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

The concentration of Hg in black shales is typically in the range of 10s to 100s of ppb and, in some instances, can exceed 1,000 ppb. This range is similar to that of coal but there is comparatively little information available on the occurrence of Hg in black shales. In this study we consider the concentration of Hg in black shales of the Late Devonian-Early Mississippian Bakken Formation, North Dakota (USA). The Bakken Formation is a major oil producer, but it is also representative of other North American black shales and thus can provide information on the biogeochemical cycling of Hg in black shales in general.

Mercury concentrations were determined by direct mercury analyzer for 33 samples of the Bakken Formation. Concentrations range from 31 to 395 ppb with the exception of one outlier of 792 ppb that is associated with a pyritic lag deposit. By comparing the concentration of Hg to that of other trace metals (V, Mo, Zn) we infer that most of the Hg in Bakken black shales was derived from normal seawater. There is a positive correlation between the concentration of Hg and both total organic carbon and total sulfur, thus complicating efforts to determine the association of Hg in the black shale. Mercury enrichment in the sulfide lag interval indicates that a significant fraction of Hg is associated with pyrite, consistent with work from the modern ocean demonstrating the pyritization of Hg along with a suite of transition metals (Mo, Zn, Co, and Cu). However, in our sample set, wherever Hg exceeds 200 ppb, it is largely decoupled from total sulfur, indicating a potential role for organic carbon as a host for Hg where its concentrations are highest. In some cases the fraction of Hg associated with organic carbon may exceed that associated with sulfides. Determining the concentration and occurrence of Hg in Bakken shale helps determine its origin and potential mobility during disturbance of this interval for oil and gas production.

Bituminous oil sands in northern Alberta and Saskatchewan (Canada) represent the worlds third largest reserve after Venezuela and Saudi Arabia. Rapid development of these oil sands deposits over the last three decades has raised concern about human health and environmental impacts. Alteration of the mercury cycle is of concern because local peoples have traditionally relied on hunting and fishing for food in this region and Hg consumption advisories exist for walleye in the Athabasca River near major developments and for gull and tern eggs in the Peace-Athabasca Delta, located ~150 km north of the major development area. As part of the Canada-Alberta Joint Oil Sands Monitoring program, we have been examining sources of mercury to the oil sands region, including atmospheric deposition potentially originating from bitumen upgrading facilities and fugitive dust from sources such as open pit mines, and leakage of tailings ponds, which are waste water storage dykes. To quantify atmospheric deposition to the region, we have been utilizing dated lake sediment cores and annual sampling of the springtime snowpack at sites located varying distances from the major development since 2012. Our measurements demonstrate that aerial loadings of numerous organic and inorganic contaminants, including total and methylmercury, increase with proximity to the major developments exhibiting a bulls-eye pattern on the landscape. To determine if methyl mercury is produced within snowpacks or deposited directly in that form, potential rates of methylmercury production in snowpacks and melted snow were quantified in 2015 using mercury stable isotope tracer experiments. At the four sites examined, methylation rate constants were low in snowpacks (km=0.0010.004 /d) and non-detectable in melted snow, except at one site (km=0.0007 /d), suggesting that in situ production is unlikely an important source of methyl mercury to oil sands region snowpacks. To determine the impacts of snowmelt on aquatic ecosystems, we also measured total and methyl mercury in five tributaries that have varying degrees of development on their catchments at a high frequency over the ice-free season in 2012-2014. Results to date suggest that while total mercury loads are tightly linked with hydrologic discharge and proximity to developments and bitumen deposits, methyl mercury loads increase in late summer when methylation activity is likely highest. Currently, we are coupling snowpack and river measurements and using a GIS approach to tease apart natural from anthropogenic sources to these rivers. Preliminary data from tailings ponds sampling will also be presented.

There are currently about 170 billion barrels of economically recoverable oil in the Athabasca Oil Sands Region (AOSR), ranking Canada as the world’s largest bitumen reserve and third largest oil reserve following Saudi Arabia and Venezuela. During extraction and processing, 0.5 to 4.0 barrels of sodium-rich water for each barrel of oil are produced. This sodium-rich water and slurry, referred to as tailings, are released in great volumes (1 million m^3/day) as waste into dykes referred to as tailings ponds. To date, several studies have demonstrated the presence of toxic compounds in tailings water, including volatile organic compounds, polycyclic aromatic hydrocarbons, and naphthenic acids. With no previous studies (to the best of our knowledge) examining mercury concentration in tailings ponds, four ponds were sampled for concentrations of total mercury and methylmercury. Samples were collected from surface waters, water columns, and bottom sediment samples at Horizon, the only pond at Canadian Natural Resources Ltd., and three ponds at Syncrude Canada Ltd. (Southwest In-pit, Mildred Lake Settling Basin, and Southwest Sand Storage). Total mercury concentrations in water samples were low overall, ranging from 0.16 - 0.57 ng/L. Southwest In-pit was noticeably lower than all other ponds with an average mercury concentration of 0.21 ± 0.04 ng/L, while the other ponds had concentrations >0.30 ng/L. Sediment mercury concentrations at Horizon and Southwest in pit were low at 31.4 ± 0.83 and 39.2 ± 1.88 ng/g, respectively. At MLSB, sediment samples marginally exceeded the Threshold Effect Level for freshwater set by Canadian Council of Ministers of the Environment (130 ng/g), with average concentration of 134 ± 69.1 ng/g. Methylmercury concentrations in water samples were also quite low (0.01 – 0.08 ng/L) at all ponds with the exception of Mildred Lake (0.41 ± 0.11 ng/L). Methylmercury in sediments at both Horizon and SWIP were quite low (0.15 ± 0.06 ng/L). Mildred Lake, however, had sediment methylmercury concentrations 3x higher (0.46 ± 0.04 ng/L), which is still within background levels. Due to high particle content and complex matrices, depth profile samples could only be analyzed in Southwest In-pit, which showed no evidence of changing mercury or methylmercury concentrations with depth. As a result of these analytical issues, further studies are recommended for water column analysis, primarily at Mildred Lake Settling Basin.

It is generally known that mercury is a naturally occurring element associated with oil and gas fields in the Pacific Rim region, particularly in the Gulf of Thailand. Petroleum products sourced from this region contain plenty of mercury resulting in decomposition on the metal surfaces, and finally complicated management both at offshore platform and at onshore treatment facilities. At the time of decommissioning, subsea pipeline management could be either leave in place or transport to onshore disposal facilities. Leaving in place is a preferred option due to engineering and cost attractiveness. However, one of key concerns of this option is to ensure the idle pipelines to be left in situ must be clean enough so that residuals will not cause any significant impacts to marine biota. Qualitative and quantitative analysis of the Hg contaminant were performed on both scale and bulk of the pipeline using several diagnostic techniques including energy-dispersive X-ray spectroscopy (EDS), X-ray diffraction (XRD), X-ray fluorescence spectrometry (XRF) and inductively coupled plasma spectrometry (ICP). The XRD spectra of the scale surface indicated the existing forms of mercury in the pipeline as Hg0, HgCl2, Hg2Cl2, HgS, and HgSe. Relative elemental concentration for each element of interest was visualized using EDS but limited by the low detection limits of approximately 1000 ppm. It was also found that XRF was not appropriate for Hg quantitation especially in the carbon steel matrix with high roughness and irregularity. Analytical procedure and validation was developed and established based on the ICP technique so that the Hg level within the pipeline sample could be accurately obtained. The ICP results confirmed that mercury existed only on surface and scale, and did not penetrate into bulk. Moreover, the mercury contaminant was inhomogeneously distributed over the representative scale samples. Highly corroded surface caused by pipeline leak was liable to have high level of mercury contaminated.

The Fort McMurray area in Alberta, Canada, is rich in bitumen with oil sands mining beginning in the late 1960s and expanding exponentially in recent decades. These operations have been controversial given the size and extent of the open pit mines and tailings ponds located along a ca. 75 km stretch of the Athabasca River. The Athabasca Delta and western Lake Athabasca (WLA), ca. 130 km downstream, support significant fisheries and wildlife populations. Among the many concerns expressed with oil sands activities is the enhanced release of mercury (Hg) into the environment from emissions, the disturbed landscape, seepages, etc. potentially affecting Hg concentrations in fish; an early study reported increased Hg in Walleye from the Athabasca River based on three years of sampling over 1976-2005. Subsequent expanded studies conducted under the Joint Oil Sands Monitoring Program are providing enhanced understandings of the influence of oil sands activities on Hg in the environment. Here we present highlights of these studies with a special interest in fish.

Hg emissions from the oil sands industry increased from ca 20 kg/yr in the late 1990s to reach 135 kg in 2010 with a subsequent decline; these releases were relatively small when compared to the ca. 600 kg/yr Hg release by coal-fired power plants (Wabamun Lake area) to the south. Hg deposition to the snowpack was primarily in particulate form and highest (1,000 ng/m2) close to the developments. In contrast, Hg concentrations were highest (ca. 100-200 ng/g) in lake sediments more than 50 km distant from the development and lowest (ca. 50-70 ng/g) at sites closest to the developments; concentrations were similar to lakes in the Northwest Territories ((NWT). While Hg concentrations in sediment cores have been increasing since the 1800s, peak concentrations were reached in the early 2000s and have been declining in recent years.

Hg concentrations in Walleye and Lake Whitefish from the Athabasca River and in Lake Trout, Walleye, Northern Pike and Burbot from WLA remained relatively stable and at historic levels. Hg concentrations in predatory fish tend to exceed 0.5 g/g as fish reach 10 years of age as also has been observed in small to medium size lakes in the NWT. Spatial variations in fish Hg concentration are primarily relatable to fish size, age, diet; lake size; and differences between lake and river systems. Concentrations are similar to those observed in fish from similar habitats in the NWT.

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.

Processing of gas and oil that is rich in mercury can be problematic and needs to be monitored and managed on an ongoing basis. The three main issues resulting from the presence of mercury are:
- Liquid Metal Embrittlement (LME) [a type of corrosion]
- Catalytic poisoning (an issue for downstream refining processes)
- Environmental issues (includes worker exposure and release to the environment)

For these reasons accurate quantification of mercury in produced gas and oil is essential. Ideally, determination of mercury in gases, condensates and oils should be carried out at source (at the well site or processing facility). Due to the volatile nature of mercury and its affinity to adsorb onto various sample vessels, transportation and subsequent delays between sampling and analysis is not advised.

It is possible to purchase inert coated stainless steel sampling equipment and cylinders (most commonly these are based on a thin surface silica chemistry based layer). The inert coating helps to minimise losses of mercury but it does not eliminate losses. This presentation highlights the extent of the losses that can be encountered when utilising inert coated sampling equipment and demonstrates the importance of analysing samples as soon as possible after sampling.

A theory for the mechanism of loss of mercury within an inert coated sampling cylinder is also put forward.

MERCAWAY(SM) process technology is a straightforward and commercially proven process for reducing mercury (Hg) content in crude oil and condensate using easily understood chemistry and well-proven unit operations. There are four key steps in the MERCAWAY(SM) processing scheme as follows: 1) Desanding to remove sand and other particles, 2) Reaction to convert mercury to filterable particulate HgS, 3) Filtration to separate HgS particles from treated crude/condensate stream, and 4) Polishing to adsorb most of the remaining mercury after the filtration step.

In this presentation, we will describe an application of MERCAWAY(SM) in the Oil and Gas industry. In this application, greater than 98% of the mercury was removed to produce a condensate with less than 50 ppb Hg. The process disposes the high mercury filter cake generated in this process by slurrying it with produced water and injecting back undergrounded in a closed system. This approach minimizes potential exposure to operation and maintenance personnel.

Onshore, offshore, mercury has been part of the oil and gas process for many years. With mercury levels found across the world dealing with contaminated products can be unique to every situation. In the upstream markets mercury can be higher in end of life reservoirs, dealing with contamination has changed to accommodate this. With tighter country oil and gas markets downstream and storage have also been affected to some degree and methods specifically adapted to each process, area and service stream executed. Upstream and downstream within the oil and gas market, increased contamination comes increased risk. There are safe and efficient ways in dealing with possible exposure and dealing with mercury. Hygiene and personnel safety is put at risk during maintenance and shutdown/turnaround areas, which lead to major health concerns and legalities throughout operational life. Methods, equipment and chemistry have all been specially adapted to dealing with mercury and successful processes in removing mercury from the work environment. This dialogue investigates mercury decontamination technologies and how specialised processes have worked across the industry under specific country and client standards. The specific, specialised and unique chemistry processes and equipment used for each individual case.

The operators of oil and gas fields containing mercury have to deal with a wide scope of mercury contaminated wastes, including but not limited to PPE, sludge, various liquids and mercury adsorbents. The mercury appears in various forms and the concentrations range from a few tens of ppm to several percent.

The treatment, respectively the recycling, of materials with such a wide variety of physical and chemical characteristics implies adapted technologies to decontaminate the waste itself but also to treat the off gas generated.

A special focus will be given to the technology for the recycling of Hg adsorbents (Hg guard beds, activated carbon) and the challenges linked to the recycling of this waste stream.

The mercury recovered from the wastes also represents a great challenge. Whilst some uses for high purity mercury still exist, these are restricted and therefore a safe and sustainable disposal route for the mercury is a growing demand, especially for companies with a high corporate responsibility.