Synthesis and Characterization of Antibacterial Silver Nanoparticle

Apr 24, 2013 - Silver nanoparticle (AgNP)-impregnated rice husks/rice hush ash (RHs/RHA) were successfully synthesized, and their potential applicatio...
0 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/est

Synthesis and Characterization of Antibacterial Silver NanoparticleImpregnated Rice Husks and Rice Husk Ash Di He,† Atsushi Ikeda-Ohno,†,‡ Daniel D. Boland,† and T. David Waite*,† †

School of Civil and Environmental Engineering, University of New South Wales, Sydney, NSW 2052, Australia Institute for Environmental Research, Australian Nuclear Science and Technology Organization, Menai, NSW 2234, Australia

‡

S Supporting Information *

ABSTRACT: Silver nanoparticle (AgNP)-impregnated rice husks/rice hush ash (RHs/RHA) were successfully synthesized, and their potential application as antibacterial materials in water disinfection was investigated with particular attention given to the use of both white rice husk ash (WRHA) and black rice husk ash (BRHA) produced by the combustion of RHs as AgNP supports. AgNPs, with diameter of ∼20 nm, were anchored tightly onto RHA, with the emplacement of the AgNPs on these supports increasing the antibacterial activity of the AgNPs through diminution in the extent of nanoparticle aggregation. Ag K-edge XANES analysis revealed that AgNPimpregnated RHs/RHA are composed of both Ag(0) and Ag(I) species with the Ag(I)/Ag(0) ratio following the order WRHA (65:35) > RHs (59:41) > BRHA (7:93). Sodium thioglycolate, a strong Ag(I) ligand, significantly affected the bactericidal activities of AgNP-impregnated RHs/RHA, suggesting that Ag(I) released from AgNP-impregnated RHs/RHA plays an important role in disinfection. The rate constants of oxidative and dissociative dissolution of Ag(0) and Ag(I) species associated with BRHA are 5.0 × 10−4 M−1s−1 and 1.0 × 10−5 s−1, respectively, while those associated with WRHA are 7.0 × 10−2 M−1s−1 and 2.0 × 10−4 s−1 respectively, demonstrating that the rate of dissolution of silver associated with BRHA is particularly slow. As such, the bactericidal “lifetime” of this material is long and exhibits a lower health risk as a result of release of Ag(I) to consumers than does AgNP-impregnated WRHA.

■

INTRODUCTION The synthesis of nanosized particles is a growing research field in the material and chemical sciences, in accordance with the extensive development of nanotechnology.1,2 The size-induced properties of nanoparticles (NPs) enable the development of new applications or the addition of flexibility to existing systems in many areas including catalysis, optics, and microelectronics.3−6 Silver nanoparticles (AgNPs) in particular have attracted considerable attention because of their catalytic, optical, conducting, and antibacterial properties.1,3,4,7,8 The advantages of AgNPs as antibacterial agents include their stability, durability, heat resistance, and broad spectrum antibacterial activity.1,9 Usually, smaller AgNPs with higher specific surface area are preferred due to their high bactericidal ability;10 however, AgNPs with diameters of less than 20 nm tend to aggregate with resultant decrease in antibacterial performance. Additionally, separation of nanosized particles from the solution phase renders their application as a water treatment technology impractical. To solve these problems and improve antibacterial properties, a wide range of materials including titanium dioxide,11 silica,12 aluminum oxide,13 zeolites,14 and activated carbon fibers15 have been employed to support AgNPs such that ultrafine AgNPs can be maintained in nonaggregated form and, prior to release of treated water, separated from the solution phase more readily. The possibility © 2013 American Chemical Society

also exists that the release time of dissolved Ag(I) from AgNPimpregnated supporting materials can be substantially delayed resulting in improved potential for long-term antibacterial applications.16,17 Supporting materials with high specific surface area, complexation capacity, and chemical durability but low cost will be particularly attractive for applications in water disinfection. The Food and Agriculture Organization (FAO) reports that the world annual rice production is approximately 582 million tons with 25 mass % or 145 million tons of this production present as rice husks (RHs).18 These husks are of limited commercial interest and cause serious pollution problems. As such, interest exists in finding applications for these residues. It is well-known that RHs have a low calorific value (3585 kcal kg−1) and a high (20−22 mass %) ash content.19 The rice husk ash (RHA) contains nearly 95 mass % silica and is a potential renewable source of silica. Of particular interest is the potential for use of RHA for the immobilization of AgNPs, since silica is expected to be a good candidate for supporting material.1 Controlled burning is a cheap method of extracting the silica Received: Revised: Accepted: Published: 5276

September 25, 2012 April 7, 2013 April 24, 2013 April 24, 2013 dx.doi.org/10.1021/es303890y | Environ. Sci. Technol. 2013, 47, 5276−5284

Environmental Science & Technology

Article

adventitious carbon. The XPS peak analysis was undertaken using Avantage software. Fourier transform infrared (FTIR) spectra of RHs/RHA were collected with an FTIR spectrophotometer (PerkinElmer Spectrum 100) equipped with an attenuated total reflectance (ATR) cell. The FTIR spectrometer was scanned from 4000 to 650 cm−1 at 1.0 cm−1 intervals. All spectra were obtained by subtraction of the background spectra from the sample spectra and normalized after acquisition to a maximum of 1.0 for comparative purpose. Synthesis of AgNP-Impregnated Rice Husks/Rice Husk Ash. AgNPs were formed and impregnated on RHs/ RHA through chemical reduction of silver nitrate under dark conditions. First, [Ag(NH3)2]+ was formed in the presence of various amounts of RHs/RHA (0.2, 0.6, and 2.0 g) by adding 1.0 mL of 0.5 M AgNO3 stock solution and 0.14 mL of aqueous ammonia (28%, w/w) into 43.9 mL MQ. This mixture was stirred for 1 h, and then, 5.0 mL of 0.5 M glucose stock solution was added to the mixture to reduce [Ag(NH3)2]+ to AgNPs on the surface of the RHs/RHA. After stirring for 1 h, AgNPimpregnated RHs/RHA were harvested and the resulting products were thoroughly washed. The products obtained were classified as Type I (dosage of RHs/RHA is 4 g L−1 during impregnation of AgNPs), Type II (12 g L−1), and Type III (40 g L−1), respectively. Characterization of AgNP-Impregnated Rice Husks/ Rice Husk Ash. Transmission electron microscopy (TEM) using a Philips CM200 TEM microscope was employed to obtain the size distribution and the electron diffraction pattern of AgNPs formed on RHA. TEM sample preparation involved placing a drop of the AgNP-impregnated RHA suspension (50% v/v ethanol) onto a Formvar-covered copper grid (230mesh) and allowing it to dry overnight. N2 adsorption− desorption isotherms were measured at liquid nitrogen temperature (77 K) using a surface area and porosimetry analyzer (Micromeritics Tristar). The samples were pretreated heating at 150 °C for 3 h under a vacuum to remove moisture. Solid phase Ag speciation was investigated using Ag K-edge X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) spectroscopy. Ag Kedge spectra were collected on the XAS beamline at the Australian Synchrotron under the ring operating condition of 3 GeV (top-up mode). A Si(311) double crystal was employed to obtain monochromatic X-rays from the synchrotron. The spectra were collected in fluorescence mode using N2-filled ionization chambers (I0 and I2) and a 100-element Ge fluorescence detector (IF). The measurements were performed in vacuum at 12−13 K. A reference sample of Ag(0) metal was measured in transmission mode using N2-filled ionization chambers under ambient conditions. Reference compounds were diluted with boron nitride powder to form pellets with minimum self-absorption effects. An appropriate amount of the standard compound was mixed with boron nitride (∼1−2 wt % Ag) to acquire sufficient fluorescence signals. The RH/RHA samples were mounted onto sample holders without dilution. The energy of each spectrum was corrected by the first inflection point of Ag(0) metal foil (25.5140 keV). The collected absorption spectra were treated according to a standard procedure21 using the program WinXAS (version 3.2).22 Disinfection Test of AgNP-Impregnated Rice Husks/ Rice Husk Ash. To investigate the bactericidal effect of AgNPimpregnated RHs/RHA, colonies of Escherichia coli K12 and

from RHs for possible commercial use. The burning of RHs in sufficient air leads to production of white rice husk ash (WRHA) containing almost pure (≥95%) silica in a hydrated amorphous form while the combustion of RHs in insufficient air results in formation of black rice husk ash (BRHA) which contains different amounts of carbon and silica depending upon the precise conditions of combustion.18 While not examined extensively, the impregnation of AgNPs on RHA with different properties (i.e., specific surface area, porosity, and residual carbon) would be expected to affect the silver loading, speciation, and dissolution rate, resulting in a range of bactericidal activities. Affordable water filters made up of RHA with nanoparticulate silver incorporated on the silica support material have been developed and marketed by Tata Corporation in India over the last two years20 and would appear to be as, if not more, cost-effective than alternate technologies based on slow release of chlorine (such as the Pureit water filter marketed by Hindustan Unilever). Additionally, while this AgNP-containing household water disinfection device is substantially less expensive than similar capacity units utilizing UV or reverse osmosis-based technologies, precise details of the nature and efficacy of the AgNP-impregnated RHA are unavailable. Therefore, in this study, we examine (i) surface chemistry of RHs/RHA; (ii) loading and speciation of silver impregnated on RHs/RHA as a function of preparation conditions; (iii) antibacterial properties of AgNP-impregnated RHs/RHA using Escherichia coli K12 and Staphylococcus aureus as target bacteria; and (iv) kinetics and mechanisms of silver dissolution from AgNP-impregnated RHs/RHA.

■

MATERIALS AND METHODS Materials. Rice husks (RHs) obtained from SunRice Corporation in Australia were washed with Milli-Q water (MQ) until the supernatant solution became clear. Subsequently, the RHs were soaked in 0.5 M nitric acid for 12 h to remove trace metals and then rinsed with MQ until the supernatant solution reached pH 7. The washed RHs were dried in a 105 °C oven until a constant mass was reached. The RHs were then ashed in porcelain crucibles at 973 K with exposure to this temperature maintained for 2 h (Carbolite Furnance). The burning of RHs in sufficient air led to production of WRHA while the combustion of RHs in insufficient air resulted in production of BRHA. Stock solutions of 0.5 M Ag(I) (∼100% present in the form of Ag+) and Dglucose were prepared by dissolving appropriate amounts of silver nitrate (AgNO3) and D-glucose in sterilized MQ, respectively. For the adjustment of pH, concentrated nitric acid or sodium hydroxide was employed. Experiments were performed in 10.0 mM phosphate buffered solutions (pH 7.0) at a controlled room temperature of 22 °C unless otherwise stated. Characterization of Rice Husks/Rice Husk Ash. Analysis of the elements present at the surface of RHs/RHA was performed by X-ray photoelectron spectroscopy (XPS) (ESCALA250Xi, Thermo Scientific, UK) using a monochromated Al K alpha source. The powdered RHA samples were compressed onto a sample holder while RH samples were mounted on a sample holder by utilizing double-sized adhesive tape. The base pressure inside the analysis chamber was usually maintained at better than 2 × 10−9 mbar, and a pass energy of 20 eV was used for recording all elemental scans. Binding energy was referenced to the C1s line at 285.1 eV from 5277

dx.doi.org/10.1021/es303890y | Environ. Sci. Technol. 2013, 47, 5276−5284

Environmental Science & Technology

Article

Staphylococcus aureus, collected and isolated from municipal wastewater, were incubated in 20.0 mL of 1× minimal media A (MMA) containing 0.2% casamino acid with a shake speed of 150 rpm at 37 °C for 18 h. The cultures were subsequently centrifuged at 2000 rpm for 10 min, and the separated cells were washed 3 times with 40.0 mL of 50.0 mM phosphatebuffered saline (PBS, pH 7.0). The bacteria were then resuspended in 10.0 mL of 10.0 mM phosphate buffer (pH 7.0) at an initial population of ∼1 × 105 CFU mL−1. An appropriate amount of AgNP-impregnated RHs/RHA was then added to the bacterial suspensions such that a final concentration of 1 g L−1 was achieved. After 1 h exposure time, 0.25 mL of neutralizer (10% sodium thioglycolate and 14.6% sodium thiosulfate) was added to prevent inactivation of bacteria by dissolved Ag(I) during the incubation period. The neutralizer was prepared as previously described.23 The number of viable cells was determined by plate counting (CFU mL−1). The concentration of dissolved Ag(I) following filtration of the suspension of E. coli and AgNP-impregnated RHs/RHA (0.2 μm Millex GN, Millipore) was measured using inductively coupled plasma (ICP) (Varian AX; Varian, Australia). The role of dissolved Ag(I) in bactericidal activity of AgNP-impregnated RHs/RHA to E. coli was examined by exposing E. coli for 1 h to 1 g L−1 AgNP-impregnated RHs/RHA in the presence of 2.5% neutralizer. Dissolution Kinetics Modeling. The rate and extent of dissolution of AgNP-impregnated RHs/RHA were quantified by monitoring the appearance of dissolved Ag(I) by ICP (Varian AX; Varian, Australia). An appropriate amount of AgNP-impregnated RHs/RHA was added to pH 7.0 phosphate buffered solutions at a final concentration of 1 g L−1 followed by filtration (0.2 μm Millex GN, Millipore) of the suspension after certain periods. To confirm that no colloidal silver was released from AgNP-impregnated RHs/RHA, the filtrates were further centrifuged for 45 min at 4000 rpm using Amicon centrifugal ultrafilters (Amicon Ultra-15 3K, Millipore, MA) containing porous cellulose membranes with a nominal pore size of 1−2 nm. Minimal difference in silver concentration between filtrates and centrifugal supernatant indicates that ionic silver dominates the released silver (data not shown). The dissolution kinetics under anaerobic conditions were investigated following the addition of an appropriate amount of AgNP-impregnated RHs/RHA into phosphate buffered solutions which had been sparged with argon gas for 2 h beforehand. The anaerobic experiments were performed in an anaerobic chamber (Plas Laboratories, USA) in which the oxygen concentration was ∼0.1%. Kinetic modeling was undertaken using KinTek professional version 3.0. This dynamic kinetic simulation program allows multiple data sets to be fit simultaneously to a single model based on numerical integration of the rate equations describing the reaction mechanism, which facilitates the exploration of initial parameters that serve as the starting point for nonlinear regression in fitting data and the relationships between individual constants and observable reactions.24

carcass and biomass assembled around it. After combustion of the RHs in sufficient air, the corrugated structure shrank and became more compact due to the mineralization of volatile organic products.18 The hard residue was made up of almost pure SiO2 (with a carbon content of only 2.6 atomic %, as shown in Table 1). The structure of the combustion product Table 1. Surface Chemical Analysis on the Elements of Rice Husks/Rice Husk Ash composition (at. %)a

RHsb

BRHAb

WRHAb

C O Si N

59.8 28.8 7.6 3.8

40.9 36.7 21.8 0.6

2.6 62.8 34.4 0

a

at. % represents atomic percentage. bRHs represents rice husks; BRHA and WRHA represent black rice husk ash and white rice husk ash, respectively.

obtained in insufficient air medium (BRHA) was also corrugated but, due to the lower percentage of volatile organic products mineralized,18 the solid residue contained significantly more carbon (40.9 atomic %, as documented in Table 1). It can be observed from Table 1 that the silicon atomic percentage follows the order WRHA (34.4%) > BRHA (21.8%) > RHs (7.6%) while the carbon content follows the order RHs (59.8%) > BRHA (40.9%) > WRHA (2.6%). Characterization of Rice Husks/Rice Husk Ash. FTIR spectroscopy provides information on the chemical structure and surface functional groups of the samples. It can be seen from Figure 1 that the RHs are characterized by a broad band

Figure 1. FTIR spectra of RHs, BRHA, and WRHA.

between 3700 and 3000 cm−1. The absorption band with maximum at about 3349 cm−1 can be attributed to either the stretching vibrations of O−H bonds in water molecules or O− H groups present in cellulose, hemicellulose, and lignin.25,26 The sharp absorption band observed at 2924 cm−1 and the less intense band at 2854 cm−1 can be attributed to the asymmetric and symmetric stretching vibrations of the aliphatic C−H bonds in −CH3 and −CH2 groups in the structures of cellulose, hemicellulose, and lignin, respectively.25,26 The peak at 1730 cm−1 is related to CO stretching vibrations of the bonds in the aldehyde groups of hemicellulose while the band appearing

■

RESULTS AND DISCUSSION Combustion of Rice Husks. Photographs of unground RHs and both WRHA and BRHA after thermal treatment of RHs at 973 K in sufficient and insufficient air, respectively, are shown in Supporting Information Figure S1. As reported by Genieva et al.,18 the outer epidermis of RHs is well organized and has a corrugated structure with relatively stable Si−O 5278

dx.doi.org/10.1021/es303890y | Environ. Sci. Technol. 2013, 47, 5276−5284

Environmental Science & Technology

Article

Figure 2. TEM images of AgNP-impregnated (a) WRHA and (b) BRHA; size distribution of AgNP impregnated on (c) WRHA and (d) BRHA. Note that the dosages of RHA are 40 g L−1 during impregnation of AgNPs (Type III AgNP-impregnated RHA).

at ∼1600 cm−1 is attributed to vibrations corresponding to aromatic CC stretching and asymmetric COO− stretching vibration.25,26 The intense band at 1047 cm−1 corresponds to the stretching vibrations of silicon−oxygen tetrahedrons.18 The relatively weak absorption band at 795 cm−1 is most likely associated with the symmetric and asymmetric vibrations of the Si−O bonds in the silicon−oxygen network.18 After combustion of RHs in insufficient air, the FTIR spectrum of BRHA (Figure 1) differed from that of RHs with the bands at 3349, 2924, 2854, and 1730 cm−1 disappearing. These changes can be explained by the decrease in aliphatic C−H and O−H content of the RHs and the associated transformation into CC and COO− groups. The thermal treatment of RHs in sufficient air (WRHA) resulted in further changes in the FTIR spectrum (Figure 1). In this case, only bands at 1060 and 804 cm−1 are observed with these two bands corresponding to the Si−O stretching vibrations, showing that organic matter is no longer present. In summary, the FTIR spectra of RHs/RHA show that the combustion of RHs in insufficient air partially transform aliphatic carbon functional groups into CC groups and/or oxidizes hydroxyl groups to carboxylate groups while combustion of RHs in sufficient air results in the formation of almost pure amorphous silica. In addition to the characterization of surface functional groups of the samples, BET surface area and pore size parameters of RHA were investigated, as shown in Supporting Information Table S1 and Figure S2, suggesting that BRHA has a higher BET surface area (∼255 m2g−1) compared to that of

WRHA (∼133 m2g−1) (more details are shown in Supporting Information Section S2). Synthesis of AgNP-Impregnated Rice Husks/Rice Husk Ash. In view of the significant proportion of silica present in the RHA, the mechanism of association of AgNPs with RHA is likely to be similar to the mechanism of association of AgNPs with silica.27 The first step is the deprotonation of hydroxyl ligands (−OH) of silica (pKa of amorphous silica is ∼7.028). Alkaline conditions (at least pH 9.0) are required to deprotonate Si−OH to the nucleophilic form (Si−O−) with this form particularly amenable to electrostatic interaction with electrophilic Ag+.1 These first two steps may be represented as follows: Si − O − H + OH− → Si − O− + H 2O

(1)

Si − O− + Ag + → Si − O − Ag

(2) +

The third step involves the reduction of Ag and the growth of AgNPs at the surface of RHs/RHA. Aqueous ammonia was used as base to adjust the pH of suspensions and as a ligand to form the [Ag(NH3)2]+ complex while glucose was chosen as the reductant to reduce Ag+.29 As will be demonstrated in the next section (Figure 2), the AgNPs so formed were attached tightly to the silica surface. The reduction of Ag+ with resultant formation of Ag0 may be represented as follows: 2OH− + 2[Ag(NH3)2 ]+ + C6H12O6 → H 2O + 2Ag 0 + 4NH3 + C6H12O7 5279

(3)

dx.doi.org/10.1021/es303890y | Environ. Sci. Technol. 2013, 47, 5276−5284

Environmental Science & Technology

Article

Characterization of AgNP-Impregnated Rice Husks/ Rice Husk Ash. Transmission electron microscopy (TEM) indicates that the particles produced by the reduction of [Ag(NH3)2]+ by glucose on RHA were indeed nanoparticulate with diameters ranging from 10 to 35 nm (Figure 2a,b). On the basis of the size distribution analysis (Figure 2c, d), the particles formed on WRHA were slightly larger (with average diameter of 21.0 ± 0.5 nm) than those produced on BRHA (18.3 ± 0.5 nm). TEM electron diffraction images of nanoparticles produced on both WRHA and BRHA (Supporting Information Figure S3a,b) display d-spacings of 2.36 Å (1/4.24 nm), 2.04 Å (1/4.89 nm), and 1.20 Å (1/8.09 nm) which are consistent with the d-spacings reported for zero-valent silver (2.36, 2.04, and 1.18 Å).4,30 There is no evidence for the presence of the solid silver species Ag(I)2O which displays d-spacings of 2.72, 2.35, and 1.67 Å31 in either BRHA or WRHA samples. We conclude therefore that the TEM data presented here provide conclusive evidence for the formation of zero-valent silver nanoparticles on the surface of RHA. XANES is sensitive to the density of states available for the photoelectron excitation of absorbing atoms, reflecting the oxidation state of the atoms.21 Figure 3a shows the Ag K-edge

Ag(I) species. Detailed analysis of their EXAFS spectra indicates that the Ag species impregnated in RH/RHA are mainly composed of Ag(0) metal (with the potential presence of Ag(I) silicate and Ag(I) citrate, the residual of the carboxylate groups in RHA) but no silver oxide species (see Supporting Information Figures S4 and S5 and Table S2), which is consistent with our TEM observations (Figure 2). It should be noted that silicate and citrate are expected to interact with Ag(I) via their oxygen atoms in a unidentate fashion with similar bond distances.32−35 Considering this similarity in coordination chemistry, here we select Ag(I) citrate as representative of Ag(I) species in the RHs/RHA samples. Consequently, it is possible to estimate the Ag(0) and Ag(I) ratios in the RH/RHA samples from their edge positions by linear interpolation of the edge values of Ag(0) metal and Ag(I) citrate. As shown in Figure 3b, the Ag(0)/Ag(I) ratio increases in the following order: BRHA (93:7) > RHs (41:59) > WRHA (35:65). The reason why BRHA as a supporting material induces particularly extensive reduction of Ag(I) to Ag(0) is unclear at this stage but may well be associated with the presence of carbon residuals on BRHA since CC groups may enhance the reduction of Ag(I) to Ag(0). The amount of AgNPs immobilized on RHs/RHA is influenced by the concentration of AgNO3 used in the reaction and the surface area of the RHs/RHA. At low AgNO3 concentrations, it is to be expected that only a portion of the active sites of RHs/RHA would be laden with AgNPs; in contrast, at high concentrations of AgNO3, the capacity of the RHs/RHA to maintain separation between AgNPs could well be exceeded with subsequent aggregation of AgNPs on RHs/ RHA and possible resultant decrease in reactivity of AgNPs.27 The atomic percentage of silver in AgNP-impregnated BRHA and WRHA increased from 0.23 to 0.43% and from 0.43 to 0.77%, respectively, with a decrease in the dosage of RHA from 40 to 4 g L−1 during the AgNP-impregnation process (i.e., increase in the dosage ratio of AgNO3 to RHA), as shown in Supporting Information Figure S6a (further details regarding silver loading on RHA are illustrated in Supporting Information Figure S7 and Table S3). Although the dosage of RHs used in the synthesis of AgNP-impregnated RHs decreased from 40 to 4 g L−1, the silver atomic percentage (∼0.15%) did not change significantly (Supporting Information Figure S6a), showing that all the surface sites could be laden with Ag, even at the lowest Ag dosage ratio. Bactericidal Activity of AgNP-Impregnated Rice Husk/ Rice Husk Ash. The results of tests of the ability of AgNPimpregnated RHs/RHA to inactivate E. coli are shown in Figure 4a. For AgNP-impregnated RHs and WRHA, Type I−III showed >4 log inactivation of E. coli after 1 h exposure, whereas only Type I BRHA (4 g L−1 BRHA added during impregnation of AgNPs) exhibited >4 log inactivation while Types II and III showed 1.7 and 1.6 log inactivation, respectively. Sodium thioglycolate, a strong Ag(I) ligand, has previously been found useful in preventing the inactivation of E. coli by dissolved Ag(I) (Supporting Information Figure S8).23,36 Therefore, sodium thioglycolate was used in examining the contribution of dissolved Ag+ to the overall bactericidal ability of AgNPs. The neutralizer (10% sodium thioglycolate and 14.6% sodium thiosulfate23) indeed decreased the inactivation effect of AgNPimpregnated RHs and WRHA on E. coli (Figure 4b), indicating that release of dissolved Ag(I) from AgNP-impregnated RHs and WRHA plays an important role in the inactivation of E. coli. However, this neutralizer had minimal inhibitory effect on the

Figure 3. (a) Ag K-edge XANES spectra for reference Ag compounds (Ag(0) metal, Ag(I)2O, and Ag(I) citrate) and AgNP-impregnated RHs/RHA samples and (b) their calculated edge positions. The Ag(I) ratios in (b) are estimated by the linear interpolation based on the edge positions of Ag(0) metal and Ag(I) citrate. Note that the dosages of RHs/RHA are 40 g L−1 during impregnation of AgNPs (Type III AgNP-impregnated RHs/RHA).

XANES spectra of the AgNP-impregnated RHs and RHA samples, along with the reference compounds of Ag(0) metal, Ag(I)2O, and Ag(I) citrate (as trisilver(I) citrate). The edge position, which is defined as the first inflection point of a XANES spectrum, of Ag(0) metal is 25.5140 keV, while the edge positions of Ag(I)2O and Ag(I) citrate are calculated to be 25.5170 and 25.5171 keV, respectively. The RH and RHA samples exhibit edge positions between these values (25.5142− 25.5160 keV), indicating that they are a mixture of Ag(0) and 5280

dx.doi.org/10.1021/es303890y | Environ. Sci. Technol. 2013, 47, 5276−5284

Environmental Science & Technology

Article

Information Figure S9) was indeed larger than that of Type III (18.3 and 21.0 nm for AgNP-impregnated BRHA and WRHA, respectively, Figure 2c,d). These size-dependent bactericidal activities imply that the bactericidal performances are dependent not only on the silver loadings but also on other properties (such as particle size) of AgNP-impregnated RHs/ RHA. In addition, the toxicity of AgNP-impregnated RHs/ RHA to S. aureus has been investigated, as shown in Supporting Information Figure S10. The almost identical log removal rates for S. aureus and E. coli for similar AgNP RHs/RHA loadings indicate that AgNP-impreganted RHs/RHA have strong bactericidal effects not only on Gram negative bacteria (E. coli) but also on Gram positive bacteria (S. aureus). Rate of Release of Ag(I) from AgNP-Impregnated RHs/RHA. It has been demonstrated here that AgNPimpregnated RHs/RHA can be produced by the glucosemediated reduction of [Ag(NH3)2]+ in the presence of RHs/ RHA with the resultant assemblage exhibiting strong bactericidal activity toward E. coli and S. aureus. In view of the possibility that the bactericidal ability and lifetime of AgNPimpregnated RHs/RHA depend on the release of dissolved Ag(I) from silver-impregnated RHs/RHA, we examine the kinetics of release of Ag(I) species from the assemblages produced. Results of studies of silver dissolution from AgNPimpregnated RHs/RHA are shown in Figure 5. Seven hours after addition of 1 g L−1 AgNP-impregnated RHs/RHA into pH 7.0 phosphate buffered solutions, the concentrations of dissolved Ag(I) released from AgNP-impregnated RHs and WRHA were 1.2 and 5.3 mg L−1, respectively, while the Ag+ concentration from AgNP-impregnated BRHA was significantly lower at only 9.3 μg L−1. Such different dissolution behavior of the silver-impregnated RHs/RHA could be achieved if there is a significant difference in the silver speciation associated with RHs/RHA. As reported by Liu and Hurt,38 the oxygenation rate of AgNPs is rather slow under neutral conditions (with a second-order rate constant of ∼0.014 M−1 s−1), and therefore, AgNP-impregnated BRHA, where zero-valent silver dominates the RHA-associated silver speciation (Figure 3), would be expected, as observed, to exhibit relatively slow dissolution kinetics. The observed minimal effect of oxygen on silver dissolution kinetics (Figure 5) also suggests that the oxidative dissolution of AgNPs associated with RHs/RHA is not the key to release of dissolved Ag(I) from the RHs/RHA surface. Rather, the dissolved Ag(I) concentration may be associated more with the dissociation of Ag(I) species from RHs/RHA surface sites. In addition, the concentration of dissolved Ag(I) also depends on the characteristics of the RHs/RHA. As demonstrated in Supporting Information Figures S11 and S12, BRHA shows a substantially stronger capacity for absorption of dissolved Ag+ than does WRHA and RHs, with this difference possibly associated with its higher specific surface area (as shown in Supporting Information Figure S2 and Table S1) and residual carbon groups.18 In summary, the concentration of dissolved Ag(I) is likely to be determined by (i) oxidative dissolution of AgNPs associated with RHs/RHA and transport from the RHs/RHA matrix to solution (reaction 1 in Table 2), (ii) dissociation of Ag(I) associated with RHs/RHA and transport from the RHs/RHA matrix to solution (reaction 2 in Table 2) and (iii) adsorption equilibrium of dissolved Ag(I) onto RHs/RHA (reaction 3 in Table 2; additional details are shown in Supporting Information Figures S11 and S12). On the basis of the above discussion, we have developed a

Figure 4. Effects of AgNP-impregnated RHs/RHA on inactivation of E. coli (a) in the absence of neutralizer and (b) in the presence of neutralizer. Error bars are sample standard deviation from triplicate measurements. Experimental conditions: initial counts of E. coli are ∼105 CFU mL−1; the initial dosages of AgNP-impregnated RHs/RHA are 1.0 g L−1.

bactericidal ability of Types II and III AgNP-impregnated BRHA toward E. coli with this lack of effect most likely attributed to minimal release of dissolved Ag(I) from AgNPimpregnated BRHA (Supporting Information Figure 6b), with some other mode of AgNP-mediated bactericidal action dominating in these cases. For example, AgNPs have been reported to directly cause cell death as a result of pitting of bacterial cell membranes, thereby increasing membrane permeability and disturbing respiration.37 Alternatively, AgNP dissolution and Ag internalization may occur following direct contact between bacteria and AgNPs. Log reduction of E. coli followed the order Type I < Type II < Type III in the presence of the neutralizer, with the mass ratio of Ag to RHs/RHA increasing in the order Type I > Type II > Type III; the greatest extent of AgNP aggregation and thus the lowest bactericidal ability would be expected for Type I and the highest bactericidal activity for Type III. The hypothesis that higher loading of Ag(0) results in a greater extent of surfaceassociated AgNP aggregation is confirmed by the TEM images which show that the average size of AgNP particles on Type I AgNP-impregnated RHA (26.1 and 30.4 nm for AgNPimpregnated BRHA and WRHA, respectively, Supporting 5281

dx.doi.org/10.1021/es303890y | Environ. Sci. Technol. 2013, 47, 5276−5284

Environmental Science & Technology

Article

simplified kinetic model for the release of dissolved Ag+ from AgNP-impregnated RHs/RHA. Assuming that the transport of Ag+ from RHs/RHA matrix to solution is not rate-limiting and that adsorption of Ag+ to RHs/RHA follows Langmuirian behavior, the reaction set in Table 2 is able to satisfactorily describe the silver release and uptake kinetics (Figure 5 and Supporting Information Figure S12), demonstrating that the rate of release of Ag(I) species associated with RHs/RHA follows the order WRHA (2.0 × 10−4 s−1) > RHs (2.0 × 10−5 s−1) > BRHA (1.0 × 10−5 s−1) with the oxidative dissolution of AgNPs impregnated on RHs/RHA following the same order; namely, WRHA (7.0 × 10−2 M−1s−1) > RHs (3.0 × 10−2 M−1s−1) > BRHA (5.0 × 10−4 M−1s−1). It should be noted however that while ready release of Ag(I) (as a result of either rapid oxidation of Ag(0) groups associated with RHA or dissociation of Ag(I) groups associated with RHA) appears to account for the acute bactericidal activity of AgNP-impregnated RH and WRHA (as confirmed by the effect of the Ag(I)binding neutralizing agent), the lower bactericidal activity of AgNP-loaded BRHA, especially at low loadings of Ag where the neutralizer had little effect on bactericidal activity, appears to be decoupled from Ag(I) release. Possible bactericidal mechanisms for AgNP-impregnated BRHA include direct damage to cell membrane by AgNPs associated with BRHA or AgNP dissolution and resultant silver uptake by bacteria only on direct contact between AgNPs and bacteria. Although a much slower dissolution rate of Ag(I) from AgNP-impregnated BRHA leads to a lower bactericidal activity, a much longer lifetime of AgNPs associated with BRHA results in a long-term bactericidal performance. Environmental Implications. Although AgNP-impregnated RHs and WRHA exhibit very strong bactericidal behavior toward E. coli (Figure 4a) and S. aureus (Supporting Information Figure S10), these antibacterial activities result from the relatively rapid rate of release of Ag+ from AgNPs and Ag(I) species impregnated on RHs/WRHA (Figure 5b) with this rapid release rate potentially posing significant health risks to consumers (US EPA health guideline for silver is 0.1 mg L−1). Compared to RHs and WRHA, however, BRHA exhibits several advantages as a water filtration material including (i) Ag(I) is particularly prone to reduction to AgNPs following addition of ammonia and glucose, resulting in slow silver dissolution from Ag species associated with BRHA with associated extended disinfection lifetime, (ii) low steady-state concentration of dissolved Ag+ contributing to minimal health risk of residual Ag+ to users, and (iii) high specific surface area resulting in long contact time for filtration (see Supporting Information Figure S2 and Table S1). As noted earlier, the reason why BRHA as a supporting material induces the

Figure 5. Effects of oxygen on the release of dissolved silver from AgNP-impregnated (a) BRHA and (b) WRHA and RHs. Error bars are sample standard deviation from triplicate measurements. Symbols are experimental data; lines are model values where (a) [Ag(0)≡BRHA]0 = 1.1 × 10−4 M and [Ag(I)≡BRHA]0 = 8.5 × 10−6 M; (b) [Ag(0)≡WRHA]0 = 5.8 × 10−5 M, [Ag(I)≡WRHA]0 = 1.1 × 10−4 M, [Ag(0)≡RH]0 = 1.7 × 10−5 M, and [Ag(I)≡RH]0 = 2.5 × 10−5 M as determined from total Ag content and XAS results. Experimental conditions: The initial dosages of AgNP-impregnated RHs/RHA are 1.0 g L−1. Note that the dosages of RHs/RHA are 40 g L−1 during impregnation of AgNPs (Type III AgNP-impregnated RHs/RHA).

Table 2. Proposed Model for Silver Dissolution of AgNP-Impregnated RHs/RHAa k no.

reaction

RH

WRHA

BRHA

1 2

Ag(0)≡RH/RHAb + O2 → Ag+ + RH/RHA Ag(I)≡RH/RHAb → Ag+ + RH/RHA

3.0 × 10−2 M−1s−1 2.0 × 10−5 s −1

7.0 × 10−2 M−1s−1 2.0 × 10−4 s−1

5.0 × 10−4 M−1s−1 1.0 × 10−5 s−1

3

Ag + + RH/RHA ⇌ Ag(I)‐RH/RHA c

k+

NA

1.8 M−1

k−

Kads = k+/k−

1.1 × 105 M−1

The dosages of RHs/RHA are 40 g L−1 during impregnation of AgNPs (Type III AgNP-impregnated RHs/RHA). bAg(0)≡RH/RHA and Ag(I)≡RH/RHA represent zero-valent Ag nanoparticles and Ag(I) species impregnated on RH/RHA, respectively, following reduction of [Ag(NH3)2]+ by glucose on RH/RHA with initial concentration determined from total Ag content and XAS results. cAg(I)-RH/RHA represents the complex formed following adsorption of Ag+ to RH/RHA. a

5282

dx.doi.org/10.1021/es303890y | Environ. Sci. Technol. 2013, 47, 5276−5284

Environmental Science & Technology

Article

Notes

reduction of Ag(I) to Ag(0) is unclear at this stage. The presence of carbon residuals on BRHA could be one possible explanation, since CC groups might enhance the reduction of Ag(I) to Ag(0) on BRHA to some extent. As such, further work on modification of the BRHA structure (e.g., increase in the content of CC groups) with subsequent examination of the effect on reduction of Ag(I) to Ag(0) would be beneficial in clarifying the role of certain organic functional groups in this redox process. Additionally, further work on the mechanism of bactericidal action of such materials is required in view of the retention of bactericidal activity in the presence of agents which deactivate the impact of Ag+ ions. A good example of AgNP-impregnated BRHA is the Swach water filter developed and marketed by Tata Corporation in India which sells for around US$40 and is claimed to be capable of treating 3000 L of water per cartridge. In terms of AgNPimpregnated BRHA, bactericidal properties depend on silver loading while the material lifespan is determined by Ag release kinetics. Tata Swach shows around 1.0 log inactivation of E. coli after 1 h exposure while our AgNP-impregnated BRHA exhibits approximately 1.7 log reduction of E. coli, which can be explained by the lower silver loading (0.08%) of Tata Swach compared to our AgNP-impregnated BRHA (0.23%), as shown in Supporting Information Table S4. The low silver loading of Tata Swach could result from a low dosage ratio of Ag(I) salts to BRHA, which is an effective way to reduce the production cost. On the other hand, Ag K-edge XANES analysis (see Supporting Information Figure S13) shows that the ratio of Ag(0) to Ag(I) associated with Tata Swach (95/5) is slightly higher than that of our BRHA sample (93/7) with the higher proportion of Ag(0) possibly resulting from the use of sodium borohydride (a strong reductant) in producing the Tata Swach.20 However, the synthesis method using glucose as a weak reductant and ammonia as a catalyst is particularly promising in view of the low chemical cost compared to sodium borohydride. Optimization in cost versus performance may be achieved by further investigation of performance as a function of silver loading and the ratio of Ag(0) to Ag(I) achieved for particular production conditions. In addition, continuous flow tests need to be undertaken to quantify the removal rates of pathogens, including bacteria and viruses, and the material lifetime for different water quality conditions.

■

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Dr. Bin Gong from the Solid State and Elemental Analysis Unit, the University of New South Wales, for assistance with XPS analysis and acknowledge funding support from the Australian Research Council through Discovery Project DP120103222. XAS measurements were undertaken on the XAS beamline at the Australian Synchrotron under proposal number 4940. We acknowledge the beamline staff for technical support.

■

(1) Kim, Y. H.; Lee, D. K.; Cha, H. G.; Kim, C. W.; Kang, Y. S. Synthesis and characterization of antibacterial Ag-SiO2 nanocomposite. J. Phys. Chem. C 2007, 111, 3629−3635. (2) Murphy, C. J.; Sau, T. K.; Gole, A.; Orendorff, C. J. Surfactantdirected synthesis and optical properties of one-dimensional plasmonic metallic nanostructures. MRS Bull. 2005, 30, 349−355. (3) He, D.; Jones, A. M.; Garg, S.; Pham, A. N.; Waite, T. D. Silver nanoparticle-reactive oxygen species interactions: Application of a charging-discharging model. J. Phys. Chem. C 2011, 115, 5461−5468. (4) He, D.; Garg, S.; Waite, T. D. H2O2-mediated oxidation of zerovalent silver and resultant interactions among silver nanoparticles, silver ions, and reactive oxygen species. Langmuir 2012, 28, 10266− 10275. (5) Farrer, R. A.; Butterfield, F. L.; Chen, V. W.; Fourkas, J. T. Highly efficient multiphoton-absorption-induced luminescence from gold nanoparticles. Nano Lett. 2005, 5, 1139−1142. (6) Jana, N. R.; Gearheart, L.; Murphy, C. J. Wet chemical synthesis of high aspect ratio cylindrical gold nanorods. J. Phys. Chem. B 2001, 105, 4065−4067. (7) Berti, L.; Alessandrini, A.; Facci, P. DNA-templated photoinduced silver deposition. J. Am. Chem. Soc. 2005, 127, 11216−11217. (8) Southward, R. E.; Thompson, D. W.; StClair, A. K. Control of reflectivity and surface conductivity in metallized polyimide films prepared via in situ silver(I) reduction. Chem. Mater. 1997, 9, 501− 510. (9) Chudasama, B.; Vala, A. K.; Andhariya, N.; Upadhyay, R. V.; Mehta, R. V. Enhanced antibacterial activity of bifunctional Fe3O4-Ag core-shell nanostructures. Nano Res. 2009, 2, 955−965. (10) Choi, O.; Hu, Z. Q. Size dependent and reactive oxygen species related nanosilver toxicity to nitrifying bacteria. Environ. Sci. Technol. 2008, 42, 4583−4588. (11) Akhavan, O. Lasting antibacterial activities of Ag-TiO2/Ag/aTiO2 nanocomposite thin film photocatalysts under solar light irradiation. J. Colloid Interface Sci. 2009, 336, 117−124. (12) Oh, S. D.; Lee, S.; Choi, S. H.; Lee, I. S.; Lee, Y. M.; Chun, J. H.; Park, H. J. Synthesis of Ag and Ag-SiO2 nanoparticles by gammairradiation and their antibacterial and antifungal efficiency against Salmonella enterica serovar Typhimurium and Botrytis cinerea. Colloids Surf., A 2006, 275, 228−233. (13) Chang, Q. Y.; Yan, L. Z.; Chen, M. X.; He, H.; Qu, J. H. Bactericidal mechanism of Ag/Al2O3 against Escherichia coli. Langmuir 2007, 23, 11197−11199. (14) Inoue, Y.; Hoshino, M.; Takahashi, H.; Noguchi, T.; Murata, T.; Kanzaki, Y.; Hamashima, H.; Sasatsu, M. Bactericidal activity of Agzeolite mediated by reactive oxygen species under aerated conditions. J. Inorg. Biochem. 2002, 92, 37−42. (15) Chen, S. X.; Liu, J. R.; Zeng, H. M. Structure and antibacterial activity of silver-supporting activated carbon fibers. J. Mater. Sci. 2005, 40, 6223−6231. (16) Lee, D.; Cohen, R. E.; Rubner, M. F. Antibacterial properties of Ag nanoparticle loaded multilayers and formation of magnetically directed antibacterial microparticles. Langmuir 2005, 21, 9651−9659.

ASSOCIATED CONTENT

S Supporting Information *

Additional text describing BET and EXAFS analysis of AgNPimpregnated RHs/RHA; eleven figures showing appearance of RHs and RHA, k space and R space of silver species impregnated on RHs/RHA, silver loading impregnated on RHs/RHA, TEM morphology of our AgNP-impregnated RHA and Tata Swach, bactericidal ability of AgNP-impregnated RHs/RHA to S. aureus, adsorption behaviors of RHA, nitrogen adsorption−desorption of AgNP-impregnated RHA and XANES spectra of Tata Swach, respectively; four tables showing the EXAFS structural parameters for Ag standards, the surface chemical analysis of AgNP-impregnated RHA, BET surface area, and pore size distribution as well as bactericidal activities of Tata Swach, respectively. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. 5283

dx.doi.org/10.1021/es303890y | Environ. Sci. Technol. 2013, 47, 5276−5284

Environmental Science & Technology

Article

(17) Zhang, L. Z.; Yu, J. C.; Yip, H. Y.; Li, Q.; Kwong, K. W.; Xu, A. W.; Wong, P. K. Ambient light reduction strategy to synthesize silver nanoparticles and silver-coated TiO2 with enhanced photocatalytic and bactericidal activities. Langmuir 2003, 19, 10372−10380. (18) Genieva, S. D.; Turmanova, S. C.; Dimitrova, A. S.; Vlaev, L. T. Characterization of rice husks and the products of its thermal degradation in air or nitrogen atmosphere. J. Therm. Anal. Calorim. 2008, 93, 387−396. (19) Chaudhary, D. S.; Jollands, M. C. Characterization of rice hull ash. J. Appl. Polym. Sci. 2004, 93, 1−8. (20) Rautaray, D.; Sastry, M.A Process for Forming a Rice Husk Ash Composition, International Patent Application No. PCT/IN2008/ 000812; World Intellectual Property Organization, 2010; Publication No. WO/2010/010570. (21) Prins, R.; Koningsberger, D. E. X-Ray Absorption: Principles, Applications, Techniques for EXAFS, SEXAFS, and XANES; WileyInterscience: New York, 1988. (22) Ressler, T. WinXAS: a program for X-ray absorption spectroscopy data analysis under MS-Windows. J. Synchrotron Radiat. 1998, 5, 118−122. (23) Tilton, R. C.; Rosenberg, B. Reversal of silver inhibition of microorganisms by agar. Appl. Environ. Microb. 1978, 35, 1116−1120. (24) Johnson, K. A.; Simpson, Z. B.; Blom, T. Global Kinetic Explorer: A new computer program for dynamic simulation and fitting of kinetic data. Anal. Biochem. 2009, 387, 20−29. (25) He, D.; Guan, X. H.; Ma, J.; Yu, M. Influence of different nominal molecular weight fractions of humic acids on phenol oxidation by permanganate. Environ. Sci. Technol. 2009, 43, 8332− 8337. (26) He, D.; Guan, X.; Ma, J.; Yang, X.; Cui, C. Influence of humic acids of different origins on oxidation of phenol and chlorophenols by permanganate. J. Hazard. Mater. 2010, 182, 681−688. (27) Zhang, X. L.; Niu, H. Y.; Yan, J. P.; Cai, Y. Q. Immobilizing silver nanoparticles onto the surface of magnetic silica composite to prepare magnetic disinfectant with enhanced stability and antibacterial activity. Colloids Surf., A 2011, 375, 186−192. (28) Santra, S.; Yang, H.; Dutta, D.; Stanley, J. T.; Holloway, P. H.; Tan, W. H.; Moudgil, B. M.; Mericle, R. A. TAT conjugated, FITC doped silica nanoparticles for bioimaging applications. Chem. Commun. 2004, 2810−2811. (29) Panacek, A.; Kvitek, L.; Prucek, R.; Kolar, M.; Vecerova, R.; Pizurova, N.; Sharma, V. K.; Nevecna, T.; Zboril, R. Silver colloid nanoparticles: Synthesis, characterization, and their antibacterial activity. J. Phys. Chem. B 2006, 110, 16248−16253. (30) Jones, A. M.; Garg, S.; He, D.; Pham, A. N.; Waite, T. D. Superoxide-mediated formation and charging of silver nanoparticles. Environ. Sci. Technol. 2011, 45, 1428−1434. (31) Xiao, Y.; Li, Y.-S.; Swihart, G. H. A close look at silver oxide colloid for surface-enhanced Raman scattering. Talanta 2002, 58, 755−760. (32) Linke, C.; Jansen, M. Subvalent ternary silver-oxides - synthesis, structural characterization, and physical-properties of pentasilver orthosilicate, Ag5SiO4. Inorg. Chem. 1994, 33, 2614−2616. (33) Klein, W.; Jansen, M. Synthesis and crystal structure of silver nesosilicate, Ag4SiO4. Z. Anorg. Allg. Chem. 2008, 634, 1077−1081. (34) Sagatys, D. S.; Smith, G.; Bott, R. C.; Lynch, D. E.; Kennard, C. H. L. Preparation and crystal-structure of polymeric ammonium silver(I) citrate hydrate, (NH4[Ag2(C6H5O7)(H2O)])N. Polyhedron 1993, 12, 709−713. (35) Fischer, A. Trisilver(I) citrate. Acta Crystallogr. E 2011, 67, M255−U1507. (36) Park, H.-J.; Kim, J. Y.; Kim, J.; Lee, J.-H.; Hahn, J.-S.; Gu, M. B.; Yoon, J. Silver-ion-mediated reactive oxygen species generation affecting bactericidal activity. Water Res. 2009, 43, 1027−1032. (37) Shahverdi, A. R.; Fakhimi, A.; Shahverdi, H. R.; Minaian, S. Synthesis and effect of silver nanoparticles on the antibacterial activity of different antibiotics against Staphylococcus aureus and Escherichia coli. Nanomedicine: NBM 2007, 3, 168−171.

(38) Liu, J. Y.; Hurt, R. H. Ion release kinetics and particle persistence in aqueous nano-silver colloids. Environ. Sci. Technol. 2010, 44, 2169−2175.

5284

dx.doi.org/10.1021/es303890y | Environ. Sci. Technol. 2013, 47, 5276−5284