US1292937A - Blast-furnace practice. - Google Patents

Description

H. B. WEAVER & J. GAYLEY.

BLAST FURNACE PRACTICE.

APPLICATION FILED Aue.a. 1918.

1,292,937. Patented Jan. 28, 1919.

2 SHEETS-SHEET 1.

H. B. WEAVER 6r]. GAYLEY.

Patented Jan. 28, 1919.

2 SHEETS-SHEET 2.

5] nvemtozs: 42mm,

Has

UNITED STATES PATENT OFFICE.

HARRY B. WEAVER, 0F CATASAUQUA, PENNSYLVANIA. AND JAMES GAYLEY, OF

NEW YORK, N. Y,

BLAST-FUBNACE PRACTICE.

Specification of Letters Patent.

Patented Jan. 28, 1919.

To all whom it may con ern:

Be it known that we, Haney B. \VEAVER, a citizen of the United States. residing at atasauqua, Lehigh county, State of Pennsylvania. and JAMES GAYLRY, a citizen of the United States. residing at Yo. 555 Park arenue. in the city. county. and State of New York. have invented certain new and useful Imprm'ements in Blast-Furnace Practice; and we do hereby declare the following to he a full. clear and exact description of the invention such as will enable others skilled in the art to which it appertaius to make and use the same.

The eonstit uents of the customary charges regularly fed into an iron blast furnace consist of iron ore, limestone and coke, and contain varying percentages of alkali eom pounds, i. (a, compounds of otassium and sodium. Descending in the furnace from a colder to a hotter temperature these alkali emupounds encounter the highest tempera ture in the zone of combustion in the neighborhood of the twyers where the heated air blast comes in contact with the incandescent and an important chemical reaction results.

The carbon of the coke and the nitrogen of the air blast cannot combine at the existing teunu-rature without the aid of an intermediar The alkali-metals found in the furnace provide the most effective medium for bringing the nitrogen and carbon into combination in the combustion zone and are present in said zone. The fact that potassium cyanid is formed in the blast furnace has been known. in fact, for a long time,

but has been regarded merely as an interesting chemical phenomenon, and, in so far as we are aware, it has not been known that it exists in the blast furnace in sufficient quantity to make it available for any useful purpose. nor that there were any other like compounds present whose withdrawal would be feasible without detriment to the blast furnace operation. where certain essential conditions must be maintained in order to permit the furnace to discharge its required functions in the production of its prescribed metallurgical product, that is to say. (in a pig iron blast furnace) its output of pig iron.

The present invention is based upon the discovery that, in addition to potassium cyanid, there are present in the furnace gases, at and in the neighborhood of the twyers, other compounds in which the alkali metals are associated. and that the potassium cjyanid and these other compounds are present in such quantity in the furnace and in such as' sociatioa with the reducing gases constantly rising from the zone of intense combustion at or near the twyers that it feasible and remunerative to withd raw from the furnace a portion of the said reducing gases and to abstract therefrom by condensation or otherwise the compounds of the alkali metals so as to recover them in la rgre quantity for useful purposes, and without diverting from the blast furnace any such tllnOLTl'll/Of the reducing gases as are required for serving their preparator function for the charge in the upper part of the furnace shaft. Ac cordingly. the invention provides an opcration wherein the blast furnace delivers its normal output of pig iron and a normal slag, preserves the necessary conditions for the required reactions in the combustion zone, permits the reducing gases rising from the combustion zone to exercise the required reducing action in the furnace shaft, and, coincident with these several necessary provisions for the successful operation of the iron blast furnace, permits the withdrawal and recovery, on a commercial scale, of highly raluable compounds of the alkali metals, some of which were not known to exist in the con)- bustion zone of the blast furnace and which, in their entirety, were not known to be present in sutlieient quantity to justify an attempt to remove them. or that paying quantities of any of such compounds were recoverable while still preserving standard blast furnace conditions.

The recovery of these valuable constituents. in an efi'ective way, and without detriment to the normal functioning of the blast furnace is dependent upon factors of the furnace operation which we have taken into careful account in working out the practice constituting the subject-matter of the present invention.

Thus, the compounds of the alkali metals with carbon and nitrogen are formed, as hereinbefore indicated, in the zone of combustion in the neighborhood of the twyers,

where the heated air blast comes in contact with the incandescent fuel. On their formation, they pass upward in the furnace with the large volume of reducing gases, thus meeting the colder charges of materials produced and before any important precipitation or disintegration takes place in their upward travel. This zone of production is located in the neighborhood ofand in close proximity to the twyers, and, if the vaporized compounds are not withdrawn at or near their place of formation, they are lost. The urpose of the present invention is to prevent such loss, and to recover the products for useful purposes.

Accordingly, in carrying out the inven- V tion, a portion of the reducing gases containing a corresponding quantity of the vaporized compounds is withdrawn from the furnace at a point in the neighborhood of the twyers, and preferably above them,

care being taken to restrict't quantity of the gases Withdrawn to an amount which will not interfere with the requirements of the reducing gases in the upper zone or shaft of the blast furnace. The furnace operator is enabled to judge of the limitations in this regard, by noting, from time to time, the character and particularly the color of .the slag, which he may tap off, for that pur so, as a overnin test, at intervals, at tilib cinder gotch of the blast furnace. This slag (in a pig-iron blast furnace) should normally be grav or white in color, thus indicating that only minute antities of iron appear in the slag, and that the blast furnace is actin 7 under proper conditions which require t at practically all of the iron in the carefully calculated ingre- -dients of the charge shall be recovered in the metal output of the furnace as pi iron. If the operator finds that the slag su stantially retains its normal gray or white color and substantially its normal temperature and fluidity, he will know that the amount of the reducing gases withdrawn is within the limits that can be tolerated for acceptable blast furnace conditions, and also that the formation of the recoverable alkali metal com ounds is progressing normally in the com ustion zone. If, on the other hand, he finds that the slag, in the test tap,

has assumed oris beginning to assume an abnormal darker tint, indicative of the as-- Sage of oxid of iron into the slag, he wi be advised that the necessary conditions for the proper functioning of the blast furnace are being interfered with by an excessive withdrawal of the reducing gases, and also that the formation, composition, and output of the desired condensable products are suffering abnormal irregularities, correspondingly lnterferin with the intent and purposes of the invention. He will, accordingly, be able to correct the irregularities referred to, by appropriately limiting stricting the amount of the reduclng gas withdrawn, per unit of time, or by temporarily restoring, either wholly or partially, the usual blast furnace conditions, until the taking of a test at the cinder notch shows that the slag'has been brought back to normal. v

In some instances, the operator may prefer to withdraw continuously the poltion of the reducing gases which it is found feasible to abstract for the recovery of the compounds of alkali metals recoverable without interfering with the necessary conditions of the blast furnace operation. In other cases, the operator may effect the withdrawal of such quantities of the reducing gases, atintervals. Either of these modes of operation are within the broad purview of t e intended practice. Continuous or practically continuous withdrawal of the gases will be more advisable, in those instances where the quantity of alkali in the charge is relatively large. Intermittent withdrawal is to be. recommended where the quantity of alkali in the charge is relatively small,the large capacity of the blast furnaces, with the high heats employed and the perfect reduction of the ores, causing even the ordinarily small content of alkali in the charge to accumulate-in considerable quantity, and it may thus build up gradually in succeeding time periods. so to make it feasible and even desirable to remove it apart. oreover, by the intermittent instead of the continuous withdrawal of the gas, furnace gas for the stack will be correspondingly conserved, thereby affording a Wider safety factor for the blast furnace operation as a whole. It will also be under stoodthat the recovery of the potassium cyanid and other alkali metal compounds contained inthe withdrawn gases may be effected in any of the methods known to the art; that is to say, either by condensation. settling, screening, solution. electric precipitation. or the like. as will readily occur to those skilled in the art.

The withdrawal of the gases containing the products to be recovered in accordance with our invention may. be through apertured plates or blocks inserted in the furor re-' eriodically at suitable intervals ion nace wall and )rovided with a water-corded passageway. '0 each of these piatcs or blocks conneetcd a conduit pipc leading to the condensing or recovering apparatus.

In order to avoid comlensutitm of the oondensubie material in tlic conduit pipe it may have a heat insulating acket cithcr on thc. inside or on the ij'llbfiilli but if the insulation is on inside it .mue he of a a torial resistant to the amino, of the g. and the vaporized compounds carried thereby.

The condensing may be conducted in one or more chambers and in one or more units as may be desired. Where the withdrawal of the gases is to take place continuously, it is. necessary to have more than one set of chambers or units, so that when one set is being cleaned, another set ,is available for operation. After the gases pass through the condensing apparatus, they are preierabl}. led through a flue of suitable length to permit them to cool down to a relatively lQ'iV temperature whereupon tl cy are led through a filter, such as a bag filter. or other analogous apparatus, for the further removal of the fume which carries values. In passing through the condensing apparatus, the gases will be reduccd 1n tempera ture, but if not sufficiently rcduccd in temperature to be ied immediately to the filter, their passage through the due chow referred to. or through any like device. may be resorted to. as will be rcadiiy undi Md.

The final exit from the Eflt'ffl'fi paratus, practically freed of its solid material, is not returned to the shaft of the fun nace, inasmuch the intmrbictinn oi reintively ooldgas into the hot gases within the furnace would accompiish no useful pun pose, and would even be delctcrious for [lie reason that it would absorb heat. Act-or ingly, the final exit gas may citlier be wasted 01- may be led by a conduit to the stoves or boilers of the plant for heating purposes, as the necessities of the plant may require, or it can be put to other appropriate uses.

The air supplied to a blast furnace usually enters the tivyers at a pressure of from 10 to 20 pounds per square inch. Accordingly. the pressure existing within the fur nacc will, in most instances. suiiicc For insuring thc outflow of tlic portion of the gases to be withdrawn therefrom. and this outflow may conveniently be controiled by an adjustable valve or damper at the outlet of the condensing apparatus, the ratc of outflow of the withdrawn gases being so adjusted that the composition of the usual exit. gases from the furnace top is not apprw ciably changed.

Referring further to the uses and practice of the invention, we may tnlrc. as an oxen:- ple. thereof, on n verage type of pigdron blast furnace producing 400 tons of pig: iron daily. In such blast furnace there will be consumed daily approximately 1300 tons of coke, ore and linn stone, so that :1 relatively small content of alkali in this aggregate will amount. in a days operation. to a considerable quantity. and may be withdrawn by the practne of tin present invention. at a rule appr prints in tin: :illzai contents of the meta; .ai charged.

In a given instance. the charge ol a blast furnace showed on analysis an average of potash (calculated as K 0) 0.28%, and an average of soda {calculated as No. 0) 0.357.. These contents of potash and soda amounted per ton of iron produced to pounds of potash and 35 pounds of soda. Qucli a fun naoe charge can be enriched in potash, if desired. by th addition of a pot'nsli-cm1taining mineml. such as green sand tusuall} containing about 8' of potash). providcd the t'urnace operations are not. disturlocd from their proper function of producing pig iron. including the proper reduction of the iron ore, as indicated by the gray or white color of the slag. As hcrcinbctore noted, small amount of iron oxid in the slag suflices to turn it black. thereby indieating abnormally iow tempcraturc conditions in the furnace. Moreover. such a slug necessarily an oxidizing and its pres rncc in the furnace tends to prevent the formation of cyanogen compounds which arc among the most powerful reducing agents known, so that the production of :1 black or oxidizing slag of itself indicates condi- News in the furnace that arc iinfavorab'lc do the production of some of the compounds which it is the purpose of the invention to recover,

The accompanying drawings illustrate col" tuin forms of appuratus adapted for currya suitable condensing system for the rceov cry of the alitalionctal and nitrogen compounds;

Fig. 2 represents :1 similar VitW showing a modified communication of piping ho tivccn the blast furnace and the first chamber of the condensing! system:

Fig. 3 represents a sectional plan view of Fig. 2.

Rcfcrring to the drawings. 3 rcprcscnts a pigriron blast furnace. ,uliicli. as ordinarilv constructed. is from 80 to 100 ft. high. 4 rcprcscnts the twvcrs for the admission of the int air blast 5 thc slap: outlet or notch. and 6 thriron notch. 7 rcprcscnts cooled pintcs or blocks inserted in the hearth and bush wail of the furnace. through which the poition of the reducing gases which. it is permissible to remove without disturbing the :isscntial blast furnace conditions are'lso withdrawn together with their content of the compounds of the alkali metals which are to be recovered.

The oints of withdrawal of the gases are not limlted to the particular location of any of-the plates or blocks shown, but may be at any point in the wall inclosing the combustion zone. Thus, as indicated in Fig. 3,,

they may be spaced at substantially equal distances apart around the periphery of the furnace, andmay all lead to the same initial chamber of the condenser system; in such case, instead ct; connecting the-several blocks to a common bustle-pipe, individual pipes 9 will connect them severally, in straight-line direction, to the condenser chamber, so that it will be corres ondinglyconvenient to remove any possi le obstructions that might condense or otherwise form therein, and so that any individual pipe may be readily removed when occasion may require and subduit 9, leading to the first condensin chamher 10 of the condensing system. he several blocks may, as shown in said figure, be connected to a manifold 8, for convenience and compactness of apparatus, instead of being connected directly to the conduit 9 without using the manifold.

The gases entering the condenser 10, at an opening 11, deposit the heavier and more readily condensablc products therein and pass out through an outlet 12. In the construction of chamber 10, a water-container 13, which may be k t supplied from any convenient source of water (as. for instance, the valved pipe 13) is so arranged that the Water supplied thereto will flow over the upper edge of the container and down over the periphery of chamber 10, thereby cooling the gases within the chamber correspondingly and accelerating the condensation thereof. The waste water is collected in a gutter 14 at the low part of the chamber 10 and flows oil to waste at the discharge lip 15. 16 is an air valve which permits the air in the chamber 10 to escape when the gases are admitted to the chamber. 17 are lugs for lifting 'the chamber from its position. when desired, and substituting a fresh chamher. and, for this purpose, the joints 18 and 19 in the piping should preferably be 'ball joints. Valves 20 and 21 are used for regulating the flow of the gas.

The gas asses from chamber 10 throu n conduit 22 into chamber 23, which may a so be water-cooled and wherein a further condensation of the condensable compounds takes place, and at a lower temperature than in chamber 10. The material condensed in chamber 23 may be removed at the hopper bottom which has an aperture provided with a suitable valve, as, for instance, the conical' pivoted and counter-weighted valve 24.

T e velocity of the gases in the chamber 23 is comparatively low so that a considerable quantit of the condensable compounds is recovere in said chanrber. From the chamber 23, the gases next pass through the conduit 25, their flow being controlled by the adjustable valve 26. The conduit 25 conducts the gases to a'bag filter or similar effective device in chamber 27-. The ba filter 28 will remove practically all of the remaining compounds which are carried along in suspension by the gases. The bag filter is provided with a ring 32 connected thereto, and operating handles 29 extend through the walls of the chamber so that the workmen mayshake the bag filter by means of these handles,-thereby dislodging the material within the .bag which material thereupon falls into the receptacle 30. The gases, after havingpassed through the filter, make their exit at 31 and can lie-carried oil to any convenient place of use.

It will be noted that the exit pipes from thechambers 10 and 23 extend down to the lower portion of said chambers, so as to give the gases a longer path to travel before taking their exit therefrom. In passing from the chamber 23 to the chamber 27, the gas must be so far lowered in temperature that it will not consume or injuriously affect the material of the filter bag. This can be done by cooling the conduit 25 in any appropriate manner, as, for instance, by Water jets evternally applied thereto, or, (in substitution or connection therewith) by extending the length of the conduit so that the gases will have a sufiicient length of travel to efi'ect the necessary reduction in tern erature.

The condensing system i lustrated is typical of many others that might be substituted therefor. The quickly condensable com ounds, which are heavier, are deposited in c amber 10. Those that are lighter and require further cooling, are deposited in chamber 23. and the compounds in suspen sion are recovered in chamber 27. The condensation and collection of the compounds may therefore be said to be fractional. and the temperature of the gases is lowered successively from one chamber to another to permit this fractional recovery of the compounds. 7

A typical instance will illustrate the prac tical results obtained by the practice of the invention. Thus, operations were conducted on a pig-iron blast furnace 80 ft. high and 17 ft. 6. in. in diameter at the bosh and with a hearth 11 ft. 6 in. The data for a weeks operation of this furnace were as follows:

Volume of air blast per minute:23,000 cu. ft. Temperature of air blast l200-1300 degrees Fahr.

30 per cent. of the ore charge was Lake Superior ores. and per cent. New Jersey magnetites. The vaporized compounds of the alkali metals were withdrawn from a point above the twyers and led to a. condensing apparatus. and from there to a bag filter.

The material. collected in that part of the apparatus preceding the filter was found. on analysis, to give the following percentages of potassium cyanid and other alkali metal compounds, to wit:

Potassium cyanid (KCN) :24.35)?

Carbonate of potash (K CO 31.272

The material obtained from the filter portion of the apparatus gave the following pen centages of potassium cyanid and other alkali-metal compounds. to wit:

Potassium cyanid (KCN) :49.05; Carbonate of potash (KfiOQ QlB-fifi. Soda (Na O):6.20EZ.

These analyses illustrate the commercial and industrial capabilities of the invention. They also illustrate, in a striking manner, the fact hereinbefore indicated. that there are present in the hearth and bosh of the iron'blast furnace not only alkali cyanid but other compounds in which the alkali metals are associated and that these are there present in large quantity and in recoverable form as potassium. carbonate, and also as sodium compounds (calculated as (Na O). In so far as we are aware, even the presence of these additional compounds, as such, in the combustion zone of the iron blast furnace, has never before been established, nor their recovery effected.

It will, of course, be understood that by the expression iron blast furnaces as used in the claims. we do not intend to exclude iron blast furnaces whoseproduct may vary from pi iron. inasmuch as the blast furnace charge is frequently made up in such manner as to result in a product containing notable quantities of alloying metals. such as manganese, silicon. or the like.

In further explanation of the characteristic feature of the presentinvention, we may point out that, in blast furnace practice, it is absolutely necessary to maintain the thermal equilibrium or heat balance, which is relatively sensitive, in order to obtain the standard metal product of the blast furnace operation. The thermal equilibrium can, perhaps, best be understood from the point of view that the blast furnace consists of two separate pieces of apparatus, the one superimposed upon the other; the upper apparatus, is that in which, occur the preheating and reduction of the iron ore and other in gredients of the charge. while the lower apparatus is the hearth and bosh. in which the melting of the iron and cinder and the removal of the final traces of oxygen, etc, are carried on. these being operations which can only occur at a certain (high) critical temperature.

The gas resulting from the combustion of the fuel in the lower apparatus must supply the heat necessary for both sets of reactions. or the furnace will not function according to its prescribed intent.

It so happens that the ordinary heat requirements of these two pieces of apparatus are proportioned to one another in approximately the same relations as the heat available from the gases of combustion. These gases yield a. small quantity of high temperature heat in cooling from their theoretical combustion temperature to the critical temperature. and a much larger quantity of heat in cooling from the critical temperature to the temperature at which they are discharged. There is generally an excess for the latter purpose: in fact. the conditions of safest furnace operation demand that there shall be such an excess. 'This is for the following reason:

The amount. of heat used in the shaft of the furnace is, roughly, five or six times the amount used in the hearth. Broadly speaking. about 5,000 thermal units per pound of iron are used in the shaft and about 1.000 thermal units per pound of iron in the hearth. A deficiency of a given number of thermal units would, therefore. be about five times as great. measured as a. percentage, in the hearth as in the shaft. Thus, if the descending charge were short in its heat requirement by 500 thermal units to the pound of iron at the base of the shaft, this would only be a deficiency of about 10% but when the charge entered the hearth this would eonstitute about a. 50']? deficiency of the. heat requirement in \that region. This would be absolutely ruindus to the work of the furnace and would put it out of working commission completely. From this reasoning it can be seen that it is vital that the charge shall enter the hearth without any de ficiency in the heat which it should receive in the shaft and this limits radically the amount of gas which can be abstracted from the base of the shaft. Generally}.

where coke is used as the fuel in the charge, the surplus of thermal units in the shaft is considerable,roughly speakin from 10 per cent. up to 20 per cent, un er ordinary conditions. This surplus can be abstracted from the shaft without detriment to the working of the furnace. For instance, it is well known that a certain amount of water can be added to the charge without affecting the work of the furnace, but if this amount be exceeded, the effect becomes highly injurious.

It will therefore be understood that in order to abstract the potassium cyanid, the potassium carbonate, and the sodium compounds, and nevertheless maintain the normal operation of the furnace for its primary purpose, which is emential for commercial operations. the amount of heat abstracted must be within the margin of surplus availablewithin the shaft.v

It is therefore possible to operate the furnace in either of the two ways hereinbefore described, for the recovery of the potassium cyanid, potassium carbonate and sodium compounds, without upsetting the thermal equilibrium. Thus, an amount of gas within the safe margin indicated may be continuously abstracted from the bosh; or, intermittently, relatively large portions of the gas may be abstracted for limited periods, the periods of withdrawal being separated by suitable time intervals, so that the drain upon the furnace shall not exceed the margm of safety. Any greater abstraction than this by either method would result in upsetting the thermal equilibrium of the furnace and would cause the conditions in the hearth to pass from completely reducing to partly oxidizing, owing to the resence of a considerable quantity of hig ly oxidizing iron oxid, which passes into the slag when the conditions in the hearth are not sufliciently reduging. This would not only upset the work of the furnace and spoil the iron roduced, but would probably materially re uce the amount of cyanid produced, since the oxidizing action of the slag in such case would tend to destroy it.

We may say further. that we regard the intermittent withdrawal of the gases for the recovery of the alkali compounds as presenting particular advantages, even in those instances where it would be feasible to withdraw the gases continuously. For instance, the removal of the gases continuously involves the removal of a correspondingly small percentage of gases, per unit of time; so that there is a tendency to clog up the small aperture or apertures through which the gases are withdrawn, and, at the same time, the relatively slight draft on the zone of concentration of the alkali compounds, tends to limit the withdrawal of the compounds to a more or less local region in the vicinity of the aperture or apertures. On the other hand, the intermittent operation, by permitting the accumulation of the alkali compounds up to acertain point of saturation, enables the operator to remove a larger proportion of the gas, with a high percentage of the alkali compounds, during a relatively short interval, and the attendant high velocity of the withdrawn gases not only tends to prevent clogging of the apertures through which they pass, but also insures a deeper draft on the zone of concentration of the alkali compounds; so that, by the intermittent process, practised at such intervals as will not disturb the thermal conditions of the blast furnace, a high recovery of the alkali com ounds is assured, even though the analysis of the furnace char e is relatively low in alkali.

What we claim is:

1. The method of operating iron blast furnaces, which consists in carrying on the blast furnace process to produce iron and slag as usual, and to produce by a heated air blast supplied under pressure an atmosphere of reducing gas in contact with the ignited fuel, withdrawing from the furnace such portions of the reducing gases as are not necessary for maintaining the required reducing conditions in the stack. and abstracting from said withdrawn gases the compounds of the alkali metals contained therein.

i 2. The method of operating iron blast furnaces, which consists in carrying on the blast furnace process to produce iron and sing as usual, and to produce by a heated air blast supplied under pressure an atmospllcrc of reducing gas in contact with the ignited fuel, withdrawing from the furnace such portions of the reducing gases as are not necessary for maintaining the required reducing conditions in the stack, an abstracting from said withdrawn gases he compounds of the alkali metals contained therein, the withdrawal of the reducing gases being effected at a point in their u ward travel below that at which the said compounds would be precipitated or disassociated by the colder materials in the furnace.

3. The method of operating iron blast furnaces, which consists in carrying on the blast furnace process to produce iron and slag as usual, and to produce by a heated air blast supplied under pressure an atmosphere of reducing gas in contact with the ignited fuel, withdrawin from the furnace such portions of the re ucing gases as are not necessary for maintaining the required reducing conditions in the stack, and abstracting from said withdrawn gases the compounds of the alkali metals contained therein, the withdrawal of the reducing gases taking place periodically so as to perfurnaces, which air blast,

mit the building up or accumulating of said compounds therein in the periods of time between withdrawals.

4. The method of operating iron blast furnaces, which consists in carrying on the blast furnace process to produce iron and slag as usual, and to produce by a heated air blast supplied under pressure, an atmosphere o reducin gas in contact with the ignited fuel, regu ating the operation so as to maintain sla of a light color, withdrawing from the reducing gases as are not necessary for maintaining the required reducing conditions in the stack, and abstracting from said withdrawn gases the compounds of the alkali metals contained therein.

5. The Method of operating iron blast consists in carrying on the blast furnace process to produce iron and slag as usual, and to produce by a heated supplied under pressure, an atmosphere of reducing gas in contact with the ignited fuel, withdrawing from the furnace a portion of the reducing gases containing alkali carbonate at a rate which will not destroy the thermal equilibrium of the furnace, and abstracting from said withdrawn gases the alkali carbonate contained therein.

6. The method of operating iron blast furnaces, which consists in carrying on the blast furnace process to produce iron and slag as usual, and to produce by a heated air blast supplied under pressure, an atmosphere o reducing gas in contact with the ignited fuel, withdrawing from the furnace urnace such portions of thea (portion of reducing gases containing a. so ium compound at a rate which will not destroy the thermal equilibrium of the furnace, and abstracting from said withdrawn gases the sodium compound contained therein.

7. The method of operating iron blast furnaces, which consists in carrying on the blast furnace process to produce iron and slag as usual and to produce by a heated air blast supplied under pressure an atmosphere of reducing gas in contact with the Ignited fuel, permitting the building up or accumulation of alkali compounds within said gases, and periodically withdrawing the major part of the alkali compounds thus accumulated together with such limited amount of reducing gases as will not destroy thethermal equilibrium of the furnace, and abstracting the alkali compounds from the withdrawn gases.

In testimony whereof I, HARRY B. WEAV- ER, aliix August, 1918.

HARRY B. WEAVER.

Witnesses:

E. C. Koons, A. E. BROWN. In testimony whereof I, JAMES GAYLEY, afiix my signature on this second day of August, 1918. i

' JAMES GAYLEY. Witnesses:

WALTER S. Rune, F. W. Your.

my signature on this sixth day of

Images