Soybean peroxidase-catalyzed removal of phenylenediamines and benzenediols from water

Enzyme and Microbial Technology 45 (2009) 253–260

Contents lists available at ScienceDirect

Enzyme and Microbial Technology journal homepage: www.elsevier.com/locate/emt

Soybean peroxidase-catalyzed removal of phenylenediamines and benzenediols from water Mohammad Mousa Al-Ansari a , A. Steevensz a , N. Al-Aasm a , K.E. Taylor a,∗ , J.K. Bewtra b , N. Biswas b a b

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

a r t i c l e

i n f o

Article history: Received 9 February 2009 Received in revised form 22 June 2009 Accepted 4 July 2009 Keywords: Soybean peroxidase Phenylenediamines Catechol Resorcinol Hydroquinone Sodium dodecyl sulfate Aluminum sulfate Wastewater Hazardous waste

a b s t r a c t Crude soybean peroxidase (SBP), isolated from soybean seed coats (hulls), catalyzes the oxidative polymerization of hazardous aqueous phenylenediamines and benzenediols in the presence of hydrogen peroxide. Experiments were conducted to investigate the optimum operating conditions including pH, hydrogen peroxide-to-substrate concentration ratio and the minimum SBP concentration required to achieve at least 95% conversion of these pollutants in synthetic wastewaters. The substrate conversion and hydrogen peroxide consumption were monitored over the period of the reactions. Polyethylene glycol (PEG) was ineffective as an additive in enhancing the conversion efficiency. The enzymatically generated polymeric products from phenylenediamines could be removed with the aid of a surfactant, sodium dodecyl sulfate (SDS), whereas the polyvalent metal cation salt, aluminum sulfate (alum), was able to remove the products from benzenediols, except hydroquinone. Enzyme-catalyzed polymerization with SBP and subsequent removal of the polymeric products generated can provide an alternative means to the conventional methods for treating many aromatic wastewater pollutants, including the title compounds.

1. Introduction Phenylenediamines (o-, m-, p-phenylenediamines or PD) and benzenediols (catechol, resorcinol, hydroquinone) are widely used as commodity synthetic intermediates and monomers. They occur variously in the process water of many industries including those producing polymers, resins, coatings, paints, dyestuffs, photographic developing agents, agrochemicals, and pharmaceuticals [1,2]. Those compounds have been shown to have an adverse on human health upon ingestion and inhalation, causing severe infections and even death [1]. The health effect extends to animals and aquatic life [1]. Thus, the discharge of these compounds, except resorcinol, into surface waters is regulated and monitored [2]. Insofar as the affected industries are in compliance with discharge limits, these compounds are not of special environmental and/or public health importance. However, their treatment is a matter of substantial activity, since the total on- and offsite disposal in the U.S. in 2007 for catechol, hydroquinone, o-PD, m-PD and p-PD were 2121, 406,203, 10,330, 112,291 and 35,208 pounds, respectively [2] (these statistics exclude air emis-

∗ Corresponding author. Tel.: +1 519 253 3000x5031; fax: +1 519 973 7098. E-mail addresses: [email protected] (M.M. Al-Ansari), [email protected] (A. Steevensz), [email protected] (N. Al-Aasm), [email protected] (K.E. Taylor), [email protected] (J.K. Bewtra), [email protected] (N. Biswas). 0141-0229/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.enzmictec.2009.07.004

© 2009 Elsevier Inc. All rights reserved.

sions and surface-water discharges, thus representing amounts treated). Conventional treatment methods for phenylenediamines and benzenediols include various microbiological approaches, adsorption methods and advanced oxidation processes. While these compounds are not particularly bio-refractory, these methods suffer from various drawbacks such as high cost, incomplete removal, formation of hazardous byproducts, low efficiency, high energy requirements and/or applicability only in a low concentration range [3]. Our long-term goal is to determine if, and if so, under what circumstances, enzymatic treatment with a peroxidase or laccase may overcome these drawbacks in a cost-effective manner [4,5]. These enzyme classes catalyze the oxidization of phenols and anilines to aromatic radicals in the presence of hydrogen peroxide (for peroxidases) or oxygen (for laccases). Those radicals diffuse from the active site of the enzyme into solution where they couple non-enzymatically to form dimers. If the dimers are soluble and still phenolic or anilino, they become substrates for another enzymatic cycle, forming higher oligomers. The cycle continues until the polymer generated reaches its solubility limit and precipitates out of solution, later to be removed by filtration or sedimentation [6]. Horseradish peroxidase (HRP) has been found to be effective in removal of phenols and aromatic amines with 95% or higher efficiency in wastewater [7]. The main drawback in using HRP is its unavailability in large quantity at a price appropriate for waste treatment. As a result, a number of other peroxidases have

254

M.M. Al-Ansari et al. / Enzyme and Microbial Technology 45 (2009) 253–260

been investigated such as Coprinus cinereus peroxidase (CCP), Coprinus macrorhizus peroxidase (CMP), Arthromyces ramosus peroxidase (ARP; it has been suggested that these three microbial peroxidases are essentially the same [8]) and soybean peroxidase (SBP) [4,9–14]. Laccases have also been investigated because of their ability to oxidize phenols and anilines, analogous to peroxidases [4,15,16]. SBP, an oxidoreductase extracted from the soybean seed coat [6], has several features that recommend it for wastewater treatment. Above all, SBP can be cheaper (certainly than HRP, arguably than the microbial peroxidases), since it is easily extracted from the soybean seed coats which are themselves a byproduct of the soybean processing industry [14]. A crude SBP extract has been found to be more efficient than the purified one [17] (as are the peroxidases cited above) and it is active over a broad range of pH (3.5–8.0) [9,11,18,19]. SBP has a higher thermal stability (being active at 70 ◦ C) than other peroxidases [20] and it is less susceptible to irreversible inactivation by hydrogen peroxide compared to HRP [19] and ARP [9]. On the one hand, its catalytic efficiency (kcat /KM ) was found to be about 20-fold higher than that of HRP for the oxidation of the assay substrate ABTS [18], on the other hand it has been reported to have slower catalytic activity and lower phenol removal efficiency than HRP [19]. Nonetheless, inactivation represents one of the drawbacks in SBP’s application as with other peroxidases. There are three possible pathways by which peroxidases are deactivated: irreversible inactivation, a form of suicide inhibition, due to the free radicals generated during the catalytic process [21]; inactivation due to endproduct polymers formed during the catalytic process [22] which may adsorb the enzyme and co-precipitate it when they exceed their solubility limit; inactivation can be due to excess hydrogen peroxide which is another form of suicide inhibition [23,24]. Additives such as polyethylene glycol (PEG), gelatin and certain polyelectrolytes [25] have been found to limit the inactivation, with PEG better than others in terms of the minimum effective concentration required, lack of interference with the removal efficiency, being easily separated from solution as a co-precipitate with the enzymatic products formed, non-toxic and more cost-effective [22,25,26]. The mechanism by which PEG protects the enzyme is not fully understood but it is believed to follow the “sacrificial polymer” theory by reacting with the free radicals generated during the catalytic process and/or the polymeric products of radical coupling, instead of these products’ reacting/associating with the enzyme and precipitating it [22]. The polymeric products normally reach their solubility limit and are separated from solution by filtration or sedimentation. However, in some cases the polymers remain soluble, in which case coagulating/flocculating agents can be employed to induce removal of such (usually colored) products. In the most common approach, hydrolyzing coagulants such as aluminum sulfate (alum), polyaluminum chloride (PAC), ferric sulfate and ferric chloride are used in removing a wide range of impurities such as colloidal particles, natural organic matter (NOM) and dissolved organic substances (DOC) [27–29]. Surfactant-mediated separation methods have been used for removal of organic species in solution. In particular, adsorptive micellar flocculation (AMF) with sodium dodecyl sulfate (SDS) and alum combined [30] was used to remove colored polymeric products after the enzymatic treatment of diphenylamine and of diaminotoluenes, while neither SDS nor alum alone were effective in removing these polymeric products [31,32]. In keeping with our long-term goal given at the outset, the hypothesis to be tested here is that SBP is an efficient catalyst for treatment of the target compounds. Preliminary optimization of SBP-catalyzed polymerization of three phenylenediamines (o-, m- and p-phenylenediamines) and three benzenediols (catechol, resorcinol and hydroquinone) is carried out

in order to determine the best removal of those polymers from solution. 2. Materials, analytical equipment and methods 2.1. Materials Crude dry solid SBP (E.C. 1.11.7, Industrial Grade lot #18541NX); (Rz value of 0.75 ± 0.10) was obtained from Organic Technologies (Coshocton, OH). Liquid ARP (SP-502, activity 2000 U/mL) was a developmental preparation of Novozymes Inc. (Franklinton, NC). Dry solid bovine liver catalase (E.C. 1.11.1.6, lot #120H7060, 19,900 U/mg solid) was purchased from Sigma Chemical Company Inc. (St. Louis, MO). Polyethylene glycol (PEG), average molecular mass of 3350 g/mole, was obtained from Sigma Chemical Company Inc. The aromatic compounds (all having a purity ≥98.0%) were obtained from Aldrich Chemical Corporation (Milwaukee, WI). All other chemicals were of analytical grade and purchased either from Sigma Chemical Company Inc., Aldrich Chemical Corporation or BDH Inc. (Toronto, ON). 2.2. Analytical equipment Aromatic compound concentrations were analyzed with a high performance liquid chromatography (HPLC) System obtained from Waters Corporation (Milford, MA). The system consisted of binary HPLC pump, autosampler, dual  absorbance detector and C18 reverse phase column (5 ␮M, 4.6 mm × 150 mm) operated by Breeze software. Elutions were isocratic with a mobile phase consisting of methanol and 50 mM phosphate buffer pH 6.7 for phenylenediamines (60:40 for o-PD (at 290 nm), 40:60 for m- (at 290 nm), and p-PD (at 310 nm)). The mobile phase for benzenediols consisted of acetonitrile and 0.1% acetic acid (40:60 for catechol (at 276 nm), 30:70 for resorcinol (at 274 nm), 20:80 for hydroquinone (at 290 nm), and 20:80 for p-benzoquinone (at 248 nm)). The injection volume was 10 ␮L and flow-rate was 1.0 mL/min. A Hewlett Packard Diode Array Spectrometer (Model 8452A), with wavelength range between 190 and 820 nm and 2 nm resolution and controlled by Hewlett Packard Vectra ES/12 computer, was used to measure sample absorbance. Quartz semi-microspectrometer cells with 1 cm optical path length were supplied by Hellma Canada Limited (Concord, ON). A Shimadzu TOC-V CSH Total Carbon Analyzer (TOC), supplied by Shimadzu Scientific Instruments (Columbia, MD), was used to measure the total organic carbon in the samples (corrected for inorganic carbon (TIC), if necessary). The microfiltered (0.2 ␮m) samples were purged by using nitrogen gas; acidified with phosphoric acid, oxidized with oxygen, and detected by using a non-dispersive infrared spectrophotometer (NDIR). The combustion chamber had a temperature between 680 and 700 ◦ C. 2.3. Analytical methods 2.3.1. Colorimetric assay for SBP activity Enzyme catalytic activity (U) is defined as the number of micromoles of hydrogen peroxide converted per minute at pH 7.4 and 23 ◦ C. The enzyme activity was determined by monitoring the initial rate of color formation at 510 nm resulting from the oxidative coupling of phenol and 4-aminoantipyrine (AAP) in the presence of hydrogen peroxide when using SBP as catalyst [4]. 2.3.2. Colorimetric assay for aromatic amines The phenylenediamine concentrations were determined by measuring the color intensity at ∼430 nm resulting from the nucleophilic substitution on trinitrobenzenesulfonic acid (TNBS) by the

M.M. Al-Ansari et al. / Enzyme and Microbial Technology 45 (2009) 253–260

phenylenediamine in the presence of sodium sulfite at pH 7.4 (phosphate buffer) and 23 ◦ C [33]. HPLC analysis and the TNBS color test for phenylenediamines gave comparable results. The o-PD and pPD deviated from the TNBS color test by 5–10% in the substrate conversion, while m-PD showed similar percent conversion. 2.3.3. Colorimetric assay for aromatic phenols The concentrations of phenolic compounds in solution were determined by measuring the color intensity resulting from the electrophilic substitution of AAP on the para-position of phenolic compounds followed by the oxidation of the intermediate by potassium ferricyanide at a pH between 8.0 and 10.0. The relatively stable product of such reaction had a pink quinoneimine chromophore, analogous to that in the enzyme assay above, with an adsorption maximum at 510 nm [16]. 2.3.4. Colorimetric assay for hydrogen peroxide This assay determined the concentration of hydrogen peroxide present in solution by measuring the color intensity resulting from oxidative coupling of phenol and AAP in the presence of hydrogen peroxide at pH 7.4 with ARP as the enzymatic catalyst. The product of such reaction produced a pink quinoneimine chromophore, identical to that in the enzyme assay above, with an absorption maximum at 510 nm [4]. 2.3.5. Buffer preparation The buffers used in the pH optimization study were prepared according to a previous study [34]. Two main buffers were used in the study depending on the choice of pH range. Acetic acid–sodium acetate buffer was used in the pH range of 3.0–5.8 and monobasic–dibasic sodium phosphate buffer was used in the pH range of 5.6–9.0. pH higher than 9.0 was adjusted by either using boric acid sodium borate buffer or phosphate buffer adjusted with NaOH. 3. Experimental protocol The first part of this study was to optimize the reaction conditions for the removal of aromatic substrate in the presence of SBP and hydrogen peroxide. The experiments were carried out in 30 mL open batch reactors at room temperature of about 23 ◦ C. The components of the sample mixture were added in the following order to a total volume of 20.0 mL: water (distilled or tap water), acetate or phosphate buffer to 40 mM, aromatic substrate to 1.0 mM, SBP to appropriate concentration, and hydrogen peroxide to appropriate concentration to initiate the reaction. The batch reactors were stirred gently for 3 h open to the atmosphere, after which the samples were quenched with excess catalase to a concentration of 62.5 U/mL to quickly consume any residual hydrogen peroxide, microfiltered and then tested for the residual concentration by using a colorimetric assay while taking into account the any necessary color correction, or HPLC or TOC. All reactions were carried out

255

in triplicate. The study was designed to achieve at least 95% conversion of aromatic substrate by optimizing the following parameters: pH, SBP concentration, hydrogen peroxide-to-substrate concentration ratio and PEG concentration. Experiments were also conducted to monitor the substrate and hydrogen peroxide consumption over time. The second part of this study was the removal of colored polymeric products that were produced from the SBP enzymatic reaction. The experiments were conducted under the optimum enzymatic conditions to achieve at least 95% conversion of aromatic substrates. Alum or SDS was added to the desired concentration and allowed to stir for 30 min and then the precipitate was allowed to settle for 1–2 h before the supernatant was analyzed with a UV–vis-spectrophotometer. The study was designed to achieve a high removal of polymeric colored products by optimizing the following parameters: pH (sodium hydroxide and sulfuric acid were used for adjustment), SDS concentration and alum concentration. 4. Results and discussion The following sections present the optimization of the two stages of this study, first the enzymic conversion of each of the six aromatic monomers into polymeric products, then the removal of those colored products from the solution. The best conditions are collected in Table 1 for each of the parameters studied. For brevity, the graphical results presented are for one diamine, ophenylenediamine, and for one diol, catechol, but are exemplary for all six compounds, which were studied in the same detail. 4.1. Conversion of the aromatic monomers by enzymatic reaction 4.1.1. Optimum pH The pH range studied for phenylenediamines and benzenediols was 3.2–11.0. The hydrogen peroxide concentration was maintained in sufficient amount for the enzymatic reaction of 1.0 mM substrate and the experiments were conducted under both “stress” and sufficient SBP concentrations for 3.0 h. The term “stress” is used throughout this paper to imply a stringency imposed by some agent, SBP concentration here, to provide easier discerning of an effect due to the varied parameter, in this case pH. The optimum pH for the enzymatic conversion of phenylenediamines was in the acidic region, 4.5–5.6, while the optimum pH for benzenediols was in the alkaline region, 6.5–8.0, except for hydroquinone which showed a broad optimum range (Table 1). In Fig. 1A, a sharp reduction in the percent conversion of phenylenediamines (o-PD) is seen in the basic region when compared to the acidic region, possibly due to a change in ionization of catalytic residues of the enzyme at high pH. By contrast, the apparent increase in the percent conversion of the benzenediols seen at pH 10 (catechol, Fig. 1B) is inferred to be due to chemical oxidation by hydrogen peroxide and/or oxygen to a quinone structure. HPLC analysis confirmed the formation of p-benzoquinone from hydroquinone enzymatically at

Table 1 Optimum conditions for the enzymatic conversion and removal of colored products for phenylenediamines and benzenediols. Aromatic compound

Optimum Optimum [H2 O2 ] pH for 1.0 mM substrate

Minimum SBP concentration required for 95% conversion of 1.0 mM substrate

Minimum SDS required for removal of polymeric products (unfiltered samples)

Minimum alum required for removal of polymeric products (unfiltered samples)

% Removed Working pH range

o-PD m-PD p-PD Catechol Resorcinol Hydroquinone

4.5–5.2a 5.4a 5.6a 6.5–7.5b 7.5–8.25b 4.0–6.5b

0.002 U/mLa 0.01 U/mLa 0.005 U/mLa 0.025 U/mLb 0.2 U/mLb 0.005 U/mLb

0.5 mM (144 mg/L) 0.25 mM (72 mg/L) 0.2 mM (58 mg/L) N.E. N.E. N.E.

N.E. N.E. N.E. 1.35 mM (36 mg/L) 0.8 mM (22 mg/L) 0.7 mM (19 mg/L)

90–95% 90–95% 90–95% 90–95% 80–85% 55–60%

a b

TNBS color test. HPLC; N.E. – no effect.

1.5 mMa 2.0 mMa 1.5 mMa 2.5 mMb 2.0 mMb 1.5 mMb

2.5–5.1 1.8–10.8 2.2–11.0 4.5–8.5 2.2–8.5 5.0–8.0

256

M.M. Al-Ansari et al. / Enzyme and Microbial Technology 45 (2009) 253–260

Fig. 1. pH optimization for enzymatic conversion of (A) 1.0 mM o-PD with 2.0 mM H2 O2 and 40.0 mM buffer; tested by TNBS color test. (B) 1.0 mM catechol with 2.5 mM H2 O2 and 40.0 mM buffer; tested by HPLC.

pH 5.2 (see discussion of carbon analysis in Section 4.2.4, below) and chemically at pH 10.0. Benzoquinone is toxic and its release is regulated just as for its diol precursor (and catechol) [1,35]. Therefore, conversion of such a substrate to a quinone form would not solve the problem, since it is not a substrate for further enzymic reaction followed by precipitation. Previous work with SBP and anilino compounds had shown an optimum pH similar to that of phenylenediamines. 2,4- and 2,6-diaminotoluenes had an optimum pH range 5.0–5.5 [9]; aniline and toluidines (o-, m- and p-) had an optimum pH range 5.0–8.0 [12], whereas, SBP and phenolic compounds had shown an alkaline optimum pH range similar to that for benzenediols. Phenol, chlorophenols (o-, m- and p-), cresol (o-, m- and p-), 2,4dichlorophenol and bisphenol A had optima in the pH range of 6.0–8.0 [11]. It is concluded that the optimum pH depends not only on the proper ionization of the catalytic residues of SBP during the enzymatic reaction but also on the type of the aromatic substrate. These experiments were necessary to find an environment for the enzyme to show its maximum catalytic activity to be used for the subsequent optimization experiments. 4.1.2. Optimum hydrogen peroxide-to-substrate concentration ratio The molar concentrations of hydrogen peroxide studied were between 0.1 and 8.0 mM for enzymatic reaction of 1.0 mM substrates at the previously determined optimum pH. Initial SBP concentrations were varied from stress conditions (defined in Section 4.1.1) to sufficient concentration for 95% conversion required for 3 h. The phenylenediamines and benzenediols consumed hydrogen peroxide in the concentration range of 1.5–2.5 mM (Table 1). The two para-substrates, p-phenylenediamine and hydroquinone, exhibited the minimum hydrogen peroxide demand for 95% con-

version. The hydrogen peroxide demand was more than what is expected theoretically for all the substrates (1 mol of hydrogen peroxide for 2 mol of the aromatic functional group). However, since the soluble dimers produced become substrates of the enzyme [6], further cycles of polymerization require additional peroxide. This is consistent with observations for other phenolic and anilino compounds [9,11]. As the hydrogen peroxide to substrate ratio increased beyond the optimum value, benzenediol compounds demonstrated a greater decrease in conversion as compared to phenylenediamines (Fig. 2). This finding was not observed at the optimum SBP condition for phenylenediamines but only under stress conditions (Fig. 2A for o-PD). Benzenediols showed peroxide deactivation under both conditions (Fig. 2B for catechol). Generally, peroxidases become deactivated as hydrogen peroxide concentration increases beyond the optimum. This may be due to the transformation of the active form of the enzyme (Compound II) to a catalytically slow form (Compound III) or the formation of an inactive form of the enzyme due to the free radicals released from Compound I and Compound II of the peroxidase that have the potential to react back with the active site [6]. Compound III formation is unlikely to be dependent on the type of reducing substrate (diol or diamine), hence it is speculated that the benzenediol radicals had a higher ability than phenylenediamine radicals to diffuse to the active site of the enzyme and deactivate it. This apparent structure–function relationship will be discussed in the following section. The deactivation of SBP in the presence of benzenediols can be overcome by the addition of extra SBP above the optimum (0.05 U/mL SBP), as shown in Fig. 2B for catechol. The foregoing results with phenols and anilines, showing differential sensitivity of SBP to deactivation by excess hydrogen peroxide, are consistent with previous studies: aniline compounds, such as diaminotoluenes, showed no enzyme deactivation as the

Fig. 2. Effect of hydrogen peroxide concentration on enzymatic conversion of (A) 1.0 mM o-PD and 40.0 mM acetate buffer of pH 5.1; tested by TNBS color test. (B) 1.0 mM catechol and 40.0 mM phosphate buffer of pH 7.28; tested by HPLC.

M.M. Al-Ansari et al. / Enzyme and Microbial Technology 45 (2009) 253–260

257

Fig. 3. Minimum SBP concentration required for 95% conversion of (A) 1.0 mM o-PD, 40.0 mM acetate buffer of pH 5.1; tested by TNBS color test. (B) 1.0 mM catechol, 40.0 mM phosphate buffer of pH 7.28; tested by HPLC.

hydrogen peroxide was increased [9] beyond the optimum, while phenolic compounds such as phenol, chlorophenols, cresols and 2,4-dichlorophenol showed deactivation at hydrogen peroxide-tosubstrate concentration ratios higher than the optimum [11]. 4.1.3. Minimum SBP concentrations and PEG effect The SBP concentration ranges studied for 1.0 mM phenylenediamines and benzenediols at the optimum pH values for 3 h were 0.0001–0.1 and 0.01–1.0 U/mL, respectively. Experiments were also conducted to investigate the effect of PEG3350 in the conversion of phenylenediamines. The phenylenediamines required less SBP than the corresponding concentration of benzenediols for 95% conversion (Table 1; Fig. 3) and within the isomeric series the meta-isomers of each exhibited the highest SBP demand (discussed below). As SBP concentration increased above the minimum required for 95% conversion, no significant effect was observed on the percent conversions of phenylenediamines and benzenediols; however, phenylenediamines exhibited better precipitate formation. In comparing phenols and anilines in terms of the minimum SBP concentrations required for 95% conversion, previous studies showed that the lower requirement for aniline compounds was mirrored in the parent compounds with phenol 0.9 U/mL [11] and aniline 0.25 U/mL [12]. In addition, 2,4- and 2,6-diaminotoluenes, required 0.01 and 0.1 U/mL of enzyme, respectively [9]. For the phenolic compounds 2-,3- and 4-chlorophenols, 2-,3- and 4-cresols and 2,4-dichlorophenol the required SBP concentrations were reported as 0.23, 0.65, 0.20, 0.60, 0.75, 0.60 and 0.7 U/mL, respectively [11]. In the current study, the minimum SBP requirement was also investigated in tap water without buffer, with pH adjusted by NaOH or H2 SO4 . Similar results were obtained: catechol, resorcinol and hydroquinone had SBP requirements of 0.025, 0.25 and 0.01 U/mL, respectively. The conversion (not removal) of hydroquinone was limited to 90%, excess SBP did not improve the conversion efficiency. A structure–function relationship can be deduced for the foregoing relative SBP requirements based on a qualitative ranking of the respective radical reactivities and the assumption that more reactive radicals cause more enzyme inactivation. Relative reactivities are inferred from the inverse ranking of radical stabilities which are estimated from the corresponding homolytic bond dissociation energies (O–H for phenols [36], N–H for anilines [37]; the higher the bond dissociation energy, the more reactive is the radical and the more damaging is it expected to be to the enzyme, hence a higher concentration of enzyme is needed to effect treatment). For the enzyme requirement data reported and cited above, the following trends in radical reactivities are consistent with the relative requirements: phenoxyl radical is more reactive than aniline radical; ortho- and para-oxyl and -amino substituents on these two parent radicals are strongly stabilizing (amino- more stabiliz-

ing than oxyl-, the substituted radicals are less toxic to the enzyme); the corresponding meta-substituted parent radicals are much less stabilized, leading to a greater enzyme requirement than the o- and p-isomers. PEG showed an insignificant effect on the conversion of phenylenediamines and benzenediols (Fig. 3). Previous work in this lab with 2,4- and 2,6-diaminotoluenes, also showed no PEG effect [9], whereas, phenolic compounds such as phenol, chlorophenols, cresols and 2,4-dichlorophenol showed a significant PEG effect in the range 20–400 mg/L of PEG [11]. These results are not surprising in view of the fact that the enzymic reactions of diols and diamines reported here, in contrast to those of simple phenols and anilines, do not form precipitates, hence, not benefitting from the addition of a “sacrificial polymer” to protect the enzyme. 4.1.4. Aromatic substrate conversion and hydrogen peroxide consumption with time Under the optimum conditions, 65–95% of the phenylenediamines and 60–85% of the benzenediols were converted during the initial 15 min of the reaction, with phenylenediamines showing slightly higher rates. The conversion of the substrates continued at a slower rate after the first 15 min, possibly due to competitive conversion of dimers and higher oligomers and/or due to enzyme inactivation caused by the reactive radicals. The rate of reaction should be directly proportional to the enzyme concentration [6], which was confirmed by experiments like those shown in Fig. 4 for o-PD as an example. Thus, an increase in SBP concentration can significantly reduce the time for the completion of the reaction, albeit at some expense. In parallel with the foregoing, for two of the phenylenediamines (o-PD and p-PD), 75–90% of the hydrogen peroxide was converted during the initial 15 min of the reaction, while 45–75% of the

Fig. 4. Conversion of o-PD and hydrogen peroxide consumption over time; 1.0 mM o-PD, 1.5 mM H2 O2 and 40.0 mM acetate buffer pH 5.1; tested by TNBS and hydrogen peroxide color tests.

258

M.M. Al-Ansari et al. / Enzyme and Microbial Technology 45 (2009) 253–260

Fig. 5. Removal of polymeric colored products. (A) Effect of SDS concentration on the removal of m-PD polymeric color products; total volume 22.0 mL and pH 5.4; tested at 304 nm by direct absorbance. (B) Effect of alum concentration on the removal of resorcinol polymeric colored products; total volume 22.0 mL and pH 7.0; tested at 440 nm by direct absorbance.

hydrogen peroxide was converted within the same time for two benzenediols (catechol and resorcinol). In all cases, during the first 15 min, hydrogen peroxide was consumed at a faster rate than the substrate as shown for o-PD in Fig. 4. The higher rate of consumption of the hydrogen peroxide likely resulted from the concurrent conversion of dimers and higher oligomers. 4.2. Removal of the colored and polymer products The enzymatic conversion of phenylenediamines and benzenediols resulted in color formation without precipitate formation with the exception that about 50% of the color from p-PD could be removed by a 0.2 ␮m filter. It is speculated that color arose from quinone-like products remaining in solution after the enzymatic reaction [31]. Furthermore, the expected polymeric products of radical coupling are surmised not to have precipitated because of the high number of hydrophilic and polar functional groups they carried (for example, a dimer would still have 3 or 4 hydroxyl or amino groups for a coupling of O/N to C or C to C, respectively). Therefore, removal of these soluble products, colored or not, is important. Preliminary experiments showed that sodium dodecyl sulfate (SDS) alone was effective in removing the phenylenediamine products but not the benzenediol ones; conversely alum alone was effective in removing the latter but not the phenylenediamine products (detailed below). In both cases it is speculated that the dominant influence is the cationic nature of the polyaniline products, in attracting the anionic dodecyl sulfate monomer or micelles and in repelling the cationic alum hydrolysis products (a chargeneutralization rather than a sweep-floc mechanism [27,28]). 4.2.1. Removal of the polymer products of phenylenediamines with SDS In this study, SDS alone was found to be able to settle the polymer products of phenylenediamines without using alum. Experiments were conducted to investigate the effect of SDS on the polymeric colored product removal from phenylenediamine reaction supernatants after 3.0 h of enzymatic treatment under the optimal conditions for 95% conversion of the monomers. The supernatants were yellow-orange (max 434 nm), dark reddish/black (max 304 nm) and dark brownish/black (max 444 nm) for o-, mand p-PD, respectively. SDS was added to remove color over a SDS concentration range of 0.005–0.5 mM (1.44–144 mg/L) while the pH of the solution was kept at that of the enzymatic reaction. The samples were stirred briefly and flocs were allowed to settle. The supernatants, microfiltered and unfiltered, were analyzed at their absorbance maxima. The minimum SDS required to remove >90% of the color after settling (unfiltered samples) was found to be

0.50 mM (144 mg/L), 0.25 mM (72.1 mg/L) and 0.20 mM (57.7 mg/L) for o-, m- and p-PD, respectively (Table 1). The difference in filtered and unfiltered samples in Fig. 5A (m-PD as an example) implies that the SDS–polymer complexes were formed even at lower SDS concentrations but did not settle until enough SDS was added. The critical micelle concentration (CMC) of SDS in water is 8.3 mM, dropping to 1.5 mM in the presence of 50.0 mM phosphate buffer [38,39]. The CMC was not determined in this study but it is speculated that settling of a colored precipitate was related to aggregation of micelles or of electrostatic complexes (anionic dodecyl sulfate with cationic polyanilines). If micellar, the polymer product might have bridged the SDS micelles together either by partition of the polymer into the micelles or by the electrostatic interaction of the amino groups of the protonated polymer onto the surface of the micelles [31] to cause precipitation. 4.2.2. Removal of the polymer products of benzenediols with alum The effect of alum on the removal of polymeric colored products from benzenediol reactions was investigated. In initial experiments, after enzymatic treatment of catechol and resorcinol in phosphate buffers, it was observed that gel was formed in an acidic pH of 5.5, but no gel formation occurred in the alkaline pH. The gel was inefficient in removing the polymeric products for resorcinol but was able to remove about 70% of the polymeric products from catechol at pH 5.5. It is suspected that, in the presence of phosphate buffer, aluminum hydroxide gel was not formed but rather aluminum phosphate gel was formed. Control experiments confirmed this phenomenon: with alum and distilled water, a gel-like struc-

Fig. 6. Effect of Alum and SDS on removal of 1.0 mM aromatic monomers in the presence of 2.7 mM alum, 0.5 mM SDS and pH adjusted with NaOH or H2 SO4 ; tested by HPLC for catechol, resorcinol and hydroquinone and by TNBS color test for o-PD, m-PD and p-PD.

M.M. Al-Ansari et al. / Enzyme and Microbial Technology 45 (2009) 253–260

259

Fig. 7. Monitoring the total organic carbon for (A) 1.0 mM phenylenediamines (Note: the first, second and fourth bars in each group are from experimental measurements; the third bars are defined as 100% for the sum of the measured TOC after enzymatic reaction plus the theoretical TOC for the amount of SDS added; this bar is to be compared only with the fourth bar in its group) and (B) 1.0 mM benzenediols under the optimum enzymatic reaction conditions, optimum SDS and alum concentrations (Table 1); tested by TOC and TIC measurements.

ture was observed in the pH range between 4.8 and 9.2, while in the presence of phosphate buffer, gel formation was observed between 3.7 and 6.1. Consequently, the enzymatic reactions were run in tap water instead of phosphate buffer for 3.0 h under optimal conditions, as discussed previously, and then the alum was added. After the enzymatic reaction of SBP with benzenediols, the solution color changed to dark brownish/black, red and light red for catechol, resorcinol and hydroquinone, respectively. The absorbance max was at 440 nm and no precipitates were observed. The aluminum ion concentration range studied was 0.27–2.2 mM (7.3–59.4 mg/L), the pH of the solution was adjusted to 7.0, the samples were stirred briefly and the flocs were allowed to settle. The supernatants, both microfiltered and unfiltered, were analyzed for absorbance at 440 nm. Approximately 90–95% of the colored product from catechol and resorcinol were removed by microfiltration when using 0.7 mM (19 mg/L) alum and 0.6 mM alum (16 mg/L), respectively, while 80–90% of colored product for catechol and resorcinol were removed by settling when using 1.35 mM alum (36.4 mg/L) and 0.8 mM alum (21 mg/L), respectively (Table 1; Fig. 5B for resorcinol as an example). The results illustrate that the floc formed at lower aluminum ion concentrations did not settle until enough alum was added. Only 60% of the colored products for hydroquinone were removed with 0.7 mM alum (19 mg/L; Table 1; for further discussion of hydroquinone, see Section 4.2.4, below). 4.2.3. Effect of SDS and alum on the removal of aromatic monomers Control experiments were conducted to investigate the effect of 2.7 mM alum and 0.5 mM SDS in removing 1.0 mM aromatic monomers at pH of 5.0 and 7.0 (Fig. 6). The results demonstrate no significant effect of SDS on the removal of any of the aromatic compounds at both pH values. Similarly, alum showed no significant effect in the removal of resorcinol, o-PD and m-PD at both pH values, while 5–15% removal was observed for hydroquinone and p-PD. The most significant effect was observed for catechol in which 40% removal was observed at pH 7.0 and about 10% at pH 5.0. The results are consistent with previous work in which resorcinol showed no removal, while catechol showed the greatest removal with ferric hydroxide gel [40]. 4.2.4. Monitoring the total organic carbon after treatment After the optimal enzymatic reaction for phenylenediamine isomers, 5–30% removal of the carbon was observed (Fig. 7A; only pairwise comparison is intended in the sets of four bars, 1 with 2 and 3 with 4). Hence it is speculated that most of the polymeric reaction products produced were still in a soluble form (low molecular mass and/or ionic) that could not be removed by microfiltration. After the flocculation with SDS, about 65–80% of the carbon was

removed. The remaining carbon may have also included any SDS that did not precipitate after the treatment. Since the SDS carbon added was at most 52% of the total (for o-PD), these proportions of removal indicate substantial settling of the diamine carbon. After the optimal enzymatic reaction for the benzenediol isomers, no significant removal of the carbon was observed as shown in Fig. 7B, while after the flocculation with alum, about 80% of the carbon was removed in the catechol and resorcinol batch reactors, but only 20% of the carbon was removed in hydroquinone sample. Since alum cannot remove most of the products of hydroquinone oxidation, although it is moderately effective at removing colored products (Section 4.2.2, above), it is inferred that the preponderance of its enzymatic oxidation products are still at the monomer stage, presumably the semiquinone and quinone. This inference is supported by HPLC at completion of these reactions, showing a near-stoichiometric correspondence of the hydroquinone converted (retention time, r.t., 3.19 min) with the benzoquinone formed (r.t. 4.80 mn), the only other product being a more polar compound (r.t. 1.20 min) in lower abundance than the residual hydroquinone. 5. Conclusions Crude SBP has been shown to be a suitable enzyme for the peroxidase-catalyzed polymerization of phenylenediamines and benzenediols, except hydroquinone. The maximum conversion of these substrates was dependent on the optimal conditions which were substrate-specific. The phenylenediamines required a lower SBP concentration than the benzenediols for 95% conversions of the substrates, and all of these compounds had lower SBP requirements than for equivalent conversion of analogous monofunctional compounds. PEG showed an insignificant effect on the conversion of phenylenediamines and benzenediols. The removal of greater than 90% of the phenylenediamine-derived polymeric products was achieved by using SDS alone. The removal greater than 90% of the catechol- and resorcinol-derived polymeric products was achieved by using alum alone. Thus, SBP-catalyzed conversion can form the basis of a cost-effective alternative to conventional processes for removal of five of the six compounds tested from industrial wasteand process water. Acknowledgment The authors wish to acknowledge the Ontario Ministry of Agriculture and Food New Directions Program for funding this research. References [1] TOXNET: Toxicology Data Network, Hazardous substance data bank (HSDB) 2009, website: http://toxnet.nlm.nih.gov [accessed May 2009].

260

M.M. Al-Ansari et al. / Enzyme and Microbial Technology 45 (2009) 253–260

[2] U.S. Environmental Protection Agency, Toxics Release Inventory (TRI) data 2007, website: http://www.epa.gov/tri [accessed May 2009]. [3] Nicell JA, Al-Kassim L, Bewtra JK, Taylor KE. Wastewater treatment by enzyme catalysed polymerization and precipitation. Biodeterior Abstr 1993;7(1):1–8. [4] Steevensz A, Al-Ansari MM, Taylor KE, Bewtra JK, Biswas N. Comparison of soybean peroxidase with laccase in the removal of phenol from synthetic and refinery wastewater samples. J Chem Technol Biotechnol 2009;84:761–9. [5] Ibrahim MS, Ali HI, Taylor KE, Biswas N, Bewtra JK. Enzyme-catalyzed removal of phenol from refinery wastewater: feasibility studies. Water Environ Res 2001;73(2):165–72. [6] Dunford HB. Heme peroxidases. New York: John Wiley & Sons Inc.; 1999. p. 1–36, 270–319, 414–454. [7] Klibanov AM, Alberti BN, Morris ED, Felshin LM. Enzymatic removal of toxic phenols and anilines from waste waters. J Appl Biochem 1980;2:414–21. [8] Kjalke M, Andersen MB, Schneider P, Christensen B, Schulein M, Welinder KG. Comparison of structure and activities of peroxidases from Coprinus cinereus, Coprinus macrorhizus and Arthromyces ramosus. Biochim Biophys Acta 1992;1120:248–56. [9] Patapas J, Al-Ansari MM, Taylor KE, Bewtra JK, Biswas N. Removal of dinitrotoluenes from water via reduction with iron and peroxidase-catalyzed oxidative polymerization: a comparison between Arthromyces ramosus peroxidase and soybean peroxidase. Chemosphere 2007;67:1485–91. [10] Al-Kassim L, Taylor KE, Nicell JA, Bewtra JK, Biswas N. Enzymatic removal of selected aromatic contaminants from wastewater by a fungal peroxidase from Coprinus macrorhizus in batch reactors. J Chem Technol Biotechnol 1994;61:179–82. [11] Caza N, Bewtra JK, Biswas N, Taylor KE. Removal of phenolic compounds from synthetic wastewater using soybean peroxidase. Water Res 1999;33(13):3012–8. [12] Mantha R, Biswas N, Taylor KE, Bewtra JK. Removal of nitroaromatics from synthetic wastewater using two-step zero-valent iron reduction and peroxidase-catalyzed oxidative polymerization. Water Environ Res 2002;74(3):280–7. [13] Ikehata K, Buchanan ID, Smith DW. Treatments of oil refinery wastewater using crude Coprinus cinereus peroxidase and hydrogen peroxide. J Environ Eng Sci 2003;2:463–72. [14] Wilberg K, Assenhaimer C, Rubio J. Removal of aqueous phenol catalyzed by a low purity soybean peroxidase. J Chem Technol Biotechnol 2002;77:851–7. [15] Bollag JM, Shuttleworth KL, Anderson DH. Laccase-mediated detoxification of phenolic compounds. Appl Environ Microbiol 1988;54:3086–91. [16] Modaressi K, Taylor KE, Bewtra JK, Biswas N. Laccase-catalyzed removal of bisphenol-A from water: protective effect of PEG on enzyme activity. Water Res 2005;39(18):4309–16. [17] Flock C, Bassi A, Gijzen M. Removal of aqueous phenol and 2-chlorophenol with purified soybean peroxidase and raw soybean hulls. J Chem Technol Biotechnol 1999;74:303–9. [18] Kamal JK, Behere DV. Activity, stability and conformational flexibility of seed coat soybean peroxidase. J Inorg Biochem 2002;94:236–42. [19] Wright H, Nicell JA. Characterization of soybean peroxidase for the treatment of aqueous phenols. Bioresour Technol 1999;70:69–79. [20] McEldoon JP, Dordick JS. Unusual thermal stability of soybean peroxidase. Biotechnol Prog 1996;12(4):555.

[21] Klibanov AM, Tu T, Scott KP. Peroxidase-catalyzed removal of phenols from coal-conversion waste waters. Science 1983;221:259–61. [22] Nakamoto S, Machida N. Phenol removal from aqueous solutions by peroxidasecatalyzed reaction using additives. Water Res 1992;26(1):49–54. [23] Arnao MB, Acosta M, Del-Rio JA, Varon R, Garcia-Canovas FA. kinetic study on the suicide inactivation of peroxidase by hydrogen peroxide. Biochim Biophys Acta 1990;1041:43–7. [24] Baynton KT, Bewtra JK, Biswas N, Taylor KE. Inactivation of horseradish peroxidase by phenol and hydrogen peroxide: a kinetic investigation. Biochim Biophys Acta 1994;1206:272–8. [25] Wu Y, Taylor KE, Biswas N, Bewtra JK. Comparison of additives in the removal of phenolic compounds by peroxidase-catalyzed polymerization. Water Res 1997;31(11):2699–704. [26] Harris JM. Introduction to biotechnical and biomedical application of poly (ethylene glycol). In: Harris JM, editor. Poly (ethylene glycol) chemistry – biotechnical and biomedical applications. New York: Plenum Press; 1992. p. 1–14. [27] Duan J, Gregory J. Coagulation by hydrolyzing metal salts. Adv Colloid Interface Sci 2003;100(102):475–502. [28] Elibeck WJ, Mattock G. Chemical process in waste water treatment. New York: John Wiley and Sons Inc.; 1987. p. 11–39, 232–293. [29] Bolto BA, Dixon DR, Gray SR, Ha C, Harbour PJ, Le N, Ware AJ. The use of soluble organic polymers in waste treatment. Water Sci Technol 1996;34(9):117–24. [30] Paton P, Talens FI. Effect of pH on the flocculation of SDS micelles by Al3+ . Colloid Polym Sci 2001;279:196–9. [31] Patapas J. The removal of dinitrotoluene from wastewater via zero-valent iron reduction and enzyme-catalyzed oxidative polymerization with Arthromyces ramosus peroxidase; 2007 (PhD Dissertation, University of Windsor, Windsor, ON). [32] Saha B, Taylor KE, Bewtra JK, Biswas N. Laccase-catalyzed removal of diphenylamine from synthetic wastewater. Water Environ Res 2008;80(11):2118–24. [33] Wang L, Biswas N, Bewtra JK, Taylor KE. A simple colorimetric method for analysis of aqueous phenylenediamines and aniline. J Environ Eng Sci 2005;4(6):423–7. [34] Gomori G. Preparation of buffers for use in enzyme studies. In: Colowick SP, Kaplan NO, editors. Methods in enzymology. New York: Academic Press Inc.; 1955. p. 138–46. [35] Smith MT, Yager JW, Steinmetz KL, Eastmond DA. Peroxidase-dependent metabolism of benzene’s phenolic metabolites and its potential role in benzene toxicity and carcinogenicity. Environ Health Perspect 1989;82:23–9. [36] Bordwell FG, Cheng J-P. Substituent effects on the stabilities of phenoxyl radicals and the acidities of phenoxyl radical cations. J Am Chem Soc 1991;113:1736–43. [37] Jonsson M, Lind J, Eriksen TE, Merényi G. Redox and acidity properties of 4substituted aniline radical cations in water. J Am Chem Soc 1994;116:1423–7. [38] Boschke FL, editor. Topics in Current Chemistry: Micelles. Berlin/Heidelberg: Springer-Verlag; 1998. p. 6–57, 66–76. [39] Fuguet E, Rafols C, Roses M, Bosch E. Critical micelle concentration of surfactants in aqueous buffered and unbuffered systems. Anal Chim Acta 2005;548:95–100. [40] Lefebvre L, Legube B. Coagulation–flocculation by ferric chloride of some organic compounds in aqueous solution. Water Res 1993;27(3):433–47.