Accumulation of airborne pollutants (PAH, chlorinated hydrocarbons, heavy metals) in various plant species and humus

Accumulation of airborne pollutants (PAH, chlorinated hydrocarbons, heavy metals) in various plant species and humus

Environmental Pollution (Series A) 36 (1984) 295-310 Accumulation of Airborne Pollutants (PAH, Chlorinated Hydrocarbons, Heavy Metals) in Various Pla...

706KB Sizes 2 Downloads 20 Views

Environmental Pollution (Series A) 36 (1984) 295-310

Accumulation of Airborne Pollutants (PAH, Chlorinated Hydrocarbons, Heavy Metals) in Various Plant Species and Humus W. Thomas,* A. Riihlingt & H. Simon* * Lehrstuhl ffir Hydrologie, Universit~it Bayreuth, PO Box 3008, D-8580 Bayreuth, Federal Republic of Germany t Swedish Environmental Research Institute, Aneboda, S-360 30 Lammhult, Sweden

ABSTRACT Vegetation and raw humus samples from an industrial area in Sweden were analysed for P A H ( benzo-ghi-perylene, benzo-a-pyrene, indeno-cdperylene, fluoranthene), chlorinated hydrocarbons (at- and 7-BHC, HCB, polychlorinated biphenyls) and heavy metals (Zn, Pb, Cu, Cd, V). The results of a principal components analysis indicate various accumulation patterns related to uptake rate and surface of individual species as well as to the physicochemical properties of the individual pollutants. While the highest P A H levels were found in the humus, followed by mosses and lichens, elevated chlorinated hydrocarbon accumulations were also detected in some higher plants. The heavy metals investigated showed a distinct pattern of uptake. Comparison of these three groups of pollutants shows the influence of different atmospheric transport phases on the uptake of trace substances by vegetation. Furthermore it can be seen that lower plants, such as mosses and lichens, which are often usedfor atmospheric pollution monitoring, can be used to determine the pollution levels of other species.

INTRODUCTION The analysis of micropollutants in natural organisms has been shown to be a suitable method for atmospheric pollution monitoring. Lower 295 Environ. Pollut. Ser..4. 0143-1471/84/$03.00 © ElsevierApplied Science Publishers Ltd, England, 1984. Printed in Great Britain


W. Thomas, A. Riihling, H. Simon

plants, especially mosses and lichens, clearly reflect regional differences in deposition levels because of their high accumulation rates for a number of pollutants. Since the early investigations of trace metal monitoring using mosses by Riihling & Tyler (1968), a number of regional studies have been carried out. Although metal uptake by means of particle filtering and retention on the plant surface, especially in highly polluted areas, seemed to be an important process of pollutant enrichment (Little, 1973; Chamberlain, 1983), the incorporation of heavy metals into plant tissues was observed by Skaar et al. (1973). In addition to heavy metals, a number of trace organics were analysed in moss samples along a transect in central Europe (Thomas & Herrmann, 1980). The pollution pattern recorded was also clearly related to various emission sources. Furthermore, a number of quantitative relations between the pollution levels of the atmosphere (precipitation, suspended particulate matter) and accumulation rates in plants were computed for heavy metals (Pilegaard, 1979; Hanssen et al., 1980) as well as for organic pollutants (Thomas, 1983). Prediction models of atmospheric deposition based on moss data were evaluated for PAH and a number of metals. The relationships between the accumulation rates of heavy metals in various plants have been described by Tyler (1972) and Little & Martin (1972). The uptake of Hg, which is characterised by chemical and physical properties different from other heavy metals, was investigated by Huckabee et al. (1983). Polychlorinated biphenyls were found to be adsorbed in large amounts by soil and it was possible to show that only a small proportion was absorbed onto plants (Strek & Weber, 1982). Polyaromatic hydrocarbons are released to the environment during processes of incomplete combustion of fossil fuels. The amount of PAH produced and the relationship between the concentrations of the individual substances of this group of pollutants depends on the kind of fuel used as well as on the pressure and temperature conditions inherent in the combustion processes. The atmospheric transport of PAH takes place in the particle phase (Neff, 1979). A number of PAH have been shown to be carcinogenic, for example benzo-a-pyrene. Polychlorinated biphenyls (PCB) are mainly used as stabilising substances in the chemical industry and as additives in electrical transformer liquids. Highly chlorinated PCB with more than five chlorine

Accumulation of airborne pollutants in plants


atoms appear to be potentially more damaging to human health than, for example, tetrachlorobiphenyl. Like chlorinated pesticides they have higher volatilisation rates than PAH and occur in the atmosphere as gases. The relationship between gas and particle phases depends on temperature and humidity (Lee, 1976). The chlorinated hydrocarbon HCB is used as a pesticide as well as a stabilising substance in paint and plastic industries. Like the well-known pesticide 7-BHC (Lindane), the application of these kinds of chemical substances is restricted in most European countries because of their effects on health and their long residence times in the environment. The ~BHC isomere is a by-product of Lindane. Because of their chemical properties, PCB and chlorinated pesticides may be cycled between the atmosphere and vegetation/soil for up to about 20 years. The purpose of this investigation can be summarised as follows: 1.

2. 3.

To present concentration levels of a number of organic and inorganic micropollutants with different chemical properties for various plant species and raw humus samples. To compare the accumulation pattern of the micropollutants with respect to different atmospheric transport processes. To compare the accumulation pattern of the micropollutants with respect to the higher and lower plants commonly used for atmospheric pollution monitoring.


Investigation area and sampling Stenungsund is a village with several petrochemical factories. It is situated on the Swedish west coast, about 30 km north of G6teborg (Fig. 1). There are about 9000 inhabitants in the central part of the community. The area around Stenungsund is formed of Archaean rock and is divided by a network of valleys. The soils of the coastal rocky plateau are acid and deficient in nutrients. The area has a maritime climate with a mean annual precipitation of about 730 mm, and a mean temperature of 6 °C. January is the coldest m o n t h of the year with a mean temperature of - 1 °C, and July the warmest at 16 °C. The mean duration of snow cover is 50 days, Calluna heath is the characteristic vegetation of the coastal rocky


W. Thomas, A. Riihling, H. Simon








2km I


Fig. 1.

Location of the investigation area. (1, 3 and 4 are chemical factories-petrochemistry, plastics; 2 is an oil fired power plant.)

hills. On deeper soils coniferous forests are common and deciduous woods often border arable land and pastures. Sampling was performed in September 1982 in the close vicinity of the industrial area of Stenungsund. Samples of soil and plant material for analysis for organic compounds were placed in aluminium boxes, whereas those for heavy metal analysis were kept in polyethylene bags. The

Accumulation of airborne pollutants in plants



List of Samples Investigated with Abbreviations


Sample name/species Litter from deciduous trees Litter from coniferous trees Deschampsia flexuosa, leaves

Cladonia rangijerina Hypnum cupressiforme Rhacomitrium lanuginosum Russula cyanoxantha Juniperus communis, needles Calluna vulgaris Vaccinium myrtillus Betula verrucosa, leaves Betula verrucosa, bark Corylus avellana, leaves Pinus sylvestris, 1st year needles Pinus sylvestris, 2nd year needles

specification of the various samples and the abbreviations used are shown in Table 1.

Chemical residue analysis Organic micropollutants were analysed by a combined method for PAH and chlorinated hydrocarbons, following the procedure of Thomas (1979). The extraction of homogenised sample material was carried out on a Soxhlet-apparatus (24h) using acetone as solvent. After mixing with aqueous NaESO 4 solution there followed a further liquid/liquid extraction stage using CHCI 3. The volume-reduced extracts were afterwards cleaned by column chromatography (aluminium oxide/silica gel). The elution of the columns was done in two stages, for PCB and HCB in the first elution and PAH and chlorinated pesticides in the second. The effective separation of PCB and HCB from BHC and the other chlorinated hydrocarbon pesticides can be seen in Fig. 2. The determination and quantification of the individual organic pollutants was carried out by the following methods.

I,lL Thomas, 4. Rffhling, H. Simon


. . BOse . . . . . . . . . .Pe,.'t~ . . . . . . . . . .=. . . .32E . . . . .' ,. ~. .'.~. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . b)

2,2',3,5. 6" pent aehlo roblphenyl

Jl,k/ 50 Bo~e


J, 15EI

l ,L 20(I




362.00 ....................................................................... F'Eo~


t/ //

C) 2, 2',3,3',6,6'- hexachloroblpheny!

I 50







Fig. 2. Gas chromatogram (ECD-Ni63) of the first elution from organic trace substance analysis and mass spectra of quantified PCBs in a lichen sample. (a) S = solvent; 1 = HCB; 2 = 2,2',3,5,6-pentachlorobiphenyl; 3,4 = pentachlorobiphenyls; 5 = 2,2',3,3',6,6'-hexachlorobiphenyl; 6,7,8,9,10,11 = hexachlorobiphenyls; 12,13,14 = heptachlorobiphenyls. (b)Mass spectrum of peak no. 2 from (a). (c) Mass spectrum of peak no. 5 from (a).

Accumulation of airborne pollutants in plants


PAH Benzo-ghi-perylene, benzo-a-pyrene, indeno-cd-perylene, and fluoranthene were separated by high performance thin layer chromatography (HPTLC) using RP18-nanoplates (Merck). The plates were developed with acetonitrile and the fluorescent intensity was measured using a chromatogram spectrophotometer (Zeiss) following Hermann (1978). The excitation wavelength was 365 nm. Quantification was carried out by comparison with standard substances.

Chlorinated pesticides The chlorinated hydrocarbons 7-BHC (the pesticide Lindane), its isomere ~-BHC, and HCB were separated by GC-ECD-analysis using capillary columns (OV 1-phase). Quantification was carried out by comparing with peak intensity of standard solution MX7 (Nanogen Standards).

Polychlorinated biphenyls The following two PCB-substances were determined by GC-ECDanalysis: 2,2',3,5,6-pentachlorobiphenyl (5-PCB), 2,2',3,3',6,6'-hexa.chlorobiphenyl (6-PCB). Separation by GC with capillary column was followed by ECD-Ni63 detection (see Fig. 2). The quantification was carried out by comparing with standard substances. Confirmation of the results of the gas and thin-layer chromatography was done by mass-spectrometry using a HP 5992 Quadropole equipped with capillary column. Further details of the organic trace substance analysis with recovery rates, etc., can be seen in Thomas (1979). Heavy metals were analysed by atomic absorption, using a Varian instrument (model AA6) after digestion of the samples with a mixture of perchloric acid and nitric acid.

RESULTS Organic trace substance concentrations

The results of organic trace substance analysis in vegetation and humus samples are given in Figs. 3 and 4. The four polyaromatic hydrocarbons and all the chlorinated hydrocarbons showed concentrations in the ng g-1 range with highest absolute values for fluoranthene, which is characterised by high data


W. Thomas, A. Riihling, H. Simon








.,i I,,, 30 , I I

. . . .





Fig. 3.

. I I

. . .


Concentrations of PAH (ngg ') in different plant species and humus from

Stenungsund (see Table 1 for explanation of abbreviations). (A) Fluoranthene; (B) 3,4benzopyrene; (C) benzo(ghi)perylene;(D) indeno(l,2,3-cd)pyrene. variation between 2 ng g- 1 (Betula, Vaccinium) and 170 ng g- 1 (raw humus). The other three PAH ranged between 1 and 100 ng g- 1. As mentioned above, chlorinated hydrocarbon levels were detected in the lower ng g-1 level in the vegetation samples with, generally, highest values for e-BHC, followed by 7-BHC, PCBs and HCB. The relationships between the e- and 7-BHC levels differ from results of moss analyses from central Europe, in that there the levels of Lindane (7-BHC) were 2-5 times lower in samples (Thomas & Herrmann, 1980).

Accumulation of airborne pollutants in plants



i li,,,l,l,,,


Fig. 4. Concentrations of chlorinated hydrocarbons (ng g-1) in different plant species and humus from Stenungsund (seeTable 1 for explanation of abbreviations). (A) ~-BHC; (B) ?-BHC; (C) HCB; (D) 5-PCB; (E) 6-PCB. The gas chromatographic and mass spectrometric analysis of the PCB fraction from the vegetation samples showed a number of substances of the PCB group. All identified peaks together made up a concentration of about 20 ng g-1 for the lichen sample from Stenungsund (Fig. 2). The comparison of data from all vegetation samples for PCBs was made with 2 PCB-peaks, the 2, 2', 3, 5, 6-pentachlorobiphenyl (Fig. 2b) and 2, 2', 3, 3',6,6'-hexachlorobiphenyl (Fig. 2c). Although the PCB with five chlorine atoms was generally more highly enriched in most of the samples, the total number of PCBs with six chlorines dominated. From the PCBpattern of the samples it can be concluded that Stenungsund PCB emissions results from the use o f a PCB mixture with about 60 ~o chlorine. According to Strek & Weber (1982) these kinds of PCB are leachable from soil in lower amounts than less chlorinated biphenyls.

W. Thomas, A. Riihling, H. Simon


TABLE 2 Variation of Data from PAH-Analysis Within One Sample of Litter from Coniferous Trees--Results of Five Replicates (in ng g- 1)

Sampling site Stenungsund



43 42 43 41 43 42 0.8



53 52 48 49 55 51 2.6



43 47 42 52 51 47 4.1

170 165 150 164 173 164 7.9

Data variation within one sample

The analysis of intraspecies variation for the trace organics detected was carried out for the raw humus sample of coniferous forest from the Stenungsund site. Five subsamples were analysed separately. As the results of these analyses indicate (Tables 2 and 3), relatively small variations within one sample were found in comparison with the differences between sites.

TABLE 3 Variation of Data from Chlorinated Hydrocarbon-Analysis Within One Sample of Litter from Coniferous Trees Results of Five Replicates (in ngg -1)

Sampling site Stenungsund

I I1 II1 IV V £ s


11"3 12"8 11"7 12'8 12'2 12"2 0"6

6"1 7.6 5"7 7.8 8.6 7"2 1.1


2-1 1.6 2.1 1.7 2.2 1.9 0.2

2,2',3,5,62,2',3,3',6,6'Pentachloro- Hexachlorobiphenyl biphenyl 8'7 10.7 8'1 9.2 11'2 9-6 1.2

2"1 2.9 3"3 2.2 3.1 2'7 0"5

Accumulation of airborne pollutants in plants




. . . Illtll ] ,,I I i,il B






~}~~~-~ - -

~8~~~, 0




Fig. 5. Concentrations of heavy metals (#g g- 1) in different plant species and humus

from Stenungsund(see Table 1 for explanation of abbreviations). (A) Zn; (B) Pb; (C) Cu; (D) Cd, (E) V. Heavy metal concentrations Although a number of studies exist on heavy metal accumulation in different vegetation samples, the results of metal analyses are also presented in Fig. 5 for comparison with the trace organics. This comparison can help to detect the effect of the different atmospheric emission processes and chemical properties of all three pollutant groups.


W. Thomas, A. Riihling, H. Simon

The heavy metals Zn, Pb, Cu, Cd and V were not excessively elevated at both investigation sites compared to areas near metal smelting factories (Little & Martin, 1972).

Principal component analysis Principal component analysis using the data for all pollutants was carried out for the investigation site (Table 4). One can then determine which groups of pollutants are characterised by similar concentration ratios of the individual species, despite different absolute values. On the other hand, there are only slight correlations in accumulation behaviour between different major pollutants. Thus high loadings of different pollutants combined on one principal component can result from (a) a similar deposition and atmospheric transport behaviour and (b) comparable processes for the accumulation of these pollutants in the samples. The matrix loading presented here is characterised by a clear structure with > 9 5 ~ of the variance on six principal components (PC). The TABLE 4

Matrix of Significant (1 ~ Level) Organic Trace Substance and Heavy Metal Concentrations (the Principal Component Loadings) in Different Plant Species and Humus (Varimax-Rotated Version) Investigation Area Stenungsund Principal component oJvariance

1 45"3

Benzo-ghi-perylene 3,4-Benzo-a-pyrene Indenopyrene Fluoranthene ~-BHC 7-BHC HCB Pentachlorobiphenyl Hexachlorobiphenyl Cd Cu Pb V Zn

0.93 0.96 0.82 0-97

2 20"5

3 11"1

4 7.4

5 6"2

6 5"9 Z = 9 6 ' 4 %

0'88 0"96 0"91 0'90 0'93 0.92 0"96 0.91 0.93 0"88

Accumulation of airborne pollutants in plants


greatest amount of information is given by the first PC with all PAH and the metals Pb and V. Similar accumulation behaviour can also be recognised from the pollutants of the second PC, ~-, y-BHC, 5-PCB, separated from the uptake pattern of Zn and Cu (third PC). The remaining three pollutants are each characterised by their own data structure. DISCUSSION As indicated by the first principal component, the accumulation behaviour of all four PAH as well as between PAH, Pb and V is very similar. Figures 3 and 5 show that these pollutants are mainly characterised by an elevated accumulation in the raw humus samples, nearly equal for coniferous and deciduous forests, According to the absolute concentrations these samples are followed by lower plants, mosses and lichens. The comparison of the three groups of samples, humus, lower plants and higher plants, shows that the accumulation rate integrated by a sample, as well as the surface-to-mass ratio of the species, are the dominant factors for the enrichment of these groups of pollutants. On the other hand accumulation through root systems seems not to be important. The accumulation rate of the raw humus can be regarded as representing a relatively long time for all these pollutants which are stable and undergo no leaching processes. The non-polar PAH are emitted and distributed in the atmosphere adsorbed to particulate matter. Their physicochemical properties ensure that they are deposited on to plant surfaces and humus over a longer period. Microbiological decomposition in higher amounts has not been detected; the only loss of substance occurs by means of photochemical processes (Neff, 1979). In studies by Thomas (1981) it was shown that both PAH and Pb are translocated in the atmosphere over longer distances than most other metals, which means that they are adsorbed to particles of relatively small size classes. This is in agreement with the investigation of Goodman & Roberts (1971), who found that Pb showed a more gradual regional decrease of atmospheric concentration values around a point-source than Zn and Cd. In the present investigation also, Zn, Cd and Cu are deposited differently from PAH, Pb and V. As has been shown above, and according to Little & Martin (1972) and Chamberlain (1983), the contamination of vegetation by lead takes place


W. Thomas, •. Riihling, H. Simon

mainly by retention of particulate matter. That means that the surface of a plant acts as a kind of mechanical filtering system and is more effective the larger the surface-to-mass ratio. These kinds of accumulation processes are also the dominant factors for the PAH uptake by plants. The other three metals measured, Zn, Cu and Cd, show a completely different accumulation pattern from the first group. They are characterised by lower accumulation in the humus samples and lower plants than in the other species, a selective enrichment of Zn and Cu in fungi at Stenungsund, and partly elevated Zn and Cd levels in birch leaves and bark. A different atmospheric transport mechanism with adsorption of these heavy metals to a wider range of particle size classes, and the different behaviour in sorption and retention of the particles, is responsible for this fact. According to Rfihling & Tyler (1970), mosses have a lower capacity for the sorption of Zn in relation to Pb: furthermore, relations between humus and moss accumulations of Zn, Cd, and Cu are nearly equal (Tyler, 1972). A selective uptake of some metals by fungi was also found by Tyler (1980). Yet another different accumulation pattern for the individual substances can be seen for the chlorinated hydrocarbons with high loadings of ~- and 7-BHC as well as 6-PCB on the second, and for HCB and 5-PCB on the fifth and sixth principal components. BHCs and 6-PCB pattern are characterised by the highest enrichment values onto the birch bark, followed by the moss Hypnum and the raw humus of coniferous forest. Juniperus and Vaccinium show elevated levels as do the pine-needle samples. In relation to the other pollutant groups significant differences exist between 1- and 2-year-old pine-needles. Enrichment factors between 1.1 and 2.9 of second/first year needles could be detected for chlorinated hydrocarbons. Although volatilisation rates of these pollutants are relatively high, and therefore a higher amount occurs in the atmosphere as gas, it is obvious that a number of species can accumulate them to a high level. It is also obvious that the highest residue values were found in biological material which contains etheric oils such as terpenes, etc. It is well known that these pollutants are fat soluble and are therefore retained by these plant materials, i.e. protected against remobilisation. CONCLUSIONS As indicated by a principal components analysis, the pollutants analysed have different accumulation patterns in vegetation and raw humus

Accumulation of airborne pollutants in plants


samples. The enrichment of PAH, Pb and V at the two sampling sites is characterised by the deposition of atmospheric particulate matter with relatively small particle sizes. The main factors for the enrichment of these substances by the different species are their surface-to-mass ratios. The other metals, Zn, Cd, and Cu, are characterised by lower retention capacities of the vegetation and a selective uptake mechanism for the fungi species analysed. Chlorinated hydrocarbons can be accumulated most effectively by species that contain fatty and aromatic substances in which these pollutants are soluble. The use of lower plants as monitoring organisms of atmospheric pollutants is especially valuable for PAH and heavy metals. Chlorinated hydrocarbon emissions can probably be monitored more sensitively by pine needles, birch bark, etc.

ACKNOWLEDGEMENTS The authors are grateful to the German Research Society (DFG) and the University of Bayreuth for financial support of the investigations. Furthermore we would like to thank E. Chinta and K. Keil for preparing the figures, E. Misch for typing and M. Tyzenhouse for reviewing the manuscript.

REFERENCES Chamberlain, A. C, (1983), Fallout of lead and uptake by crops. Atmos. Environ., 17, 693-706. Goodman, G. T. &Roberts, T. M. (1971), Plants and soils as indicators of metals in the air. Nature, Lond., 231,287-92. Hanssen, J. E., Rambaek, J. P., Semb, A. & Steinnes, E. (1980). Atmospheric deposition of trace elements in Norway. In Proceedings oJthe International Conjerence on the Ecology oj Impact Acid Precipitation, Norway, March 1980, 116-17. Herrmann, R. (1978). Regional patterns of polycyclic aromatic hydrocarbons in NE-Bavarian snow and their relationships to anthropogenic influence and air flow. Catena, 5, 165-75. Huckabee, J. W., Sanz Diaz, F., Janzen, S. A. & Solomon, J. (1983). Distribution of mercury in vegetation at Almad+n, Spain. Environ. Pollut., Set. A, 30, 211-24.


W. Thomas, ~. Riihling, H. Simon

Lee, R. E. (1976). Air pollutionjrom pesticides and agricultural processes. Ohio, CRC Press. Little, P. (1973). A study of heavy metal contamination of leaf surfaces. Environ. Pollut., 5, 159-72. Little, P. & Martin, M. H. (1972). A survey of zinc, lead and cadmium in soil and natural vegetation around a smelting complex. Environ. Pollut., 3, 241-54. Neff, J. M. (1979). Polycyclic aromatic hydrocarbons in the aquatic environment. London, Applied Science. Pilegaard, K. (1979). Heavy metals in bulk precipitation and transplanted Hypogymnia physodes and Dicranoweisia cirrata in the vicinity of a Danish steelworks. Water, Air, Soil Pollut., 11, 77-91. Rfihling, A. & Tyler, G. (1968). An ecological approach to the lead problem. Bot. Notiser, 121,321-42. Riihling, A. & Tyler, G. (1970). Sorption and retention of heavy metals in the woodland moss Hylocomium splendens (Hedw.) Br. et Sch. Oikos, 21, 92-7. Skaar, H., Ophus, E. & Gullvhg, B. M. (1973). Lead accumulation within nuclei of moss leaf cells. Nature, Lond., 241,215-16. Strek, H. J. & Weber, J. B. (1982). Behaviour of polychlorinated biphenyls (PCBs) in soils and plants. Environ. Pollut., Ser. A., 28, 291-312. Thomas, W. (1979). Monitoring organic and inorganic trace substances by epiphytic mosses--a regional pattern of air pollution. In Proceedingsiofthe 13th International Conferences on Trace Substances in Environmental Health, Columbia, June 1979, 285-9. Thomas, W. (1981). Concentrations and total inputs of PAH, chlorinated hydrocarbons and trace metals in bulk precipitation--comparison of suburb and rural stations. Dtsch. Gewiisserkundl. Mitt., 25(5/6), 120-9. Thomas, W. (1983). l]ber die Verwendung von Pflanzen zur Analyse r~iumlicher Spurensubstanz-Immissionsmuster. Staub-Reinhalt. Lufi., 43, 141-8. Thomas, W. & Herrmann, R. (1980). Nachweis von Chlorpestiziden, PCB, PCA und Schwermetallen mittels epiphytischer Moose als Biofilter entlang eines Profils durch Mitteleuropa. Staub-Reinhalt. Lufi, 40, 440-4. Tyler, G. (1972). Heavy metals pollute nature, may reduce productivity. Ambio, 1, 52-9. Tyler, G. (1980). Metals in sporophores of basidiomycetes. Trans. Br. Mycol. Soc., 74, 41-9.