Soil from farmland, in Beijing (39˚59' - 116˚21'), was selected for this study. After the removal of the uppermost soil layer, soils samples were collected at a depth of 5 - 15 cm ground surface soil, the most populated soil layer by microorganisms [35] . It was then transported in sterile polyethylene bags to the laboratory where it was air-dried at room temperature, homogenized, and sieved (mesh size 2 × 2 mm) for the removal of the large particles. The physicochemical characteristics of this soil were previously analyzed [36] as follows: pH, 7.26; Zn, 3.3 mg・kg⁻¹; Fe, 16.5 mg・kg⁻¹; Total organic matter, 17.1 g・kg⁻¹; total N, 76.2 mg・kg⁻¹; P, 78.9 mg・kg⁻¹; and K, 144.5 mg・kg⁻¹. Samples were collected in July 2016 and stored in sterile polyethylene bags and kept at 4˚C.

Cu was used as CuSO₄・5H₂O compound. Copper sulfate (CuSO₄) is one of the most used copper compounds and one of the most contributors to environmental pollution by this element. Xanthates, PAX and SIAX, were provided in the solid and analytical form from Beijing General Research Institute of Mining and Metallurgy. Both copper and collectors stock solutions were prepared using distilled water and were kept at 4˚C during the experimental time.

Data analysis was conducted using the Origin 8.5 (OriginLab, MA, USA) software, Excel, and Spss Statistics 17.0 software packages. Comparison between results was performed through the Pearson correlation coefficient analysis, with a statistical significance of P < 0.05 or P < 0.01. All treated samples were compared to the control sample. The power-time curves allowed obtaining the peak-time (T_peak) values of the microbial growth and the corresponding peak-power (P_peak) values. The microbial growth rate constant (k) was calculated through linear fitting equation using the semilogarithmic conversion of the heat flow rate data [36] [37] [38] (Equation (1)) while the inhibitory ratio (I, in %) was calculated on the basis of K as follows (Equation (2)) :

P t = P 0 exp ( k t ) or ln P t = n P 0 + k t (1)

I ( % ) = K 0 − K c K 0 × 100 (2)

where P₀ and P_t are, respectively, the heat output at time 0 and time t; I (%) is the inhibitory ratio and K₀ and K_c are the growth rate constant of the control and the growth rate constant of soil microbes under c dose of toxicant, respectively. K and I are the most used parameters for a quantitative evaluation of the microbial response to a given stressful condition [37] [39] . On this basis, the mixture inhibitory ratio was compared to that of the individuals’ mixture components to determine whether the exposure to the mixture could be more harmful or not with respect to the individual effects of the components of the mixture. The mixture type-reaction was determined by calculating the ratio (R) between the observed mixture inhibitory ratio (MI) and the sum of the inhibitory ratios of the mixture components as follows (Equation (3)):

R = M I ∑ n = 1 i I i (3)

where R represents the ratio between the observed mixture inhibitory ratio (MI) and the additive inhibitory ratios of the mixture components, Ii represents the inhibitory ratio of i-mixture components alone. R = 1 indicates a sample additivity; R < 1 and R > 1 correspond to an antagonistic action and a synergistic action between the mixture components, respectively.

The power-time curves corresponding to the soil microbial activity in the control sample and treated samples with different doses of Cu, PAX, and SIAX are shown in Figures 1(a)-(c). As it was reported in the previous similar studies, samples exhibited differences in the curves profiles and duration of the microbial growth phases (lag, exponential, stationary and death phases) indicating a dose-effect relationship [37] [38] .

The lag and stationary phases were relatively extended in relation to the increase in the dose of pollutants. Different samples exhibited an interruption of the decline phase or death phase by reactivation and increase of the soil microbial growth, particularly for the samples treated with xanthates, individually or in the mixtures. Such a situation may be due to the growth of the surviving microbes that grow slowly (e.g. Fungi) [40] or take much time to adapt to pollutants. In addition, it is known that the first peak (Peak I) generally occurs at the exhaustion-time of glucose supplement [41] . Thus, the observed inflexion and secondary peaks (Peak II) can also be explained by the usage by the surviving organisms of secondary sources of metabolites, including through the decomposition of dead sensitive microbes, as well as the usage of the residual ammonium sulfate by some autotrophic microorganisms [37] [40] .

The thermokinetic parameters and the corresponding statistics are presented in Table 1. The peak-time relatively increased with the increase in pollutants dose, but without a significant relationship (e.g. 14.88 h for the control sample, 16.2 h for 600 μg∙Cu∙g⁻¹ soil). The peak-power (P_peak) significantly decreased (P < 0.05 or P < 0.01) in relation to the increase in pollutants dose (Table 1 and Table 2). The highest and lowest peak-power values were observed with the control sample (1112.10 μW・g⁻¹) and the highest dose of chemicals (436.55, 310.45, and 264.13 μW・g⁻¹ for 200 μg∙PAX・g⁻¹ soil, 200 μg∙SIAX・g⁻¹ soil and 600 μg∙Cu・g⁻¹ soil, respectively). For all treatments, the microbial growth rate constant (K) significantly decreased with the pollutants dose. In contrast, the inhibitory ratio (I) significantly increased with the increasing treatment dose (Table 1 and Table 2). Based on the inhibitory ratio (I) which is an ideal indicator of the toxicity effect [42], and the inhibitory concentrations (IC 12.5, IC 25 and IC 50) (Table 1), the highest and lowest toxicity were observed with SIAX and Cu, respectively (e.g. IC 50 was 83.95 μg・g⁻¹, 158.3 μg・g⁻¹, and 447.5 μg・g⁻¹ for SIAX, PAX and Cu, respectively).

Different reasons may explain the observed lower toxicity level for copper than for xanthates.

In contrast to xanthates which are manmade organic chemicals recently produced, mainly for the mineral flotation purposes [13], Cu is naturally occurring in the environment. Therefore, soil microorganisms could have developed more homeostasis and resistance mechanisms along with their evolutionary pathway [43] [44] than it is for xanthates. Moreover, Cu plays an essential role of micronutrient for the living organisms. Specifically, Cu acts as a prosthetic group involved in a variety of metabolic pathways [24] . However, above a threshold concentration, Cu becomes harmful and negatively affects the microbial diversity, biomass, and functions [24] [45] . Such toxicological properties confer to Cu a

wide range of use as a biocide and its application in medicine. With respect to xanthates, it was reported that even a very low dose, less than 1 mg・L⁻¹, is harmful to aquatic organisms [26] . From recent findings, xanthates, including sodium ethyl xanthate and sodium isobutyl xanthate, affected some moderately thermophilic, mineral-oxidizing acidophilic bacteria and mineral bioleaching bacteria (e.g. Leptospirillum sp., Acidimicrobium ferrooxidans, Acidithiobacillus caldus and Sulfobacillus sp.) and archaeon (e.g. Ferroplasma sp.), with a high sensitivity to less than 100 μg・mL⁻¹, and even to less than 1 μg・mL⁻¹ [46] [47] [48] . The activity of soil enzymes, including FDA hydrolyase and catalase (an antioxidative stress enzyme), was negatively affected by the exposure to sodium isobutyl xanthate and sodium isopropyl xanthate in the amounts ranging between 5 to 100 µg・g⁻¹ soil [49] . However, such a toxicity level can be influenced by the fate of xanthates and their decomposition products in the environment. In fact, unlike metals which are non-biodegradable and persistent elements, various organic compounds, including xanthates, are subject to various degradation

T_peak = peak time; P_peak = peak power; K = microbial growth rate constant; I = Inhibitory ration; IC = Inhibitory concentration. In case of SIAX, the peak time and peak power are given for both first peak (Peak I) and second peak (Peak II) of the power time curves, respectively.

processes, the most important being the microbial-mediated decomposition [13] [31] [50] . Unfortunately, such chemicals decomposition does not always result in the detoxification of the affected system (production of less or non-toxic by-products), but can also lead to the formation of new toxic elements (equally or more adverse than those from which they originate) or nonbiodegradable or persistent compounds [51] . The moisture content and temperature, as well as the existence of potentially degrading microbial strains, of our experimental conditions are favorable for such xanthates decomposition. Xanthates decomposition leads to the generation of various toxic compounds, including carbon disulfide (CS₂), hydrogen sulfide (H₂S), hydrogen peroxide (H₂O₂) [13] [31] [52] which can all affect the exposed microbial community. Human health-based studies have reported that CS₂ was associated with a depletion of anti-oxidative defense mechanisms, with a decreased level of the plasma antioxidant α-tocopherol and a decreased activity of antioxidant enzymes (i.e. glutathione peroxidase and catalase), the reduction in cytochrome P-450 and cytochrome c-reductase [53] . Exposure to CS₂ resulted in clinical effects on male sexual function and semen quality [54] . Due to its toxicity to soil bacteria, CS₂ is less biodegradable in soil and is used as soil disinfectant. Excessive H₂S is cytotoxic and lead to oxidative stress [55] . H₂S is harmful to the metal-containing enzymes

**Correlation is significant at the 0.01 level (2-tailed). *Correlation is significant at the 0.05 level (2-tailed).

[56] and can also interfere with Cytochrome c oxidase [57], inhibiting, therefore, the mitochondrial function or blocking the energy production. Such toxic effect was compared to that of cyanide and carbon monoxide [56] . The decomposition of H₂S is associated with the production of hydrogen peroxide and oxygen depletion (Equation (4)):

O 2 + H 2 S → S o + H 2 O 2 (4)

H₂O₂ is an oxidative stress inducer and can lead to bacterial cell damage and death [58] . However, the high mobility of most of those xanthates decomposition products may affect their residence time and contact with the exposed organisms while it can lead to air and water pollution. For example, investigation have reported a high volatility and partition to air from soils and waters for CS₂ [59], H₂S [59] [60] and COS [60] . Due to their low sorption capacity to soils and high water solubility, they are likely to be transported through moist soils and aquatic environment via water infiltration and reach the deep part of the Earth and the groundwater. In this study, despites the high inhibitory ratio or high toxic effect observed with SIAX, exposure to this chemical was generally characterized by a remarkable microbial growth after the first decline or death phase (Figure 1(c)). Such situation may be due to a rapid degradation or to a generation of less harmful degradation products than PAX. Based on this assumption, sodium isoamyl xanthate should be ecologically preferred to potassium amyl xanthates.

The power-time curves corresponding to microbial activity under the binary mixtures of the equitoxic doses of copper and xanthates, PAX and SIAX, are shown in Figure 2(a) and Figure 2(b). Those corresponding to the microbial exposure to the binary mixtures of a fixed dose of copper (200 μg・g⁻¹ soil) and different xanthates doses (25, 50 and 100 μg・g⁻¹ soil) are shown in Figure 3(a) and Figure 3(b).

The corresponding thermokinetic parameters and the related statistical analysis are shown in Table 3. As for the single toxicity, the curves profiles exhibited lag, exponential, stationary and death phases of the microbial growth. The lag and stationary phases were relatively extended in relation to the increasing dose of the pollutants.

For all treated samples the peak-time (T_peak) was higher than that of the control sample (Table 3) and was increased in relation to the applied dose. For the same dose of copper (200 μg・g⁻¹ soil), increasing the xanthates dose (from 25 to 100 μg・g⁻¹ soil) in the mixture was followed by a decrease in the peak-power (P_peak) and microbial growth constant (K). Conversely, the inhibitory ratio (I) increased in the same treatment conditions.

For the same dose of copper (200 μg・g⁻¹ soil), increasing the xanthates dose (from 25 to 100 μg・g⁻¹ soil) in the mixture was followed by a decrease in the peak-power (P_peak) and microbial growth constant (K). Conversely, the inhibitory ratio (I) increased in the same treatment conditions. For all cases, the observed mixture inhibitory ratio (MI) was higher than that of the corresponding individual mixture components, meaning that the microbial communities are more affected by the simultaneous exposure to copper and xanthates than by the exposure to the single chemicals. For a mixture containing a copper dose of 200 µg・g⁻¹ and a xanthates dose of 100 µg・g⁻¹, the corresponding inhibitory ratio was 53.85% compared to 38.5% for xanthates and 23.1% for copper. For example, in case of the mixture containing 200 µg・Cu・g⁻¹ and a xanthates dose of 25, 50 and 100 µg・g⁻¹, the ratio (R) between the mixture inhibitory ratio and the sum of the individual components inhibitory ratios was less than 1, corresponding to an antagonistic mixture-type reaction. However, with respect to the mixture of the equitoxic amounts, synergism was observed in soil treated with the lower chemicals doses (R < 1).

From these results, the microbial activity was not completely suppressed by any of the applied doses, including by the components whose sum of the individual inhibitory ratios equals 100% (Table 3). This may simply mean that copper and xanthates may have different sites of action and do not additive actions. In particular, it can be justified by the existence of some microbial strains that

can adapt to all applied individual and combined chemicals doses, as well as to the formed complex copper/xanthates and to the xanthates associated decomposition products. It is well known that metals are mostly toxic in their free ionic form, mainly Cu²⁺ in case of copper, which can interact with the polar group (S=C-S⁻) of xanthates [13] [61], reducing the interaction of both chemicals with microbes. Such affinity of xanthates for metals has some ecological implications. Xanthated materials are capable of sorption for the metal, as they are for other minerals [62] [63] . On this basis, Potassium ethyl xanthate was successfully tested for the treatment of copper-rich wastewater [64] . Depending on the chemicals species and doses, xanthates can negatively or positively affect the bacteria involved in the mineral bioleaching, including those belonging to the

Obs.MI = observed mixture inhibitory ratio; AI = additive inhibitory ratio; K = microbial growth rate constant; I = Inhibitory ratio; Ia = inhibitory ratio of copper alone; Ib = Inhibitory ratio of xanthates alone; R = ratio between the observed mixture inhibitory ratio and the additive inhibitory ratio of the mixture components. For all cases, the observed mixture inhibitory ratio (MI) is higher than the individual copper and xanthates inhibitory ratios.

family Thiobacteriaceae [65] . It should also be mentioned that xanthates degradation products have the ability to interact with other soil elements, including metals. H₂S has the ability to interact with heavy metals such as Nickel, Copper, Iron, and Zinc to form insoluble metal sulfides [60] while carbonyl sulfide is corrosive to metals [66] . Such interactions are likely to impacts on their availability and interactions with the exposed organisms. On this basis, it can be understood that the observed mixture toxicity effect in this study may result from different processes involving copper, xanthates and the xanthates decomposition products. Accordingly, the risk assessment and management of the pollution associated with metals, copper in this study, and xanthates should take into consideration this complexity of various elements.