Summary of literature review on measurement of IONS in PM2.5

Directive 2008/50/EC on ambient air quality and Cleaner Air for Europe set limit values for particulate matter PM10 and fine particulate matter PM2.5. The reference measurement methods have already been standardized by CEN, the European Committee for Standardization (EN 12341:1998 and EN 14907:2005 ). Both methods are currently under revision and will be merged (prEN 12341:2012 ). The Directive also requires the measurement in the PM2.5 fraction of inorganic components (anion and cations) and elemental and organic carbon (EC/OC) with the objective to support air quality assessment and management. In order to ensure better data comparability, validated standard methods for these measurements are equally required. This report represents the summary of the literature review required for the validation of the method for anions and cations in PM2.5 or more specific the water soluble fraction of nitrate, sulphate, chloride, ammonium, sodium, potassium , calcium and magnesium. Quartz fiber and Teflon filters are considered. Major uncertainty aspects of the method, the inlet system, evaporation from the filter, reactive gas retention, storage, blanks and extraction are reviewed. In addition equivalence with the EMEP method is considered and guidelines for laboratory tests are given. It is concluded that the contribution of artefacts induced by inlet system, the storage of the filters and the extraction procedure, to the uncertainty of the individual components is negligible. The filter blanks and the retention of precursor gases contribute to a minimum extent to the uncertainty. Blanks (field and laboratory) need to be taken into consideration on a continuous basis. Evaporation leads to considerable uncertainty contribution for nitrate and ammonium, and to a lesser extent for chloride. Table 1 Individual sources of uncertainty Nitrate is a major component of PM2.5 in terms of mass fraction and exhibits a considerable artifact potential. This potential is due to evaporation of ammonium nitrate (NH4NO3) from the filter by changes in the temperature, humidity, timing and precursor gases ammonia and nitric acid. The volatilization may lead in exceptional cases to complete evaporation. Mid and south of Europe are most vulnerable for this artifact. On the other hand overestimation may occur by the substitution of chloride associated with sodium by nitric acid. This effect is less effective in terms of PM2.5 mass. The quantity of evaporated or absorbed mass is strongly dependent on start and stop time of the sampling period. The major fraction of ammonium is associated with nitrate and sulphate. Ammonium sulphate has no vapour pressure and will therefore remain entirely on the filter. Ammonium associated with nitrate may entirely evaporate. It is concluded that ammonium is considerable vulnerable for artefacts but less than nitrate. Furthermore ammonium represents a much smaller PM2.5 mass fraction than nitrate. Chloride is predominantly present as sea salt. Chloride may be substituted by nitrate on filters in polluted coastal areas. This process usually takes place by changing wind directions during sampling. Already collected sea salt chloride being part of marine originated air can get exposed later on to nitric acid being part on polluted continental air mass. In general this effect is very limited for Europe. And most certainly when the effect on the PM2.5 mass fraction is considered. Blanks are considered from two different angles, the anions and cations levels extracted from unused filters and the standard deviations of these levels. Both are of importance concerning the quality of the method. Teflon exhibits very low blank levels for all components, but will not be used for the method because of practical reasons (CEN/TC 264/WG34, document N75). Various quartz fiber brands were tested and showed on average satisfying results. All components when converted into PM2.5 concentrations below 0.1 µg/m3, except for Na in Whatman filters (0.5 µg/m3) and Cl in Albet filters (0.2 µg/m3). As the laboratory averaged blank concentration is subtracted from sample result the uncertainty is dependent on the standard deviation of the blank rather than the blank level itself. Considering the standard deviations all components remained below 40 ng/m3. This is sufficient for background area’s throughout Europe, even in Finland. Brand related restrictions cannot be made, because brand specific qualities are bound to change in time. Currently the use of Whatman QMA should be discouraged because of the high Na blanks and the uncertainty to what level the water soluble Na per batch may vary. In conclusion, considering all artefacts only the impact of the loss of nitrate by evaporation is severe and not well understood. Therefore Laboratory tests (Work Package 2) (7) will focus on parameterization of this phenomenon. Comparison of the standard with the EMEP (SPM) reference method is difficult since the difference in size cut off makes it hard to assess the potential differences. In addition, the Teflon filter used for collecting aerosols in the filterpack method is not necessarily comparable with the single filter PM2,5 when it comes to positive and negative bias like evaporating and absorption of gases. These differences are dependent on chemical composition and meteorology, and the comparability between these methods will therefore vary between sites.