- Email: [email protected]
Attachment surface energy effects on nitrification and estrogen removal rates by biofilms for improved wastewater treatment
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
Attachment surface energy effects on nitrification and estrogen removal rates by biofilms for improved wastewater treatment Mohiuddin Md. Taimur Khan a,b, Timothy Chapman c, Kristin Cochran c, Andrew J. Schuler a,* a
Department of Civil Engineering, University of New Mexico, Albuquerque, NM 87131-0001, USA Center for Molecular Discovery, University of New Mexico, Albuquerque, NM 87131-0001, USA c Scientific Laboratory Division, New Mexico Department of Health, Albuquerque, NM 87106, USA b
article info
abstract
Article history:
Submerged biofilm systems, such as integrated fixed-film activated sludge (IFAS) and
Received 13 September 2012
moving bed bioreactors (MBBRs), are increasingly being used for domestic wastewater
Received in revised form
treatment, often to improve nitrification. Little is known about whether and how biofilm
14 January 2013
attachment surface chemical properties affect treatment performance, although surface
Accepted 20 January 2013
chemistry is known to affect attachment in other systems, and work with pure strains has
Available online 8 February 2013
suggested that attachment of nitrifying bacteria may be enhanced on high surface energy surfaces. The objective of this research was to systematically evaluate the effects of surface
Keywords:
chemistry on biofilm quantity and rates of nitrification and estrogen removal. Biofilms
Biofilms
were grown on four plastic attachment surfaces with a range of hydrophobicity and sur-
Endocrine disrupting compounds
face energy values (nylon, melamine, high-density-polyethylene [HDPE], and acetal poly-
Hormones
meric plastic) by immersing them in a full scale nitrifying activated sludge wastewater
Nitrification
treatment system, followed by batch test experiments. The attachment surface water
Surface energy
contact angles ranged from 53 to 98 and surface energies ranged from 48.9 to 20.9 mJ/m2.
Wastewater
Attachment surface hydrophilicity and surface energy were positively correlated with total biomass attachment, with more than twice as much biomass on the highest surface energy, most hydrophilic surface (nylon) than on the lowest surface energy, least hydrophilic surface (acetal plastic). Absolute and specific nitrification rates were also correlated with hydrophilicity and surface energy (varying by factors of 5 and 2, respectively), as were absolute and specific removal first order rate constants of the hormones estrone (E1), bestradiol (E2) and 17a-ethynylestradiol (EE2). These results suggested that attachment surface chemistry may be a useful design parameter for improving biofilm performance for removal of ammonia and endocrine disrupting hormones from wastewater. Further research is required to verify these results at longer time scales and with typical media geometries. ª 2013 Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: þ1 (505) 277 4556; fax: þ1 (505) 277 1988. E-mail addresses: [email protected], [email protected] (A.J. Schuler). 0043-1354/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.watres.2013.01.036
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
1.
Introduction
Submerged-growth biofilm systems, such as integrated fixed film activated sludge (IFAS) and moving bed bioreactors (MBBRs) are increasingly being used for wastewater treatment. IFAS is an adaptation of the activated sludge process in which plastic media, typically free-floating and a few cm in diameter, are included in bioreactors to provide surfaces for biofilm growth. IFAS biofilms can increase capacity and improve nitrification (ammonia oxidation) (Randall and Sen, 1996). Higher nitrifier activities reported in IFAS biofilms than activated sludge suspended (planktonic) biomass (Kim et al., 2011; Onnis-Hayden et al., 2007) are thought to be due to the longer solids residence time associated with biofilms, as nitrifying bacteria are slow-growing autotrophs. MBBRs are similar to IFAS systems, but without planktonic biomass. Suspended plastic media in IFAS and MBBR systems are typically composed of readily-extrudable plastics, such as highdensity-polyethylene (HDPE). Microconstituent chemicals are wastewater contaminants that include industrial chemicals, pharmaceuticals, and hormones. Chemicals that disrupt the endocrine systems of wildlife are a particular concern. Natural hormones, such as estrone (E1) and b-estradiol (E2), as well as the synthetic contraceptive hormone 17a-ethinylestradiol (EE2), are largely responsible for endocrine disruption in domestic wastewater effluents (Johnson and Sumpter, 2001). Hormones are removed from wastewater primarily in wastewater treatment biological reactors, but this process is not complete (Johnson and Sumpter, 2001; Ternes et al., 1999). Hormone removal from wastewater has been linked to nitrifying bacteria (Andersen et al., 2003; Dytczak et al., 2008; Khunjar et al., 2011; Shi et al., 2004; Vader et al., 2000; Yi and Harper, 2007). Whether removals are directly due to nitrification is uncertain, in part because nitrifying systems typically have relatively high biomass concentrations due to their typically long solids residence times, which could also increase removal rates. There has been little published research describing media surface chemistry effects on biofilm development and IFAS or MBBR performance. Bacterial adhesion is a complex process that includes an initial physicochemical interaction phase, followed by molecular and cellular interactions, and these are affected by the characteristics of the bacterial cells, the attachment surface, and environmental factors, such as the presence of serum proteins (An and Friedman, 1998). Surface chemistry strongly affects bacterial attachment, with different microorganisms preferentially attaching to surfaces with different chemistries (Ista et al., 2004, 2010; Khan et al., 2011; Liu et al., 2008). Initial attachment has been shown to affect later biofilm structure, adhesion strength and detachment (Busscher et al., 1995; Lackner et al., 2009). The driving force for bacterial attachment has been suggested to be the interfacial surface tension between a cell and a surface, less the interfacial surface tensions between the cell and the liquid and the liquid and the surface (Ista et al., 2004). Attachment surface chemistry is known to affect bacterial attachment (reviewed in Van Loosdrecht et al., 1990), including that of pure heterotroph strains in biofilm reactors
2191
(Dimitrov et al., 2007). Reports for nitrifying bacteria have been inconsistent: more hydrophilic surfaces gave higher nitrification rates in a pure culture study with the ammoniaoxidizing bacteria (AOB) Nitrosomonas europaea and Nitrobacter winogradskyi (Kim et al., 1997), but the opposite trend was also reported (Sousa et al., 1997). Pure strains of the AOB N. europaea and Nitrosospira multiformis preferably attached to higher surface energy (a measure of the surface characteristics that can drive attachment) self-assembled monolayer surfaces (Khan et al., 2011). In a mixed culture nitrifying system, surfaces modified with amine groups produced more biofilm than unmodified or methyl group-modified surfaces (Lackner et al., 2009). It was hypothesized that increasing attachment surface energy can increase biofilm activity with respect to nitrification by increasing biofilm mass and/or the nitrifier biofilm fraction, and that it can improve hormone removal, either because of improvements in nitrification or because of an increase and/or change in heterotrophic biomass. The objective of this study was to test these hypotheses, with a focus on early-phase biofilm development (over 8 days). The experimental approach was to culture biofilms on plastic surfaces with a range of chemical properties by immersing them in a full scale biological wastewater treatment system, and then analyzing them in laboratory batch tests.
2.
Methods
2.1.
Biofilm development and batch tests
Four attachment surface plastics were evaluated: acetal polymer plastic (Ensinger Plastic, PA, USA), high density polyethylene (HDPE) (Sheffield Plastics, MA, USA), melamine (Regal Piedmont Plastics LLC, NM, USA), and nylon (type MC 907 natural, Regal Piedmont Plastics LLC, NM, USA). These were selected to provide a broad range of wettability (hydrophilicity, as determined by water contact angle values) and surface energies. Two sheets of each of the 4 materials (20 20 cm, 0.16 cm thick) were fastened to a PVC pipe rack in a parallel configuration with 2.5 cm spacing between each sheet. An additional sheet of plastic was attached to the top and bottom of the stack so that all sheets had similar hydraulic conditions. The rack was placed at a 2 m depth in the aerobic reactor of a full-scale activated sludge system of the City of Albuquerque Southside Water Reclamation Facility, with the sheets oriented horizontally, for 8 days. This 230,000 m3/d average daily flow system includes a nitrification/denitrification Modified LudzackeEttinger configuration. Typical parameter values in this reactor were: total suspended solids (TSS), 4.3 g/L; NH4-N, 16 mg/L; NO3-N, 14 mg N/L; NO2-N, 0.7 mg N/L; dissolved P, 2.7 mg/L; COD 20 mg/L, dissolved oxygen controlled at 2 mg/L, and temperature, 17e22 C. After 8 days the sheets were carefully removed and each was transferred to a separate batch reactor containing 2.2 L of synthetic wastewater based on the recipe in Schuler and Jenkins (2003), except there was no nitrification inhibitor and the initial ammonia nitrogen concentration was 132 mg N/L to
2192
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
insure it was not limiting during the batch tests. The 8 day time frame was selected to target early-phase biofilm development. Estrone (E1), b-estradiol (E2) and 17a-ethynylestradiol (EE2) (>99% purity, SigmaeAldrich, USA) were added to the synthetic media to a final concentration of 1000 ng/L each. Continuous aeration was provided for 6 h, after which time the plastic sheets were removed. A 5-ml sample was collected from each batch reactor at 2-h intervals for nitrogen compound analysis. Samples for E1, E2 and EE2 measurements were collected at 6 h. All rates and rate constants were calculated based on the final (6 h) concentrations. All samples were immediately filtered through 0.45 mm microfiber filters (Whatman, USA) and stored at 4 C or analyzed immediately.
2.2.
Solids and nitrogen compounds
Before batch testing, a 7.6 cm 7.6 cm area was scraped from one side of each plastic sheet using a sterilized razor blade to remove biomass. This was filtered through a pre-weighed glass fiber filter (Whatman, USA) and the total attached solids (TAS) were measured by Standard Method 2540 D (American Public Health Association, 2005). Experimental replicates were included by testing two sheets of each plastic type. NH3-N, NO3-N and NO2-N were measured in 0.45 mm filtered samples using Hach methods 10031, 10020 and 10019 (Hach Company, USA), respectively. Each sample was measured in duplicate or triplicate.
2.3.
Estrogens
Hormones were measured by New Mexico Department of Health, Scientific Laboratory Division personnel. Samples were filtered through 0.45 mm filters and 2 L of each were stored in amber glass bottles at 4 C without preservative. Liquideliquid extraction was performed at room temperature according to US EPA Method 3510C. Briefly, target analytes were spiked into 2 fortified blank quality control samples at 500 ng/L and 700 ng/L (recoveries were 88e101%, 78e82%, and 89e95%, for E1, E2, and EE2, respectively). Progesterone was spiked as a surrogate into all samples and quality control samples at 700 ng/L for quality control. A measured volume (approximately 2 L) of each sample was poured into a separatory funnel and 60 ml of dichloromethane were added after rinsing sample bottle. Separatory funnels were vigorously shaken for 1 min and they were allowed to separate for three to 5 min. The solvent layer was drained through a bed of anhydrous sodium sulfate into a 500 ml Erlenmeyer flask. Extraction with 60 mL dichloromethane was repeated, including shaking, separation, and passage through anhydrous sodium sulfate. The pH of the remaining solution was adjusted to 12 with 6N sodium hydroxide, and was extracted twice more with 60 mL dichloromethane. The pH of the aqueous solution was adjusted to w2 with concentrated sulfuric acid and it was extracted twice more with 60 mL dichloromethane. These extractions produced w420 mL of dichloromethane, with the extracted analytes, which was placed in a 35 C water bath and evaporated with ultra-high purity nitrogen gas until the final volume was w1 ml. Hormones were measured by gas chromatography-mass spectrometry (Varian 3800 with a type 1079 injector, an RXI-5 30 m
fused silica capillary column, and a Saturn 2000 ion trap mass spectrometer) with 50 ml injections. The detector scanned for ions in the range 40 m/z to 650 m/z. Compounds were identified by correlation to predetermined retention times using individual standards at known concentrations and to a reference spectra.
2.4.
Surface energy
The plastic contact angles were determined by goniometry for the three probe liquids ultrapure water, diiodomethane (99%, SigmaeAldrich, USA), and formamide (>99.5%, SigmaeAldrich, USA). The sessile drop technique was used (Rame´Hart Instrument Co., Goniometer Model# 400-22-300 with DROPimage Standard, NJ) as previously described (Khan et al., 2011). At least five 1.5 mL droplets were measured with each liquid. Surface energies were calculated from the measured contact angles as summarized in Liu et al. (2008). Briefly, the total surface energy (gtotal) is the sum of apolar (Lifshitz-van der Waals; LW) and polar (Lewis acid-base; AB) components: gtotal ¼ gLW þ gAB
(1) AB
The acid-base component (g ) is the geometric mean of the electron-donor (g) and electron-acceptor (gþ) parameters for the applied probe liquid or the substrata, given by: pffiffiffiffiffiffiffiffiffiffiffiffiffiffi gAB ¼ 2$ gþ $g
(2) LW
þ
The relationships between g , g , and g for a surface (x) and for a probe liquid (L) is provided by the YoungeDupre´ equation: qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffi ffi þ LW g þ 2$ gþ gL ðcos qL þ 1Þ ¼ 2$ gLW x $gL x $gL þ 2$ x $gL
(3)
where gL is for the liquidevapor interaction, qL is the probe liquidesolid contact angle. Reference values for the probe liquids (gL, gþ L and gL ) were from Good and van Oss (1992) and confirmed with the supplier. Applying Eq. (3) and contact angle measurements for the three probe liquids water (W), diiodomethane (D) and formamide (F), the three unknown þ surface energy components (gLW x ; gx and gx ) of each surface were calculated using Eq. (4). 131 0 ffi pffiffiffiffiffiffiffi pffiffiffiffiffiffiffi 192 3 82 0 pffiffiffiffiffiffiffiffi þ gLW g gw $½cosðqW Þþ1 = gLW < Wffi W pg W x p ffiffiffiffiffiffiffiffi ffiffiffiffiffiffi p ffiffiffiffiffiffi 6 B C 7 4 gþ 5 ¼ 42$@ gLW g $@ gD $½cosðqD Þþ1 A gþ x D ffi D D A5 pffiffiffiffiffiffiffiffi ; : pffiffiffiffiffiffi pffiffiffiffiffiffi gF $½cosðqF Þþ1 g g g gLW x F F F 2
(4)
3.
Results
3.1.
Plastic surface properties
The liquid contact angles (q values) for the probe liquids, the calculated surface energy parameters, and the total surface energies for the four plastic surfaces are shown in Table 1. Nylon was the most hydrophilic surface (it had the lowest water contact angle), and it had the highest total surface energy. The acetal plastic surface was the most hydrophobic surface, and it had the lowest total surface energy. HDPE,
2193
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
Table 1 e Contact angles (q) and surface energy (g) parameters of the four plastic surfaces. The subscripts W, D and F denote water, diiodomethane and formamide, respectively. Average values ± one standard deviation are shown, as calculated from at least three measurements. Plastic surfaces qW Nylon Melamine HDPE Acetal
Surface energy parameters (mJ/m2)
Contact angle (degrees)
53 84 90 98
qD 2 0 1 0
36 31 51 46
gLW
qF 0 1 1 0
36 38 76 94
2 0 1 1
which is commonly used for IFAS media, was the second most hydrophobic surface and it had the second lowest total surface energy.
3.2. Mass of attached biofilm and nitrogen transformations The quantity of attached biofilm was correlated with both surface energy (Fig. 1) and attachment surface hydrophilicity (indicated by cos[qW]; R2 ¼ 0.95), ranging from 0.21 0.03 mg/ cm2 (acetal) to 0.57 0.06 mg/cm2 (nylon). Ammonia removal fluxes (mass rate per attachment surface area), ranged from 0.55 to 2.53 g N/(m2*d), with the lowest on acetal plastic and the highest on nylon (the time series data for the batch tests is shown in Supplemental Fig. S1). Ammonia fluxes were correlated with both surface energy (Fig. 2) and cos(qW) values (R2 ¼ 0.78, Fig. S2 in Supplementary Data). The change in ammonia concentration in the experimental control (an HDPE sheet with no biofilm growth) was zero (Fig. S1). The positive correlation between nitrification rates and surface energy values was likely due in part to the correlation between attached biomass with surface energy (Fig. 1), as more biomass is expected to produce higher removal rates. The specific rates of nitrification (expressed as the mass of ammonia removed per unit of attached biomass per unit time) were therefore calculated to assess whether increasing surface energy increased the nitrifying activity per unit biofilm mass, and these were also correlated with the attachment surface energy (Fig. 2). This correlation suggested that as
Fig. 1 e The relationships between attachment surface energy of the four attachment surfaces and attached biomass.
41.4 43.8 27.1 13.7
0.1 0.2 0.5 0.2
gþ 0.6 2.4 0.4 1.3
g 0.2 0.1 0.2 0.1
22.7 0.1 6.4 9.9
gAB 2.2 0.03 1.0 0.4
7.5 1.9 3.3 7.2
1.0 0.1 0.8 0.8
gtotal 48.9 45.7 30.4 20.9
1.0 0.3 0.8 0.7
surface energy increased, not only did the total mass of biofilm increase, but the biofilm composition changed, with nitrifier activity per mass of biofilm increasing with the attachment surface energy. The correlation between cos[qW] values and specific ammonia consumption rates was much weaker (R2 ¼ 0.34; Supplemental Fig. S2), suggesting that surface energy was a better predictor of total and specific nitrification rates than was hydrophilicity. Nitrification occurs in a two step process: ammonia oxidation to nitrite, and nitrite oxidation to nitrate, which are performed by AOB and nitrite-oxidizing bacteria (NOB), respectively (Tchobanoglous et al., 2003). NOx (the sum of nitrate and nitrite) production is an indication of AOB activity, since NOB transform the nitrite produced by AOB to nitrate. NOx production rates were consistent with ammonia removal, as both the NOx fluxes and the specific rates of production were correlated with the attachment surface energies (Fig. 3), ranging from 0.23 to 0.98 g N/(m2*d) and 4.5e7.1 mg N/ (g TAS*h), respectively. There were no clear trends in nitrite accumulation rates relative to surface energy, with higher rates on HDPE and melamine than the other surfaces (Supplementary data Fig. S3). Because nitrite is produced by AOB and consumed by NOB, nitrite accumulation indicates higher AOB than NOB activity, and correlations with surface energy were not necessarily expected.
Fig. 2 e The relationship between surface energy values and the ammonia removal rates expressed as fluxes to the media surface and as specific rates (per total attached solids, or TAS).
2194
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
Fig. 4 e The relationship between surface energy and hormone removal first order rate constants (k). Fig. 3 e The relationship between surface energy values and the NOx (the sum of nitrate and nitrite) production rates expressed as fluxes to the media surface and as specific rates.
3.3.
of 0.5 for the specific EE2 removal rate constant indicated a weaker correlation with surface energy than for E1 and E2 (R2 values of 0.73 and 0.82, respectively).
Estrogen removal
Hormone removal rates in aerobic biological wastewater treatment processes have generally been considered to follow first order kinetics (Eq. (5)), and so the first order reaction rate constant (k, units 1/time) has been used to compare reaction kinetics between studies (reviewed in de Mes et al., 2005). Ct ¼ Co ekt
(5)
where Co is the initial hormone concentration and Ct is the concentration at time t. Because biomass concentrations also affect estrogen removal rates, k values are of limited use for comparing different studies, and specific reaction rate constants (per microbiological mass) have also been calculated (de Mes et al., 2005): 0
Ct ¼ Co eXk t
(6)
where X is the biomass concentration and k0 is the specific first order rate constant (1/[time*biomass concentration]). The k and k0 values were calculated by solving Equations (5) and (6), respectively. Ct was the hormone concentration at the end of 6 h batch test and X was the biofilm mass per volume of the batch reactor. The k values followed similar trends as the ammonia and NOx rates (Figs. 2 and 3), with positive correlations between attachment surface energy values and reaction rate constants for E1, E2, and EE2 (Fig. 4). The control sample (an HDPE sheet with no biofilm) yielded zero removal for all hormones, indicating losses due to volatilization and sorption to the batch reactor walls were negligible. Given the greater amounts of biofilm on the higher surface energy surfaces (Fig. 1), these results were not surprising. However, the positive correlations found for the specific reaction rate constant (k0 ) values (Fig. 5) suggested that not only was there more biofilm on the higher surface energy surfaces, but also that it was more effective per unit mass than the biofilms on the lower surface energy surfaces, with increasing removal rates per unit biomass with increasing surface energy, although the relatively low R2 value
4.
Discussion
4.1.
Biofilm attachment and nitrogen transformations
The range of attached biomass values (Fig. 1) was similar to the 0.24e0.60 mg/cm2 reported by Regmi et al. (2011) for biofilms grown on virgin polyethylene media (AnoxKaldnes K3) in a full scale system. The finding that the overall amount of attached biomass increased with increasing surface energy was consistent with previous reports of greater adhesion by both pure strains of heterotrophic bacteria and AOBs (Dimitrov et al., 2007; Khan et al., 2011), and that amine modification increased the strength of wastewater biofilm adhesion (Lackner et al., 2009). The range of ammonia nitrogen fluxes across the four surface types (0.55e2.53 g N/(m2*d; Fig. 2) was comparable to previous studies: 0.2e1.0 g N/(m2*d) in an MBBR with polyethylene carriers (Hem et al., 1994), 0.4e1.0 g N/(m2*d) in an MBBR (Zimmerman et al., 2003), and 0 to approximately 2 g N/
Fig. 5 e The relationship between surface energy and specific hormone removal first order rate constants (k0 ).
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
(m2*d) in an IFAS system with HDPE carriers (AnoxKaldnes type K1) (Lydmark et al., 2007). The range of specific nitrification rates (10.9 for HDPE to 21.2 mg NH3-N/(g TAS*h) for melamine; Fig. 2), was similar to but somewhat higher than some previous reports, with a range of 0.53e5.9 mg NH3-N/(g TAS*h) for IFAS media (Kim et al., 2011; Onnis-Hayden et al., 2007). Regmi et al. (2011) reported approximately 2e10 g NOx-N/ (g TAS*h) in a full scale IFAS system with HDPE media, and NH3-N removal rates were likely higher, as not all NH3-N is transformed to NOx. The observed rates of NOx production (Fig. 3) were also similar to the 0.28e1.12 g NO3-N/(m2*d), or 2.4e5.9 mg g NO3-N/(g TAS*h) by the attached phase of a full scale IFAS system with HDPE media reported by OnnisHayden et al. (2011), Regmi et al. (2011) reported 0.90 g NOxN/(m2*d) in a full scale system with HDPE media, as well as the specific rates noted above, and Kim et al. (2011) reported 0.058e1.35 mg N/(g TSS*h) nitrate production in an IFAS system with HDPE media. Factors such as relative heterotroph/ nitrifier content (due to varying COD/N ratios, for example), mixing rates, media geometries, and substrate concentrations likely explain differences between studies. Furthermore, this study focused on early-stage biofilm development, and utilized fixed-in-place flat sheets rather than floating media with interior surfaces; in light of these differences, the similarity in values between studies was notable. The correlations noted above support the hypothesis of preferential attachment of AOB to higher surface energy substrata, as demonstrated in our previous study using pure AOB strains and well-defined surfaces created using selfassembled monolayers (SAMs) where the adhesion rates of the AOBs N. europaea and N. multiformis were correlated with surface energy (Khan et al., 2011). Similarly, Kim et al. (1997) demonstrated that rates of nitrification were higher on a hydrophilic surface than more hydrophobic surfaces using pure strains of AOB (N. europaea and N. winogradskyi). It could be argued that enrichment of nitrifiers on the higher surface energy surfaces was due to these surfaces providing a stronger bond with the attached phase in general. This would provide a thicker biofilm with less sloughing, consistent with reports of stronger wastewater biofilm adhesion to amine-modified surfaces (Lackner et al., 2009), and the provision of a more stable, longer retention time environment could be favorable to slow-growing nitrifiers. However, the current study analyzed biofilms grown over only 8 days, which makes this explanation less likely, and instead suggests enrichment of AOB by a more specific increase in attachment rates and/or attachment strength on higher energy surfaces. It may be significant that the higher energy surfaces (melamine and nylon) both included amine groups, as in Lackner et al. (2009). Because nitrifiers take up ammonia as an electron donor for energy production, it may be that they have an affinity for surfaces containing amine groups. Indeed, the data presented here is consistent with a hypothesis that increasing surface energy led to greater overall biomass (Fig. 1) and ammonia removal (Fig. 2), while the presence of amine groups led to greater enrichment of nitrifiers (melamine and nylon, which had similar specific rates) relative to surfaces without amine groups (acetal and HDPE, which also had similar specific rates, and were much lower than the amine-containing surface specific rates). Further research is required to
2195
determine whether the increasing specific rates were due to changes in the surface energy or the presence of amine groups (or some other factor); this could be tested by comparing high surface energy surfaces with and without amine groups.
4.2.
Estrogen removal
The data in Figs. 4 and 5 suggest that higher surface energy attachment surfaces could be used to produce biofilms with improved estrogen removal abilities. A wide range of estrogen removal rates have been reported in previous studies, with most previous work performed on suspended growth systems, and their applicability to submerged biofilms systems is not known. An excellent review was provided by de Mes et al. (2005), with calculation of the specific rate constants (k0 ) by the authors, thereby accounting for variations in test duration and biomass concentrations between studies. Reaction rate k0 values for estrogen removal have ranged over several orders of magnitude across different activated sludge studies (for example, from 0.22 to 768 L/(g TAS*d) for E2), encompassing the ranges of values shown in Fig. 5. These varying results in previous studies were likely due to differences in sludge characteristics such as solids residence times, which have been suggested to improve removal of estrogens (Clara et al., 2005). The initial concentrations selected for use in the current study (1000 ng/L) were somewhat higher than typically reported in full scale wastewater influents (for example, Muller et al. (2008) reported maximum concentrations of 119, 28, and 38 ng/L of E1, E2, and EE2, respectively, in raw sewage and primary effluent, not including conjugated forms) because the rates of uptake were unknown in advance, and it was necessary to provide sufficient quantities that the hormone concentrations were not likely to fall below detection limits during the batch tests, while still yielding measurable changes in concentration. The initial concentrations were also orders of magnitude lower than those used in many previous studies (see review by de Mes et al. (2005)). If the assumption of first order kinetics is reasonably accurate the results reported here should be relevant for lower, more common concentrations as well. Previous work has generally reported that EE2 was more persistent than E1 and E2 in suspended growth wastewater treatment systems (de Mes et al., 2005). This was not the case for the data shown in Fig. 5, where EE2 rate constants were comparable to those of E1 and E2, although its rate constants were less than the average of the three estrogens on all surfaces except for acetal plastic. However, others have reported EE2 removal at higher rates, although k0 values were not calculated. For example, 80% of EE2 was removed in conventional activated sludge systems, which was similar to E1 removal (Gunnarsson et al., 2009). Similarly, Layton et al. (2000) reported that 80% of EE2 was removed by activated sludge biomass, and although only 20% was mineralized to CO2, 75% of its estrogenic activity was removed. Ternes et al. (1999) reported that 99.9% of E2, 83% of E1 and 78% of EE2 were removed in a Brazilian activated sludge plant, but a German plant removed 64% of E2 and negligible amounts of E1 and EE2. It therefore appears that EE2 may be resilient in some systems, but for unknown reasons other systems with different
2196
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
biomass qualities remove EE2 at higher rates, as was found in this study. EE2 also had the lowest R2 value (0.50) for the linear correlation between k0 and surface energy values of the 3 hormones, indicating the weakest correlation (Fig. 5), and suggesting that surface energy may have played less of a role in its removal compared to the other hormones. Little work has been performed on EE2 removal in submerged biofilm systems, but Cargouet et al. (2004) also reported similar removal percentages for E1 (55%), E2 (43%) and EE2 (45%) in a biofilm treatment system, and Gunnarsson et al. (2009) demonstrated that an MBBR used as a polishing step provided greater than 91% removal for E2 and EE2 (less than the limit of detection), and 100% for E1. The potential for improved EE2 removal by biofilm systems indicated by these studies is important to practice, as it has been reported to be 11e27 times more potent than E2 for endocrine disruption in fish (Thorpe et al., 2003). It has been suggested that some estrogens can be removed from wastewater by nitrifiers, possibly by cometabolic activity of the nitrification enzyme ammonia monooxygenase (Dytczak et al., 2008; Khunjar et al., 2011; Shi et al., 2004; Vader et al., 2000; Yi and Harper, 2007); by heterotrophic bacteria that may use the hormones as electron donors (Gaulke et al., 2008); and by adsorption to the biomass (Andersen et al., 2005). As noted, the correlations between the rate constant k values and surface energy (Fig. 4) were attributable at least in part to the correlation between attached biomass and surface energy (Fig. 1). However, the correlations between the specific rate constant (k0 ) and surface energy values (Fig. 5) suggested that estrogen removal activity increased per unit of biomass with increasing surface energy, suggesting one or more of the above mechanisms (cometabolism by nitrifiers, heterotroph biodegradation, or adsorption) increased as well on a per mass basis. The findings that both specific nitrification (Figs. 2 and 3) and specific hormone removal (Fig. 5) rates increased with increasing substrata surface energy suggested that the mechanism of estrogen removal by nitrifiers was at least plausible for this system. Ammonia and estrogen removal as fluxes were well correlated (R2 values > 0.84; Supplemental Fig. S4), due in part to the greater biomass on the higher surface energy surfaces (Fig. 1). Correlations between the specific rates of ammonia and estrogen removal were much weaker (R2 values 0.19, 0.31 and 0.58 for E1, E2, and EE2, respectively; Supplemental Fig. S5), casting doubt on the hypothesis that nitrifiers were primarily responsible for hormone removal in this system. It is also possible that the correlation between specific estrogen removal and substrata surface energy was attributable to differences in the adsorption properties of the biofilms grown on each surface, which would suggest the substrata surface energy affected the adsorptive properties of the attached biofilm. For example, higher surface energy biofilms could have formed on the higher surface energy plastics. It has been estimated that sorption plays a relatively small role in removal in suspended growth systems, at only 1.5e1.8% of the total estrogen loading in typical activated sludge domestic wastewater plants (Andersen et al., 2005), based on calculated values using sorption coefficients. However, Layton et al. (2000) reported that 80% of EE2 was removed within minutes
in a batch test but with little change over 24 h thereafter, which was consistent with removal by adsorption. Although only 20% was mineralized to CO2, this occurred continuously without further change in aqueous EE2, which may suggest adsorbed EE2 was biodegraded during the experiment. IFAS biofilms have been reported to be more hydrophobic, have less EPS and have EPS with different composition than suspended growth flocs (Mahendran et al., 2012); such differences may have been responsible for observations that biofilm systems provided better removal of EE2 than suspended growth systems (Cargouet et al., 2004; Gunnarsson et al., 2009). The effects of attachment surface chemistries on biofilm adsorption characteristics is not known, but such phenomena could have occurred by, for example, influencing the intracellular extracellular polymeric substance (EPS) matrix quantity and composition. The possibility that differences in adsorptive properties may have played a role in the observed differences in hormone removal across the surfaces requires further research and could be tested by including inhibited controls. Gaulke et al. (2008) suggested that abiotic transformation of EE2 by nitration may have been responsible for reported hormone removals in some previous studies; they showed this to occur in samples with high nitrite concentrations. Because nitrite concentrations were less than 1.5 mg/L in this study (Supplemental Fig. S3) this mechanism of hormone removal appears unlikely in this study. While these results demonstrated that higher surface energy plastics provided improved nitrification and hormone removal rates, both on absolute and per unit biomass bases, further research is required to better understand the precise mechanisms by which this occurred. This study focused on early-stage (8 day) biofilm development, and so future work should also determine the extent to which the observed phenomena persist in systems operated for longer periods. Finally, more realistic, free floating IFAS/MBBR media should be considered, which would better represent common practice in full scale systems.
5.
Conclusions
This study provided a systematic analysis of how important functional behaviors of mixed culture biofilms are affected by attachment surface chemistry in wastewater treatment systems. Biofilms grown on flat plastic sheets with a range of chemical characteristics in a full scale treatment system aerobic reactor demonstrated that attachment surface energy was positively correlated with biofilm mass, nitrification, and estrogen removal. Nitrification and hormone removal rates were correlated with surface energy both as fluxes to the biofilms and as a specific rates (per mass of biofilm), suggesting that not only was more biofilm produced on the higher surface energy surfaces, but that the specific activity of the biofilm increased as well, demonstrating an enrichment of nitrifiers on the higher surface energy surfaces. IFAS and MBBR media is commonly made of low surface energy plastics, such as HDPE. Future design may benefit from consideration of alternative materials, including high energy plastics such as nylon, or surface coatings. Improvements in treatment performance may justify increased costs in media
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
materials and manufacture. The generally good removal of EE2 by all biofilms in this study warrants further investigation. Future work should also consider biofilms grown over longer terms, and with media geometries more commonly used in practice.
Acknowledgments This work was supported by the Water Environment Research Foundation (Paul L. Busch Award) and by the National Science Foundation (Grant # 0852469).
Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.watres.2013.01.036.
references
American Public Health Association, American Water Works Association and Water Environment Federation, 2005. Standard Methods for the Examination of Water and Wastewater (Washington, D.C). An, Y.H., Friedman, R.J., 1998. Concise review of mechanisms of bacterial adhesion to biomaterial surfaces. Journal of Biomedical Materials Research 43 (3), 338e348. Andersen, H., Siegrist, H., Halling-Sorensen, B., Ternes, T.A., 2003. Fate of estrogens in a municipal sewage treatment plant. Environmental Science and Technology 37 (18), 4021e4026. Andersen, H.R., Hansen, M., Kjolholt, J., Stuer-Lauridsen, F., Ternes, T., Halling-Sorensen, B., 2005. Assessment of the importance of sorption for steroid estrogens removal during activated sludge treatment. Chemosphere 61 (1), 139e146. Busscher, H.J., Bos, R., Vandermei, H.C., 1995. Initial microbial adhesion is a determinant for the strength of biofilm adhesion. FEMS Microbiology Letters 128 (3), 229e234. Cargouet, M., Perdiz, D., Mouatassim-Souali, A., TamisierKarolak, S., Levi, Y., 2004. Assessment of river contamination by estrogenic compounds in Paris area (France). Science of the Total Environment 324 (1e3), 55e66. Clara, M., Kreuzinger, N., Strenn, B., Gans, O., Kroiss, H., 2005. The solids retention timeda suitable design parameter to evaluate the capacity of wastewater treatment plants to remove micropollutants. Water Research 39, 97e106. de Mes, T., Zeeman, G., Lettinga, G., 2005. Occurrence and fate of estrone, 17b-estradiol and 17a-ethynylestradiol in STPs for domestic wastewater. Reviews in Environmental Science and Technology 4, 275e311. Dimitrov, D., Hadjiev, D., Nikov, I., 2007. Optimisation of support medium for particle-based biofilm reactors. Biochemical Engineering Journal 37, 238e245. Dytczak, M.A., Londry, K.L., Oleszkiewicz, J.A., 2008. Biotransformation of estrogens in nitrifying activated sludge under aerobic and alternating anoxic/aerobic conditions. Water Environment Research 80 (1), 47e52. Gaulke, L.S., Strand, S.E., Kalhorn, T.F., Stensel, H.D., 2008. 17aethinylestradiol transformation via abiotic nitration in the presence of ammonia oxidizing bacteria. Environmental Science and Technology 42 (20), 7622e7627.
2197
Good, R.J., van Oss, C.J., 1992. In: Shrader, M.E., Loeb, G.I. (Eds.), Modern Approaches to Wettability, Theory and Applications. Plenum Press, New York. Gunnarsson, L., Adolfsson-Erici, M., Bjorlenius, B., Rutgersson, C., Forlin, L., Larsson, D.G.J., 2009. Comparison of six different sewage treatment processes-Reduction of estrogenic substances and effects on gene expression in exposed male fish. Science of the Total Environment 407 (19), 5235e5242. Hem, L.J., Rusten, B., Odegaard, H., 1994. Nitrification in a movingbed biofilm reactor. Water Research 28 (6), 1425e1433. Ista, L.K., Callow, M.E., Finlay, J.A., Coleman, S.E., Nolasco, A.C., Simons, R.H., Callow, J.A., Lopez, G.P., 2004. Effect of substratum surface chemistry and surface energy on attachment of marine bacteria and algal spores. Applied and Environmental Microbiology 70 (7), 4151e4157. Ista, L.K., Mendez, S., Lopez, G.P., 2010. Attachment and detachment of bacteria on surfaces with tunable and switchable wettability. Biofouling 26 (1), 111e118. Johnson, A.C., Sumpter, J.P., 2001. Removal of endocrinedisrupting chemicals in activated sludge treatment works. Environmental Science and Technology 35 (24), 4697e4703. Khan, M.M.T., Ista, L.K., Lopez, G.P., Schuler, A.J., 2011. Experimental and theoretical examination of surface energy and adhesion of nitrifying and heterotrophic bacteria using self-assembled monolayers. Environmental Science and Technology 45 (3), 1055e1060. Khunjar, W.O., Mackintosh, S.A., Skotnicka-Pitak, J., Baik, S., Aga, D.S., Love, N.G., 2011. Elucidating the relative roles of ammonia oxidizing and heterotrophic bacteria during the biotransformation of 17 alpha-ethinylestradiol and trimethoprim. Environmental Science and Technology 45 (8), 3605e3612. Kim, H.-S., Schuler, A.J., Gunsch, C.K., Pei, R.T., Gellner, J.W., Boltz, J.P., Freudenberg, R.G., Dodson, R., 2011. Attached and suspended phase nitrifying microbial community structures and functions in integrated fixed film activated sludge. Water Environment Research 83 (7), 627e635. Kim, Y.H., Cho, J.H., Lee, Y.W., Lee, W.K., 1997. Development of a carrier for adhesion of nitrifying bacteria using a thermodynamic approach. Biotechnology Techniques 11 (11), 773e776. Lackner, S., Holmberg, M., Terada, A., Kingshott, P., Smets, B.F., 2009. Enhancing the formation and shear resistance of nitrifying biofilms on membranes by surface modification. Water Research 43 (14), 3469e3478. Layton, A.C., Gregory, B.W., Seward, J.R., Schultz, T.W., Sayler, G.S., 2000. Mineralization of steroidal hormones by biosolids in wastewater treatment systems in Tennessee USA. Environmental Science and Technology 34 (18), 3925e3931. Liu, Y., Gallardo-Moreno, A.M., Pinzon-Arango, P.A., Reynolds, Y., Rodriguez, G., Camesano, T.A., 2008. Cranberry changes the physicochemical surface properties of E. coli and adhesion with uroepithelial cells. Colloids and Surfaces B-Biointerfaces 65 (1), 35e42. Lydmark, P., Almstrand, R., Samuelsson, K., Mattsson, A., Sorensson, F., Lindgren, P.-E., Hermansson, M., 2007. Effects of environmental conditions on the nitrifying population dynamics in a pilot wastewater treatment plant. Environmental Microbiology 9 (9), 2220e2233. Mahendran, B., Lishman, L., Liss, S.N., 2012. Structural, physicochemical and microbial properties of flocs and biofilms in integrated fixed-film activated sludge (IFFAS) systems. Water Research 46 (16), 5085e5101. Muller, M., Rabenoelina, F., Balaguer, P., Patureau, D., Lemenach, K., Budzinski, H., Barcelo, D., De Alda, M.L., Kuster, M., Delgenes, J.P., Hernandez-Raquet, G., 2008. Chemical and biological analysis of endocrine-disrupting hormones and estrogenic activity in an advanced sewage
2198
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
treatment plant. Environmental Toxicology and Chemistry 27 (8), 1649e1658. Onnis-Hayden, A., Dair, D., Johnson, C., Schramm, A., Gu, A.Z., 2007. Kinetics and nitrifying populations in nitrogen removal processes at a full-scale integrated fixed-film activated sludge (IFAS) plant. In: Proceedings of Proceedings of the Water Environment Federation Annual Conference (WEFTEC), Washington, D.C. Onnis-Hayden, A., Majed, N., Schramm, A., Gu, A.Z., 2011. Process optimization by decoupled control of key microbial populations: distribution of activity and abundance of polyphosphate-accumulating organisms and nitrifying populations in a full-scale IFAS-EBPR plant. Water Research 45 (13), 3845e3854. Randall, C.W., Sen, D., 1996. Full-scale investigation of an integrated fixed-film activated sludge (IFAS) process for enhanced nitrogen removal. Water Science and Technology 33 (12), 155e162. Regmi, P., Thomas, W., Schafran, G., Bott, C., Rutherford, B., Waltrip, D., 2011. Nitrogen removal assessment through nitrification rates and media biofilm accumulation in an IFAS process demonstration study. Water Research 45 (20), 6699e6708. Schuler, A.J., Jenkins, D., 2003. Enhanced biological phosphorus removal from wastewater by biomass with varying phosphorus contents, part I: experimental methods and results. Water Environment Research 75 (6), 485e498. Shi, J., Fujisawa, S., Nakai, S., Hosomi, M., 2004. Biodegradation of natural and synthetic estrogens by nitrifying activated sludge and ammonia-oxidizing bacterium Nitrosomonas europaea. Water Research 38 (9), 2323e2330.
Sousa, M., Azeredo, J., Feijo, J., Oliveira, R., 1997. Polymeric supports for the adhesion of a consortium of autotrophic nitrifying bacteria. Biotechnology Techniques 11 (10), 751e754. Tchobanoglous, G., Burton, F.L., Stensel, H.D., 2003. Wastewater Engineering: Treatment and Reuse. Metcalf & Eddy, Inc., McGraw-Hill, New York. Ternes, T.A., Stumpf, M., Mueller, J., Haberer, K., Wilkena, R.-D., Servosb, M., 1999. Behavior and occurrence of estrogens in municipal sewage treatment plants e I. Investigations in Germany, Canada and Brazil. Science of the Total Environment 225, 81e90. Thorpe, K.L., Cummings, R.I., Hutchinson, T.H., Scholze, M., Brighty, G., Sumpter, J.P., Tyler, C.R., 2003. Relative potencies and combination effects of steroidal estrogens in fish. Environmental Science and Technology 37 (6), 1142e1149. Vader, J.S., van Ginkel, C.G., Sperling, F., de Jong, J., de Boer, W., de Graaf, J.S., van der Most, M., Stokman, P.G.W., 2000. Degradation of ethinyl estradiol by nitrifying activated sludge. Chemosphere 41 (8), 1239e1243. Van Loosdrecht, M.C.M., Lyklema, J., Norde, W., Zehnder, A.J.B., 1990. Influence of interfaces on microbial activity. Microbiological Reviews 54 (1), 75e87. Yi, T., Harper, W.F., 2007. The link between nitrification and biotransformation of 17 alpha-ethinylestradiol. Environmental Science and Technology 41 (12), 4311e4316. Zimmerman, R.A., Richard, D., Bradshaw, A.T., Craddock, P.P., 2003. Pilot-scale evaluation of separate-stage nitrification using an attached-growth, moving-bed media process. Water Environment Research 75 (5), 422e433.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
Attachment surface energy effects on nitrification and estrogen removal rates by biofilms for improved wastewater treatment Mohiuddin Md. Taimur Khan a,b, Timothy Chapman c, Kristin Cochran c, Andrew J. Schuler a,* a
Department of Civil Engineering, University of New Mexico, Albuquerque, NM 87131-0001, USA Center for Molecular Discovery, University of New Mexico, Albuquerque, NM 87131-0001, USA c Scientific Laboratory Division, New Mexico Department of Health, Albuquerque, NM 87106, USA b
article info
abstract
Article history:
Submerged biofilm systems, such as integrated fixed-film activated sludge (IFAS) and
Received 13 September 2012
moving bed bioreactors (MBBRs), are increasingly being used for domestic wastewater
Received in revised form
treatment, often to improve nitrification. Little is known about whether and how biofilm
14 January 2013
attachment surface chemical properties affect treatment performance, although surface
Accepted 20 January 2013
chemistry is known to affect attachment in other systems, and work with pure strains has
Available online 8 February 2013
suggested that attachment of nitrifying bacteria may be enhanced on high surface energy surfaces. The objective of this research was to systematically evaluate the effects of surface
Keywords:
chemistry on biofilm quantity and rates of nitrification and estrogen removal. Biofilms
Biofilms
were grown on four plastic attachment surfaces with a range of hydrophobicity and sur-
Endocrine disrupting compounds
face energy values (nylon, melamine, high-density-polyethylene [HDPE], and acetal poly-
Hormones
meric plastic) by immersing them in a full scale nitrifying activated sludge wastewater
Nitrification
treatment system, followed by batch test experiments. The attachment surface water
Surface energy
contact angles ranged from 53 to 98 and surface energies ranged from 48.9 to 20.9 mJ/m2.
Wastewater
Attachment surface hydrophilicity and surface energy were positively correlated with total biomass attachment, with more than twice as much biomass on the highest surface energy, most hydrophilic surface (nylon) than on the lowest surface energy, least hydrophilic surface (acetal plastic). Absolute and specific nitrification rates were also correlated with hydrophilicity and surface energy (varying by factors of 5 and 2, respectively), as were absolute and specific removal first order rate constants of the hormones estrone (E1), bestradiol (E2) and 17a-ethynylestradiol (EE2). These results suggested that attachment surface chemistry may be a useful design parameter for improving biofilm performance for removal of ammonia and endocrine disrupting hormones from wastewater. Further research is required to verify these results at longer time scales and with typical media geometries. ª 2013 Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: þ1 (505) 277 4556; fax: þ1 (505) 277 1988. E-mail addresses: [email protected], [email protected] (A.J. Schuler). 0043-1354/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.watres.2013.01.036
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
1.
Introduction
Submerged-growth biofilm systems, such as integrated fixed film activated sludge (IFAS) and moving bed bioreactors (MBBRs) are increasingly being used for wastewater treatment. IFAS is an adaptation of the activated sludge process in which plastic media, typically free-floating and a few cm in diameter, are included in bioreactors to provide surfaces for biofilm growth. IFAS biofilms can increase capacity and improve nitrification (ammonia oxidation) (Randall and Sen, 1996). Higher nitrifier activities reported in IFAS biofilms than activated sludge suspended (planktonic) biomass (Kim et al., 2011; Onnis-Hayden et al., 2007) are thought to be due to the longer solids residence time associated with biofilms, as nitrifying bacteria are slow-growing autotrophs. MBBRs are similar to IFAS systems, but without planktonic biomass. Suspended plastic media in IFAS and MBBR systems are typically composed of readily-extrudable plastics, such as highdensity-polyethylene (HDPE). Microconstituent chemicals are wastewater contaminants that include industrial chemicals, pharmaceuticals, and hormones. Chemicals that disrupt the endocrine systems of wildlife are a particular concern. Natural hormones, such as estrone (E1) and b-estradiol (E2), as well as the synthetic contraceptive hormone 17a-ethinylestradiol (EE2), are largely responsible for endocrine disruption in domestic wastewater effluents (Johnson and Sumpter, 2001). Hormones are removed from wastewater primarily in wastewater treatment biological reactors, but this process is not complete (Johnson and Sumpter, 2001; Ternes et al., 1999). Hormone removal from wastewater has been linked to nitrifying bacteria (Andersen et al., 2003; Dytczak et al., 2008; Khunjar et al., 2011; Shi et al., 2004; Vader et al., 2000; Yi and Harper, 2007). Whether removals are directly due to nitrification is uncertain, in part because nitrifying systems typically have relatively high biomass concentrations due to their typically long solids residence times, which could also increase removal rates. There has been little published research describing media surface chemistry effects on biofilm development and IFAS or MBBR performance. Bacterial adhesion is a complex process that includes an initial physicochemical interaction phase, followed by molecular and cellular interactions, and these are affected by the characteristics of the bacterial cells, the attachment surface, and environmental factors, such as the presence of serum proteins (An and Friedman, 1998). Surface chemistry strongly affects bacterial attachment, with different microorganisms preferentially attaching to surfaces with different chemistries (Ista et al., 2004, 2010; Khan et al., 2011; Liu et al., 2008). Initial attachment has been shown to affect later biofilm structure, adhesion strength and detachment (Busscher et al., 1995; Lackner et al., 2009). The driving force for bacterial attachment has been suggested to be the interfacial surface tension between a cell and a surface, less the interfacial surface tensions between the cell and the liquid and the liquid and the surface (Ista et al., 2004). Attachment surface chemistry is known to affect bacterial attachment (reviewed in Van Loosdrecht et al., 1990), including that of pure heterotroph strains in biofilm reactors
2191
(Dimitrov et al., 2007). Reports for nitrifying bacteria have been inconsistent: more hydrophilic surfaces gave higher nitrification rates in a pure culture study with the ammoniaoxidizing bacteria (AOB) Nitrosomonas europaea and Nitrobacter winogradskyi (Kim et al., 1997), but the opposite trend was also reported (Sousa et al., 1997). Pure strains of the AOB N. europaea and Nitrosospira multiformis preferably attached to higher surface energy (a measure of the surface characteristics that can drive attachment) self-assembled monolayer surfaces (Khan et al., 2011). In a mixed culture nitrifying system, surfaces modified with amine groups produced more biofilm than unmodified or methyl group-modified surfaces (Lackner et al., 2009). It was hypothesized that increasing attachment surface energy can increase biofilm activity with respect to nitrification by increasing biofilm mass and/or the nitrifier biofilm fraction, and that it can improve hormone removal, either because of improvements in nitrification or because of an increase and/or change in heterotrophic biomass. The objective of this study was to test these hypotheses, with a focus on early-phase biofilm development (over 8 days). The experimental approach was to culture biofilms on plastic surfaces with a range of chemical properties by immersing them in a full scale biological wastewater treatment system, and then analyzing them in laboratory batch tests.
2.
Methods
2.1.
Biofilm development and batch tests
Four attachment surface plastics were evaluated: acetal polymer plastic (Ensinger Plastic, PA, USA), high density polyethylene (HDPE) (Sheffield Plastics, MA, USA), melamine (Regal Piedmont Plastics LLC, NM, USA), and nylon (type MC 907 natural, Regal Piedmont Plastics LLC, NM, USA). These were selected to provide a broad range of wettability (hydrophilicity, as determined by water contact angle values) and surface energies. Two sheets of each of the 4 materials (20 20 cm, 0.16 cm thick) were fastened to a PVC pipe rack in a parallel configuration with 2.5 cm spacing between each sheet. An additional sheet of plastic was attached to the top and bottom of the stack so that all sheets had similar hydraulic conditions. The rack was placed at a 2 m depth in the aerobic reactor of a full-scale activated sludge system of the City of Albuquerque Southside Water Reclamation Facility, with the sheets oriented horizontally, for 8 days. This 230,000 m3/d average daily flow system includes a nitrification/denitrification Modified LudzackeEttinger configuration. Typical parameter values in this reactor were: total suspended solids (TSS), 4.3 g/L; NH4-N, 16 mg/L; NO3-N, 14 mg N/L; NO2-N, 0.7 mg N/L; dissolved P, 2.7 mg/L; COD 20 mg/L, dissolved oxygen controlled at 2 mg/L, and temperature, 17e22 C. After 8 days the sheets were carefully removed and each was transferred to a separate batch reactor containing 2.2 L of synthetic wastewater based on the recipe in Schuler and Jenkins (2003), except there was no nitrification inhibitor and the initial ammonia nitrogen concentration was 132 mg N/L to
2192
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
insure it was not limiting during the batch tests. The 8 day time frame was selected to target early-phase biofilm development. Estrone (E1), b-estradiol (E2) and 17a-ethynylestradiol (EE2) (>99% purity, SigmaeAldrich, USA) were added to the synthetic media to a final concentration of 1000 ng/L each. Continuous aeration was provided for 6 h, after which time the plastic sheets were removed. A 5-ml sample was collected from each batch reactor at 2-h intervals for nitrogen compound analysis. Samples for E1, E2 and EE2 measurements were collected at 6 h. All rates and rate constants were calculated based on the final (6 h) concentrations. All samples were immediately filtered through 0.45 mm microfiber filters (Whatman, USA) and stored at 4 C or analyzed immediately.
2.2.
Solids and nitrogen compounds
Before batch testing, a 7.6 cm 7.6 cm area was scraped from one side of each plastic sheet using a sterilized razor blade to remove biomass. This was filtered through a pre-weighed glass fiber filter (Whatman, USA) and the total attached solids (TAS) were measured by Standard Method 2540 D (American Public Health Association, 2005). Experimental replicates were included by testing two sheets of each plastic type. NH3-N, NO3-N and NO2-N were measured in 0.45 mm filtered samples using Hach methods 10031, 10020 and 10019 (Hach Company, USA), respectively. Each sample was measured in duplicate or triplicate.
2.3.
Estrogens
Hormones were measured by New Mexico Department of Health, Scientific Laboratory Division personnel. Samples were filtered through 0.45 mm filters and 2 L of each were stored in amber glass bottles at 4 C without preservative. Liquideliquid extraction was performed at room temperature according to US EPA Method 3510C. Briefly, target analytes were spiked into 2 fortified blank quality control samples at 500 ng/L and 700 ng/L (recoveries were 88e101%, 78e82%, and 89e95%, for E1, E2, and EE2, respectively). Progesterone was spiked as a surrogate into all samples and quality control samples at 700 ng/L for quality control. A measured volume (approximately 2 L) of each sample was poured into a separatory funnel and 60 ml of dichloromethane were added after rinsing sample bottle. Separatory funnels were vigorously shaken for 1 min and they were allowed to separate for three to 5 min. The solvent layer was drained through a bed of anhydrous sodium sulfate into a 500 ml Erlenmeyer flask. Extraction with 60 mL dichloromethane was repeated, including shaking, separation, and passage through anhydrous sodium sulfate. The pH of the remaining solution was adjusted to 12 with 6N sodium hydroxide, and was extracted twice more with 60 mL dichloromethane. The pH of the aqueous solution was adjusted to w2 with concentrated sulfuric acid and it was extracted twice more with 60 mL dichloromethane. These extractions produced w420 mL of dichloromethane, with the extracted analytes, which was placed in a 35 C water bath and evaporated with ultra-high purity nitrogen gas until the final volume was w1 ml. Hormones were measured by gas chromatography-mass spectrometry (Varian 3800 with a type 1079 injector, an RXI-5 30 m
fused silica capillary column, and a Saturn 2000 ion trap mass spectrometer) with 50 ml injections. The detector scanned for ions in the range 40 m/z to 650 m/z. Compounds were identified by correlation to predetermined retention times using individual standards at known concentrations and to a reference spectra.
2.4.
Surface energy
The plastic contact angles were determined by goniometry for the three probe liquids ultrapure water, diiodomethane (99%, SigmaeAldrich, USA), and formamide (>99.5%, SigmaeAldrich, USA). The sessile drop technique was used (Rame´Hart Instrument Co., Goniometer Model# 400-22-300 with DROPimage Standard, NJ) as previously described (Khan et al., 2011). At least five 1.5 mL droplets were measured with each liquid. Surface energies were calculated from the measured contact angles as summarized in Liu et al. (2008). Briefly, the total surface energy (gtotal) is the sum of apolar (Lifshitz-van der Waals; LW) and polar (Lewis acid-base; AB) components: gtotal ¼ gLW þ gAB
(1) AB
The acid-base component (g ) is the geometric mean of the electron-donor (g) and electron-acceptor (gþ) parameters for the applied probe liquid or the substrata, given by: pffiffiffiffiffiffiffiffiffiffiffiffiffiffi gAB ¼ 2$ gþ $g
(2) LW
þ
The relationships between g , g , and g for a surface (x) and for a probe liquid (L) is provided by the YoungeDupre´ equation: qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffi ffi þ LW g þ 2$ gþ gL ðcos qL þ 1Þ ¼ 2$ gLW x $gL x $gL þ 2$ x $gL
(3)
where gL is for the liquidevapor interaction, qL is the probe liquidesolid contact angle. Reference values for the probe liquids (gL, gþ L and gL ) were from Good and van Oss (1992) and confirmed with the supplier. Applying Eq. (3) and contact angle measurements for the three probe liquids water (W), diiodomethane (D) and formamide (F), the three unknown þ surface energy components (gLW x ; gx and gx ) of each surface were calculated using Eq. (4). 131 0 ffi pffiffiffiffiffiffiffi pffiffiffiffiffiffiffi 192 3 82 0 pffiffiffiffiffiffiffiffi þ gLW g gw $½cosðqW Þþ1 = gLW < Wffi W pg W x p ffiffiffiffiffiffiffiffi ffiffiffiffiffiffi p ffiffiffiffiffiffi 6 B C 7 4 gþ 5 ¼ 42$@ gLW g $@ gD $½cosðqD Þþ1 A gþ x D ffi D D A5 pffiffiffiffiffiffiffiffi ; : pffiffiffiffiffiffi pffiffiffiffiffiffi gF $½cosðqF Þþ1 g g g gLW x F F F 2
(4)
3.
Results
3.1.
Plastic surface properties
The liquid contact angles (q values) for the probe liquids, the calculated surface energy parameters, and the total surface energies for the four plastic surfaces are shown in Table 1. Nylon was the most hydrophilic surface (it had the lowest water contact angle), and it had the highest total surface energy. The acetal plastic surface was the most hydrophobic surface, and it had the lowest total surface energy. HDPE,
2193
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
Table 1 e Contact angles (q) and surface energy (g) parameters of the four plastic surfaces. The subscripts W, D and F denote water, diiodomethane and formamide, respectively. Average values ± one standard deviation are shown, as calculated from at least three measurements. Plastic surfaces qW Nylon Melamine HDPE Acetal
Surface energy parameters (mJ/m2)
Contact angle (degrees)
53 84 90 98
qD 2 0 1 0
36 31 51 46
gLW
qF 0 1 1 0
36 38 76 94
2 0 1 1
which is commonly used for IFAS media, was the second most hydrophobic surface and it had the second lowest total surface energy.
3.2. Mass of attached biofilm and nitrogen transformations The quantity of attached biofilm was correlated with both surface energy (Fig. 1) and attachment surface hydrophilicity (indicated by cos[qW]; R2 ¼ 0.95), ranging from 0.21 0.03 mg/ cm2 (acetal) to 0.57 0.06 mg/cm2 (nylon). Ammonia removal fluxes (mass rate per attachment surface area), ranged from 0.55 to 2.53 g N/(m2*d), with the lowest on acetal plastic and the highest on nylon (the time series data for the batch tests is shown in Supplemental Fig. S1). Ammonia fluxes were correlated with both surface energy (Fig. 2) and cos(qW) values (R2 ¼ 0.78, Fig. S2 in Supplementary Data). The change in ammonia concentration in the experimental control (an HDPE sheet with no biofilm growth) was zero (Fig. S1). The positive correlation between nitrification rates and surface energy values was likely due in part to the correlation between attached biomass with surface energy (Fig. 1), as more biomass is expected to produce higher removal rates. The specific rates of nitrification (expressed as the mass of ammonia removed per unit of attached biomass per unit time) were therefore calculated to assess whether increasing surface energy increased the nitrifying activity per unit biofilm mass, and these were also correlated with the attachment surface energy (Fig. 2). This correlation suggested that as
Fig. 1 e The relationships between attachment surface energy of the four attachment surfaces and attached biomass.
41.4 43.8 27.1 13.7
0.1 0.2 0.5 0.2
gþ 0.6 2.4 0.4 1.3
g 0.2 0.1 0.2 0.1
22.7 0.1 6.4 9.9
gAB 2.2 0.03 1.0 0.4
7.5 1.9 3.3 7.2
1.0 0.1 0.8 0.8
gtotal 48.9 45.7 30.4 20.9
1.0 0.3 0.8 0.7
surface energy increased, not only did the total mass of biofilm increase, but the biofilm composition changed, with nitrifier activity per mass of biofilm increasing with the attachment surface energy. The correlation between cos[qW] values and specific ammonia consumption rates was much weaker (R2 ¼ 0.34; Supplemental Fig. S2), suggesting that surface energy was a better predictor of total and specific nitrification rates than was hydrophilicity. Nitrification occurs in a two step process: ammonia oxidation to nitrite, and nitrite oxidation to nitrate, which are performed by AOB and nitrite-oxidizing bacteria (NOB), respectively (Tchobanoglous et al., 2003). NOx (the sum of nitrate and nitrite) production is an indication of AOB activity, since NOB transform the nitrite produced by AOB to nitrate. NOx production rates were consistent with ammonia removal, as both the NOx fluxes and the specific rates of production were correlated with the attachment surface energies (Fig. 3), ranging from 0.23 to 0.98 g N/(m2*d) and 4.5e7.1 mg N/ (g TAS*h), respectively. There were no clear trends in nitrite accumulation rates relative to surface energy, with higher rates on HDPE and melamine than the other surfaces (Supplementary data Fig. S3). Because nitrite is produced by AOB and consumed by NOB, nitrite accumulation indicates higher AOB than NOB activity, and correlations with surface energy were not necessarily expected.
Fig. 2 e The relationship between surface energy values and the ammonia removal rates expressed as fluxes to the media surface and as specific rates (per total attached solids, or TAS).
2194
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
Fig. 4 e The relationship between surface energy and hormone removal first order rate constants (k). Fig. 3 e The relationship between surface energy values and the NOx (the sum of nitrate and nitrite) production rates expressed as fluxes to the media surface and as specific rates.
3.3.
of 0.5 for the specific EE2 removal rate constant indicated a weaker correlation with surface energy than for E1 and E2 (R2 values of 0.73 and 0.82, respectively).
Estrogen removal
Hormone removal rates in aerobic biological wastewater treatment processes have generally been considered to follow first order kinetics (Eq. (5)), and so the first order reaction rate constant (k, units 1/time) has been used to compare reaction kinetics between studies (reviewed in de Mes et al., 2005). Ct ¼ Co ekt
(5)
where Co is the initial hormone concentration and Ct is the concentration at time t. Because biomass concentrations also affect estrogen removal rates, k values are of limited use for comparing different studies, and specific reaction rate constants (per microbiological mass) have also been calculated (de Mes et al., 2005): 0
Ct ¼ Co eXk t
(6)
where X is the biomass concentration and k0 is the specific first order rate constant (1/[time*biomass concentration]). The k and k0 values were calculated by solving Equations (5) and (6), respectively. Ct was the hormone concentration at the end of 6 h batch test and X was the biofilm mass per volume of the batch reactor. The k values followed similar trends as the ammonia and NOx rates (Figs. 2 and 3), with positive correlations between attachment surface energy values and reaction rate constants for E1, E2, and EE2 (Fig. 4). The control sample (an HDPE sheet with no biofilm) yielded zero removal for all hormones, indicating losses due to volatilization and sorption to the batch reactor walls were negligible. Given the greater amounts of biofilm on the higher surface energy surfaces (Fig. 1), these results were not surprising. However, the positive correlations found for the specific reaction rate constant (k0 ) values (Fig. 5) suggested that not only was there more biofilm on the higher surface energy surfaces, but also that it was more effective per unit mass than the biofilms on the lower surface energy surfaces, with increasing removal rates per unit biomass with increasing surface energy, although the relatively low R2 value
4.
Discussion
4.1.
Biofilm attachment and nitrogen transformations
The range of attached biomass values (Fig. 1) was similar to the 0.24e0.60 mg/cm2 reported by Regmi et al. (2011) for biofilms grown on virgin polyethylene media (AnoxKaldnes K3) in a full scale system. The finding that the overall amount of attached biomass increased with increasing surface energy was consistent with previous reports of greater adhesion by both pure strains of heterotrophic bacteria and AOBs (Dimitrov et al., 2007; Khan et al., 2011), and that amine modification increased the strength of wastewater biofilm adhesion (Lackner et al., 2009). The range of ammonia nitrogen fluxes across the four surface types (0.55e2.53 g N/(m2*d; Fig. 2) was comparable to previous studies: 0.2e1.0 g N/(m2*d) in an MBBR with polyethylene carriers (Hem et al., 1994), 0.4e1.0 g N/(m2*d) in an MBBR (Zimmerman et al., 2003), and 0 to approximately 2 g N/
Fig. 5 e The relationship between surface energy and specific hormone removal first order rate constants (k0 ).
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
(m2*d) in an IFAS system with HDPE carriers (AnoxKaldnes type K1) (Lydmark et al., 2007). The range of specific nitrification rates (10.9 for HDPE to 21.2 mg NH3-N/(g TAS*h) for melamine; Fig. 2), was similar to but somewhat higher than some previous reports, with a range of 0.53e5.9 mg NH3-N/(g TAS*h) for IFAS media (Kim et al., 2011; Onnis-Hayden et al., 2007). Regmi et al. (2011) reported approximately 2e10 g NOx-N/ (g TAS*h) in a full scale IFAS system with HDPE media, and NH3-N removal rates were likely higher, as not all NH3-N is transformed to NOx. The observed rates of NOx production (Fig. 3) were also similar to the 0.28e1.12 g NO3-N/(m2*d), or 2.4e5.9 mg g NO3-N/(g TAS*h) by the attached phase of a full scale IFAS system with HDPE media reported by OnnisHayden et al. (2011), Regmi et al. (2011) reported 0.90 g NOxN/(m2*d) in a full scale system with HDPE media, as well as the specific rates noted above, and Kim et al. (2011) reported 0.058e1.35 mg N/(g TSS*h) nitrate production in an IFAS system with HDPE media. Factors such as relative heterotroph/ nitrifier content (due to varying COD/N ratios, for example), mixing rates, media geometries, and substrate concentrations likely explain differences between studies. Furthermore, this study focused on early-stage biofilm development, and utilized fixed-in-place flat sheets rather than floating media with interior surfaces; in light of these differences, the similarity in values between studies was notable. The correlations noted above support the hypothesis of preferential attachment of AOB to higher surface energy substrata, as demonstrated in our previous study using pure AOB strains and well-defined surfaces created using selfassembled monolayers (SAMs) where the adhesion rates of the AOBs N. europaea and N. multiformis were correlated with surface energy (Khan et al., 2011). Similarly, Kim et al. (1997) demonstrated that rates of nitrification were higher on a hydrophilic surface than more hydrophobic surfaces using pure strains of AOB (N. europaea and N. winogradskyi). It could be argued that enrichment of nitrifiers on the higher surface energy surfaces was due to these surfaces providing a stronger bond with the attached phase in general. This would provide a thicker biofilm with less sloughing, consistent with reports of stronger wastewater biofilm adhesion to amine-modified surfaces (Lackner et al., 2009), and the provision of a more stable, longer retention time environment could be favorable to slow-growing nitrifiers. However, the current study analyzed biofilms grown over only 8 days, which makes this explanation less likely, and instead suggests enrichment of AOB by a more specific increase in attachment rates and/or attachment strength on higher energy surfaces. It may be significant that the higher energy surfaces (melamine and nylon) both included amine groups, as in Lackner et al. (2009). Because nitrifiers take up ammonia as an electron donor for energy production, it may be that they have an affinity for surfaces containing amine groups. Indeed, the data presented here is consistent with a hypothesis that increasing surface energy led to greater overall biomass (Fig. 1) and ammonia removal (Fig. 2), while the presence of amine groups led to greater enrichment of nitrifiers (melamine and nylon, which had similar specific rates) relative to surfaces without amine groups (acetal and HDPE, which also had similar specific rates, and were much lower than the amine-containing surface specific rates). Further research is required to
2195
determine whether the increasing specific rates were due to changes in the surface energy or the presence of amine groups (or some other factor); this could be tested by comparing high surface energy surfaces with and without amine groups.
4.2.
Estrogen removal
The data in Figs. 4 and 5 suggest that higher surface energy attachment surfaces could be used to produce biofilms with improved estrogen removal abilities. A wide range of estrogen removal rates have been reported in previous studies, with most previous work performed on suspended growth systems, and their applicability to submerged biofilms systems is not known. An excellent review was provided by de Mes et al. (2005), with calculation of the specific rate constants (k0 ) by the authors, thereby accounting for variations in test duration and biomass concentrations between studies. Reaction rate k0 values for estrogen removal have ranged over several orders of magnitude across different activated sludge studies (for example, from 0.22 to 768 L/(g TAS*d) for E2), encompassing the ranges of values shown in Fig. 5. These varying results in previous studies were likely due to differences in sludge characteristics such as solids residence times, which have been suggested to improve removal of estrogens (Clara et al., 2005). The initial concentrations selected for use in the current study (1000 ng/L) were somewhat higher than typically reported in full scale wastewater influents (for example, Muller et al. (2008) reported maximum concentrations of 119, 28, and 38 ng/L of E1, E2, and EE2, respectively, in raw sewage and primary effluent, not including conjugated forms) because the rates of uptake were unknown in advance, and it was necessary to provide sufficient quantities that the hormone concentrations were not likely to fall below detection limits during the batch tests, while still yielding measurable changes in concentration. The initial concentrations were also orders of magnitude lower than those used in many previous studies (see review by de Mes et al. (2005)). If the assumption of first order kinetics is reasonably accurate the results reported here should be relevant for lower, more common concentrations as well. Previous work has generally reported that EE2 was more persistent than E1 and E2 in suspended growth wastewater treatment systems (de Mes et al., 2005). This was not the case for the data shown in Fig. 5, where EE2 rate constants were comparable to those of E1 and E2, although its rate constants were less than the average of the three estrogens on all surfaces except for acetal plastic. However, others have reported EE2 removal at higher rates, although k0 values were not calculated. For example, 80% of EE2 was removed in conventional activated sludge systems, which was similar to E1 removal (Gunnarsson et al., 2009). Similarly, Layton et al. (2000) reported that 80% of EE2 was removed by activated sludge biomass, and although only 20% was mineralized to CO2, 75% of its estrogenic activity was removed. Ternes et al. (1999) reported that 99.9% of E2, 83% of E1 and 78% of EE2 were removed in a Brazilian activated sludge plant, but a German plant removed 64% of E2 and negligible amounts of E1 and EE2. It therefore appears that EE2 may be resilient in some systems, but for unknown reasons other systems with different
2196
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
biomass qualities remove EE2 at higher rates, as was found in this study. EE2 also had the lowest R2 value (0.50) for the linear correlation between k0 and surface energy values of the 3 hormones, indicating the weakest correlation (Fig. 5), and suggesting that surface energy may have played less of a role in its removal compared to the other hormones. Little work has been performed on EE2 removal in submerged biofilm systems, but Cargouet et al. (2004) also reported similar removal percentages for E1 (55%), E2 (43%) and EE2 (45%) in a biofilm treatment system, and Gunnarsson et al. (2009) demonstrated that an MBBR used as a polishing step provided greater than 91% removal for E2 and EE2 (less than the limit of detection), and 100% for E1. The potential for improved EE2 removal by biofilm systems indicated by these studies is important to practice, as it has been reported to be 11e27 times more potent than E2 for endocrine disruption in fish (Thorpe et al., 2003). It has been suggested that some estrogens can be removed from wastewater by nitrifiers, possibly by cometabolic activity of the nitrification enzyme ammonia monooxygenase (Dytczak et al., 2008; Khunjar et al., 2011; Shi et al., 2004; Vader et al., 2000; Yi and Harper, 2007); by heterotrophic bacteria that may use the hormones as electron donors (Gaulke et al., 2008); and by adsorption to the biomass (Andersen et al., 2005). As noted, the correlations between the rate constant k values and surface energy (Fig. 4) were attributable at least in part to the correlation between attached biomass and surface energy (Fig. 1). However, the correlations between the specific rate constant (k0 ) and surface energy values (Fig. 5) suggested that estrogen removal activity increased per unit of biomass with increasing surface energy, suggesting one or more of the above mechanisms (cometabolism by nitrifiers, heterotroph biodegradation, or adsorption) increased as well on a per mass basis. The findings that both specific nitrification (Figs. 2 and 3) and specific hormone removal (Fig. 5) rates increased with increasing substrata surface energy suggested that the mechanism of estrogen removal by nitrifiers was at least plausible for this system. Ammonia and estrogen removal as fluxes were well correlated (R2 values > 0.84; Supplemental Fig. S4), due in part to the greater biomass on the higher surface energy surfaces (Fig. 1). Correlations between the specific rates of ammonia and estrogen removal were much weaker (R2 values 0.19, 0.31 and 0.58 for E1, E2, and EE2, respectively; Supplemental Fig. S5), casting doubt on the hypothesis that nitrifiers were primarily responsible for hormone removal in this system. It is also possible that the correlation between specific estrogen removal and substrata surface energy was attributable to differences in the adsorption properties of the biofilms grown on each surface, which would suggest the substrata surface energy affected the adsorptive properties of the attached biofilm. For example, higher surface energy biofilms could have formed on the higher surface energy plastics. It has been estimated that sorption plays a relatively small role in removal in suspended growth systems, at only 1.5e1.8% of the total estrogen loading in typical activated sludge domestic wastewater plants (Andersen et al., 2005), based on calculated values using sorption coefficients. However, Layton et al. (2000) reported that 80% of EE2 was removed within minutes
in a batch test but with little change over 24 h thereafter, which was consistent with removal by adsorption. Although only 20% was mineralized to CO2, this occurred continuously without further change in aqueous EE2, which may suggest adsorbed EE2 was biodegraded during the experiment. IFAS biofilms have been reported to be more hydrophobic, have less EPS and have EPS with different composition than suspended growth flocs (Mahendran et al., 2012); such differences may have been responsible for observations that biofilm systems provided better removal of EE2 than suspended growth systems (Cargouet et al., 2004; Gunnarsson et al., 2009). The effects of attachment surface chemistries on biofilm adsorption characteristics is not known, but such phenomena could have occurred by, for example, influencing the intracellular extracellular polymeric substance (EPS) matrix quantity and composition. The possibility that differences in adsorptive properties may have played a role in the observed differences in hormone removal across the surfaces requires further research and could be tested by including inhibited controls. Gaulke et al. (2008) suggested that abiotic transformation of EE2 by nitration may have been responsible for reported hormone removals in some previous studies; they showed this to occur in samples with high nitrite concentrations. Because nitrite concentrations were less than 1.5 mg/L in this study (Supplemental Fig. S3) this mechanism of hormone removal appears unlikely in this study. While these results demonstrated that higher surface energy plastics provided improved nitrification and hormone removal rates, both on absolute and per unit biomass bases, further research is required to better understand the precise mechanisms by which this occurred. This study focused on early-stage (8 day) biofilm development, and so future work should also determine the extent to which the observed phenomena persist in systems operated for longer periods. Finally, more realistic, free floating IFAS/MBBR media should be considered, which would better represent common practice in full scale systems.
5.
Conclusions
This study provided a systematic analysis of how important functional behaviors of mixed culture biofilms are affected by attachment surface chemistry in wastewater treatment systems. Biofilms grown on flat plastic sheets with a range of chemical characteristics in a full scale treatment system aerobic reactor demonstrated that attachment surface energy was positively correlated with biofilm mass, nitrification, and estrogen removal. Nitrification and hormone removal rates were correlated with surface energy both as fluxes to the biofilms and as a specific rates (per mass of biofilm), suggesting that not only was more biofilm produced on the higher surface energy surfaces, but that the specific activity of the biofilm increased as well, demonstrating an enrichment of nitrifiers on the higher surface energy surfaces. IFAS and MBBR media is commonly made of low surface energy plastics, such as HDPE. Future design may benefit from consideration of alternative materials, including high energy plastics such as nylon, or surface coatings. Improvements in treatment performance may justify increased costs in media
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
materials and manufacture. The generally good removal of EE2 by all biofilms in this study warrants further investigation. Future work should also consider biofilms grown over longer terms, and with media geometries more commonly used in practice.
Acknowledgments This work was supported by the Water Environment Research Foundation (Paul L. Busch Award) and by the National Science Foundation (Grant # 0852469).
Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.watres.2013.01.036.
references
American Public Health Association, American Water Works Association and Water Environment Federation, 2005. Standard Methods for the Examination of Water and Wastewater (Washington, D.C). An, Y.H., Friedman, R.J., 1998. Concise review of mechanisms of bacterial adhesion to biomaterial surfaces. Journal of Biomedical Materials Research 43 (3), 338e348. Andersen, H., Siegrist, H., Halling-Sorensen, B., Ternes, T.A., 2003. Fate of estrogens in a municipal sewage treatment plant. Environmental Science and Technology 37 (18), 4021e4026. Andersen, H.R., Hansen, M., Kjolholt, J., Stuer-Lauridsen, F., Ternes, T., Halling-Sorensen, B., 2005. Assessment of the importance of sorption for steroid estrogens removal during activated sludge treatment. Chemosphere 61 (1), 139e146. Busscher, H.J., Bos, R., Vandermei, H.C., 1995. Initial microbial adhesion is a determinant for the strength of biofilm adhesion. FEMS Microbiology Letters 128 (3), 229e234. Cargouet, M., Perdiz, D., Mouatassim-Souali, A., TamisierKarolak, S., Levi, Y., 2004. Assessment of river contamination by estrogenic compounds in Paris area (France). Science of the Total Environment 324 (1e3), 55e66. Clara, M., Kreuzinger, N., Strenn, B., Gans, O., Kroiss, H., 2005. The solids retention timeda suitable design parameter to evaluate the capacity of wastewater treatment plants to remove micropollutants. Water Research 39, 97e106. de Mes, T., Zeeman, G., Lettinga, G., 2005. Occurrence and fate of estrone, 17b-estradiol and 17a-ethynylestradiol in STPs for domestic wastewater. Reviews in Environmental Science and Technology 4, 275e311. Dimitrov, D., Hadjiev, D., Nikov, I., 2007. Optimisation of support medium for particle-based biofilm reactors. Biochemical Engineering Journal 37, 238e245. Dytczak, M.A., Londry, K.L., Oleszkiewicz, J.A., 2008. Biotransformation of estrogens in nitrifying activated sludge under aerobic and alternating anoxic/aerobic conditions. Water Environment Research 80 (1), 47e52. Gaulke, L.S., Strand, S.E., Kalhorn, T.F., Stensel, H.D., 2008. 17aethinylestradiol transformation via abiotic nitration in the presence of ammonia oxidizing bacteria. Environmental Science and Technology 42 (20), 7622e7627.
2197
Good, R.J., van Oss, C.J., 1992. In: Shrader, M.E., Loeb, G.I. (Eds.), Modern Approaches to Wettability, Theory and Applications. Plenum Press, New York. Gunnarsson, L., Adolfsson-Erici, M., Bjorlenius, B., Rutgersson, C., Forlin, L., Larsson, D.G.J., 2009. Comparison of six different sewage treatment processes-Reduction of estrogenic substances and effects on gene expression in exposed male fish. Science of the Total Environment 407 (19), 5235e5242. Hem, L.J., Rusten, B., Odegaard, H., 1994. Nitrification in a movingbed biofilm reactor. Water Research 28 (6), 1425e1433. Ista, L.K., Callow, M.E., Finlay, J.A., Coleman, S.E., Nolasco, A.C., Simons, R.H., Callow, J.A., Lopez, G.P., 2004. Effect of substratum surface chemistry and surface energy on attachment of marine bacteria and algal spores. Applied and Environmental Microbiology 70 (7), 4151e4157. Ista, L.K., Mendez, S., Lopez, G.P., 2010. Attachment and detachment of bacteria on surfaces with tunable and switchable wettability. Biofouling 26 (1), 111e118. Johnson, A.C., Sumpter, J.P., 2001. Removal of endocrinedisrupting chemicals in activated sludge treatment works. Environmental Science and Technology 35 (24), 4697e4703. Khan, M.M.T., Ista, L.K., Lopez, G.P., Schuler, A.J., 2011. Experimental and theoretical examination of surface energy and adhesion of nitrifying and heterotrophic bacteria using self-assembled monolayers. Environmental Science and Technology 45 (3), 1055e1060. Khunjar, W.O., Mackintosh, S.A., Skotnicka-Pitak, J., Baik, S., Aga, D.S., Love, N.G., 2011. Elucidating the relative roles of ammonia oxidizing and heterotrophic bacteria during the biotransformation of 17 alpha-ethinylestradiol and trimethoprim. Environmental Science and Technology 45 (8), 3605e3612. Kim, H.-S., Schuler, A.J., Gunsch, C.K., Pei, R.T., Gellner, J.W., Boltz, J.P., Freudenberg, R.G., Dodson, R., 2011. Attached and suspended phase nitrifying microbial community structures and functions in integrated fixed film activated sludge. Water Environment Research 83 (7), 627e635. Kim, Y.H., Cho, J.H., Lee, Y.W., Lee, W.K., 1997. Development of a carrier for adhesion of nitrifying bacteria using a thermodynamic approach. Biotechnology Techniques 11 (11), 773e776. Lackner, S., Holmberg, M., Terada, A., Kingshott, P., Smets, B.F., 2009. Enhancing the formation and shear resistance of nitrifying biofilms on membranes by surface modification. Water Research 43 (14), 3469e3478. Layton, A.C., Gregory, B.W., Seward, J.R., Schultz, T.W., Sayler, G.S., 2000. Mineralization of steroidal hormones by biosolids in wastewater treatment systems in Tennessee USA. Environmental Science and Technology 34 (18), 3925e3931. Liu, Y., Gallardo-Moreno, A.M., Pinzon-Arango, P.A., Reynolds, Y., Rodriguez, G., Camesano, T.A., 2008. Cranberry changes the physicochemical surface properties of E. coli and adhesion with uroepithelial cells. Colloids and Surfaces B-Biointerfaces 65 (1), 35e42. Lydmark, P., Almstrand, R., Samuelsson, K., Mattsson, A., Sorensson, F., Lindgren, P.-E., Hermansson, M., 2007. Effects of environmental conditions on the nitrifying population dynamics in a pilot wastewater treatment plant. Environmental Microbiology 9 (9), 2220e2233. Mahendran, B., Lishman, L., Liss, S.N., 2012. Structural, physicochemical and microbial properties of flocs and biofilms in integrated fixed-film activated sludge (IFFAS) systems. Water Research 46 (16), 5085e5101. Muller, M., Rabenoelina, F., Balaguer, P., Patureau, D., Lemenach, K., Budzinski, H., Barcelo, D., De Alda, M.L., Kuster, M., Delgenes, J.P., Hernandez-Raquet, G., 2008. Chemical and biological analysis of endocrine-disrupting hormones and estrogenic activity in an advanced sewage
2198
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 1 9 0 e2 1 9 8
treatment plant. Environmental Toxicology and Chemistry 27 (8), 1649e1658. Onnis-Hayden, A., Dair, D., Johnson, C., Schramm, A., Gu, A.Z., 2007. Kinetics and nitrifying populations in nitrogen removal processes at a full-scale integrated fixed-film activated sludge (IFAS) plant. In: Proceedings of Proceedings of the Water Environment Federation Annual Conference (WEFTEC), Washington, D.C. Onnis-Hayden, A., Majed, N., Schramm, A., Gu, A.Z., 2011. Process optimization by decoupled control of key microbial populations: distribution of activity and abundance of polyphosphate-accumulating organisms and nitrifying populations in a full-scale IFAS-EBPR plant. Water Research 45 (13), 3845e3854. Randall, C.W., Sen, D., 1996. Full-scale investigation of an integrated fixed-film activated sludge (IFAS) process for enhanced nitrogen removal. Water Science and Technology 33 (12), 155e162. Regmi, P., Thomas, W., Schafran, G., Bott, C., Rutherford, B., Waltrip, D., 2011. Nitrogen removal assessment through nitrification rates and media biofilm accumulation in an IFAS process demonstration study. Water Research 45 (20), 6699e6708. Schuler, A.J., Jenkins, D., 2003. Enhanced biological phosphorus removal from wastewater by biomass with varying phosphorus contents, part I: experimental methods and results. Water Environment Research 75 (6), 485e498. Shi, J., Fujisawa, S., Nakai, S., Hosomi, M., 2004. Biodegradation of natural and synthetic estrogens by nitrifying activated sludge and ammonia-oxidizing bacterium Nitrosomonas europaea. Water Research 38 (9), 2323e2330.
Sousa, M., Azeredo, J., Feijo, J., Oliveira, R., 1997. Polymeric supports for the adhesion of a consortium of autotrophic nitrifying bacteria. Biotechnology Techniques 11 (10), 751e754. Tchobanoglous, G., Burton, F.L., Stensel, H.D., 2003. Wastewater Engineering: Treatment and Reuse. Metcalf & Eddy, Inc., McGraw-Hill, New York. Ternes, T.A., Stumpf, M., Mueller, J., Haberer, K., Wilkena, R.-D., Servosb, M., 1999. Behavior and occurrence of estrogens in municipal sewage treatment plants e I. Investigations in Germany, Canada and Brazil. Science of the Total Environment 225, 81e90. Thorpe, K.L., Cummings, R.I., Hutchinson, T.H., Scholze, M., Brighty, G., Sumpter, J.P., Tyler, C.R., 2003. Relative potencies and combination effects of steroidal estrogens in fish. Environmental Science and Technology 37 (6), 1142e1149. Vader, J.S., van Ginkel, C.G., Sperling, F., de Jong, J., de Boer, W., de Graaf, J.S., van der Most, M., Stokman, P.G.W., 2000. Degradation of ethinyl estradiol by nitrifying activated sludge. Chemosphere 41 (8), 1239e1243. Van Loosdrecht, M.C.M., Lyklema, J., Norde, W., Zehnder, A.J.B., 1990. Influence of interfaces on microbial activity. Microbiological Reviews 54 (1), 75e87. Yi, T., Harper, W.F., 2007. The link between nitrification and biotransformation of 17 alpha-ethinylestradiol. Environmental Science and Technology 41 (12), 4311e4316. Zimmerman, R.A., Richard, D., Bradshaw, A.T., Craddock, P.P., 2003. Pilot-scale evaluation of separate-stage nitrification using an attached-growth, moving-bed media process. Water Environment Research 75 (5), 422e433.