Composition and heat treatment of steel

Preface

In preparing the matter that enters into this book no attempt has been made to go into details on the subjects of the ores, their melting down into iron, or refining the iron in steel making. This part has merely been covered in a general way in order to lead up to and give a better understanding of the effect of the elements present in and added to steels of the various grades and kinds.

An attempt has been made to cover all the materials that have been used, either commercially or experimentally, for the purpose of making better steel and improving the standard brands so they will have greater strengths; withstand strains and stresses better; possess a longer wearing surface; have a greater electrical resistance, conductivity, or magnetism; attain a greater hardness, ductility, resiliency, or malleability; be capable of taking larger cuts on other metals or machining them faster; produce a metal that can be easier rolled, hammered, pressed, drawn, forged, welded, or machined into shape; be non-corrosive, or, in fact, make a better metal for any of the many uses to which it is put. The effect these materials or elements have had upon the carbon and alloyed steels has been told as well as the data at hand would permit, and hints have been injected, as to what might be expected from many of the elements, in order to stimulate further investigations and experiments. Results have been obtained in this way in the very recent past that are truly wonderful, yet these are liable to sink into insignificance before the discoveries that may be made in the near future.

The different chemical compositions that can be made from the elements here Hsted and described are so numerous that it seems hopeless to expect that all of them will ever be compounded, and tests made, and the results recorded. However, with the many individuals that are working along these lines some combinations are bound to be made that will prove to be beneficial, and doubtless some steels will be produced that will cause as great a revolution as "Mushet" or "Bessemer" steels did in their respective lines. A verj^ few of the possible quaternary steels have been tried, i.e., alloys made by combining four different elements with the ferrite, and therefore many are yet to be investigated in the many different percentages in which it is possible to combine them. And this does not take into consideration the compositions that are possible with six, eight or more elements.

V vi PREFACE

Following the ingredients of and materials used in steel, comes the heat-treatment, as the two have moved along parallel Hnes, in the many investigations, experiments, and improvements that have been made, and seem to be inseparable. Each change in composition seems to have altered the heat-treatment, and each improvement in heat-treatment seem to have altered the percentage, that is best to use, of some one or more element. Many different methods and various kinds of materials have been experimented with and consequently a great deal of useful information has been obtained and many improvements of a radical nature made. New methods, new materials, and new apparatus have thus been brought into use for the heat-treatment of steel. These have enabled the hardener to get more definite, positive, and uniform results, and in this way the metal has been improved to a great extent.

All of the information that could be obtained on this phase of steel making and working has therefore been recorded as carefully as possible. This also suggests ideas that would indicate that there is still room for important improvements or discoveries. One of these is the attaching of a positive and negative wire of an electrical circuit to the piece of steel to be hardened and place it in a quenching bath. The current can then be turned on, the piece heated, the current turned off, and the piece quenched without moving it or allowing the air to strike the metal and oxidize it. Another instance is the possibilities suggested by carbonizing steel with gases or chemicals and thus doing away with the old laborious method of packing the steel pieces in bone and charcoal. Still another is the 30-minute annealing of high speed steel and the possibility of a similar method being applied to carbon steel.

In gathering together the data necessary to add to my own, very little credit has been given to individuals, as to make this correct is not only a laborious but a hopelessly impossible task. To illustrate this I have seen professors claim as their own discoveries, new principles, new methods, etc., that were developed and perfected by students in their classes, and shop foremen and superintendents claim as theirs, inventions made by men in the shop. Two important discoveries that developed into new kinds of steel were made through the mistakes of workmen in steel mills. Two men on the same job added the correct percentage of a material and thus this element was twice as large as it was thought would give good results. In fact, it was believed that it would injure the metal to add more than a certain percentage, but when this maximum percentage was doubled the metal was given properties that were very beneficial for certain purposes. None of us can add but a mite to the knowledge that we have obtained from others and because we are enabled to write it so it will be recorded in books and papers does not give us the privilege of claiming to be the originators of certain ideas,

PREFACE vii

principles, discoveries, or inventions. Every one who has worked in the 4teel mill or the laboratory is entitled to a part of the credit for any new ideas or information that may happen to be enclosed between the two covers of this book. To pick out a few individuals and give less credit than this would be working an injustice and stating an untruth, and to name all that should be given credit is a physical impossibility.

E. F. Lake.

Contents

CHAPTER PAGE

I The Making of Pig Iron 1-12

Ordinary blast furnace reduction of ores, 1-3; Use of excess gas, 4; Conveying molten metal in ladle cars, 5; Making pig beds in sand floor, 6; Casting pigs in iron molds, 7-8; Electric blast furnace, 9-12.

CHAPTER I The Making of Pig Iron

The iron that forms the base for all steel, as well as iron, products is first obtained from its ores, as a commercial product, from a blast furnace similar to that shown in Fig. 1. It is then in the form of an iron that contains a large amount of carbon, both in the graphitic and combined state. This makes it too weak and brittle for most engineering purposes, but about one-third of the total product is run out of the blast furnace into pigs of iron that is used only for castings that are to be subjected to compressive, transverse or very slight tensile strains, such as bed plates or supporting parts for machinery, water pipes, stove plates, car wheels, etc.

The various kinds of steels are relatively increasing in proportion to the amount of pig iron used. To-day about two-thirds of this product is being turned into steel through purification by either the Bessemer, open-hearth, puddling, crucible or electric methods. The carbon content is reduced to any desired point, the graphitic carbon being eliminated by any of these processes, and the silicon and manganese are oxidized out by the accompanying reactions, or as a condition precedent to the reduction of the carbon. The two impurities of the metal which are the greatest bane to engineers and steel makers alike are phosphorus and sulphur. These are reduced by either the basic open-hearth, puddling, crucible, or electric processes.

In making steel, the operation begins by making pig iron from the iron ore, which is a natural iron rust or a combination of iron and oxygen. The oxygen is removed by combining iron ore, coke, and limestone in a furnace, as shown in Fig. 2, and heating them to a high temperature by injecting superheated air into the bottom of the furnace. The coke is burned by the oxygen in the air; a part of it aids in maintaining this high temperature while the rest is useful in removing the oxygen from the ore.

This superheated air is usually heated with the gases which have been conducted from the top of the furnace to the bottom through the pipe shown to the left of the furnace in Fig. 2, and passed through a hot blast stove. Four of these stoves are shown grouped in pairs, to the left of the blast furnace, in Fig. 1. They are about the same hight as the furo a

o3 m o

l-H

^^^ Coke

Iron Ore O Lime

Drops of Slag Drops of Ii-on Molten Slag

=-Sj=_?. Molten Iron

^iJffi

Fig. 2. — Details of blast furnace. Condition of charge at different

levels.

nace, which is something over 100 feet, and are round steel tanks that have a comparatively small annular fire-brick chamber in the center for nearly the hight of the tank. This chamber is surrounded with brick work that is filled with flues. The gas from the furnace comes in at the bottom of the central annular chamber; burns on its passage up this; comes down through the flues in a heated condition, thus warming up the brick work, and then passes out the chimney as waste product.

When the brick work is heated properly, the gas from the furnace is turned off and the blower engine turned on to blow a blast of air through the flues into the furnace. This heats the air in its passage up through the central chamber and down through the flues, and makes it a hot blast when it enters the tuyeres of the furnace. Thus it does not have a coding effect on the materials being melted in the blast furnace. Four hot stoves are used with each furnace, so that three can be burning gas and warming up, while the fourth is having the air blast sent through it into the furnace.

The gas, which is a product of combustion of the materials in the blast furnace, comes down through the pipe A (Fig. 2), which is called the downcomer, leaves most all of its accumulated dirt at B, and then passes out of the pipe C.

One-third of this gas is all that is needed to keep the stoves hot and the balance was formerly wasted. It is now being sent through gas producers, and used to run the blower engine. While the blast furnace is thus made to furnish its own fuel for power and heat, it is still sending out more gas than can be used for these purposes, and new uses are being found for it in other parts of the iron and steel mills.

The blast furnace is charged from a car that is brought up to the top, on an inclined track, with a cable. The charge is dumped from the car into the top of the furnace through a hopper and bell. This charge consists of coke, iron ore, and limestone, which are dumped into the furnace in the order named. The change that takes place in these as they pass down through the furnace is plainly shown in Fig. 2.

The coke serves as a fuel for generating the heat that melts the iron ore, and the limestone unites with the earthy materials as the ore is being reduced to a molten state. The resultant slag is run off from the top of the iron through a hole in the side of the furnace immediately below the tuyeres. The metallic iron melts and collects in the hearth below this slag, and is tapped out of another hole, close to the bottom. From this it is run through channels into molds that form it into "sows" and "pigs," or the molten metal is tapped from the furnace into ladle cars as shown in Fig. 3, in which it is taken to furnaces for conversion into steel.

While the metal is in contact with the white-hot coke in the furnace it absorbs a certain amount of carbon, some of which is chemically combined with the iron and another part is held in suspension as graphite.

o

If the "sows" and "pigs" are cooled slowly it tends to make the carbon take the form of graphite. When such iron is broken it has a gray or black appearance showing loose scales of graphite, and the iron is soft and tough.

If the metal is cooled quickly, or chilled as soon as it comes from the furnace, the carbon has a tendency to be kept in the combined state. When fractured such metal will be white and hard.

Of this product about 20% is made into gray-iron castings, 3% into malleable-iron castings, 3% is purified in puddling furnaces to make wrought iron, and the balance, or 74%, is converted into steel by the various

Fig. 4. — Section of sand bed for casting pig iron.

processes. Of the latter 52% is converted by the Bessemer process, 19% in the basic open-hearth furnace, 2% in the acid open-hearth furnace, and 1% by the crucible process. Recently the electric furnace has been brought into use, and this promises to take a certain percentage for conversion into the finer grades of steel.

The older method of casting the blast into pigs, and which is used now by some, was to have a casting floor in front of the blast furnace that was composed of silica sand. In this sand, impressions or molds were made for the pigs and these connected with runners called "sows,"

which in turn were connected to a main runner from the furnace. Fig. 4 shows how the fioor was laid out. In front of and under the tap hole of the furnace a small hole, A, was formed into which the iron was run. From this the main runner for the iron ran down the center of the casthouse. Branching off on either side of this were the sows with the pigs leading off from the sows.

In the main runner, immediately back of each sow, a dam was formed, as shown by the two lines at B, and sow C, with its pigs, was run full of iron. Then this was dammed and the stream of molten iron turned into sow D, which when filled was dammed and the stream turned into sow E. This was continued until the whole blast was run off, and then the pigs, sows, and runner were broken up, loaded into cars, and the floor remolded, ready for the next tapping. At the entrance to each pig and at stated intervals in the sows and runner, as shown by the double lines, a dam was formed that about half filled these, so as to make the metal thinner at this point and thus allow it to be broken more easily into nearly standard sizes and weights.

Automatic machines into which the pigs are cast, cooled, and then dumped into cars are now used at most blast furnaces, as the saving in labor over the cost of the machine is a big item; the pigs are more uniform in size, thus facilitating handling, piling, and storing, and they are free from the adhering silicious sand that is especially objectionable in the basic open-hearth furnace.

The pig molding machines are made in several styles, the most common forms of which are a revolving frame with the pig molds in a continuous series around its annular outer edge, as shown in Fig. 5, and a series of molds attached to an endless chain which carries them in a straight line from where they are poured to the cars into which they are dumped, as shown in Fig. 6. In the latter, the empty molds travel back to the ladle underneath the filled ones, and in both the molds are sprayed with thick lime water, long enough before they are filled to allow the water to be dried out by the heat of the mold, and leave it covered with a coating of lime so the molten metal will not stick to the cast-iron mold.

ELECTEIC BLAST FURNACE

The experiments that have for some time been carried on for the electric production of pig iron seem to be fast approaching a successful culmination, and we may in the near future see this method used commercialljr, especially where an adequate water power is available.

The Noble Steel Company in California were the pioneer experimenters in this country. Their first attempt was a 1500 kilowatt, three-phase, resistance type of furnace that was completed in July, 1907, but after running it a short time the mechanical difficulties which presented themselves made this type of furnace impractical commercially, and it was abandoned.

A 160 kilowatt furnace of a different type was then constructed and run for 40 days. From this run data were gathered that were used in the construction of the present 1500 kilowatt furnace, shown in Fig. 7. The data obtained would indicate that with one ton of charcoal, costing about $9, three tons of pig iron could be produced with about 0.25 electric horse-power-year per ton.

The quantity of carbon used for the electric smelting of iron ores is only about one-third of that required for the ordinary blast furnace.

Fig. 6. — Double strand, endless chain, pig casting machine.

Thus charcoal can be used and the product will be charcoal pig iron, which, owing to its comparative purity, would demand a higher price than the ordinary product.

In the furnace shown in Fig. 7, the ore with its proper fluxing materials is brought in, in cars, and fed into the preheater, A, where it is dried and heated by the products of combustion piped from the combustion chamber of the furnace, B, through the flue, C. After drying it is dumped in the scale car, D, which runs around the top of the stack, on a circular track, so it can alternately take a charge of ore and flux from A, and carbon from the hopper, E, weigh them and charge them into the furnace in the proper proportions through the usual hopper and bell in the top of the stack.

Six electrodes (G) are arranged equidistantly around the furnace, and the electric current passing through between them melts the charge, the metal and slag collecting in the crucible at the bottom of the furnace, from which they are drawn as in ordinary practice. All the necessary heat is supplied by electrical energy, and thus no blast is blown in. This causes all of the solid carbon to be used for reduction, excepting of course the small amount that is dissolved into the pig iron. Above the level of the charge, however, are small openings at F, for admitting the correct

Fig. 7. — Electric pig-iron furnace at Noble Steel Co., Heroult, Cal.

amount of air, through valves, to burn the gases that result from the reduction of the ores in the lower part of the furnace.

In Fig. 8 is shown the combination of electric and blast furnace that has resulted from several years of study and experiments conducted by three Swedish engineers at. the Domnarfvet Iron Works in Sweden. In this a large crucible with an arched roof is formed at /. An opening is left in the center of the roof, and over it is constructed a stack (J) very similar to the ordinary blast furnace; in fact, the only difference being that the bosh is contracted more at K, where the charge enters the crucible. This was made necessary by the fact that too large an opening would not retain the required heat in the crucible part of the furnace without increasing the power to generate the electrical energy to too high a point. It was feared that this contraction would result in the furnace clogging

Fig. 8. — Electric furnace at Domnarfvet Iron Works, Sweden.

and an overhang form in the bosh, but with a continuous run of three months no such condition was apparent.

Three electrodes (L) that project into the crucible through a water-jacketed stuffing-box in the arched roof conduct a three-phase alternating current of 40 volts to the charge. The current is from 8000 to 9500 amperes, and the load 480 to 500 kilowatts. The arched roof over the crucible gave considerable trouble in the earlier experiments by being overheated, but this is now preserved by taking the gaseous products of combustion from the top of the furnace at M, and blowing them back through tuyeres at N, which are provided with peep-holes so the roof can be examined and the volume of gas increased or diminished as desired.

In the three months' run, that was terminated July 3, 1909, by the general strike in Sweden, it was demonstrated that the electrodes did not need readjusting oftener than once a day, and in one case an electrode was not touched for five days; that the consumption of energy was remarkably uniform even though the short run did not enable the furnace to approach its working condition until near the end; that the charge moved with regularity into the melting chamber; free spaces were maintained underneath the arched roof next to the outer wall, and the gases kept the roof effectively cooled.

This furnace was first constructed as an induction, but was later changed to a resistance furnace. It is started and worked the same as the ordinary blast furnace, and in the experiments so far only coke has been used, but charcoal can be used as well as the "prime ore briquettes" and "slig" which they get in Sweden. It has produced 2 tons of iron per electrical horse-power-year, but conditions would indicate that this could be increased to 3 tons. It being easy to make a pig iron in this furnace with a low carbon content, if the molten iron was transferred directly to electric refining ftirnaces, it would greatly reduce the time consumed in converting it into steel. It now takes a comparatively long time to reduce the carbon to the percentage required in the electric converting furnace when ordinary blast-furnace iron is used.

The carbon in the experiments with the Swedish furnace averaged about 1.80% in the three months' run, while in some previous experiments it ran as high as 3.20%, and in one tapping it was as low as 1%. The silicon varied between 0.20 and 0.07%, but in one case was 4.40%. The sulphur content has been as low as 0.005%, with 0.50% of sulphur in the coke that was used.

The ability of the electric furnace to reduce the impurities to a minimum may result in its becoming a prominent factor in the reduction of the ore, and the conversion of this into steel as soon as experience teaches the operators to control the carbon and silicon, and it is demonstrated that it is practical commercially. In Denmark there will soon be started another ore furnace, and in Canada negotiations are well advanced for an electric iron-ore reduction and steel plant with a capacity of 5000 horse-power.

CHAPTER II Bessemer Process of Converting Iron into Steel

In converting the blast-furnace iron into steel the Bessemer process is the one most used, as it is the cheapest method, although the improvements in the open-hearth method have been such that it is replacing the Bessemer, in some places, for the cheap production of steel, and the product which it turns out is much better.

In the Bessemer process a converter similar to that shown in Figs. 9 and 10 is used, it being pear-shaped and open at the small end. It is hung on trunnions so the metal can be easily poured in and out. Into this is poured the melted pig iron, which is usually taken direct from the furnace, although it is sometimes remelted, and through this is blown cold air in fine sprays in the proportions of about 25,000 cubic feet of cold air per minute to every 10 tons of molten metal, which is the usual charge for a converter.

Curious as this may seem to the uninitiated, the cold air raises the temperature of the molten metal to such a high degree that it is often necessary to inject steam or add scrap to cool off the metal. This rise in temperature is due principally to the silicon in the iron which comes in contact with the oxygen of the air and produces a state of combustion that oxidizes out the silicon and manganese in about four minutes, after which the carbon begins to oxidize to carbon monoxide (CO), which boils up through the metal and is forced out of the mouth of the converter in a long bright flame that gradually diminishes, until at the end of six minutes more the carbon has been reduced to about 0.04% and the flame dies away.

Except for the impurities which poison the metal, namely phosphorus and sulphur, it has become for all practical purposes a pure metal that is very brittle. This makes it necessary to add certain ingredients that will toughen, strengthen, and harden it so as to make it useful and workable.

Carbon is added in different percentages while it is in the molten state to give it the proper degree of hardness and strength.

Manganese is added for the purpose of reducing the sulphur and oxygen which the metal has absorbed during the process of conversion, as the oxygen would render it unfit for use, and the sulphur if present in too large a percentage would make the metal "hot short," and thus difficult to roll or forge when heated. The manganese desulphurizes the molten metal by causing a gradual liquefication of manganese sulphide,

Fig. 9. — Bessemer converter purifying the metal.

which rises to the surface, allowing the sulphur to oxidize into the air. Silicon is also added for the purpose of freeing the metal from blow-

Shoulder

Fig. 10. — Bessemer converter tilted to pour finished metal into ladle.

holes by reducing the gases, as these work to the top under the influence of the chemical action caused by silicon. Of the last two elements but

BESSEMER PROCESS OF CONVERTING IRON INTO STEEL from one-half to two-thirds of the percentages added will be found in the finished product, owing to their combining with the impurities and oxidizing out with them.

These ingredients are added to the charge in the converter by first mixing them with iron in the proper amounts in a small cupola in the converter house, called the spiegel cupola. It is called recarburizing

Bessemer converters at Lackawana Steel Co.

the bath. Pig iron high in manganese, called spiegeleisen, is mixed with other iron high in silicon and carbon until the right percentages of all three are obtained to give the 10 tons of metal in the converter its desired composition. After this the converter is tilted, as shown in Fig. 10, and the steel poured into ladles. These may be tilting ladles, as shown in Fig. 11, or bottom-pour ladles, as shown in Fig. 12. In these, as the name signifies, the metal is poured from the ladle by tilting it or through a hole in the bottom. This latter saves turning the ladle over and also draws the molten metal away from the slag. From the ladles the metal is run into cast-iron ingot molds located on cars, and when they have become solid the ingots are often taken to the rolls and rolled into the desired shapes while still hot. The ingot molds are usually kept going

Fig. 11.— Tilting ladles.

through the converter house in a steady stream; coming in one side empty, getting filled and going out the other side.

Blast-furnace iron with about 1% of silicon is preferred for the Bessemer process, as the lower this is kept the shorter will be the blow, and as it is the chief slag producer it will reduce the iron loss by limiting the amount of slag made. If too low, however, the blow will be cold and take too much time to purify. It is only by working rapidly and allowing no

BESSEMER PROCESS OF CONVERTING IRON INTO STEEL unnecessary wait between blows, thus keeping the converter and ladles very hot, that as low as 1% of silicon in the pig iron can be used successfully.

In the Bessemer process it is almost impossible to reduce the phosphorus, and especially is this so with the acid Bessemer process which is the one used in this country. The percentage of phosphorus which goes into the furnace in the ore and coke will come out in the finished product after it has passed through the converter; hence, it is necessary to use ores and coke that are low in phosphorus if a good steel is to be produced.

In Europe, however, the basic Bessemer converter is used to a certain extent. This is the same as the acid, except that the lining of the

. :if

Fig. 12. — Bottom-pour ladles.

furnace is of some basic material such as limestone, containing from 30 to 40% of magnesia and burned hard, commonly called dolomite. The chemical action of this causes the phosphorus to slag off, but ores which are low in silicon must be used.

The Bessemer process being the cheapest way of converting iron into steel, all of the cheaper and ordinary grades of steel are made by it, such as steel rails, structural shapes, wire, merchant bar, etc.

Where several Bessemer converters are in operation it takes quite a number of blast furnaces to supply them, as it takes two furnaces to furnish iron enough for one converter. As the product of each furnace is liable to vary in chemical composition, it is necessary to have some means of mixing the product of the different furnaces, before they are poured into the converter, if a uniform grade of steel is to be taken out of the converter. For that reason a large reservoir, shaped something like the housewife's chopping bowl, but with a large spout on it, and capable of holding from 200 to 500 tons, is used. Into this is poured the metal from the various furnaces and from it are taken the charges for the cpnverters.