Characteristics of environmental samples before and after bioremediation
Table 1 lists the parameters of the samples collected from the upper aquifer (12 m) at three-time points. In sample 1, before bioremediation, the content of nitrate ions reached 2517 mg/L. Against this background, in an oxidizing environment, a high content of uranium up to 1.1 mg/L and plutonium up to 0.7 Bq/L was observed. The content of organic matter did not exceed 5.9 mg/L. The suspension contained a significant amount of clay particles. Uranium in sample 1 was predominantly in dissolved form or nanoaggregates less than 5 nm in size (Fig. 1).
Percentage distribution of uranium in the filtrate during sequential filtration of samples 1 and 3. Concentrations of U in the filtrates were determined by the ICP-MS method.
In sample 2, a year after the injection of organic matter, the content of nitrate ions reached 320 mg/L, while the values of the redox potential continued to remain in the reduction region (− 175 mV) as they were 3 months after bioremediation. The content of organic matter reached 57.5 mg/L. The uranium content dropped to 80 μg/L, and the plutonium content was below the detection limit of the device.
2 years after injection (sample 3), the content of nitrate ions increased to 970 mg/L, the redox potential entered the oxidizing region and reached + 70 mV, while no significant release of uranium into solution occurred. According to the distribution scheme of uranium (Fig. 1), most of it was associated with large particles of more than 400 microns in size of clay and ferruginous nature. The plutonium content was below the sensitivity of the method. Thus, despite the fact that after a single injection of organic matter, after two years the content of nitrate ions increased markedly and the value of the redox potential returned to the oxidizing region. Nevertheless, it should be mentioned no significant remobilization of uranium and plutonium occurred. It is important to note that according to the data in Table 1, a decrease in the content of suspended matter was observed in the course of bioremediation. A discussion of the content of organic matter in the suspended matter will be carried out in the next section.
Figure 2 shows electronic maps of micrographs of a filter with a maximum pore size after filtration of sample 3. It has been established that U is mainly associated with large particles (suspensions) of aluminosilicate and ferrous nature. The distribution of Al, Si, Fe and U on the surface of the filter cake was fairly uniform.
Electron micrographs of the filters with a pore size of 2400 nm surface after sample 3 filtration with elements maps (A) Al, (B) Si, (C) Fe, (D) U (SEM EDX analysis).
Although at low plutonium concentration it was not possible to see it by the SEM EDX method on clays, it is well known that clay minerals montmorillonite and kaolinite could have been carrier phases for Pu39. In work on the analysis of colloidal transport of radionuclides in groundwater at Yucca mountain40 uranium was found to be dominantly associated with an unidentified phase rich in Si and Fe while Pu was shown to be preferentially adsorbed onto Mn-oxides in the presence of Fe-oxides.
Laboratory simulation of biogenic associative colloids formation in environmental water samples, stimulated by H2
In a laboratory experiment with environmental samples, molecular hydrogen was used to stimulate microbial processes in order to avoid changing the content of the organic matter.
Filtration studies (step-by-step filtration, Fig. 3) revealed that only 8% of organic matter in sample 1 was represented by suspended particles over 1200 nm in size. These were bacterial cells and other large particles (fulvic and humate acids, etc.). More than 50% of organic matter was in soluble form or in the form of colloidal particles up to 100 nm. In general, the distribution of organic matter in sample 3 was similar to sample 1—about 60% of organic matter was in dissolved or colloidal form and about 10% in the form of large particles.
Organic matter distribution by particle size (nm) in samples 1 and 3 before and after (B) microbial activation. Organic matter in the filtrate after each filtration step was measured using an Elementar Vario EL III CHN analyzer.
An organic carbon content of 100 and 200 mg/L was observed in samples 1 and 2, respectively, after microbial activation by molecular hydrogen.
After day 30 of incubation in sample 1 and after microbial processes, there was a noticeable increase in the content of large organic particles; their contribution reached 50%. In this case, the content of dissolved organic matter and organic particles of colloidal size decreased noticeably (their total contribution did not exceed 10% probably due to their consumption or aggregation into larger fractions). The content of organic particles with a size range of 220–450 nm had noticeably increased.
In sample 3, a noticeable decrease in dissolved and colloidal organic matter was also noted; the content of organic particles of 220–100 nm and particles of 1200–400 nm increased markedly. We believe that the increase in organic particles in both samples in the range of 100–1200 nm is associated with an increase in the content of bacterial cells. Changes in the intensity of light scattering provided the most relevant information (Table 2).
In sample 1, before stimulation, the intensity of light scattering was at its maximum in the filtrate at 450–220 nm. In the filtrate less than 10 nm, light scattering was not detected. In filtrates larger than 450 nm and 220–50 nm, the values of the light scattering intensity were close. After microbial activation with hydrogen, a tenfold change in the intensity of light scattering was observed in the filtrate with particles larger than 2400 nm. Also, there was an almost twofold increase in filtrates with a particle size of 450–2400 nm, which is probably associated with the appearance of cells in the solution.
In sample 2, before microbial activation, the maximum intensity of light scattering was observed in the filtrate with particle sizes in the range of 450–1200 nm. After microbial activation, the intensity of light scattering significantly increased in all filtrates. It is important to note that the light scattering of particles with a size characteristic of colloids (50–100 nm) increased by more than 10 times. The different behavior after hydrogen activation of two samples can probably be explained by the fact that in sample 3 the microbial community was initially more active after the injection of organic matter into the formation. In both samples, a noticeable increase in the content of coarse suspensions may indicate the agglomeration of clay suspensions by microbial polysaccharides. According to Ivanov et al.41, a similar process is observed for soil and clay particles.
Laboratory simulation of the formation of biogenic associative colloids in model and environmental water samples with actinides
The second series of experiments was carried out to evaluate the behavior of U, Np, and Pu upon activation of microbial processes. At the first stage of the laboratory simulation, a significant enlargement of large particles possibly caused by the agglomeration of natural clay and ferruginous particles due to microbial polysaccharides in natural samples was found. An important task of the second stage of the work was to assess the contribution of ferruginous and clay particles to the distribution of actinides over particles with different sizes in model solutions.
When activating the microbial community in groundwater, a mixture of whey and acetate was used. However, in a laboratory simulation of this process, we decided not to use such a complex multicomponent substrate like whey. The whey contained a lot of organic suspensions and its use in this experiment would have led to even more uncertainties. A mixture of highly soluble sodium acetate and glucose substrates was added to the samples.
Table 3 shows the data on the content of polysaccharides and proteins in solutions during microbial processes in samples.
No significant increase of cells or polysaccharide content was recorded in samples with no organic matter additions. A low protein content was found in the sample NWO, which indicates that some content of cells remained in it after bioremediation. An increase in the concentration of the biomass, with peak values on day 10 and polysaccharides on day 15, was observed in all samples with additions of organic matter (O) (Table 3). The maximum accumulation of polysaccharides and protein was observed for the natural sample.
On the 30th day of the experiment, there was no visible sediment in the MW sample, in the rest of the samples, there was a large amount of sediment at the bottom of the test tubes. At the same time, the solution looked almost transparent in both the MW model water sample and the MWIO sample with added iron. The average hydrodynamic radii of colloidal particles were obtained on days 3, 7, 14, 21, and 28 of the experiment (Table 4). In model water samples without added organic compounds, colloidal particles were not formed. However, by the end of the experiment, particle formation was observed. This was probably due to the transformation of colloidal matter originating from the natural water aliquot or as a result of low microbial activity.
In the presence of glucose and acetate, the emergence of the colloidal phase and a gradual increase in particle size were observed from the fifth day of incubation. The average stable hydrodynamic radii of the particles amounted to ~ 100 nm. In the presence of clay, stable colloids with the average hydrodynamic radii of 80–90 nm were formed. Stimulation of microbial processes with glucose and acetate resulted in increased particle size and partial sedimentation (samples MWO through day 20, MWIO through day 15, and NWO through day 30). After that, the sedimentation of large particles took place, and particles of smaller sizes remained in the solution.
The addition of iron to the model system resulted in the formation of the particles with hydrodynamic radii of ~ 100 nm. The stimulation of the biological processes resulted in increased particle size, the formation of new particles (by day 21), and complete particle sedimentation by day 30.
An important parameter used to evaluate the stability of colloidal particles in the system is the value of particles’ zeta potential. When no organic matter was added, the charge of preliminarily filtered 100–50 nm particles equaled − 29, − 26.2 mV in model water, and − 16, − 12 mV in natural water, which indicates low stability of such particles (see Table 2 Supplementary). A shift in charge of particles towards zero and positive values was observed when microbial processes were running, and this hints at the stabilization of particles in the solution.
The diagrams of actinide distribution by size of colloidal particles in solutions of different nature before and after microbial stimulation on day 30 are shown in Fig. 4.
Actinide distribution by size of colloidal particles in solutions of different nature depending on the incubation time, normalized % in the filtrate. (I-before, II-after microbial stimulation on day 30). Actinides (233U, 237Np, and 239Pu) were added in the concentrations of 10–8 M/l per sample. Concentrations of 233U 239Pu were determined by liquid scintillation (Tri-Carb-3180 TR/SL liquid scintillation spectrometer) (“Perkin-Elmer,” USA).
In the model water Pu(IV) forms true colloidal associates (up to 50%) due to deep hydrolytic polymerization. Np(V) was also partially sorbed due to slight disproportionation (by 10%). U(VI) was a stable component of soluble carbonate complexes. In the model water, increased pH and decreased Eh result in the occurrence of 99% Pu, 30% Np, and 10% U within large colloidal particles. Ultrafiltration, however, is not suitable for the assessment of the possible actinide reduction and biosorption contribution to the process of colloid formation.
The microbiota and clay promote the stabilization of Pu, U, and Np in large colloidal particles. The addition of iron had no effect on actinide colloid formation, although iron caused a significant increase in neptunium colloid formation in the presence of the microbiota. This is probably due to the formation of iron-polysaccharide complexes42, which also have a high ability to chelate actinides.
In Bentley43, high-efficiency Pu(V) adsorption onto colloids about 100 nm in size–including hematite, silica, and montmorillonite in natural and synthetic Yucca Mountain water–was mentioned. Pu sorption onto iron oxides such as hematite is strong and irreversible, but only 50% of the Pu(V) was sorbed on the montmorillonite. The time dependence of the sorption onto the clay suggests a more complicated interaction than occurs with oxide minerals. They also discussed the desorption of Pu(IV) and Pu(V) from various colloids: Pu(IV) exhibits a greater tendency to desorb than Pu(V).
According to Silva and Nitsche44, conditions for Np(V) (intrinsic) colloid formation in environmental waters would usually not be achieved, and Np pseudocolloid transport may be more important. In this case, Np (V and IV) could be sorbed onto environmental colloids such as Fe(OH)3 SiO2, humic and clay particles45.
Thus, microbial processes may result in the coagulation of natural colloids due to the development of a weak negative surface charge. Tinnacher, et al.39 also mentioned that organic matter can modify the surface charge and characteristics of particle and colloid aggregates, depending on their size. For example, large surface-active organic molecules such as polysaccharides act to bind colloid particles together and thus cause colloid instability. In addition, microbial polymers and exopolymers themselves can not only lead to adhesion of colloids and suspended particles, but also increase the sorption of actinides on them due to various organic functional groups38,46,47,48.
After termination of the biogenic processes, the formation of large uranium-, plutonium-, and neptunium-containing particles associated with cells49 and large biopolymers (protein-polysaccharide biofilms)50,51 was observed. This results in decreased migration activity of the actinides since it is known that the transport of colloids is strongly influenced by their size and water filtration parameters. With an increase in the size and charge of a colloidal particle, the risk of its migration is significantly reduced52.
The stability of colloidal solutions depends on many factors: the size and concentration of particles of a substance, temperature, and the presence of electrolytes. An increase in the content of two and three charged ions in systems can lead to rapid coagulation of colloidal particles. The addition of bivalent cations (e.g., Ca2+, Mg2+) during bioremediation should potentiate this process and may become an efficient mechanism for decreasing the risk of active migration of radionuclide-associated particles.
Thermodynamical modelling of the species of radionuclide occurrence in the course of biotransformation
It is well known that the nucleation process of biogenic iron oxyhydroxides (leading to their mineralization) is closely related to the organic matter of exopolysaccharides of biofilms. Through strong mineral binding (high feq), microbial polymers can decrease the nucleation barriers for ferrihydrite and direct nucleation on the polymers53. The mineralization process in microbial exopolysaccharide sediments of iron and associated actinides54 can serve as a reliable anti-migration biogeochemical barrier, an important consequence of bioremediation.
To assess the possible contribution of biomineralization processes to the increase in the size of suspended particles which we observed in experiments, as well as to theoretically determine the forms of actinides during the development of the microbiota, thermodynamic modelling was carried out. The data on pH and Eh changes during the process of colloid formation and actinide incorporation into associative particles (and possibly into true colloidal particles as well) are listed in Table 3 (supplementary).
The most notable pH and Eh changes occurred in the presence of the microbiota, which was probably due to an increase in the number of bacteria. The pH increased moderately, while Eh values changed to negative, potentially creating the conditions for a shift of actinides’ oxidation states to lower ones. Since Ac(IV) is the most sorbed form of actinides, this may promote their association with colloidal materials of various natures55.
Speciation of elements, including dissolved species and the phase saturation indices, was calculated for 500 µg/L U, Pu, and Np in the Sample NWO (natural sample 2) (Table 5). The species of actinides and iron after microbial processes were calculated with an account for the following parameter changes: pH increase by 1, Eh decrease by 100 mV, complete denitrification, and sulfate reduction.
In accordance with previous thermodynamical modelling experiments after microbial processes, ferrous iron in hydroxide, sulphide, and carbonate forms were formed, and precipitation of goethite, pyrrhotite, siderite, troilite, and ferrihydrite mineral phases occurred56,57,58,59. These new sorption phases could cause additional actinide removal from solutions60,61,62,63,64. Thus, one of the reasons for a significant increase in the size of actinide-containing particles in model samples with the addition of iron and organic matter, as well as in samples of natural water, may be the formation of ferruginous minerals.
Prior to microbial treatment, U was expected to be present as di- and tricarbonate complexes. Np occurred as a neptunoyl ion or as a relatively poorly soluble hydroxo complex. Plutonium was expected to occur as sulfate and as a hydroxo complex. Microbial processes resulted in uranium remaining as a tricarbonate complex ore as a poorly soluble hydroxide. Plutonium and neptunium were present in all aquifers as oxyhydroxides, which can be attached to mineral surfaces and various hydroxyl phases65,66.
It is important to note that although thermodynamic modelling predicted the formation of soluble carbonate phases of uranium and neptunium after microbial processes, in fact, no significant increase in the contribution of soluble fractions was observed in the laboratory experiment. It should be added that the solution contains supersaturated calcium and magnesium carbonate phases (calcite, aragonite, dolomite), which can also be sorption phases for most actinides67. In addition, it is known that carbonate phases such as calcite are actively formed in biofilm exopolysaccharides, contributing to their biomineralization68. Thus, the formation of carbonate polysaccharide aggregations in the solution can also contribute to their sedimentation together with actinides.
Source: Ecology - nature.com
