Download (Initial page layout)

Ozone and Aquifer Recharge and Recovery Hybrid for Attenuation of Bulk and
Trace Organic Contaminants
Min K. Yoon* and Gary L. Amy
King Abdullah University of Science and Technology (KAUST), 4700 KAUST. Thuwal,
23955, Saudi Arabia
(E-mail: [email protected]; [email protected])
*Corresponding author
Abstract
The attenuation of bulk organics and trace organic contaminants (TOrCs) during
an aquifer recharge recovery (ARR) coupled ozone (O3) treatment hybrid was
evaluated in soil-batch and bench-scale ozonation simulations for wastewater
reclamation and reuse. In this work, the role of biodegradation in the removal of
bulk organics and TOrCs was investigated during soil passage alone (i.e., ARR),
after ozonation (i.e., ARR-O3) and prior to ozonation (i.e., O3-ARR). ARR
simulations alone and ARR coupled ozonation hybrids were compared using
percent removal of bulk organic characters and a representative subset of TOrCs
(i.e. contaminants of emerging concern). In oxic ARR simulations alone, higher
molecular weight polysaccharides and proteins (i.e., biopolymers) were easily
removed while lower molecular weight humic substances and aromatic organic
matter were not easily removed over a short-term ARR operation. In ARR
coupled ozonation hybrids, removal of bulk organics and TOrCs was effective and
rapid compared to ARR alone. A higher reduction of effluent-derived organic
matter, including aromatic organic matter and humic substances, was observed in
the ARR-O3 hybrid followed by the O3-ARR hybrid. An enhanced attenuation of
recalcitrant TOrCs was determined by a slight increase of ozone dose (O3: DOC =
1) and/or a post-ozone treatment (i.e. ARR-O3). In addition, the carcinogenic
wastewater by-product, N-Nitrosodimethylamine (NDMA), was eliminated below
the method reporting limit (< 5ng/L) in both ARR alone and ARR hybrids.
Keywords
Wastewater treatment, Aquifer recharge and recovery, Ozonation, Trace organic
contaminant
INTRODUCTION
Elimination processes during soil passage, via both oxic and anoxic biodegradation as well as
sorption, provide an effective barrier for attenuation of effluent-derived organic carbon. However,
some trace organic contaminants (TOrCs) are not easily removed or require a long residence time or
a unique redox condition. Previous findings on TOrC removal in ARR are summarized as follows:
1) removal of degradable TOrCs (with the exception of diclofenac and propyphenazone) under
aerobic porous infiltration condition is equal or better compared to anoxic porous infiltration
conditions, 2) biodegradable dissolved organic carbon (BDOC) can serve as a co-substrate in the
co-metabolic transformation of TOrCs, 3) concentration and character of bulk organic carbon
present in effluent affects the degradation efficiency for TOrCs during the recharge operations, 4)
redox conditions and degradation potential as a function of temperature are key factors in
influencing performance (Hoppe-Jones et al. 2010; Maeng et al. 2010).
In the work reported herein, an oxidation (ozonation) process was tested prior to, and after, soil
passage for the removal of dissolved organic carbon (DOC) and recalcitrant TOrCs and for effective
drinking water treatment. Ozonation can convert refractory DOC such as humics/fulvics into more
biodegradable components. Studies have shown that oxidation improves the biodegradability of
poorly degradable organic compounds in ARR systems because of co-metabolism. In the case of
using ozone as a pre-treatment, it can 1) reduce TOrCs loading in a subsequent ARR by eliminating
the precursors and 2) make the matrix more amenable to biotransformation (i.e. creation of BDOC),
thereby achieving further TOrC reductions. In a post-ozonation strategy, ARR will remove a
significant amount of effluent organic matter (EfOM) and reduce the concentrations of
contaminants and specific TOrCs. This will reduce the oxidant scavenging potential of the matrix
and increase the overall efficiency of the downstream ozone process. Thus, ozonation can be an
effective treatment strategy for both sequences with ARR.
Selected TOrCs were used as representative indicators for oxidation and aquifer recharge recovery
(ARR) conditions. Due to the low concentrations of these TOrCs in typical wastewater effluent,
trace analysis of the TOrCs was required, and performed using solid phase extraction (SPE) and
liquid chromatography-tandem mass spectrometry (LC-MS/MS) and gas chromatography-tandem
mass spectrometry (GC-MS/MS). The attenuations of bulk organics and TOrCs in different hybrids
of O3 and ARR (O3-ARR) vs. ARR and O3 (ARR-O3) were examined to understand parameters and
conditions controlling their removal as an effort to assess ARR wastewater reuse strategies. Both
configurations were considered because the O3-ARR approach could provide a barrier for
eliminating transformation products (oxidation by-products) and the latter ARR-O3 approach could
result in a lower oxidant demand and lead to lower costs. The amount and character of BDOC were
important factors influencing the fate of TOrCs during ARR. BDOC fractions and their rates (slow
vs. rapid) were compared between ARR and different ARR hybrids in this work. Water quality
parameters were evaluated using DOC and specific ultraviolet absorbance (SUVA) augmented by
further characterization with innovative EfOM characterization techniques of liquid
chromatography - organic carbon detection and - organic nitrogen detection (LC-OCD-OND) and
3D fluorescence excitation-emission matrix (F-EEM). The objective of this study was to determine
the most effective combination of ozone and aquifer recharge recovery as a treatment hybrid.
METHODS
Experimental Setups
In bench-scale tests, Milli-Q water was ozonated in 2 L glass flasks by bubbling ozone from oxygen
gas produced by an LAB2B ozone generator (Ozonia, Degremont Technologies). The dissolved
ozone in the stock solution was measured periodically at a wavelength of 258 nm (Δ ε = 2950 M-1
cm-1) (Rakness et al. 1996) and was confirmed by the indigo trisulfonate method (Bader and Hoigné
1981). A specific volume of the ozonated water was transferred to batch reactors that contained
secondary effluent from the Al-Ruwais Wastewater Treatment Plant (ARWWTP; Jeddah, Saudi
Arabia) at room temperature (21°C). Applied (transferred) ozone doses took into account dilution
of the stock solution with tested water. The ARWWTP wastewater (after primary treatment) was
treated in activated sludge aeration tanks. The typical water quality of the ARWWTP secondary
effluent is summarized in Table 1. Ozone to DOC (O3: DOC) ratios of 0.5 and 1 mg/mg were
selected because previous studies have demonstrated efficacy in oxidizing a wide range of
contaminants over this range (Sundaram 2009; Gerrity et al. 2011). For the soil component, the
batch sands were prepared in 1 L glass bottles using 150g (dry weight) of washed silica sand (grain
sizes 0.5 - 1 mm; bulk density 1.5 - 1.6 g/cm3; sand porosity 40%). The sands were then bioacclimated for 8 weeks with secondary effluent from the ARWWTP. During this period, DOC and
UV254 absorbance were monitored until steady state was achieved. After the acclimation period, the
soil batch reactors were fed with secondary effluent or ozonated secondary effluent until a range of
retention times had been achieved (3 hrs., 1, 3, 5, 8, 12 days). The reactors were kept on a shaker
table to ensure aerobic conditions. Water samples were then collected to assess biodegradability of
bulk organics and TOrCs or post-ozonated. This experimental setup allowed for evaluations of bulk
organic and TOrC attenuation with oxic ARR alone, O3-ARR and ARR-O3. To differentiate
microbial biodegradation and sorption, abiotic batch reactors were also prepared with sodium azide
at 20 mM (Maeng 2010) and monitored over 5 days. Sorption isotherms were generated using a
linear regression analysis with TOrC concentrations ranging from 0.5 to 10 μg/L. An overview of
experimental approach is provided in Figure 1.
Table 1. ARWWTP Secondary Effluent Water Quality
Water Quality Parameters
pH
Conductivity (µS/cm)
Dissolved Organic Carbon (mg/L)
UV254 Absorbance (cm-1)
Turbidity (NTU)
Fluoride (mg/L)
Chloride (mg/L)
Nitrite (mg/L)
Bromide (mg/L)
Nitrate (mg/L)
Phosphate (mg/L)
Sulfate (mg/L)
7.1
3300
5.9
0.148
0.47
0.21
746.24
< MRL
1.44
28.24
11.28
411.29
Figure 1. Simplified Schematic for Experimental ARR Scenarios
Analysis of Bulk Organics
DOC was measured using a Shimadzu TOC analyzer and BDOC (operationally defined as the
change in DOC by biodegradation under the conditions investigated) was measured weekly during
the bio-acclimation period (8 weeks). In addition, UV254 and SUVA were monitored as indicators
of aromaticity and humic substances using a Shimadzu UV-Vis Spectrophotometer (UV-2550).
LC-OCD-OND-UV254nm based on the Gräntzel thin-film UV reactor (DOC-Labor Dr. Huber,
Germany) was used to evaluate major molecular-size fractions of natural organic matter such as
slowly degrading biopolymers vs. rapidly degrading low molecular weight acids and neutrals during
treatment processes. Protein- (i.e., soluble microbial product-like), and humic- and fulvic-like
EfOM in water samples were characterized using fluorescence excitation-emission matrix (F-EEM)
using a FlouroMax4 (Horiba, USA). The spectrofluorometer included a 150-W xenon lamp. Slit
width for both excitation and emission monochromators was set to 5 nm and a 5 nm increment was
used. Analyses were performed in a 1-cm quartz cuvette at room temperature (21 °C). Samples
were analyzed in signal/reference mode where the fluorescence emission intensity is normalized to
the intensity of the lamp at the particular excitation wavelength applied.
Analysis of Trace Organic Contaminants (TOrCs)
A representative set of TOrCs were selected based on several factors, including structural and
chemical properties (e.g. functional groups, polarity, aromaticity, etc.), use classes (e.g. antibiotic,
fragrance, pharmaceutical, etc.), high frequency of environmental occurrence (Kolpin et al. 2002;
Benotti et al. 2008), resistance to natural treatment processes (e.g. biodegradation, photolysis, etc.)
and engineering treatment processes (e.g. adsorption, oxidation, etc.) (Ternes et al. 2002; Huber et
al. 2003; Westerhoff et al. 2005), and amenability to existing analytical methods. Each of
seventeen target TOrCs and the corresponding classification, CAS numbers, physiochemical
properties (i.e., molecular weight, water solubility, pKa and log P), ozone rate constant and method
reporting limit are listed in Table 2. Additionally, tris (1-chloro-2-propyl) phosphate (TCPP) was
added because it was used instead of tris (2-chloroethyl) phosphate (TCEP) in the abiotic
experiment.
All samples were collected in amber glass bottles preserved with sodium azide (1 g/L) and ascorbic
acid (50 mg/L). The samples were processed with solid phase extraction and analyzed with GCMS/MS (for NDMA analysis) and LC-MS/MS with isotope dilution according to the previously
published studies (Vanderford et al. 2003; Trenholm et al. 2006; Vanderford and Snyder 2006).
Table 2. Target TOrCs with Their CAS No. and Physicochemical Properties
Compound
Use Class
CAS No.
29122-68-7
1912-24-9
80-05-7
298-46-4
134-62-3
15307-86-5
25812-30-0
15687-27-1
57-53-4
81-14-1
22204-53-1
57-41-0
125-33-7
Molecular
Weight
(g/mol)
266.3
215.7
228.3
236.3
191.3
296.2
250.3
206.3
218.3
294.2
230.3
252.3
218.3
Water
Solubility
(mg/L)
1.33 x 104
27
120
18
912
2.37
28
21
3400
5
15.9
71
500
Atenolol
Atrazine
Bisphenol A
Carbamazepine
DEET
Diclofenac
Gemfibrozil
Ibuprofen
Meprobamate
Musk Ketone
Naproxen
Phenytoin
Primidone
Beta-Blocker
Herbicide
Plasticizer
Anticonvulsant
Pesticide
NSAID
Lipid Regulator
NSAID
Anti-Anxiety
Fragrance
NSAID
Anticonvulsant
Anticonvulsant
Sulfamethoxazole
TCEP
TCPPb
Triclosan
Trimethoprim
Antibiotic
Flame Retardant
Flame Retardant
Anti-Microbial
Antibiotic
723-46-6
115-96-8
13674-84-5
3380-34-5
738-70-5
253.3
285.5
327.6
289.5
290.3
610
7000
1200
10
400
pKa
logP (exp)a
kO3 (M-1 s1 c*
)
kOH (M-1
s-1)d*
MRLe
(ng/L)
9.2
1.7
9.6, 10.2
13.9
0.7 (est.)
4.15
4.42
4.91
14 (est.)
N/A
4.15
8.3
11.1 (est.),
12.2 (est.)
5.5
N/A
N/A
7.98 (est.)
7.12
0.16
2.61
3.32
2.63 (est.)
2.18
4.51
3.80 (est.)
3.97
0.7
3.82 (est.)
3.18
2.47
0.91
2 x103
6
7 x105
3 x105
< 10
1x106
2x104
10
<1
<1
2 x105
< 10
< 10
8 x 109
3 x 109
1 x 1010
9 x 109
5 x 109
8 x 109
1 x 1010
7 x 109
4 x 109
1 x 109
1 x 1010
6 x 109
7 x 109
25
10
50
10
25
25
10
25
10
100
25
10
10
0.89
1.44
2.59
4.76
0.91
3 x106
<1
N/A
4 x107
3 x105
6 x 109
7 x 108
N/A
1 x 1010
7 x 109
25
200
200
25
10
Note: All physicochemical properties obtained from Snyder et al., 2009
a
log P (exp) : Other than indicated explicitly, all are experimental values obtained from ALOGPS 2.1 (PHYSPROP database)
b
The compound, TCPP was used instead of TCEP in abiotic experiment.
c*
d*
Second order rate constant for ozone
Second order rate constant for hydroxyl radical
Method reporting limit (MDL) : Based on 3-5 times the calculated MDL in DI, n=12 (TCEP based on 5 times average background)
* Sources: (Ternes et al. 2002; Huber et al. 2003; Westerhoff et al. 2005; Gerrity et al. 2011)
e
RESULTS AND DISCUSSION
Bulk Organic Characterization
A net reduction of DOC was observed in all systems (i.e., O3-ARR, ARR- O3 and ARR alone) and
the highest reduction was achieved for O3-ARR (29%) followed by ARR-O3 (24%), both at O3:
DOC=1, O3-ARR (23%) at O3: DOC = 0.5, ARR alone (17%), and then ARR-O3 (16%) at O3: DOC
= 0.5. The UV254 absorbance and specific UV254 absorbance (SUVA = UV254 absorbance/DOC)
data indicated that the O3-ARR and ARR-O3 were superior to ARR alone. For UV254 absorbance,
ARR-O3 provided 60% and 72% reduction at low (O3: DOC = 0.5) and high (O3: DOC=1) O3 doses,
respectively, and O3-ARR provided 49% and 57% at low and high O3 doses, respectively. However
only a 28% reduction in UV254 absorbance was observed for ARR alone. The summary of percent
(%) removal of DOC and UV254 is provided in Table 3. Overall, DOC removal was higher with
pre-ozonation due to the increase of BDOC during oxidation, and UV254 removal was higher in
post-ozonation due to the more efficient destruction of aromatic humic substances under conditions
of lower ozone demand.
Table 3. Percent (%) DOC and UV254 Removal in ARR Combinations.
ARRs
O3: DOC
% DOC Removal
% UV254 Removal
ARR alone
0
17
28
O3-ARR
0.5
23
49
O3-ARR
1
29
57
ARR-O3
0.5
16
60
ARR-O3
1
24
72
This result shows that the oxidation (i.e., ozone) transforms the aromaticity of the dissolved organic
matter as well as larger humic substances. The SUVA values of various ARR effluents were
observed over 12 days of residence time as shown in Figure 2. The pre- (O3-ARR) and post- Ozone
(ARR-O3) hybrids showed lower SUVA values than ARR alone while ARR-O3 hybrids yielded the
lowest SUVA values. The difference in SUVA was about 1 to 1.5 (Lmg-1m-1) between ARR alone
and ARR hybrids. Proportionally, SUVA went down noticeably after eight day of retention time
due to a reduction of aromatic humic substance. This may be attributable to refractory DOC of
humic/fulvic substance taking longer to degrade than BDOC and/or a potential transformation of
humic substances.
Figure 2. Specific Ultraviolet Absorbance (SUVA) of Various ARR Effluents
LC-OCD-OND-UV254nm was used to characterize the attenuation of bulk organics with respect to
their molecular weights ranging from ≈0.1 to ≈20 kDa. Previous work has established categorizing
effluent organic matter into major fractions of different sizes and chemical fractions: biopolymers,
humic substances, building blocks, low molecular weight organic acids and neutrals, and
hydrophobic organic carbon (Huber et al. 2011). With an increased ozone dose, the organic carbon
concentration in the biopolymers decreased while the organic carbon concentration in the humics,
building blocks, and low-molecular-weight humics and acids fractions increased. Our data agreed
with previous studies that large, hydrophobic organic content is transformed into smaller
hydrophilic organic content during ozonation (data not shown). In comparison with ARR effluent
alone, biopolymers were decreased while humics and building blocks were decreased slightly but a
refractory, low-molecular-weight acids component remained fairly constant (Drewes et al. 2006).
3D Fluorescence excitation-emission matrices (F-EEMs) provide an illustrative and sensitive
representation of EfOM that could not be captured by other organic parameters such as DOC and
UV254nm. F-EEM spectra can be divided into three main regions associated with specific organic
fractions indicated in Figure. 3-A. Region 1 represents protein-like fluorescence whereas Region 2
represents fulvic-like fluorescence and Region 3 humic-like fluorescence (Henderson et al. 2009).
While ARR alone (Figure 3-B) provided a minimal reduction in both regions of fulvic- and humiclike fluorescence, a substantial decrease of fluorescence in all regions was observed in O3-ARR and
ARR- O3 hybrids. Decreases in fluorescence are associated with improvements in water quality;
ARR- O3 hybrids shown in Figure 3-E and F, showed even further reduction (close to fluorescence
of in Milli-Q) compared to O3-ARR hybrids, shown in Figure 3-C and D. Similar to the reduction in
UV254 absorbance and SUVA, ARR-O3 was superior to O3-ARR.
Figure 3. Fluorescence Excitation-Emission Matrices (F-EEMs) for: (A) Secondary Effluent Where
1 = Protein-like, 2 = Fulvic-like, 3 = Humic-like Fluorescence, (B) ARR Effluent, (C) O3-ARR
Effluent (O3: DOC=0.5), (D) O3-ARR Effluent (O3: DOC=1.0), (E) ARR-O3 Effluent (O3:
DOC=0.5), and (F) ARR-O3 Effluent (O3: DOC=1.0)
F-EEMs can also quantitatively compare water quality changes in EfOM and associated treatment
efficacy. The key peaks and their percent reductions in fluorescence intensity were evaluated as
shown in Table 4. While protein-like peaks (e.g., soluble microbial products or biopolymers) were
easily degraded, fulvic- and humic-like peaks were not easily degraded in ARR alone. With the
benefit of ozone, fulvic- and humic-like peaks were removed in all ARR hybrids. Overall, higher
reductions in all three key peaks were observed in ARR-O3 hybrids over O3-ARR hybrids in which
(%) reductions were 84% or higher at low ozone dose (O3: DOC=0.5) and 90% or higher at high
ozone dose (O3: DOC = 1).
Table 4. Key Peaks and Their (%) Reductions in Fluorescence Intensity
ARRs O3: DOC (1) Protein-like (2) Fulvic-like (3) Humic-like
ARR
0.0
47 %
9%
3%
O3-ARR
0.5
75 %
51 %
51 %
O3-ARR
1.0
80 %
61 %
60 %
ARR-O3
0.5
89 %
84 %
88 %
ARR-O3
1.0
93 %
90 %
94 %
Trace Organic Contaminant Attenuation
The selected TOrCs were categorized as easily biodegradable (i.e., Bisphenol A, Diclofenac,
gemfibrozil, ibuprofen, and naproxen), moderately biodegradable (i.e., atenolol, DEET, triclosan,
and trimethoprim), and poorly biodegradable (i.e., atrazine, carbamazepine, phenytoin, primidone,
meprobamate, sulfamethoxazole, and TCEP), based on their removals under ARR retention times of
3 hrs., 5 days (i.e., 120 hrs.), and 12 days (i.e., 288 hrs.) as shown in Figure 4. This result indicates
that the attenuation of TOrCs varies from hours to days, and therefore, specific TOrCs removal can
be targeted based upon the residence time of ARR. TOrC oxidation by ozone alone was then
monitored for the ARWWTP secondary effluent with both ozone doses, O3 to DOC ratios of 0.5
and 1. Figure 5. illustrates the removal of TOrCs by ozonation alone. The results have shown that
TOrCs with their O3 rate constants, kO3 (M-1s-1), higher than 103 were readily removable (> 80%)
while TOrCs with their O3 rate constants, kO3 (M-1s-1), equal or lower than 10 were not. As a result,
the latter group of TOrCs required a higher ozone dose for a further removal (> 80%). Overall, this
result is generally similar with previous bench-scale oxidation data of TOrC for typical secondary
effluent. While ozonation provided the fast elimination of TOrCs, biodegradation of poorly
degradable TOrCs via ARR requires a long residence time from weeks to months.
RT = 3 hrs
RT = 120 hrs
RT = 288 hrs
1.20
1.00
C/C0
0.80
0.60
0.40
0.20
I. Easily Biodegradable
II. Moderately biodegradable
Figure 4. TOrC Attenuation by ARR Alone
zi
rb
ne
am
az
ep
M
in
ep
e
ro
ba
m
at
e
P
tra
A
Ca
TC
E
yt
oi
n
Pr
im
Su
id
lfa
on
m
e
et
ho
xa
zo
le
Ph
en
T
EE
D
et
ho
pr
im
Tr
ic
lo
sa
n
A
te
no
lo
l
Tr
im
Bi
sp
he
no
lA
D
ic
lo
fe
na
G
c
em
fib
ro
zi
l
Ib
up
ro
fe
n
N
ap
ro
xe
n
0.00
III. Poorly biodegradable
03:DOC=0.5
O3:DOC=1
1.20
kO3 > 103 M-1s-1
kO3 ≤ 10 M-1s-1
1.00
C/Co
0.80
0.60
0.40
0.20
Tr
im
Ca eth
rb opr
am im
az
G epi
Su em ne
lfa
f
m ibro
et
ho zil
xa
D zol
ic
lo e
fe
N nac
ap
ro
x
A en
t
Bi eno
sp lol
he
no
Ib l A
up
ro
Ph fen
en
y
Pr toin
im
id
on
e
M DE
ep
ro ET
ba
m
a
A te
tra
zi
ne
TC
EP
0.00
Figure 5. TOrC Attenuation by Ozone Alone
To investigate the relative contributions of oxidation and biodegradation/sorption to TOrC removal
during oxidation-ARR hybrid processes, target TOrC removal by ozonation and/or ARR was
compared. Figure 6 shows the percent (%) removal of TOrCs by ARR (RT = 288 hrs.) and
ozonation (O3: DOC = 0.5 and1). The additional oxidation provided by an increase in O3: DOC
from 0.5 to 1.0 is also illustrated with arrows. This summary figure infers that a majority of the
TOrCs can be easily removed by ozone alone or a combination of ozone and ARR. Some of the
compounds were resistant to ARR but were susceptible to ozone (carbamazepine,
sulfamethoxazole, primidone, and phenytoin), but only DEET proved to be resistant to ozone but
moderately susceptible to ARR. Several compounds demonstrated resistance to both ozone and
ARR (meprobamate, atrazine, and TCEP). In these treatment hybrids, both ozone doses of O3: DOC
= 0.5 vs. O3: DOC = 1 and the short residence times of ARR were strategically combined for
potential synergies. Figure 7 compares ARR alone and ARR hybrids for the attenuation of
recalcitrant TOrCs (kO3 ≤ 10) and illustrates the impacts of varying ozone dose and hybrid process.
All recalcitrant TOrCs showed a further attenuation either or both by the increased ozone dose and
post-ozone treatment except for TCEP. TCEP, a flame retardant, was difficult to remove by either
the relatively low degrees of ozonation (O3: DOC ≤ 1) or the short term ARR operation (RT = 12
days). Although previous literature suggests that pre-ozonation provides greater potential for
cometabolism, post-ozonation (i.e., ARR-O3) achieved lower effluent concentrations for the
recalcitrant TOrCs, except for TCEP. Therefore, implementing some form of biological filtration
prior to ozonation appears to remove the bioamenable compounds that react rapidly with ozone,
thereby reducing oxidant scavenging (demand). This increases the efficacy of downstream
ozonation despite similar O3: DOC ratios in both hybrids.
O3: DOC = 0.5
% Removal by O3
O3: DOC = 1.0
I
II
II
I
IV
% Removal by ARR
Figure 6. TOrC Removal by ARR (RT = 12 days) and Ozone (O3: DOC = 0.5 or 1)
Note: I: O3; II: O3 or ARR; III: ARR; IV: neither
Figure 7. Attenuation of Recalcitrant TOrCs (kO3 ≤ 10) with Different Treatment Hybrids
TOrCs Behavoir during Abiotic Conditions: Sorption Isotherms
Solid-water distribution coefficients (Kd) were calculated based on slopes of linear sorption
isotherms over a defined concentration range. Low TOrC concentrations (spiked) in the 0.5-10 μg/L
ranges were used to simulate the current ARR process as well as raw municipal wastewaters for
many TOrCs. The majority of the isotherms followed a linear regression with correlation
coefficients of R2 ≥ 0.95 as shown Table 5. For low sorbing compounds, the data did not fit a linear
isotherm and a single point Kd was determined. The results shown that all of the negatively charged
compounds, such as diclofenac, ibuprofen, naproxen, phenytoin, sulfamethoxazole, and triclosan,
had low Kd values (Kd < 1) except for gemfibrozil, (Kd = 2.81; R2 = 0.98). Two positively charged
compounds, atenolol and trimethoprim yielded relatively lower Kd values than expected, 1.09 and
1.76, respectively. This may be due to their low log P values (< 1), which indicate they are more
hydrophilic than other positively charged compounds. Overall, Kd values measured in this work
were generally 2 to 3 order of magnitudes lower than previously reported Kd values for the sorption
potential to wastewater primary and activated sludge solids. However, the abiotic behaviour of
these TOrCs seems to follow the relative trends of (higher vs. lower) sorption reported in literature
for the sludge solids (Stevens-Garmon et al. 2011). This is perhaps related to the silica sand’s (SiO2
≥ 96.3%) negatively charged surface, which results in a stronger association occurs between its
surface and a positively charged compounds, than with a neutral or a negatively charged compound
(Schwarzenbach et al. 2005). Interestingly, all of the easily biodegradable TOrCs (i.e. bisphenol A,
diclofenac, gemfibrozil, ibuprofen and naproxen) in ARR, determined earlier, were the compounds
with low Kd values, except for gemfibrozil. Thus, the dominant mechanism of TOrC attenuation for
the ARR appears to be biodegradation with a possible exception of gemfibrozil. This is particularly
important considering that sorption sites will eventually be exhausted in ARR applications after
continuous exposure to treated effluent.
Table 5. Measured Kd Values Based on Sorption Isotherms
Compound
Atenolol
Atrazine
Kd (L/kg) R2
1.09
1.00
2.57
0.99
BPA
0.67
N/A
Carbamazepine
1.11
1.00
DEET
3.23
0.97
Diclofenac
0.87
N/A
Gemfibrozil
2.81
0.98
Ibuprofen
0.66
N/A
Musk Ketone
0.99
N/A
Naproxen
0.71
N/A
Phenytoin
0.67
1.00
Primidone
1.11
1.00
Sulfamethoxazole 0.68
0.99
TCPP
2.41
0.99
Triclosan
0.74
0.95
Trimethoprim
1.76
0.99
* Kd values in italics are based on single point calculations thus R2 values are not applicable.
**Meprobamate was excluded because its analytical standard was unavailable.
NDMA Formation and Destruction
N-Nitrosodimethylamine (NDMA) is a mutagenic and carcinogenic wastewater by-product that has
been found in wastewater effluent even after disinfection/oxidation processes. Due to its
demonstrated carcinogenicity, the state of California, USA has recently established a notification
limit of 10 ng/L and a response limit of 300 ng/L. Reported concentrations are typically in a range
of 5 to 25 ng/L in secondary effluent and higher in primary effluent; with a median of 26 ng/L in the
U.S. (Pehlivanoglu-Mantas and Sedlak 2006) and a median of 5 ng/L in Switzerland (Krauss et al.
2009). In our wastewater effluent (Jeddah, Saudi Arabia) with an ambient level of 7 ng/L of
NDMA, a slight formation of NDMA was observed with an increase of a total NDMA of 12 ng/L
after ozonation (O3: DOC = 1). However, the ARR experiments indicated that NDMA was
eliminated in all ARR simulations and its hybrids below its method reporting limit (i.e., < 5 ng/L).
The MRL was achieved after 12 days of soil treatment even when NDMA was spiked at ~500 ng/L.
This result is also consistent with a previous field study (Zhou et al. 2009). Therefore, ARR is an
effective treatment barrier against NDMA and may eliminate the need for UV photolysis or UVAOP in indirect portable reuse (IPR) applications.
CONCLUSIONS
ARR is a robust and multi-objective process capable of achieving substantial removal of many bulk
and trace organic contaminants. The more bioamenable compounds are removed rapidly during soil
passage, but more recalcitrant compounds may require long residence times or unique redox
conditions. Ozone and oxidation process can eliminate many TOrCs, however ozone generally
achieves incomplete oxidation (i.e. limited mineralization), is unable to destroy some compounds
(e.g., TCEP), and only transforms other compounds into simpler oxidation by-products. Therefore,
ARR is an effective secondary barrier to protect human health against these persistent contaminants.
The ARR and isotherm experiments in this study indicated that biodegradation was more significant
than sorption during ARR, and these collective processes may also be supplemented with dilution
(mixing) in applications that limit recycled water contributions. Therefore, there is potential for
significant reductions in TOrC concentrations during ARR with a variety of mechanisms. With
respect to the California Department of Public Health regulations for spreading applications, the
biodegradation and sorption pathways must achieve at least 90% removal of indicator compounds
(e.g., selected TOrCs). Published research indicates that biodegradation of target contaminants is
often more efficient and complete in natural ecosystems comprised of diverse microbial
communities. There are often complex interactions, feedback, and “communication” between the
microbial species that promote degradation pathways. Also, it is often unnecessary to supplement
the system with oxygen, organic substrates, or specific microbial species, although exceptions do
exist. Thus, nonspecific degradation of TOrC mixtures, which is important in IPR applications, can
often be achieved with ambient conditions.
For wastewater reclamation and reuse strategies, ARR and oxidation hybrids of ozonation prior to
or after a short-term ARR operation showed positive outcomes. Both O3-ARR (i.e., pre-ozonation)
and ARR-O3 (i.e., post-ozonation) were more effective than ARR alone in addressing a variety of
bulk organics and TOrCs. O3-ARR was superior for overall removal of DOC, while ARR-O3 was
more effective in reducing UV absorbance, fluorescence, and TOrC concentrations. Finally, ARR
proved to be an effective attenuation strategy for ambient NDMA and any NDMA that formed
during ozonation. These results also indicate that the reductions in NDMA during aquifer recharge
operation may eliminate the need for UV photolysis or UV-AOP in typical groundwater
replenishment treatment schemes (i.e., full advanced treatment). Future research should identify
which variables are most critical in terms of TOrC attenuation during ARR or whether
travel/storage time alone essentially controls the process. Furthermore, future research should
identify strategies to develop engineered aquifer recharge operation to replace natural systems and
ultimately eliminate the environmental buffer. This would facilitate the transition from indirect
potable reuse to direct potable reuse.
REFERENCES
Bader, H. and J. Hoigné (1981). "Determination of ozone in water by the indigo method." Water
Research 15(4): 449-456.
Benotti, M. J., R. A. Trenholm, et al. (2008). "Pharmaceuticals and Endocrine Disrupting
Compounds in U.S. Drinking Water." Environmental Science & Technology 43(3): 597-603.
Drewes, J. E., D. M. Quanrud, et al. (2006). Character of Organic Matter in Soil-Aquifer Treatment
Systems, ASCE.
Gerrity, D., S. Gamage, et al. (2011). "Pilot-scale evaluation of ozone and biological activated
carbon for trace organic contaminant mitigation and disinfection." Water Research 45(5): 21552165.
Henderson, R. K., A. Baker, et al. (2009). "Fluorescence as a potential monitoring tool for recycled
water systems: A review." Water Research 43(4): 863-881.
Hoppe-Jones, C., G. Oldham, et al. (2010). "Attenuation of total organic carbon and unregulated
trace organic chemicals in U.S. riverbank filtration systems." Water Research 44(15): 4643-4659.
Huber, M. M., S. Canonica, et al. (2003). "Oxidation of Pharmaceuticals during Ozonation and
Advanced Oxidation Processes." Environmental Science & Technology 37(5): 1016-1024.
Huber, S. A., A. Balz, et al. (2011). "Characterisation of aquatic humic and non-humic matter with
size-exclusion chromatography - organic carbon detection - organic nitrogen detection (LC-OCDOND)." Water Research 45(2): 879-885.
Kolpin, D. W., E. T. Furlong, et al. (2002). "Pharmaceuticals, Hormones, and Other Organic
Wastewater Contaminants in U.S. Streams, 1999−2000: A National Reconnaissance."
Environmental Science & Technology 36(6): 1202-1211.
Krauss, M., P. Longrée, et al. (2009). "Occurrence and removal of N-nitrosamines in wastewater
treatment plants." Water Research 43(17): 4381-4391.
Maeng, S. K. (2010). Mutiple Objective Treatment Aspects of Bank Filtration. Delft, Delft
University of technology and of the Academic Board of the UNESCO-IHE Institute for Water
Education. DOCTOR: 210.
Maeng, S. K., E. Ameda, et al. (2010). "Organic micropollutant removal from wastewater effluentimpacted drinking water sources during bank filtration and artificial recharge." Water Research
44(14): 4003-4014.
Pehlivanoglu-Mantas, E. and D. L. Sedlak (2006). "The fate of wastewater-derived NDMA
precursors in the aquatic environment." Water Research 40(6): 1287-1293.
Rakness, K., G. Gordon, et al. (1996). "Guideline for measurement of ozone concentration in the
process gas from an ozone generator." Ozone-Science & Engineering 18(3): 209-229.
Schwarzenbach, R. P., P. M. Gschwend, et al. (2005). Sorption III: Sorption Processes Involving
Inorganic Surfaces. Environmental Organic Chemistry, John Wiley & Sons, Inc.: 387-458.
Stevens-Garmon, J., J. r. E. Drewes, et al. (2011). "Sorption of emerging trace organic compounds
onto wastewater sludge solids." Water Research 45(11): 3417-3426.
Sundaram, V., Emerick, R.W., Shumaker, S.E. (2009). Field evaluation of MF-Ozone-BAC process
train for the removal of microconstituents from wastewater effluent. 24th Annual WateReuse
Symposium.
Ternes, T. A., M. Meisenheimer, et al. (2002). "Removal of Pharmaceuticals during Drinking Water
Treatment." Environmental Science & Technology 36(17): 3855-3863.
Trenholm, R. A., B. J. Vanderford, et al. (2006). "Broad range analysis of endocrine disruptors and
pharmaceuticals using gas chromatography and liquid chromatography tandem mass spectrometry."
Chemosphere 65(11): 1990-1998.
Vanderford, B. J., R. A. Pearson, et al. (2003). "Analysis of Endocrine Disruptors, Pharmaceuticals,
and Personal Care Products in Water Using Liquid Chromatography/Tandem Mass Spectrometry."
Analytical Chemistry 75(22): 6265-6274.
Vanderford, B. J. and S. A. Snyder (2006). "Analysis of Pharmaceuticals in Water by Isotope
Dilution Liquid Chromatography/Tandem Mass Spectrometry†." Environmental Science &
Technology 40(23): 7312-7320.
Westerhoff, P., Y. Yoon, et al. (2005). "Fate of Endocrine-Disruptor, Pharmaceutical, and Personal
Care Product Chemicals during Simulated Drinking Water Treatment Processes." Environmental
Science & Technology 39(17): 6649-6663.
Zhou, Q., S. McCraven, et al. (2009). "Field evidence of biodegradation of NNitrosodimethylamine (NDMA) in groundwater with incidental and active recycled water
recharge." Water Research 43(3): 793-805.