Graphene oxide nanoparticles and hematite colloids behave oppositely in their co-transport in saturated porous media

Highlights

• Hematite inhibited GO mobility by heteroaggregation and providing retention sites.

• GO significantly enhanced hematite transport through multi-mechanisms.

• Highly mobile GO acted as a carrier of hematite colloids.

• Low pH or high ionic strength decreased the mobility of both particles.

• Ca²⁺ magnified transport-enhancement effect of GO on hematite transport.

Abstract

Since iron oxide minerals are ubiquitous in natural environments, the release of graphene oxide (GO) into environmental ecosystems can potentially interact with iron oxide particles and thus alter their surface properties, resulting in the change of their transport behaviors in subsurface systems. Column experiments were performed in this study to investigate the co-transport of GO nanoparticles and hematite colloids (a model representative of iron oxides) in saturated sand. The results demonstrated that the presence of hematite inhibited GO transport in quartz sand columns due to the formation of less negatively charged GO-hematite heteroaggregates and additional deposition sites provided by the adsorbed hematite on sand surfaces. Contrarily, GO co-present in suspensions significantly enhanced the transport of hematite colloids through different mechanisms such as the increase of electrostatic repulsion, decreased physical straining, GO-facilitated transport of hematite (i.e., highly mobile GO nanoparticles served as a mobile carrier for hematite). We also found that the co-transport behaviors of GO and hematite depended on solution chemistry (e.g., pH, ionic strength, and divalent cation (i.e., Ca²⁺)), which affected the electrostatic interaction as well as heteroaggregation behaviors between GO nanoparticles and hematite colloids. The findings provide an insight into the potential fate of carbon nanomaterials affected by mineral colloids existing in natural waters and soils.

Introduction

Graphene oxide (GO), a new two-dimensional carbon-based materials, has received much attention due to its outstanding physicochemical characteristics, such as optical, mechanical, and thermal properties (Chen et al., 2012a). The growing use of GO nanoparticles has raised global concerns due to their potential adverse effects and environmental risks (Li et al., 2019a; Zhao et al., 2014). Many previous studies reported that GO could induce toxic effects on microorganisms such as bacteria and mammalian cells (Akhavan and Ghaderi, 2010;Liu, 2011; Yang et al., 2013). With the wide use of GO nanomaterials in many fields such as drug delivery, energy storage devices, and photocatalysis (Bullo et al., 2019; Cheng et al., 2018; Jaya Seema et al., 2018; Lalwani et al., 2013), colloidal GO particles will be inevitably released into the natural environment including soil, groundwater and surface water (Lu et al., 2019; Zhao et al., 2014). In this case, there is a very high probability that GO will interact with natural minerals such as clays and iron oxides (Feng et al., 2019; Zhao et al., 2015).

Iron oxides including goethite and hematite are abundant in natural environment including soil and sediment (e.g., the concentrations in the soil and sediment samples are in the range from a few μg/L to several hundred mg/L) (Li et al., 2019b; Wang et al., 2015a, 2015bbib_Wang_et_al_2015abib_Wang_et_al_2015b). Since iron oxides can interact with various substances, they have been exhibited considerable effect on the fate and transport of metal ions (e.g., lead (Hassellov and von der Kammer, 2008), arsenic (Yean et al., 2005), and copper (Madden et al., 2006; Wang et al., 2011a, Wang et al., 2011b)) and engineered nanoparticles (e.g., silver nanoparticles (El-Badawy et al., 2013) and GO (Chen et al., 2019; Duster et al., 2016; Qi et al., 2019; Wang et al., 2017)). For example, Yean et al. (2005) found that desorption of arsenic from smaller magnetite nanoparticles exhibited stronger desorption hysteresis due to the formation of inner-sphere surface complexes (i.e., highly stable iron-arsenic complexes). In a recently published study (Ma et al., 2018), it was reported that the interaction between humic acid and ferrihydrite colloid highly regulated the transport and deposition of arsenic in porous media. Our recent study showed that the heterogeneity (i.e., iron oxide coating on sand surfaces) in porous media significantly enhanced GO deposition, mainly due to increased electrostatic attraction between GO nanoparticles and porous media as well as the surface roughness (Qi et al., 2019). Therefore, the transport and fate of iron oxides in subsurface water systems has drawn significant attentions in past several decades (Li et al., 2019b; Wang et al., 2020).

To date, several studies have explored the transport behaviors of iron oxide colloids or GO nanoparticles individually in saturated porous media such as sand and soil. Different environmental factors such as solution chemistry (e.g., pH (Lanphere et al., 2013; Pawlowska et al., 2017; Qi et al., 2014a; Zhuang and Jin, 2008), ionic strength (Feriancikova and Xu, 2012; Legg et al., 2014; Pawlowska et al., 2017; Wang et al., 2015a, 2017bib_Wang_et_al_2015abib_Wang_et_al_2017), and divalent cation (Lanphere et al., 2014; Xia et al., 2015, 2017bib_Xia_et_al_2015bib_Xia_et_al_2017; Yang et al., 2016)), pore-water velocity (Carstens et al., 2017; Qi et al., 2014b; Wang et al., 2015a; Zhang et al., 2018), dissolved organic matter (Liao et al., 2017a, 2017bbib_Liao_et_al_2017abib_Liao_et_al_2017b; Qi et al., 2014c; Shen et al., 2019; Wang et al., 2012), and surfactants (Fan et al., 2015a; Pawlowska et al., 2017; Wang et al., 2019a) have been investigated and were found to considerable effect on the transport of iron oxide colloids or GO nanoparticles in the subsurface environment.

Recently, an overwhelming majority of the previous studies are limited to the transport of single GO or iron oxide in porous media. Nevertheless, the understanding of the co-transport behaviors of these two suspended particles in porous media has not been systematically investigated. Considering that the mobile colloidal phase of iron oxides is known to play important roles in the transport and fate of colloids/biocolloids (e.g., titanium dioxide (Fisher-Power and Cheng, 2018), hydroxyapatite (Wang et al., 2015a), plastic particles (Li et al., 2019b), dissolved organic matter (Carstens et al., 2018), bacteria (Georgopoulou et al., 2020; Yang et al., 2016), and viruses (Zhuang and Jin, 2008)), it is, therefore, anticipated that the transport of GO nanoparticles in the aquifer media is, in part, affected by the presence of iron oxide colloids. More importantly, it is necessary to note that the co-transport properties of GO nanoparticles and iron oxide colloids may be markedly different from co-transport of other nanoparticles/colloids (e.g., TiO₂ and biocolloids) and iron oxide particles in porous media due to the unique physical geometry and surface chemistry of GO (Dreyer et al., 2010).

Furthermore, attention has been paid to the heteroaggregation behaviors between GO and natural minerals including montmorillonite (Zhao et al., 2015), kaolinite (Huang et al., 2016; Sotirelis and Chrysikopoulos, 2017; Syngouna et al., 2020), goethite (Huang et al., 2016; Zhao et al., 2015), and hematite (Feng et al., 2017, 2019bib_Feng_et_al_2017bib_Feng_et_al_2019). The information obtained from these studies demonstrated that physicochemical properties (e.g., surface charge and functional groups) of mineral particles and solution chemistry could influence the GO−mineral heteroaggregation behaviors. For example, Zhao et al. (2015) reported that GO considerably promoted the dispersion of goethite particles via heteroaggregation. They also found that the dominant force operating in GO−goethite heteroaggregation was electrostatic attraction. Normally, GO−mineral heteroaggregation might result in the destabilization of GO nanoparticles; however, GO−goethite heteroaggregates were still able to suspend in aqueous phase, probably due to excellent dispersibility of GO (Zhao et al., 2015). In addition, Feng et al. (2017) found that the heteroaggregation rates of negatively charged GO and positively charged GO with hematite first increased and then decreased with increasing GO/hematite mass concentration ratios. As the heteroaggregation of GO and iron oxide colloids can possibly change their structures and properties, this process is expected to influence their co-transport characteristics in porous media.

Thus, the goal of this project was designed to explore the co-transport properties of GO nanoparticles and hematite colloids (selected as a model iron oxide) in saturated sand columns. Different GO concentrations (3, 5, and 10 mg/L) and different hematite concentrations (3, 5, and 10 mg/L) were used in column experiments. Moreover, the effects of solution chemistry (i.e., pH (5.0–9.0), ionic strength (0–20 mM NaCl), and divalent cation (Ca²⁺)) on the co-transport of GO and hematite were also explored. Both breakthrough curves as well as retention profiles of GO nanoparticles and hematite colloids in co-transport experiments were examined. The possible mechanisms contributing to co-transport and deposition behaviors were proposed and discussed. This work firstly focused on the co-transport properties of GO nanoparticles and natural mineral colloids and could be helpful in refining our understanding on the mutual effect GO and iron oxide when they transport in porous media.