An important property of aqueous solutions is pH because it affects chemical and biochemical properties such as chemical reactions, equilibrium conditions, and biological toxicity. With the increasing uptake of fossil fuel CO2 into the oceans, a decrease in pH is important to consider at this time. Unfortunately, many different methods for assessing pH have been used by different groups. The objectives of this review were to (1) briefly examine the concept of pH as it was introduced and developed, up to the current scientific developments, assumptions, and recommendations, (2) critically assess the various approaches that different scientific groups have adopted for pH, balancing their preferences and arguments, (3) compare measuring vs. modeling pH, and (4) issue recommendations on an optimized approach or approaches for pH.
The main conclusions of this review are: (1) pH definitions and conventions are highly variable, which leads to highly variable estimates of pH. For example, for seawater at SA = 35.165 g/(kg soln), t = 25 °C, P = 1.0 atm, and fCO2 = 3.33E-4 atm, model calculated pH values varied from 8.08 to 8.33 on the various pH scales; (2) An acceptable nomenclature is needed to keep pH variability unambiguous, due to alternative definitions and conventions. A nomenclature example is given in this paper. It is the (still unsolved) task of international bodies such as IUPAC or IOC to develop and promote such widely recognized conventions; (3) pH can be accurately estimated based on measurement (potentiometric, spectrophotometric) and modeling approaches. Accuracy via different definitions and conventions clearly requires consistency with respect to experimental measurements, equilibrium constants, activity coefficients, and buffer solutions that are used for specific approaches; (4) “Total” pH accuracy that includes the Bates-Guggenheim convention is ± 0.01 pH units. Removing the Bates-Guggenheim convention from the accuracy calculation can lead to “conventional” accuracies of ± 0.004 pH units; (5) pH extensions to high solution concentrations are capable using the Pitzer modeling approach. Modeling can, in principle, lead to pH estimates that are more accurate than measurements, which is illustrated with two Pitzer models for natural waters made up of the major components of seawater. But this principle still needs to be proven; (6) It is recommended that ocean scientists use the free concentration or activity of the proton to examine the effect of pH on processes in the oceans.
Marion G. M., Millero F. J., Camões M. F., Spitzer P., Feistel R. & Chen C.-T. A., in press. pH of seawater. Marine Chemistry. Article (subscription required).