lines in gas/gas equilibria, but we have used the method to estimate gas/liquid and liquid/liquid lines for several hydrocarbon systems with success [2,25]. We next remark on the relationship of our curves with experiment. Unfortunately this is not clearcut for the following reasons: (a) Conformal solution theory is not strictly valid and tends to break down for a mixture whose species differ widely in size, mass, and interaction energy [13,14]. Unfortunately such mixtures are the more likely to show gas/gas equilibria, especially mixtures with helium. Our predictions of type I behavior for mixtures of He-C2H6, HeCO2 and He-Xe (for which V2/V4 and T2/T30) where not too satisfactory. Predictions of type II behavior for the mixtures Ne-Kr, Ne-Xe and Ne-CH, were somewhat better. Also the model correctly gave the negative result that mixtures whose components are not disparate, e.g., ArKr, Ar-Xe and Kr-Xe, do not show gas/gas equilibria. 22 22 11 (b) Real mixtures do not obey corresponding states, although the method can be generalized to include the socalled shape factors introduced by Leland and coworkers [6]. (c) The procedure-common with all mixture procedure available at this time-requires interaction parameters such as the 12 and 12 of equation (3) and the numerica properties of a mixture are often, but not necessarily extremely sensitive to a particular choice of and n. W have set and ʼn equal to unity but their influence on the critical lines is quite pronounced, which is demonstrated in figures 4 and 5. The general tendency is: lowering 12 shifts the critical line towards a higher temperature, (the same qualitative effect is obtained by increasing the critical temperatures ratio with the critical volumes ratio held constant): increasing 12 increase the curvature of the critical line. We would like to thank Professor J. Stecki for many helpful comments and Dr. W. R. Smith and Dr. Howard Hanley for discussions. We are also very grateful to Karen Bowie for preparing the manuscript. 6. References [1] Sengers, J. V., and Levelt Sengers, J. M. H., Progress in Liquid Physics, C. A. Croxton, Ed., Wiley, Chichester, U. K., pp. 74-103 (1978). [2] Hanley, J. H. M., and Stecki, J., to be published. [3] Mollerup, J., and Rowlinson, J. S., Chem. Eng. Sci. 29, 1373 (1974). [4] Mollerup, J., Adv. Cryo. Eng. Vol. 20 (Plenum Press, New York). [5] Watson, I. D., and Rowlinson, J. S., Chem. Eng. Sci. 24, 1575 (1969). [6] Leland, T. W., Chappelear, P. S., and Gamson, B. W., A.I.Ch.E. J. 8, 482 (1962); Leland, T. W., Rowlinson, J. S., and Sather, G. A., Trans. Farad. Soc. 64, 1447 (1968). [7] Van Konynenburg, P. H., Ph.D. Thesis, U.C.L.A. (1968). [8] Hicks, C. P., and Young, C. L., J. C. S. Faraday II 73, 597 (1977). [9] Gosman, A. L., McCarty, R. D., and Hust, J. G., Nat. Stand. Ref. Data Ser., Nat. Bur. Stand. (U.S.), 27, (Mar. 1969) Ser. 9] Kammerling Onnes, H., and Keesom, W. H., Proc. Roy. Acad. Sci. Amst. 9, 786 (1907); Proc. Roy. Acad. Sci. Amst. 10, 231 (1907). 0] Krichevskii, I. R., Acta Physicochimica U.R.S.S. 12, 480 (1940). 1] Zandbergen, P., Knaap, H. F., and Beenakker, J. J. M., Physica 33, 379 (1967). 2] Scott, R. L., and Van Konynenburg, P. H., Disc. Farad. Soc. 49, 87 (1970). 3) Scott, R. L., Ber. Bunsenges. Phys. Chem. 76, 296 (1972). 4] Hurle, R. L., Jones, F., and Young, C. L., J. C. S. Faraday II 73, 613 (1977). 5] Hurle, R. L., Toczylkin, L., and Young, C. L., J. C. S. Faraday II 73, 618 (1977). 5] Prigogine, I., and Defay, R., Chemical Thermodynamics (Longmans, London, 1954). JOURNAL OF RESEARCH of the National Bureau of Standards Systematic Errors in an Isoperibol Solution Calorimeter Measured with Standard Reference Reactions Marthada V. Kilday* National Bureau of Standards, Washington, D.C. 20234 July 23, 1980 Systematic errors in an isoperibol calorimeter of a widely-used design, amounting to about 0.5 percent of the endothermic enthalpy of solution of SRM 1655 (KCl) in H2O, were found to be the result of errors in heat leak corrections due to inadequate stirring and commonly used calorimetric procedures. Other systematic errors were found in measurements of the enthalpy of solution of the exothermic reaction of tris(hydroxymethyl)aminomethane in aqueous HCl solution. Recommended procedures are summarized. Key words: Calorimetry; enthalpy of solution; KCl; solution calorimetry; standard reference materials; 1. Introduction Since the beginnings of calorimetry it has been cognized that compromises must be made with ideal nditions in order to design calorimeters for practical plications. It is always hoped that these compromises do t effect the accuracy of the measurements. Design criteria d possible sources of systematic errors in calorimeters ve recently been discussed a priori in a series of papers West, Churney, and Armstrong [1-4].' They also suggest perimental tests to determine whether some systematic rors are likely to occur during calorimetric measurements. ecent work [5] indicated disagreement in results obtained - three calorimeters believed to be capable of easurements of highest accuracy of highest accuracy and experimental precision of 0.1 percent or less. Two of these were operibol calorimeters of the same design used by different entists at different locations; the values measured for the Enter for Thermodynamics and Molecular Science, National Measurement Doratory. igures in brackets indicate literature references at the end of this paper. he sample certified for the enthalpy of solution is designated SRM 1655. It is en from the same lot as SRM 999, the primary analytical standard, and is ilable through the Office of Standard Reference Materials at the National Bureau Standards (U.S.), Washington, D.C. 20234. Available through the Office of Standard Reference Materials at the National reau of Standards (U.S.), Washington, D.C. 20234. enthalpy of solution of KBr (from the same sample) differed by nearly 0.4 percent. The value obtained with the same sample of KBr using our adiabatic calorimeter differed by an additional 0.4 percent. It was apparent that an investigation of the sources of these errors was needed and this was the motivation in this work. It can be regarded as an incomplete work because more experimental work is needed before the specific reasons for some of the results can be given. The standard reference reactions used in this work are (1) the solution of NBS Standard Reference Material (SRM) 16552, KCl, in H2O [6], and (2) the solution of SRM 724a2, tris(hydroxymethyl)aminomethane (TRIS), in 0.1 mol.dm HCl solution [7]. The first reaction is endothermic and the second is exothermic. -3 The enthalpy of solution measurements were made in a commercially-available isoperibol calorimeter which had been previously modified [8] to make possible addition of measured amounts of electrical energy during the experiments. Results were obtained in this system using various calorimetric procedures in common use at high and low stirring rates. These results are compared with those obtained in a platinum-lined silver, vacuum-jacketed, adiabatic solution calorimeter [6,7,9] for which it has been demonstrated [9] that negligible corrections for heat transfer with the environment are required. |