| Abstract |
Diisodecyl phthalate (DIDP) has recently been proposed to be an industrial reference fluid for moderately high viscosity at atmospheric pressure. Moreover, some results of the viscosity and also of the density of DIDP at high pressures have been reported recently. The main aim of the present work is to contribute with necessary data to enable the eventual proposal of DIDP to be a reference fluid for viscosity at high pressures. In this work, we describe measurements of the density of DIDP at 12 temperatures from (283.15 to 363.15) K and at pressures from (0.1 to 65) MPa. The measurements were performed using an Anton Paar GmbH model DMA HP vibrating tube densimeter with a reading device (model DMA 5000 unit). Prior to the DIDP measurements, the apparatus was calibrated with water and toluene. The sample of DIDP was obtained from Merck KGaA, with a minimum purity of 99.8 \% by gas chromatography (GC). The results of the density of DIDP were correlated by a modified Tait equation. The relative root-mean-square deviation (rmsd) of the reference data from the correlation is +/- 0.005 \% To with practically zero bias. The effects of the viscosity of the sample on the density measurements are discussed and taken into account in the estimated combined uncertainty of the present results. However, no corrections of the present measurements have been performed. The relative uncertainty of the present measurements is estimated to be less than 0.3 \% at a 2 sigma level. This estimate is partly based on the tests performed with dichloromethane at temperatures within the range (283.15 to 333.15) K and pressures from (0.1 to 65) MPa. The present results are compared with literature density data of DIDP samples with purity equal to or greater than 99.8 \% (GC), performed at either atmospheric pressure or pressures higher than 0.1 MPa. The literature data sets comprise measurements performed using vibrating tube instruments and pycnometry and results calculated from speed of sound measurements. The literature data were found to be, generally, in agreement with the present results, within their nominal relative uncertainty, except for one of the vibrating tube data sets that has a maximum relative deviation (-0.4 \%), which is slightly larger than its claimed relative uncertainty. |
