US2614028A - Method of superheating titanium tetrachloride - Google Patents

Description

Oct. 14, 1952 H. H. SCHAUMANN 2,614,028

METHOD OF SUPERHEATING TITANIUM TETRACHLORIDE Filed July 16, 1947 i H H M 5 HEATED souos COOLED HOT 8FLUE GAS souos 501.105

co L 3 HALIDE L J r COOL HALIDE ST REA M I MB X: T AIR 1 co L 501.105 & FUEL. saga: GASES SPARK 12 L DRAINS INVENTOR. HOLGER H. SCHAUMANN HIS AGE Patentecl Oct. 14, 1952 METHOD OF SUPERHEATING TITANIUM TETRACHLORIDE lilolger H. Schaumann, Newark, Del., assignor to E. I. du Pont de Nemours & Company, Wilmington, Del., a corporation of Delaware Application July 16, 1947, Serial No. 761,227

1 Claim.

This invention relates to methods for heating gaseous materials, and more particularly to the vaporizing and superheating of metal halides. It further concerns the application theret of certain fluidized solid techniques.

Many important industrial operations utilize metal halides in various forms. Principal among these is the preparation of metal oxides by the treatment of metal ores with halogen vapors and the subsequent oxidation of the resulting gaseous metal halides. High temperatures are required to initiate these oxidation reactions, so that heat must usually be added to the reactants. It is well recognized that the most efiicient method of supplying this initial heat is by preheating the halide vapors and the oxidizing gas, each separately, prior to mixing. The temperature to which they are preheated, and the accuracy with which such temperatures can be controlled, will determine to a large extent the quality of the metal oxide product formed upon their admixture. For a pigment-useful material, it is best to preheat the oxidizing gas and separately to superheat the metal halide vapors to a temperature of at least about 75 C. above their boiling pointand preferably to approximately 350 C. or more above their boiling point. The metal halides may at the start be at a temperature below their boiling point, and so in liquid or solid form. They must then of course be vaporized and those vapors subsequently superheated, an operation requiring an even greater output of energy by the heating means employed. It has been exceedingly difiicult so to heat large volumes of these halides, because of the effect of the high temperatures, together with the highly corrosive nature of the halogens, on the materials of construction of suitable apparatus. Such halides have usually been heated by passage through conventional heat-exchangers wherein they are circulated around a complex structure of metal tubes through which is passing a hot fluid. The fluid releases its heat content to the halide, whereby the former is cooled while the latter is correspondingly heated. Halide vapors objectionably corrode the said metal tubes, rendering the process excessively expensive. Heat transfer is relatively slow, so that to handle large volumes of material on a commercial scale, the equipment must be of very large size, thus adding still more to its cost. Other well-known methods are objectionable on similar grounds. None of them allows optimum efficiency in operations like the aforesaid vaporizing and superheating of metal halides.

2 Some metal halides, such as chromium, iron, or aluminum chlorides or the like, are vaporized at normal pressures directly from the solid state. Conventional heat-exchange equipment cannot generally be used for such materials, as it is difficult to handle solids therein. The solids tend overcome these and other disadvantages of prior art processes. A further object is to eliect sublimation of those metal halides which pass directly from the solid to the vapor state upon the application of heat. A particular object is to heat metal halides above their boiling point prior to their oxidation to form metal oxides. Another object is to heat large volumes of such halides in equipment of relative simplicity and of relatively small'size. Yet another object is to minimize opportunity for corrosion of the equipment by the halide vapors. A further object is to heat iron halides above their boiling point prior to their oxidation to form iron oxides.

The above and other objects are attained by the following invention which in its broader aspects comprises contacting a stream of metal halide with particles of a hot solid material which has substantially no chemical efiect thereon, and which is itself essentially unaifected thereby. The said halide stream is caused to flow at'sucha velocity that the said added solids become entrained and fluidized therein. Heat transfer is efiected between the halide and the hot solids, the former being heated while the latter are cooled. If the former were initially in solid or liquid form, it is vaporized; if already in the gas phase, it is further heated.

In one preferred embodiment of the invention, relatively small particles of a solid material such as quartz sand or other form of silica are heated by any desired means and then entrained in a stream of relatively cool iron chloride. The iron chloride may be in the form of a vapor or in the form of a solid entrained in a stream of an inert carrier gas. The solid particles give up their heat to the said iron chloride, thus superheating it (preliminarily vaporizing it if it were ini- 3 tially a solid). The corresponding cooled added solids are separated from the resulting halide vapors by conventional means, to be reheated and recycled to effect heating of further quantities of iron chloride.

The choice of the particular solids to be used as the heating means is governed primarily by two critical factors. It is essential to the invention that the material be one which has substantially no chemical effect on the halides. It is also essential that it be itself substantially unaffected by the highly corrosive, high temperature conditions obtaining in the process. Natural forms of silica, such as quartz sand, are frequently satisfactory. Another excellent material is a quantity of the oxide of the same metal itself. With the latter there is even less opportunity for contamination of the metal halide vapors.

The degree of fluidity imparted to the heating solids is relatively immaterial and Will depend upon the type of equipment used. The velocity of the injected halide stream may be so low that the solids are caused simply to oscillate gently within a small area, in a manner analogous to the boiling of a liquid. Means may be provided for bleeding off this bubbling bed as it becomes cooled and adding fresh, hot material. Again, the gas velocity may be such that the solids are actually entrained and conveyed through the system with the halide current, to be separated therefrom later by some conventional device. It is generally true that with usual commercial equipment and moderate solids content, gas velocities of from approximately A.; to 2 feet per second are satisfactory to establish a bubbling bed, and of 15 to 75 feet per second for a conveying bed. The diameter and length of the particular chamber through which the materials are passing, the volume and particle size of the heating solids employed, and various other factors must be considered in determining these rates. If the halide is in solid form, it or the hot solid is first entrained in an inert carrier gas, thereby providing the stream in which the other solid ma be fluidized to effect intimate contact. Separation of the solids from the vapors after heat-exchange has been effected, may be carried out in any manner known to the art, such as by cyclone separators or the like.

One specific system utilizing a conveying-type contactor for cyclic operation is shown in the accompanying semi-schematic drawing. In operation, valves 1 and 8 are closed initially and 9 is opened. Cold solids in chamber 2 (shown for convenience as a cyclone-type separator) drop downward through line and valve 9 into line 6, where they are fluidized in a stream of hot gases resulting from the combustion at l2 of any desired fuel. The solids are conveyed upward through line 6, being heated in transit by the hot gases, and returned to chamber 2, where they are separated therefrom. The flue gases exit from the system. Valve 9 is then closed, valves 1 and 8 opened, and a stream of either solid, liquid or vaporous metal halide is injected into 1ine'3. The heated solids in chamber 2 drop down through 1 into the halide current flowing through line 3, are entrained therein and conveyed into chamber I, which may also be a cyclone-type separator, yielding their heat content to the cool halide on the way, thus vaporizing and superheating it. The thus-heated halide vapors are separated out in I, the correspondingly cooled solids dropping back into line 4 and passingthrough'valve 8, again entering the stream of hot combustion gases flowing through line 6, there to be reheated for re-use. In continuous operation, solids passing downward through line 5 contain trapped fiue gases in the voids between the particles; similarly, solids in line 4 contain trapped halide vapors. To obviat this diificulty, conventional seal pots may be introduced into these two lines if desired, say at areas l0 and I I. These cause the solids to assume a free conical surface at the normal angle of repose. A purge of an inert gas may then be passed therethrough, removing the flue gas or halide from the voids and maintaining substantially complete purity in the system. The dilution by this added inert gas has a negligible effect on the entire process.

One of the principal reasons for the efficiency of this fluidized solids technique is the fact that a large surface area is provided by the relatively finely divided material. For a given volume of solids, for instance, the smaller the individual particles thereof, the greater the hot surface exposed for contact with the vapors, hence the more rapid the heating effect. The quantity of solids used is interdependent with their particle size, with the degree of heating desired, the character of the gases, the type of bed, and the length of time the gases and the solids are in contact. For a given rate of heating, a correspondingly lesser quantity of solid material is theoretically required the higher the surface area of its discrete particles. Practically speaking, however, a degree of subdivision is reached beyond which the relative influences of these factors are impossible to measure. The particles are preferably sufiiciently large, of course, to be palpable and to flow easily. It is desirable that the suspension of solids in the gases be relatively fluid, so that the gas velocity or pressure required to set it and to continue it in motion be not excessive. An especially good type of solids has been found to be quartz sand of 20+30 mesh (or about 760 microns diameter) particle size; but any size of pellets or granules or the like may be operable depending upon the various other conditions of operation. For instance in commercial production, a broad range of 30 microns to 1 inch diameter pellets can be used effectively. It has usually been found that suspensions having a solids content greater than 50 pounds of such sand particles per pound of metal halide vapor will be too dense for practical handling, the gas velocities required to establish a conveying bed being excessively high. Hence from about 1 to 50 pounds of solids per pound of vapor are best, and from about 5 to 30 pounds of solids per pound of metal halide are often sufficient.

Any desired means for heating the solid material prior to its use in the present process may be employed. Previously mentioned conventional heat-exchangers are often used. One good method, referred to hereinabove and illustrated in the accompanying figure, is to effect combustion of a fuel and air or oxygen and to entrain the solids in the stream of hot gaseous combustion products. A fuel may be chosen which will give almost any desired temperature, and the solids will be heated to varying degrees depending on such choice. For instance, a particularly high heat is obtained by burning acetylene and oxygen; a lower heat, by burning producer gas and air. Various oils, natural and artificial gases and the like ar also useful. Preferably, a fuel should be selected which is low in ash. With such a technique the temperature of the resulting hot solids can be quite accurately controlled,

5 so that their subsequent heat-exchange with the relatively cold metal halides will be similarly ascertainable; Alernatively, the hot solids might come from some separate cooling operation, wherein they have been used cold, to cool some hot material. They have been themselves heat ed thereby, and in this condition may be circulated tothe present system. It is further possible that after their use in the present process, whereby they are again cooled, they may be recycled for re-use in the aforementioned quenching system. i

Apparatus for separating the solids from the gasesmay be of any type conventionally used. Cyclonesare schematically shown in the accompanyin drawing. However, equally operable un-- der various conditions are filters, settling chambers, electrostatic precipitators, and the like,

Essentially any volatilizable metal halide. will be operable in theinvention, e. g., iron chlorides, bromides,- iodides; aluminum, chromium, zirconium, titanium halides and the like. The fluorides, bromides and iodides are used relatively infrequently. The chlorides are preferred on grounds of availability, cost, etc. However, it is to be understood that any one of the said halides, or mixtures thereof, can be employed if desired. The temperature ranges obtained in the process will naturally vary, depending on the particular halide, because of the difierence in boiling point of each compound.

The following examples are given to illustrate this invention, but not in any way to limit its scope:

Erample I Apparatus similar to that shown in the accompanying drawing was used in this experiment. The material of construction was silica 'Pertinent dimensions of the various members were: overall heightl5 feet; line 62 inches I. D.; lines 4 and 51% inches; and line 3- inch I. D. Chamber 2 was a 9 inch cyclone; chamber i, a 3 inch cyclone. At areas l and III were inserted 6 inch by 6 inch seal pots. Purified solid iron chloride, entrained in nitrogen as a conveying gas and en route to an oxidation chamber for reaction with oxygen to form iron oxides, was circulated to the system. Quartz sand particles having an average diameter of 700 microns had been added to chambers l and Z in an amount to fill the two standpipes 4 and 5. Valves l, 8, and 9 were new devices known as jet valves and described in a co-pending application by C. H. Winter, Jr., Serial Number 699,194, filed September 25, 1946. Valves 7 and 3 were closed and 9 was opened. Sand dropped downward from chamber 2 through line and out through valve 0 into line 6. Meanwhile, air and propane gas were mixed and ignited to produce a flame at 2. The hot combustion gases therefrom travelled through 0, picking up the sand particles and conveying them upward through 6 back into chamber 2. Heat-exchange occurred between the hot gases and the cold sand, whereby the latter was heated to a temperature of about 700 C. The rate of flow of this flue gas was roughly 300 pounds per hour or 5 pounds per minute, and its temperature was about 1030 C. It was cooled by the exchange to about 700 C. and exited at the stack of cyclone 2, while the similarly heated sand dropped back into line 5.

Valve 0 was at this point closed, and 'l and 8 were opened. The flow of the iron chloride stream into line 3' wasbegun. This stream was at a temperature of about C., and it was dethe solid state at 20 C. to the vapor at 500 0.,

is 200 P. C. U. per pound. The heat required to raise the temperature of 125 pounds of it from 20 to500 C. was calculated from this to be (200x or 25,000 P. C. U. per hour. (1 P. C. U.

is the amount of heat required to raise the temperature of 1 pound of water 1 degree centigrade; 1 P. C. U.=1.8 B. t. u.) The sand, havinga specific heat of .25, the heat liberated by 1 poundof it incoolin from 700 C. to 500C. was found to be ('700-500) (.25) or 50 P. C. U. Thus to furnish the 25,000 P. C. U. required to heat 125 pounds of the iron chloride to 500 C., it was contacted with 25,000/50 or 500 pounds of sand per hour, or 4 pounds of sand perv pound of chloride.

The cool iron chloride stream entered line 3 and passed upward therethrough. Simultaneously hot sand particles from chamber 2 dropped through line 5 and out of valve 1 into this line, where they were picked up and entrained in the FeClanitrogen current. i was so regulated that the sand entered line 3 at a e to give the requisite proportion of i pounds per pound of chloride. The solid iron chloride was heated by the contact withthe hot sand and vaporized by the time the mass reached chamber i. In if, thevapors and the sand were separated, and the former passed out of the stack at 500 0., ready for the oxidation reaction. The sandwhichremained in I had of course given up some of its heat to the iron chloride and been cooled to 500 C. This material dropped downward through line Aland out Example I I The apparatus of Example I was again used, with the single variation that conduit 3 was 19 inches in I. D. This time titanium tetrachloride vapor was superheated in the equipment prior to its reaction with an oxygen-containing gas to prepare titanium dioxide pigment.

Quartz sand particles, averaging 700 microns in diameter, were circulated through the system by the procedure of Example I. They were heated in line 0 to a temperature of 1000 C. by hot combustion gases at about 1100 C. resulting from the burning at :2 of air and acetylene. TiCl4 vapor at 138 C. was then introduced into the system through line 3 at a, velocity of 250 pounds per hour, and was contacted with the hot sand. Two pounds of sand per pound of vapor or 500 pounds of sand per hour were added, thus heating the vapor from 138 C. to 800 C. The superheated vapor was then separated from the correspondingly cooled sand in chamber I, the former to be passed onward to the oxidation reaction, the latter to be reheated as in Example I and reused.

Example III The heating system and procedure of Example I were employed to vaporize and superheat A1C13. 240 pounds per hour of similar quartz sand were circulated through the equipment, being heated in line 6 by contact with gases at 1030 C. resulting from the combustion of city gas in air and flowing through line 6 at the rate of 150 pounds per hour. A stream of AlC13 powder at 20 C., entrained in nitrogen in proportions to give a flow through the system of 100 pounds per hour of AlCls and pounds per hour of N2, entered at conduit 3. Heat exchange occurred between the A1013 and the hot sand, whereby the former was sublimed and its vapors superheated to about 500 C., and the latter was cooled to about the same temperature. The hot vapors were separated out in chamber I, and the cooler solids were returned to line 6 to be reheated and recycled as in Example I. p

The many advantages of the present invention become obvious to those skilled in the art. A high degree of heating eiliciency is obtained for a comparatively small volume of the heating medium; and the equipment required is correspondingly simpler and of relatively smaller size than for any system generally in use today. The large area, of hot surface provided by the solids allows optimum heat-exchange conditions. The technique easily lends itself to cyclic operation-s, because the heating solids may be efficiently separated from the vapors and reheated and reused. The solids have no chemical effect on the vapors and are not themselves adversely affected. There is no danger of objectionable partial oxidation or hydrolysis of the pure halide vapors, but they exit in the same condition of purity in which they entered the system. The process is of such eflicacy that it is not necessary to prevaporize the metal halide unless desired. This has previously been practically essential in other methods, in order to lessen the burden placed upon the usual heat-exchangers. However, in the present invention the solid or liquid form may be used, and can beboth vaporized and then superheated with essentially no more difficulty than mere superheating alone. This new operation is thus more effective and less costly than those complex processes heretofore employed.

I claim:

A method for superheating titanium tetrachloride to 800 C. which comprises entraining in titanium tetrachloride vapor at a temperature of 138 C. quartz sand particles at a temperature of 1000 C. and which average 700 microns in diameter, utilizing in the entraining operation 3. velocity of 250 pounds per hour of TiCh vapor and 2 pounds of sand per pound of titanium tetrachloride vapor, flowing the resulting mixture in a confined, elongated stream as a fluidized suspension and until desired heat exchange between said chloride vapors and sand is effected, and then separating said sand from the superheated titanium tetrachloride vapors.

HOLGER H. SCHAUMANN.

REFERENCES CITED,

The following references are of record in the file of this patent:

UNITED STATES PATENTS

Images