This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Hormonal contamination from dairy farms is of great concern for the protection of the environment and human health. The constructed wetland system (CWS) is well-known natrual wastewater treatment system are and their efficiency are limited in removal steroid hormone. An efficient polishing system is required to assist in the complete treatment of dairy farm herd effluents prior to release into the environment. This paper compares the efficient removal of estrogens and androgens from HPLC water and CWS treated dairy farm herd effluents by various reactive materials including Connelly zero-valent iron (ZVI), Gotthart Maier ZVI, Tübingen ZVI, granular activated carbon (GAC) and organo clay. The study was carried out using a lab-based batch-test combined with hormone measurement using highly sensitive estrogen and androgen reporter gene assays (RGAs).
This preliminary study showed that each 1g of man-made materials efficiently removed over 99 % of -17Î²-ï€estradiol (17Î²-E2) or testosterone (T) over a two weeks period from 50 mL of HPLC grade water spiked at a concentration of 1000 ng L-1 17Î²-E2. In comparison, 1g of man-made materials efficiently removed over 97% of estrogenic load and 93% of androgenic load from 50 ml of dairy farm herd effluent water at a concentration of 51.5 ng L-1 17Î²-E2 equivalents (EEQs) and 123.96 ng L-1 T equivalents (TEQs). Connelly ZVI and GAC have stronger efficiencies on removal T than 17Î²-E2. In comparison, Gotthart Maier ZVI and Tübingen ZVI are more suitable for removal 17Î²-E2. Organo clay has very similar performance on 17Î²-E2 and T. Connelly ZVI and GAC have the most rapid reduction process for estrogens and androgens respectively. Organo clay was able to achieve a final level of EEQs which was below the lowest observable effect concentration (LOEC) of E2 (10 ng L-1). Alkaline condition has a negative effect on hormonal removal efficiency of Connelly ZVI Gotthart Maier ZVI and GAC, whereas the organo clay presents higher treatability at neural and acidic condition. Taken together, these results have demonstrated an excellent performance for organo clay, GAC and Connelly ZVI in removing estrogens and androgens. Therefore, these man-made materials may be useful adsorbents for the advanced treatment of residual natural hormones in treated dairy farm herd effluents.
Kai, what about the differences between estrogen and androgen treatment behavior for the materials? Some removed androgens quicker than estrogens I thought? I thought this was an important point you were going to discuss as highlighted by Debbie?
Keywords: dairy farm herd effluents, estrogens, androgens, reporter gene assay, man-made materials
Over the past twenty years, considerable scientific concerns and public debate have been expressed over the potential health risk posed by endocrine disrupting compounds (EDCs) which can alter the normal function of the endocrine system in wildlife and humans. Potential adverse effects of EDCs have concentrated mainly on reproductive and sexual development, immune function, the nervous system, thyroid function and hormone-related cancers. In addition, the concern of EDCs has been heightened by a number of human and experimental animal studies (Damstra et al., 2002). Natural estrogens have been of high concern as low part per trillion concentrations (10~100 ng L-1) of these compounds may cause adverse an effect on reproductive biology of aquatic species by interfering with the normal function of their endocrine system (Routledge et al., 1998). Androgens are another group of natural disrupting steroid hormones that have not been investigated fully. In vivo testing demonstrated that the sexual differentiation of fish can be influenced by both exogenous estrogens (Tabata et al., 2001) and androgens (Katiadaki et al., 2002).
Due to EDCs effects, in recent year, natural estrogenic and androgenic contaminants from the farm wastes have generated wide interest because of their endocrine disrupting effects. Dairy cows have been identified as the largest estrogens contributors in the environment discharging 90% of estrogens in the US (Lange et al., 2002), UK (Johnson et al., 2006) and New Zealand (Sarmah et al., 2006). The latest survey of 18 US dairy farms reported that an estrogenic response was detected from 1.3 to 670 ng L-1 of EEQ (17Î²-estradiol (17Î²-E2) equivalents) and the 17Î²-E2 (24 ng L-1) was the most potent estrogenic compounds found in dairy farm herd effluents. Androgens in dairy waste are limited reported and the lowest observable effect level (LOEL) in the environment is still unknown. The highest level of androgenic activity was found in manure from non-lactating, pregnant dairy cows (1737 ng g-1 dairy manure wet) (Lorenzen et al., 2004). In a dairy lagoon, the testosterone (T) was measured as high as 650 ng L-1. In literature review, 17Î²-E2 and T are the most frequently detected in dairy farm herd effluents and they have higher endocrine disrupting effects than other detected compounds (Kolodziej et al., 2004; Arnon et al., 2008; Sarmah et al., 2006; Zheng et al., 2008).
Drinking water is a potential source for human exposure to EDCs and is therefore also a target for external measurements (Schenck et al., 2003). The Environment Agency of England and Wales proposed the lowest observable effect concentration (LOEC) of E2 in surface water was 10 ng L-1 and predicted no effect concentration (pNEC) was 1 ng L-1(Young et al., 2002). Surface waters destined for drinking water can be contaminated by a variety of natural steroid hormones (e.g. estrogens, androgens) as shown by surveys carried out on water samples taken near dairy farms (Soto et al., 2002). Furthermore, similar concentrations of these hormones were found in fish within the same water system, providing further evidence of the potential for dairy farm herd effluents in contaminating water systems (Kolodziej et al., 2004). In addition, estrogens and androgens were also detected in ground water samples under a dairy waste water lagoon acting as another hormonal contamination pathway to water resource (Aronon et al., 2008).
Overall, the removal of estrogens and androgens from dairy farm herd effluents is a new challenge for the protection of water resources. Considering the potential impact of EDCs, it is highly important to remove them from dairy farm herd effluents before discharge. Current data shows that a vertical-flow with three ponds wetland system has a limited capacity in removing the natural estrogens. The highest removal efficiency was 67.8Â Â±Â 28.0%, 84.0Â Â±Â 15.4% for estrone (E1) and 17Î²-E2 respectively (Song et al., 2009). Therefore, advanced treatment processes after the CWS treatment step may be required to meet with the LOEC or pNEC standard.
Zero-valent iron (ZVI) has previously been used for groundwater remediation in permeable reactive barriers (Powell et al., 1998). There has been a great deal of interest in the degradation of aquatic contaminants using ZVI due to its applicability under different geochemical conditions, operational simplicity and low cost maintenance (Tyrovola et al., 2007). Although numerous studies have been carried-out on the treatment of organic and inorganic contaminants in aqueous solution by ZVI (Tyrovola et al., 2007; Shin et al., 2008; Mu et al., 2004), no research has been conducted on its removal efficiency of naturally occurring hormones.
Granular activated carbon (GAC) is a well-known absorption process for removing EDCs and pharmaceuticals (Kim et al., 2007; Snyder et al., 2007). Absorption experiments have demonstrated that GAC has a very high absorption capacity with a maximum absorption constant of 9290 mL g-1 for estrone and 12200 mL g-1 for 17Î²-E2 (Zhang et al., 2005). However, there is no data regarding GAC treatability in the removal androgens.
Clay is of interest as a solid catalytic matrix constituting of various mineral ions has the potential to act as catalytically active regents (Varma et al., 2002). Previous reports of effective absorption of lipophilic (Gianotti et al., 2008) and estrogenic mycotoxin zearalenone (Lemke et al., 1998) by organo clay led to the hypothesis that this technology could be useful in binding EDCs compounds such as estrogens and androgens. Our results provide sufficient evidence in support of this hypothesis. Absorption studies have shown that organo clays can effectively bind to zearalenone an endocrine disruptor with estrogenic activity (Ryu et al., 2003). Zearalenone is attracted into clay's interlayer at a neutral pH. At a neutral pH, zearalenone binds to the hydroxyl group on the edge of the clay via ion-dipole interaction and electrostatic attraction to the excess exchanged surfactant cations (Lemke et al., 1998).
In order to clarify the efficiencies of ZVI, GAC and organo clay in removal steroid hormones, the batch-test studies are the essential component of the design process to demonstrate the suitability of these reactive materials to treat androgens and estrogens. This test has an advantage of relatively quick, and also only small amounts of reactive materials and water are needed. A batch test is performed by placing a known weight of material and water containing targeting compounds in a number of bottles that are shaken over a known length of time. Samples are removed at regular intervals and the contaminant concentrations are determined by a screen assay.
So then, a sensitive and high-throughput combined screen assay for monitoring the change of hormonal activity in water samples is required. In particular, this method should allow the detection of total biological activity in low-level individual or mixture samples, which estimate the level of potential environmental risk. On our batch-test study, reporter gene assay (RGAs) was used as a screen assay for demonstrating the data of batch tests. RGAs incorporating relevant receptors and a reporter gene such as luciferase can detect compounds by measuring biological activity. This cell-based assay present perfectly suitable for testing large quantities samples and provide very sensitive response reflecting potential EDCs effect in mono-substance or complex mixture of different compounds (Willemsen et al., 2004). RGAs have previously been applied for a number of environmental samples, including the detection of estrogenic activity in sediment (Houtman et al., 2007) and the characterization of androgenic activity in wastewater and surface water (Blankvoort et al., 2005).
The aim of this project was to clarify the treatability of ZVI, GAC and organo clay in removal estrogens and androgens. Consequently, the applicability of these reactive materials was investigated in a filed study with preliminary treated dairy farm herd effluents.
2. Materials and methods
2.1. Chemicals and materials
The standards 17Î²-E2 and T were purchased from Sigma-Aldrich, UK, at highest purity (â‰¥98%). HPLC grade water was brought from Sigma-Aldrich. Stock solutions of standards were prepared at a concentration of 1000 ng L-1 in 100% methanol (Sigma-Aldrich) and were stored at -20oC to minimise evaporation and degradation. DMEM cell culture medium (with/without phenol red), penicillin, general fetal bovine serum, hormone-depleted serum and Trypsin Express were obtained from the GIBCOÂ®. A Luciferase Assay system (Cat. 1501) which was purchased from Promega Corporation. A list of the five man-made materials used can be seen in Table 1 including details of the manufacturer, mesh size and mass.
2.2. Batch test analysis
Time scale batch tests were carried out by mixing 1g of each material with 50 ml of spiked standard (1000 ng L-1) in 50ml HPLC grade water which was also used as the negative control. The experiments were shaken with continuous agitation on an orbital shaker for two weeks at room temperature. 50ml of spiked water sample without man-made materials was included as the positive control. 1ml of water sample was taken at regular intervals (e.g. 0min, 1min) and dried under nitrogen steam at 40oC. Hormonal analysis was performed using RGAs.
2.3. RGAs procedure
RGAs were used to measure the estrogenic and androgenic activity in water samples. An estrogenic responsive cell line (ER-Luc) was previously produced (Willemsen et al., 2002) by stably co transfecting the human MCF-7 breast cancer cell line with the MAR-Vit-Luc reporter plasmid. AR-Luc cell line (AR-Luc) was (Willemsen et al., 2002) obtained by stable integration of a pSV-AR0 expression vector coding for the human androgen receptor in to T47D human breast carcinoma cell line. Cells were seeded in a 96 welled plate (Greiner Bio-One) and incubated overnight at 37oC in assay media (DMEM, 10% hormone depleted serum). The next day assay media containing spikes or samples was added to the cells and again incubated 24h for the estrogen assay or 48h for the androgen assay at 37oC. The supernatant was then discarded and the cells washed with PBS prior to lysis and luciferase measurement on a luminometer (Berthold).
2.4. Sample collection, transport and storage
The field samples for application of man-made materials was collected from the first pond of a constructed wetland system (CWS) connected to a dairy unit facility in Northern Ireland. 1 L of dairy farm herd effluents was sampled by a polyethylene terephthalate (PETE) plastic bottle (Davidson & Hardy Ltd) (bisphenol-A free) and fresh effluents were used for batch-tests immediately upon arrival at the laboratory. How were samples collected, containers, storage prior to analysis????
2.5. Batch test analysis of dairy herd effluents
A solid phase extraction (SPE) pretreatment process was developed to minimise cytotoxic components and concentrate the steroid hormones in the batch test samples, before their application to the RGA systems. During the SPE process, 1ml of dairy herd effluents was diluted in 9ml of HPLC grade water and extracted using cartridges which contained a mixture of hydrophobic and hydrophilic packing (OASIS HLB, waters, Milford, MA). Each cartridge was sequentially conditioned with 3 mL of Tert-butylmethyl ether, 3 mL of methanol and 3 mL of HPLC grade water. The 10mL of diluted sample were passed through the HLB cartridges with the aid of vacuum to control the flow at ~1drop/s. After the samples flowed through the cartridges, the HLB cartridge was rinsed with 3 mL 5% methanol. Finally, hormones were eluted with 6 mL of 10% methanol in Tert-butylmethyl ether and evaporated under nitrogen at 40Â°C. The dried extracts were reconstituted in 200 Î¼L of methanol. 10 Î¼L of extract was diluted in 1mL of assay medium and the ER/AR level of water extracts were tested by RGAs as described in section 2.3.
2.6. Statistical analysis
The RGA data was fitted with the sigmoidal dose-response curve equation to a four-parameter Hill plot, Y= Min Response+ (Max response-Min response)/ (1+X/EC50) ^ slope of the liner, where X is the concentration of the standards, Y represent the response of hormonal activity and the range is 0~100% (Blankvoort et al., 2005; Willemsen et al., 2004). For the standards the calibration curve was obtained in the presence of an increasing dose of agonist. To compare the level of hormonal activity in the test samples the equivalent standard concentration of each sample was calculated using the light response of the samples and the EC50 of the calibration curve. The level of the samples hormonal activities were determined by their corresponding values on the standard curve.
To compare the level of hormonal activities of the test samples, the equivalent standard concentration of each sample was calculated using the calibration curve of either 17Î²-E2 or T. The estrogenic contents in the dairy herd effluentswas calculated as EEQs, defined as the amount of standard estrogen 17Î²-E2 inducing an equivalent response in the estrogenic assay as observed with the test samples. In the same way, the androgenic dose was calculated according to TEQs.
Estrogenic or androgenic treatment efficiency of man-made materials was calculated as follows:
% Estrogenic/Androgenic treatment efficiency = (initial EEQs/TEQs - final EEQs/TEQs) / initial EEQs/TEQs
3. Results and discussion
3.1. RGAs standard curve
The typical estrogenic and androgenic response curves for were shown as Fig. 1. Standard curve of ER-Luc cell line was set up with increasing concentration of 17-Î²-E2 and the EC50 was 0.002 ng mL-1 with a fold induction of 5, corresponding with the previous data 0.002 ng mL-1 (Vickie et al., 2004). Standard curve of AR-Luc cell line was set up with increasing concentration of T, EC50=0.16 ng mL-1with a fold induction of 47. This result was corresponded with published data with an EC50 0.23 ng mL-1 (Willemsen et al., 2004). These two assays provide high sensitivity for screening the hormonal activity in the batch-test samples.
3.2. Time-scale batch test
The results of the batch test indicating the decrease in rate of adsorption of 17Î²-E2 and T are shown in Fig. 2. ZVI is the first time applied to remove steroid hormone from water samples and three ZVI demonstrated ability to reduce the concentration of 17Î²-E2 and T. Reaction with Gotthart Maier ZVI produced the most rapid decreasing rate of 17Î²-E2 concentration, followed by Tübingen ZVI and Connelly ZVI. Most of 17Î²-E2 (~95%) was removed by Gotthart Maier ZVI within 45 minutes reaction and the final concentration of 17ß-E2 was 1.58 ng L-1 in the spiked HPLC grade water. Conversely, Connelly ZVI has better performance in removal T than other two ZVI and the final treated concentration of T was measured as 0.38 ng L-1 in the spiked water with Connelly ZVI. The mechanism of reaction may have been caused by the oxidation of steroid hormones with ZVI oxidant or hydrogen peroxide (H2O2). As shown in scheme 1, ZVI is quickly oxidized to ferrous (Fe2+) and ferric (Fe3+). Reaction of ZVI with oxygen can produce oxidant Fe3+ and H2O2 which lead to the formation of reactive oxygen species capable of oxidizing contaminants that cannot be reduced by ZVI (Kang et al., 2009; Keenan et al 2008). Moreover, recent studies have demonstrated that phenolic EDCs like estrogens can be removed by oxidizer such as Ferrate (Fe6+) (Lee et al., 2005) and UV/ H2O2 (Chen et al. 2007) in both drinking water and wastewater treatment systems. Our results indicated that GAC also has a strong capacity in the absorption of T and produced a more rapid decrease in the rate of T absorption than other materials in our experiment. Fig. 2b shows around 95% of T were removed by GAC during 20 minutes reaction and the final concentration of T was 2.34 ng L-1. The reaction of organo clay with 17Î²-E2 represented a more rapid decline rate than the other man-made materials and the final concentration of the hormones in the spiked water sample was 0.4 ng L-1 for 17Î²-E2 and 1.22 ng L-1 for T.
Taken together, GAC and organo clay showed a very rapid reaction time with the T in HPLC grade water and with over 95% of the spiked standard absorbed in 20 min. In comparison, Gotthart Maier ZVI and organo clay have also a short time of 45 min and 30min respectively for removing most of the dose of spiked 17Î²-E2 in HPLC grade water. In addition, as table 2 shown, although 17Î²-E2 and T belong to steroid hormonal groups which have similar physical properties the performance of the hormonal treatability of each material was specific for estrogens and androgens which are distinguished by the C-18 group located in the configuration of carbon ring at position C16 and C17. Figure 2 demonstrate that Connelly ZVI and GAC have more rapid treatability on removal T than 17Î²-E2. In contrast, Gotthart Maier ZVI and Tübingen ZVI showed better performance on removal 17Î²-E2. Only organo clay did not have significant difference between two reactions with 17Î²-E2 and T. Accordingly, our results suggest hormonal treatability should be considered by each individual hormonal compound.
3.3 Affect of pH on the performance of the reactive materials
In order to clarify the performance of man-made materials in acid, neutral and alkaline conditions, a further batch test was carried out to investigate the hormone removal efficiency of the five selected man-made materials at three different pH values. Each 1g material was reacted with 50ml of spiked 17Î²-E2 (1000 ng L-1) or T (1000 ng L-1) at pH 4, pH 7 and pH 12. The stability of 17Î²-E2 (Figure 3a) and T (Figure 4a) in HPLC grade water were not affected by the different pH conditions. At pH 4 and pH7, three ZVI reaction solutions had the same brown color. However, at pH 12, the reaction solution showed came out black, yellow and clear for Connelly ZVI, Gotthart Maier ZVI and Tübingen ZVI respectively. The corresponding graphs (Fig. 3 c and Fig. 4c) show the reduction of 17Î²-E2 or T by Gotthart Maier ZVI was delayed at higher pH condition and the rapid reduction of 17Î²-E2 Gotthart Maier ZVI was produced at neutral pH condition. In comparison, Connelly ZVI performed a more rapid reduction of spiked 17Î²-E2 at pH 4. Fig.3 d and Fig. 4d shows Tübingen ZVI efficiency in removal 17ß-E2 or T was not significantly affected by the pH conditions. The lasted study indicated that the corrosion of ZVI by oxygen is the key to oxidation of organic compounds (Kang et al., 2009). Moreover, another study showed the reactive oxidant production of ZVI was affected by different pH conditions. H2O2 and Fe (II) produced hydroxyl radical (OH-) by the Fenton reaction at acid condition. In contract, oxygen was in charged of the ZVI oxidation (Keenan et al., 2008). H2O2 is a stronger oxidizer than O2, which can give better results on the oxidation of organic compounds. It may be the reason of pH influence on Connelly ZVI and Gotthart Maier ZVI.
Fig. 3e and Fig. 4e shows organo clay started reacting with 17Î²-E2 or T more quickly at neutral and alkaline condition than at acid condition. It may caused by disruption of ion exchange at higher pH condition. A previous study has indicated that the absorption of organo clay is dependant on the exchange cation hydrophobicity. The efficiency of this reaction was decreased by the interference of ion exchange in the interlayer below the pH 7 (Lemke et al., 1998).
As shown in Fig. 3f and Fig. 4f, the adsorption for 17Î²-E2 or T did not change significantly when there was a pH increase from pH 4 to pH 7. However, the adsorption rate for 17Î²-E2 or T was decreased at pH 12. Similar results were reported by Zhang et al (2005) for adsorption of 17Î²-E2 onto GAC and it indicated that the pH value in the water phase can influence the surface charge of GAC on the adsorbent particles (Zhang et al., 2005). In addition, the sorption studies of phenol by activated carbon demonstrated that increasing hydrogen ion lead to the neutralization of negative charge on the surface of carbon, thus reducing donor-acceptor interaction between the aromatic ring of phenolic compounds and activated carbons (Beker et al., 2010). In all, these results suggest that pH is an important parameter affecting on the hormonal removal efficiency of ZVI, GAC and organo clay.
3.4. Application of reactive materials to wastewater treatment
Results from batch test experiments using dairy herd effluents are shown in Fig. 5. This test was carried out to identify the clean up efficiency in a real environmental condition. The selected dairy farm herd effluents were discharged from a dairy unit and contain large quantities of nutrients (phosphate (P) (670 mg L-1), calcium (Ca) (30 mg L-1), magnesium (Mg) (7.6 mg L-1), sodium (49mg L-1), potassium (K) (43 mg L-1) and biochemical oxygen demand (BOD) (105 mg L-1). Their pH value was 6.91. The initial level of EEQs and TEQs were measured in dairy herd effluents found to be 51.5 ng L-1 EEQs and 123.96 ng L-1 TEQs, respectively.
The estrogenic removal efficiency was 98.27 %, 97.02 %, 98.13 %, 98.63 % and 99.14 % for Connelly ZVI, Gotthart Maier ZVI, Tübingen ZVI, GAC and organo clay. After having reacted with 1g of each man-made material for a two week period, only the spiked sample treated by organo clay got final EEQs under the requirement of LOEC. Connelly ZVI represented the most rapid reaction with estrogens in waste water and reduced EEQs level below 200 ng L-1 within one hour. This finding showed the possibility of the application of ZVI-clay technology (Connelly-GPM, Inc) for maximum and rapid removal of estrogens from the dairy herd effluents. The androgenic removal efficiency was 98.45 %, 94.83 %, 93.29 %, 99.68 % and 99.84 % for Connelly ZVI, Gotthart Maier ZVI, Tübingen ZVI, GAC and organo clay. The dairy herd effluents treated by organo clay have lower TEQs than other treated water samples. In addition, the results showed that GAC reduced TEQs below 200 ng L-1 within 15 min, suggesting that GAC was the most suitable for androgenic treatment in dairy waste water.
By comparison to the reduction of estrogenic and androgenic activity in spiked HPLC grade water, most materials have slower decreasing rates of EEQs and TEQs in waste water, suggesting a negative influence of the wastewater matrix on their efficiencies. This is not unexpected due to the presence of various organic and inorganic sediments in effluents, which can also bind to reactive materials and it leads to reduction of the effective surface area of materials for adsorbing steroid hormones. Activated carbon efficacy was greatly reduced by the presence of natural organic matter which competes for binding sites and can block pores within activated carbon structure (Snyder et al., 2007). Moreover, the microorganisms in pond water may consume dissolved oxygen in waste water, which is essential for production of oxidants in ZVI action (Keenan et al., 2008) (Scheme 1). Dissolved oxygen is an important factor for EDCs removal efficiency and the aerobic condition was higher than in anaerobic condition (Furuichi et al., 2006; Ermawati et al., 2007). The organic compounds oxidation reaction was completely inhibited in the absence of dissolved oxygen (Kang et al., 2009). Thus, microbial activity in dairy farm herd effluents may lead to a smaller formation of reactive oxygen species capable of oxidizing estrogens and androgens. However, recent research reported that ZVI nano-particles can efficiently inactivate bacteria (Diao et al., 2009; Lee et al., 2009) which may provide more oxygen for ZVI oxidation. ZVI Connelly Iron showed a better performance after the first hour of reaction with estrogens in dairy farm herd effluents than in HPLC grade water. Gotthart Maier ZVI had almost the same rate of decrease of androgenic activity in dairy farm herd effluents and in HPLC water. The results were useful for the selection of suitable materials in environmental condition, which had minimum negative effects by the waste water matrix.
Hormonal contamination from dairy farms is of great concern with respect to environmental protection and human health. The removal of natural estrogens and androgens from dairy farm herd effluent is of importance due to their potential endocrine disrupting effects and the risk of contaminating the environment and food chain drinking water resources. ZVI, GAC and organo clay all exhibited the capacity to remove estrogens and androgens from synthetic water (HPLC grade water) and field samples (dairy farm herd effluents). In the five materials tested, Connelly ZVI and organo clay perform better with respect to maximum and fast removal of estrogens from dairy farm effluents. The ZVI-Clay technology has been used as an in-situ approach and a recent batch experiment has shown that the combination of bentonite clay and ZVI provided a more efficient reduction of trichloroethylene, chromium, and nitrate (Lee et al., 2006). In contrast, GAC is the most suitable man-made material for the removal of androgens in dairy farm herd effluents. Overall, this preliminary study demonstrated a small amount of man-made material may cause a rapid reduction of the hormonal activity in the short term. According to our finding, the removal of estrogens and androgens has an ancillary benefit when applying these man-made materials for reducing hormonal contaminants in dairy farm herd effluents. It is also important to consider how to dispose of used materials as well as the issue of regeneration associated with ZVI and GAC, which may lead indirectly to potential environmental damage. Thus careful consideration must be given to the actual cost and benefits of these processes which are used for the removal of trace hormones.
We would also like to thank the Department of Agriculture and Rural Development and the Henry Lester Trust for funding.