1 Introduction

Crude oil, as the dominant primary energy resource and raw material in modern industrial systems, is constantly exploited, processed, transported, and stored, which inevitably results in oil wastewater effluents as well as oil spills (Emeka Ndimele et al. 2018). According to statistics from 1970 to 2021, approximately 5.87 million tonnes of oil were lost as a result of tanker incidents globally (ITOPF 2022). Some well-known catastrophic accidents have caused extensive concern for petroleum pollution, such as the Deepwater Horizon spill that lost 779 million liters of oil (Atlas et al. 2011), the Hebei Spirit spill that led 10,900 tons of crude oil to escape into the sea (Yim et al. 2011), and the Sanchi spill that released almost 1 million barrels of oil (Chen et al. 2020a, b). Oil spill is a serious marine environmental problem and a pressing social problem for the government. When spilled oil enters the marine environment, these petroleum hydrocarbons cause the mass mortality of sea birds, seals, sea otters, and other sea life within a few weeks (Barron 2012). In the long term, the oil will give rise to the decline and eventual extinction of fish stocks, damage to the tissue of the digestive gland of krill, and chlorophyll-deficient propagules in mangroves (Veldkornet et al. 2020). As a persistent organic pollutant, hydrocarbons may cause extensive or permanent damage to marine ecosystems (Chandra et al. 2013).

For that reason, a sequence of classical physical, chemical, and bioremediation techniques including in situ burning, dispersants, physical containment, skimmers and adsorbents, and biodegradation have been applied to clean the oil slicks (Liu et al. 2017; Tran et al. 2021). However, how to remove oil more effectively from seawater is a critical problem, as it is of great importance to marine oil spill response. Even though these techniques are partly successful, there are a few environmental issues and weaknesses that limit their application to thoroughly remove oil, for example, the ecotoxicity of chemical dispersants, the release of toxic matter by burning, and incomplete results of oil collection (Bianco et al. 2021). Nevertheless, from an economic and ecological perspective, sorption methods, including natural organic/mineral materials, synthetic organic, and inorganic materials, are the most suitable to remove and collect oil slicks, preventing spilled oil from further drift and diffusion. Ideal adsorbents should have the following features: inexpensive and easily available, porous, with high specific surface area, environmentally friendly, and easily recoverable (Gurav et al. 2021a, b). Natural sorbents, especially carbonaceous materials, are the most environmentally friendly and easily available, compared with other materials (Gupta et al. Gupta and Tai 2016).

Biochar (BC) is a carbon-rich solid material produced through pyrolysis at specific temperatures (300–900℃) in an oxygen-limited environment (Liang et al. 2021). The precursor materials of biochar generally include agricultural wastes, kitchen wastes, animal manure, crustacean shells, and municipal solid wastes. Thermochemical conversion of solid waste into biochar can bring multifunctional benefits to the circular economy in addition to climate change mitigation and carbon sequestration. As a carbon–neutral tool for sustainable development, biochar is widely used in soil amendment, delivery of agrochemicals and microbes, catalyst production, environmental remediation, building material manufacturing, and feed formulation (Wang et al. 2021). In the aspect of contamination remediation, biochar is an almost ideal adsorbent. It has been used in the removal of organic contaminants containing antibiotics (Liang et al. 2022), dyes (Wathukarage et al. 2019), persistent organic pollutants (POPs) (Shi et al. 2016; Zhang et al. 2018), heavy metals (Agrafioti et al. 2014), nitrogen and phosphorus (Dai et al. 2020), based on its porosity, high surface area, and abundant functional groups, etc.

In recent years, biochar has gradually attracted much attention on adsorbing oil. In previous studies, various biochars have been studied as adsorbents for the purification of oil pollution, such as rice husk BC (Angelova et al. 2011), peat-derived BC (AlAmeri et al. 2019), and sawdust BC (Silvani et al. 2017). The oil sorption efficiency of pristine BC to a great extent relies on its hydrophobicity, which guarantees that biochar does not sink into the aqueous phase (Madhubashani et al. 2021). The enhancement of hydrophobicity by modifying the biochar surface can facilitate the sorption of the oil. Hence, research focused on biochar modification had been conducted by researchers. Fir BC, red pine BC, and lettuce BC decorated with lauric acid, coconut oil, and polydimethylsiloxane (PDMS) have been studied to improve sorption (Chen et al. 2020a, b; Gurav et al. 2021a, b; Navarathna et al. 2020). However, there are few research concerns about the oil sorption variation resulting from the difference in precursor materials and pyrolysis temperature. Accordingly, we investigated the characteristics of the inherent properties and sorption capacity of biochars produced from different representative feedstock materials and pyrolysis temperatures. The water repellency of biochar has been unsatisfactory for a long time. n-Octadecyl trimethoxy silane (OTMS) is a hydrophobic modification reagent with excellent performance and eco-friendliness. Hydrolysis and condensation reactions easily occur to form a hydrophobic film layer on the material surface, which makes the material have better water resistance. Then, biochar modification was performed by employing n-Fe3O4 and OTMS for controllability and hydrophobicity. The surface morphology, surface element distribution, BET surface area, functional groups, crystal structure, magnetic properties, thermal analysis, flotation, and wettability of the prepared adsorbents were investigated. The performances of the sorption capacity, availability, sorption kinetics, and sorption isotherm were tested. The target adsorbent is expected to float on the water to adsorb spilled oil individually or as a complement to other techniques, which is beneficial to the thoroughness of oil spill clean-up work.

2 Materials and methods

2.1 Feedstock and chemicals

Cornstalk (CS), corncob (CC), Sophora sawdust (SS), and rice husk (RH) were collected from a local farmland and wood-working factory (Qingdao, China). The n-octadecyl trimethoxy silane (OTMS) was obtained from Rhawn (China). Ferrous sulfate heptahydrate (FeSO4·7H2O), ferric chloride hexahydrate (FeCl3·6H2O), sodium hydroxide, and ethanol were purchased from Sinopharm Chemical Reagent Limited Corporation (Shanghai, China). Diesel and engine oil were provided by a local petrol station (Qingdao, China). Crude oil was obtained from an oilfield in the Bohai Sea (China).

2.2 Preparation of biochar

The process of biochar preparation is shown in Additional file 1: Fig. S1. The feedstock material, with impurities removed, was chopped into small chunks, washed with deionized water 3 times, and dried at 105℃ overnight. A crucible was filled with feedstock material and was wrapped in aluminum foil. Then pyrolysis was performed at 350℃, 500℃, and 650℃ in a muffle furnace (Zhonghuan, SX-G07103) under oxygen-limited conditions achieved by nitrogen gas (N2) purging in advance. The heating rate of the reactor was 10℃ min−1 and the target pyrolysis temperature was maintained for 4 h. After the reactor was cooled to room temperature, the biochars were removed and milled, passing through a 0.55 mm mesh screen for later use. According to the pyrolysis temperature and feedstock, the prepared pristine biochars were referred to as CSBC350, CSBC500, CSBC650, CCBC500, SSBC500, and RHBC500. Furthermore, the specific biochars were screened into three different sizes, small (0.55–0.25 mm), medium (0.25–0.15 mm), and large (< 0.15 mm).

2.3 Modification of biochar

The magnetic biochar was prepared as described by Mohan (2014). The main process is presented in Fig. 1a. FeCl3·6H2O (1.80 g) and FeSO4·7H2O (1.85 g) were dissolved in 150 mL distilled water for 5 min at 60℃ to prepare the Fe2+/Fe3+ solution. Then, 2.0 g of CSBC350 was slowly added to the Fe2+/Fe3+ solution and stirred for 30 min, forming a mixed suspension. After that, 1 M NaOH was carefully added to the mixed suspension until the pH reached ~ 11, and the suspension successively became brown and black at ~ pH 6 and ~ pH 10, respectively. Subsequently, the suspension was heated to 80℃ for 1 h and aged at room temperature for 2 h. Then, a magnet was used to separate the magnetic biochar (MBC), and the product was washed several times with distilled water and ethanol. Finally, the prepared MBC was filtered by a vacuum suction filter and dried at 50℃ in a vacuum drying chamber. The fabrication of hydrophobic magnetic biochar was as follows: the magnetic biochar and OTMS were mixed and heated for 3 h in 100 mL of 25% aqueous ethanol solution, centrifuged for 10 min at 6000 rpm and washed several times with ethanol. The obtained biochar was named OMBC.

Fig. 1
figure 1

General scheme for the preparation of the hydrophobic magnetic biochar (a). The schematic diagram of the silylation reaction mechanism for the magnetic biochar modification with OTMS (b)

Full size image

2.4 Sorption experiment

The adsorptive performance of several pristine biochars was determined in a pure oil system. The detailed steps were performed according to the standard test method for the sorption performance of sorbents as given by ASTM F726 (Bazargan et al. 2015). A certain quantity of biochar was placed on a preweighed stainless steel grid and immersed in the oil for 1 h to equilibrate at room temperature. Afterward, before being weighed, the excess oil was drained from the grid for 30 min until no drops fell. The sorption capacity was obtained by subtracting the sorption capacity of the grid. BC and modified biochars (MBC, OMBC) were employed for the removal of petroleum oil from the water surface. Measurement of the adsorptive capacity in the oil-water system was similar to that in the oil system. In brief, 200 mL of freshwater was poured into a beaker (250 mL) and 2.0 g of oil was added to simulate spilled oil. The stainless steel grid carrying 0.2 g of adsorbent was placed in the oil and was allowed to float on the surface of water for 2 h at room temperature. Next, the excess oil was allowed to drain off the grid for 30 min, and the grid was dried in a drying oven at 60℃ to remove adsorbed water. In the same way, the blank net was used as a control to measure the real sorption capacity. For the modified biochars (MBC, OMBC), a magnet was used to separate and collect the biochars after sorption equilibrium.

The sorption capacity of these adsorbents was calculated as follows (Eq. 1):

$${Q = }\frac{{{m}_{{c}} { - m}_{{b}} { - m}_{{a}} }}{{{m}_{{a}} }}$$
(1)

where \({Q}\) is the mass sorption capacity of the adsorbent (g g−1), \({m}_{{a}}\) is the weight of the adsorbent (g), \({m}_{{b}}\) is the mass sorption capacity of the control grid (g), and \({m}_{{c}}\) is the final weight of the grid and adsorbent after sorption (g).

The experiments to determine the effect of salinity, pH, and temperature on the sorption of crude oil by OMBC were carried out by changing a single condition based on the above description. After reaching adsorption equilibrium, a magnet was used to collect the OMBC oil mixture. Then, 20 mL of n-hexane was used to elute the adsorbed oil for three times, and the sorbent was washed again with deionized water. Finally, the cleaned sorbent was dried and weighed.

2.5 Sorption kinetics and isotherm models

The kinetics and isotherm models for oil sorption on biochar and modified biochar were studied in a water–oil system at 25℃. The adsorbent kinetics were determined by adding 200 mL of distilled water and 2 g of oil to a 250 mL glass beaker and varying the contact time from 0 to 240 min. For the isotherm study, adsorbent uptake capacities were determined by changing the quantity of each added oil from 0.30 to 7 g with 0.20 g of material while holding other factors constant. The detailed contents of the models are presented in the Supplementary Information.

2.6 Characterization methods

The surface morphology and elemental compositions of the adsorbent material were observed through scanning electron microscopy (SEM, TESCAN, TESCAN MIRA4, Czech Republic) equipped with an energy dispersive spectrometer (EDS, Xplore). The N2 adsorption–desorption experiments were conducted using a surface area analyzer (Micromeritics, APSP 2460, USA). The surface area and pore size of the biochars were obtained via the Brunauer–Emmett–Teller (BET) and Barrett-Joyner-Halenda (BJH) analysis. The composition of CSBC350, MBC, and OMBC was analyzed using X-ray diffractometry (XRD, Rigaku, SmartLab-SE, Japan) from 5° to 90°. The changes in the functional groups of the samples were determined by Fourier transform infrared spectroscopy (FT-IR, MAGNA-560, Nicolet, USA) within the range of 4000 to 400 cm−1 in transmittance mode. The hydrophobicity of the CSBCs and modified BC was measured by a contact angle measuring device (CA, OCA 20, Data Physics ES, Germany) at room temperature by releasing 2 μL of water on the sample surface. The magnetic properties of the samples were recorded by vibrating sample magnetometer (VSM, Lake Shore, 7404, USA) in a field of 15,000 to 15,000 Oe. The thermal stability of adsorbent was characterized by thermogravimetric analysis (TGA and DTG, METTLER TOLEDO, TGA2/DSC3, Switzerland) under nitrogen conditions from 30 to 800℃. The iron content of the water sample was measured by inductively coupled plasma-mass spectrometry (ICP‒MS, Agilent, 7700, USA).

3 Results and discussion

3.1 Characterization of the adsorbents

3.1.1 Surface morphology and surface area analysis

The surface appearances of the CSBCs produced at three pyrolysis temperatures (350℃, 500℃, and 650℃) were observed by SEM, as shown in Fig. 2. The images show that all CSBCs had an obvious three-dimensional macropore structure, regardless of the preparation temperature. However, there were still some evident structural distinctions that can be found among them. Figure 2 shows that with the increase in pyrolysis temperature from 350℃ to 650℃, the porous structure of CSBCs appeared to gradually shrink, and even collapse. Specifically, CSBC350 had a tidy, full, and sharply marginated cellular pore structure (Fig. 2a and b). CSBC500 appeared as a disordered, fragmented surface under continuous pyrolysis (Fig. 2c and d). Furthermore, the situation of CSBC650 was more prominent (Fig. 2e and f): numerous pores existed on the surface, and full holes became crevices, which is attributed to the formation of fine stomates that connect and destroy the original structure (Xiong et al. 2021). The properties and structure of biochar rely on the type and nature of the lignocellulosic feedstock used. Agricultural residues, such as corn stalks, contain more cellulose and hemicellulose and less lignin. During the preparation process, the cellulose and hemicellulose, which needed a low pyrolysis temperature (200–400℃) first decomposed to form pores. Subsequently, lignin gradually decomposed with increasing pyrolysis temperature (> 500℃), which left a portion of air holes (Yaashikaa et al. 2019). Under high temperature conditions, cellulose and hemicellulose were completely gasified, and lignin began to change, which caused the catastrophic collapse of the porous structure.

Fig. 2
figure 2

The SEM images of the CSBC350 (a, b), CSBC500 (c, d), and CSBC650 (e, f)

Full size image

The SEM images of MBC and OMBC are shown in Fig. 3. As shown in Fig. 3a, abundant magnetic particles and their aggregates were deposited on the surface of the biochar. Likewise, there was a depositional plane at the opening of the channel, as seen in Fig. 3b. Notably, there was no blocking phenomenon in the porous structure, and the original porous construction was maintained to a large extent. As seen in Fig. 3c, the magnetic nanoparticles had a lamellate structure, with a nano magnitude size of less than 100 nm (Karunanayake et al. 2019; Navarathna et al. 2019). After further modification using OTMS, the MBC appeared to have a smoother surface that coated the rough magnetic layer (Fig. 3d). At the same time, the macropores left by pyrolysis had well-defined borders as before, as shown in Fig. 3d and e. The surface element compositions of the OMBC were analyzed by EDS spectroscopy. The distribution of elemental C, O, Fe, and Si was observed from the corresponding SEM‒EDS mapping images (Fig. 3 g), with their peaks appearing at approximately 0.267, 0.521, 0.710, and 1.74 keV, respectively (Additional file 1: Fig. S2). Elemental C and O were the main component elements, mainly from the biochar substrate. Elemental Fe and Si were found in trace amounts in the biochar and were mainly a result of the magnetic nano iron oxide and the OTMS coating, respectively. In summary, according to their elemental mapping images (Fig. 3g), iron oxide nanoparticles were evenly distributed on the surface of the biochar, and homogeneously covered by the OTMS. The modification reaction process is presented in Fig. 1b. The -SiOCH3 functional group of the OTMS hydrolysis became -SiOH. Then, the dehydration reaction occurred between the -SiOH and the hydrophilic -OH functional groups on the surface of the MBC.

Fig. 3
figure 3

SEM micrographs of MBC (a–c) and OMBC (d–f). A selected area on the OMBC for EDS measurement (f). Elemental mapping images of OMBC for C, O, Fe, and Si (g)

Full size image

The specific surface area and pore size of the biochars were obtained via the Brunauer–Emmett–Teller (BET) and Barrett-Joyner-Halenda (BJH) analyses. The detailed data are shown in Table 1.

Table 1 BET surface area and pore size of biochars
Full size table

The biochar obtained at low pyrolysis temperatureshas more macropores, relatively few mesopores and micropores, and a smaller BET surface area. With the increase of the pyrolysis temperature, the hemicellulose and lignin in the cornstalk gradually volatilized, which resulted in the shrinkage or fragmentation of macropores and the increase in the ratio of mesopore and micropore, thus BET surface area and pore volume increased. This is the reason why the CSBC650 had a higher BET surface area and pore volume, reaching 208.3194 m2 g−1 and 0.096402 cm3 g−1, respectively. However, at higher pyrolysis temperatures, the macropores and mesopores were reduced, and the micropores became the main pores. Hence, the average pore diameter and mesopore diameter were smaller for CSBC650, 1.8256 nm and 3.2387 nm, respectively. The above results were also verified in Fig. 2 SEM images.

In addition, the BET surface area, pore volume, and BJH pore diameter of MBC were increased to 48.2159 m2 g−1, 0.209931 nm, and 17.1995 nm, due to the deposition of magnetic nanoparticles on the surface and the macropore's inner wall of CSBC350 (Fig. 3a–c). However, the modification of OTMS caused the rough surface of MBC to become smooth, destroying the existing mesoporous and microporous structures. Therefore, the BET surface area, pore volume, and pore diameter of OMBC were decreased compared with before the modification.

3.1.2 FT-IR and XRD analysis

Figure 4a demonstrates the FT-IR spectra of CSBC350, CSBC500, CSBC650, MBC and OMBC. From the spectrogram of the CSBCs, it can be easily concluded that the organic functional groups either faded away or were eliminated, and the curves became smooth as the pyrolysis temperature increased. These organic and inorganic functional groups were predominated by –OH (3406 cm−1), aliphatic CH stretching (2927 cm−1), ester C=O (1698 and 1562 cm−1), aromatic C=C (1580 cm−1), and C–O (1200 to 1060 cm−1) (Qian et al. 2016; Xie et al. 2021). The biochar produced at lower temperatures had lower aromaticity, compared with the biochar prepared at higher temperatures, which was an aromatic compound-rich carbon material with lower polarity (Keiluweit et al. 2010; Sun et al. 2011). For the MBC spectra, the most noteworthy was the Fe–O characteristic peak (575 cm−1), which indicates the success of magnetic loading on CSBC350 (Tapia et al. 2019). In the OMBC spectrum, the peaks at approximately 2917 and 2846, 1466 cm−1 were attributed to aliphatic CH and CH2 stretching vibrations. Furthermore, the peaks at 1027, 790 cm−1, and 880 cm−1 were associated with Si–O–Si (Villegas et al. 1988) and Si–C (Zhou et al. 2018) stretching vibrations, respectively. Simultaneously, the peak at around 575 cm−1 was also from the Fe–O band. The primary peaks in the OMBC spectrographic curve strongly suggested that the magnetic nanoparticles and the long-chain hydrophobic alkyl groups of OTMS had been grafted onto the surface of biochar as expected.

Fig. 4
figure 4

The FT-IR characterization of the CSBCs and OMBC (a). The XRD spectra of CSBCs, MBC, and OMBC (b)

Full size image

XRD analysis was used to determine the crystalline structures of the samples. As shown in Fig. 4b, the peak at around 22.7° in the XRD pattern of CSBC350 was the diffraction from the transformation of the lignocellulosic crystal structure to a more carbonaceous structure after pyrolysis (Regmi et al. 2012). This broad peak disappeared after magnetite precipitated due to magnetite’s highly intense crystalline peaks (Navarathna et al. 2020). It can be seen from the spectra of CSBCs that the peaks at 28.41°, 40.51°, 50.25°, and 58.66°, etc. indicate the presence of potassium chloride, especially in CSBC500. This is due to the release of potassium from biomass in the form of potassium chloride during the pyrolysis process (Fatehi et al. 2015). It is more pronounced at 500℃ and alleviated due to volatilization at higher pyrolysis temperatures (650℃) (Jensen et al. 2000). For both modified adsorbents, it was completely obvious that there were multiple typical iron oxide diffraction peaks. The peak shapes of MBC and OMBC indicates that a precipitated layer of magnetite particles had formed. The major peak at 35.62° was attributed to the crystalline plane of Fe3O4 (Karunanayake et al. 2017). The other peaks at 30.16° (220), 43.24° (400), 53.79° (422), 57.15° (511), 62.83° (440), and 74.26° (620) were also observed. Remarkably, the peak at 21.5° (002) in OMBC was attributed to SiO2 (Alchouron et al. 2020).

The crystallite sizes of the Fe3O4 deposited on MBC and OMBC were analyzed through the Scherrer equation (Eq. 2) (Navarathna et al. 2020). The magnetite crystallite sizes were deduced from the peak of D311 for MBC, and OMBC, which were 8.65 nm and 8.85 nm, respectively.

$${D}_{{{hkl}}} { = }\frac{{{k}\lambda }}{{\beta {\text{cos}}\theta_{{{hkl}}} }}$$
(2)

where \({k}\) is the shape factor, with \({k}\) = 0.89 for cubic crystals, \({\uplambda }\) is the XRD analysis wavelength (nm), \({\upbeta }\) is the full width at half-maximum value (rad) for XRD lines, and \({\uptheta }\) is half the diffraction angle 2 \({\uptheta }\).

3.1.3 Magnetic moment and thermal analysis

The magnetic properties of OMBC were investigated by vibrating sample magnetometry (VSM). The loading of magnetic nanoparticles offered OMBC paramagnetic behavior with no hysteresis while removing the magnetic field (Fig. 5a). The magnetic moment illustrates that the saturation magnetization of OMBC was 24.21 emu g−1, which was less than value of iron oxide (66.75 emu g−1). The reason for this phenomenon is the coating effect of OTMS on Fe3O4 and the low proportion of magnetic particles. However, the saturation magnetization of OMBC was sufficiently high to be separated by a magnet. Similarly, Raj et al. had prepared a magnetic adsorbent nanocomposite material, in which the magnetization was 15.00 emu g−1 (Raj et al. 2015). Yin et al. had produced carbonized Fe3O4@cotton for use in the oil-water separation and the saturation magnetization was 14.19 emu g−1, indicating outstanding magnetic properties (Yin et al. 2021). Furthermore, the OMBC iron loading rate was 16.86%, which is slightly lower than the theoretical value (18.26%).

Fig. 5
figure 5

Magnetic hysteresis loop of magnetic Fe3O4 and OMBC (a). Thermogravimetric analysis curves (10℃ min−1 heating rate under N2 atmosphere) of CSBC (350℃), MBC, and OMBC (b)

Full size image

The thermal gravimetric analyses (TGA) results of CSBC, MBC, and OMBC at 30–800℃ are shown in Fig. 5b. In the initial stage, CSBC, MBC, and OMBC had a small amount of weight loss below 150℃ to varying degrees, which was caused by the evaporation of physically adsorbed water. During this period, it is noticeable that the OMBC lost relatively little weight, which may be a result of its hydrophobic nature. A weight loss occurred in the temperature range of 170–260℃ was mainly due to the dihydroxylation and the loss of unreacted oxygen-containing species on the OMBC (Yang et al. 2021). In the temperature range of 350–550℃, there was great weight loss for CSBC (15.85%) and MBC (9.87%), which was attributed to the pyrolysis of the biochar produced at 350℃. Over the same temperature range, the OMBC dramatically lost almost 31% of its weight, which was mainly attributed to the thermal decomposition of the OTMS chain grafted on the MBC surface (Liu et al. 2015). This phenomenon exactly indicates that the alkyl chains (–CH2(CH2)16CH3) of OTMS were efficaciously grown on the surface of the magnetic biochar. The weight loss at temperatures above 550℃ resulted from the thermal decomposition of the biochar substrate, whereby almost 74%, 69%, and 54% of the weight was lost for MBC, CSBC, and OMBC, respectively, at temperatures up to 800℃.

As seen in the CSBC curve, three stages typically existed in the process. In the first stage (below 200℃), the decrease in mass was owning to the evaporation of moisture and light volatiles (Cárdenas-Aguiar et al. 2017). The next stage was from approximately 200 to 500℃, during which CSBC had a significant mass loss. As the primary process, polymerized organic compounds, such as hemicellulose and cellulose, were volatilized and decomposed at a fast rate. In the third stage above 500℃, the slow weight loss was caused by the decomposition of lignin and other organic matter with stronger chemical bonds, and residual organic matter was further transferred into biochar gradually (Yuan et al. 2019). Overall, although the weight loss can be caused by the modifying agent, the TG curve for the OMBC corresponded to the change in the weight of the biochar during the pyrolysis process.

3.1.4 Wettability measurement

As depicted in Additional file 1: Fig. S3a, the wettability of the CSBCs and OMBC was studied by water contact angle (WCA). Hydrophilicity is defined as a water contact angle less than 90° and vice versa. The WCA of CSBC350 was 86.23° (< 90°), appearing hydrophilic. However, it was significant that the WCA of the CSBCs gradually increased with increasing pyrolysis temperature (Additional file 1: Fig. S3a). The hydrophilicity and hydrophobicity of biochar are determined by the organic groups on the surface (Yuan et al. 2019). However, a higher pyrolysis temperature leads to a decline in polar functional groups on the biochar surface, which enhances the hydrophobicity of the biochar (Ahmad et al. 2014). In detail, the biochar produced at low temperatures had an abundance of hydrophilic groups, such as oxhydryl and carboxyl groups. In contrast, the biochar produced at higher temperatures was aromatic compound rich with high hydrophobicity. In Additional file 1: Fig. S3a, the water contact angle of OMBC decorated with OTMS reached 138.00°, which indicates a stronger hydrophobicity compared with the substrate material (CSBC350). A large number of long alkyl chains have been grafted on the surface of OMBC and in the crevices, which result in the hydrophobic characteristic that allows the material to float on the surface of the water. Additional file 1: Fig. S3b shows the situations of CSBC350, MBC, and OMBC in water after stirring to test the floatability. CSBC350 and MBC were mixed into the water phase, and OMBC was still floating on the surface yet. The silane reagent’s hydrocarbon chains on the surface of OMBC prevent water from adhering to the surface and entering the pore network, whereas the OMBC remains buoyant and does not sink. The excellent hydrophobicity and floatability help to guarantee efficient oil sorption on the water surface. In addition, even when entering the water, since hydrocarbons are entrapped inside the pore network, biochar particles also provide a favorable environment for the colonization of oil-degrading microorganisms (Nguyen et al. 2013).

3.2 Oil sorption capability

The oil sorption ability of different biochar was researched. We selected three representative oils with different densities and viscosities to study the adsorption properties of biochar on typical oils. Figure 6a shows that there were obvious differences in the oil sorption ability among the biochars produced from different biomass materials. At the same pyrolysis temperature (500℃), the CSBC had the maximum oil sorption ability for crude oil (8.45 g g−1), being almost 2 times that of CCBC and SSBC, while RHBC was the most inefficient. Based on the results, the CSBCs produced at different temperatures were employed to adsorb the crude oil in the pure oil system. CSBC350 had the strongest ability to adsorb oil, as seen in Fig. 6a. The main reason for this result was that the CSBC350 had a more complete macropore structure to hold the oil, as mentioned before. The oil sorption ability of biochar is closely associated with the pyrolysis temperature and feedstock material (Lian et al. 2017). Generally, feedstock materials, such as agricultural stalks that are composed of cellulose and hemicelluloses, only need low preparation temperatures to form a fully porous structure. Conversely, wood-like materials, containing a high proportion of lignin and celluloses, are often pyrolyzed at high temperatures to achieve volatilization and form pores (Yaashikaa et al. 2019). In addition, the biochar particle size is a significant and neglected factor in the sorption of organic pollutants, so the correlation between removal efficiency and biochar particle size has aroused controversy (Jin et al. 2022). The CSBC350 was sieved into three different sizes: large (0.55–0.25 mm), medium (0.25–0.15 mm), and small (< 0.15 mm). Different types of oil were used to measure the oil sorption ability of the three biochar sizes in the pure oil environment. As shown in Fig. 6b, the oil sorption ability of biochar in all sizes was the greatest for crude oil, with the results all being greater than over 10 g g−1. The medium-sized biochar has the best sorption to crude oil and diesel oil. The small-sized biochar had a higher ability to adsorb engine oil. Overall, the medium-sized biochar had a better sorption effect on the three kinds of oil. The differences in the sorption of the three oils by different sizes were mainly due to the pore size. The undersized biochar broke the initial macroporous structure, which made it inefficient for oil sorption (Jin et al. 2022).

Fig. 6
figure 6

The oil capacity of biochar and modified biochar. The crude oil sorption capacity of biochar that pyrolyzed from various feedstock materials prepared at 500℃ and the corn stock biochar that was pyrolyzed at different temperatures (a). The sorption capacity of different sizes of CSBC350 for three oils. (Small: 0.55–0.25 mm, Medium: 0.25–0.15 mm, and Large: < 0.15 mm) (b). The sorption capacities of CSBC (350℃), MBC, and OMBC for three types of oil in the oil–water system (c)

Full size image

The experimental results of CSBC350, MBC, and OMBC oil sorption in the oil-water system are shown in Fig. 6c. It was obvious that the oil sorption ability of MBC presented an obvious downward trend, due to the hydrophilicity of the nano iron oxide and magnetite deposition enhancing the adsorbent weight. The MBC decorated by OTMS had an oil sorption ability that was almost equal to that of CSBC350, reaching 8.54 g crude oil g−1, 5.03 g engine oil g−1, and 4.30 g diesel oil g−1. Although the oil sorption capacity of CSBC350 was slightly greater than that of OMBC, the capacity sharply declined when the CSBC faced the turbulence of the water wave, which can provide a greater probability of contact with water. However, OMBC can selectively adsorb oil on the water surface in a floating posture due to its hydrophobic surface. Regarding oil types, crude oil always has the highest level of sorption capacity followed by engine oil and diesel oil, regardless of the size of the original biochar or modification. In general, the difference in oil sorption capacity is related to the density and viscosity of the oil. The greater the density of the oil is, the greater the quality of the oil stored in the material space and the greater the calculated oil sorption capacity. The viscosity is another important influencing factor. Although a higher viscosity of oil hinders the sorption of oil molecules into the internal structure of the material, for macroporous materials, a higher viscosity is conducive to the sorption of oil molecules on the surface of the material. In this case, the sorption effect is greater than the absorption effect. Therefore, increasing the viscosity of the oil can increase the oil absorption of the biochar (Ceylan et al. 2009; Zhu et al. 2011). The viscosity of crude oil is relatively high, and it easily attaches and accumulates inside the biochar. Engine oil and diesel oil have lower densities and viscosities and, of course, lower storage ability in biochar.

3.3 Sorption kinetics and sorption isotherm experiments

As seen in Fig. 7, the sorption of crude oil by CSBC and OMBC reached equilibrium at 60 min, while the sorption equilibrium of diesel oil and engine oil was completed within 10 min. During the sorption of crude oil, the sorption was rapid at the beginning, and after 30 min the sorption was sluggish. The reason for this process is that initially the surface of biochar contained many blank active sites, which contribute to the fast progress of sorption. With continuous sorption, the sites were occupied by crude oil, and the crude oil had a high viscosity, hence, the sorption gradually slowed down. The increased sorption after 30 min was probably attributed to the breakage of crude oil droplets, which reduces the diameter of the crude oil droplets causing more interfacial pore area for the sorption to occur (Stang et al. 1994). However, in the presence of low viscosity oils, such as diesel and engine oil, the sorption was relatively fast. The pseudo-first-order and pseudo-second-order models were used to simulate oil sorption onto CSBC and OMBC, as shown in Fig. 7, and the results and detailed parameters are summarized in Additional file 1: Table S1. From the fitting results, the sorption kinetics of the CSBC and OMBC both coincided with the pseudo-second-order model, and the correlation coefficients (R2) were high (> 0.99). When comparing the theoretical equilibrium sorption capacity produced from the two kinetic models, the theoretical capacities fitted from the pseudo-second-order model were closer to the experimental values. These fitting results illustrate that the sorption process was a chemical sorption process, not just simple physical sorption.

Fig. 7
figure 7

The pseudo-first-order kinetics model of oils sorption on CSBC (350℃) (a) and OMBC (c). The pseudo-second-order kinetics model of oil sorption on CSBC (350℃) (b) and OMBC (d)

Full size image

The sorption isotherms of CSBC and OMBC are shown in Fig. 8. The sorption isotherms are significant for describing the interaction between adsorbate and adsorbent (Sidik et al. 2012). Nonlinear Langmuir and Freundlich isotherm models were employed to analyze the experimental isotherm data. The detailed contents are listed in Additional file 1: Table S2. The Langmuir isotherm model was most suitable for the sorption of three oils, with regression coefficient values (R2) that were both greater than 0.95 for CSBC and greater than 0.80 for OMBC, which indicates that the sorption of oils was monolayer coverage onto the homogenous distribution of active sites on the CSBC or OMBC surface, where all active sites are energetically equivalent (Gurav et al. 2021a, b). OMBC’s maximum oil sorption capacity (qm, 9.25 g crude oil g−1, 4.05 g diesel oil g−1, 4.41 g engine oil g−1) calculated from the Langmuir model was close to the experimental value (qe, Exp, 8.77 crude oil g g−1, 4.01 diesel oil g g−1, 4.44 engine oil g g−1). Certainly, the qm values of CSBC were also nearly the same as the experimental values.

Fig. 8
figure 8

The Langmuir isotherm model of oil sorption on CSBC (350℃) (a), OMBC (c). The Freundlich isotherm model of oil sorption on CSBC (350℃) (b), OMBC (d)

Full size image

3.4 Availability of sorbent

The oil sorption capacity of OMBC in the seawater-oil system was measured, as shown in Fig. 9a. The sorption ability of OMBC for the three oils showed a slight decrease compared with the performance in the deionized water-oil environment (Fig. 6c). A similar result was also reported by other researchers (Navarathna et al. 2020). Furthermore, the effect of salinity on OMBC sorption of crude oil was investigated. Figure 9b shows that the effect of salinity change on the sorption capacity of crude oil was small, but generally showed a trend of decreasing and then increasing in the range of 30–120‰. The reason for this may be that the hydrophobicity of the oil is enhanced under the action of static electricity, which makes the oil more easily adsorbed. In addition, the influences of pH and temperature on crude oil sorption were surveyed. As Fig. 9c shows, the sorption ability of OMBC in the low pH range was greater than that in the high pH range, and this result may be related to the charge change. When the temperature was gradually increased, the viscosity of crude oil decreased, and the sorption capacity increased slightly (Fig. 9d).

Fig. 9
figure 9

Sorption of oils by OMBC in the seawater-oil system (a). Sorption of OMBC on crude oil under different conditions, salinity (b), pH (c), and temperature (d)

Full size image

The recovery rate of the sorbent is an important parameter for magnetic oil sorbents. Hence, the recovery rate of OMBC after oil sorption was tested, and the results show that the recovery rate was almost stable at approximately 75% regardless of the oil type (Additional file 1: Fig. S4a). Furthermore, considering that the adsorbent will be used in the water environment, there may leach supported metals. Hence, after the sorption of crude oil by OMBC, the iron contents of deionized water and seawater were measured by ICP‒MS (Additional file 1: Fig. S4b). Compared with the blank water sample, it can be seen that iron was leached out neither in deionized water or seawater environment after OMBC adsorbed crude oil.

Table 2 summarizes the capacities of oil sorbents reported in the literature. It can be observed that our sorbent exhibited a great sorption performance, in comparison with previous sorbent capacities.

Table 2 Comparison with the oil sorption capacity of biochar in literatures
Full size table

3.5 Sorption mechanism

Biochar is an ideal sorption material due to its high porosity, high specific surface area, abundant surface organic functional groups, and outstanding cation exchange capacity and has extensive application in the remediation of soil pollution, water treatment, and carbon sequestration. The differences in the biochar characteristics were important reasons for the different sorption mechanisms. As shown in Fig. 10, the oil sorption mechanisms of biochar were mainly dominated by hydrophobic forces, pore intercepts, π–π electron attraction, hydrogen bonds, and electrostatic attraction. As another critical factor affecting oil sorption capability, the hydrophobicity of the adsorbent is determined by the heterogeneity of the morphology and surface chemistry and can be represented by water contact angle (Doshi et al. 2018). Through the aforementioned results of WCA analysis, the OMBC was identified as a hydrophobic material and can specifically adsorb oil due to the fatty acid functional groups on the surface and prevent the attachment and entry of water. Moreover, the three-dimensional multiporous structure of the biochar is one of the essential points that dominates the oil removal efficacy. Specifically, this three-dimensional structure increases the pore volume and provides more effective sites for oil sorption (Ha et al. 2016). For instance, higher pyrolysis temperatures can lead to the destruction of the partial pore structure of rice husk biochar, which results in a further decrease in removal (Zhao et al. 2013), similar to CSBC. The SEM figure of OMBC shows that the porous structure was unbroken, even though modification occurred. With the sorption of the surface hydrophobic layer, the intact pore structure associated with the presence of macropores can hold more oil. Moreover, hydrogen bonds are formed between polar components of the oil (especially resins and asphaltenes) and the biochar’s oxygen-containing functional groups, including –COOH and –OH (Jiang et al. 2016). The graphene content of biochar is electron rich and interacts with electron-deficient components of the oil, enhancing the π–π electron effect (Ahmad et al. 2014). In addition, electrostatic interactions may have been created between the positively charged components of the oil and the negatively charged adsorbent sites (Xu et al. 2011).

Fig. 10
figure 10

The schematic diagram for sorption mechanisms of oil by OMBC

Full size image

4 Conclusion

Biochars prepared from different feedstock materials at the same pyrolysis temperature (500℃) were used to determine the adsorbing capacity of crude oil (3.01–8.47 g g−1). The results showed that CSBC had a relatively higher sorption ability. Further research showed that CSBC350 had a fully porous structure and was an excellent adsorbent to oils, especially medium-size biochar. The OMBC modified by the magnetic particles and OTMS presented outstanding hydrophobicity and satisfactory controllability. Additionally, OMBC demonstrated a high oil sorption capacity to crude oil of 8.54 g g−1 in the oil-deionized water system. The kinetics model and isotherm model were conducted to study the sorption process. The pseudo-second-order kinetics and the Langmuir isotherm models were fitted compatibly to the sorption process. And OMBC maintained stable sorption capacity in different environmental conditions and had good environmental applicability and availability. For that reason, when facing a complex hydrodynamic environment, OMBC could be an excellent oil adsorbent and could be employed as an effective measure to combine with other physical methods to eliminate oil contaminants in the emergency handling of oil spills. After adsorbing oils, in contrast to other adsorbents, material based on biochar was a high-performance and economical fuel, which realizes resource reuse and is in accord with the goal of carbon neutrality.