(632e) Mercury Traces Removal From Natural Gas: Optimization of Guard Bed Adsorption Properties

Mercury (Hg) is a natural contaminant found in the earth’s crust where its concentration can range from 10 to 20 000 ppb. Mercury is thus released in the environment from a variety of natural sources, including volcanic, geothermal activities or wildfires; but also from anthropogenic activities as a total of about 2000 tons of mercury is estimated to be released each year from fossil fuel combustion or metal production [1]. For instance, natural gas production frequently generates hydrocarbons streams containing traces level of mercury [2], especially in the Southeast region of Asia where Hg concentration can reach up to 300 ppb. In these conditions, mercury is predominantly found as elemental mercury (Hg°) in the gas phase. Though these levels are rather low, the impact on the industrial equipment and human health can be serious. For instance, mercury has a strong ability to form amalgams with Al-based LNG (Liquefied Natural Gas) cryogenic exchanger leading to corrosion issues and potential industrial disaster like the one encountered in 1973 in Skikda, Algeria [3]. In addition, mercury is harmful for human health and numerous incentives have been issued to control and limit its emissions from anthropogenic sources.

Hence, mercury removal is an ongoing issue and natural gas streams are usually decontaminated using guard beds protecting downstream equipment. Traditionally, these guards beds are made of beads of activated carbon impregnated with elemental sulfur (S) [4]. Mercury will chemically bond with sulfur to form mineral cinnabar (HgS) within the porosity of the activated carbon. Mercury is thus immobilized in a non-hazardous form and guard beds are designed to decrease traces level of Hg down to 1 ppb for several years before being disposed and retreated.  

However, these types of guard beds suffer from many drawbacks such as possible sulfur loss during mercury removal operation and are prone to capillary condensation for wet gas [4]. One way to avoid theses problems is to optimize the interaction between the carrier support and the active phase deposited within the solid. For instance, pore size distribution can be tuned to avoid capillary condensation issues and the nature of the active phase and mineral support can be adequately chosen to obtain strongly bound and finely dispersed mercury-reactive nanoparticles. 

In this work, mercury adsorbents are prepared by dispersing copper sulfide (CuS) within the porosity of alumina carrier supports (Al₂O₃). In practice, the solids are prepared by wetness impregnation method with a solution of copper precursor, following by thermal treatment leading to supported copper oxide. The finely dispersed copper sulfide nanoparticles phase is finally obtained by a sulfidation procedure before mercury removal evaluation. We investigate the effect of copper oxide dispersion by varying the nature and/or the textural properties of the solid support. Scanning electron microscopy and X-ray diffraction experiments showed that the deposited copper oxide particles are more efficiently dispersed on the alumina surface for highly porous volumes and specific mineral phases. Mercury removal performances are then assessed through pilot plant experiments where a feed of elemental mercury is passed through a fixed bed of adsorbents. These experiments are finally interpreted using a dedicated mass transfer model and mercury removal mechanisms are studied through numerical simulations.

[1] UNEP Chemicals branch, 2008. The global atmospheric mercury assessment: sources, emissions and transport. UNEP-Chemicals, Geneva.

[2] U.S. Environmental protection agency, 2001. Mercury in petroleum and natural gas: estimation of emissions from production, processing and combustion. EPA-600/R-01-066.

[3] Kinney, G. T. Skikda LNG plant solving troubles. Oil & Gas J. Sept, 15, 1975.

[4] McNamara, J.D. and Wagner, M.J. Process effects on activated carbon performance and analytical methods used for low levels mercury removal in natural gas applications. Gas. Sep. Purif., 1996, 10, pp. 137-140.