Improved Methods for Gas Mixture Viscometry Using a Vibrating Wire Clamped at Both Ends

• Published in: Journal of chemical and engineering data Vol. 59; no. 5; pp. 1619 - 1628
• Main Authors: Locke, Clayton R., Stanwix, Paul L., Hughes, Thomas J., Kisselev, Austin, Goodwin, Anthony R. H., Marsh, Kenneth N., May, Eric F.
• Format: Journal Article
• Language: English
• Published: American Chemical Society 08.05.2014
• ISSN: 0021-9568; 1520-5134
• DOI: 10.1021/je500083n

We present a clamped vibrating-wire instrument and the associated methods of measurement and analysis that enabled gas-mixture viscosity measurements at densities up to 110 kg·m–3 with a standard uncertainty of 0.09 μPa·s, which is a relative uncertainty of 0.60 %. The vibrating-wire was clamped at both ends and operated in the steady-state mode to make the apparatus more compact and allow operation over a broad range of conditions. New modifications to the method include an interleaved measurement protocol to minimize errors arising from fluctuations in temperature and pressure, and optimization of the signal-to-noise while ensuring that the driven wire’s response remained in the linear regime. The wire’s radius was determined from calibration measurements with He, and the viscometer’s performance was verified with N2, CO2, and CH4. The discrepancies between the measured pure fluid viscosities and those predicted with models implemented in the software REFPROP 9.1 were smaller than 1 %; literature data for these fluids exhibit similar deviations. Viscosities of (1 – x)­C3H8 + xCH4 with x = 0.9452, and (1 – x)­CO2 + xCH4 with x = 0.57 were also determined at pressures between (1.5 and 6.5) MPa and temperatures of (280, 303 and 328) K. The largest rms deviation of 3.6 % of the measured viscosities relative to those calculated with the extended corresponding states model implemented in REFPROP occurred for CH4 + CO2 at a temperature of 328 K.