a1 Department of Earth and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, Mass., USA.
Two thermodynamic schemes have been developed to describe the CO2 system in sea water. The ‘pure’ treatment is based on the theory of dilute solutions and is general. The ‘pseudo’ treatment uses ‘stoichiometric’ constants related to the acidity constants of Bronsted (1928). It is empirical and specific to the medium in which the measurements are made. The pure thermodynamic treatment developed from the early attempts to characterise the various equilibria involving CO2, carbonic acid and its anions and was much influenced by the work of biochemists and physiologists on the blood plasma system. These attempts were initiated before the concepts of complete dissociation or of activities were developed. The subsequent application of these concepts was immediate and fruitful. By 1927, especially as a result of the work of Warburg and of Hastings and Sendroy, a consistent scheme had been developed for the NaCl—NaHCO3—Na2CO3 system. This work had also been extended to phosphate, citrate and blood plasma solutions and the discrepancies between the apparent activity coefficients discussed and tentatively ascribed to complex formation.
Johnston and co-workers developed the thermodynamic treatment of the carbonate system in natural waters and by 1920 understood the relation between ΣCO2, alkalinity and carbonate solubility and had identified and solved the major analytical difficulties.
The discovery of the interference due to complexing prompted various workers to determine empirical 'apparent' or stoichiometric constants. These are derived from the experimental data by assuming that a function analogous to the pure constant, but defined in terms of total species concentrations rather than activities, will behave like the pure constant over the temperature, salinity and pressure range of interest and will not vary with the concentration of any of the species it describes. Starting with the work of Moberg and Buch in the early 1930s, a complete scheme has been developed based on this treatment. The apparent constants for carbonic and boric acids in sea water are now known with sufficient accuracy to allow detailed study of the effect of water chemistry on the formation and dissolution of carbonates under any oceanic conditions.