4c: Anthropogenic emissions: Monitoring and analysis to support mitigation

Atmospheric mercury (Hg) has both natural and anthropogenic sources. The latter include artisanal gold mining, coal combustion, chlor-alkali industry, production of cement and certain metals, crematoria, waste disposal and recycling (especially Hg containing products), and production of Hg containing devices (batteries, fluorescent light bulbs). While many urban sources have been remediated or relocated outside of cities in developed countries, legacy sources may remain and poorly identified, continuing sources may exist. We deployed our recently developed passive air sampler for gaseous Hg at a total of 161 sites across the Greater Toronto Area (GTA), with a greater density of sampling sites in the City of Toronto itself. Prior to sampling, we identified likely source locations grouped as either waste disposal and recycling facilities, crematoria, or hospitals and dental clinics. The deployment included 103 additional sites at the homes of graduate students, staff, and faculty of the University of Toronto Scarborough, and other locations where the relative spatial density of deployments was low. Additionally, a downwind transect of samplers was deployed at a regional Hg disposal and recovery facility located near Kitchener, Ontario. Mean gaseous Hg concentration near waste disposal and recycling facilities (1.9 ± 0.3 ng m-3), crematoria (1.8 ± 0.2 ng m-3), and hospitals and dental clinics (1.9 ± 0.3 ng m-3) sites are all slightly, but significantly (p < 0.05), higher than at the remaining sites (1.6 ± 0.2 ng m-3). The three potential source groupings are not significantly different from each other (p > 0.05). Geospatial interpolation of the data reveals slightly elevated concentrations in the central business district of the City of Toronto, an area that also includes six hospitals, and a general decline of gaseous mercury concentrations toward the eastern Greater Toronto Area. Gaseous Hg concentrations decrease along a transect of increasing distance from the Hg disposal and recovery facility near Kitchener, ranging from 13.7 (~10 m SW of the central building) to 1.5 ng m-3 (2.2 km SSE of the site). This study demonstrates that passive sampling using this method can be sufficiently precise to spatially distinguish small (<0.2 ng m-3) gaseous Hg concentration differences at and around global background levels. Furthermore, the sampler can be used as a valuable tool for identifying point sources in urban areas that may previously have been described as diffuse source regions.

The objectives of this study were to: (1) characterize the hourly and seasonal variations of atmospheric total gaseous mercury (TGM) concentrations, (2) identify the relationships between TGM and co-pollutants concentrations, (3) characterize high TGM concentration events by distinguishing between long-range transport events and local events, and (4) identify likely source directions and locations of TGM using conditional probability function (CPF), conditional bivariate probability function (CBPF) and potential source contribution function (PSCF) for Gwangyang City, Jeollanam-do, a province in southern South Korea. Gwangyang City is heavily industrialized with large steel manufacturing facility and there are iron and steel manufacturing facilities including electric and sintering furnaces using coking around the sampling site. In addition, there is Yeosu City which includes a large petrochemical complex from the west of the sampling site.

TGM concentrations were measured every 5 min during spring (11 May-6 June 2016), and fall (2-10 November 2016) on the roof of the public bath house using a Tekran 2537B. Hourly meteorological data and concentrations of SO2, NO2, O3, CO, PM10 and PM2.5 were obtained from the Korea Meteorological Administration (KMA) and the National Air Quality Monitoring Network (NAQMN), respectively.

The TGM concentration was statistically significantly higher in fall (1.4 ± 1.2 ng m-3) than in spring (0.8 ± 0.6 ng m-3) (p<0.01). The TGM concentrations in spring were simultaneously positively correlated with SO2 (r = 0.31, p<0.01), NO2 (r = 0.22, p<0.01), PM10 (r = 0.31, p<0.01) and PM2.5 (r = 0.23, p<0.01). Compared to spring, the TGM concentrations in fall have strong positive correlation with CO (r = 0.94, p<0.01), NO2 (r = 0.53, p<0.01), PM10 (r = 0.71, p<0.01) and PM2.5 (r = 0.26, p<0.01) due to an extreme pollution episode with very high TGM concentrations.

A total of 3 high TGM concentration events were identified based on ΔTGM/ΔCO slopes and correlations between TGM and CO concentrations; 2 long-range transport events and 1 local event. Backward trajectory analysis showed that air parcels arrived from China during long-range transport events.

CPF, CBPF and PSCF plots indicated that the main sources of TGM were west of the sample site and included iron and steel manufacturing facilities and the east of the site including cement production facilities. Southeast of the sampling site was also identified as a likely source area of TGM where Yeosu City, the location of a large petrochemical complex and known TGM source.

The study of mercury (Hg) in its different chemical forms in the environment has developed greatly in the US, Canada, Japan and other countries. However in Mxico, few studies have been performed in urban, rural coastal and mining areas. The objective of this study was to evaluate the concentration of mercury (Hg) in particulate matter (PM10) in a mining zone the 2014 to 2015 years. Therefore, the people who live around this mining region are the most affected because they are exposed to the mercury in a chronicle way which is also in high concentrations. Chronic exposure to this metal can get to affect human health causing severe damage to the immune system. Estimating the atmospheric concentrations of PM10-bounded Hg in mining areas is crucial for evaluating adverse health impacts. In the current study, a combination of measurements and multivariate statistical tools was used to investigate the influence of mines activities on variations in the concentrations of particulate mercury (HgP) in ambient air. San Joaqun is a mining region south of the Sierra Gorda of Queretaro, Mexico. The concentrations of Hg in PM10 were measured simultaneously at two air sites. The comparatively low coefficients of divergence (COD) that were found in the majority of samples highlight that site-specific effects are of minor importance. A principal component analysis (PCA) revealed that 37.74 %, 13.51 %, and 11.32 % of the total variances represent mines emissions, vehicular exhausts. This analysis allows determining the extent of contamination by mercury within each environmental component, as well as the detection of its maximum and minimum thresholds. For a better understanding of mercury distribution in each component, box and whiskers diagrams were used through STATISTICA Version 10 software. Emphasis is given to the associations between Hg concentrations and local meteorological conditions.

Mercury is a global pollutant with serious harmful impacts on human and ecosystem health. It is emitted into the atmosphere from many sources such as fossil fuel combustion, incineration, and landfill. Some sources/processes are either poorly documented or unknown. Cremation processes are one of these that have not been studied and are currently unaccounted for in the United States Environmental Protection Agency’s National Emissions Inventory. The objective of this study was to characterize the temporal variation of total gaseous mercury (TGM) concentrations and identify an unknown, seemingly highly localized source of gaseous mercury causing very high episodic concentrations. TGM measurements were conducted from the State University of New York College of Environmental Science and Forestry (SUNY-ESF) campus in downtown Syracuse, New York during the time period of summer 2013 – fall 2015. A complete annual cycle was observed, with lowest concentrations (1.36 ng m^-3) in September, and highest (1.57 ng m^-3) in January, with an annual average amplitude of 0.21 ng m^-3. Concentrations appeared to be decreasing continuously throughout the study period, with decreases of 0.12 ng m^-3 and 0.18 ng m^-3 for summer 2013-2014 and 2014-2015, respectively; 0.14 ng m^-3 and 0.05 ng m^-3 for fall 2013-2014 and 2014-2015, respectively; and 0.08 ng m^-3 for winter 2014-2015. Diurnal cycles were observed with daily maximums at 13:00-16:00 UTC (1.55 ng m^-3 - 1.65 ng m^-3) in winter-spring, 1:00 UTC (1.4 ng m^-3 – 1.7 ng m^-3) and 12:00-16:00 UTC (1.3 ng m^-3 – 1.52 ng m^-3) in summer-fall. The concentrations above the seasonal 99th percentile values under calm (< 2 m s^-1) and southeasterly wind conditions were associated with probable Hg emissions from a nearby crematorium, located approximately 890 meters to the southeast of the monitoring station. The total emission of mercury from this source was estimated to be 0.51 lbs, 1.64 lbs, and 0.49 lbs for 2013, 2014, and 2015, respectively. These were compared to nearby sources documented in the EPA’s National Emissions Inventory 2011, including the Syracuse Steam Station (0.3 lbs), Onondaga County Resource Recovery Facility (7.7 lbs) and Bristol-Myer Squibb Company (9.52E-02 lbs). Further study is warranted to determine the extent of mercury emissions from crematoriums across the United States.

Wastewater entering municipal wastewater treatment plants (WWTPs) contains mercury (Hg) from domestic, commercial, and industrial sources. During treatment, almost all of that Hg (>98%) ends up in the sludge residual stream, and Hg concentrations in the treated wastewater effluent are typically quite low (<10 ng/L). Many larger WWTPs in metropolitan areas incinerate their sludge onsite. U.S. Environmental Protection Agency (EPA) regulations for existing fluidized bed reactor (FBR) sewage sludge incinerators limit Hg concentrations in the stack emissions to 37,000 ng/DSCM (at 7% O2). The limit for new fluidized bed reactors (FBRs built after October 14, 2010) is much lower (1000 ng/DSCM). At the Metropolitan Wastewater Treatment Plant (St. Paul, Minnesota, USA), dewatered primary and waste secondary sludge is incinerated in three FBRs at a rate of approximately 230 dry tons/day. The process trains for each FBR include energy recovery and air pollution control (APC) equipment, including activated carbon addition to the exhaust gas stream for Hg control, and baghouses for combustion ash and carbon removal. Here, we use EPA Method 30B to measure the in-stack Hg concentrations for one of these FBRs, and attempt to correlate Hg emission levels with particular process operating parameters. The results show that the APC equipment is capable of removing Hg to very low levels. Actual measured Hg concentrations in the stack gas ranged from 11 to 300 ng/DSCM (at 7% O2), increasing with higher dust levels in the baghouse outlet stream, and decreasing with increasing carbon feed rate. Actual Hg mass emission rates ranged from 0.01 to 0.25 g/day for this particular FBR.

Because mercury (Hg) is a toxic global pollutant dispersed through the atmosphere, measuring it is important to apportion sources and determine spatial and temporal trends in deposition and air concentrations. Here, we modified a popular and inexpensive quadcopter to collect gaseous mercury (Hg) on gold-coated quartz cartridges. The gold traps were analyzed in the laboratory using cold vapor atomic fluorescence spectrometry (USEPA Method IO-5). The purpose was to determine the feasibility of using the method to measure ambient Hg concentrations at precise locations aloft and to detect emissions from a known point source. To that end, the quadcopter was outfitted with a pump, three gold coated quartz cartridges, one colorimetric SO2 (Drager) tube, and syringe filters (0.2 µm, PTFE). Flight times averaged 15 minutes, limited by battery life, and yielded >5 pg of Hg, well above the <0.1 pg detection limit. Method precision averaged 12% RSD (range 4.3% to 28%). Demonstrating that the technique can measure elevated emissions of Hg from a point source, we measured progressively higher concentrations nearer a pool of elemental Hg placed atop a ladder in an open field. We applied the method near two coal fired power plants (CFPPs) and a petroleum refinery in the mid-south USA. Concentrations (mean ± SD) near the rural Red Hills Power Plant were 1.5 ± 0.2 ng m-3 downwind and 1.3 ± 0.1 ng m-3 upwind. Concentrations near the Allen Fossil Plant were 1.9 ± 0.1 ng m-3 downwind and 2.0 ± 0.1 ng m-3 upwind. However, because of distance from the stacks (dilution) we do not know if sampling occurred in the actual power plant plumes, and these concentrations likely reflect ambient air rather than the near-field plume. Concentrations near the refinery were also similar downwind and upwind (3.2 ± 0.6 ng m-3 and 3.3 ± 0.9 ng m-3, respectively), but elevated compared to the rural location. Overall, this study demonstrates that highly maneuverable multicopters can be used to probe Hg concentrations aloft. Because of portability, the method may be particularly useful for evaluating Hg emissions from landscapes and transient sources, such as biomass burning, which are poorly characterized and leading to uncertainties in ecosystem budgets. However, before using multicopters for air sampling, air space and flight restrictions must be carefully considered.

Event-based precipitation samples were collected at a downtown industrial and a predominantly upwind rural location in the Cleveland, Ohio metropolitan area from July 2009 through December 2010 to investigate the potential local total mercury (Hg) wet deposition enhancement in a region with a high concentration of coal combustion sources. Total Hg wet deposition for the 18-month period was 6.8 μg/m2 (n = 81) at the rural site and 10.7 μg/m2 (n = 98) at the urban site demonstrating a significant (p = 0.046) 37% enhancement in deposition between the two sites. Large deposition events (>0.2 μg/m2) occurred predominantly from May through October [n = 16 (urban), n = 10 (rural)] and represented from 2 to 8% of total 18-month deposition per event. At the downtown urban site, the average Hg precipitation concentration was 53% higher for these large deposition events. Concurrently measured precipitation events delivered in aggregate 2.4 times more total Hg wet deposition to the urban site compared to the rural site. Hg rainfall concentrations for concurrent events with similar precipitation depth were 2-4 times higher at the urban site and suggest scavenging of local Hg emissions. Further evaluation of these events revealed 83% more total Hg deposition at the urban site from January to December 2010 compared to July to December 2009, while there was 26% more at the rural site during these same time periods. The larger increase in deposition at the urban site in 2010 may be evidence of increased local emissions from sources that were known to be offline during this study period because of an economic recession.

In recent years, national attention has been given to the emission characteristics of mercury compounds and their behavior in the air. Therefore, it is necessary to conduct a national-level research about the inventory of domestic mercury compound emission sources and to develop distribution factors to calculate the emission quantity. This study had three objectives: (1) process for conducting a relative accuracy test audit of Continuous Emission Monitor; (2) study mercury behavior in incineration sources; (3) evaluate the influence of emission sources on ambient point. To understand mercury behavior at medical waste incinerator (MWI), their concentrations at the stack were monitored continuously for three month. During the same period we conducted a relative accuracy test audit of Mercury Continuous Emission Monitor (Hg CEM) using Ontario-Hydro Method and Method 30B. The relative accuracy of Hg CEM with respect to Ontario-Hydro Method and Method 30B as reference method result was 9.38 % and 7.38 % (20.00 % limit). As such, the Hg CEM data passed the RATA. Mercury concentration in flue gas of MWI was 1.05 ~ 280 ug/Sm3, and many cases of short term high concentration occurred. Mercury speciation testing using mercury CEM and Ontario-Hydro Method of showed 26 % elemental Hg, 74 % oxidized Hg in stack. To figure out how mercury affects ambient site from the emission source, we selected ambient site using Industrial Source Complex Short Term 3 and measured mercury continuously for a month and compared evaluation with equipped CEM at MWI. Mercury concentration of ambient site shows mostly clean level (0.90 ~ 2.99 ng/Sm3), but in the monitoring period, the episode was occurred twice that considered to have an affected from the emission source. Further studies on the effects of emission sources on the ambient site will be needed in the future.

The objectives of this study were to: (1) characterize the diurnal and seasonal variations of atmospheric total gaseous mercury (TGM) concentrations, (2) investigate the relationships between TGM and co-pollutants concentrations, and (3) identify likely source directions and locations of TGM using conditional probability function (CPF), conditional bivariate probability function (CBPF) and potential source contribution function (PSCF) for Ansan City, South Korea. There are two national industrial complexes (Sihwa and Banwol) which consist of the manufacturing, electronics industries, the petroleum refineries and the steel industry around the sampling site.

TGM concentrations were measured every 5 min during spring (11-19 May 2015), summer (28 July-04 August 2015), and fall (13-21 October 2015) on the roof of the Choji high school using a Tekran 2537B. Hourly meteorological data and concentrations of NO2, O3, CO, PM10 and SO2 were obtained from the Korea Meteorological Administration (KMA) and the National Air Quality Monitoring Network (NAQMN), respectively.

The TGM concentration was statistically significantly higher in spring (2.7 ± .0 ng m-3) than other seasons (p<0.01), followed by fall (2.5 ± 0.8 ng m-3) and summer (1.9 ± 0.5 ng m-3).The TGM concentrations generally showed a consistent increase around 7:00 due to local emissions related to increased traffic, industrial activities, and activation of local surface emission sources.

The TGM concentrations were simultaneously positively correlated with CO (r = 0.42, p<0.01), NO2 (r = 0.39, p<0.01) and PM10 (r = 0.51, p<0.01), suggesting that combustion processes are an important source.

There was a significantly positive correlation between TGM and CO in this study (r = 0.42, p<0.01), suggesting that TGM and CO were affected by similar anthropogenic emission sources. However, the observed ΔTGM/ΔCO was 0.0031 ng m-3 ppbv-1 in spring, 0.0026 ng m-3 ppbv-1 in summer, 0.0030 ng m-3 ppbv-1 in fall, which are significantly lower than that indicative of Asian long-range transport (0.0046–0.0056 ng m-3 ppbv-1), suggesting that local sources are more important than those of long-range transport in this study.

CPF, CBPF and PSCF plots for TGM concentrations higher than the upper 25th percentile show high source probabilities in the direction of other industrial complexes including the Incheon port, and a coal-fired power plant as well as two national industrial complexes (Sihwa and Banwol).

Gaseous Elemental Mercury (GEM) concentrations and total and leached mercury contents on paints, plasters, roof tiles, concretes, metals, dust and wood structures were determined in the main buildings and structures of the abandoned word-class Hg-mining district of Abbadia San Salvatore (Siena, Central Italy), in order to understand how to proceed for the forthcoming remediation activities, the main aim being that to recover this site for museum and public green purposes. The mining complex covers a surface of about 65 ha and contains mining structures and managers and workers buildings. In this work, nine surveys of GEM measurements were carried out from July 2011 to August 2015 and more detailed measurements were performed in February, April, July, September and December 2016 in the buildings devoted to the production of liquid mercury. GEM concentrations showed a strong variability in terms of space and time mostly depending on the distance from the building hosting driers, furnaces and condensers and ambient temperature, respectively. Surveys carried out in the hotter period (from June to September) showed strikingly high GEM concentrations despite the fact that the mining activity stopped in 1982. In fact, in some of the contaminated sites GEM values reached concentrations that saturated (>50,000 ng m3) the GEM measurement device (Lumex 915+). Concentrations of total (in mg kg1) and leached (in g L1) mercury from different building materials, e.g. brick, rust, furniture, paint, plaster, concrete, showed for the same type of material highly variable values in dependence on the edifice or mining structure from which they were collected. Significantly high total and leached mercury concentrations, up to 46,580 mg kg1 and 4,470 mg L1, respectively, were measured. The obtained results are of relevant interest for the operational cleanings to be carried out during the reclamation activities. Operators are to wear appropriate personal protective equipments and act with machineries (e.g. hydro-blasters) to avoid the dispersion of GEM and reactive mercury in the environment during the removal of the building materials. This is highly recommended for both the operators safety and that of the inhabitants living nearby, the urban centre of Abbadia San Salvatore bordering the former mining area. Continuous acquisition of GEM data is suggested and samples of urine, blood and hair for mercury concentrations should be collected from the operators prior and after the reclamation since several months are likely necessary to complete the cleaning activity particularly in the most contaminated sites.

In the framework of ongoing research projects and programmes (i.e., GMOS, UNEP F&T) aiming to develop advanced sensors for major atmospheric pollutants, and having as overarching goal to assure a full operational capability of global observing systems for persistent pollutants such as mercury a novel sensors with promising sensing features for environmental applications have been designed and tested. The aim of this paper is to present novel kinds of sensors with promising sensing features for environmental applications, exploiting both the combination of gold affinity for mercury and nanosized frameworks of the sensing materials. Specifically, in the present study, conductive sensors working at room temperature and based on composite nanofibrous electrospun scaffolds of titania easily decorated with gold nanoparticles by photocatalysis under UV-light irradiation, have been developed to obtain nanostructured hybrid materials, capable of entrapping and detecting Gaseous Elemental Mercury (GEM) traces. The size and the shape of these nanostructures have been demonstrated to be key parameters in defining the properties of the resulting sensors, because of the strict relationship between the surface and the bulk of the sensing material which is extremely reduced in size. The increase in the number of binding sites has been confirmed to be a successful strategy to ensure sensitivity at trace level. SEM, AFM, TEM and HR-TEM analyses have been performed to characterise the morphology and the nano-sized structure of these composite materials. Different electrical and sensing features of the resulting chemosensors have been achieved by tuning fibres roughness and gold nanoparticle size. A suitable measuring chamber for mercury detection have been designed and developed in order to improve the sensing feature of the sensor. Thus few minutes of air sampling were sufficient to detect the concentration of mercury in the air without using traps (LOD ~ 1 ppb). Longer measurements allowed the sensor to detect lower concentrations of GEM (tens of ppt). A short thermal treatment (450°C, 3min) was necessary to completely desorb mercury from AuNPs. The resulting chemosensors are expected to be very stable over time, robust and resistant to the interference that may be caused by common solvents and by VOCs commonly present in ambient air.

Waste incineration plants have been declared by the Minamata Convention as one of the major industrial sources of mercury (Hg) emissions. Elemental mercury (Hg0) is released from the incinerator (850-1200 ºC) into the flue gas and, as the temperature goes down, Hg undergoes a large number of homogeneous and heterogeneous oxidation processes. Hg0 is converted either to oxidized mercury (Hg2+) compounds and/or Hg adsorbed compounds (Hgp) onto particles. Effective and efficient mercury control technologies are needed to meet the increasingly stricter mercury emission regulations. Dry flue gas cleaning methods, using solid sorbents based either on soda or calcium hydroxide, are used in municipal solid waste (MSW) incineration but their efficiency on mercury removal needs to be further studied. Injection of activated carbon is a potential method for capturing mercury which is removed downstream, in a particulate matter control device, such as electrostatic precipitators (ESP) or fabric filter (FF). The efficiency of mercury removal from the flue gas is substantially affected by its speciation, flue gas composition and process conditions (e.g. temperature, air pollution control units). Many studies are found in the literature related to mercury behaviour under conditions of coal combustion, however, there is a lack of studies connected with MSW conditions. This study discusses the effect of temperature and gas components, present in typical flue gas from MSW incineration, on both mercury oxidation and capture by mineral and carbon-based sorbents. The study was carried out by means of a laboratory scale device that simulates the gaseous mercury behaviour in flue gas at temperatures ranged between 150-300 ºC. The results show the influence of HCl, SO2, NOx and H2O vapour in the gaseous transformation of mercury identifying the major reaction pathways. Under the simulated conditions, mercury (II) chloride (HgCl2), mercury (II) oxide (HgO) and elemental mercury (Hg0) are thermodynamically relevant species. In presence of CO2 and O2, the minor gas components NO, SO2, and HCl are involved in homogeneous oxidation of mercury and therefore, in the retention capacity. The higher temperature, the lower efficiency of Hg capture (i.e. higher emissions of Hg). SO2 is oxidized to SO3 at higher temperature. This effect, together with adsorption effects of CO2 and water vapour, inhibit mercury adsorption as there could be a competition for the same surface binding sites. The study provides a basis for the development of new strategies for mercury removal in the air pollution control devices of MSW incineration plants.