News

Electroslag Remelting Process

• satyendra
• March 7, 2020
• 4 Comments
• Electroslag remelting, ESR, IESR, ingot casting, Macro segregation, Micro segregation, PESR, slag, VAC-ESR, yield,

Electroslag Remelting Process

Electroslag remelting (ESR) is the abbreviated name given to electroslag consumable electrode remelting and secondary refining process. It is a secondary steelmaking process which is used for remelting and refining of steels and special alloys normally used for critical applications in aircraft, thermal and nuclear power plants, and defence hardware, etc. The main purpose of the ESR process is to control the non-metallic inclusions in the steel, remove segregations and shrinkage, and produce more homogenous ingots. The ESR process is normally essential for heavy steel ingots.

ESR has been known since the 1930s, but it took around 30 years before it became an acknowledged process for the mass production of high-quality ingots. At the end of the 1960s, the concept of using ESR plants to manufacture large forging ingots gained acceptance. Increasing demand for larger electrical power generating units required forging ingots weighing 100 tons or more for manufacturing of generator and turbine shafts. The ESR technology is of interest not only for the production of smaller weight ingots of tool steels and super-alloys, but also of heavy forging ingots. A large ESR furnace, commissioned in the early 1970s, allows manufacturing ingots of 2,300 mm diameter and 5,000 mm length weighing upto 165 tons. The furnace operates with ingot withdrawal employing four consumable electrodes remelted simultaneously in the large diameter mould and replacing of the consumed electrodes with subsequent ones and as many as necessary to produce the desired ingot weight.

The ESR process is one of the most important new processes developed. The main advantage of the process is the refining which can be obtained by melting through a slag of controlled composition, and the special control over solidification. This control reduces dendrite arm spacing, micro-segregation, and porosity, which leads to a sound ingot. Also, the automatic melt control system (AMC) with its inherent features, ease of operation and its accuracy and repeatability of control, helps in producing ingots with excellent properties, including (i) homogeneous, sound and directionally solidified structure, (ii) high degree of cleanliness, (iii) free of internal flaws (e.g. hydrogen flakes), (iv) free of macro-segregation, and (v) smooth ingot surface resulting in a high ingot yield.

Hence, the ESR process is an appropriate process for high-quality materials which include (i) ball bearing steels, steel rollers, tool steels, wear resistant steels for low and high working temperatures, and high-speed steels for high performance, (ii) highly alloyed stainless steels, corrosion- and acid resistant steels, and steels used for high-temperature applications,  (iii) steels used in defence as well as in aviation and aerospace technology, (iv) steels used for medical, pharmaceutical, and chemical industries, and (v) steels used in off-shore, power and aerospace engineering,  nuclear reactor components.

The various advantages of ESR process compared to conventionally cast ingots include (i) dense structure of ingot without inner shrink hole or porosity, (ii) higher cleanliness (less and smaller inclusions), (iii) homogenous structure and chemical analysis over the whole ingot length, (iv) higher yield of remelted material to final product, (v) general improvement of mechanical properties, (vi) smooth surface (in general no treatment of surface before hot working necessary), (vii) controlled solidification (improved macro- and microstructure), and (viii) protection of remelted material against the oxidizing atmosphere by means of the slag bath.

Nowadays, steelmaking technology has improved a lot and it enables the production of high-purity liquid steel. However, during ingot casting the re-oxidation of the liquid steel occurs, which increases the inclusion content. Segregations on the macro and micro scales are also the features of the ingot casting. These cause anisotropy in the mechanical properties in the steel. The ESR process almost completely removes the macro-segregation phenomenon in heavy steel ingots, and ensures a more homogeneous chemical composition and a finer microstructure with fewer and more evenly distributed non-metallic inclusions than in the cast ingots. The low speed of remelting, combined with the water-cooled copper mould in the ESR process, ensures a particularly homogeneous and balanced, stable solidification. The segregations within an ingot produced by the ESR process are thus much lower (or even eliminated) compared to the open cast continuous cast billets or conventional cast ingots. For this reason, most segregation-sensitive steels are ESR processed for homogenization. The influence of ESR on remelted steel is shown in Fig 1.

Fig 1 Influence of ESR process on the properties of remelted steel

ESR process

ESR process is a continuous process. In this process, during the remelting of the consumable electrode, refining and solidification of the steel take place at the same time. Cast, rolled or forged steel ingots can be used as a consumable electrode. The ESR process is based on an electrical current running through an electrode through the liquid slag and ingot. Due to the high electrical resistance of the slag, the slag heats up and melts. The complete remelting process is carried out in a water-cooled copper mould, which allows the remelted ingot to solidify quickly and very uniformly.

The consumable electrode is immersed in a pool of the liquid slag in a water-cooled mould where the slag heat gradually melts the tip of the electrode. An electric current (usually AC) passes through the slag, between the electrode and the ingot being formed and superheats the slag so that drops of metal are melted from the electrode. These drops travel through the slag to the bottom of the water-cooled mould where they solidify. The mould with the slag pool is moving upwards as the new ingot is formed. The new ingot of refined material builds up slowly from the bottom of the mould. It is homogeneous, directionally solidified mass and is free from the central unsoundness which can occur in conventionally cast ingots as they solidify from the outside inwards.

Directional solidification is to be ensured in the process over the entire ingot cross-section and length to avoid interior defects, such as macro-segregation, shrinkage cavities and non uniform distribution of inclusions. By maintaining the correct remelting rate and slag temperature, directional solidification can be achieved for ingot with very large diameters.

Normally the ESR process offers very high, consistent, and predictable product quality. Finely controlled solidification improves soundness and structural integrity. Ingot surface quality is improved by the formation of a solidified thin slag skin between ingot and mould wall during the remelting operation. This is why ESR is recognized as the preferred production method for high-performance steels. Liquefied steel drips from the electrode tip and is refined when passing through the liquid slag, with oxides and sulphur getting bound in the slag. After passing through the slag, the steel cools down and solidifies again into a remelted ingot.

The design of the mould can be in the form of fixed long moulds or collar-type moulds. The use of collar-type moulds with movable moulds or a movable base plate, gives the possibility of producing ingots of any required length (Fig 2). Furthermore, the ESR process enables the production of ingots (i) with the AC-current used as remelting energy (from 3 kA to 92 kA), (ii) with the ingot weights from 100 kg to above 200 metric tons depending on material being remelted, and (iii) with ingot diameters ranging from 80 mm to greater than 2000 mm. Also ingots of round, square, and rectangular (slab) shapes can be produced by the process.

Fig 2 Schematic representation of the ESR process

Due to the superheated slag which is continuously in touch with the electrode tip, a liquid film of metal forms at the electrode tip. As the developing droplets pass through the slag, the steel is cleaned of non-metallic impurities which are removed by chemical reaction with the slag or by physical flotation to the top of the molten pool. The remaining inclusions in ESR are very small in size and evenly distributed in the remelted ingot.

In spite of directional dendritic solidification various defects, such as the formation of tree ring patterns and freckles, can occur in remelted ingots. It is important to note that white spots normally do not occur in an ESR ingot. The dendrite skeletons or small broken pieces from the electrode are to pass the superheated slag and have enough time to become molten before they reach the solidification front. This prevents the white spots in the ingots.

The ingot surface is covered by a thin slag skin and hence, needs no conditioning prior to forging. Electrodes for remelting can be used in the as-cast condition.

There are generally three possible sources for nonmetallic oxides inclusions. These are (i) inclusions inside the electrode, (ii) dissolved oxygen (O2) and de-oxidation agents such as aluminum (Al), and silicon (Si), and (iii) reactions between electrode and the process slag.

The most critical inclusions in steel are oxides. In principle, the chemical compositions of inclusions are influenced by the seven refining sites shown in Fig 2, but the three most important refining sites are three (site 1 to site 3). Reaction site 1 is thought to be the phase boundary with the largest refining potential, from a kinetic point of view. While it is proposed that inclusions dissolve into the surrounding material, temperatures of around 1800 deg C to 2000 deg C are necessary. According to site 1, these temperatures are not reached at this phase boundary. Instead, the metal droplets detach shortly after reaching the liquidus temperature and are subsequently overheated in the slag bath. However, the dwell time for the liquid steel underneath the electrode is with around 10 seconds significantly higher than in the slag bath itself (around 0.1 second). Because of this, there is enough time for the chemical refining reactions to occur underneath the electrode rather than in the slag bath. Hence, in the static ESR process, the reaction site 2 plays only an inferior role.

Concerning refining site 3, there have to be considered different phenomena in case of inclusion behaviour. On the one hand, there occur exchanges of O2 and deoxidizing agents on the interface slag/metal pool, on the other hand, the solubility of O2 decreases with advancing solidification of the steel in the mushy zone. Hence, there are inclusions precipitated, especially with regards to Al. It is not sure if refining mechanisms such as flotation occur as this phenomenon is governed by different influencing factors, beginning with the transport of these inclusions through the mushy zone and eventually the dissolution in the slag medium. If the process is not carried out under a protective gas atmosphere, the other refining sites (except 6) are of importance because of a potential O2 pickup and a subsequent increase of O2. For a sufficient desulphurization, the refining site 5 is of importance because here the sulphur (S) picked up by the slag forms SO2 which is transferred into the gas phase and removed.

The ESR process can be automated from melt initiation, through power build-up, steady melt rate period, reduced melt rate period to maintain pool profile, hot-tapping sequences, and melting termination. Close control of all remelting parameters is needed for reproducible production of homogeneous ingots. For fulfilling of the most stringent material quality specifications, ESR process furnaces normally make use of computer controlled process automation. Logic control functions, continuous weighing of the ingot, closed loop control of the process parameters, and data acquisition and their management are handled by dedicated computer systems. These computer systems communicate via field bus or specific interfaces. An operator interface PC (OIP) acting hierarchically as master of the control system is utilized as the interface between operator and ESR process. The OIP serves for process visualization, featuring parameter indications, graphic displays and soft keys for operator commands, editing and handling of remelting recipes, data acquisition and storage as well as for generation of the melt records

Variations of the ESR process

A fully coaxial furnace design is required for remelting of segregation-sensitive steels in order to prevent melt stirring by stray magnetic fields. Different variations of the ESR process have been developed for ensuring the ever-increasing demands for material properties. Shielding of the melt space with protective atmosphere has been the latest trend in recent years. Remelting under increased pressure to increase the N2 content in the ingot is another variation of ESR.

Three ESR process variations which have been developed are namely (i) remelting under inert gas atmosphere (IESR), (ii) remelting under increased pressure (PESR), and (iii) remelting under reduced pressure (VAC-ESR).

The IESR process consists of electroslag remelting under a fully enclosed protective atmosphere of inert gas at atmospheric pressure. It is a variation of the ESR process where the inert argon (Ar) gas protects the slag and steel from oxidation and the absorption of nitrogen (N2) and hydrogen (H2) from the air. The inert gas atmosphere frees the ESR process from H2 pick-up problem and the influence of seasonal atmospheric changes. In addition it allows remelting under O2 free inert gas The oxidation of the electrode is almost completely avoided, thus the process provides better cleanliness of the ingot. However, due to the absence of O2 in the furnace atmosphere, desulphurization is not optimal. Two furnace concepts are available, one with a protective hood system of relative tightness, the other with a fully vacuum-tight protective hood system which allows the complete exchange of air against an inert gas atmosphere prior to starting the remelting process.

The PESR consists of electroslag remelting under increased pressure. In recent years, N2 has become increasingly attractive as an inexpensive alloying element for enhancing the properties of steel. In austenitic steel, N2, particularly in dissolved form, increases yield strength by forming a super-saturated solid solution. With ferritic steel grades, the aim is to achieve a fine dispersion of nitrides comparable to the microstructure obtained by quenching and tempering iron (Fe) -carbon (C) steels. For the production of these new materials, it is essential that a sufficiently high amount of N2 above the solubility limit under normal pressure is introduced into the liquid steel and that N2 loss is prevented during solidification. As the solubility of N2 is proportional to the square root of its partial pressure, it is possible to introduce large amounts of N2 into the melt and allow it to solidify under higher pressure. This has been established by the electroslag remelting process at an operating pressure of 42 kg/sq cm.

Due to the extremely short residence time of the metal droplets in the liquid phase during remelting, the N2 pick-up via the gas phase is insufficient. The N2 is, hence, supplied continuously during remelting in the form of solid N2-bearing additives. The high pressure in the system serves exclusively to retain the N2 introduced into the liquid steel. The pressure level depends on the composition of the steel and on the desired N2 content of the remelted ingot.

Electroslag remelting under vacuum (VAC-ESR) is another newly developed process. It is a variation of the ESR which also provides vacuum degassing of the melt. In the VAC-ESR process, the remelting is carried out under vacuum using a slag. Problems of oxidation of the melt do not arise. In addition, dissolved gases such as H2 and N2, can be removed and the danger of white spots is reduced to a minimum. The process is suitable for the remelting of super-alloys and titanium alloys.

The parameters of the process

The heat needed for the operation of the ESR process is generated in the slag bath by the Joule effect. The quality of the remelted ingot is influenced by (i) the electrical characteristics, (ii) heat balance, and (ii) the electrode/ingot diameter. The requirement of energy input for the ESR process is normally in the range of 1000 kWh/t and 1500 kWh/t of steel. The slag bath is considered to be a variable resistor. Its resistance is determined by the electrode distance, the effective slag resistivity, and by the electrical current path. The normal slag depth is usually around 100 mm.

The shape of the liquid pool is influenced by the heat input in the process. The higher is the distance between the consumable electrode and the remelted ingot, the smoother is the heat distribution in the slag. When determining the electrode distance, it is necessary to take into account that a shorter current path indicates a higher current with concentrated heat generation under the electrode tip and an undesirable deepening of the metal pool. On the other hand, a longer current path requires a high voltage, which causes more even heat generation and a flatter, more favourable pool profile.

The operating voltages in the ESR process are normally around 40 V or lower. The electrical circuit for the ESR process can be either AC (alternating current) or DC (direct current). Single phase AC- ESR process gives optimum refinement and melt rate for the ingots having diameters 200 mm or more. The DC-ESR process needs a lower melt rate for the refinement of the steel. However, when the refinement of the steel is not the main requirement, the DC-ESR process gives the highest melting rates per unit of power consumed. The present practice is to use a single-phase AC power supply and low electrode/ingot diameter ratio normally in the range of 0.4 to 0.7. Generally 50 Hz (hertz) or 60 Hz frequency is used for the AC operation. However, for very large ingots, where reactivity is more important, it is better to use low-frequency power (in the range of 5 Hz to 10 Hz) for improved efficiency.

Optimum melting rates and energy inputs are dependent on the ingot diameter. The optimum conditions for the maximum permissible melt rate at the lowest possible power are normally determined by the equation ‘melt rate = constant × power × fill ratio (area) × mould area / electrode distance’. Many operating practices consider melt rate as proportional to the ingot diameter, which is obtained at a melt rate of around 0.004 kg/min/mm. Fig 3 shows the effect of voltage and current on the rate of melting for an ingot of 240 mm diameter. It can be seen from the relationship that for a given current and ingot size, there is an optional voltage which corresponds to a maximum melt rate.

Fig 3 Effect of voltage and current on the rate of melting

Role of slag in ESR process

The slag has an important role to play in the ESR process from the control-of-inclusions point of view. The chemical and physical properties of slag also have a great effect on the removal of inclusions. The role of slag includes (i) generation of Joule heat for the melting of the electrode, (ii) refinement of the liquid steel through the absorption of non-metallic inclusions, (iii) desulphurization of the steel, (iv) protection of the steel from contamination, (v) providing lubrication for the copper mould/solidifying steel shell interface, and (vi) controlling the horizontal heat transfer between the solidifying steel and the mould.

Slags for ESR are usually based on calcium fluoride (CaF2), lime (CaO) and alumina (Al2O3). Magnesia (MgO), titanium oxide (TiO2) and silica (SiO2) can also be added, depending on the steel to be remelted. The CaF2 in the slag increases the solubility of basic components (CaO and MgO) of the slag and hence increases the effective sulphide capacity of the slag.

To perform its intended functions, the slag is required to have some well-defined properties, such as (i) its melting point is to be lower than that of the metal to be remelted, (ii) it is to be electrically efficient, (iii) its composition is to be selected to ensure the desired chemical reactions, and (iv) it is to have suitable viscosity at remelting temperature. Tab 1 gives characteristics of different types of ESR slags.

As given in Tab 1, the concentrations of CaF2 can vary from 0 % to 100 % of the mass fractions. The remaining slag constituents are mostly used for decreasing the basicity. The slag chemical composition is changed during the ESR process, due to the formation of volatile fluoride, the precipitation of high-melting-point phases, and the reaction in the ESR process. The changes in composition affect the slag’s metallurgical properties and eventually affect the quality of the remelted ingot. The quantity of consumed slag per ton of steel depends on the remelted ingot diameter.

Many of the slags used in ESR can be described with the ternary CaF2-CaO-Al2O3 system. The main feature is a eutectic corresponding to compositions with roughly equal proportions of CaO and Al2O3. This identifies the slags with liquidus temperatures in the range 1350 deg C to 1500 deg C, which make them suitable for melting of a wide range of alloys, including steels and super alloys. In the case of slag with 70 % CaF2 and 30 % Al2O3, the CaO is excluded as much as possible in order to prevent the pick-up of H2, while there are no problems with the presence of the two liquids. The binary CaO-Al2O3 system on the other hand, has only a limited range of slags with suitable melting characteristics, while the binary CaF2-CaO system is used in cases where a high degree of desulphurization is required.

However, its disadvantage is having a low resistivity. High CaO contents also increase the risk of moisture retention or H2 pick-up. A certain amount of SiO2 addition into the ESR slag in the case of the drawing-ingot-type ESR process is important for improving the lubrication performance, controlling Si and Al content in the liquid steel, and modifying oxide-type inclusions. Also, the addition of SiO2 suppresses the crystallization temperature of CaF2-Al2O3-CaO slags. Further, the MgO and SiO2 in CaF2-containing slags affect the surface tension of the slag.

Although CaF2 is a crucial component in any ESR slag and it greatly decreases the melting temperature of the slag systems, it is insoluble in oxide phases. Slag properties, such as electrical conductivity, thermal conductivity, density, viscosity, and surface tension play an important role in effective melting and metal refining. Slag resistivity affects the operating characteristics and economics of the ESR process. Al2O3 increases the resistivity of the slag and promotes good heat generation, thus enabling a reduction of the slag bulk content, which also reduces the heat loss due to the reduced area of contact between the slag and the mould wall.

Slags are sometimes referred as ‘long’ and ‘short’ slags when the slag viscosity is considered. Long slags remain fluid over a wide range of temperatures and are likely to give thin slag skins and hence good ingot surfaces. Short slags rapidly become viscous on cooling and are likely to give thick slag skins and poor ingot surfaces. High CaF2 contents promote short slags, whereas SiO2 and MgO contents favour long slags.

Thermodynamics of the ESR process

In the case of ESR process of steel in an air atmosphere, chemical reactions take place and change the chemical composition of the as-cast ingot. The levels of some elements, such as cobalt (Co), nickel (Ni), chromium (Cr), molybdenum (Mo), tungsten (W), and carbon (C) remain unchanged after remelting. However, the content of Si (silicon), O2, and S can be changed from 10 % to 80 %, while the content of Al and Ti (titanium) can vary depending on the melting conditions (decrease or increase). Hence, some measures are required to be taken to prevent the losses of elements. This can be achieved by using special ESR variations. Another way is to control the slag composition by regular additions to the slag, which is desirable due to the steady melting conditions.

The oxidation of the elements can be prevented by deoxidation slag during the melting process achieved by additions of Al. The O2 potential of the slag determines the chemistry of the ESR process. It affects the removal of S and the non-metallic inclusions. O2 reacts with some elements in the steel and suppresses the pick-up of H2. In the slag, O2 is mostly in combined form as FeO, MnO and SiO2. To determine the O2 content in the steel, it is essential to find the relationship between FeO in the slag and the O2 in the remelted ingot. However, due to the very low solubility of FeO in CaF2 slags, its activity is very high. The O2 content can be determined by the thermodynamic analyses of the reactions between O2 and the active components.

Si and Mn are elements which can react with the O2 present in the steel and from the slag. When Si is the strongest deoxidizer, the O2 content of the steel is determined by the Si content. At constant temperature and Si content in the steel, the O2 content of the steel is higher at higher activity of the SiO2 in the slag, or by lowering the basicity of the slag. Al losses in the remelted ingot are small, especially at high Al2O3 content in the slag. On the other hand, the presence of Al2O3 in the slag reduces the oxidation of Si. The reaction between Si in the electrode and Al2O3 in the slag also controls the oxidation of Al in the remelted ingot. Hence, Al content in the remelted ingot depends on the content of Al2O3 in the slag and the content of Si in the electrode, temperature and chemical composition of the steel.

The content of Al in the remelted ingot decreases when CaF2-Al2O3-CaO slags with increased SiO2 content are used. When Al is used for deoxidation, upto 15 % of the added Al is transferred to the liquid steel. The content of Ti in the remelted steel depends on the content of Al and Ti in the consumable electrode, the content of Al2O3 and TiO2 in the slag, and the O2 potential in the gas phase above the slag. The equilibrium between the Al and Ti content in the electrode at different TiO2 contents of the slag is shown in Fig 4. For the given content of Al in electrode, the loss of Ti can be minimized by the addition of TiO2 to the slag. At high contents of Al, the TiO2 in the slag is reduced by the Al and hence, Al also regulates the Ti/TiO2 ratio.

Fig 4 Equilibrium between the Al and Ti content in the electrode at different TiO2 contents in the slag

In the early stages of the development of the ESR process, the removal of S has been was considered as one of the main objectives. The rate of desulphurization increases with the basicity of the slag. S transfer takes place mainly at two interfaces, according to the two reactions namely (i) slag/metal reaction, and (ii) gas/slag reaction (Fig 5).

Fig 5 Interfaces for the transfer of sulphur

 A thermodynamic analysis of the reactions given in Fig 5 shows that the desulphurization is related to (i) the concentration of O2 ions in the slag, (ii) the partial pressure of O2 in the gas phase, and (iii) the chemical composition of the steel. The transfer of S from the steel to the slag is promoted by the high slag basicity and low concentration of O2 in the steel. On the other hand, the S transfer from slag to gas is promoted by a high partial pressure of O2 in the atmosphere and the low basicity of the slag. The ability of the slag to take S is defined in terms of its S capacity. The S capacity for the CaF2-CaO-Al2O3 system increases as the CaF2 content is increased and by increasing the amount of CaO to the saturation limit.

In the case of ESR under protective inert gas atmosphere, the S remains in the slag and there is build up of the S as the process continues. In such cases, the S capacity is the ruling factor, and the slag composition is to be adjusted in order to continue its desulphurizing action to the end of the process. This means that the slag/metal ratio assumes greater importance.

Solidification and structure of the ingot

The solidification structure of the ingot produced by the ESR process is a function of the local solidification time and the temperature gradient at the liquid/solid interface. For achieving a directed dendrite primary structure, a relatively high temperature gradient at the solidification front is required to be maintained during the complete remelting period.

Macrostructure of the ingots produced by the ESR process is different from the macrostructure of conventionally cast ingots due to the different method of heat transfer and heat removal. The growth direction of the dendrites is a function of the metal pool during solidification. Thus, the gradient of dendrites with respect to the ingot axis increases with melting rate. In extreme cases, the growth of directed dendrites can come to a stop. The ingot core then solidifies non-directionally in equiaxed grains, which leads to segregation and micro shrinkage. Even in the case of directional dendritic solidification, the micro segregation increases with the dendrite arm spacing. A solidification structure with dendrites parallel to the ingot axis yields optimal results. However, this is not always possible.

A good ingot surface needs a minimum input of energy and hence, a minimum rate of melting. Increase in the rate of melting increases the difference between the gradient of the solidus and liquidus isotherms and leads to the increased pool depth. Thus, grains growth takes place in radial direction instead of vertical direction. Increasing the melting rate causes a finer grain structure and changes the growth direction of the columnar structure from the axial to radial growth and deeper liquid pool at very high melting rates. Increasing the temperature of the liquid slag also results in a coarser columnar grain structure and a reduced thickness of the refined equiaxed grain layer, both at the surface and the bottom of the ESR ingot. In spite of directional dendritic solidification, defects such as tree-ring patterns, freckles and white spots can occur in a remelted ingot.

Macro-segregation and porosity structures in the middle of the ingot are usually uncommon in the ESR ingots. A major characteristic of the ESR process is its ability to produce steel with reduced micro-segregation. This is linked with the local solidification time and dendrite-arm spacing. Steel in the ESR process normally freezes in a columnar manner, which gives less micro-segregation than equiaxed structures. The greater is the temperature gradient, the smaller is the distance between the dendritic arm spacing and the lower is the chemical heterogeneity in the micro areas. In ESR process, the temperature gradients are greater than in the case of the conventional casting. Hence, the secondary dendrite-arm spacing is smaller in case of ESR process than in conventional casting of the ingots.

The effect of decreasing the segregation effect is shown in Fig 6, where a comparison of microstructures before and after ESR processing has been made for a hot-work tool steel. The microstructure in both cases is tempered martensite. The difference in segregation bands is apparent. While the segregations are evident in the consumable electrode, they are almost completely eliminated in the remelted ingot.

Fig 6 Segregation in the hot worked tool steel remelted by the ESR process

The effect of local solidification time on the dendrite spacing shows that the dendrite-arm spacing is decreased as the cooling rate is increased. Besides a more homogeneous composition and compact solidification structure, the removal of non-metallic inclusions is an important characteristic of the ESR process. In general, inclusions easily initiate micro-voids and cracks at the inclusion/steel interface, which can be the origin of fatigue fracture or other defects. Also, ESR processed steel is not an exception. Many factors influence the formation of non-metallic inclusions in ESR processed steel, including furnace atmosphere, content of inclusions in the consumable electrode, slag amount and its composition, power input, melting rate, filling ratio etc. Most of the non-metallic inclusions occur due to the reactions between O2 and the elements such as Mn (manganese), Si, and Al. Deoxidization of the slag during electroslag has an important influence on the non-metallic inclusions formation in the ESR processed ingot. It can be seen that the lowest number of inclusions is attained in ESR with the lowest viscosity and the highest interfacial tension. However, the absence of large inclusions is typical for the ESR process.

The removal of non-metallic inclusions during ESR process takes place at the tip of the electrode, where mainly absorption and dissolution of non-metallic inclusions in the slag take place. As the electrode tip is heated towards its melting point, the inclusions in the electrode are re-dissolved before the steel melts. Any other inclusions, such as larger exogenous inclusions in the electrode, are not dissolved in the solid metal and get exposed to the slag when the electrode tip becomes molten. If the slag composition is suitable, the temperature is high enough and the residence time is long enough, the non-metallic inclusions dissolve in the slag. Though, at this point there can be further reactions due to the difference in equilibrium constants, as well as the possibility of the flotation of large inclusions. The steel at this point is free from non-metallic inclusions, but can have in solution elements which produce inclusions by reaction during the freezing time (S removal reaction). The removal efficiency of inclusions increases with the reduced melting speed. It has been seen that a multi-component slag (CaF2, CaO, Al2O3, SiO2, and MgO) has a better capacity for controlling the amount of inclusions. Most non-metallic inclusions for multi-component slag are MgO-Al2O3 inclusions, while mainly Al2O3 inclusions exist when using conventional 70 % CaF2 – 30 % Al2O3 slag. Furthermore, the maximum inclusion size for multi-component slags has been found to be smaller than for conventional binary slag.