Description

The invention relates to a taphole assembly, which is typically used in a metallurgical vessel (like a steel-making converter) for the discharge of molten metal. A generic taphole assembly is disclosed in US 4,427, 184. It is comprised of a series of refractory blocks (bricks), each defining a central bore and the bores being aligned to define respective and successive sections of the taphole installation.

The invention starts from this generic taphole design.

It has been practice in the steel industry to use taphole bricks (blocks) made of a refractory ceramic material based on magnesia (MgO). This is in particular true for taphole assemblies in contact with molten metal in basic steel making furnaces, such as a so-called BOF (basic oxygen furnace).

Although taphole assemblies of this type have proved successful for long there is a continuous demand for improvements.

Indeed there is a need for improvement with respect to so-called clogging, i. e. reducing or even avoiding a built-up of steel and/or slag at the inner surface of the taphole bricks, in particular at the bore hole of the terminal brick, being the brick which is the lowermost one when the taphole assembly is in its tapping position.

Such adherences reduce the inner cross section of the bore hole (tapping channel) and thus reduce the efficiency and quality of the metal melt flowing through the taphole.

During trials and after working intimately with taphole assemblies of the type described it has been the experience that keeping the cold face brick (which is the terminal brick according to the previous definition) warmer will reduce cold face adherences (clogging).

One alternative to "keep the terminal brick warmer" is to heat the said brick, for example by induction, but this will cause additional costs and amendment of the overall tapping method. Insofar the invention takes a different approach, namely to wrap the terminal brick with an insulation layer. During said tests mentioned it was found that the thermal conductivity of said insulation layer must be at least 20 % less than the thermal conductivity of the adjacent MgO-based refractory material.

Such wrapping keeps the overall brick warmer during tapping. The inventors have seen a reduction in temperature loss of up to 200°C.

In its most general embodiment the invention relates to a taphole assembly for a metallurgical vessel, comprising the following features:

- it is made of at least two ring shaped bricks,

- the bricks are made of an MgO based refractory material and

arranged one after the other to form a continuous tapping channel,

- a terminal brick at a lower end of the taphole assembly, when the metallurgical vessel is in a tapping position, is a multi-layer brick, comprising an outer layer, made of a material which has a thermal conductivity of at most 80% of the thermal conductivity of an inner MgO based layer.

The outer layer of the terminal brick (terminal brick being equivalent to an end section of a monolithic one-piece taphole) can be made of an Al ₂ 2 ₃ -based refractory material, for example a monolithic refractory, comprising at least 60wt.-%, >75wt.-% or >90wt.-% alumina, i.e. a high alumina refractory.

According to one embodiment the outer layer (the insulating layer) of said terminal brick has a wall thickness, which is between 70 and

130wt.-% of the wall thickness of the inner MgO based layer.

In other words: The wall thickness (being the wall thickness measured in a radial direction to the tapping channel) of the insulation layer may be about the same as the inner MgO based layer, bigger or smaller, depending on the insulation effect required.

As mentioned above comparative tests have been made.

Test A with a taphole assembly, made of generic MgO bricks (ring shaped blocks according to US 4,427, 184).

Test B with the proviso that the terminal brick was split into two layers of identical thickness.

The inner layer was made of the same MgO material as all other bricks, while the outer layer was made of a refractory material comprising 98wt.-% Al ₂ O ₃ (alumina). Both assemblies were installed in a BOF.

During tapping the inner layer of the terminal brick according to test B had a temperature of about 190° higher than that of the terminal brick of test A.

The tapping channel may have any cross-section. A cross-section of the channel, which gets smaller between the inlet end, when the vessel is in a tapping position, and its outlet end, is one favorable option.

The outer layer of the terminal brick may have a wall thickness such that its outer surface is substantially flush with corresponding outer surfaces of the other bricks of the taphole assembly. This makes it easier to integrate the assembly into the adjacent refractory lining of the

metallurgical vessel and to replace the assembly and/or the terminal brick. The use of such terminal brick with an outer insulation layer, for example the mentioned high alumina envelope, allows the following effects:

- increases the temperature of the inner brick layer during a tapping campaign,

- reduces the appearance of adherences at the inner taphole wall,

- avoids knocking off of adherences,

- increases the overall taphole life,

- improves the tapping stream (none or reduced cold face growth),

- reduces the need for taphole changes or repairs and

- is easy to realize and fit.

Further features of the invention will be apparent from the sub claims and the other application documents.

The invention will now be described by way of an example, wherein the only figure shows an axial (longitudinal) cross-section of one

embodiment of a taphole assembly.

The taphole assembly as shown comprises a series of refractory ring- shaped bricks (blocks) 1 , 2, 3, 4, each defining a central bore and the bores of adjacent bricks being aligned to define respective and successive sections of the taphole (channel).

With respect to blocks 1 , 2, 3 reference is made to US 4,427, 184 for further details.

All of said bricks 1 , 2, 3 are made of an MgO based refractory standard material comprising:

- 97wt.-% fused magnesia (<4mm) - at least one solid carbon carrier

- remainder: minor components.

Both are bonded by means of a liquid pitch binder. This material is mixed at about 150°C, then pressed into bricks and finally tempered at about 300°C.

Lowermost brick 4 differs from bricks 3 as it is made of two distinct layers, an inner layer 4i surrounding a bore 6 of said brick 4 and an outer layer 4o, surrounding said inner layer 4i, wherein the overall shape of brick 4 corresponds to that of brick 3.

While inner layer 4i is made of the same MgO-based material as brick 3 the outer layer 4o is made of a refractory material comprising:

- 70wt.-% tabulas and calcined alumina (<6mm)

- 25wt.-% MA-Spinell (<lmm) (MA: magnesia-alumina)

- 5wt.-% calcium aluminate cement

with a chemical analysis acc. EN ISO 12677 of A1 ₂ 0 ₃ = 91 ,5wt.-%, CaO = 2wt.-% and MgO = 6wt.-%, remainder: impurities.

The thermal conductivity of inner layer 4i is: 10,0 W/mK at 500°C and 5,0 W/mK at 1200°C.

The thermal conductivity of outer layer 4o is: 3,3 W/mK at 500°C and 2,7 W/mK at 1200°C. wherein the thermal conductivity is always established under reducing atmosphere (intert gas atmosphere) and according to the Dr. Klasse method, disclosed in "Berichte der Deutschen Keramischen Gesellschaft e. V., edition 6/57, pages 183-189" and referred to in UNITECR

proceeding 279, Vienna 2015, copy of which is attached. In practice, during tapping, a corresponding steel melt enters the tapping assembly via brick 1 , flows through corresponding bores of bricks 2, 3 into the bore 6 of brick 4 and leaves the tapping assembly via brick 4.

The new design of terminal brick 4 with outer insulation layer 4o allows to keep the terminal brick 4 during casting (tapping) at a temperature which is between 100°C and 200°C higher compared with the

temperature within the terminal brick 4 in a standard tapping assembly as shown in US 4,427, 184.

Caused by said temperature increase adherences along an inner wall W of bore 6/tapping channel C could be minimized characteristically.

It is within the scope of the invention to design (construct) at least one more brick of a taphole currently in the same way as said terminal brick.

Proceeding 279

HEAT CONDUCTIVITY MEASUREMENT OF REFRACTORIES-COMPARISON

OF METHODS AND CRITICAL REVIEW OF APPLYING RESULTS

Bernd Lorenzoni'. Rainer Neubock ¹ Michael Klikovich ² ,

'RHIAG; TC Leoben, Austria

2RHI AG; Vienna, Austria

ABSTRACT put to the hot wire and its temperature at two specific intervals

The most frequently used measurement methods for heat of time after the heating current is switched on. The variation in conductivity are critically discussed, namely the Hot Wire Tests temperature of the hot wire is a function of the thermal conduc(ASTM C1113, DIN EN 993-14 and 15), Laser Flash, ASTM tivity of the material from which the test pieces are made. The C201 and Dr. Klasse. The methods expose differences in the valmethod is used for materials with a thermal conductivity up to 1.5 ues generated on the same material and have different levels of W/(mK) and the minimum sample size is two bricks of 200 mm uncertainty. Not only the measurement type but also the sampling x 100 mm x 50 mm or equivalent, recommended dimensions are can have a significant influence on the results. Furthermore the 230 mm x 114 mm x 64 mm. Measurements can be carried out influences on the heat transfer calculation have to be taken careup to 1250 ^e C. [2]

fully into account when designing a lining concept.

DIN EN 993-15 (parallel)

INTRODUCTION The main difference to the aforementioned hot wire methods is

Thermal conductivity is one of the key properties of refracthe position where the temperature increase after the application tory products. It is used to determine the design of the overall reof the current to the heating wire is measured. The thermocouple fractory lining and the steel construction, but also the dimensions is placed 15 mm beside the hot wire with limbs parallel to it. of potential cooling equipment of all industrial high-temperature The method is used for materials with a thermal conductivity applications. It can even be critical in the decision to install a of up to 25 W/(mK) and the minimum sample size is again two certain refractory material or not. bricks of 200 mm x 100 mm x 50 mm or equivalent, recommended dimensions are 230 mm x 114 mm x 64 mm. Measurements

Hot wire methods can be carried out up to 1250 ^e C. [3]

All variations of the hot wire methods belong to the transient

type of methods for measuring thermal conductivity. A time deFlash method

pendent change of the sample temperature is measured. The hot The laser flash analysis belongs to the transient type of methods wire methods can only be used for non-carbonaceous, dielectric for measuring the thermal conductivity. It is not values obtained refractories and it is difficult to make accurate measurements of in an equilibrium state but the change of the temperature in deanisotropic materials but they have the advantage of being applipendence of time that is measured and used for the calculations. cable to unshaped (e.g. granular or powdered) samples. The sample is located in the middle of the furnace which can be heated electrically from room temperature up to 1400 °C (2550

ASTM C1113 °F). Measurements are always executed in an inert gas atmo¬

The test specimens consist of two 228-mm (9-in.) straight brick sphere. The sample size is the smallest of the methods described or equivalent which are heated in a furnace to specific temperin this article. A disc with approximately 20 mm (0.8 in) in diature levels (up to 1500 °C). When equilibrium conditions are ameter and a height of several millimeters is usually used for the reached a constant electrical current is applied to a pure platitest. The upper and lower surfaces of a sample material with a num wire placed between two brick. The rate at which the wire naturally bright color have to be blackened with graphite, to preheats is dependent upon how rapidly heat flows from the wire vent the laser light being reflected at the surface. Such a reflection into the constant temperature mass of the refractory brick. The would significantly deteriorate the precision of the measurement rate of temperature increase of the platinum wire is accurately de[4]. When the furnace and the sample have reached the desired termined by measuring its increase in resistance in the same way measuring temperature level a laser beam heats the lower surface a platinum resistance thermometer is used. A Fourier equation is of the sample for a few milliseconds. This results in an increase of used to calculate the thermal conductivity λ based on the rate of the sample temperature of only a few degrees Celsius. A detector temperature increase of the wire and power input. This method measures the time dependant temperature rise at the upper suris also called platinum resistance thermometer technique and is face of the sample and a computer program calculates the thermal used for materials with a thermal conductivity of up to 15 W/ diffusivity. There are several mathematical models which can be (mK). [1] used for the calculation. At each temperature level of the furnace the laser heats the specimen several times and an average value is

DIN EN 993-14 (cross array) determined for the thermal diffusivity. Between each "shot" the

In addition to the setup described above a thermocouple is welded sample needs to cool to the furnace temperature. In order to calto the center of the hot wire. The limbs of the thermocouple are culate the thermal conductivity the density and the specific heat perpendicular to the hot wire. capacity of the tested material must be known.

The thermal conductivity is calculated from the known power inThe laser flash analysis is best suited for fine grained carbon containing refractory materials, for example many isostatically measurements can be carried out in air (oxidizing conditions) or pressed products fulfill these criterions. Due to using a transient can be executed in an inert gas atmosphere. Inert gas atmosphere measuring method and the small sample size it is the fastest methit is commonly used for magnesia-carbon refractories as they are od described in this paper. A complete measurement is usually usually too coarse grained for laser flash.

completed within a day. As the specimen size is significantly smaller than one for the

ASTM method the equilibrium condition is usually achieved

ASTM C201 faster and as a result complete measurement is normally com¬

The ASTM C201 method belongs to the steady state methods for pleted within two days.

deteimining thermal conductivity. The heating chamber is placed

above the sample and can be heated electrically over a temperaDifferences in the values generated

ture range from 205 up to 1S40 °C (400 to 2800 °F). A silicon There are ongoing discussions about the different measuring carbide slab is positioned between the sample and the heating methods regarding their precision and the comparability of the rechamber to provide uniform heat distribution. Otherwise areas sults [7]. A simple comparison of measured thermal conductivity of the specimen closer to the heating elements would get hotter values from RHI-devices with literature values (e.g.[8]) cannot than areas farther away. Below the sample the copper calorimeter be done based only on the knowledge of the main chemical comassembly is located which measures the quantity of heat flowponents, as the density/porosity and other structural parameters ing through the test specimen. The water circulation through the obviously have a huge influence on the results. It is clear that not calorimeter is such that adjacent passages contain incoming and all refractory materials can be tested accurately in all of described outgoing streams of water to maintain a uniform heat distribution devices. For example many insulating materials often have a high at the lower side of the sample. Also the four independent parts porosity and quite large pores that negatively affect the precision of the calorimeter assembly have an individually adjustable rate of the results when tested with laser flash. An unexpected anisot- of water flow. The sample consists of three straight bricks and six ropy regarding thermal conductivity of a material can also lead to soap bricks and must be free of broken corners or edges. This is quite different values if the direction of the measured heat flux is important for a precise measurement. Of the test methods covered not identical. A slight electrical conductivity occurring at elevated in this article this method has the largest sample size. The matetemperatures might also deteriorate the result of a hot wire test. rial must cover the whole calorimeter assembly (approximately

456 by 342 mm or 18 by 13.5 in). Heat conductivity measurement of MgO-C-materials

When measuring the thermal conductivity at a specific temperaDue to their carbon content MgO-C refractories usually cannot ture the whole system has to fulfill certain equilibrium conditions be measured with the hot wire methods. At elevated temperatures that are specified in the ASTM standard [5]. As a result after evthe carbon chemically reacts with the platinum heating wire and ery temperature change of the furnace it takes up to twelve hours the thermocouples. Additionally the electrical conductivity of to reach those conditions. In the original setup of the device this the material makes it impossible to fulfill certain criterions for a was controlled by the operator. As this was very time consuming valid measurement (e.g. the current that should only heat the wire (and technically not up to date) the decision was made to upwould also go through the material and the temperature calculagrade the device. Now the equilibrium conditions are electronition from the now erroneously measured resistance of the wire cally monitored and additionally the measured data is recorded would also be faulty). The ASTM C201 device could be used by a computer. These changes have resulted in a halving of the for the measurement of MgO-C materials however the device is time for a measurement and improved accuracy. Due to the large not completely sealed. Additionally the long duration of a measample size and the small area in the center of the sample where surement will lead to carbon burnout and a decarburized sample the data for the calculation is collected, radial losses of heat don't surface changes the thermal conductivity. With the laser flash deplay a role for the precision with this experimental setup. The vice carbon containing refractories can be measured provided the duration of a complete measurement is dependent on the number grains in the specimen are not too large (grain size < 1 mm). The of temperature levels to be tested. Usually it takes three days up Dr. Klasse devices can be used for carbon containing materials. to one week. This method is best suited for insulating refractory Before the thermal conductivity of a magnesia-carbon refractory materials. material can be measured the sample must be coked. Otherwise volatile components that are set free when the material reaches

Dr. Klasse method certain temperatures for the first time severely damage the equip¬

The Dr. Klasse method is also a steady state method for thermal ment.

conductivity measurements. Again the whole system must reach For this paper a MgO-C material containing 14% carbon was a thermodynamic equilibrium state at each temperature step to chosen to be measured in the Dr. Klasse device and in the laobtain a valid result. At the bottom of a Dr. Klasse device a disk ser flash device. Four different samples were measured in the Dr. shaped electrical heating plate is positioned. For a homogenous Klasse apparatus, one sample was measured twice. The sample heat distribution an aluminium nitride slab is placed onto the for the laser flash was cut from a sample that was previously meaheating plate and the sample is placed directly above this slab. It sured in the Dr. Klasse device. Results are shown in Fig.l. The is also disk shaped with a diameter of 100 mm (4 in) and a height differences between the four samples measured in the Dr. Klasse of 25 mm (1 in). Two grooves are cut in the sample for the therapparatus seem to be significant. The two measurements of the mocouples. To be able to calculate the thermal conductivity of the identical sample are almost indistinguishable. The biggest differsample a reference disk of same dimensions and with confirmed ence is between the Dr. Klasse and the laser flash result. Other thermal conductivity is placed on top of the disk pile. To miniMgO-C materials with a different carbon content not included in mize radial heat loss the area around the pile is filled with granuthis paper showed similar behavior. As mentioned above the laser lar insulating refractory material [6]. With the Dr. Klasse devices flash method is more suited for fine grained materials (grain size < lmm). In this case the MgO-C material contains large magnesia

grains, which bridge completely through the measurement height

of the sample. The grains act like "a heat conductivity highway".

As the thermal conductivity of magnesia is very high at low temperatures mis alters the result significantly.

Fig.2: Simulated temperature profile ofthe ladle wall (case 1)

0

£ 0 300 600 900 1200

Temperature t/°C

Fig. 1: Comparison of results of thermal conductivity of

a magnesia carbon refractory with 14% carbon. The Dr.

Klasse measurements are denoted "Ml" to "M4-2". The flash

measurement "LFA". M4-1 and M4-2 is the same sample

measured twice.

Consequences for lining concepts

Based on these results simulations for a ladle wall were carried out. These simulations were executed for a wall consisting

of 162.4 mm MgO-C at the hot face followed by 64 mm fired

bauxite brick, then 10 mm of insulating material and a IS mm

steel shell. The main focus was on the outside temperature of the Fig.3: Simulated temperature profile ofthe ladle wall (case 4) steel shell and the heat flow density through the wall. For the first

simulation (case 1 , Fig.2) the thermal conductivity ofthe MgO-C Table 1 : Comparison of simulation results for a ladle wall with material was selected with 8.31 W/(mK). For the second simuladifferent thermal conductivities. Simulation parameters that tion (case 2) this value was increased to 12.12 W/(mK). For the were not changed for all simulations are not included.

third simulation (case 3) the thermal conductivity ofthe MgO-C

part was identical to the first case however the thermal conductivity ofthe insulating material was raised from 0.11 to 0.22 W/

(mK). In the fourth case (Fig. 3) both λ values were the higher

ones. The liquid steel temperature was 1650 °C and the ambient temperature was 37 °C for all simulations. The heat-transfer

coefficient between the outer steel shell and ambient air was 7 W/

(m ² K). To enable comparative analysis, a theoretical stationary

model was used Typically this type of stationary modeling results in steel shell temperatures and heat flow densities 10 to 25%

higher than when modeled using transient conditions. Results are

comprised in Table 1. CONCLUSION

Not all thermal conductivity measuring methods are applicable to all refractory materials. In addition the sample selection has an influence on the results as the materials are never completely homogeneous. In some cases variances in thermal conductivity have only a minor influence on parameters like the outside temperature of a metallurgical vessel. In other cases variances in the λ value can have a major impact on a lining concept. The simulations in this paper clearly show that the increasing the thermal conductivity ofthe hot face material has a much smaller influence than increasing the insulating material thermal conductivity. The heat flow density is also significantly affected by the change of the insulating material parameters. REFERENCES

[1] ASTM C1113/C1113M-09 Standard Test Method for Thermal

Conductivity of Refractories by Hot Wire (Platinum Resistance Thermometer Technique)

[2] EN 993-14 Methods of testing dense shaped refractory products Part 14: Determination of thermal conductivity by the hot-wire (cross-array) method

[3] EN 993- IS Priifverfahren fur dichte geformte feuerfeste Er- zeugnisse-Teil IS: Bestimmung der Warmeleitfahigkeit nach dem HeiBdraht- (Parallel-) Verfahren

[4] DIN EN 821-2:1997 Messung der TemperaturleitfShigkeit mit dem Laserflash- (oder Warmepuls-) Verfahren

[S] ASTM C201-93(2013) Standard Test Method for Thermal

Conductivity of Refractories

[6] Klasse, F., Heinz, A. and Hein, J. Vergleknsverfahren zur

Ermittlung der Warmeleitfahigkeit keramischer Werkstoffe.

Presented at the Jahrestagung der DKG, Wiesbaden, Germany, 17. June 1956

[7] Wulf, R, Ph.D. Thesis, Technischen Universitat Bergakade- mie Freiberg, Germany (Warmeleitfahigkeit von hitzebestan- digen und feuerfesten Dammstoffen-Untersuchungen zu Ur- sachen fur unterschiedliche Messergebnisse bei Verwendung verschiedener Messverfahren)

[8] Routschka, G., Wuthnow, H.; Handbook of Refractory Materials, 4* edition, Vulkan Verlag (2012)

Data citation:

- Proceeding title, see above

- Proceeding no., see above

- Authors' names, see above

- Source: USB-Stick UNITECR2015 - 14th Biennial Worldwide Congress

- ISBN 978-3-9815813-1-7

- urn:nbn:de:101:l-201506294612