuous Casting of Steel in Developing Countries, China, (1993), 832.
15. H. Gilles, M. Byrne, T. Ruso, G. DeMasi: ISS Process Technology Conf. Proceedings, (1990), 123.
16. H. Sakai, T. Kawashima, T. Shiomi, K. Watanabe, T. Iida: Molten Slags, Fluxes and Salts ‘97 Conference, (1997),787.
17. K. Watanabe, M. Suzuki, K. Murakami, H. Kondo, A. Miyamoto, T. Shiomi, "The effect of crystallization of mold powder on the heat transfer in continuous casting mold", Tetsu to Hagane, 83(1997),.31.
18. M. Bhamra, M. Charlesworth, S. Wong, D. Sawyers-Villers and A. Cramb: ISS Electric Furnace Conference, (1996), 551.
19. R. Bommaraju: ISS Steelmaking Proceedings, (1991), 131.
20. F.Ordway: J.Res. Natl. Bur. Stand., 48(1952), 512.
21. J.H. Welch: J. Sci. Instrm. , 31(1954), 458.
22. T. Yanagase, et.al.: Proc. of Inter. Conf. Sci. and Techn. Iron and Steel, (1971), 516.
23. Y. Ohta, K. Morinaga, T. Yanagase: Bulletin Japan Inst. Metal., 19(1980),139.
24. E. Asayama, H. Takebe and K. Morinaga: ISIJ International , 33, (1993), 233.
25. Y. Kashiwaya, D. Jukawa, K. Ishii: JASMAC-8 (1992).
26. Y. Murayama, Y. Kashiwaya, K. Ishii: CAMP-ISIJ, 7(1994), 891.
27. T. Kuranaga, Y. Kashiwaya, K. Ishii: Intrn. Sympo. on Adv. Mat. and Tech. for 21st Cent. JIM ‘95 Fall Meeting, Honolulu( 1995).
28. H. Scholze: "Glass", Springer-Verlag, (1991)
29. 29. D. R. Uhlman: J. of Am.Ceram. Soc., 66 (1983), p 95
30. Yoshiaki Kashiwaya Carlos E. Cicutti, Alan W. Cramb, Kuniyoshi Ishii: "Development of Double and Single Hot Thermocouple Technique for In-Situ Observation and Measurement of Mold Slag Crystallization", ISIJ International, Vol. 38, No.4, 1998, p 348-356
31. Yoshiaki Kashiwaya, Carlos E. Cicutti and Alan W. Cramb: Crystallization Phenomena of Mold Slag for Continuous Casting, ISIJ International, 1998, Vol 38, No 4, 1998, p 357-65
32. Yoshiaki Kashiwaya, Carlos E. Cicutti, and Alan W. Cramb:"Crystallization Behaviour of Slags", Steelmaking Conference, 1998, Vol 81, p 185-191
33. C. Orrling, A. Tilliander, Y. Kashiwaya and A. W. Cramb: "Melting and Solidification of Slags", Steelmaking Proceedings, 1999
34. Y. Kashiwaya, C. Cicuitti and Alan W. Cramb: "Development of Techniques for Observation of Flux Crystallization", Electric Furnace Conference Proceedings, 1997, p 617-622.
35. C. Cicuitti , Y. Kashiwaya and A. W. Cramb: "Mold Flux Crystallization", Proceedings of the 11th Steel Symposium, IAS, 1997, p9-20