Partition coefficients in calcite

Morse JW, Bender ML (1990) Partition coefficients in calcite Examination of factors irrflnencing the validity of experimental resnlts and their application to natural systems. Chem Geol 82 265-277 Mucci A, Morse JW (1990) The chemistry of low temperature abiotic calcites Experimental studies on coprecipitation, stability and fractionation. Rev Aquatic Sci 3 217-254 Musgrove ML, Barmer JL, Mack LE, Combs DM, James EW, Cheng H, Edwards RL (2001) Geochronology of late Pleistocene to Holocene speleothems from central Texas Implications for regional paleoclimate. Geol Soc Am Bull 113 1532-1543... 

One of the most controversial topics in the recent literature, with regard to partition coefficients in carbonates, has been the effect of precipitation rates on values of the partition coefficients. The fact that partition coefficients can be substantially influenced by crystal growth rates has been well established for years in the chemical literature, and interesting models have been produced to explain experimental observations (e.g., for a simple summary see Ohara and Reid, 1973). The two basic modes of control postulated involve mass transport properties and surface reaction kinetics. Without getting into detailed theory, it is perhaps sufficient to point out that kinetic influences can cause both increases and decreases in partition coefficients. At high rates of precipitation, there is even a chance for the physical process of occlusion of adsorbates to occur. In summary, there is no reason to expect that partition coefficients in calcite should not be precipitation rate dependent. Two major questions are (1) how sensitive to reaction rate are the partition coefficients of interest and (2) will this variation of partition coefficients with rate be of significance to important natural processes Unless the first question is acceptably answered, it will obviously be difficult to deal with the second question. 

A major problem in the application of Mg2+ partition coefficients in calcite to natural carbonates has been the failure to predict commonly observed compositions of carbonate cements. The Mg content of shallow water Holocene marine cements is typically 1.5 to 3 times that predicted using experimentally measured partition coefficients, and assuming that the solution from which the cements formed had the same Mg to Ca ratio as normal seawater. This problem has been the basis for arguments against the use of experimentally determined partition coefficients (e.g., Given and Wilkinson, 1985a,b). Other observations of natural... 

The many problems that plague partition coefficients in calcite and aragonite (see Chapter 3) probably also are important for dolomite, but comparable experimental data are not available for dolomite at near Earth surface temperatures. Additional complications are likely to arise in dolomite in because of its degree... 

Morse J.W. and Bender M.L. (1989) Partition coefficients in calcite an examination of factors influencing the validity of experimental results and their application to natural systems. Chem. Geol., (in press). 

Gascoyne M (1983) Trace element partition coefficients in the calcite-water system and their paleoclimatic significance. J Hydrol 61 213-222... 

On a brighter side, the model of Kretz (1982) for the partitioning of cations between calcite and dolomite has shown impressive correlations between observational data and predicted partition coefficients (see Table 3.2). He proposes, based on the earlier work of Jacobsen and Usdowski (1976), that different partition coefficients exist for a trace cation at Mg and Ca sites in dolomite, which are principally influenced by the size differences between the trace cation and calcium or magnesium. He has made a number of predictions of metal ion partition coefficients in dolomite. It will be interesting to see if future work bears them out. 

The study of the coprecipitation of Mg2+ in calcite has been an active area of research, frequently marked by controversy over experimental results and their applicability to natural systems. The literature on this topic is prolific, and we will not attempt to review all of it (for a general review see Mackenzie et al 1983). The literature is divided into studies where direct measurement of the solids formed from solutions have been made, studies where properties of the solids have been inferred from their interactions with solutions (e.g., Schoonmaker et al 1982), and papers where authors have estimated values of the partition coefficient by deduction (e.g., Lahann and Seibert, 1982 Given and Wilkinson, 1985a). Here the discussion will be confined to the experimental studies where the compositions of the solids have been directly determined. 

Mucci and Morse (1983) found no statistically significant dependence of the partition coefficient for Mg in calcite precipitated from seawater on reaction rate over a range of seawater supersaturations from about one half ( 2 =3) to close to three times ( 2=17) that of typical surface seawater. However, their experiments had durations of from only a few hours to days. The compositions of magnesian calcites grown very near equilibrium ( 2=1.2) over periods of several months were determined by Mucci et al. (1985). They found excellent agreement with the results of Mucci and Morse (1983) even though different experimental techniques were used. The rate independence of the partition coefficient of Mg in calcite, therefore, has been found to be independent of reaction rate over about 12 orders of magnitude in seawater. 

Several studies have been conducted on the influence of temperature on the partition coefficient of Mg in calcite (see Figure 3.2 for summary). The recent studies of Burton and Walter (1987), Mucci (1987) and Oomori et al. (1987) are all in good agreement, whereas the results of earlier studies scatter significantly. The... 

An important aspect of the coprecipitation of Mg2+ with calcite clearly demonstrated by the studies of Mucci and Morse (1983) is the strong dependence of the partition coefficient on the mole % MgCC>3 in the calcite. The mole % MgCC>3 has also been observed to influence strongly the coprecipitation of Sr2+ and other coprecipitates. 

The results of Lorens (1981) for the partition coefficient of Sr in calcite, in 0.7 M NaCl solution, and at different precipitation rates, are presented in Figure... 

Unfortunately, there is generally a large scatter in the values obtained for these partition coefficients. A possible reason for this scatter, as shown by the study of Lorens (1981), is probably the major effect of precipitation rate on the values of the partition coefficients. It is interesting to note that the partition coefficients for the transition and heavy metals in calcite, studied by Lorens (Cd2+, Mn2+, Co2+), have a negative linear log partition coefficient- log precipitation rate relation, whereas Sr2+ has a positive relation. This behavior may be explained by the fact that the transition metal carbonates are isostructural with calcite, whereas strontium carbonate is isostructural with aragonite. Also, as precipitation rates increase, partition coefficients tend towards unity. 

For models to be useful in studies of natural carbonate minerals, they must also be able to deal with the problem of multiple coprecipitates. The importance of this statement in terms of Sr2+, Na+ and SO42- concentrations has previously been discussed in this chapter. This problem is made particularly formidable by the observations of Angus et al. (1979). They measured the partition coefficients of several ions in calcite and aragonite using electron spin spectroscopy (GSR),... 

---