Thermal decpmposition of 1.1-6.6 m aqueous perchloric acid at 295-322" yields oxygen, chlorine, and a small quantity of hydrochloric acid. The initial rates of decomposition have been measured, using a perchlorate ion activity electrode, and show 3.5 order dependence on the stoichiometric perchloric acid concentration; the corresponding rate coefficient and activation energy at 300.0" are 3.1 x lo-' tn-2.5 s-' and 84 kcal mole-', respectively. The order of reaction with respect to time is less than 3.5 and depends upon the initial perchloric acid concentration; this effect is due to acceleration of the decomposition by the reaction products, notably chloride ion. It is suggested that the ratedetermining step in the decomposition of aqueous perchloric acid is the reaction of two undissociated HC10, molecules to give H 2 0 , CIO,, and C10,. Perchloric acid is used extensively in physical studies of phenomena in aqueous solution, since it is the strongest of the common acids and also because its anion has remarkably little tendency to form complexes with metal ions. The aqueous acid is, however, thermodynamically unstable (AGO = -26.3 kcal/mole-' at 25" for the hypothetical 1 rn solution (1)) with respect to decomposition to oxygen, chlorine, and water, and it is clear that the kinetics of this decomposition must be known if the acid is to be used in extending studies of aqueous electrolyte solutions to high temperatures.Accordingly, the present study was designed primarily to provide empirical rate data on the decomposition of aqueous perchloric acid as a function of temperature, and over a concentration range which would include the commonlychosen "swamping electrolyte" values, 1-6 m. A literature search for such information was unsuccessful, although it has been reported (2) that no chlorine, DOCI, DCIO,, or DCIO, can be detected spectroscopically in 1 m DCIO, after several days in a titanium vessel with sapphire windows at 250". On the other hand, the decomposition of 83-100% HCIO, has received considerable attention (3-8), as has the decomposition of HC10, vapor (9-12).
ExperimentalAll the stock chemicals used in this study were "Baker Analyzed" reagent grade, with the exception of ammonium nitrate (Matheson, Coleman and Bell) and sodium nitrate (Fisher Certified reagent). The perchloric acid (72%) was used without further purification; the experi-'To whom correspondence should be addressed. mental results were independent of the manufacturer's batch. Lithium perchlorate trihydrate was made from lithium carbonate and perchloric acid. Distilled water was passed through Barnstead deionizer and organic removal cartridges before use. Red mercuric oxide was used as the primary standard in acidimetry. Spectrophotometric measurements were made using a Cary Model 15 spectrometer.The pressure vessels used consisted of a cylindrical body (2.54 cm i.d. x 4.60 cm 0.d.) of ATL-40 commercially pure titanium, 0.2% palladium alloy (Atlas Titanium, Welland, Ontario), with a plug of the same material, such that the cavity volum...