US 3681972 A
The oxygen concentration in a metal melt, such as a steel melt, is determined by comparing the melting point temperature of a sample taken from the melt with the melting point temperature of the same metal but without any free oxygen, other impurities being taken into account. The difference is related to the oxygen concentration. This is preferably done by taking two samples from the melt, fixing the free oxygen in one of them by a reducing agent and determining the difference between the melting point temperatures of the treated and untreated samples.
Claims available in
Description (OCR text may contain errors)
United States Patent Mahanty et al.
 3,681,972 51 Aug. 8, 1972  PROCESS AND DEVICE FOR DETERMINING THE OXYGEN CONCENTRATION IN METAL MELTS  Inventors: Bhaskar Chandra Mahanty, Karlsruhe; Gustav Mahn, Wolfenbuttel, both of Germany  Assignee: Salzgitter I-Iuttenwerk Aktiengesellschaft, Salzgitter-Drutte, Germany  Filed: June 8, 1970  Appl. No.: 44,312
 US. Cl. ..73/l9, 73/ 17 R, 73/DIG. 9, 75/58, 23/230 PC, 23/253 PC  Int. Cl......G0ln 1/12, GOln 25/06, GOln 31/12  Field of'Search ..73/l9, 25, 26, 27, 23, 354, 73/17 R, DIG. 9; 75/60, 58, 57; 266/1;
Primary Examiner-Richard C. Queisser Assistant ExaminerC. E. Snee, lll Attorney-McClew and Toren  ABSTRACT The oxygen concentration in a metal melt, such as a steel melt, is determined by comparing the melting point temperature of a sample taken from the melt with the melting point temperature of the same metal but without any free oxygen, other impurities being taken into account. The difference is related to the oxygen concentration. This is preferably done by taking 23/230 253 PC two samples from the melt, fixing the free oxygen in one of them by a reducing agent and determining the  References cued difference between the melting point temperatures of UNITED STATES PATENTS the treated and untreated samples.
3,574,598 4/1971 Kern et al. ..75/60 7 Claims, 3 Drawing Figures PATENTEmus 81012 3.681.972
SHEET 1 0r 2 F l G- l o/rrspmcs or m: MELT/1V6 POINT TEMPERATURE a; REDUCED mvo yum-was SAMPLES w "c (cu/e1 5 ro/e 044 65 srsa) INVEN TORJ 11135111112 cunflhua 1111111110] BY usTrw 111 111 14 BACKGROUND c This invention relates generally to a process for determining the oxygen concentration in metal melts, and particularly steel melts. Metal melts are usually at temperatures above 1,500C and consequently, in the past, the samples for analysis first of all have had to be cooled down to room temperature before the analysis could be done.
In one known method a sample, for example a steel sample, is removed from the melt by dipping a sampling spoon into the melt. The sample which has been withdrawn in this way is rapidly cooled to room temperature, after which the surface layers, containing oxides and impurities, are cut away on a lathe. The resultant cooled and cleaned sample is then melted again in a carbon boat under vacuum in a hot gas extraction apparatus at 1,600 to 1,800 C. The oxygen escapes in the form of carbon monoxide mixed with nitrogen and hydrogen, and the gas mixture is fed to a gas analysis apparatus and the oxygen content determined. Alternatively the carbon monoxide content in the gas mixture can be determined by gas chromatography. As a still further alternative, the carbon monoxide can be oxidized with oxygen to carbon dioxide on a platinum wire. The carbon dioxide is passed through a solution of potassium hydroxide in a gas analysis apparatus and the oxygen content determined volumetrically.
Alternatively the analysis can, if desired, be performed without using a hot extraction apparatus (vacuum extraction). The sample is heated in a carbon boat and the carbon monoxide escaping is carried away in a current of argon. The mixed gases are passed over palladium asbestos. The carbon monoxide is oxidized to carbon dioxide, which is absorbed in a solution containing barium perchlorate and a little barium hydroxide. Barium carbonate is formed and the barium hydroxide consumed is replaced electrolytically from the barium perchlorate in the solution. The oxygen content is calculated from the quantity of electric current consumed in the electrolytic process.
These known methods of analysis consume a considerable amount of time. The sample has to be extracted from the melt, mechanically prepared and the gases extracted by hot extraction. The expensive hot extraction apparatus is usually located in a laboratory at some distance from the furnace containing the melt. A further disadvantage of this method is that gases escape during the cooling of the sample to room temperature and these are not measured. Attempts have been made to remove this latter disadvantage by using, for extracting the sample from the melt, an evacuated quartz ampoule one end of which is closed by a meltable plug. When the ampoule is immersed in the melt the plug melts and a sample of the metal melt is sucked into the ampoule. This method 'doescapture the gases released from the sample during cooling, but the quartz ampoules or small tubes are easily damaged. Moreover they only retain their vacuum for a short time and soon become useless if stored for any length of time. A further disadvantage of this method is that two analysis are required, one on the gases released during cooling and a second analysis on the solid cooled sample.
Finally, all the known methods for determining the oxygen content of a metal melt involve lost time between the taking of the sample and the presentation of the analytical results, and consequently the results arrive too late for readjusting the refining process being carried out on the melt.
SUMMARY The present invention relates generally to a method and apparatus for determining the concentration of oxygen in metal melts such as steel melts. More particularly it relates to a new and useful method and apparatus by which this determinationcan be carried out quickly and at the place of the melt. I
It is thus a general object of the present invention to provide a new and useful method for determining the oxygen concentration in a melt which is rapidenough to provide within afew seconds analytical results which can be used for controlling the refining of the melt. The solution to this problem is based on recognition of the fact that the melting point of a pure metal is lowered by the presence of impurities, for example the dissolved oxygen. The lowering of the melting point depends on the natures and concentrations of the impurities, including the concentration of oxygen. The oxygen concentration in a melt can therefore, in principle, be calculated from the lowering of the melting point.
It is thus a further object of invention'to provide for the measuring of the melting point temperature of a sample of the metal melt containing free oxygen and for the comparison of this-temperature with the melting point of said metal containing no oxygen.
A still further object of the invention is to provide a method and apparatus for measuring the melting point temperatures of two samples taken from the metal melt containing free oxygen, one of the samples having had its free oxygen fixed by the use of a reducing agent. The lowering of the melting point of said metal due to the presence of oxygen is calculated as the difierence between the two melting points. Finally the oxygen concentration in the melt is read ofi from a calibration curve plotting oxygen concentration against temperaturedifierence. 7 An excess of the reducing agent added to the second sample is preferably used, to ensure that all the oxygen is eflectively fixed or stabilized. ln-the case of a steel melt, a particularly suitable reducer is aluminum, because aluminum'oxide has a particularly high free reaction enthalpy, and the excess of aluminum required to stabilize the oxygen concentrations encountered in practice in steelmelts has only a negligible influence on the melting point of the steel, a 'fact which can be derived from the phase diagrams for iron and alul-loweve'r, to obtain precise results there should not be too great an excess of aluminum. For other metal melts, different reducing substances should be used, in each case the reducing agent having little, if any, influence on the melting point of the metal. In all cases the amount of reducing agent used should be only slightly in excess of the stoichiometric quantity.
. An alternative method in accordance with the invention assumes that the specific lowering of the melting point temperature corresponding to the concentration, produced by each of the impurities in the metal, apart from the oxygen, is known. On this basis the total lowering of the melting point of the metal produced by all the impurities, other than oxygen, can be calculated once the impurities have been analyzed and quantitized. Any, further lowering of the melting point must be due to the concentration of oxygen. The melting point temperature of a sample of melt is therefore'measured and this substracted from the known melting point temperature of the metal in its pure form. From this value the calculated lowering due to the impurities other than oxygen is then substracted and the difference is the lowering of the melting point produced by the oxygen content. The oxygen concentration in the melt can then be calculated using a known proportionality factor, or can be read off from the above mentioned calibration curve. However, this method -assumes that a precise analysis has previously been made showing the concentrations of all the impurities other than oxygen in the melt. The preferred method is therefore the first of the above alternatives, in which two samples of the melt are taken, and the oxygen concentration is obtained from the difference between the melting points of an untreated sample and a sample in which the oxygen has been fixed by a reducing agent.
Yet another object of the invention therefore is to provide for use in the preferred method a sampling lance having a pair of pans or receptacles at its lower end for withdrawing the two samples from the melt, each pan being equipped with a thermocouple element which is sensitive to temperature. Electric leads and compensation leads of the thermocouples are preferably led through a drilling passing upwards through the lance to means for determining the temperatures sensed by the thermocouples. Preferably this is connected to a temperature difference computer, which can itself be followed, if desired, by an indicating device which may indicate directly the oxygen concentration in the melt.
These, together with other objects and advantages of the invention, will become apparent as the invention becomes better understood from the following description and the accompanying drawings.
DRAWINGS FIG. 1 is a graph plotting the oxygen concentration in a steel melt as a function of the difference between the melting points of reduced and unreduced samples from the melt in C (on the abscissa);
FIG. 2 is a diagrammatic picture of a concentration measuring apparatus; and,
FIG. 3 is a modification of this apparatus.
DESCRIPTION The method of analysis of the-present invention can be applied in practice, for example, to a steel melt as follows. A sample is withdrawn from the melt by means of a sampling device and its melting point is measured by observing the hesitation point on the cooling temperature curve, using a thermocouple and a temperature recorder. At the same time, or .just afterwards a reference sample is taken from the melt and killed, i.e. its free oxygen is fixed or stabilized, for example by adding 0.2 percent of aluminum. The melting-point of the reference sample is also measured. The difference between the two melting pointsis calculated and the oxygen concentration in themelt is read off from the calibration curve shown in FIG. I. This method was used to make the following test. 'Using a sampling device, or sampling spoon, samples were taken from Melt 1 Melt 2 Melt 3 Melt 4 killed sample 1532.6 1533.0 1532.0 1530.1 (melting pt. in C) unkilled sample l528.2 l524.5 1525.6 1526.3 (melting pt. in C) difference (C) 4.4 8.5 6.4 3.8 0, from 0.062 0.123 0.094 0.055 diagram. %0,by hot 0.065 0.ll7
Comparing the oxygen concentrations derived from the curve shown in FIG. 1 with those obtained by the hot extraction method, it will be seen that the difference is not more than 5 percent, and it is of course an open question which of the two methods is the more accurate.
A simplification is obtained by dipping two sampling spoons into the melt together, instead of taking the two samples separately. FIG. 2 shows a device according to the invention which can be used for this purpose. The device consists of a sampling lance 4 whose lower end is adapted for dipping into a metal melt and is provided with two sampling pans 5, 6 mounted next to each other. Each sampling pan contains a thermocouple element 7, 8 whose leads are led up through a central drilling 9 in the lance 4 to a connection block 10, in which the thermocouple leads are connected over contacts 12, 13 to cables whose other ends are connected to a temperature determining and recording device 14, which is itself connected to a difference computer 14, which is followed by an indicator l6 calibrated to show oxygen concentrations directly.
Before making the test a piece of a reducing agent 17 is introduced into one of the measuring pans 6, the quantity of reducing agent being sufficient to ensure complete reduction of the free oxygen in the sample entering this pan. The sampling lance 4 is then dipped into the melt, so that both the measuring pans 5, 6 are filled at the same time. The quantity of oxygen in the melt running into the measuring pan 6 is immediately captured by the reducing agent already present in this pan. The sampling lance is then withdrawn from the melt and the two samples are allowed to cool. During the cooling the temperature of each sample decreases continuously down to the melting point, or freezing point of the sample. At this point the falling temperature curves of the two samples as measured by the device 14 hesitate indicating that the melting points have been reached. The two melting points, or hesitation points are transmitted to the difference computer 15, which calculates the difference between them. This difference is transmitted to the indicating instrument 16, which is calibrated on the basis of the diagram shown in FIG. 1, so that it directly indicates the oxygen concentration in the melt.
If desired the measuring pans can be mounted one above the other as represented in FIG. 3. The method according to the invention for determining the oxygen concentrations of a melt, used in conjunction with the device represented in the drawings is capable of delivering the desired analytical result in a period of time short enough to allow the result to be used in controlling the process. An important point to observe is that the oxygen concentration is measured directly at the furnace or crucible containing the melt. A sample can therefore be taken just before tapping the furnace, allowing the quantity of reducing agents which need to be added to be calculated exactly. This makes it possible to dose the reducing agents preciselyto suit the precise oxygen concentration in the melt, so that there is obtained a final product of higher purity and better quality, which shows an analysis more precisely in agreement with the specified analysis.
In the other variation of the process according to the invention only a single sample need be tested. The sample is taken from the melt and its melting point measured in the unkilled state. The lowering of the melting point from that of the metal in its pure state due to oxygen is then calculated, making allowance for the influence of the other impurities, whose concentrations have previously been determined by the customary analysis in a laboratory. Using this method a test was made on a steel melt containing 0.05 percent carbon, 0.20 percent manganese, 0.020 percent phosphorous and 0.015 percent sulphur. The melting point of an unkilled sample was measured and found to be l,528 C. The steel analysis showed the above concentrations of non-iron substances besides oxygen in the melt and the table shows the amounts by which each of these quantities reduces the melting point of steel from its value in the pure form.
0.05 percent carbon 3. 1 C
0.20 percent manganese 0.8 C
0.020 percent phosphorous 0.6 C
0.015 percent sulphur 0.4 C
The total melting point lowering produced by these substances in the steel is therefore 4.9 C, and this is added to the measured melting point of l,528 C to compensate for the influence of these substances. On the basis of a melting point of l,536 C for pure iron, that is to say oxygen free iron, the melting point lowering produced by the oxygen in the melt is therefore:
1,536C( 1,528C+4.9 C)=3.l C.
On the basis of the curve shown in FIG. 1 this temperature difference of 3.1 C corresponds to an oxygen concentration of approximately 0.042 percent oxygen in the sample. In addition to the above substances in the melt any other substances present must also be allowed for although elements present only in traces may be neglected.
Both the methods described above can be used for rapidly determining the oxygen concentration in a melt, although the second method assumes that a precise analysis of the melt has previously been made.
It is now deemed obvious that there has been provided a method and apparatus for determining the oxygen concentration in metal melts. The invention, in its broader aspects, is not limited to the specific steps and apparatus shown and described, but departures may be made therefrom within the scope of the accompanying claims without departing from the principles of the invention and without sacrificing its chief advantages.
What is claimed is:
1. A process for determining the oxygen concentration in metal melts comprising the steps of taking a first sample from said metal melt containing free oxygen, determining the melting point temperature of said sample, taking a second sample from said metal melt containing free oxygen, adding to said second sample -a reducing agent which has little effect, if any, on the melting point of said metal to fix said free oxygen in said second sample, thereby to provide said metal containing no free oxygen, and determining the melting point temperature. of said treated second sample, the difference between the melting point temperatures of said first and second samples being directly related to the oxygen concentration in said metal melt.
2. A process as set forth in claim 1, wherein a slight excess of said reducing agent is used over the amount which is necessary to fix the free oxygen in said second sample.
3. A process as set forth in claim 1, wherein said metal melt is steel and said reducing agent used is aluminum.
4. A process as set forth in claim 1, including the step of analyzing and determining the quantities of impurities other than oxygen in said metal melt to obtain the amount by which said impurities reduce the melting point temperature of said metal, and wherein said melting point temperature of said sample is compared with the melting point temperature of said metal in its pure state minus the decrease in temperature due to said impurities, the difference between said two melting point temperatures being directly related to said oxygen concentration in said metal melt.
5. A device for determining the oxygen concentration in metal melts comprising a sampling lance having an upper end, and a lower end adapted for dipping into a metal melt, a pair of measuring pans mounted on said lower end of said sampling lance, said pans being adapted to receive samples from the metal melts, one of said pans containing reducing means for its metal melt sample, a pair of thermocouple measuring elements located one in each of said measuring pans, conductors from said thermocouple elements leading to said upper end of said sampling lance, means for determining the temperatures measured by said thermocouple elements, and means connecting said conductors to said temperature determining means. I
6. A device as set forth in claim 5, wherein said sampling lance is provided with a longitudinal drilling extending between said upper and lower ends and said conductors pass through said drilling.
7. A device as set forth in claim 5, further including means for computing the difference between the temperatures measured by said thermocouple elements and said temperature determining means, said computing means having an indicator device.