2k: Mercury in the oil and gas industry, from initial exploration through production to end-of-life decommissioning

Detection of mercury in early stages of oil and gas project development allows for better incorporation of facility designs and HES procedures to manage potential risks associated with produced fluids. The ability to detect mercury during exploration and appraisal drilling activities has improved through enhancements to mercury analysis of downhole samples, and increased knowledge of the key geologic risk factors. This presentation will discuss an improvement to the analysis of downhole fluid samples for mercury content, and illustrate various limitations with conventional analysis methods. The improved analysis method includes a post-wash of samples chambers to capture mercury which is retained in sample chambers (even with so-called inert coatings) this retention can range from minimal to concentrations exceeding 1 ppm wt. of the original reservoir fluid. Under some circumstances, conventional analysis of downhole samples can significantly under report mercury concentrations. The importance of rigorous QA/QC protocols will be highlighted. Reservoir mercury results will be shared that indicate that bottom-hole temperature might correlate with mercury concentration in produced fluids, and that the equilibrium of mercury and mercury sulfide in the reservoir could be the basis for this correlation.

The determination of Hg in natural gas is a well-established procedure which is routinely used by many laboratories at gas processing plants worldwide. ISO 6978 and ASTM 6350 are the two methods that are most commonly used and these methods are based on gold amalgamation with atomic absorption/fluorescence spectrometry. These methods however were developed and originally validated for dry export gas. In this case, the sample matrix is relatively simple as the sample has undergone numerous gas processing purification steps including dehydration, acid gas treatment, removal of contaminants such as hydrogen sulphide and mercury and also separation of heavier hydrocarbons. When the methods are applied to sample point locations upstream of gas treatment numerous issues may be encountered and as such special sampling precautions have to be made to ensure that the accuracy of the method is not compromised. Data will be presented from an inter-laboratory field study highlighting the complications of determining Hg in raw untreated natural gas.

Laboratory studies were subsequently conducted using a simulated raw natural gas stream containing saturated water and BTEX components. The gas stream was spiked with a known concentration of elemental Hg using a dynamic Hg generator system. The effect of these components on the accuracy of gold amalgamation with atomic fluorescence was studied. Several types of gold adsorbents including silica coated with gold nanoparticles and gold-platinum wire were studied at different collection temperatures. This work led to the development of a new sampling arrangement which was tested in the field on wet untreated natural gas. Field data will be presented from gas plant in Egypt where upstream measurements of Hg are conducted as process monitoring on site. This accuracy of the measurement is crucially important to the oil gas and petrochemical industry because of the requirements of these sites to fully understand the fate and transport of Hg across the processing plant.

Elemental mercury is a natural contaminant of hydrocarbons streams (natural gas, gas condensates, crude oil). In particular, mercury is a safety issue in LNG plants, as mercury forms amalgams with aluminum-based alloys, leading to corrosion issues on cryogenic exchangers required to liquefy natural gas. In hydrocarbon cuts, it will poison refining catalysts, especially those containing noble metals.

In natural gas, mercury is essentially present as elemental mercury. However, in crude oil or even in gas condensates, mercury is not necessarily only under this form, especially if they are submitted to air. Indeed, it could be in combination with sulfur. For instance, some streams contain solid particles that have been identified as being HgS. In other case, they are some mercury compounds that cannot be stripped out by gaseous streams. They are considered as refractory compounds. Although their nature are not very well known, it is supposed that at least one part of them are mercury mercaptide, that is to say the result of the reaction between mercury and mercaptans.

Yet, solutions to eliminate elemental mercury are based on redox reaction between elemental mercury and sulfur or metallic sulfur. But if mercury is already at a +II oxidation state as it is the case in HgS or mercury mercaptides, these reactions cannot occur anymore. Consequently either a specific solution has to be developed to trap these others types of mercury compounds, or they have to be transform into elemental mercury before being trapped. This second option is the preferred one as mercury can be under many forms having different reactivity.

The way IFPEN has optimized solutions dedicated to the total elimination of mercury whatever its form is based on the comparison of the behavior of synthetic and real crude oil. The objective was double : verifying that our hypotheses on the nature of refractory mercury compound was reasonable and developing an efficient method to eliminate mercury from all kind of hydrocarbon streams. This implies to decompose mercury compounds and to trap elemental mercury.

Mercury is a naturally occuring contaminant found in almost all oil and gas reservoirs, existing in a range of concentrations and in a number of different forms.

This causes a number of issues when the oil is processed, including corrosion and the potential for worker exposure and release to the environment. There are a number of options for removal of mercury from liquid hydrocarbons, which are predominantly focused on removal of elemental mercury. In order to understand how much mercury will be removed by a given process the mercury needs to be speciated to ascertain which forms are present. Due to the losses of mercury observed over time when samples are stored and the proven transformation of mercury from one species to another in relatively short time periods in liquid hydrocarbons, it is desirable to perform speciation analysis immediately after sampling. This limits the use of analytical molecular speciation techniques such as GC-ICP-MS since these instrumental techniques are not portable.

The international standard test method UOP-938 allows on-site functional speciation using the physical and chemical properties of mercury species to categorise mercury as; Elemental / soluble ionic / soluble non-ionic / insoluble.

This presentation details work carried out to better understand which mercury compounds are included in the groups labelled soluble ionic and soluble non-ionic as there is very little literature information available on this subject. This study has provided information on which compounds fall into which category, which is valuable in the evaluation of whether a removal technology will in fact remove the mercury (without having to instigate a live trial), and where the mercury can be expected to partition within the hydrocarbon processing system.

Trace concentrations of mercury are found in all hydrocarbon reservoirs across the world presenting risks to personnel, process equipment, and the environment. The effects of mercury on personnel has been well researched with exposure limits in place for both inhalation and skin absorption exposure routes. The effect of mercury in oil and gas processing equipment has also been well documented due to a number catastrophic, mercury related incidents such as Skikida (Algeria) and Moomba (Australia). In both of these incidents the occurrence of Liquid Metal Embrittlement resulted in critical failure of an aluminum component within the process system. These failures led to explosions, which caused significant damage to the facilities and in the case of Skikida a significant loss of life. Learnings from these incidents have resulted in improved understanding and accepted industry standards for the removal of mercury from feed streams in both oil and gas processing facilities. To date, however no such standards have been developed for the disposal of mercury or mercury associated waste streams that are generated through scrubbing or maintenance activities at these facilities. Waste streams can include mercury contaminated PPE, piping/plant, equipment, sludges, scale and mercury guard bed adsorbents. Treatment processes for mercury waste streams are varied and requirements for disposal are largely driven by regulation. This has resulted in variation of mercury waste management practices not only between geographical regions, but also from country to country.

This session investigates mercury processing technologies and how the limited availability of dedicated mercury treatment facilities has limited standardization of mercury waste management across the industry. It also explores the suitability of the current hub and spoke model utilized by the major mercury recyclers given the requirementsfor international trans-country movements of mercury waste under the Basel Convention.

Hydroxyapatite-based adsorbents have been pointed out as high capacity adsorbents for the mercury removal from liquid or vapor phase. In this work, hydroxyapatites modified with metal sulfides were successfully employed for adsorption of trace elemental mercury (Hg0) from gaseous hydrocarbon streams such as natural gas.

Dynamics of mercury adsorption by the synthesized adsorbents was investigated in a fixed-bed adsorption process, one of most promising technologies for elemental mercury (Hg0) removal from gaseous streams. Four different adsorption tests were carried out with the following operational conditions, respectively for the T-1, T-2, T-3, and T-4 tests: initial mercury concentrations of 9.46, 8.96, 12.5, and 11.2 ng.mL-1; adsorbent metal content of 2.1, 2.1, 5.4 and 4.8 %; bed length of 1.25, 0.5, 0.5, and 1.0 cm. Other conditions are the same for all tests: temperature and diameter of reactor of 301K and 0.5 cm, respectively, and gas flow rate of 30mL.min-1.

A mathematical model were developed based on experimental information of nitrogen physisorption at 77K, X-ray diffraction (XRD), X-ray fluorescence (XRF), scanning electron microscopy (SEM), and energy dispersive X-ray spectroscopy (EDS) techniques of the raw and modified adsorbent and thermal desorption of used adsorbent samples. The proposed model includes intraparticle diffusion and chemical reaction between Hg0 and two types of metal sulfides present in the internal surface of adsorbent particle.

Model parameters such as bed porosity, particle porosity and intraparticle diffusion coefficient were calculated based on adsorbent characterization results. Kinetic and maximum capacity parameters were estimated from the experimental breakthrough curves of the two longer adsorption tests (T-1 and T-3, with time duration of 95 and 49 days, respectively). Calculated results were in good agreement with experimental breakthrough curves and estimated model parameters were statistically significant. Indeed, these parameters were validated with the experimental breakthrough curves of the two other adsorption tests (T-2 and T-4) and the similarity between experimental and calculated results corroborate the proposed phenomenology of mercury removal by hydroxyapatite-based adsorbents.

Even though it is just a metal, in industries where Mercury is a naturally occurring contaminant or even a required process component its dynamic and phenomenal cyclic characteristics are still not widely known or fully understood.

Global industry continues to drive forward with production, research, development, adaptation, mitigation, treatment and disposal of mercury, mercury compounds and mercury-contaminated products, but there are still large areas of the industrial process that have yet to be fully or properly addressed, although it could be argued that is mainly due to it being directly related to end-of-life on the process and containment infrastructure. The issue is carbon steel.

Carbon steel acts as a scavenger, capturing any mercury or mercury compounds from within process streams and retaining it, in various conditional states, along the surface of the steel, within layers of corrosion, areas of deposition and within the surface matrix of the steel substrate.

In particular, Mercury contamination can cause problems to offshore processing platforms and other downstream hydrocarbon production facilities, as well as the intermediate infrastructure. The Oil Gas industry is aware of the cycle phenomenon whereby Mercury is absorbed into the surface matrix of carbon steel pipelines during the process and transportation of contaminated hydrocarbon mediums and its subsequent ability to desorb back out from the steel surface and return to those same, or new uncontaminated, streams. However, the fate of the carbon steel pipelines, vessels and equipment falls under further limitations with lengthy, costly and typically unverifiable results until now!

As a growing concern among operators and companies around the world the decommissioning, disassembly and disposal of contaminated assets, subsea pipelines, platforms and refineries has always had its limits with current legislation, and most commercial applications, only dealing with the decontamination of free (adsorbed) Mercury from system internals prior to its decommissioning.

It is now possible to fully remove mercury from the surface matrix of carbon steel, in a repeatable manner, which ensures that the decontaminated carbon steel sections meet the strict acceptance criteria for recycling purposes using a licensed steel smelter.

Understanding the contaminant and how it behaves under certain conditions has been key to developing a selection of decontamination methods and applications that can now be applied quickly and methodically in remote locations as well as sensitive and dynamic environments while also providing flexibility and certainty to meet limited, but licensed, final waste disposal routes and best practices.

Mercury has to be removed from hydrocarbon streams due to safety issues and to protect health, environment and equipment. We are going to focus on a non-regenerative adsorbent based on the chemical reaction between mercury and the sulphur of active phase (metal sulphide) to form non-hazardous and very stable cinnabar (HgS). The limitation factor of this chemisorption mechanism is the mercury diffusion. Mercury diffusion issues lead to an increase of the axial dispersion inside the vessel which can be responsible for a premature mercury breakthrough. Therefore adsorbents have to be designed to minimize the diffusion issues.

This innovative technology consists of an active phase highly dispersed on an optimized alumina carrier. The specificity of these adsorbents is to maximize the number of active sites accessible for the chemical reaction with the mercury. Indeed, the active sites number has been maximized and the porosity design has been tailored. This alumina carrier is highly mechanically resistant and thanks to its way of preparation, the active phase is strongly linked to the alumina carrier. Different adsorbents devoted to specific gas or liquid application have been developed.

Experience in the manufacturing of mercury removal adsorbents is completed with licensing and associated technical services activities, leading to a mercury removal global offer from design to disposal. This complete integrated offer covers: mercury analyses, pilot tests, on-site tests, process selection, performances and lifetime projections, process design package delivery, skid-mounted packages delivery, adsorbent loading supervision, unit follow-up and adsorbent unloading supervision and spent adsorbent disposal logistic.

Feasibility study of the adsorbent solution and performances simulation are also required for selection of the most appropriate adsorbent for specific operating conditions. The modeling tool takes into account diffusion limitation and provides mercury adsorption profile within adsorbent bed and mercury breakthrough curves versus time. This model fits with pilot tests and industrial units results. On-site tests are also very useful to confirm this selection.

In this presentation the mercury removal performances of the latest adsorbent developed with this technology, AxTrap 283, will be discussed. Then, based on case studies, the interest of the modeling tool and of on-site tests to predict the mercury adsorption profile within adsorbent bed will be highlighted.