3.10  Sampling of heavy metals in precipitation

3.10.1  Introduction

The Protocol to the Convention on long-range transboundary air pollution (LRTAP) on heavy metals was signed on 24th of June 1998 in Aarhus in Denmark. It targets three particularly harmful metals: cadmium, lead and mercury. According to one of the basic obligations, Parties will have to reduce their emissions of these three metals below their levels in 1990 (or an alternative year between 1985 and 1995). The main sources for atmospheric deposition of heavy metals are mines, smelting plants and metal industry of various types as well as burning of coal and other fossil fuels. Additive of lead in gasoline is historically one significant source of lead; however, international agreements have decreased this problem.

From 1999, heavy metals have been part of the EMEP program; and it is therefore a need for harmonisation and standard procedures for sampling and analysis of heavy metals. The recommendations are mainly based on previous work of EMEP (NILU and IVL, 1993) and the conclusions from the EMEP and WMO-GAW workshops in Durham, Beekbergen, Moscow and Aspenäs (EMEP, 1993; EMEP, 1996b; EMEP, 1997a; EMEP, 1997b) and at the two first task force of measurements and modelling, TFMM (EMEP, 2000).

Due to the special properties of mercury, this needs a different sampling technique than the other elements and separate chapters have been written for sampling of mercury, chapter 3.12.

3.10.2  Siting criteria

Generally, as for the main components, the heavy metal measurements should not be influenced by emissions from local sources nor by local circulation effects and formation of stagnant air pockets, which cannot be reproduced in the regional scale models. The siting should follow the criteria written in Chapter 2; in addition, further precautions have to be taken to prevent local sources as e.g. metal dust from pumps, metal surfaces, building materials, paint etc.

3.10.3  Sampling procedure

The sampling of heavy metals follows to a large extent the same procedures as for the main components in precipitation, Chapter 3.1, but due to the sensitivity for contamination extra precautions are needed. The concentrations of heavy metals in precipitation are typically only a few nanograms per millilitre and it is important that the standard procedures are followed carefully.

3.10.3.1  Equipment

It is recommended to use a wet only collector for precipitation sampling. When choosing which wet-only sampler to use, it is important that no parts of the precipitation collector are made of metal. Further that all parts can easily be cleaned and be of known composition. As for material, high-density polyethylene collectors are recommended. Among good wet-only collectors are the ARS type from Eigenbrodt in Germany and the one from MISU in Sweden. Other wet only collectors can be used if they have proven to be of the same or better quality.

Different sampler designs have different sampling efficiency, which may lead to incomparable results when calculating the wet deposition. A parallel measurement of precipitation amounts should be made to identify any discrepancies. It is strongly recommended to use a rain gauge in parallel with the sampling equipment. The difference in precipitation amount between the two collectors should not be greater than 10%. If systematic errors are found, the sampler design should be reconsidered. The precipitation amounts from both collectors should be reported to CCC.

Bulk collector may be used when proven to be quantitatively equivalent to the recommended method. Bulk collectors tend to give too high metal concentrations due to dry deposition, but in some areas there are practically no difference between the two types of collector i.e. in the Nordic countries. When a bulk collector can be used, it is recommended to use a sampler with separate collective funnel and collection bottle for easy cleaning. In addition the funnel should have high walls. Some of the traditional bulk collectors have a metal ring in the top; this has to be taken away before it can be used for heavy metal sampling. If a bird ring is needed it should be made of polyethylene. In the field comparison of heavy metals in precipitation performed in Deuselbach Germany in 1995 a variety of different bulk precipitation collectors were compared (Winkler and Roider, 1997). It was discovered that the use of a fine meshed net (<1 mm) in the funnel neck caused troubles with wetting loss, sometimes hindered water from draining into the bottle and caused difficulties in sealing the funnel and bottle. It is therefore recommended to avoid using this type of net. However, in order to prevent insects leaves etc. to enter the collection bottle one can use a sieve made of i.e. polycarbonate with larger grid size. The sieve should be free and not tied up in the funnel neck. The sieve can for example just be the filter backing used in the filter folder system. For sites with snow in wintertime it is recommended to use the rain sampler also in this season, since a snow collector (cylindrical bucket) are difficult to clean thoroughly. However, for places with extremely much snow and strong wind it might be necessary to use a snow collector in the wintertime.

3.10.3.2  Cleaning

The concentration of heavy metals in precipitation samples and extracts from filters are very low, in the interval 0.01-10 ng/ml. Therefore, it is very important that all the equipment are thoroughly cleaned to prevent contamination The precipitation bottles should be thoroughly cleaned between use. They should be stored minimum one day in 2% nitric acid and then washed several times (minimum three times) with de-ionized water, dried corked and packed in two clean plastic bags and zipped until used in the field. All reusable lab-ware should be cleaned in the same manner. The funnels should be washed in the field between every sampling and every month sent to laboratory for more thorough cleaning.

3.10.3.3  Standard operating procedure

Sample bottles, measuring equipment etc should always be handled with care to prevent contamination. Plastic gloves should be used when collecting the samples and the inside of the funnel or the tip of the collector should not be touched. All bottles should be kept in double plastic bags during transport and storage. The station observer must wear clean clothes that do not give any dust or other kind of pollution.

In order to prevent contamination, the precipitation bottle should be sent directly to the laboratory without transferring any precipitation into smaller transport bottles, which are usually done when measuring the main components in precipitation. Immediately after disconnection the sample bottle must be closed e.g. with a screw cap and sent to the laboratory. The precipitation amount is measured by weight. The empty sampling bottle with screw cap is weighed before use and then weighed after the sampling period is finished.

The precipitation must be conserved in nitric acid which is added either before or just after (Chapter 3.10.4) the sampling. To prevent growth of algae in the sample the acid should be added before the precipitation sampling. Before sending the precipitation bottle to the field it should then be filled with e.g. 2 ml concentrated HNO3. Choosing this sampling strategy it is, however, important to consider whether i.e. a possible evaporation of nitric acid may influence other measurements at the site.

Some trace metals may absorb on the surface of the funnel. Therefore, the funnel should be washed with an exact volume (200 ml) of acidic water (1% HNO3), which is collected in a separate collection bottle. This is analysed to study the influence of absorption. This is especially important in the beginning of the sampling program, and if it turns out to have a significant influence this needs to be included in the sampling procedure as written in cursive below.

The standard sampling procedure is:

Storage of equipment in the field should always be in plastic bags to prevent contamination and kept on as clean and dust-free place as possible. Especially it should be avoided to let the equipment be in contact with or close to metal surfaces as copper, zinc, aluminium etc. since these may often give off metallic dust.

The sampling procedures described above are similar for both wet-only and bulk collectors. However, when using bulk collector for snow sampling, the funnel is often full before the end of the sampling period. The station observer must therefore take the collector (both funnel and precipitation bottle) indoor whenever it is full, and close the funnel with a polyethylene lid. The lid must have been cleaned before use and it should be kept on during the entire melting process. While this sample is melting another collector and funnel is installed, and when the sampling period is finished, the samples are poured together in the last collector. Before pouring, the sample should be shaken to include possible solid residue. If the amount is too much, both collectors are sent to the laboratory. A major drawback of the bulk sampling approach is the likely reasons for contamination due to insects, bird droppings or other material in the sampling vessels. This is especially a problem for extended sampling periods. The risks of contamination are kept under control by using two or three parallel samplers. Contaminated samples can then be identified and discarded.

3.10.4  Conservation and filtering precipitation samples

3.10.4.1  Sample storage

Precipitation samples should be stored in the dark and refrigerated. A storage time up to 6 months can be acceptable providing that long time stability is checked. This includes the testing of blanks of samples stored for long time periods. However, to detect problems with contamination on an early stage, it is recommended to analyse as soon as possible after sampling.

After measuring the sampling volume by weighing the storage bottles, nitric acid should be added (this can also be added before sampling,). This can be done by adding one ml of suprapure concentrated nitric acid to each 100 ml precipitation. This will dissolve the metals that could be adsorbed to the walls of the container. The equipment used for this procedure must be carefully washed with 1% HNO3 before use and plastic gloves must be worn. Acidified samples should be stored in the collection bottle for at least 24h before being transferred to acid cleaned storage bottles. The samples should be stored refrigerated (4oC) until analysis.

Filtering the samples should generally be avoided to prevent contamination; however it may be necessary for samples containing too much non-dissolved material, which is often the case for precipitation samples in South and Central Europe. Procedures for filtering are described in Chapter 3.10.4.2.

3.10.4.2  Filtration of precipitation

Acidify the collected precipitation in the sampling bottle as described in the section above. Equipment for vacuum filtration should be used. All equipment used for filtration must be thoroughly cleaned in 2% HNO3. A cellulose acetate filter with 0.45 mm pore size (e.g. Sartorius, No. 1106-50-N) should be used. If sufficient amount of precipitation has been collected, filter about 20-50 ml of the sample to rinse the filter and discard the filtrate. Then filter 100 ml of the sample and transfer into an acid cleaned bottle for analysis. If less than 120 ml precipitation is collected, filter 20-50 ml 1% HNO3 s.p. to rinse the filter and discard the filtrate. Then filter the collected precipitation and transfer into an acid cleaned bottle for analysis. Change filter and rinse the filtration flask with 1% HNO3 s.p. between samples. Filtered blank samples should be prepared as a control for possible contamination during filtration. 

3.10.5  Field blanks

Two extra sampling bottles are brought to the site; one containing about 100 ml diluted HNO3 (or HCl for Hg sampling), pH 3 to 4, and one empty. After removing the regular sample bottle and washing the funnel as in the ordinary exchange procedure, the empty bottle is installed and the diluted acid is poured through the sampling device. The bottle is capped and brought to the laboratory for analysis. Field blank samples should be taken regularly, at least four times a year. If blank values exceed 20% of the concentration normally measured at the site measures should be taken to reduce the blanks (i.e. exchange or cleaning of sampling devices). The yearly average blank values are used to determine the detection limit and should be reported to CCC.

3.10.6  Measuring the influence of dry deposition

After a fixed period, i.e. one week without any rain, 100 ml diluted HNO3 (or HCl for Hg sampling) are poured into the funnel and collected in the empty precipitation bottle. The bottled is disconnected and sent to the laboratory for analysis. The metal content will then indicate the importance of dry deposition. This type of field blank should be done regularly and especially important when a new station is installed, afterwards it should be done yearly. This exercise is especially important when considering the use of bulk collector for precipitation sampling.

3.10.7  Quality assurance

The low ambient concentration of trace elements will easily cause wrong measurements if strict precautions are not taken to prevent contamination and other sources of errors. The laboratories collecting trace element data for EMEP should therefore have a QA procedure, which is designed for their own sampling and analytical procedures. The QA procedures should among others include:

3.10.8  References

EMEP (1993) Proceedings of the First Workshop on emissions and modelling of atmospheric transport of persistent organic pollutants and heavy metals. Durham, N. C., United States, 6-7 May 1993. Lillestrøm, Norwegian Institute for Air Research (EMEP/CCC-Report 7/93).

EMEP (1996) Proceedings of the EMEP Workshop on European monitoring, modelling and assessment of heavy metals and persistent organic pollutants. Beckbergen, Netherlands, 3-5 May 1994. Bilthoven (RIVM Report 722401013).

EMEP (1997a) EMEP-WMO Workshop on strategies for monitoring of regional air pollution in relation to the need within EMEP, GAW and other international bodies. Aspenäs, Sweden, 2-4 June 1997. Kjeller, Norwegian Institute for Air Research (EMEP/CCC-Report 10/97).

EMEP (1997b) Report and proceedings of the workshop on the assessment of EMEP activities concerning heavy metals and persistent organic pollutants and their further development. Volume I. Moscow, Russian Federation 14-26 September 1996. Geneva (EMEP/MSC-E Report 1/97, WMO/GAW report No 117).

UN-ECE (2001) Measurements and modelling. (EB.AIR/GE.1/2001/4). http://www.unece.org/env/documents/2001/eb/ge1/eb.air.ge.1.2001.4.e.pdf

NILU and IVL (1993) EMEP-long term plans, Annex 1: Measurement program for heavy metals. Lillestrøm, Norwegian Institute for Air Research (EMEP/CCC-Note 2/93).

Winkler, P. and Roider, G. (1997) HELCOM-EMEP-PARCOM-AMAP Field intercomparison of heavy metals in precipitation 1995. Berlin, Umweltbundesamt (Report 104 08 540).

WMO (1971) Guide to meteorological instrument and observing practices. Geneva (WMO No. 8 TP 3).


Last revision: November 2001