US2779666A - Halide detector - Google Patents

Description

Explorin Hose 9 Jan. 29, 1957 c. c. ANTHES 2,

HALIDE DETECTOR Filed May 22, 1953 Z [5551 18 i5 w l7 W M n 30 :ml I .1 Flame /IIIIIIIII A Acefylene INVENTOR CLIFFORD ANTHES 1A 0 EY I HALIDE DETECTOR Clifford C. Anthes, Union, N. .L, assignor to Union Car- ];(ide and Carbon Corporation, a corporation of New ork Application May 22, 1953, Serial No. 356,761

8 Claims. (Cl. 23-254) This invention relates to a halide detector such as is used for the discovery of leaks around domestic and commercial refrigerators using a halide gas in common practice today. An object of this invention is to enhance the sensitivity of the detector. Another object is to prolong its life before renewal of a copper-containing reaction disc becomes necessary. A further object is to provide a detector of this sort which is more convenient to handle for a prolonged period without becoming objectionably hot to the hand.

Heretofore, a commercial detector of this sort has been of the general type disclosed in the patent to Tuel No. 2,200,523, dated May 14, 1940 and entitled Gas Detector. This prior device has a flame nozzle to which a fuel gas is supplied under pressure for aspirating air to a flame nozzle for directing a flame through a perforate disc of copper. A light and wind shield of cylindrical shape encloses the flame and reaction disc. When a halide gas in suitable proportions is mixed with the aspirated air, the flame issuing from the copper reaction disc becomes a distinctive color indicating the presence of a halogen or compound thereof in the air supplied in concentrations of about 150 parts of the halide gas per million parts of air and stronger concentrations of the halide.

According to this invention it has been discovered that a stainless steel reaction disc beside the one of copper is able to impart a different and distinctive yellowish color to the flameat concentrations of the halide gas far below those, for which the copper disc alone is able to provide an indication. The stainless steel disc is about 1000% more sensitive and able to detect the presence of a halide atconcentrations of about 15 parts per million parts of aspirated air. Thus the stainless steel disc functions at concentrations of the gas being sought, well below those for which the copper has been found sensitive. In event that the halide content of the supplied airrises to a value strong enough for the copper disc to function as an indicator, the copper disc takes over, masking the yellowish color with a bluish green, which turns to a purple with still stronger halide concentrations.

Where the former copper disc had an average life of perhaps 14 hours, it has. been discovered that by using a reaction plate of silicon copper its life may be prolonged about' seven times that of the prior copper plate.

'Referring'to the drawing, Fig. l is a longitudinal section through a. preferred embodiment of this invention.

States Patent Fig. 2 is a perspective showing how the two reaction discs of Fig. 1 fit together.

As shown in Fig. 1, the device illustrated is provided with a fuel gas inlet connection 10 and an air connection 11.. Any one of a number of fuel. gases may be used,

such as acetylene, natural gas, and many others, and

advantageously the gas selected can be among the less expensive types. Air is aspirated by the fuel gas into the mixing chamber '12 from which a passageway 13 extends to a usual type flame nozzle. The flame may be of any conventional shape and preferably is fairly colorless for the flame color tests and may be of the general shape illustrated by the dotted lines of Fig. 1 of the drawing. A copper containing disc 14 receives the direct impact of the flame upon it and a stainless steel disc 15 is held contiguous the disc 14 by a flange 16 on the disc 14, which flange is crimped over to hold the discs 14 and 15 together.

In Fig. 2, the shape of these discs will be apparent from the drawing, the disc 14 has a groove through its flange 16 and extensions on the disc 15 project in the groove through the flange.

A supporting arm 17 of stainless steel, at one end of which the disc 15 as formed, is provided with a sharp bend as illustrated and a screw 18 secures this arm 17 for supporting these discs from the usual light and wind shield 21 illustrated as being of generally cylindrical shape and extending around the flame, to make color changes in the flame more readily perceptible and to prevent drafts distorting the flame. The copper-containing disc 14 is provided with a perforation 19 substantially coaxially aligned with the flame nozzle, while the disc 15 has a larger perforation 20 also coaxial as illustrated.

The light and wind shield 21 is provided with a window 22 through which the flame may be viewed. A bushing 23 is provided with the flange illustrated, against which the shield 21 may be held by a thermally insulating sleeve 24 held in place by the spring 25 illustrated. The tube walls around the passageway 13 within the sleeve 24 are of stainless steel to reduce heat conduction to the main body portion of the inlet connections 10 and 11.

Discs of screening material 26 serve to filter out any foreign matter in the fuel gas before such gas passes through the orifice 27 creating a sub-atmospheric pressure in the mixing chamber 12.

In operation the detector of this invention may housed in substantially the same way as is contemplated for that According to one phase of this invention it has been.

discovered that; when an ordinary l8-8 stainless steel disc commercially designated as the type 304'and containing a very small amount of sodium in presumably non-vola'tileform at a dull red heat is placed beside a copper disc ingthegnanner illustrated, this detector becomes about ten timesmore sensitive to the presence of halides in theair; In concentrations of gas containing ahalogen or compounds thereof of at least about the strength of 15 parts per million parts "of air a yellowish color is discernible in the flame due to the presence of the halide gas and the stainless steel disc. In concentrations between the aforementioned minimum of about 15 parts per million and the concentration at which a copper containing disc begins to function, i. e. about parts per million, the stainless steel disc of this-invention provides the new and more sensitive detector by strength the eflect is at first a blending of the colors due to the flame indication from each disc. This blending is at first a light blue. Later with increasing concentrations the color from the copper containing discbecomes dominating and substantially masks the color due to this stainless steel.

Another use for the detector of this'invention is in connection with a test of a vacuum chamber where halogen containing compounds may be placed around the outside of the chamber and the detector of this invention used in connection .with the exhaust gases from a vacuum pump exhausting the chamber, the gas tightness of which is being tested. a

In operation it has been found that the best results were obtainable with the discs of the approximate shape illustrated. Due to the copper containing disc 14 being a better conductor of heat, it was found the flame should impinge upon the side of the disc 14 around its central orifice for a substantial area to supply heat conducted away from the inner portion of the disc 14. The stainless steel disc 15 being much less thermally conductive than the .disc14 did not seem .to need so much heat applied to it-exceptclosely around its perforation. It is estimated that both discs were brought to adull red heat around their perforations. Perhaps the average temperature of each disc was in the neighborhood of 1100 F. to 1500" F. The precise temperature of the disc does not appear to be critical.

A noteworthy feature of this invention is the use of a supporting arm 17,of stainless steel whereby the poor thermal conductivity of the stainless steel enables less heat to be transmitted to the light and wind shield 21. Another worthwhile feature of the apparatus illustrated is the useof thermal insulating material 24 providing a convenient handle grip for an operator.

A number of different batches of type 304 stainless steel from different manufacturers were tried and all found to be about equally suitable as a sensitive indicator of the presence of halides. A typical formula for this type 304 stainless steel has the following maximum values reported by the manufacturer for the-elements mentioned: .08% carbon; 2.00% manganese; 1.00% silicon; .04% phosphorus; .03% sulphur; 18.00% to 20.00% chromium; 8.00% 'to 11.00% nickel and the balance iron. The necred heat was not reported. a

A visual spectograph analysis of the yellow color due to the reaction of the halide containing gas with the stainless steel proved by the wave length of the color tested that such distinctive yellow'color was due to the'presence of sodium. At first on heating, the yellow color was present, showing that traces of sodium probably in volatile form at the temperature of dull red heat, were present without any halide in the flame. burned off in about a minute leaving no indication of any more sodium being present in the stainless steel. n the presence of'a member selected from the group consisting of halogens and halides in the air and flame, the yellow colored flame again appeared indicating sodium. As the presence of the halide gas continued the stainless steel reaction disc gave the sodium yellow colored flame. However, on continued use it was noted the disc became less sensitive to the halide gas. It wasfound that knocking or jarring the stainless steel disc resensitized it. This was found to be dueto the formation of a metal oxide coating on'the surface at the dull red heat, and this oxide coating was so untenacious that it was easily jarred 01f.

Efforts to determine how small the traces of sodium in non-volatile form may be have so far been inconclusive. It is believed that sodium in very small quantities in steels has been extremely diflicult if not impossible of quantitative determination. This is confirmed by the literature which gives compositions of steel alloys to one one hundredth of a percent for many of the elements but makes This volatile sodium 'essary presence of sodium in non-volatile form at a dull I no mention of sodium. The analysis given above for the 304 type of stainless'steel was taken from data supplied by a steel manufacturer. This mentions carbon, phosphorous, and sulphur to hundredth of a percent. Yet traces of sodium have been proven to be present when the yellow color is present without a halide present and again after this probably volatile sodium has been burned off, further sodium-is shown to have been present as an indicator of the presence of a halide in the concentration range mentioned.

Many metals show traces of sodium capable of reacting with weaker concentrations of halide gas than can be indicated by the copper of prior art halide indicators. The field of possibly usable sensitive indicators is believed to be very large because most metals are believed to contain traces of both volatile and non-volatile sodium at the dull red heat. For example many types of non-corrosive steel alloys were tried and found suitable at first. Likewise many bronzes and brasses. Among metals which have not been found appropriate may be mentioned electrolytic iron, because of its extreme purity. The prior art copper does not show a yellow sodium flame with weaker concentrations. Aluminum is not appropriate for use as the reaction disc because it melts at too low a temperature. It therefore is believed that an essential of an equivalent of the type 304 stainless steel mentioned is that it must have a fusion temperature well above the 1500 F. mentioned above as a'possible limit for a dull red heat. Another essential appears to be the possession of at least traces of non-volatile sodium, which must be present to react with the halide or in the presence of a halide and give the yellow color. One reasonable theory is that this non-volatile sodium material may be a sodium oxide. Text-books show such oxide to sublime at a temperature between 2300 F. and 2400 F. By increasing the flame temperature using oxygen and acetylene and raising the temperature of the reaction disc its sodium content that was not volatile as the dull red heat showed the sodium yellow without any halide gas supplied to the flame. It seems reasonable to conclude that such was due to the sodium that was non-volatile at the dull red heat becoming gaseous at the temperatures above'2400 F.

Another essential for acommercial equivalent of the type 304 stainless steel is an ability to retain its sensitivity. Many steels and other metals possessing traces of sodium that is non-volatile at a dull red heat and having a fusion temperature well over 1600 F. were found to be sensitive to halides for only a short time. The reason appears to be because a tenacious oxide coating forms at the dull red heat which appears to blanket the non-volatile sodium at such temperature. an example. Traces of the non-volatile sodium after the volatile sodium had been burned off at the dull red heat were appropriately sensitive to the halides but its sensitivity did not last. Even the preferred material, the type 304 stainless steel was not free from this handicap of losing its sensitivity. But the oxide coating on this type 304 stainless steel was fortunately found to be untenacious and capable of being jarred or knocked off easily. Removal of the oxide coating had the effect of renewing the sensitivity of the sodium bearing metal to the halide gas. Hence an easily removable oxide coating is a desirable characteristic. One possible reason Why some oxidemetal coatings are untenacious is that the presence of internal stresses due to unequal thermal expansion in the oxide and metal overcome the binding action of the coating. Again the permeability of the oxide coating may aifect the sensitiveness of the halide for the non-volatile sodium. The lack of sensitivity due to the absence of non-volatile sodium was largely overcome by impregnating the metals with sodium, by heating the metals with a 5% sodium mercury amalgam in an inert atmosphere and actually fusing the metals (copper at 2400 F. and stainless steel at 2800" F.) and allowing the fused metals to solidify in the inert atmosphere.

No proof is available that the impregnated sodium becomes oxidized but it is thought that may be the case. At any rate this impregnation is believed to get more sodium into the metal and with such a greater quantity of non-volatile sodiumthe metal is not believed to become desensitized as rapidly by an oxide coating at the dull red heat.

Next to the type 304 stainless steel mentioned above, a

Commercially pure iron is such amaese desirable material found for the reaction disc 15 is a 25-20 stainless steel known commercially as type 310. A typical formula for the type 310 stainless steel is reported by the manufacturer to be as follows: .25 carbon; 2.00% manganese; 1.50% silicon; 24.00% to 26.00% chromium; 19.00% to 22.00% nickel and the balance iron. It is believed that substantially all stainless steels or non-corrosive steels having the small amount of non-volatile sodium at the dull red heat should function to give a sensitive indication of the presence of a halide gas in low concentrations. With some such steels it was noticed that the the sensitivity decreased as an oxide coating was formed, the sensitivity being restored upon scraping off the oxide coating. Some stainless steels such as the types 304 and 310 mentioned possessed an untenacious oxide coating of such character that it fell off with silght jars or impacts, without the necessity for actually scraping off such coating. The stainless steel disc appears to have a very long life, dependent only upon the length of use and the loss in metal from the oxidation referred to above.

Another feature of this invention resides in the copper containing disc 14. To prolong the life of the copper disc around sevenor eight times what its life formerly was, it has been found that an alloyed copper such as silicon bronze sold by American Brass Company under the designation 1012 not only has the longer life refered to, due probably to the slower oxidation of the copper, but that this slower oxidation does not affect the sensitivity of the copper to the presence of halide gases. Atypical formula for the particular copper containing disc referred to is reported by the manufacturer to be as folows: 95.6% copper; .40% lead; 3.00% silicon; and 1.00% manganese. More specifically in this invention the disc 14 has been found to have an effective life of slightly over 100 hours before another disc of the same material should be substituted. Without any halide gas present in the flame, the only color discernible is that due to the flame itself, which is a very pale blue or almost colorless appearance.

The temperature of the discs is not believed to be critical so long as they-are at least at a dull red color mentioned, however, their temperatures should not be near that at which either of the discs may fuse. The flame is that of an ordinary pre-mixed flame having an inner cone as illustrated and preferably also as nearly colorless as is reasonably obtainable.

It was found that National Bureau of Standards stand-- ard sample 123A of 18-11 stainless steel did not give a satisfactory halide indication. Another portion of the same sample melted down andprepared in the same manned but treated with a small amount of aforementioned sodium amalgam gave a good indication.

In this last case 20 grams of sample 123A was treated with .01 gram of the sodium amalgam before melting.

Pure copper and also electrolytic iron impregnated with sodium in a similar manner gave like results. Not all metals have been tried out but it is believed that all metals having a melting point high enough to be satisfactory and an untenacious oxide coating could be impregnated with small amounts of sodium to possess the same in nonvolatile form at a dull red heat to give satisfactory indicaany convenient'manner and secured to the shield 21. Copper impregnated with a little sodium in form that is non-volatile at a dull red heat has the advantage of not appearing to become desensitized by an oxide coating that needs to be jarred or scraped off. Electrolytic iron impregnated with said sodium amalgam gave a satisfactory test. The metals that were impregnated with sodium to be in non-volatile form at a dull red heat seem to become desensitized to a less extent by an oxide coating. The reason for this is not known. The inert gas used was argon. Perhaps the sodium was impregnated into the metal in metallic form and then became oxidized or otherwise rendered non-volatile at the dull red heat, by uniting with oxygen in the air or elsewhere, or by uniting with some unknown material to render it non-volatile at that temperature. Possibly it is the greater quantity of nonvolatile sodium that is responsible for the lessened appearance of a metal becoming desensitized by an oxide coating.

I claim:

1. In a halide detector having a flame nozzle, a metal reaction member adjacent said nozzle for being heated thereby, a fuel gas connection to said nozzle, and a connection for introducing air to said nozzle, said nozzle being of the type for heating said reaction member to a dull red heat with a substantially colorless flame, the com bination therewith of the improvement for enhancing the sensitivity of said detector, said improvement including said reaction member having a small amount of sodium in non-volatile form at a dull red heat in quantity to give no yellow color to a flame from said nozzle in the absence of a gas selected from the group consisting of halogens and halides in air supplied to said flame after any volatile sodium has been burned off at said dull red heat, but in quantity large enough to give a yellow color to said flame at said dull red heat when a gas selected from the group consisting of halogens and halides is present in about at least 15 parts per million parts of air by volume, said reaction member forming at said dull red heat an untenacious metal oxide coating, and said metal reaction member having a melting temperature well above about 1600 F.

2. In a halide detector having a flame nozzle, a metal reaction member adjacent said nozzle for being heated thereby and containing no non-volatile sodium at a dull red heat, a fuel gas connection to said nozzle, and a connection for introducing air to said nozzle, said nozzle being of the type for heating said reaction member to a dull red heat with a substantially colorless flame, the: combination therewith of the improvement for enhancing the sensitivity of said detector, said improvement including another metal reaction member adjacent the first to be heated by said flame nozzle and having sodium in non-volatile form at said dull red heat in quantity to give no yellow color to a flame in the absence of a gas selected from the group consisting of halogens and halides in air supplied to said flame after any volatile sodiumhas been burned off at said dull red heat but to give a yellow flame at said dull red heat when gas selected from the group consisting of halogens and halides is present in at least about 15 parts per million. parts of air by volume, said second mentioned reaction member forming at said dull red heat an untenacious metal oxide coating and having a melting temperature well above 1600 F.

3. A detector according to claim 2 in which said first reaction member is a copper silicon alloy and said second reaction member is a chromium-nickel stainless steel alloy. V

4. A detector according to claim 2 in which said sec- 0nd reaction member is a stainless steel containing traces When only one reaction disc of sodium impregnated copper is used, it may be mounted in a 'of sodium in non-volatile form ata dull red heat.

5. A detector according to claim 2 in which said first reaction member is largely of copper, in disc form provided with a perforation substantially axially aligned with said nozzle, and said second reaction. member is a disc contiguous the first reaction member, between said first reaction member and said flame nozzle, and provided with a coaxial perforation larger .than that in said first reaction members.

6. A detector according to claim in which alight and wind shield is provided around said nozzle and reaction members, the first such member has a flange around a substantial part of the second reaction member, and a support of less thermal conductivity than that of said first reaction member is secured to said shield.

7. A halide detector comprising a metal reaction disc having a fusion temperature Well above 1600 F., a flame nozzle for heating said disc to adull red heat with a substantially colorless flame, a supply connection for fuel gas to said nozzle, means for aspirating air to said fuel gas, and means for controlling the supply of fuel gas to said nozzle, the combination therewith of the improvement for enhancing the sensitivity of said detector to the presence of a gas selected from the group consisting of halogens and halides in said aspirated air and fuel gas, said improvement including at least a trace of sodium in nonvolatile form at a dull red heat in said disc.

8. A bi-metallic reaction disc unit for a halide detector comprising a copper alloy disc containing no nonvolatile sodium at a dull red heat and a stainless steel alloy disc containing traces of sodium in non-volatile form at a dull red heat, said copper alloy disc having a peripheral flange crimped over a peripheral portion of the stainless steel disc With the central portion of each disc perforate, the perforate portion of the stainless steel disc being larger than the perforate portion of the copper alloy disc, and a supporting arm from the stainless steel disc by means of which said unit may be carried by and secured to a light and wind shield around a detector flame, said stainless steel disc being of approximately an 18-8 type containing about .08% carbon, 2.00% manganese, 1.00% silicon, .04% phosphorous, .03% sulphur, 18.00% to 20.00% chromium, and 8.00% to 11.00% nickel, and the balance iron.

References Cited in the file of this patent UNITED STATES PATENTS 438,117 Shapleigh Oct. 7, 1890 1,305,025 Tait May 27, 1919 1,607,245 Duhme Nov. 16, 1926 1,990,706 Midgley Feb. 12, 1935 2,017,915 Dasher Oct. 22, 1935 2,106,147 Hull Jan. 18, 1938 2,134,552 Gaugler Oct. 25, 1938 2,200,523 Tuel May 14, 1940 2,270,442 Jares Jan. 20, 1942 2,314,882 Hensel'et al. Mar. 30, 1943 2,532,687 Weichselbaum Dec. 5, 1950 2,562,874 Weichselbaum July 31, 1951 2,589,116 Nolcken Mar. 11, 1952 2,635,105 Tanner Apr. 14, 1953 FOREIGN PATENTS 678,806 Great Britain Sept. 10, 1952 OTHER REFERENCES Vogel, Qualitative Chemical Analysis, 3rd Edition, 1945, pages 106, 107, 339. Published by Longmans, Green & Co., New York City.

Claims (1)

1. IN A HALIDE DETECTOR HAVING A FLAME NOZZLE, A METAL REACTIOIN MEMBER ADJACENT SAID NOZZLE FOR BEING HEATED THEREBY, A FUEL GAS CONNECTION TO SAID NOZZLE, AND A CONNECTION FOR INTRODUCING AIR TO SAID NOZZLE, SAID NOZZLE BEING OF THE TYPE FOR HEATING SAID REACTION MEMBER TO A DULL RED HEAT WITH A SUBSTANTIALLY COLORLESS FLAME, THE COMBINATION THEREWITH OF THE IMPROVEMENT FOR ENCHANCING THE SENSITIVITY OF SAID DETECTOR, SAID IMPROVEMENT INCLUDING SAID REACTION MEMBER HAVING A SMALL AMOUNT OF SODIUM IN NON-VOLATILE FORM AT A DULL RED HEAT IN QUANTITY TO GIVE NO YELLOW COLOR TO A FLAME FROM SAID NOZZLE IN THE ABSENCE OF A GAS SELECTED FROM THE GROUP CONSISTING OF HALOGENS AND HALIDES IN AIR SUPPLIED TO SAID FLAME AFTER ANY VOLATILE SODIUM HAS BEEN BURNED OFF AT SAID DULL RED HEAT, BUT IN QUANTITY LARGE ENOUGH TO GIVE A YELLOW COLOR TO SAID FLAME AT SAID DULL RED HEAT WHEN A GAS SELECTED FROM THE GROUP

Images