MIL

LEN

NIU

M S

TEEL

200

8

104

RH metallurgySince its introduction more than 40 years ago vacuum degassing has developed from being used primarily for hydrogen removal to an essential part of secondary steelmaking, with an ability to refine, alloy, heat and deoxidise steel. Metallurgical reactions are influenced by thermo-chemistry and vessel design as well as vacuum level.

During the past 30 years there has been a tremendous improvement in the secondary processing of steel. In

the first stage of this development simple ladle metallurgical processes were developed to:

` Improve deoxidation control` Remove inclusions by gentle bath stirring` Desulphurise steel by synthetic slags and injection

metallurgy` Modify inclusions primarily by calcium

In the next stage of development, ladle furnaces were implemented in order to improve productivity and the quality of steel. These have been used to:

` Reheat steel and control temperature` Buffer heats for sequence casting` Add large quantities of alloys` Homogenise composition and temperature` Produce ultra clean steels with an extended gentle

gas stirring` Desulphurise and, in some cases, dephosphorise steel

with synthetic slags

Since its introduction more than 40 years ago, vacuum

Authors: Dieter Tembergen, Rainer Teworte and Robert RobeySMS Mevac

degassing had been used primarily for hydrogen removal. Today, however, vacuum degassing processes are also used for secondary refining and have become increasingly important in the modern steelmaking industry, mainly because of the rising demand for cold-rolled sheet with improved mechanical properties. Pressure dependent reactions are the reason for the treatment of liquid steel under vacuum (see Table 1).

The dominating vacuum degassing system is the RH type degasser (see Figure 1). This process was named after the companies Ruhrstahl and Heraeus where the process was developed. The amount of steel requiring vacuum treatment will continue to grow and every international competitive integrated plant and many electric furnace plants will require vacuum degassing if they are to remain competitive.

RH/RH TOP PROCESSThe mechanism of the vacuum treatment of liquid steel is shown in Figure 1. The unit consists mainly of a refractory lined reaction vessel and two steel pipes attached to the bottom of this vessel. These are the inlet and outlet snorkels. Both of them are completely refractory lined on the inside but only the lower part is refractory coated on the outside. The inlet snorkel is equipped with a number of gas injection pipes arranged in the lower section in one or two levels and equally distributed around the circumference.

The degassing process is started after both snorkels are sufficiently immersed into the melt. Before snorkel immersion the injection of inert gas, usually argon, is started in the gas pipes of the inlet snorkel. Having achieved the required immersion depth, the reaction vessel is evacuated by means of a vacuum pump system which is connected to the reaction vessel via an off-take duct. The density of liquid steel is assumed as 6.94t/m3 at 1,600°C. The atmospheric pressure exerted on the ladle surface causes the steel in the snorkels to rise to a barometric height of approximately 1.45m under deep vacuum conditions.

Due to the economic benefits achieved by using secondary metallurgical equipment the processes of steel treatment

r Fig 1 Schematic view of the reaction area of the RH process

STEELMAKING

MIL

LEN

NIU

M S

TEEL

200

8

105

a

RH KINETICSRH vacuum processes, in general, do not reach equilibrium and the amount of hydrogen, carbon and nitrogen removed are governed by kinetic considerations.Decarburisation The decarburisation mechanism is fairly complex as the reaction kinetics depend both on the circulation rate and rate of decarburisation. The effect of bath mixing on decarburisation is well known and many studies have been performed to enhance this reaction. Based on a simple simulation, molten steel in vacuum refining equipment can be divided into two vessels (see Figure 3): a reaction vessel, where the decarburisation occurs, and a mixing vessel.

Between the two vessels, the molten steel circulates at constant speed. The mass balance of carbon in the mixing and the reaction vessel is represented by the following equations [3]:

in vacuum conditions are under continuous development. The development of vessel geometry (size and shape), cross-section of snorkels and production capacities of RH units is demonstrated in Figure 2 [2]. It shows the vessel design of RH units in Thyssen’s steelmaking shops since the beginning of RH technology.

RH PROCESS CHARACTERISTICSWhen the RH process was introduced, the primary target was the reduction of hydrogen in liquid steel. The first results were not as successful as expected due to an insufficient vacuum in the vessel, however, the application of steam ejector vacuum pumps in the early 1960s enabled sufficiently low pressure to be reached, leading to hydrogen contents of less than 1ppm.

The application of the RH process for decarburisation was introduced at the end of the 1970s. Today, extremely low final carbon contents of less than 20ppm can be obtained, as required for the production of automotive sheets. The addition of alloying elements during degassing has the advantage of achieving high yields for alloys and high accuracy of chemical analysis of steel due to the absence of air and the avoidance of metal-slag reactions.

Further developments were the use of gaseous oxygen during RH treatment in RHO, RH-OB, RH-KTB, RH-MESID and MFB processes. The targets for these processes were to accelerate the decarburisation reaction, to reheat liquid steel by alumino-thermic reaction, to remelt skulls, to keep vessel at high temperature by converting generated CO gas into CO2 during the decarburisation period, and to heat the refractory lined vessels between treatments. Recently, some RH TOP lances have also been used for blowing powder onto the liquid steel for reducing sulphur or carbon contents to the lowest contents. Today, all these processes, except RH-OB, are called RH TOP.

Parameter RH technology Tank & ladle technology RH-OB, RH-TOP RH VOD, VD-OB Tank LadleCarbon content – < 20 < 20 < 50 30-40 30-40 final product, ppmDecarburisation rate Very high Sufficient High 2-30% less than for LC steels with RH-TOP/VODDecarburisation < 13 < 15 Depends on 15-20 20 time, min initial C content Hydrogen removal All systems have sufficient performanceSteel cleanliness All systems improve steel cleanliness / VD-OB, VOD, tank and ladle need soft stirring periodDesulphurisation Powder injection No Very efficient by lance – seldomChemical heating Yes No Yes No NoRelative capital costs 1 0.8-0.9 0.4-0.6 0.4-0.5 0.3-0.4Operating costs →Decreasing

r Table 1 Comparison of various degassing systems

r Fig 2 Various RH units (ThyssenKrupp Steel, 1958-92)

MIL

LEN

NIU

M S

TEEL

200

8

106

(1)

(2)

where : V, v are volumes of mixing vessel and reaction vesselak is volumetric coefficient for decarburisationC1 and C2 are carbon concentrations in mixing vessel and reaction vesselCe is carbon concentration in molten steel in equilibrium with partial pressure of CO in gas phaseQ’ is recirculation flow rate of molten steelt is time

If equations 1 and 2 are solved, assuming that:

C1=C2=C10 at t=0 (3)

and v is considerably smaller than V, the following equations are obtained as approximate solutions:

(4)

(5)where:

is mass transfer coefficient for carbon in liquid steelk is apparent rate constantVM is v+VA is surface area of metal exposed to the vacuum

The effects of the circulation rate Q’ and the volumetric coefficient of decarburisation on the apparent rate constant are calculated and shown in Figure 4. The figure shows the presence of three regions:

1. Reaction control region where k increases slightly by increasing the circulation rate

2. Recirculation control region where k increases slightly by increasing the volumetric coefficient ak

3. Mixed control region where k is affected equally by the circulation speed Q’ and the volumetric coefficient ak

The improvement of the operational conditions to increase k (shortening the decarburisation time), can be investigated using Figure 4. If the recirculation rate is small, the apparent rate constant, k, is controlled by the circulation and can be improved by increasing the recirculation rate. This can be accomplished by increasing the snorkel diameter or, in some cases, by increasing the flow rate of the lift gas. In this case, the volumetric coefficient of decarburisation is sufficiently high. In other words, bath mixing has a strong influence in the early and middle periods of the decarburisation process, when decarburisation proceeds rapidly with decreasing vacuum pressure.

(1), (2) and (3) in this diagram represent three phases of an investigation to increase the decarburisation rate at a 270t RH plant. In the beginning of the investigation the conditions were in phase (1). As the size of the snorkel was increased to increase the recirculation rate, the conditions were improved to phase (2). Finally, as ak was augmented by agitating molten steel in the vessel more vigorously and increasing the interfacial reaction between gas and molten steel, the conditions were improved to phase (3).

r Fig 3 Two vessel model of bath mixing

r Fig 4 Apparent rate constant for decarburisation in a RH as a function of recirculation rate Q’ and mass transfer parameter ak (volumetric coefficient)

STEELMAKING

MIL

LEN

NIU

M S

TEEL

200

8

107

a

DEGASSINGHydrogen and nitrogen removal in RH TOP plants follow similar equations. In general, the rate constant for hydrogen is greater than for carbon because the mass transfer coefficient for hydrogen is a factor of three times higher than carbon. However, at low hydrogen levels He may become important and must be taken into account.

Nitrogen removal is controlled by liquid phase mass transfer and chemical kinetics. The chemical aspects depend on the oxygen and sulphur content. This is because oxygen and sulphur segregate to the surface of the lift gas bubble and impede the formation of N2. In general, the nitrogen removal rate increases as the oxygen and sulphur contents decrease. The rate is slow compared to hydrogen removal. If oxygen and sulphur are low, the rate is controlled by liquid phase mass transfer, which is also slow. The rate constant is about one third of that of hydrogen.

LIFT GAS, SNORKEL GEOMETRY AND STEEL RECIRCULATIONAs the RH process is based on the exchange of molten steel between the ladle and the RH vessel, the rate of steel recirculation determines the velocity of the metallurgical reactions and the duration of the process assuming a defined metallurgical target. Melt circulation depends on the geometry of the equipment such as snorkel diameter, the radius of the equipment, and the position and number of lift gas tuyeres.

In state-of-the-art RH facilities, heavy splashing of the steel melt inside the vacuum vessel is a further limiting factor which is playing an increasingly important role. This effect occurs at extremely high decarburisation rates, which correspond to high k values.

A typical example of the change of carbon content with time is given in Figure 5. The decarburisation reaction was analysed assuming that it would the first order reaction of equation 4, where Ce=0. It indicates that the decarburisation reaction can be divided into three different stages with different k values. It is clear from the equations that the apparent decarburisation rate, k, can be obtained from the slope of a straight line shown in a -ln([Ct]/[Cini] versus time plot.

In stage 1, where the evacuation of the vacuum vessel and the recirculation of the melt have commenced, the value of k is the smallest of the three stages. In stage 2, the k value is the largest, and approximately 80-90% of the total carbon removed during the RH process is removed within this step, indicating that this is the main decarburisation step. When decarburisation reaches stage 3, the carbon content approaches a critical value which is different for each RH facility. The critical carbon content between stage 2 and stage 3 is approximately 20-50ppm.

In the past, many investigations have shown that the decarburisation rate of stage 2 can be influenced by the recirculation rate and the pump down behaviour.

r Fig 5 Changes of carbon content and -ln([Ct]/[C0]) with the decarburisation time in a RH unit

MIL

LEN

NIU

M S

TEEL

200

8

108

STEELMAKING

where:U is steel recirculation rate

is diameter of the up-snorkel is diameter of the down-snorkel

G is the flow rate of lift gasH is the length of snorkel used by the lift gas

(plume zone height)

The parameter H has been evaluated several times by users of the Ono formula and is defined as the length of the inlet snorkel used by the lift gas. With increased H-value the velocity of the up-stream in the inlet snorkel will also be increased. Equation (6) allocates the strongest influence to the snorkel inner diameter.

These theoretical investigations were considered for the engineering and design of SMS MEVAC’s new RH units, with the recommendation for an optimised relation between the ladle capacity and process related lift gas rates shown in Figures 6 and 7.

CONCLUSIONSThe RH process, now in industrial use for nearly 50 years, has been continuously developed to satisfy ever more stringent demands of the steel producer. Further developments are ongoing to provide an economic tool for decarburisation and hydrogen removal to even lower levels within a shorter treatment time and thus make the RH ‘fit for the future’. With achievable high levels of cleanness and accurate composition and temperature control, the RH process is a key facility in modern steelmaking. MS

Dieter Tembergen and Rainer Teworte are with SMS Mevac, Essen, Germany, and Robert Robey is with SMS Mevac UK, Winsford, UK.

CONTACT: thilo.sagermann@sms-demag.de

REFERENCES[1]F-J Hahn and HP Haastert, “Entwicklung der RH-Vakuummetallurgie für Stähle mit tiefen Kohlenstoffgehalten bei der Thyssen Stahl AG”, in Stahl und Eisen, 113(12), pp103-8 (1993).[2]RJ Fruehan, Vacuum Degassing, Iron & Steel Society (1990).[3]T Kuwabara et al, (trans.) ISIJ Int., 28, pp305-14 (1988).[4]W Lutz, Stahl und Eisen, 115(8) pp89-93 (1995).[5]B Kleimt et al, Stahl und Eisen, 115(8) pp75-81 (1995).[6]K Ono et al, Denki Seiko 56(7), pp149-57 (1981).

The first simple functional relations between steel recirculation rate, snorkel inner diameter and argon flow rate were published from 1963 to 1965. In 1971 Hupfer et al [1] tried to describe the influence of geometrical factors of the snorkel design on the steel recirculation rate. The calculation formulas developed by Kuwabara [2], Lutz [3] or Kleimt [4] also consider the effect of the vacuum on steel circulation.

The empirically derived formula (see equation (6) for the determination of steel recirculation rate, defined by Ono et al [5] considers the so-called ‘plume zone’ (H), the distance between point of lift gas injection and the refractory lined vessel bottom.

(6)

r Fig 6 Snorkel diameter vs heat size

r Fig 7 Lift gas flow vs snorkel diameter