Method of manufacturing aluminum fluoride anhydride

BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a method of manufacturing aluminum fluoride anhydride, and more particularly to a method of effectively manufacturing aluminum fluoride anhydride having an average particle diameter of 100 .mu.m or more by means of a wet method.
2. Description of the Related Art
Aluminum fluoride anhydride is very useful as a flux in electrolytic reduction of alumina, or as a raw material for ceramics such as an optical glass. As the manufacturing method of aluminum fluoride anhydride, there are known two methods. i.e. a dry method wherein anhydrous hydrogen fluoride is used as a source of fluorine for manufacturing aluminum fluoride anhydride, and a wet method wherein a fluosilicic acid solution or a hydrofluoric acid solution is used as a source of fluorine for manufacturing aluminum fluoride anhydride.
The aluminum fluoride anhydride which is obtained from the dry method contains 88 to 92% of aluminum fluoride, and has a bulk density of 1.5 to 1.6 g/cm.sup.3, a shape of angular aggregated body similar to the raw aluminum hydroxide and a relatively large size, i.e. 80 to 100 .mu.m in average particle diameter
By contrast, aluminum fluoride anhydride to be obtained from the wet method has a content of aluminum fluoride which is higher than that obtainable by the dry method mentioned above, i.e. 95 to 98% as the aluminum fluoride anhydride according to the wet method is manufactured via an intermediate product of aluminum fluoride trihydrate. However, the aluminum fluoride anhydride manufactured by this wet method is porous, as pores are formed in the particle as a result of the dehydration of crystallization water during the manufacturing process, thereby lowering the bulk density of aluminum fluoride anhydride to about 0.8 g/cm.sup.3.
Moreover, the shape of aluminum fluoride anhydride obtained from this wet method is poor in uniformity, as the particle of aluminum fluoride anhydride is formed through aggregation and crystal growth.
Thus, the size of the particle of aluminum fluoride anhydride generally ranges from 60 to 70 .mu.m in average particle diameter, and is generally smaller than that of aluminum fluoride anhydride obtainable in the dry method mentioned above. Since the particles of aluminum fluoride anhydride obtained from the wet method are light and fine as mentioned above, there are pointed out many problems for using these particles of aluminum fluoride anhydride for flux in electrolytic reduction of alumina. For example, when the particles of aluminum fluoride anhydride obtained from the wet method are to be charged into an electric furnace for flux in electrolytic reduction of alumina, the fluidity of the particles of aluminum fluoride anhydride during pneumatic transportion occasionally becomes too poor to charge a predetermined amount of them into the furnace. There is another problem that the particles of aluminum fluoride anhydride charged into aluminum electrolyzing cell are easily scattered by the inner air stream within the furnace, and the particles thus scattered escape out of the furnace through an exhaust stream, thereby losing a relatively large amount of aluminum fluoride anhydride.
Due to these problems, it has been considered that the particles of aluminum fluoride anhydride to be obtained from the wet method are not suited for use in electrolytic reduction of alumina requiring a large amount of aluminum fluoride anhydride, and in fact the particles of aluminum fluoride anhydride obtained from the wet method have scarcely been utilized for the electrolytic reduction of alumina.
In view of handling and the prevention of generation of dust, if aluminum fluoride anhydride is to be used for electrolytic reduction of alumina, it is generally desired to use aluminum fluoride anhydride having an average particle diameter of 100 .mu.m or more with the particles having 44 .mu.m or less being controlled to not more than 1%. However, a technique of effectively producing, by way of a wet method, aluminum fluoride anhydride having a controlled particle size as mentioned above has not been available to date.
Fluosilicic acid, which is a raw material useful for producing aluminum fluoride anhydride by a wet method, is by-produced in large amount in the manufacture of phosphoric acid from for example phosphate rock. To date, this fluosilicic acid has been scarcely utilized, and mostly discarded as useless. However, in view of saving natural resources, and reducing environmental pollution, it is desired to find a way to utilize the fluosilicic acid.
For responding the need as mentioned above, there have been already proposed various methods to manufacture aluminum fluoride anhydride through a wet method. For example, East German Patent No. 88,080 describes a method of manufacturing aluminum fluoride anhydride wherein aluminum fluoride trihydrate is produced from a supersaturated aluminum fluoride solution, which is obtained by reacting fluosilicic acid with aluminum hydroxide. In this method the generation of fine particles during the step of calcination/dehydration are controlled to a minimum by controlling the agitation speed during the crystallization thereof. However, the method employed In this East German Patent uses no seed crystal, and moreover the crystallization is conducted only once, thus failing to effectively produce aluminum fluoride anhydride of large particle size.
East German Patent No. 248,349 describes a method of precipitating aluminum fluoride trihydrate of large particle size from a super-saturated solution of aluminum fluoride, the composition and concentration of the solution being controlled respectively to 3<F/Al.ltoreq.6, and 1-2.5 mol/L or less. However this method is defective in that the yield of aluminum fluoride trihydrate is very low, i.e. 55%.
Further, U.S. Pat. No. 3,057,681 discloses a method of crystallizing AlF.sub.3.3H.sub.2 O in a batch or continuous method from a super-saturated solution of aluminum fluoride, wherein the crystallization is performed by using a seed crystal, and fine particles contained in the resultant product are removed therefrom to re-use as a seed crystal.
In this U.S. Patent however, nothing is taught about the amount of particles to be used as a seed crystal, which is one of the very important crystallization factors, and, on the contrary, all of the fine particles settled in a thickener are re-used as a seed crystal for the subsequent crystallization process, thereby permitting a lot of fine particles to be produced in every crystallization step, thus raising a problem.
In view of this problem, the method proposed in this U.S. Patent is not suited for efficiently producing AlF.sub.3.3H.sub.2 O of larger particle size, thereby making this technique difficult to adopt to the real industrial production of AlF.sub.3.3H.sub.2 O.
SUMMARY OF THE INVENTION
An object of this invention is to provide a method of manufacturing AlF.sub.3.3H.sub.2 O of larger particle diameter in high efficiency from a super-saturated solution of aluminum fluoride by using a wet method and by utilizing, as a fluorine source, a by-product produced in the processing of phosphate rock. Another object of this invention is to provide a method of manufacturing aluminum fluoride anhydride having an average particle diameter of 100 .mu.m or more, which is suited for utilization in electrolytic reduction of alumina.
Another object of this invention is to provides a method of manufacturing in high efficiency aluminum fluoride anhydride having a larger particle diameter.
According to this invention, there is provided a method of manufacturing aluminum fluoride anhydride, which comprises the steps of preparing a super-saturated solution of aluminum fluoride; adding into the super-saturated solution of aluminum fluoride m seed crystal of AlF.sub.3.3H.sub.2 O containing not more than 5% of fine particles of 40 .mu.m or less in particle diameter in such a ratio that the total surface area of the seed crystal is in the range of 40-100 m.sup.2 per 1 kg of AlF.sub.3.3H.sub.2 O to be precipitated during a time period between the moment of initiating crytallization and the moment where an initial concentration of the super-saturated solution of aluminum fluoride is reduced to 1.6%; adjusting the initial concentration of the super-saturated solution of aluminum fluoride to 8 to 15%; heating the resultant super-saturated solution of aluminum fluoride to a temperature ranging from 75.degree. to 106.degree. C.; maintaining the super-saturated solution of aluminum fluoride at this temperature under agitation, thereby allowing crystals to be precipitated from the super-saturated solution of aluminum fluoride through a batch crystallization method to obtain a slurry of AlF.sub.3.3H.sub.2 O; separating AlF.sub.3.3H.sub.2 O having an average particle diameter of 100 .mu.m or more from the slurry of AlF.sub.3.3H.sub.2 O; drying the AlF.sub.3.3H.sub.2 O thus separated and dehydrating the AlF.sub.3.3H.sub.2 O thus dried.
According to this invention, there is further provide a method of manufacturing aluminum fluoride anhydride, wherein the AlF.sub.3.3H.sub.2 O having a relatively large particle diameter is separated from the slurry of AlF.sub.3.3H.sub.2 O in above-mentioned steps and re-used as a seed crystal, and the above-mentioned steps of producing the AlF.sub.3.3H.sub.2 O crystal are repeated until aluminum fluoride anhydride having a desired particle diameter is obtained.
According to this invention, there is further provided a method of manufacturing aluminum fluoride anhydride, wherein the seed crystal of AlF.sub.3.3H.sub.2 O containing not more than 5% of fine particle of 40 .mu.m or less in particle diameter is used in a form of slurry. According to this invention, there is further provided a method of manufacturing aluminum fluoride anhydride, wherein the seed crystal of AlF.sub.3.3H.sub.2 O containing not more than 5% of fine particle of 40 .mu.m or less in particle diameter is used in a form of filter cake of slurry.
According to this invention, the temperature of the super-saturated solution of aluminum fluoride may preferably be controlled to range from 85.degree. to 95.degree. C.
According to this invention, the step of precipitation is conducted under agitation so that the AlF.sub.3.3H.sub.2 O is not accumulated at the bottom of a crystallizer. According to this invention, the steps of obtaining AlF.sub.3.3H.sub.2 O crystal may be repeated until AlF.sub.3.3H.sub.2 O having a particle diameter of 100 .mu.m or more occupies not less than 60% of whole AlF.sub.3.3H.sub.2 O produced.
Details of this invention will be more apparent from the following explanations.

Additional objects and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objects and advantages of the invention may be realized and obtained by means of the instrumentalities and combinations particularly pointed out in the appended claims.
BRIEF DESCRIPTION OF THE DRAWINGS
The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate presently preferred embodiments of the invention, and together with the general description given above and the detailed description of the preferred embodiments given below, serve to explain the principles of the invention.
FIG. 1 is a scanning type electron microscopic photograph of aluminum fluoride particles having an average particle diameter of 128 .mu.m as obtained by the present invention;
FIG. 2 is an optical microscopic photograph of aluminum fluoride particles shown in FIG. 1;
FIG. 3 is a scanning type electron microscopic photograph of aluminum fluoride particles having an average particle diameter of 136 .mu.m as obtained by the present invention; and
FIG. 4 is an optical microscopic photograph of aluminum fluoride particles shown in FIG. 3.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
The inventors have made a detailed observation on the mechanism of crystal growth of AlF.sub.3.3H.sub.2 O in a super-saturated solution of aluminum fluoride in the process of achieving this invention. As a result, it has been confirmed by the inventors that the crystal growth of AlF.sub.3.3H.sub.2 O is proceeded by the lamination of a large number of fine AlF.sub.3.3H.sub.2 O crystals on the surface of an AlF.sub.3.3H.sub.2 O seed crystal.
It is more importantly confirmed that in the case of an aluminum fluoride solution which is higher in super-saturation degree, a large number of fine AlF.sub.3.3H.sub.2 O crystals are more actively laminated on the surface of an AlF.sub.3.3H.sub.2 O seed crystal, and the surface of an AlF.sub.3.3H.sub.2 O crystal after the growth is more excellent in smoothness. This is considered to be a unique phenomenon in the crystal growth of AlF.sub.3.3H.sub.2 O which is first discovered by the present inventors.
It has been generally considered according to the conventional knowledge about the crystallization that when the supersaturation degree of a solution is high, a large number of new crystals are generated in the solution, so that the growth of seed crystal added in advance thereto is greatly inhibited. Accordingly, the fact newly found by the inventors is considered to be a unique phenomenon peculiar to an aluminum fluoride solution of high super-saturation degree, which is quite different from what has been generally considered with regard to the crystal growth.
To be more specific, it has been found by the present inventors that, as far as the growth of AlF.sub.3.3H.sub.2 O seed crystal is concerned, when a solution of high super-saturation degree, for example a solution of an aluminum fluoride having a concentration of more then 8% is employed for the crystal growth, it is possible to obtain AlF.sub.3.3H.sub.2 O crystal of large particle diameter. It has been confirmed upon observation of the cross-section of the AlF.sub.3.3H.sub.2 O crystal of large particle diameter thus obtained that the growth of the crystal is extended radially and uniformly just like that of the annual ring of tree from the seed crystal forming a center core. It has been also confirmed that the surface of the crystal thus obtained is very smooth.
This invention is based on these new findings, and therefore the crystal growth is initiated using an aluminum fluoride solution of high concentration. The aluminum fluoride solution of high concentration to be used in this invention can be obtained by reacting aluminum hydroxide with fluosilicic acid. It is advantageous in this invention to carry out the crystallization by using a seed crystal of larger particle diameter, since the growth of seed crystal is effected by the lamination of a large number of fine AlF.sub.3.3H.sub.2 O crystals. However, the employment of extremely large seed crystal may invite another problem that the production of such an extremely large seed crystal itself is rather difficult.
Therefore, AlF.sub.3.3H.sub.2 O crystal containing not more than 5% of fine particle of 40 .mu.m or less in particle diameter are employed as a seed crystal according to this invention. The reason for limiting the content of fine particle of 40 .mu.m or less in particle diameter to 5% or less is that if a lot of such a fine particle is included in the seed crystal, a large amount of aluminum fluoride in the super-saturated solution is consumed for the growth of the fine particle, thereby inhibiting the effective growth of AlF.sub.3.3H.sub.2 O of larger particle size.
There is no other requirement for the seed crystal to be useful in this invention as long as the above requirements for a seed crystal are met. However, it is more preferable to employ a seed crystal containing not more than 5% of fine particle of 50 .mu.m or less in particle diameter.
As a method of removing fine particles of 40 .mu.m or less in particle diameter, a method employing a hydraulic classifier, or a method utilizing a decantation may be adopted. As a seed crystal, either AlF.sub.3.3H.sub.2 O crystals to be produced from a super-saturated solution of aluminum fluoride without using a seed crystal, or AlF.sub.3.3H.sub.2 O crystals to be produced from a super-saturated solution of aluminum fluoride by using a seed crystal may be used.
The average particle diameter of AlF.sub.3.3H.sub.2 O crystals to be produced from a super-saturated solution of aluminum fluoride without using a seed crystal is generally in the range of 70 to 80 .mu.m. However, these particles may be treated so as to be useful as a seed crystal by selectively removing fine particles therefrom to obtain a crystal of AlF.sub.3.3H.sub.2 O containing not more than 5% of fine particles of 40 .mu.m or lees in particle diameter. There is no limitation as to the average particle diameter of the seed crystal useful in this invention. When a seed crystal of large size is employed, crystals of larger size may be naturally obtained. If it is desired to obtain a large crystal having a particle diameter of about 100 .mu.m in one crystallizing process, the use of a seed crystal having an average particle diameter of about 80 .mu.m is required.
According to this invention, a seed crystal of AlF.sub.3.3H.sub.2 O is added to a super-saturated solution of aluminum fluoride in such a weight ratio that when the amount of the seed crystal is converted to the total surface area of the seed crystal, it corresponds to the range of 40-100 m.sup.2 per 1 kg of AlF.sub.3.3H.sub.2 O to be precipitated during a time period between a moment of initiating the crystallization and a moment where an initial concentration of the super-saturated solution of aluminum fluoride is reduced to 1.6%. At the same time, the initial concentration of the super-saturated solution of aluminum fluoride is adjusted to 8 to 15%.
Namely, when the concentration of super-saturated solution of aluminum fluoride and the amount of the super-saturation solution is determined, the amount of AlF.sub.3.3H.sub.2 O to be crystallized during a time period between a moment of initiating the crystallization and a moment where an initial concentration of the super-saturated solution of aluminum fluoride is reduced to 1.6% is naturally determined. Accordingly, a seed crystal of AlF.sub.3.3H.sub.2 O is added to a super-saturated solution of aluminum fluoride in such a weight ratio that when the amount of the seed crystal is converted to the total surface area of the seed crystal, it falls in the range of 40-100 m.sup.2 per 1 kg of AlF.sub.3.3H.sub.2 O to be precipitated. In this case, the initial concentration of the super-saturated solution of aluminum fluoride is adjusted to range from 8% to 15%. The terms of the initial concentration of the super-saturated solution of aluminum fluoride is meant herein a concentration as measured after the seed crystal is added to the super-saturated solution of aluminum fluoride.
It is required in this invention to adjust the concentration of super-saturated solution of aluminum fluoride at the initiation of crystallization to range from 8% to 15%, and at the same time it is important to select the amount of seed crystal to be added to the super-saturated solution.
It is also required in order to effectively crystallize AlF.sub.3.3H.sub.2 O of such a large particle size as intended according to this invention to have both of the concentration of the super-saturated solution of aluminum fluoride and the amount of the seed crystal to be added in the super-saturated solution of aluminum fluoride satisfying the conditions mentioned above. In other words, it is impossible to effectively crystallize AlF.sub.3.3H.sub.2 O of such a large particle size as intended according to this invention if only one of the conditions is met. If the concentration of the super-saturated solution of aluminum fluoride is lower than 8%, a crystal having a large number of columnar or needle-like crystal attached on its seed crystal can be obtained. Additionally, under such a condition of low concentration, the crystals which have been precipitated may impinge upon one another, or impinge against the wall of a crystallizer causing a stripping of columnar or needle-like crystal, thereby inhibiting the growth of excellent crystals.
On the other hand, if the concentration of super-saturated solution of aluminum fluoride at the initiation of crystallization exceeds 15%, it will give rise to the generation of a large number of new nuclei due to the excessive concentration of the super-saturated solution. If a large number of new nuclei are generated, an efficient crystal growth of seed crystal is no longer possible. By limiting the initial concentration of the super-saturated solution of aluminum fluoride at the initiation of crystallization to range from 8% to 15%, it is possible to obtain AlF.sub.3.3H.sub.2 O crystals which are high in density and smoother in surface texture. It is more preferable to limit the initial concentration of the super-saturated solution of aluminum fluoride at the moment of the initiation of crystallization to range from 9% to 13%.
When the total surface area of the seed crystal to be added to the super-saturated solution is smaller than 40 m.sup.2 per 1 kg of AlF.sub.3.3H.sub.2 O to be precipitated, the quantity of the seed crystal is too little in relative to the quantity of the super-saturated solution thereby permitting a large number of additional new crystal nuclei to be generated. Further, in this case, the amount of newly precipitated crystal to be shared to the growth of fine particles 40 .mu.m or less will exceed 10%. If the amount of newly precipitated crystal to be shared to the growth of fine particles 40 .mu.m or less exceeds over 10%, it is no longer possible to efficiently obtain crystal of larger size. In addition to that, it also gives rise to a new problem of the disposal of a large number of fine particles to be generated during the precipitation step.
Furthermore, if the total surface area of the seed crystal to be added to the super-saturated solution exceeds over 100 m.sup.2 per 1 kg of AlF.sub.3.3H.sub.2 O to be precipitated, the generation of new crystal nuclei becomes constant so that any increase in surface area exceeding this upper limit is no longer effective in inhibiting the generation of new crystal nuclei. Also, the magnitude of crystal growth of seed crystal in one crystallization process in this case is minimized, and the concentration of solid materials within a crystallizer is so increased as to make the operation of crystallization process difficult to continue, thereby raising another problem.
The total surface area of a seed crystal is determined from a measured value of particle size distribution, or from a value calculated from an average particle diameter.
The seed crystal may be added to a super-saturated solution of aluminum fluoride either in the form of slurry, which can be produced by removing a sufficient amount of fine particles from a slurry obtained upon finishing the crystallization until the slurry no longer contains more than 5% of fine particles 40 .mu.m or less in particle diameter, and then concentrating the slurry to such an extent that the slurry contains 30% or more of AlF.sub.3.3H.sub.2 O, or in the form of filter cake, which can be obtained by filtering the last-mentioned slurry.
The crystallization of the super-saturated solution of aluminum fluoride is carried out by heating the solution to a temperature ranging from 70.degree. C. to 106.degree. C., which is a boiling point of aluminum fluoride solution. The crystallization may proceed even if the temperature of the solution is lower than 70.degree. C., but the rate of crystallization is low and inadvisable in view of productivity. The upper limit in temperature is the boiling point of aluminum fluoride solution as mentioned above. In one embodiment, the temperature for carrying out the crystallization of the solution is in the range of from 85.degree. C. to 95.degree. C. This heating can be conducted by the conventional method, for example by blowing a heated steam into the solution.
The growing of the seed crystal is performed while agitating the heated super-saturated solution of aluminum fluoride. This agitation is conducted for preventing AlF.sub.3.3H.sub.2 O from accumulating at the bottom of a crystallizer drum. When a flat-blade type agitator is to be employed, the agitation may be conducted by using an agitating blade having a length 0.6 to 0.9 times as long as the diameter of a crystallizer, at the blade-tip velocity of 0.4 to 3 m/sec., more preferably 0.7 to 2 m/sec. If the agitation is too vigorous, the crystals of AlF.sub.3.3H.sub.2 O may impinge upon each other so as to generate a large number of secondary nuclei. Therefore, it is advisable to avoid a vigorous agitation.
The method of agitation is not limited to the use of afore-mentioned flat-blade type agitator, but can be selected from various methods, i.e. such as those using other kind of agitation blade, or those utilizing a foam-blowing agitation. It is advisable in any of the agitation methods to perform the agitation in such a manner as to avoid the accumulation of AlF.sub.3.3H.sub.2 O at the bottom of a crystallizer. If AlF.sub.3.3H.sub.2 O is accumulated at the bottom of a crystallizer, the accumulated AlF.sub.3.3H.sub.2 O becomes a big block and adheres to the bottom of the crystallizer.
The time period required for the crystallization is dependent on the amount of seed crystal employed, the concentration of the solution, the temperature of crystallization, and so on, but generally ranges from 3 to 6 hours.
In subsequent to the crystallization process of the solution, particles of larger particle size exceeding 40 .mu.m in particle diameter are separated from particles of fine particle size 40 .mu.m or less in the resultant slurry. This separation can be performed by using the conventional wet type classifier. In another method, this separation may be conveniently performed by utilizing a decantation. The AlF.sub.3.3H.sub.2 O particles of larger particle size thus separated are then subjected to filtration, drying and dehydration.
The filtration can be conducted by using for example a centrifugal separator. Then, the drying is conducted at a temperature of 100.degree. to 150.degree. C. The dehydration is conducted for removing crystallization water from AlF.sub.3.3H.sub.2 O. As a method of dehydration, either a single-step method wherein the temperature of AlF.sub.3.3H.sub.2 O is raised in a single step up to 600.degree. C., or a two-step method wherein the temperature of AlF.sub.3.3H.sub.2 O is raised in the first step to about 300.degree. C. thereby removing most of crystallization water, and then the temperature is raised in the second step up to 600.degree. C. to remove remainder of crystallization water may be employed.
By finishing these steps, it is possible to ultimately obtain aluminum fluoride anhydride of larger crystal size. According to another method of this invention, the above-mentioned processes are performed under the same conditions as explained above as a first step.
Namely, the processes of preparing a super-saturated solution of aluminum fluoride; adding into the super-saturated solution of aluminum fluoride a seed crystal of AlF.sub.3.3H.sub.2 O containing not more than 5% of fine particle of 40 .mu.m or less in particle diameter in such a ratio that the total surface area of the seed crystal is in the range of 40-100 m.sup.2 per 1 kg of AlF.sub.3.3H.sub.2 O to be precipitated during a time period between a moment initiating a crystallization and a moment where an initial concentration of the super-saturated solution of aluminum fluoride is reduced to 1.6%; adjusting the initial concentration of the super-saturated solution of aluminum fluoride to 8 to 15%; heating the resultant super-saturated solution of aluminum fluoride to a temperature ranging from 75.degree. to 106.degree. C.; maintaining the super-saturated solution of aluminum fluoride at this temperature under agitation, thereby allowing crystal to be precipitated from the super-saturated solution of aluminum fluoride through a batch crystallization method are repeated.
In this case, the seed crystal of AlF.sub.3.3H.sub.2 O employed in this first step is hereinafter referred to as a primary seed crystal, and this first crystallization step is hereinafter referred to as a primary crystallization.
An AlF.sub.3.3H.sub.2 O crystal is separated from the slurry obtained from the primary crystallization, and processed to obtain a secondary seed crystal containing, as in the case of the primary seed crystal, not more than 5% of fine particle having a particle diameter of 40 .mu.m or less.
The secondary crystallization is performed using the secondary seed crystal under the same conditions as in the case of the primary crystallization. Further, a AlF.sub.3.3H.sub.2 O crystal is separated from the slurry obtained from the secondary crystallization, and processed to obtain a tertiary seed crystal containing, as in the case of the secondary seed crystal, not more than 5% of fine particle having a particle diameter of 40 .mu.m or less. Subsequently, the tertiary crystallization is performed using the tertiary seed crystal under the same conditions as those in the secondary crystallization. Likewise, a AlF.sub.3.3H.sub.2 O crystal is separated from the slurry obtained from the previous crystallization, and processed to obtain a new seed crystal. Subsequently, another crystallization is additionally performed using this crystal under the same conditions as those in the previous crystallization. By repeating these crystallizing processes, it is possible to gradually enlarge the particle size of AlF.sub.3.3H.sub.2 O to be obtained from the slurry of AlF.sub.3.3H.sub.2 O. It is preferable in this invention to repeat these crystallization processes at least once in subsequent to the primary crystallization. If these crystallization processes are repeated three times or more, preferably four times or more, it is possible to obtain an AlF.sub.3.3H.sub.2 O crystal having an average particle diameter of 100 .mu.m or more, which is large enough for a practical use. It is preferable to repeat these crystallization processes until not less than 60% of AlF.sub.3.3H.sub.2 O to be crystallized becomes 100 .mu.m or more in particle diameter.
After the final crystallization process is finished, an AlF.sub.3.3H.sub.2 O crystal having an average particle diameter of 100 .mu.m or more is separated in the same manner as explained above from the slurry of AlF.sub.3.3H.sub.2 O, and is then dried and dehydrated in the same manner as explained above thereby to obtain aluminum fluoride anhydride of large particle size.
Even in this method of obtaining an AlF.sub.3.3H.sub.2 O crystal of large particle size by repeating the crystallization processes, it is possible to utilize as a seed crystal a slurry of AlF.sub.3.3H.sub.2 O obtained as a result of crystallization, or to utilize a filter cake thereof. The temperature for carrying out the crystallization of the super-saturated solution of aluminum fluoride is preferably in the range of 85.degree. to 95.degree. C.
When a crystal of spherical shape is employed as a seed crystal, an AlF.sub.3.3H.sub.2 O crystal to be obtained after crystallization is also spherical in shape, and therefore excellent in fluidity in a transportation route and easy in handling. An example of the conditions for obtaining a spherical AlF.sub.3.3H.sub.2 O seed crystal is as follows.
The concentration of the super-saturated solution of aluminum fluoride should be 8% or more; the temperature of the solution during the crystallization should be 70.degree. C. or more, preferably 85.degree. C. to 100.degree. C.; and the crystallization is conducted for 4 to 6 hours thereby to obtain a slurry. The slurry is then subjected to a classification process thereby removing fine particles to obtain a spherical AlF.sub.3.3H.sub.2 O seed crystal.
The agitation in this case is performed at the blade-tip velocity of 0.4 to 3 m/sec., more preferably 0.7 to 2 m/sec., if a flat-blade type agitator is employed. Other conditions including the method of recovering AlF.sub.3.3H.sub.2 O from a slurry in subsequent to the crystallization are the same as mentioned hereinabove. The particle diameter of aluminum fluoride anhydride obtained after drying and dehydrating AlF.sub.3.3H.sub.2 O particles should preferably be as such that includes not more than 1% of fine particles which have been passed through a 44 .mu.m sieve. By limiting the content of the fine particle as mentioned above in the final product, it has become possible to substantially inhibit the generation of dust when the final product is transportated.
This invention will be further explained with reference to the following experiments. In this experiments, the super-saturated solution of aluminum fluoride was prepared by mixing together a super-saturated solution of aluminum fluoride having a concentration of 17% and a solution of aluminum fluoride having a concentration of 1.3%. The separation of fine particles from AlF.sub.3.3H.sub.2 O slurry after the crystallization, and the measurement of particle size distribution were conducted as follows.
(Preparation of the super-saturated solution of aluminum fluoride)
1500 g of fluosilicic acid solution containing 20% of H.sub.2 SiF.sub.6 was charged into a plastic wide-mouthed bottle having an inner volume of 2000 cm.sup.3, and then the plastic bottle was settled in a water bath maintained at the temperature of 95.degree. C. Then, the fluosilicic acid solution was subjected to agitation by using a flat-blade type agitator having a blade length of 8 cm and at the blade-tip velocity of 1.3 m/sec. when the temperature of the solution within the plastic bottle was raised up to 70.degree. C., 323 g of aluminum hydroxide containing 98% of Al(OH).sub.3 was added to the solution over a period of 5 minutes. 20 minutes after the initiation of the addition of aluminum hydroxide, a solid material was separated by filtering. The filtered solid material was then washed twice with a 250 g of hot water, thereby obtaining a filtrate and a wash liquid. The concentration of the resultant super-saturated solution of aluminum fluoride was 17%.
(Method of separating large particles and fine particles)
Particles having a particle diameter exceeding 40 .mu.m are herein defined as large particles (simply referred to as "large particles" in the following experiments and (Examples), and particles having a particle diameter of 40 .mu.m or less are herein defined as fine particles (simply referred to as "fine particles" in the following experiments and Examples).
The separation of fine particles was conducted by using a hydraulic classifier. The velocity of the upward flow at the overflowing portion of the hydraulic classifier was set to 0.93 mm/sec. The temperature of the liquid was 25.degree. C., the density of the liquid was 1.02 g/cm.sup.3, and the viscosity of the liquid was 0.01 g/cm.sec. Further, the density of AlF.sub.3.3H.sub.2 O in the separated liquid was 2.09 g/cm.sup.3. The density of solid material in the classifier was set to about 5%, and the time of the separation was set to 20 minutes.
(Measurement of particle size distribution and Calculation of average particle diameter)
The measurement of particle size distribution was conducted by using a granulometer, HR-850 (Cilas-, Alcatel Co. Ltd. ). The average particle diameter was indicated by the value of a particle diameter which corresponds to 50% in the particle size distribution. The measurement of particle size distribution of aluminum fluoride anhydride was conducted by sieve.
EXPERIMENT NO. 1
(Concentration of a super-saturated solution )
Various kinds of solutions each differing in the concentration of aluminum fluoride from one another were prepared by mixing together a super-saturated solution of aluminum fluoride having a concentration of 17% and a solution of aluminum fluoride having a concentration of 1.3%. With these super-saturated solutions of aluminum fluoride, the growth of seed crystal was conducted under the following conditions.
______________________________________Crystallizing conditions:______________________________________Crystallizer: Polyethylene wide-mouthed bottle (Volume: 1000 cm.sup.3, diameter: 9 cm)Agitator: Flat-blade type agitator (blade length: 8 cm)Blade-tip velocity: 1.3 m/sec.Temperature of solution: 93.degree. C.Crystallizing time: 5 hoursAlF.sub.3 super-saturated solution: 600 g (Concentration is indicated in Table 1)Particle diameter of seed crystal: 89 .mu.m in average diameterAmount of seed crystal: 75 m.sup.2 in surface area per 1 kg of AlF.sub.3.3H.sub.2 O to be precipitated. The surface area is calculated from the average particle diameter of seed crystal.______________________________________
The slurry of AlF.sub.3.3H.sub.2 O resulting from the crystallization was subjected to classifying treatment to separate large particles from fine particles, and each of the particles were filtered thereby recovering each of them. The large particles thus obtained were observed with a scanning type electron microscope to determine the feature of the particle. Results are shown in Table 1.
The concentration of the super-saturated solution of aluminum fluoride at the initiation of the crystallization was determined by measuring the volumes of a super-saturated solution of aluminum fluoride having a concentration of 17% and a solution of aluminum fluoride having a concentration of 1.3%, and by calculating water attached to the seed crystal. The amount of newly precipitated AlF.sub.3.3H.sub.2 O was determined by measuring the amount of AlF.sub.3.3H.sub.2 O to be precipitated during a time period between a moment of initiation of crystallization and a moment where the concentration of aluminum fluoride in the mother liquor after crystallization (a liquid which is left after the removal of AlF.sub.3.3H.sub.2 O) is reduced to 1.6%. The symbols noted on the surface feature of crystal and on the generation of fine particles in the column of judgment in Table 1 were based on the criteria indicated on the margin of Table 1. The symbols noted on the summary in the column of judgment were determined as follows. Namely, if both of the surface feature of crystal and the generation of fine particles were marked with the symbol of a double circles, the judgment of the summary was marked with a double circle, if either the surface feature of crystal or the generation of fine particles was marked with the symbol of X, the judgment of the summary was marked with X, and if one of the surface feature of crystal and the generation of fine particles was marked with the symbol of double circle, and the other is marked with the symbol of single circle, the judgment of the summary was marked with a single circle. The same marking system as explained above was adopted also in the following tables.
TABLE 1______________________________________Initial Judgment Concentration Genera- of Alf.sub.3 Super- tion of saturated Surface FineNo. Solution (%) Feature Particle Summary______________________________________1 4.3 .times. .times. .times.2 6.1 .times. .largecircle. .times.3 8.0 .largecircle. .circleincircle. .largecircle.4 8.9 .circleincircle. .circleincircle. .circleincircle.5 10.1 .circleincircle. .circleincircle. .circleincircle.6 11.4 .circleincircle. .circleincircle. .circleincircle.7 12.9 .circleincircle. .circleincircle. .circleincircle.8 14.9 .circleincircle. .largecircle. .largecircle.9 16.4 .circleincircle. .times. .times.______________________________________
Surface feature:
.circleincircle. . . . Growth of columnar or needle-like crystal was not admitted at all on the surface of seed crystal.
.largecircle. . . . Growth of fine columnar crystal was slightly admitted on the surface of seed crystal.
X . . . A lot of needle-like crystals were admitted on the surface of seed crystal.
Generation of fine particles:
.circleincircle.. . . Amount of classified fine particles of 40 .mu.m or less was 5% or less based on the total amount of newly precipitated crystal.
.largecircle. . . . Amount of classified fine particles of 40 .mu.m or less was more than 5% and not more than 10% based on the total amount of newly precipitated crystal.
X . . . Amount of classified fine particles of 40 .mu.m or less was more than 10% based on the total amount of newly precipitated crystal.
As apparent from Table 1, when the concentration of super-saturated solution of aluminum fluoride at the initial moment of crystallization was in the range of from 8.0 to 16.4%, the generation of needle-like crystal was hardly recognized on the surface of seed crystal, and at the same time only a little amount of columnar crystal was admitted on the surface of seed crystal. Further, if the concentration of super-saturated solution of aluminum fluoride at the initial moment of crystallization was in the range of from 8.0 to 14.9%, the generation of fine particles was very limited, and a lot of excellent crystals were obtained.
EXPERIMENT NO. 2
(Amount of seed crystal)
A super-saturated solution of aluminum fluoride having a concentration as indicated below was prepared by mixing together a super-saturated solution of aluminum fluoride having a concentration of 17% and a solution of aluminum fluoride having a concentration of 1.3%. With this super-saturated solution of aluminum fluoride, the growth of seed crystal containing not more than 5% of fine particle of 40 .mu.m or less in particle diameter was conducted by changing the amount of seed crystal as shown in the following Table 2. Other crystallizing conditions were selected as follows.
______________________________________Crystallizing conditions:______________________________________Crystallizer: the same as in Experiment No. 1Temperature of solution: 93.degree. C.Crystallizing time: 5 hoursInitial concentration ofAlF.sub.3 super-saturated solution: 11.4%AlF.sub.3 super-saturated solution: 417 gParticle diameter of seed crystal: 89 .mu.m in average diameter______________________________________
The slurry of AlF.sub.3.3H.sub.2 O resulting from the crystallization was subjected to classifying treatment to separate large particles from fine particles, and each of the particles were filtered thereby recovering each of them. Results are shown in Table 2.
TABLE 2__________________________________________________________________________ Average Share of Particle Crystalliza- Diameter tion Concentration Dia- For of Solid In Meter For Genera- Judge- SlurryAmount of Seed Crys- of Growth tion of ment Initial FinalCrystal Crystalli- tal Large of Fine Genera- con- Con- Surface zation Mate- Par- Seed Parti- tion of cent- cent- Area* Weight Ratio rial ticle Crystal cle Fine ration rationNo. (m.sup.2 /kg) (g) (%) (.mu.m) (.mu.m) (%) (%) Particle (%) (%)__________________________________________________________________________10 25 53.4 87.6 10 114 84.1 15.9 .times. 11.4 26.011 40 85.5 88.7 105 107 92.1 7.9 .smallcircle. 17.0 30.712 50 106.9 88.9 103 105 95.3 4.7 .circleincircle. 20.4 33.613 75 160.3 89.4 10 100 96.9 3.1 .circleincircle. 27.8 39.714 100 213.8 90.2 98 98 97.6 2.4 .circleincircle. 33.9 44.815 125 267.2 90.3 96 96 97.7 2.3 .circleincircle. 39.1 49.1__________________________________________________________________________ *Surface area: An amount of seed crystal par 1 kg of newly precipitated AlF.sub.3.3H.sub.2 O which is represented by the total surface area thereof. **Crystallization ratio: The crystallization ratio is calculated from the chemical analysis of Al in the supersaturated solution of AlF.sub.3 at th moment of starting the crystallization, and in the solution of AlF.sub.3 at the moment of finishing the crystallization. ***Generation of fine particles: The same a in Experiment 1.
As apparent from Table 2, when the seed crystal was employed at a ratio of 25 m.sup.2 per 1 kg of AlF.sub.3.3H.sub.2 O, the amount of newly precipitated crystal to be shared to the growth of the seed crystal was relatively small, whereas the generation of new fine particles was increased.
By contrast, if the seed crystal was employed at a ratio exceeding 100 m.sup.2 per 1 kg of AlF.sub.3.3H.sub.2 O, the amount of newly precipitated crystal to be shared to the generation of fine particles was decreased to become constant. However, in this case, the degree of crystal growth of seed crystal per crystallization process was very limited, and additionally the concentration of solid material in the slurry was increased, thereby making the operation rather difficult. Accordingly, in order to efficiently manufacture a crystal of large particle size, it is preferable to employ a seed crystal of AlF.sub.3.3H.sub.2 O in such a ratio that the total surface area of the seed crystal is in the range of 40-100 m.sup.2 per 1 kg of AlF.sub.3.3H.sub.2 O to be precipitated.
EXPERIMENT NO. 3
(Effect of fine particles in the seed crystal)
Seed crystal of large particle size, which was obtained by removing fine particles 40 .mu.m or less in average particle size by means of a hydraulic classification from the slurry of AlF.sub.3.3H.sub.2 O obtained by performing the same crystallizing process as that used for No. 6 of Experiment No. 1, and fine particles having a particle size of 40 .mu.m or less which was separated from a slurry obtained by performing the same crystallizing process as mentioned above were mixed at the ratios indicated in Table 3, thereby preparing samples of seed crystal differing from each other in content of fine particle of 40 .mu.m or less in diameter. These seed crystals were subjected to crystal growth, and the relationship between the amount of the fine particles in the seed crystal and the generation of fine particles resulted from the precipitation was examined. In this experiment, all of the crystallizer, the agitator, temperature and time were the same as those in Experiment No. 1. Results were shown in Table 3.
TABLE 3__________________________________________________________________________ AlF.sub.3 SuperSeed Crystal Saturated Share of Weight (Dry) Solution Crystallization 40 .mu.m Large Fine Initial For Growth For or Surface Par- Par- Concen- of Seed Generation Less Area ticle ticle Total traion Weight Crystal of FineNo. (%) (m.sup.2 /kg) (g) (g) (g) (%) (g) (%) Particle (%)__________________________________________________________________________16 3.6 75 150.0 0.0 150 11.0 365 97.8 2.217 5.0 76 147.8 2.2 150 11.0 365 92.2 7.818 7.0 77 145.0 5.0 150 11.0 365 64.6 15.419 10.0 79 140.0 10.0 150 11.0 365 62.9 37.1__________________________________________________________________________ NOTE: In No. 16, only large particles obtained by hydraullc classificatio were employed, but it was found that it contained 3.6% of fine particle having a diameter of 40 .mu.m or less as a result of measurement of the particle size distribution.
As apparent from Table 3, when the content of seed crystal having a diameter of 40 .mu.m or less was more than 5%, the ratio of newly precipitated crystal to be shared to the growth of the fine seed crystal was increased. However, if the content of seed crystal having a diameter of 40 .mu.m or less was controlled to not more than 5%, the growth of fine seed crystal could be inhibited thereby making it possible to obtain crystals of larger particle size.
EXPERIMENT NO. 4
(Relationship between crystallization temperature and crystallization velocity)
A solution comprising 295 g of a super-saturated solution of aluminum fluoride having a concentration of 17%, 145 g of a solution of aluminum fluoride having a concentration of 1.3%, and 163 g of seed crystal having an average particle diameter of 77 .mu.m, and containing not more than 5% of fine particle 40 .mu.m or less in particle diameter and 8.1% of moisture were charged into a crystallizer of the same type as used in Experiment No. 1.
Then, this solution was maintained at the temperatures as indicated in Table 4, and allowed to crystallize over a period of 5 hours while agitating it with a flat-blade type agitator at the blade-tip velocity of 1.3 m/sec. The crystallization ratio was determined by withdrawing a portion of slurry at every one hour, separating solid material from the slurry, and analyzing aluminum concentration. Results are shown in Table 4.
TABLE 4______________________________________ Crystallization Time (h)Crystallization Ratio ofTemperature Crystallization (%)No. (.degree.C.) 2 3 4 5______________________________________20 55 -- 58.5 63.7 80.521 65 74.9 84.2 87.4 88.922 75 78.4 86.7 88.9 90.523 85 85.7 88.1 89.7 90.524 95 86.6 88.0 89.8 89.8______________________________________
As apparent from Table 4, when the crystallization time was over a period of 4 hours, the crystallization ratio become substantially constant at a crystallization temperature of 75.degree. C. or more.
EXAMPLE 1
Seed crystal was prepared as follows.
900 g of a super-saturated solution of aluminum fluoride having a concentration of 17% was charged into a polyethylene wide-mouthed bottle 12 cm in diameter and 2000 cm.sup.3 in volume. Then this bottle was immersed in a water bath kept at a temperature of 95.degree. C. The solution was allowed to crystallize over a period of 6 hours while agitating it with a flat-blade type agitator at the blade-tip velocity of 1.3 m/sec. Resultant slurry was then cooled approximately to the room temperature, and classified by using a hydraulic classifier to remove fine particles having a diameter of 40 .mu.m or less. After allowing the solution to stand still, a portion of the supernatant liquid was removed to obtain a slurry containing AlF.sub.3.3H.sub.2 O at a concentration of 50%. The average particle diameter of the AlF.sub.3.3H.sub.2 O crystal in the slurry was 83 .mu.m.
By using the AlF.sub.3.3H.sub.2 O crystal as a seed crystal, the crystal growth of the seed crystal was conducted as follows. 200 g of a slurry containing the above seed crystal was charged into a polyethylene wide-mouthed bottle 12 cm in diameter and 1000 cm.sup.3 in volume. Then this bottle was immersed in a water bath kept at a temperature of 95.degree. C.
The slurry was agitated with a flat-blade type agitator having a blade length of 8cm at the blade-tip velocity of 1.3 m/sec. Then, 335 g of a super-saturated solution of aluminum fluoride having a concentration of 17% and 85 g of a solution of aluminum fluoride having a concentration of 1.3% were charged into the slurry thereby obtaining a super-saturated solution having concentration of 11.4%.
Then, this solution was subjected to crystallization over a period of 4 hours. In this case, the total surface area of the seed crystal employed was 40 m.sup.2 per 1 kg of newly precipitated AlF.sub.3.3H.sub.2 O.
The removal of fine particles having a diameter of 40 .mu.m or less was conducted in the same manner as explained above. The remainder of slurry containing large particles was filtered to obtain 202 g of filter cake. This filter cake of large particle size was found to have 7.8% of moisture, and an average particle diameter of 101 .mu.m.
A slurry containing fine particles 40 .mu.m or less in diameter was filtered to obtain a filter cake, which was then dried at a temperature of 105.degree. C., for 4 hours to obtain 4.2 g of fine particles. The amount of the resultant fine particles Was 4.9% based on the total amount of the newly precipitated crystal. The large particles thus obtained were dried at a temperature of 105.degree. C., and dehydrated in two steps, i.e. at 300.degree. C., and at 600.degree. C. to obtain aluminum fluoride anhydride. The resultant aluminum fluoride anhydride was found to contain 1% or less of particles that could be passed through a 44 .mu.m mesh.
EXAMPLE 2
The same processes as those of Example 1 were repeated except that the slurry was agitated at the blade-tip velocity of 0.5 m/sec at the crystallization step, thereby obtaining a seed crystal having an average particle diameter of 89 .mu.m. The seed crystal thus obtained was allowed to grow in the same manner as in Example 1. However, in this case, the total surface area of the seed crystal employed was 60 m.sup.2 per 1 kg of newly precipitated AlF.sub.3.3H.sub.2 O, and 300 g of a slurry containing 50% of water was employed. The slurry contained 312 g of a super-saturated solution of aluminum fluoride having a concentration of 17% and 23 g of a solution of aluminum fluoride having a concentration of 1.3%.
The resultant slurry was subjected to hydraulic classification in the same manner as in Example 1, thereby obtaining 247 g of filter cake of large particles having a diameter of more than 40 .mu.m, which was found to have 7.7% of moisture, and an average particle diameter of 102 .mu.m. When fine particles having a diameter of 40 .mu.m or less was processed in the same manner as in Example 1, 3.4 g (dry weight) of fine particles was obtained. The amount of the fine particles was 4.2% based on the total amount of the newly precipitated crystal.
The large particles thus obtained were dried and dehydrated in the same manner as in Example 1, thereby obtaining aluminum fluoride anhydride. The resultant aluminum fluoride anhydride was found to contain 1% or less of particles that could be passed through a 44 .mu.m mesh.
EXAMPLE 3
The same processes as those of Example 2 were repeated to obtain a seed crystal. The seed crystal thus obtained was allowed to grow in the same manner as in Example 1. However, in this case, the total surface area of the seed crystal employed was 40 m.sup.2 per 1 kg of newly precipitated AlF.sub.3.3H.sub.2 O, and 300 g of a slurry containing 50% of water was employed. The super-saturated solution of aluminum fluoride containing 475 g of a super-saturated solution of aluminum fluoride having a concentration of 17% and 110 g of a solution of aluminum fluoride having a concentration of 1.3% was used in this example.
The resultant slurry was subjected to hydraulic classification in the same manner as in Example 1, thereby obtaining 284 g of filter cake of large particles having a diameter of more than 40 .mu.m, which was found to have 7.8% of moisture, and an average particle diameter of 106 .mu.m. When fine particles having a diameter of 40 .mu.m or less were processed in the same manner as in Example 1, 11 g (dry weight) of fine particles was obtained. The amount of the fine particles was 8.9% based on the total amount of the newly precipitated crystal.
The large particles thus obtained were dried and dehydrated in the same manner as in Example 1, thereby obtaining aluminum fluoride anhydride. The resultant aluminum fluoride anhydride was found to contain 1% or less of particles that could be passed through a 44 .mu.m mesh.
EXAMPLE 4
As a primary seed crystal, AlF.sub.3.3H.sub.2 O particles containing not more than 5% of fine particles having an average diameter of 40 .mu.m or less and 8.1% of moisture was employed. The total surface area of the seed crystal per 1 kg of newly precipitated AlF.sub.3.3H.sub.2 O, the weight (dry basis) of seed crystal employed, the concentration of the slurry at the initial moment of crystallization, and the weight of the super-saturated solution of aluminum fluoride were as shown in Table 5. The slurry was agitated at the blade-tip velocity of 0.7 m/sec by using a flat-blade type agitator. The crystallization was conducted under the same conditions as in Example 1.
Large particles exceeding 40 .mu.m in diameter were separated in the same manner as in Example 1 from the slurry of AlF.sub.3.3H.sub.2 O. The average particle sizes of the large particles were as shown at No.4 in Table 5. The sharing ratio of crystallization, and the judgments to the crystal on the basis of Experiments are also shown at No.4 in Table 5.
The large particles obtained as shown at No.4 in Table 5 were used as a seed crystal in another crystallization process, which was conducted under the same conditions as mentioned above. In the same manner, large particles obtained In the previous crystallization process were repeatedly employed as a seed crystal in the next crystallization process, thereby obtaining larger particles. Results of them are shown at Nos. 5 to 7 in Table 5.
TABLE 5__________________________________________________________________________Seed Crystal Average Share ofAver- AlF.sub.3 Super Particle Crystallizationage Saturated Ratio Diameter Genera- Judgement Par- Sur- Solution of Diameter Growth tion of Genera- ticle face Initial Crys- of of Fine tion of Dia- Area Weight Concen- talli- Large Seed Par- Fine meter (m.sup.2 / (Dry) tration Weight zation Particle Crystal ticle Surface Par-No. (.mu.m) kg) (g) (%) (g) (%) (.mu.m) (%) (%) Feature ticle Summary__________________________________________________________________________4 78 40 120 11.1 700 88.6 96 95.4 4.6 .circleincircle. .circleincircle. .circleincircle.5 96 40 120 11.0 570 88.2 115 98.1 1.9 .circleincircle. .circleincircle. .circleincircle.6 115 40 150 10.9 590 87.9 133 98.0 2.0 .circleincircle. .circleincircle. .circleincircle.7 133 40 150 11.3 510 87.4 151 97.7 2.3 .circleincircle. .circleincircle. .circleincircle.__________________________________________________________________________
As shown in Table 5, when the amount of the seed crystal corresponds to 40 m.sup.2 in total surface area of the seed crystal per 1 kg of newly precipitated AlF.sub.3.3H.sub.2 O, it was possible by repeating the crystallization process to obtain large particles having an average particle size of as large as 151 .mu.m. In this case, it was also possible to control the sharing ratio of precipitated material so as to limit the generation of line particles to less than 10%, to allow the most of the precipitated material to be consumed for the growth of large particles. Further, in this case, the degree of the crystal growth to be attained by one crystallization process was found to be as high as about 20 .mu.m, and the surface feature of the crystal thus obtained was also excellent.
The large particles thus obtained were dried and dehydrated in the same manner as in Example 1, thereby obtaining aluminum fluoride anhydride. The resultant aluminum fluoride anhydride was found to contain 1% less of particles that could be passed through a 44 .mu.m mesh.
EXAMPLE 5
A crystallization process was repeated four times by selecting the same conditions as employed in Example 4 with respect to the crystallizer, the shape of the agitator, the blade-tip velocity and the crystallization temperature, with other conditions being selected as indicated in Table 6. Results are shown in Table 6.
TABLE 6__________________________________________________________________________ AlF.sub.3 Super Average Share ofSeed Crystal Saturated Particle CrystallizationAverage Solution Ratio of Diameter Growth Generation Judgement Particle Surface Weight Initial Crystalli- Diameter of Seed of Fine Generation Diameter Area (Dry) Concentration Weight zation of Large Crystal Particle Surface of FineNo. (.mu.m) (m.sup.2 /kg) (g) (%) (g) (%) Particle (.mu.m) (%) (%) Feature Particle Summary__________________________________________________________________________ 8 78 75 150 11.3 465 88.0 90 97.7 2.3 .circleincircle. .circleincircle. .circleincircle. 1 9 90 75 300 11.1 800 87.8 101 96.6 3.4 .circleincircle. .circleincircle. .circleincircle.10 101 75 300 11.1 720 88.5 111 96.9 3.1 .circleincircle. .circleincircle. .circleincircle.11 111 75 300 11.4 650 89.0 121 98.2 1.8 .circleincircle. .circleincircle. .circleincircle.__________________________________________________________________________
As apparent from Table 6, the results were almost the same as those of Example 4 (Table 5). However, the generation of fine particles resulting from the sharing of precipitation was found to be small.
The large particles thus obtained were dried dehydrated in the same manner as in Example 1 to obtain aluminum fluoride anhydride. The resultant aluminum fluoride anhydride was found to contain 1% or less of particles that could be passed through a 44 .mu.m sieve.
EXAMPLE 6
A crystallization process was repeated four times by setting the blade-tip velocity of a flat-blade type agitator to 2.0 m/sec. and selecting the same conditions as employed in Example 4 with respect to the crystallization drum and the crystallization temperature, with other conditions being selected as indicated in Table 7. Results are shown in Table 7.
TABLE 7__________________________________________________________________________ AlF.sub.3 Super Average Share ofSeed Crystal Saturated Particle CrystallizationAverage Solution Ratio of Diameter Growth Generation Judgement Particle Surface Weight Initial Crystalli- Diameter of Seed of Fine Generation Diameter Area (Dry) Concentration Weight zation of Large Crystal Particle Surface of FineNo. (.mu.m) (m.sup.2 /kg) (g) (%) (g) (%) Particle (.mu.m) (%) (%) Feature Particle Summary__________________________________________________________________________12 78 75 150 11.2 465 88.1 88 93.7 6.3 .circleincircle. .smallcircle. .smallcircle.13 88 75 300 11.1 855 88.5 99 95.2 4.8 .circleincircle. .circleincircle. .circleincircle. 614 99 75 300 11.0 735 87.4 110 94.4 5.6 .circleincircle. .smallcircle. .smallcircle.15 110 75 300 11.3 660 88.3 120 93.9 6.1 .circleincircle. .smallcircle. .smallcircle.__________________________________________________________________________
As apparent from Table 7, it was possible to obtain a crystal having an average particle diameter of more than 100 .mu.m. At the same time, it was possible to minimize the amount of the generation of fine particles resulting from the sharing of precipitation.
The large particles thus obtained were dried and dehydrated in the same manner as in Example 1 to obtain aluminum fluoride anhydride. The resultant aluminum fluoride anhydride was found to contain 1% or less of particles that could be passed through a 44 .mu.m mesh.
EXAMPLE 7
A crystallization process was repeated four times by setting the blade-tip velocity of a flat-blade type agitator to 1.3 m/sec. and selecting the same conditions as employed in Example 4 with respect to the crystallizer and the crystallization temperature, with other conditions being selected as indicated in Table 8. In this case, the surface area of the seed crystal and the concentration of the super-saturated solution of aluminum fluoride at the moment of initiating the crystallization were changed as indicated in Table 8. Results are shown in Table 8.
TABLE 8__________________________________________________________________________ AlF.sub.3 Super SaturatedSeed Crystal Solution Share of Aver- Ini- Ratio Average Crystallization age tial of Particle Genera- Judgement Par- Sur- Con- Crys- Diameter Growth tion of Genera- ticle face cen- tal- Large Fine of Fine tion of Dia- Area Weight tra- liza- Par- Par- Seed Par- Fine meter (m.sup.2 / (Dry) tion Weight tion ticle ticle Crystal ticle Surface Par-No. (.mu.m) kg) (g) (%) (g) (%) (.mu.m) (.mu.m) (%) (%) Feature ticle Summary__________________________________________________________________________16 78 75 300 13.4 765 90.4 89 14 96.3 3.7 .circleincircle. .circleincircle. .circleincircle.17 89 75 300 13.3 670 90.4 99 12 95.5 4.5 .circleincircle. .circleincircle. .circleincircle.18 99 75 300 13.1 600 89.8 110 12 95.1 4.9 .circleincircle. .circleincircle. .circleincircle.19 110 75 300 13.1 540 90.1 120 11 95.2 4.8 .circleincircle. .circleincircle. .circleincircle.__________________________________________________________________________
EXAMPLE 8
A crystallization process was repeated four times under the same conditions as employed in Example 7 except that the concentration of the super-saturated solution of aluminum fluoride at the moment of initiating the crystallization were changed as indicated in Table 9. Results are shown in Table 9.
TABLE 9__________________________________________________________________________ AlF.sub.3 Super SaturatedSeed Crystal Solution Share of Aver- Ini- Ratio Average Crystallization age tial of Particle Genera- Judgement Par- Sur- Con- Crys- Diameter Growth tion of Genera- ticle face cen- tal- Large Fine of Fine tion of Dia- Area Weight tra- liza- Par- Par- Seed Par- Fine meter (m.sup.2 / (Dry) tion Weight tion ticle ticle Crystal ticle Surface Par-No. (.mu.m) kg) (g) (%) (g) (%) (.mu.m) (.mu.m) (%) (%) Feature ticle Summary__________________________________________________________________________20 78 75 150 9.0 590 84.3 89 13 98.0 2.0 .circleincircle. .circleincircle. .circleincircle.21 89 75 150 9.0 520 83.5 100 13 97.3 2.7 .circleincircle. .circleincircle. .circleincircle.22 100 75 180 9.1 550 83.7 111 12 97.9 2.1 .circleincircle. .circleincircle. .circleincircle.23 111 75 180 9.0 500 83.7 121 12 97.8 2.2 .circleincircle. .circleincircle. .circleincircle.__________________________________________________________________________
The large particles thus obtained were dried and dehydrated in the same manner as in Example 1 to obtain aluminum fluoride anhydride. The resultant aluminum fluoride anhydride was found to contain 1% or less of particles that could be passed through a 44 .mu.m mesh.
EXAMPLE 9
A crystallization process was repeated four times under the same conditions as employed in Example 7 except that, (1) as a primary seed crystal, AlF.sub.3.3H.sub.2 O particles having an average particle diameter of 78 .mu.m, and containing not more than 5% of fine particles having an average diameter of 40 .mu.m or less and 8.1% of moisture was employed in an amount corresponding to 75 m.sup.2 in total surface area of the seed crystal per 1 kg of newly precipitated AlF.sub.3.3H.sub.2 O, and (2) the temperature of solution was maintained at 83.degree. C. As a result, AlF.sub.3.3H.sub.2 O crystals having an average particle diameter of 123 .mu.m were obtained. Results are shown in Table 10.
TABLE 10__________________________________________________________________________ AlF.sub.3 Super SaturatedSeed Crystal Solution Share of Aver- Ini- Ratio Average Crystallization age tial of Particle Genera- Judgement Par- Sur- Con- Crys- Diameter Growth tion of Genera- ticle face cen- tal- Large Fine of Fine tion of Dia- Area Weight tra- liza- Par- Par- Seed Par- Fine meter (m.sup.2 / (Dry) tion Weight tion ticle ticle Crystal ticle Surface Par-No. (.mu.m) kg) (g) (%) (g) (%) (.mu.m) (.mu.m) (%) (%) Feature ticle Summary__________________________________________________________________________24 76 75 150 11.1 460 89.1 88 14 98.6 1.4 .circleincircle. .circleincircle. .circleincircle.25 88 75 180 11.8 500 90.3 101 15 98.3 1.7 .circleincircle. .circleincircle. .circleincircle.26 101 75 180 11.0 430 89.9 113 14 97.3 2.7 .circleincircle. .circleincircle. .circleincircle.27 113 75 300 11.6 600 89.0 123 12 98.1 1.9 .circleincircle. .circleincircle. .circleincircle.__________________________________________________________________________
The large particles thus obtained were dehydrated in the same manner as in Example 1 to obtain aluminum fluoride anhydride. The resultant aluminum fluoride anhydride was found to contain 1% or less of particles that could be passed through a 44 .mu.m mesh.
EXAMPLE 10
A crystallization process was repeated four times under the same conditions except that, as a primary seed crystal, AlF.sub.3.3H.sub.2 O particles having an average particle diameter of 78 .mu.m, and containing act more than 5% of fine particles having an average diameter of 40 .mu.m or less and 8.1% of moisture were employed in an amount corresponding to 50 m.sup.2 in total surface area of the seed crystal per 1 kg of newly precipitated AlF.sub.3.3H.sub.2 O. The crystallization was conducted for 4 hours under the conditions: 12.1% in concentration of AlF.sub.3 at the moment of initiating the crystallization; 93.degree. C. in crystallization temperature; and 1.3 m/sec. In blade-tip velocity. Further Nos. 29-31 in Table 11 were conducted. Results are shown in Table 11. As apparent from Table 11, a crystal growth of about 14 to 16 .mu.m in thickness was recognized in one step of crystallization.
TABLE 11__________________________________________________________________________ AlF.sub.3 Super Average Share ofSeed Crystal Saturated Particle CrystallizationAverage Solution Ratio of Diameter Growth Generation Judgement Particle Surface Weight Initial Crystalli- Diameter of Seed of Fine Generation Diameter Area (Dry) Concentration Weight zation of Large Crystal Particle Surface of FineNo. (.mu.m) (m.sup.2 /kg) (g) (%) (g) (%) Particle (.mu.m) (%) (%) Feature Particle Summary__________________________________________________________________________28 78 50 120 12.1 500 89.1 93 95.1 4.9 .circleincircle. .circleincircle. .circleincircle. .29 93 50 150 11.8 535 89.3 107 94.3 5.7 .circleincircle. .largecircle. .largecircle.30 107 50 150 11.8 470 88.9 123 93.9 6.1 .circleincircle. .largecircle. .largecircle.31 123 50 150 12.1 435 88.7 138 95.7 4.3 .circleincircle. .circleincircle. .circleincircle.__________________________________________________________________________
EXAMPLE 11
The amount of seed crystal employed was either 50 m.sup.2 or 60 m.sup.2 in total surface area of the seed crystal per 1 kg of newly crystallized AlF.sub.3.3H.sub.2 O. The average particle diameter and weight of the seed crystal were those as indicated in Table 12. By using the seed crystal, the crystallization process was repeated four times in the same manner as employed in Example 10. As a result, AlF.sub.3.3H.sub.2 O crystals having an average particle diameter of 131 .mu.m were obtained, and a crystal growth of about 13 .mu.m in thickness was recognized in one step of crystallization. Results are shown in Table 12.
TABLE 12__________________________________________________________________________ AlF.sub.3 Super Average Share ofSeed Crystal Saturated Particle CrystallizationAverage Solution Ratio of Diameter Growth Generation Judgement Particle Surface Weight Initial Crystalli- Diameter of Seed of Fine Generation Diameter Area (Dry) Concentration Weight zation of Large Crystal Particle Surface of FineNo. (.mu.m) (m.sup.2 /kg) (g) (%) (g) (%) Particle (.mu.m) (%) (%) Feature Particle Summary__________________________________________________________________________32 78 60 120 11.4 445 88.8 91 94.7 5.3 .circleincircle. .largecircle. .largecircle.33 91 60 150 11.8 495 89.4 104 96.9 3.1 .circleincircle. .circleincircle. .circleincircle.34 104 60 150 11.3 420 88.6 118 97.1 2.9 .circleincircle. .circleincircle. .circleincircle.35 118 50 150 11.2 445 89.2 131 95.2 4.8 .circleincircle. .circleincircle. .circleincircle.__________________________________________________________________________
EXAMPLE 12
(1) Preparation of seed crystal
Crystallization of seed crystal was carried out for 6 hours in the same manner as in Example 1 except that 1200 g of a super-saturated solution of aluminum fluoride having a concentration of 17% was charged into the bottle, and the agitation with the flat-blade type agitator having 8.0 cm blade was conducted at the blade-tip velocity of 1.0 m/sec.
Resultant slurry after crystallization was filtered to obtain 323 g of a filter cake containing 7.7% of moisture. This filter cake was then converted into a slurry having 50% solid, which is then treated with a hydraulic classifier to remove fine particles, and thereafter filtered again. As a result, 317 g of large particles containing 9.6% of moisture and having an average particle diameter of 83 .mu.m was obtained. Details are shown at No. 36 in Table 13.
(2) A first crystallization
The crystallizing process as mentioned above was repeated by using as a seed crystal the large particles obtained above. As a crystallizing liquid, a solution denoted at No. 37 in Table 13 and comprising a mixture of a super-saturated solution of aluminum fluoride and a solution of aluminum fluoride having a concentration of 1.3% was employed. Other crystallization conditions were as shown below.
______________________________________Crystallizing device: The same as in Experiment No. 4Temperature of slurry: 93.degree. C.Crystallizing time: 4 hoursBlade-tip velocity 1.3 m/sec.______________________________________
After finishing the crystallization, the slurry was subjected to classification and filtering thereby to recover large particles and fine particles respectively. The ratio of sharing of newly precipitated crystal between that shared to the growth of seed crystal and that shared to the generation of fine particles was also determined. Additionally, judgments on the surface feature of the crystal were also conducted,
(3) Second to sixth crystallizations
The processes starting from the crystal growth to the classification of fine particles were repeated by using as a seed crystal large particles obtained in the above first crystallization under the same crystallizing conditions as explained above. The crystallization ratio, average particle diameter and sharing of precipitated material in each crystal growth were calculated as shown in Table 13.
TABLE 13__________________________________________________________________________ AlF.sub.3 Super SaturatedSeed Crystal Solution Share of Aver- Ini- Ratio Average Crystallization age tial of Particle Genera- Judgement Par- Sur- Con- Crys- Diameter Growth tion of Genera- ticle face cen- tal- Large Fine of Fine tion of Dia- Area Weight tra- liza- Par- Par- Seed Par- Fine meter (m.sup.2 / (Dry) tion Weight tion ticle ticle Crystal ticle Surface Par-No. (.mu.m) kg) (g) (%) (g) (%) (.mu.m) (.mu.m) (%) (%) Feature ticle Summary__________________________________________________________________________36 -- 0 0 17.0 1200 86.2 83 26 96.3 3.7 -- -- --37 83 60 150 10.6 522 88.1 92 16 96.9 3.1 .circleincircle. .circleincircle. .circleincircle.38 92 70 150 11.4 417 89.1 103 14 96.5 3.5 .circleincircle. .circleincircle. .circleincircle.39 103 70 150 11.4 368 89.1 111 16 95.5 4.5 .circleincircle. .circleincircle. .circleincircle.40 111 80 170 11.0 354 89.4 120 12 97.0 3.0 .circleincircle. .circleincircle. .circleincircle.41 120 70 180 10.4 407 68.2 128 13 95.2 4.8 .circleincircle. .circleincircle. .circleincircle.42 128 70 180 10.2 392 B7.9 136 13 96.3 3.7 .circleincircle. .circleincircle. .circleincircle.__________________________________________________________________________
As apparent from Table 13, it is possible according to this invention to steadily enlarge the particle size of a crystal by repeating the crystallization process without inviting excessive generation of fine particles in the process. At the same time, a crystal having a smooth surface and an excellent feature can be obtained.
A scanning type electron microscopic photograph of aluminum fluoride particles having an average particle diameter of 128 .mu.m obtained as indicated at No. 41 in Table 13 is shown in FIG. 1: and an optical microscopic photograph thereof is shown in FIG. 2. Further, a scanning type electron microscopic photograph of aluminum fluoride particles having an average particle diameter of 136 .mu.m obtained as indicated at No. 42 in Table 13 is shown in FIG. 3; and an optical microscopic photograph thereof is shown in FIG. 4.
EXAMPLE 13
A crystallization process was repeated four times under the same conditions as employed in Example 4 except that, the agitation was conducted at the blade-tip velocity of 1.3 m/sec. under the conditions indicated in Table 14. As a result, AlF.sub.3.3H.sub.2 O crystals having an average particle diameter of 153 .mu.m were obtained. Results are shown in Table 14.
TABLE 14__________________________________________________________________________ AlF.sub.3 Super SaturatedSeed Crystal Solution Share of Aver- Ini- Ratio Average Crystallization age tial of Particle Genera- Judgement Par- Sur- Con- Crys- Diameter Growth tion of Genera- ticle face cen- tal- Large Fine of Fine tion of Dia- Area Weight tra- liza- Par- Par- Seed Par- Fine meter (m.sup.2 / (Dry) tion Weight tion ticle ticle Crystal ticle Surface Par-No. (.mu.m) kg) (g) (%) (g) (%) (.mu.m) (.mu.m) (%) (%) Feature ticle Summary__________________________________________________________________________43 78 40 120 11.8 640 88.8 95 23 91.9 8.1 .circleincircle. .smallcircle. .smallcircle.44 95 42 150 11.8 630 89.1 113 21 92.9 7.1 .circleincircle. .smallcircle. .smallcircle.45 113 40 150 11.2 590 88.0 132 21 91.6 8.4 .circleincircle. .smallcircle. .smallcircle.46 132 40 180 10.5 650 87.0 153 24 90.3 9.7 .circleincircle. .smallcircle. .smallcircle.__________________________________________________________________________
EXAMPLE 14
The amount of seed crystal employed was 100 m.sup.2 in total surface area of the seed crystal per 1 kg of newly precipitated AlF.sub.3.3H.sub.2 O. By using the seed crystal, the crystallization process was repeated six times in the same manner as employed in Example 13 under the conditions indicated in Table 15. As a result. AlF.sub.3.3H.sub.2 O crystals having an average particle diameter of 128 .mu.m were obtained. Results are shown in Table 15.
TABLE 15__________________________________________________________________________ AlF.sub.3 Super SaturatedSeed Crystal Solution Share of Aver- Ini- Ratio Average Crystallization age tial of Particle Genera- Judgement Par- Sur- Con- Crys- Diameter Growth tion of Genera- ticle face cen- tal- Large Fine of Fine tion of Dia- Area Weight tra- liza- Par- Par- Seed Par- Fine meter (m.sup.2 / (Dry) tion Weight tion ticle ticle Crystal ticle Surface Par-No. (.mu.m) kg) (g) (%) (g) (%) (.mu.m) (.mu.m) (%) (%) Feature ticle Summary__________________________________________________________________________47 78 100 200 10.4 450 86.9 86 9 98.1 1.9 .circleincircle. .circleincircle. .circleincircle.48 86 100 200 11.3 410 88.4 95 8 97.9 2.1 .circleincircle. .circleincircle. .circleincircle.49 95 100 400 11.3 740 88.7 104 8 97.9 2.1 .circleincircle. .circleincircle. .circleincircle.50 104 100 400 11.1 690 88.3 111 8 97.5 2.5 .circleincircle. .circleincircle. .circleincircle.51 111 100 400 10.7 670 87.3 119 8 97.3 2.7 .circleincircle. .circleincircle. .circleincircle.52 119 100 400 10.7 630 87.7 128 9 96.9 3.1 .circleincircle. .circleincircle. .circleincircle.__________________________________________________________________________
EXAMPLE 15
A crystallization process was repeated six times under the same conditions as employed in Example 13 except that, as a primary seed crystal, AlF.sub.3.3H.sub.2 O particles having an average particle diameter of 78 .mu.m, and containing not more than 5% of fine particles having an average diameter of 40 .mu.m or less and 8.1% of moisture were remployed under the conditions indicated in Table 16. As a result, large AlF.sub.3.3H.sub.2 O crystals having an average particle diameter of 141 .mu.m were obtained. Results are shown in Table 16.
TABLE 16__________________________________________________________________________ AlF.sub.3 Super SaturatedSeed Crystal Solution Share of Aver- Ini- Ratio Average Crystallization age tial of Particle Genera- Judgement Par- Sur- Con- Crys- Diameter Growth tion of Genera- ticle face cen- tal- Large Fine of Fine tion of Dia- Area Weight tra- liza- Par- Par- Seed Par- Fine meter (m.sup.2 / (Dry) tion Weight tion ticle ticle Crystal ticle Surface Par-No. (.mu.m) kg) (g) (%) (g) (%) (.mu.m) (.mu.m) (%) (%) Feature ticle Summary__________________________________________________________________________53 78 75 300 14.7 670 91.0 89 14 93.9 6.1 .circleincircle. .smallcircle. .smallcircle.54 89 75 300 14.9 570 91.7 99 14 91.7 8.3 .circleincircle. .smallcircle. .smallcircle.55 99 75 300 14.7 520 91.5 110 13 92.4 7.6 .circleincircle. .smallcircle. .smallcircle.56 110 75 600 15.1 920 91.8 120 15 91.2 8.8 .circleincircle. .smallcircle. .smallcircle.57 120 75 600 14.9 860 91.5 13 13 91.6 8.4 .circleincircle. .smallcircle. .smallcircle.58 130 75 600 14.9 800 91.6 141 13 90.3 9.7 .circleincircle. .smallcircle. .smallcircle.__________________________________________________________________________
EXAMPLE 16
Seed crystal was prepared as follows. 600 g of a super-saturated solution of aluminum fluoride having a concentration of 17% was charged into a polyethylene wide-mouthed bottle 9 cm in diameter and 1000 cm.sup.3 in volume. Then the crystallization of AlF.sub.3.3H.sub.2 O was conducted in the same manner as in Example 1.
The agitation during the crystallization was conducted by using a flat-blade type agitator having a blade length of 8 cm at the blade-tip velocity of 2.1 m/sec. The slurry of AlF.sub.3.3H.sub.2 O thus obtained was subjected to classification by using a hydraulic classifier to separate fine particles having a diameter of 40 .mu.m or less, and then filtered to obtain a filter cake. As a result, the filter cake comprising AlF.sub.3.3H.sub.2 O particles having an average particle diameter of 58 .mu.m, and containing not more than 5% of fine particles having an average diameter of 40 .mu.m or less and 8.1% of deposited water was obtained and used as a primary seed crystal. The crystal growth of seed crystal was carried out by repeating a crystallization process four times under the same conditions as in Example 13, thereby obtaining large AlF.sub.3.3H.sub.2 O particles having an average particle diameter of 129 .mu.m. The AlF.sub.3.3H.sub.2 O particles thus obtained were spherical in shape and high in surface denseness. Results are shown in Table 17.
TABLE 17__________________________________________________________________________ AlF.sub.3 Super SaturatedSeed Crystal Solution Share of Aver- Ini- Ratio Average Crystallization age tial of Particle Genera- Judgement Par- Sur- Con- Crys- Diameter Growth tion of Genera- ticle face cen- tal- Large Fine of Fine tion of Dia- Area Weight tra- liza- Par- Par- Seed Par- Fine meter (m.sup.2 / (Dry) tion Weight tion ticle ticle Crystal ticle Surface Par-No. (.mu.m) kg) (g) (%) (g) (%) (.mu.m) (.mu.m) (%) (%) Feature ticle Summary__________________________________________________________________________59 58 40 100 11.4 750 88.3 75 24 93.9 6.1 .circleincircle. .smallcircle. .smallcircle.60 75 40 100 11.4 580 88.1 93 22 92 8 7.2 .circleincircle. .smallcircle. .smallcircle.61 93 40 100 11.3 480 87.5 111 22 92.1 7.9 .circleincircle. .smallcircle. .smallcircle.62 111 40 150 11.4 590 87.4 129 21 90.7 9.3 .circleincircle. .smallcircle. .smallcircle.__________________________________________________________________________
EXAMPLE 17
The AlF.sub.3.3H.sub.2 O filter cake obtained in Example 10 was dried in a hot air circulating type thermostatic chamber (300.degree. C.) for one hour to remove water therein. As a result, this aluminum fluoride was dehydrated to have a composition of AlF.sub.3.3H.sub.2 O. This aluminum fluoride was transferred to a platinum dish, which was then kept in an electric furnace to be heated at 600.degree. C. for one hour thereby to finish a complete dehydration thereof. The content of aluminum fluoride, average particle size, amount of particles passed through a 44 .mu.m mesh, bulk density, angle of repose, and factor "a" of the empirical equation relating pressure to powder volume were measured on this sample, and samples obtained in Examples 1 and 3. The factor "a" of the compression fluid equation is identical to the constant "a" of the following empirical equation about the compression of particles, which represents a relationship between a pressure applied from outside to a particle layer and a volume change. Smaller the value of "a", larger the fluidity of particles (Particle technology handbook, Japan Particle Technology Society, Daily Industry Newspaper Co., pages 145-146, Feb. 28, 1986).
P/C=1/ab+P/a
P . . . Pressure (Pa)
C . . . Degree of decrease in apparent volume (-)
C=(V.sub.0 -V)/V.sub.0
V.sub.0 . . . Initial volume (m.sup.3)
V . . . A volume under pressure (m.sup.3)
a, b . . . Constants
Results are shown in Table 18.
In this Table 18, No. 63 indicates the results obtained in Example 1; No. 64 indicates the results obtained in Example 3; to produce aluminum fluoride anhydride the product of A Co. was aluminum fluoride anhydride obtained by a wet method; and the product of B Co. was aluminum fluoride anhydride obtained by a dry method.
TABLE 18__________________________________________________________________________ Bulk Density Angle of Repose Average Injec- Ex- Particle tion haust AlF.sub.3 Diameter 44 .mu.m> Loose Packed Method MethodNo. (%) (.mu.m) (%) (g/cm.sup.3) (g/cm.sup.3) (.degree.) (.degree.) Factor "a"__________________________________________________________________________63 98.3 103 0.0 0.78 0.89 31 32 0.14064 98.3 107 0.0 0.78 0.90 31 32 0.13565 98.5 140 0.0 0.80 0.90 30 32 0.129Product 97.5 63 7.9 0.75 0.94 35 42 0.207of A Co.(WetMethod)Product 91.4 95 0.0 1.39 1.64 33 35 0.152of B Co.(DryMethod)__________________________________________________________________________
As shown in this Table 18, as far as the angle of repose and the constant "a" are concerned, the products Nos. 63 to 65 according to this invention are equal to or lower than those of aluminum fluoride anhydride obtained by the dry method. As apparent from these results, aluminum fluoride anhydride having a large average diameter to be obtained according to the wet method of this invention is excellent in fluidity, and generates only a very small amount of dust since it contains not more than 1% of fine particles 44 .mu.m or less in diameter.
Additional advantages and modifications will readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific details, and illustrated examples shown and described herein. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents.

Number	Name	Date
1797994	Morrow	Mar 1931
2801907	Scott	Aug 1957
3057681	Gernes	Oct 1962

Number	Date	Country
88 080	Feb 1974	DDX
139833	Jan 1980	DDX
248 349 A1	Aug 1987	DDX
275 855 A1	Feb 1990	DDX
45-36207	Nov 1970	JPX
5-229817	Sep 1993	JPX
782423	Sep 1957	GBX
1074665	Jul 1967	GBX

	Number	Date	Country
Parent	542037	Oct 1995
Parent	237588	May 1994

Method of manufacturing aluminum fluoride anhydride

Information

Patent Number

Date Filed

Date Issued

Inventors

Original Assignees

Examiners

Agents

CPC

US Classifications

Field of Search

US

International Classifications

Abstract

Description

Claims

Priority Claims (1)

Parent Case Info

US Referenced Citations (3)

Foreign Referenced Citations (8)

Non-Patent Literature Citations (6)

Continuations (2)

Entry
2nd International Congress on Phosphorous Compounds Proceeding, Apr. 21-25, 1980, Boston, USA pp. 799-808, J. Y. Morlock et al.
Perry, "Chemical Engineers' Handbook", Third edition, 1950, pp. 1050-1061.
Derwent abstract for: DD 248,349 (Aug. 1987).
Derwent abstract for: DD 275,855 (Feb. 1990).
Derwent abstract for: JP 5-229,817 (Sep. 1993).
Derwent abstract for: DD 88,080 (Feb. 1974).