CHEMICAL MEASUREMENT PROGRAM

(C. Temme, J. Baukau, G. Blöcker, A. Caba, C. Caliebe, W. Gerwinski, R. Gioia, E. Hudson, A. Jahnke, Z. Xie)

Reactions mechanisms of mercury and selected persistent organic pollutants (POPs) in air, water, and snow

Several international leading groups of Environmental Chemistry were joining the RV Polarstern on ARK-XX/1 2004. Their common interest was the detection of trace organic contaminants and mercury species in remote environments of the Northern Hemisphere, to investigate the environmental cycling and fate of key global pollutants. The Polarstern with her conditions has been found to be well suited to act as a ‘clean ship’ for the sampling of these trace compounds.

The chemical research program during ARK-XX/1 was focused on two major topics:

1. Determination of mercury and related volatile organic compounds (VOCs) and dissolved organic carbon (DOC) in different compartments and the calculation of the air/sea-exchange during Arctic summer
2. Determination of selected POPs in air, water, and snow

1. Mercury and the VOC/DOC connection

Mercury (Hg) is outstanding among the global environmental pollutants of continuing concern. The element and many of its compounds behave exceptionally in the environment due to their volatility and capability for methylation, in contrast with most of the other heavy metals. Long-range atmospheric transport of mercury, its transformation to more toxic methylmercury compounds, and their bioaccumulation in the aquatic food chain have motivated intensive research on mercury as a pollutant of global concern.

The international process study on board of Polarstern with the transect from Germany to the North Atlantic helped to examine the temporal end of Atmospheric Mercruy Depletions Events (AMDEs) during Artic summer and the spatial distribution of the relevant areas in the north Atlantic Ocean. The fate of mercury during polar summer in the Arctic was investigated with several different methods for the detection of mercury species in air, water, snow, and ice.

Two Tekran gas-phase mercury vapour analysers (Model 2537A) were installed on Polarstern for the determination of Gaseous Elemental Mercury (GEM) and Reactive Gaseous Mercury (RGM species in the Arctic. RGM was measured with a Tekran 1130 mercury speciation unit which gave one Tekran 2537A mercury vapour analyser the ability to concurrently monitor both elemental (GEM) and reactive gaseous mercury (RGM) species in ambient air at the pg/m³ level. The sample inlet was located on the upper deck of Polarstern in front of the GKSS container. The air was sampled at a flow rate of 1.2 L/min with a 0.45 mm PTFE filter in front of the sample inlet of the analyser.

The results of GEM measurements and ground-level ozone concentrations for the time period 16^th of June to 14^th of July are preliminary data. The arithmetic mean of all GEM measurements during this cruise lag is (1.6 ± 0.1) ng m^-³ and is in good agreement with mean summer concentrations from other polar sites like Alert, Canada or Ny Ålesund, Svalbard.

Contaminations from the ships plume will be eliminated after the cruise according to the PODAS data of wind directions and wind speed relative to the ships course.

In addition the second Tekran analyser in the wet lab was used to investigate the sea/air exchange of GEM during Arctic summer. The analyser was connected to a 20 L glas bottle, called “Equilibrator”, indirectly measuring the Dissolved Gaseous Mercury (DGM) concentrations in the sea water with the help of the Henrys law constant.

We found that the Atlantic Ocean is a potential source for GEM in the atmosphere during this time of the year. The flux of GEM from the sea to the atmosphere was in the range of 0.1 to 3 ng/m²/h depending on the model used for calculation. We found significantly higher fluxes when we entered the most westerly part of the 75^oN transect (July 1-3) where the ocean was covered with sea ice.

The oxidative properties of the lower atmosphere (which determine mercury speciation and thus transport) are also reflected in the profile of volatile organic compounds (VOCs) it contains. Dissolved organic carbon (DOC) in surface waters is thought to be the ultimate source of these VOCs over the open sea.

Air samples were taken for VOC analysis (McGill University). Water samples were taken for photochemical and photooxidation experiments on surface water DOC, to further investigate whether these explain the VOCs found in the air samples. Alltogether, 25 whole air samples (6 L) have been taken. Most of these (14) were taken along the 75°N transect (June 24-July 3), to match the extensive water characterization of this transect. A further 7 were used to profile the transition from the North Sea (58°N, June 16) to the Norwegian Sea (72°N, June 22).

Water samples were taken to match the time/location of each air sample. To date, more than 50 samples have been taken (again, mostly along the 75°N transect) and filtered (0.2 µm). The additional samples will serve as checks on the effects of sample size, storage and materials, sampling method (Niskins vs. Glass Sphere Water Samplers (GSWS) vs. Polarstern's in situ pump in the keel), as well as adding 3 intermediate and 4 deep water samples to the data set, which allows us to expose the analytical methods to a wider range of DOC.

Aerosols are important to this study because they represent another interface between the sea surface and the atmosphere, on which DOC may give volatile compounds through photochemistry. However, the cascade impactor used to sample them required long collection times and a favorable wind direction to avoid contamination by the ship itself. Still, 9 samples were collected, each consisting of 9 size fractions. Sampling times ranged from 6 to 40 hours, but were typically 12 to 24 hours.

Towards the end of the 75°N transect, snow was sampled on ice floes in the East Greenland Current to complement the air and water sampling by the air chemistry group. Samples were taken on July 2 and 3^rd at 74^o58.08´N / 13^o38.46´W and 75^o08.52´N / 16^o45.11´W, respectively. Four 250-mL samples were collected by ultra-clean methods for mercury and radionucleide analysis. These will suppplement the extensive snow sampling for these materials which will take place during ARK-XX/2. Four 1-L samples were collected for experiments on the photochemical generation of VOCs (to be done at McGill University). Finally, 40 L of snow were taken in GSWS on each occasion. These will be analysed for POPs by the BSH group.

2. Persistent organic pollutants (POPs)

By combining short-term atmospheric samples with the collection of representative water samples across different region of the North Atlantic / Arctic circle, answers are sought as to whether atmospheric transport or the marine phytoplankton productivity are controlling the transport and settling flux of persistent organic pollutants.

Air Sampling

Four different modified High Volume air samplers were used to collect air on board of RV Polastern during ARK-XX/1. POPs such as polychlorinated biphenyls (PCBs), polybrominated dyphenyls ethers (PBDEs), organochlorine Pesticides (HCHs, HCBs , DDTs etc.), polyflourinated compounds (PFCs), Nonylphenol (NP), and combustion-derived polychlorinated dibenzo-p-dioxins, furans (PCDD/Fs) and Total Suspended Particulate (TSP) will be investigated. Totally 45 air samples were collected from Bremerhaven to Longyearbyen during the ARK-XX/1 (see table 1 - 3). Table 1 shows samples collected using the Lancaster University High-Volume air sampler, which will be used to determine PCBs and PBDEs concentrations. Table 2 represents samples collected using the High Volume air sampler from University of Bremen and the samples collected for the determination of TSP (Total Suspended Particulate) using a TSP sampler belonging to Lancaster University. Samples from the Bremen Hi-vol will be used to investigate the concentration of dibenzo-p-dioxin and furans in the atmosphere. Table 3 shows sampling volumes and locations for the samples taken with the GKSS High-Volume air sampler in order to determine PFCs and NP in air.

Sample ID	Latitude	Longitude	Air volume (m^3)	Comments
ARK-XX 1	55.5-58.5 N	7.1-4.8 E	287	North Sea
ARK-XX 2	58.6-61.1 N	4.7-3.99E	193	North Sea
ARK-XX 3	61.4-63.4N	4.0-5.06 E	196	North Sea
ARK-XX 4	63.5-65.7 N	5.08-6.98 E	200	North Sea
ARK-XX 5	65.9-68.1 N	6.98-8.94 E	191	North Sea
ARK-XX 6	68.2-69.2N	9.03-13.2 E	190	North Sea
ARK-XX 7	69.2-71.4 N	13.2-14.8 E	188	North Sea
ARK-XX 8	72.00N	14.72E	199	HMMV
ARK-XX 9	72.00N	14.72E	192	HMMV
ARK-XX 10	72.00N	14.72E	196	HMMV
ARK-XX 11	71.1-74.6 N	16.6-10.6 E	365	75º Latitude transect
ARK-XX 12	74.6-74.55N	10.6E-4.4W	411	75º Latitude transect
ARK-XX 13	75.0-76.2 N	4.5-3.4W	458	75º Latitude transect-LYB
ARK-XX 14	76.20-78.9 N	3.7W-5.4E	490	On the way to LYB
ARK-XX 15	79.4-79.8 N	4.2E-3.20	771	Hausgarten
ARK-XX 16	79.4-79.35N	3.2-5.11E	396	Hausgarten
ARK-XX FB1	68.1N	8.9E	Field blank	North Sea
ARK-XX FB2	69.2N	13.2 E	Field blank	North sea
ARK-XX FB3	72.0 N	14.7 E	Field blank	HMMV
ARK-XX FB4	74.5 N	4.4 W	Field blank	75º Latitude transect
ARK-XX FB5	79.4N	3.2 E	Field Blank	Hausgarten

Table 1. Samples collection for the Lancaster High-Volume sampler during ARK-XX/1.

Sample ID (Bremen High-VOL)	Sample ID (TSP sampler)	Latitude	Longitude	Flow rate (ft^3/min)	Comments
ARK-XX R1	TSP ARK-XX 1	57.2-61.7 N	5.3-3.9E	13	North Sea
ARK-XX R2	No TSP sample	61.9-66.3 N	3.9-7.4 E	13	North Sea
ARK-XX R3	TSP ARK-XX 2	66.5-69.35 N	7.4-13.7 E	14	North Sea
ARK-XX R4	TSP ARK-XX 3	71.9-72.01N	14.72 E	19	HMMV
ARK-XX R5	TSP ARK-XX 4	71.10-74.6N	16.6-10.6E	20	To the 75ºtransect
ARK-XX R6	TSP ARK-XX 5	74.6-74.55N	10.6E-4.3W	20	75ºtransect
ARK-XX R7	TSP ARK-XX 6	75.0-76.1N	4.47-3.4W	19	75ºtransect to LYB
ARK-XX R8	TSP ARK-XX 7	76.2-78.91	3.1W-5.1E	19	To LYB
ARK-XX R9	TSP ARK-XX 8	79.6 N-78.9	5.18E-4.8 E	20	Hausgarten
ARK-XX RFB1	TSP ARK-XX FB1	69.38N	13.85 E	Blank	North Sea
ARK-XX RFB2	TSP ARK-XX FB2	72.0 N	14.72 E	Blank	HMMV
ARK-XX RFB3	TSP ARK-XX FB3	74.6N	4.3W	Blank	75ºtransect

Table 2. Samples collection for the Bremen High-Volume sampler and the TSP sampler during ARK-XX/1.

Table 3. Samples collection for the GKSS High-Volume sampler during ARK-XX/1

GFFs (Glass Fiber Filter) precombusted at 450 °C overnight, were used to capture the particles and particle bound species, and two cylindrical 3 inch diameter polyurethane foam (PUF) adsorbants were used downstream to capture the gas-phase contaminants. Prior to use, the PUFs for the Lancaster and Bremen High-Volume sampler were cleaned with an Accelerated Solvent Extraction (ASE) system using dichloromethane (DCM) as solvent. After cleaning, the PUFs were desiccated under vacuum to remove excess solvent and stored frozen in pre-cleaned aluminum tins. The PUF adsorbant and XAD resin for NP and PFCs sampled with the GKSS high-vols were precleaned by soxhlet extraction with acetone/hexane and ethyl acetate respectively. After samples collection, GFFs and PUFs (and XAD resin) were sealed and stored frozen until they will be analysed. Because low concentration levels are expected to be found in Artic regions, all the pre-cleaning procedures was performed in a clean room at Lancaster University and GKSS.

The overall ship’s track from the North Sea into the remote Artic region is a great opportunity to monitor the air concentrations of POPs from source regions to polar areas. We expect to see spatial trends in air concentrations that provide some evidence for the global fractionation theory. It is generally assumed that the atmosphere can serve as pathway for the delivery of these pollutants to water and terrestrial surfaces. Therefore, POPs can undergo long-range atmospheric transport with high volatile compounds condensing in colder (polar) regions and less volatile compounds condensing in warmer regions close to sources. This will probably not occur in one step, but in a number of steps of volatilization followed by deposition, followed by seasonal fluctuations in temperature. The effect of this will be a relative enrichment of the more volatile compounds in cold areas. Criteria for global fractionation behaviour of chemicals are various physical-chemical properties such as vapour pressure, the octanol-air partition coefficient, the octanol-water partition coefficient and the Henry’s Law Constant.

PCBs are a class of compounds with a variety of different physical-chemical processes. There are 209 congeners depending from the number of chlorine atoms on the molecule and the position that these atoms occupy on the molecule. Therefore, with all these differences in physical-chemical properties, PCBs are ideal to investigate and find evidence of the global fractionation theory. These properties are very dependent on temperature and will therefore greatly influence the global transport of POPs. The temperature dropping during this cruise track, from 11 ºC to – 2 ºC is a rare opportunity to estimate the temperature dependence of POPs in the atmosphere.

Growth in interest on PBDE flame retardants has been as exponential as their apparent increase in the environment over the past 20-25 years in North America and Europe. Toxicological studies of limited PBDE congeners indicate that they are potential thyroid disruptors and developmental neurotoxicants. However, there is still very little information on PBDE contamination and its spatial trend over the regions of the world. The investigation, which will follow air sample collection on Polarstern, will attempt to comprehensively understand the spatial and temporal trend of contamination by PBDEs emerging compounds in the atmosphere.

Polyfluorinated organic acids and their derivatives are produced by industry in very large quantities and are used for many purposes. Perfluoroalkyl sulfonates are used e.g. as surfactants and surface protectors in carpets, leather, paper, packaging and upholstery. In addition, some sulfonated and carboxylated PFCs have been used in or as fire fighting foams, alkaline cleaners, shampoos, and insecticide formulations. Due to the large production quantities and the persistence in the environment, polyfluorinated compounds are meanwhile globally distributed. Perfluorooctanesulfonic acid (PFOS) has been detected in blood of ringed seals, other long chain perfluorinated chemicals have been detected in polar bears, arctic foxes, ringed seals, mink, birds and fishes collected in the Arctic.

Because of the findings of polyfluorinated compounds in Arctic biota samples, it is of special interest to investigate their long range transport. Due to their high polarity, a transport by the water phase is likely, especially since some of the PFCs have been found in North Sea water. Some precursors of PFOS and PFOA are highly volatile and can lead to an increased input of PFCs from the atmosphere to remote areas. The investigation of the wide scale distribution of polyfluorinated acids in the sea water of the North Sea and Arctic Ocean is a perfect complement to the simultaneous measurements in the atmosphere. The cruise was quite optimal for these investigations as it ranged from the likely sources (European continent) to remote areas without direct inputs.

Short-term air samples (12 hr) were collected in the North Sea, from Bremerhaven to Tromso. These samples are useful to investigate the day-night cycle (if there is any) of PCBs over the ocean and whether the marine phytoplankton is controlling the atmospheric concentrations of POPs in the marine atmosphere. We were expecting to collect more short-term atmospheric samples during the 75° latitude transect and up to the Hausgarten, but this was not possible since the ship was stopping and starting most of the time. When taking an air sample good care should be taken to minimize the risk of contamination from the ship exhaust and this was quite impossible when the ship stopped and was not kept in the wind.

All the laboratory analysis of extractions and clean up will be performed in a clean room at Lancaster University, UK, and GKSS-Research Centre in Geesthacht, Germany. Field blanks were collected in order to define blank-based limit of detection (LOD) for each sampling matrix (PUF, XAD and GFF) as the average contaminant mass of field blanks plus 3 standard deviations.

Passive air sampling

Five passive air sampler were deployed on board of the ship to look at the ship background concentrations of POPs. Passive samplers were deployed on the Peildeck near the hi-vol, in the GKSS container where matrix are treated before and after sampling, on the back of the ship, in the –30 ºC freezer on the F deck where air samples are stored and in the wet lab, where water sampling is performed.

Polyurethane Foam (PUF) samplers are used to passively sample air. The advantage of passive sampler compared with active samplers is that they are cheap because they are not power consuming and they are easy to deploy.

The surface of the PUF comes into contact with vapour phase species in the atmosphere and it will respond through three different steps:

· Initial linear uptake of the compound to the surface

· A curvilinear portion of the uptake as equilibrium is approaching

· Equilibrium between the air concentration and the surface.

The mass of the compound held by the surface when it is at equilibrium with the air will depend on:

1. Temperature
2. Type of the surface
3. Physico-chemical properties of POPs

The uptake is also influenced by:

• The size or capacity of the subsurface compartment.
• The suface area
• The thickness of the sampler.

These factors above can be varied depending on the compound of interest, the deployment/sampling time and the sensitivity of the analytical instrumentation.

Time to reach equilibrium with the sampler device vary between compounds and increases with K_oa.

Previous studies on RV Polarstern, where passive sampler were also deployed, showed that the ship does not contribute to PCB contamination and this allowed us to conduct research on POPs on this research vessel even at very low concentration such as the Artic regions. We expected to see the same results about the ship background contamination during this cruise leg.

Water Sampling

Water sampling was performed simultaneously to the air sampling in order to investigate the mechanisms controlling the air-water exchange flux.

The water samples were taken from the clean seawater system (stainless steel pipe) of Polarstern (11m depth) as well as directly from the surface water (20m depth) with the glas sphere water samplers (GSWSs) from the BSH. We expect the air concentration of these compounds to be at equilibrium with the water concentration in the Artic region, given the remoteness of this area.

ARK-XX/1 Sampling Overview for contaminants in Seawater (BSH)
Sampling	Number of stations	Number of SPE-extractions	Number of LLE-extractions
Sampling locations winch	8	40	8
Sampling locations pipe	12	18	5
Snow samples	2	4	0
Quality control, blanks		6	5
Quality control, recovery		4	0
Total samples	22	62	13