1 Introduction

Woody agricultural wastes are generated by biomass byproducts which come from agriculture residues (Boldrin and Christensen 2010; Shi et al. 2013). In the past decades, large amounts of woody agricultural waste have been collected for landfill or energy generation in China (Shi et al. 2013). The decomposition in landfills and incomplete combustion contribute significantly to the emission of greenhouse gases (CO2, CH4, N2O) in the atmosphere and therefore cause negative environmental impacts (Calabro 2009; Semrau 2011). In order to decrease the greenhouse gases release, different methods prior to landfills and combustion have been developed (Kan et al. 2016). One of these methods is the thermal decomposition of woody agricultural waste (oxygen-limited pyrolysis) to produce biochar which could reduce greenhouse gases release (Lehmann and Joseph 2009; Mitchell et al. 2013).

Biochar could remain in soil as a stable carbon (Lehmann and Joseph 2009). Greater persistence of biochar compared to uncharred biomass could decrease the CO2 emissions (Dong et al. 2018). More importantly, biochar has a porous structure and special physicochemical properties that allow it to be used as an alternative adsorbent to remove pollutants from water and air (Chemerys and Baltrėnaitė 2018). It has shown to be able to reduce the risk of toxic compound in environment contamination (Park et al. 2016; Pathirana et al. 2019; Yang and Jiang 2014; Zhou et al. 2016). Bazan-Wozniak et al. (2020) reported that the maximum sorption capacity of the biochars was 146 mg/g for methylene blue, 103 mg/g for methyl red, and 81 mg/g for nitrogen dioxide was 81 mg/g (Bazan-Wozniak and Pietrzak 2020). Chen et al (2019) also demonstrated the ability of biochar to remove heavy metals and organic pollutants from water (Chen et al. 2019; Huang et al. 2019; Shi et al. 2019).

However, different factors such as feedstock and pyrolysis parameters affected biochar properties which are related to the adsorption capacity. For instance, the biochar produced from pine wood had much higher O/C than that of eucalyptus wood, suggesting a greater degree of oxygenated structure in pine wood tissues(Mitchell et al. 2013), while the pH value of pine wood biochar was much lower than that of eucalyptus wood (Balwant Singh et al. 2010; Enders et al. 2012). Differences in structural composition of lignocellulosic biomass make it difficult to predict biochar properties during the pyrolysis. As demonstrated by Chemerys and Baltrėnaitė (2018), interactions between lignin and cellulose during pyrolysis reduce biochar formation, while the interactions between cellulose and hemicellulose do not significantly influence the formation of pyrolysis products (Chemerys and Baltrėnaitė 2018). Thus, it is important to assess the properties of biochar from different feedstocks and pyrolysis temperatures. The various properties of biochar play a vital role in the adsorption capacity and mechanism (Baoliang Chen et al. 2008). Some mechanisms such as complexation, cation exchange, precipitation, electrostatic interactions, π–π electron donor–acceptor interactions and hydrogen bonding have been proposed (Baoliang Chen et al. 2008, 2015; Kwak et al. 2019; Samsuri et al. 2013; Zhang et al. 2013). However, most research provided sorption mechanisms for pollutants as a group, lacking a comparison of the main mechanisms for removal of different pollutants. Meanwhile, there is little consensus on the influence of physico-chemical properties of woody agricultural wastes-derived biochars on heavy metal and organic pollutant adsorption capacity.

As such, the objectives of this study are: (1) to characterize the physical and chemical properties of biochars from mulberry wastes and cinnamon wastes at different pyrolysis temperatures including 350 °C, 450 °C, 550 °C, and 650 °C (2) to evaluate the adsorption capacity of biochars for removing organic and inorganic pollutants; (3) to quantify the effect of physico-chemical properties of biochars the adsorption of heavy metals and organic pollutants and compare the factors affecting adsorption of inorganic and organic pollutants onto biochars.

2 Material and methods

2.1 Materials

Atrazine (AT) was purchased from Beijing North Weiye Metrology Technology Research Institute. Cadmium nitrate tetrahydrate (Cd(NO3)2-4H2O) was purchased from Sonopharm Chemical Reagent Co., Ltd. All other used chemicals were of analytical grade and purchased from DAMAO chemical reagent factory in Tianjin, China. Stock solution of Cd2+ (1000 mg/L) was made by dissolving suitable amount of Cd in ultrapure water. Atrazine stock solution (1000 mg/L) was made by dissolving suitable amount of methanol.

2.2 Preparation and characterization of biochars

The mulberry wastes (pH = 8.1) and cinnamon wastes (pH = 4.5) were grounded to pass a 2 mm sieve and pyrolyzed in a biomass carbonization furnace with an oxygen-limited condition at 350 ℃, 450 ℃, 550 ℃ and 650 ℃, respectively. The biochars were cooled in a steel tank after pyrolysis. After preparation, the biochars were sieved through a #200 mesh sieve (< 75 um) and stored in sealed plastic sample bags. The selected properties of biochars were listed in Table 1. For typographical convenience, the biochars were named MB350, MB450, MB550, MB650, CB350, CB450, CB550 and CB650 to indicate their feedstocks of mulberry waste and cinnamon waste and their pyrolysis temperatures of 350 °C 450 °C, 550 °C and 650 °C, respectively.

Table 1 The selected properties of (mulberry biochar) MB and (cinnamon biochar) CB
Full size table

2.3 Characterization of biochars

The Brunauer–Emmett–Teller (BET) surface area and total pore volume (TP) were calculated based on the N2 sorption–desorption isotherms using a gas analyzer (Quantachrome Quadrasorb, USA). A field emission scanning electron microscopy (SEM) instrument (S-3400 N-II, Hitachi, Japan) was used to observe the surface morphology of each biochar. The Fourier transform infrared spectroscopy (FTIR) was investigated on a spectrophotometer (VETREX 70, Bruker). The main element composition was confirmed by Elementar (Vario EL cube). Ash content was investigated by calculating mass loss of biochars after being heated at 800 °C for 8 h. The zeta potential (ZP) was determined by using a micro-electrophoresis apparatus (JS94G+, Shanghai). The cation exchange capacity (CEC) of biochar was measured by using the silver thiourea methods described by Dohrmann (2006). The concentration of total P was measured by the phosphomolybdate colorimetric assay using a UV–vis spectrometer (V1800, UNICO, China).

2.4 Adsorption

The adsorption isotherms of Cd and AT were investigated to reveal the adsorption characteristics of wood waste derived biochars. All the batch adsorption experiments were performed in 25 ml vials by mixing a certain amount of biochar with Cd or AT solution of varying initial concentration ranging from 0.06 to 19 mg/L for Cd solution and from 5 to 32 mg/L for AT solution, respectively. 0.05 g of biochar was mixed with 20 ml of Cd or AT solution. The pH values of the solution were adjusted by adding 0.01 mol/L HNO3 or 0.01 mol/L NaOH until the pH value reached 7. The residual concentration of Cd was given by atomic adsorption spectrometer (AAS, PEAA700, USA). The residual concentration of AT was determined by High Performance Liquid Chromatography (HPLC) (Shimadzu. Japan). All of the adsorption tests were replicated three times.

The amount of adsorbate was calculated using the following equation:

$$q=\frac{{C}_{0}-{C}_{e}}{m}\times V$$
(1)

where \({C}_{0}\) (mg/L) and \({C}_{e}\) (mg/L) are the initial and equilibrium concentration of the Cd or AT, respectively, \(V\) (L) is the volume of the Cd or AT solution, \(m\) is the amounts of biochars.

2.5 Isotherm model fittings

The Freundlich isotherms model and Langmuir isotherms model were chosen to predict the adsorption behavior. The equations were:

$${q}_{e}={K}_{F}{C}_{e}^\frac{1}{n}$$
(2)
$${q}_{e}=\frac{{K}_{L}{q}_{m}{C}_{e}}{1+{K}_{L}{C}_{e}}$$
(3)
$${R}_{L}=\frac{1}{(1+{C}_{0}{K}_{L})}$$
(4)

where \({q}_{e}\) is the adsorption capacity, \({C}_{e}\) is the residual concentration of adsorbate, \({K}_{F}\) and \({K}_{L}\) are the Freundlich and Langmuir adsorption coefficient, \(\frac{1}{n}\) indicates the surface heterogeneity or adsorption intensity, \({q}_{m}\) represents the maximum adsorption capacity. RL is to determine whether the adsorption is favorable for the Langmuir adsorption. \({C}_{0}\) is the lowest initial concentration of solutes in the solution.

3 Results and discussion

3.1 Characterization of biochars

The basic properties of biochars were firstly investigated and presented in Table 1. The biochar pH values varied significantly among feedstocks. Specifically, pH values of mulberry biochar (MB) were much higher than that from cinnamon biochar (CB). Similar phenomenon was also observed in ash content. The correlation analysis showed that the pH was positively related to ash content (P < 0.05). It suggested that the high pH of MB was attributed to its high mineral ash content. Zeta potentials of biochars were negative, reflecting that the surfaces of both biochars are negatively charged. No significant difference in the zeta potential was found between MB and CB, which meant the zeta potential was not affected by the type of feedstock.

The element analysis in Table 2 showed that all of the biochars were carbon rich materials. The C content of biochars ranged from 68.35 to 83.95%. The high carbon content may indicate high hydrophobicity of biochar properties that favors for adsorption (Liu et al. 2015). The atomic ratio of H/C, O/C and (O + N)/C was utilized to provide information about the level of aromatic structure, hydrophilicity and polarity index, respectively. For all feedstocks, the lowest values of H/C were obtained from MB650 and CB650, which indicated the pyrolysis was a process of aromaticity enhancement. Meanwhile, the highest values of O/C and (O + N)/C were noted at MB350 and CB350, which indicated the stronger hydrophilicity and polarity at low pyrolysis temperature.

Table 2 Element analysis of (mulberry biochar) MB and (cinnamon biochar) CB
Full size table

Functional groups of biochars were determined by the FTIR (Fig. 1). The peak at 3440 cm−1 represented the stretching vibration of O–H and N–H. The peaks ranging from 2850 to 2950 cm−1 were assigned mainly to the aliphatic C-H and C-H2 in biopolymers (Baoliang Chen et al., 2008). The characteristic for C=O stretches of carboxyl groups and carbonyl groups in esters bonds and aromatic rings was found at the peaks at1600–1700  cm−1, which might be shifted due to the protonation and polarity (Yao et al. 2015; Zięba-Palus et al. 2017). The aliphatic C–H and C–H2 diminished when heating to 450 ℃, indicating the decrease of contents of nonpolar groups. The peaks at 1600–1700 cm−1 became weak with pyrolysis temperature increasing and disappeared at 650 ℃, due to the thermal destruction of ester C=O and aromatic C=O. This suggested the polar groups were eliminated upon heating. Moreover, the peak at 1386 cm−1 might be ascribed to the –OH bending vibration or COO– stretching vibration (Rich and Marechal 2008; Zhang et al. 2017). The peak at 1040 cm−1 was probably attributed to the stretching vibration of C–O and C–C. The peaks at 880–753 cm−1 were assigned to aromatic C–H which were increased with increasing pyrolysis temperature.

Fig. 1
figure 1

The fourier transform infrared (FTIR) spectroscopy of (a) mulberry waste biochar (MB) and (b) cinnamon waste biochar (CB). The 350, 450, 550 and 650 represent the pyrolysis temperature of 350 ℃, 450 ℃, 550 ℃ and 650 ℃, respectively

Full size image

The BET surface area of biochars ranged from 2.89 to 8.05 m2/g (Table S1). The CB650 produced the highest BET surface area for biochar. For all feedstocks, the high temperature promoted the BET surface area. Moreover, it has been reported that micropore play the key role in the specific surface area (Chu et al. 2018). As listed in Table S1, the highest value of micropore volume was observed at CB650, which was consistent with the BET area. The average pore diameter for all biochars was in the range of 2–50 nm, indicating the pore system were mainly consist of mesopores (Lehmann and Joseph 2009). The SEM surface morphologies of biochars produced from the two feedstocks were shown in Fig S1. All of the biochars presented rough structure and irregular plates. Meanwhile, the pore structure was also observed in all images of biochars. Biochar at 650 ℃ clearly had more abundant pore structure than biochar at 350 ℃. MB650 and CB650 therefore had higher total pore volumes and micropore volumes. This was located in consistent with results of N2 sorption–desorption analysis.

3.2 Adsorption isotherms

In order to characterize the biochar adsorption behavior and performance, the adsorption isotherms of Cd and AT were investigated and shown in Figs. 2 and 3, respectively. The sorbed amount was increased with the initial concentration of solutes, indicating biochars derived from mulberry and cinnamon wastes were effective adsorbents for removing heavy metals and organic pollutants from water. Moreover, it was found that the pyrolysis temperature significantly affected the biochar adsorption performance. For AT adsorption, no matter which the type of feedstock, the highest sorbed amount was obtained from biochar at 650 ℃.

Fig. 2
figure 2

Adsorption isotherms of Cd onto mulberry waste biochar (MB) and cinnamon waste biochar (CB). The 350, 450, 550 and 650 represent the pyrolysis temperature of 350 ℃, 450 ℃, 550 ℃ and 650 ℃, respectively

Full size image
Fig. 3
figure 3

Adsorption isotherms of AT onto mulberry waste biochar (MB) and cinnamon waste biochar (CB). The 350, 450, 550 and 650 represent the pyrolysis temperature of 350 ℃, 450 ℃, 550 ℃ and 650 ℃, respectively

Full size image

Two adsorption models including Freundlich model and Langmuir model were selected to describe the experimental data (Figs S2 and S3). Langmuir model assumes that the surface of adsorbents is uniform and no interaction between adsorbate molecules on adjacent sites exists. Freundlich model is an empirical model, which focuses on the assumption that the adsorption is multilayer and heterogeneous (Wang et al. 2018). The related parameters are listed in Table S2. The choice of isotherm depends on the determination coefficient R2. The closer its value is to 1, the better the fitting is. The R2 values of characterizing the fit to the Cd adsorption isotherms were 0.96–0.99 for Langmuir isotherm and 0.90–0.99 for the Freundlich isotherm, respectively. For AT adsorption, the R2 values were 0.98–1 for Langmuir isotherm and 0.99–1 for the Freundlich isotherm, respectively. The results indicated that the adsorption data of Cd and AT were generally well correlated with Langmuir and Freundlich models. It can be speculated that a multilayer adsorption occurred only because it fitted well with Freundlich model.

The maximum adsorption capacity of Cd onto biochars from Langmuir model varied from 416.34 to 5306.41 mg/kg, and was in the order: MB650 > MB550 > MB450 > MB350 >  > CB350 > CB550 > CB650 > CB450 (Table S2). The MB exhibited a significantly higher adsorption capacity than CB, suggesting the adsorption of Cd was critically affected by the feedstock. Specifically, there may be several reasons for the higher adsorption capacity of MB as following: (1) The properties of biochars showed that the pH of MB was much higher than that of CB. According to Wang et al. (2018), the pH could influence the charge on the surface of material and the degree of ionization and speciation of Cd. When the pH > zero point of charge of biochar, the surface of biochar was deprotonated and the electrostatic attraction occurred between adsorbent and Cd (Wang et al. 2018). (2) The high ash content of MB may provide alkali ions such as PO43−, CO32− which enhanced the formation of Cd precipitation. Xiaoqiang Cui et al. (2016) investigated the Cd adsorption on biochar and confirmed the presence of Cd precipitation based on the XPS analysis (Cui et al. 2016). (3) Since MB has the higher CEC values (Table 1) than CB, metal ions are reserved on the MB through complexes with carboxyl and hydroxyl groups. These metals could be exchanged by Cd through cation exchange during the adsorption process, which promoted the adsorption capacity of MB. For CB, the Fig. 2b shows that the CB350 had a similar capacity with CB650 and C550, higher than CB450. The different adsorption capacities in CB350 and CB450 were probably attributed to the different pH values. Table 1 shows that the pH value of CB350 is 6.37, which is higher than that of CB450. The previous study proved that the high pH of biochar would promoted adsorption of Cd. Additionally, the ash content and CEC of CB350 were also higher than those of CB450. The high ash content may provide alkali ions which resulted in the formation of Cd precipitation. The high CEC was favorable for Cd exchanging with cations on the surface of biochar. Thus, CB350 had higher adsorption capacity than CB450.

The maximum adsorption capacity of AT on biochars was in the range of 238.05–538.74 mg/kg. Different from Cd adsorption, the MB did not exhibit a stronger adsorption capacity than CB, indicating the type of solutes affected the adsorption behavior of biochars. The 1/n values of AT adsorption from Freundlich model exhibited a transition from 0.72–0.89 to 0.50 when the pyrolysis temperature was increased to 650℃, which suggested an increasing non-linearity for AT adsorption at high temperatures. Previous studies found that most biochars are not fully carbonized (Baoliang Chen et al. 2008). Specifically, the carbonized organic matter (condensed domain) is expected to behave as an adsorbent and noncarbonized organic matter (amorphous domain) as a partition phase (Chiou and Kile 1998). The decreased non-linearity for AT adsorption was probably attributed to the amorphous domain of biochar generating linear isotherms due to partitioning. As listed in Table S2, 1/n values below 650 ℃ were larger than 0.7, probably due to the existence of noncarbonized organic matter at the temperature ranging from 350 to 550 ℃, However, with the pyrolysis temperature increased to 650 ℃, the 1/n values were significantly decreased, probably due to the increased carbonization of amorphous domain in biochar at 650 ℃ to produce charcoal-like materials. This was in line with the lowest H/C values for biochars at 650 ℃ in elements analysis.

3.3 Quantify the influence of physico-chemical properties of biochars on Cd and AT adsorption

According to the isotherms in Figs. 2 and 3, the adsorption capacity of biochar was markedly influenced by the feedstock and temperature. To elucidate and predict the controlling factors in adsorption of heavy metals and organic solutes onto woody agricultural waste biochars, the relationships between adsorption parameters (qm) and physiochemical properties were investigated by using principal component analysis (PCA) and pearson correlation analysis. PCA was an efficient multivariate data analysis technique and could be used to reduce the dimensionality of large sets of data to streamline the representation of the data field in question (Pathirana et al. 2019). In this study, the raw data matrix consisted of 8 biochars and 14 variables including adsorption capacity of AT and Cd, pH, total organic carbon (TOC), cation exchange capacity (CEC), ash content, specific surface area (SSA), total pore volume (TP), micropore volume (MV), average pore diameter (AP), zeta potential (ZP), O/C, H/C and (O + N)/C). The PCA results are shown in Fig. 4. The first two PCs accounted for 82.9% of the total variance. Moreover, PCA showed an cute angle between qm of Cd and ash content, suggesting strong correlation between ash content and metal adsorption capacity. Pearson correlation analysis in Table S3 confirmed that ash content had high positive correlation with Cd adsorption capacity, which was consistent with the results of PCA. The strong correlation between Cd adsorption capacity and ash content implied that precipitation was an important mechanism for metal adsorption (Park et al. 2016; Zhao et al. 2013). SSA did not have a significant correlation with the Cd adsorption capacity. A similar phenomenon was found by Chaamila et al. (2019) who reported that adsorption is a surface phenomenon, SSA alone does not exert a prominent influence on the adsorption of metals, if materials with different functional group densities and same particle size are used (Pathirana et al., 2019).

Fig. 4
figure 4

Principle component analysis (PCA) biplot consisting of physico-chemical properties and adsorption capacities of biochars. SSA specific surface area, TP total pore, AP average pore diameter, MV micropore volume, CEC cation exchange capacity, TOC total organic carbon

Full size image

Figure 4 also shows the relationship between AT adsorption capacity and biochars properties. An acute angle between AT adsorption capacity and total pore volume was observed, which indicated strong correlations. Pearson correlation analysis (Table S3) confirmed the significant correlations between qm of AT and MV and TP. This suggested and that SSA the AT adsorption was related to the biochar porosity. Pore filling might affect the AT adsorption.

3.4 Modeling the Cd and AT adsorption

Partial least square (PLS) model was used to predict the linear relationship between properties of biochars and adsorption capacity. The prediction model in this study was developed by utilizing qm values of Cd and AT along with 14 parameters of biochars with PLS regression methods reported by Dengmiao Cheng et al. (2018). PLS of normalized values for 8 data in the training set was achieved using of the 14 inputs. To assess the importance of each physicochemical parameter with model, the variable importance in the projection (VIP) values were summarized in Table S4. In PLS analysis, the parameters with a larger VIP (> 1) have an above average influence on the results of model. In the current study, parameters whose VIP were larger than 0.8 were chosen to model the Cd and AT adsorption. The equations were obtained as following:

$${\text{q}}_{{\text{m}}} \left( {{\text{Cd}}} \right) = - {2257}.0{8} + {215}.{\text{24pH}} + {177}.{7}0{\text{ash}} + {46}.{\text{17CEC}} + 0.{\text{58ZP}} + {5}0{129}.{\text{19TP}} + {92}.{\text{26AP}} - {81}0.{\text{68H/C}} - {15}.{\text{85TOC}}$$
(5)
$${\text{q}}_{{\text{m}}} \left( {{\text{AT}}} \right) = {295}.{7}0 + {1}0.{\text{27pH}} + {4}.{\text{81ash}} + 0.{\text{85CEC}} + {2}0.{\text{93BET}} + {1}0{7671}.{\text{57MV}} + {76}0{1}.{\text{68TP}} + {3}.{\text{39AP}} - {119}.{\text{56H}}/{\text{C}} - {1}.{\text{36TOC}}$$
(6)

Based on the two models, the plots of the observed against predicted qm values in different biochars are described in Fig. 5. The plots gathered round the 1:1 line, illustrating the models were appropriate for predicting the data obtained from biochar adsorption. Moreover, Table S5 compares the results acquired by the PLS models. The R2 value of AT was slightly higher than that of Cd, and the MSE and RMSE of AT was smaller than those of Cd. As a result, PLS model was more suitable for AT adsorption than Cd adsorption.

Fig. 5
figure 5

The line fitting from the plots of the observed versus predicted qm values a Cd adsorption; b AT adsorption

Full size image

3.5 Possible Cd and Atrazine adsorption mechanism

The adsorption of Cd onto biochars is mainly via the following mechanisms: (1) metal precipitation (2) cation exchange between protons or metals; (3) complexation of functional groups with metals (Tong et al. 2011; Zhao et al. 2020). In the current study, the PCA analysis showed the strong correlations between Cd adsorption capacity and ash content, implying that the Cd precipitation probably affected the Cd adsorption process. The high ash content could provide alkali ions such as PO43+, CO32−, SiO42−. These species could form precipitation with Cd. To further explore this mechanism, XRD spectra of biochars before and after Cd adsorption were performed as shown in Fig. 6. It was observed that new characteristic diffraction reflections were indexed to the precipitation of CdH2P2O7 and Cd3P6O18. This implied that Cd precipitation with phosphate may be the key factor for Cd adsorption onto biochars. Similar result was also noted in lead (Pb) adsorption onto P-rich biochar (Yan et al., 2020), where the formation of hydroxypyromorphite (Pb–P–Ocompound) was the main Pb adsorption mechanism. In order to confirm this result, the phosphorus (P) concentration of biochars was determined and listed in Table 1. The relationship between P concentration and Cd adsorption capacity was analyzed as shown Fig. 7. A significantly positive correlation between P concentration and Cd adsorption capacity (R2 = 0.93, P < 0.01) indicated that Cd precipitation with phosphate probably dominated the Cd adsorption mechanism. Meanwhile, XPS spectra of biochar before and after Cd adsorption were employed to quantify the different P forms present on the surface (Fig. 7). It was found that the binding energy of P2p peak was changed from 130.6 to 129.6 eV after Cd adsorption. This further implied that the Cd precipitation with phosphate played a critical role in Cd adsorption on biochar. On the other hand, functional groups such as carboxyl groups could form complexation with Cd which promoted the adsorption capacity (Li et al. 2017; Wang et al. 2018; Zhang et al. 2015a). In order to reveal the influence of functional groups on Cd adsorption, the C1s peaks of biochar were deconvoluted into five peaks including 281.04 eV, 282.21 eV, 284.12 eV, 289.43 eV and 292.29 eV which were related to C=C bonds, hydroxyl groups (C–OH), carbonyl groups (C=O) and carboxyl groups (COOH) (Fig S4) (Inglezakis and Poulopoulos 2006). After Cd adsorption, the positions of these peaks did not changed. Meanwhile, the FTIR spectra of biochar (Fig S5) showed that the characteristics spectral bands at 1600 cm−1 and 1386 cm−1 remained unchanged before and after Cd adsorption. This indicated the complexation between Cd and functional groups on the surface of biochar was not the main driving force for Cd adsorption.

Fig. 6
figure 6

The XRD spectrum of biochar before and after Cd adsorption

Full size image
Fig. 7
figure 7

The relationship between total P concentration and Cd adsorption capacity (a), The binding energy of P 2p/5 before (b) and after (c) Cd adsorption onto biochar

Full size image

Because AT was a weak base with pKa of 1.68, it existed as an non-ionic organic compound at the pH in this study. Thus, the electric interaction became weak and the surface negative charge on biochar was not liable for the AT adsorption. Figure 4 confirmed the weak correlation of AT adsorption capacity with pH values and ash content. Previous studies reported that adsorption of AT onto biochars mainly involved two main mechanisms: (1) amorphous domain generates linear isotherms due to partitioning; (2) condensed aromatic domain generates non-linear isotherms due to a specific interaction or adsorption (Wang et al. 2020). The presence of moderately nonlinear (1/n = 0.4 ~ 0.5) and low nonlinear (1/n = 0.7 ~ 0.9) implied both partitioning and adsorption contributed to the AT adsorption onto biochars (Baoliang Chen et al. 2008). Furthermore, the moderately nonlinear isotherms were obtained at biochar from 650 ℃ pyrolysis temperature. This is because the high temperature increased carbonization of biochar to produce charcoal-like materials with higher porosity and surface where the specific adsorption play dominant role in biochar. The PCA analysis showed a strong correlation between AT adsorption capacity and TP. This indicated a pore-filling mechanism was involved in AT adsorption (Zhang et al. 2013, 2015b). In addition, adsorption capacity of AT onto carbon materials is always influenced by H-bonding and π–π electron donor–acceptor interactions (Lonappan et al. 2018; Sadasivam and Reddy 2015). To reveal this mechanism, comparison of the FTIR spectra of biochar before and after AT adsorption were performed as shown Fig S6. It showed that the peak for biochar at 1600 cm−1 moved to 1634 cm−1 after AT adsorption, suggesting that O-containing functional groups such as carboxyl groups might form H-boding interactions with AT molecule. Specifically, the heterocyclic ring of AT was electron donor and acceptor which could form H-bonding with H and O on the surface of biochars. H-bonding interactions were also found in biochar derived from animal product, peanut-shell, raw sludge (Liu et al. 2015; Wang et al. 2020; Zhang et al. 2015b). Based on the above analysis, the Scheme 1 depicted the difference in adsorption mechanism between organic pollutants and heavy metals.

Scheme 1
scheme 1

Adsorption mechanism of Cd and AT onto biochar

Full size image

3.6 Environmental significance

Agricultural wastes have increased more than three times over the last 50 years because of the expansion of soil and the accelerated growth of population (Duque-Acevedo et al. 2020). Reclamation of agricultural wastes is economic and ecofriendly due to the unique chemical composition, availability in abundance, renewability, and low cost. In this study, two woody agricultural wastes including mulberry waste and cinnamon waste were reclaimed and pyrolyzed into biochar. The Kf values of Cd and AT onto biochars were located in range of 161.15–2533.57 L/kg and 14.42–120.30 L/kg, respectively. These values were higher than the literature values of Cd (0.04-40L/kg) in soil and AT (4.02 L/g) in soil (Martin et al. 2012; Usman 2008). It implied woody agricultural waste had great potentials as remediation tool for amendment of Cd and AT polluted soils. Furthermore, the model fitting and mechanism study will help to devise an approach or strategy for recycling the waste woody agricultural biomass and converting them to effective biosorbents for treatment of water and soil contaminated with potentially toxic compounds.

4 Conclusions

In this study, two woody agricultural waste- derived biochars including mulberry waste biochar and cinnamon waste biochar were synthesized at four pyrolysis temperatures. The characterization of biochars showed that the physical and chemical properties of biochars were influenced by the type of raw materials and the pyrolysis temperature. The BET surface area was increased with increasing pyrolysis temperature. However, hydrophilicity and polarity were decreased with increasing pyrolysis temperature. Moreover, the adsorption isotherms of Cd and atrazine onto these biochars were performed. Freundlich model and Langmuir model described the experimental data well. The adsorption capacity values of Cd adsorption on MB were higher than that on CB. For all feedstocks, the biochars at 650 ℃ pyrolysis temperature gave the highest adsorption capacity. Besides, the correlation between adsorption capacity and physico-chemical properties of biochars were analyzed. The strong correlation between P concentration and Cd adsorption capacity indicated that metal precipitation was involved in the Cd adsorption. The strong correlation between pore volume and AT indicated that a pore-filling mechanism contributed to the AT adsorption. In addition, PLS model was appropriate for predicting the data obtained from AT adsorption.