Bioresource Technology 179 (2015) 339–347

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Removal and toxicity reduction of naphthenic acids by ozonation and combined ozonation-aerobic biodegradation Eleni Vaiopoulou 1, Teresa M. Misiti, Spyros G. Pavlostathis ⇑ School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0512, USA

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Ozone reacted preferentially with

naphthenic acids (NAs) of higher cyclicity and MW.  Ozone reactivity with acyclic/ monocyclic model NAs depended on molecular structure.  NA branching, tertiary and quaternary carbons reduced ozone reactivity.  Semibatch ozonation of commercial NAs mixture followed pseudo firstorder kinetics.  Ozonation-biodegradation removed NAs by 89% and decreased toxicity by 15-fold.

a r t i c l e

i n f o

Article history: Received 14 November 2014 Received in revised form 14 December 2014 Accepted 15 December 2014 Available online 20 December 2014 Keywords: Aerobic biodegradation Bioavailability MicrotoxÒ toxicity Naphthenic acids Ozonation

a b s t r a c t A commercial naphthenic acids (NAs) mixture (TCI Chemicals) and five model NA compounds were ozonated in a semibatch mode. Ozonation of 25 and 35 mg/L NA mixture followed pseudo first-order kinetics (kobs = 0.11 ± 0.008 min1; r2 = 0.989) with a residual NAs concentration of about 5 mg/L. Ozone reacted preferentially with NAs of higher cyclicity and molecular weight and decreased both cyclicity and the acute MicrotoxÒ toxicity by 3.3-fold. The ozone reactivity with acyclic and monocyclic model NAs varied and depended on other structural features, such as branching and the presence of tertiary or quaternary carbons. Batch aerobic degradation of unozonated NA mixture using a NA-enriched culture resulted in 83% NA removal and a 6.7-fold decrease in toxicity, whereas a combination of ozonation-biodegradation resulted in 89% NA removal and a 15-fold decrease in toxicity. Thus, ozonation of NA-bearing waste streams coupled with biodegradation are effective treatment processes. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Naphthenic acids (NAs) are a complex group of alkyl-substituted acyclic, monocyclic and polycyclic carboxylic acids. Classical NAs have a general formula of CnH2n+ZO2, where n is the carbon ⇑ Corresponding author. Tel.: +1 404 894 9367; fax: +1 404 894 8266. E-mail address: [email protected] (S.G. Pavlostathis). Present address: Laboratory of Microbial Ecology and Technology, Faculty of Bioscience Engineering, University of Ghent, 9000 Ghent, Belgium. 1

http://dx.doi.org/10.1016/j.biortech.2014.12.058 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.

number and Z is the number of hydrogen atoms lost due to ring formation, also known as hydrogen deficiency. Z is zero or a negative, even integer, and reflects the degree of cyclicity. NAs are anionic surfactants, have pKa values in the range of 5–6, and their molecular weight (MW) typically varies between 140 and 500 Da (Whitby, 2010). Recently, use of advanced mass spectrometric techniques has shown that oil sands process-affected water (OSPW), as well as oil field and refinery wastewaters contain oxidized NAs (CnH2n+ZOx, where x is three or more oxygen atoms) in addition to classical NAs which contain 2 oxygen atoms (Headley

340

E. Vaiopoulou et al. / Bioresource Technology 179 (2015) 339–347

et al., 2013; Pereira et al., 2013a; Sun et al., 2014; Wang et al., 2013a). The range of NA structures has recently been expanded to include other polar and N–S-containing heteroatomic species and aromatic species found in the oil sands acid extractable fraction (Headley et al., 2011a,b, 2013; Sun et al., 2014). NAs are found in OSPW, crude oil and petroleum products, as well as in refinery process water and wastewater. Typical NA concentrations are 40–120 mg/L in OSPW (Kannel and Gan, 2012), 0.9–3.6 mg/g in heavy crude oil, 4.2–40.4 mg/L in refinery desalter brine, and 4.5–16.6 mg/L in the influent of refinery wastewater treatment systems (Misiti et al., 2013b). NAs are corrosive and among the most toxic components of OSPW, refinery process water and wastewater, acutely toxic to both aquatic and terrestrial species (Whitby, 2010). To mitigate NA toxicity and treat OSPW, various removal and degradation technologies have been tested. The most well documented and cost effective method investigated is biodegradation. Model NAs, commercial NA mixtures and NAs in refinery process water can be aerobically degraded by indigenous or enriched bacterial cultures, although some types of NAs are recalcitrant (Choi et al., 2014; Hwang et al., 2013; Misiti et al., 2013c; Scott et al., 2005). However, previous reports have shown that complete NA biodegradation may not be feasible (Misiti et al., 2013a,b). Thus, combination of advanced oxidation and biodegradation processes may be more effective resulting in very low NA concentrations, thus achieving a higher quality effluent. The toxicity and biodegradability of NAs depend strongly on their structure. Complex NAs, i.e., those with higher cyclicity, branching and high MW, higher degree of alkyl-substituted aliphatic chains, tertiary substituents at positions other than the b-position relative to the carboxylic group of the main carbon chain, methyl substitution on the cycloalkane rings, are the most recalcitrant and thus persistent NAs (Han et al., 2008; Kannel and Gan, 2012; Misiti et al., 2013c, 2014; Scott et al., 2005; Smith et al., 2008; Whitby, 2010). Advanced oxidation processes (AOPs) have been suggested as the best alternative method for the degradation of recalcitrant NAs (Kannel and Gan, 2012). AOPs, such as ozonation, UV radiation, and other catalytic processes create oxidizing conditions by the generation of highly reactive oxidizing species, mainly hydroxyl radicals (OH). Photolysis using natural UV radiation is very limited in degrading NAs; however, photocatalysts, such as TiO2, microwaves and H2O2 combined with UV radiation are effective in oxidizing NAs (Afzal et al., 2012; Headley et al., 2010; Mishra et al., 2010). Ozonation of NAs is considered the most promising AOP with a great potential for OSPW remediation, and has been suggested as a pre-treatment step to biodegradation (Brown et al., 2013; Choi et al., 2014; Gamal El-Din et al., 2011; Hwang et al., 2013; Kannel and Gan, 2012; Martin et al., 2010; Pereira et al., 2013b; Perez-Estrada et al., 2011; Scott et al., 2008; Sun et al., 2014; Wang et al., 2013b). This suggestion is based on the ability of ozone to degrade recalcitrant NAs with relatively high MW and cyclicity, leading to a reduction of their toxicity and an increase of their biodegradability. However, NAs are not expected to be completely degraded by ozone due to their complex nature, whereas formation of potentially more toxic and/or hazardous by-products cannot be excluded. NA structure, MW and the rate of generation of hydroxyl radicals strongly affect the effectiveness of NA degradation by ozone (Hwang et al., 2013; Kannel and Gan, 2012; Pereira et al., 2013b; Perez-Estrada et al., 2011; Wang et al., 2013b). In spite the fact that NAs are found in refinery waste streams where are proven difficult to treat and create operational problems, such as corrosion and toxicity (Dorn, 1998; Misiti et al., 2013b; Whitby, 2010), relatively less is known about the fate and degradation of NAs in such streams compared to OSPW and other waste streams related to oil sands processes. The impetus of the

work reported here was the observation that aerobic biodegradation processes, typically employed for the treatment of refinery wastewater, were not able to completely remove NAs resulting in low, residual NA concentrations (Misiti et al., 2013a,b). Ozonation coupled with biodegradation may result in a more effective treatment of NA-bearing waste streams, potentially leading to water reclamation and reuse. The objectives of this study were to assess: (a) the degradation of a commercial NA mixture and select model compounds by ozonation; (b) the effectiveness of ozonation relative to NA molecular structure; and (c) the toxicity and biodegradability of the ozonated NA mixture. 2. Methods 2.1. Chemicals A commercial mixture of NA sodium salt, purchased from TCI Chemicals (Tokyo Chemical Industry Co., Ltd., Tokyo, Japan), was used in this study. The mixture consists of mostly 0, 1 and 2 ring NA structures (i.e., Z = 0, 2, 4) with carbon numbers (n) ranging between 10 and 25 (Misiti et al., 2013a). The fraction of 0, 1, and 2 ring NA structures in the TCI NA mixture was 10.5%, 34.6%, and 35.4%, respectively. The TCI NA mixture is representative of the types of NAs found in refinery wastewater streams, having a similar NA congener distribution to refinery desalter brine, which was identified as the main source of NAs in refinery wastewater (Misiti et al., 2013b). A stock solution of 8 g NA/L in deionized water (DI) was prepared and used in all assays. NA concentrations mentioned in this work refer to NAs and not the NA salt. The NA concentrations used in this study are of the same order of magnitude to those found in refinery desalter brine and wastewater influent streams (Misiti et al., 2013b). To investigate the significance of NAs structure relative to the ozone reactivity, five model NA compounds were used in this study: octanoic acid (OCT; n = 8 acyclic, unbranched), 2-ethylhexanoic acid (EHA; n = 8 acyclic, branched), 1-methyl-1-cyclo-hexane carboxylic acid (1MCH; n = 8 monocyclic with a quaternary carbon), 4-methyl-1-cyclo-hexane carboxylic acid (4MCH; n = 8 monocyclic with two tertiary carbons), and 2,2-dicyclo-hexylacetic acid (DCH; n = 14 dicyclic with a tertiary carbon) (Table 1). All model compounds were of >99% purity, purchased from Sigma– Aldrich (St. Louis, MO). Stock solutions of 10 g/L of each model compound were prepared in 1 N NaOH and used for the preparation of working solutions, which were then adjusted to pH 7 with 1 N HCl. 2.2. Ozonation An ozone generator (model CD 06-A; Aqua-Flo Inc., Baltimore, MD) was used to produce ozone gas from extra dry, high purity oxygen. The electrical discharge of the ozonator was set at maximum output (120 V, 60 Hz, 30 W) and the oxygen flow rate fed to the ozonator was kept constant at 100 L/h during all experiments. Ozonation was performed at room temperature (22– 24 °C) in a modified 1-L Pyrex graduated cylinder reactor (5.6 cm inner diameter  20.3 cm water column height). Ozone gas was fed to the bottom of the glass reactor through a fine pore size glass sparger. Excess ozone was constantly provided for the whole duration of each ozonation test. The steady-state gas-phase O3 concentration was 9.3 ± 0.2 mg/L at 23 °C and 1 atm, regardless if the reactor was empty, filled with DI water or NAs solution. Ozonation was conducted in a semibatch mode (liquid batch, ozone continuous) as follows. The reactor was filled up to 500 mL with either a TCI NA mixture or a model NA working solution and

E. Vaiopoulou et al. / Bioresource Technology 179 (2015) 339–347 Table 1 Molecular formula and structure of model NA compounds used in this study. Model compound

Molecular formula

Octanoic acid (OCT)

C8H16O2

2-Ethylhexanoic acid (EHA)

C8H16O2

1-Methyl-1-cyclo-hexane carboxylic acid (1MCH)

C8H14O2

4-Methyl-1-cyclo-hexane carboxylic acid (4MCH)

C8H14O2

2,2-Dicyclo-hexylacetic acid (DCH)

C14H24O2

Structure

341

onated NA mixture (control) and one amended with ozonated NA mixture, were prepared in 500 mL Erlenmeyer flasks (total liquid volume 400 mL). Salt medium, ammonium chloride, and inoculum were added to the NA mixtures. The salt medium consisted of (in g/ L): K2HPO4, 1.07; KH2PO4, 0.524; CaCl22H2O, 0.068; MgCl22H2O, 0.135; MgSO47H2O, 0.268; FeCl24H2O, 0.068; and 0.67 mL/L trace metal stock solution (Misiti, 2012). The ozonated NA mixture was prepared by ozonation of a 25 mg/L TCI NA mixture for 60 min. The resulting initial biomass and ammonium chloride concentrations were 110 mg VSS/L and 25 mg N/L in both culture series. The initial NA concentration was 25 and 4.3 mg/L in the unozonated and ozonated culture series, respectively. The initial NA concentration used in the bioassay was chosen based on typical NA concentrations found in the influent of oil refinery activated sludge wastewater treatment systems (4.5–16.6 mg NA/L) as previously reported (Misiti et al., 2013b). The cultures were mixed with a magnetic stirrer, aerated with pre-humidified compressed air passed through a fine pore diffuser (DO P 7 mg/L), and incubated at room temperature (22–24 °C). Total and liquid-phase NAs, pH, total and soluble COD were measured. 2.4. Analytical methods

the ozonator was switched on. Working solutions of both the commercial NA mixture and model compounds were prepared from stock solutions diluted with DI water and buffered with 2%, v/v 0.5 M phosphate buffer (K2HPO4 and KH2PO4; pH 7.2). NA solutions with a relatively low initial NA concentration (25–35 mg/L) were used due to foaming during ozonation. The duration of ozonation varied from 2.5 to 60 min, depending on initial NA and model compound concentration. Immediately after ozonation, the ozonator was switched off and the glass reactor sparged with oxygen for 5 min to purge all residual ozone from the liquid phase. Oxygen, as opposed to reduced agents (e.g., thiosulfate), was used to quench the ozonated solution in order to avoid interference with the measurement of chemical oxygen demand (COD) and possibly with the MicrotoxÒ test. To ensure undisturbed conditions of ozonation, the above-described ozonation procedure was repeated for each ozonation duration assessed, typically 2.5, 5, 10, 15, 20, 30 and 60 min, and for replicate ozonation runs (independent runs). Ozonated samples were collected and then analyzed for pH, COD, dissolved organic carbon (DOC), volatile fatty acids (VFAs), non-ionizable organic acids, NAs and model compounds. In addition to TCI NA mixtures, ozonation of the selected five model NA compounds was carried out at an initial concentration between 22 and 25 mg/L. 2.3. Bioassay A batch bioassay was performed to compare the biodegradability of the ozonated TCI NA mixture to that of the untreated (unozonated) mixture. The inoculum for this bioassay was obtained from a mixed, NA-enriched culture maintained for over 3 years fed with the untreated TCI NA mixture as the sole carbon and energy source (Misiti et al., 2013c, 2014). In order to remove residual and biomass-adsorbed NAs, the inoculum was prepared by washing a sample of the culture with 0.5 M phosphate buffer (pH 7.2), followed by biomass separation by settling, and this procedure was repeated once more. The residual NAs concentration in the washed inoculum, extracted and measured as described in Section 2.4, below, was 3.7 mg/L. Two culture series, one amended with unoz-

Total and soluble COD, pH, and DOC were determined following procedures outlined in Standard Methods (APHA, 2012). VFAs (C2 to C7) were measured by gas chromatography (GC) with flame ionization detection (Agilent 6890 Series; Agilent Technologies, Inc.) and non-ionizable organic acids were measured by HPLC with refractive index detection (HP 1100 Series; Hewlett Packard) (Misiti, 2012). All liquid-phase parameters were measured after the samples were centrifuged at 10,000 rpm for 15 min and the supernatants filtered through 0.2 mm PTFE filters. To measure the gas-phase ozone concentration, gas samples were collected in a quartz cuvette (1  1  3 cm) and absorbance at 254 nm was measured using a Hewlett–Packard 8453 UV/VIS spectrophotometer equipped with a diode array detector. Pure oxygen gas was used as the blank sample. Then, the gas-phase ozone concentration was calculated from absorbance measurements according to the method described by Rakness et al. (1996). NA concentrations and congener distributions were determined using a pair-ion extraction method (PIX) after the sample pH was raised to 10, followed by liquid chromatography/mass spectrometry (LC/MS; direct infusion; electron spray ionization in negative mode) (Misiti et al., 2013b). Benzyl tributyl ammonium (BTBA) and p-toluene sulfonate (pTS) were used as the counter-ion and surrogate standard, respectively. The NA concentration was determined relative to the surrogate standard (pTS). Analysis and quantification of the model NAs followed the same procedures used for the NA mixture. The acute toxicity of NA-bearing samples was assessed using the standard MicrotoxÒ test as previously described (Misiti et al., 2013a). The effective sample strength (% v/v) that results in light emission at 50% of the control (EC50) was used as the descriptor of toxicity. 3. Results and discussion 3.1. Ozonation of TCI NA mixture Samples of the TCI NA mixture at initial concentrations of 25 and 35 mg NAs/L were exposed to excess ozone for up to 60 min (Fig. 1). Very little pH change was observed during ozonation, which remained between 6.9 and 7.0. After 60 min ozonation, the remaining NA concentration was equal to about 5 mg/L, regardless of the initial NA concentration. Progressively lower deg-

342

E. Vaiopoulou et al. / Bioresource Technology 179 (2015) 339–347

radation rates during ozonation have been attributed to the accumulation of refractory by-products (Rivas et al., 2012), as well as to fast direct ozone reactions followed by slower indirect ozone reactions by hydroxyl radicals produced from the decomposition of ozone (Nöthe et al., 2009). However, given that in the present study ozone was continuously supplied, the observed NA degradation more likely was a combination of direct and indirect ozone reactions. This is in agreement with Wang et al. (2013b), who reported oxidation of NAs by the combined effect of ozone and hydroxyl radicals in short reaction times (0–100 min), whereas the role of hydroxyl radicals become dominant at longer ozonation times. Hydroxyl radicals are considered to be the main contributors to the NAs degradation by abstracting hydrogen atoms from C–H bonds (Gamal El-Din et al., 2011; Perez-Estrada et al., 2011; Wang et al., 2013b). It was recently shown that although molecular ozone does not react directly with NAs, its decomposition in NAcontaining wastewaters is followed by secondary reactions that contribute to NAs degradation (Wang et al., 2013b). Based on the residual NAs measured at 60 min ozonation, the extent of NA conversion ranged from 81% to 85% (Table 2), not a significant difference (p = 0.05) between the two NA samples. Other studies have also reported a NA fraction that remains unaffected by ozone (Choi et al., 2014; Gamal El-Din et al., 2011; Scott et al., 2008). Scott et al. (2008) reported that continuous bubbling with ozone resulted in the degradation of NAs in OSPW from an initial concentration of 60 down to 20 mg NAs/L in 50 min and to 2 mg NAs/L after 130 min of ozonation, which corresponds to about 66% removal in less than 60 min of ozonation. Given the fact that ozonation in the present study was performed with a continuous supply of ozone, which resulted in a constant, steady-state ozone concentration (9.3 ± 0.2 mg/L in the gas-phase) throughout the tests, pseudo first-order kinetics were assumed. As mentioned above, after 30 min continuous ozonation of two NA mixture sam-

A

45 40 35

CONCENTRATION (mg NA/L)

30 25 20 15 10 5 0 30

B

25 20 15 10 5 0

0

10

20

30

40

50

60

70

TIME (min) Fig. 1. NA concentration over the course of 60-min semibatch ozonation of the TCI NA mixture starting with approximately 35 (A) and 25 (B) mg NA/L. Error bars represent mean values ± one standard deviation (n = 3). Lines are model fit to the experimental data (see text).

Table 2 Extent of NA conversion and COD, DOC removal after 60-min ozonation of the TCI NA mixture tested at two different initial NA concentrations. Initial NA (mg/L) 23.8 ± 2.4a 35.7 ± 4.1 a

NA conversion (%)

COD removal (%)

DOC removal (%)

81 ± 4 85 ± 2

42 ± 5 46 ± 5

14 ± 1 11 ± 1

Mean ± standard deviation (n = 3).

ples at two different initial NA concentrations resulted in a low NA concentration, which remained practically stable after an additional 30 min of ozonation (Fig. 1). Taking into account the residual, non-reactive NA concentration, the following equation was used to evaluate NA degradation kinetics:

C t ¼ C o expðkobs tÞ þ C r

ð1Þ

where Co, Ct and Cr is the NA concentration initially, at time t, and residual (mg/L); kobs is the pseudo first-order degradation rate constant (min1). Using the data shown in Fig. 1B, non-linear regression was conducted based on Eq. (1) using SigmaPlot Ver. 12 (Systat Software, Inc., San Jose, CA). The value of k was equal to (estimate ± standard error) 0.11 ± 0.008 min1 (r2 = 0.989), which corresponds to a mean half-life time of 6.3 min. Fig. 1 shows the fit of Eq. (1) to the two sets of data obtained by the ozonation of two NA mixture samples at two different initial concentrations using the same k value, further demonstrating that pseudo first-order kinetics described well the semibatch ozonation process used in the present study. In a previous study by Perez-Estrada et al. (2011) on the ozonation of OSPW in a batch system with limitless supply of ozone and a commercial NA mixture (Merichem Chemicals) in a batch system with a saturated ozone solution, the kinetics of NA degradation was pseudo first-order (k 6 0.025 min1) and second-order (k 6 0.08 L/mg-min), respectively. First-order kinetics in both studies is attributed to the excess ozone, and thus presence of hydroxyl radicals in abundance, throughout the tests. As a result, the rate of ozonation depended mainly on the NA concentration. Based on data reported by Perez-Estrada et al. (2011), at a pH 8, which resulted in the highest extent of the commercial NA degradation by ozone, the half-life time was approximately 1.7 min. The difference in ozonation kinetics between the Merichem and the TCI NA mixture used in the present study is attributed to the difference in both the chemical composition of the two mixtures and the ozonation apparatuses used. To our knowledge, this is the first report on the degradation kinetics of the commercial TCI NA mixture under excess ozone conditions. The NA distribution as a function of ozonation time is shown in Fig. 2. Consistent with the decrease in the total NA concentration (Fig. 1), even 10 min of ozonation resulted in a significant decrease of NA abundance and changes in carbon (n) and Z numbers. Beyond 30 min of ozonation no significant change was detected (Fig. 2). Fig. 3 shows that n and Z gradually decreased, resulting in residual NAs with a lower MW and cyclicity. It is noteworthy that the most abundant NAs initially and also after 60 min of ozonation were mono- (Z = 2) and di-cyclic NAs (Z = 4) (Fig. 3). Moreover, NAs with six (Z = 12), five (Z = 10) and four (Z = 8) rings were completely removed at 30 min of ozonation. Specifically, the hexacyclic (Z = 12) NAs were the first to be degraded and totally depleted after 10 min of ozonation, whereas 30 min of ozonation eliminated tetra- and penta-cyclic NAs (Z = 8 and 10). By the end of 60 min ozonation only NAs with three or less rings remained. At the end of 60 min of ozonation, and considering that conversion of higher cyclicity NAs to lower ones due to ring opening more likely took place, the net removal of NAs with Z = 0, 2, 4, 6, 8, 10 and 12 was 69%, 77%, 80%, 83%, 100%, 100%

343

E. Vaiopoulou et al. / Bioresource Technology 179 (2015) 339–347

and 100%, respectively. It is noteworthy that the extent of NA removal is proportional to the number of rings, implying that NAs with a higher MW and cyclicity are more prone to degradation by ozone and were thus preferentially oxidized. These findings are in agreement with other NA ozonation studies. Scott et al. (2008) reported that continuous ozonation of OSPW degraded NAs and shifted their distribution to NAs with a lower MW and cyclicity. Other researchers also found that ozone preferentially degraded NAs of higher cyclicity, while compounds with lower cyclicity were produced, possibly due to ring opening (Gamal El-Din et al., 2011; Perez-Estrada et al., 2011; Wang et al., 2013b). It was demonstrated that 90% removal of total NAs by ozonation corresponded to 74%, 94% and 99% removal of monocyclic, dicyclic and tetracyclic NAs, respectively, whereas acyclic and monocyclic NAs increased (Gamal El-Din et al., 2011). As shown in Table 2, in spite the fact that ozonation for 60 min resulted in a significant removal of total NAs (81–85%), a lower COD and a much lower DOC removal was observed. Thus, ozonation resulted in partial oxidation of NAs and conversion to other organic compounds, achieving a very low extent of NA mineralization. Similarly to our findings, Scott et al. (2008) reported that continuous ozonation of OSPW resulted in about 70% and >95% removal of NAs after ozonation for 50 and 130 min, respectively, but ozonation for 130 min resulted in about 50% COD decrease and the total organic carbon concentration did not change significantly. Similarly, 90% of OSPW NAs were removed by ozonation, but DOC decreased slightly indicating oxidation but no mineralization (Brown et al., 2013). It has been reported that the major intermediate byproducts resulting from ozonation of OSPW were oxidized NAs, such as hydroxy- or keto-NAs (Martin et al., 2010; Wang et al., 2013b) and unaffected non-NAs carboxylic acids (Pereira et al., 2013b), which might be responsible for remaining

To further investigate the role of NA structure on ozone reactivity, five model compounds were ozonated for either 10 or 60 min. Working solutions of the model compounds were prepared in 1 N NaOH and buffered with 0.5 M phosphate buffer (pH 7.2). The initial pH of OCT, EHA, 1MCH, 4MCH and DCH solutions was 7.42, 7.44, 7.35, 7.43 and 7.59, respectively, and remained unchanged regardless of the duration of ozonation. Limited removal (below 14%) of all model compounds tested was observed in 10 min of ozonation (Table 3) compared to about 53% NA removal of the NA mixture at a comparable initial NA concentration of 25 mg/L. Ozonation of the model compounds for 60 min resulted in higher removal ranging from 25% to 40% (Table 3), which is again lower than the observed removal for the NA mixture (81%). The observed difference in ozone reactivity between the five model compounds and the commercial NA mixture is attributed to the above-discussed observation that NAs with a higher MW and cyclicity are more prone to degradation by ozone and are thus preferentially oxidized. DCH, which has the highest number of rings and carbons among the five compounds tested, was degraded the most (about 40%)

pTS

30 20

80

10 60 0 25 20 15 n 10 5

40

-6 -10-8

0 -4 -2

Z

20 0

C

RELATIVE ABUNDANCE (%)

100

100

pTS

30 20

80

10 60

0 25 20 15 n 10 5

40

-6 -4 -10-8

-2 0

Z

20 0

D 100

pTS 30 20

80

10

60

0 25 20 15 n 10 5

40 20 0 150

-6 -10-8

0 -4 -2

Z

200

250

300

m/z

350

400

RELATIVE ABUNDANCE (%)

RELATIVE ABUNDANCE (%)

3.2. Ozonation of model NA compounds

B

A

RELATIVE ABUNDANCE (%)

toxicity after ozone treatment. Lower COD and DOC reduction compared to the extent of NAs removal was also observed by Hwang et al. (2013) and Gamal El-Din et al. (2011) who assessed the effectiveness of OSPW ozonation. Ozonation of the Merichem NA mixture and OSPW resulted in the production of oxy-NAs, which were degraded at higher utilized ozone dose (Sun et al., 2014). The limited ozone-induced NA mineralization observed in the present study indicates production of oxidized intermediates. VFAs and non-ionizable organic acids were not detected in samples of TCI NA mixture ozonated for up to 60 min.

100

pTS 30 20

80

10

60

0 25 20 15 n 10 5

40 20 0 150

-8 -6 -12-10

-4 -2

0

Z

200

250

300

350

400

m/z

Fig. 2. NA distribution as a function of ozonation time of the TCI NA mixture at an initial NA concentration of 25 mg/L ((A) 0 min; (B) 10 min; (C) 30 min; and (D) 60 min ozonation; pTS, p-toluene sulfonate used as the surrogate standard).

E. Vaiopoulou et al. / Bioresource Technology 179 (2015) 339–347

160 140 120 100 80 60 40 20 0 160 140 120 100 80 60 40 20 0 160 140 120 100 80 60 40 20 0 160 140 120 100 80 60 40 20 0

A1

B1

C1

D1

0

-2

-4

-6

-8

-10

-12

ABUNDANCE

ABUNDANCE

344

80 70 60 50 40 30 20 10 0 80 70 60 50 40 30 20 10 0 80 70 60 50 40 30 20 10 0 80 70 60 50 40 30 20 10 0

Z NUMBER

A2

B2

C2

D2

10 12 14 16 18 20 22 24 26

CARBON NUMBER (n)

Fig. 3. Effect of ozonation time on the cyclicity (1) and carbon number (n) distribution (2) of the TCI NA mixture at an initial NA concentration of 25 mg/L ((A) 0 min; (B) 10 min; (C) 30 min; and (D) 60 min ozonation).

after 60 min ozonation (Table 3). The faster and higher removal of DCH (14% and 40% removal in 10 and 60 min, respectively) in comparison to the other model compounds is attributed to the higher number of H atoms (and carbons) that are available for abstraction, which is the main indirect reaction of ozone with organic compounds (Minakata et al., 2009; Afzal et al., 2012). These results are in agreement with those of other studies which reported increased NA oxidation by ozone with increasing number of carbons and rings (Perez-Estrada et al., 2011; Wang et al., 2013b). However, among the four model compounds with n = 8, the two acyclic compounds were converted by ozone in 60 min more than the two monocyclic compounds (Table 3). The effect of a saturated ring on the reactivity of model NAs with hydroxyl radicals was assessed by Afzal et al. (2012) by comparing the linear decanoic acid (C10H20O2) to the monocyclic 4-propylcyclohexanoic acid (C10H18O2, Z = 2) and found that the cyclic compound was more reactive than the linear by more than 13%. The higher reactivity of the cyclic NA was attributed to the higher reactivity of H on the two tertiary carbons compared to the primary and secondary carbons in the linear decanoic acid. Their results agreed well with estimates of rate constants based on previously reported methods

Table 3 Conversion of five model NA compounds by ozonation.a Model compound

OCT EHA 1MCH 4MCH DCH

Conversion (%) with ozonation for n

Z

10 min

60 min

8 8 8 8 14

0 0 2 2 4

13.1 8.1 4.1 1.2 13.8

32.4 34.7 25.1 29.4 40.3

a The initial concentration of the five model compounds varied between 22 and 25 mg/L.

(Minakata et al., 2009). The lower reactivity of the two monocyclic compounds compared to the two acyclic compounds observed in the present study may be attributed to the position of the alkyl group resulting in one quaternary (for 1MCH) and one tertiary (for 4MCH) carbon in the two monocyclic compounds, features that can affect the reactivity of these compounds with hydroxyl radicals as shown previously (Minakata et al., 2009; Afzal et al., 2012). On the other hand, branching in the acyclic EHA may have

E. Vaiopoulou et al. / Bioresource Technology 179 (2015) 339–347

contributed to a higher reactivity, though the distance between the tertiary carbon and the carboxylic group plays also a role (Minakata et al., 2009). Relative to the effect of branching, a comparison of the reactivity of the linear OCT to that of EHA, which has an ethyl substituent at the a-position (tertiary carbon), resulted in mixed results. OCT was transformed by 13% and 32% with 10 and 60 min ozonation, respectively, whereas EHA was transformed by 8% with 10 min ozonation and had a slightly higher extent of transformation with 60 min ozonation (Table 3). An alkyl group substitution increases the reactivity with hydroxyl radicals as a result of a decrease in the activation energy (Minakata et al., 2009), but such an effect was not pronounced in the present study in the case of EHA. This result may be attributed to the fact that a tertiary carbon in the a-position of acyclic NAs results in lower reactivity and degradation rate in comparison with linear acyclic NAs, which in turn is attributed to the lower reactivity of hydrogen atoms in a-positions (Perez-Estrada et al., 2011; Afzal et al., 2012). To assess the effect of a quaternary carbon, the extent of transformation of 1MCH and 4MCH by ozonation was compared. Both isomers are monocyclic, have the same carbon number, but differ in the methyl position: 1MCH at a-position, resulting in a quaternary carbon, and 4MCH at d-position, resulting in a tertiary carbon. Although both compounds were the least reactive among all five model compounds tested, as shown by the limited degradation with 10 min ozonation, 60 min ozonation resulted in 25% and 29% transformation of 1MCH and 4MCH, respectively (Table 3). The limited reactivity of both compounds is expected as the introduction of a quaternary carbon at a-position reduces NAs reactivity (Perez-Estrada et al., 2011; Afzal et al., 2012). On the other hand, a quaternary carbon in cyclic NAs has no hydrogen atom for abstraction and thus, reduces reactivity. Similar work showed that quaternary carbons result in lower ozone reaction rates (Perez-Estrada et al., 2011; Afzal et al., 2012). Comparing total removal efficiencies, the higher reactivity of 4MCH compared to 1MCH could be attributed to the methyl-substitution at d-position, which permits H abstraction. The limited number of existing studies underlines the need for further investigation on the effect of NA molecular structure on the effectiveness of ozonation which was assessed in the present study by using model compounds and the commercial TCI NA mixture. 3.3. Biodegradation of ozonated and unozonated NA mixture Ozonation tests conducted in the present study as well as in previous studies have shown that for the most part NAs are transformed rather than mineralized and thus, ozone persistent NAs and by-products accumulate in the ozonated solution. Thus, the toxicity and biodegradability of the ozonated solutions need to be assessed. In order to assess the effect of ozonation on the aerobic biodegradation of the commercial TCI NA mixture, a 25 mg/L NA sample was ozonated for 60 min and its NA distribution as well as biodegradation compared with those of the unozonated NA mixture. Ozonation for 60 min resulted in a decrease of the NA concentration from 25 to 4.7 mg/L, which corresponds to about 81% NA conversion, while the COD and DOC concentrations decreased by about 50% and 12%, respectively. As shown in Fig. 4, ozonation resulted in a decrease of cyclicity and a shift in the median carbon number because of the decrease in the abundance of NAs with carbon number between 12 and 16, which, relatively speaking, were converted the most by the ozone treatment. Incubation for 3 days under aerobic conditions with the NAenriched culture inoculum resulted in a decrease of the liquid-phase NA concentration from 25 to 3.1 mg/L and from 4.3 to 2.7 mg/L in the culture series amended with the unozonated and ozonated NA mixture, respectively. The final total NA concentration, i.e., both

345

liquid- and biomass-associated NAs, was 4.2 and 3.6 mg/L in the culture series amended with the unozonated and ozonated NA mixture, respectively. In a previous study conducted with the same TCI NA mixture and the same NA-enriched mixed culture used in the present study, a similar level of residual NAs was achieved even after extending the aerobic incubation time beyond 3 days (Misiti et al., 2013c). The results of the present study show that the NA-enriched culture was able to access and degrade NA compounds in both the unozonated and ozonated NA mixture. It is noteworthy that the combination of ozonation-biodegradation achieved practically the same extent of NA conversion/removal as biodegradation alone. Likewise, the extent of soluble NA-associated COD removal (i.e., accounting for residual, background soluble COD), was 90% and 89% in the culture series amended with the unozonated and ozonated NA mixture, respectively. Thus, aerobic biodegradation achieved about the same extent of NA oxidation regardless if the NA mixture was ozonated or not. It is noteworthy that in the case of both unozonated and ozonated NA samples, a residual NA concentration persisted after 3 days of incubation, which is attributed to very low concentrations of individual NA compounds in the mixture. These results are in agreement with those of Misiti et al. (2013c), who reported that the residual NAs remaining after batch aerobic biodegradation assays with the same NA-enriched culture were not necessarily inherently recalcitrant but rather existed in too low individual or group concentrations, below a threshold value. However, the rate of mineralization is expected to be significantly faster in the case of the ozonated NA mixture. Indeed, biodegradation of NAs in biofilm reactors showed that both the extent and rate of removal and mineralization were higher with ozonated compared to unozonated OSPW samples (Choi et al., 2014). Fig. 4 shows the abundance of NAs relative to the number of rings (i.e., Z number) before and after the bioassay for both the unozonated and ozonated NA mixtures. In both cases, acyclic NAs were degraded significantly. Moreover, mono- and dicyclic NAs were removed by 90% and 70% from the unozonated and ozonated NA mixture, respectively, implying an affinity of the NAenriched culture for the Z = 2 and 4 NAs. These results are in agreement with previous biodegradation studies conducted with the same NA-enriched culture (Misiti et al., 2013c), in which unozonated NAs with Z numbers equal to 2 (mono-) and 4 (di-cyclic) were degraded at the highest extent (about 90%). The median carbon number of both the unozonated and ozonated NA mixtures increased after the bioassay (Fig. 4), a direct result of the preferential biodegradation of the lower MW NAs. These results agree with those of previous biodegradation studies conducted with OSPW and commercial NA mixtures (Choi et al., 2014; Misiti et al., 2013c; Scott et al., 2005; Han et al., 2008; Wang et al., 2013b).

3.4. Microtox toxicity The toxicity of unozonated and ozonated samples of the TCI NA mixture used in the above-discussed bioassay (initial concentration 25 mg NA/L), before and after the bioassay (filtered samples) was measured by the standard acute MicrotoxÒtoxicity assay for a 5- and 15-min exposure. The 5-min EC50 values are reported here as there was negligible difference between the 5- and 15-min exposure EC50 values. The EC50 values were 27.7% and 92.6% (v/v) for the unozonated and ozonated NA mixture, respectively, before the bioassay. Thus, ozonation alone lowered the NA mixture toxicity by a 3.3-fold. Results of previous studies based on the standard MicrotoxÒ assay have shown that ozonation of OSPW and commercial NAs mixtures reduced NA toxicity (Scott et al., 2008; Martin et al., 2010; Gamal El-Din et al., 2011; Sun et al., 2014; Wang et al., 2013b), which has been attributed to a decrease in the concentration of lower MW NAs.

E. Vaiopoulou et al. / Bioresource Technology 179 (2015) 339–347

120

A1

100

Unozonated Ozonated

40

60

30

40

20

0 120

B1

100

10 0 60

40

60

30

40

20

20

10 0

-2

-4

-6

-8

-10

-12

B2

50

80

0

Unozonated Ozonated

50

80

20

A2

60

ABUNDANCE

ABUNDANCE

346

0

10

Z NUMBER

12

14

16

18

20

22

CARBON NUMBER (n)

Fig. 4. Effect of 60-min ozonation and combined ozonation-biodegradation on the cyclicity (1) and carbon number (n) distribution (2) of the TCI NA mixture at an initial NA concentration of 25 mg/L ((A) and (B) before and after biodegradation, respectively).

After the aerobic bioassay, the 5-min EC50 values increased to 185.1 and 416.6% (v/v) for the unozonated and ozonated samples of the TCI NA mixture, respectively. Thus, biodegradation alone lowered the NA mixture toxicity by 6.7-fold, whereas the combination of ozonation-biodegradation resulted in a 15-fold lower toxicity. EC50 values above 100% imply that NA mixtures more concentrated than the ones used in this test will result in a measurable acute MicrotoxÒ toxicity. These results show the synergistic and beneficial effect of combining ozonation with biodegradation to treat NA-bearing waste streams. The results of the MicrotoxÒ toxicity assay also agree with the above-discussed high extent of soluble COD removal and partial mineralization during the bioassay with both the unozonated and ozonated NA mixture. The results of the present study are in agreement with previous studies which reported toxicity reduction of OSPW and commercial NA mixtures by biodegradation (Brown et al., 2013; Choi et al., 2014; Misiti et al., 2013a,c; Wang et al., 2013b). The observed reduction in toxicity is attributed to the decrease of the total NA concentration, as well as to changes in the NA congener distribution (i.e., Z and n numbers, MW). However, an attempt to correlate these NA descriptors to measured MicrotoxÒ acute toxicity for refinery wastewaters was not successful, due to their complex nature and presence of other inhibitors (Misiti et al., 2013b). In the present study, NAs with lower Z and n numbers, and thus lower MW were removed by biodegradation and/or a combination of ozonation-biodegradation, resulting in decreased toxicity, in agreement with other studies, which correlated toxicity reduction to the decrease of lower MW NAs removed by biodegradation (Frank et al., 2009; Wang et al., 2013b). This work is a step forward to understand the effectiveness of NA ozonation and facilitate coupling ozonation as a pre-step to aerobic biodegradation to achieve lower residual NAs concentrations in refinery effluents and thus lower toxicity, or in other low NA concentration wastewater streams. Both processes (ozonation and aerobic degradation) have been widely implemented worldwide for other applications; thus, they are good candidates for achieving a high degree of NAs removal.

4. Conclusions Ozonation of the TCI NAs mixture followed pseudo first-order kinetics, but a residual NAs concentration was not degraded. Ozone reacted preferentially with NAs of higher cyclicity and decreased the number of rings and MW. Ozonation of model compounds confirmed that the ozone reactivity with NAs depends on the number of H atoms available for abstraction, the position of alkyl substituents in relation to the carboxylic group, and branching. Biodegradation of the unozonated NA mixture resulted in a significant toxicity reduction, but the combination of ozonation-biodegradation resulted in an even higher toxicity reduction, showing a synergistic effect of the two processes. Acknowledgements Support by the Fulbright Foundation Greece and the Prefecture of Kavala, Region of Eastern Macedonia and Thrace, Greece to E. Vaiopoulou is acknowledged. References Afzal, A., Drzewicz, P., Perez-Estrada, L., Chen, Y., Martin, J.W., Gamal El-Din, M., 2012. Effect of molecular structure on the reactivity of naphthenic acids in the UV/H2O2 advanced oxidation process. Environ. Sci. Technol. 46 (19), 10727– 10734. American Public Health Association (APHA), 2012. Standard Methods for the Examination of Water and Wastewater, 22nd ed. APHA–AWWA–WEF, Washington, DC. Brown, L.D., Perez-Estrada, L., Wang, N., Gamal El-Din, M., Martin, J.W., Fedorak, P.M., Ulrich, A.C., 2013. Indigenous microbes survive in situ ozonation improving biodegradation of dissolved organic matter in aged oil sands process-affected waters. Chemosphere 93 (11), 2748–2755. Choi, J., Hwang, G., Gamal El-Din, M.G., Liu, Y., 2014. Effect of reactor configuration and microbial characteristics on biofilm reactors for oil sands process-affected water treatment. Int. Biodeterior. Biodegrad. 89, 74–81. Dorn, P.B., 1998. Case histories – The petroleum industry. In: Ford, D.L. (Ed.), Toxicity Reduction: Evaluation and Control. Water Quality Management Library, vol. 3. Technomic Publishing Company, Lancaster, PA, pp. 183–223.

E. Vaiopoulou et al. / Bioresource Technology 179 (2015) 339–347 Frank, R.A., Fischer, K., Kavanagh, R., Burnison, B.K., Arsenault, G., Headley, J.V., Peru, K.M., Van der Kraak, G., Solomon, K.R., 2009. Effect of carboxylic acid content on the acute toxicity of oil sands naphthenic acids. Environ. Sci. Technol. 43 (2), 266–271. Gamal El-Din, M., Fu, H.J., Wang, N., Chelme-Ayala, P., Perez-Estrada, L., Drzewicz, P., Martin, J.W., Zubot, W., Smith, D.W., 2011. Naphthenic acids speciation and removal during petroleum-coke adsorption and ozonation of oil sands processaffected water. Sci. Total Environ. 409 (23), 5119–5125. Han, X.M., Scott, A.C., Fedorak, P.M., Bataineh, M., Martin, J.W., 2008. Influence of molecular structure on the biodegradability of naphthenic acids. Environ. Sci. Technol. 42 (4), 1290–1295. Headley, J.V., Peru, K.M., Mishra, S., Meda, V., Dalai, A.K., Mc-Martin, D.W., Mapolelo, M.M., Rodgers, R.P., Marshall, A.G., 2010. Characterization of oil sands naphthenic acids treated with ultra-violet and microwave radiation by negative ion electrospray Fourier transform ion cyclotron resonance mass spectrometry. Rapid Commun. Mass Spectrom. 24 (21), 3121–3126. Headley, J.V., Barrow, M.P., Peru, K.M., Fahlman, B., Frank, R.A., Bickerton, G., McMaster, M.E., Parrott, J., Hewitt, L.M., 2011a. Preliminary fingerprinting of Athabasca oil sands polar organics in environmental samples using electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry. Rapid Commun. Mass Spectrom. 25 (13), 1899–1909. Headley, J.V., Barrow, M.P., Peru, K.M., Derrick, P.J., 2011b. Salting-out effects on the characterization of naphthenic acids from Athabasca oil sand using electrospray ionization. J. Environ. Sci. Health., Part A 46 (8), 844–854. Headley, J.V., Peru, K.M., Mohamed, M.H., Frank, R.A., Martin, J.W., Hazewinkel, R.R., Humphries, D., Gurprasad, N.P., Hewitt, L.M., Muir, D.C., Lindeman, D., Strub, R., Young, R.F., Grewer, D.M., Whittal, R.M., Fedorak, P.M., Birkholz, D.A., Hindle, R., Reisdorph, R., Wang, X., Kasperski, K.L., Hamilton, C., Woudneh, M., Wang, G., Loescher, B., Farwell, A., Dixon, D.G., Ross, M., Pereira, A.S., King, E., Barrow, M.P., Fahlman, B., Bailey, J., McMartin, D.W., Borchers, C.H., Ryan, C.H., Toor, N.S., Gillis, H.M., Zuin, L., Bickerton, G., McMaster, M., Sverko, E., Shang, D., Wilson, L.D., Wrona, F.J., 2013. Chemical fingerprinting of naphthenic acids and oil sands process waters – a review of analytical methods for environmental samples. J. Environ. Sci. Health Part A 48 (10), 1145–1163. Hwang, G., Dong, T., Islam, S.Md., Sheng, Z., Pérez-Estrada, L.A., Liu, Y., Gamal El-Din, M., 2013. The impacts of ozonation on oil sands process-affected water biodegradability and biofilm formation characteristics in bioreactors. Bioresour. Technol. 130, 269–277. Kannel, P.R., Gan, T.Y., 2012. Naphthenic acids degradation and toxicity mitigation in tailings wastewater systems and aquatic environments: a review. J. Environ. Sci. Health., Part A 47 (1), 1–21. Martin, J.W., Barri, T., Han, X., Fedorak, P.M., Gamal El-Din, M., Perez, L., Scott, A.C., Jiang, J.T., 2010. Ozonation of oil sands process-affected water accelerates microbial bioremediation. Environ. Sci. Technol. 44 (21), 8350–8356. Minakata, D., Li, K., Westerhoff, P., Crittenden, J., 2009. Development of a group contribution method to predict aqueous phase hydroxyl radical (HO) reaction rate constants. Environ. Sci. Technol. 43 (16), 6220–6227. Mishra, S., Meda, V., Dalai, A.K., Mc-Martin, D.W., Headley, J.V., 2010. Photocatalysis of naphthenic acids in water. J. Water Resour. Prot. 2 (7), 644–650. Misiti, T.M., 2012. Fate and Effect of Naphthenic Acids in Biological Systems (PhD Thesis). School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA. Misiti, T.M., Tandukar, M., Tezel, U., Pavlostathis, S.G., 2013a. Inhibition and biotransformation potential of naphthenic acids under different electron accepting conditions. Water Res. 47 (1), 406–418.

347

Misiti, T.M., Tezel, U., Pavlostathis, S.G., 2013b. Fate and effect of naphthenic acids on oil refinery activated sludge wastewater treatment systems. Water Res. 47 (1), 449–460. Misiti, T.M., Tezel, U., Tandukar, M., Pavlostathis, S.G., 2013c. Aerobic biotransformation potential of a commercial mixture of naphthenic acids. Water Res. 47 (15), 5520–5534. Misiti, T.M., Tezel, U., Pavlostathis, S.G., 2014. Effect of alkyl side chain location and cyclicity on the aerobic biotransformation of naphthenic acids. Environ. Sci. Technol. 48 (14), 7909–7917. Nöthe, T., Fahlenkamp, H., von Sonntag, C., 2009. Ozonation of wastewater: rate of ozone consumption and hydroxyl radical yield. Environ. Sci. Technol. 43 (15), 5990–5995. Pereira, A.S., Bhattacharjee, S., Martin, J.W., 2013a. Characterization of oil sands process-affected waters by liquid chromatography Orbitrap mass spectrometry. Environ. Sci. Technol. 47 (10), 5504–5513. Pereira, A.S., Islam Shahinoor, M.D., Gamal El-Din, M.S., Martin, J.W., 2013b. Ozonation degrades all detectable organic compound classes in oil sands process-affected water; an application of high-performance liquid chromatography/Orbitrap mass spectrometry. Rapid Commun. Mass Spectrom. 27 (21), 2317–2326. Perez-Estrada, L.A., Han, X.M., Drzewicz, P., Gamal El-Din, M., Fedorak, P.M., Martin, J.W., 2011. Structure–reactivity of naphthenic acids in the ozonation process. Environ. Sci. Technol. 45 (17), 7431–7437. Rakness, K., Gordon, G., Langlais, B., Masschelein, W., Matsumoto, N., Richard, Y., Robson, C.M., Somiya, I., 1996. Guideline for measurement of ozone concentration in the process gas from an ozone generator. Ozone Sci. Eng. 18 (3), 209–229. Rivas, F.J., Beltran, F.J., Encinas, A., 2012. Removal of emergent contaminants: integration of ozone and photocatalysis. J. Environ. Manage. 100, 10–15. Scott, A.C., MacKinnon, M.D., Fedorak, P.M., 2005. Naphthenic acids in Athabasca oil sands tailings waters are less biodegradable than commercial naphthenic acids. Environ. Sci. Technol. 39 (21), 8388–8394. Scott, A.C., Zubot, W., MacKinnon, M.D., Smith, D.W., Fedorak, P.M., 2008. Ozonation of oil sands process water removes naphthenic acids and toxicity. Chemosphere 71 (1), 156–160. Smith, B.E., Lewis, C.A., Belt, S.T., Whitby, C., Rowland, S.J., 2008. Effects of alkyl chain branching on the biotransformation of naphthenic acids. Environ. Sci. Technol. 42 (24), 9323–9328. Sun, N., Chelme-Ayala, P., Klamerth, N., McPhedran, K.N., Islam, M.S., Perez-Estrada, L., Drzewicz, P., Blunt, B.J., Reichert, M., Hagen, M., Tierney, K.B., Belosevic, M., Gamal El-Din, M., 2014. Advanced analytical mass spectrometric techniques and bioassays to characterize untreated and ozonated oil sands process-affected water. Environ. Sci. Technol. 48 (19), 11090–11099. Wang, B.L., Wan, Y., Gao, Y., Yang, M., Hu, J., 2013a. Determination and characterization of oxy-naphthenic acids in oilfield wastewater. Environ. Sci. Technol. 47 (16), 9545–9554. Wang, N., Chelme-Ayala, P., Perez-Estrada, L., Garcia-Garcia, E., Pun, J., Martin, J.W., Belosevic, M., Gamal El-Din, M., 2013b. Impact of ozonation on naphthenic acids speciation and toxicity of oil sands process-affected water to Vibrio fischeri and mammalian immune system. Environ. Sci. Technol. 47 (12), 6518–6526. Whitby, C., 2010. Microbial naphthenic acid degradation. Adv. Appl. Microbiol. 70, 93–125.