OUP user menu

Hexadecane mineralization and denitrification in two diesel fuel-contaminated soils

Réal Roy, Charles W. Greer
DOI: http://dx.doi.org/10.1111/j.1574-6941.2000.tb00694.x 17-23 First published online: 1 April 2000

Abstract

The effect of nitrate, ammonium and urea on the mineralization of [14C]hexadecane (C16H34) and on denitrification was evaluated in two soils contaminated with diesel fuel. In soil A, addition of N fertilizers did not stimulate or inhibit background hexadecane mineralization (4.3 mg C16H34 kg−1 day−1). In soil B, only NaNO3 stimulated hexadecane mineralization (0.91 mg C16H34 kg−1 day−1) compared to soil not supplemented with any nitrogen nutrient (0.17 mg C16H34 kg−1 day−1). Hexadecane mineralization was not stimulated in this soil by NH4NO3 (0.13 mg C16H34 kg−1 day−1), but the addition of NH4Cl or urea suppressed hexadecane mineralization (0.015 mg C16H34 kg−1 day−1). Addition of 2 kPa C2H2 did not inhibit the mineralization process in either soil. Denitrification occurred in both soils studied when supplemented with NaNO3 and NH4NO3, but was not detected with other N sources. Denitrification started after a longer lag in soil A (10 days) than in soil B (4 days). In soil A microcosms supplemented with NaNO3 or NH4NO3, rates of denitrification were 20.6 and 13.6 mg NO3 kg−1 day−1, respectively, and in soil B, they were 18.5 and 12.5 mg NO3 kg−1 day−1, respectively. We conclude that denitrification may lead to a substantial loss of nitrate, making it unavailable to the mineralizing bacterial population. Nitrous oxide was an important end-product accounting for 30–100% of total denitrification. These results indicate the need for preliminary treatability studies before implementing full-scale treatment processes incorporating commercial fertilizers.

Keywords
  • Soil
  • Bacterium
  • Nitrate
  • Ammonium
  • Hexadecane
  • Denitrification
  • Nitrous oxide

1 Introduction

In the context of the biodegradation of organic pollutants, such as petroleum hydrocarbons, in terrestrial environments, addition of commercial fertilizers is a common practice to stimulate the soil indigenous microflora in degrading the target pollutants (biostimulation) [1, 2]. In a study of the Exxon Valdez oil spill in Alaska, it was found that the most critical factor for successful bioremediation was the concentration of NH4+ in the pore water following the application of an oleophilic fertilizer [3]. A commercial fertilizer typically contains three forms of N: urea, nitrate and ammonia. Although these nutrients may all be directly assimilated by various soil microorganisms, including those involved in the degradation of petroleum hydrocarbons, they may also be transformed by other members of the soil microbial community [4]. For instance, mineralization of urea may lead to the production of NH4+ which in turn may be oxidized to NO2 and NO3 by nitrifying bacteria [5]. Ultimately, NO2 and NO3 may be dissimilated to N2 by denitrifying bacteria under anoxic conditions [6]. This last process leads to a loss of N fertilizer from terrestrial ecosystems that may represent up to 1 kg N ha−1 year−1 in fertilized agricultural fields [79]. In the context of bioremediation of contaminated soil, little is known about the fate of N fertilizer.

Denitrification may also be important in the context of mineralization of petroleum hydrocarbons under anoxic conditions [10]. A growing number of studies are showing that, under denitrifying conditions, several aromatic compounds, for instance toluene, xylene, phenols, cresols, and naphthalene may be degraded [11, 6]. Only recently, Bregnard et al. [12] have shown the degradation of an aliphatic compound, pristane, under denitrifying conditions. In the field, anoxic microsites, in otherwise oxic soils, frequently occur in soil aggregates [13, 14]. Under such conditions, in a diesel-contaminated soil, denitrification may occur and contribute to the biodegradation of petroleum hydrocarbons. In this case, the addition of a commercial fertilizer containing nitrate may also act as a source of electron acceptors for indigenous denitrifying diesel-degrading bacteria.

The objective of the present study was to evaluate the effect of various nitrogen fertilizers (urea, NH4Cl, NaNO3 and NH4NO3) on the mineralization of hexadecane (C16H34) in aerobic microcosms of two different soils with similar levels of C10-C50 contamination. Denitrification which may be responsible for loss of N fertilizer in soil was also evaluated during the mineralization process using the acetylene (C2H2) blockage technique [15, 16].

2 Materials and methods

2.1 Soils

The characteristics of the two soils used for these studies are presented in Table 1. Standard procedures were used for the determination of total petroleum hydrocarbon (C10-C50) [17], total carbon [18], total Kjeldahl nitrogen [19] and total phosphorus [20], which were performed either in our lab or by the Soil Testing Lab of McGill University. The water content was measured by an Electronic Moisture Analyzer MA30 (Sartorius, Goettingen, Germany). Both soils were sampled from diesel-contaminated areas. Soil A was from a military installation in Bagotville, Que., Canada, and soil B was from a gas station in Montreal, Que., Canada. Soils were stored at 4°C until used. Both soils had similar levels of petroleum hydrocarbon contamination (2000 and 3800 mg kg−1 soil), and the total carbon and total mineral nitrogen contents were higher in soil A than in soil B. The C/N ratios, on a mass basis, were 32 and 81 in soils A and B, respectively. Total phosphorus and water contents were lower in soil A than in soil B.

View this table:
1

Summary of the characteristics of the soils studied

SoilC10-C50 (mg kg soil−1)C (mg kg soil−1)N (mg kg soil−1)P (mg kg soil−1)C/NH2O content (%)
A2 00044 0001 388633254
B3 80024 3003009308175

2.2 Soil microcosms

Experiments were performed in aerobic soil microcosms consisting of 10 g (wet weight) of soil in 100-ml glass serum bottles capped with teflon-lined rubber stoppers and sealed with aluminum crimps. Unlabeled hexadecane (50 mg ml−1), dissolved in hexane, was added as 20-μl aliquots to all soil microcosms to a final concentration of 100 mg kg−1 soil. Sterile stock solutions of NaNO3 (1 M), NH4Cl (1 M), NH4NO3 (0.5 M) and urea (0.5 M) were added separately in 400-μl aliquots to soil microcosms and no change in soil pH greater than 0.3 units occurred following the additions. These N supplements were equivalent to 560 mg N kg−1 wet soil, which is in the upper range for fertilization studies, but typical for denitrification assessments [21, 22]. In control flasks, 400 μl sterile distilled water was added instead of nitrogenous salts solutions. For each treatment, triplicate flasks were prepared. Moreover, two sets of microcosms were prepared: one set for the determination of [14C]hexadecane mineralization and the other for the determination of denitrification rates using the acetylene blockage technique [15]. All microcosms were incubated statically in the light at 25°C. Sterile microcosms were prepared as previously described for the control. Soil microcosms were sterilized for 1 h on two consecutive days. Hexadecane and C2H2 (2 kPa) were added to these microcosms.

2.3 Hexadecane mineralization

The mineralization experiments followed a modified technique from Whyte et al. [23]. A 10-ml test tube containing 0.5 M KOH (1 ml) was placed inside the soil microcosms (CO2 trap). A mixture of [1-14C]hexadecane (specific activity 2.2 mCi mmol−1 or 0.081 GBq mmol−1) (Sigma–Aldrich, Mississauga, Ont., Canada) with unlabeled hexadecane (50 mg ml−1 in hexane) was added to each soil microcosm as 20-μl aliquots. The amount of radioactivity added to soil microcosms was equivalent to 1948.5 Bq with a total concentration of hexadecane of 462 μmol kg−1 (104 parts per million (ppm)). The KOH was replaced periodically and its radioactivity content was determined by liquid scintillation spectrometry (Tri-carb 2100TR, Packard Instruments, Meriden, CT, USA) [24]. Rates of mineralization were computed as the cumulative fraction of [14C]CO2 produced from the initial [14C]hexadecane added. We also prepared a series of KOH trap-microcosms with C2H2 (2 kPa) to evaluate the effect of this alkyne on hexadecane mineralization.

2.4 Denitrification and respiration rates

Denitrification was measured by determination of N2O production in the presence of C2H2 (2 kPa) in soil microcosms without KOH traps [15, 18]. Acetylene is an inhibitor of the N2O reductase of denitrifying bacteria that prevents N2O reduction to N2[25, 26]. Respiration was measured by accumulation of CO2 in the headspace of flasks. Gas samples (0.5 ml) were withdrawn from the headspace of the soil microcosms and analyzed by gas chromatography (GC). The concentration of each gas in the headspace was calculated based on a standard curve and using the gas law to calculate the amount (mol) of each gas. Since the soils were not slurried, we considered that the dissolved fraction of the various gases would be marginal compared with the headspace.

2.5 Analytical methods

Gases (N2O, CO2) were measured by GC either on a HP6890 GC (Hewlett–Packard, Palo Alto, CA, USA) with a thermal conductivity detector (TCD) or on a SRI 8610C gas chromatograph (SRI, Torrance, CA, USA) with a TCD, a flame ionization detector (FID) and an electron capture detector (ECD) in parallel. The HP6890 was configured in the following manner: the TCD was set at 225°C and the oven at 60°C; a 2 m×3.1 mm stainless steel column packed with HaysepQ (Supelco, Bellefonte, PA, USA); helium (ultra high purity grade, Prodair, Montreal, Que., Canada) was the carrier gas flowing at 40 ml min−1. The detection limit of the TCD for N2O was 800 ppmv and the response was linear up to 16 kPa.

The SRI 8610C GC had the following configuration: the TCD was set at 100°C, the FID at 150°C and the ECD at 250°C; the oven was set at 60°C, and each detector was connected to a separate column (2 m×3.1 mm stainless steel packed with Porapak Q (Supelco)). Helium (ultra high purity grade, Prodair) was used as carrier gas with flow rates of 23 ml min−1 for the TCD, 20 ml min−1 for the FID and 30 ml min−1 for the ECD. The detection limits were: N2O, 8 ppmv (ECD) or 200 ppmv (TCD); CH4, 20 ppmv (FID) or 800 ppmv (TCD); and CO2, 300 ppmv (TCD). The TCD response was linear for N2O, CH4, CO2 (up to 20 kPa), the FID response was linear for CH4 up to 1.5 kPa, and the ECD response was linear for N2O up to 300 ppmv.

For gas determinations, 0.5 ml of the gas samples was injected into the GC system with simultaneous integration of peaks using PeakSimpleII software (SRI). Gas standards were injected at the beginning and at the end of each day of analysis. A gas standard which contained 8033 ppmv of each of the following gases: CH4, CO2, N2O and C2H2, was prepared at the beginning of each day of analysis. Calibrated gas standards (990 ppmv N2O in N2; 1010 ppmv CH4 and 978 ppmv C2H4 in N2) (Prodair) were also used.

2.6 Statistical analysis

Rates of hexadecane mineralization were computed by linear regression of the amount of hexadecane mineralized over time using Excel 5. The time period used for the regression was different for each soil. For soil A, data from day 3 to day 6 showing maximum mineralization were used while for soil B, data from day 4 to day 32 were used. The equality of these rates (slopes) was tested for each soil independently by using the method described in Sokal and Rohlf [27]. For both soils, the rates (slopes) between the various treatments were not all equal. Rates computed for each treatment were then compared using the T′ method described in [27] because the sampling times were all the same for a given soil.

3 Results

3.1 Hexadecane mineralization

Mineralization of hexadecane in soil A began immediately after the addition of substrate (Fig. 1.I). Mineralization approached a maximum of ca. 60% within 30 days of incubation. Addition of the various N fertilizers had no effect on the mineralization of hexadecane by the indigenous microbial population in this soil. Rates of hexadecane mineralization (days 3–6) varied from 6.7 to 9.4 mg C16H34 kg−1 day−1 (Table 2). Addition of C2H2 (2 kPa) did not inhibit the mineralization of hexadecane in this soil (Fig. 1.II and Table 2).

1

Mineralization of [14C]hexadecane (100 ppm) in soil A (left) and in soil B (right) in the absence (I, III) and in the presence (II, IV) of C2H2 (2 kPa). Each data point is the average of triplicate flasks±1 S.E.M.

View this table:
2

Maximum rates of hexadecane mineralization in two soils with similar levels of C10-C50 contamination and with various N sources

SoilWaterNH4ClNaNO3NH4NO3UreaSterile
A−C2H28.8 (0.4)a,b9.4 (1.0)a9.3 (0.3)a8.9 (0.4)a,b7.2 (0.5)a,b
+C2H28.1 (0.5)a,b7.9 (0.8)a,b8.1 (0.2)a,b8.4 (0.4)a,b6.7 (1.0)a,b0.00 (0.00)b
B−C2H20.17 (0.01)B,C0.015 (0.001)C0.91 (0.01)A0.13 (0.02)B,C0.02 (0.00)C
+C2H20.20 (0.02)B,C0.011 (0.001)C0.85 (0.03)A0.28 (0.07)B0.01 (0.00)C0.00 (0.00)C
Rates are expressed in mg C16H34 kg−1 day−1. Each value represents the regression coefficient with its S.E.M. in parentheses. Values with different letters indicate significant differences (P=0.05) between treatments for a given soil based on unplanned comparisons among a set of regression coefficients using the T′ method [27]. Lower case and upper case indicate independent testing of soil A and soil B.

Mineralization of hexadecane in soil B (Fig. 1.III) occurred after a brief lag and at a slower rate than in soil A (Fig. 1.I). In the absence of any N supplement, the rate of hexadecane mineralization (days 4–32) was 0.17±0.01 mg C16H34 kg−1 day−1 (Table 2). Addition of NaNO3 increased the rate of hexadecane mineralization while NH4Cl reduced it when compared to the soil with no N supplement (Table 2). Addition of urea to this soil also decreased the rate of hexadecane mineralization to the same extent as did NH4Cl. Supplementing the soil with NH4NO3 did not have a significant effect on the hexadecane mineralization rate (0.13±0.02 mg C16H34 kg−1 day−1) when compared to the unsupplemented soil. The addition of C2H2 (2 kPa) to the headspace of microcosms with soil B did not inhibit hexadecane mineralization (Fig. 1.IV). Regardless of the N supplement, rates of hexadecane mineralization were independent of the presence of C2H2 (Table 2).

3.2 Denitrification and loss of N fertilizer

In the presence of C2H2, the accumulation of N2O was detected after a lag of 10 days in microcosms with soil A when supplemented with NaNO3 or NH4NO3, and to a much lesser extent when NH4Cl or urea were added (Fig. 2.I). Rates of denitrification were higher with NH4NO3 (1.7±0.2 μmol N2O flask−1 day−1) as the N supplement. Considering the quantity of NO3 added initially, it was possible to estimate the amount of N supplement that was lost through denitrification in this soil. It was found that when NH4NO3 was used, loss of N reached 35% of the initial NO3 concentration after 30 days of incubation, while it reached only 10% within the same incubation period when NaNO3 was added as supplement.

2

Denitrification in soil A (left) and in soil B (right) supplemented with hexadecane (100 ppm) in the absence (I, III) and in the presence (II, IV) of C2H2 (2 kPa). Each data point is the average of triplicate flasks±1 S.E.M.

Production of N2O in the presence of C2H2 was observed in soil B, only when supplemented with NaNO3 or NH4NO3 but not when this soil was supplemented with other N sources (Fig. 2.III), and the initial time lag before the initiation of denitrification was shorter (4 days). In soil B, denitrification occurred concurrently with hexadecane mineralization, as indicated by their highly significant (P≤0.01) correlation (r2=0.868) [27, 28], whereas in soil A, hexadecane mineralization was completed before denitrification had started. Between 24% (NaNO3) and 34% (NH4NO3) of the N supplement was lost through denitrification after 32 days incubation.

In the absence of C2H2, N2O accumulated in microcosms of soil A (Fig. 2.II). N2O was emitted from the 16th day of incubation and was especially significant when NaNO3 (0.34±0.09 μmol N2O flask−1 day−1) or NH4NO3 (0.47±0.06 μmol N2O flask−1 day−1) were used as N supplements. The molar ratio of N2O/N2 in this soil was 0.3. The N2O production rates represent a loss of 9% (NH4NO3) or 3% (NaNO3) of the initial NO3 supplement after 30 days of incubation.

Accumulation of N2O as a free intermediate in the microcosm headspace was even more significant in microcosms from soil B (Fig. 2.IV). N2O was emitted from day 4 and thereafter increased continuously in the headspace. N2O production rates were similar following the addition of NaNO3 or NH4NO3 (1.1±0.0 μmol flask−1 day−1). No N2O production was observed when other N supplements were added to this soil. The N2O/N2 molar ratios were 1.0 or 0.7 when NH4NO3 or NaNO3 were used as N supplements, respectively. Of the initial NO3 applied, the amount lost as N2O after 32 days of incubation was 33% (NH4NO3) or 15% (NaNO3).

4 Discussion

The C/N ratio is often used as a criterion to decide if N fertilizer would be required for effective bioremediation of a contaminated soil. This ratio is a useful indicator of the N limitation of a system, yet it is not sufficient to predict the outcome of the bioremediation of a site following application of N fertilizers. For instance, in this study, we investigated two soils with similar petroleum hydrocarbon contamination levels. A lower C/N ratio in soil A (32) than in soil B (81) indicated that the former soil was less N-limited than the latter. Accordingly, mineralization of hexadecane in soil A was 51 times faster than in soil B without nitrate addition. Addition of N fertilizer did not stimulate or inhibit the mineralization of hexadecane in soil A. These results suggest that hexadecane mineralization in soil A was not N-limited, but it was P-limited if the C/P ratio (698) is considered in this soil.

In soil B, addition of NH4NO3, NH4Cl or urea as N sources did not stimulate hexadecane mineralization, unlike other reports generally showing stimulation of petroleum biodegradation following addition of NH4NO3 or other ammonium salts [1]. Even more surprising, addition of NH4Cl or urea inhibited hexadecane mineralization. A number of reasons may explain this observation: a pH shift [29] following nitrification, nitrite accumulation from nitrification [30] or inhibition of petroleum-degrading bacteria by NH4+ or Cl ions [31, 32]. The exact mechanism is not known. In this soil, the C/N ratio poorly predicted the effect of N supplementation on hexadecane degradation. These results suggest that the addition of N fertilizers, although improving the C/N ratio, may not necessarily improve the biodegradation of pollutants in contaminated soils.

In contrast to the other fertilizers, the addition of NaNO3 to soil B stimulated hexadecane mineralization. Although it was almost ten times lower than in soil A. Stimulation of hexadecane mineralization by NaNO3 in soil B may be explained by: (1) the use of NO3 as a N source by bacteria involved in hexadecane mineralization, or (2) as an electron acceptor, in the absence of O2, by denitrifying bacteria involved in the degradation of hexadecane. In order to evaluate the second possibility, we measured denitrification by the C2H2 blockage technique in soils A and B. Because C2H2 may affect monooxygenases [5] and possibly those involved in the degradation of alkanes [33] or more simply the growth of certain bacteria [34], it was important to assess the effect of acetylene on hexadecane mineralization in soils A and B. The fact that acetylene had no significant effect on the rate of hexadecane mineralization in soil A or B with no N supplement validated the use of the C2H2 blockage technique to measure denitrification [35, 15].

A strong correlation between denitrification and hexadecane mineralization in soil B supplemented with NaNO3 would suggest that denitrifying bacteria may have played a role in the mineralization of hexadecane in soil B. However, if denitrifiers were involved in hexadecane mineralization, it may be expected that rates of hexadecane mineralization in soil B supplemented with NaNO3 would decrease in the presence of C2H2 because under this condition, denitrifying bacteria are deprived of some energy generation by the blockage of nitrous oxide reductase. No significant decrease of hexadecane mineralization was observed in the presence of C2H2. Moreover, no significant increase of hexadecane mineralization was observed when NH4NO3 was applied to soil B although denitrification activity was similar to that observed with NaNO3. Based on these considerations, it is more likely that assimilation rather than denitrification of NO3 stimulated hexadecane mineralization in soil B when supplied with NaNO3.

Assessing denitrification in soils A and B was also of interest because denitrifiers may be possible competitors with petroleum-degrading bacteria that use NO3 as N source. Denitrifiers may also be responsible for the local loss of N fertilizers through the coupling of denitrification and nitrification. In both tested soils, significant denitrification occurred following supplementation with NaNO3 or NH4NO3. Denitrification also occurred in soil A when treated with NH4Cl or urea, but the denitrification rates were very low compared to those with NaNO3 and NH4NO3 and may be related to the presence of endogenous NO3. In soil A, denitrification started after hexadecane mineralization was almost complete, unlike in soil B where denitrification started at the same time as hexadecane mineralization. Rates of denitrification and loss of N fertilizer measured by the C2H2 blockage technique were similar for both soils studied with either NH4NO3 or NaNO3. These results suggest that denitrification may be responsible for a significant loss of added fertilizer.

In summary, the application of N fertilizers may not always lead to increased biodegradation of petroleum hydrocarbons. A specific N fertilizer (here NaNO3 with soil B) may be required to specifically stimulate the mineralization of aliphatic petroleum hydrocarbons. The use of other N fertilizers such as NH4+ or urea may lead to a suppression of the mineralization activity in some soils, the mechanism of which is unknown. The use of the C/N ratio to determine the type and quantity of fertilizer to be added for soil bioremediation may not be sufficient. The loss of N fertilizer through denitrification may be significant (10–30% of initial NO3 input) and much of this loss may be via N2O emissions (30–100%).

Acknowledgements

We thank Suzanne Labelle, Anca Mihoc, Chantale Beaulieu and Stéphane Deschamps for their technical assistance. We also thank Prof. Roger Knowles for providing a critical review of the manuscript.

References

  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
View Abstract