Influence of the Nonionic Surfactant Brij 35 on the Bioavailability of

JOSÉ MANUEL GARCIA, †. LUKAS Y. WICK, ‡. AND HAUKE HARMS* ,†,‡. Swiss Federal Institute for Environmental Science and. Technology (EAWAG), CH ...
0 downloads 0 Views 90KB Size
Environ. Sci. Technol. 2001, 35, 2033-2039

Influence of the Nonionic Surfactant Brij 35 on the Bioavailability of Solid and Sorbed Dibenzofuran JOSE ´ MANUEL GARCIA,† L U K A S Y . W I C K , ‡ A N D H A U K E H A R M S * ,†,‡ Swiss Federal Institute for Environmental Science and Technology (EAWAG), CH 8600 Du ¨ bendorf, Switzerland, and Swiss Federal Institute of Technology Lausanne (EPFL), IATE-Pe´dologie, GR-Ecublens, CH 1015 Lausanne, Switzerland

The effect of the nonionic surfactant Brij 35 on the bioavailability of solid and Teflon-sorbed dibenzofuran for Sphingomonas sp. strain HH19k was studied in simple model systems. Growth with dibenzofuran and dibenzofuranspecific oxygen uptake in surfactant-free media and with Brij 35 above the critical micelle concentration (cmc) were compared with dissolution and desorption in the absence of bacteria. Brij 35 accelerated dissolution and biodegradation of solid dibenzofuran by a factor of 2. It also enhanced the initial desorption rate of dibenzofuran from Teflon by this factor. Continuously decreasing desorption rates were attributed to slow diffusion of dibenzofuran inside Teflon, leading to depletion of dibenzofuran in the exterior of the Teflon particles. Surprisingly, Brij 35 slowed the initial biodegradation of desorbing dibenzofuran. We propose two processes that led to low bioavailability of sorbed dibenzofuran in the presence of surfactant. First, desorbing dibenzofuran rapidly accumulated in surfactant micelles, leading to reduced truly water-dissolved dibenzofuran concentration as the factor controlling the biodegradation rate. Second, Brij 35 suppressed the contact between bacteria and Teflon. This increased the average diffusion distance of dibenzofuran to the bacteria, which in turn flattened the gradient of the dissolved dibenzofuran concentration between the sorbent and the cells as the driving force for desorption.

Introduction Many biodegradable chemicals accumulate in soils and sediments because they are poorly available to microorganisms. Limited bioavailability arises for instance when an environmental chemical is mainly present in a physical state that cannot be taken up by microbes (1, 2). Therefore, primarily poorly water-soluble and strongly sorbing contaminants defy biodegradation because only small fractions of these compounds occur in the exclusively bioavailable water-dissolved state. Limited bioavailability can also occur when environmental chemicals and microorganisms are spatially separated (3). This is the case when pollutants are absorbed in natural organic matter (4, 5) or adsorbed to the walls of submicron-sized pores inside soil particles, which * Corresponding author phone: +41-21-693 3773; fax: +41-21693 5670; e-mail: [email protected]. † Swiss Federal Institute for Environmental Science and Technology (EAWAG). ‡ Swiss Federal Institute of Technology Lausanne (EPFL). 10.1021/es001552k CCC: $20.00 Published on Web 04/11/2001

 2001 American Chemical Society

exclude microorganisms (6, 7). When a pollutant is present as a nonaqueous liquid or as a solid, the high volume to surface ratio of droplets and crystals allows the direct access of only a limited number of microorganisms to only a limited fraction of the chemical. It has often been shown that in such cases the homogenization and dispersion of the pollutants may increase the biodegradation rates (8-10). A way to disperse pollutants is the addition of surfaceactive compounds. Surfactants are amphiphilic molecules that enrich at interfaces and form micelles when exceeding the critical micelle concentration (cmc). Surfactants are known to improve the pollutant transfer into the water phase by decreasing the interfacial tension between water and hydrophobic pollutants and by accumulating hydrophobic compounds in micelles (11, 12). Numerous studies have been conducted in which surfactants were used to improve contaminant biodegradation in soils. While in some cases increased biodegradation rates were observed (10), negative results (13) or no effects (14) were obtained in other studies. The failure of surfactants was attributed to toxic effect (15) and the utilization of the surfactant rather than the target chemical (10, 12). However, the complexity of soils often prevented the unequivocal identification of the mechanisms of action of the surfactants (16). More conclusive results on the effect of surfactants could be obtained from studies conducted in defined laboratory systems. Tiehm (11) elucidated the influence of the physicochemical properties of surfactants on their toxicity for bacteria, and Liu et al. (16) found that micelle-solubilized naphthalene was degraded by mixed cultures of bacteria. Volkering et al. (12) showed that naphthalene in micellar state was not directly available to bacteria but had to partition into the aqueous phase before it was taken up. Little is known about the mechanism of surfactant influences on sorbed compounds. Theoretically, surfactants could mobilize compounds that are sorbed inside of soil particles if (i) surfactant micelles were able to enter and leave soil particles; (ii) the presence of lipophilic micelles as an additional hydrophobic phase in the water phase would drive the desorption; and/or (iii) the surfactant would disaggregate the sorbing particles, thereby increasing their interfacial area with the aqueous phase. To our knowledge, mobilization and degradation of sorbed compounds by surfactants in a well-defined soil-model system has not been studied yet. In a previous study, it was shown that dibenzofurans adsorbed inside of porous Teflon material (17). The sorption kinetics and the kinetics of degradation of desorbing dibenzofuran by Sphingomonas sp. HH19k resembled those reported for hydrophobic compounds in soils (5, 18). The desorption was biphasic due to sorption-retarded intraparticle diffusion, i.e., the relatively fast exchange of the fraction located in the outermost regions of the Teflon particles was followed by the slow exchange of the dibenzofuran located in the particle centers. Here, we used the same experimental system to study the influence of the nonionic surfactant Brij 35 on the bioavailability of solid and Teflon-sorbed dibenzofuran for strain HH19k. Brij 35 above the cmc favored the dissolution and bioavailability of solid dibenzofuran. Although it also enhanced desorption of dibenzofuran from Teflon, it reduced the bioavailability of sorbed dibenzofuran. We propose that two mechanisms led to this surprising finding: accumulation of dibenzofuran in surfactant micelles and suppression of bacterial adhesion to Teflon. VOL. 35, NO. 10, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2033

Materials and Methods Bacterium and Cultivation. Dibenzofuran-degrading Sphingomonas sp. strain HH19k (19) was used in all degradation and adhesion experiments. The growth medium and the cultivation method were as described before (19). Dibenzofuran (1 g L-1) was used as the growth substrate. Because of its low solubility of only 5 mg L-1 (19), dibenzofuran was mainly present as crystals. Cells were harvested in the early stationary growth phase. Residual dibenzofuran crystals were removed by filtration through a cellulose tissue. After being centrifuged and washed in phosphate-buffered saline (PBS) with an ionic strength of 100 mM (19), the cells were resuspended in the same buffer. Chemicals and Sorbent. The nonionic alkyl poly(ethylene glycol) ether surfactant Brij 35 [C12H25O(CH2CH2O)23H] and dibenzofuran were purchased from Aldrich (Steinheim, Germany). Brij 35 was chosen because it had been found to be nontoxic to bacteria (11). Brij 35 did not serve as carbon source for growth of strain HH19k. Its cmc in water is reported to be 120 mg L-1 (11). Dibenzofuran was used either in crystalline form as received or as thick coatings on polystyrene granules. The specific surface area A (area per mass) of dibenzofuran crystals of 9 × 10-2 m2 g-1 was calculated on the basis of microscopically measured edge lengths x of about 100 randomly chosen crystals as

A ) 6x2n

(1)

where n, the number of dibenzofuran crystals per gram, was determined by weighing 1000 crystals. The average crystal edge length x of 5.5 × 10-6 m corresponded to 4.85 × 106 crystals g-1. The specific density of dibenzofuran of 1.2 g cm-3 was determined from the rise of the water level in a calibrated flask upon submerging a known mass of dibenzofuran. Volume reduction due to dissolution was considered negligible because of the low water solubility of dibenzofuran. To obtain solid dibenzofuran with smaller specific surface area, crystalline dibenzofuran was melted at 100 °C, and polystyrene granules of about 1.5 mm diameter were repeatedly ducked into the fluid dibenzofuran. In this way, smooth coatings of dibenzofuran on the polystyrene were obtained. The dibenzofuran-coated granules were spherical with diameters of 1.9 ( 0.4 mm corresponding to a surface area of 4.7 × 10-5 ( 1.8 × 10-6 m2 per bead. During dissolution and solubilization experiments, generally less than 1% of the solid dibenzofuran was released. Hence, the surface areas were assumed to remain constant during the experiments. The molecular diffusivities Dm (m2 s-1) of dibenzofuran and Brij 35 of 6.5 × 10-10 and 1.8 × 10-10 m2 s-1, respectively, in water at 25 °C were estimated by using the Wilke-Chang equation and the group molar volume contribution method of Le Bas (20). The diffusivity of Brij 35 micelles De (m2 s-1) was calculated with the Stokes Einstein equation:

De ) kBT/6πµRe

(2)

where kB is the Boltzmann constant (1.38 × 10-23 J K-1), T is the absolute temperature (K), and µ is the water viscosity (8.9 × 10-4 kg m-1 s-1). When spherical micelles of an average radius Re of 4 × 10-9 m (21), corresponding to 104 molecules of Brij 35 per micelle were assumed, De at 25 °C was 6.1 × 10-11 m2 s-1. Teflon 350 particles as an artificial sorbent of dibenzofuran were obtained from Dolder AG (Basel, Switzerland). A fraction with diameters ranging between 250 and 500 µm diameter (mean diameter, 382 µm) was obtained by sieving. The external and total surface areas of the porous particles were 0.073 and 1.2 m2 g-1, respectively. The internal porosity was reported to be 0.005, and the internal surface area was regarded as inaccessible for bacteria because of pore 2034

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 10, 2001

diameters