PDF Archive

Easily share your PDF documents with your contacts, on the Web and Social Networks.

Send a file File manager PDF Toolbox Search Help Contact


J Environ Sci Health, 40, 2005, 535 551 .pdf



Original filename: J Environ Sci Health, 40, 2005, 535-551.pdf
Title: TJ1385-05-46557.tex
Author: fdz

This PDF 1.3 document has been generated by Textures¨: LaserWriter 8 8.6.5 / Acrobat Distiller 5.0.5 for Macintosh, and has been sent on pdf-archive.com on 03/11/2015 at 02:17, from IP address 71.17.x.x. The current document download page has been viewed 257 times.
File size: 352 KB (17 pages).
Privacy: public file



Download original PDF file








Document preview


Journal of Environmental Science and Health, 40:535–551, 2005
C Taylor & Francis Inc.
Copyright
ISSN: 1093-4529 (Print); 1532-4117 (Online)
DOI: 10.1081/ESE-200046557

Evidence for Tin Inhibition
of Enhanced Biological
Phosphorus Removal at
a Municipal Wastewater
Treatment Plant
Sierra Rayne,1 Sheila Carey,2 and Kaya Forest1
1
2

Ecologica Environmental Consulting, Victoria, Canada
Kelowna Wastewater Treatment Facility, City of Kelowna, Kelowna, Canada

Concentrations of 34 metals were determined in the concentrated waste activated sludge
from a dissolved air flotation unit at an advanced municipal wastewater treatment plant
(WWTP) with biological nutrient removal. Reduction in enhanced biological phosphorus
removal (EBPR) efficiency was observed at total tin concentrations greater than 4 µg L−1
in the solids fraction of the mixed liquor suspended solids. No influence on carbon or
nitrogen removal efficiency was found by elevated tin concentrations. Other process
control variables and metal concentrations were not correlated with reduced EBPR
efficiency on dates with elevated tin levels. The known high contributions of organotin
species toward total tin in activated sludges from other municipal WWTPs, and the high
toxicity of these compounds, suggests elevated organotin levels may be responsible for
the observed reduction in EBPR efficiency.
Key Words: Tin; Phosphorus uptake; Biological nutrient removal; Activated sludge;
Municipal wastewater treatment.

INTRODUCTION
Concerns over eutrophication in receiving waters have led to significant advancements in biological nutrient removal (BNR) at municipal wastewater
treatment plants (WWTPs) over the past several decades.[1,2] In particular,
phosphorus removal has received much attention because in many aquatic
systems it is the limiting nutrient for algal growth.[3−5] Of the two methods

Address correspondence to Sierra Rayne, Ecologica Environmental Consulting, Victoria,
Canada; E-mail: srayne@shaw.ca

536

Rayne, Carey, and Forest

for phosphorus removal in wastewaters—chemical and biological—the latter means is generally preferred because it does not generate metals-based
sludges that may be difficult to dispose of in economic and environmental terms. The ability to achieve enhanced biological phosphorus removal
(EBPR) in WWTPs using anaerobic release followed by aerobic “luxury” uptake was first recognized in the 1960s,[6,7] and since then there have been
numerous studies investigating the mechanisms and optimization of this
process.[8,9]
However, the wide variability in domestic wastewater quantity and quality
can reduce BNR in municipal WWTPs, and understanding the chemical species
and concentrations responsible for potential BNR inhibition is important from
both source and process control perspectives. Metals, both in inorganic or organic form, are known to affect the quality of wastewater treatment. Although
some of the raw influent metal loading is removed during primary sedimentation by precipitation and adsorption processes, the soluble, adsorbed, and suspended metal species may travel into secondary biological stages where they
can become closely associated with the biomass.[10−22] As a result of toxic metal
loadings, microbial community structures may be changed, with a resulting loss
of floc structures, microbial viability, and nutrient treatment efficiency.[23−27]
Despite the quantity of research into metals-based toxicity toward wastewater
microorganisms, the large number of different metal species and their differing
toxicities—including synergistic and antagonistic effects in mixed streams—
makes predictive assessments regarding real-world WWTP performance difficult. Thus, in conjunction with traditional laboratory and pilot-scale research
on metal-based toxicities in synthetic and controlled source wastewaters, empirical correlations from full-scale WWTPs are valuable to help guide and focus
research efforts.
In the Okanagan Valley of south central British Columbia, Canada, excessive nutrient loadings—particularly phosphorus—from wastewater and agricultural sources in the 1960s led to extensive limnologic studies and construction/upgrades of advanced WWTPs throughout the region in the 1970s
and 1980s.[28−36] In Kelowna, the largest city in the region, the municipal
WWTP (see Fig. 1 for schematic) was one of the first plants in North America specifically designed to include a sequence of anaerobic, anoxic, and aerobic
environments (the original Bardenpho configuration has been subsequently
changed to an A2 O process train) for the removal of carbon, nitrogen, and
phosphorus.[35−37] At present, the Kelowna wastewater system serves ∼1050
institutional users and an estimated population of ∼70,000 with a 10 ha WWTP
capable of treating 40 million liters per day.[38] Because of the importance for
consistent long-term BNR in the study region, we sought to investigate whether
individual metals, or combinations thereof, may be responsible for the variations in BNR efficiency observed over an 18-month period at the Kelowna
WWTP.

537

Figure 1: Treatment process flow diagram for the Kelowna WWTP.

538

Rayne, Carey, and Forest

MATERIALS AND METHODS
Samples were collected between April 30, 2002, and September 16, 2003. Sampling of raw influent, primary effluent, and final effluent was performed by
collecting 24-h composite samples in refrigerated samplers. Raw influent samples were taken every 15 min. Primary effluent and final effluent samples were
flow rated and collected every 5 to 15 min. All other samples were grab samples taken between 7:30 and 9:00 AM when influent flows were low. Samples
were filtered immediately and analyzed the same day, except for the composite
samples which were stored at 4◦ C for up to three days before analysis. Soluble
reactive phosphorus (SRP), chemical oxygen demand (COD), total solids (TS),
volatile suspended solids (VSS), and total suspended solids (TSS) were analyzed at the Kelowna WWTP laboratory. SRP was determined using a Technicon Auto-Analyzer II (Technicon Instrument Corporation; Tarrytown, NY, USA)
and Technicon method 94-79W. COD, TS, VSS, and TSS were determined using
standard methods 5220, 2540B, 2540E/2540G, and 2540D, respectively.[39] Total Kjeldahl nitrogen (TKN) analyses were performed by CARO Environmental
Laboratories (Kelowna, BC, Canada). Metals analyses were performed by Norwest Labs (Surrey, BC, Canada). Mercury was analyzed by cold vapor atomic
absorption using standard method 3112B.[39] All other metals were analyzed by
ICP-AES using U.S. EPA method 200.15.[40] All lab facilities are certified by the
Canadian Association for Environmental Analytical Laboratories (CAEAL) for
analyses performed as part of this study. Metal concentrations were determined
in the dissolved air flotation (DAF) sludge on a dry-weight basis and were converted to concentrations in the solids fraction of the mixed liquor suspended
solids (MLSS) by the following formula:

QCent
Sg × Sc × Q
WAS
CMLSS = CDAF ×
MLSSC × SVI
where CMLSS is the metal concentration in the solids fraction of the MLSS on
a mass per unit volume basis, CDAF is the metal concentration in the DAF
sludge on a mass per unit mass dry-weight basis, Sg is the specific gravity of
the DAF sludge, Sc is the percent water content of the DAF sludge expressed
as a decimal, QCent is the flow of concentrated sludge sent from the DAF unit to
the centrifuge for dewatering, QWAS is the total flow of waste activated sludge
sent from the aerobic stage of the BNR reactor to the DAF unit, MLSSC is the
concentration of MLSS on a mass per unit volume basis, and SVI is the sludge
volume index on a volume per unit mass basis.

RESULTS AND DISCUSSION
The performance of the Kelowna WWTP was investigated over a 504-day period between April 30, 2002, and September 16, 2003, with regard to BNR

Tin Inhibition of Biological Phosphorus Removal

efficiencies and metal concentrations. Over the study period, the advanced
WWTP had generally high levels of nutrient removal, with average removal
efficiencies of 96%, 93%, and 99% for chemical oxygen demand (COD), total
Kjeldahl nitrogen (TKN), and soluble reactive phosphorus (SRP), respectively
(Table 1). However, concentrations of total tin in the solids fraction of the mixed
liquor suspended solids (MLSS) were found to influence percent removal of SRP
with a sigmoidal relationship at concentrations ≥4 µg L−1 (Fig. 2). No corresponding influence on percent removals of TKN or COD in the bioreactor was
observed.
Between tin concentrations of 4 and 60 µg L−1 , SRP removal dropped from
∼99.5% to 98.7%, or a loss of ∼0.4 log units of removal efficiency. A tin concentration of 79 µg L−1 was observed on one of the study dates (April 30, 2002;
denoted with a square symbol in Fig. 2) with a corresponding low SRP removal
of 96.8%. On this day, operational secondary clarifiers were being changed and
could have resulted in some phosphorus release from the clarifier being taken
out of service. However, there did not appear to be a significant plant upset because of this event, as COD and TKN removal efficiencies appeared relatively
unaffected and MLSS and SVI values remained within the range obtained on
the other study dates. As well, there is no apparent explanation for a high tin
concentration on this date that can be attributed to clarifier changeover. If this
data point is valid, it represents a significant loss of SRP removal by up to 0.8 log
units when tin concentrations reach 80 µg L−1 in the MLSS. Selective interference with the microbial enzymes and inherently low diversity of the microbial
community responsible for EBPR—while not affecting removal efficiencies of
carbon—has been previously reported for inorganic forms of other metals (e.g.,
Zn, Cd, Pb).[23] While little other work appears to be done in this area of selective EBPR inhibition, the results of our work and that of Hao and Chang[23]
suggest future studies are warranted to identify the specific influence of, and
redefine toxic limits for, many compounds known to be present in wastewaters
but for which only carbon removal inhibition limits are known.
The effect of tin concentrations in the MLSS on EBPR is also demonstrated by the exponential relationship between WWTP effluent COD:SRP ratio
(COD:SRPeff ) and tin concentration (CSn ) at values >4 µg L−1 (COD:SRPeff =
2.957 × exp(−0.0701 × (CSn − 91.0)) + 337; R2 = 0.987) (Fig. 3). No influence
on COD:SRPeff was observed at concentrations <4 µg L−1 . This appears to be
further evidence that EBPR is inhibited at these higher tin concentrations,
whereas carbon removal is not, and that the aerobic stage microorganisms responsible for EPBR (e.g., Acinetobacter sp.)[1,41] are not able to sequester phosphorus to the extent observed at lower tin concentrations, resulting in the lower
effluent COD:SRPeff ratio.
The potential tin species responsible for this EBPR inhibition can be broadly
grouped into two classes: inorganic tin and organotins. In general, inorganic tin
species are considered to have low toxicity and bioavailability, especially when

539

540
6.83
3.71
6.61
10.9
1.81
28.5
877
281
616
186

±
±
±
±
±
±
±
±
±
±
a

0.04 to 7.57 ± 0.10
0.12
0.72
0.8
0.73
3.8
127
79
108
4

a Values given are the daily minimum and daily maximum.
bPercent changes in this column are due to the primary treatment
cPercent changes in this column are overall through the plant.

Note: All error bars are 95% confidence limits about the mean.

pH
3−
−1
PO3−
4 (mg PO4 − P·L )
Total P (mg P·L−1 )
NH3 (mg NH3 -N·L−1 )

−1
NO−
3 (mg NO3 -N·L )
TKN (mg N·L−1 )
Total COD (mg·L−1 )
Total BOD (mg·L−1 )
TSS (mg·L−1 )
Alkalinity (mg·L−1 )

Influent

process.

7.42
4.51
4.62
23.3
0.16
29.7
258
111
56.1
259

±
±
±
±
±
±
±
±
±
±
0.05
0.12 (+21.6 ± 6.0%)b
0.32 (−30.1 ± 10.1%)
0.8 (+113 ± 11%)
0.12 (−91.1 ± 7.3%)
2.0 (4.2 ± 10.0%)
19 (−70.6 ± 5.5%)
36 (−60.4 ± 10.7%)
10.1 (−90.9 ± 2.7%)
5 (+39.0 ± 5.3%)

Primary effluent

Table 1: Performance characteristics of the WWTP during the study period.

6.65
0.04
0.13
0.80
2.50
2.10
32.9
2.45
0.60
157

±
±
±
±
±
±
±
±
±
±

0.09 to 6.75 ± 0.09a
0.01 (−98.9 ± 0.3%)c
0.02 (−98.0 ± 0.5%)
0.46 (−92.6 ± 4.5%)
0.18 (+38.9 ± 47.6%)
0.52 (−92.6 ± 2.5%)
1.8 (−96.2 ± 0.7%)
0.92 (−99.1 ± 0.5%)
0.98 (−99.9 ± 0.3%)
5 (−20.8 ± 0.8%)

WWTP effluent

Tin Inhibition of Biological Phosphorus Removal

Figure 2: Influence of tin concentration in the solids fraction of the MLSS on removal
efficiencies of SRP, TKN, and COD in the BNR reactor.

541

542

Rayne, Carey, and Forest

Figure 3: Influence of tin concentration in the solids fraction of the MLSS on the effluent
COD:SRP ratio.

compared to other metals.[42,43] For example, the EC50 for inorganic tin toward
Daphnia magna is 22 mg L−1 , a concentration ∼200-fold higher than the corresponding EC50 for inorganic copper. Conversely, organotin species are known to
have a high acute toxicity and bioavailability.[43,44] Potential sources of organotins in municipal wastewater include the following: leaching from PVC pipes
and food packaging where they are used as stabilizers; as industrial catalysts
for producing polyurethane foams, plasticizers, lubricants, heat transfer fluids, and silicone rubbers; as antimicrobials, molluscicides, acaricides, and agricultural biocides; and as wood preservatives and glass coatings.[45−51] A large
number of organotin compounds are used commercially, and they are generally
represented by the formula Rn SnX4−n with n = 0–3, the R groups are typically
alkyl (e.g., methyl, butyl, octyl) or aryl moieties (e.g., phenyl) covalently bound
to the central tin atom, and X is the counterion (e.g., −OH, −OR ).[43,48,52] Furthermore, the organotins are more bioaccumulatory than inorganic tin, with log
Kow values up to 4[53,54] and log BCFs up to 4.5 in algae,[55] and are known to
preferentially accumulate in the organic solids fraction of wastewaters.[56,57] As
well, the trialkyltins in general are refractory to biological treatment and are
not substantially degraded under either aerobic or anaerobic conditions.[47] —
which is potentially not surprising given their high toxicities to aquatic life—
thereby allowing higher concentrations to build up in the recycled activated
sludge. For these reasons, normalizing organotin concentrations (and those of
other hydrophobic metal species) to the solids fraction of the MLSS (i.e., mass
per unit volume of settled MLSS, as in the current study—see “Materials and
Methods” section for details) may help elucidate and identify the potential toxic
influence of these compounds in wastewaters by accounting for the inherent
variability of MLSS concentrations and thereby investigating the contaminant
concentrations that microorganisms are actually exposed to.

Tin Inhibition of Biological Phosphorus Removal

Because of the rigorous and expensive sample collection and analytical
methods required to speciate and quantify the many different toxicologically
relevant organotin compounds in wastewaters,[58,59] it is unlikely WWTP staff
will have access to these resources for routine monitoring. However, previous
reports have shown organotin contributions to total tin in activated sludge
can range from ∼50% to >90%,[57,60] and thus, total tin (with the known nontoxicity of inorganic tin) may be a suitable surrogate for organotin concentrations in preliminary assessments of metals-based toxicity toward wastewater
treatment. With the relatively wide range of acute and chronic toxicities depending on the nature of organic substituents,[43,48,52] on the other hand, more
comprehensive studies into tin-based WWTP inhibition will require sampling
and analysis methods with speciation capacity. While we do not have the analytical data necessary to confidently speculate on the likely tin compound(s)
responsible for the observed EBPR inhibition, we note that acute and chronic
lethal concentrations in the range of ∼ 0.1 − 1 µg Sn L−1 have been reported[46]
for the most toxic member of the organotins, tributyltin, with decreased growth
and other harmful effects observed at concentrations as low as 0.002 µg Sn
L−1 .[44] Previous reports have observed methyl-, butyl-, and phenyl- substituted organotins in wastewater sludges with total concentrations in the range
from ∼ 0.1 − 10µg Sn g−1 dry-weight (dw).[45,47,56,57,60,61] The dry-weight total
tin concentrations we observed in the DAF sludge (0.25–9.8 µg Sn g−1 dw)
prior to our MLSS normalization process are in this range, thereby suggesting
the need for other municipal WWTPs with BNR removal to examine potential relationships between tin concentrations (either with total tin acting as
a surrogate for organotins, or preferably with direct analysis for organotins)
in the bioreactor MLSS and nutrient removal efficiencies. Although the MLSS
concentrations for total tin we observed in the Kelowna WWTP are below the
range previously shown for organotin (specifically, tributyltin oxide) inhibition
of the activated sludge process,[62] this toxic threshold range was determined
using COD/TOC/BOD removal and oxygen uptake end points, which may not be
sensitive to any disruptions of the EBPR microbial community. As such, further
work appears necessary to determine the end points for EBPR (and possibly
nitrification/denitrification) disruption by the various organotin species so that
effective source and process control regimes can be developed for the optimization of BNR at municipal WWTPs.
Because many other factors can affect EBPR in advanced WWTPs, we
sought to investigate whether other process variables may help explain the
low SRP removal on dates when tin concentrations in the MLSS were elevated.
No correlations were observed between SRP removal and any of the following variables known to influence EBPR: aerobic stage MLSS concentration,
sludge age, temperature in the aerobic stage, pH in the aerobic stage, sludge
volume index in the aerobic MLSS, or the BNR reactor influent COD:SRP ratio
(COD:SRPin ; i.e., the primary effluent COD:SRP ratio) (Fig. 4). Values for SRP

543


Related documents


PDF Document j environ sci health 40 2005 535 551
PDF Document ijeas0403014
PDF Document j environ eng sci 4 2005 353 367
PDF Document 17n19 ijaet0319377 v7 iss1 143 152
PDF Document world water treatment equipment market
PDF Document wetlands wastewatertreatment


Related keywords