Published online 9 August 2006
Published in J Environ Qual 35:1779-1783 (2006)
DOI: 10.2134/jeq2005.0345
© 2006 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
TECHNICAL REPORTS
Organic Compounds in the Environment
Sorption–Desorption of Carbamazepine from Irrigated Soils
C. F. Williamsa,*,
C. F. Williamsb and
F. J. Adamsena
a USDA-ARS, U.S. Arid Land Agricultural Research Center, 21881 N. Cardon Ln., Maricopa, AZ 85239
b Department of Plant and Animal Science, Brigham Young University, Provo, UT 84602
* Corresponding author (cwilliams{at}uswcl.ars.ag.gov)
Received for publication September 12, 2005.
 |
ABSTRACT
|
|---|
Th anti-seizure medication carbamazepine is often found in treated sewage effluent and environmental samples. Carbamazepine has been shown to be very persistent in sewage treatment, as well as ground water. Due to environmental persistence, irrigation with sewage effluent could result in carbamazepine contamination of surface and ground water. To determine the potential for leaching of carbamazepine, a series of adsorption and desorption batch equilibrium experiments were conducted on irrigated soils. It was found that carbamazepine adsorption to biosolid-amended (T) soils had a KD of 19.8 vs. 12.6 for unamended soil. Based on adsorption, carbamazepine leaching potential would be categorized as low. During desorption significant hysteresis was observed and KD increased for both soils. Desorption isotherms also indicate a potential for irreversibly bound carbamazepine in the T soil. Results indicate that initial removal of carbamazepine via adsorption from irrigation water is significant and that desorption characteristics would further limit the mobility of carbamazepine through the soil profile indicating that carbamazepine found in sewage effluent used for irrigation has a low leaching potential.
Abbreviations: T, soil with biosolid application • UT, soil without biosolid amendment • AD, adsorption • DS, desorption • KD, distribution coefficient • EC, electrical conductivity • SAR, sodium adsorption ratio • SPE, solid phase extraction
|
INTRODUCTION
|
|---|
RECENTLY a number of pharmaceutically active compounds have been found at very low concentration in environmental samples. Kolpin et al. (2002) surveyed 139 streams across the USA for organic wastewater contaminants and found that 80% of the streams contained at least one target compound. They also detected 86% of the target compounds in at least one sample. The sampling was spread geographically through 30 states and covered every geographic region of the USA. Research in Germany determined that 31 different pharmaceutical compounds have been found in samples from 40 different rivers and streams (Ternes, 2001). It was also found that at least one compound was found in every sample.
In addition to surface waters, ground water samples have been found to contain quantifiable amounts of many pharmaceuticals. Ternes (2001) reported that 15% of 233 ground water samples contained at least one pharmaceutical or primary metabolite. Individual stream analysis by Kolpin et al. (2004) indicates that a major source of pharmaceuticals found in surface waters are a result of urban inflows to streams and rivers. They found that during low stream flow periods, when return flows from urban centers contribute significant flow to the overall stream, the concentration of pharmaceuticals increased downstream from the inputs. This would suggest that a major source of pharmaceuticals found in environmental water samples originate from municipal sources.
The chemical structure of carbamazepine (5H-dibenz(b,f)azepine-5-carboxamide) is shown in Fig. 1
. It is a commonly prescribed drug used to control seizures in the treatment of epilepsy and is sometimes used in the treatment of bipolar disorder (Thacke, 2005). A low estimate of generic carbamazepine production in the USA is approximately 35 000 kg annually and does not include name brand production (Thacke, 2005). It has been found that 2 to 3% of the ingested drug is excreted unchanged in the urine (Clara et al., 2004) resulting in an estimated 700 kg of the drug entering sewage systems annually in the USA. Carbamazepine has been found to be very resistant to degradation during sewage treatment. Möhle and Metzger (2001) found that typical treatment processes removed only 7% of the carbamazepine entering the treatment system. They also found that carbamazepine was present in effluent samples from nine different sewer treatment plants that were sampled throughout a year. As a result of carbamazepines widespread use and resistance to degradation in sewer treatment plants it is also commonly found in environmental samples at relatively high concentrations. Tixier et al. (2003) detected carbamazepine in surface waters in Switzerland at concentrations up to 0.95 µg L–1. Surface waters directly downstream from treated sewage effluent outfalls have also been shown to have carbamazepine concentrations as high as 0.26 µg L–1 (Kolpin et al., 2004).
Carbamazepine is also very stable in the environment. Clara et al. (2004) found that no appreciable reduction of carbamazepine occurred over 120 d during ground water recharge. They used a calibrated model to determine travel time from the point of infiltration and then took samples over time at intervals corresponding to the distance traveled by the infiltrated water. Small reductions in the concentration of carbamazepine in the ground water were linked to dilution effects. Carbamazepine has also been found to have an environmental dissipation half-life of 328 d and was classified as highly persistent (Löffler et al., 2005). This dissipation half-life was determined for a water sediment system likely to be the terminus for a sewage treatment facility outfall (i.e., river or stream) and can explain the prevalence of carbamazepine in the environment.
In the arid southwestern USA sewage effluent is often seen as a valuable water resource. Mutually beneficial arrangements where irrigated agriculture will transfer water to cities for use in exchange for monetary compensation and the use of treated effluent from the cities sewer treatment plant are becoming more common. As a result, understanding environmental fate of waste water contaminants found in sewage effluent in soil is becoming more important. In particular the fate of pharmaceutically active compounds in soil systems is becoming a topic of research. Previous investigation of pharmaceuticals has focused on river and stream systems and hydrologically connected ground water (Clara et al., 2004; Löffler et al., 2005; Kolpin et al., 2002, 2004). Conclusions drawn from research on these systems are not transferable to terrestrial systems where effluent is applied as irrigation water. Irrigated soils offer a biologically active zone where the oxygen status of the microenvironment can change from aerobic to anaerobic depending on moisture status (Luxmoore et al., 1970). This changing environment can potentially result in differences in the persistence for a compound such as carbamazepine. In addition carbamazepine sorption to soils should be affected by the type and amount of organic matter found in the soil. Fields adjacent to sewage treatment facilities are often used as a disposal site for the biosolids produced. Addition of biosolids can affect the sorption behavior of organics applied to the soil (Williams et al., 1999), and therefore, the mobility (Nelson et al., 1998). In general as organic matter increases, sorption of organics to the solid phase increases, reducing mobility (Nelson et al., 2000). Therefore, the addition of biosolids has the potential to decrease the mobility of carbamazepine in soils irrigated with sewage effluent.
The sorption of carbamazepine to sediments and aquifer material has been investigated (Löffler et al., 2005; Clara et al., 2004); however, sorption characteristics to terrestrial soils has not. The objectives of this study were to establish parameters that can help predict the soil leaching potential of carbamazepine, found in sewage effluent applied as irrigation water by: (i) determining carbamazepine sorption and desorption characteristics to an irrigated soil; (ii) develop sorption isotherms for a typical soil that has undergone effluent irrigation and biosolid application; and (iii) desorption isotherms to characterize the leaching potential of carbamazepine in irrigated soils.
|
MATERIALS AND METHODS
|
|---|
Carbamazepine (Fig. 1) is an established drug for the treatment of seizures caused by epilepsy. Other uses include treatment of bipolar disorder and pain management. Carbamazepine was purchased from Sigma-Aldrich Co. (St. Louis, MO).
The soil chosen was an Airport silt loam (fine-silty, mixed, active, mesic Aquic Natrixeroll) that had been under 10 yr of continuous sod production in Davis County, Utah. The production schedule requires applied biosolids to be plowed into the field. One year following application, turfgrass is planted, and then harvested the following year. The result is a 2-yr rotation of 1 yr fallow followed by a 1 yr turf crop. For this study the surface 18 cm of soil was collected from two adjacent sites before biosolid application. The treated soil (T) had received bi-annual amendments of 137 Mg (dry wt.) sewage sludge ha–1 for four consecutive rotations. The untreated soil (UT) had never received biosolid application. The T soil was collected 2 yr following the last sludge application. Soils were air-dried, sieved to 2.0 mm, and some of their physical properties measured by standard techniques (Table 1).
View this table:
[in this window]
[in a new window]
|
Table 1. Soil physical properties of Airport silt loam (Aquic Na-trixeroll) used to determine carbamazepine sorption charac-teristics.
|
|
Sorption isotherms were measured on T and UT soils. Adsorption (AD) was determined by batch equilibrium by preparing carbamazepine solutions of 37.5, 25.0, 12.5, and 5.0 µg L–1 in a water solution made by adding NaCl and CaCl2 to 18 M
water to create an electrical conductivity (EC) of 1 dS m–1 and a sodium adsorption ratio (SAR) of 2. Additionally, a control treatment with no carbamazepine was used. Unless otherwise stated, this water was used in all studies reported here. Carbamazepine sorption was performed by placing 20 mL of each carbamazepine solution in 50 mL Teflon centrifuge tubes containing 5 g soil. Each treatment was replicated three times. Centrifuge tubes were shaken for 2 h at 17°C, centrifuged for 15 min at 2000 x g, and 10 mL of supernatant removed for analysis. Preliminary studies showed that from 98 to 100% of adsorption occurred within 1.75 h. It was determined that 2 h would be used for equilibration to minimize potential loss due to degradation.
Desorption was determined by adding 10 mL of carbamazepine free water to the tubes containing the soil and remaining solution from the adsorption experiments. Centrifuge tubes were shaken on a reciprocating shaker for 2 h at 17°C, centrifuged for 15 min at 2000 x g, and 10 mL of supernatant removed for analysis. The removed supernatant was again replaced with carbamazepine free water and the processes repeated two more times for a total of three desorption events. A total of three different desorption isotherms (DS1, DS2, and DS3) were determined from three sequential desorption events.
Solid phase extraction (SPE) was used for sample concentration and clean up before analysis. Oasis HLB (Waters Co., Milford, MA) SPE cartridges were preconditioned in succession with 5 mL of MTBE, 5 mL of MeOH, and 5 mL of water followed by air-drying for 5 min. Cartridges were loaded with 10 mL of supernatant followed by 5 min drying. Carbamazepine was eluted using two successive 2-mL aliquots of MTBE. Samples were then evaporated to <0.25 mL and brought to 1 mL with acetonitrile.
Soils were analyzed for background carbamazepine and adsorbed carbamazepine following the final desorption step for mass balance. Pre-existing carbamazepine was measured by placing 20 mL of acetonitrile in 50 mL Teflon centrifuge tubes containing 5 g of soil. Centrifuge tubes were shaken for 2 h at 17°C, centrifuged for 15 min at 2000 x g, and 10 mL of supernatant removed. Removed supernatant was evaporated to <0.5 mL and then brought to 1 mL volume with acetonitrile. Adsorbed carbamazepine was determined following the final desorption step by centrifuging the soil and remaining solution for 15 min at 2000 x g. Supernatant was removed and the tube weighed to account for water remaining in the pore space and sufficient acetonitrile added to bring the total liquid volume to 20 mL. Centrifuge tubes were shaken for 2 h at 17°C, centrifuged for 15 min at 2000 x g, and 10 mL of supernatant removed. Removed supernatant was evaporated to <0.25 mL and then brought to 1 mL volume with acetonitrile.
Carbamazepine analysis was performed using a Micromass Quatro Micro LC-MS-MS. Separation was performed using a Waters 2.1 by 30 mm XTerra MS C18 column with a 2.5 µm stationary phase (Waters Co., Milford, MA). Operating conditions were 0.25 mL min–1, with a binary mobile phase of 0.1% formic acid in acetonitrile and 0.1% formic acid in water. Initial conditions were 10:90 acetonitrile/water, followed by isocratic flow for 1.5 min, at 1.5 min a linear gradient from 10:90 acetonitrile/water to 90:10 acetonitrile/water was applied over 3.5 min, followed by 1.5 min isocratic flow at 90:10 acetonitrile/water. These conditions resulted in carbamazepine having a retention time of 4.95 ± 0.05 min. Carbamazepine was quantified using electrospray (+) MS-MS of the transition 237.1 (m/z) 
194.06 (m/z).
Data analysis was performed using the linear form of the Freundlich equation (Eq. [1]):
 | [1] |
where CS is the amount of carbamazepine sorbed per mass of soil, CL is the solution phase concentration of carbamazepine, Kf is the adsorption coefficient, b is the intercept term, and N accounts for the degree of nonlinearity in the sorption isotherm. The linear form reduces N to 1 and the resulting slope of the equation is the adsorption coefficient and referred to as the distribution coefficient KD.
|
RESULTS AND DISCUSSION
|
|---|
Pre-existing carbamazepine in the T and UT soils were below detection (0.05 µg kg–1 soil). Both the T and UT soil had previously been irrigated with sewage effluent and the T soil had been treated with biosolids. Due to the ubiquitous nature and persistence of carbamazepine in effluent and the environments surrounding discharge, the lack of carbamazepine in both soils was unexpected. As the effluent used for irrigation had not been analyzed for carbamazepine one possible explanation may be that it never contained any. A second possible explanation is that the actively growing root zone near the surface under turfgrass is biologically active and undergoes wetting and drying in such a way that numerous different biological populations may have the opportunity to metabolize and transform carbamazepine more effectively than the organisms within the sewer treatment plant. Final analysis of the soil following the last desorption event also resulted in an average mass balance of between 97.2 and 104.4% of the applied carbamazepine. Thus, all of the applied carbamazepine can be accounted for.
Results of carbamazepine adsorption to the T and UT soils are shown in Fig. 2
. Adsorption was linear and a KD of 19.8 (T) and 12.6 (UT) are reported (Table 2). The T soil had a KD greater than the UT soil. Organic sorption in soils is generally thought to be governed by the organic matter content of the soil. The organic content of the T soil is greater than the UT soil, but if KD is divided by the fraction of organic C in the soil the resulting KOC for the T (KOC = 1250) soil is still much higher than the UT (KOC = 885) soil. Kim and Hage (2005) found that ionic strength had a significant effect on the sorption of carbamazepine to proteins in plasma. It was found that an order of magnitude increase in ionic strength led to a 50% increase in carbamazepine sorption to albumin. The effect was attributed to an increase in the nonpolar interactions between the organic sorbent and carbamazepine as the aqueous phase became more polar. The salinity in the T soil (1.48 dS m–1) is significantly higher than the UT soil (0.49 dS m–1) and could be the reason for some of the increased sorption. Additionally, others have shown that biosolid-derived soil organic matter can increase the adsorption of organics due to differences in the nature of the organic matter (Nelson et al., 1998; Williams et al., 1999).
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 2. Adsorption isotherm for carbamazepine to Airport silt loam with (T) and without biosolid (UT) application. Error bars represent ± one standard error of the mean.
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Distribution coefficients determined for adsorption, and desorption events of carbamazepine on Airport silt loam.
|
|
Generally a KD > 5 would indicate a compound with low leaching potential (Whitford et al., 2001). Based on this assessment the potential mobility of carbamazepine in soil is low. By contrast Ternes et al. (2002) found that carbamazepine exhibited no significant sorption to sand and gravel taken from aquifers and ground water recharge sites. Thus, in surface waters or ground water, influenced by surface water through highly permeable material, carbamazepine may be highly mobile. The T and UT soils have a significant clay and organic C content which would lead to increased sorption over sand and gravel. The potential mobility of carbamazepine in irrigated soils is expected to be much lower than in highly permeable subsurface unconsolidated material.
Desorption isotherms are shown for each treatment in Fig. 3
and KD is reported in Table 2. In all cases hysteresis was observed in the desorption of carbamazepine from both soils resulting in an increase in KD for sequential desorption events. A linear fit to the Freundlich equation was the best fit for desorption over the range of the experimental conditions, however the intercept from the linear regression was not zero indicating either, nonlinear desorption, or the presence of nonreversible desorption. Another indication of nonlinearity is that correlation coefficients were generally lower for desorption events than adsorption events. Nonlinearity of desorption isotherms can also be caused by additive experimental errors in supernatant removal and replacement. The overall ionic strength of the solution can also change as it approaches equilibrium with the water solution used for the batch treatments resulting in another possible cause for nonlinearity in the desorption isotherms. Even with these possible errors included, the increase in KD seen from adsorption to desorption indicates a difference between adsorption and desorption.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 3. Distribution coefficient (KD) for desorption of carbamazepine from Airport silt loam with (a) and without biosolid (b) application. Desorption was conducted from each soil treatment at each carbamazepine level (37.5, 25, 12.5, and 5.0 µgL–1) following adsorption by replacing half of the solution and subsequent equilibration. A total of three consecutive desorption events were conducted (DS1, DS2, and DS3). Error bars represent ± one standard error of the mean.
|
|
According to the KD for adsorption, carbamazepine has a low potential for mobility. During desorption the KD increased in all cases and would further limit the leaching potential of carbamazepine. These results are contrasting to the ubiquitous nature of carbamazepine in the environment. This would suggest that the mechanism for environmental contamination would not be expected to be from effluent percolate leaching from irrigated fields. In addition it would appear that the persistence of carbamazepine in turfgrass irrigated with sewage effluent is low as indicated by the lack of carbamazepine found in the soil before treatment. Conversely, a high KD for carbamazepine would indicate that the potential for contamination of surface waters if surface erosion is high. In this case the carbamazepine is adsorbed to soil solids that are entrained in runoff water and transported to surface streams where desorption can occur.
Carbamazepine exhibited significant hysteresis between the adsorption and desorption phases of sorption to irrigated soils. Soils amended with biosolids also increased the initial adsorption of carbamazepine. Results presented here provide evidence that carbamazepine present in treated sewage effluent has a low potential for leaching beyond the root zone in irrigated soils. Further investigations into the transport and degradation of carbamazepine in soils are needed to determine the environmental fate of carbamazepine in sewage effluent used for irrigation to ensure that ground water is protected from potential contamination.
|
NOTES
|
|---|
Trade names are included for the benefit of the reader and imply no endorsement or preferential treatment of the product listed by the USDA or Brigham Young University.
|
REFERENCES
|
|---|
- Clara, M., B. Strenn, and N. Kreuzinger. 2004. Carbamazepine as a possible anthropogenic marker in the aquatic environment: Investigations on the behaviour of carbamazepine in wastewater treatment and during groundwater infiltration. Water Res. 38:947–954.[Medline]
- Kim, H.S., and D.S. Hage. 2005. Chromatographic analysis of carbamazepine binding to human serum albumin. J. Chromatogr. B. 816:57–66.
- Kolpin, D.W., E.T. Furlong, M.T. Meyer, S.D. Zaugg, L.B. Barber, and H.T. Buxton. 2002. Pharmaceuticals, hormone, and other organic wastewater contaminants in U.S. streams, 1999–2000: A national reconnaissance. Environ. Sci. Technol. 36:1202–1211.[Medline]
- Kolpin, D.W., M. Skopec, M.T. Meyer, E.T. Furlong, and S.D. Zaugg. 2004. Urban contribution of pharmaceuticals and other organic wastewater contaminants to streams during differing flow conditions. Sci. Total Environ. 328:119–130.[CrossRef][Medline]
- Löffler, D., J. Römbke, M. Meller, and T.A. Ternes. 2005. Environmental fate of pharmaceuticals in water/sediment systems. Environ. Sci. Technol. 39:5209–5218.[Medline]
- Luxmoore, R.J., L.H. Stolzy, and J. Letey. 1970. Oxygen diffusion in the soil plant system part 1 a model. Agron. J. 62:317–322.[Abstract/Free Full Text]
- Möhle, E., and J.W. Metzger. 2001. Drugs in municipal sewage effluents: Screening and biodegradation studies. p. 192–205. In C.G. Daughton and T.L. Jones-Lepp (ed.) Pharmaceuticals and personal care products in the environment: Scientific and regulatory issues. Am. Chem. Soc., Washington, DC.
- Nelson, D.W., and L.E. Sommers. 1982. Total carbon, organic carbon and organic matter. p. 539–581. In A.L. Page (ed.) Methods of soil analysis. Part 2. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI.
- Nelson, S.D., J. Letey, W.J. Farmer, and C.F. Williams. 2000. Stability and mobility of napropamide complexed with dissolved organic matter in soil columns. Soil Sci. Soc. Am. J. 29:1856–1862.
- Nelson, S.D., J. Letey, W.J. Farmer, C.F. Williams, and M. Ben-Hur. 1998. Facilitated transport of napropamide by dissolved organic matter in sewage sludge amended soil. J. Environ. Qual. 27:1194–1200.[Abstract/Free Full Text]
- Ternes, T. 2001. Pharmaceuticals and metabolites as contaminants of the aquatic environment. p. 39–54. In C.G. Daughton and T.L. Jones-Lepp (ed.) Pharmaceuticals and personal care products in the environment: Scientific and regulatory issues. Am. Chem. Soc., Washington, DC.
- Ternes, T.A., M. Meisenheimer, D. Mcdowell, F. Sacher, H.-J. Brauch, B. Haist-Gulde, G. Preuss, U. Wilme, and N. Zulei-Seibert. 2002. Removal of pharmaceuticals during drinking water treatment. Environ. Sci. Technol. 36:3855–3863.[Medline]
- Thacke, P.D. 2005. Pharmaceutical data eludes environmental researchers. Environ. Sci. Technol. News. Available at http://pubs.acs.org/subscribe/journals/esthag-w/2005/mar/science/pt_pharmdata.html (posted 16 Mar. 2005; accessed 23 Aug. 2005; verified 9 June 2006).
- Tixier, C., H.P. Singer, S. Oellers, and S.R. Müller. 2003. Occurrence and fate of carbamazepine, clofibric acid, diclofenac, ibuprofen, ketoprofen, and naproxen in surface waters. Environ. Sci. Technol. 37:1061–1068.[Medline]
- Whitford, F., J. Wolt, H. Nelson, M. Barrett, S. Brichford, and R. Turco. 2001. Pesticides and water quality: Principles, policies, and programs. Purdue Pesticide Programs 35. Purdue Coop. Ext., Purdue Univ., West Lafayette, IN.
- Williams, C.F., W.J. Farmer, J. Letey, and S.D. Nelson. 1999. Molecular weight of DOM-napropamide complex as affected by napropamide–soil application methods. J. Environ. Qual. 28:1429–1435.[ISI]
TOP
ABSTRACT
INTRODUCTION
