Mercury is a highly toxic pollutant that occurs naturally in soil and rock throughout the earth’s crust, including the formations that comprise oil and gas reservoirs. The growth of industrialization in the past century has produced rapid increases in anthropogenic mercury emissions, originating from processes such as coal burning, mining, smelting, and more. Mercury exists in several forms: elemental (metallic) mercury, inorganic mercury compounds, and organic mercury compounds, whereby the level of toxicity varies with the form of mercury [1]. As the only common metal which exists in liquid form at ambient conditions, mercury and its compounds behave exceptionally in the environment compared to most of the other heavy metals. Mercury exhibits unique properties including high volatility and a capacity for methylation, leading to the formation of considerably more toxic methylmercury, which can be accumulated in the aquatic food chain and in living organisms.
In addition to environmental and health concerns, the presence of mercury in many industrial processes poses serious corrosion risks through liquid-metal embrittlement (LME) with metals, as well as equipment fouling risks due to the precipitation of insoluble salts [1]. As a naturally occurring element, mercury is present in virtually all oil and gas reservoirs, largely as elemental mercury or inorganic compounds (e.g. mercury sulphide and mercury chloride). The concentration of mercury in crude oil and natural gas is highly dependent on geologic location and varies between approximately 0.01 ppb and 10 ppm (mass). Mercury species from crude oil may accumulate in processing equipment over time, causing operational concerns and posing critical exposure hazards for maintenance personnel. In refining processes, mercury can present serious risks through amalgamation with other metals, the poisoning of catalysts, and cracking (LME) with metals such as aluminum, which is the primary alloy component in many LNG & LPG cryogenic heat exchangers [1,2].
The development of technologies to mitigate risks from mercury requires an accurate prediction of the possible persistence of mercury in oil and gas processing. This capability rests on a foundation of complex mercury chemistry. OLI’s comprehensive database is equipped with an extensive library of species and relevant parameters to model the solubility, volatility, and reactivity of this critical chemistry.
Leveraging OLI to predict mercury removal from oil & gas streams
OLI has developed advanced technologies based on its proprietary thermodynamic model, the Mixed-Solvent Electrolyte (MSE) model [3]. The model provides an accurate prediction of mercury partitioning in vapour, aqueous, and oil phases, enabling the identification of optimal removal conditions through process simulation.
The most prevalent form of mercury found in oil and gas reservoirs and in refineries is elemental mercury, which can react to form mercuric sulphide (HgS) and other forms of inorganic mercury during production and processing.
The MSE database provides solubility predictions for:
- Elemental mercury (Hg0) in key components of oil & gas process streams, including
- water
- hydrocarbons
- alcohols (e.g., methanol, isopropanol, ethylene glycol)
- amines (e.g., monoethanolamine (MEA) and methyldiethanolamine (MDEA))
- CO2
- mixtures of the above
- Inorganic mercury compounds such as HgCl2, HgS, HgCO3, HgO, and HgF2 in aqueous solutions
OLI Version 12.5 featured significant mercury-containing component additions to the MSE database, including:
- Dimethylmercury
- Mercuric nitrate (Hg(NO3)2)
- Mercuric nitrate mercuric oxide (Hg(NO3)2.2HgO)
- Mercuric nitrate monohydrate (Hg(NO3)2.H2O)
- Chloro(methyl)mercury
- Dimercury(I) formate
- Dimercury(I) glycolate
- Dimercury(I) oxalate
- Dimercury(I) oxalate monohydrate
- Mercury(II) formate
- Mercury(II) glycolate
- Mercury(II) oxalate
- Methylmercury hydroxide
- Methylmercury nitrate
How reliable is OLI technology for predicting mercury properties?
The OLI model for mercury removal predictions has been developed based on detailed analysis of the chemistry and systematic evaluation of experimental mercury solubilities in key components of process streams. The model predicts accurate results of the solubility of elemental mercury and its partitioning between aqueous, liquid-hydrocarbon, and vapour phases. The effects of temperature, pressure, and composition of mixtures on the mercury distribution can also be predicted.
OLI’s first-principles predictions are grounded in real data. Figures 1-4 include direct comparisons of experimental and OLI-predicted data for key mercury compounds and their critical properties. Among many other data points, this experimental data was employed in regressing thermodynamic and interaction parameters of the corresponding species in the MSE database.
For example, shown in Figure 1 is the solubility of Hg0 in water as a function of temperature and pressure. Figures 2 and 3 illustrate OLI predictions of mercury solubility in vapor and liquid C4H10, respectively. Figure 4 demonstrates the accuracy of the model in predicting the solubility of HgS in aqueous solutions. Overall, the predicted results are in good agreement with literature data.
Figure 1. Solubility of elemental mercury in water as a function of temperature and pressure.
Figure 2. Solubility of Hg0 in liquid C4H10.
Figure 3. Solubility of Hg0 in gaseous C4H10.
Figure 4. Solubility of HgS in aqueous solutions as a function of temperature, pressure, and Na2S concentrations.
OLI has developed unique sets of thermodynamic model parameters for each group of solvents. These parameters are established based on available experimental data in each solvent group, and enable the prediction of solubilities in a solvent for which experimental data are not available. Solubilities of mercury have been compared for solvents from the same group (e.g. alkanes with different carbon numbers) and among different groups of solvents (e.g. alkanes, alcohols, and water) with general regularities that have been well reproduced [4]. Particularly, the MSE model with such parameters can be applied when pseudo-components are introduced in process simulations. Illustrated in Figure 5 are the Hg0 solubility results in n-alkanes as a function of carbon number at different temperatures.
Figure 5. Hg0 solubility in liquid hydrocarbons as a function of carbon number, at different temperatures.
The model can predict mercury distribution in mixtures of hydrocarbons and other components, which is critical for effective mercury remediation. Figure 6a demonstrates a stream configuration in OLI Studio: Stream Analyzer for a sample gas mixture containing primarily CH4 with other hydrocarbon components, water, CO2, N2, and ~ 1 ppm Hg. By running a temperature survey, engineers can identify the required specifications for mercury remediation by evaluating the predicted Hg0 phase distribution and dropping out condition, plotted in Figure 6b.
(a)
(b)
Figure 6. (a) Example mercury-containing hydrocarbon stream in OLI Studio: Stream Analyzer. (b) Predicted Hg0 distribution among solid, vapor, aqueous (Liquid-1), and liquid hydrocarbon (Liquid-2) phases for the representative system, calculated in OLI Studio: Stream Analyzer.
Conclusion
OLI software is equipped with a first-principles thermodynamic framework, validated by experimental data, to predict complex mercury distribution, methylation, and salt precipitation. Together, this creates a chemistry engine that helps oil & gas teams predict and remediate mercury persistence across industrial process streams, enabling safer working environments and equipment reliability.
Contact OLI for more information or to schedule a meeting with an OLI expert.
References
- S.M. Wilhelm and D. A. Kirchgessner, “Mercury in Petroleum and Natural Gas–estimation of Emissions from Production, Processing, and Combustion”, United States Environmental Protection Agency, National Risk Management Research Laboratory, 2001.
- R. Coade and D. Coldham, “The interaction of mercury and aluminium in heat exchangers in a natural gas plants”, International journal of pressure vessels and piping 83.5 (2006): 336-342.
- P. Wang, A. Anderko, and R. D. Young, “A speciation-based model for mixed-solvent electrolyte systems”, Fluid Phase Equilibria 203.1-2 (2002): 141-176.
- P. Wang and A. Anderko, “Modelling Solubility and Solution Chemistry of Mercury in Water, Alcohols, and Hydrocarbons”, ISSP-18, July 15-20, 2018.