- Email: [email protected]
Silver nanoparticles-loaded activated carbon fibers using chitosan as binding agent: Preparation, mechanism, and their antibacterial activity
Applied Surface Science 394 (2017) 457–465
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Full length article
Silver nanoparticles-loaded activated carbon fibers using chitosan as binding agent: Preparation, mechanism, and their antibacterial activity Chengli Tang a,∗ , Dongmei Hu b , Qianqian Cao a , Wei Yan c , Bo Xing a a
College of Mechanical and Electrical Engineering, Jiaxing University, Jiaxing 314001, PR China College of Mechanical Science and Engineering, Jilin University, Changchun 130022, PR China c Department of Environmental Science and Engineering, Xi’an Jiaotong University, Xi’an 710049, PR China b
a r t i c l e
i n f o
Article history: Received 26 May 2016 Received in revised form 1 October 2016 Accepted 17 October 2016 Available online 21 October 2016 Keywords: Silver nanoparticles Chitosan Activated carbon fibers Molecular dynamics simulation Binding agent
a b s t r a c t The effective and strong adherence of silver nanoparticles (AgNPs) to the substrate surface is pivotal to the practical application of those AgNPs-modified materials. In this work, AgNPs were synthesized through a green and facile hydrothermal method. Chitosan was introduced as the binding agent for the effective loading of AgNPs on activated carbon fibers (ACF) surface to fabricate the antibacterial material. Apart from conventional instrumental characterizations, i. e., scanning electron microscope (SEM), X-ray diffraction (XRD), Fourier Transform Infrared Spectroscopy (FT-IR), zeta potential and BrunauerEmmett-Teller (BET) surface area measurement, molecular dynamics simulation method was also applied to explore the loading mechanism of AgNPs on the ACF surface. The AgNPs-loaded ACF material showed outstanding antibacterial activity for S. aureus and E. coli. The combination of experimental and theoretical calculation results proved chitosan to be a promising binding agent for the fabrication of AgNPs-loaded ACF material with excellent antibacterial activity. © 2016 Elsevier B.V. All rights reserved.
1. Introduction The rapid development of nanotechnology enables the controllable synthesis of metal silver to nano-scale [1–3]. Silver nanoparticles (AgNPs), as a broad spectrum antimicrobial agent, have been used extensively in therapeutic applications such as catheters [4,5], surgical devieces [6] and wound dressings [7–9]. However, the antibacterial efficiency of AgNPs depends on both colloidal stability and particle size [10]. Additionally, the practical application of AgNPs in colloidal form are often confronted with the problem of aggregation and loss of antibacterial activity [11–13]. Therefore, the stability of AgNPs is an important issue for their practical application as antibacterial material [14,15]. An effective way to prevent AgNPs from aggregation is to deposite AgNPs on specific surface to fabricate AgNPs-loaded materials. Apart from the better stability, these kinds of materials are easy to be recycled and reused compared with the colloidal form of AgNPs. The common supporting materials (substrates) are SiO2 , zeolite and carbon materials [16–18]. Activated carbon fiber is an
∗ Corresponding author. E-mail address: [email protected] (C. Tang). http://dx.doi.org/10.1016/j.apsusc.2016.10.095 0169-4332/© 2016 Elsevier B.V. All rights reserved.
ideal supporting material for AgNPs loading because of the huge specific surface area, proper micropores and excellent adsorption capacity, which have been widely used in water treatment field [19,20]. The strong adherence of AgNPs to the supporting surface is critical for the practical feasibility of the AgNPs-loaded materials. Binding agent is needed to guarantee the effective connection of AgNPs and supporting surface. For instance, butane tetracarboxylic acid was applied as the cross-linking agent to bind AgNPs on the surface of cotton [21]. As a natural polysaccharide, chitosan is rich in amino groups, which results in high percentage of nitrogen (6.89%) [22]. Chitosan has wide applications in medical science, food industry, cosmetic industry, agriculture and environmental field [23–25]. In this paper, we introduced chitosan as the binding agent for the first time to prepare AgNPs-loaded ACF. Chitosan is a natural polysaccharide, which has advantages over synthetic ones when used for water purification. In the research method, we combined the two aspects of experimental result and theoretical calculation together. Specifically, molecular dynamics simulation method was applied except for the conventional instrumental characterizations of scanning electron microscope (SEM), energy dispersive spectrum (EDS), X-ray diffraction (XRD), Fourier Transform Infrared Spectrum (FT-IR), zeta potential and Brunauer-Emmett-Teller (BET)
458
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
surface area. Instrumental characterizations were carried out to observe the morphology, crystalline structure, chemical bonds, surface charge and specific surface area of the material. Besides, molecular dynamics simulation method was also applied to explore the loading mechanism of AgNPs on ACF surface. This combination of experimental result and theoretial calculation may provide a new way for the material science researchers in mechanism studying. The AgNPs-loaded ACF material was also tested for antibacterial activity to evaluate its use potential. 2. Experimental section 2.1. Materials and reagents All the chemical reagents of sodium hydroxide (NaOH), hydrogen chloride (HCl) were purchased from Shanghai Chemical Reagent Co. Ltd (Shanghai, China). Silver nitrate (AgNO3 ) was primary reagent. Maltose was biochemical reagent purchased from Beijing AOBOX biological technology Co. Ltd (Beijing, China). All the chemicals were used as received without further purification. All the ultrapure water used in this study was produced by EPED system (Nanjing, China). ACF cloth was purchased from Nantong Yongtong Environmental Protection Science and Technology Co., Ltd (Nantong, China). 2.2. Synthesis of silver nanoparticles Silver nanoparticles used in this study were prepared by hydrothermal method similarly as described by Tang et. al [26]. Typically, 15 mL of AgNO3 aqueous solution (0.02 M) and 15 mL of maltose aqueous solution (0.02 M) were added to 30 mL of deionized water with magnetic stirring. Then the mixture was adjusted to pH value of about 11 by NaOH (1 M) and transferred into a Teflonlined stainless steel autoclave (100 mL volume) and maintained at 160 ◦ C for 15 min in an electronic oven. After the autoclave cooled down naturally, the precipitate at the bottom of the reactor was sufficiently washed with water and then dried at 60 ◦ C for further usage.
by a pore size and specific surface area analyzer (BET, Builder, SSA4300). The loading amount of AgNPs on ACF surface as well as Ag+ ions concentration released in the disinfection process was checked by the Inductively Coupled Plasma (ICP) measurement in ICPE-9000 system (Shimadzu, Japan). 2.5. Coarse-grained molecular dynamics simulation Coarse-grained molecular dynamics simulation was carried out to explore the interactions between the chitosan, AgNPs and ACF surface, aiming at revealing the loading mechanism of AgNPs on ACF surface. All simulations were conducted in reduced LJ units. The masses of chitosan monomer, silver nanoparticle and counterion were set to 1.0 to reduce the equilibrium time. Furthermore, implicit solvent model was used. A flat surface consists of uniformly distributed wall particles was used to mimic the carbon surface. Single chitosan chain was coarse-grained as a “bead-spring” model. Unit positive charge was carried by each chitosan monomer. AgNPs were represented by face-centered cubic lattice structural particles with a fixed radius of 8. To simulate the negative charge property of AgNPs, unit negative charges were randomly distributed on the nanoparticle surface. The distance between negative charges and nanoparticle center was greater than or equal to 7.92. In this study, the number of monomers in single chitosan chain was set to 100. The number of silver nanoparticles was to be 8. The sizes of simulation box was Lx = Ly = 50, Lz = 100. 2.6. Antibacterial activity of the AgNPs-loaded ACF Antibacterial property of the as-prepared chitosan-grafted ACF and AgNPs-loaded ACF materials were tested by the disk-diffusion method [27]. The water disinfection experiment was carried out in a circulated system (Fig. S1) using the AgNPs-loaded ACF material as the filter. Typical gram-positive and gram-negative bacteria of Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli) were selected as the model bacterial with initial bacterial concentration of 7 × 105 cfu mL−1 . The antibacterial efficiency was determined by comparing the inflow and effluent bacterial concentration via plate counting method [28].
2.3. Preparation of the AgNPs-loaded ACF
3. Results and discussion
To carry out attachment of chitosan onto ACF, pre-weighed quantity of ACF were put in acetic acid solution (0.4%, v/v) containing 0.2% (wt) chitosan, 0.2% (wt) crosslinker citric acid, and 0.15% (wt) sodium dihydrogen phosphate and esterification was carried out at 180 ◦ C for 10 min. Then chitosan-grafted ACF was placed at 60 ◦ C in an oven until the ACF were completely dry. A dry pre-weighed piece of chitosan-grafted ACF was put in AgNPs suspension prepared by dissolving 20 mg of AgNPs in 100 mL of ultrapure water for next 24 h. Then the ACF were washed in deionized water and dried at 60 ◦ C in an oven for further characterization and use.
3.1. Preparation and characterization of AgNPs-loaded ACF
2.4. Characterization of the AgNPs-loaded ACF The morphologies of the samples were observed by a JSM-6390A scanning electron microscope (SEM, JEOL, Japan) with an EDS system. The crystalline structures of all the samples were examined by X-ray diffraction meter (XRD, X’pert PRO with Cu K␣ radiation). FT-IR spectra of the as-prepared materials was collected by using a FT-IR spectrophotometer (Vertex 70, Germany) in the range of 4000–400 cm−1 at a resolution of 2 cm−1 . Zeta potential of the samples (dispersed sample powders in deionized water) were measured by Malvern zetasizer NanoZS 90 (Malvern, England). BET surface area of ACF before and after AgNPs loading were measured
3.1.1. SEM observation The surface morphologies of the differently treated ACF were observed by SEM images. The as-obtained ACF surface is smooth and clean as shown in Fig. 1(a), while numberable particles were found on the smooth ACF surface after the ACF was impregnated in AgNPs suspension (Fig. 1(b)). When the ACF was immersed in chitosan acetic acid solution (0.4%, v/v, containing 0.2% chitosan, 0.2% citric acid, and 0.15% sodium dihydrogen phosphate), followed by impregnation in AgNPs suspension for 24 h, there exists plentiful particles on the ACF surface (Fig. 1(c), (d)). As can be seen, the AgNPs loaded on ACF surface present piles of 1–5 particles. Since no additional dispersant or stabilizer was used, the AgNPs tend to accumulate in the water suspension because of their huge specific surface area and surface energy. We can also see from the SEM image (Fig. 1(c)) that the AgNPs distributed all around the ACF surface, which will be beneficial to the antibacterial process. SEM images of the three samples show apparently that the chitosantreated process increased the particle number on ACF surface. 3.1.2. EDS analysis EDS measurement was carried out to confirm the elemental composition of the particles on the ACF surface, the results are
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
459
Fig. 1. SEM images of bare ACF (a), ACF impregnated in AgNPs suspension (b), and AgNPs-loaded ACF (c), (d).
Fig. 2. EDS results of AgNPs-loaded ACF (the inserted table is the elemental atoms results).
shown in Fig. 2. As can be seen from Fig. 2(b) and Fig. 2(c), there are carbon, oxygen and silver elements peaks in the EDS spectrum taken from spot 1, while there are only carbon and oxygen peaks taken from spot 2. The elemental atoms results in the inserted table also confirm that the particles formed on the ACF surface is composed of silver.
3.1.3. XRD analysis XRD analysis was applied to confirm the crystalline structure of the particles on ACF surface. As shown in Fig. 3, the broad XRD reflection peaks at 2 angle of 25.8◦ and 44◦ could be attributed to the diffraction of the amorphous structure of ACF. The four diagnostic diffraction peaks at 2 angle of 38.2◦ , 44.4◦ , 64.6◦ and
460
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
Fig. 3. XRD pattern of the ACF and AgNPs-loaded ACF.
77.5◦ correspond to the diffraction of (111), (200), (220) and (311) planes of face-centered-cubic silver, which evidently revealed that the particles loaded on the ACF surface are of metallic silver Ag0 . Besides, the peaks belong to amorphous carbon remained unchanged, indicating that the loading process could not affect the carbon crystallography structure. Both of the EDS and XRD results demonstrated that AgNPs have been successfully loaded onto the ACF surface. Besides, the ICP measurement shows that the AgNPs loaded on ACF surface weighed to be 0.076 mg g−1 ACF. 3.1.4. FT-IR spectra FT-IR technology was used to explore the interaction between the composites in AgNPs-loaded ACF material so as to reveal the loading mechanism. The infrared spectra of the as-obtained ACF, chitosan-grafted ACF and AgNPs-loaded ACF are shown in Fig. 4. There is a broad peak in the range from 3707 cm−1 to 3024 cm−1 (peak at 3433 cm−1 ) in the ACF IR spectrum, which was the stretching vibration peak of OH group (OH ). For the chitosan-grafted ACF and AgNPs-loaded ACF, the peak became broader. Because of the N H stretching vibration (NH ) of amino group, which is at ca. 3400 cm−1 , it overlapped with OH stretching vibration, resulting in a broader peak. For the chitosan-grafted ACF, the peak is in the range from 3782 cm−1 to 2997 cm−1 (peak at 3439 cm−1 ), and AgNPs-loaded ACF from 3703 cm−1 to 3006 cm−1 (peak at 3429 cm−1 ) [29]. The new peaks at wavenumber of 2931 cm−1 and 2916 cm−1 , which is nearly invisible in ACF IR spectrum, was assigned to the stretching vibration (CH ) of chitosan backbone [30]. Compared with ACF, the peak intensity at ca. 1640 cm−1
Fig. 4. FT-IR spectra of the ACF, chitosan-grafted ACF and AgNPs-loaded ACF.
got enhanced for chitosan-grafted ACF because of the primary amine in chitosan [30]. But there showed an apparent weakness for the AgNPs-loaded ACF. The obvious lower peak intensity after AgNPs loading is supposed to caused by the conbination of -NH2 with AgNPs in the obtained AgNPs-loaded ACF composites. The peak at 1519 cm−1 , 1531 cm−1 and 1521 cm−1 for ACF, chitosan-grafted ACF and AgNPs-loaded ACF underwent the similar experience. There are also new peaks at wavenumber of 1080 cm−1 and 1033 cm−1 for chitosan-grafted ACF and AgNPs-loaded ACF, which are the absorption peaks of the deformation vibration of N H (␦NH ) in amide I [30]. The band located 1145 cm−1 and 1151 cm−1 for chitosan-grafted ACF and AgNPs-loaded ACF are related to asymmetric vibrations of C O in oxygen bridge, resulting from deacetylation of chitosan [31]. The whole IR results of the three materials demonstrated that the chitosan was successfully coated on the ACF surface and it was favorable for the AgNPs loading because of the abundant NH2 groups in chitosan molecule. 3.1.5. Zeta potential of different materials Zeta potential of different suspensions were carried out to determine the surface charge of the corresponding sample. As shown in Table 1, the zeta potential of the as-obtained ACF and hydrothermally formed AgNPs (in aqueous solution) valued to be −20.3 mV, indicating that both of the ACF and AgNPs surface are negatively charged. Since both of the ACF surface and AgNPs are negatively charged, it is difficult for them to attract each other without assis-
Fig. 5. N2 adsorption-desorption isotherm and BJH pore size distribution (inserted) of the ACFs (a) and AgNPs-loaded ACFs (b).
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
461
Fig. 6. Typical adsorption process of AgNPs on chitosan-grafted ACF surface (a)–(f), and the final conformation of silver nanoparticle adsorption (g). Purple beads: coarsegrained non-charged silver nanoparticles, blue beads: negatively charged silver particles, green beads: chitosan monomers, silver beads: carbon surface. To show a clear snapshot, carbon surface and counterions are not included in (g). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Table 1 Zeta potential of different materials.
Zeta potential/mV
ACF
Chitosan
AgNPs
Chitosangrafted ACF
AgNPs-loaded ACF
−20.3
98.6
−20.3
33.2
−2.89
tance. The chitosan exhibited positively charged surface with a zeta potential of 98.6 mV. The positive surface charge could be imparted by protonated amino groups (-NH3 + ) of chitosan molecule. The total zeta potential of the chitosan-grafted ACF valued to be 33.2 mV, indicating that the ACF surface was completely coated with a layer of chitosan. When the chitosan-grafted ACF was loaded with AgNPs, the composite was negatively charged again with the zeta potential of −2.89 mV. The negative zeta potential of the AgNPs-loaded ACF
462
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
supported the above assumption that the chitosan served as the binding agent for the successful loading of AgNPs on ACF surface. 3.1.6. BET surface area measurement BET surface area of the raw ACF and AgNPs-loaded ACF was calculated according to BET method. The N2 adsorption-desorption isotherms of ACF and AgNPs-loaded ACF and the corresponding pore size distribution are shown in Fig. 5. Both of the hysteresis loops before (Fig. 5(a)) and after (Fig. 5(b)) AgNPs-loading resemble type H4 in IUPAC lassification [32], indicating that the slit pores and laminated structure of the ACF remained unchanged after AgNPs loading. The BET surface area of ACF was calculated to be 1020 m2 g−1 . There was a decrease of 19.3% in the specific surface area, but the AgNPs-loaded ACF remained the excellent adsorption capacity of ACF with a specific area of 822.480 m2 g−1 . The pore size distribution results show that the pores number at radius of 80–90 nm decreased deastically, while no obvious change was obversed with other pore sizes. The number decreased was probably caused by the chitosan coated on the ACF surface. Although admitting that there was an apparent decrease in pores number at radius of 80–90 nm, the AgNPs-loaded ACF remained excellent adsorption capacity of ACF, which is extremely useful for bacteria adsorption and inactivation. 3.2. Coarse-grained molecular dynamics simulation Coarse-Grained molecular dynamics simulation was carried out to study the mechanism of AgNPs loading on single chitosan chain in dilute salt solution at molecular scale. To model the soft chitosan chain, finite extensible nonlinear elastic (FENE) “spring” was used to simulate the chemical bonds between adjacent chitosan monomers [33].
UFENE = −0.5KR02 ln 1 −
r 2
(1)
R0
where, r is the distance between two bonded beads, R0 = 1.5 is the maximum bond length, and K is a spring constant which is chosen to be 30.0 εLJ /( LJ 2 ) [34]. εLJ and LJ are reduced LJ units. Shortand long-range interactions between non-bonding coarse grained beads are represented by Lennard-Jones potential and Coulombic potential. Long-range par of Coulombic interaction is calculated by Particle-particle particle-mesh solver [35].
ULJ =
⎧ 12 6 ⎨ 4ε − ⎩
r
r
(2)
r ≥ rc
0,
Ucoul rij = kB TZi Zj
r < rc
B rij
(3)
where ε is the depth of potential well and is the finite distance at which inter-particle potential is zero, rc is the cutoff radius of Lennard-Jones potential. All LJ interactions are truncated at rc = 2.5. In Eq. (3), kB is the Boltzmann constant. Zi , Zj are the charge valences of two particles which are separated by a distance of rij . The Bjerrum length B = e2 /(4ε0 εr kB T) is the distance at which the electrostatic energy between two elementary charges is comparable in magnitude to the thermal energy kB T [36]. Periodic boundary conditions are used in x and y directions. Fixed boundary is used in z direction with a virtual impassible wall. The length of time step is 0.005 for all simulations. The temperature of system is maintained at T* = 1.2Tkb /εLJ for 2000 to accelerate the simulation. Then the system temperature is reduced to T* = 1.0Tkb /εLJ and runs for 10,000 to reach an equilibrium state. Finally, another 5000 runs for data collection. All simulations are conducted by NVT ensemble and Nose-Hoover thermostat. The damp parameter is set to 0.5.
Fig. 7. Radius of gyration Rg of chitosan chain as a function of time. The legend shows the average Rg and standard deviation of Rg during the equilibrium state.
Fig. 6(a)–(f) show the typical adsorption processes of AgNPs on chitosan-grafted ACF. As the AgNPs were dispersed in water suspension without any additional stabilizer, 1–5 particles tend to accumulate to form a cluster to reduce the huge surface energy. These small clusters will bind to the chitosan chains in the subsequent process as can be seen from the submitted dynamic video. This result was consistent with the experimental results shown in Fig. 1. The chitosan chain undergoes from stretched to collapsed state. The electrostatic attracrion leads to adsorption of AgNPs on ACF surface. (The detailed adsorption process of the AgNPs on chitosan-grafted ACF can be seen from the dynamic vedio submitted with the manuscript.) The more detailed image can be seen from the snapshot of the final conformation for AgNPs adsorbing on single chitosan chain as shown in Fig. 6(g). AgNPs adsorbed on chitosan chains due to the electrostatic interactions from the charges carried on polymer chains and silver nanoparticles. Fig. 7 shows the radius of gyration of chitosan chain. Here, several typical conformations are shown to illustrate the conformational transition of chitosan chain. During the initial stage, the chitosan chain would firstly shrink due to the solvent effect and then stretch a little which is mainly caused by the electrostatic interactions between chitosan chain and AgNPs. During the adsorption stage, chitosan chain would repeatedly shrink and stretch itself to adsorb AgNPs until equilibrium state. The single chitosan chain binds silver nanoparticles together like a rope. Fig. 8 presents the number density distribution of polymer monomers p (z), silver nanoparticles np (z), and counterions c (z) as a function of z. Peaks of values of p (z) and np (z) indicate the area of the complexes of chitosan chain and silver nanoparticles. As shown in Fig. 8, the centers of adsorbed silver nanoparticle clusters locate at about z = 13 which is about 1.5 times of silver nanoparticle radius. Moreover, most counterions uniformly distribute in the simulation box. This indicates that the salt ions distrubute uniformly. Fig. 9 shows the radial distribution g(r) between polymer monomers and silver nanoparticles which also indicates the size of the silver nanoparticle cluster. The size of silver cluster is about 10 in radius which is correspondent with the density distribution function of silver nanoparticle np (z). 3.3. Proposed loading mechanism of AgNPs on ACF surface Based on the above characterizations and simulation results, we proposed the possible loading mechanism of the AgNPs on ACF surface. Chitosan was successfully grafted on the ACF surface by the
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
463
Fig. 9. Radial distribution function g(r) between polymer monomers and AgNPs. Fig. 8. Number density distribution of polymer monomers p (z), nanoparticles np (z), and counterions c (z). Notably, the number density of nanoparticles is counted by coarse-grained silver particles not the number of nanoparticles.
esterification reaction occurred between the ACF, citric acid and chitosan. AgNPs were adsorbed on ACF surface by their electrostatic attraction with amino group, which was abundant in chitosan molecule. The loading process of AgNPs on ACF surface is illustrated by the schematic diagram in Fig. 10. 3.4. Antibacterial activity of the AgNPs-loaded ACF 3.4.1. Inhibition zone test Inhibition zone test was carried out to qualitatively investigate the antibacterial property of the AgNPs-loaded ACF material by using standard Kirbye Bauer approach. Both of the typical grampositive and gram-negative bacteria of S. aurous and E .coli and were adopted. The raw ACF and chitosan-grafted ACF were also investi-
Fig. 10. Schematic diagram of the loading process of AgNPs on ACF surface.
gated for comparison. The results are shown in Fig. 11. Clearly, there is no inhibition zone around the raw ACF for neither the E. coli in Fig. 11(a) nor the S. aureus in Fig. 11(d), indicating that there is no antibacterial activity for ACF itself, which is tend to be microbial contaminated when used as filtration material. For the chitosangrafted ACF, there shows a ring around the material where the colour is of less grayness (Fig. 11(b) and (e)). The partially inhibited growth of the bacteria is supposed to be caused by the chitosan
Fig. 11. Inhibition zones of E.coli for bare ACF (a), chitosan-grafted ACF (b), AgNPs-loaded ACF (c). Inhibition zones of S. aureus for bare ACF (d), chitosan-grafted ACF (e), AgNPs-loaded ACF (f). The agar plates were observed after 24 h of incubation at 37 ◦ C. The cell concentration was 107 cfu ml−1 .
464
C. Tang et al. / Applied Surface Science 394 (2017) 457–465 Ackonwledgements
This work is financially supported by National Natural Science Foundation of China under Grant No. 31500801 and the Fundamental Research Funds for the Central Universities (2011JDGZ15). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.apsusc.2016.10. 095. References
Fig. 12. Bactericidal rate of the chitosan-grafted ACF and AgNPs-loaded ACF in circulated water filtration system.
[37]. The chitosan-grafted ACF can achieve inhibition effect for only a portion of the bacteria. While for the AgNPs-loaded ACF, there is no bacteria growth around the material, indicating the enhanced antibacterial activity by loading of AgNPs on the chitosan-grafted ACF (Fig. 11(c) and (f)). The inhibition zones were measured to be 20.5 mm for E. coli and 20.0 mm for S. aureus, respectively.
3.4.2. Antibacterial experiments in circulated system Antibacterial activity of the prepared AgNPs-loaded ACF was further tested in a circulated system as shown in Fig. S1. The experiment of chitosan-grafted ACF was also carried out for comparison. Since the untreated ACF showed no antibacterial activity as demonstrated in 3.4.1, ACF was not included in the circulated system experiment. The results in Fig. 12 show that the AgNPs-loaded ACF can reach 100% of bactericidal efficiency in 3.5 min (cell concentration: 5 × 103 cfu ml−1 , circulating flow rate: 1.2 L min−1 ), while the chitosan-grafted ACF can achieve the bactericidal efficiency of 82.31% at 30 min. The enhanced bactericidal property is caused by the AgNPs. Free Ag+ ions in the treated water was measured to be 0.01 ppm, which was below the United States Environmental Protection Agency (US EPA) and World Health Organization (WHO) drinking water standards [38]. Both of the excellent bactericidal efficiency and the Ag+ ion release rate of the prepared AgNPs-loaded ACF material demonstrated its usability for water treatment.
4. Conclusions Chitosan was used as the binding agent for the effective loading of AgNPs on ACF surface. The amount of AgNPs loaded on ACF surface can reach 0.076 mg g−1 ACF. Molecular dynamics simulation result demonstrated that the AgNPs were adsorbed on ACF surface due to the electrostatic attraction between the AgNPs and chitosan. Chitosan was demonstrated to be a promising binding agent for AgNPs loading on ACF surface from both of the experimental and theoretical calculation perspective. The AgNPs-loaded ACF material showed excellent antibacterial activity to both of the of S. aureus and E. coli. The 100% of bactericidal efficiency in circulated system and the trace amounts of Ag+ ions released certificated the usability of the AgNPs-loaded ACF material in water treatment field, especially for the outdoors circumstances.
[1] I. Sondi, B. Salopek-Sondi, Silver nanoparticles as antimicrobial agent: a case study on E. coli as a model for gram-negative bacteria, J. Colloid Interface Sci. 275 (2004) 177–182. [2] J.R. Morones, J.L. Elechiguerra, A. Camacho, et al., The bactericidal effect of silver nanoparticles, Nanotechnology 16 (2005) 2346–2353. [3] D.T. Schoen, A.P. Schoen, L. Hu, et al., High speed water sterilization using one-dimensional nanostructures, Nano Lett. 10 (2010) 3628–3632. [4] K.N.J. Stevens, O. Crespo-Biel, E.E.M. van den Bosch, et al., The relationship between the antimicrobial effect of catheter coatings containing silver nanoparticles and the coagulation of contacting blood, Biomaterials 30 (2009) 3682–3690. [5] K.N.J. Stevens, S. Croes, R.S. Boersma, et al., Hydrophilic surface coatings with embedded biocidal silver nanoparticles and sodium heparin for central venous catheters, Biomaterials 32 (2011) 1264–1269. [6] H. YeonáLee, H. KunáPark, Y. MiáLee, et al., A practical procedure for producing silver nanocoated fabric and its antibacterial evaluation for biomedical applications, Chem. Commun. 28 (2007) 2959–2961. [7] S. Silver, L.T. Phung, G. Silver, Silver as biocides in burn and wound dressings and bacterial resistance to silver compounds, J. Ind. Microbiol. Biotechnol. 33 (2006) 627–634. [8] Q. Shi, N. Vitchuli, J. Nowak, et al., One-step synthesis of silver nanoparticle-filled nylon 6 nanofibers and their antibacterial properties, J. Mater. Chem. 21 (2011) 10330–10335. [9] J. Cui, C. Hu, Y. Yang, et al., Facile fabrication of carbonaceous nanospheres loaded with silver nanoparticles as antibacterial materials, J. Mater. Chem. 22 (2012) 8121–8126. [10] K. Chamakura, R. Perez-Ballestero, Z. Luo, et al., Comparison of bactericidal activities of silver nanoparticles with common chemical disinfectants, Colloids Surf. B 84 (2011) 88–96. [11] Z. Li, L. Fan, T. Zhang, et al., Facile synthesis of Ag nanoparticles supported on MWCNTs with favorable stability and their bactericidal properties, J. Hazard. Mater. 187 (2011) 466–472. [12] R. Kumar, H. Münstedt, Silver ion release from antimicrobial polyamide/silver composites, Biomaterials 26 (2005) 2081–2088. [13] H. Kong, J. Jang, Antibacterial properties of novel poly (methyl methacrylate) nanofiber containing silver nanoparticles, Langmuir 24 (2008) 2051–2056. [14] P. Mulvaney, T. Linnert, A. Henglein, Surface chemistry of colloidal silver in aqueous solution: observations on chemisorption and reactivity, J. Phys. Chem. 95 (1991) 7843–7846. [15] A. Henglein, Colloidal silver nanoparticles: photochemical preparation and interaction with O2 , CCl4 , and some metal ions, Chem. Mater. 10 (1998) 444–450. [16] S.J. Kazmi, M.A. Shehzad, S. Mehmood, et al., Effect of varied Ag nanoparticles functionalized CNTs on its anti-bacterial activity against E. coli, Sens. Actuators A 216 (2014) 287–294. [17] A. Niu, Y. Han, J. Wu, et al., Synthesis of one-dimensional carbon nanomaterials wrapped by silver nanoparticles and their antibacterial behavior, J. Phys. Chem. C 114 (2010) 12728–12735. [18] C. Tang, W. Sun, W. Yan, Green and facile fabrication of silver nanoparticles loaded activated carbon fibers with long-lasting antibacterial activity, RSC Adv. 4 (2014) 523–530. [19] M. Abe, K. Kawashima, K. Kozawa, et al., Amination of activated carbon and adsorption characteristics of its aminated surface, Langmuir 16 (2000) 5059–5063. [20] P. Chingombe, B. Saha, R.J. Wakeman, Sorption of atrazine on conventional and surface modified activated carbons, J. Colloid Interface Sci. 302 (2006) 408–416. [21] M. Montazer, F. Alimohammadi, A. Shamei, et al., Durable antibacterial and cross-linking cotton with colloidal silver nanoparticles and butane tetracarboxylic acid without yellowing, Colloids Surf. B 89 (2012) 196–202. [22] M. Ranjbar-Mohammadi, M. Arami, H. Bahrami, et al., Grafting of chitosan as a biopolymer onto wool fabric using anhydride bridge and its antibacterial property, Colloids Surf. B 76 (2010) 397–403. [23] G. Zhang, R. Qu, C. Sun, et al., Adsorption for metal ions of chitosan coated cotton fiber, J. Appl. Polym. Sci. 110 (2008) 2321–2327. [24] P. Miretzky, A.F. Cirelli, Hg (II) removal from water by chitosan and chitosan derivatives: a review, J. Hazard. Mater. 167 (2009) 10–23.
C. Tang et al. / Applied Surface Science 394 (2017) 457–465 [25] A. Jimtaisong, N. Saewan, Utilization of carboxymethyl chitosan in cosmetics, Int. J. Cosmet. Sci. 36 (2014) 12–21. [26] C. Tang, W. Sun, J. Lu, et al., Role of the anions in the hydrothermally formed silver nanowires and their antibacterial property, J. Colloid Interface Sci. 416 (2014) 86–94. [27] B. Xin, L. Jing, Z. Ren, et al., Effects of simultaneously doped and deposited Ag on the photocatalytic activity and surface states of TiO2 , J. Phys. Chem. B 109 (2005) 2805–2809. [28] G.M. Balestra, I.J. Misaghi, Increasing the efficiency of the plate counting method for estimating bacterial diversity, J. Microbiol. Methods 30 (1997) 111–117. [29] D. Han, L. Yan, W. Chen, et al., Preparation of chitosan/graphene oxide composite film with enhanced mechanical strength in the wet state, Carbohydr. Ploym. 83 (2011) 653–658. [30] H.V. Tran, L. Dai Tran, T.N. Nguyen, Preparation of chitosan/magnetite composite beads and their application for removal of Pb (II) and Ni (II) from aqueous solution, Mater. Sci. Eng. C 30 (2010) 304–310. [31] F.A. Abd-Elmohdy, Z. El Sayed, S. Essam, et al., Controlling chitosan molecular weight via bio-chitosanolysis, Carbohydr. Ploym. 82 (2010) 539–542.
465
[32] F. Rouquerol, J. Rouquerol, K. Sing, Adsorption by Powders and Porous Solids, Principles, Methodology and Applications, Academic Press, 1999 (p. 205). [33] K. Kremer, G.S. Grest, Dynamics of entangled linear polymer melts: a molecular dynamics simulation, J. Chem. Phys. 92 (1990) 5057–5086. [34] D.S. Wood, V. Koutsos, P.J. Camp, Computer simulations of surface deposition of amphiphilic diblock copolymers driven by solvent evaporation, Soft Matter 9 (2013) 3758–3766. [35] R.W. Hockney, J.W. Eastwood, Computer Simulation Using Particles, CRC Press, Florida, 1988. [36] Q.Q. Cao, M. Bachmann, Polyelectrolyte adsorption on an oppositely charged spherical polyelectrolyte brush, Soft Matter 9 (2013) 5087–5098. [37] Y.-J. Jeon, J.Y. Kamil, F. Shahidi, Chitosan as an edible invisible film for quality preservation of herring and Atlantic cod, J. Agric. Food. Chem. 50 (2002) 5167–5178. [38] B. Zhang, J. Sun, C. Bi, et al., A highly selective ratiometric fluorescent chemosensor for Ag+ based on a rhodanineacetic acid–pyrene derivative, New. J. Chem. 35 (2011) 849–853.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Full length article
Silver nanoparticles-loaded activated carbon fibers using chitosan as binding agent: Preparation, mechanism, and their antibacterial activity Chengli Tang a,∗ , Dongmei Hu b , Qianqian Cao a , Wei Yan c , Bo Xing a a
College of Mechanical and Electrical Engineering, Jiaxing University, Jiaxing 314001, PR China College of Mechanical Science and Engineering, Jilin University, Changchun 130022, PR China c Department of Environmental Science and Engineering, Xi’an Jiaotong University, Xi’an 710049, PR China b
a r t i c l e
i n f o
Article history: Received 26 May 2016 Received in revised form 1 October 2016 Accepted 17 October 2016 Available online 21 October 2016 Keywords: Silver nanoparticles Chitosan Activated carbon fibers Molecular dynamics simulation Binding agent
a b s t r a c t The effective and strong adherence of silver nanoparticles (AgNPs) to the substrate surface is pivotal to the practical application of those AgNPs-modified materials. In this work, AgNPs were synthesized through a green and facile hydrothermal method. Chitosan was introduced as the binding agent for the effective loading of AgNPs on activated carbon fibers (ACF) surface to fabricate the antibacterial material. Apart from conventional instrumental characterizations, i. e., scanning electron microscope (SEM), X-ray diffraction (XRD), Fourier Transform Infrared Spectroscopy (FT-IR), zeta potential and BrunauerEmmett-Teller (BET) surface area measurement, molecular dynamics simulation method was also applied to explore the loading mechanism of AgNPs on the ACF surface. The AgNPs-loaded ACF material showed outstanding antibacterial activity for S. aureus and E. coli. The combination of experimental and theoretical calculation results proved chitosan to be a promising binding agent for the fabrication of AgNPs-loaded ACF material with excellent antibacterial activity. © 2016 Elsevier B.V. All rights reserved.
1. Introduction The rapid development of nanotechnology enables the controllable synthesis of metal silver to nano-scale [1–3]. Silver nanoparticles (AgNPs), as a broad spectrum antimicrobial agent, have been used extensively in therapeutic applications such as catheters [4,5], surgical devieces [6] and wound dressings [7–9]. However, the antibacterial efficiency of AgNPs depends on both colloidal stability and particle size [10]. Additionally, the practical application of AgNPs in colloidal form are often confronted with the problem of aggregation and loss of antibacterial activity [11–13]. Therefore, the stability of AgNPs is an important issue for their practical application as antibacterial material [14,15]. An effective way to prevent AgNPs from aggregation is to deposite AgNPs on specific surface to fabricate AgNPs-loaded materials. Apart from the better stability, these kinds of materials are easy to be recycled and reused compared with the colloidal form of AgNPs. The common supporting materials (substrates) are SiO2 , zeolite and carbon materials [16–18]. Activated carbon fiber is an
∗ Corresponding author. E-mail address: [email protected] (C. Tang). http://dx.doi.org/10.1016/j.apsusc.2016.10.095 0169-4332/© 2016 Elsevier B.V. All rights reserved.
ideal supporting material for AgNPs loading because of the huge specific surface area, proper micropores and excellent adsorption capacity, which have been widely used in water treatment field [19,20]. The strong adherence of AgNPs to the supporting surface is critical for the practical feasibility of the AgNPs-loaded materials. Binding agent is needed to guarantee the effective connection of AgNPs and supporting surface. For instance, butane tetracarboxylic acid was applied as the cross-linking agent to bind AgNPs on the surface of cotton [21]. As a natural polysaccharide, chitosan is rich in amino groups, which results in high percentage of nitrogen (6.89%) [22]. Chitosan has wide applications in medical science, food industry, cosmetic industry, agriculture and environmental field [23–25]. In this paper, we introduced chitosan as the binding agent for the first time to prepare AgNPs-loaded ACF. Chitosan is a natural polysaccharide, which has advantages over synthetic ones when used for water purification. In the research method, we combined the two aspects of experimental result and theoretical calculation together. Specifically, molecular dynamics simulation method was applied except for the conventional instrumental characterizations of scanning electron microscope (SEM), energy dispersive spectrum (EDS), X-ray diffraction (XRD), Fourier Transform Infrared Spectrum (FT-IR), zeta potential and Brunauer-Emmett-Teller (BET)
458
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
surface area. Instrumental characterizations were carried out to observe the morphology, crystalline structure, chemical bonds, surface charge and specific surface area of the material. Besides, molecular dynamics simulation method was also applied to explore the loading mechanism of AgNPs on ACF surface. This combination of experimental result and theoretial calculation may provide a new way for the material science researchers in mechanism studying. The AgNPs-loaded ACF material was also tested for antibacterial activity to evaluate its use potential. 2. Experimental section 2.1. Materials and reagents All the chemical reagents of sodium hydroxide (NaOH), hydrogen chloride (HCl) were purchased from Shanghai Chemical Reagent Co. Ltd (Shanghai, China). Silver nitrate (AgNO3 ) was primary reagent. Maltose was biochemical reagent purchased from Beijing AOBOX biological technology Co. Ltd (Beijing, China). All the chemicals were used as received without further purification. All the ultrapure water used in this study was produced by EPED system (Nanjing, China). ACF cloth was purchased from Nantong Yongtong Environmental Protection Science and Technology Co., Ltd (Nantong, China). 2.2. Synthesis of silver nanoparticles Silver nanoparticles used in this study were prepared by hydrothermal method similarly as described by Tang et. al [26]. Typically, 15 mL of AgNO3 aqueous solution (0.02 M) and 15 mL of maltose aqueous solution (0.02 M) were added to 30 mL of deionized water with magnetic stirring. Then the mixture was adjusted to pH value of about 11 by NaOH (1 M) and transferred into a Teflonlined stainless steel autoclave (100 mL volume) and maintained at 160 ◦ C for 15 min in an electronic oven. After the autoclave cooled down naturally, the precipitate at the bottom of the reactor was sufficiently washed with water and then dried at 60 ◦ C for further usage.
by a pore size and specific surface area analyzer (BET, Builder, SSA4300). The loading amount of AgNPs on ACF surface as well as Ag+ ions concentration released in the disinfection process was checked by the Inductively Coupled Plasma (ICP) measurement in ICPE-9000 system (Shimadzu, Japan). 2.5. Coarse-grained molecular dynamics simulation Coarse-grained molecular dynamics simulation was carried out to explore the interactions between the chitosan, AgNPs and ACF surface, aiming at revealing the loading mechanism of AgNPs on ACF surface. All simulations were conducted in reduced LJ units. The masses of chitosan monomer, silver nanoparticle and counterion were set to 1.0 to reduce the equilibrium time. Furthermore, implicit solvent model was used. A flat surface consists of uniformly distributed wall particles was used to mimic the carbon surface. Single chitosan chain was coarse-grained as a “bead-spring” model. Unit positive charge was carried by each chitosan monomer. AgNPs were represented by face-centered cubic lattice structural particles with a fixed radius of 8. To simulate the negative charge property of AgNPs, unit negative charges were randomly distributed on the nanoparticle surface. The distance between negative charges and nanoparticle center was greater than or equal to 7.92. In this study, the number of monomers in single chitosan chain was set to 100. The number of silver nanoparticles was to be 8. The sizes of simulation box was Lx = Ly = 50, Lz = 100. 2.6. Antibacterial activity of the AgNPs-loaded ACF Antibacterial property of the as-prepared chitosan-grafted ACF and AgNPs-loaded ACF materials were tested by the disk-diffusion method [27]. The water disinfection experiment was carried out in a circulated system (Fig. S1) using the AgNPs-loaded ACF material as the filter. Typical gram-positive and gram-negative bacteria of Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli) were selected as the model bacterial with initial bacterial concentration of 7 × 105 cfu mL−1 . The antibacterial efficiency was determined by comparing the inflow and effluent bacterial concentration via plate counting method [28].
2.3. Preparation of the AgNPs-loaded ACF
3. Results and discussion
To carry out attachment of chitosan onto ACF, pre-weighed quantity of ACF were put in acetic acid solution (0.4%, v/v) containing 0.2% (wt) chitosan, 0.2% (wt) crosslinker citric acid, and 0.15% (wt) sodium dihydrogen phosphate and esterification was carried out at 180 ◦ C for 10 min. Then chitosan-grafted ACF was placed at 60 ◦ C in an oven until the ACF were completely dry. A dry pre-weighed piece of chitosan-grafted ACF was put in AgNPs suspension prepared by dissolving 20 mg of AgNPs in 100 mL of ultrapure water for next 24 h. Then the ACF were washed in deionized water and dried at 60 ◦ C in an oven for further characterization and use.
3.1. Preparation and characterization of AgNPs-loaded ACF
2.4. Characterization of the AgNPs-loaded ACF The morphologies of the samples were observed by a JSM-6390A scanning electron microscope (SEM, JEOL, Japan) with an EDS system. The crystalline structures of all the samples were examined by X-ray diffraction meter (XRD, X’pert PRO with Cu K␣ radiation). FT-IR spectra of the as-prepared materials was collected by using a FT-IR spectrophotometer (Vertex 70, Germany) in the range of 4000–400 cm−1 at a resolution of 2 cm−1 . Zeta potential of the samples (dispersed sample powders in deionized water) were measured by Malvern zetasizer NanoZS 90 (Malvern, England). BET surface area of ACF before and after AgNPs loading were measured
3.1.1. SEM observation The surface morphologies of the differently treated ACF were observed by SEM images. The as-obtained ACF surface is smooth and clean as shown in Fig. 1(a), while numberable particles were found on the smooth ACF surface after the ACF was impregnated in AgNPs suspension (Fig. 1(b)). When the ACF was immersed in chitosan acetic acid solution (0.4%, v/v, containing 0.2% chitosan, 0.2% citric acid, and 0.15% sodium dihydrogen phosphate), followed by impregnation in AgNPs suspension for 24 h, there exists plentiful particles on the ACF surface (Fig. 1(c), (d)). As can be seen, the AgNPs loaded on ACF surface present piles of 1–5 particles. Since no additional dispersant or stabilizer was used, the AgNPs tend to accumulate in the water suspension because of their huge specific surface area and surface energy. We can also see from the SEM image (Fig. 1(c)) that the AgNPs distributed all around the ACF surface, which will be beneficial to the antibacterial process. SEM images of the three samples show apparently that the chitosantreated process increased the particle number on ACF surface. 3.1.2. EDS analysis EDS measurement was carried out to confirm the elemental composition of the particles on the ACF surface, the results are
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
459
Fig. 1. SEM images of bare ACF (a), ACF impregnated in AgNPs suspension (b), and AgNPs-loaded ACF (c), (d).
Fig. 2. EDS results of AgNPs-loaded ACF (the inserted table is the elemental atoms results).
shown in Fig. 2. As can be seen from Fig. 2(b) and Fig. 2(c), there are carbon, oxygen and silver elements peaks in the EDS spectrum taken from spot 1, while there are only carbon and oxygen peaks taken from spot 2. The elemental atoms results in the inserted table also confirm that the particles formed on the ACF surface is composed of silver.
3.1.3. XRD analysis XRD analysis was applied to confirm the crystalline structure of the particles on ACF surface. As shown in Fig. 3, the broad XRD reflection peaks at 2 angle of 25.8◦ and 44◦ could be attributed to the diffraction of the amorphous structure of ACF. The four diagnostic diffraction peaks at 2 angle of 38.2◦ , 44.4◦ , 64.6◦ and
460
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
Fig. 3. XRD pattern of the ACF and AgNPs-loaded ACF.
77.5◦ correspond to the diffraction of (111), (200), (220) and (311) planes of face-centered-cubic silver, which evidently revealed that the particles loaded on the ACF surface are of metallic silver Ag0 . Besides, the peaks belong to amorphous carbon remained unchanged, indicating that the loading process could not affect the carbon crystallography structure. Both of the EDS and XRD results demonstrated that AgNPs have been successfully loaded onto the ACF surface. Besides, the ICP measurement shows that the AgNPs loaded on ACF surface weighed to be 0.076 mg g−1 ACF. 3.1.4. FT-IR spectra FT-IR technology was used to explore the interaction between the composites in AgNPs-loaded ACF material so as to reveal the loading mechanism. The infrared spectra of the as-obtained ACF, chitosan-grafted ACF and AgNPs-loaded ACF are shown in Fig. 4. There is a broad peak in the range from 3707 cm−1 to 3024 cm−1 (peak at 3433 cm−1 ) in the ACF IR spectrum, which was the stretching vibration peak of OH group (OH ). For the chitosan-grafted ACF and AgNPs-loaded ACF, the peak became broader. Because of the N H stretching vibration (NH ) of amino group, which is at ca. 3400 cm−1 , it overlapped with OH stretching vibration, resulting in a broader peak. For the chitosan-grafted ACF, the peak is in the range from 3782 cm−1 to 2997 cm−1 (peak at 3439 cm−1 ), and AgNPs-loaded ACF from 3703 cm−1 to 3006 cm−1 (peak at 3429 cm−1 ) [29]. The new peaks at wavenumber of 2931 cm−1 and 2916 cm−1 , which is nearly invisible in ACF IR spectrum, was assigned to the stretching vibration (CH ) of chitosan backbone [30]. Compared with ACF, the peak intensity at ca. 1640 cm−1
Fig. 4. FT-IR spectra of the ACF, chitosan-grafted ACF and AgNPs-loaded ACF.
got enhanced for chitosan-grafted ACF because of the primary amine in chitosan [30]. But there showed an apparent weakness for the AgNPs-loaded ACF. The obvious lower peak intensity after AgNPs loading is supposed to caused by the conbination of -NH2 with AgNPs in the obtained AgNPs-loaded ACF composites. The peak at 1519 cm−1 , 1531 cm−1 and 1521 cm−1 for ACF, chitosan-grafted ACF and AgNPs-loaded ACF underwent the similar experience. There are also new peaks at wavenumber of 1080 cm−1 and 1033 cm−1 for chitosan-grafted ACF and AgNPs-loaded ACF, which are the absorption peaks of the deformation vibration of N H (␦NH ) in amide I [30]. The band located 1145 cm−1 and 1151 cm−1 for chitosan-grafted ACF and AgNPs-loaded ACF are related to asymmetric vibrations of C O in oxygen bridge, resulting from deacetylation of chitosan [31]. The whole IR results of the three materials demonstrated that the chitosan was successfully coated on the ACF surface and it was favorable for the AgNPs loading because of the abundant NH2 groups in chitosan molecule. 3.1.5. Zeta potential of different materials Zeta potential of different suspensions were carried out to determine the surface charge of the corresponding sample. As shown in Table 1, the zeta potential of the as-obtained ACF and hydrothermally formed AgNPs (in aqueous solution) valued to be −20.3 mV, indicating that both of the ACF and AgNPs surface are negatively charged. Since both of the ACF surface and AgNPs are negatively charged, it is difficult for them to attract each other without assis-
Fig. 5. N2 adsorption-desorption isotherm and BJH pore size distribution (inserted) of the ACFs (a) and AgNPs-loaded ACFs (b).
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
461
Fig. 6. Typical adsorption process of AgNPs on chitosan-grafted ACF surface (a)–(f), and the final conformation of silver nanoparticle adsorption (g). Purple beads: coarsegrained non-charged silver nanoparticles, blue beads: negatively charged silver particles, green beads: chitosan monomers, silver beads: carbon surface. To show a clear snapshot, carbon surface and counterions are not included in (g). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Table 1 Zeta potential of different materials.
Zeta potential/mV
ACF
Chitosan
AgNPs
Chitosangrafted ACF
AgNPs-loaded ACF
−20.3
98.6
−20.3
33.2
−2.89
tance. The chitosan exhibited positively charged surface with a zeta potential of 98.6 mV. The positive surface charge could be imparted by protonated amino groups (-NH3 + ) of chitosan molecule. The total zeta potential of the chitosan-grafted ACF valued to be 33.2 mV, indicating that the ACF surface was completely coated with a layer of chitosan. When the chitosan-grafted ACF was loaded with AgNPs, the composite was negatively charged again with the zeta potential of −2.89 mV. The negative zeta potential of the AgNPs-loaded ACF
462
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
supported the above assumption that the chitosan served as the binding agent for the successful loading of AgNPs on ACF surface. 3.1.6. BET surface area measurement BET surface area of the raw ACF and AgNPs-loaded ACF was calculated according to BET method. The N2 adsorption-desorption isotherms of ACF and AgNPs-loaded ACF and the corresponding pore size distribution are shown in Fig. 5. Both of the hysteresis loops before (Fig. 5(a)) and after (Fig. 5(b)) AgNPs-loading resemble type H4 in IUPAC lassification [32], indicating that the slit pores and laminated structure of the ACF remained unchanged after AgNPs loading. The BET surface area of ACF was calculated to be 1020 m2 g−1 . There was a decrease of 19.3% in the specific surface area, but the AgNPs-loaded ACF remained the excellent adsorption capacity of ACF with a specific area of 822.480 m2 g−1 . The pore size distribution results show that the pores number at radius of 80–90 nm decreased deastically, while no obvious change was obversed with other pore sizes. The number decreased was probably caused by the chitosan coated on the ACF surface. Although admitting that there was an apparent decrease in pores number at radius of 80–90 nm, the AgNPs-loaded ACF remained excellent adsorption capacity of ACF, which is extremely useful for bacteria adsorption and inactivation. 3.2. Coarse-grained molecular dynamics simulation Coarse-Grained molecular dynamics simulation was carried out to study the mechanism of AgNPs loading on single chitosan chain in dilute salt solution at molecular scale. To model the soft chitosan chain, finite extensible nonlinear elastic (FENE) “spring” was used to simulate the chemical bonds between adjacent chitosan monomers [33].
UFENE = −0.5KR02 ln 1 −
r 2
(1)
R0
where, r is the distance between two bonded beads, R0 = 1.5 is the maximum bond length, and K is a spring constant which is chosen to be 30.0 εLJ /( LJ 2 ) [34]. εLJ and LJ are reduced LJ units. Shortand long-range interactions between non-bonding coarse grained beads are represented by Lennard-Jones potential and Coulombic potential. Long-range par of Coulombic interaction is calculated by Particle-particle particle-mesh solver [35].
ULJ =
⎧ 12 6 ⎨ 4ε − ⎩
r
r
(2)
r ≥ rc
0,
Ucoul rij = kB TZi Zj
r < rc
B rij
(3)
where ε is the depth of potential well and is the finite distance at which inter-particle potential is zero, rc is the cutoff radius of Lennard-Jones potential. All LJ interactions are truncated at rc = 2.5. In Eq. (3), kB is the Boltzmann constant. Zi , Zj are the charge valences of two particles which are separated by a distance of rij . The Bjerrum length B = e2 /(4ε0 εr kB T) is the distance at which the electrostatic energy between two elementary charges is comparable in magnitude to the thermal energy kB T [36]. Periodic boundary conditions are used in x and y directions. Fixed boundary is used in z direction with a virtual impassible wall. The length of time step is 0.005 for all simulations. The temperature of system is maintained at T* = 1.2Tkb /εLJ for 2000 to accelerate the simulation. Then the system temperature is reduced to T* = 1.0Tkb /εLJ and runs for 10,000 to reach an equilibrium state. Finally, another 5000 runs for data collection. All simulations are conducted by NVT ensemble and Nose-Hoover thermostat. The damp parameter is set to 0.5.
Fig. 7. Radius of gyration Rg of chitosan chain as a function of time. The legend shows the average Rg and standard deviation of Rg during the equilibrium state.
Fig. 6(a)–(f) show the typical adsorption processes of AgNPs on chitosan-grafted ACF. As the AgNPs were dispersed in water suspension without any additional stabilizer, 1–5 particles tend to accumulate to form a cluster to reduce the huge surface energy. These small clusters will bind to the chitosan chains in the subsequent process as can be seen from the submitted dynamic video. This result was consistent with the experimental results shown in Fig. 1. The chitosan chain undergoes from stretched to collapsed state. The electrostatic attracrion leads to adsorption of AgNPs on ACF surface. (The detailed adsorption process of the AgNPs on chitosan-grafted ACF can be seen from the dynamic vedio submitted with the manuscript.) The more detailed image can be seen from the snapshot of the final conformation for AgNPs adsorbing on single chitosan chain as shown in Fig. 6(g). AgNPs adsorbed on chitosan chains due to the electrostatic interactions from the charges carried on polymer chains and silver nanoparticles. Fig. 7 shows the radius of gyration of chitosan chain. Here, several typical conformations are shown to illustrate the conformational transition of chitosan chain. During the initial stage, the chitosan chain would firstly shrink due to the solvent effect and then stretch a little which is mainly caused by the electrostatic interactions between chitosan chain and AgNPs. During the adsorption stage, chitosan chain would repeatedly shrink and stretch itself to adsorb AgNPs until equilibrium state. The single chitosan chain binds silver nanoparticles together like a rope. Fig. 8 presents the number density distribution of polymer monomers p (z), silver nanoparticles np (z), and counterions c (z) as a function of z. Peaks of values of p (z) and np (z) indicate the area of the complexes of chitosan chain and silver nanoparticles. As shown in Fig. 8, the centers of adsorbed silver nanoparticle clusters locate at about z = 13 which is about 1.5 times of silver nanoparticle radius. Moreover, most counterions uniformly distribute in the simulation box. This indicates that the salt ions distrubute uniformly. Fig. 9 shows the radial distribution g(r) between polymer monomers and silver nanoparticles which also indicates the size of the silver nanoparticle cluster. The size of silver cluster is about 10 in radius which is correspondent with the density distribution function of silver nanoparticle np (z). 3.3. Proposed loading mechanism of AgNPs on ACF surface Based on the above characterizations and simulation results, we proposed the possible loading mechanism of the AgNPs on ACF surface. Chitosan was successfully grafted on the ACF surface by the
C. Tang et al. / Applied Surface Science 394 (2017) 457–465
463
Fig. 9. Radial distribution function g(r) between polymer monomers and AgNPs. Fig. 8. Number density distribution of polymer monomers p (z), nanoparticles np (z), and counterions c (z). Notably, the number density of nanoparticles is counted by coarse-grained silver particles not the number of nanoparticles.
esterification reaction occurred between the ACF, citric acid and chitosan. AgNPs were adsorbed on ACF surface by their electrostatic attraction with amino group, which was abundant in chitosan molecule. The loading process of AgNPs on ACF surface is illustrated by the schematic diagram in Fig. 10. 3.4. Antibacterial activity of the AgNPs-loaded ACF 3.4.1. Inhibition zone test Inhibition zone test was carried out to qualitatively investigate the antibacterial property of the AgNPs-loaded ACF material by using standard Kirbye Bauer approach. Both of the typical grampositive and gram-negative bacteria of S. aurous and E .coli and were adopted. The raw ACF and chitosan-grafted ACF were also investi-
Fig. 10. Schematic diagram of the loading process of AgNPs on ACF surface.
gated for comparison. The results are shown in Fig. 11. Clearly, there is no inhibition zone around the raw ACF for neither the E. coli in Fig. 11(a) nor the S. aureus in Fig. 11(d), indicating that there is no antibacterial activity for ACF itself, which is tend to be microbial contaminated when used as filtration material. For the chitosangrafted ACF, there shows a ring around the material where the colour is of less grayness (Fig. 11(b) and (e)). The partially inhibited growth of the bacteria is supposed to be caused by the chitosan
Fig. 11. Inhibition zones of E.coli for bare ACF (a), chitosan-grafted ACF (b), AgNPs-loaded ACF (c). Inhibition zones of S. aureus for bare ACF (d), chitosan-grafted ACF (e), AgNPs-loaded ACF (f). The agar plates were observed after 24 h of incubation at 37 ◦ C. The cell concentration was 107 cfu ml−1 .
464
C. Tang et al. / Applied Surface Science 394 (2017) 457–465 Ackonwledgements
This work is financially supported by National Natural Science Foundation of China under Grant No. 31500801 and the Fundamental Research Funds for the Central Universities (2011JDGZ15). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.apsusc.2016.10. 095. References
Fig. 12. Bactericidal rate of the chitosan-grafted ACF and AgNPs-loaded ACF in circulated water filtration system.
[37]. The chitosan-grafted ACF can achieve inhibition effect for only a portion of the bacteria. While for the AgNPs-loaded ACF, there is no bacteria growth around the material, indicating the enhanced antibacterial activity by loading of AgNPs on the chitosan-grafted ACF (Fig. 11(c) and (f)). The inhibition zones were measured to be 20.5 mm for E. coli and 20.0 mm for S. aureus, respectively.
3.4.2. Antibacterial experiments in circulated system Antibacterial activity of the prepared AgNPs-loaded ACF was further tested in a circulated system as shown in Fig. S1. The experiment of chitosan-grafted ACF was also carried out for comparison. Since the untreated ACF showed no antibacterial activity as demonstrated in 3.4.1, ACF was not included in the circulated system experiment. The results in Fig. 12 show that the AgNPs-loaded ACF can reach 100% of bactericidal efficiency in 3.5 min (cell concentration: 5 × 103 cfu ml−1 , circulating flow rate: 1.2 L min−1 ), while the chitosan-grafted ACF can achieve the bactericidal efficiency of 82.31% at 30 min. The enhanced bactericidal property is caused by the AgNPs. Free Ag+ ions in the treated water was measured to be 0.01 ppm, which was below the United States Environmental Protection Agency (US EPA) and World Health Organization (WHO) drinking water standards [38]. Both of the excellent bactericidal efficiency and the Ag+ ion release rate of the prepared AgNPs-loaded ACF material demonstrated its usability for water treatment.
4. Conclusions Chitosan was used as the binding agent for the effective loading of AgNPs on ACF surface. The amount of AgNPs loaded on ACF surface can reach 0.076 mg g−1 ACF. Molecular dynamics simulation result demonstrated that the AgNPs were adsorbed on ACF surface due to the electrostatic attraction between the AgNPs and chitosan. Chitosan was demonstrated to be a promising binding agent for AgNPs loading on ACF surface from both of the experimental and theoretical calculation perspective. The AgNPs-loaded ACF material showed excellent antibacterial activity to both of the of S. aureus and E. coli. The 100% of bactericidal efficiency in circulated system and the trace amounts of Ag+ ions released certificated the usability of the AgNPs-loaded ACF material in water treatment field, especially for the outdoors circumstances.
[1] I. Sondi, B. Salopek-Sondi, Silver nanoparticles as antimicrobial agent: a case study on E. coli as a model for gram-negative bacteria, J. Colloid Interface Sci. 275 (2004) 177–182. [2] J.R. Morones, J.L. Elechiguerra, A. Camacho, et al., The bactericidal effect of silver nanoparticles, Nanotechnology 16 (2005) 2346–2353. [3] D.T. Schoen, A.P. Schoen, L. Hu, et al., High speed water sterilization using one-dimensional nanostructures, Nano Lett. 10 (2010) 3628–3632. [4] K.N.J. Stevens, O. Crespo-Biel, E.E.M. van den Bosch, et al., The relationship between the antimicrobial effect of catheter coatings containing silver nanoparticles and the coagulation of contacting blood, Biomaterials 30 (2009) 3682–3690. [5] K.N.J. Stevens, S. Croes, R.S. Boersma, et al., Hydrophilic surface coatings with embedded biocidal silver nanoparticles and sodium heparin for central venous catheters, Biomaterials 32 (2011) 1264–1269. [6] H. YeonáLee, H. KunáPark, Y. MiáLee, et al., A practical procedure for producing silver nanocoated fabric and its antibacterial evaluation for biomedical applications, Chem. Commun. 28 (2007) 2959–2961. [7] S. Silver, L.T. Phung, G. Silver, Silver as biocides in burn and wound dressings and bacterial resistance to silver compounds, J. Ind. Microbiol. Biotechnol. 33 (2006) 627–634. [8] Q. Shi, N. Vitchuli, J. Nowak, et al., One-step synthesis of silver nanoparticle-filled nylon 6 nanofibers and their antibacterial properties, J. Mater. Chem. 21 (2011) 10330–10335. [9] J. Cui, C. Hu, Y. Yang, et al., Facile fabrication of carbonaceous nanospheres loaded with silver nanoparticles as antibacterial materials, J. Mater. Chem. 22 (2012) 8121–8126. [10] K. Chamakura, R. Perez-Ballestero, Z. Luo, et al., Comparison of bactericidal activities of silver nanoparticles with common chemical disinfectants, Colloids Surf. B 84 (2011) 88–96. [11] Z. Li, L. Fan, T. Zhang, et al., Facile synthesis of Ag nanoparticles supported on MWCNTs with favorable stability and their bactericidal properties, J. Hazard. Mater. 187 (2011) 466–472. [12] R. Kumar, H. Münstedt, Silver ion release from antimicrobial polyamide/silver composites, Biomaterials 26 (2005) 2081–2088. [13] H. Kong, J. Jang, Antibacterial properties of novel poly (methyl methacrylate) nanofiber containing silver nanoparticles, Langmuir 24 (2008) 2051–2056. [14] P. Mulvaney, T. Linnert, A. Henglein, Surface chemistry of colloidal silver in aqueous solution: observations on chemisorption and reactivity, J. Phys. Chem. 95 (1991) 7843–7846. [15] A. Henglein, Colloidal silver nanoparticles: photochemical preparation and interaction with O2 , CCl4 , and some metal ions, Chem. Mater. 10 (1998) 444–450. [16] S.J. Kazmi, M.A. Shehzad, S. Mehmood, et al., Effect of varied Ag nanoparticles functionalized CNTs on its anti-bacterial activity against E. coli, Sens. Actuators A 216 (2014) 287–294. [17] A. Niu, Y. Han, J. Wu, et al., Synthesis of one-dimensional carbon nanomaterials wrapped by silver nanoparticles and their antibacterial behavior, J. Phys. Chem. C 114 (2010) 12728–12735. [18] C. Tang, W. Sun, W. Yan, Green and facile fabrication of silver nanoparticles loaded activated carbon fibers with long-lasting antibacterial activity, RSC Adv. 4 (2014) 523–530. [19] M. Abe, K. Kawashima, K. Kozawa, et al., Amination of activated carbon and adsorption characteristics of its aminated surface, Langmuir 16 (2000) 5059–5063. [20] P. Chingombe, B. Saha, R.J. Wakeman, Sorption of atrazine on conventional and surface modified activated carbons, J. Colloid Interface Sci. 302 (2006) 408–416. [21] M. Montazer, F. Alimohammadi, A. Shamei, et al., Durable antibacterial and cross-linking cotton with colloidal silver nanoparticles and butane tetracarboxylic acid without yellowing, Colloids Surf. B 89 (2012) 196–202. [22] M. Ranjbar-Mohammadi, M. Arami, H. Bahrami, et al., Grafting of chitosan as a biopolymer onto wool fabric using anhydride bridge and its antibacterial property, Colloids Surf. B 76 (2010) 397–403. [23] G. Zhang, R. Qu, C. Sun, et al., Adsorption for metal ions of chitosan coated cotton fiber, J. Appl. Polym. Sci. 110 (2008) 2321–2327. [24] P. Miretzky, A.F. Cirelli, Hg (II) removal from water by chitosan and chitosan derivatives: a review, J. Hazard. Mater. 167 (2009) 10–23.
C. Tang et al. / Applied Surface Science 394 (2017) 457–465 [25] A. Jimtaisong, N. Saewan, Utilization of carboxymethyl chitosan in cosmetics, Int. J. Cosmet. Sci. 36 (2014) 12–21. [26] C. Tang, W. Sun, J. Lu, et al., Role of the anions in the hydrothermally formed silver nanowires and their antibacterial property, J. Colloid Interface Sci. 416 (2014) 86–94. [27] B. Xin, L. Jing, Z. Ren, et al., Effects of simultaneously doped and deposited Ag on the photocatalytic activity and surface states of TiO2 , J. Phys. Chem. B 109 (2005) 2805–2809. [28] G.M. Balestra, I.J. Misaghi, Increasing the efficiency of the plate counting method for estimating bacterial diversity, J. Microbiol. Methods 30 (1997) 111–117. [29] D. Han, L. Yan, W. Chen, et al., Preparation of chitosan/graphene oxide composite film with enhanced mechanical strength in the wet state, Carbohydr. Ploym. 83 (2011) 653–658. [30] H.V. Tran, L. Dai Tran, T.N. Nguyen, Preparation of chitosan/magnetite composite beads and their application for removal of Pb (II) and Ni (II) from aqueous solution, Mater. Sci. Eng. C 30 (2010) 304–310. [31] F.A. Abd-Elmohdy, Z. El Sayed, S. Essam, et al., Controlling chitosan molecular weight via bio-chitosanolysis, Carbohydr. Ploym. 82 (2010) 539–542.
465
[32] F. Rouquerol, J. Rouquerol, K. Sing, Adsorption by Powders and Porous Solids, Principles, Methodology and Applications, Academic Press, 1999 (p. 205). [33] K. Kremer, G.S. Grest, Dynamics of entangled linear polymer melts: a molecular dynamics simulation, J. Chem. Phys. 92 (1990) 5057–5086. [34] D.S. Wood, V. Koutsos, P.J. Camp, Computer simulations of surface deposition of amphiphilic diblock copolymers driven by solvent evaporation, Soft Matter 9 (2013) 3758–3766. [35] R.W. Hockney, J.W. Eastwood, Computer Simulation Using Particles, CRC Press, Florida, 1988. [36] Q.Q. Cao, M. Bachmann, Polyelectrolyte adsorption on an oppositely charged spherical polyelectrolyte brush, Soft Matter 9 (2013) 5087–5098. [37] Y.-J. Jeon, J.Y. Kamil, F. Shahidi, Chitosan as an edible invisible film for quality preservation of herring and Atlantic cod, J. Agric. Food. Chem. 50 (2002) 5167–5178. [38] B. Zhang, J. Sun, C. Bi, et al., A highly selective ratiometric fluorescent chemosensor for Ag+ based on a rhodanineacetic acid–pyrene derivative, New. J. Chem. 35 (2011) 849–853.