Steel Hardening by Quenching and Tempering
Steel Hardening by Quenching and Tempering
In general terms, streel parts are manufactured from steels which are in their softest possible condition. This makes cutting, bending, and stamping etc. as easy as possible. However, in the final application, the steel part is required to be hard, wear resistant, tough, or strong or some combination of these. It is the hardening carried out by quenching and tempering processes which transform the steel part from one condition to the other condition.
Hardening is carried out by quenching of steel, which consists of cooling it fast from a temperature above the Ac3 transformation temperature. The quenching is necessary for suppressing the normal breakdown of austenite into ferrite and cementite (pearlite), and for causing a partial decomposition at such a low temperature so as to produce the new phase called martensite. For achieving this, steel needs a critical cooling velocity, which is greatly reduced by the presence of alloying elements. In such case hardening of steel occurs with mild quenching.
Martensite is a super-saturated metastable phase and has body centered tetragonal (bct) lattice instead of body centred cubic (bcc) lattice. After steel is quenched, it is normally very hard and strong but brittle. Martensite looks needle like under microscope because of its fine lamellar structure. The martensite of quenched steel is exceedingly brittle and highly stressed. Hence, cracking and distortion of the work-piece are liable to occur after quenching. Retained austenite is unstable and since it changes dimensions. It is hence necessary to warm the steel below the critical range in order to relieve stresses and to allow the arrested reaction of cementite precipitation to take place. This is known as tempering.
Quenching and tempering are a combined heat treatment process which consists of quenching and high temperature tempering. It is a common process used in the manufacture of steel parts. During the quenching and tempering process, as-quenched steels are tempered for achieving optimum mechanical properties tailored to fit specific application. The final properties of a quenched and tempered steel part depend on the micro-structure which evolves during the quenching and tempering process. In practice, the quenching and tempering process is designed by empirically determining and by trial-and-error methods.
The heat treatment strategies greatly influence the micro-structural morphology, resulting in different final properties of steels. Among them, precise control of temperature and duration for the quenching and tempering processes is imperative for achieving the desired balance of constitutive micro-structural and mechanical properties. For the majority of the plain carbon and low-alloy steels, ferrite and pearlite can be acquired through slow cooling.
Normally, the pearlite always encompasses the lamellar clusters of ferrite and cementite. With a higher cooling rate, an upper bainite micro-structure can be achieved from austenite. The upper bainite normally includes ferrite platelets hindered by austenite or carbides. Meanwhile, these types of micro-structures can be observed and characterized through their particular ‘C-curve’ based on the well-known time–temperature transformation (TTT) diagrams. It has been documented that carbides can precipitate from the carbon-supersaturated austenite, while they can also precipitate from enriched ferrite plates in the lower bainite. In contrast, martensite is obtained when a further increase in the cooling rate is applied, which normally involves a number of carbon-supersaturated ferrite platelets with a high density of dislocations or twin boundaries.
It is to be noted that the residual austenite always exists along the martensite lath boundaries, and can be more homogeneous and detached compared to the original structure. After tempering, the residual austenite disintegrates and maintains the state of fine and even distribution. In general, low tempering temperatures can simply lessen the internal stresses, diminishing brittleness while mostly preserving the hardness levels. However, in some low-alloy steels containing certain elements, such as chromium and molybdenum, hardness can increase even when tempered at low temperatures.
Several steels with high contents of such alloying elements act similarly to precipitation-hardening alloys, which show the opposite effects observed in quenching and tempering processes. In comparison, tempering at higher temperatures tends to cause a higher reduction in hardness, but leads to increased elasticity and plasticity by sacrificing a certain quantity of strength. However, the steel can suffer from another embrittlement phase, known as ‘temper embrittlement., which arises once the steel is kept for too long within the temperature interval of temper embrittlement. When heated above this temperature, the steel normally quickly cools down to avoid the occurrence of temper embrittlement.
Steels can be hardened by the simple expedient of heating to above the Ac3 transformation temperature, holding long enough for ensuring the achievement of uniform temperature and solution of carbon in the austenite, and then cooling fast (quenching) as shown in Fig 1.
Fig 1 Hardening of steel by quenching and tempering
Complete hardening depends on cooling so fast that the austenite, which otherwise decomposes on cooling through the Ar1 temperature, is maintained to relatively low temperatures. When this is accomplished, the austenite transforms to martensite on cooling through the Ms-Mf temperatures range. Fast cooling is necessary only to the extent of lowering the temperature of the steel to well below the nose of the S-curve (TTT curve). Once this has been accomplished, slow cooling from then on, either in oil or in air, is beneficial in avoiding distortion and cracking.
Special treatments, such as time quenching and martempering, are designed to bring about these conditions. As martensite is quite brittle, steel is rarely used in the as-quenched condition, that is, without tempering.
The maximum hardness which can be achieved in completely hardened plain carbon and low-alloy structural steels depends mainly on the carbon content. The relationship of maximum hardness to carbon content is shown in Fig 2.
Fig 2 Relationship of carbon and maximum achievable hardness of quenched steel
Previous discussion of the formation of martensite has neglected the influence of mass. It is to be realized that even with a sample of relatively small dimensions, the rate of abstraction of heat is not uniform. Heat is always abstracted from the surface layers at a faster rate than from the interior. In a given cooling medium, the cooling rate of both the surface and interior decreases as the dimensions of a sample increase and the possibility of exceeding the critical cooling rate becomes less. For overcoming this, the steel can be quenched in a medium having a very high rate of heat abstraction, such as iced brine, but, even so, several steels have a physical restriction on the maximum size amenable to complete hardening regardless of the quenching medium.
The marked effect which mass has upon the hardness of quenched steel can be shown by measuring the hardness distribution of different size rounds of the same steel quenched in the same medium. Curves showing the distribution of hardness in a series of round bars of different sizes of 0.5 % carbon steel are shown in Fig 3. The quenching medium used here is water, and the quenching temperature is 830 deg C. The rate of cooling is decreased as the diameters of the rounds is increased. Only the 40 mm diameter round has been hardened completely through the cross section, whereas with the 100 mm diameter round the critical cooling rate has not been achieved even at the surface.
Fig 3 Variation in hardness
Tempering (sometimes called drawing) is the process of reheating hardened (martensitic) or normalized steels to some temperature below the lower critical (Ac1) temperature. The rate of cooling is immaterial except for some steels which are susceptible to temper brittleness. As the tempering temperature is increased, the martensite (Fig 4A) of hardened steel passes through stages of tempered martensite (Fig 4B, and Fig 4C) and is gradually changes into a fine structure termed as bainite (Fig 4D). These changes are accompanied by a decreasing hardness and increasing toughness.
Fig 4 Microstructures and corresponding hardness of quenched and tempered high carbon steel
The tempering temperature depends upon the desired properties and the purpose for which the steel is to be used. If considerable hardness is necessary, the tempering temperature is required to be low. If considerable toughness is needed, the tempering temperature is required to be high. The effect of tempering on the hardness of fully hardened carbon steels is shown in Fig 5a.
Fig 5 Effect of tempering temperature and time on the hardness
Temper brittleness is manifested as a loss of toughness (observed only by impact tests of notched bars) after slow cooling from tempering temperatures of 595 deg C or higher, or after tempering in the temperature range between around 455 deg C and 595 deg C. It is most pronounced in alloy steels which contain manganese or chromium and normally can be prevented by fast quenching from the tempering temperature. The presence of molybdenum is beneficial in counteracting the tendency toward temper brittleness.
Proper tempering of a hardened steel needs a certain quantity of time. At any selected tempering temperature, the hardness drops rapidly at first, gradually decreasing more slowly as the time is prolonged. The effect of time at different tempering temperatures upon the resident hardness of a eutectoid carbon steel is shown in Fig 5b. Short tempering periods are normally undesirable and is to be avoided. Good practice needs at least half an hour (or, preferably 1 hour to 2 hour) at tempering temperature for any hardened steel.
The necessity for tempering a steel promptly after hardening cannot be over emphasized. If fully hardened steel is allowed to cool to room temperature during hardening there is danger that the steel can crack. Carbon steels and the majority of the low alloy steels are to be tempered as soon as they are cool enough to be held comfortably in the bare hands. Steels are not to be tempered before they cool to this temperature since in some steels the Mf temperature is quite low and untransformed austenite can be present. Part of all of this residual austenite transforms to martensite on cooling from the tempering temperature so that the final structure consists of both tempered and untempered martensite. The brittle untempered martensite, together with the internal stresses caused by its formation, can easily cause failure of the heat-treated steel part. When it is possible that such a condition exists, a second tempering treatment (double tempering) is to be given to temper the fresh martensite formed on cooling after the initial tempering treatment.
If structural steels are to be used in the normalized condition, the normalizing operation is frequently followed by heating to a temperature of around 650 deg C to 700 deg C. The purpose of this treatment, which is also designated as tempering, is to relieve internal stresses resulting on cooling from the normalizing temperature and to improve the ductility.
Hardening – Hardening is a heat-treating operation necessary to impart hardness to a steel part. This treatment consists of heating the steel to a selected hardening temperature (austenitizing temperature), and holding it at this temperature, followed by cooling or quenching at a rate fast enough to develop the desired hardness. When the steel is austenitized, its lattice structure is a face-centered cubic (fcc). The reason for heating and holding the steel in the austenite range is to dissolve carbides cementite into a matrix, then the steel is quenched. At this stage, the trapped carbon causes a shift in atoms to form a body-centered tetragonal (bct) structure.
The shift of atoms and the trapped carbon creates a stressed lattice structure. This stressed structure, called martensite, is hard and brittle. It is responsible for the high hardness of the steel. The hardness finally achieved depends on how much carbon has been available, how much of it is dissolved, the temperature adopted, and the rate of cooling. It is subsequently tempered for reducing the induced quenching stresses caused by the formation of martensite.
The main object of hardening is to develop high hardness and to increase the wear life of the steel parts. The higher is the hardness, the higher is the wear resistance of the steel part. For example, spindles, gears, shafts, tools and dies, and high-speed steel tools, etc., need high hardness. It is also necessary for structural steels to have the needed mechanical properties such as tensile strength, ductility, and elasticity etc. These properties can be achieved by hardening, followed by tempering.
Hardened and tempered steel parts possess better mechanical properties compared to the annealed and normalized steel parts. Since hardening is the last operation, the steel part is almost close to the finished state. Hence, it is not to be attempted without proper equipment, facilities, and suitably trained operators.
The hardening temperature for specific grades of steels, achieved on the basis of a series of practical trials, is given in the manuals of the majority of the steel manufacturers. Recommended practices giving ranges of hardening temperature relevant to steels, listed in the international standards, can also be adopted.
The hardening temperature to be used depends on the chemical composition, carbon content, and section thickness of the steel. The temperature used for hypo-eutectoid steel is in the range of 20 deg C to 50 deg C above the Ac3 critical temperature, and for hyper-eutectoid steels 30 deg C to 50 deg C above the Ac1 critical temperature. If a steel containing, for example, 0.5 % carbon (ferrite and pearlite structure) is heated to a temperature below the lower critical temperature Ac1, it does not change the original structure of the steel containing pearlite and ferrite. Heating to a temperature above the lower critical temperature Ac1 but below the upper critical temperature Ac3 changes pearlite to austenite, without affecting the free ferrite. Quenching from this temperature produces a semi-hard steel, since the austenite is transformed to martensite, but the free ferrite remains unchanged.
The presence of ferrite in addition to brittle martensite does not give appreciable toughness. If the steel is heated slightly above the Ac3 critical temperature and held at this temperature for a sufficient time to bring about complete diffusion and equalization, the steel is transformed, giving the smallest austenite grain size possible. An effective quenching from this range produces martensite and the maximum hardness possible.
Heating to a temperature considerably above the higher range of critical temperature tends to increase the grain size. Subsequent quenching produces a martensite structure, whose properties, even after tempering, results in a low impact strength. This can lead to warping and cracking during the quenching operation.
In case of hardening of the hyper-eutectoid steels, the process consists of heating the steel to a temperature 30 deg C to 50 deg C above the critical temperature Ac1 which is the austenite plus carbide region. This is followed by rapid quenching to produce a fine-grained martensite with sufficient undissolved carbides. The quenched structure has a higher hardness compared to the martensite, since the hardness of carbides is higher. If too little carbide is dissolved in austenite, the resultant hardness is low.
The quantity of carbide which can be dissolved in austenite is proportional to the austenitizing temperature. The quantity of carbide increases progressively as the temperature is increased, and the grain size of the steel also increases. If too much carbide is dissolved, grain growth can occur with a corresponding decrease in hardness as well as toughness. If the steel is heated above the Acm temperature, the resultant structure has only austenite grains. In such a case, the grain growth is larger, and hence, the martensite becomes coarser. The resultant hardness is low because of the presence of an excessive quantity of retained austenite in the quenched structure and also since no more carbide is present in the steel structure.
Hence, proper precautions are needed to be taken to get a fine grained martensite with sufficient undissolved carbides. The treatment temperature for specific grades of steels is given in the manuals of all steel manufacturers. This information is based on a series of practical trials. In addition, the recommended practices in international standards can also be used. They contain a hardening temperature range to which the steel is to be heated.
Operations before hardening – It is very important that the steel parts to be hardened are to be free from scales, grease, and oil etc. in order to achieve the desired hardness. Hence, such steel parts are to be cleaned thoroughly. Steel parts having holes, particularly tool steels, can be packed with clay, asbestos, and steel inserts wherever necessary, so that no hardening can occur in the holes. Special attention is not needed where holes are relatively large and tools are to be quenched in such a manner that the internal surfaces of the holes are hardened completely. The steel parts are to be placed in suitable fixtures before placing them in the hardening furnace. This is done to avoid distortion. Small steel parts can be heated in a suitably designed basket for ensuring uniform hardness.
Plain carbon and low-alloy steels can be heated right up to the hardening temperature in one step, without any pre-heating. Steel parts of large size and complicated shapes can be pre-heated to avoid distortion and cracking because of the uneven temperature differentials between the core and the surfaces. This is more common, particularly in tool steels, since the thermal conductivity of tool steels is very low. For this reason, these steel parts are preheated to 500 deg C to 600 deg C. At this temperature, the internal stresses generated by uneven heating between the core and the case are gradually relieved as a result of local plastic deformations. The remaining internal stresses become negligible. Hence, at temperatures above 500 deg C to 600 deg C, steel parts can be heated faster. Pre-heating is also necessary when the hardening temperature is high, since the shortest possible time at high temperatures minimizes scaling and decarburization. Complicated steel parts or tools made of highly alloyed steels are to be pre-heated in two steps before subjecting them to the austenitizing temperature.
It is also very important that steel parts have a homogeneous and fine-grained structure before hardening. For example, if the steel part has a coarse-grained structure before hardening, it can lead to a non-uniform hardness, a higher degree of distortion, and quenching cracks can also develop during hardening.
Prior to the hardening, the steel parts possessing a coarse-grained structure are to be normalized for a short period. Normalizing refines the grain structure of steels, and helps to avoid the previously mentioned defects. For ensuring the development of a high and uniform hardness in tool steels in the as-hardened condition, the steel, prior to hardening is required to have a lamellar pearlite structure instead of a globular structure. The reason for this preference is that the transformation of globular structure into austenite occurs at a considerably slower rate than the lamellar pearlite. Yet another reason is that steels of this type do not possess a higher depth of hardness. If it is necessary to harden a tool steel to a higher depth, it is then necessary to have a lamellar pearlite structure. For this, a normalizing operation can be done in the temperature of 780 deg C to 800 deg C.
Before hardening, the steel part is also to be free from the cementite network since it leads to a further increase in brittleness of the hardened steel. Such a steel is to be subjected to the normalizing operation for overcoming this defect.
Holding time in the hardening bath – The time taken to reach the hardening temperature depends on several factors such as the heating equipment and the hardening bath. The rate of heating in a salt bath is considerably faster than it is possible in a furnace with atmosphere since the heat transfer from a liquid to a solid body occurs at a faster rate. When the desired hardening temperature is reached, sufficient soaking time is to be given. A visual inspection can be done through a small hole in front of the furnace to assess whether the steel part has reached the desired hardening temperature in relation to the walls of the furnace. If the surface of the steel part is darker, its temperature is lower, and it is to be allowed to remain in the furnace for some more time. When the steel part achieves the furnace temperature throughout, the colour of the steel part is indistinguishable from that of the furnace wall. When the surface of the steel part has reached the needed temperature, the holding time is to be counted. The same principle applies to salt bath heating. If the salt bath is maintained properly, it is possible to match easily the colour of the steel part to the colour of the transparent liquid salt. After the component has reached a uniform temperature, it is to be held at this temperature and subsequently quenched to get the martensite structure.
The time of holding at the hardening temperature depends on the type of steel and the selected temperature within the prescribed range for plain shaped parts. In majority of the cases, the upper limit of the hardening temperature is selected. However, if the cross-section shows large differences, the lower temperature range is preferred. Holding times are, hence, longer in the first case than in the later case. For offsetting the risk of overheating worsening of grain, carbon and low-alloy steels are to be kept for a far shorter period at the hardening temperature than, for example, in the case of high-alloy, special hot working steels containing carbides which need more time for the dissolution of carbides necessary for the optimum retention of hardness.
Hardening furnaces – The hardening treatment needs a suitable furnace which serves as a heating equipment. The furnaces used for hardening need to have accurate temperature control and a suitable atmosphere for ensuring an accurate treatment. It is important that the atmosphere used is neutral during the hardening operation and it is not to cause decarburization or carburization on the surface of the component. A decarburized layer gives a lower hardness which can result in the choice of an incorrect tempering temperature. Decarburization can lead to cracking in tool steels. In certain cases, the decarburized layer can be removed by grinding, if the component displays adequate tolerance. However, several tools are normally finished to size and hence the same is to be maintained throughout the hardening operation. Scaling and decarburization cannot be tolerated in such cases. For getting an optimum hardening on steel parts, it is necessary to select the best furnace with suitable heating media.
The types of heat treatment furnaces available for hardening are (i) salt bath furnace, (ii) muffle furnace, (iii) vacuum furnace, and (iv) fluidized bed furnace. These furnaces derive their names because of the particular heating medium used. For example, different salts are used as a heating media in a salt bath furnace, whereas a vacuum or a fluid bed is used in the case of the vacuum furnace and the fluidized bed furnace respectively. In all these furnaces, heating is normally done with electrical energy using electrodes, coils, or heating elements housed in a suitable container, depending on the particular heating media, namely salt, and gas etc.
Since the hardening process depends very much on the correct preparation of the hardening media, a brief description of the different media used is given below.
Salt baths – Neutral salts are used for hardening steels. These salts are needed to heat the steel without causing either carburization or decarburization. Salts are available in the market as proprietary salts. They can be used for the medium temperature and high temperature range of hardening.
Salts used for hardening different types of steel at medium temperature are chlorides of sodium, potassium, barium and calcium. They vary widely in composition. There is no single mixture which can cover the whole range of temperatures needed for all types of steels. Three or four combinations of salts, normally available under trade names, cover the entire heating range needed for all steels. Manufacturers of heat treatment salts do not specify the composition of the salt, on the other hand, they give detailed information on properties and applications.
Although chloride salts are inert to steel, they provide the best protection to steel when first melted. In the course of operation, molten salts gradually react with air and form oxy-chlorides and oxides, resulting in decarburization on the work-piece. Decarburization is aggravated in the higher temperature range, if held for a longer time in the bath. Hence, salt baths are to be periodically rectified, the frequency being determined by the cleanliness of the work-piece being treated. Rectification can be done in several ways, for example, by the addition of boric acid (2.4 %), which converts the oxy-chlorides into a sludge which can be periodically removed.
Decarburization can be detected by surface hardness tests, by the use of test files or by the microscopic examination. The degree of rectification can be checked by heating and quenching high-carbon test-pieces. Any sign of low surface hardness indicates the need for rectification. Since the salts used for the medium temperature bath are available commercially, it is desired to follow the recommendations given by the salt manufacturers for the addition of rectifiers.
It is important that the bath is covered by a layer of pure graphite powder for preventing the heat losses, ensuring thereby a reducing atmosphere. A salt containing a mixture of 20 % to 30 % sodium chloride and 70 % to 80 % of barium chloride, with a melting point of 640 deg C, can be used at temperature ranges between 750 deg C and 950 deg C. This salt mixture can be used for treatment of carbon and low-alloy steels, hardening of precision tools, dies, gauges and other steel parts which are finished to size prior to hardening, and not subjected to further operations. It is desired to cover the bath with pure graphite powder for increasing the pot life and reducing the quantity of rectifiers to be added.
For maintaining the salt bath in the original condition, it is necessary to remove scales or rust present on the steel parts. Presence of rust or scales on work-piece leads to an excessive quantity of oxides in the melt, necessitating an increased addition of the rectifying agent. However, it can be de-oxidized by the addition of borax. Activated carbon is also used as a cover to form an insulating blanket, for reducing the heat losses and to prevent decarburization.
A salt containing a mixture of 80 % barium chloride and 20 % sodium chloride, with a melting point of 750 deg C, is used in the temperature range of 810 deg C to 1,100 deg C for hardening high alloy steels. Rectification or deoxidation of the bath is done by adding 4 % to 5 % borax.
A salt mixture of 50 % barium chloride and 50 % sodium chloride, having a melting point of 660 deg C, can be used in the working range of 700 deg C to 1,000 deg C for hardening of tool steels and constructional steels. This salt can also be used as a pre-heating salt for hot-work steels and high-speed steels. Deoxidation of the bath is done by adding borax.
Salt mixtures of 50 % to 60 % barium chloride, 15 % to 25 % sodium chloride and 20 % to 30 % potassium chloride with a melting point of 620 deg C, are used in the temperature range of 650 deg C to 1,000 deg C. Such mixtures can be used for hardening, high temperature tempering, normalizing, and also along with a carburizing salt for the preparation of the bath. Carburized components can be annealed for machining purposes using this salt bath.
High temperature salt baths have a temperature range of 1,000 deg C to 1,300 deg C. Barium chloride salts with a melting point of 960 deg C are normally used in the temperature range of 1,000 deg C to 1,300 deg C. Barium chloride and 3 % to 5 % magnesium fluoride having a melting point of 950 deg C, can be used in the temperature range of 1,050 deg C to 1,300 deg C. The rectifiers used for de-oxidation are borax, ferro-silicon, and magnesium fluoride. In the first 6 hours to 8 hours of operation, there is no tendency towards oxidation of the molten salt.
Sometimes the salts are mixed with a sufficient quantity of rectifiers. However, for the rectification of bath salts in the course of operation, it is necessary to add 0.6 % to 0.8 % borax, 1 % to 2 % of ferro-silicon or 0.6 % -0.8% magnesium fluoride per shift of 8 hours. The addition of magnesium fluoride provides full protection to the bath, the electrodes, and also to the bath-lining. In fact, ferro-silicon can cause erosion of bath-lining and electrodes. For this reason, it is better to add both ferro-silicon and borax in identical proportion. These salts can also be used for treatment of hot-work steels. The decarburizing tendency of the bath is aggravated by high temperatures. Decarburizing can be checked by treating a test piece, quenching it in oil and examining it with a file. A file soft surface indicates the need for more rectification.
The initial investment cost of a salt bath furnace is low compared with the other heating methods. Salt bath provides uniform heating for all the work-piece in the furnace, whereas in atmospheric furnaces, the parts in the centre of the charge lag considerably behind those on hearth or on the outer edges of the charge in reaching the desired temperature. The heat-transfer rate from salt to metal is much higher than that obtainable by heat radiation in atmospheric furnaces. When the charge is withdrawn, it is coated with a thin film of liquid salt to protect the metal oxidation while being transferred to the quench. Conduction heating in salt bath is more rapid and uniform than radiation heating. Yet, no thermal shock is encountered.
Work handling in salt bath is especially well supported because of the high density of the medium. It lowers the apparent weight of the work-piece, hence, the tendency towards bending distortion is reduced. Long, slender components can be hung and quenched vertically and this helps in preventing distortion. Because of the stirring action of the salt bath, the temperature throughout the bath is uniform. The salt-bath temperature can be controlled accurately. Salt baths are ideal for the hardening of die steels and high-speed steel, since they are capable of being quenched very rapidly below the pearlitic nose, hence assuring high hardness.
The disadvantage in salt baths is the disposal of dross and effluent containing barium waste. A separate room is to be provided for storing the salts. Salt baths are very corrosive. Corrosion leads to the deterioration of the plant, since the salts are hygroscopic. Salt fumes condensed on the equipment produce thin films which, when cooled, rapidly hydrate and cause accelerated rusting. Salt baths call for a good ventilation system. Furnaces are to be provided with a separate exhaust system to suck the fumes from the bath during operation. Operating personnel are to be subjected to periodical medical examination. An electrically heated pre-heating furnace is necessary to pre-heat the work-piece in air for removing the water which can be present, and for eliminating the risk of salt spitting.
Muffle furnaces – Endothermic atmospheres muffle furnaces and sealed quench furnaces mainly use either an endothermic or exothermic atmosphere. An endothermic atmosphere is produced through a suitable generator using an appropriate gas and air. The product obtained consists of mainly nitrogen, hydrogen, and carbon-monoxide, with small quantities of carbon-dioxide and methane.
The gas normally used is propane which is mixed carefully in a balanced proportion with air, and passed into a chamber which is filled with a nickel-bearing catalyst. The chamber is heated externally to a temperature of around 1,040 deg C. The reaction takes place in an endothermic atmosphere which is also called the carrier gas. The carrier gas has a carbon potential of 0.4 %, hence a majority of structural steels can be heated in direct contact with this gas, without any risk of change in the carbon content on their surface.
Steels containing a higher percentage of carbon, such as carbon tool steels and high-speed steels, can be hardened by increasing the carbon potential by the addition of propane directly into the furnace. The needed potential of the carrier gas is to be the same as that of the carbon dissolved in the steel at the hardening temperature. It is very important to carefully control the carbon potential in the furnace, using either the dew point or infra-red radiation method for avoiding decarburization or carburization.
The exothermic atmosphere is produced by combining a fuel gas (natural gas, or liquefied petroleum gas etc.) with varying quantities of air in a suitable generator to get a range of gas mixtures. Heat is developed because of the reaction of the gases and hence it is called exothermic. The resultant gas contains a mixture of hydrogen and carbon monoxide (accounting for 25 %), 5 % to 12 % carbon dioxide, and the balance is made up by nitrogen plus water vapour. In cases where carbon dioxide and water vapour are detrimental, it is necessary to remove them, normally either by absorption or through molecular sieves. The resultant gas is mainly dry nitrogen, with or without controlled residues of hydrogen and carbon monoxide.
Nitrogen is generated mainly from air by liquefaction. It can also be prepared in an exothermic generator by burning a mixture of air and fuel gas. This gas can be used as a neutral atmosphere in the muffle furnace, sealed-quench furnace, and fluidized furnace etc.
Fluidized bed furnaces – These furnaces can also be used for the hardening operation. The steel parts to be treated are heated in a fluidized bed. The furnace consists of a refractory-lined sheet-shell carrying a retort, which is normally filled with particles of fine-grained aluminium oxide. Aluminium oxide particles are fluidized by a controlled flow of gas blown upwards through the base of the retort. The controlled flow of gas blown upwards through a porous bottom of the retort creates a fluidizing effect such that the bed acts like a fluid. When heat is applied the bed provides a rapid heat transfer medium, almost similar to that of molten salt. The fluidized gas performs two tasks simultaneously namely (i) it creates the operating conditions of the fluidized bed, and (ii) it also determines the furnace atmosphere.
Heating of the furnace can be external or internal, by electricity or gas. The heating medium used for hardening is neutral gas (nitrogen). Use of nitrogen provides a bright surface which does not need decarburization. Fluidized furnaces are ideal for the production situations such as those needed in tool rooms.
Vacuum furnace – Hardening, using a vacuum medium, is gaining importance, particularly for hardening tool steels and super-alloys since the vacuum medium confers better hardening properties on components, together with less distortion and a brilliant surface appearance. The essential difference between vacuum hardening and other hardening methods is that it provides absolute neutral conditions with little effort, and prevents the occurrence of surface reactions such as oxidation, decarburization, and carburization etc. during hardening. A vacuum pressure range of 1 micrometre to 500 micrometres is needed for the hardening treatment.
Quenching – Quenching refers to the process of rapidly cooling steel parts from the austenitizing or solution treating temperature, typically from within the range of 815 deg C to 870 deg C for carbon and low alloy steel. Stainless and high-alloy steels can be quenched for minimizing the presence of grain boundary carbides or to improve the ferrite distribution but the majority of steels including carbon, low-alloy, and tool steels, are quenched to produce controlled quantities of martensite in the micro-structure.
Successful hardening normally means achieving the needed micro-structure, hardness, strength, or toughness while minimizing residual stress, distortion, and the possibility of cracking. For preventing cracking, surface stresses after quenching are to be either compressive or at a relatively low tensile level. In general, the use of a less drastic quenching suited to the hardenability of the steel results in lower distortion and higher freedom from cracking.
The selection of a quenching medium depends on the hardenability of the particular steel, the section thickness and shape involved, and the cooling rates needed to achieve the desired micro-structure. The most common quenching media are either liquids or gases. The liquid quenching media which is normally used include (i) oil which can contain a variety of additives, (ii) water, (iii) aqueous polymer solutions, and (iv) water which can contain salt or caustic additives. The most common gaseous quenching medias are inert gases including helium, argon, and nitrogen. These quenching medias are sometimes used after austenitizing in a vacuum.
The ability of a quenching media to harden steel depends on the cooling characteristics of the quenching media. Quenching effectiveness is dependent on the steel composition, type of quenchant, or the use conditions of the quenching media. The design of the quenching system and the thoroughness with which the system is maintained also contribute to the success of the process.
Fundamentally, the objective of the quenching process is to cool steel from the austenitizing temperature sufficiently fast to form the desired micro-structural phases, sometimes bainite but more frequently martensite. The basic function of the quenching media is to control the rate of heat transfer from the surface of the steel part being quenched. For minimizing distortion, long cylindrical objects are to be quenched vertically, flat sections edgeways, and thick sections are to enter the bath first.
The rate of heat extraction by a quenching media and the way it is used substantially affects the performance of the quenching media. Variations in quenching practices have resulted in the assignment of specific names to some quenching techniques such as (i) direct quenching, (ii) time quenching, (iii) selective quenching, (iv) spray quenching, (v) fog quenching, and (vi) interrupted quenching.
Direct quenching refers to quenching directly from the austenitizing temperature and is by far the most widely used practice. The term direct quenching is used to differentiate this type of cycle from more indirect practices which can involve carburizing, slow cooling, reheating, followed by quenching.
Time quenching is used when the cooling rate of the steel part being quenched needs to be abruptly changed during the cooling cycle. The change in cooling rate can consist of either an increase or a decrease in the cooling rate depending on which is needed to achieve the desired results. The normal practice is to lower the temperature of the steel part by quenching in a medium with high heat removal characteristics (for example, water) until the steel part has cooled below the nose of the time-temperature-transformation (TTT) curve, and then to transfer the steel part to a second medium (for example, oil), so that it cools more slowly through the martensite formation range. In some applications, the second medium can be air or an inert gas. Time quenching is very frequently used to minimize distortion, cracking, and dimensional changes.
Selective quenching is used when it is desirable for certain areas of a part to be relatively unaffected by the quenching media. This can be accomplished by insulating an area to be more slowly cooled so the quenching media contacts only those areas of the steel part which are to be rapidly cooled.
Spray quenching involves directing high-pressure streams of quenching liquid onto areas of the work-piece where higher cooling rates are desired. The cooling rate is faster since the droplets of the quenching media formed by the high-intensity spray impact the surface of the steel part and remove heat very effectively. However, low-pressure spraying, in effect a flood-type flow, is preferred with certain polymer quenching medias.
Fog quenching utilizes a fine fog or mist of liquid droplets in a gas carrier as the cooling agent. Although similar to spray quenching, fog quenching produces lower cooling rates because of the relatively low liquid content of the stream.
Interrupted quenching refers to the fast cooling of the steel from the austenitizing temperature to a point above the Ms temperature where it is held for a specified period of time, followed by cooling in air. There are three types of interrupted quenching namely (i) austempering, (ii) marquenching (martempering), and (iii) isothermal quenching. The temperature at which the quenching is interrupted, the length of time the steel part is held at temperature, and the rate of cooling can vary depending on the type of steel and work-piece thickness. Comparisons of direct and interrupted quench cycles are shown in Fig 6.
Fig 6 Comparison of cooling rates and temperature gradients
Austempering consists of fast cooling the steel part from the austenitizing temperature to around 230 deg C to 400 deg C (depending on the transformation characteristics of the particular steel involved), holding at a constant temperature for allowing isothermal transformation, followed by air cooling. Austempering is applicable to the majority of the medium-carbon steels and alloy steels. Low-alloy steels are normally restricted to 9.5 mm or thinner sections, while more hardenable steels can be austempered in sections up to 50 mm thick. Molten salt baths are normally the most practical for austempering applications. Oils have been developed which suffice in some cases, but molten salts possess better heat-transfer properties and eliminate the fire hazard.
The marquenching (martempering) process is similar to austempering in that the work-piece is quenched fast from the austenitizing range into an agitated bath held near the Ms temperature. It differs from austempering in that the work-piece remains at temperature only long enough for the temperature to be equalized throughout the work-piece. When the temperature has achieved equilibrium but before transformation begins, the work-piece is removed from the salt bath and air cooled to room temperature. Oils are used successfully for marquenching, but molten salt is normally preferred because of its better heat-transfer properties. Cooling from the marquenching bath to room temperature is normally conducted in still air. Deeper hardening steels are susceptible to cracking while martensite forms if the cooling rate is very fast. Alloy carburizing steels, which have a soft core, are insensitive to cracking during martensite formation, and the rate of cooling from the Ms temperature is not critical. Marquenching does not remove the necessity for subsequent tempering. The structure of the metal is essentially the same as that formed during direct quenching.
Isothermal quenching is also similar to austempering in that the steel is rapidly quenched through the ferrite and pearlite formation range to just above Ms temperature. However, isothermal quenching differs from austempering in that two quench baths are used. After the first quench, and before transformation has time to begin, the work-piece is transferred to a second bath at a somewhat higher temperature where it is isothermally transferred, followed by cooling in air.
Methods of quenching – After the component is held at the hardening temperature for a desired length of time, it is taken out for cooling, or quenching, in order to get a hard martensite structure. The rate of cooling is required to be controlled, so that the formation of soft pearlite or bainite is prevented.
The medium used for quenching depends upon the chemical composition of the steel, the hardness needed, permissible degree of distortion, and the complexity of the component. Water, oil, brine, molten salt and polymer quenching medias are some of the quenching media in use.
The grade of steel, section thickness, distortion allowed, and the properties to be imparted to the steel part govern the method to be adopted. The different quenching methods are (i) direct quenching, (ii) martempering, (iii) austempering, (iv) delay quenching, (v) time quenching, and (vi) die quenching.
In direct quenching method, steel parts held at the hardening temperature for the needed length of time are directly quenched in water or oil, for getting the hard martensite structure. The disadvantages of this method are (i) higher distortion and, (ii) presence of cracks because of the very high rate of cooling in the martensite range. However, this method is deemed suitable for mild steels, low-carbon and medium-carbon steels.
The main object of martempering is to minimize distortion and to eliminate cracking during hardening. It is to be noted that martempering cannot prevent volume changes which are inevitable in any hardening treatment. However, it can help to reduce erratic changes to a minimum, so that the minimum finishing allowances can be maintained on the steel parts. In this method, the steel part is heated to the hardening temperature in the normal manner, and quenched in a molten salt-bath, instead of using water or oil. The temperature of martempering bath is kept a little above the Ms temperature point of steel being treated. The steel part is soaked in the bath until its temperature is uniform throughout, but the steel part is not kept long enough for transformation into softer bainite. The steel part is then taken out of the bath and allowed to cool in air.
Austempering consists of heating steel to above the Ac3 transformation temperature and then quenching into a hot bath held at a temperature below that at which fine pearlite forms (the nose of the S-curve), but above the Ms temperature. The product of isothermal decomposition of austenite in this temperature region is bainite. This constituent combines relatively high toughness and hardness. A typical micro-structure of austempered steel is shown in Fig 3e. This austempering process has certain limitations which make it impracticable for use with several steels. In order to ensure a uniform structure (and hence uniform properties), it is essential that the entire cross-section of the steel be cooled fast enough so that even the centre escapes transformation at the nose of the S-curve.
In carbon steels the time needed to start transformation at the nose of the S-curve is extremely short, so that only relatively small sections (about % mm maximum thickness) can be successfully hot quenched in austempering baths. The time needed for transformation of the austenite of alloy steels to fine pearlite is normally longer, and hence larger sections can be successfully austempered (around 25 mm maximum). However, the time needed for transformation to bainite frequently becomes inordinately long with several alloy steels and the process of austempering, hence, is normally impracticable for these steels.
Delay quenching is the term applied to a quenching process in which the steel parts, after removal from the hardening bath, are quenched at a suitable low temperature, resulting in a minimum of distortion. The method can be applied to the high-speed steels, hot-worked steels, and case-hardened steels.
Time or interrupted quenching method is mainly adopted for steels of low hardenability which need water-quenching, and for those of higher hardenability when the steel parts are big. It minimizes internal strains, resulting reduced distortion and breakage during or after hardening. Steel parts subjected to this method yield a soft and rough core with a progressively tougher and hardened outer zone.
Steel parts treated by this method are quenched from the austenitizing temperature in a suitable quenching medium until the temperature is 60 deg C to 80 deg C below the Ms temperature. They are withdrawn after a fixed time before they achieve the bath temperature throughout the sections. Finally, they are quenched in an oil bath or cooled in air. The initial quenching produces a martensite structure to a depth depending on the time allowed in the quenching bath. Since cooling continues at a much-reduced rate after the steel parts is withdrawn, the core gets transformed into a non-martensite structure. The martensite formed at the outer rim is tempered by the heat stored in the interior of the steel part. The process yields a soft and tough core, with a progressively toughened and hardened outer part. The structure of the hardened surface is like that achieved in continuous hardening.
Die quenching method can be used for thin disks, gears, flats, long slender rods, and other delicate parts which get excessively distorted when they are quenched in a conventional liquid medium. The process is best adopted for symmetrical objects. Normally, in steel parts having irregular sections, the transformation to martensite in quenching tends to produce warping. If good contact can be made between the hot austenite steel and anything else which abstracts heat rapidly, quenching can be accomplished. By pressing the austenitized steel part between closely fitting dies some of the warpage can be mechanically removed.
Role of alloying elements in quenching – Since the work-pieces treated are frequently relatively big and since the alloying elements have the general effect of lowering of the temperature range at which martensite is formed, the thermal and transformational stresses set up during quenching tend to be higher in the alloy steel work-pieces than those encountered in quenching of the smaller work-pieces of plain carbon steels. In general, the higher stresses result in distortion and risk of cracking.
Alloying elements, however, have two functions which tend to offset these disadvantages. First and probably most important is the capacity to permit use of lower carbon content for a given application. The decrease in hardenability accompanying the decrease in carbon content can be readily offset by the hardenability effect of the added alloying elements and the lower carbon steel shows a much lower susceptibility to quench cracking. This lower susceptibility results from higher plasticity of the low carbon martensite and from the normally higher temperature range at which martensite is formed in the lower carbon materials. Quench cracking is seldom encountered in steels containing 0.25 % carbon or less, and the susceptibility to cracking increases progressively with increasing carbon content.
The second function of the alloying elements in quenching is to permit slower rates of cooling for a given section, because of increased hardenability, hence normally decreasing the thermal gradient and, in turn, the cooling stress. It is to be noted, however, that this is not altogether advantageous, since the direction, as well as the magnitude, of the stress existing after the quench is important in relation to cracking.
Tempering of steel – It is a process in which previously hardened or normalized steel is normally heated to a temperature below the lower critical temperature and cooled at a suitable rate, mainly to increase ductility and toughness, but also to increase the grain size of the matrix. Steels are tempered by reheating after hardening to achieve specific values of mechanical properties and also to relieve quenching stresses and to ensure dimensional stability. Tempering normally follows quenching from above the upper critical temperature. However, tempering is also used to relieve the stresses and reduce the hardness developed during welding and to relieve stresses induced by forming and machining.
Tempering during manufacture imparts shock resistance with only a slight decrease in hardness. The micro-structural changes accompanying tempering include loss of acicular martensite pattern and the precipitation of tiny carbide particles. This micro-structure is referred to as tempered martensite.
Principal variables associated with tempering which affect the micro-structure and the mechanical properties of a tempered steel include (i) tempering temperature, (ii) time at temperature, (iii) cooling rate from the tempering temperature, and (iv) composition of the steel, including carbon content, alloy content, and residual elements.
In a steel quenched to a micro-structure consisting necessarily of martensite, the iron lattice is strained by the carbon atoms, producing the high hardness of quenched steels. Upon heating, the carbon atoms diffuse and react in a series of distinct steps that eventually form cementite or an alloy carbide in a ferrite matrix of gradually decreasing stress level. The properties of the tempered steel are mainly determined by the size, shape, composition, and distribution of the carbides which form, with a relatively minor contribution from solid-solution hardening of the ferrite. These changes in micro-structure normally decrease hardness, tensile strength, and yield strength but increase ductility and toughness.
Under certain conditions, hardness can remain unaffected by tempering or can even be increased as a result of it. For example, tempering a hardened steel at very low tempering temperatures can cause no change in hardness but can achieve a desired increase in yield strength. Also, those alloy steels which contain one or more of the carbide-forming elements (chromium, molybdenum, vanadium, and tungsten) are capable of secondary hardening, i.e., they can become somewhat harder as a result of tempering.
Temperature and time are inter-dependent variables in the tempering process. Within limits, lowering temperature and increasing time can normally produce the same result as raising temperature and decreasing time. However, minor temperature changes have a far higher effect than minor time changes in typical tempering operations. With few exceptions, tempering is done at temperatures between 175 deg C and 705 deg C and for times from 30 minutes to 4 hours.
Based on x-ray, dilatometric, and micro-structural studies, there are three distinct stages of tempering, even though the temperature ranges overlap. The first stage is the formation of transition carbides and lowering of the carbon content of the martensite to 0.25 % (100 deg C to 250 deg C). The second stage is the transformation of retained austenite to ferrite and cementite (200 deg C to 300 deg C). The third stage is the replacement of transition carbides and low-temperature martensite by cementite and ferrite (250 deg C to 350 deg C). An additional stage of tempering (fourth stage, precipitation of finely dispersed alloy carbides, exists for high-alloy steels. It has been found that the first stage of tempering is frequently preceded by the redistribution of carbon atoms, called auto-tempering or quench tempering, during quenching and / or holding at room temperature. Other structural changes take place because of carbon atom rearrangement preceding the classical first stage of tempering.
Martensite transformation is associated with an increase in volume. During tempering, martensite decomposes into a mixture of ferrite and cementite with a resultant decrease in volume as tempering temperature increases. Since a 100 % martensitic structure after quenching cannot always be assumed, volume cannot continuously decrease with increasing tempering temperature.
The retained austenite in plain carbon and low-alloy steels transforms to bainite with an increase in volume, in second stage of tempering. When certain alloy steels are tempered, a precipitation of finely distributed alloy carbides occurs, along with an increase in hardness, called secondary hardness, and an increase in volume. With the precipitation of alloy carbides, the Ms temperature (temperature at which martensite starts to form from austenite upon cooling) of the retained austenite increases and transform to martensite during cooling from the tempering temperature.
Several empirical relationships have been made between the tensile strength and hardness of tempered steels such that the measurement of hardness is commonly used to evaluate the response of a steel to tempering. Fig 7 shows the effect of tempering temperature on tensile and yield strengths, elongation, and reduction in area of a plain carbon steel (AISI 1050) held at temperature for 1 hour. It can be seen that both room temperature hardness and strength decrease as the tempering temperature is increased. Ductility at ambient temperatures, measured by either elongation or reduction in area, increases with tempering temperature.
Fig 7 Effect of tempering temperature on steel properties
Majority of the medium-alloy steels show a response to tempering similar to that of carbon steels. There is no decrease in ductility in the temperature range of tempered martensite embrittlement (also known as 260 deg C embrittlement or one-step temper embrittlement) since the tensile tests are performed on smooth, round samples at relatively low strain rates. However, in impact loading, catastrophic failure can result when alloy steel is tempered in the tempered martensite embrittlement range (260 deg C to 370 deg C).
Whereas elongation and reduction in area increase continuously with tempering temperature, toughness, as measured by a notched-bar impact test, varies with tempering temperature for the majority of steels, as shown in Fig 7. Tempering at temperatures from 260 deg C to 320 deg C decreases impact energy to a value below the value achieved at around 150 deg C. Above 320 deg C, impact energy again increases with increasing tempering temperature. Both plain carbon and alloy steels respond to tempering in this manner. The phenomenon of impact energy minima centered around 300 deg C is called tempered martensite embrittlement or 260 deg C embrittlement.
The diffusion of carbon and alloying elements necessary for the formation of carbides is temperature and time dependent. The effect of tempering time on the hardness of a steel tempered at different temperatures is shown in Fig 5b. The changes in hardness are around linear over a large portion of the time range when the time is presented on a logarithmic scale. Fast changes in room-temperature hardness occur at the start of tempering in times less than 10 seconds. Less rapid, but still large, changes in hardness occur in times from 1 minute to 10 minutes, and smaller changes occur in times from 1 hour to 2 hours. For consistency and less dependency on variations in time, steel parts are normally tempered for 1 hour to 2 hours. The levels of hardness produced by very short tempering cycles, such as in induction tempering, are quite sensitive to both the temperature achieved and the time at temperature.
By the use of an empirical tempering parameter developed by Holloman and Jaffe, the approximate hardness of quenched and tempered low-alloy and medium-alloy steels can be predicted. The parameter is ‘T (c + log t)’, where ‘T’ is the temperature in degrees Kelvin, ‘t’ is the time in seconds, and ‘c’ is a constant which depends on the carbon content of the steel. Reasonably good correlations are achieved except when considerable quantities of retained austenite are present.
Another factor which can affect the properties of a steel is the cooling rate from the tempering temperature. Although tensile properties are not affected by cooling rate, toughness (as measured by notched-bar impact testing) can be decreased if the steel is cooled slowly through the temperature range from 375 deg C to 575 deg C, especially in steels which contain carbide-forming elements. Elongation and reduction in area can be affected also. This phenomenon is called temper embrittlement.
Leave a Comment