Environmental significance of the interaction between titanium dioxides and soil solutions

Nanotechnology, especially in the field of photocatalysis, has witnessed rapid advancements, with titanium dioxide being one of the most widely used photocatalysts. As the use of products containing photoactive nanomaterials increases, concerns have arisen regarding their potential release into the environment over time. This release can impact soil, groundwater, and surrounding ecosystems, resulting in nanoparticles being dispersed in water and eventually depleted from the system. This study aimed to investigate how different soil solutions affect the structural, textural properties, and photocatalytic activity of titanium dioxide-based, commercial reference Evonik Aeroxide P25. The Regosol soil solution, characterized by acidic pH, low ionic content, and high organic matter content, induced nanoparticle aggregation and bandgap changes. In addition, the acidic pH hindered the adsorption process, potentially affecting the photocatalytic processes. In contrast, the Chernozem soil solution, with slightly alkaline pH, high ionic content, and low organic matter content, did not significantly alter the morphology or structure of the material. However, various organic compounds were absorbed on the surface, reducing the availability of active sites. The study highlights the importance of understanding the influence of soil solutions on nanomaterials, as it impacts their properties and environmental risks. Results show that the material is still activated, i.e., it can exert its photoactive effect on the environment. This sheds light on the challenges posed by nanoparticles in soil, particularly in terms of their toxicity and consequences for the surrounding ecosystems. The study underlines the need for further research in this area to assess potential risks and optimise the use of


Background
The widespread use of TiO 2 in nanotechnology can be attributed to its cost-effectiveness and photochemical stability [1].TiO 2 has a wide bandgap, enabling it to absorb ultraviolet (UV) light and generate electron-hole pairs efficiently.Electrons and holes can participate in reduction and oxidation reactions, facilitating the degradation of organic pollutants through direct oxidation or the generation of reactive free radicals, such as hydroxyl radical (•OH) [2].Due to its outstanding efficiency, TiO 2 nanoparticles (NPs) have gained widespread success in practical applications and are commercially produced on a large scale.Estimates suggest that nanosized TiO 2 will dominate the market by 2025, with a global production of 14 million tons [3,4].
A notable product in this category is Evonik's AEROXIDE ® TiO 2 P-25 (P25), the most widely used commercial TiO 2 for researching photocatalytic processes, and to develop self-cleaning surfaces [5,6].When exposed to UV light, P25 degrades organic compounds, in which is the basic procedure for a self-cleaning surface [7,8].This characteristic significantly reduces the need for manual cleaning, particularly in outdoor applications, making these surfaces highly advantageous.Common applications include self-cleaning glass in architecture [9], automotive coatings, and outdoor signage [10][11][12].
With the increasing use of products containing photoactive nanomaterials, their erosion and corrosion over time can result in their release into the soil [13][14][15].Naturally, TiO 2 NPs also can be present in soils and sediments due to the weathering and erosion of rocks containing titanium minerals [16].The increasing presence of TiO 2 NPs in soil is raising concerns, and identifying them poses challenges due to limitations in extraction and separation methods [17].The expected concentration of TiO 2 NPs in soil is ~ 50 mg•kg −1 in the European Union [18].In the United States, sewage sludge has shown concentrations ranging from 1.8 to 6.4 mg•kg −1 of TiO 2 NPs, while in the UK, the concentration was found to be 305 mg•kg −1 [19,20].Moreover, various models have been developed to forecast the fate of TiO 2 NPs in the environment, and the outcomes consistently indicate a rise in their concentration over time [13,21,22].
In the process of entering the soil, rainfall plays a crucial role by actively interacting with the soil, facilitating the removal of TiO 2 NPs from various surfaces [23][24][25].The three-phase soil dynamics occur on the surface, where TiO 2 NPs either bind to the solid phase of the soil or disperse, becoming part of the soil solution [26].However, the interaction of TiO 2 NPs with soil solutions extends beyond surface processes.Leaching serves as another significant pathway, where water transports NPs into soil pores, especially when they are not strongly attached to soil particles [27,28].This underscores the potential for TiO 2 NPs to move and redistribute within the soil environment, influenced by environmental factors, such as rainfall and leaching processes.What does not bind to the surface is subsequently washed down into the deeper layers, ultimately reaching the groundwater [29,30].
Previous studies have shed light on the behavior NPs in soil solutions and the associated mechanisms.For instance, a prior investigation shows that the enhanced stability of suspended TiO 2 concentrations, attributed to factors, such as a higher zeta potential, promotes their heightened mobility through diverse layers of the soil [31].In the case of Ag NPs, their stability is influenced by the sorption of solutes, such as Ca 2+ from the soil solution, where dissolved organic matter (DOM) exhibits a concentration-dependent stabilizing effect [26,32].Furthermore, the interaction between CuO NPs and DOM highlights that hydrophobic DOM stabilizes CuO NPs more effectively than hydrophilic DOM in the presence of Ca 2+ [28,33,34].
Soils differ in composition, texture, and genetic processes.For example, the Chernozem soil (mainly found in Eastern Europe) has been characterized by the World Reference Base of Soil Resources with a mollic horizon (dark in color with an organic matter content of at least 0.6%), a depth of occurrence of at least 20 cm, and a texture of sandy-loam or loamy.Due to the organic matter content and the texture, the biological activity of Chernozem soils is high, so their crop yield is favorable too.Therefore, this soil type is usually utilized as arable land [35].Another example is the Regosol soil type, covering approximately 2% of the continental land area on Earth, mainly including the hill slopes [36] in Northern China, the Middle East, and Eastern Europe.Regosol soils are found under natural vegetation and crops and can be characterized by high organic matter content (> 2%), sandy texture, and reddish brown color.Based on the information provided above and considering their abundance in Europe, these two soil types were chosen for the investigations.
While existing research has primarily concentrated on the impact of NPs in aquatic environments, there is a recognized gap in our understanding of their effects on soil solutions.This study contributes to this area by underscoring the importance of investigating the relationship between NPs and soil solutions.The aim is to provide valuable insights into the environmental risks associated with NPs, particularly addressing the limited knowledge about how the physicochemical properties of soil solutions influence the catalytic activity of TiO 2 NPs.Further exploration in this field is essential for developing a more comprehensive understanding of the environmental implications of these NPs.
Therefore, the aim of this study is to assess the impact of different soil solutions on the structural properties of the widely used commercial reference material, Evonik Aeroxide P25 TiO 2 , with a specific focus on environmental protection considerations.Two distinct artificially prepared soil solutions, designed to replicate the characteristics of Chernozem and Regosol soils, were employed in the investigation.The study examines the interaction between these soil solutions and Evonik Aeroxide P25, emphasizing changes in photocatalytic activity as well as several structural and textural characteristics, including morphology, crystalline composition, surface features, and optical properties.By concentrating on the environmental implications, the research aims to contribute valuable insights into how TiO 2 materials behave in different environmental conditions.

Materials
The following TiO 2 NPs were used during the experiments: Evonik Aeroxide P25 (anatase:rutile = 89%:11%) was purchased from VWR, Hungary, and was used in all the experiments without prior treatment or modification.Ultra-pure Millipore Milli-Q (MQ) water was used for preparing the soil solutions.To determine the photocatalytic activities, phenol (VWR, > 99%) was used as a model pollutant.
KBr (99.0%,Millipore Sigma, Uvasol) was used to prepare the samples for infrared spectroscopy (IR).0.1 mM and 0.01 mM NaOH and HCl were used to adjust the pH before the zeta (ζ) potential measurements.For the determination of the organic matter content (OM) of the soils, H 2 SO 4 (95%, VWR) and 0.33 M K 2 Cr 2 O 7 (VWR) were used.
For the flow injection analysis (FIA) measurements, NH 4 Cl buffer (pH = 8.5) was used as a reagent to determine NO 3 − , for PO 4 3− determination, SnCl 2 and (NH 4 ) 6 Mo 7 O 24 were used.For the ion chromatography (IC) measurement, 0.5 M Na 2 CO 3 was used as the eluent composition.

Soil sampling and analysis
Soil samples (10 Regosol and 10 Chernozem topsoil samples) were collected (Fig. 1) from an agricultural land near Szeged, Hungary and from a forest near Tallya, Hungary.The soil samples were taken from 0 to 20 cm depths in September 2021 and in January 2022.Sampling was performed by mixing topsoil samples from 0 to 10 cm and 10 to 20 cm layers.The soil samples were dried in air for 10 days and sieved with a 2 mm diameter sieve.Primary soil parameters such as pH, electrical conductivity (EC), total salt content, texture, and OM content were determined following the Hungarian standards (MSZ), and can be found in Supplementary information.

Soil solution extraction and analysis
Soil solutions were obtained using the methodology developed by Klitzke et al. [32] and Qiu et al. [33].The soil solutions were prepared using Milli-Q water, applying a solid-to-solution ratio of 1:2.5.The mixtures were shaken with a tube rotator (40 rpm) for 18 h, centrifuged at 3700 rpm for 30 min and subsequently filtered with a 0.45 μm cellulose nitrate membrane.
The pH of the soil solution was measured with a digital pH meter (Inolab pH 720), while the EC was determined using a conductivity meter (Orion 3-Star, Thermo Electron Corporation).COD values were determined following the Hungarian Standards (details in the Supplementary information).

Characterization of nanomaterials
X-ray diffraction (XRD) patterns were recorded with a Rigaku Miniflex II diffractometer equipped with a graphite monochromator using Cu-Kα radiation (λ = 1.5406Å).Data points were taken in the 2θ = 20-40° range at a scan speed of 1 min -1 .Primary crystallite sizes were calculated using the Scherrer equation [37].
The morphology of the samples was characterized with a Hitachi S-4700 Type II scanning electron microscope (SEM) after coating them with gold NPs (< 10 nm).
An IR spectrometer (Bruker Equinox 55) was used to examine the changes on the surface of TiO 2 NPs.The samples and KBr powder were pressed into pellets, and the IR spectra were recorded with a resolution of 2 cm −1 .To determine the bandgap values and also the color changes following the interaction with soil extracts JASCO-V650 spectrophotometer with an integration sphere (ILV-724) was used to record the diffuse reflectance spectra (DRS) of the samples (λ = 220-800 nm).BaSO 4 was used as a reference, and the bandgap energy was calculated based on the Kubelka-Munk method [38].
The ζ potential of the NPs was determined with a Horiba SZ-100 Nanoparticle Analyzer (Retsch Technology GmbH, Germany).The measurements were performed in a carbon electrode cell.The ζ potential values were determined using the Smoluchowski model.The measured dispersion concentration was 0.001 w/v%.
The photocatalytic activity of P25 NPs was evaluated by phenol degradation experiments.P25 NPs (1 g•L -1 ) were added to a phenol solution (c 0 = 0.1 mM), and the suspension was sonicated for 10 min.Then, the suspension was poured into a double-walled glass vessel (V = 100 mL) surrounded by six fluorescent UV tubes (Vilber-Lourmat T-6L UV-A, 6W).To reach adsorption-desorption equilibrium, we stirred the suspension in the dark for 10 min before switching on the lamps.The concentration of phenol was followed by an HPLC system consisting of a Merck Hitachi L-4250 UV-Vis detector and a Merck Hitachi L-7100 low-pressure pump (0.7 cm 3 •min −1 flow rate; detection at 210 nm) using a 50:50 (v/v) methanol/ water mixture as an eluent.

Soil solution experiments
The ten Chernozem and ten Regosol pre-prepared soil solutions were merged, and throughout the experiments, a singular Chernozem and a Regosol representative homogeneous sample was employed.The P25 NPs stock suspension (10 g•L −1 ) was prepared using Milli-Q water under ultrasonication (10 min).The final concentration of the P25 NPs in the soil solutions was 1 g•L −1 after blending.The established suspension concentration was determined based on experimental reproducibility and typical catalytic applications [6,7].Subsequent to adding the P25 stock suspension to the soil solution, the resultant suspension underwent mixing with a magnetic stirrer under light protection.Following a 1-, 4-and 24-h interaction between P25 NPs and the soil solutions, the suspensions underwent centrifugation for 10 min at 3700 RPM (1531 RCF).A schematic representation of the steps of the methodology is shown in Additional file 1: Fig. S1.

Evonik aeroxide P25 characterization before immersion in soil solution
Before immersion in soil solutions, the properties of P25 was investigated.The SEM micrograph shows that the P25 sample contained particles of different shapes due to its polycrystalline structure (Fig. 2a).The XRD pattern shows that it contained anatase and rutile phases based on the diffraction peaks at 25.6° and 38.0° for the former and at 27.2° and 36.1° for the latter (Fig. 2b).The average crystallite size of P25 was calculated to be 24.8 nm based on the Scherrer equation.Due to their diminutive size, small particles exhibit ease of suspension in aqueous media [39].In addition, this characteristic also implies a propensity for these particles to diffuse into cell membranes, potentially causing harm to living organisms [40][41][42][43].The excitability of the material, such as P25, is determined by its bandgap, which, in the case of P25, is in the UV region, specifically in the UV-A range, and sunlight contains a significant amount of UV-A radiation.Based on the DR spectrum of P25 (Fig. 2c), we found that the bandgap was 3.15 eV, which is consistent with literature data [44].Lastly, the degradation of phenol is highly efficient, even at a concentration as low as 0.1 mM, as depicted in its overall properties (Fig. 2d).This implies its capability to degrade any organic matter, potentially leading to a significant local concentration change upon release into the environment.This high catalytic activity can be attributed to the (i) synergistic effect between the anatase and rutile phases and their different surface energies [45], (ii) relatively high specific surface area, resulting in many active sites on the surface [46], and (iii) efficient excitability [47].

Properties of soil solutions and their possible effects on the behavior of TiO 2 NPs
The pH, IS, COD, and inorganic ion content of regosol and chernozem soil solutions were investigated (Table 1, Additional file 1: Tables S2, S3).
The pH values of the soil solution were different: the average pH of the Regosol soil solution (REG) was ~ 4.95  (acidic), while that of the Chernozem soil solution (CH) was ~ 7.74 (slightly alkaline).It is known that pH significantly influences the isoelectric point (IEP) and ζ potential of catalysts.Significant differences were observed between the COD concentrations of the soil solutions.The COD of the CH was 20.9 mg•L -1 , while for the REG, a much higher value (165 mg•L -1 ) was measured.The significant differences between pH and COD values are due to areal characteristics.Regosol soil samples (the basis of REG) were collected from forest areas with high biomass turnover and soil acidification by microbial decomposition products.COD values also represent the organic matter content of aqueous media [48].The high organic matter content strongly affects the photocatalytic activity of P25, because various organic compounds can adsorb onto its surface, hindering access to its active sites [49].This impedes the interaction between the photocatalyst and the target pollutants or substrates.In addition, organic molecules may compete with the target pollutants for adsorption onto the P25 surface.If the organic molecules have a higher affinity for the surface, they can effectively displace the target pollutants, diminishing the photocatalytic activity towards those pollutants [50].
The behavior of the P25 photocatalyst can be affected the presence of specific ions too, so their concentrations in soil solutions were measured too.Na + and Ca 2+ concentrations were low for REG and high for CH.Various authors reported increased sedimentation and aggregation at similar values for TiO 2 NPs in water [39,51,52].The K + and Mg 2+ concentrations in the REG were much higher than in the CH, which could be due to the geochemical characteristics (e.g.bedrock) the area from which the Regosol soil samples were collected.
Differences in the anion concentrations were also significant.The concentration of NO 3 -in the REG was 21.55 mg•L -1 , while in the CH, a much higher value (74.61 mg•L -1 ) was measured.Based on literature data, even 50 mg•L -1 NO 3 -the ζ potential of TiO 2 exhibited a gradual decrease, eventually nearing zero, leading to a subsequent increase in aggregate size [53].Similar PO 4 3concentrations were measured for the two types of soil solutions.A previous study shows, that even at millimolar concentrations, PO 4 3-decreases the activity of TiO 2, because it strongly adsorbs onto the TiO 2 surface [54].The differences were significant for NH 4 + values of CH (0.33 mg•L -1 ) and REG (1.39 mg•L -1 ) samples.The higher value for the latter can be associated with the intensive microbial activity in the area from which the soil samples were collected [55].The NH 3 content has similar effect to the TiO 2 activity as PO 4 3- [56].Finally, F -, Cl -, and SO 4 2-ions were considered.Significant differences were observed for the first two, while for SO 4 2-, the difference was not as notable.These ions can also bind to the active sites of catalysts, blocking access for reactant molecules and thus hindering photocatalytic reactions [57][58][59].However, it is crucial to acknowledge that the effect of SO 4 2-content on activity is highly dependent on the specific synthesis method employed.Different methods can lead to variations in the material's structure, composition, and properties, which in turn affect its reactivity or and can increase the photocatalytic activity of the material [60]

Morphology, crystal structure, particle size, and optical properties of P25 after immersion in soil solution
As the next step, the effect of soil solutions on the properties of P25 TiO 2 was investigated.Based on the XRD patterns in Fig. 3, the crystal structure of P25 did not change significantly (there were no changes in the intensity and width of the peaks) after its immersion in the soil solutions.The stability of the crystal structure suggests that if P25 NPs are released into the environment, they are likely to retain their properties and key applicability.SEM micrographs show that the morphology of NPs did not change; however, the aggregation of NPs can be observed: in the case of REG already after 1 h of interaction, and in the case of CH after 24-h interaction time (Fig. 4).This can be correlated to the organic matter content of REG, because various organic compounds with different ions can adsorb to the surface, which induces aggregation.Due to this, the particles sediment faster [61,62].
Optical changes were also observed following the interaction of P25 with CH and REG. Figure 5 shows significant spectral differences between the samples in the visible region (differently colored samples).This can indicate potential adsorption of reflectance onto the surface and alterations in attached materials affecting both color and semiconductor properties (Fig. 5a, b).However, it is important to note that even if the color changes, that may not result in significant changes in its photocatalytic properties [63].
Figure 5c summarizes the bandgap values of the P25 samples.The bandgaps decreased (suffered a red shift) by 0.15 eV and 0.5 eV after interaction with CH and REG, respectively.However, the bandgaps within the soil solution types did not show significant variations over different interaction times, with a maximum difference of 0.12 eV.Interestingly, minimal increases were observed for almost all samples after phenol degradation, compared to those interacting with soil solutions (e.g., P25_ CH_4 has 3.0 eV, while P25_CH4_AP has 3.1 eV).
We excluded the possibility that the ions in the soil solutions caused the changes in the bandgaps.Although transition metals (e.g., Fe 3+ ) can cause such changes, this Fig. 3 XRD patterns of P25 samples before and after interaction with soil solutions Fig. 4 SEM micrographs of P25 after interaction with the soil solutions happens only through incorporation into the crystal lattice (doping) [64].However, such a process cannot occur during adsorption, as it requires either a heat treatment or a hydrothermal process [65].In contrast, various organic compounds, such as humic acids or conjugated double-bond-containing systems, can accept electrons (together with catalysts), causing bandgap changes [63].Therefore, the significant color changes may have been caused by the surface-anchored organic compounds originating from the soil solutions, making the investigation of surface properties key to the present work.

Surface properties of P25
Changes in the surface chemistry of P25 NPs were investigated by ζ potential measurements after their interaction with the soil solutions.This is because ζ potential plays a crucial role in the interaction of catalysts with each other or with other components (e.g., ions, organic compounds).ζ potential refers to the electrical charges on the surface of NPs in a solution, influencing stability and aggregation [26].The ζ potential and IEP of P25 (Fig. 6) was investigated in REG, in CH, and, for comparability, in distilled water (P25_REF).The IEP refers to the pH at zero net total particle surface charge [66].
In distilled water, the IEP of P25 was ~ 5.65 pH units.However, the material's IEP in both soil solutions decreased to pH 2. This decrease is significant and indicates a substantial shift in the surface charge of the NPs.This may suggest, that a variety of dissolved ions, such as cations (Na + or Ca 2+ ) or organic species are present, which can adsorb onto the surface of the NPs and alter their surface charge.However, it is also possible that carboxyl groups from humic acids are adsorbed, which later on may suffer a deprotonation process.The reason why the IEP did not change differently in the two soil solutions despite their differences in pH can be attributed to several factors.(i) The surface chemistry and composition of Evonik Aeroxide P25 NPs are primarily responsible for determining their isoelectric point.It is possible that the functional groups on the nanoparticle surface and their affinity to ionization remained consistent across the two soil solutions, leading to a similar shift in the IEP.(ii) Both acidic and slightly alkaline soil solutions can contain species that can protonate or deprotonate the functional groups on the nanoparticle surface, thereby affecting the surface charge.These species may be present in both soil solutions, leading to a similar change in the IEP.
Investigating the surface chemistry of nanomaterials in the aquatic environment using IR spectroscopy can provide valuable insights into their molecular interactions and surface functional groups.It allows the identification and characterization of adsorbed species, such as organic compounds or ions, which can influence the behavior of NPs.
As the next step, the IR spectra of the samples were recorded (Fig. 7, Additional file 1: Figs.S1, S2).Before analysis, it is important to clarify that the interaction time with the soil solutions did not influence the adsorption of various substances on the surface of the P25 NPs.Henceforth in the analysis, we will refer to the specific groups of samples as follows: those interacting with Chernozem soil solutions as P25_CH and P25_CH_AP, and those interacting with Regosols as P25_REG and P25_REG_AP.
Vibrations of Ti-O bonds in the 400-900 cm -1 region were detected.Since the P25 samples were not oxidized/ reduced (due to the lack of conditions that enable such reactions), the IR spectra were normalized to the M-O (400-900 cm −1 ) region.
Vibrations representative of CO 2 bonds at 2350 cm -1 was observed in all samples, originating from the sample preparation process and irrelevant to the analysis [67].
Bands at 1650 cm -1 were attributed to O-H bonds.This band was present not only in the samples exposed to soil solutions but in P25_REF too, which can be attributed to the presence of physisorbed water [68].The intensity of this region varied in conjunction with that of the 3000-3500 cm -1 region ("water band") [69].The intensity of the previous two regions did not change in parallel in our samples.If the intensity of these two regions does not increase in parallel, it suggests that a compound with additional O-H groups has been adsorbed onto the surface.In the P25_AP, P25_REG_AP, and P25_CH_AP samples, the significant changes between these regions could be attributed to the presence of phenol intermediates (e.g., hydroquinone, resorcinol, pyrocatechol) originating from the degradation process [7,70,71].
The bands identified so far were observed in each sample; however, differences in other regions were also observed and analyzed.In the 1600-1700 cm -1 region, bands with high intensities were observed in the P25_ REG and P25_CH samples.These bands can be attributed to C=O and C=C vibrations originating from carboxyl groups (e.g., humic acids).In addition, N-H bonds were also detected in the 1580-1650 cm -1 region, Fig. 7 IR spectra of the samples after 4-h interaction with the soil solutions in the 900-3900 cm -1 region which may be attributed to different proteins/peptides from residues of animal and plant origins [72].P25_CH, P25_REG, and P25_REG_AP had highly intensive bands in the 1000-1100 cm -1 region.These can mainly be attributed to Si-O bonds (1070-1030 cm -1 ), since silicon oxides can be found in high amounts in the soil as inorganic compounds of different minerals [72].In the same samples, C-O vibrations were also identified in the 1080-1100 cm -1 and 1250-1310 cm -1 regions, which can be ascribed to long carbon-chain molecules in soil solutions (e.g., lignin-based substances from the decomposition of plants [73]).
In the 1310-1380 cm -1 region, O-H bonds were observed in the P25_CH_AP, P25_REG_AP, and P25_ AP samples, corresponding to the phenol degradation experiments.Another type of O-H bonds in the 1400-1420 cm -1 region could also be found in each sample, apart from P25_REF.For the P25_CH and P25_REG samples, this band may originate from the soil solutions (e.g., humic acid) [72], while in the other samples, this band may originate from the phenol degradation intermediates.Moreover, in the 1380-1390 cm -1 region, C-H bonds were identified, corresponding most likely to the aldehyde and alkaline groups of lignin in P25_CH and P25_REG [74].However, it is worth noting that properly evaluating this region is challenging, because several vibrational bands overlap with each other.In P25_AP and P25_REG_AP, the intensive C=O bands in the 1690-1700 cm -1 region (aldehydes) also result from the phenol degradation experiments.Furthermore, intensive C=O bands were observed in the 1750-1820 cm -1 region in all samples except for P25_REF and P25_CH_AP [26].Same as before, in this region, many vibrational bands overlap (e.g., aldehydes, carboxyl acids, esters, and anhydrides), which could originate from the phenol degradation process and the soil solutions.In the 2860-3000 cm -1 region, vibrations of C-H bonds were found with high intensity for P25_REG and P25_AP and with low intensity for P25_ CH and P25_REG_AP [26].

Evaluation of photocatalytic activity after immersion in soil solution
The photocatalytic activity of P25 before and after its interaction with soil solutions was investigated via phenol degradation under UV-A light irradiation for 180 min.Previous adsorption studies have confirmed that the adsorption of phenol on P25 is negligible (~ 2-4%), so the adsorption-desorption equilibrium was reached in under 10 min.
As shown in Fig. 8, P25_REF degraded phenol completely after ~ 90 min.After its interaction with the CH, the activity decreased, and complete decomposition only occurred after 150 min regardless of the interaction time.
In contrast, after interacting with REG, the photocatalytic activity decreased drastically.For instance, in the case of P25_REG_1, only ~ 55% conversion was achieved after 180 min, while for P25_REG_4 and P25_REG_24, it was only about 60% and 62% respectively.It is clear that the cause for the decreasing photocatalytic activities must be related to the chemical compounds in the soil solutions.As discussed in the previous sections, P25 NPs underwent several changes after interacting with soil solutions.
In the P25_CH samples, no changes were observed in its morphological, structural, and semiconductive properties.However, as discussed in Sect."Surface properties of P25", several organic compounds were adsorbed on the surface of the samples (e.g., carboxyl groups from humic acids, amines from pectines).Such organic compounds adsorbed on the surface can strongly influence photocatalytic activity in several ways.They can act as a physical barrier, reducing the surface area available for catalytic reactions by occupying active sites and undergoing degradation themselves instead of the target pollutant [49].In addition, they can act as electron traps, decreasing the overall efficiency of photocatalysts.This effect is particularly significant for polar compounds, such as hydroxyl and carboxyl groups, which tend to adsorb onto the active surface of catalysts [75].In addition, the CH solution had high F -, Cl -, and NO 3 -contents, and these ions can adsorb to the surface of P25, blocking active sites.F -ions can form surface complexes with TiO 2 , leading to surface passivation.This can reduce the availability of active sites, decrease the surface area available for adsorption, and decrease photocatalytic activity [55,57,59,76,77].However, Na + concentration in the soil solution was also high, which can positively affect the activity of P25.Na + can facilitate the adsorption of reactant molecules on the catalyst surface, promoting the availability of active sites for catalytic reactions [76].
After the interaction of P25 with the REG, aggregation, and bandgap changes were observed.These changes can be attributed to the high organic matter content of the REG (the REG had about eight times higher COD than the CH, as shown in Table 1).This statement is supported by the results presented in Sect."Surface properties of P25", where it was described that several organic compounds were indeed adsorbed on the surface of the samples.However, as shown in Fig. 7, the intensity of each region in the IR spectrum is significantly higher than that after interaction with the CH.Besides the organic matter content, the acidic pH of the REG is important to consider.The positive surface charge of P25 in this pH range can influence the adsorption of various ions and consequently impede the accessibility of active sites [61].This charge-related effect may contribute to the hindrance of adsorption and affect the overall efficiency of photocatalytic processes [78].

Conclusion
This study focused on investigating the properties of two different typed soil solutions and their effects on P25 TiO 2 NPs.Both soil solution induced aggregation of P25, reducing stability and promoting sedimentation.Optical changes were observed, but they may not significantly affect the photocatalytic properties of P25.The high organic matter content strongly affected the photocatalytic activity of P25, as organic compounds adsorbed onto its surface, hindering access to active sites.The investigation of inorganic ions in the soil solutions revealed notable differences in their concentrations, which influenced the ζ potential and aggregation of TiO 2 NPs.The evaluation of photocatalytic activity demonstrated that the presence of chemical compounds in the soil solutions affected the efficiency of P25.Organic compounds adsorbed on the surface acted as physical barriers and electron traps, reducing the availability of active sites and overall efficiency.In addition, inorganic ions could block active sites and decrease photocatalytic activity.Consideration of surface properties and the presence of various adsorbed compounds is essential for comprehending and optimizing the photocatalytic activity of NPs in soil environments.Our findings indicate that the duration of interaction did not exert a significant influence on the properties of TiO 2 NPs.However, chemical parameters, particularly the organic matter content, demonstrated a notable impact.This suggests that the behavior and characteristics of TiO 2 NPs are more strongly governed by the chemical composition of the environment they interact with rather than the duration of interaction.This highlights the stability of TiO 2 in the environment, emphasizing that its properties are more dependent on the chemical composition of the surroundings rather than the duration of interaction.Presently, despite their low concentrations, these materials demonstrate a high accumulation potential [79].Most studies maintain TiO 2 , including P25, to be inert and non-toxic upon environmental release.However, experimental results suggest an initial inhibition of TiO 2 activity, followed by an improvement over time.Notably, Chernozem exhibited comparable degradation efficiency to pure/reference P25 after interacting with soil solution.It is important to note that the best catalysts are UV-active and have a much higher quantum efficiency than visible light.At 3% UV light, UV catalysts can achieve a much higher degradation efficiency than a visible light-active catalyst in sunlight [80,81].This could potentially pose a significant environmental and ecological risk, as it continues to degrade the surrounding organic matter once released into the environment, consequently altering the chemical composition of the different environmental matrices, focused on the topsoil layers.Research has primarily focused on understanding how chemical parameters in soil/soil solution affect substance behavior.Future investigations will delve into reusability, and the influence of physical soil factors (e.g., texture) on TiO 2 behavior to enhance our comprehension of its environmental applicability and potential risks.

Fig. 1 3 −
Fig. 1 Location of the sampling areas

Fig. 2 a
Fig. 2 a SEM-micrograph, b XRD-pattern, c DR spectrum (with bandgap evaluation), and d phenol degradation curve of P25 TiO 2

Fig. 5
Fig. 5 DR spectra of the P25 samples and color changes before and after 4-h interaction with soil solutions: a Regosol soil solution, b Chernozem soil solution, c bandgap values of the samples

Table 1
Physicochemical parameters of soil solutions