Experimental (p, Vm, T) for pure CO2 between 220 and 450 K

“…State-of-the-art PVT measurements can now provide second virial coefficients B ( T ) to t 0.2 cm3/mo1a.2% if B = -100 cm3/mol (Holste et al, 1987). Adsorption errors associated with a highly polar component, whether pure or in a mixture, cause uncertainty in B( T ) to rise by one or two orders of magnitude even after correction of the data for adsorption errors .…”

Section: State-of-the-art In Pvt Measurementsmentioning

confidence: 98%

Optimum mixture compositions for measurement of cross virial coefficients

Eubank

Hall

1990

AIChE Journal

View full text Add to dashboard Cite

Random error propagation is examined to determine cross second and third virial coefficients from binary gas density measurements by both the indirect and direct methods. Determination of the cross second virial coefficient B,, from Bm at a single composition via the indirect method is optimized when y:el = y:e2, where e, and e, are the pure component uncertainties. When all errors are equal, the equimolar mixture is optimum as expected.Corresponding analysis of the cross third virial coefficients, C1,2 and C, , , ,show that the optimum compositions are not (V3, V3) as expected for equal error distribution, but rather (%, 3/4). Equations presented allow determination of the two optimum compositions, when em varies with composition. Minimization of the combined errors in B,,, C, , , , and C,22via mixture measurements at two compositions provides results near (V3, V3), when all errors are equal. Methods for direct measurement of B,, are reexamined. Optimal compositions are found using third virial coefficient correlations.

show abstract

“…Again, FIGURE 2. Relative deviations drn of the results of measurements of the amount-of-substance density rn of CO2 from the equation of state of Span and Wagner (2) plotted against pressure p for temperatures from 233 K to 523 K. w, Present work; W, Brachthauser et al; (4) r, Holste et al (5) it is possible to discern some systematic trends but they are always within the estimated experimental error. It is therefore possible to conclude that the measurements reported here are consistent with earlier results in the overlapping range of temperature.…”

Section: Resultsmentioning

confidence: 90%

“…It can be seen that while there is a tendency for the experimental density to be larger than that predicted by the equation of state, the difference is always less than 20.1 per cent. Figure 2 contains the relative deviations of earlier measurements of the amount-of-substance density of Brachthauser et al (4) and Holste et al, (5) that extend up to temperatures of 523 K, from the equation of state of Span and Wagner. Again, FIGURE 2.…”

Section: Resultsmentioning

confidence: 95%