Recently, nano-Mn3O4 was applied to leaves by spraying a suspension of 20 mg/L, after 20 days from the sowing of the seeds of squash plants until 40 days from sowing,improving the growth plant yield but decreasing the yield of fruit. Another study reported that nano-Mn2O3 applied at 6 mg/kg/plant promoted nutrient fixation in wheat, but it was found that manganese could affect plants in subtle ways when applied in soil rather than via foliar.Although effects have been noted when these NMs are applied in the soil, the mechanisms related to the effect and transport of nano-MoO3 and nano-Mn3O4 in RE have not been studied in detail. Therefore, studies about aggregation and dissolution of nanoMoO3 and nano-Mn3O4 in RE can contribute to understand the interactions between NMs and roots. In this work, we characterized and determined the aggregation, stability and dissolution, in soybean root exudates, of four nanoparticles proposed to be used as nanofertilizers or nanopesticides: nano-CeO2, nanoMn3O4, nano-Cu2 and nano-MoO3. We also studied the early aggregation of these NPs in RE via single particle inductively coupled plasma mass spectrometry . This study also aimed to determine whether the use of these nanoenabled agricultural products may result in the environmental buildup of metal ions or nanoparticles in the roots-soil interface.Pregerminated seedlings were grown in vermiculite in glass jars until the full expansion of the first true leaf . Then, the plants were transplanted to vermiculite in culture boxes. Hoagland nutrient solution at 10% was supplemented to the plants every alternate day. Pregermination and culturing occurred at 20 °C in an environmental growth chamber with a 16 h photo period. After 28−30 days of culturing, the soybean seedlings were removed from the culture box,mobile vertical grow racks and the roots were washed thoroughly with NANOpure water for RE collection. The procedure for collection of RE was modified from the protocol described by Zhao et al. .

Briefly, the seedlings were placed for 12 h in new metal free tubes containing 40 mL of autoclaved 0.1 mM CaCl2 as the solution to collect RE. This solution was aerated continuously during the 12 h period of collection. After 12 h, the 0.1 mM CaCl2 solution with secreted metabolites was filtered using a syringe filter to remove larger particles and microorganisms.Analysis of the soybean RE composition was performed in triplicate via liquid chromatography coupled to triple quadrupole mass spectrometry , based on the methods developed in previous studies.The main characteristics of the NPs considered in this study are presented in Table 1. Plots are presented in the Supporting Information. The morphology and primary size of NPs were different but all in the nanoscale . As can be observed, Mn3O4 and MoO3 NPs were generally spherical. CeO2 and Cu2 NPs were mostly nanorods and nanobars. Although aggregates were observed for all NPs, the primary size of NPs, for example, diameter, was smaller than 60−70 nm. The nanorods of CeO2 and Cu2 had mean diameters of 8 and 50 nm respectively, while the average diameters of Mn3O4 and MoO3 nanospheres were roughly 40 and 60 nm. Determination of the LSPR helps to understand the photocatalytic activity of NPs. For NPs of CeO2, Cu2, and MoO3, the surface plasmon resonance peaks were localized in the ultraviolet B region at 311, 293, and 291 nm, respectively . For Mn3O4, the LSPR was in the visible region with a peak at 511 nm, where photocatalytic activity of this material can occur. The XPS spectra of the four NPs are presented in SI Figures S2 and S3. In the four graphs, a strong peak is observed at 529−530 eV, which corresponds to O 1s and this indicates that oxygen ions associated with an oxidation state 2− are bound to transition metals,such as those used in our experiments. For CeO2 NPs, a primary peak for Ce 3d was observed at 883 eV, confirming the presence of Ce in the binding energy Ce 3d5/2. The primary peaks observed for Mn3O4 NPs were in the region Mn 2p3/2, with binding energy that corresponds to the presence of MnO and Mn2O3 and whose oxidation states are Mn and.

For Cu2 NPs, the binding energy was measured in the region Cu 2p3/2 at 933.5 eV and corresponds to Cu .In the case of MoO3 NPs, the primary peak was observed in the region Mo 3d3/2 at 233 eV and correlates with the oxidation state Mo .The binding energies of these materials indicated they are bound to oxygen in the upper energy levels, and to hydroxide in the case of Cu2 NPs. Moreover, there were no carbon-based coatings added to the NPs used in the experiments since there were no peaks in the small region 284−288 eV in all four NPs. Based on the XRD analyses , Cu2 and MoO3 exhibited orthorhombic crystalline structure, and MoO3NPs were associated with the mineral phase alpha . For CeO2 NPs, the XRD spectra shows strong peaks at , and that are consistent with literature for a cubic crystalline structure, in the well-known phase of ceria. Lastly, the XRD pattern of Mn3O4 was associated with a tetragonal structure in databases and corresponds to hausmannite phase. This phase is known to have manganese as Mn2+ and Mn3+, which was confirmed in the XPS analysis, and to be a paramagnetic mineral.However, in contact with RE and soil leachate the charge is generally negative for these NPs. Moreover, the charge of all four NPs when exposed to RE and soil leachate was almost independent of NM composition. In RE, the charge was from −28.8 mV to -34.2 mV at the pH of 6.8. In soil leachate , the charged ranged between −30.5 mV and −42.1 mV. The charge in the surface has been related to the electrolytes and organic molecules present in the media.Hence, detailed analysis of electrolytes and composition of organic matter in soil and exudates of agricultural setting will contribute to better understanding of the final charge of nanoenabled products.Soil leachate and soybean RE were analyzed for organic acids. From a total of 15 organic acids analyzed, only 3 were found in the soil leachate and 5 in the RE . Malic acid was the most abundant in the soil leachate, with lower concentration of fumaric acid and gallic acid . In the case of soybean RE, ascorbic acid had the highest concentration, followed by glutaric , salicylic , malic , and ferulic acids.

The concentration of these organic acids was low compared to the concentration of the NPs. The positive charge of metal NPs can be neutralized by trace amounts of organic acids in suspensions with pH lower than 8,as we observed for CeO2 and Cu2 NPs; while the charge of Mn3O4 and MoO3 NPs did not change substantially . The charge of the CeO2 and Cu2 NPs became more negative than the point of zero charge ,vertical garden growing which can be due to the presence of other organic ligands not analyzed such as humic acids and amino acids, among others. Since these organic ligands have a negative charge at pH range of 6−8,their presence in both media, soil leachate and RE, may contribute to maintaining Mn3O4 and MoO3 NPs negatively charged. The modulation of the charge of the NPs due to the presence of organic ligands in soil and in the roots zone can increase the mobility and the dissolution of metal based NPs;hence, further research is needed for systematic evaluation of low molecular weight organic acids and amino acids influence the mobility and transformations of NPs in the root-soil interface.All four NPs aggregate in DI water with time , as noted from the first few hours, although there were clear differences at initial state of aggregation. This was expected for Mn3O4 and Cu2 given their ZP close to PZC at the experimental pH, which indicates weak electrostatic repulsions between particles and leading to the aggregation; but unexpected for CeO2 and MoO3 NPs given their larger positive and negative ZP, respectively. For these cases, although CeO2 and MoO3 NPs form smaller aggregates and generally aggregate at a slower rate, the relatively high concentration of both NPs may explain their aggregation over time. Since aggregation behavior is concentration-dependent, these results may be different if the studies are performed at much lower or higher concentrations. The concentrations employed in this study are within the range expected to be used in nanoagrochemical applications. After the aggregation experiments it became clear that 5 days were sufficiently long for colloidal systems containing Cu2, CeO2 and MoO3 NPs to became stable; however, the system containing Mn3O4NPs could require more time to reach colloidal stability due to aggregation seems to be ongoing along 5 days. From a nanoagrochemical application, the relatively fast aggregation of these bare NPs could result in significant deposition of the particles, and less bio-availability.The aggregation of NPs exposed to RE varied significantly . For Cu2 and MoO3 NPs, the aggregate size increased over the 5 days, albeit at different rates, from 518 ± 43 nm to 938 ± 32 nm, and from 372 ± 14 nm to 690 ± 65 nm, respectively. Conversely, CeO2 and Mn3O4 NPs disaggregated in RE over time, decreasing from 289 ± 5 nm to 129 ± 10 nm, and from 761 ± 58 nm to 143 ± 18 nm, respectively.

Some of the observed disaggregation can be attributed to the presence of organic ligands, such as detected organic acids, in RE. Moreover, the size of the aggregates decreased substantially for both NPs during the first 24 h, indicating that the larger aggregates were likely soft agglomerates, held together by weaker attractive physical interactions such as van-der-Waals or hydrogen bridge forces.Due to this, the size distribution of all NPs in the early period was analyzed via spICP-MS and presented in a following section. However, a period of 5 days was insufficient to reach colloidal stability for all NPs in RE, since they continued to aggregate or disaggregate. Since NPs with potential use as nanoagrochemicals can reach the roots and soil organisms and the size of clusters or aggregates is an important factor in the interaction at the nano/bio interface,it is important characterize the size of the NPs in RE, which is a representative matrix of the intended application, rather than determining their size in DI water. For example, aggregate size has been shown to influence the apoplastic transport of CeO2 and CuO NPs. In soil leachate , all the NPs disaggregated over 5 days. Similar to the results in RE, the disaggregation was more noticeable in the first 24 h. In general, the rate of disaggregation decreased substantially after 3 days and the size of aggregates generally stabilized after that time. However, in the case of Cu2 and MoO3 NPs, more time would be needed to reach colloidal stability of the systems. The availability of organic ligands in soil leachate that can be adsorbed onto particle surfaces likely interferes with the weak aggregation, providing a barrier to aggregation.The size distribution of these NPs in RE was measured via spICP-MS for 6 h, at 2 h intervals . For CeO2 NPs, although the change in size distribution was not as substantial as for the other NPs, there was a clear shift to smaller sizes after first 6 h compared to immediately after spiking , with single particles as small as 9 nm . For Mn3O4 NPs, the shift to smaller sizes was more significant, and the size distribution became narrower with time . Conversely, for Cu2 NPs the size distribution shifted to larger particles with time, and the distribution broadened and became multi-modal. For MoO3 NPs the mean size increased over time, and the size distribution became wider with time, but the shape of the distribution was similar to the original one at spiking. It is important to note that for the spICP-MS analysis, the shape of the NPs was assumed to be spherical, since that is the only option in the current algorithm. Disaggregation in the presence of soybean RE occurred in NPs with low rates of dissolution: CeO2 and Mn3O4 NPs. The spICP-MS size distributions correlated well with the observed decrease in aggregate size of these NPs in RE as measured by the Zetasizer . This can be attributed to the binding of negatively charged organic acids, which can interfere with the weak attractive forces.In addition, this is the first time that Mn3O4 NPs stability in soil leachate and RE is reported; presenting low rates of dissolution and smaller aggregates size over time in both media.