Journal of Environmental Quality 30:2081-2090 (2001)
© 2001 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
TECHNICAL REPORT
Plant and Environment Interactions
Plant Contamination by Organic Pollutants in Phytoremediation
Kijune Sunga,
M.Yavuz Corapcioglu*,a,
M.C. Drewb and
C.L. Munsterc
a Dep. of Civil Engineering, Texas A&M Univ., College Station, TX 77843-3136
b Dep. of Horticultural Science, Texas A&M Univ., College Station, TX 77843-2133
c Dep. of Agricultural Engineering, Texas A&M Univ., College Station, TX 77843-2117
* Corresponding author (yavuz{at}acs.tamu.edu)
Received for publication December 15, 2000.
 |
ABSTRACT
|
|---|
Phytoremediation is a remediation technique that involves plant uptake, transformation, accumulation, and/or volatilization of soil- and aqueous-phase contaminants or by the stimulation of microbial cometabolic activity in the rhizosphere of the plant. Even when the principal mechanism is by stimulation of bacteria, any resultant plant contamination cannot be overlooked. For the purpose of modeling, a two-compartment plant model has been developed. The model divides the plant into the shoot compartment (which can be harvested) and the root compartment (into which contaminants can accumulate). Numerical experiments were conducted to investigate model behavior and to determine important parameters affecting plant contamination. Johnsongrass [Sorghum halepense (L.) Pers.] was used to evaluate the model behavior. The contaminants TNT (2,4,6,-trinitrotoluene) and chrysene were selected on the basis of their contrasting aqueous-phase solubilities. The results indicate that plant contamination and soil remediation by plants depend on soil properties such as soil organic carbon content, the physicochemical properties of the contaminants such as the octanol–water partition coefficient, and plant properties. The most important factor affecting plant contamination is bioavailability. As bioavailability increased, the concentrations in root and shoot compartments were predicted to increase. Microbial activities and plant contamination are closely related, which suggests that plants and microorganisms can have complementary roles in phytoremediation.
|
INTRODUCTION
|
|---|
PLANTS play an important role in the fate and transport of organic and inorganic contaminants in vegetated soil (Paterson et al., 1990). They can be major pathways for entry of pesticides and hazardous contaminants into the food chain. As these contaminants move to higher trophic levels, they are accumulated in the body of each organism and the concentration may magnified if the organism cannot metabolize or discharge the contaminant. Bioconcentration is of great interest because it affects human health and the ecological system. The "itai itai" disease is a well-documented example of the consequence of bioconcentration of cadmium-contaminated rice (Oryza sativa L.) (Masters, 1974).
Phytoremediation is the technique that uses trees and plants to enhance bioremediation. The principal mechanisms are either stimulation of soil microbial activity and degradation of contaminants and/or plant uptake of contaminants or their degradation products. Phytoremediation is potentially always associated with plant contamination; therefore, information about contaminant distribution and concentration in the plant is essential in predicting the effectiveness of a phytoremediation operation to remove and process these contaminated plants. If a contaminant transported into the plant is mainly localized in the root, the plant plays a significant role as a stabilizer rather than as an extractor in this case because most of the contaminants are still in the subsurface. If the contaminant is transported to the aboveground part, we can easily remove the contaminant from the soil by disposing of the shoot. Therefore, protection of the remediation sites from access by animals and humans is needed. Also, monitoring and treatment of the shoot may be necessary when concentrations are high. The contaminant distribution in the plant is determined by plant, soil, and contaminant properties.
Several investigators have used modeling and experiments with plants and soil to study organic contaminant fate and transport in plants (Briggs et al., 1982; McFarlane et al., 1990; Trapp et al., 1990; Wang and Jones, 1994; Paterson et al., 1994; Chang and Corapcioglu, 1998; Burken and Schnoor, 1998). In modeling efforts, the plant has been divided into one or three compartments to predict plant contamination (Behrendt and Bruggemann, 1993; Trapp and Matthies, 1995; Matthies and Behrendt, 1995). However, the one-compartment plant model may be too simple and the three-compartment model may be too sophisticated to evaluate contaminant concentration in the plant at phytoremediation sites. Furthermore, soil may not be treated as a single box because soil and environmental properties, and contaminant concentrations, vary with depth. In the present study, we developed a two-compartment plant model that can be applied to contaminant fate in phytoremediation. It also reflects the role of microorganisms in phytoremediation and plant contamination. This model was developed to be applied to field conditions and considered various factors that may be encountered in actual field conditions. Successful estimation of contaminant fate and transport in plants at a remediation site would assist in decisions relating to field maintenance and plant disposal and could help to reduce the maintenance and monitoring costs. The model results could be used as indicators of degree of soil contamination (Erdman and Christenson, 2000) and in risk assessment studies as well.
|
MODEL DEVELOPMENT
|
|---|
The model divides the plant into aboveground shoot and root compartments (Fig. 1a)
. The contaminant mass flow from the soil into the root varies with the soil depth since the contaminant concentration and soil physical conditions are different. To link the plant compartment model to contaminant fate and transport in the soil, the root compartment was discretized (Fig. 1b). Because the most important mechanism of contaminant transport into the plant is through the water phase, any contaminant transport into the plant through the gas phase is neglected.
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 1. Schematic of the plant contamination model. (a) Components of the two-compartment plant model. (b) Linkage of the two-compartment plant model to the soil model.
|
|
Contaminant Mass Balance in Rhizosphere Soil
The water-phase mass balance equation in the soil that includes contaminant advection and dispersion, sorption onto soil and roots, and biodegradation by microorganisms in the soil can be written as (Sung et al., unpublished data, 2000):
 | [1] |
where 
rhw is the water content (cm3 cm-3), qw is water flux (cm3 cm-2 h-1), DHw is the hydrodynamic dispersion coefficient (cm2 h-1), Crhw is the mass concentration of contaminant dissolved in pore water (g cm-3), as is the first-order kinetic rate coefficient (h-1), 
b is bulk density (g cm-3), k1 is the distribution coefficient (g cm-3), Crhs is the sorbed concentration (g g-1), km is the nongrowth substrate utilization rate (g g-1 h-1), Crb is the microorganism concentration in soil (g cm-3), Crhp is the primary substrate concentration in the soil (g cm-3), Krhw is the half saturation constant for contaminant (g cm-3), Krhp is the half-saturation constant for primary substrate based on soil water phase (g cm-3), and Ki is the inhibition coefficient (cm3 g-1).
Contaminant Mass Balance in the Root Compartment
The temporal mass balance equation for contaminant in the root can be expressed as:
 | [2] |
where Vr,i is the root volume of soil section i (cm3), Cr,i is the contaminant concentration in the root (g cm-3), Smrsp,i is the rate of absorption from root zone water phase into the root, where it accumulates (g h-1), Smrup,i is the rate of contaminant uptake into the plant's transpiration stream (g h-1), Srbp,i is the rate of contaminant transport in the transpiration from the lower root sections (g h-1), Srup,i is the rate of contaminant transport toward the upper root sections (g h-1), which becomes the rate of contaminant transport to shoot compartment, Srh, at the root and shoot boundary, and Srd,i is the rate of contaminant degradation in the root compartment (g h-1). The term Smrsp,i represents the contaminant that has crossed the root cell plasma membranes and remains inside the cell of the root. This fraction is therefore "accumulated" by the root and does not get transported further. The term Smrup,i represents the contaminant that has crossed the root cell plasma membranes and then becomes transported radially to reach the xylem. From the xylem, it is carried by mass flow with the transpiration stream toward the leaves. This fraction does not remain in the root. Neither of the above fractions involves active transport (Briggs et al., 1982; Clarkson, 1974; Salt et al., 1998). The term Smrsp,i can be defined by the kinetic equation:
 | [3] |
where Krw is the root concentration factor, Crhw,i is the water-phase contaminant concentration in the soil in the root zone (g cm-3), and Knrw is the mass-transfer rate coefficient normalized by volumetric root content (h-1). The volume-based root concentration factor, Krw, can be described as a function of octanol–water partition coefficient (Kow) to estimate the maximum contaminant distribution between the root and soil water under equilibrium conditions (Briggs et al., 1982; Trapp, 1995):
 | [4] |
This relationship, originally developed for the accumulation by roots of intact barley (Hordeum vulgare L.), was obtained empirically for a series of O-methylcarbamoyloximes and substituted phenylureas of different lipophilicity (Briggs et al., 1982). The rate of accumulation by the root compartment increases as root volume and concentration gradient increase. The volume of the root section can be calculated as:
 | [5] |
where 
i is the volume of roots per unit volume of soil in section i (cm3 cm-3) and VT,i is the total soil volume of section i (cm3).
Although water and contaminant may eventually reach the conducting tissue (xylem) for transport by mass flow to the shoot, they travel across the root by different mechanisms. If the contaminant travels more slowly to the xylem than does water, the concentration of contaminant in the xylem sap will be less than it was in the external solution (soil water). This difference in rates of transport to the xylem is embodied in the transpiration stream concentration factor, Tscf, which is the ratio between the concentration of contaminant in the transpiration stream in the xylem and its concentration in the external solution. For organic molecules, the Tscf depends upon the specific chemical properties (Briggs et al., 1982) and can be written as:
 | [6] |
The contaminant transport into the transpiration stream at ith root section, Smrup,i, can be defined as:
 | [7] |
where Uw,i is the volumetric root water uptake rate in section i (cm3 cm-3 h-1). Water flow through the soil–plant–atmosphere system can be described with total water head differences and total resistance in the system based on an electrical analog model (Chang and Corapcioglu, 1997; Marion and Tracy, 1988; Molz, 1981). If we assume that all roots on the plant have the same water uptake ability irrespective of age, and that root water uptake rate is mainly affected by soil water content, a stress index due to water deficiency (Feddes et al., 1978) can be applied as:
 | [8] |
where Umax is the maximum water uptake rate (cm3 cm-3 h-1), Uw is water uptake rate, and 
() is water stress index. Water uptake rate under various water contents may be expressed as the product of rooting density and root water uptake rate depending on soil water content:
 | [9] |
where qmax is a maximum water uptake rate per unit length of root (cm3 cm-1 h-1), Ld,i is rooting density in section i, defined as the length of roots per unit volume of soil (cm cm-3). The maximum water uptake rate can be estimated using:
 | [10] |
where Ev is the maximum transpiration rate from a unit soil surface area (cm3 cm-2 h-1).
Root growth and distribution are closely related to environmental conditions such as soil temperature, nutrients, availability of water, aeration, and soil strength (Russell, 1977). Rooting density declines exponentially with depth of soil z, as measured from the ground surface, at least for some species and soil conditions (Gerwitz and Page, 1974). Root distribution in terms of rooting density, as modified by heat stress, can be written as:
 | [11] |
where LT is total root length (cm), A is the cross-sectional area normal to the z axis (cm2), f is constant over soil depth at a given time period for plant growth (cm-1), zm is the rooting depth at time t (cm), and Ssfr is an overall heat stress index varying from 0 to 1. A stress index of 1 indicates that there is no stress while a stress index of 0 means that the roots are severely affected. This index accounts for any depression of root growth above a threshold temperature.
The contaminant transport from the lower root section to the nth root section through xylem flow, Srbp, can be defined as:
 | [12] |
The contaminant transport to the lower root section from the nth root section, Srup, is the sum of the contaminant transport from the lower root section and from the nth root section through xylem flow, and is defined as:
 | [13] |
The term Srd can be defined by introducing the first-order degradation rate constant, Krc (h-1) (Trapp and Matthies, 1995):
 | [14] |
Substitution of Eq. [3], [7], [12], [13], and [14] into Eq. [2] yields:
 | [15] |
The total contaminant concentration in the root compartment can be estimated by averaging the contaminant concentration over the entire root:
 | [16] |
Contaminant Mass Balance in the Shoot Compartment
The shoot refers to the aboveground plant parts. Contaminants can be transported to the shoot compartment from the root through xylem flow. The contaminants can be metabolized in the root and in the shoot so that they are transformed or degraded. Contaminants might be transported back to the root from the shoot in the phloem, but that is not considered here since the mass flow in the phloem is insignificant compared with xylem flow (Trapp, 1995). The loss of contaminant from the shoot as the gas phase was not incorporated into this model because nonvolatile organic contaminants were used here. In a lumped parameter model, the mass balance equation for the rate of contaminant accumulation in the shoot can be expressed as:
 | [17] |
where Mh is the dry mass of shoot (g), Ch is the contaminant concentration of the shoot compartment (g g-1), Srh is the rate of contaminant transport to the shoot compartment by xylem flow (g h-1), and Shd is the rate of contaminant degradation in the shoot compartment (g h-1). The contaminant transport from the root compartment to the shoot compartment, Srh, can be defined as:
 | [18] |
The contaminant degradation term, Shd, includes degradation and transformation to other compounds. It can be defined by introducing a first-order degradation rate constant, Khc (h-1) (Trapp and Matthies, 1995):
 | [19] |
Substitution of Eq. [18] and [19] into Eq. [17] yields:
 | [20] |
The mass of the shoot compartment can be estimated using a factor, ßh, which is the ratio (dry mass) between the shoot and the total plant (Feddes et al., 1978):
 | [21] |
where Mr is the root dry mass (g). Rengasamy and Reid (1993) presented the relation between root dry matter and rooting density as an empirical equation:
 | [22] |
where LTd is rooting density defined as the ratio of total root length to total soil volume (cm cm-3). The values for a and b are constants dependent on the type of plant species. The ratio of shoot mass to total plant mass is highly variable even in the same species, depending on temperature, nutrient supply, and moisture (Davidson, 1969a,b). If a constant value of ßh is assumed to apply throughout the life of a plant, the mass of the shoot compartment can be defined using Eq. [21] and [22]:
 | [23] |
Microbial Growth in the Rhizosphere
The rhizosphere is the zone of interaction between plant roots and soil microorganisms (Lynch, 1990; Walton et al., 1994). Plants release organic substrates such as exudates, mucilage, and dead cells to stimulate microbial growth (Anderson et al., 1993). Since the rhizosphere is a region of dynamic interaction between roots and soil microorganisms, microbial growth in the rhizosphere is stimulated by the continual input of readily assimilable organic substrates from roots. If the supply of a primary substrate is considered to be limiting, microbial metabolism and subsequent microbial growth are assumed to follow the Monod equation, where the rate of microbial growth with bacterial decay is described by (Newman and Watson, 1977):
 | [24] |
where Cm is microbial concentration in the soil (g cm-3), Crhp is the primary substrate concentration in the soil (g cm-3), Krhp is the half-saturation constant for primary substrate based on the soil water phase (g cm-3), Kd is the first-order endogeneous decay coefficient that includes cell maintenance and death (h-1), and µm (T) is the apparent microbial growth rate (h-1). When the soil has a primary substrate source from roots, the soil itself, and an additional substrate source, the mass balance equation for primary substrates in the vegetated soil can be expressed as:
 | [25] |
where r is the radial distance from the root axis (cm), Drm is the effective diffusion coefficient (cm2 h-1), YP is the yield coefficient of microorganisms for utilizing primary substrate (mass of microorganism per unit mass of consumed substrate) (g g-1), I is the additional substrate conversion rate (h-1), Ci is the concentration of additional substrate for available microorganisms (g cm-3), and Cme is the microbial concentration at steady state (g cm-3).
Case Study
The proposed plant contamination model was applied to a vegetated soil to simulate the fate and transport of contaminant in a plant. Numerical experiments investigated the model behavior using parameters obtained from experiments under field conditions with real irrigation, weather, and plant data. Physicochemical parameters such as bulk density, saturated hydraulic conductivity, organic matter content, and soil texture of the Weswood soil were measured and results are given in Table 1. Soil texture is classified as a silt loam with organic matter content of 0.8%.
The simulation was conducted for 40 d. We assumed that plants started to grow after the contaminant was mixed with soil. Johnsongrass was selected as the plant to evaluate the model behavior as well as practical effectiveness of phytoremediation. Johnsongrass is a perennial summer-growing grass that spreads by seeds and rhizomes and has an extensive root system. It was originally introduced to the USA as a pasture grass (Hampton and Cameron, 1980). To get the plant data, Johnsongrass roots were separated from the soils, and total root length, rooting density, and root volume were analyzed using root analysis software, WINRHIZO, developed by Regent Instruments (Regent Instruments, 1998).
Two contaminants, TNT and chrysene, were selected on the basis of their contrasting aqueous-phase solubility. Their physicochemical properties are listed in Table 2. For Rcf and Tscf values of TNT, Burken and Schnoor's calculated values (3.50 and 0.65) are used (Burken and Schnoor, 1998; Thompson et al., 1998). The initial total concentration was 2.46 mg kg-1 soil, which was the estimated maximum concentration for chrysene to prevent precipitation of chrysene under assumed experimental conditions. We assumed that once a contaminant was transported into the plant, it did not leach back into the soil since the mass transport from root to soil is by diffusion only and it is insignificant compared with mass transport from soil to root for a nonvolatile contaminant (Paterson et al., 1994). Dry biomass of root and shoot compartments for Johnsongrass was measured using collected root and shoot samples. The estimated value of ßh using experimental data was 0.574 (Corapcioglu et al., 1999). The benchmark parameters used in numerical experiments and initial conditions are listed in Table 3. We assumed that qmax is constant for the purpose of studying the sensitivity of the model.
|
RESULTS AND DISCUSSION
|
|---|
Effect of Chemical Properties
Figure 2
illustrates simulation results, showing the effect of contrasting contaminant properties on plant contamination. TNT has a higher solubility and a lower octanol–water partition coefficient than chrysene. The same initial concentrations were used for simulations for both contaminants, though the mass distributions for each phase were different due to different chemical properties. The effects of the contaminant solubility and hydrophobicity on plant contamination were evident. Figure 2a shows that chrysene would be expected to have a higher root concentration than TNT, reflecting the greater lipophilic properties of chrysene. This greater concentration in the root for chrysene was predicted even though TNT concentration in the water phase was higher than chrysene. Because of the low concentration in the water phase, chrysene reached its maximum value at around 20 d, and the concentration then started to decrease due to degradation in the plant under the conditions of this particular simulation. However, TNT showed a different pattern. The concentration of TNT in the water phase decreased in the initial period due to degradation by microorganisms in the soil. The root concentration was therefore high in the initial period and then no more contaminant moved into the root. Subsequently, the concentration decreased because of degradation in the plant.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2. Temporal variation of (a) contaminat concentration in the root compartment, (b) contaminant concentration in shoot compartment, and (c) total mass of root and shoot.
|
|
The shoot concentration showed a different pattern than the root concentration (Fig. 2b). The TNT concentration in the shoot was higher than for chrysene, reflecting the relationship between octanol–water partition coefficient and transpiration stream concentration factor. TSCF has high values when the log Kow value is between 1 and 3, and the log Kow of TNT falls into this range (Briggs et al., 1982). Furthermore, because TNT concentration in the water phase is higher than that of chrysene, more contaminant potentially can move with the transpiration stream. The initial TNT concentration in the water phase was high and decreased rapidly while the shoot concentration decreased due to dilution and degradation in the shoot compartment. The shoot concentration of chrysene increased first and then kept a relatively constant value due to the same degree of uptake and metabolization since the initial chrysene concentration in the water phase was low and did not vary during the simulation. Contaminant concentration in the water phase and the octanol–water partition coefficient are the main parameters that determine plant contamination, though these two parameters are not independent. If the chemical has a high octanol–water partition coefficient, the contaminant concentration in the water phase becomes low. The log Kow also affects the root concentration factor, Krw, which indicates the degree of contaminant partition between root and pore water.
Effect of Sorption
Figure 3
shows the relation between sorption kinetics and plant contamination. TNT was used to investigate sorption effects and other model behavior for plant contamination. As the first-order sorption kinetic rate s increased, the root concentration was high due to high contaminant concentration in the water phase (Fig. 3a). The concentration in the shoot shows the same tendency (Fig. 3b). Higher contaminant concentrations in the water phase allow more contaminant to move with the transpiration stream to the shoot. However, because water-phase concentration of TNT decreases from the initial values, the shoot concentration also decreases due to dilution mainly caused by slow uptake of contaminant mass relative to mass flux of water, and by degradation in the plant. Both root concentration and shoot concentration show that they start decreasing and approach steady state concentration values regardless of s. That is because the contaminant in the water phase does not change much except during the initial period and contaminant degrades mostly in root and shoot compartments.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 3. The effect of the first-order sorption kinetic rate coefficient s for TNT concentration in (a) root compartment and (b) shoot compartment.
|
|
Figures 4a and b
represent the effect of soil properties on TNT transport into the plant. Soil organic carbon content is the most important soil parameter that determines the hydrophobic contaminant distribution between solid surfaces and the water phase. In soil that is low in organic carbon, this relates to high water-phase concentrations, especially during the initial period, with more contaminant being transported into the root as well as to the shoot. For the first few days, different degrees of contaminant sorption by the soil organic carbon content affected the predicted concentration of contaminant in the root and shoot (Fig. 4) and the effect of soil organic carbon content was large relative to that of sorption rate coefficient (Fig. 3). After contaminant concentration in the water phase decreases, the concentration in the plant also decreases and approaches steady state concentration values under assumed experimental conditions.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 4. The effect of the soil organic carbon content foc on TNT concentration in (a) root compartment and (b) shoot compartment, and (c) the effect of the mass transfer rate coefficient between soil and root Knrw on TNT concentration in the root compartment.
|
|
The mass transfer rate coefficient between soil and root, Knrw, is the parameter that also determines the contaminant partition into the root compartment. Root and soil properties should be the main factors determining the value of Knrw. Figure 4c shows the effect of the root and soil mass transfer rate coefficient on root concentrations. Although root concentrations are high when Knrw is large, root concentration starts to decrease for all cases with time as the water-phase concentration decreases and approaches steady state concentration values.
Degradation in the Plant
Figure 5a
illustrates the effects of the first-order degradation rate in the root compartment, Krc, on TNT concentrations. As Krc increases, the root concentration decreases rapidly. Figure 6a also shows that although Krc is zero and there is no contaminant degradation in the root compartment, the root concentration keeps decreasing because of the dilution effect of root growth.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 5. TNT degradation in the plant for various values of (a) the first-order degradation rate coefficient in the root compartment, Krc, and (b) the first-order degradation rate coefficient in shoot compartment Khc.
|
|
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 6. The effect of the microbial biomass concentration on TNT concentration in (a) root compartment and (b) shoot compartment, and (c) the temporal variation of microbial concentration.
|
|
Figure 5b illustrates the effects of the first-order degradation rate in shoot compartment, Khc, on TNT concentrations. As Khc increases, the shoot concentration decreases quite quickly. It also shows that because of the plant growth with time, the shoot concentration decreases even if Khc is zero.
Responsibility for Contaminant Degradation by Both Plants and Microorganisms
Plants and microorganisms will take up a contaminant from the soil water phase, so that the water-phase concentration changes in the soil water due to microbial activities may affect plant contamination, and vice versa. Figure 6 shows the temporal variation of TNT concentrations in the root and shoot with changes in the microbial concentration. A constant microbial biomass concentration with time was assumed (Fig. 6c). As biomass concentration increased in soil, the TNT concentration decreased in the root compartment as well as in the shoot compartment. Bioavailability in the soil decreased as contaminant was degraded due to microbial biodegradation. The results indicate that the contaminant transport into the plant mainly depends on the soil water–phase concentration as well as microbial concentrations that can determine bioavailability in soil. We assumed a constant biomass in the soil for simulations given in Fig. 6, but Fig. 7
showed the effect of temporal variation of microbial concentration on the plant contamination (i.e., contaminant degradation by plants) due to changing of the additional substrate concentration. Additional substrate caused increases in the microbial biomass concentration (Fig. 7c). Figure 7 also illustrates the resultant root and shoot concentration due to temporal biomass changes. The root concentration and the shoot concentration in the early period were almost the same regardless of additional substrate concentration due to the fact that microbial biomass concentrations and TNT mass in the water phase were almost the same for each treatment early on. However, as the biomass concentration started changing with time, the water-phase concentration also changed. High microbial biomass resulted in not only a low concentration in the water phase but also low concentrations in root and shoot (Fig. 7).
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 7. The effect of the additional substrate concentration, CI0, on TNT concentration in (a) root compartment and (b) shoot compartment, and (c) the temporal variation of microbial concentration.
|
|
The maximum specific growth rate, µmax, is another factor affecting the degree of plant contamination. Because µmax affects the contaminant degradation rate as well as the microbial biomass concentration in cometabolic biodegradation, the resultant plant concentrations are influenced by µmax. Figure 8
illustrates the change of TNT concentration in plants as the µmax varies. High µmax caused fast biodegradation of contaminant in the water phase and fast growth of microbial biomass concentration. Therefore, the small maximum specific growth rate value would be expected to yield high root and shoot concentrations as well as high bioavailability. It indicates that the plant role in soil remediation becomes more important when the microbial activity is low, as expected.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 8. The effect of the maximum specific growth rate, µmax, on TNT concentration in (a) root compartment and (b) shoot compartment.
|
|
|
CONCLUSIONS
|
|---|
The two-compartment plant model system appears to be a reasonable approach to simulate plant contamination and soil remediation by plants. The results quantitatively indicate that plant contamination depends on soil properties such as soil organic carbon content, the physicochemical properties of the contaminants such as octanol–water partition coefficient, and various plant properties. Although TNT may not be mobile in plant tissue, as other studies mentioned (Thompson et al., 1998), this study shows the possibility and pattern of the plant contamination quantitatively under the field conditions.
The major conclusion from the numerical simulation is that the most important factor affecting plant contamination is contaminant mass dissolved in the water phase, defined as bioavailability (contaminant mass in the water phase/total contaminant mass in the soil) in this study. Physicochemical properties of chemicals and soil are the main properties that determine bioavailability in the soil. As bioavailability increased because of less contaminant being partitioned to the solid phase, the concentrations in root and shoot compartments increased as well. The octanol–water partition coefficient is an important parameter that determines not only contaminant bioavailability but also the tendency of the contaminant to partition into the root and shoot compartments. If the contaminant has a high octanol–water partition coefficient, the chemical is likely to be adsorbed to solid surfaces, causing a low contaminant concentration in the water phase and bioavailability. However, the ability of the contaminant to cross the lipid bilayer of the plasma membrane of the root cells also increases as the octanol–water partition coefficient increases. This explains how, even when the contaminant concentration in the water phase is low, the root can absorb appreciable amounts of contaminant mass. Contaminant transport into the shoot is more of a function of solubility in the water phase in the transpiration stream as well as bioavailability.
Soil properties also are important variables that affect plant contamination. If the soil has a high organic carbon content, more contaminant mass partitions into the solid phase, making it less available in pore water for absorption by plant roots. Therefore, concentrations in root and shoot compartments are predicted to be low when the soil has a high sorption capacity. The properties that determine the rate of contaminant desorption from the solid phase also are important factors that determine bioavailability in unsaturated soils.
One consideration in the use of plants in remediation is the complementary roles of plants and microorganisms for contaminant degradation. The results show that the contaminant concentration in plants decreases as parameters determining microbial activity, such as microbial concentration, maximum specific growth rate, and additional substrates, increase. If the microbial activity increases, then microorganisms degrade more contaminants and bioavailability in the soil is reduced. Plants can be contaminated in phytoremediaiton. However, this result suggests that the role of contaminant uptake by plants can be negligible when the soil already has a high microbial activity and consequently the soil water–phase concentration is low. It also suggests that plants and microorganisms act in a complementary manner in soil remediation. We assumed that qmax is constant for the purpose of studying the sensitivity of the model and we recognize that under natural conditions, qmax would vary with transpiration rate. We also assumed that an active population of microorganisms is present that can immediately metabolize or cometabolize the pollutants under study. However, it may be that a period of selection or adaptation is required before the parameters given in Table 3 are applicable.
The stress factor for root water uptake is:
where hw is the pressure head [L], the subscripts max and min denote the limiting maximum and minimum head (measure of water tension) at which the root can take up water, and the subscripts lt and ut denote the lower and upper threshold limit for optimum conditions, respectively.
The heat stress factor for root distribution is:
where SfT is the stress index for temperature. The subscript alt is the point at which slope changes, the subscripts max and min denote the limiting maximum and minimum temperature, and the subscripts lt and ut denote the lower and upper threshold limit for optimum conditions. A1 is the slope for the first-phase response.
The estimation of overall heat stress for the lifetime of the plant is:
|
ACKNOWLEDGMENTS
|
|---|
This research was funded through grants from the USEPA (R8254-01-0) and Gulf Coast Hazardous Substance Research Center (107TAM0696).
|
REFERENCES
|
|---|
- Anderson, T.A., E.A. Guthrie, and B.T. Walton. 1993. Bioremediation in the rhizosphere. Environ. Sci. Technol. 27:2630–2636.
- Behrendt, H., and R. Bruggemann. 1993. Modeling the fate of organic chemicals in the soil plant environment: Model study of root uptake of pesticides. Chemosphere 27:2325–2332.
- Briggs, G.G., R.H. Bromilow, and A.A. Evans. 1982. Relationships between lipophilicity and root uptake and translocation of non-ionised chemicals by barley. Pestic. Sci. 13:495–504.
- Burken, G.J., and J.L. Schnoor. 1998. Predictive relationships for uptake of organic contaminants by hybrid poplar trees. Environ. Sci. Technol. 32:3379–3385.
- Chang, Y.Y., and M.Y. Corapcioglu. 1997. Effect of roots on water flow in unsaturated soils. J. Irrig. Drain. Eng. ASCE 123:202–209.
- Chang, Y.Y., and M.Y. Corapcioglu. 1998. Plant-enhanced subsurface bioremediation of nonvolatile hydrocarbons. J. Environ. Eng. ASCE 124:162–169.
- Clarkson, D. 1974. Ion transport and cell structure in plants. McGraw Hill, London.
- Corapcioglu, M.Y., R.L. Rhykerd, C.L. Munster, M.C. Drew, K. Sung, and Y.Y. Chang. 1999. Phytoremediation and modeling of land contaminated by hydrocarbons. p. 9–14. In A. Leeson and B.C. Alleman (ed.) Phytoremediation and innovative strategies for specialized remedial applications. Battelle Press, Columbus, OH.
- Davidson, R.L. 1969a. Effects of root/leaf temperature differentials on root on some pasture grasses and clover. Ann. Bot. 33:551–559.
- Davidson, R.L. 1969b. Effects of soil nutrients and moisture on root/shoot ratios in Lolium perenne L. and Trifolium repens L. Ann. Bot. 33:571–577.[Abstract/Free Full Text]
- Erdman, J.A., and S. Christenson. 2000. Elements in cottonwood trees as an indicator of ground water contaminated by landfill leachate. Ground Water Monit. Remediat. 20:120–126.
- Feddes, R.A., P.J. Kowalik, and H. Zaradny. 1978. Simulation of field water use and crop yield. John Wiley & Sons, New York.
- Gerwitz, A., and E.R. Page. 1974. An empirical mathematical model to describe plant root systems. J. Appl. Ecol. 11:773–781.
- Hampton, M.D., and C.C. Cameron. 1980. Johnsongrasss sorghum halepense—Biology significance and control. Farm Prod. Practice 345:2–3.
- Lynch, J.M. 1990. The rhizosphere. John Wiley & Sons, New York.
- Marion, M.A., and C.C. Tracy. 1988. Flow of water through root-soil environment. J. Irrig. Drain. Eng. ASCE 114:588–604.
- Masters, G.M. 1974. Introduction to environmental science and technology. John Wiley & Sons, New York.
- Matthies, M., and H. Behrendt. 1995. Dynamics of leaching, uptake, and translocation: The simulation model network atmosphere–plant–soil. p. 215–243. In S. Trapp and J.C. McFarlane (ed.) Plant contamination: Modeling and simulation of organic chemical process. Lewis Publ., Boca Raton, FL.
- McFarlane, C., T. Pfleeger, and J. Fletcher. 1990. Uptake and disposition of nitrobenzene in several terrestrial plants. Environ. Toxicol. Chem. 9:513–520.
- Molz, F.J. 1981. Models of water transport in the soil–plant system: A review. Water Resour. Res. 17:1245–1260.
- Newman, E., and A. Watson. 1977. Microbial abundance in the rhizosphere: A computer model. Plant Soil 48:17–56.
- Paterson, S., D. Mackay, and C. McFarlane. 1994. A model of organic chemical uptake by plants from soil and the atmosphere. Environ. Sci. Technol. 28:2259–2266.
- Paterson, S., D. Mackay, D. Tam, and W.Y. Shiu. 1990. Uptake of organic chemicals by plants: A review of processes, correlations and models. Chemosphere 21:297–331.
- Regent Instruments. 1998. Win/Mac Rhizo manual. Regent Instruments, Quebec, Canada.
- Rengasamy, J., and J.B. Reid. 1993. Root system modification of faba beans (Vicia faba L.) and its effects on crop performance 1. Responses of root and shoot growth to subsoiling, irrigation and sowing date. Field Crops Res. 33:175–196.
- Russell, R.S. 1977. Plant and root systems: Their function and interaction with the soil. McGraw–Hill Book Co., New York.
- Salt, D.E., R.D. Smith, and I. Raskin. 1998. Phytoremediation. p. 643–668. In L. Russel et al. (ed.) Annual review of plant physiology and plant molecular biology. Annual Reviews, Palo Alto, CA.
- Thompson, L.P., L.A. Ramer, and J.L. Schnoor. 1998. Uptake and transformation of TNT by hybrid poplar trees. Environ. Sci. Technol. 32:975–980.
- Trapp, S. 1995. Model for uptake of xenobiotics into plants. p. 107–151. In S. Trapp and J.C. McFarlane (ed.) Plant contamination: Modeling and simulation of organic chemical process. Lewis Publ., Boca Raton, FL.
- Trapp, S., and M. Matthies. 1995. Generic one-compartment model for uptake of organic chemicals by foliar vegetation. Environ. Sci. Technol. 29:2333–2338.
- Trapp, S., M. Matthies, I. Scheunert, and E.M. Topp. 1990. Modeling the bioconcentration of organic chemicals in plants. Environ. Sci. Technol. 24:1246–1252.
- Van Genuchten, M.Th. 1980. A closed form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44:892–896.[Abstract/Free Full Text]
- Walton, B.T., E.A. Guthrie, and A.M. Holylman. 1994. Toxicant degradation in the rhizosphere. p. 10–26. In T.A. Anderson and J.R. Coats (ed.) Bioremediation through rhizosphere technology. Am. Chem. Soc., Washington, DC.
- Wang, M., and K.C. Jones. 1994. Uptake of chlorobenzenes by carrots from spiked and sewage sludge-amended soil. Environ. Sci. Technol. 28:1260–1267.
TOP
ABSTRACT
INTRODUCTION
