Removal of Ca.sup.++ from MgCl.sub.2

Abstract

Removal of Ca.sup.++ values from MgCl.sub.2 brine is accomplished by use of a crystalline zeolite structure of the empirical formula Na.sub.12 [(AlO.sub.2).sub.12 (SiO.sub.2).sub.12 ].nH.sub.2 O, where nH.sub.2 O represents waters of hydration.

Description

BACKGROUND OF THE INVENTION

Aqueous MgCl.sub.2 solutions are obtained by various techniques, such as by aqueous chlorination of Mg-containing minerals (e.g., dolomite), by taking natural MgCl.sub.2 brines from natural underground deposits, or by chlorination of Mg hydrate which is formed by alkalizing seawater. These MgCl.sub.2 brines are useful for various purposes, such as being a source of MgCl.sub.2 for use as cell feed to an electrolytic cell where molten MgCl.sub.2, dissolved in a molten salt electrolyte, is electrolyzed to form Mg metal at the cathode.

One of the common impurities found in MgCl.sub.2 brines is Ca.sup.++, present as a soluble compound.

Various techniques have been used to substantially reduce the level of Ca.sup.++ ions in MgCl.sub.2 brines, such as by treatment with excess sulfate ions to suppress solubility of the Ca.sup.++ as CaSO.sub.4 which is only slightly soluble in water (whereas MgSO.sub.4 is highly soluble); after separating the insolubles there still remains a small amount of Ca.sup.++ in solution and there still remains a need for sulfate removal from the MgCl.sub.2 solution. There is a need for reducing the Ca.sup.++ content of relatively concentrated MgCl.sub.2 solutions in an efficient and expeditious manner, especially MgCl.sub.2 solutions which are intended to be the source of MgCl.sub.2 as a cell feed to an electrolytic Mg cell.

In the Journal of The American Chemical Society, Vol. 78, number 23, Dec. 8, 1956, pp. 5963-5977, there are two companion articles by D. W. Breck, et al. The articles are: Crystalline Zeolites I. The Properties of a New Synthetic Zeolite, Type A, and Crystalline Zeolites. II. Crystal Structure of Synthetic Zeolite, Type A. These published articles are incorporated herein by reference to show the preparation and analyses of Type A zeolites. These Type A zeolites are said to be represented by the formula Na.sub.12 [(AlO.sub.2).sub.12 (SiO).sub.12 ].27H.sub.2 O, characterized as cubic, a.sub.0 =12.32 A., space group O.sub.h '-Pm3m, having a 3-dimensional network consisting of cavities 11.4 A. in diameter separated by circular openings 4.2 A. in diameter. Removal of the crystal water leaves a stable crystalline solid containing mutually connected intracrystalline voids amounting to 45 Vol. % of the zeolite. The article also discloses that the sodium ions undergo cation exchange in aqueous solution and that for dipositive ions the order of decreasing selectivity of the Type A zeolite was determined to be

Zn<Sr<Ba<Ca<Co<Ni<Cd<Hg<Mg and in Table VII of the article it is disclosed that the selectivity of Ca.sup.++ is 0.72 and of Mg.sup.++ is 0.43, these materials being each tested as 0.2 N solutions at 25.degree. C. The selectivity x (100) is defined as the extent of cation exchange achieved by contacting the zeolite with a solution containing the exact equivalence of exchanging ion.

In the intervening years since the above articles were published, the sodium forms of the synthetic Type A zeolites, conforming substantially to the empirical formula

Na.sub.12 [(AlO.sub.2).sub.12 (SiO.sub.2).sub.12 ].nH.sub.2 O have become known in industry as Type 4A zeolites and are sold as such by Linde Air Products Company, Davison, and others. It is these Type 4A zeolites which are of interest in the present invention.

We have found that in highly concentrated MgCl.sub.2 brines that a zeolite of the 4A structure, Na.sub.12 [(AlO.sub.2).sub.12 (SiO.sub.2).sub.12 ].nH.sub.2 O, has an unexpectedly high selectivity for calcium ions with respect to the much higher concentration of magnesium ions. As pointed out above, the selectivities of Mg.sup.++ and Ca.sup.++ were tested by Breck, et al at 0.2 N concentrations. With our instant solutions the Mg is on the order of 25 times (or more) the 0.2 N concentration used by Breck, et al. The selectivity as listed by Breck, et al shows zeolite Type 4A to prefer Ca:Mg at a ratio of 72:43 when contacting the zeolite with an exact equivalency of exchanging ions. We have now found that the zeolite, 4A, when challenged by the high-Mg.sup.++, low-Ca.sup.++ concentrations still rejects Mg.sup.++ and removes Ca.sup.++ to low levels.

Stated another way, one would expect from Breck, et al that with a concentrated MgCl.sub.2 brine containing a small percent of Ca.sup.++ ions, the high concentration of Mg.sup.++ ions would override the tendency toward Ca.sup.++ selectivity and that the Mg.sup.++ selectivity would prevail in such a high concentration.

We have unexpectedly found that the present process, which uses a Type 4A zeolite, selectively removes low concentration Ca.sup.++ ions from a concentrated MgCl.sub.2 brine. The sodium ions which are substituted for the calcium are then readily removed as NaCl which precipitates when the brine is further concentrated. The zeolite can readily be returned (regenerated) to the sodium form with a concentrated NaCl brine.

SUMMARY OF THE INVENTION

In a MgCl.sub.2 brine, where the concentration of MgCl.sub.2 is in the range of about 30% or more by weight, Ca.sup.++ ions are reduced from a given level to a lower level by contacting the brine with a synthetic crystalline zeolite which conforms substantially to the empirical formula Na.sub.12 [(AlO.sub.2).sub.12 (SiO.sub.2).sub.12 ].nH.sub.2 O, where nH.sub.2 O represents waters of hydration.

DETAILED DESCRIPTIONS

The Type A (or Type 4A) synthetic crystalline zeolite structure of interest in the present invention is as shown above. The number of waters of hydration is indefinite and non-critical and the value of n in the empirical formula is generally in the range of about 5 to about 27.

The MgCl.sub.2 brine has a concentration of at least about 30% MgCl.sub.2 by wt., preferably about 34% to about 38%; though the concentration can be as high as 38% at 100.degree. C.

The amount of Ca.sup.++ ion (as CaCl.sub.2) in the MgCl.sub.2 brine before contact with the zeolite exchanger is generally in the range of about 1.0% to about 3.0% by weight, but it can be more or less.

After undergoing one or more ion exchange treatments with the zeolite, the MgCl.sub.2 brine product generally contains less than about 0.4% Ca.sup.++ as CaCl.sub.2 by weight (based on 100% MgCl.sub.2) preferably less than about 0.13% CaCl.sub.2, and is generally in the range of about 0.06% to about 0.10% CaCl.sub.2, but it can be lower.

In carrying out the present process, the zeolite is contained in a vessel, such as a column, and is completely wetted or covered with a NaCl brine.

The NaCl brine is preferably washed out or pushed out by using a pure MgCl.sub.2 solution (e.g., 3-30% conc.) which contains virtually no Ca.sup.++. Then the Ca-contaminated MgCl.sub.2 feed stream is passed through the bed of zeolite until analysis shows the Ca.sup.++ is breaking through with the MgCl.sub.2 effluent.

The flow of MgCl.sub.2 feed stream is stopped and the MgCl.sub.2 in the zeolite is replaced by NaCl brine which, because of the high concentration of Na.sup.+ ions causes the zeolite to convert from the Ca.sup.++ form back to the Na.sup.+ form and the Ca.sup.++ ions flow out (as CaCl.sub.2) with the effluent. The NaCl brine-containing zeolite is once again ready for contact with MgCl.sub.2 brine.

As with the general practice of ion exchange processes, well known to practitioners, the effluent from the zeolite can be cut into fractions to isolate portions which contain the various concentrations of brines.

The temperature at which the ion exchange process is operated may be in the range of about 20.degree. C. to about 120.degree. C., preferably in the range of about 70.degree. C. to about 95.degree. C.

Analyzing of the fractions for Ca.sup.++ and Mg.sup.++ may be done by standard versene titration.

EXAMPLE 1

Zeolite of Type 4A structure is prepared from 113.6 gm. of Na.sub.2 SiO.sub.3.9H.sub.2 O in 450 cc H.sub.2 O to which is added 85.2 gm of a solution prepared from 5 mol of Bayerite (Al(OH).sub.3) and 7.5 mol. of 50% NaOH. The cold (ambient temp.) mixture is well stirred to form a fine dispersion of creamy texture and then heated to 90.degree. C. for about 15 minutes; the mixture has a volume of 610 cc. After removing 305 cc of the mixture, 20 gm of 1-micron zeolite (sold as VALFOR.RTM.) is stirred in at 90.degree. C. for 15 minutes, then the removed portion (305 cc) is added in about 40 cc increments every 4 minutes. After final addition, heating is continued for about 15 minutes. The slurry is filtered and well washed with H.sub.2 O to yield agglomerated 4A zeolite with particle size from about 5- 25 microns. Each agglomerate is made up of numerous particles of the 5-25 micron size.

About 100 cc of the agglomerated zeolite 4A is tested in a column of about 8" in height which is equipped with a jacket to permit circulation of a temperature-controlled heat transfer fluid. The zeolite is retained on a medium glass frit within the column.

The column is first treated with a saturated NaCl brine, then a flow of 21% MgCl.sub.2 (pure), then a solution containing 4.85 molar MgCl.sub.2, 0.102 molar CaCl.sub.2, and NaCl up to saturation. This is followed by the 21% MgCl.sub.2 (about 2.6 molar Mg.sup.++) (pure), then saturated NaCl brine. Flow rate is 3.34 cc/min. at 85.degree.-90.degree. C. Effluent samples (cuts) of 10 cc each are collected and analysed by standard versene titration for Ca.sup.++ and Mg.sup.++. The results are shown in Table I. In Table I, "DO" means "ditto"; a series of dashes indicates no analysis made.

TABLE I______________________________________ Flow Molar Molar Flow Molar Molar Re-Cut Ca.sup.++ Mg.sup.++ Remarks Cut Ca.sup.++ Mg.sup.++ marks______________________________________1 0.052 0.264 21% 28 nil -- Mg/Ca MgCl.sub.2 feed2 0.048 0.092 21% 29 nil -- Mg/Ca MgCl.sub.2 feed3 0.044 0.076 21% 30 0.012 4.855 Mg/Ca MgCl.sub.2 feed4 0.032 0.081 Mg/Ca 31 -- -- Mg/Ca feed feed5 0.036 0.076 Mg/Ca 32 0.024 -- Mg/Ca feed feed6 0.036 0.088 Mg/Ca 33 -- -- Mg/Ca feed feed7 0.024 0.108 Mg/Ca 34 0.024 4.916 Mg/Ca feed feed8 0.032 -- Mg/Ca 35 -- -- Mg/Ca feed feed9 0.020 -- Mg/Ca 36 0.032 -- Mg/Ca feed feed10 0.024 0.136 Mg/Ca 37 0.032 -- Mg/Ca feed feed11 0.020 0.988 Mg/Ca 38 0.032 4.888 Mg/Ca feed feed12 0.016 1.404 Mg/Ca 39 -- -- Mg/Ca feed feed13 0.020 2.212 Mg/Ca 40 0.044 4.836 Mg/Ca feed feed14 0.016 3.120 Mg/Ca 41 -- -- 21% feed MgCl.sub.215 0.016 3.834 Mg/Ca 42 -- -- 21% feed MgCl.sub.216 nil 3.92 Mg/Ca 43 -- -- 21% feed MgCl.sub.217 nil 4.54 Mg/Ca 44 -- -- NaCl feed18 nil 4.54 Mg/Ca 45 -- -- NaCl feed19 nil 4.64 Mg/Ca 46 0.048 4.862 NaCl feed20 nil 4.74 Mg/Ca 47 0.064 -- NaCl feed21 nil -- Mg/Ca 48 0.084 4.686 NaCl feed22 nil -- Mg/Ca 49 0.140 4.43 NaCl feed23 nil -- Mg/Ca 50 0.184 3.926 NaCl feed24 nil -- Mg/Ca 51 0.196 3.544 NaCl feed25 nil -- Mg/Ca 52 0.168 3.182 NaCl feed26 nil 4.856 Mg/Ca 53 0.124 2.772 NaCl feed27 nil -- Mg/Ca feed______________________________________

EXAMPLE 2

A second test was made under nearly identical conditions as Example 1 above, but the feed brine was 0.518 molar Ca.sup.++ and 3.99 molar Mg.sup.++. Even with this feed of higher Ca.sup.++, lower Mg.sup.++ the Ca.sup.++ was preferred.

EXAMPLE 3

A commercial sample of Linde.RTM. 4A zeolite molecular sieve was ground wet in a Waring.RTM. blender and screened to -50 mesh and the very fine material removed by washing. This product was tested in the same column of Example 1 using the same brine as in Example 2. Again, the zeolite displayed the ability to remove Ca.sup.++ from the MgCl.sub.2 solution at a surprisingly high selectivity.

EXAMPLE 4

The effect of particle size of type 4A zeolites was tested using screenings which were retained on a 30-mesh screen (U.S. Standard Sieve) but which passed through a 10-mesh screen, compared with screenings which passed through the 30-mesh screen and screenings which passed through a 50-mesh screen. The smaller particles which passed through the 30-mesh screen and the 50-mesh screen were more efficient than the larger particles and Ca.sup.++ breakthrough was not encountered until greater volumes of feed were passed through, using substantially the same exchange column, process, and feed of Example 1 above.

EXAMPLE 5

A sample of commercially-available type 4A zeolite was wet ground and screened to collect the portion passing through a 20-mesh screen and collected on a 50-mesh screen.

The same column, process (90.degree. C.) and brine feed as used in Example 1 above was used here except that water was added to dilute the Ca.sup.++ from a 0.102 molar feed to a 0.096 molar feed; also, the saturated 26% NaCl brine was also diluted 4 volumes/1 volume H.sub.2 O for use as the regenerate. Data is given in Table II below for cuts 11-99. The first cut was 50 cc and all others were 10 cc. The density of cuts 2-10 were measured as follows:

______________________________________ g/cc g/ccCut Density Cut Density______________________________________2 1.165 7 1.2763 1.164 8 --4 1.214 9 1.3145 1.249 10 1.3166 1.273______________________________________

In addition, the molar concentration of Mg.sup.++ in cut 10 was measured as being 4.22. In Table II a series of 3 dashes means no analysis was made.

TABLE II__________________________________________________________________________ Molar Molar Molar Molar Molar MolarCut Ca.sup.++ Mg.sup.++ Cut Ca.sup.++ Mg.sup.++ Cut Ca.sup.++ Mg.sup.++__________________________________________________________________________11 .004 -- 41 .085 5.06 72 .021 --12 .005 -- 42 .079 -- 73 -- --13 .005 -- 43 .083 5.06 74 .022 --14 -- 4.81 44 .064 -- (upflow MgCl.sub.2)15 .008 -- 45 .054 5.09 7516 .009 -- 46 .071 -- 7617 -- -- 47 .119 4.89 77 * .033 .02918 .017 -- 48 .197 4.153 7819 .020 -- 49 .295 3.33 7920 -- -- 50 .307 2.37 80 .003 .03721 -- -- 51 .263 1.75 81 .001 .04022 .024 -- 52 .238 1.40 82 -- .17023 .025 -- 53 .194 1.04 83 -- 2.9524 -- -- 54 .157 .86 84 -- 3.7025 .028 -- 55 .103 .58 85 -- 4.1626 .027 -- 56 .09 -- 86 .001 4.4227 -- -- 57 .065 -- 87 -- --28 -- -- 58 88 -- --29 .033 -- 59 89 -- --30 .035 -- 60 * .065 -- 90 -- --31 .037 -- 61 91 -- --32 .047 -- 62 92 .009 4.9533 .0488 -- 63 .046 -- 93 -- --34 .048 -- 64 94 .006 --35 .050 -- 65 95 .013 --36 .047 -- 66 * .037 -- 96 .017 5.0837 .053 -- 67 97 .027 --38 .055 -- 68 98 .030 --39 .055 -- 69 .031 -- 99 .0280 5.0840 .055 5.02 70 .032 --(downflow NaCl) 71 .026 --__________________________________________________________________________ *Cuts combined before analysis.

The foregoing examples illustrate the practice of the invention, but the invention is not limited to the particular embodiments illustrated.

Claims

1. A process for substantially reducing the amount of a Ca.sup.++ ion impurity in a MgCl.sub.2 brine, where the MgCl.sub.2 is at a concentration of at least about 30% by weight,

said process comprising contacting said MgCl.sub.2 brine with a crystalline synthetic zeolite having the empirical formula Na.sub.12 [(AlO.sub.2).sub.12 (SiO.sub.2).sub.12 ].nH.sub.2 O where nH.sub.2 O represents waters of hydration, and

recovering from said zeolite the MgCl.sub.2 brine with the Ca.sup.++ ion impurity reduced to a lower level than the initial impurity level.

2. The process of claim 1 wherein the particle size of the zeolite is in the range of about 5 to about 25 microns, said particles being agglomerated.

3. The process of claim 2 wherein the agglomerates are of a size which pass through a 10-mesh screen.

4. The process of claim 2 wherein the agglomerates are of a size which pass through a 30-mesh screen.

5. The process of claim 2 wherein the agglomerates are of a size which pass through a 50-mesh screen.

6. The process of claim 1 wherein there is performed the additional steps of regenerating the zeolite to the Na.sup.+ form by washing it with NaCl brine, thereby also substantially removing the Ca.sup.++ from the zeolite, the regenerated zeolite being re-usable in the process.

7. The process of claim 1 wherein the initial Ca.sup.++ ion impurity level of the MgCl.sub.2 brine is as much as about 3.0% (as CaCl.sub.2) and the Ca.sup.++ ion impurity level of the MgCl.sub.2 brine after treatment with the zeolite is lower than about 0.4% (as CaCl.sub.2).

8. The process of claim 1 wherein the value of n in the formula is a value in the range of about 5 to about 27.

9. The process of claim 1 wherein the concentration of MgCl.sub.2 in the MgCl.sub.2 brine is in the range of about 30% to about 38% by weight.

10. The process of claim 1 wherein the concentration of MgCl.sub.2 in the MgCl.sub.2 brine is in the range of about 34% to about 38% by weight.