1 Introduction

Biochar, the man-made charcoal, is produced from pyrolyzed biomass, and has been used for various environmental and agricultural purposes, including contaminant removal from wastewater, carbon sequestration, and soil remediation (Chen et al. 2022; Lehmann and Joseph 2015; Liu et al. 2022). The estimated global production of biochar is around 220 million tons per year (Woolf et al. 2010). A large fraction of biochar-based materials can be physically degraded into micro- and nano-sized fine particles in the environment (Sigmund et al. 2018; Spokas et al. 2014). Several studies have shown that biochar fine particles possess high colloidal stability and mobility in aquatic environments (Wang et al. 2013a, b, 2019). For example, even in sand columns saturated with 50 mM NaCl or 10 mM CaCl2, breakthrough of biochar fine particles still reached nearly 40% (Wang et al. 2019). Additionally, biochar fine particles (in particular, the nano-sized ones) exhibit strong adsorption affinities toward many classes of environmental contaminants, including polycyclic aromatic hydrocarbons, endocrine disruptor compounds, poly- and perfluoroalkyl substances, antibiotics, and herbicides (Kah et al. 2017; Lian and Xing 2017; Sigmund et al. 2020; Zhang et al. 2020). Consequently, biochar fine particles may serve as carriers of environmental contaminants and significantly enhance the transport of the adsorbed contaminants. Understanding the co-transport of biochar fine particles and organic contaminants is critical for minimizing the potential negative impact of biochar application in agricultural and environmental practices.

Thus far, only two published studies have addressed the effects of biochar fine particles on the transport of organic contaminants in subsurface environments (Hameed et al. 2021; Zhu et al. 2021). Biochar fine particles may increase or decrease the mobility of organic contaminants, depending on the properties of the biochar particles and the organic contaminants of concern, as well as the hydrogeological and solution chemistry conditions. For example, it was observed that biochar nanoparticles (BCNPs), especially of smaller sizes and those derived from biochars prepared at lower pyrolysis temperatures, significantly enhanced the transport of phenanthrene, atrazine, and oxytetracycline in soil columns saturated with 10 mM NaCl, an effect attributed to the vehicle effects of BCNPs, as well as the competition of dissolved organic carbon (released from the BCNPs) for the adsorption sites on soil grains (Hameed et al. 2021). In contrast, BCNPs inhibited the transport of ciprofloxacin (CIP) in sand columns saturated with 10 mM CaCl2 (pH 6.5) and the transport-inhibition ability increased with the increasing pyrolysis temperature of the biochar (Zhu et al. 2021). The inhibited transport was attributed to the competition of the BCNPs for the sorption sites of CIP, and to the deposition of CIP-bearing BCNPs, as the sorption of CIP increased the aggregation of the BCNPs and thus, suppressing the transport of the nanoparticles (Zhu et al. 2021).

Biochar materials can undergo various aging processes in the environment, as affected by sunlight irradiation, freeze-thaw and dry-wet cycles, chemical oxidation/reduction, and biological transformation (Sigmund et al. 2018; Wang et al. 2020). Aging may significantly affect the physicochemical properties of biochar (Hale et al. 2011; Liu et al. 2019a; Mia et al. 2017; Sigmund et al. 2018; Sun et al. 2013; Wang et al. 2019, 2020; Yang et al. 2021; Zimmerman 2010) and consequently, influence the adsorption properties and mobility of the BCNPs released from the biochar. Previous studies showed that aging can alter the surface and structural properties of biochar materials (e.g., surface functional groups and pore volume), thereby affecting the relative importance of different adsorptive mechanisms such as pore filling, electrostatic interaction, H-bonding, π-π interactions, and Lewis acid-base interactions (Hale et al. 2011; Wang et al. 2020; Yang et al. 2021). Aging can also affect the colloidal stability and mobility of BCNPs (Ma and Chen 2020; Ma et al. 2021; Wang et al. 2019). For instance, treating pinewood biochar with HNO3 increased its hydrophilicity; accordingly, the nanoparticles of the treated biochar were more mobile due to enhanced negative surface charge (Wang et al. 2019). Our previous studies demonstrated that even a slight alteration of biochar surface functional groups (e.g., from sulfide reduction or leaching of organic carbon) may affect the mobility of BCNPs (Ma and Chen 2020; Ma et al. 2021). In particular, the loss of surface carboxyl groups alleviates the deposition of particles via cation bridging, a prominent deposition mechanism in groundwater. To date, only one published study has addressed the effects of aging (H2O2 oxidation) on BCNPs-mediated transport, showing that the nanoparticles from the aged biochar were more effective in inhibiting CIP transport, in that, CIP was more adsorptive to the aged biochar, resulting in more significant aggregation and retention of CIP-bearing BCNPs (Zhu et al. 2021).

The objective of this study was to further understand the effects of BCNPs on the transport of common organic contaminants in saturated porous media, as well as the effects of sulfide reduction and leaching of organic carbon, two relatively mild but widely occurring aging processes, on the contaminant-mobilizing ability of BCNPs. Note that sulfide is prevalent both in the environment and in anaerobic wastewater treatment units (Kent et al. 2014; Rickard and Luther 2007), and leaching of organic carbon readily occurs in the environment (Dittmar et al. 2012; Jaffé et al. 2013). Pinewood and rice-straw biochars (pyrolyzed at 300 °C) were selected as the model biochars (Qu et al. 2016) to obtain the BCNPs. The physicochemical properties of the nanoparticles of as-prepared and aged biochars were characterized. The transport properties of six model organic compounds of varied physicochemical properties, including naphthalene (NAPH), pyrene (PYR), trimethoprim (TMP), CIP, sulfamethoxazole (SMX), and bisphenol A (BPA), as affected by different BCNPs were examined in 0.5 mM NaCl or in synthetic groundwater. The effects of BCNPs on the transport of the ionizable organic compounds (i.e., TMP, CIP, SMX, and BPA) were also compared under different pH conditions. Batch adsorption and desorption experiments were carried out to understand the adsorptive interactions between the BCNPs and the contaminants.

2 Materials and methods b

2.1 Materials

Pinewood (PW) and rice straw (RS) were ground and then pyrolyzed under oxygen-limited conditions using a muffle furnace. The pyrolysis temperature was programmed with a heating rate of 5 °C/min and holding a temperature of 300 °C was held for 2 h. The obtained biochars were ball-milled and then passed through 100-mesh sieves. These biochar samples are referred to as PW300 and RS300, respectively. To obtain sulfide-reduced or leached biochar samples, the as-prepared materials were treated with 0.1 mM Na2S or washed repeatedly with deionized (DI) water, following the methods developed in our previous studies (Ma and Chen 2020; Ma et al. 2021). The detailed procedures are given in Supplementary Information (SI). The sulfide-reduced and leached biochar samples are referred to as Na2S-PW300 and RS300OCD, respectively (the suffix OCD stands for organic-carbon-deficient).

14C-labeled naphthalene (2.15 GBq/mmol), trimethoprim (2.06 GBq/mmol), and ciprofloxacin (2.19 GBq/mmol) were purchased from Moravek Inc. (California, USA). 14C-labeled pyrene (2.18 GBq/mmol) was purchased from American Radiolabeled Chemicals (Missouri, USA). 14C-labeled sulfamethoxazole (0.35 GBq/mmol) and bisphenol A (0.74 GBq/mmol) were synthesized using 14C-labeled aniline hydrochloride and phenol as precursors, respectively (Li et al. 2014; Wu et al. 2022). The physicochemical characteristics of the compounds are given in Table S1 and Fig. S1. Nonlabeled naphthalene, pyrene, trimethoprim, ciprofloxacin, sulfamethoxazole and bisphenol A (all with purity > 99%) were purchased from Sigma−Aldrich (Missouri, USA).

Lufa soil (d50 = 295 μm), containing 86.0% sand, 11.5% silt, and 2.5% clay, was purchased from Lufa Speyer (Speyer, Germany). The values of fractional organic carbon (fOC) and uniformity of the soil were 0.71% and 0.46, respectively. Additional properties of the soil can be found in our previous papers (Liu et al. 2019b).

Artificial groundwater (AGW) was prepared following the literature (Chowdhury et al. 2013), and the recipe is given in Table S2.

2.2 Preparation and characterization of BCNPs suspensions

The BCNPs stock suspensions were prepared by using the methods mentioned in the literature (Chen et al. 2018, 2021; Lian et al. 2019; Wang et al. 2013a, b, 2019; Yang et al. 2017). Briefly, 2 g of an as-prepared or aged biochar sample was first added to 500 mL DI water, gently stirred for 2 min, and then sonicated at 50 W for 5 min to disperse the suspension. Afterward, the suspension was left quiescently for 72 h to settle the large particles (Qu et al. 2016). Next, the remaining suspension was filtered through 1-μm membrane filters (Pall, USA), and the filtrate was then passed through 0.45-μm membrane filters (Pall, USA). Finally, the particles retained on the filters were added to 500 mL DI water, stirred, and sonicated again for 5 min. The obtained stock suspensions were designated as BCNPs suspensions and were labeled as PW300_NPs and Na2S-PW300_NPs, RS300_NPs and RS300OCD_NPs, respectively. The mass concentrations were determined according to the measured total organic carbon contents (TOC-5000A, Shimadzu Scientific Instruments, USA) and elemental compositions (Vario MICRO cube, Elementar, Germany).

The physical dimensions and morphologies of the BCNPs were measured with atomic force microscopy (AFM) (D3100M, Digital Ltd., USA) and transmission electron microscopy (TEM) (JEM-2100, JEOL, Japan). The surface chemistry properties were characterized with X-ray photoelectron spectroscopy (XPS) (PHI5000 VersaProbe, Ulvac-Phi, Japan) and Fourier transform infrared (FTIR) transmission spectroscopy (NEXUS 870, Nicolet, USA). The surface area and micropore volume were measured using a Micromeritics ASAP2010 accelerated surface area and porosimetry system (Micromeritics Co, USA). The hydrophobicity was characterized by determining the water contact angle and n-dodecane–water partition coefficient (Mitzel et al. 2016; Song et al. 2011).

2.3 Column transport experiments

Lufa soil (~ 3.5 g) was dry-packed into Omnifit borosilicate glass columns (10 cm × 0.66 cm, Bio-Chem Valve Inc., USA) with two stainless-steel screens (10 μm, Valco Instruments Inc., USA) on both ends. The column experiments were run in an upward mode using syringe pumps (KDS-200, KD Scientific, USA). Once packed, the column was first purged with CO2 at low pressure for 30 min, and then flushed with 100 mL DI water (3 mL/h) followed by 180 mL background electrolyte solution. The porosity and dead volume were determined with tracer tests.

The experimental protocols of the column experiments are summarized in Table S3. To prepare the influents, a BCNPs stock suspension was first diluted with a background electrolyte of 0.5 mM NaCl or AGW in amber EPA vials to give the working BCNPs concentrations of ~ 20.0 mg/L. A stock solution of an organic contaminant in methanol was added immediately to each vial to give a contaminant concentration of ~ 10 μg/L. The volume of methanol was kept below 0.1% (v/v) to avoid co-solvent effects. The vials were sealed with Teflon-lined screw caps and equilibrated by tumbling end-over-end at 3 rpm for 7 d. The hydrodynamic diameter (Dh) and ζ potential of the BCNPs in different influents were measured by dynamic light scattering and electrophoretic mobility, respectively, using a ZetaPALS (Malvern Instruments, U.K.); these data are summarized in Table S4.

In a typical column experiment, the influent was pumped into the column from a glass syringe (100 mL, SGE Analytical Science, Australia). After ~ 60 pore volumes (PV), the influent was switched to the respective background solution for another 15 PV of injection. Effluent samples were collected every 3 PV. Each collected sample was split into two aliquots to measure the concentrations of BCNPs and contaminants. The concentrations of the BCNPs were determined using a UV − vis spectrophotometer (UV-2401, Shimadzu Scientific Instruments, USA), based on the pre-determined calibration curves (Figs. S2 and S3). The contaminants were quantified by determining radioactivity using a liquid scintillation counter (LS6500, Beckman Coulter, USA) (the detailed procedures are given in the SI).

2.4 Batch adsorption and desorption experiments

The adsorption and desorption isotherms of the contaminants to and from the BCNPs prepared using as-prepared and aged biochars, as well as sorption isotherms to Lufa soil, were conducted using a batch adsorption approach (Liu et al. 2018) (see SI for detailed procedures). For the convenience of comparing degrees of desorption hysteresis, the hysteresis index (HI) was calculated as follows (Huang and Weber 1997):

$$\textrm{HI}=\frac{q_{\textrm{e}}^{\textrm{d}}-{q}_{\textrm{e}}^{\textrm{s}}}{q_{\textrm{e}}^{\textrm{s}}}\ \left|\textrm{T},{C}_{\textrm{e}}\right.$$
(1)

where \({q}_{\textrm{e}}^{\textrm{d}}\) is adsorbed-phase concentration observed in the desorption experiment that is in equilibrium with an aqueous phase concentration Ce, and \({q}_{\textrm{e}}^{\textrm{s}}\) is the adsorbed-phase concentration calculated from Ce assuming that desorption is reversible. T specifies the temperature of the experiments. An HI value of 0 indicates that desorption is completely reversible; the higher the HI value, the greater degree of desorption hysteresis.

2.5 Statistical analysis

Significant differences were determined by one-way analysis of variance (ANOVA) using SPSS 17.0 software (IBM Corporation, USA) and the differences were considered significant at p < 0.05.

3 Results and discussion

3.1 Physicochemical characteristics of nanoparticles derived from as-prepared and aged biochar materials

Selected physicochemical properties of the nanoparticles prepared using the as-prepared and the aged biochars are summarized in Table 1. Both PW300_NPs and RS300_NPs—the nanoparticles of as-prepared biochars—had high C contents (73.9% and 65.2%, respectively) and abundant surface O-functional groups, including epoxy/hydroxyl, carbonyl, and carboxyl groups, as evidenced by the XPS and FTIR results (Figs. S4 and S5). The n-dodecane−water partition coefficients of PW300_NPs and RS300_NPs were 0.44 and 0.40, respectively, and the water contact angles were 92.7° and 86.3°. The TEM images show that the BCNPs were irregular in shape, with average particle sizes of 530 ± 124 nm (PW300_NPs) and 543 ± 145 nm (RS300_NPs) (Fig. S6). The AFM images confirm that the BCNPs consisted of sheet-like materials, with average lateral sizes of 400–600 nm and heights of 5–10 nm (Fig. S7). The micropore volumes of PW300_NPs and RS300_NPs were 0.12 and 0.07 cm3/g, respectively.

Table 1 Selected physicochemical properties of biochar nanoparticles (BCNPs) prepared using as-prepared and aged biochars
Full size table

The nanoparticles derived from the aged biochars had similar physical appearance to those from the respective as-prepared biochars, but had different surface chemistry properties (Table 1). The TEM and AFM images showed that there were no statistical differences (p > 0.05) in physical dimensions between nanoparticles of aged and those of as-prepared biochars (Figs. S6 and S7). The XPS results (Fig. S4) showed that aging resulted in changes in the distribution of surface O-functional groups. For example, a decrease of the carboxyl groups (from 8.75% for PW300_NPs to 4.61% for Na2S-PW300_NPs, and from 7.64% for RS300_NPs to 5.73% for RS300OCD_NPs) was observed. This was corroborated by the changes of the FTIR spectra (−C=O stretching of –COOH at ~ 1710 cm− 1 or C=O bonds of –COOH at 1622 cm− 1; Fig. S5). Additionally, the nanoparticles of sulfide-treated biochars had higher hydrophobicity, as evidenced by the higher n-dodecane−water partition coefficients (0.63 of Na2S-PW300_NPs vs. 0.44 of PW300_NPs) and higher the water contact angles (108° vs. 92.7°) (Table 1). The nanoparticles of sulfide-treated biochars also had lower BET surface area (28.5 m2/g of Na2S-PW300_NPs vs. 35.3 m2/g of PW300_NPs), but similar micropore volume (Table 1). In comparison, the hydrophobicity, BET surface area, and micropore volume were statistically indistinguishable between the nanoparticles of the OC-deficient biochar and those of the as-prepared biochar (p > 0.05, Table 1).

3.2 Ability of BCNPs to mediate transport of organic contaminants varies significantly with physicochemical properties of organic contaminants

The effects of BCNPs (~ 20.0 mg/L) on the transport of organic contaminants in sandy soil (saturated with 0.5 mM NaCl, pH 6.5) are shown in Fig. 1. The BCNPs resulted in significant mobilization of hydrophobic, nonpolar contaminants and positively charged polar contaminants, but slightly inhibited the transport of neutral and negatively charged hydrophilic, polar compounds. For example, in the presence of PW300_NPs the maximum breakthrough (C/C0) of PYR increased from 1.71% (in the absence of biochar) to 83.1%, and that of TMP from 2.95% to 81.6% (Fig. 1b and c). In comparison, the maximum C/C0 value of BPA decreased from 96.3% (without biochar) to 90.5% (with biochar) (Fig. 1f).

Fig. 1
figure 1

Effects of BCNPs on transport of NAPH, PYR, TMP, CIP, SMX, and BPA in 0.5 mM NaCl-saturated sandy soil at pH 6.5. The acronyms NAPH, PYR, TMP, CIP, SMX, and BPA stand for naphthalene, pyrene, trimethoprim, ciprofloxacin, sulfamethoxazole, and bisphenol A, respectively. The left panel shows the breakthrough curves of BCNPs, and the right panel breakthrough curves of the respective contaminants. The term “control” represents the contaminant transport experiments carried out in the absence of BCNPs

Full size image

It has been shown that the specific extent to which carbonaceous (nano)particles can mediate the transport of an organic contaminant depends not only on the relative adsorption affinity of the contaminant between the particles and the porous media, but also on how irreversibly the contaminant is bonded to the particles (Hofmann and von der Kammer 2009; Liu et al. 2018, 2019b; Qi et al. 2014; Wang et al. 2012; Zhang et al. 2011). NAPH and PYR—the two nonionic, nonpolar, hydrophobic compounds—were bound much more strongly to the BCNPs than to the soil grains (Fig. S8), consistent with the previous studies demonstrating that polycyclic aromatic hydrocarbons interact strongly with biochars via hydrophobic and π-π interactions (Kah et al. 2017; Lian and Xing 2017; Sigmund et al. 2020; Zhang et al. 2020). Additionally, both NAPH and PYR (in particular, the latter) exhibited desorption hysteresis from the BCNPs (Fig. 2), likely due to the entrapment of these compounds in the micropores of the BCNPs – it has been established that nonionic, highly hydrophobic organic contaminants (e.g., pyrene) tend to enter the porous regions of carbonaceous materials and become irreversibly trapped (Han et al. 2016; Pignatello et al. 2017). For both TMP and CIP a considerable fraction of the molecules was positively charged under the pH tested (Fig. S1), and these two compounds were mostly retained in the soil columns in the absence of the BCNPs (Fig. 1c and d), mainly due to the electrostatic attraction to the negatively charged soil grains (Chen et al. 2014; Golet et al. 2003; Sun et al. 2018; Zhang et al. 2014). The presence of the BCNPs significantly enhanced the transport of these contaminants (Fig. 1c and d), as both compounds were much more adsorptive to the BCNPs than to the soil grains (Fig. S8), and both exhibited strong desorption hysteresis, as indicated by the high HI values (Fig. 2). Note that the concentrations of PYR, TMP and CIP in the effluents of the column experiments correlated very well with the concentrations of the eluted BCNPs (Fig. S9), further indicating that only the BCNPs-adsorbed contaminants could break through (Liu et al. 2018; Zhang et al. 2011). A previous study reported inhibited transport of CIP in sand by biochar (Zhu et al. 2021); the opposite trend in this study was apparently due to the strong sorption of CIP to the soil grains. Notably, the BCNPs slightly inhibited the transport of SMX and BPA (Fig. 1e and f). The two compounds exhibited higher mobility than the BCNPs (Fig. 1). Thus, being associated with the BCNPs inhibited their transport, that is, the contaminants associated with the BCNPs were retained due to the deposition of the BCNPs during the co-transport process (Ren and Packman 2004). This was particularly evident for BPA, which was not only adsorbed strongly to the BCNPs (Fig. S8), but also exhibited desorption hysteresis (Fig. 2f).

Fig. 2
figure 2

Adsorption (black symbols) and desorption (red symbols) data of NAPH, PYR, TMP, CIP, SMX, and BPA to and from BCNPs (20 mg/L), using one-step desorption experiments. The acronyms NAPH, PYR, TMP, CIP, SMX, and BPA stand for naphthalene, pyrene, trimethoprim, ciprofloxacin, sulfamethoxazole, and bisphenol A, respectively. Hysteresis index (HI) values were calculated using Eq. 1. The error bars represent the mean deviations of duplicates

Full size image

Note that the effects of BCNPs on the transport of ionizable organic compounds can be sensitive to solution pH, as pH not only affects mobility of BCNPs, but also speciation of the contaminants (Fig. S1), and consequently, contaminant-BCNPs interactions. Thus, we compared the effects of the BCNPs on the transport of the four ionizable compounds (i.e., TPM, CIP, SMX, and BPA) under two pH conditions (Fig. 3). Notably, the transport-enhancement effects of the BCNPs on TMP or CIP were both lower at higher pH (i.e., pH 8.0 or 8.5), even though the BCNPs were more mobile (Fig. 3a and b). Increasing pH promotes the deprotonation of the functional groups of biochar and ionic compounds – at the higher pH values tested the mass fraction of the cationic species of TMP decreased from 80.7% to 11.7%, and the mass fraction of the anionic species of CIP increased from 0.4% to 36.4% (Fig. S1). This decreased electrostatic attraction for TMP to the negatively charged BCNPs and increased electrostatic repulsion between the anionic species of CIP and the BCNPs (Kah et al. 2017; Lian and Xing 2017). Consequently, the binding of these two compounds to the BCNPs was weakened, resulting in lower contaminant-mobilizing abilities of BCNPs at higher pH values. Similarly, the mass fraction of the anionic species of SMX increased from 38.7% to 86.5% with increasing pH from 5.5 to 6.5 (Fig. S1), resulting in enhanced electrostatic repulsions between SMX− species and nanoparticles. Therefore, a slightly weaker inhibitory effect on SMX transport by the BCNPs was observed at pH 6.5 than at pH 5.5, due to the weakened interaction between SMX and the BCNPs at the higher pH. In comparison, the BCNPs had a negligible effect on the transport of BPA, largely because the neutral species was the dominant one at both pH values (6.5 and 9.0) (Fig. S1). These results further highlight the role of contaminant-BCNPs interactions in governing the abilities of BCNPs in mediating transport of co-existing contaminants.

Fig. 3
figure 3

Effects of BCNPs on transport of TMP, CIP, SMX, and BPA in 0.5 mM NaCl-saturated sandy soil as a function of pH. The acronyms TMP, CIP, SMX, and BPA stand for trimethoprim, ciprofloxacin, sulfamethoxazole, and bisphenol A, respectively. The left panel shows the breakthrough curves of BCNPs, and the right panel breakthrough curves of the respective contaminants

Full size image

3.3 Aging of biochar affects ability of BCNPs in mediating transport of organic contaminants mainly by altering mobility of BCNPs

To qualitatively understand the effects of aging-induced alteration of physicochemical properties of BCNPs on contaminant-mobilizing ability of BCNPs, the transport properties of two model organic compounds, PYR and BPA, with different properties (nonpolar, nonionic vs. polar, ionic), mediated by the as-prepared and aged BCNPs, were examined. The BCNPs derived from the aged biochar materials exhibited different abilities in mediating transport of organic contaminants, compared with the nanoparticles derived from the as-prepared biochars. Both of the aged BCNPs (i.e., Na2S-PW300_NPs and RS300OCD_NPs) enhanced the transport of PYR in artificial groundwater-saturated sandy soil to a higher extent than their respective counterpart (i.e., PW300_NPs and RS300_NPs, the nanoparticles derived from the as-prepared biochars) (Figs. 4a and 5a). In the absence of BCNPs, PYR exhibited minimal breakthrough in saturated sandy soil, due to the strong sorption to the soil (Fig. S8). The presence of PW300_NPs and RS300_NPs increased the breakthrough of PYR to approximately 21.8% and 32.7%, respectively (Figs. 4a and 5a), whereas at the same nanoparticles concentration Na2S-PW300_NPs were able to increase the breakthrough of PYR to 35.6% and RS3000OCD_NPs to 44.4%. As discussed earlier, the presence of BCNPs inhibited the transport of BPA, due to the deposition of BPA-bearing BCNPs. Compared with PW300_NPs, Na2S-PW300_NPs (the BCNPs derived from the aged biochar) inhibited the transport of BPA to a lesser extent (Fig. 4b). Similar patterns were observed between RS300_NPs and RS300OCD_NPs, that is, RS300OCD_NPs inhibited BPA transport to a less degree (Fig. 5b).

Fig. 4
figure 4

Effects of sulfide reduction on contaminant-mobilizing abilities of BCNPs: transport of a PYR and b BPA in AGW-saturated sandy soil. The acronyms PYR and BPA stand for pyrene and bisphenol A, respectively. The biochar receiving treatment of sulfide reduction is indicated with the prefix “Na2S”. The left panel shows the breakthrough curves of BCNPs, and the right panel breakthrough curves of the respective contaminants. The term “control” represents the contaminant transport experiments carried out in the absence of BCNPs

Full size image
Fig. 5
figure 5

Effects of leaching of organic carbon on contaminant-mobilizing abilities of BCNPs: transport of a PYR and b BPA in AGW-saturated sandy soil. The acronyms PYR and BPA stand for pyrene and bisphenol A, respectively. The biochar receiving treatment to leach out the organic carbon is indicated with the suffix “OCD”, which stands for organic-carbon-deficient. The left panel shows the breakthrough curves of BCNPs, and the right panel breakthrough curves of the respective contaminants. The term “control” represents the contaminant transport experiments carried out in the absence of BCNPs

Full size image

Under the relatively mild aging conditions involved in this study, aging affected the ability of the BCNPs in mediating contaminant transport primarily by altering the mobility of the BCNPs, rather than the interactions between the BCNPs and the contaminants. As shown in Figs. 4 and 5 (left-hand-side plots), the BCNPs of aged biochars exhibited greater mobility in artificial groundwater than those of the as-prepared biochars. The increase in mobility was mainly due the loss of surface carboxyl groups from the aging process (sulfide reduction or leaching of organic carbon), leading to alleviated particle deposition via cation bridging (i.e., divalent metal cations acting as the bridging agents in linking the surface functional groups of the soil grains and the BCNPs) (Ma and Chen 2020; Ma et al. 2021). Accordingly, the retention of BCNPs-associated contaminants would be less significant, due to the alleviated deposition of the BCNPs (Figs. 4 and 5). Notably, previous research has shown that the binding of organic contaminants to biochar as affected by aging might also influence the co-transport behaviors of BCNPs and contaminants (Zhu et al. 2021). To quantitatively understand the interactions between BCNPs and organic contaminants as affected by the two aging processes, we examined the adsorption and desorption of PYR and BPA as affected by aging of biochar. As shown in Fig. 6, the adsorption-desorption isotherms of the contaminants on the BCNPs of as-prepared and aged BCNPs nearly overlap, indicating that the aging processes involved in this study had overall negligible effects on the affinities of the BCNPs for the two organic contaminants. Note that an apparent lack of response of biochar-contaminant interactions to the physicochemical changes of the biochars was partly the net result of the counterbalancing effects of several altered processes. For example, the increased surface hydrophobicity of the BCNPs (as evidenced by the values of the n-dodecane−water partition coefficient and water contact angle in Table 1) from sulfide reduction would increase contaminant adsorption (Chen et al. 2008; Han et al. 2016; Wang et al. 2006; Yang et al. 2021). However, the decreased surface area would inhibit adsorption (Kwon and Pignatello 2005; Lattao et al. 2014; Nguyen et al. 2007; Wang and Xing 2007). Additionally, the loss of surface O-functional groups could have suppressed BPA adsorption due to the weakened hydrogen bonding between BPA and the BCNPs (Han et al. 2013; Mitchell and Simpson 2013). Moreover, the small effects of the treatments on the micropore volumes of the biochars, a critical property controlling the degree of irreversible desorption or desorption hysteresis, were in line with the overall negligible changes of the contaminant-binding capability upon the treatments. Overall, the changes of the interactions between the BCNPs and the organic contaminants in our experiment appeared to have played insignificant roles in determining the ability of BCNPs in mediating contaminant transport.

Fig. 6
figure 6

Comparison of the extents of irreversible adsorption of contaminants to the BCNPs and aged BCNPs: a PYR and b BPA to PW300_NPs and Na2S-PW300_NPs; and c PYR and d BPA to RS300_NPs and RS300OCD_NPs. The acronyms PYR and BPA stand for pyrene and bisphenol A, respectively. The biochar receiving treatment of sulfide reduction is indicated with the prefix “Na2S”. The biochar receiving treatment to leach out the organic carbon is indicated with the suffix “OCD”, which stands for organic-carbon-deficient. The filled symbols are adsorption data and hollow symbols are desorption data. The hysteresis index (HI) values were calculated using Eq. 1. The error bars represent the mean deviations of duplicates

Full size image

4 Conclusions

This study shows that BCNPs may significantly affect transport of organic contaminants in saturated porous media, and environmental aging of BCNPs can further affect their contaminant-mobilizing abilities. The ability of BCNPs to mediate transport of organic contaminants varies significantly with the physicochemical properties of contaminants. For nonpolar, hydrophobic compounds and positively charged compounds, BCNPs can act as a carrier and enhance the mobility of contaminants under solution chemistry conditions commonly encountered in subsurface environments. The strong binding of these contaminants by BCNPs (which often results in desorption hysteresis) is the main factor contributing to the contaminant-mobilizing abilities of BCNPs. BCNPs may inhibit the transport of some of the negatively charged or neutral, hydrophilic organic compounds that exhibit high mobility in porous media; this is particularly important if the BCNPs can strongly bind the contaminants, as this increases the retention of the contaminants due to the deposition of contaminant-bearing BCNPs in the porous media. Under relatively mild aging conditions (e.g., sulfide reduction and leaching of organic carbon), aging of biochars affects the contaminant-mobilizing ability of BCNPs primarily by altering the mobility of the BCNPs, rather than altering the interactions between the BCNPs and the organic contaminants. Findings of this work underline the significant effects of BCNPs on the fate and transport of environmental contaminants, and further highlight the important role of aging in affecting environmental behaviors and effects of biochar materials. Further studies are needed to understand how BCNPs may mediate transport of other important environmental contaminants, under a wider range of environmental conditions. One limitation of this study is the use of artificial BCNPs of relatively uniform properties. Further studies should be conducted by expanding the properties of BCNPs (e.g., wider size distribution), and preferably, using in situ formed or field-aged BCNPs.