Risk of colloidal and pseudo-colloidal transport of actinides in nitrate contaminated groundwater near a radioactive waste repository after bioremediation
Characteristics of environmental samples before and after bioremediationTable 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).Figure 1Percentage 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.Full size imageIn 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.Figure 2Electron 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).Full size imageAlthough 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.Figure 3Organic 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.Full size imageAn 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).Table 2 The intensity of light scattering (kHz) by suspended particles of different fractions before and after day 30 of the ongoing microbial process in the stratal water (Light scattering intensity was determined by Zetasizer Nano ZS, Malvern Panalytical).Full size tableIn 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 actinidesThe 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.Table 3 Polysaccharide (A) (mg/L) and protein (B) (mg/ml) concentrations in the model solutions during incubation. Polysaccharide determination was carried out by the phenol–sulfuric acid method according to Dubois 34. Protein content was measured with the Folin phenol reagent according to Lowry 35.Full size tableNo 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.Table 4 Hydrodynamic radii of colloidal particles during the experiment, nm (The measurement accuracy was at least 2%.).Full size tableIn 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.Figure 4Actinide 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).Full size imageIn 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 Bentley More
