US2731330A - Determination of oxygen in titanium and titanium alloys - Google Patents

Description

Jan. 17, 1956 M. CODELL ET AL DETERMINATION OF OXYGEN IN TITANIUM AND TITANIUM ALLOYS Filed Dec. 10, 1954 SMMV INVENTORS MAURICE CODELL BY GEORGE NORW/TZ WfWdMQw m QM4fidm ATTORN 5- United States Patent fi 2,731,330 Patented Jan. 17, 1956 ice DETERMINATION OF OXYGEN IN TITANIUM AND TITANIUM ALLOYS Philadelphia, Pa, America as repre- Maurice Codell and George Norwitz,

assignors to the United States of sented by theSecretary of the Army The invention described herein may be manufactured and used by or for the Government for governmental pur poses without the payment of any royalty thereon.

This invention relates to a method for quantitative determination of oxygen in titanium and its alloys. The ductility, hardness and grain structure of titanium and its alloys are influenced materially by oxygen content and for that reason quantitative determination of contained oxygen is of the utmost technological importance.

Oxygen in titanium and its alloys has been determined by a vacuum fusion method. According to this method a sample is fused in the presence of carbon. The oxygen is "converted to carbon monoxide which is then oxidized to carbon dioxide. Two techniques are employed. In one, a low melting point alloy of the sample and iron is formed; in the other, a low melting point alloy of the sample and tin is formed. Both techniques require expensive apparatus and highly trained technicians. An additional disadvantage is that when certain alloying elements are present in a sample, e. g., manganese, such elements volatilize upon fusion of the sample, condense downstream in the apparatus train, and may recombine with the oxygen to vitiate results.

Another known method is that of chlorination. According to this method dry chlorine is passed over the sample at 400 C.; the titanium is volatilized as titanium tetrachloride and the oxygen remains as titanium dioxide. The oxygen is then calculated by determining the titanium of the titanium dioxide colorimetrically. in this method a correction must be made for carbon in the sample in view of the reaction whereby contained oxygen is lost as carbon monoxide. The chlorination reaction must be controlled carefully to avoid low results which are caused by attack of chlorine upon titanium dioxide, with attendant production of gaseous (free) oxygen. Another characteristic of this method that may produce low values is that when certain metals such as molybdenum are present in a sample, oxygen is lost as oxychloride. Dry hydrogen chloride can be used in lieu of chlorine; in any event, values obtained by chlorine and hydrogen chloride volatilization techniques are in poor agreement with values obtained by vacuum fusion analyses.

Oxygen in titanium but not in its alloys has been determined also through the phenomenon of additive hardening effects of oxygen and nitrogen. Given Brinell hardness and nitrogen content, oxygen content is determinable.

Additional methods are: By radioactivation wherein the specimen is subjected to fast neutron bombardment for 30 seconds followed by observation of activity after a known time interval, based upon detection of approximately 6.1 to 7.1 mev. gamma activity with a half-life of 7.35 seconds resulting from the O (n,p)N reaction; this method requires a series of calibrated titanium specimens of known oxygen content. By means of the mass spectrometer wherein a known amount of O tracer is added to the sample; the oxygen is extracted by vacuum fusion as carbon monoxide, and the ratio 0 to O is established. By difference, in oxygen-titanium alloys containing rather large amounts of oxygen and only insignificant amounts of other elements, wherein the sample is oxidized to titanium dioxide by igniting in oxygen, or the titanium itself is determined volumetrically.

Attempts have been made to determine oxygen-in titanium spectrographically through the phenomenon of emission of characteristic TiO spectral bands when titanium is sparked in a vacuum; so far the method has proved of limited value. Suggestions have been made that oxygen in titanium might be determined with fluorine or non-aqueous hydrogen fluoride. Using fluorine at elevated temperatures, the oxygen probably would be converted to gaseous oxygen and using hydrogen fluoride under like conditions, the oxygen probably would be converted to Water. It is known that fluorine attacks titanium at C. Non-aqueous hydrogen fluorde attacks titanium only very slightly at room temperature and reacts so slowly with titanium oxide at 550 C. that volatilization is still incomplete after two hours. Apparatus for fiuorination techniques would need to be constructed of special materials, possibly nickel. It is another disadvantage that commercial fluorine and non-aqueous hydrogen fluoride always contain respectively some oxygen and some water. A method for determination of oxygen in titanium oxide and other oxides by the use of bromine trifiuoride has been described. According to preliminary work, the method is applicable to determination of oxygen in titanium metal. However, certain elements in titanium alloys appear to present difliculty.

Oxygen in steels, iron powder, bismuth and copper is determined by reduction with hydrogen. This is impracticable with titanium. On treatment with hydrogen, titanium absorbs hydrogen with the result that the contained oxides are merely reduced to lower oxides. Solution and electrolytic techniques, frequently used in the case of steels, whereby the metal is dissolved and the oxide remains, are inapplicable to titanium for the reason that the oxides present in the latter dissolve nearly as readily as the metal itself. Because titanium does not amalgamate, amalgamation techniques used to determine oxygen in sodium and lead, whereby the metal is amalgamated and the oxide remains unattached, are inapplicable. For the reason that titanium has nearly as great an afiinity for oxygen as it has for aluminum, the method of melting with metallic aluminum and then determining the aluminum oxide formed, as is used in the case of steels, is inapplicable. On melting aluminum with a titanium alloy containing oxygen, an alloy of titanium, oxygen and aluminum is formed by the oxygen remains tied to the titanium.

The method of our invention comprises conversion of the metal of titanium and its alloys to metallic bromides and the oxygen to carbon monoxide. More particularly our method comprises: reaction of titanium or its alloys, in the presence of carbon in excess of stoichiometric quantity, with bromine vapor introduced into the reaction by a purified, inert carrier gas; separation of reaction products to isolate oxygen as carbon monoxide; conversion of carbon monoxide to carbon dioxide; and measurement of oxygen as carbon dioxide. The equation of the reaction is believed to be:

Our method will be described according to its three principal phases and corresponding zones of apparatus train. These are:

1. Purification of carrier gas in a gas purification zone. 2. Reaction of sample with bromine vapor in presence of carbon in a reaction zone.

'rier gas is charged with bromine vapor.

Accordingly, with reference to the diagram forming a part of this application, an inert carrier gas, e. g., U. S. Army Class A helium, is released from a container 1 by a needle valve 2 through a flow meter 3 whereby rate of flow of the gas may be observed and adjusted. A preferred flow rate is about 130 milliliters per minute. The illustrated flow meter is a mercury manometer type; the mercury is illustrated at 4. A mercury blow-ofi valve 5 is next inserted in the apparatus train as a safety measure against build-up of hazardous pressure. Thence, the gas passes through a tube 7 containing copper turnings 1i) upstream and copper oxide 11 in wire form downstream. In this tube which in turn is contained in an electric heater 6 and sealed from the atmosphere by rubber stoppers at both ends as at numeral 8, the carrier gas is purified of contained oxygen in passing over the copper turnings, and carbon monoxide, hydrogen and hydrocarbons are converted to carbon dioxide and water in passing over the copper oxide. Glass wool 9 is used at both ends of tube 7 to holdreagents in place. From the tube 7 which is maintained at a temperature of 500 C., the gas passes to a drying tower 12 and therein successively through glass wool 9, sodium hydroxide-on-asbestos 13, anhydrous magnesium perchlorate 14 and again glass wool, for removal of carbon dioxide, sulphur dioxide and water. The foregoing constitutes the purification-of the-carriergas-phase and corresponding zone of apparatus train.

Through a washing bottle 15, sealed with a ground glass joint 16 and arranged as a safety trap to protect against a reversal of flow, the purified gas passes by way of a ball-joint type connection 17 through bromine 19 and supernatant 98% sulphuric acid 20 in a second 'washing bottle 15. A by-pass of the bromine and sulcarrier gas to a point of equilibrium prior to each determination. In passing through the bromine the car- In passing through the sulphuric acid the bromine is relieved of traces ofmoisture. The bromine is thus introduced into 21 and situated in a heater equipped with a pyrometer 26, Within the reaction tube 22 the sample in admixture with carbon is contained in a boat 24 which may be of gold, platinum or silica.- at 8-15 C. and is continued until conversion of the entire; sample has been attained as evidenced by a disappearance of cloudiness in the vapors efiiuent from the reaction tube. The foregoing constitutes the reaction phase and corresponding zone of apparatus train.

Theproducts of reaction emanating from tube 22 are passed 11110 an Erlenmeyer trap ZSwhich -iscooled moderately with ice-water contained in tray 29 so as to achieve partial condensation of metallic bromides and bromine. Thefiow continues from side arm 23 successively through two cylindrical traps 31 havinginlet tubes 32 widened at their lower extremities. The first of the traps is shallowly cooled with a Dry Ice-in-alcohol bath-34-contained in a beaker 33; the second trap is more. deeply; immersed ing the same. refrigerant which is also contained in a beaker 33. By means-of the traps 28and 31, the metallic bromides and bromine are condensed and frozen out. The widened extremities of inlet tubes 32 avoided occlusion by frozen products. Through a third cylindrical flask 31, serving as a bubbler for observing rate of flow at that point in the apparatus trainandcontaining. sulphuric acid 20, the uncondensed produets :flo .v intoasecond drying tower l2'containing glass wool 9, sodium hydroxide-on-asbestos 13,- anhydrous magnesium. perchlorate 14 anda second layer of-vglass wool. Again, thefunction ofthe glass wool is to hold reagents in place. In this second drying tower sub- The reaction is carried out 4 stances having acidic properties are removed by the sodium hydroxide, and the resultant water of reaction by the anhydrous magnesium perchlorate. Thence the flow is through a tube 35 situated in a heater 36, and containing granulated silver 37. Tube 35 is maintained at a temperature of 600 C. The last traces of brominated carbon compounds are removed by the silver in this tube. The efiiuent of tube 35 which consists of carbon monoxide and possibly hydrogen liberated from the sample is passed through tube 38 positioned in heater 39 and containing copper oxide 11 in wire form. Tube 38 is maintained at a temperature of 500 C. The carbon monoxide and hydrogen are converted herein to carbon dioxide and water which pass to a third drying tower 12 containing anhydrous magnesium perchlorate 14 supported by glass wool 9. The water is removed, leaving only carbon dioxide and the inert carrier gas to enter an absorption. bulb or weighing bottle 40. The reagents in bottle 40 are sodium hydroxide on asbestos 13 for absorption of carbon dioxide by conversion to sodium carbonate, and anhydrous magnesium perchlorate 14 for absorption of the water produced in absorbing the carbon dioxide. The weighing bottle is exhausted to atmosphere through a safety trap 41 and final bubb'ler 42 containing sulphuric acid 20. The foregoing constitutes the separation and conversion phase and corre sponding zone of apparatus train. 7

According to our method acarbon blank is deducted. The blank which is of amagnitude of 0.9 milligram of carbon dioxide per hour (equivalent to about 0:33 milligram of oxygen per hour) is consistent and repro ducible. It is obtained by passing bromine over 1 .5' grams of heated carbon for three hours. Typical hourly blank readings obtained are as-follows:

Average hourly blank for 3 runs 0.88 mg.

Standard deviation for all hourly blanks- 1.08 mg.

A preferred practice of our method comprises preparation 0f a'sample by drilling, chipping or st'rippinginto pieces of about 0.02 inch thickness. The pieces of sample are then' cleaned with carbon tetrachloride and dried thoroughly in an oven at 40 0. Two grams of sample and 1.5 grams of graphite.powder-, by way of example, are then placed in a boat; half the graphite powder is first placed in the bottom of the boat, the sample is then placedthereon and'the balance of the graphite is usedto cover the sample. 7

The silver. utilized in the separation of last'traces of:

b'rominated hydrocarbons is specially. prepared; 450" grams of silver nitrate are dissolved'in asolution of10 milliliters of concentrated nitric acid in 3.5 liters of, water.

A copper tube or'a thick sheet of copper rolled into tubunatant'liquidis then decanted. The silver is then'was'h'ed" several times with' dilute nitric acid approximately. '1 paif' of concentrated acid to 200 parts tapwater) and"thereafter washed with dilute ammonium hydroxide"(approximately 1 part'of concentrated ammonium"hydroxide"to' l0-parts tap 'water) until the bluecopperamine'color is" no ldngerpcrceptibl. Finaliy'the silver is washedsev cral ti-rneswith tap water, placed in atrayj'andoven 'dried The silver is then "spreadjthinly' at C1 for 1 hour. in porcelain-dishesand baked at 750 C. for l near:

Aftericooling, the silver=is handcrumbled,- and-charged into tube 35 as follows: A wad of glass wool is inserted into one end of the tube. The wadded end of the tube is supported on a thick glass rod so as to hold the wad in place during packing. The silver is inserted through the opposite end of the tube in small portions that are successively compressed with a steel rod 0.2 inch in diameter and a tapping hammer. Care must be exercised not to pack the silver so tightly that flow of gas therethrough would be prevented. Upon completion of the packing operation, the glass wool wad is removed.

Certain precautions must be observed in the practice of our method. Leaking connections between elements of the apparatus train must be avoided. Leakage upstream of the reaction tube 22 can produce results that are high, and leakage downstream of the reaction tube can cause very low results. Connections upstream of the reaction tube should be tested with ammonium hydroxide which when brought close to a leaking connection will form a white cloud of ammonium bromide. Downstream of the reaction tube, connections should be tested with soap solution which, upon completion of testing, is rinsed off with water from a wash bottle. The bubbler should be watched as an indicator of leakage in the apparatus; any fluctuation or faltering in bubble rate is indicative of leakage. To achieve a leak-free laboratory apparatus train, it is essential that ground-glass joints, ball joints and stop-cocks be ground together with a paste made of 600 mesh carborundum and water. All stopcocks should be lightly coated with a fluorinated hydrocarbon type grease. The weight of carbon dioxide recovered in the weighing bottle 40 having been determined, the percent oxygen content of the sample is computed according to the equation:

where:

W=Weight of CO2 (grams) B=Blank (grams CO2 per hour) T=Time for run (hours) M =Weight of sample (grams) The results obtained by our method differ on the average only 0.007% from results obtained with the vacuum fusion method. The reproducibility and reliability of our method are reflected by an average standard deviation in results of 0.009%.

We claim:

1. A process for quantitative determination of the oxygen content of titanium comprising: purification of an inert gas by successive contact with copper, copper oxide, sodium hydroxide and anhydrous magnesium perchlorate; saturation of the purified gas with bromine vapor by passing said gas as a carrier through liquid bromine; dehydration of the bromine vapor by conducting the bromine-saturated gas through concentrated sulphuric acid; reaction of the bromine with a sampleimrnersed in carbon in excess of stoichiometric quantity by conducting the bromine-saturated gas over the sample; separation from the products of reaction of metallic bromide and unreacted bromine by condensation in a first separation step; separation from the products of reaction of brominated carbon compounds by passing over silver in a second separation step; conversion of the remaining products of reaction to carbon dioxide and water by passing over copper oxide; removal of the water by passing through anhydrous magnesium perchlorate; and collection of the carbon dioxide as sodium carbonate and water by passing through sodium hydroxide and anhydrous magnesium perchlorate.

2. A process for quantitative determination of the oxygen content of titanium alloys comprising: purification of an inert gas by successive contact with copper, copper 6 I oxide, sodium hydroxide and anhydrous magnesium perchlorate; saturation of the purified gas with bromine vapor by passing said gas as a carrier through liquid bromine; dehydration of the bromine vapor by conducting the bromine-saturated gas through concentrated sulphuric acid; reaction of the bromine with a sample immersed in carbon in excess of stoichiometric quantity by conducting the bromine-saturated gas over the sample; separation from the products of reaction of metallic bromides and unreacted bromine by condensation in a first separation step; separation from the products of reaction. of brominated carbon compounds by passing over silver in a second separation step; conversion of the remaining products of reaction to carbon dioxide and water by passing over copper oxide; removal of the water by passing through anhydrous magnesium perchlorate; and collection of the carbon dioxide as sodium carbonate and water by passing through sodium hydroxide and anhydrous magnesium perchlorate.

3. A process for quantitative determination of the oxygen content of titanium comprising: purification of an inert gas at a rate of about milliliters per minute by successive contact with copper and copper oxide maintained at about 500 0., sodium hydroxide and anhydrous magnesium perchlorate; saturation of the purified gas with bromine vapor by passing said gas as a carrier through liquid bromine; dehydration of the bromine vapor by conducting the bromine-saturated gas through sulphuric acid; reaction of the bromine with a sample immersed in carbon in excess of stoichiometric quantity and maintained at about 815 C., by conducting the bromine-saturated gas over the sample; separation from the precincts of reaction of metallic bromide and unreacted bromine by progressive exposure to refrigerant at from about 0" C. to about minus C.; separation from the products of rection of brominated carbon compounds by passing over silver maintained at about 600 0; conversion of the remaining products of reaction to carbon dioxide and water by passing over copper oxide maintained at about 500 C. removal of the water by passing through anhydrous magnesium perchlorate; and collection of the carbon dioxide as sodium carbonate and water by passing through sodium hydroxide and anhydrous magnesium perchlorate.

4. A process for quantitative determination or" the oxygen content of titanium alloys comprising: purification of an inert gas at a rate of about .130 milliliters per minute by successive contact with copper and copper oxide maintained at about 500 0, sodium hydroxide and anhydrous magnesium perchlorate; saturation of the purified gas with bromine vapor by passing said gas as a carrier through liquid bromine; dehydration of the bromine vapor by conducting the bromine-saturated gas through sulphuric acid; reaction of the bromine with a sample immersed in carbon in excess of stoichiomctric quantity and maintained at about 815 C., by conducting the brominosaturated gas over the sample; separation from the products of reaction of metallic bromides and unreacted bromine by progressive exposure to refrigerant at from about 0 C. to about minus 170 C.; separation from the products of reaction of brominated carbon compounds by passing over silver maintained at about 600 C.; conversion of the remaining products of reaction to carbon dioxide and water by passing over copper oxide maintained at about 500 C. removal of the water by passing through anhydrous magnesium perchlorate; and collection of the carbon dioxide as sodium carbonate and water by passing through sodium hydroxide and anhydrous magnesium perchlorate.

References Cited in the file of this patent Titanium Barksdale, page 85, The Ronald Press Co. (1949).

Claims (1)

1. 4. A PROCESS FOR QUANTITATIVE DETERMINATION OF THE OXYGEN CONTENT OF TITANIUM ALLOYS COMPRISING: PURIFICATION OF AN INERT GAS AT A RATE OF ABOUT 130 MILLILITERS PER MINUTE BY SUCCESSIVE CONTACT WITH COPPER AND COPPER OXIDE MAINTAINED AT ABOUT 500* C., SODIUM HYDROXIDE AND ANHYDROUS MAGNESIUM PERCHLORATE; SATURATION OF THE PURIFIED GAS WITH BROMINE VAPOR BY PASSING SAID GAS AS A CARRIER THROUGH LIQUID BROMINE; DEHYDRATION OF THE BROMINE VAPOR BY CONDUCTING THE BROMINE-SATURATED GAS THROUGH SULPHURIC ACID; REACTION OF THE BROMINE WITH A SAMPLE IMMERSED IN CARBON IN EXCESS OF STOICHIOMETRIC QUANTITY AND MAINTAINED AT ABOUT 815 * C., BY CONDUCTING THE BROMINE-SATURATED GAS OVER THE SAMPLE; SEPARATION FROM THE PRODUCTS OF REACTION OF METALLIC BROMIDES AND UNREACTED BROMINE BY PROGRESSIVE EXPOSURE TO REFRIGERANT AT FROM ABOUT 0* C. TO ABOUT MINUS 170* C.; SEPARATION FROM THE PRODUCTS OF

Images