| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Assistant Professor of Biological and Agricultural Engineering, Kansas State Univ., 147 Seaton Hall, Manhattan, KS 66507
b Associate Professor of Civil Engineering, Purdue Univ., 1284 Civil Engineering Building, West Lafayette, IN 47907
c Associate Professor of Agronomy, Purdue Univ., 1150 Lily Hall of Life Sciences, West Lafayette, IN 47907-1150
Corresponding author (sllhutch{at}ksu.edu)
Received for publication July 2, 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: EC, electrolytic conductivity • GC, gas chromatography • TPH, total petroleum hydrocarbons
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Phytoremediation can use several different mechanisms for remediating contaminants including phytoaccumulation, phytostabilization, phytodegradation, phytovolatilization, and rhizosphere degradation (McCutcheon, 1998). The primary phyto-mechanism used is dependent on the compound being degraded. Because of the chemical characteristics of petroleum hydrocarbons, it is not likely that petroleum hydrocarbons will be taken up into the plant, thus the primary phytoremediation mechanism is rhizosphere degradation or plant-assisted bioremediation (Briggs et al., 1982). Therefore, one of the key components of petroleum sludge phytoremediation is microbial activity in the soil because of its effect on carbon mineralization rates. The size and function of the microbial community are influenced by soil moisture content, aeration, and substrate and nutrient availability. The interaction of these factors creates specialized environments that select for specific microorganisms capable of degrading petroleum hydrocarbons.
The activities of the heterotrophic microbial community in soils are driven primarily by the oxidation of organic carbon that enters the soil ecosystem as root exudates, plant litter, manure, compost, or industrially produced waste materials (e.g., petroleum by-products and industrial wastes) (Tate, 1997, p. 93–121 and 171–228). In addition to an energy source, cellular growth requires a variety of macronutrients including nitrogen, phosphorus, and sulfur (Tate, 1997, p. 93–121 and 171–228).
The inorganic nutrients that are most often limiting in the bioremediation of hazardous organic compounds are nitrogen (N) and phosphorus (P). Although not all bioremediation systems respond to nutrient additions, the addition of N fertilizer has been observed to enhanced bioremediation of oil contamination in aquatic systems and soils (Lin and Mendelssohn, 1998; Churchill et al., 1995; Glaser, 1991; Rasiah et al., 1992). Carbon mineralization rates increased in response to the addition of fertilizer, indicating the importance of sufficient nutrients in enhancing oily waste decomposition in soil (Rasiah et al., 1992). Graham et al. (1995) conducted a laboratory experiment on the application of various nutrients to soil to optimize biodegradation of xylenes, anthracene, phenanthrene, and n-hexadecane. Biodegradation was enhanced by nutrient addition, and each type of hydrocarbon reacted differently to various levels of fertilization.
In all vegetated systems, plants compete with microorganisms for required nutrients including carbon, hydrogen, oxygen, potassium, calcium, magnesium, nitrogen, phosphorus, sulfur, iron, zinc, copper, magnesium, chlorine, boron, and molybdenum (Gardner et al., 1985, p. 98–131). Plants are capable of using light as their energy source to transform CO2 and H2O into organic compounds and use inorganic nutrients drawn from the environment to synthesize required amino acids and vitamins (Raven et al., 1986, p. 515–540).
Degradation of petroleum hydrocarbons appears to be enhanced in the rhizosphere, and plants with a healthy and extensive root system are speculated to be beneficial to bioremediation. Root biomass generally increases as the aboveground biomass increases. Large healthy plants often support extensive root systems that are capable of pumping more water through the rhizosphere, transporting more nutrients through the soil system, and providing increased surface area for microbial activity.
There is also evidence to suggest that (i) the availability of organic chemicals to soil microorganisms declines with time; (ii) freshly added chemicals are available to soil microorganisms whereas identical, aged compounds are not metabolized; (iii) organic compounds incubated in sterile soils become increasingly less available to subsequently added microorganisms; (iv) aged compounds often are resistant to extraction; and (v) sorption and desorption of hydrophobic compounds often require long time periods to reach equilibrium (Alexander, 1995). Studies conducted on organic pesticides demonstrated that the compounds sequestered into the soil's organic fraction and molecules that diffuse into internal micropores and other remote sites are effectively unavailable to microorganisms and the biological food chain (Atlas and Bartha, 1993, p. 383–412). Despite the presence of presumably toxic concentrations of organic contaminants as determined through rigorous chemical extractions, aged compounds may not be bioavailable (Alexander, 1995; Loehr and Webster, 1996). Thus, biodegradation of contaminants may slow to imperceptible rates. However, the presence of vegetation may enhance remediation due to soil disturbance by roots and increased microbial activity.
The objective of this study was to determine the effect of vegetation and rates of N and P fertilization on contaminant degradation in aged petroleum sludge. A greenhouse phytoremediation experiment was conducted using petroleum sludge from an oil refinery in California. Plant shoot and root biomass, microbial counts, and dissipation of contaminants were measured.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Soil Characterization and Preparation
The medium ("soil") used in this experiment was a petroleum sludge generated during the refining process at a California oil refinery. The sludge initially contained 48800 mg TPH kg-1 (Table 1) with carbon chains ranging from C10 to C35. The petroleum sludge had been held in clay-lined oxidation ponds for more than 40 yr.
|
Prior to packing the sludge into greenhouse pots, the petroleum sludge was ground to pass a 6-mm sieve to obtain a homogenous mixture that was subsequently air-dried and thoroughly mixed using a large volume mixer for each replicate.
Greenhouse pots (7.5 L) were filled with 5 kg of petroleum sludge with a moisture content of 10% at a bulk density of 1 g cm-3. The bottom of the pot was lined with 5 mm of glass wool and a 4-cm layer of gravel to allow for improved drainage and to minimize loss of petroleum sludge from the pot. Vinyl pans were placed under each pot to catch leachate. A sample was taken from each pot to determine initial total petroleum hydrocarbon (TPH) concentration and variance among pots.
Plant Species and Fertilization
Grass plugs of tall fescue and bermuda grass from the Kansas State University horticultural turf farm were planted in each vegetated treatment. Prior to planting, the petroleum sludge was saturated and allowed to settle. A 7.5-cm-diam., 4-cm-thick turf plug, obtained using a soil sampler, was placed in each pot after the clean soil was washed from the roots.
Fertilization rates were chosen based on a C to N to P ratio of 100:10:1 (Cookson, 1995) used to calculate the total fertilizer N and P for bioremediation of soils contaminated with organic compounds. Using this ratio, the maximum fertilization rate was calculated assuming that the entire mass of C in TPH was labile, but the maximum amount of N or P to be applied was reduced by 50% to account for the short duration of the experiment. Thus, the highest fertilization application rate (F6, Table 2) was 2500 mg N kg-1 and 250 mg P kg-1. One treatment was no fertilizer added (F1), and F2 was the maintenance addition of N and P as recommended for maintaining turf grasses. Rates F3, F4, and F5 were various combinations of N and P at rates below the maximum.
|
Plant Maintenance
The pots were irrigated using a drip irrigation system. Evapotranspiration was estimated using a modified pan evaporation technique (Soil Conservation Service, 1993) where the weight of evaporated water from a clear plastic pan was measured and recorded three times weekly. The flow rate from the irrigation system was calibrated four times throughout the experiment to ensure consistent water application along the drip irrigation lateral. Each plant treatment received water applications scheduled to match water use. Immediately prior to the addition of fertilizer, the unvegetated control pots received 1 L of water while vegetated treatments required 1.5 L of water to produce a leachate and remove excess salts. The leachate was analyzed for available N and TPH, as described below. Due to controlled irrigation, there was no leaching between samplings.
Aboveground biomass was cut every 7 to 8 wk to a height of 5 cm. Biomass was collected, dried, and reported as total dry weight at each of the two sampling dates (6 and 12 mo). Roots were collected when the pot was sacrificed at the end of each time period. At that time, the petroleum sludge was homogenized by hand mixing for chemical and microbial analysis. A subsample of the petroleum sludge was air-dried and ground to pass a 0.25-mm sieve for TPH analysis as described below. A moist sample was retained and stored at 4°C for subsequent microbial analysis.
Extraction and Analysis
Petroleum Sludge Extraction
Total petroleum hydrocarbons was extracted from the petroleum sludge using a sequential shaking extraction (Schwab et al., 1999). The petroleum sludge was air-dried and mechanically ground to pass through a 0.25-mm mesh sieve. One gram of air-dried petroleum sludge was spiked with 100 µL of 100 mg L-1 tetracosane and shaken on a reciprocal shaker with 10 mL Optima grade dichloromethane (Fisher Scientific, St. Louis, MO) for 30 min then centrifuged at 2000 rpm for 10 min to settle the petroleum sludge. The supernatant was collected in 60 mL glass jars. This procedure was repeated two more times, and the mass of the final extractant was recorded. A 1.5-mL sample was transferred into a 2-mL gas chromatography (GC) vial, spiked with 10 µL of 1000 mg L-1 
-androstane, capped with a Teflon-lined septa, and stored at 4°C until analysis by GC.
Leachate Extraction
Leachate was collected from one replicate of each treatment prior to fertilizer application and analyzed for electrolytic conductivity (EC), TPH, and nitrate. The EC was measured using a Model 31 Conductivity Bridge (Yellow Springs Instruments, Yellow Springs, OH). Before nitrate analysis, TPH was extracted using solid phase C18 cartridges (J&W Scientific, Folsom, CA). The cartridge was conditioned with 5 mL dichloromethane (DCM), 5 mL methanol, and 5 mL deionized water distilled in the presence of potassium permanganate. Then 10 mL of leachate was loaded onto the cartridge. The effluent was collected for NH+4 and NO-3 analysis. The TPH was eluted from the C18 cartridges with 3 mL DCM. The sample was dried with anhydrous sodium sulfate. A 1-mL aliquot of the eluent was transferred to GC vials, spiked with androstane, capped with a Teflon-lined septum, and stored at 4°C until analysis.
Gas Chromatographic Analysis
All extracted samples were analyzed on a Hewlett–Packard (Palo Alto, CA) 5890A Gas Chromatograph (GC) equipped with a flame ionization detector (FID), HP Chemstation Integration Software, and an HP7673A autosampler utilizing a DB-TPH column. The DB-TPH column, designed specifically for analysis of TPH, had an inside diameter of 0.32 mm, a length of 30 m, and a 0.25-µm film thickness. The fuel and carrier gas for the FID was H2. The carrier gas was delivered at 67 µL s-1 and the fuel gas at 670 µL s-1. Nitrogen was used as the make-up gas at a flow rate of 530 µL s-1. Air was supplied as the oxidant at a rate of 7.0 µL s-1. The initial oven temperature was maintained at 40°C for 2 min, then increased at 0.2°C s-1 to 320°C and maintained at 320°C for 1 min for leachate samples and 10 min for petroleum sludge samples. The temperature of the injection port was 250°C and the detector temperature was 350°C. The splitless injection volume was 2 µl. The TPH concentrations were measured using integrated GC areas and then converted to concentrations using external standard calibration curves.
Microbial Activity
Microbes were enumerated using viable heterotrophic plate counts by the spread plate method (Pepper et al., 1995). Ten grams (dry weight) of petroleum sludge were diluted in 95 mL of 0.2% tetrasodium pyrophosphate for soil dispersal (Alef and Nannipieri, 1995, p. 503–505), shaken on a rotary shaker at 150 rpm for 30 min, and allowed to settle for 10 min. The solute was diluted serially in 0.85% sodium chloride (NaCl), transferring 1 mL of solution into 9 mL of NaCl using a sterile 1-mL syringe. Two dilutions were selected for the plating procedure to obtain a concentration of microorganisms that had from 30 to 300 colony forming units per plate. Each sample was diluted in triplicate, and the average value of the three replicates was used for calculations. Petri plates of tryptic soy agar were inoculated with an aliquot of 0.1 mL. The plates were surface spread using an alcohol flame sterilized spreader (Wollum, 1982). Once inoculated, the plates were inverted to prevent water from settling on the plate surface, and incubated at 25°C for 96 h. Plates were counted for colonies at 24-h intervals for 72 h.
Statistical Analysis
The greenhouse experiment was arranged in a split plot design. Analysis of variance was performed by SAS (SAS, Cary, NC). Unless stated otherwise, all main effects or interactions discussed below are statistically significant (P < 0.05).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Prior to each fertilizer application, the pots were leached with excess irrigation to reduce salt buildup. The amount of leachate generated varied with vegetation treatment from an average of 500 mL in the unvegetated control to less than 100 mL in the higher fertilization treatments for both bermuda and fescue. In all treatments, most samples contained no detectable NH+4 or NO-3, with the remaining samples containing up to a maximum of 5 mg L-1 NH4–N and 4 mg L-1 NO3–N. The small quantities of nitrogen found in the leachate indicate that the applied fertilizer was being used by the plants and microorganisms. Measured EC varied during the experiment and with treatment (Fig. 1) . The EC of the unvegetated treatment was consistently higher throughout the experiment, indicating that more salt was accumulated and leached from this treatment.
|
|
Plant Biomass
Over the course of the experiment, bermuda generated almost twice as much biomass as fescue in all fertilizer treatments (Fig. 3)
. Growth in both plant species was significantly less at the lower fertilization rates, indicating that N and perhaps P were limiting for plant growth under these circumstances. Although neither N nor P was added in the absence of the other, the high initial Bray P concentration (191 mg P kg-1) suggests that P was not limiting.
|
Bermuda had more root biomass than fescue for all treatments (Fig. 4) , which is consistent with the shoot biomass. During the first 6 mo, root biomass was highly variable and neither plant species showed any trend related to fertilization. After 1 yr, the root biomass of bermuda was almost constant with no effect from the fertilization rates. Fescue, on the other hand, produced the greatest biomass with intermediate fertilization (F3 and F4) and biomass was significantly less at low and high fertilization rates.
|
|
|
|
|
Fertilization had a significant effect on changes in TPH from 6 to 12 mo. Within the unvegetated control treatment, the effect of fertilization on percent degradation was not significant at P < 0.05, but degradation was greater for F5 and F6 compared with F1 and F2 at P < 0.10. For fescue, degradation was greater in treatments F3, F5, and F6 compared with F1 and F2 (P < 0.05); only F5 had greater degradation than F1 and F2 for bermuda. Relative to the unvegetated controls, percent degradation was greater with bermuda for all treatments except F6 (which was significantly greater at P < 0.10). Degradation in the fescue pots was significantly greater than in the unvegetated controls for F2, F3, F5, and F6. After initially equivalent degradation in all treatments during the first 6 mo, TPH reduction was slower in the unvegetated control than both vegetated treatments during the second 6 mo.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The time required for shoot and root development also contributed to the lack of significant differences between vegetated treatments during the first 6 mo. Neither plant species showed signs of distress due to toxicity, but both were slow to establish with bermuda producing an average accumulated dry biomass of 74 g shoots and 30 g roots, while fescue produced only 35 g of shoots and 20 g of roots. During the second 6 mo of the experiment, above- and belowground biomass increased by an average of 300% for both plant species, which increased transpiration rates, and the enhanced removal of water from the soil profile resulted in greater aeration.
During the second 6 mo of the experiment, microbial counts increased by an order of magnitude in all treatments, perhaps a result of decreased soil toxicity. Significantly higher bacterial counts in fescue may account for the increased degradation rate observed in fescue during the latter half of the study. However, bacterial plate counts may not represent petroleum-degrading microbes, which makes these data difficult to interpret in terms of phytoremediation.
The control and maintenance fertilization treatments (F1 and F2) were associated with significantly smaller percent TPH degradation at 6 and 12 mo, but only for the fescue and bermuda treatments. The lack of an effect of fertilizer on TPH degradation in the unvegetated treatment would suggest that these materials were not N and P deficient for bioremediation. The presence of plants appears to stimulate microbial TPH degradation and requires more N and P than when plants are absent. This is further illustrated by the residual, total N in the different treatments at the end of the experiment (Table 4). For F6, increases in total N were the greatest in the unvegetated pots and the least in bermuda. Responses were similar for total P. This is most likely the result of the very high use of N for biomass production in the bermuda, intermediate use by fescue, and the absence of fertilizer demand by plants in the unvegetated pots. Nearly 50% of the N and P applied in F6 could be found in this tissue of bermuda and 25% in fescue (data not shown).
Because previous studies have demonstrated successful bioremediation with a wide range of nutrient application in freshly contaminated soils (Churchill et al., 1995; Glaser, 1991; Graham et al., 1995; Rasiah et al., 1992), an aged contaminated system may be carbon limiting, not N and P limiting. If the organic contaminant was sequestered in microsites and not bioavailable, the reduction rate would be dependent upon contaminant release, not the addition of N and P. This would account for the lack of differences in degradation in response to fertilization and the slower degradation in the second half of the study for the unvegetated pots. Mineralization of organic N and P have supplied adequate nutrients for biodegradation in the absence of plants.
Another explanation for the observed differences in TPH reduction between vegetated and unvegetated treatments is lack of O2 in the unvegetated soil profile. Roots create macropores and preferential channels that facilitate the transport of gases and liquids through the soil column. As the belowground biomass increased, the soil column was more likely to maintain soil, air, and water conditions that support active aerobic microbial activity.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The decline in contaminant degradation observed in the unvegetated control was typical of unplanted bioremediation, where carbon mineralization is enhanced through tillage and nutrient addition. The initial sludge processing was similar to tillage in the field. Because tillage is unable to continue reducing aggregate size to expose all contaminated surfaces, TPH degradation slows considerably with time. Plant roots are able to penetrate the soil and break up large aggregates, and contaminated surfaces are continually being exposed and higher degradation rates maintained. This was apparent in the latter stages of the experiment where the average degradation rate in the vegetated treatments (27%) was 1.7 times the rate in the unvegetated control (16%).
Microbial activity was assessed at each harvest date using viable heterotrophic plate counts. Increased microbial numbers were observed in the planted treatments, but fertilization rates did not significantly affect microbial populations. Because the sludge had a high level of organic N and total P, microorganisms may have been able to obtain the nutrients required for growth from the sludge.
The highest rate of fertilization (F6) ultimately applied 2500 mg kg-1 N and 250 mg kg-1 P, or a ratio of 100:5:0.5, which was half of the target ratio of 100:10:1 for C to N to P suggested by Cookson (1995). This rate of fertilization produced excessive N and P fertilization when one considers overall TPH degradation, plant biomass production, and residual N in the soil. Nitrate and total N accumulated in the soil except in the bermuda pots, which produced very high quantities of shoot biomass. For this petroleum sludge, rate F3 resulted in remediation equivalent to F6 with no nitrate accumulation in the soil. The relatively low demand for fertilizer was probably the result of the presence of significant concentrations of total N and P, incomplete bioremediation of the contaminant, and efficient use of nutrients. Thus, for this material, we would recommend a C to N to P ratio of 100:2:0.2 for phytoremediation. For other petroleum-contaminated sites, initial assessment of the soil and contaminant will be useful in determining the ideal fertilization rate.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. E. Olson, A. Castro, M. Joern, N. M. DuTeau, E. A. H. Pilon-Smits, and K. F. Reardon Comparison of Plant Families in a Greenhouse Phytoremediation Study on an Aged Polycyclic Aromatic Hydrocarbon Contaminated Soil J. Environ. Qual., August 31, 2007; 36(5): 1461 - 1469. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. D. Parrish, M. K. Banks, and A. P. Schwab Effect of Root Death and Decay on Dissipation of Polycyclic Aromatic Hydrocarbons in the Rhizosphere of Yellow Sweet Clover and Tall Fescue J. Environ. Qual., January 1, 2005; 34(1): 207 - 216. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. T. Koivula, M. Salkinoja-Salonen, R. Peltola, and M. Romantschuk Pyrene Degradation in Forest Humus Microcosms with or without Pine and its Mycorrhizal Fungus J. Environ. Qual., January 1, 2004; 33(1): 45 - 53. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Hutchinson, A. P. Schwab, and M. K. Banks Phytoremediation of Aged Petroleum Sludge: Effect of Irrigation Techniques and Scheduling J. Environ. Qual., September 1, 2001; 30(5): 1516 - 1522. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal |
and 12 mo 
comparing vegetation treatments