J O U R N A L OF THE AMERICAN CHEMICAL SOCIETY (Registered in U. S. P a t e n t Office)
(0Copyright, 1957, by the American Chemical Society) MAY 7, 1957
VOLUME 79
NUMBER 9
PHYSICAL AND INORGANIC CHEMISTRY [CONTRIBUTION FROM THE
DEPARTMENT OF CHEMISTRY,
UNIVERSITY O F WISCONSIN]
Precipitation of Calcite and Aragonite BY JOHN L. WRAYAND FARRINGTON DANIELS RECEIVED OCTOBER22, 1956 The two common polymorphs of calcium carbonate, calcite and aragonite, have been precipitated by mixing soluble carbonate solutions with solutions of calcium ions under conditions of controlled temperature, concentration and aging. Results from these experiments have provided new information regarding the factors affecting the formation of these two crpstalline varieties.
Introduction use in chromatography and the efficiency in such Although the preparation of calcium carbonate work depends not only on the size of the particles, has been described many times in the chemical and but also on the calcite-aragonite ratio. Again, geological literature, the products usually have not calcite and aragonite are found in nature and their been well defined and they have varied consider- origin is important in the interpretation of geologic ably with the experimental conditions. Among the phenomena. important early workers who have attempted to Experimental determine the factors controlling the formation of Calcium carbonate was precipitated in 500-ml. roundthe various crystalline varieties of calcium carbon- bottom flasks by mixing soluble carbonate solutions with solutions of calcium ions with rapid mixing by a motorate are Credner,‘ and Johnson, Merwin and Wil- driven stirrer. Reagent grade calcium nitrate manufacliamson.2 A comprehensive summary of much of tured by Merck and Co. was used. Spectrographic analysis the early work was compiled by M e l l ~ r . ~More of selected impurities in this material given in parts per recently, studies of artificial calcium carbonate million are as follows: MgO, 80; FezOs, 20; MnO, 3; SrO, 30; Pb The pH of the 0.1 M C a ( N 0 3 ) ~was 6.3 a t 30" 30Y0 calcite and 7oY0 aragonite, but changes durand increased to 9.1 with the addition of the 1 M ing two hours digestion to more than 90% calcite. T\'azC03. The grain size varied from about 10 to A t 50' the precipitate is mostly aragonite. At temperatures between 1 5 and Z O O , with suitable 20 f i . The effect of increasing the concentrations of the periods of digestion, one can obtain practically any reagents, using 1 M solutions of both Ca(N03)Z ratio of calcite to aragonite. Although the preand NazC03, was to give a finer grained precipitate cipitate suspended in solution may go rather rapidly (5-10 u ) and products that contain a much greater to the more stable calcite, this change is stopped proportion of calcite and vaterite than aragonite by filtering and drying the material. The rate of when precipitated a t high temperatures and al- change in the dry state is indefinitely slow in the lowed to age only for brief periods. For example, laboratory. The temperature a t which dry aragonite changes a precipitate containing 7Oyocalcite, 20Y0 vaterite and 10% aragonite was obtained a t 70" after 6 spontaneously into calcite a t atmospheric pressure minutes aging when 100 ml. of 1 M NazC03 was is quite high (usually above 400"). At very high pressures, aragonite is thermodynamically st:hlc : I ~ ~ t to ~ c ino r ~ lrli 0' 1 :i c,lisojii
PRECIPITATION OF CALCITE AND ARAGONITE
X a y 5 , 1957
at lower temperatures, even a t room temperature. The phase equilibrium diagram for the different crystalline varieties of calcium carbonate has been fully worked There is much geological and laboratory evidence to show that the aragonite crystal structure is favored by the inclusion in its lattice of cations as impurities which are larger than the calcium ion such as strontium, barium, and The ionic radii of the cations of the calcite and aragonite groups are given in Table IV.
Fe
Zn
..
Mn
Cd
140
YI L
L
+E
IONIC RADIIOF CATIONS, ANGSTR~M UNIT^ Mg
18.0
-100
TABLE IV
Calcite 0.78 0.83 0.83 type Aragonite type .. ,.
203 3
Ca
Sr
Pb
Ba
c
0
0.91 1.03 1.06 ,
.
..
..
..
c YI
al 0
.-
n ,.
1.06 1.27 1.32 1.43
In this Laboratory, i t was foundg that aragonite generally has a higher content of strontium than calcite, and that the formation of aragonite is induced by adding strontium, barium or lead under conditions, namely, high pH, such that they will be co-precipitated with tlie calcium carbonate. Only in case these ions are incorporated into the lattice would they be expected to produce the aragonite. I t was pointed out that the presence of these larger ions under the proper conditions tends to produce the aragonite, rather than that the aragonite is produced from other causes and takes up the strontium, barium, or lead which happen to be in solution. In studying the aging process, it is important to distinguish three different stageslo in the mechanics of crystallization : (a) an aggregation velocity in which the ions are brought together and precipitated; (b) an orientation velocity in which the ions in colloidal or amorphous material of the precipitated particles form crystals; and (c) a rate of recrystallization. In the present work the aggregation velocity is considered to be infinitely rapid, the orientation is probably complete in a few seconds or minutes, and the recrystallization takes place in minutes, hours or days, and longer depending on conditions. With the facts and generalizations discussed here, i t is possible to explain the experimental results and to point out the factors which affect the production of calcite or aragonite. The formation of the unstable aragonite having a larger crystal structure than calcite is favored by the incorporation of larger-sized cations, Sr ++, Ba++ or Pb++, into the CaC03 lattice. The concentration of these ions may be very low, because once a seed crystal of aragonite or calcite is formed, the rest of the crystal growth tends to follow the same crystal pattern. (5) J. C. Jamieson, J . Chem. Phys., 21, 1385 (1953). (6) G.T.Faust, A m . Min., 56, 222 (1950). (7) 11. C. Bloom a n d M. J. Buerger, 2. Krist., 96, Abt. A , , 365 (1937). ( 8 ) Pi. L. Bragg. "Atomic Structure of Minerals," Cornell University Press, Ithaca, h7.Y.,1037,p. 118. (9) E. J. Zeller and J. I.. Wray, Bull. Ameu. Assoc. Pelvol. Geol., 40, 140 (1956). (10) I. M. Kolthoff and E. B Sandell, "Textbook of Quantitative Inorganic Analysis," 3rd Ed., T h e Macmillan Co., New York, X. Y., 1952, p. 113.
60
.-
-
A
10 Calcite
30
50
70
I
90 Aragonite
Percent Aragonite
Fig. 1.-Influence of temperature and aging on the crystal form of precipitated calcium carbonate.
At low pH values, strontium, barium and lead carbonates do not precipitate, and the first colloidal particles of calcium carbonate to form do not contain these larger ions. Calcite is precipitated under these conditions. At higher pH values, however, these larger ions are co-precipitated with the calcium carbonate and there is a tendency to form aragonite. At the instant of precipitation, the strontium ions as well as the calcium ions are brought from a large volume of solution into the small volume of the first colloid particles to form, thus giving in these aggregates a high concentration. I n a matter of seconds or minutes, the colloid aggregate undergoes orientation leading t o crystal formation. If there is sufficient time between precipitation and crystallization, the strontium (and barium and lead) ions can diffuse out of the colloidal particles into the strontium-depleted solution which surrounds them. Factors which tend to shorten this time for escape of strontium ions tend to produce aragonite, and those which lengthen the time before crystallization give more opportunity for the escape of strontium ions and thus tend to produce calcite. At the higher temperatures, the rate of crystallization is accelerated rapidly and the strontium ions do not have time to escape and thus they are incorporated into the CaC03 lattice giving the larger structure-aragonite. In series 1, calcite is formed a t temperatures below 40' and aragonite is formed a t temperatures above 50'. At temperatures between these limits, the incorporation of strontium in the lattice is intermediate and a mixture of aragonite and calcite is formed. The greater the concentration of precipitating
2034
JAMES
S. JOHNSON
AND
KURT-4. KRAUS
Vol. 79
ions the greater the tendency to form colloidal M Sr(NO& I n this flask there was no trace of precipitates and the longer the time available for calcite formation after 20 hours, whereas in the expulsion of strontium ions before crystallization. flask to which strontium nitrate had not been Accordingly, when all other conditions are the added, about 50'% of the aragonite had changed to same, an increase in concentration tends to favor calcite in the same period of time. calcite formation rather than aragonite, as was The fact that similar results were obtained with foulid when 1 molar solutions were mixed. calcium nitrate of two different manufacturers Thus far only the first two stages have been dis- probably indicates that after a certain minimum cussed-the formation of aggregates and the concentration of strontium and other critical imorientation or first crystallization. The work of purities are reached, higher concentrations have de Keyser and Dugueldre4 was concerned only with but little further effect. At first sight, it would be expected that the these first two steps, but extensive changes from aragonite to calcite can take place by further con- recrystallization from aragonite to calcite would tact with the solution from which the precipitates go faster a t high temperatures, but an examination were formed. The transformation of the solid of Fig.1 shows that a t 40' the 95y0 calcite changes aragonite to solid calcite all in the same crystal is to 100yo calcite in half an hour, a t 45' it changes an exceedingly slow process a t room temperature, from 30 to 100% in about 10 hours, and a t 50" but if these crystals are in contact with water, the it changes only from zero to about 35% in 18 more soluble aragonite goes into solution and re- hours. The number of calcite crystals to act as crystallizes as the less soluble calcite. Moreover, nuclei for further growth is much less in the prethe redissolved calcium carbonate produced by the cipitates formed a t the higher temperatures. The solution of the aragonite brings its strontium into slowness to go from aragonite to calcite a t the solution where i t is diluted so that it is less likely higher temperatures is thus explained as due to the relative absence of calcite seed crystals. This to produce aragonite. Spectrographic analyses showed a higher per hypothesis was checked by taking out a precipitate cent. of strontium (80 p.p.m.) in the 70% aragonite of 50% aragonite and 50% calcite slowly changing precipitated a t 45' (Fig. 1) than in the recrystal- a t 45'. When this precipitate in its solution was lized calcite (25 p.p.m. strontium) after 10 hours heated to 70' the change to calcite was complete in two hours, whereas a t 45' it required ten hours. of standing in the solution. The removal of strontium ions from aragonite If sufficient seed crystals are present, the higher which is undergoing recrystallization is believed temperature does give a faster rate of transformaresponsible for the formation of calcite. This tion to the stable calcite, as would be expected. Acknowledgment.-The authors are indebted to hypothesis was confirmed with additional experiments. Pure aragonite precipitated a t 50' and its the Atomic Energy Commission, Contract ATsurrounding solution was divided equally and (11-1)-178,and to a grant for fundamental research placed in two separate flasks and maintained a t by the E. I. du Pont de Nemours Company. 50'. To one of these was added 100 ml. of 0.1 MADISON, WISCOSSIS [CONTRIBUTION FROM THE OAK
RIDGEN.4TIONAL LABORATORY, CHEMISTRY DIVISION, O A K RIDGE,T E S N . ]
Equilibrium Ultracentrifugations of Acidic Hg(I),Hg(II),In(III),and Au(II1) Solutions' B Y JAMES
s. JOHNSON
AND
KURTA.
aL4US
RECEIVED DECEMBER 1, 1956 Equilibrium ultracentrifugations confirm that In(II1) in HBr and Au(II1) in HC1 exist as mononuclear complexes. The results contradict recent suggestions based on solvent extraction data that dimers of In(II1) and Au(II1) exist in the aqueous phase. The method was checked by comparing apparent molecular weights of Hg(1) and Hg(I1) in nitrate solutions.
I n recent years the extractability of several metal ions from aqueous into organic phases has been reported to increase with decreasing concentration of the metal. For example, from 3.7 M NCI, the extraction coefficient of Ga(II1) into his-2-chloroethyl ether increases manyfold as the aqueous Ga(II1) concentration is decreased from Similar changes in extractaM to lo-* bility have been observed for Au(TI1) in HCI, In(111) in HBr and Tl(II1) in HCL3 It has been (1) This document is hased on work performed for t h e U. S. Atomic Euergy Commission a t the Oak Ridge National Laboratory. (2) W. E.Bennett and J. W. Irvine, J r . , in Progress Report, Laboratory for Nuclear Science, hlassachusetts Institute of Technology, F e b . , 1953, AECU 2404, p. 2 3 . ( 3 ) J . W. Irvine, Jr , R . J. Dietz, L. C. Bogar, G. S . Golden and R. Ii. Herber, in Progreqs IZeport. 1,abnratory for L-urlear Science. Massachusetts I n i t i t U t e c t f Te:,hiiology, \ T : i y . l!)dZ, AI'ICG '2043, p. IS.
suggested2 that such results could be explained by the formation of unextractable dimers of the metal ion species in the aqueous phase a t the higher metal concentrations, but alternate explanations have also been Recently Irving and Rossotti have attributed this change in extractability for In(II1) from HBr solutions t o dimerization.' We have attempted to decide between alternate explanations by molecular weight measurements with the equilibrium ultracentrifuge. (4) R . H. Herber, VI. E. Bennett, D.R . Bentz, I,. C. Bogar, R . I . Dietz, Jr., G. S . Golden a n d J. W. Irvine, Jr.. Abstracts, 126th Meeting Am. Chem. Soc., Sept., 1954, p. 33R. (5) J. Saldick, J . Phys. Chem., 60, 500 (19.50). ( G ) A. M. Poskanzer, private communication. ( i ) H. Irving and F. J. C. Rossotti, J. Chcm. Sac., 1938 (1955).
