OF MAKING SUCH PRODUCTS

This invention relates to an engineering ferrous metal product and to a method of making such a product. More particularly, but not exclusively, the product is a roll for use in rolling mills. The rolling mill may be, for example, used for rolling strip or sheet metal or maybe a calendering mill for paper.

An object of the invention is to provide a new and improved engineering ferrous metal product and a new and improved method of making such a product.

According to a first aspect of the invention we provide a method of making an engineering ferrous metal product comprising performing an electroslag remelting operation on a consumable electrode element comprising engineering ferrous metal and in which an ingredient, additional to said consumable electrode element, is added to the melt pool.

The consumable electrode element may comprise a complete consumable electrode and said additional ingredient may be added to the melt pool separately from the electrode.

Alternatively, the consumable electrode element may comprise a first part of a consumable electrode, the consumable electrode having a second part of a different composition to the first part and the second part providing said additional ingredient.

The first part of the consumable electrode may comprise an inner part of the electrode and the second part of the consumable electrode may comprise an outer part of the electrode.

From a first more specific facet the invention relates to a roll for use in rolling mills and to a method of making such a roll. The first more specific facet of the invention is particularly, but not exclusively, concerned with roughing work rolls of a hot strip rolling mill for rolling steel strip or sheet.

In strip or sheet metal rolling, requirements for better gauge accuracy, surface properties and productivity have lead to the development of high

performance rolls. Such rolls are required to have high resistance to rolling stress induced damage to the roll surface such as surface wear, cracking and spalling.

Hitherto monoblock forged and hardened steel rolls have generally been used for cold rolling and bimetallic rolls have generally been used for hot rolling because of the inability to produce monoblock rolls of metal providing the surface properties desired for hot rolling. Such bi-metallic rolls comprise a shell providing a surface layer of the roll made of metal of a composition to provide the desired surface properties and a core made of different metal to provide the desired core properties.

Typically the shell metal comprises high alloy steel or high alloy cast iron and the core metal comprises nodular cast iron or plain steel.

There are several techniques capable of producing such bi-metallic rolls, for example:-

Centrifugal Casting; in which metal of shell composition is poured into a spinning mould and allowed at least substantially wholly to solidify followed by pouring of metal of core composition into the mould and allowing the core metal then to solidify.

Bimetallic Electroslag Remelting; in which a water cooled mould is disposed concentrically around a solid metal arbour made of metal of core composition and at least one consumable electrode of shell composition is inserted into the gap between the core and the mould and melted by electroslag remelting and the mould moved upwardly as the shell metal solidifies at the bottom of the mould.If desired, the arbour and the mould may be rotated.

Continuous Pouring of Cladding; in which a water cooled mould is disposed concentrically around a sold metal arbour of core composition and liquid metal of shell composition is poured into the gap between the mould and the arbour and the mould is moved upwardly as the shell metal solidifies at the bottom of the mould.

Bimetallic Hot Isostatic Pressing; in which a solid arbour of core composition is coated with metal powder, heated in a furnace and then pressed

using a high isostatic pressure pressing operation such as explosive forming. The metal powder is normally produced as an extruded cylindrical shell and slid onto the arbour prior to the heating and hot isostatic pressing operation.

Weld Cladding; in which a submerged arc welding operation is performed to weld a wire of shell composition spirally around an arbour of core composition whilst rotating the arbour.

Bimetallic rolls, whether made by one of the methods described hereinbefore or by any other method all have a distinct chemical interface between the core metal and the shell metal and a consequent distinct physical interface. Depending upon the manufacturing technique and the relative compositions of the shell and core metal, there may be a transition zone, of the order of 2mm to 3mm radial width, for example, where there is a steep composition gradient.

In addition, monoblock rolls have been heat treated to improve the surface properties of the roll, for example by heating to austenitising temperature and then quenching to produce a martensitic/bainitic shell which is then tempered and a pearlitic/ferritic core. Again, because of the nature of the heat treatment operation, there is a sharp interface between the shell and the core although there may be a transition zone of relatively steep physical gradient of relatively short radial width, for example, of the order of 5mm to 10mm.

All such rolls have been found to suffer from spalling, that is to say the loss of one or more pieces from the surface layer of the roll. Roll surface damage can induce microcracks that subsequently propogate by fatigue, resulting in a surface spall. Sub-surface damage is known to generate fatigue cracking at the shell metal-core interface regions where fatigue properties are reduced, which cracks propagate and cause the loss of pieces of metal from the surface layer.

An object of the first more specific facet of the present invention is to provide a roll for use in rolling mills whereby the above mentioned problems of spalling is overcome or is reduced. A further object of the first more specific facet of the invention is to provide a method of making such an improved roll.

According to a first aspect of the first more specific aspect of the invention, we provide a method of making a roll for use in a rolling mill comprising performing an electroslag remelting operation on a consumable electrode which comprises an inner part of a core composition and an outer part of a shell composition.

The electroslag remelting operation may produce an ingot comprising an inner part of a first composition and a surface part of a second composition, the second composition being different to the first composition and the metal of the roll between the inner part and the surface layer having a composition which changes without discontinuity.

The consumable electrode may comprise an inner body provided with an external cladding.

The external cladding may comprise a plurality of discrete elements, which may be elongate elements, such as slats, fixed by welding or otherwise to an external surface of the inner body.

The discrete elements may extend longitudinally of the inner body or may be wound helically around the body or may extend circumferentially around the body or may be in the form of a loop or part loop which lies in a plane which is inclined to the longitudinal axis of the body at less than a right angle.

Alternatively, the external cladding may comprise a sleeve which may be fixed, by welding or otherwise, to the external surface of the body.

The discrete elements or the sleeve may be solid and may be of constant composition throughout their cross-section and length, or they may vary in composition across their cross-section and/or along their length.

Where discrete elements are provided they may be provided in different numbers at different positions along the length of the body.

Alternatively, the discrete elements or the sleeve may comprise a composite element of different composition at different positions across the cross- section of the element and/or along the length of the element.

Each discrete element or the sleeve may comprise a hollow element containing material of different composition to the material of the hollow element.

Where the cladding is a discrete element each discrete element may comprise a tube to provide said hollow element.

The material inside the or each hollow element may be a powder.

Where the additional ingredient comprises powder means may be provided to release the powder in controlled manner. For example by reducing the cross-section of a hollow component in which the powder is contained to divide the powder into discrete amounts for release into the melt pool sequentially as the component and electrode element are melted.

The powder may have a mesh size of less than 3mm, but may have a mesh size up to 5mm and may lie in the range 2 - 5mm.

The powder may comprise at least one or chromium, tungsten, titanium, molybdenum, niobium and vanadium combined with carbon as an inter- metallic compound or compounds or as a solid solution or solutions.

Each discrete hollow element or hollow sleeve may comprise steel, such as mild steel or stainless steel.

Alternatively, the external cladding may comprise a shell part of a bimetallic rolling mill roll the core of which provides the inner body. The bimetallic rolling mill roll may have been made in any desired manner including, for example, Centrifugal Casting, Bimetallic Electroslag Remelting, Continuous Pouring of a Cladding, and Bimetallic Hot Isostatic Pressing or Weld Cladding.

Where the cladding comprises discrete elements, at least one of the elements may be of different composition to the other elements.

The inner body may be of constant composition throughout its cross- section or it may be of varying composition across at least part of its cross-section.

Further alternatively, the consumable electrode may comprise a body having an inner part having a first composition and an outer part having a second composition, which is different from the first composition and the metal of the

body between the inner part and the outer part having a composition which changes from the first composition to the second composition without discontinuity.

The consumable electrode may comprise a scrap roll according to the first aspect of the invention.

The electroslag remelting operation may be performed at a relatively slow rate or a relatively high rate.

These variable rates may be achieved by using a relatively high voltage and a relatively low current, or vice versa.

The slag may have a relatively high melting point.

The ratio of the diameter of the electrode to the diameter of the mould cavity may be relatively high, for example, greater than 0.8.

The melting point of the outer part of the consumable electrode may be higher than the melting point of the inner part of the consumable electrode and the melting point of the inner part may be higher than the melting point of the slag.

In this case, as the consumable electrode is melted, droplets of metal from the outer part of the consumable electrode because of their spatial and temperature proximity to the mould wall, solidify relatively rapidly at or adjacent to the mould wall so that the ingot is provided with a surface part of a composition similar to that of the outer part of the consumable electrode.

Alternatively, the melting point of the inner part of the consumable electrode may be higher than the melting point of the outer part of the consumable electrode and the melting point of the outer part may be higher than the melting point of the slag.

In this case, as the consumable electrode is melted, the droplets of metal from the outer part of the consumable electrode because of their spatial proximity to the wall of the mould, solidify relatively rapidly at or adjacent to the mould wall so that the ingot is provided with a surface part of a composition similar to that of the outer part of the consumable electrode.

In each case, because the droplets of metal from the outer part of the electrode are made of a metal of higher alloy content than the metal of the inner part of the electrode, they are denser and they drop relatively rapidly and directly downwardly from the outer part of the electrode and hence are disposed adjacent to the mould wall and hence they solidify relatively rapidly to provide the ingot with a surface layer having a composition similar to that of the outer part of the consumable electrode.

Where the external cladding is hollow and contains an alloy and/or alloy-carbide, the droplets of alloy and/or alloy-carbide, if molten, from the hollow elements, because of their spatial and temperature proximity to the mould wall, solidify relatively rapidly at or adjacent to the mould wall whereby the region of the re-melted ingot adjacent the mould wall has a relatively high and uniform distribution of alloy and/or alloy carbide.

At or adjacent to the centre of the ingot there may be a relatively low uniform distribution of alloy and/or alloy carbide.

The melting point of the alloy and of the alloy carbides may be greater than the melting point of the slag and of the inner part of the consumable electrode.

The outer part of the ingot may be hardened by at least one of a dispersion hardening effect of the alloy carbides, phase transformation, or cell size transformation.

Alternatively, the outer part of the rolled ingot may be hardened, additionally, or alternatively, by a secondary hardening heat treatment operation.

Preferably, the electroslag remelting operation is performed so that the floor of the molten bath is relatively flat. This may be achieved by inter-relating the voltage, current and the slag composition of the electroslag remelting operation.

Alternatively or in addition, the cooling rate of the mould may be adjusted as may be the rate of movement of the mould.

The additional ingredient may comprise a high alloy steel or cast iron.

The additional ingredient may comprise a powder comprising at least one of chromium, tungsten, titanium, molybdenum, niobium and vanadium combined with carbon as an inter-metallic compound or compounds or as a solid solution or solutions or ferro-titanium.

The additional ingredient may comprise a powder having a mesh size of less than 5mm.

A second more specific facet of the invention relates to an engineering ferrous metal product and to a method of making such a product. More particularly, but not exclusively, the product comprises a roll for use in rolling mills such as roughing work rolls of a hot strip rolling mill for rolling steel strip or sheet, or a calendering mill for paper.

In strip or sheet metal rolling, requirements for better gauge accuracy, surface properties and productivity have led to the development of high performance rolls. Such rolls are required to have high resistance to rolling stress and reduce damage to the roll surface such as surface wear, cracking and spalling. Similar properties are required for rolls for other applications and for other products.

An object of the second more specific facet of the present invention is to provide an engineering ferrous metal product having a relatively hard and tough surface layer. A further object of the second more specific facet of the invention is to provide a method of making such a product.

According to a first aspect of the second more specific facet of the invention we provide a method of making an engineering ferrous metal product comprising performing an electroslag remelting operation on a consumable electrode element comprising engineering ferrous metal and in which an addition to form sub-microscopic carbides, nitrides and/or carbo-nitrides is made to the melt pool.

The second more specific facet of the invention may be provided together with or without the first more specific facet of the invention.

In the second facet:-

Preferably the addition forms sub-microscopic titanium and/or niobium carbides, nitrides and/or carbo-nitrides.

The additions may comprise at least one of titanium carbide, niobium carbide, mixed carbides of titanium and tungsten, mixed carbides of any of the above mentioned carbides or ferro titanium.

For example, the carbide may be a mixed tungsten titanium carbide of the kind (TiW)C where the ratio of titanium to tungsten is about 1:1 by weight.

The alloy carbide may comprise Ti and W in the ratio range:

1 : 1 to 1 : 1.17

The alloy carbide may contain nitrogen, for example up to about 0.1% nitrogen.

The alloy carbide preferably comprises:-

30 - 42 -Ti

35 - 49%-W 9%-C and usual residuals

The alloy carbide particles preferably have a density which matches that of the engineering ferrous metal.

By "matches" we mean a density preferably lying in the range 6 - 8 gms per cc. This is to be compared with a typical density of 7 gms per cc. for cast iron and steel. More preferably, the alloy carbide particles have a density of ± 5% of the density of the engineering ferrous metal to which they are added.

The alloy carbide particles have a very low co-efficient of thermal expansion compared to the co-efficient of thermal expansion of engineering ferrous metals. Accordingly, if relatively large alloy carbide particles were present in the engineering ferrous metals this would give rise to high stresses on cooling, and leading to thermal fatigue.

The alloy carbide particles preferably have a maximum dimension of up to 10 microns and preferably 0.1 - 5 microns and more preferably 0.1 - 2 microns.

The amount of alloy carbide particles added is such as to achieve up to 20% by volume of alloy carbide particles in the solid metal. Generally, the alloy carbide content may be in the range 0.1 to 20% by volume, and may be in the range 1 to 20% and more preferably 3 to 10% when a hardening effect based on the law of mixtures is provided. However, the alloy carbide content may be lower, e.g. down, for example from about 1%, to about 0.5% or 0.1% or less when a hardening effect based on a modification of the transformation of the microstructure is provided.

The engineering ferrous metals are preferably steel or cast iron having a carbon content lying in the range 0.3 - 3.8%.

The engineering ferrous metals may contain chromium and may have a chromium content which is greater than or equal to 1%.

The engineering ferrous metal and/or the additions may contain nitrogen, for example up to 0.1% nitrogen in total in the as cast product.

The solid carbide particles may be coated with a metal which allows wetting to occur between the particles and the liquid engineering ferrous metal. By "wetting" we mean the ability of the liquid engineering ferrous metal to wet the coating metal. More particularly, for example, where the interfacial tension between the liquid engineering ferrous metal and the solid coating metal is such that the contact angle therebetween is 0° - 90°C.

The wettability of the coated particles and the density of the alloy carbide particles each promote a uniform distribution of the carbide particles in the liquid engineering ferrous metal which is retained when the metal solidifies.

By a "uniform distribution" we mean an even distribution throughout the section of a casting made of the engineering ferrous metal with no significant segregation. The solid carbide particles are not orientated in any direction and are distributed across all phases of the micro-structure.

The coating metal preferably comprises iron or an iron carbon alloy, but may comprise an alloy of two or more elements selected from the group

comprising iron, nickel, copper, titanium and carbon, or may be nickel or copper and usual incidentals.

The coating metal may comprise nitrogen, for example up to about 0.1% nitrogen.

If the coating is iron, then because iron has a higher melting point than the engineering ferrous metal to which it is to be added, which would inhibit wettability, it is preferred to add an appropriate amount of at least one alloying element such as carbon, nickel, copper or titanium to the iron to produce an alloy having a melting point which matches the operating temperature of the ferrous metal. By "matches" we mean that the melting point of the coating and said operating temperature are preferably within approximately 20 - 30°C of each other.

By "operating temperature" we mean the temperature of the engineering ferrous metal whilst the coated carbide particles are added.

The iron coating may contain up to 3.5% carbon.

The additions to provide the sub-microscopic carbide, nitride or carbide-nitride particles may be made into the melt pool separately from the electrode.

In this second facet of the invention, the electroslag remelting operation may be performed so as to achieve or aim to achieve a uniform composition throughout the cross-section of the ingot.

Where the coating metal, and/or the alloy carbide and/or the engineering ferrous metal contains nitrogen it is considered that titanium carbo- nitrides may be precipitated in the microstructure of the engineering ferrous metal. The total nitrogen content of all the components may be limited to about 0.1% nitrogen.

According to a second aspect of the invention we provide an engineering ferrous metal product when made by performing an electroslag remelting operation on a consumable electrode element comprising engineering ferrous metal in which an ingredient, additional to said consumable electrode element, was added to the melt pool.

According to a second aspect of the first more specific facet of the invention, we provide a roll for use in a rolling mill comprising an inner part

having a first composition and a surface part having a second composition, which is different from the first composition and the metal of the roll between the inner pan and the surface part having a composition which changes from the first composition to the second composition without discontinuity.

The composition may change continuously between the surface part and the inner part and the change in composition may be relatively uniform.

The roll may have a composition gradient between the surface part and the inner part of less than 40% per 100mm in the radial direction at any position along the longitudinal extent of the roll.

The roll may have a relatively uniform transition region between the surface part and the inner part and the roll may have substantially the same elastic modulus throughout its cross-section.

The total alloying elements may reduce from about 15% in the surface pan to less than 3% at the middle of the inner part.

The carbon content may be less than 0.9 throughout and the ratio of molybdenum to carbon in the surface part may be greater than 2.0 such that M7C3 type carbides are formed. The presence of such M7C3 carbide permits secondary hardening of the roll to be achieved, with suitable heat treatment.

The surface part may comprise carbides of at least one of chromium, tungsten, titanium, molybdenum, niobium and vanadium combined with carbon as an inter-metallic compound or compounds and/or as a solid solution or solutions.

The surface part may have a relatively high distribution of alloy and/or alloy carbides.

The roll may have a relatively uniform transition region between the surface part and the inner part.

The roll may have the same elastic modulus throughout its cross- section.

Alternatively, the roll may have an elastic modulus that increases in proportion to the amount of the alloy and/or alloy carbide through the roll cross- section.

According to a second aspect of the second more specific facet of the invention we provide an as-cast engineering ferrous metal product, at least a surface region of which has a microstructure comprising feathery or a sheath-like bainite with sub-microscopic carbide, nitride and/or carbo-nitride particles distributed in the bainite and/or bainitic ferrite matrix.

Preferably, the sub-microscopic particles are of titanium and/or niobium carbide and/or carbo-nitride.

By "electroslag remelting" we mean a process in which a consumable electrode is melted beneath an electrical conductive slag to form a melt pool in a moving mould from the lower end of which a continuously cast ingot emerges; the electrode is heated by thermal conduction from the slag and the slag is heated by electrical conduction from the electrode through the slag and melt pool to a counter-electrode provided by the ingot.

By "engineering ferrous metals" we mean cast iron and steel.

By "sub-microscopic" we mean a particle size of less than 1 micron.

Two examples of the invention will now be described in more detail with reference to the accompanying drawings, wherein:-

FIGURE 1 is a diagrammatic side elevation of an electroslag remelting apparatus for use in a method embodying the present invention;

FIGURE 2 is a transverse cross-section through a consumable electrode for use in the apparatus of Figure 1;

FIGURE 3 is a diagrammatic fragmentary cross-section through the apparatus of Figure 1 when in use;

FIGURE 4 is a graph showing the variation in composition in a transverse section of roll made according to the present invention;

FIGURES 5a and 5b are graphs showing the variation in composition in a transverse section of prior art bimetallic rolls made by hot isostatic pressing and centrifugal casting, respectively,

FIGURE 6 is a diagrammatic side elevation of another embodiment of an electroslag remelting apparatus when in use in a method embodying the present invention, and

FIGURE 7 is a transverse cross-section through a consumable electrode for use in the apparatus of Figure 6.

FIGURE 8 is a photomicrograph of an outer part of a roll embodying the invention at a magnification of x 100, and

FIGURE 9 is a photomicrograph of an outer part of a roll of a similar base composition to the roll shown in Figure 8 but not embodying the invention at a magnification of x 100.

Referring to Figures 1 to 5b, an electroslag remelting apparatus is shown in Figure 1 and is of essentially conventional kind comprising a cylindrical water cooled mould 10 which is movable vertically upwardly or downwardly. An electrode holder 11 holds, for example by being welded thereto, the upper end of a consumable electrode 12. Initially, the bottom end of the consumable electrode is immersed in molten slag contained between a bottom plate 13 and the wall of the mould 10. Electric current is then passed to cause the lower end of the electrode to melt and as the droplets of metal fall from the lower end of the electrode 12 they pass through the slag and are refined and then solidify to form an ingot 14.

The basic operation of the electroslag remelting operation is conventional as is the plant and hence further discussion is not necessary.

In accordance with this embodiment, the consumable electrode 12 comprises an electrode element which comprises a first part of the electrode and comprises an inner body 20 of cylindrical configuration having welded to its external surface 21 a second, outer, part of the electrode 12 which comprises a cladding 22 which comprises a plurality of discrete elements 23 in the form of

slats and which provides an ingredient, additional to the material of the electrode element 20, which is added to the melt pool during the electroslag remelting operation.

The inner body 20 is made of low alloy steel, cast iron or mild steel such as 0.2%C steel, whilst the cladding is made of high alloy steel or cast iron or material of any desired suitable composition.

For example, the cladding may have a composition lying in the range:-

Carbon 0.5% - 3%

Aluminium 0.0% - 1%

Molybdenum 0.5% - 5%

Vanadium 0.0%) - 12%

Chromium 0.5% - 15%

Tungsten 0.0% - 8%

Silicon 0.0% - 3%

Titanium 0.0% - 8%

Nickel 0.0% - 5%

Iron and usual residuals - balance.

Preferably, the cladding has a composition lying in the range:-

Carbon 0.5% - 3%

Aluminium 0.0% - 0.1%

Molybdenum 0.5% - 5.0%

Vanadium 0.1% - 12%

Chromium 0.5% - 15%

Tungsten 0.1% - 8%

Silicon 0.0%? - 3%

Titanium 0.1% - 8%

Nickel 0.1% - 5%

Iron and usual residuals - balance.

In the present example, the slats have the following composition:

Carbon 1%

Iron and usual residuals - Balance

The inner body may comprise a low alloy steel which may comprise:

Carbon 0.4%

Chromium 1%

Iron and usual residuals - Balance

If desired, the cladding may comprise other relatively high alloy engineering ferrous metal such as:-

Carbon 3%

Molybdenum 5%

Vanadium 8%

Nickel 2%

Iron and usual residuals - balance Carbon 1%

Vanadium 12%

Iron and usual residuals - balance Carbon 2%

Tungsten 5%

Titanium 5%

Iron and usual residuals - balance Carbon 3%

Silicon 2%

Iron and usual residuals - balance

If desired, the inner body may comprise other relatively low alloy ferrous engineering metal such as:-

Carbon 0.2%

Chromium 0.5%

Iron and usual residuals Balance

or

Carbon 3%

Silicon 2%

Iron and usual residuals Balance or plain cast iron or mild steel.

Although the case of cladding in the form of discrete slats has been described and illustrated in Figure 2, if desired, the cladding may comprise a sleeve, for example, a cylindrical sleeve mounted, for example, by welding or by virtue of being a push fit, on an inner cylindrical body.

Further alternatively, the consumable electrode may comprise a bimetallic body, that is to say a body having a core and a shell bonded together by, for example, Centrifugal Casting, Bimetallic Electroslag Remelting, Continuous Pouring of Cladding, Weld Cladding or in any other way and the bimetallic body may comprise a scrap bimetallic roll.

Further alternatively, the electrode may comprise a scrap roll made in accordance with the present invention may comprise another body comprising an inner part having a first composition and an outer part having a second composition, the second composition being different from the first composition and the metal of the body between the inner part and the surface layer having a composition which changes without discontinuity.

In a still further alternative, the electrode may comprise an E.S.R. or other spray coated composite electrode.

Referring now to Figure 3, a consumable electrode 12 of the same construction as described and illustrated with reference to Figure 2, comprising an inner part 20 and a cladding 22 of slats 23, is melted in the electroslag remelting apparatus shown in Figure 1.

As shown in Figure 3, the lower end of the disposable electrode 12 is immersed in a bath 30 of slag and beneath the slag bath 30 is a bath 31 of molten metal from which the metal solidifies in normal dendritic structure as shown at 32 to form an ingot 33.

In the present example the metal of the cladding 22 has a melting point which is 40°C below the melting point of the inner part 20 and 60°C above the melting point of the slag 30. The melting point of the cladding 22 is, in the present example, 1530°C.

The inner part may have a melting point lying in the range 1160°C to 1600°C and the cladding may have a melting point lying in the range 1160°C to 1600°C.

The melting point of the inner and outer parts may differ by 20°C to 60°C.

The slag may have a melting point which differs from the lower of the melting point of the inner part and the cladding by 20°C to 60°C.

The mould is water cooled to a temperature lying in the range of 15°C to 65°C and the molten metal pool 31 has a temperature lying in the range 1400°C to 1600°C may be in the range 1160°C to 1600°C.

As the electrode 12 melts, droplets of liquid metal from the cladding fall generally vertically downwardly and because of the relatively close proximity to the wall of the mould 10 and the relative low melting point of the metal from the cladding it is cooled and begins to solidify at and adjacent to the wall of the mould 10 so that the metal at the surface of the ingot 33 has a composition substantially similar to the composition of the metal of the cladding.

The metal droplets falling from the middle of the bottom of the inner part fall vertically downwardly and thus the metal solidifying at the centre of the ingot has a composition substantially similar to the composition of the inner part of the electrode.

Between these two extremes, there is limited mixing of the metal from the two regions so that a change in composition between the two extremes takes place without discontinuity. As shown in Figure 4, which plots the composition of carbon and chromium as typical elements of the composition. It will be seen that the composition changes substantially uniformly and continuously between the centre of the inner part and the surface and, in particular, there is no sharp discontinuitv.

The composition at the surface of the ingot is:-

Carbon 0.8 %

Molybdenum 2.2 %

Vanadium 0.5 %

Chromium 12.0 %

Iron and usual residuals - Balance

The composition at the centre of the ingot is:-

Carbon 0.5%

Molybdenum 0.3%

Vanadium 0.2%

Chromium 2.0%

Iron and usual residuals - balance.

It will also be seen that the composition gradient does not exceed 40% per 100mm in the radial direction and this condition applies at any position along the longitudinal extent of the roll although only one such position is illustrated in Figure 4.

In contrast, in previously known bimetallic rolls there is a sharp discontinuity as illustrated in Figures 5a. and 5b. for a hot isostatic pressing roll and a centrifugally cast bimetallic roll respectively.

The resultant ingot has a uniform transition between the outer layer and the inner part and has substantially the same elastic modulus throughout its cross- section with the absence of any definite boundary or discontinuity between parts of different composition.

The total alloying elements reduced from about 15 wt.% in the surface part to less than 3% at the middle of the inner part.

The carbon content is less than 0.9 throughout and the molybdenum to carbon ratio in the surface part is greater than 2.0 such that M7C3 carbides are formed in the ingot to permit of secondary hardening of a roll made from the ingot when suitably heat treated in known manner.

The resultant ingot has uniform axial chemical and physical properties, has superior oxidation, wear and mechanical properties compared with known rolls which all suffer from the interface problems described hereinbefore.

The voltage and current used in the electroslag remelting operation are manipulated so that there is a relatively high voltage and a relatively low current, or vice versa and in addition the composition of the slag is adjusted so as to provide a relatively viscous slag. As a result turbulence in the slag and the melt pool are minimised so that the desired composition gradient explained above is achieved.

In addition, the voltage and current used are manipulated so that the floor of the melt pool is relatively flat.

The metal droplets leaving the cladding of the electrode are more dense than the metal droplets leaving the inner body of the electrode because they are more highly alloyed. This results in the droplets from the cladding of the electrode falling downwardly adjacent to the wall of the mould and, because the floor of the melt pool is relatively flat, there is relatively little tendency for these droplets to run towards the middle of the melt pool before they have solidified as a result of their spatial and temperature proximity to the mould wall.

The composition of the slag is adjusted to reduce the ionic capacity of the slag by adjusting the balance of the silicon, calcium and aluminium in the slag to reduce the tendency for electromagnetic stirring as well as the composition of the slag being adjusted to provide a relatively higher viscosity.

Typically the slag has the following composition:- 33V3% CaO, 33V3% CaF ₂ , 33Vs% A1,0 ₃ .

However the slag may have a composition, for example, lying in the range to 20% CaO, 80%o CaF ₂ , 0% A1 ₂ 0 ₃ or 45% CaF, 45% A1 ₂ 0 ₃ , 10% SiO ₂ .

The cooling rate provided by the water cooling to the mould may be adjusted by adjusting the raw temperature of the mould to lie in the range 15°C to 65°C. In addition, the rate of movement of the mould may also be adjusted.

Typically, the electrode has a diameter which is about 0.9% of the diameter of the mould cavity.

The electrode is arranged so that approximately 20%, by weight, of the electrode comprises cladding and the balance is provided by the inner body. This ratio may lie in the range 10 to 40%.

Although in the above described example the melting point of the cladding has been described as being lower than the melting point of the metal of the inner body, if desired this may be reversed. In such a case metal droplets falling from the cladding of the electrode will still solidify out adjacent to the mould wall without significant mixing but other metal from the inner body, because of the greater density of the metal from the cladding, causes the droplets to fall downwardly adjacent to the mould wall together with the hereinbefore mentioned precautions to avoid mixing of low slag viscosity, low slag electromagnetic stirring and relatively flat melt pool base.

After the ingot has been produced by the above described electroslag remelting operation, appropriately shaped neck portions are attached to the ingot and the ingot is machined to produce a rolling mill roll. The rolling mill roll may be used for any desired purpose depending upon the selected composition but in the present example it is intended as a roughing work roll for the reducing stands of a hot strip rolling mill for rolling steel strip. Other typical applications are rolls for a cold rolling mill or back-up rolls for hot or cold rolling mills.

The modulus of elasticity of the ingot and hence of the resultant rolling mill roll varies in accordance with the composition change across the cross-section of the ingot but there is a relatively uniform change of modulus across the cross- section substantially proportional to the change of composition equivalent to less than 10% per 100mm of cross-section.

Referring now to Figures 6 and 7, a second embodiment is illustrated in which the electroslag remelting apparatus is the same as shown in Figure 1 and the same reference numerals have been used to refer to the corresponding parts.

In accordance with the second embodiment, the consumable electrode 12 comprises an inner body 120 having welded to its external surface 121 a hollow cladding 122 which comprises a plurality of discrete tubular elements 123 in the form of tubes 124 containing powder 125. The tubes and the powder provide an ingredient, additional to the tube of the inner body 120, which is added to the melt pool during electroslag remelting.

The inner body 120 is made of low alloy steel, cast iron or mild steel such as 0.2% carbon steel or any other suitable engineering ferrous metal.

The tubes 124 are made of mild steel in the present example but may be made of stainless steel and, if desired, the tubes may be made of high alloy steel or cast iron or indeed in any material of the composition described for the slats of the first embodiment.

However, the tubes may be provided solely to hold the powder or other alloying addition and so could be composed of material of the same composition as the inner body or indeed may be made of suitable non-metallic material, such as silica, of a melting point so as to melt in the bath at a desired rate to release the alloying addition. The above described materials are applicable to hollow cladding of any suitable configuration.

In the present example the tubes have the following composition:

Carbon 0.2%

Iron and usual residuals - balance

The powder has a mesh size of less than 3mm, but may have a mesh size up to 5mm and may lie in the range 2 - 5mm.

The powder may comprise any desired alloying additions such as one or more elements, intermetallic compound, or alloy.

For example the powder may comprise 0-100% of any one or more of the following elements:-

Carbon

Molybdenum

Chromium

Tungsten

Titanium

Vanadium

Niobium

If desired the powder may be made of a powdered alloy of any one of the compositions or range of compositions described as being suitable for the cladding in connection with the first embodiment or may be made of alloy of other compositions or of components which provide such compositions.

The powder may comprise alternatively or in addition metallic carbide(s) such as carbides of at least one of chromium, molybdenum, tungsten, titanium and vanadium and niobium.

In another example, the powder may comprise:-

Some of the carbon is combined with the Tungsten and the Titanium as intermetallic compound but the Tungsten and Titanium are combined as a solid solution and a proportion of the Carbon present is as free graphite.

In the present example, the powder has the following composition:

Alternatively, the powder mixture has the following composition:- Chromium 2%

One or more of the above mentioned alloying elements may be present in the powder, either combined with the carbon as an inter-metallic compound or compounds, or as a solid solution or solid solutions.

The powder may comprise solid carbide particles coated with metal which allows wetting to occur between the particles and the metal of the melt pool.

The coating metal is preferably iron or iron carbon alloy or may be nickel or copper.

If the coating is iron, because iron has a higher melting point than the melt pool metal to which it is to be added, which would inhibit wettability, it is preferred to add an appropriate amount of carbon or other alloying element such as nickel to the iron to produce an alloy having a melting point which matches that of the melt pool metal. By "matches" we mean that the melting points are preferably within approximately 20°C-30°C of each other.

Typically, the iron coating contains 3-3.5% carbon.

Alternatively, the particles may be coated with iron or with an iron carbon alloy having a lower carbon content where the particles dwell in the melt sufficiently long for the carbon to diffuse into the iron of the coating and so produce a composition which has a melting point which matches that of the metal of the melt pool.

The coated carbide particles preferably have a density which matches that of the melt pool. By "matches" we mean a density preferably lying in the range 6-8 grams per cc.

The carbides are preferably selected from the group comprising chromium, molybdenum, titanium, tungsten, niobium, vanadium or are mixed carbides thereof such as Cr7C3, (CrMo)7C3 and/or carbo-nitrides or mixed carbo-nitrides.

The carbides preferably have a composition so that the above mentioned matching density is achieved.

The amount of carbide particles added is such as to achieve up to 0.5 - 20% by weight of carbide particles in the surface part.

The tubes 124 in the present example are 16mm OD, 13mm ID but may lie in the range from 6mm ID to 28mm ID with appropriate OD.

The tubes are provided with restrictions at intervals along their length so as to hold discrete amounts of powder at positions along the length of the tubes, for example, such restrictions may be provided every 10cm.

In the present example the restrictions are provided by crimping the tubes so as to close or substantially to close their bore but, of course, such restrictions may be provided in any other desired way.

In the present example the tubes 124 are attached to the inner body 120 so as to extend longitudinally thereof parallel to the longitudinal axis of the inner body but, if desired, a single tube or tubes may be wound helically around the inner body or one or more tubes may extend circumferentially around the body so as to lie in a plane which is perpendicular to the longitudinal axis of the inner body so long as the tubes are of the correct size to hold the correct amount of powder to be released into the melt at the longitudinal position of the body.

Further alternatively, one or more tubes may be arranged to extend around the inner body in a plane or respective plane which is inclined to the longitudinal axis of the inner body at less than a right angle.

Although in this case cladding is in the form of discrete elements provided by the tubes 124 containing powder 125, if desired, the cladding may comprise a hollow sleeve, for example, a sleeve comprising inner and outer generally cylindrical walls inter-connected by generally annular shaped walls with the space between the walls containing powder such as that described hereinbefore and the

sleeve being mounted, for example, by welding or by virtue of being a push fit on the inner body. In this case the hollow sleeve is provided with restrictions in its wall at suitable positions along the length of the sleeve so as to divide the powder into discrete amounts for release into the melt pool sequentially as the sleeve and inner body are melted.

The consumable electrode 12 is melted in the electroslag remelting apparatus shown in Figure 3 and as in the case of the first embodiment, the lower end of the electrode 12 is immersed in a bath 30 of slag and beneath the slag bath 30 is a bath 31 of molten metal from which the metal solidifies in normal dendritic structure, as shown at 32 to form an ingot 14.

In the present example the melting point of the additional ingredient, for example Tungsten, Titanium solid solution which can at least partly melt in the melt bath and the alloy carbides is greater than the melting point of the slag 30 and of the inner body 120 of the consumable electrode 12. The melting point of the slag, the inner body and cladding tube/sleeve may be related as described in connection with the first embodiment and the mould may be similarly water cooled whilst the molten melt pool may have a temperature lying in a similar range to the first embodiment.

As the electrode 12 melts, droplets of liquid metal and metal carbide from the tubes 124 and powder 125 fall generally vertically downwardly and because of the relatively close spacial and temperature proximity to the wall of the mould 110 the droplets solidify relatively rapidly at or adjacent to the mould wall so that the metal at the surface of the ingot 14 has a relatively high uniform distribution of alloy and alloy carbides as well as having a composition of matrix substantially similar to the composition of the metal of the tubes 124. Where the carbide is solid it is trapped by the solidifying metal adjacent the mould wall.

As in the case of the first embodiment, the metal droplets falling from the middle of the bottom of the inner part fall vertically downwardly and thus the metal solidifying at the centre of the ingot has a composition substantially similar to the composition of the inner part of the electrode.

Between these two extremes there is limited mixing of the material from the two regions so that a change in composition between the two extremes takes place without discontinuity. Accordingly, the ingot has a relatively uniform transition region between the inner part and the surface part. The ingot, and hence any resulting product such as a roll as described hereinbefore, has an elastic modulus which increases in direct proportion to the amount of the distribution of the alloy/alloy carbide through the ingot cross-section.

In the present example the composition at the surface of the ingot is:

Carbon 0.5%

Molybdenum 0.2%

Vanadium 0.9%

Tungsten 1.2%

Chromium 0.5%

Titanium 0.15%

Iron carbide and usual residuals balance

The composition at the centre of the ingot is:

Carbon 0.5 %

Molybdenum 0.1%

Tungsten 0.3%

Titanium 0.03%

Vanadium 0.2%

Chromium 0.5%

Iron carbide and usual residuals balance

The variation in composition is believed to arise due to the relatively higher diffusion of chromium and carbon compared with the relatively lower diffusion and more dense tungsten and titanium.

As in the case of the first embodiment, the composition gradient does not exceed 40% per 100mm in the radial direction and this condition applies with any position along the longitudinal extent of the roll.

Other features of the method and apparatus of this embodiment are similar to those of the first described embodiment and hence further discussion is not required.

A third embodiment will now be described, which is a modification of the second embodiment. The structure and composition of the consumable electrode and the hollow elements are as described in connection with the second embodiment except that the inner body 120, or the tubular elements 123 or the additional ingredient, hereinafter to be described, individually or together, contain nitrogen in an amount or amounts to provide a total of 0.001 to 0.1% nitrogen and, in addition, may provide a total of 0.001 to 0.5% niobium. If desired, such amounts of nitrogen and niobium may be provided in the first and second embodiments.

In this embodiment the powder in the tubular elements comprises at least one of, titanium carbide, a mixed carbide of titanium and at least one other element, such as a mixed carbide of titanium and tungsten such as previously described, niobium carbide, or mixed carbides of any of the above mentioned carbides, and/or ferro-titanium. Such additions may also be provided in the second embodiment if desired.

The particles of the additional ingredient powder have a mesh size lying in the range up to 5 mm.

The additional ingredient may include other components such as those described in connection with the second embodiment.

The amount of such powder lies in the range 0.5% - 2% by weight.

In this embodiment the ESR operation may be performed in a manner in which mixing occurs in the melt pool so that there is little or no variation in composition across the cross-section of the resultant ingot.

Where the addition comprises titanium carbide or a mixed carbide of titanium and other elements such as tungsten, as the carbide particles fall from the ends of the tube through the slag into the melt pool they are heated and caused to wholly or substantially go into solution in the metal in the melt pool.

As the metal in the melt pool is cooled and approaches the relatively rapidly upwardly advancing solidification face, titanium carbo-nitrides (TiCN) and/or titanium nitride sub-microscopic particles are precipitated in the interdendritic pools occurring on solidification. The titanium carbo-nitride typically contains about 90% titanium and less than 10% carbon and nitrogen. The tungsten carbide, when present in the initial carbide, is believed to go into solid solution with the matrix metal, but some precipitation of very fine sub-microscopic particles of tungsten carbide or tungsten oxy-carbide may take place.

Where the addition comprises ferro-titanium, the ferro-titanium particles go into solution in the melt pool and there are precipitated sub-microscopic titanium carbo-nitride particles and/or titanium nitride particles and/or titanium carbide particles are precipitated, again in the interdendritic pools. In each case larger, microscopic, particles of a respective carbide or carbides may also be precipitated and/or some of the added ingredient particles may not go into solution, but so long as they are sufficient titatium containing sub-microscopic particles the effects on microstructure hereinafter to be described are obtained.

By performing the ESR operation with appropriate parameters and by using additions as described above in connection with the third embodiment, together with appropriate amounts of additions, a perturbation of the as-cast microstructure, particularly in the surface region of the ingot, compared with that which normally occurs is achieved. A feathery or sheath-like upper bainitic structure or bainitic ferrite, as shown in Figure 8, is obtained instead of the conventional microstructure, as shown in Figure 9 which comprises pro-eutectoid ferrite at original austenite grain boundaries with a matrix of lamellar pearlite.

It is considered that the titanium of the sub-microscopic titanium carbo- nitride or other titanium or niobium containing sub-microscopic particles which are precipitated during the ESR operation pin the austenite grain boundaries, which interferes with the normal pearlite precipitation to give the above described bainitic/ferritic structure.

Of course, where appropriate additions to produce such sub-microscopic titanium carbo-nitride etc. particles are made in the second or first embodiments, then a similar microstructure is obtained, although in such cases the effect will be concentrated at the surface due to the composition gradient.

If desired, additional ingredient to produce such sub-microscopic particles of titanium carbo-nitride etc. may be added to the melt pool in other ways to that described hereinbefore. For example, by adding particles in any desired manner, for example through a lance or in discrete amounts from a suitable dispensing means. Alternatively, the particles may be contained in a hollow container separate from the electrode and which is fed into the melt pool as a desired rate.

Further alternatively the appropriate ingredient may be fed into the melt pool in other than powder form either in the form of slats or other solid elements fixed to the electrode element or separately therefrom. Example

In this example a consumable electrode was melted as described previously. The consumable electrode comprises an inner body part with tubes containing powder mounted thereon.

The body part has the following composition:-

The tubes have the following composition:-

Carbon 0.2%

Iron and usual residuals balance

The powder has the following composition:-

Carbon 10%

Tungsten 36%

The particle has a mesh size of less than 2mm.

The wt.% of powder addition was 1% of the finished ingot.

The ESR operation was carried out under standard operating conditions without any attempt being made to obtain the composition gradient. In fact, some composition gradient was noted, in particular with regard to tungsten and titanium.

The resulting ingot has a microstructure in a surface region as shown in Figure 8 and has the following composition:-

Figure 9 is a microstructure of a similar surface region of another ingot cast in the same way as in the above described example using an electrode of the same composition as the body part described above but without the tubes or powder addition. The resultant ingot, shown in Figure 9, has a composition which corresponds to that of the electrode which was remelted.

ED AX analysis of individual carbide particles shows a strong presence of titanium with only traces of tungsten. This suggests that solution of tungsten from the particles has occurred in the melt. This would retard the pearlite transformation cooling allowing a bainitic structure to develop and thus giving a greater hardness due to the modification of phase transformations in this way.

Moreover, since the low carbon steel of this example contained about 0.1% nitrogen it is considered that at least some of this nitrogen has formed titanium

carbo-nitride particles and hence that at least some of the above mentioned carbide particles comprise such titanium carbo-nitride particles which may also contain tungsten.

In the third embodiment the composite electrode may comprise first and second parts which are not inner and outer parts respectively. For example, the second part may comprise a bore or bores or other hollow or hollows provided within the first part. For example, a central axially extending bore containing powder or solid additional ingredient. Further alternatively, the composite electrode may comprise a plurality of individual components, such as a plurality of rods, slats or the like, of the same composition range as described above for the first part and assembled with a plurality of individual other components, such as tubes or other hollow elements containing powder or solid additional ingredient, or assembled with tubes and/or rods, slats or the like of solid metal and, of the same composition ranges as described above for the additional ingredient.

In this specification the term "usual residuals" includes nitrogen when present.

When nitrogen is present it is typically present up to 0.1% and generally when present, is present in the range 0.001-0.1%.

In this specification compositions are expressed in % by weight.

The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in the terms or means for performing the desired function, or a method or process for attaining the disclosed result, may, separately or in any combination of such features, be utilised for realising the invention in diverse forms thereof.