| Abstract
| - Over the past few years, concerns have been raised in theliterature about the accuracy of the Karl Fischer (KF)method for assessing moisture in transformer mineraloils. To better understand this issue, the performance ofa static headspace capillary gas chromatographic (HS-CGC) technique was compared to that of KF coulometrictitration by analyzing moisture in samples containingknown amounts of water and various samples obtainedfrom the National Institute of Standards and Technology(NIST). Two modes of adding samples into the KF vesselwere used: direct injection and indirect injection via anazeotropic distillation of the moisture with toluene. Underthe conditions used for direct injection, the oil matrix wastotally dissolved in the anolyte, which allowed the moisture to be titrated in a single-phase solution rather thanin a suspension. The results have shown that when HS-CGC and combined azeotropic distillation/KF titration arecalibrated with moisture-in-oil standards, a linear relationis observed over 0−60 ppm H2O with a correlationcoefficient better than 0.9994 (95% confidence), with theregression line crossing through zero. A similar relationcan also be observed when calibration is achieved bydirect KF addition of standards prepared with octanol-1,but in this case an intercept of 4−5 ppm is noted. Theamount of moisture determined by curve interpolation inNIST reference materials by the three calibrated systemsranges from 13.0 to 14.8 ppm for RM 8506 and 42.5 to46.4 ppm for RM 8507, and in any case, the results wereas high as those reported in the literature with volumetricKF titration. However, titration of various dehydrated oiland solvent samples showed that direct KF titration isaffected by a small bias when samples contain very littlemoisture. The source of error after correction for the largesample volume used for the determination (8 mL) is about6 ppm for Voltesso naphthenic oil and 4 ppm for toluene,revealing a matrix effect on the measurement. Finally, theresults revealed that HS-CGC is a good technique formeasuring moisture in oil samples and that the use ofazeotropic vapors for introducing moisture into the titratoralmost completely eliminates the matrix effect observedwith the oil components. Direct KF injection could alsobe used provided the system is calibrated with moisture-in-oil standards prepared in the same matrix which is tobe used for the determination.
|
