Measurements of pH and Redox Potential Distributions in TNT-Contaminated Plant-soil Systems Using Microelectrode Techniques
H. Pang and T.C. Zhang
125B Engineering Building, Dept. of Civil Engineering, University of Nebraska-Lincoln at Omaha Campus, Omaha, NE68182-0178., Phone: (402)554-3784, Fax: (402)554-3288
Abstract
The pH and redox potential profiles in TNT-contaminated soils with and without plants were investigated using microelectrode techniques. The new pH cocktail and double-barreled structure greatly improved the performance of the pH microelectrode. For soil without plants, there is almost no pH difference at different locations with different heights; while for the TNT-contaminated soils with plants there exist pH profiles. The soil immediately near the root of the plant has the lowest pH value. The pH value increases as the distance between the measuring point and the plant roots increases. The pH gradient (the increased pH value over the unit distance) decreases with an increase of the distance between the measuring point and the plant roots. These results show that the plant presence can greatly affect the pH distribution. In vegetated soil, the redox potentials in the layer nearest the plant roots are higher than those in the bulk soil without plants. The redox potentials in the central part of the plant are lower than those in the soil around the plant and soil without the plant. The redox potentials in the soil without plants decrease with an increase of depth.
Keywords: microelectrode, pH, redox potential, soil, TNT
Introduction
A serious environmental problem facing the EPA is the presence of contaminated
soil at sites where munitions were formerly manufactured, stored, or demilitarized (Jenkins
and Walsh, 1992). In the U.S., there are at least 50 former military installations that have
3,000 
or more of contaminated soil. TNT is the primary contaminant found at these
sites, and is known to be mutagenic and carcinogenic to human health (McCormick, et al.,
1981; Kaplan and Kaplan, 1982). Thus, soil contaminated with TNT needs to be
remediated. Incineration is the only available and proven technology successful in
remediating highly toxic soils contaminated with TNT; however, it is expensive with a cost
approaching US $800/
kg of contaminated soil (Funk, et al., 1993).
Bioremediation is gaining more and more interest due to its characteristics of in situ treatment and low cost (Preuss, et al., 1993). The success of a bioremediation strategy is determined by the interrelationships among organisms (including both microbes and plants), substrate (contaminant), and the soil-water environment. Therefore, to advance bioremediation into a predictable and controllable technology, it is imperative to understanding the complex interplay of chemistry, physics, and microbial ecology and how the environment permits and ultimately limits remediation. Measurement of soil microenvironments, such as pH, redox potentials, microbial communities, and other parameters, may provide information to explain many of the complex reactions taking place in soil systems.
One powerful tool that can be used to study soil microenvironments is microelectrode techniques. Traditionally, microelectrode techniques were used in the field of neurobiology to measure the intracellular and extracellular conditions (Admstrong and Garcia-diaz; Thomas, 1978). In recent decades, microelectrode techniques have become some of the most powerful analytical techniques for characterizing chemical and metabolic activity gradients in biofilms (Zhang and Bishop, 1994; Revsbech, 1989). However, very few studies have reported on the reliability of microelectrodes for the measurements of pH, redox, or other parameters in soil-plant systems. Tiedje's group used O2 microelectrodes with a sensing tip size of 1 to 3 µm to measure oxygen profiles and denitrification rates directly in soil aggregates (Sexstone, et al., 1985). Flessa and Fischer (1992) used redox potential microelectrodes with a Pt wire (0.5 mm in diameter) to measure the redox in microsites of the rhizosphere of rice continuously flooded, and found redox potential markedly increased close to the root tips. Conkling and Blancear (1989) and Yang, et al. (1995 and 1994) used glass microelectrodes to measure the microscale pH spatial distribution in soil cores. They reported that soil heterogeneity and spatial dependence of pH in microscale environments exist and vary with sampling site and scale. These studies demonstrate that bulk soil parameters measured by conventional procedures may not reveal variability that could have important impacts on plant roots and microbial growth.
The objectives of this project are to study the microscale environments of TNT-contaminated soil and to provide the optimum remediation strategy for the TNT-contaminated sites. In this paper, we present how to improve the pH microelectrode for measuring soil samples, and the preliminary results of pH and redox potential distributions in TNT-contaminated soil with and without plants.
Materials and Methods
Soil samples
The TNT-contaminated soil samples both with and without plants were collected from the former Nebraska Ordnance Plant. In that Superfund Site, there is a narrow piece of soil lot on which no plants grow, probably due to the high concentration of TNT in the soil. Soil in this lot was collected as the nonvegegated soil samples. On the edge of this lot, plants grow luxuriantly, which shows the tolerant capacity of the plant for toxic contaminants. Plants together with the soil around them were collected as vegetated soil samples. After being brought back to the laboratory, the plants were taken care of for 10 days by watering and shining until they looked as fresh and active as when they were on site. The nonvegegated soil was kept in the same extro-environment as the soil with plants.
Experimental design
In order to investigate the distribution of pH and redox potentials in the soils near
the plant roots, a soil core of 
with a plant in the center of it was
collected and constructed (see Fig. 1). Along the vertical center line of each side, four
points were chosen for the measurements; that is, 0.5cm below the top (H1), 3cm below
the top (H2), 6cm below the top (H3), and 9cm below the top (H4), respectively. After
measuring the 16 points (4 points on each side) for pH and redox potential, the soil sample
was cut to 
, with no change in the core height. After measuring the
corresponding 16 points in this second core, the soil sample was cut again to
. pH and redox potential were then measured for the 16 points in this
new core. For one point, the redox potential was measured first, then measurement of pH
followed. The penetration depth of the pH or redox potential microelectrode was 3 to 4
mm at each point. The readings obtained from these three cores were used to analyze the
pH and redox potential distributions of the soil around the plant.
The central part of the vegetated soil was a column with a diameter of 4cm and a height of 4.5cm; in the center of this soil column was a plant called Lambs Quarters, a common weed. Three points in this soil column with the plant, that is, 0.5cm below the top, 2.5cm below the top, and 4cm below the top, respectively, were measured to represent the situation of the central part.
For the soil without plants, three columns with a diameter of 2cm and a height of 10cm were collected at three randomly chosen sites, one column from each site. For each column, pH and redox potential were measured at different heights, 0.5cm below the top (H1), 4cm below the top (H2), and 8cm below the top (H3), respectively.
Microelectrodes
Double-barreled pH ion-selective microelectrodes were used to measure the pH profiles. A double-barreled glass tubing with an I. D. of 0.8 mm and an O.D. of 1.5 mm was pulled, beveled, and salinized subsequently. One barrel, used as the working probe, then was backfilled with a hydrogen ion-selective cocktail. The composition of the pH cocktail was 2.85 wt% hydrogen ion ionophore (Fluka, No. 95292), 5 wt% polyvinyl chloride (molecular weight PVC, Fluka, No. 81392), 0.5 wt% potassium tetrakis(4-chlorophenyl)borate (Fluka, No60591), and 91.65 wt% bis(1-butylpentyl)adipate (BBPA) (Fluka, No. 02150). The mixture was diluted by 2 to 3 times its weight with tetrahydrofuran (Fluka,. 87369) to stabilize the PVC. The composition of this pH cocktail was developed in our laboratory specifically for soil samples. The new composition of the pH cocktail is quite different from that suggested by Fluka (1988), but this new cocktail is much more stable and has a slope above 50 mV per pH unit. The barrel filled with the pH cocktail was then backfilled with a buffer solution (pH = 7), consisting of 0.04M KH2PO4, 0.023M NaOH, and 0.015M NaCl (Ammann, et al., 1981). Another barrel, used as the reference probe, was filled with agar (0.5% to 1%)/3M KCl saturated with AgCl. A piece of Ag/AgCl wire was inserted into each barrel to transport the milivolt signal generated between the working electrode and reference electrode to a Chemical Microsensor (see below). In order to offset the effect of soil structure on milivolt readings, the pH microelectrode was calibrated using the soil standard series which had the same soil structure as the tested samples. Part of nonvegegated soil was taken out and put into three small containers (10 mL plastic beakers). Each small container then was soaked with a pH buffer solution (pH = 4, 7, or 10). The pH values of these small containers were measured using a macro-pH probe (Beckman Instrument, Fullerton, CA), specifically designed for soil application, and the pH values measured were used as the standard pH to make the calibration curve for our pH microelectrodes.
Redox potentials were measured using a commercial redox electrode with a tip O.D. of 800 µm (Microelectrodes Inc., Bedford, NH), coupled with a self-made reference electrode (Ag/AgCl) with a tip size of 20 µm. The reference electrode was made in the same way as mentioned above.
Measurement systems
The tray holding the soil sample was put on a vibration isolation table (Technical Manufacturing Corporation, Peabody, MA) inside a Faraday cage. The microelectrodes (pH or redox electrodes) were positioned manually using a 3-D micromanipulator (World Precision Instrument, Sarasota, FL). The signal generated from the microelectrodes was measured using a chemical microsensor (Diamond General, Ann Arbor, MI). The signal could then be recorded and analyzed. Figure 2 shows the experimental setup used in this study.
Results
Fabrication of microelectrodes suitable for soil samples
To apply the microelectrode technique to soil samples, we first had to find the suitable microelectrode, including a usable tip size and a good cocktail composition. Three tasks were conducted in this study for this purpose.
First, tests were conducted to find a suitable tip size for microelectrodes used in this study. The moisture content and the soil composition can affect the suitable tip size of the microelectrode. Therefore, for different soils or the same soils with different moisture, the tip size of the probes will be different. For the soil samples used in this study, the suitable tip sizes were evaluated using the pulled-glass probes to penetrate the soils. After several trials, the suitable tip size (the O.D. of the tip) for the soil samples of interest (moisture high content 10%) in this study was found to be 35 to 40 µm.
Second, we developed a new pH cocktail in this study. Figure 3 shows the
comparisons of the cocktail reported by Fluka (1988) and the new cocktail developed in
this research using a pH microelectrode with a tip size of 35 µm. The cocktail reported by
Fluka consisted of 1% hydrogen ion ionophore I, in 33% PVC, 65.5% BBPA, and 0.5%
potassium tetrakis. The slope produced by the pH microelectrode filled with this cocktail
was found to be around -30 to -40 mV/pH (-33.8mV/pH in Figure 3). According to the
theory of the ion-selective microelectrode (Admstrong and Garcia-diaz, 1980), the ideal
slope, even considering the effect of the interference ion, should be above -50 mV/pH
when the temperature is 20C. The lower slope represents poor selectivity. The tip size is
one of the factors affecting the performance of the microelectrode (Henriksen, et al.,
1990). In previous biofilm studies, the tip size was usually 1 to 5 µm (Zhang and Bishop,
1995), which is much smaller than the size used for soil samples. In addition, the Fluka's
cocktail is very sticky and is difficult to backfill into the probe. Therefore, we developed a
new cocktail for the soil application. By changing the relative concentration of the four
chemicals, eventually, the optimum composition of cocktail in our research was found.
The slope for all the pH microelectrodes WAS in the range of -52
2 mV/pH units
(Figure 3). The performance of the pH microelectrode filled with this new cocktail was
very stable and satisfactory.
Third, we evaluated the effect of the distance between two electrodes on signal readings and developed double-barreled microelectrodes. Initially, a separated macro reference electrode (Microelectrode Inc., Bedford, NH) was coupled to a working microelectrode. Table 1 lists the different readings for the same point measured when the position of the reference electrode changed. It is shown that for the soil samples, which usually have low conductivity, the distance between the working electrode and reference electrode can greatly affect the final reading. Therefore, the distance between these two electrodes must be fixed and, if possible, reduced. For this reason, the double-barreled glass tubing was used to make the pH microelectrode. For the Redox electrode, the reference electrode was bound to the working probe by tape.
pH distributions
The pH readings for the vegetated soil and nonvegetated soil are shown in Tables 2, 3, and 4. The average pH values in Table 2 and pH values in Table 3 are used to show the pH profiles in TNT-contaminated soil with a plant. As shown in Figure 4, pH readings decreased as the soil nears the plant. In core 1, the outer most layer, the pH readings range from 5.58 to 5.91, with an average of 5.75. In the third core, the one nearest to the plant, the pH readings range from 5.38 to 4.72, and the average value is only 5.03. The average pH value in core 2 is 5.46, which is 0.29 less than the one in the first core and 0.43 higher than the one in core 3. The pH difference between the core 3 and the central part is 0.95, a large difference. The gradient of pH decrease becomes higher as the soil approaches the center of the plant, which can been judged from the inducing slopes. From the pH readings of the third core and center part in Figure 4, we find that the two points about 3cm below the top have the lowest pH value, compared with the other points in the same layer (or part). It was found that these two points were in the part of soils which contain many tiny roots. The above results show that: (1) the soil immediately near the root of the plant has the lowest pH value; (2) pH value increases as the distance between the measuring point and the plant roots increases; and (3) the increased pH value over the unit distance decreases as the measuring point becomes further away from the plant roots, and eventually there is no pH difference existing among different points.
From Table 4, we can see that there is almost no pH difference at locations with different heights. The only difference between the vegetated soil and nonvegetated soil is the presence of plants. It can be concluded that plants are the reason for the existence of microsacle pH profiles in the vegetated soil. Therefore, plants can change the characteristics of the site.
Redox potential
Tables 5, 6, and 7 show the original redox potential measurments for soil both with
and without plants. Because the 
value of a given oxidation-reduction system is pH-dependent, in order to compare the results among soils with different pH, it is necessary to
convert the measured redox potential value to the same pH basis, generally, pH 7. When
the temperature is 25C, an increase in pH by one unit causes a decrease in 
by 60 mV
(Yu and Ji, 1993). Therefore, equivalent redox potential at pH 7(
) can be calculated
using the following equation:
where 
is the original 
and pH is the pH value of the measuring point.
Tables 8, 9, and 10 show the converted 
for soils both with and without plants.
From Table 5, making a comparison among the average redox potentials of each layer, we
can see that redox potential increases as the layer approaches the plant. The redox
potential in the central part, however, is much lower than those in the surrounding area,
even lower than those in soil without plants. Currently, we don't know the reasons; we
need more experiments to figure out this phenomenon. For the soil without plants, there
is almost no redox potential difference among different locations. It seems that the redox
potential decreases as the depth increases, probably due to the transportation of oxygen
from the air to soil.
Conclusion
The results of our study show that the microelectrode techniques can be used to measure the pH and redox potentials in soil samples successfully. The modified pH cocktail and double-barreled structure greatly improved the performance of the pH microelectrodes. The pH profiles of TNT-contaminated soil with a plant are totally different from the soil without plants. For the vegetated soil, the lowest pH value exists in the soil immediately near the plant roots; pH increases as the distance between the measuring point and the plant roots increases. The increased pH over a unit distance (the pH gradient) decreases as the measured point becomes further away from the plant roots, and eventually there exists no pH difference at locations with different heights, which is the pH distribution in soil without plants. It is the plant that changes the characteristics of the site. In vegetated soil, the redox potentials in the layer nearest the plant roots are higher than those in the bulk soil. The redox potentials in the central part of the plant are lower than those in the soil around the plant and the soil without plants. The redox potential in the soil without plants decreases with an increase of depth.
Acknowledgment
We gratefully acknowledge Professors P.J. Shea, G.L. Horst, and S.D. Comfort of the Agronomy Department at the University of Nebraska-Lincoln (UNL) for their help to get the plants and soils used in this study. This work is being funded partially by the NSF/EPSCoR Program; the College of Engineering and Technology, the Center for Infrastructure Research, and the Civil Engineering Department of the UNL are providing the matching funds.
Reference
Admstrong, W.M., and J.F. Garcia-diaz, 1980. Ion-selective microelectrode: theory and technique, Federation Proc., 39, pp. 2851-2859.
Ammann, D., F. Lanter, and R.F. Steiner, 1981. Neutral carrier-based hydrogen ion selective microelectrode for extra- and intracellular studies, Analytical Chemistry, 53, pp. 2267-2269.
Conkling, B.L., and R.W. Blanchar, 1989. Class microelectrode techniques for in situ pH measurement, Soil Sci. Soc. Am. J., 53, pp. 58-62.
Flessa, H., and W.R. Fischer, 1992. Plant-induced changes in the reodx potentials of rice rhizospheres, Plant and Soil, 143, pp. 58-62.
Fluka, 1988. Inonophores, pp. 15-16.
Funk, S.B., D.J. Roberts, D.L. Crawford, and R.L. Crawford, 1993. Initial-phase optimization for biorestoration of munitions compound-contaminated soils, Appl. Environ. Microbiol., 59, pp. 2171-2177.
Henriksen, G.H., A.J. Bloom, and R.M. Spanswick, 1990. Measurement of met fluxes of ammonium and nitrate at the surface of barley roots using ion-selective microelectrode, Plant Physiology, 93, pp. 271-280.
Jenkins, T.F., and M.E. Walsh, 1992. Development of field screen methods for TNT, 2,4-DNT, and RDX in soil, Talanta, 39, pp. 419-428.
Kaplan, D.L., and A.M. Kaplan, 1982. 2,4,6-trinitrotoluene-surfactant complexes: Decomposition, mutagenicity, and soil leaching studies, Environ. Sci. Technol., 16, pp. 566-571.
McMorimick, N.G., J.H. Cornell, and A.M. Kalpan, 1981. Bioremediation of bexahydro-1,3,5-trinitro-1,3,5-triazine, Appl. Environ. Microbiol, 42, pp. 817-823.
Preuss, A., J. Fimpel, and G. Diekert, 1993. Anerobic transformation of 2,4,6-trinitrotoluene (TNT), Arch. Microbiol., 159, pp. 345-353.
Revsbech, N.P., 1989. Microsensors: spatial gradients in biofilms, In Structure and Function of Biofilms, W.G. Characklis and P.A. Wilderer (eds.), John Wiley & Sons, Ltd. pp. 129-144.
Thomas, R. C.., 1978. Ion-selective intracellular microelectrodes: How to make and use them. Academic press, New York.
Sexstone, A.J., N.P. Revsbech, T.B. Parkin, and J. M. Tiedje, 1985. Direct measurement of oxygen profiles and denitrification rates in soil aggregates, Soil Sci. Soc. Am. J., 49, pp. 645-651.
Yang, J., R.W. Blanchar, and R.D. Hammer, 1994. Calibration of glass microelectrode for in situ pH measurements. Soil Sci. Soc. Am. J., 58, pp. 354-357.
Yang, J., R.D. Hammer, and R.W. Blanchar, 1995. Microelectrode pH spatial distribution in the Ap horizon of Mexico silt loam, Soil Science, 160, (5), pp. 371-375.
Yu, T.R. and G.L. Ji, 1993. Electrochemical methods in soil and water measurement, Pergamon press, pp. 297-313.
Zhang, T.C., Y.C. Fu, and P.L. Bishop, 1994. Structure, activity and composition of biofilms, Wat. Sci. Tech., 29, (7), pp. 335-344.
Zhang, T.C., and P.L. Bishop, 1995. Competition for substrate and space in biofilm.
Wat. Environ. Res., 67, pp. 992-1003.
Table 1. Effect of the distance between the working and reference microelectrode on
milivolt readings.
| Distance, cm | 0.5 | 1.5 | 2.5 | 1.5 |
| Voltage, mV | 88 | 93 | 102 | 75 |
Table 2. pH profiles of TNT-contaminated soil with a plant.
| Side 1 | Side 2 | Side 3 | Side 4 | Average | ||
| H1 | 5.87 | 6.12 | 6.08 | 5.54 | 5.90 | |
| I | H2 | 6.22 | 5.51 | 6.52 | 5.37 | 5.91 |
 |
H3 | 5.78 | 5.81 | 5.47 | 5.32 | 5.60 |
| H4 | 5.88 | 5.69 | 5.23 | 5.50 | 5.58 | |
| H1 | 5.33 | 5.61 | 5.97 | 5.80 | 5.68 | |
| II | H2 | 5.13 | 5.01 | 6.16 | 5.60 | 5.48 |
 |
H3 | 4.98 | 5.11 | 5.62 | 5.49 | 5.30 |
| H4 | 5.01 | 5.12 | 5.76 | 5.63 | 5.38 | |
| H1 | 5.55 | 5.19 | 5.24 | 5.52 | 5.38 | |
| III | H2 | 5.66 | 4.91 | 4.09 | 4.22 | 4.72 |
 |
H3 | 5.73 | 5.02 | 4.46 | 4.79 | 5.00 |
| H4 | 5.58 | 4.93 | 4.87 | 4.72 | 5.03 |
H1: 0.5cm below the top; H2: 3cm below the top; H3: 6cm below the top; H4: 9cm below the top. The
whole height of the sample is 11cm.
Table 3. pH readings in the central part of vegegated soil.
| H1 | H2 | H3 |
| 4.32 | 3.81 | 4.10 |
H1: 0.5cm below the top; H2: 2.5cm below the top; H3: 4cm below the top.
Table 4. pH readings in TNT-contaminated soil without plants.
| Column 1 | Column 2 | Column 3 | |
| H1 | 6.33 | 6.28 | 6.30 |
| H2 | 6.29 | 6.31 | 6.27 |
| H3 | 6.30 | 6.27 | 6.32 |
H1: 0.5cm below the top; H2: 4cm below the top; H3: 8cm below the top.
Table 5. Original redox potentials (mV) in TNT-contaminated soil with a plant.
| Side 1 | Side 2 | Side 3 | Side 4 | ||
| H4 | 336.3 | 283.9 | 280.3 | 331.5 | |
| I | H3 | 315.7 | 292.6 | 297.7 | 342.9 |
| H2 | 295.7 | 298.4 | 293.6 | 341.7 | |
| H1 | 280.7 | 289.2 | 295.7 | 352.0 | |
| H4 | 360.0 | 414.5 | 377.0 | 421.5 | |
| II | H3 | 378.8 | 250.1 | 428.6 | 424.7 |
| H2 | 293.1 | 359.5 | 398.3 | 402.0 | |
| H1 | 391.0 | 332.0 | 383.6 | 422.6 | |
| H4 | 397.5 | 432.2 | 395.5 | 421.2 | |
| III | H3 | 404.4 | 415.3 | 430.5 | 426.7 |
| H2 | 424.1 | 452.6 | 430.8 | 409.2 | |
| H1 | 375.8 | 399.3 | 406.6 | 396.9 |
H1: 0.5cm below the top; H2: 3cm below the top; H3: 6cm below the top; H4: 9cm below the top.
Table 6. Original redox potenials (mV) in the central part of the vegegated soil.
| H1 | H2 | H3 |
| 215.1 | 433.6 | 399.9 |
H1: 0.5cm below the top; H2: 2.5cm below the top; H3: 4cm below the top.
Table 7. Original redox potentials (mV) in TNT-contaminated soil without plants.
| 1 | 2 | 3 | |
| H1 | 310.1 | 307.4 | 327.6 |
| H2 | 288.7 | 273.2 | 298.8 |
| H3 | 274.5 | 250.1 | 263.9 |
H1: 0.5cm below the top; H2: 4cm below the top; H3: 8cm below the top.
Table 8. Redox potential profiles (mV) of TNT-contaminated soil with a plant. (based
on pH 7)
| Side 1 | Side 2 | Side 3 | Side 4 | Average | ||
| H4 | 269.1 | 205.3 | 174.1 | 241.5 | 222 .5 | |
| I | H3 | 242.5 | 221.2 | 205.9 | 242.1 | 227.9 |
| 232.7 | H2 | 248.9 | 209.0 | 264.8 | 243.9 | 241.7 |
| H1 | 212.9 | 236.4 | 240.5 | 264.4 | 238.6 | |
| H4 | 240.6 | 301.7 | 302.6 | 339.3 | 296.1 | |
| II | H3 | 257.6 | 136.7 | 345.8 | 334.1 | 268.6 |
| 284.8 | H2 | 180.9 | 240.1 | 347.9 | 318.0 | 271.7 |
| H1 | 290.8 | 248.6 | 321.8 | 350.6 | 302.9 | |
| H4 | 312.3 | 308.0 | 267.7 | 284.4 | 293.1 | |
| III | H3 | 328.2 | 296.5 | 278.1 | 294.1 | 299.2 |
| 295.5 | H2 | 343.7 | 327.2 | 256.2 | 242.4 | 292.4 |
| H1 | 288.8 | 290.7 | 301.0 | 308.1 | 297.2 |
H1: 0.5cm below the top; H2: 3cm below the top; H3: 6cm below the top; H4: 9cm below the top. The
whole height is 11cm.
Table 9. Redox potentials (mV) in the central part of vegegated soil.
(based on pH 7)
| H1 | H2 | H3 |
| 54.3 | 242.2 | 225.9 |
H1: 0.5cm below the top; H2: 2.5cm below the top; H3: 4cm below the top.
Table 10. Redox potential profiles of TNT-contaminated soil without plants.
(based on pH 7)
| 1 | 2 | 3 | |
| H1 | 269.9 | 264.2 | 285.6 |
| H2 | 246.1 | 231.8 | 255.0 |
| H3 | 232.5 | 206.3 | 223.1 |
H1: 0.5cm below the top; H2: 4cm below the top; H3: 8cm below the top.
Figure 1. Shape
of a soil core.
Figure 2. Experimental setup, the microelectrode system.
Figure 3. Comparison between the Fluka's pH cocktail and the cocktail developed in this study.
Figure 4. pH profiles as a function of verticle depth in TNT-contaminated soil with a plant.
Figure 5. pH profiles over the distance.