Fixation of nanoscale iron (nfe) on resin granules for environmental applications

Transcription

1 Fixation of nanoscale iron (nfe) on resin granules for environmental applications Aik.Toli 1 *, K. Chalastara 1, Ch. Mystrioti 1, A. Xenidis 1, N. Papassiopi 1 1 Sch. of Mining and Metallurgical Eng., National Technical University of Athens, 15780, Greece *Corresponding author: katerinatoli@metal.ntua.gr, Tel , Fax Abstract The objective of present study is to obtain the fixation of nano zero valent iron (nzvi) particles on a permeable matrix and evaluate the performance of this composite material for the removal of Cr(VI) from contaminated waters. The experiments were carried out using the cationic resin Dowex 50WX2 as porous support of the iron nanoparticles. The work was carried out in two phases. The first phase involved the fixation of nzvi on the resin matrix. The resin granules were initially mixed with a FeCl3 solution to obtain the adsorption of Fe(III). Then the Fe(III) loaded resin (R- Feads) was treated with a polyphenol solution to obtain the reduction of trivalent iron to the elemental state. The second phase was focused to the investigation of Cr(VI) reduction kinetics using the nanoiron loaded resins (R-nFe). The parameters investigated included concentration of Cr(VI) and amount of resin. Keywords: nano zero valent iron (nzvi), cation exchange resin (Dowex 50WX2), nzvi loaded resin, hexavalent chromium (Cr(VI)) 1. INTRODUCTION Zero valent iron (ZVI) is considered as a very efficient reagent and has been used for the remediation of contaminated natural waters or industrial effluents for a very wide spectrum of contaminants, both organics and inorganics. At nanoscale, ZVI effectiveness is strongly enhanced, but there are many difficulties in the management and final disposal of the resulting colloidal suspension. On the other hand, nzvi cannot be used for the treatment of contaminated water under flow conditions, e.g. through a permeable reactive barrier or a filter consisting of a permeable reactive bed. For this reason, several researchers investigated the possibility to fix nzvi on the matrix of another material easier to handle and separate from the aqueous phase. Various natural minerals or synthetic materials have been tested as supporting matrices including bentonite, kaolinite, chitosan and ion exchange resins [1-5]. The objective of present work was to obtain the incorporation of iron nanoparticles in the matrix of a cationic resin and evaluate the performance of this composite material for the removal of Cr(VI) from contaminated waters. In the natural environment chromium exists in both the trivalent Cr(III) and the hexavalent Cr(VI) oxidation state. Trivalent chromium as a trace element is of great importance for health as it helps in the metabolism of glucose. It also takes part in the metabolism of hydro carbonates and lipids. Hexavalent chromium on the other hand is carcinogenic and mutagenic. Extended contact can cause severe problems on the digestive system, the liver, the lungs, the kidneys, the skins and the embryos. Therefore the scientific investigation concerning remediation of polluted waters is increased day by day. Many methods have been developed to this direction including the use of zero-valent iron (ZVI) [6-10]. Elemental iron at milli- and micro-meter scale has been widely used

2 in permeable reactive barriers or in pump and treat installations for the remediation of Cr(VI) polluted waters. Only a small percentage of the elemental iron is utilized for the reduction of Cr(VI) to Cr(III) in these installations due to the passivation of iron surface [7]. The use of nzvi fixed on a porous substrate is an innovative alternative, which can combine the benefits of treatment under flow conditions with the high reactivity and the high degree of utilization of nano-scale iron. In this study, the porous material used as supporting matrix was the cationic resin Dowex 50WX2. It consists of a styrene and divinylbenzene (DVB) copolymer, where DVB creates cross-linkages between the long polystyrene chains. The selected resin has a low percentage, 2%, of DVB crosslinkages, which means that it has high internal porosity in the order of 80%. The functional group acting as cation exchanger is the sulfonic acid group (-SO3H). The elemental iron nanoparticles were produced by reducing adsorbed Fe(III) cations with polyphenol compounds, i.e. green tea extract and gallic acid. It is noted that in the previously published studies, the researchers applied dense borohydrite solutions to obtain the reduction of adsorbed iron [3-5]. This is the first attempt to evaluate alternative, environmentally friendly means for the step of adsorbed iron reduction. The resin granules embedded with iron nanoparticles, R-nFe, were used for treating water containing hexavalent chromium. 2. MATERIALS AND METHODS The resin Dowex 50WX2 Na + form, mesh, is a strongly acidic cation exchange resin containing sulfonic groups, and was purchased from Sigma Aldrich, China. All chemical reagents were of analytical reagent grade. Potassium dichromate (K2Cr2O7), iron chloride (FeCl3.6H2O) and gallic acid (3,4,5 trihydroxybenzoic acid) were purchased from Mallinckrodt Chemical Works, USA, Merck, Germany and Alfa Aesar, Germany, respectively. Also commercially available dry leaves of green tea (Twinings of London) were used as sources of polyphenols. 2.1 Synthesis of R-nFe The embedded resin was developed in two steps. The first step involved the adsorption of trivalent iron cations on the resin and the second step the reduction of adsorbed iron by the polyphenol solution, either green tea extract (GT) or gallic acid (GA). In the following text, the resin loaded with adsorbed Fe(III) is denoted as R-Fe and the resin containing nzvi is denoted as R-nFe. For the adsorption step, 20 g of Dowex 50WX2 resin were mixed with 200 ml of iron chloride solution, 0.05 M, and agitated at 100 rpm in an orbital agitator for 4 hours, to homogeneously obtain ferric iron adsorption in the resin. Afterwards, the resin was washed with deionized water (DI), 300mL, to remove residual chloride solution. The second step, aiming at the reduction of adsorbed Fe(III), involves mixing of the loaded and washed R-Fe resin with 150 ml of polyphenol solution and 100 ml of DI, and agitation of the suspension for 20 hours. The green tea extract was prepared by immersion of 20 g/l of GT leaves in water preheated at a temperature of 80 o C for 5 minutes. The extract was separated from the leaves by vacuum filtration, using a filter paper of 0.45μm pore size. The extract was analyzed with the Folin Ciocalteau method (ISO ) and was found to contain 1.9 g/l of gallic acid equivalents. The synthetic gallic acid solution was prepared by dissolving the chemical reagent in deionized water to obtain a concentration of 2 g/l. The nzvi loaded resin (R-nFe), was recovered by filtering the suspension, rinsed with 100 ml of DI, and submitted to freeze drying for the removal of moisture and bound water. Water loss during freeze drying was equivalent to 80% of the total wet weight of the resin. The effectiveness of Fe(III) adsorption was evaluated by analyzing the residual iron in the aqueous solution after the first treatment step. The effectiveness of Fe(III) reduction during the 2 nd step was

3 estimated indirectly by mixing R-nFe with a Cr(VI) solution at a molar ratio nfe/cr(vi)=1/1 mol/mol and analyzing residual Cr(VI) after 1, 2 and 4 hours. Namely, taking into consideration that dry R-nFe contained 1.92 mmol of iron per gram, the experiments were carried out mixing 0.2 g of dry R-nFe with 200 ml of an aqueous solution containing 100 mg/l Cr(VI) (1.92 mm). Analysis of iron and total chromium was carried out by flame atomic absorption spectrometry (AAS, Perkin Elmer 2100). The concentration of Cr(VI) was determined colorimetrically with 1,5- diphenylcarbazide at 540nm, using a UV- Visible spectrophotometer Hitachi (U1100). 2.2 Cr(VI) reduction. Kinetic experiments Batch experiments were performed to evaluate the kinetic law determining the reduction of hexavalent chromium at different Cr(VI) concentrations and resin levels. The standard conditions were: chromium concentration 15 mg/l and resin amount 4 g/l. The investigated ranges of values were 5-25 mg/l for Cr(VI) and 1-6 g/l for resin. Preliminary tests were carried out using different starting ph values, i.e. 3, 4, 5 and 6, but it was found difficult to maintain this value during the whole duration of the test. In all the cases, ph drifted towards the value of 3. For this reason all the experiments were carried out at ph 3. Temperature was kept constant at 25 o C. The batch experiments were conducted in Erlenmeyer flasks (500 ml), mixing 100 ml of Cr(VI) solution with the appropriate amount of R-nFe resin. The suspensions were placed into an incubator, maintaining constant temperature at 25 o C and applying a gentle agitation of 100 rpm. Samples of the aqueous phase were taken after 2, 5, 10, 20, 40, 60 and 90 minutes of agitation and analyzed for hexavalent chromium. Each experiment was duplicated under identical conditions. 3. RESULTS AND DISCUSSION 3.1 Synthesis of R-nFe The results of the first adsorption step indicated that resin particles were loaded with 21.58±0.54 mg of Fe(III) per gram of wet resin. The composition of the aqueous phase following the treatment with the polyphenol solutions is shown in Table 1. As seen in the table this treatment caused a very slight desorption and the residual embedded Fe remained essentially constant, i.e mg Fe per gram of wet resin. Table 1. Initial and final composition of polyphenol solutions during the second step of R-nFe synthesis and calculated iron desorption. Initial adsorbed Fe(III) mg/g wr (wet resin). Resin (wet) to aqueous phase ratio S/L=80 g wr /L Initial solution Final solution Resin ph ORP ph ORP Fe Desorbed Residual Fe Fe mv mv mg/l mg/g (wr) mg/g (wr) Green Tea Gallic acid As seen in Table 1, the ph of final solutions are 3.59 and 2.30 for GT and GA respectively, values more acidic compared to the initial solutions. This is in agreement with reaction 1 describing the reduction of Fe(III) to the elemental state Fe(0) under the action of polyphenols. According to the reaction, 3 moles of H + are generated per mole of reduced iron. 2Fe R C 6 H 3 (OH) 2 2Fe R C 6 H 3 O 2 + 6H + (1)

4 In reaction (1) [R'-C6H3(OH)2] denotes a diphenol and [R'-C6H3O2] the quinone produced from the reduction of diphenol. As previously mentioned, the effectiveness of Fe(III) reduction during the treatment of resins with the polyphenol solutions was estimated indirectly by mixing R-nFe with a Cr(VI) solution at a molar ratio nfe/cr(vi)=1/1 mol/mol and analyzing residual Cr(VI). The results are presented in Figure 1. Chromium (VI) was not reduced when the solution was mixed with the resin treated with GT extract. This is an indication that GT was not efficient in reducing adsorbed Fe(III). A possible explanation is that GT polyphenols, corresponding to relatively large molecules, e.g. epigallocatechin gallate with MW 458 g/mol, cannot diffuse effectively through the internal porosity of resin particulates and, thus, cannot reach the adsorbed Fe(III) cations. On the contrary the resin treated with GA (R-nFe-GA) was found able to reduce Cr(VI). After 4 hours, 84% of Cr(VI) was reduced Cr(VI), mg/l Green Tea extract Gallic Acid time (h) Figure 1. Reduction of Cr(VI) by the Fe embedded in the resin and treated with GT extract and gallic acid. Tests carried out at molar ratio nfe/cr(vi)=1/1 mol/mol. According to the stoichiometry of reaction (2) one mole of Fe(0) is able to reduce one mole of Cr(VI). This indicates that at least 80% of the iron embedded in R-nFe-GA was reduced to the elemental state during the treatment with GA. Fe 0 + HCrO H + Fe +3 + Cr H 2 O (2) The solution of 4 hours had a ph of 5.6 and was also analyzed for total chromium and total iron. Total Cr was found approximately equal to Cr(VI) and total Fe was below AAS detection limit. This finding suggests that Cr(III) and Fe(III) produced during the reduction of Cr(VI) according to reaction (2), remain in the solid phases. The trivalent cations are either adsorbed on the sulfonate groups of the resin or precipitate as mixed Fe(OH)3.xCr(OH)3. Precipitation of mixed hydroxides occurs usually at ph values >5 [11]. Taking into consideration the better performance of gallic acid the subsequent experiments were carried out using the resin treated with gallic acid. 3.2 Reduction of Cr(VI). Kinetic experiments. Effect of initial Cr (VI) concentration The effect of initial Cr(VI) concentration on the rate of Cr(VI) removal was investigated in the range of 5-25 mg/l. The experiments were carried out mixing the aqueous solution with 4 g/l R- nfe-ga. The embedded elemental iron was in stoichiometric excess with respect to Cr(VI), so that

5 nfe consumption was no more than 7% to 34% of the total amount. Figure 3a shows the time evolution of Cr(VI) in the aqueous phase for all the tests carried out with different initial Cr(VI) concentrations. Cr(VI), mg/l time(min) (a) 5 mg/l 10 mg/l 15 mg/l 20 mg/l 25 mg/l Figure 3. (a) Time evolution of Cr(VI) at different initial Cr(VI) concentration. (b) Linear regression of ln(c/c0) versus time suggesting that Cr(VI) reduction follows a kinetic law of first order with respect to Cr(VI) concentration y = x R² = (b) y = x R² = y = x R² = y = x R² = y = x R² = time(min) 5 mg/l 10 mg/l 15 mg/l 20 mg/l 25 mg/l Assuming that chromate reduction follows a kinetic law of first order with respect to Cr(VI), the time evolution of Cr(VI) concentration is described by equations 3 and 4 in differential and integrated form respectively: dc dt = k 1C (3) ln ( C C 0 ) = k 1 t (4) where C0 is the initial concentration of Cr(VI), C is the concentration at time t and k1 is the chemical constant. In figure 3b, the experimental data were plotted in a graph with the values of time in axis X and the values of ln(c/c0) in axis Y. As seen in the figure, the data can be described with satisfactory precision by straight lines with intersection equal to zero, as expected by equation (4). The correlation coefficient R 2 resulted from the least square linear regression ranges from to 0.989, which is close to unit. So, the assumption of 1 st order kinetics was confirmed and the constant k1 was calculated from slope of the lines. The values of k1 ranged from to min -1. The average value is min - 1 and the standard deviation ± min -1. Effect of resin dose The effect of resin dose on Cr (VI) removal was investigated in the range of g/l of dry resin (dr). It is noted that the weight of freeze dried resin was only 20% compared to the weight of the corresponding wet resin. As a consequence, taking as basis the dry resin weight, the content of nfe corresponds to 107 mg nfe per gram dr. All experiments were performed using initial concentration of Cr(VI) 15 mg/l and adjusting solution ph to 3. The reduction of Cr(VI) as a function of time is illustrated in Figure 4(a). As seen in the figure increasing the dose of resin

6 Cr(VI), mg/l resulted in faster reduction. As previously the experimental data were plotted as ln(c/c0) versus time (figure 4b) and were found to satisfy the linear relationship, with correlation coefficient R 2 ranging from to 0.994, which is close unit. The values of kinetic constant k1 were calculated by the slopes of the lines and are presented in Table g/l g/l g/l 6 g/l y = x R² = y = x R² = y = x R² = g/l g/l y = x R² = g/l 6 g/l time(min) time(min) (a) Figure 4. (a) Time evolution of Cr (VI) concentration and (b) linear regression of ln(c/c0) versus time in the experiments carried out at different resin doses. (b) Table 2. Values of first order kinetic constant k1 calculated at the different doses, CR, of the resin CR k1 g dr/l min As seen in the Table, there is a systematic increase in the value of constant k1 when the quantity of resin increases. If the reduction of Cr(VI) follows a kinetic law, which is of n order with respect to the resin (and indirectly to the embedded nζvi), then the pseudo first order constant k1 will incorporate the influence of resin concentration according to equation 5. k 1 = k 2 C R n ln (k 1 ) = ln(k 2 ) + n ln(c R ) (5) (6) The logarithmic form of equation 5 suggests that the experimental data, expressed as ln(k1) versus ln(cr), must follow a linear trend with slope corresponding to the order n (equation 6). The linear relationship is confirmed in the graph of Figure 5. Moreover the slope of line was found to be n=1.034 indicating that the reaction is of first order with respect to the resin and essentially to the incorporated nzvi.

7 y = 1.034x R² = Figure 5. Logarithm of constant k1 versus logarithm of resin concentration (CR, g/l) The intersection of the line (-4.079) in Figure 5 corresponds to the logarithm of the second order constant k2. Consequently the value of constant is k2 = in L (g dr) -1 min -1. The 2 nd order chemical constant can be expressed with respect to the embedded nfe, taking into consideration that the resin contains 107 mg of nfe per gram of dry resin. In this case k2 = L(mg nfe) -1 min -1. Therefore, the kinetic equation describing the reduction of Cr(VI) from the embedded nzvi in the Dowex resin can be expressed either by equation 7a or by equation 7b: dc dt = k 2 C R C with k2= L (g dr) -1 min -1, when CR is the amount of dry resin (dr) in g/l, or dc dt = k 2 C nfe C with k2 = L(mg nfe) -1 min -1, when CnFe is the amount of nfe in mg/l (7a) (7b) 4. CONCLUSIONS Two polyphenol solutions, i.e. green tea extract and gallic acid, were tested for the reduction of adsorbed Fe(III) in order to obtain incorporation of nzvi particles on the porous matrix of Dowex 50WX2. Green tea was found to be inefficient, probably due to the relatively big size of the contained polyphenol molecules, but GA molecules were able to reach adsorbed Fe(III) and reduce the cations to the elemental state Fe(0). The nanoiron loaded resin was used for the treatment of Cr(VI) contaminated waters. The parameters investigated were Cr(VI) concentration in the range 5-25 mg/l and amount of resin (1-6 g of dry resin per liter). The experiments were carried out at ph=3 and at constant temperature 25 o C. In the majority of experiments reduction of Cr(VI) was completed within 60 minutes. It was found that the reduction follows a kinetic law of first order with respect to Cr(VI) and to the embedded nanoiron. AKNOWLEDGEMENTS Mystrioti Christiana would like to thank IKY fellowships of excellence for post graduate students in Greece Siemens program and its financial support. Toli Aikaterini is grateful to EDEIL NTUA of excellence for graduate students in scholarship programs. The whole team would like to thank

8 (CHARM) LIFE 10 ENV/GR/ a project titled Chromium in Asopos Groundwater System: Remediation Technologies and Measures (CHARM). REFERENCES 1. Shi, L.N., Zhang, X., Chen, Z.L., Removal of Chromium (VI) from wastewater using bentonite-supported nanoscale zero-valent iron. Water Research 45, Zhang, X., Lin, S., Chen, Z.L., Megharaj, M., Naidu, R., Kaolinite-supported nanoscale zero-valent iron for removal of Pb 2+ from aqueous solution: reactivity, characterization and mechanism. Water Research 45, Park HS., Park YM, Yoo, KM, Lee SH., Reduction of nitrate by resin-supported nanoscale zero-valent iron. Water Science and Technology 59, Shu, H.Y., Chang, M.C., Chen, C.C., Chen, P.E., Using resin supported nano zerovalent iron particles for decoloration of Acid Blue 113 azo dye solution. Journal of Hazardous Materials 184, Fu F., Ma J., Xie L., Tang B., Han W., Lin S Chromium removal using resin supported nanoscale zero-valent iron. Journal of Environmental Management, 128, Owlad M, Aroua M. et al., Removal of Hexavalent Chromium-Contaminated Water and Wastewater. Water Air Soil Pollution 200, Gheju, M., Hexavalent chromium reduction with zero-valent iron (ZVI) in aquatic systems. Water Air Soil Pollution 222 (1-4), O Carroll D, Sleep B. et all, Nanoscale zero valent iron and bimetallic particles for contaminated site remediation. Advances in Water Resources, 51, Mystrioti, C., Papassiopi, N., Xenidis, A., Application of Iron Nanoparticles Synthesized by Green Tea for the Removal of Hexavalent Chromium in Column Tests. Journal of Geoscience and Environment Protection, 2, Mystrioti, C., Sparis, D., Papassiopi, N., Xenidis, A., Dermatas, D., & M. Chrysochoou, M Assessment of Polyphenol Coated Nano Zero Valent Iron for Hexavalent Chromium Removal from Contaminated Waters. Bulletin of Environmental Contamination and Toxicology, 94, Papassiopi, N., Vaxevanidou, K., Christou, C., Karagianni, E., and Antipas, G., Synthesis, characterization and stability of Cr(ΙΙΙ) and Fe(ΙΙΙ) hydroxides. Journal of Hazardous Materials, 264,