Published in J. Environ. Qual. 33:1562-1567 (2004).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA
SHORT COMMUNICATIONS
Automated and Continuous Redox Potential Measurements in Soil
Michel Vorenhouta,*,
Harm G. van der Geestc,
Daan van Marumb,
Kees Wattelb and
Herman J. P. Eijsackersa
a Institute of Ecological Science, Vrije Universiteit, De Boelelaan 1085, 1081 HV Amsterdam, the Netherlands
b Electronics Department, Faculty of Earth and Life Sciences, Vrije Universiteit, De Boelelaan 1085, 1081 HV Amsterdam, the Netherlands
c Aquatic Ecology and Ecotoxicology, Faculty of Science, University of Amsterdam, Kruislaan 320, 1098 SM, Amsterdam, the Netherlands
* Corresponding author (michel.vorenhout{at}ecology.falw.vu.nl).
Received for publication July 22, 2003.
 |
ABSTRACT
|
|---|
Redox potential (Eh) describes the electrical state of a matrix. In soils, Eh is an important parameter controlling the persistence of many organic and inorganic compounds. A popular, but also criticized, manual measuring method makes use of a small tip of Pt placed on a copper wire that is placed in the soil; a reference electrode is placed in the same soil at a fixed distance. Fluctuations in redox potential values measured in the soil can be very large and depth-dependent. This will be overlooked when making single-point measurements. We developed the datalogger Hypnos 2.0 for continuous redox potential and temperature measurements at various depths in the soil and without disturbance of the site. Hypnos is field-deployable, relatively cheap, and runs on batteries. The datalogger can use a "sleep mode" between sampling events. In sleep mode, there is no constant voltage on the Pt wire or the reference electrode, but there is only a short pulse during sampling. We did not measure an effect of this short pulse on the measured redox potential. In sandy soils in mesocosms and in a salt marsh soil we measured changes in the Eh as large as from –400 to +100 mV within 4 d, and daily cycles of 200 mV. Both absolute redox potential values and their diurnal variations were depth-dependent. Because single redox measurements are insufficient in describing redox conditions in some soil systems, Hypnos can be a powerful tool when studying the effects of fluctuating redox conditions on metal availability and pollutant degradation.
Abbreviations: Eh, redox potential
|
INTRODUCTION
|
|---|
REDOX POTENTIAL is an important parameter in environmental quality research. It indicates the tendency of an environment to receive or give electrons, or as stated by Mitsch and Gosselink (1993), "a measure of the electron pressure/availability in a solution." Redox potential is, together with pH, a driving variable for speciation of heavy metals (Salomons and Stigliani, 1995), and it describes the potential for degradation of organic substances. The redox potential is measured as voltage between the environment and a standard reference electrode. A popular, but also criticized, manual method makes use of a small tip of Pt placed on a copper wire that is placed in the soil; a reference electrode is placed in the same soil at a chosen distance (Cogger et al., 1992; Mansfeldt, 2003). After waiting for this system to stabilize (up to 2 d), the potential between the Pt tip and the reference electrode is measured. Then the measured potential is transformed into a standardized Eh (i.e., the redox potential relative to the standard hydrogen reference electrode).
The manual method is simple, but has several disadvantages. First of all, it can be very impractical to measure the redox potential in wetlands at all times due to flooding, weather conditions, and location of the sample spot. Also, since Eh can be a highly fluctuating value and sensitive to water pressure in the soil, one can never exclude the possibility that the presence of the operator close to the measuring spot in a field setting might induce errors in the redox values by building up pressure in the soil and changing the redox potential. Furthermore, fluctuations in redox conditions on a longer time scale time due to changes in flooding, weather conditions, and vegetation type or other causes are not conveniently recorded by the measuring method described above.
The redox potential is related to the concentration of several redox pairs in the soil; the most important are given in Table 1. Oxygen is the first acceptor that plays a large role. Oxygen diffuses into the soil, but can also be produced by plants and leak into the soil by radial oxygen loss from roots (Adema and Grootjans, 2003; Aldridge and Ganf, 2003; Jespersen et al., 1998). The amount of dissolved O2 produced by this process is time- and weather-dependent. Simultaneously, other redox pairs can fluctuate in concentrations as well. Consequently, fluctuations in redox potential values measured in the soil can be very large, but they may be overlooked (Mansfeldt, 2003). When studying redox behavior in situ, the fluctuations can be of great interest, however. For this purpose, Eshel and Banin (2002) have demonstrated the use of a setup that continuously measures the redox potential at ground water level. Their setup only functions within the vicinity of a bored well, which is not usable in situations where the importance of vegetation is studied. Another concept has been used by Tokunaga et al. (2001)(2003), but they did not measure continuously in situ. Their daily measurements were performed on a soil column in a laboratory using permanently installed Pt electrodes. This excluded the variations in the Eh values caused by weather conditions, growth of vegetation, and/or flooding. Seybold et al. (2002) measured the Eh every hour in a wetland at –20 and –50 cm, where Eh did not show much variation. Variations in the top layer of the soil (0 to –20 cm) were not measured. Seybold et al. (2002) used a combination of Pt electrodes, a control unit, and a multiplexer. The Hypnos datalogger we have developed incorporates these functions in one. Van Bochove et al. (2002) showed the use of an interface to be able to maintain a continuous current between the reference electrode and redox microelectrodes. They state that a continuous current is important. Their setup also used separate components, like Seybold et al. (2002). Swerhone et al. (1999) showed that the placement of the redox probes under reduced or oxidized conditions for a period of up to 3 yr should not impose problems regarding reliability of the measurements.
This paper describes the development of an experimental design for continuous redox measurements at various depths in the soil, with low disturbance of the site. We also show that daily variations of Eh can be large and that variations at short time intervals exist. We argue that such variations should be taken into account in long-term studies.
|
The Hypnos Redox and Temperature Logger
|
|---|
We have developed Hypnos, a continuous Eh and temperature measuring device for soil, to be able to measure redox potential at various depths in the soil over time without disturbance of our sites by the presence of the researcher. Hypnos incorporates a control unit, a multiplexer, storage memory, and Eh and temperature probes in one. It is a field-deployable, relatively cheap datalogger. The datalogger is connected to two probes and a reference electrode. Configuration and data collection is performed with a mobile computer. An electrical scheme of Hypnos is shown in Fig. 1. Hypnos has a maximum of four redox inputs and a maximum of 32 temperature inputs for each connected probe. Data are processed by a central PIC processor (PIC16F877) and stored in 256 Kbit ferroelectric nonvolatile random access memory (FRAM) serial memory. All devices are waterproof and electronically autonomous. They run on 9-V batteries providing a life span of at least 1 yr at a sampling interval of 6 min.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1. Electrical layout of Hypnos. Its main characteristics are industrial temperature range components (–20 to +70°C); maximum of eight redox inputs; very low power consumption (input current typically = 1 pA); electrostatic discharge protected input of 2 kV; surface mount assembly techniques; ferroelectric nonvolatile random access memory (FRAM); water resistance: IP67. MUX, multiplexer; RTC, real-time clock.
|
|
Figure 2 shows the layout of the probes. Probes are made of inert Perspex tubes (Stoutperspex, Rotterdam, the Netherlands) sealed with a pointed tip that is connected to the tube with ACRIFIX 192 glue (Röhm, Darmstadt, Germany). Three Pt wires for redox measurements (0.3-mm diameter with a 5-mm contact to the soil, purity 99.95%; Engelhard-CLAL/Drijfhout BV, Amsterdam, the Netherlands) and digital temperature sensors at the same three depths (Model DS18B20, ±0.5°C accuracy from –10°C to +85°C; Dallas Semiconductor, Dallas, TX) are installed. The Pt wires do not extend the probe surface. The temperature sensors are glued to the tube. Sensors and the Pt wires are connected to the datalogger using shielded computer cable (lengths of up to 15 m have been used). The top of the sensor is sealed with vulcanizing tape. The potential between the Pt wire and calomel reference electrode (R2; Sentek, Braintree, UK) is measured at time intervals preset by the user. Probes are pushed into the soil to the desired depth. The pointed tip ensures minimal disturbance of the soil layer when the probes are placed. Care is taken that organic litter is not transported down when pushing the probes into the soil.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 2. Layout of the redox–temperature probe. (A) Three-dimensional view. (B) Layout of probe, without the sensors placed. The probe consists of a Perspex tube, with holes drilled in it for the temperature sensors. At the same depths, Pt wires are placed on the outside of the probe, slightly sunken into the tube ensuring minimal damage to the soil profile when the probe is pushed into the soil. The pointed tip is glued to the tube. All sensors and Pt wires are connected to the Hypnos datalogger with isolated computer cable (connecting wires not shown).
|
|
Calibration of the datalogger has been performed with a stabilized potential allowing calibration of each redox channel. This makes each probe unique. Each separate calibration is used by the computer program, so no conversion for the measured potential difference to Eh is needed. Calibration and temporal checks after experiments were performed by placing the probe and the reference electrode in a redox buffer solution (ORP buffer solution RH28, 220 mV, pH 7 at 25°C; WTW Measurement Systems, Woburn, MA). Differences between measured values and calculated values of the reference solutions were 10 mV or less. After 8 mo of deployment, measured Eh was still within 10 mV of the original value. We concluded that our probes still function well after 8 mo of intensive use, as found by others (Austin and Huddleston, 1999). The reference electrodes did not show a change in functioning after 1.5 yr of application.
Configuration of and data collection from the datalogger are performed with the computer program Hypnos Datacollector Version 1.4, created using LabView (National Instruments Corporation, 1998). Programming features of the Hypnos Datacollector include the measurement interval and time of the first sample. The computer program communicates via an RS232 connection with the logger. At each reading, collected Eh and temperature data are plotted on the screen. In the configuration of a 6-min interval, the logger can measure independently for 6 wk before data storage meets its limit. Figure 3 shows the setup in a field setting.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 3. Overview of the setup with Hypnos in the field. The probes with redox and temperature (REDOX/TEMP) electrodes are placed at a distance of several meters from the datalogger. The datalogger has been placed on a 2-m-high stake to prevent direct contact with water during flooding of the area. The calomel reference electrode (REF) can be placed anywhere near the datalogger. This setup allows sampling without disturbance when taking data from the logger.
|
|
Verification of the Redox Measurements
The datalogger can use a "sleep mode" between sampling events. In sleep mode, there is no constant voltage on the Pt wire and the reference electrode, but there is only a short pulse during sampling. This sleep mode saves the batteries, enabling a longer sampling period. It was claimed before that it is important to maintain a current between the redox reference electrode and the Pt electrode (Van Bochove et al., 2002). As pointed out by some other authors (e.g., Böttcher and Strebel, 1988), there can be problems with polarization of the probes. Polarization can influence the value of measured Eh. To check if our measurements are influenced by the sleep mode we performed an experiment in which we placed two electrodes in a layer of organic soil. The soil had been placed at 16°C in a climate room for more than 1 yr. One of the probes was connected to a datalogger in sleep mode, the other was connected to a datalogger with a continuous voltage. After several hours, the dataloggers were switched. The switching was repeated three times. Differences between the measured Eh in sleep mode and in continuous mode were –2 to 10 mV (with a single peak to 15 mV) depending on the Eh measured (the higher the Eh, the larger the difference). Variations in the measurements in time were 15 mV around a steady state level. The dataloggers in sleep mode measured a steady state level in Eh after 24 h. The loggers in a continuous mode took less time to stabilize but still required some hours after every switch. These results led us conclude that the measurement error of the Hypnos redox logger and the connected probes is 10 mV in sleep mode, less than variation due to other sources. It was also concluded that it is not important to maintain a continuous current when using Hypnos. We have used the sleep mode in all our measurements.
|
Demonstrations
|
|---|
First deployment of Hypnos was in a mesocosm experiment, previously set up by Leendertse et al. (1996). In short, it consisted of a mesocosm with an inflow of natural saltwater from the nearby Wadden Sea of the northern Netherlands. The water was led over sedimentation basins to remove solids. Water was not modified in any other way and in general was rich in oxygen. The sandy soil (60–80% sand, 15% organic matter in the top 2 cm, 1–5% organic matter below –2 cm) was flooded twice a day up to soil level and after 6 h of flooding it was allowed to drain. Once every week the soil was flooded at 30 cm above soil level. Flooding of soils is expected to have an influence on the measured redox potential. Daily temperature cycles can be an explanation for the diurnal variation in Eh. Both Eh and temperature were sampled at –1, –4, and –9 cm depths in the soil at a 5-min interval. In Fig. 4A we show the results for the Eh at all three depths. In Fig. 4B only the temperature at –1 cm is shown for clarity.
View larger version (31K):
[in this window]
[in a new window]
|
Fig. 4. Results of two in situ measurements. The redox potential at three depths and the temperature at the top level are shown for the mesocosm experiment (A and B) and the salt marsh (C and D). Numbers indicate the depth of measurement (cm). Note the daily fluctuations and range of fluctuations at each depth over time.
|
|
A second deployment was in the salt marsh of Holwerd, Province of Friesland, the Netherlands (5°54'10'' E, 53°23'00'' N). Soil pH values ranged from 6.5 to 7.8. Eh was not corrected for pH. In Fig. 4C we show the results for the Eh at all three depths (measured at –2, –5, and –10 cm), and in Fig. 4D the temperature at –2 cm. Both were sampled at a 6-min interval. The salt marsh is a dynamic ecosystem with periodic flooding, a Sulfaquent soil (30–40% clay, 40–50% silt, 5–10% organic matter), and periodic freezing. A temperature rise will cause a subsequent rise in Eh, following the Nernst equation. Daily temperature changes can only explain part of the diurnal variation in Eh in the salt marsh. The larger shifts can be explained by the floodings appearing in the salt marsh.
We measured changes in Eh as large as from –400 to +100 mV within 4 d, and daily cycles with a range of 200 mV in sandy soils in the mesocosms. The more clay-rich soil in the salt marsh showed less variation than the sandy mesocosm soil, probably due to the lower diffusion rate of oxygen in this type of soil, and the larger retention of moisture in the top layer. The redox potentials at the three depths had variable values, but also had a different variation. The levels at –2 cm are stable in the salt marsh, but vary greatly in the periodically flooded mesocosms. The drainage of the sandy soil in the mesocosm is much better than the drainage of the clayey soil in the salt marsh and most probably explains the difference in Eh fluctuations between the mesocosm and the salt marsh (Vorenhout, unpublished data, 2003).
|
Conclusions
|
|---|
With Hypnos one can measure Eh and temperature continuously. Hypnos has been used in moist to wet conditions, but its functioning under dry conditions has not been tested up to now. Variations of Eh can be very large in wetland systems, as shown by other authors, but only by continuous measurements can one determine the possible periodicities and time amplitude of the variations. When measuring redox, the soil profile also has to be taken into account. Single redox measurements at one depth can be insufficient in describing redox conditions in soil systems. Therefore, Hypnos can be a powerful tool when studying the effects of fluctuating redox conditions that ultimately influence metal bioavailability and pollutant degradation.
|
ACKNOWLEDGMENTS
|
|---|
This research was funded by the World Wildlife Fund for Nature and the Prince Bernhard Cultural Foundation. We thank the mechanics department of the Faculty of Earth and Life Science, Vrije Universiteit, Amsterdam, for their help in constructing the devices. More information about Hypnos, including building plans, can be obtained from the corresponding author.
|
REFERENCES
|
|---|
- Adema, E.B., and A.P. Grootjans. 2003. Possible positive-feedback mechanisms: Plants change abiotic soil parameters in wet calcareous dune slacks. Plant Ecol. 167:141–149.
- Aldridge, K.T., and G.G. Ganf. 2003. Modification of sediment redox potential by three contrasting macrophytes: Implications for phosphorus adsorption/desorption. Mar. Freshwater Res. 54:87–94.
- Austin, W.E., and J.H. Huddleston. 1999. Viability of permanently installed platinum redox electrodes. Soil Sci. Soc. Am. J. 63:1757–1762.[Abstract/Free Full Text]
- Böttcher, J., and O. Strebel. 1988. Calculation of redox potential in soil and groundwater from the curve of the electrode depolarization. Z. Pflanzenernaehr. Bodenkd. 151:363–368.
- Cogger, C.G., P.E. Kennedy, and D. Carlson. 1992. Seasonally saturated soils in the Puget Lowland. 2. Measuring and interpreting redox potentials. Soil Sci. 154:50–58.
- Eshel, G., and A. Banin. 2002. Feasibility study of long-term continuous field measurement of soil redox potential. Commun. Soil Sci. Plant Anal. 33:695–709.
- Jespersen, D.N., B.K. Sorrell, and H. Brix. 1998. Growth and root oxygen release by Typha latifolia and its effects on sediment methanogenesis. Aquat. Bot. 61:165–180.
- Leendertse, P.C., M.C.T. Scholten, and J.T. van der Wal. 1996. Fate and effects of nutrients and heavy metals in experimental salt marsh ecosystems. Environ. Pollut. 94:19–29.[Medline]
- Mansfeldt, T. 2003. In situ long-term redox potential measurements in a dyked marsh soil. J. Plant Nutr. Soil Sci. 166:210–219.
- Mitsch, W.J., and J.G. Gosselink (ed.) 1993. Wetlands. 2nd ed. Van Nostrand Reinhold, New York.
- National Instruments Corporation. 1998. LabView 5.01. Natl. Instruments Corp., Austin, TX.
- Salomons, W., and W.M. Stigliani. 1995. Biogeodynamics of pollutants in soils and sediments. 1st ed. Springer, Berlin.
- Seybold, C.A., W. Mersie, J. Huang, and C. McNamee. 2002. Soil redox, pH, temperature, and water-table patterns of a freshwater tidal wetland. Wetlands 22:149–158.
- Swerhone, G.D.W., J.R. Lawrence, J.G. Richards, and M.J. Hendry. 1999. Construction and testing of a durable platinum wire Eh electrode for in situ redox measurements in the subsurface. Ground Water Monit. Rem. 19:132–136.
- Tokunaga, T.K., J. Wan, M.K. Firestone, T.C. Hazen, E. Schwartz, S.R. Sutton, and M. Newville. 2001. Chromium diffusion and reduction in soil aggregates. Environ. Sci. Technol. 35:3169–3174.[Medline]
- Tokunaga, T.K., J.M. Wan, T.C. Hazen, E. Schwartz, M.K. Firestone, S.R. Sutton, M. Newville, K.R. Olson, A. Lanzirotti, and W. Rao. 2003. Distribution of chromium contamination and microbial activity in soil aggregates. J. Environ. Qual. 32:541–549.[Abstract/Free Full Text]
- Van Bochove, E., S. Beauchemin, and G. Thériault. 2002. Continuous multiple measurement of soil redox potential using platinum microelectrodes. Soil Sci. Soc. Am. J. 66:1813–1820.[Abstract/Free Full Text]
Related articles in JEQ:
- This Issue in Journal of Environmental Quality
JEQ 2004 33: 1177-1182.
[Full Text]
TOP
ABSTRACT
INTRODUCTION
