Early and Accurate Quantification of Mercury Contaminant Levels in Gas-Condensate Reservoirs

Abstract

AbstractMercury is a trace contaminant present in natural gas and condensates. Production facilities must be designed to remove anticipated mercury contamination in natural gas streams to meet process safety, health, environmental, regulatory and sales requirements. Early and accurate quantification of mercury concentration levels is therefore critical for the safe and economic development of gas resources. This paper introduces reservoir fluid sampling and mercury analysis techniques demonstrated to accurately determine the mercury concentrations in gas-condensate reservoirs.The mercury analysis of reservoir fluid samples acquired with formation sampling tools often misrepresent mercury concentration levels measured later during production. This is due to the highly volatile and reactive nature of mercury, particularly with uncoated metallic surfaces. Laboratory experiments were conducted to develop procedures for accurate and repeatable analysis of mercury concentrations in gas sampling systems. Methane was contaminated with precise concentrations of mercury to facilitate a range of experiments examining both the contamination and scavenging behaviour of mercury in sample chambers. Uncontaminated methane was used as a baseline for background testing sample chambers and laboratory equipment.The experiments identified the presence of residual mercury contamination in sample chambers and laboratory equipment previously exposed to mercury. Cleaning procedures involving the thermal desorption of residual mercury were developed and proven to ensure sampling chambers and laboratory equipment were free of any mercury contamination. Without applying cleaning procedures, any residual mercury would contribute to incorrectly overpredicting the mercury concentrations of reservoir fluids captured in the sample chambers. Subsequent experiments showed that even inert coatings do not completely prevent mercury adsorption to internal metallic chamber walls. Heating the sample chambers to 140 °C was found to release most of the adsorbed mercury back into the gas phase, with a wash procedure required to recover all the remaining mercury and avoid incorrectly underpredicting the actual mercury concentration levels due to scavenging. By implementing a combination of procedural changes, sampling tool modifications and improved measurement techniques, it is possible to acquire reservoir fluid samples with formation sampling tools and accurately measure mercury concentrations representative of that produced from the reservoir.The case study of a gas-condensate field is used to demonstrate that by following the novel techniques introduced in this paper, reservoir fluid samples acquired from wells using formation sampling tools and analysed for mercury can accurately represent the mercury concentration levels measured later during the production phase. Processing facilities can then be appropriately designed or modified early in the field development lifecycle to remove any anticipated mercury in natural gas process streams.