The PYT properties of liquid n-octane are presented for six isotherms: -15.00, 5.00, 25.00, 45.00, 65.00, and 85.00OC. The pressure ranged from atmospheric to near freezing at each temperature. The accuracy is estimated at f0.0006 cm3/cm3. The data were fitted to the Tait equation of state at each temperature. The data were also used to obtain the Tait parameters L and J as functions of temperature. The error introduced by the temperature fitting averaged 0.041 O/O. The worst error introduced was 0.001 7 cm3/cm3.T h e PVT data for liquid n-octane, as reported in the literature, are inconsistent. Furthermore, the more accurate data do not extend over a large enough range of pressure and temperature to allow the differentiation necessary to calculate accurately thermal properties.Previous data taken in this laboratory on Clo to C16 normal alkanes showed the sylphon-bellows technique sufficiently accurate to be used for calculating properties requiring two differentiations of the data (14). In this work, n-octane is examined from -15" to 85OC a t pressures from atmospheric to near freezing.' To whom correspondence should be addressed. EXPERIMENTALApparatus. The operation of the PVT cell can be described with the aid of Figure 1. The two main components of the PVT cell are the bellows, marked B on the drawing, and the slide wire, S. The slide wire is a section of Karma wire approximately 1 inch in length and 0.010 inch in diameter. Karma is a trademark of the Driver-Harris Co., Harrison, N. J., and is an alloy of nickel, chromium, and aluminum. Karma wire was used as the slide wire because of its low temperature coefficient of resistivity (0 f 10 ppm/" C), its precision drawn diameter, and its highly uniform linear conductivity (experimentally determined as

No abstract

No abstract

No abstract

A detailed study has been made of the computing times for several common equations of state used by industry. The cubic equations are approximately equivalent to more complicated equations (e.g., perturbed hard chain and BWRS) for systems with eight or more components. Thus, it is not justified to choose cubic equations over other forms purely on the basis of computational advantage. Further, the study identifies mixing-rule summations as the dominant consumer of computing time.