Ternary and Quaternary Nanocrystalline Cu-Based Sulfides as Perspective Antibacterial Materials Mechanochemically Synthesized in a Scalable Fashion.

Twelve Cu-based ternary (Cu-Me1-S, Me1 = Fe, Sn, or Sb) and quaternary (Cu-Me2-Sn-S, Me2 = Fe, Zn, or V) nanocrystalline sulfides are shown as perspective antibacterial materials here. They were prepared from elemental precursors by a one-step solvent-free mechanochemical synthesis in a 100 g batch using scalable eccentric vibratory ball milling. Most of the products have shown strong antibacterial activity against Escherichia coli and Staphylococcus aureus bacteria. For instance, stannite Cu2FeSnS4 and mohite Cu2SnS3 were the most active against E. coli, whereas kesterite Cu2ZnSnS4 and rhodostannite Cu2FeSn3S8 exhibited the highest antibacterial activity against S. aureus. In general, stannite has shown the best antibacterial properties out of all the studied samples. Five out of twelve products have been prepared using mechanochemical synthesis for the first time in a scalable fashion here. The presented synthetic approach is a promising alternative to traditional syntheses of nanomaterials suitable for biological applications and shows ternary and quaternary sulfides as potential candidates for the next-generation antibacterial agents.

Introduction

Contemporary human society is confronted with several critical concerns, and infection diseases are among the most important ones.[1] Antibiotics worked very well against bacteria in the past. However, nowadays, antimicrobial resistance has been reaching a critical level.[2] Insight into new antibacterially active materials, not yet known to the microbes, is therefore a reasonable approach. Finding a chemically stable, non-toxic, and low-cost antibacterial material is of utmost importance. Metal sulphides mimicking safe natural minerals seem to be rising stars in this area.[1,3] More binary metal sulphides perfectly serve this role. The antibacterial potential of Ag2S,[4−8] CdS,[9−17] CuS[18−27] MoS2,[28−38] and ZnS[39−44] has been studied many times. There are few publications on the antimicrobial action of more exotic binary sulfides like SnS2,[45] CoS2,[46] NiS,[47] and In2S3.[48] Binary sulfides are most often complemented by another compounds like oxides to get the composite with good antibacterial activity. The works on the antibacterial activity of ternary and quaternary sulfides are even more scarce, namely just the antibacterial potential of Cu2SnS3,[49] Cu2ZnSnS4,[50,51] and Cu12Sb4S13[52] has been discovered so far.

There are many synthetic pathways for sulfide nanoscale production,[3,53,54] among which a scalable, solvent-free one-step methodology called mechanochemical synthesis has found an inevitable place.[55,56] One of the scalable alternatives to the lab-scale mechanochemical synthesis is eccentric vibratory milling[57] and it has been sufficiently applied to prepare both sulfides[58−65] and selenides.[66]

This article provides a comprehensive viewpoint on the antibacterial potential of twelve Cu-based ternary and quaternary sulfides mechanochemically synthesized in a scalable fashion. For the most of the products, the antibacterial activity has not been reported yet. Moreover, five of them have not even been prepared mechanochemically so far. As the eccentric vibratory mill used for the experiments in this study was located at Technical University of Clausthal, Germany, a birthplace of famous microbiologist Robert Koch,[67] the samples in this study are labeled as KOCHx (x being a sample number), as a tribute to him.

Materials and Methods

Materials

For mechanochemical synthesis of chalcopyrite CuFeS2, chatkalite Cu6FeSn2S8, stannoidite Cu8Fe3Sn2S12, skinnerite Cu3SbS3, and stannite Cu2FeSnS4, the following precursors were used: copper (Merck, Germany, 99,7% purity), antimony (Merck, Germany, 99.8% purity), tin (Nihon Seiko, Japan, 99% purity), iron (WINLAB, Germany, 99% purity), and sulfur (CG-Chemikalien, Germany, 99% purity).

Mechanochemical Synthesis

The details on the mechanochemical syntheses of the majority of the studied compounds can be found in the following publications: mawsonite Cu6Fe2SnS8,[58] rhodostannite Cu2FeSn3S8,[59] kesterite Cu2ZnSnS4,[60,65] tetrahedrite Cu12Sb4S13,[61] colusite Cu13VSn3S16,[62] mohite Cu2SnS3,[63] and famatinite Cu3SbS4.[64] The weights of precursors for all the compounds (including those newly prepared) and the corresponding milling time are provided in Table .

Table 1

Weights of the Precursors and Milling Time for the Mechanochemical Syntheses of Ternary and Quaternary Cu-Based Sulfides Synthesized in this Studya

			weight (g)					milling time (min)
sample	desired phase	phase name	Cu	Fe	Sb	Sn	S
KOCH 1	CuFeS2	chalcopyrite	34.6	30.4			35.0	720
KOCH 2[58]	Cu6Fe2SnS8	mawsonite	53.9	12.9		13.7	29.5	240
KOCH 3	Cu6FeSn2S8	chatkalite	41	6		25.5	27.5	120
KOCH 4	Cu8Fe3Sn2S12	stannoidite	39.2	12.9		18.3	29.6	120
KOCH 5[59]	Cu2FeSn3S8	rhodostannite	16	7		32.2	44.8	600
KOCH 6[60]	Cu2ZnSnS4	kesterite	28.9	14.9		27.0	29.2	360
KOCH 7[61]	Cu12Sb4S13	tetrahedrite	45.8		29.2		25.0	240
KOCH 8[62]	Cu13VSn3S16	colusite	47.3			20.4	29.4	720
KOCH 9[63]	Cu2SnS3	mohite	37.2			34.7	28.1	180
KOCH 10[64]	Cu3SbS4	famatinite	43.3		27.6		29.1	240
KOCH 11	Cu3SbS3	skinnerite	46.7		29.8		23.5	120
KOCH 12	Cu2FeSnS4	stannite	29.6	13.0		27.6	29.8	120

For colusite (KOCH8), also 2.9 g of vanadium was used for the synthesis.

The rest of the milling conditions was similar for all the samples. The mechanochemical syntheses were carried out in an industrial eccentric vibratory ball mill ESM 656-0.5ks (Siebtechnik, Germany) working under the following conditions: A 5 L steel satellite milling chamber attached to the main corpus of the mill, 83 tungsten carbide balls with a diameter of 35 mm and a total mass of 30 kg, 80% ball filling in the milling chamber, amplitude of inhomogeneous vibrations 20 mm, rotational speed of the eccenter 960 min–1, and an argon atmosphere. The total feed of reaction precursors was 100 g per batch. The milling was performed for different times which are mentioned in the appropriate sections.

Characterization

X-ray Diffraction

The phase composition of all mechanochemically prepared products was analyzed by X-ray diffraction (XRD, Rigaku SmartLab 3 kW) in a Bragg–Brentano geometry in a 2θ-angle range of 10–90° using Cu Kα radiation with 30 kV and 40 mA and a scan speed of 4° min–1. The phase identification in the acquired XRD patterns was performed using the HighScore Plus software (PANalytical B.V., The Netherlands, version 3.0e) in the 2θ-angle range of 15–65°.

ζ-Potential Measurements

ζ-potential was measured in the diluted water solution of sodium chloride (10 mM) using Zetasizer Nano ZS (Malvern, Malvern, U.K.) setup, the electrophoretic mobility of the particles being converted to zeta potential using the Smoluchowski equation built in the Malvern Zetasizer software. The measurements were performed in triplicate with at least 12 sub-runs for each sample.

Antibacterial Activity

The antibacterial properties of the samples were evaluated by the agar well diffusion method by a slight modification in the process reported in.[68] The tested bacteria (Staphylococcus aureus CCM 4223 and Escherichia coli CCM 3988) were obtained from the Czech collection of microorganisms (CCM, Brno, Czech Republic). The procedure used was as follows:

Bacteria were cultured overnight, aerobically at 37 °C in Luria–Bertani (LB) medium (Sigma-Aldrich, Saint-Louis, MO) with agitation. After this, bacteria were mixed with 50% glycerol (Mikrochem, Pezinok, Slovakia) and frozen glycerol stock cultures were maintained at −20 °C. Before the experimental use, cultures were transferred to LB medium and incubated for 24 h, and used as the source of inoculum for each experiment.

Plate count agar (HIMEDIA, Mumbai, India) medium was cooled to 42 °C after autoclaving, inoculated overnight with liquid bacterial culture to a cell density of 5 × 105 colony-forming units per mililiter (cfu/mL).

20 mL of this inoculated agar was pipetted into a 90 mm diameter Petri dish.

Once the agar was solidified, five mm diameter wells were punched in the agar and filled with 50 μL of samples prepared in the form of suspensions (prepared by dispersing 20 mg of KOCHx samples in 1 mL of distilled water). Gentamicin sulfate (Biosera, Nuaille, France) with the concentration of 30 mM was used as a positive control.

The plates were incubated for 24 h at 37 °C.

Afterward, the plates were photographed and the inhibition zones were measured by the ImageJ 1.53e software (National Institutes of Health, Bethesda, MD). The values used for the calculation are mean values obtained from 3 replicate tests.

The antibacterial activity was calculated by applying the formula reported in:[68] %Relative inhibition zone diameter (RIZD = [(IZD sample – IZD negative control)/IZD gentamicin] × 100, where RIZD is the relative inhibition zone diameter (%) and IZD is the inhibition zone diameter (mm). As a negative control, the inhibition zones of distilled water equal to 0 were taken. The inhibition zone diameter (IZD) was obtained by measuring the diameter of the transparent zone and subtracting the size of the wells (5 mm).

Results and Discussion

X-ray Diffraction

The formation of nanoparticles by mechanochemistry is well-known.[55,69] In general, two phenomena are observed during high-energy milling, mechanical activation and mechanochemical reaction.[70,71] While the first one can be considered a top-down approach (by diminishing the crystallite size down to the nanoscale), the second one might also be considered as a bottom-up strategy, as the size of the produced nanoparticles can be increased with prolonged milling. The particle size of the precursors is being reduced down to the nanoscale and simultaneously, the reaction between them occurs on the grain boundaries. The produced nanoparticles are most often present in the form of microcrystalline agglomerates; however, the X-ray diffraction usually reveals nanoscale dimensions of the individual crystallites.

The XRD patterns of all the samples synthesized in the present study and their principal identified phases are provided in Figure .

Figure 1

XRD patterns of all powders under study containing mostly ternary and quaternary sulfides: (a) KOCH 1-3, (b) KOCH 4-6, (c) KOCH 7–9, (d) KOCH 10–12. High-intensity diffraction peaks are marked in black for the main present phase, additional secondary phases are marked in grey for each system.

The mechanochemical syntheses of ternary and quaternary Cu sulfides represent an excellent scalable method to obtain a predetermined Cu–sulfide phase.[63,65] In general, all samples were composed of the desired ternary or quaternary copper sulfide as a major phase. However, secondary phases were also present either as a result of an incomplete reaction between the initial reactants or because of the fact that they represent intermediate phases formed on the reaction pathway to the desired compound. A significant peak broadening clearly shows the nanocrystalline character of the products, and this has been also confirmed by calculations from XRD data and transmission electron microscopy (TEM) analysis for seven products that have been synthesized before.[58−65]

It can be deduced that the secondary phases are present in small amounts by comparing the peak intensities of the identified phases in each processed system. Only in the synthesis of tetrahedrite (Cu12Sb4S13, KOCH7 sample), similar peak intensity of famatinite (Cu3SbS4) was registered. The presence of secondary phases is something that must be followed for scalable intentions in potential industrial applications, especially if the presence of a specific secondary phase compromises the antibacterial activity of the final product. This can be exemplified in the KOCH7 system, where neither of the two ternary Cu-based sulfides did contribute to the antibacterial response, as will be shown later.

An overview of the identified phases in each processed sample, together with their crystallographic information references or synthesis methodologies is provided in the supplementary material (Table S1).

Antibacterial Activity

Relative Inhibition Zone Diameter (RIZD) determination using the agar well diffusion method

Twelve Cu-based sulfides were subjected to antibacterial tests using the agar well diffusion method. The relative inhibition zone diameters (RIZDs) received for both bacterial strains are summarized in Figure and representative images from the experiments can be seen in the supporting file (Figure S1).

Figure 2

Relative inhibition zone diameter (RIZD) for all twelve studied samples for both S. aureus and E. coli. The positive control was antibiotic gentamicin with a concentration of 30 μg/mL and its RIZD was taken as 100%.

It can be seen that only three samples did not exhibit any antibacterial activity, namely chalcopyrite CuFeS2, chatkalite Cu6FeSn2S8, and tetrahedrite Cu12Sb4S13. Out of these three compounds, only the antibacterial potential of tetrahedrite has been revealed so far.[52] In the mentioned study, the E. coli bacteria could be almost completely destroyed by the Cu12Sb4S13 film in 10 min. The fact that tetrahedrite was not active in our case might be connected with the difference in the methodology used to prepare the material, which results in dissimilar properties (e.g., in nonidentical particle size).

For all other nine samples, at least some activity could be observed. In general, the activity was better against gram-negative E. coli (7 out of 12 investigated samples were active and RIZD values were slightly larger for the other bacteria). In the case of gram-positive S. aureus, the activity was registered in 5 out of 12 samples. Interestingly, different compounds were active against different types of bacteria, e.g., kesterite Cu2ZnSnS4 and rhodostannite Cu2FeSn3S8 were active only against gram-positive S. aureus, whereas skinnerite Cu3SbS3, famatinite Cu3SbS4, collusite Cu13VSn3S16, and mawsonite Cu6Fe2SnS8 were active only against gram-negative E. coli. The activity against both types of bacteria was detected only scarcely (in 3 out of 12 samples), namely for stannite Cu2FeSnS4, mohite Cu2SnS3, and stannoidite Cu8Fe3Sn2S12. The highest RIZD was evidenced in the case of stannite (136.25 ± 8.6%) against E. coli. Against S. aureus, the best activity out of 12 studied samples was evidenced for kesterite Cu2ZnSnS4 (69.41 ± 8.21%).

The antibacterial potential of kesterite has already been studied earlier.[50,51] The zone of inhibition (ZOI) reported in[50] was 3 and 5 mm for E. coli and S. aureus, respectively, being in accordance with better activity of this compound against S. aureus observed also in our case. In the mentioned study, 1 mg of the powder was introduced into the well, which is the same as in our case (accepting the premise that the powder was homogeneously distributed in the distilled water before testing). The mechanism of antibacterial action of kesterite is most probably connected with the electrostatic interaction due to the opposite surface charges of the compound and the bacterial cell wall.[72]

The antibacterial effect of mohite Cu2SnS3 has been revealed by Lokhande et al. upon testing it against the same bacteria as in our case (S. aureus and E. coli). The antibacterial action was stronger against E. coli, similar to our study. The authors ascribed the difference to the thickness of the walls of the two bacteria, being 80 nm thick for S. aureus and only 10 nm thick for E. coli, respectively,[73] making the former one more resistant. The authors have demonstrated the damage caused to the E. coli and S. aureus cells by providing the scanning electron microscopy (SEM) images of the bacteria after being subjected to the effect of CTS nanocrystals. While E. coli bacterial cells’ length increased as a result of the exposure to the Cu2SnS3 nanoparticles, the shrinkage of cellular texture and extrusion of intracellular fluids were observed in the case of S. aureus. Similar to kesterite, the authors also ascribe the mechanism of action to the electrostatic interaction between the oppositely charged species (positive charge of the nanoparticles and negative charge of the cell wall).[74,75]

Relationship between Antibacterial Action and ζ-Potential Measurements

The surface charge of the compounds, which is expressed by their ζ-potential, can play a decisive role in the antibacterial action. ζ-potential reflects the stability of the reaction systems in the form of colloidal dispersions.[76]

Therefore, ζ-potential measurements for all samples were performed. The measured values are summarized in Figure and the Electronic Supporting Information (Table S2). It can be seen that the whole range of ζ-potential spanning from very negative to very positive values was evidenced in our case, the most negative being in the case of Cu3SbS3 (−32.9 mV) and the most positive in the case of chalcopyrite CuFeS2 (15.3 mV). Similar results have been observed in the literature.[77−79] The presence of iron seems to lead to the positive ζ-potential values (KOCH 1–5, 12). This is a result of the dissolution of Fe2+ ions originating from iron particles and their transfer into the solution.[77] This is also the case of Zn2+ ions (KOCH 9).[80] On the other hand, the presence of antimony yields negative ζ-potential values (samples KOCH 7, 10 and 11), as evidenced for chalcostibite CuSbS2.[78] The direct correlation between the ζ-potential values and the antibacterial action can be seen in Figure .

Figure 3

Dependence of antibacterial activity expressed as RIZD on ζ-potential. The formulas of the main phases in the samples are in different color: green- active against both bacteria, red—active only against S. aureus, blue—active only against E. coli, black—not active against either bacteria.

The correlation of ζ-potential values and the RIZD detected for all samples provides us with some clues. First, it seems that a very positive ζ-potential value is not beneficial, as when it was higher than 10 mV, no antibacterial action was observed. Second, there is a markable difference in the fact that ZP values are beneficial for the antibacterial action against the given bacterial species. While quite a wide range of ZP values (from almost −35 to +8 mV) are efficient against E. coli, there is quite a narrow window for the action against S. aureus (values between +4 and +10 mV). The products with the negative ZP values are not active against S. aureus, but are only active against E. coli, so there might be some correlation between the negative surface charge of the given sulfides and the properties of the gram-negative E. coli bacteria.

In three out of four studies reporting the antibacterial activity of the multinary sulfides published so far, the authors also determined the minimum inhibitory concentration (MIC). It is defined as the lowest concentration of a drug that will inhibit the visible growth of an organism after overnight incubation. The MIC value for CZTS nanocrystals prepared in[50] was 500 μg/mL against all tested bacteria, including E. coli and S. aureus. The antibacterial action of kesterite was tested against these bacterial cell lines (among others) also in[51] and the MIC value was in the range of 128–512 μg/mL, thus the results are in accordance. Lokhande et al.[49] observed a significant reduction in the growth of bacterial colonies (reporting this to be MIC) after the introduction of 1 and 3 mL of mohite Cu2SnS3 solution (solvent unspecified) for S. aureus and E. coli, respectively; thus the activity was better against S. aureus in this case. However, as mentioned earlier, the ZOI obtained by the agar well diffusion method in that study was larger for E. coli than for S. aureus, so the results from the two complementary methods are contradictory. In the present research, we also tried to determine the MIC value, however, as our powders are black, their dispersion in the solution with bacteria caused the solution to become opaque, which hampered the proper determination of MIC, as this is done spectrophotometrically.

Conclusions

Twelve ternary and quaternary nano-sulfides were successfully prepared by a mechanochemical one-step solvent-free synthesis in a 100 g batch using an eccentric vibratory ball mill. The synthesis is perfectly feasible just by solid-state milling of the elemental precursors, as in the majority of experiments, the desired sulfides were prepared as main phases. The agar well diffusion method has shown that at the concentration of 20 mg/mL, most of the products are efficient antibacterial agents, throwing some light on the influence of the chemical composition on the antibacterial action. In general, better activity of the studied sulfides was evidenced against gram-negative E. coli; however, in two cases (kesterite Cu2ZnSnS4 and rhodostannite Cu3FeSnS8), only the activity against S. aureus was observed. The most potent agents were found to be stannite Cu2FeSnS4 and mohite Cu2SnS3. The former one exhibited a significant antibacterial action against both types of bacteria. Only three out of twelve products (namely, chalcopyrite CuFeS2, chatkalite Cu6FeSn2S8, and tetrahedrite Cu12Sb4S13) did not show any activity at the studied concentration. The investigation on the relationship between ζ-potential values and antibacterial activity has revealed that the products with the negative ζ-potential values were efficient only against E. coli bacteria, thus there might be some relationship between their negative charge and the properties of the E. coli cell wall. The present research has shown nanocrystalline sulfides as interesting alternatives to traditional antibacterial agents and also the robustness of the mechanochemical synthesis performed on a larger scale.