Environ. Sci. Technol. 2002, 36, 3224-3229
Air-Side and Plant-Side Resistances Influence the Uptake of Airborne PCBs by Evergreen Plants J O N A T H A N L . B A R B E R , * ,† GARETH O. THOMAS,† GERHARD KERSTIENS,‡ AND KEVIN C. JONES† Departments of Environmental and Biological Sciences, Institute of Environmental and Natural Sciences, Lancaster University, Lancaster, LA1 4YQ, UK
The transfer of persistent organic pollutants (POPs) from air to vegetation is an important air-surface exchange process that affects global cycling and can result in human and wildlife exposure via the terrestrial food chain. To improve understanding of this process, the uptake of gas-phase polychlorinated biphenyls (PCBs) by two slow-growing evergreen shrubs, Skimmia japonica Thunb. and Hebe “Great Orme”, was studied to investigate the influence of airside and plant-side resistances. Uptake of PCBs was monitored over periods of hours, days, and weeks. Uptake rates were higher in the smaller Hebe leaves than the Skimmia leaves. Equilibrium was not attained between air and plants in the duration of the experiments; uptake curves were indicative of a two-phase uptakesstep 1 over the order of hours and step 2 continuing steadily over days to weeks. Uptake rates (h-1) were greater in conditions simulating typical ambient wind speeds (2 m s-1) than under still air, indicating a significant impact of air-side resistance relative to plant-side resistance in still air. Wind speed is an important variable that has not been previously considered in studies of the air-plant transfer of persistent organic pollutants (POPs). Uptake rate constants increased with increasing level of chlorination (and hence KOA) both in still air and under turbulent conditions. This was inconsistent with the idea of air-side resistance dominating uptake, since diffusion rates in air decrease with molecular weight (and hence KOA). Greater uptake of particlebound PCBs may have contributed to this finding, but the most likely explanation is the previously established relationship that the permeability of cuticles increases with increasing KOA of the diffusing chemical. The findings indicate that plant-side resistance can have an important effect on uptake rates of different PCB congeners in the field, even when air-side resistance is high.
Introduction The transfer of persistent organic pollutants (POPs) from air to vegetation is the important first step leading to human and wildlife exposure via the terrestrial food chain (1). Vegetation also affects air-surface exchange processes ands * Corresponding author phone: (44)-1524 593 957; fax: (44)-1524 593 985; e-mail:
[email protected]. † Department of Environmental Sciences. ‡ Department of Biological Sciences. 3224
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 15, 2002
hencesglobal cycling of POPs (2). However, despite its importance, there are still some important uncertainties in our knowledge about the processes of air-plant exchange and the precise role of vegetation in influencing the fate of POPs. McLachlan (3) has produced an important conceptual framework, which gives guidance on how to view the transfer of POPs from air to plants. The deposition of particle bound POPs occurs as particulates settle out of the bulk air, through the atmospheric boundary layer, on to the plant leaf surface (4). Gas-phase POPs must move through the bulk air, pass through the boundary layer next to the leaf surface, and partition into the plant (5). It is generally assumed that gasphase lipophilic POPs principally partition into the cuticle layer (a polymeric lipid structure encrusted with and overlain by waxes on the plant surface) and that this is where the majority of the plants’ POP burden is retained (6). Particulatebound POPs presumably may partition off the particle into the cuticle or remain associated with the particle, whichsin turnsmay be retained on the surface, washed off, or shed along with cuticular material that is sloughed off the leaf surface as it ages or is abraded. Once POPs reach the plant surface, they may diffuse into the cuticle (7), thereby reaching interior parts of the plant. Dry gaseous deposition has been shown to be the dominant deposition process for PCBs (8), a representative class of POPs, with a wide range of physicochemical properties; their log KOA values vary between ∼8.4 for PCB 18 and ∼11.6 for PCB 203 at 10 °C (9). One continuing area of uncertainty concerns the time plants may take to reach equilibrium with gas-phase POPs in the atmosphere. Equilibrium between plant and air may be attained, and this can be observed from kinetics studies when plant concentrations stop increasing. However, constant plant concentrations may also arise under nonequilibrium steady-state conditions, where uptake is still continuing, but concentrations are exactly diluted by growth. The transfer of gas-phase POPs to vegetation has been described by a two-resistance model, composed of a boundary layer resistance around the leaf, and a resistance to diffusion within the leaf (10, 11). The supply of PCBs from the bulk air is considered the rate-limiting step in air/plant transfer in the McLachlan framework (3) and is believed to be slow enough that it may prevent plants from reaching equilibrium with the air in the lifetime of the leaf (e.g. during the growing season). This is supported by field experiments in which equilibrium was not attained after a period of months in grasses (12), or years in pine needles (13, 14). The plant leaf has been modeled as a one-compartment (12) or a two-compartment system (15) with respect to POPs, in the latter case consisting of a surface and a reservoir compartment within the leaf. However, the true spatial location of these conceptual compartments is poorly understood, and it is difficult to relate them to morphological structures within the leaf. Hung et al. (16) used a two-compartment model to describe their field experimental data for grass. The model had a small surface adsorption (surface wax) compartment that attained equilibrium with air within hours, and a larger internal compartment, corresponding to the bulk cuticle or inner leaf, which was still taking up PCBs after many days. Presumably it is possible in plants such as grasses, with a thin cuticle (i.e. a small internal compartment), to achieve equilibrium between the outer and inner compartments relatively rapidly. Indeed, some field experiments show steady-state concentrations being attained after a few days for PCBs (17) and low molecular weight PAHs in grass (18) or weeks in tree leaves for PAHs and PCBs (19, 20). 10.1021/es010275u CCC: $22.00
2002 American Chemical Society Published on Web 06/22/2002
It is interesting to consider why some studies showed steady-state concentrations being attained relatively quickly (17-20), while others did not (12-14). Possible explanations include site-, species-, or methodological-differences between the studies. One hypothesis is that wind speed may be an important variable between sites (17, 21), because the rate of supply from the bulk air and diffusion through the thinner atmospheric boundary layer would be more rapid at windier locations, for example (8, 17). Another hypothesis relates to the diversity in the thickness and structure of plant cuticles, which vary greatly (22). The limiting factor controlling plant concentration in some situations may conceivably be influenced by the nature of the plant’s lipid pool via effects on plant-side resistance or the capacity for accumulation relative to surface area (23). It has been shown that the properties of the cuticular skin, not cuticle thickness, determine the cuticular barrier properties (6). However, if the bulk cuticle is the main lipophilic phase in the leaf, then it will take longer to bring a leaf with a thick cuticle into equilibrium with the atmosphere than a leaf with a thin cuticle, as the surface/volume ratio is smaller for a thick cuticle. Cuticle structure and composition can vary between individual plants of the same species, depending on the environmental conditions under which they grow. Further complications arise in studying air-plant transfer and establishing general rules from which to derive models, because environmental conditions (i.e. wind speed, temperature, particle loadings) vary widely in space and time; plants grow and change structurally through the seasons; compound properties control partitioning between the gas, particle, leaf, and soil phases; and compounds may be prone to physical, chemical, and biochemical loss processes while on or in plant leaves. The objectives of the study presented here were to investigate the uptake of a wide range of PCBs by two slowgrowing evergreen plant species with different morphologies and to investigate the influence of wind speed on the uptake kinetics. These two objectives were addressed, to provide information on the relative importance of air-side and plantside resistance to uptake. As noted above, PCBs can be used to study uptake of gas-phase compounds, without substantial complications caused by particle deposition (4, 7). Uptake was studied by transferring plants initially grown in background air to a room with stable indoor air PCB concentrations elevated several 100-fold above typical, ambient outdoor concentrations (24). This made it possible to clearly see plant concentrations changing over periods of hours/days in response to the elevated air concentrations.
Materials and Methods Experiment 1: Comparison of Uptake by Two Plant Species. An experiment was conducted in March/April 1999 with two species of evergreen shrub, Skimmia japonica Thunb., a largeleafed waxy plant with thick cuticles, and Hebe “Great Orme” (a hybrid of H. speciosa and H. salicifolia), a small-leafed waxy plant. Plants were purchased from a rural outdoor garden center and transferred into a room (with an air volume of 350 m3) that “naturally” experiences stable, elevated PCB concentrations. The plants were maintained under a constant 2 m s-1 flow of air using electric fans (wind speed measured with a hand-held digital anemometer). This wind speed represents typical ambient conditions, which can range from 0.8 were obtained after a few hours for Hebe. With the number of data points on the line, r2 values over 0.13 were significant at p < 0.05 and r2 values over 0.22 were significant at p < 0.01. Apparently straight lines were also obtained for the Skimmia but with lower r2 values (data not shown). In the second experiment (Skimmia only) straight lines with r2 values of >0.8 were obtained for the fanned plants after several hours and the unfanned plants after 2 weeks. For a given time point in experiment 2, the gradient of the line was greater for the fanned plants than for the unfanned plants (see Figure 4 a). Although in previous studies (8, 17, 18) we have interpreted this straight line to mean that the plant has reached equilibrium with the air, uptake was clearly still occurring in this study. Therefore a straight line in these plots is not a good indication that equilibrium conditions are occuring. The correct interpretation of these plots may be that the surface compartment of the leaf has reached equilibrium or a steady state with the air, but the internal compartment(s) were still accumulating chemicals, via diffusion from the surface compartment (16). The slopes of the plots increased with time, from a value of ∼0.3 after a few hours to 0.67 after 2 weeks in fanned plants and from a value of ∼0.1 after a few hours to 0.67 after 10 weeks in unfanned plants (see Figure 4b). The difference in slopes reported by many authors (8, 17, 19, 34, 35) may indicate that in their experiments nonequilibrium still prevailed. Air-Side Resistance. The data from this study provides evidence that air-side, rather than plant-side, resistance controlled the transfer of gas-phase PCBs to these plant species under conditions of still air. As noted earlier, the VOL. 36, NO. 15, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3227
δbl (in mm), can be estimated from
δbl ) 4.0 × 10-3 *
(Lv)
1/2
(4)
where L is mean length of leaf in the downwind direction (in m), and v is ambient wind speed (in m s-1). Hebe had much smaller leaves than Skimmia, which according to eq 4 results in a smaller boundary layer. (Values for L of 0.06 m for Skimmia and 0.02 m for Hebe and a wind speed of 2 m s-1 give values of δbl ) 0.7 × 10-3 and 0.4 × 10-3 m for Skimmia and Hebe, respectively, for leaves whose axis was parallel to the wind direction). Small leaves therefore normally have higher boundary layer conductances than large leaves and so are more closely coupled to the environment (39). This could explain why Hebe plants took up PCBs more rapidly than Skimmia plants. However, this factor of 2 difference in boundary layer thickness alone does not account for the greater difference measured in uptake rate constants (Table 1a). In the bulk air, eddy diffusion dominates over molecular diffusion (39) and is the same for all molecules in a given situation, because a small pocket of air moves as a unit (36). The eddy diffusion coefficient is usually proportional to the local wind speed. For a wind speed of 2 m s-1 an eddy diffusion coefficient is generally 0.05-0.2 m2 s-1 just above the plant canopy (104-105 times higher than molecular diffusion coefficients in air) (36).
FIGURE 4. “Scavenger” plots for Skimmia plants for (a) fanned and unfanned plants after 1 week exposure and (b) unfanned plants after 1, 4, and 12 weeks exposure. air-side transfer of compounds to leaves involves two steps: the transfer from the bulk air to the leaf boundary layer air and diffusion across the boundary layer. Air immediately around the leaf is stationary, so gas-phase compound transfer is purely diffusive. Outside this static boundary layer, with increasing distance from the leaf surface, there is a transition from laminar flow parallel to the leaf to turbulent eddy movement in the bulk air (36). POP movement by molecular diffusion in the boundary layer can be estimated from
A ∆Cbl dC ) D A* * dt V δbl
(2)
where DA is the diffusion coefficient in air (in m2 s-1), A is area of leaf surface (in m2), V is the volume of the leaf (in m3), ∆Cbl is the difference in concentration across the boundary layer, and δbl is the boundary layer thickness (in m). The diffusion coefficient in air (in m2 s-1) is dependent on the molar volume (MV) of the compound and can be estimated from Schwarzenbach et al. (37) as
DA )
1.55 × 10-4 MV0.65
(3)
DA can be calculated for two contrasting congeners, PCB 18 and PCB 203, with molar volumes of 247.3 cm3 mol-1 and 351.8 cm3 mol-1 (38), to be 0.43 × 10-5 m2 s-1 and 0.34 × 10-5 m2 s-1, respectively. This small difference implies that there should only be a slight difference in uptake rate between the two congeners, if diffusion across the boundary layer is the limiting process. The boundary layer thickness around leaves, 3228
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 15, 2002
If air-side resistance limits plant uptake of POPs, it would be expected that the plants in fanned air would have a faster uptake rate than those in unfanned air, because eddy diffusion and boundary layer thickness are inversely related to wind speed (40). However, wind speed affects boundary layer thickness equally for all congeners and diffusion rates in air differ only by a factor of 2 or so. Therefore, the large difference between PCB-31 and PCB-180 in the amount that fanning the plants increased the uptake rate cannot be explained by a decrease in boundary layer thickness alone. It is possible that fanning the plants increased the amount of particles that they intercepted, which could explain the increase in uptake rate with increasing KOA, since the proportion of particle-bound PCBs increases with KOA. However a later air sampling program of the gas- and particlephase PCBs in the room where the experiments were performed showed that only the octa-chlorinated biphenyls had a fraction that were >5% particle-associated. Therefore, increased particle impaction will not have had a significant effect on the rates of uptake for most of the PCB congeners studied. Plant-Side Resistance. If plant-side resistance was limiting uptake, it should have had a predictable effect on the uptake rate constants. It has been reported that resistance to compound diffusion through the cuticle is related to molar volume for pesticides in isolated cuticles (41), which might therefore result in a decrease in diffusion rate of PCB congeners with increasing level of chlorination. However, it has also been shown that cuticular permeability increases with KOA (5, 27, 41). This property varies much more widely across the range of PCBs than molar volume (by 3 orders of magnitude in contrast to by a factor of 2), and this relationship is in agreement with the observation that the uptake rate constants increased with KOA (see Table 1). It can therefore be concluded that plant-side resistance also had a controlling effect on the uptake of gas-phase PCBs in Skimmia plants. Plant-side resistance will vary between plant species with structural differences, such as surface area per unit volume, cuticle wax morphology, and cuticular permeability (6). Total resistance to uptake of PCBs is therefore influenced by three factors: the effect of wind speed on air-side resistance, the nature of the plant leaf and its associated plant-side
resistance, and the KOA of the different PCB congeners which affects the relative importance of the air-side and plant-side resistances. General Remarks on the Effect of Air Movement on Uptake Kinetics. This study supports the earlier work by McLachlan and co-workers (4) by showing that air-side resistance limits the uptake of gas-phase POPs under stillair conditions, although our results show that plant-side resistance can also be important under natural (windy) conditions. Air-side resistance will be greatest under still air conditions and reduced under windier conditions as wind affects the rate of gas-phase POP uptake, by reducing the boundary layer thickness and/or by increasing eddy diffusion. The importance of air-side resistance will therefore vary, according to the wind speed, with uptake kinetics falling into two distinct zones, where either air-side or plant-side resistances dominate. Interspecies differences in uptake rates (as illustrated by this study) may be influenced by differences in plant morphology affecting boundary layer thickness and/ or altering the aerodynamic surface roughness of the plant (42). Several chamber studies have been conducted previously on POP uptake (e.g. 11, 32, 33). Different authors have used different air-flow rates without always considering the effect this may have on uptake kinetics and how they may relate to field conditions. Future studies should be carefully designed, to ensure that they mimic typical field conditions. The fanned treatment in this study mimicked wind speed conditions of 2 m s-1. However, sites with higher average wind speeds are commonplace, while gusty conditions can frequently give wind speeds well above 10-20 m s-1 (25). Air-side resistance will be lowered considerably under these conditions, leading to enhanced uptake rates of gas-phase POPs in cases where plant-side resistance is colimiting, but not controlling, uptake.
Acknowledgments We would like to thank the Food Contaminants Division of the UK Food Standards Agency (FSA) for funding this work. We also thank Dr. Wendy Ockenden (formerly of Lancaster University) and Dr. Tom Harner of Meteorological Services Canada (MSC), for their contributions to the discussion from which this experiment was born. We are also grateful for comments from five anonymous reviewers.
Literature Cited (1) McLachlan, M. S. Environ. Sci. Technol. 1996, 30, 252-259. (2) Bennett, D. H.; McKone, T. E.; Matthies, M.; Kastenberg, W. E. Environ. Sci. Technol. 1998, 32, 4023-4030. (3) McLachlan, M. S. Environ. Sci. Technol. 1999, 33, 1799-1804. (4) Welsch-Pausch, K.; McLachlan, S.; Umlauf, G. Environ. Sci. Technol. 1995, 29, 1090-1098. (5) Riederer, M. In: Plant contamination: Modeling and simulation of organic chemical processes; Trapp, S., McFarlane, J. C., Eds.; Lewis Publishers: London, 1995; 153-190. (6) Riederer, M. Environ. Sci. Technol. 1990, 24, 829-837. (7) Scho¨nherr, J.; Riederer, M. Rev. Environ. Contam. Toxicol. 1989, 108, 1-70. (8) Thomas, G. O.; Sweetman, A. J.; Ockendon, W. A.; Mackay, D.; Jones, K. C. Environ. Sci. Technol. 1998, 32, 936-942. (9) Harner, T.; Bidleman, T. F. J. Chem. Eng. Data 1996, 41, 895899.
(10) Paterson, S.; Mackay, D.; Gladman, A. Chemosphere 1991, 23, 539-565. (11) Ko¨mp, P.; McLachlan, M. S. Sci. Total Environ. 2000, 250, 6371. (12) McLachlan, M. S.; Welsch-Pausch, K.; Tolls, J. Environ. Sci. Technol. 1995, 29, 1998-2004. (13) Jensen, S.; Eriksson, G.; Kylin, H. Chemosphere 1992, 24, 229245. (14) Sinkkonen, S.; Welling, L.; Vattulainen, A.; Lahti, L.; Lahtipera¨, M.; Paasivirta, J. Chemosphere 1996, 32, 1971-1982. (15) Tolls, J.; McLachlan, M. S. Environ. Sci. Technol. 1994, 28, 159166. (16) Hung, H.; Thomas, G. O.; Jones, K. C.; Mackay, D. Environ. Sci. Technol. 2001, 35, 4066-4073. (17) Thomas, G. O.; Smith, K. E. C.; Sweetman, A. J.; Jones, K. C. Environ. Pollut. 1998, 102, 119-128. (18) Smith, K. E. C.; Thomas, G. O.; Jones, K. C. Environ. Sci. Technol. 2001, 35, 2156-2165. (19) Simonich, S. L.; Hites, A. Environ. Sci. Technol. 1994, 28, 939943. (20) Simonich, S. L.; Hites, A. Environ. Sci. Technol. 1995, 29, 29052914. (21) Bakker, M. I.; Tolls, J.; Kollo¨ffel, C. In Persistent, bioaccumulative and toxic chemicals I: Fate and exposure; Lipnick, R. L., Hermens, J. L. M., Jones, K. C., Muir, D. C. G., Eds.; American Chemical Society Symposium series: 2000; pp 218-236. (22) Price, C. E. In The plant cuticle; Cutler, D. F., Alvin, K. L., Price, C. E., Eds.; Academic Press: New York, 1982; pp 237-251. (23) Bo¨hme, F.; Welsch-Pausch, K.; McLachlan, M. S. Environ. Sci. Technol. 1999, 33, 1805-1813. (24) Alcock, R. E.; Halsall, C. J.; Harris, C. A.; Johnston, A. E.; Lead, W. A.; Sanders, G.; Jones, K. C. Environ. Sci. Technol. 1994, 28, 1838-1842. (25) Yin, X. J. Appl. Meteorol. 2000, 39, 1861-1865. (26) Holloway, P. J.; Baker, E. A. Plant Physiol. 1968, 43, 1878-1879. (27) Kerler, F.; Riederer, M.; Scho¨nherr, J. Physiol. Plant. 1984, 62, 599-602. (28) Thomas, G. O.; Sweetman, A. J.; Parker, C. A.; Kreibich, H.; Jones, K. C. Chemosphere 1998, 36, 2447-2459. (29) Lee, R. G. M.; Hung, H.; Mackay, D.; Jones, K. C. Environ. Sci. Technol. 1998, 32, 2172-2179. (30) Lee, R. G. M.; Jones, K. C. Environ. Sci. Technol. 1999, 33, 705712. (31) McCrady, J. K. Chemosphere 1994, 28, 207-216. (32) Bacci, E.; Gaggi, C. Chemosphere 1987, 16, 2515-2522. (33) McCrady, J. K.; Maggard, S. P. Environ. Sci. Technol. 1993, 27, 343-350. (34) Ko¨mp, P.; McLachlan, M. S. Environ. Sci. Technol. 1997, 31, 2944-2948. (35) Ockenden, W. A.; Steinnes, E.; Parker, C.; Jones, K. C. Environ. Sci. Technol. 1984, 18, 2721-2726. (36) Nobel, P. S. Physicochemical and environmental plant physiology, 2nd ed.; Academic Press: San Diego, 1999. (37) Schwarzenbach, R.; Geschwend, P.; Imboden, D. Environmental Organic Chemistry; Wiley: New York, 1993. (38) Mackay, D.; Shiu, W. Y.; Ma, K. C. Illustrated Handbook of Physical-Chemical Properties and Environmental Fate for Organic Chemicals; Lewis Publishers: Ann Arbor, 1992. (39) Grace, J. In Plants and their atmospheric environment; Grace, J., Ford, E. D., Jarvis, P. G., Eds.; Blackwell Scientific Publications: Oxford, 1981; pp 31-56. (40) Hauk, H.; Umlauf, G.; McLachlan, M. S. Environ. Sci. Technol. 1994, 28, 2372-2379. (41) Schreiber, L.; Scho¨nherr, J. Pestic. Sci. 1993, 38, 353-361. (42) Bakker, M. I. Doctoral Dissertation, University of Utrecht, 2000.
Received for review October 29, 2001. Revised manuscript received May 22, 2002. Accepted May 23, 2002. ES010275U
VOL. 36, NO. 15, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3229
