US1819479A - Method of making silicon iron compounds - Google Patents

Description

Patented Aug. 18, 1931 UNITED STATES PATENT OFFICE JAMES A. PARSONS, or DAYTON, onro, ASSIGNOR TO THE DURIRON COMPANY, me, A

CORPORATION or new YORK METHOD OF MAKING SILII ICON IRON COMPOUNDS No Drawing.

The invention relates to a method of making silicon iron compounds having definite desirable proportions of silicon and iron. The process is applicable in the making of ferrosilicon used in steel manufacture, to the making of acid resisting alloys, and to the making of other compounds of steel and silicon where it is important that the proportions of silicon to steel or iron shall be accurately fixed. The process has been developed for use particularly in connection with the making of acid resisting iron in which the amount of silicon in the alloy must be held within relatively narrow limits in order to produce a roduct which is commercially satisfactory. n this alloy, the silicon should range between 14.25% to 14.50%, since the product is too brittle if the silicon runs above the upper limit specified and lacks in acid resisting qualities if the silicon drops below the lower limit. Heretofore, these limits have been diflicult to attain, due to the fact that some silicon is lost during the melt, and to the fact that the ferrosilicon which is used to provide the silicon does not run uniform as to its silicon content. As a result, it has been the practice to first melt up a quantity of ferrosilicon, analyze a sample, so that its exact silicon content is known, and then use the ferrosilicon thus produced in the melt from which the final alloy is made.

The method hereinafter described and claimed, avoids this difficulty and produces a.-

product, without any preliminary melt, which is very close in its content of silicon to any redetermined content which may be selected.

In carrying out the process, the first part of the melt is accomplished in the usual way, preferably in an electric furnace. Steel or iron and ferrosilicon are mixed and melted with the proportions roughly adjusted so that the calculated amount of silicon is on the high side of the exact percentage desired; for instance, one or two per cent of silicon over a desired final content of 14.25%. As soon as the batch is melted, which occurs when the temperature is from 2450 degrees F. to 2550 degrees F., a sample is removed in a test spoon or ladle and cast in a chilled Application filed January 23, 1930. Serial No. 422,952.

metal mold to provide a bar one half inch in diameter and six inches long. The use ofa mold of the kind specified is desirable in order to give a test bar of exact dimensions.

The test bar, as thus provided, is now tested for electrical resistance, this being done with any suitable testing apparatus, such as a standard Kelvin bridge. This resistance, as thus determined, will indicate very accurately the amount of silicon in the melt. This has been found by running a large number of tests with a wide range of silicon contents and establishing a graph or resistivity curve which is relatively smooth and uniform. By reference to this curve or to a table covering the results of the determinations, the amount of silicon can be found by noting the resistance in the test bar. To illustrate, a resistance in the bar specified of 20 micro-ohms per cubic inch of alloy means a silicon content of 14.25% silicon; a resistance of 25'micro-ohms means a silicon content of 14.70% silicon; a resistance of 30 micro-ohms means a silicon content of 15%, etc. This electrical resistance determination of the silicon, .if properly done will not vary over one-tenth of one per cent plus or minus from that found upon chemical analyses of the alloy. The most useful portion of the graph is that portion which indicates the resistance from about 14.25% of silicon upwards, as below such point the graph curves in the reverse direction.

The determination of the silicon content can thus be accurately found in fifteen min utes or less, and during the period in which the melt is attaining its proper pouring temperature of 2800 to 2900 degrees F. Assuming that the silicon content is on the high side of that desired in the final alloy, this content is now reduced by adding to the melt an amount of iron suflicient (by calculation) to lower the silicon content to exactly the per cent desired. It is desirable to proceed in this way, rather than to work on the low side of the silicon content, and then add silicon to bring the alloy to the desired proportions, as the addition of silicon causes boiling in the melt tending to make the resulting casting porous.

The whole operation can thus be carried out without loss of time, starting with ferro silicon in which the average silicon content is only roughl known, and a resulting alloy can be secure whose content of silicon is within one-tenth of one per cent plus or minus of any predetermined content, which it is desired to produce. I have found that the carbon in the melt does not affect its resistivity where such carbon runs from 75 per cent of saturation to saturation and that the relatively small manganese content normally found in silicon steel alloys also does not affect their resistivity. It will be understood that the term iron is used in its generic sense to comprehend, in addition to ordinary iron, the various grades of steel containing more or less t'otal carbon.

In the making of commercial ferrosilicon, the raw materials are placed around the electrodes in an electric furnace and fused in the usual Way, the materials ordinarily consisting of quartz, iron turnings, anthracite coal, charcoal and coke, the constituents and their proportions depending on requirements and conditions. The silicon content has heretofore been roughly determined by examining the fracture from a sample and such content is of course not constant due to variations in the materials and to other causes. The present procedure is equally applicable here, as samples of the product can be taken from time to time, their silicon content accurately determined electrically as heretofore explained, and the necessary additions or changes made in the materials in order to give the desired silicon content. In this manner the silicon content can be kept much closer to the desired figure than has heretofore been possible.

What I claim is:

1. A method of making silicon iron compounds of substantially exact predetermined proportions in which the silicon is above 14.25% which consists in fusing together as constituents ferrosilicon and iron in the approximate calculated quantities necessary to give said predetermined proportions, taking a sample from the melt before it has reached pouring temperature, casting it into a test bar, and determining the proportion of silicon in the sample by measuring its electrical resistance during the period in which the temperature of the melt is being raised to pouring temperature and then adding to the melt a quantity of one of said constituents suflicient to bring the alloy to said predetermined proportions.

2. A method of making silicon iron compounds of substantially exact predetermined proportions in which the silicon content lies between 14.25% and 18% which consists in fusing together ferrosilicon and iron in the JAMES A. PARSONS.

Images