The determination of water in crude oil and transformer oil reference materials

Abstract

The measurement of the amount of water in oils is of significant economic importance to the industrial community, particularly to the electric power and crude oil industries. The amount of water in transformer oils is critical to their normal function and the amount of water in crude oils affects the cost of the crude oil at the well head, the pipeline, and the refinery. Water in oil Certified Reference Materials (CRM) are essential for the accurate calibration of instruments that are used by these industries. Three NIST Standard Reference Materials (SRMs) have been prepared for this purpose. The water in these oils has been measured by both coulometric and volumetric Karl Fischer methods. The compounds (such as sulfur compounds) that interfere with the Karl Fischer reaction (interfering substances) and inflate the values for water by also reacting with iodine have been measured coulometrically. The measured water content of Reference Material (RM) 8506a Transformer Oil is 12.1+/-1.9 mg kg(-1) (plus an additional 6.2+/-0.9 mg kg(-1) of interfering substances). The measured water content of SRM 2722 Sweet Crude Oil, is 99+/-6 mg kg(-1) (plus an additional 5+/-2 mg kg(-1) of interfering substances). The measured water content of SRM 2721 Sour Crude Oil, is 134+/-18 mg kg(-1) plus an additional 807+/-43 mg kg(-1) of interfering substances. Interlaboratory studies conducted with these oil samples (using SRM 2890, water saturated 1-octanol, as a calibrant) are reported. Some of the possible sources of bias in these measurements were identified, These include: improperly calibrated instruments, inability to measure the calibrant accurately, Karl Fischer reagent selection, and volatilization of the interfering substances in SRM 2721.

A Visual Measurement of Water Content of Crude Oil Based on Image Grayscale Accumulated Value Difference

“…Crude oil moisture content is a vital parameter in the petroleum exploitation and petrochemical industries, and it has significant reference value to crude oil processing, storage, transportation, and pricing. By measuring the water content of crude oil, the position of reservoir can be estimated in time [1,2,3]. For measurement of water in crude oil, there are two major categories which are online and offline [4].…”

Determination of Trace Water Content in Petroleum and Petroleum Products

“…The measurement of water is necessary for raw materials (various crude oils), intermediate refinery products, and the wide variety of finished products. Water levels affect the economics of transport, storage, refining, formulating, safety, and the proper performance of end products. − Because of the difficulty in obtaining accurate measurements in the variety of different hydrophobic and complex matrixes, there is a need for accurate reproducible standards. Thus, it is not surprising that the National Institute of Standards and Technology (NIST) has produced different reference materials to encompass different types of petroleum and petroleum products.…”

“…Thus, it is not surprising that the National Institute of Standards and Technology (NIST) has produced different reference materials to encompass different types of petroleum and petroleum products. Indeed, there has been some controversy and debate over the procedures used to produce some of the reference values even within this journal. ,− …”

“…In one case, the certified water levels listed on the Certificate of Analysis differ by almost an order of magnitude depending on whether the NIST or American Society for Testing and Materials (ASTM) method was used. ,, This was believed to be due to the fact that significant amounts of interfering compounds to Karl Fisher titration reagents were present in this standard (light sour crude oil) and that additional experimentation was done to minimize these effects with the NIST method but not with the ASTM method. ,, That notwithstanding, the NIST value still has a deviation of greater than ±13%. In another NIST reference material (transformer oil), four different reference values are provided, one of which is the consensus result from a 14-laboratory study from which some results were excluded. , The consensus result (21.2 ± 1.7 mg/kg water) differed from the NIST result minus interferences (12.1 ± 1.9 mg/kg water) and from the NIST volumetric/ASTM method (34.5 ± 2.2 mg/kg water). , The point here is not to cast dispersions on such standards, which we believe are the best available and were produced with high integrity by the best available methods and in which the variations and their possible causes are clearly indicated. − Rather, any problems are with the limited methodologies that are available to measure water in these (and other) samples.…”

“…In addition to having low concentrations of water, sample insolubility, reactive impurities, and other matrix effects, there are inaccuracies due to the extensive sample handling procedures needed as well as other instrumental effects (Table ). ,,− ,,,− Indeed, the dominant KFT response in some crude oils is from other compounds in the sample. , Thus, different dual titrations are needed and used to subtract the unwanted from the wanted responses. , Other KFT alternatives include removing the water from the sample (e.g., stripping oven KFT or by azeotropic distillation). , In either case, contamination by atmospheric moisture or the residual water in the azeotropic distillation solvent further complicates these procedures and greatly extends analysis time as well as hinders automation. , …”