Removal of dinitrotoluenes from water via reduction with iron and peroxidase-catalyzed oxidative polymerization: A comparison between Arthromyces ramosus peroxidase and soybean peroxidase

Chemosphere 67 (2007) 1485–1491 www.elsevier.com/locate/chemosphere

Removal of dinitrotoluenes from water via reduction with iron and peroxidase-catalyzed oxidative polymerization: A comparison between Arthromyces ramosus peroxidase and soybean peroxidase Joey Patapas a, Mohammad Mousa Al-Ansari a, K.E. Taylor b

a,*

, J.K. Bewtra b, N. Biswas

b

a Department of Chemistry and Biochemistry, University of Windsor, 401 Sunset Avenue, Windsor, Ont., Canada N9B 3P4 Department of Civil and Environmental Engineering, University of Windsor, 401 Sunset Avenue, Windsor, Ont., Canada N9B 3P4

Received 13 October 2006; received in revised form 11 December 2006; accepted 11 December 2006 Available online 30 January 2007

Abstract A two-step process for the removal of dinitrotoluene from water is presented: zero-valent iron reduction is coupled with peroxidasecatalyzed polymerization of the resulting diaminotoluenes (DAT). The effect of pH was examined in the reduction step: at pH 6 the reaction occurred much more rapidly than at pH 8. In the second step, optimal pH and substrate ratio, minimal enzyme concentration and effect of polyethylene glycol (PEG) as an additive for greater than 95% conversion of DAT, over a 3 h reaction period were determined using high performance liquid chromatography. Two enzymes were investigated and compared: Arthromyces ramosus peroxidase (ARP) and soybean peroxidase (SBP). The optimal pH values were 5.4 and 5.2 for ARP and SBP, respectively, but SBP was more resistant to mild acid whereas ARP was more stable in neutral solutions. SBP was found to have a greater hydrogen peroxide demand (optimal peroxide/DAT molar ratio for SBP: 2.0 and 3.0 for 2,4-diaminotoluene (2,4-DAT) and 2,6-diaminotoluene (2,6-DAT), respectively; for ARP: 1.5 and 2.75 for 2,4-DAT and 2,6-DAT, respectively) but required significantly less enzyme (0.01 and 0.1 U ml1 for 2,4-DAT and 2,6-DAT, respectively) to convert the DAT than ARP (0.4 and 1.5 U ml1 for 2,4-DAT and 2,6-DAT, respectively). PEG was shown to have no effect upon the degree of substrate conversion for either enzyme.  2006 Elsevier Ltd. All rights reserved. Keywords: Dinitrotoluene; Diaminotoluene; Enzyme treatment; Arthromyces ramosus peroxidase; Soybean peroxidase; Zero-valent iron; Hazardous waste

1. Introduction Dinitrotoluene (DNT) is found in the effluents of a number of manufacturing and processing plants, primarily in one of two isomeric forms: 2,4-DNT and 2,6-DNT. Commercially it is used in the manufacture of dyes and as a precursor to toluene diisocyanate, which is used to man*

Corresponding author. Tel.: +1 519 253 3000x3917; fax: +1 519 971 3667. E-mail addresses: [email protected] (J. Patapas), [email protected] (M.M. Al-Ansari), [email protected] (K.E. Taylor), bewtra@ uwindsor.ca (J.K. Bewtra), [email protected] (N. Biswas). 0045-6535/$ - see front matter  2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2006.12.040

ufacture polyurethane foams. DNT is also used or arises in the manufacture of munitions, as a plasticizer and as a component in propellants (Doppalapudi et al., 2003; Tchounwou et al., 2003; Shin et al., 2005). Studies have shown DNT to be toxic to humans and animals, a carcinogen, and a cause of anaemia, nerve and liver damage (Tchounwou et al., 2003). The most prevalent method for removing DNT from wastewater involves adsorption onto activated carbon (Rodgers and Bunce, 2001; Doppalapudi et al., 2003). One major drawback of this method is that it does not actually treat DNT but simply removes it from the aqueous phase. The unaltered DNT is concentrated and the

1486

J. Patapas et al. / Chemosphere 67 (2007) 1485–1491

negative health effects remain. In addition, both the removed DNT and the spent carbon need to be disposed of, which increases the overall cost of this treatment technology. Other physical/chemical methods, such as ultrafiltration, reverse osmosis, liquid–liquid extraction or resin adsorption all suffer from similar drawbacks (Rodgers and Bunce, 2001). Biological methods have been proposed for degrading DNT (Spain, 1995; Rodgers and Bunce, 2001; Doppalapudi et al., 2003) and can be divided into two categories, anaerobic and aerobic. Anaerobic treatment methods reduce DNT to diaminotoluene (DAT) via nitroso, hydroxylamino and amino-nitrotoluene intermediates (Hughes et al., 1999; Vanderloop et al., 1999). These methods are limited in that the intermediates generated can kill the microorganisms before remediation is complete and that the DAT-containing effluent is unacceptable for discharge. Aerobic treatment methods involve microorganisms such as Pseudomonas sp. (Spanggord et al., 1991), Burkholderia sp. (So et al., 2004) or a microbial consortium (Snellinx et al., 2003) and have been shown to degrade DNT. Though more effective than anaerobic biological treatment methods, aerobic biological treatment methods generate a wide range of degradation products which are not completely mineralized (Rodgers and Bunce, 2001). Fungi, such as P. chrysosporium, have been widely studied as a means of treating nitroaromatics, including DNT (Spain, 1995). These studies have shown that the enzymes and small molecules produced from these fungi are capable of mineralizing the nitroaromatic contaminants. An alternative biological method to using microbial organisms is the employment of an enzyme-catalyzed polymerization and precipitation process. This process has several advantages over conventional biological treatment methods, including: action on, or in the presence of, chemical species which are toxic to microbes; operation on a broad range of compounds; operation over a variety of reaction conditions (wide pH, temperature and salinity ranges); operation at both high and low concentrations of contaminants; reduction of sludge volume due to significantly less waste being produced; and no shock loading effect or delays with start up or shutdown due to acclimatization of microbes (Nicell et al., 1992). This method has been shown to be effective in the treatment of several phenolic compounds including phenol (Nicell et al., 1992; Al-Kassim et al., 1993, 1994), cresols (Caza et al., 1999; Steevensz et al., 2006), chlorophenols (Siddique et al., 1993) and most recently, bisphenol A (Modaressi et al., 2005). Nitrobenzene and nitrotoluenes have also been treated via this method once they have been reduced to the corresponding aniline species (Mantha et al., 2002). The bulk of this research involves peroxidases from several sources: horseradish (Nicell et al., 1992; Siddique et al., 1993), Arthromyces ramosus (Al-Kassim et al., 1993; Ibrahim et al., 2001), Coprinus macrorhizus (Al-Kassim et al., 1994) and soybean (Caza et al., 1999; Mantha et al., 2002).

The enzymes follow a modified ping-pong mechanism, often referred to as peroxidase ping-pong (Dunford, 1999), as follows: E þ H2 O2 ! E–I þ H2 O 

E–I þ AH2 ! E–II þ AH E–II þ AH2 ! E þ  AH þ H2 O

ð1Þ ð2Þ ð3Þ

The native form of the enzyme (E) will reduce hydrogen peroxide to water and is converted to Compound I (E–I), which is the active form of the enzyme. Compound I can oxidize an aromatic compound (AH2), generating an aromatic radical (AH). The enzyme is now in the Compound II form, which oxidizes a second aromatic molecule, generating a second aromatic radical and a second molecule of water. Upon release of the water molecule, the enzyme reverts back to the native form, completing the enzymatic cycle. The aromatic radicals having diffused from the active site of the enzyme will combine non-enzymatically to form dimers. If these dimers are soluble, then they too will be substrates (as long as they are phenolic or anilino compounds) and acted upon by the peroxidase, forming higher oligomers. This process will continue until the polymer generated reaches its solubility limit and precipitates from solution. The work presented here demonstrates a two-step treatment process for 2,4-DNT and 2,6-DNT. The first step involves reduction of DNT to DAT using zero-valent iron. This is a necessary first-step in the enzymatic treatment of nitroaromatics since they are not substrates of the enzyme (Mantha et al., 2002). Nitroaromatic species which are also phenolic or anilino compounds, such as p-nitrophenol or nitroaniline, must be reduced as well, since they too are not substrates of the enzyme. The second step is the enzyme-catalyzed polymerization of the DAT generated. Two different peroxidases were studied for comparison: A. ramosus peroxidase (ARP), a class II plant peroxidase (originating from a fungal source) and soybean peroxidase (SBP), a class III plant peroxidase (originating from a plant source). The aspects of interest include the optimal operating pH, the optimal molar ratio between hydrogen peroxide and DAT and the minimum enzyme concentration required for DAT treatment (95% removal within 3 h). 2. Materials and methods 2.1. Materials ARP was a developmental preparation (SP-502, activity 2000 U ml1, by the activity test below) of Novozymes Inc., (Franklinton, NC) obtained from them as a generous gift. SBP (Industrial grade, Lot # 18541NX, activity 8000 purpurogallin units g1) was purchased from Organic Technologies (Coshocton, OH). Catalase (EC 1.11.1.6, Lot # 81H7146, activity 2000 U mg1) was a gift from Novo Nordisk Bioindustries Inc. (Danbury, CT). 2,4DNT, 2,6-DNT, 2,4-DAT, 2,6-DAT, 4-aminoantipyrine (AAP) and phenol (all having a purity of 97% or greater)

J. Patapas et al. / Chemosphere 67 (2007) 1485–1491

were all supplied by Aldrich Chemical Co. (Milwaukee, WI). Monobasic sodium phosphate, dibasic sodium phosphate, sodium sulphite (all with 98% purity or greater) and hydrogen peroxide (30% w/v) were purchased from BDH Inc. (Toronto, ON). Cobalt chloride, sodium acetate (all with 99% purity) and iron filings (about 0.45 mm size) were provided by Fisher Scientific Co. (Fair Lawn, NJ). Glacial acetic acid (99% purity) and concentrated hydrochloric acid were obtained from ACP (Montreal, QC). Syringe filters (bulk, non-sterile, 0.2 lm) were purchased from Pall Gelman Laboratories (Mississauga, ON). UV–vis absorbances were measured using a Hewlett Packard Diode Array Spectrophotometer model 8452A (wavelength range 190–820 nm, 2 nm resolution), controlled by a Hewlett Packard Vectra ES/12 computer. Quartz cells with a 10 mm path length were purchased from Hellma Canada Limited (Concord, ON). High performance liquid chromatography (HPLC) analysis was accomplished with a Waters system HPLC (model 1525 binary HPLC pump, model 717 auto-sampler, model 2487 dual k absorbance detector) equipped with a C18 column (5 lm, 4.6 · 150 mm) and operated by Breeze software. Elutions were isocratic with a mobile phase comprized of methanol and water in a 60:40 ratio. Flow-rate was maintained at 1.0 ml min1, the injection volume was 10 ll and column temperature was 40 C. The UV–vis detector was set at 294 nm for 2,4-DAT and 2,4-DNT and 286 nm for 2,6-DAT and 2,6-DNT. 3. Methods 3.1. Buffer preparation The buffers used in this study were prepared according to Gomori (1955) in the pH range 4.0–8.0 (acetic acid– sodium acetate for pH 4.0–5.8, monobasic–dibasic sodium phosphate for pH 5.6–8.0). A pH meter was used to measure the pH of the mixture in a batch reactor. 3.2. Iron pre-treatment Iron filings were pre-treated as previously described (Agrawal and Tratnyek, 1996; Mantha et al., 2001). The iron was washed with hydrochloric acid (10% v/v) to remove any oxides and hydroxides from the metal surface. Subsequently, the excess HCl was removed by washing four times with carbonate buffer (0.05 M, pH 9.5) and made anaerobic by including sodium sulphite. To remove any excess alkalinity and prevent contact with oxygen, the iron was finally washed and stored in a sodium sulphite solution (20 mM, 0.1% w/w cobalt chloride). 3.3. Determination of pKa values for diaminotoluene The pKa values for the DATs were measured by titrating a 50 ml solution of 0.1 M DAT with 0.1 M HCl. Changes in pH were monitored with a pH meter.

1487

3.4. Enzyme activity assay The peroxidase activity was determined by colorimetric monitoring of the oxidative coupling between AAP and phenol in the presence of hydrogen peroxide under enzyme catalysis, as was previously described (Ibrahim et al., 2001). Enzyme activity is reported as standard units of catalytic activity per millilitre (U ml1), where 1.0 U = 1.0 lmol hydrogen peroxide converted per min. 3.5. Experimental protocol All reactions were conducted in triplicate (results shown are an average of the triplicated experimental data) at 20 C in 30 ml batch reactors. Reduction of 1.0 mM solutions of DNT, in a total volume of 20 ml, was accomplished with the appropriate amount of pre-treated iron. The solutions were made anaerobic by the presence of 1.0 mM sodium sulphite and a catalytic amount (0.1% w/w relative to sodium sulfite) of cobalt chloride. The batch reactors were sealed with rubber septa and shaken at maximum setting on a Burrel Model 75 wrist action shaker (Pittsburgh, PA). Solutions were then microfiltered to remove the iron filings. DNT and DAT concentrations were measured via HPLC as described above. The catalyzed polymerization step involved buffered 1.0 mM solutions of DAT, with appropriate amount of hydrogen peroxide and a final volume of 20 ml. Peroxidase was added to initiate the reaction. The open batch reactors were stirred with a stir bar and a magnetic stirrer for 3 h. To quench the reaction, 100 ll of catalase (200 ml1) was added to the batch reactor. All samples were microfiltered prior to concentration measurements via HPLC. 4. Results and discussion 4.1. Reduction of dinitrotoluene with iron The two most abundant isomers of DNT (2,4-DNT and 2,6-DNT) were used in this study. The reduction of these species yields the corresponding DAT (2,4-DAT and 2,6-DAT). Each nitro group is reduced to an amino group via nitroso and hydroxylamino intermediate species. This generates 2-amino-4-nitrotoluene and 4-amino-2nitrotoluene from 2,4-DNT and 2-amino-6-nitrotoluene from 2,6-DNT (Agrawal and Tratnyek, 1996; Choe et al., 2001; Oh et al., 2002). The second nitro group on each amino-nitrotoluene is subsequently reduced, via the analogous intermediate species, to generate the corresponding DAT. The need for iron pre-treatment and the maintenance of anaerobic conditions to achieve complete DNT reduction are well documented in the literature (Agrawal and Tratnyek, 1996; Mantha et al., 2001). The reagents used to create the anaerobic conditions do not interfere in the reduction reaction, but it has been reported that in their absence, 100% reduction of the nitroaromatics does not

J. Patapas et al. / Chemosphere 67 (2007) 1485–1491

occur (Mantha et al., 2001). HPLC studies show the consumption of DNT and the gradual formation of the intermediate amino-nitrotoluenes and DAT. Nitroso and hydroxylamino intermediates were not detected in this study. These intermediates must be present but are usually adsorbed on the iron surface and are short-lived, thus are seldom detected (Agrawal and Tratnyek, 1996). After the reduction reaction is complete, it has been reported that the solution contains sufficient ferrous iron and residual sodium sulphite to interfere with the enzymatic treatment (Mantha et al., 2002). Sodium sulphite has hydrogen peroxide demand, while ferrous iron will react with hydrogen peroxide, producing hydroxyl radicals in a mixture known as Fenton’s reagent (Rodgers and Bunce, 2001; Mantha et al., 2002). Aeration of this solution prior to the enzymatic step is necessary to oxidize the ferrous iron to less soluble ferric iron and convert sodium sulphite to sodium sulphate (Mantha et al., 2002). 4.2. Effect of pH on the reduction of dinitrotoluene The overall reaction for the reduction of DNT to DAT can be expressed as DNT þ 6Fe0 þ 12Hþ ! DAT þ 6Fe2þ þ 4H2 O

Concentration (mM)

Because of the involvement of hydrogen ions in the reduction, it was believed that changes in pH would affect the reaction. Batch reactors containing 1.0 g of pre-treated iron were used to reduce a 1.0 mM solution of DNT buffered in either an acidic or alkaline medium. The consumption of DNT and formation of DAT with respect to time at pH values of 6.0 and 8.0 are shown in Figs. 1 and 2, respectively. In the slightly acidic medium, all of the DNT is converted within 20 min and over 0.75 mM of DAT was detected in solution, accounting for over 75% of the theoretical DAT produced. As no further intermediate products were detected in the chromatogram after 20 min, the reduction is considered complete. The neutral, free-base form of the amino groups on DAT are capable of acting as ligands 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0

2

4

6

8

10

12

14

16

18

20

Time (minutes) Fig. 1. Consumption of DNT and formation of DAT with respect to time at pH 6.0. Conditions: 1.0 mM DNT solution in batch reactor containing 1.0 g of iron, buffered with 0.01 M phosphate buffer with a pH of 6.0. () 2,4-DNT, (j) 2,6-DNT, (m) 2,4-DAT and (d) 2,6-DAT.

Concentration (mM)

1488

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0

20

40

60

80

100

120

140

160

180

Time (minutes) Fig. 2. Consumption of DNT and formation of DAT with respect to time at pH 8.0. Conditions: 1.0 mM DNT solution in batch reactor containing 1.0 g of iron, buffered with 0.01 M phosphate buffer with a pH of 8.0. () 2,4-DNT, (j) 2,6-DNT, (m) 2,4-DAT and (d) 2,6-DAT.

and may complex with the metal (Vasudevan and Stone, 1996; Mantha et al., 2001). This adsorbed DAT may account for the missing theoretical DAT not seen in the chromatogram. In a slightly alkaline medium, however, the reaction occurs much more slowly, with over 0.2 mM of DNT (over 20%) remaining after 3 h and only 19% of the theoretical 2,4-DAT and 27% of the theoretical 2,6-DAT detected in solution. An unbuffered DNT solution has an initial pH near neutrality, however, this increased by 3.5 pH units over the course of a 3 h reaction. In a similar study involving the reduction of nitrobenzene to aniline, an increase of two pH units was attributed to the aqueous corrosion of the metal (Mantha et al., 2001). The unbuffered solution exhibited similar DNT consumption and corresponding DAT formation as the alkaline buffered solution. This suggests that buffering of the initial solution would be required to prevent pH increase and the resulting slowdown of DNT reduction. To ascertain if DNT reduction has a dependence on pH in the acidic range, batch reactors containing 150 mg of pre-treated iron were used to reduce 1.0 mM solutions of DNT, buffered between pH 3.0 and 7.0, over 5 min. The smaller quantity of iron and the short reaction time where chosen to emphasize any pH-dependence that may exist. The results of this study (data not shown) indicate that within the acidic range there is a marginal increase in DNT consumption with decreased pH. This lack of a significant pH-dependence within the acidic range can be attributed to rates which are limited by the mass transport to the reactive sites on the iron as opposed to being limited by nitro group reduction (Agrawal and Tratnyek, 1996). The tracking of DAT formation over this pH range was not conducted. It was felt that such results could not indicate an accurate pH effect for DAT formation due to DAT adsorption via ligation between the amino groups and the metal (Vasudevan and Stone, 1996; Mantha et al., 2001) and the complete adsorption of protonated DAT below the pKa of 5.12 for 2,4-DAT and 4.75 for 2,6-DAT (Agrawal and Tratnyek, 1996).

J. Patapas et al. / Chemosphere 67 (2007) 1485–1491

5. Enzymatic treatment 5.1. Effect of pH on enzymatic treatment Enzyme function may be affected if catalytically relevant amino acid residues are in the wrong ionization state or if structural modification occurs as a result of ionization changes (Petsko and Ringe, 2004). The pH range studied was between 4.0 and 10.0, with a fixed peroxide/substrate ratio of 1.5 for 2,4-DAT and ARP, 2.75 for 2,6-DAT and ARP, 2.0 for 2,4-DAT and SBP and 3.0 for 2,6-DAT and SBP (optimal substrate ratios as described below). Reduced enzyme concentrations (stress conditions) were used to emphasize any effects caused by pH changes (0.05 U ml1 ARP and 0.001 U ml1 SBP over the broad range and 0.025 U ml1 ARP and 0.005 U ml1 SBP over the narrow range). An optimal pH of 5.4 is obtained for both DAT isomers when ARP is used and an optimal pH of 5.2 is observed for both isomers when SBP is the oxidizing enzyme. In Fig. 3, it can be seen that SBP is sensitive to neutral and alkaline pH, as exhibited by the decrease in substrate conversion, whereas ARP is less sensitive to these pH changes, and the decrease in substrate conversion is less pronounced. This sensitivity in SBP can be attributed to the lower stability of the Compound I form of SBP in a neutral solution, which degrades quickly compared to the Compound I forms of other peroxidases (Nissum et al., 2001). Conversely, SBP exhibits unusual stability in acidic medium (Henriksen et al., 2001; Nissum et al., 2001) and as a result, there is a less pronounced decrease in substrate conversion compared to that with ARP. The catalytic distal histidine side chain has a pKa of 3.2 in SBP (Nissum et al., 2001) whereas the same in ARP has a pKa of 5.1 (Dunford, 1999). This makes ARP more susceptible to inactivation in an acidic medium than SBP.

5.2. Optimal peroxide/substrate ratio and minimum enzyme concentration

1489

(Dunford, 1999). However, soluble dimers and oligomers are substrates of the enzyme and their conversion to higher polymers will increase the optimal hydrogen peroxide– DAT molar ratio above the stoichiometric ratio (Nicell et al., 1992). In the limit of an infinite polymer, the expected stoichiometry would be 1 mol of hydrogen peroxide per mol of amino groups, or 2 mol of peroxide per mol of DAT. In addition, peroxidases exhibit catalase activity, which is the reduction of hydrogen peroxide to water outside of the normal enzymatic cycle (Dunford, 1999). This too will add to the hydrogen peroxide demand, but is dependant on the concentration of the enzyme in solution. As higher concentrations of enzyme will consume larger portions of hydrogen peroxide via catalase activity, the optimal hydrogen peroxide–DAT molar ratio may increase with enzyme concentration. Fig. 4 illustrates the effect of variable hydrogen peroxide/DAT molar ratios on the treatment of 2,4-DAT with ARP at different enzyme concentrations. The increase in the optimal substrate molar ratio with increased enzyme concentration is indicative of increased hydrogen peroxide demand, and increases from 1.0 for 0.1 U ml1 of enzyme to 1.5 for 0.4 U ml1 of enzyme. Controlling the amount of hydrogen peroxide is crucial for optimal operating conditions. Since hydrogen peroxide is required to convert the enzyme to Compound I, insufficient concentrations of peroxide will hinder the overall activity of the enzyme (Dunford, 1999). Conversely, an excess of hydrogen peroxide will reduce activity by converting the enzyme to the catalytically slow Compound III (Nicell and Wright, 1997; Dunford, 1999). These trends are evident from Fig. 4 as the amount of residual DAT increases above the optimal hydrogen peroxide/DAT molar ratio. Table 1 summarizes the results from the optimal substrate ratio studies for the two enzymes with both DAT isomers. SBP is shown to have a greater peroxide demand than ARP but requires significantly less enzyme for both DAT isomers. This is of particular interest since enzyme cost is a limiting factor in the application of enzymes in

Stoichiometrically, for every mol of hydrogen peroxide consumed, 2 mol of DAT amino groups are converted 60

% DAT Remaining

% DAT Remaining

100 90 80 70 60 50 40 30 3.5

4

4.5

5

5.5

6

6.5

7

7.5

8

pH Fig. 3. Effect of pH on DATe conversion in the enzymic step. Conditions: 1.0 mM 2,4-DAT in batch reactor, buffered with 0.01 M acetate or phosphate buffer, containing 0.05 U ml1 ARP or 0.001 U ml1 SBP, after 3 h reaction with the enzyme. () 2,4-DAT with ARP, (j) 2,6-DAT with ARP, (m) 2,4-DAT with SBP and (d) 2,6-DAT with SBP.

50 40 30 20 10 0 0.5

1

1.5

2

2.5

3

3.5

4

Hydrogen Peroxide/DAT Molar Ratio Fig. 4. Effect of variable peroxide concentration on the conversion of 2,4DAT as a function of enzyme concentration. Conditions: 1.0 mM 2,4DAT in batch reactor, buffered with 0.01 M acetate buffer, pH 5.4 for ARP and 5.2 for SBP, after 3 h reaction. (j) 0.1 U ml1, (d) 0.2 U ml1, () 0.3 U ml1, (m) 0.4 U ml1 ARP and (s) 0.01 U ml1 SBP.

1490

J. Patapas et al. / Chemosphere 67 (2007) 1485–1491

Table 1 Optimal hydrogen peroxide–DAT molar ratio and minimum enzyme concentration to achieve over 95% conversion of DAT over 3 h at pH 5.4 for ARP and 5.2 for SBP Enzyme

Substrate

H2O2/DAT molar ratio

Minimum enzyme conc. (U ml1)

ARP

2,4-DAT 2,6-DAT

1.5 2.75

0.4 1.5

SBP

2,4-DAT 2,6-DAT

2.0 3.0

0.01 0.1

wastewater treatment (Ibrahim et al., 2001; Modaressi et al., 2005). In addition, SBP does not exhibit a significant reduction in activity in the presence of excess hydrogen peroxide (Fig. 4). This stability towards hydrogen peroxide is likely responsible for the low enzyme concentration requirements. Catalase activity has been reported as the main pathway by which enzymes protect themselves from peroxide inactivation (Hiner et al., 2002). Greater catalase activity from SBP compared with ARP could account for the observed results, including the increased peroxide demand for SBP at optimal hydrogen peroxide–DAT molar ratios. Alternatively, the crude SBP could have a greater chemical oxygen demand. 5.3. Effect of polyethylene glycol additive Polyethylene glycol (PEG), when used as an additive, has been reported to prolong enzyme activity, thus greatly reducing the amount of enzyme required. PEG has been shown to provide various levels of protection to horseradish peroxidase (Wu et al., 1993), SBP (Caza et al., 1999), ARP (Ibrahim et al., 2001) and laccase (Modaressi et al., 2005), however, in all these studies the substrate was either phenol or a phenolic compound. Studies involving the treatment of aniline with SBP showed no prolongation of enzyme activity upon the addition of PEG (Mantha, 2001). Experiments were conducted in which variable concentrations of PEG, up to 250 mg l1, were added to the batch reactors. Optimal pH (5.4 and 5.2 for ARP and SBP, respectively) and peroxide/substrate ratio (as reported in Table 1) were maintained for the appropriate substrate and enzyme. Reduced enzyme concentration (0.05 U ml1 ARP and 0.001 U ml1 SBP) was used to emphasize any PEG effect. The results from these experiments showed no improvement between the control batch reactor and batch reactors with PEG additives (data not shown). This would suggest that suppression of enzyme inactivation by PEG is substratedependant, and while PEG is effective when the substrate is phenolic, this is not the case when the substrate is anilino. 6. Conclusions The results from the batch reactor studies have demonstrated that the reduction of DNT to DAT occurs quickly in slightly acidic solutions, requiring less than 20 min, however, DNTs react much more slowly in slightly alkaline

solutions. There was very little pH dependence on the degree of reduction within the acidic range of the pH scale and this is attributed to reactions which are limited by the mass transport to the reactive sites on the metal. Both ARP and SBP are suitable enzymes for the polymerization of DAT, however, based on the comparison experiments between them the following conclusions can be drawn: The optimal operating pH for ARP is 5.4 and for SBP it is 5.2, regardless of which DAT isomer is being treated. As the pH approaches neutrality there is a decrease in SBP activity due to the Compound I form of SBP being less stable at those pH values. Conversely, SBP is inactivated less than ARP in acidic solution due to the lower pKa of its active site. ARP has a lower hydrogen peroxide demand than SBP, however, SBP requires significantly less enzyme to treat DAT (40-fold less for 2,4-DAT and 15-fold less for 2,6DAT). This out-performance of ARP by SBP is of particular interest, since the cost of the enzyme is a limiting economic factor to the application of this technology. This stability of SBP towards inactivation might be attributed to a greater catalase activity than ARP, which is a primary means by which enzymes protect themselves from peroxide inactivation. This would also account for the increased hydrogen peroxide demand seen when SBP is employed. PEG as an additive did not show any increase in the enzyme activity compared with the control batch reactor. Since PEG has been reported to act as a protective additive for both enzymes when the substrate is phenolic, these results suggest that PEG’s effectiveness is substratedependant. Acknowledgements The authors thank the Natural Sciences and Engineering Research Council of Canada, the Department of Chemistry and Biochemistry and the Faculty of Graduate Studies and Research of the University of Windsor for their generous funding and support. References Agrawal, A., Tratnyek, P.G., 1996. Reduction of nitro aromatic compounds by zero-valent iron metal. Environ. Sci. Technol. 30, 153–160. Al-Kassim, L., Taylor, K.E., Bewtra, J.K., Biswas, N., 1993. Aromatics removal from water by it Arthromyces ramosus peroxidase. In: Welinder, K.G., Rasmussen, S.K., Penel, C., Greppin, H. (Eds.), Plant Peroxidases: Biochemistry and Physiology. University of Geneva, pp. 197–200. Al-Kassim, L., Taylor, K.E., Nicell, J.A., Bewtra, J.K., Biswas, N., 1994. Enzymatic removal of selected aromatic contaminants from wastewater by a fungal peroxidase from Coprinus macrorhizus in batch reactors. J. Chem. Technol. Biotechnol. 61, 179–182. Caza, N., Bewtra, J.K., Biswas, N., Taylor, K.E., 1999. Removal of phenolic compounds from synthetic wastewater using soybean peroxidase. Water Res. 33, 3012–3018. Choe, S., Lee, S.H., Chang, Y.Y., Hwang, K.Y., Khim, J., 2001. Rapid reductive destruction of hazardous organic compounds by nanoscale Fe0. Chemosphere 42, 367–372.

J. Patapas et al. / Chemosphere 67 (2007) 1485–1491 Doppalapudi, R.B., Sorial, G.A., Maloney, S.W., 2003. Electrochemical reduction of 2,4-dinitrotoluene in a continuous flow laboratory scale reactor. J. Environ. Eng. 129, 192–201. Dunford, H.B., 1999. Heme Peroxidases. John Wiley and Sons Inc., New York, pp. 1–57, 299–304. Gomori, G., 1955. Preparation of buffers for use in enzyme studies. In: Colowick, S.P., Kaplan, N.O. (Eds.). In: Methods in Enzymology, vol. II. Academic Press Inc., New York, pp. 138–146. Henriksen, A., Mirza, O., Indiani, C., Teilum, K., Smulevich, G., Welinder, K.G., Gajhede, M., 2001. Structure of soybean seed coat peroxidase: a plant peroxidase with unusual stability and haem– apoprotein interactions. Protein Sci. 10, 108–115. Hiner, A.N.P., Hernander-Ruiz, J., Rodriguez-Lopez, J.N., GarciaCanovas, F., Brisset, N.C., Smith, A.T., Arnao, M.B., Acosta, M., 2002. Reactions of class II peroxidases, lignin peroxidases and Arthromyces ramosus peroxidase, with hydrogen peroxide. J. Biol. Chem. 277, 26879–26885. Hughes, J.B., Wang, C.Y., Zhang, C., 1999. Anaerobic biotransformation of 2,4-dinitrotoluene and 2,6-dinitrotoluene by Clostridium acetobutylicum: a pathway through dihydroxylamino intermediates. Environ. Sci. Technol. 33, 1065–1070. Ibrahim, M.S., Ali, H.I., Taylor, K.E., Biswas, N., Bewtra, J.K., 2001. Enzyme-catalyzed removal of phenol from refinery wastewater: feasibility studies. Water Environ. Res. 73, 165–172. Mantha, R., 2001. Continuous Flow System for Combined Chemical and Enzyme-Catalyzed Removal of Nitroaromatics from Synthetic Wastewater. PhD Dissertation, University of Windsor, Windsor, ON. Mantha, R., Biswas, N., Taylor, K.E., Bewtra, J.K., 2001. A continuous system for Fe0 reduction of nitrobenzene in synthetic wastewater. Environ. Sci. Technol. 35, 3231–3236. Mantha, R., Biswas, N., Taylor, K.E., Bewtra, J.K., 2002. Removal of nitroaromatics from synthetic wastewater using two-step zero-valent iron reduction and peroxidase-catalyzed oxidative polymerization. Water Environ. Res. 74, 280–287. Modaressi, K., Taylor, K.E., Bewtra, J.K., Biswas, N., 2005. Laccasecatalyzed removal of bisphenol-A from water: protective effect of PEG on enzyme activity. Water Res. 39, 4309–4316. Nicell, J.A., Wright, H., 1997. A model of peroxidase activity with inhibition by hydrogen peroxide. Enzyme Microb. Technol. 21, 302–310. Nicell, J.A., Bewtra, J.K., Taylor, K.E., Biswas, N., St Pierre, C., 1992. Enzyme catalyzed polymerization and precipitation of aromatic compounds from wastewater. Water Sci. Technol. 25, 157– 164. Nissum, M., Schiødt, C.B., Welinder, K.G., 2001. Reactions of soybean peroxidase and hydrogen peroxide pH 2.4–12.0, and veratyl alcohol at pH 2.4. Biochim. Biophys. Acta 1545, 339–348.

1491

Oh, S.Y., Cha, D.K., Chiu, P.C., 2002. Graphite-mediated reduction of 2,4-dinitrotoluene with elemental iron. Environ. Sci. Technol. 36, 2178–2184. Petsko, G.A., Ringe, D., 2004. Protein Structure and Function. New Science Press Ltd, London, pp. 92–93. Rodgers, J.D., Bunce, N.J., 2001. Treatment methods for the remediation of nitroaromatic explosives. Water Res. 35, 2101–2111. Shin, K.H., Lim, Y., Ahn, J.H., Khil, J., Cha, C.J., Hur, H.G., 2005. Anaerobic biotransformation of dinitrotoluene isomers by Lactovovvus lactis subspecies lactis strain 27 isolated from earthworm intestine. Chemosphere 61, 30–39. Siddique, M.H., St Pierre, C.C., Biswas, N., Bewtra, J.K., Taylor, K.E., Al-Kassim, L., 1993. Immobilized enzyme catalyzed removal of 4chlorophenol from aqueous solution. Water Res. 27, 883–890 (corrigendum: 28(2), I (1994). Snellinx, Z., Taghavi, S., Vangronsveld, J., Van der Lelie, D., 2003. Microbial consortia that degrades 2,4-DNT by interspecies metabolism: isolation and characterisation. Biodegradation 14, 19–29. So, J., Webster, D.A., Stark, B.C., Pagilla, K.R., 2004. Enhancement of 2,4-dinitrotoluene biodegradation by Burkholderia sp. in sand bioreactors using bacterial haemoglobin technology. Biodegradation 15, 161–171. Spain, J.C., 1995. Biodegradation of nitroaromatic compounds. Annu. Rev. Microbiol. 49, 523–555. Spanggord, R.J., Spain, J.C., Nishino, S.F., Mortelmans, K.E., 1991. Biodegradation of 2,4-dinitrotoluene by a Pseudomonas sp. Appl. Environ. Microbiol. 57, 3200–3205. Steevensz, A., Taylor, K.E., Biswas, N., Bewtra, J.K., 2006. Enzymatic treatment of cresols using laccase SP504 with the aid of a hydrophilic synthetic polymer, polyethylene glycol (PEG). In: Forty-first Central Canadian Symposium on Water Quality Research, Burlington, Ontario, Canada, February 2006. Tchounwou, P.B., Newsome, C., Glass, K., Centeno, J.A., Leszczynski, J., Bryant, J., Okoh, J., Ishaque, A., Brower, M., 2003. Environmental toxicology and health effects associated with dinitrotoluene exposure. Rev. Environ. Health 18, 203–229. Vanderloop, S.L., Suidan, M.T., Moteleb, M.A., Maloney, S.W., 1999. Biotransformation of 2,4-dinitrotoluene under different electron acceptor conditions. Water Res. 33, 1287–1295. Vasudevan, D., Stone, A.T., 1996. Adsorption of catechols, 2-aminophenols and 1,2-phenylenediamines at the metal (hydr)oxide/water interface: effect of ring substituents on the adsorption onto TiO2. Environ. Sci. Technol. 30, 1604–1613. Wu, J., Taylor, K.E., Bewtra, J.K., Biswas, N., 1993. Optimization of the reaction conditions for enzymatic removal of phenol from wastewater in the presence of polyethylene glycol. Water Res. 27, 1701–1706.