BACTERICIDAL ADSORBENTS OBTAINED BY ION EXCHANGE MODIFICATION OF NATURAL PHILLIPSITE

Zeolite adsorbents and ion exchangers reducing the concentrations of contaminants in aqueous medium, containing bioactive metals and endowed with bactericidal properties are promising for application in environmental protection practice and medicine. Phillipsite has a high ion exchange capacity and can be used to produce such materials. Silver-, copper-, and zinc-containing micro-mesoporous zeolite materials have been prepared on the basis of natural phillipsite from the Shukhuti field, Western Georgia (Saqartvelo), using ionexchange reactions between grinded and washed by dilute hydrogen chloride solution zeolite and a salt of a corresponding transition metal in the solid phase followed by washing with distilled water. Synthesized in such way adsorbent-ion-exchangers are characterized by chemical analysis and sorption data (nitrogen adsorption-desorption isotherms at 77 K and water vapour sorption at room temperature), powder X-ray diffraction patterns, Fourier transform infra-red spectra, and scanning electron microscope images. Obtained materials keep the crystal structure and general sorption and ion-exchange properties of phillipsite, they contain up to 230 mg/g of silver, 66 mg/g of copper, and 86 mg/g of zinc, which is several times higher than the content of bioactive metals in the cationexchange forms of clinoptilolite and synthetic zeolites obtained by ion exchange in the liquid phase described in the literature. Prepared silver-, copper-, and zinc-containing phillipsites show bactericidal and bacteriostatic activity towards Escherichia coli regardless of whether the number of released ions of the bioactive metal reaches the minimum inhibitory concentration in solution. The procedure of dry ion-exchange synthesis leads to an increase in the dispersion of the material, but does not affect the developed mesoporous system of phillipsite and the total pore volume averaging 0.285 cm3/g. The compliance of proposed method for preparation of silver-, copper-, and zinccontaining forms of phillipsite with high environmental standards is confirmed by its low Sheldon’s factor E in comparison with the similar green chemistry metrics of conventional methods of the ion exchange in solutions.


INTRODUCTION
The environmental use of zeolites, alunimosilicates with the general formula Me n Si x Al n O 2(n+x) ·mH 2 O (Me = Na, K, … ½Ca, ½Mg, …), is based on the complex of their properties, especially on the capability of zeolites to enter into ion exchange reactions with the participation of Me +n ions compensating the negative charge of the crystal lattice constructed from alternating SiO 4 and AlO 4 tetrahedra. Many different studies have demonstrated effectiveness of zeolites in reducing the concentrations of contaminants (heavy metals, anions and organic matter) in water [1][2][3][4][5][6].
The complexity of aquatic systems demands special attention in the selection and preparation of materials for water purification. The chemical behaviour of natural zeolites in different aqueous environments depends on the structural characteristics and chemical composition [5,6]. The possibilities of using natural and synthetic zeolites to extract ammonium ions and heavy metals from water were well studied in the last century [7][8][9][10], but research into specific natural zeolites continues [11][12][13][14] including studies of adsorbents for the removal of organic pollutants [15]. Nowadays, modified natural zeolites are increasingly used also for biological treatment of water, precisely for surface binding of biological agents from water.
Besides, started at the beginning of the 21 st century and continuing to this day, studies have shown that natural and synthetic zeolites exchanged by ions of silver, copper, zinc, or some other transition metals exhibit antimicrobial activity toward a broad range of microorganisms [16][17][18][19][20][21][22][23][24][25][26][27][28], and silver-containing zeolites are characterized by the most powerful antibacterial activity [18,23,25]. In general, silver is considered as antibacterial agent with well-known mode of action, bacterial resistance against silver is well described [29], similarities and differences between silver ions and silver in nanoforms as antibacterial agents were discussed recently [30].
Application of such cation-exchanged and bioactive metal-containing zeolites (MZs) is helpful in the remediation of hazardous heavy metal-polluted soils [31] or in the purification of industrial wastewater [32], in such cases it is necessary to provide the sorption material with bactericidal properties in order to prevent the growth of microorganisms on its surface.
It was decided that zeolites not containing silver, copper, zinc or other transition metals (such as mercury, cadmium, chromium, and lead) are not active toward microorganisms [21,33]. It is believed that the porous zeolite structure enables metal cations to move freely from the lattice to the environment, and this seems to be responsible for their activity toward microorganisms [34], but it has been recently established that in some cases the antibacterial activity could be attributed to the MZ itself [27,28].
The aim of our study was to develop a fast, eco-friendly method for producing silver-, copper-, and zinc-containing MZs from natural phillipsite, which has a higher ion-exchange capacity ( [35] Application of phillipsite is not so wide comparing to clinoptilolite, but it is used for ammonia adsorption and removal of various heavy metals from waste waters [36], phillipsite is effective in preparation of surfactant modified zeolite [37] and in removal of picolines from aqueous solution in the broad range of concentrations [38], the calcium-enriched phillipsites are found to exhibit the capability to adsorb humic acids [39]. Possibility of preparation of phase-pure zeolite NaX with high specific surface area (up to 590 m 2 /g) and pore volume (up to 0.58 cm 3 /g) by hydrothermal recrystallization of acid-treated natural phillipsite was shown recently [40].

Materials.
Preparation of MZs by "ion exchange synthesis" was carried out using Georgian natural phillipsite-containing tuff rock from Shukhuti (Western Georgia) described in [28].
The conventional mechanical grinding of tuff to obtain < 63 µm (240 mesh) fraction leads to the formation of a multitude of micrometric crystallites (Fig. 2 a). It is easy to obtain a highly dispersed fraction, since large crystallites (with dimensions of about 50 µm) consist of smaller (about 5 µm) bound together by clay minerals (Fig. 2 b). Crushed and sieved rock was washed by diluted HCl solution (0.025 N) to remove clay impurities, and named as NPSH (natural phillipsite from Shukhuti). The high dispersion of the phillipsite sample NPSH used in solid-state ion-exchange synthesis is confirmed by the data on low-temperature sorption-desorption of nitrogen: the BET surface area is 73.5 m 2 /g, total volume of pores less than 121 nm in diameter is 0.28 cm 3 /g.
The affiliation of the zeolite phase crystal structure to the PHI type for NPSH was confirmed by comparing the experimental powder X-ray diffraction pattern ( Fig. 3) with the calculated one from the Database of Zeolite Structures of the International Zeolite Association (http://www.izastructure.org).  Analytical grade silver nitrate AgNO 3 , copper chloride CuCl 2 , and zinc chloride ZnCl 2 were purchased from Merck KGaA (Darmstadt, Germany) and used without any further purification.
Preparation of MZs. Ion exchange was carried out as follows: powder of zeolite NPSH and the corresponding salt were mixed in different weight ratios (from 1:1 to 1:6) and thoroughly grinded in an agate mortar for 5-10 min, dependent on the cationic form and weight ratio. The solid mixture was then transferred to a filter and washed with distilled water until the absence of nitrate or chloride anions, after which the modified samples were first dried in air and then at 100-105 °С in a thermostat; samples with a maximum content of corresponding are labeled as AgPSH (silver-containing phillipsite), CuPSH (coppercontaining phillipsite), and ZnPSH (zinccontaining phillipsite), the conditions of their preparation are given in Table 1. Characterization of samples. Chemical composition of raw material and prepared samples was determined by elemental analyses carried out using an atomic absorption spectrometer (model 300, Perkin-Elmer, UK) and energy dispersive X-ray (EDS) analysis. The crystalline phase was identified by powder X-ray diffraction (XRD) patterns obtained from a modernized Dron-4 X-ray diffractometer (Russia) employing the CuK α line (λ = 0.154056 nm). The samples were scanned in the 2Θ range of 5 to 50 o with a 0.02 o step at a scanning speed of 1 o /min. Fourier transform infrared spectra were collected by a 10.4.2 FTIR spectrometer (Perkin-Elmer, UK) over the range of 400-4000 cm −1 with a resolution of 2 cm −1 using the KBr pellet technique for sample preparation. The surface morphology of the samples was observed with a scanning electron microscope JSM6510LV (Jeol, Japan) equipped with an X-Max 20 analyzer (Oxford Instruments, UK) for EDS. Nitrogen adsorption-desorption isotherms were measured at 77 K using an ASAP 2020 Plus physisorption analyzer (Micromeritics, Norcross, GA, USA), after evacuation of the samples at 350 °C during 2 h; water adsorption capacity was measured under static conditions.
Metal release and antibacterial activity. The determination of the amount of metals released from MZs in normal salina solution (9 g of NaCl in 1 L of deionized water) was carried out under static conditions in a thermostatic bath (Grant Instruments OLS26 Aqua Pro) at the temperature of 37±0.1 °C, without stirring or shaking. Sampling for analysis was carried out after 1, 3, 6 and 24 h after loading 0.1 g of zeolite in 100 ml of salina.
The antibacterial activity of NPSH and MZs was tested against Gram-negative bacteria Escherichia coli. Before testing the antibacterial activity, all dry zeolite products were sterilized at 70 °C for 2 h in a dry sterilizer. No microbial contamination of the prepared samples was found.
Luria Bertani (LB) medium sterilized by autoclaving (121 °C, 15 min) prior to the antibacterial activity tests was used as a growing medium, bacteria were grown aerobically in LB broth at 37 °C for 12 h, the culture was centrifuged twice (10.000 rpm), and the cells were washed and suspended in distilled water. 1 cm 3 of the prepared biomass suspension of approximately 10 7 colony-forming units (CFU) per cm 3 was inoculated into the Schott's bottles with 100 cm 3 of autoclaved saline, and zeolite samples in a concentration of 0.1 g/100 cm 3 were added. The bottles were incubated in a thermostatic water bath with shaking at 105 rpm for 24 h at 37±0.1 °C. The number of viable cells was determined taking 0.1 mL of water + bacteria + zeolite mixture at the beginning of the experiment, after 1 h (the lag phase of bacterial growth), and after 3, 6, and 24 h (the stationary phase). The aliquots were diluted in distilled water, spread on LB agar plates and incubated at 37 °C for 24 h. Bacterial colonies were counted using microscope.
Bacteriostatic properties of natural and modified zeolite samples were determined by the disk diffusion (Kirby-Bauer) method under standard conditions using the culture of E.coli grown on Mueller-Hinton agar medium at 37 °C for overnight and placed (10 9 CFU/ cm 3 ) on Mueller-Hinton agar (3 mm deep) poured into 100 mm Petri dishes. 0.2 g of zeolite in the form of pellets with 8 mm in diameter was placed into the plates. The plates were incubated at 37 °C over 5 % CO 2 medium and, finally, the width of inhibition zone of each sample in the plates was measured at the end of the first day.
All experiments on antibacterial activity of NPSH and MZs were done in triplicate. The values obtained were averaged to give the final data with standard deviations.

Chemical composition.
Chemical composition of natural phillipsite and its modified forms with a maximum silver, copper or zinc content are listed in the Table 2  According to the elemental analysis data, when silver, copper, and zinc ions are introduced into the phillipsite crystal lattice, monovalent potassium and sodium ions are mainly displaced. Degree of replacement is quite high, the obtained modified forms contain a large amount of transition metals -up to 230 mg/g (2.14 mmol per 1 g of zeolite) of silver in the AgPSH sample, up to 66 mg/g (~1 mmol per 1 g of zeolite) of copper in the CuPSH sample, and up to 86 mg/g (~1.3 mmol per 1 g of zeolite) of zinc in the ZnPSH sample. In Table 3, obtained results are compared with the maximum possible content of silver, copper, and zinc calculated from ion-exchange isotherms measured on natural clinoptilolite from Gördes, Turkey [18], and with the literature data for different clinoptilolites and synthetic zeolites. 1.02 (6) 1.01 (6) 1.03 (7) 1.02(6) Silver ions Ag + quite easily enter the microporous structure of phillipsite, the introduction of copper Cu 2+ and zinc Zn 2+ ions requires an increased amount of salt and a longer contact with the surface of the zeolite. This can probably be explained by a slight difference in the hydration character of the ions entering the pores of the zeolite. So, an "isolated" silver ion Ag + (radius 0.115 nm) is larger than Cu 2+ and Zn 2+ ions (radii 0.073 and 0.074 nm, respectively), but the hydrated silver(I) ion contains four water molecules (Ag(H 2 O) 4 + ) in a linearly distorted tetrahedron configuration, whereas the hydrated copper(II) and zinc(II) ions contain six water molecules (M(H 2 O) 6 2+ ) and have regular octahedral configuration [42].
Crystal structure. Ion exchange reactions do not change the crystal structure of the zeolite, this is confirmed by the powder X-ray diffraction patterns of the modified samples, characteristic peaks remain in XRD patterns, only their intensities change, as shown in the Table 4.
No notable changes were observed in the IR spectra of the modified phillipsites as compared with the vibration bands of raw zeolitic mineral (Table 5), only the intensity of the broad band at 3200-3700 cm -1 corresponding to the asymmetric stretching of OH group is increased due to the larger number of water molecules in the samples containing silver, copper, and zinc. 1.75 * -maximum of broad peak; ** -shoulder at broad peak The ratio of the absorbance of asymmetric stretching vibration of the external tetrahedra with frequency ν asym to the absorbance of internal bending vibration with frequency δ was used for the evaluation of the IR spectra data for natural and modified Mexican zeolite [17]; for a mixture of clinoptilolite-heulandite and corresponding MZs, this ratio varies from 1.34 to 1.64, but for phillipsite NPSH and its modifications, this ratio varies only slightly.
A narrow absorption band at 1385 cm -1 typical for NO stretching vibrations in nitro compounds was observed in IR spectra of insufficiently washed silver-enriched phillipsite, this effect can be used to monitor the purity of silver-containing samples.
Both XRD and FTIR data show developed zeolitic crystal microporous structure in metal-containing samples, and this is confirmed by their sorption properties, due not only to micropores, but also to the presence of mesopores.
System of mesopores. The low-temperature N 2 adsorption-desorption isotherms on natural phillipsite (Fig. 4) and its modified forms demonstrate a hysteresis loop with a jump at p/p 0 = 0.4-0.5 indicating the presence of mesopores including slit-shaped pores in nonrigid aggregates of particles (H 3 type hysteresis loop) and possibly well defined cylindrical pore channels (H 1 type hysteresis loop). Average pore diameter of mesopores, calculated by the Barrett-Joyner-Halenda method using adsorption and desorption isotherm, is 22.0 and 54.4 nm, respectively.

Fig. 4. N 2 adsorption-desorption isotherms on NPSH
The system of mesopores does not change as a result of the introduction of transition metals into the phillipsite structure; some changes are observed for adsorption isotherms at low relative pressures (0.05<p/p 0 <0.25), under conditions of filling micropores.
However, the pore sizes in phillipsite crystal structure are close to the kinetic diameter of N 2 (3.64 Å), and the Brunauer-Emmett-Teller method cannot be used to estimate the surface area and volume of micropores: despite the formal suitability of this method up to p/p 0 < 0.2, it gives an average pore diameter over 15 nm, typical for mesopores and not for micropores.
Room temperature water adsorption capacity (Table 6) at the "plateau" pressure (p/p 0 = 0.40) is 7.25 mmol/g or 0.130 cm 3 /g, and is 46 % of the total water adsorption capacity 0.285 cm 3 /g, which is close to the total pore volume 0.278 cm 3 /g, determined from the lowtemperature N 2 adsorption-desorption isotherms. The total pore volume within the experimental error is preserved; the volume of micropores in the MZs available for the adsorption of water molecules decreases, especially for the silvercontaining sample AgPSH. Apparently, transition metal ions, when introduced into the zeolite structure, at least partially retain their hydration shell, effectively reducing the free micropore volume for adsorbing water molecules.
Morphology. In general, the procedure of ion-exchange synthesis leads to a significant increase in the dispersion of the material, the size of the largest crystallites does not exceed 20 µm for AgPSH (Fig. 5 a) and 30 µm for ZnPSH (Fig. 6 a) and CuPSH. Compared with the natural phillipsite, the proportion of crystallites smaller than 2 µm is increased, especially for the AgPSH sample (Fig. 5 b). In all likelihood, the increase in dispersion is associated with the removal of clay minerals (see Fig. 2 b) when washing with water after "dry" ion exchange procedure. Additional grinding with the introduction of silver can be explained by the acidic action of the NO 3 counterion.   The silver-and copper-containing crystals are sufficiently isolated, whereas the aggregation of zinc-containing crystallites (Fig. 6 b) is preserved to a greater degree, like in the natural phillipsite.
Release of metal ions. Data on leaching of metals from modified zeolites are given in Table 7. The amount of silver ions released after 6 h corresponds to their concentration of 0.067 mM, which is higher than the minimal inhibitory concentration (MIC) value for silver ions toward E. coli, 3.996 mg Ag in dm 3 [43] or 0.037 mM.
On the contrary, the amount of copper and zinc ions released after 24 h corresponds to concentration of 0.45 and 0.5 mM, respectively, lower than MIC value for copper and zinc ions toward E. coli, 1 mM [44].
Bactericidal activity. Table 8 shows the relative number of viable cells of E. coli suspended in water after their contact with natural and modified phillipsites in relation to the number of cells at the beginning of the experiment.
Taking into account the leaching of bioactive metals and their comparison with the values of the minimal inhibitory concentrations, we can conclude that the antibacterial activity of CuPSH and ZnPSH could be ascribed to the metalcontaining zeolite M-Z itself and not to the leached metal ions. The silver-containing zeolite AgPSH also exhibits a certain antibacterial activity even before the concentration of ions in the solution reaches the MIC value, and its bactericidal effect could be ascribed not only to released Ag + ions but also to AgPSH itself.
The total number of bacteria in bottles with NPSH was not significantly different than in the corresponding controls, showing that natural phillipsite had no antibacterial activity itself. Bacteriostatic activity. Results of the Kirby-Bauer test are given in the Table 9. No antibacterial action was observed for the original phillipsite.
The width of the inhibition zones of the antibacterial metal-exchanged forms AgPSH, CuPSH, and ZnPSH is of the same order of magnitude as for silver-, copper, and zinccontaining clinoptilolite-rich mineral from Gördes, Turkey, Western Anatolia [18], and exceed diameter of inhibition zones reported for copper-containing clinoptilolite from the "Holinskoe" mineral deposit, Russia, Republic of Buryatia [32].
Environmental aspect. The Sheldon's factor E, the ratio of the mass of waste per mass of product [45], is an important environmental and green chemistry metrics, and its reduction is an urgent task. Table 10 presents data on the consumption of liquid materials for various methods of ionexchange production of MZs. Information about the required quantities of zeolites, reagents and solutions have been obtained from recent publications [21], [24], [27], and [43].
Data on the amount spent on washing the target products by distilled water, as a rule, are not given in publications. In accordance with our experience [46], washing of one gram of product filtered from a solution needs at least 300 mL of water, and washing after a "dry" ion exchange requires at least 500 mL. Solid waste such as zeolites, reagents, packaging and other consumables were not taken into account. The findings suggest that the solid-state "dry" method is preferable to the "wet" method of the ion exchange in solution.

CONCLUSION
As a result of the research carried out, it has been found that solid-state ion-exchange reactions between Georgian natural phillipsite and salt of corresponding transition metal followed by washing results in zeolite materials with a significantly higher content of silver (up to 230 mg/g), copper (up to 66 mg/g), and zinc (up to 86 mg/g), than those obtained by ion exchange in solutions on synthetic zeolites and natural clinoptilolite. The introduction of hydrated silver, copper, and zinc ions into the channels and pores of the zeolite is facilitated by the developed system of macro-and mesopores in the used natural phillipsite.