3.1 Material characterizations
The Raman spectra reflect the carbon structural defects of carbon in biochar (Fig. 2). The characteristic peaks are located at approximately 1,340 cm− 1 (D band) and 1,582 cm− 1 (G band), which correspond to the carbon structure defects and the sp2 hybridization oscillations of the graphite microcrystals (Li et al. 2022 b; Sun et al. 2020; Xu et al. 2022). The ID/IG ratio reflects the graphitization of the carbon structure. The ID/IG ratios of BC, BCNP-Fe, BCHNP-Fe, and BCP-Fe are 0.63, 1.12, 1.14, and 1.34, respectively, indicating that Fe, N, and P codoping produces more carbon graphitization on biochar. The carbon structure contains more sp2 C, and the amount of sp2 C is related to the π-electron acceptance capacity of the carbonaceous adsorbents (Mei et al. 2021). The hydroquinone molecule possesses two hydroxyl groups (-OH) with electron-donating properties, which significantly increases π-electron density. (Xu et al. 2022).
Please insert Fig. 2 here.
SEM images present the surface morphology and pore microstructure of biochar (Fig. 3). Compared with the smooth surface of original corn straws (Fig. 3d), BCNP-Fe (Fig. 3a), and BCHNP-Fe (Fig. 3b) are loaded with a larger specific surface area. Porous structures, BCP-Fe (Fig. 3c) characterized by numerous spherical particulates, are observed on the surface due to the formation of iron oxide particles (Li et al. 2022 b). Irregular rods and fragments appear on BCNP-Fe and BCHNP-Fe surfaces, indicating a condensation reaction between ammonium phosphate or diammonium hydrogen phosphate and corn straw components in the pyrolysis process (Dong er al. 2020; Xu et al. 2020).
Please insert Fig. 3 here.
The functional groups of biochar are determined by infrared spectroscopy (Fig. S1). Biochar has several common features from the infrared spectral analysis. The -OH/-NH2 peak near 3,450 cm− 1 and spectral bands at 2,900-3,200 cm− 1 are attributed to the stretching vibrations of -CH3 and -CH2. N-H is at 2,365 cm− 1, with aromatic C = O near 1,740 cm− 1. The sharp peak at 1,620 cm− 1 may be attributed to the carbonyl stretching vibration, including C = N at 1,395 cm− 1 and C-H at 800 − 500 cm− 1. Fe-O appears near 670 cm− 1 of the FTIR spectrum after the Fe modification. P-O-P stretching vibrations appear at 1,060 cm− 1 after N/P doping. C = N appears near 1,400 cm− 1 (Li et al. 2020 a; Li et al. 2022 b; Qu et al. 2022).
The two diffraction peaks observed at about 26 and 44° in modified biochar correspond to the anamorphic carbon structure and the graphitic carbon structure, respectively (Fig. S2) (Li et al. 2020 b; Xu et al. 2020). H3PO4 activation increases the amorphous carbon structure for BCP-Fe, which facilitates crystal synthesis (Zhao et al. 2022). The crystal structure of Fe2.95(PO4)2(OH)2 is synthesized in BCNP-Fe and BCP-Fe to achieve the dispersion of Fe within the carbon skeleton combined with heter. This facilitates the generation of free radicals in biochar, the electronic structure, and the adsorption of hydroquinone.
The surface area and pore structure of biochar are further determined by N2 adsorption-desorption measurement (Fig. S3). The pore distribution analysis reveals that the predominant pore size of biochars ranges from 2 to 5 nm, which confirms its mesoporous. N2 adsorption isotherm of biochar belongs to type IV according to the IUPAC classification. An obvious lag line exists under high pressure, indicating the mesoporous structure in biochar (Mei et al. 2021; Zhao et al. 2022). The specific surface areas of BCNP-Fe, BCHNP-Fe, and BCP-Fe are calculated to be 23.582, 111.000, 125.880, and 262.050 m2/g, respectively (Table 1). Original biochar lacks a microporous structure. It is modified by heteroatoms embedded in the carbon network to form mesopores and micropores (Li et al. 2020 a; Li et al. 2020 b). The thermal decomposition of ammonium phosphate, diammonium hydrogen phosphate, and phosphoric acid generates gases during the modification process, which destroys the carbon structure and creates plenty of pores. The catalytic effect of iron doping during the thermal decomposition process contributes to the formation of pore structures.
Table 1
Surface area and pore parameters of BC, BCNP-Fe, BCHNP-Fe, and BCP-Fe.
Adsorbent
|
SSA
|
Smicro
|
Smeso
|
Vtotal
|
Vmicro
|
Vmeso
|
Average pore diameter
|
(m2/g)
|
(m2/g)
|
(m2/g)
|
(cm3/g)
|
(cm3/g)
|
(cm3/g)
|
(nm)
|
BCNP-Fe
|
110.00
|
71.19
|
19.24
|
0.13
|
0.03
|
0.10
|
4.84
|
BCHNP-Fe
|
125.88
|
75.31
|
21.44
|
0.11
|
0.03
|
0.08
|
3.64
|
BCP-Fe
|
262.05
|
177.57
|
43.96
|
0.20
|
0.07
|
0.13
|
3.08
|
BC
|
23.58
|
0.00
|
16.80
|
0.06
|
0.00
|
0.06
|
10.56
|
Please insert Table 1 here.
Table 2 shows biochar’s elements. C content exhibits a noticeable decrease, while N and O content demonstrates increased BCNP-Fe, BCHNP-Fe, and BCP-Fe in comparison to BC. Modified biochar materials exhibit higher O/C and (O + N)/C ratios, indicating that the modification leads to the incorporation of groups with enhanced hydrophilicity and increased polar functional groups (Li et al. 2020 a; Li et al. 2022 b). The aromaticity of biochar, decreasing with the increased H/C atom ratio,is increased to enhance the π-π bonds of aromatic molecules on the biochar surface. Besides, it weakens the π-π bonding between biochar and pollutants (Mei et al. 2021). The modification introduces Fe into biochar according to the spherical particles in the SEM images and the variation of Fe content in the elemental analysis table.
Table 2
Elemental compositions of BC, BCNP-Fe, BCHNP-Fe, and BCP-Fe.
Adsorbent
|
Ash
(%)
|
Elemental composition (%, mass-based)
|
H/C
|
O/C
|
(O + N)/C
|
C
|
H
|
O
|
N
|
S
|
Fe
|
BCNP-Fe
|
45.27
|
23.23
|
0.60
|
29.57
|
1.18
|
0.14
|
16.29
|
0.03
|
1.27
|
1.32
|
BCHNP-Fe
|
22.98
|
31.20
|
0.98
|
43.39
|
1.14
|
0.32
|
3.44
|
0.03
|
1.39
|
1.43
|
BCP-Fe
|
40.76
|
32.18
|
0.87
|
24.78
|
1.10
|
0.31
|
3.64
|
0.03
|
0.77
|
0.80
|
BC
|
53.90
|
37.17
|
0.72
|
7.25
|
0.80
|
0.16
|
0.90
|
0.02
|
0.20
|
0.22
|
Please insert Table 2 here.
3.2 Adsorption experiments
3.2.1 Adsorption kinetics
Models including pseudo-first-order, pseudo-second-order, intraparticle diffusion models, and film diffusion models were employed to investigate the adsorption performance of biochar samples at the initial hydroquinone concentration of 50 mg/L within 240 min. The adsorption rate of hydroquinone reached equilibrium within 180 min (Fig. 4). The maximum adsorption capacities of hydroquinone by BC, BCNP-Fe, BCHNP-Fe, and BCP-Fe were 83.64, 94.872, 92.928, and 96.984 mg/g, respectively (Table S2).
Please insert Fig. 4 here.
For BC, the pseudo-second-order model (Fig. 5(b)) with a higher correlation coefficient was more appropriate to describe the adsorption process in comparison with the pseudo-first-order model (Fig. 5(a)). Both pseudo-first-order and pseudo-second-order kinetic models were suitable to reveal the adsorption behavior based on the higher correlation coefficient (R2) for BCNP-Fe, BCHNP-Fe, and BCP-Fe (Table S2). The adsorption process probably combined physical and chemical adsorption (Li et al. 2022 b; Ren et al. 2021).
An intra-particle diffusion model was used to evaluate the mass transfer resistance in the adsorption of hydroquinone onto biochar. The adsorption process of hydroquinone on the BC, BCNP-Fe, BCHNP-Fe, and BCP-Fe was comprised of three stages (Fig. 5(c)). The noticeable rise was attributed to the film diffusion process for hydroquinone from the solution to the external biochar surface at the first stage. The second stage was conducive to intra-particle diffusion, with hydroquinone gradually entering the internal biochar surface through the pore structure. The last stage was the equilibrium state of adsorption, and the intraparticle diffusion rate decreased as the diffusion resistance increased (Sun et al. 2022; Xia et al. 2022). The fitted curve for all three stages was not through the origin, indicating that the adsorption of hydroquinone was influenced by various adsorption processes (Table S3).
The film diffusion models of BCNP-Fe, BCHNP-Fe, and BCP-Fe were characterized by high R2 values (R2 > 0.983) and small intercepts, indicating that the boundary liquid diffusion of biochar plays a major role in the adsorption of hydroquinone (Fig. 5(d) and Table S4).
Please insert Fig. 5 here.
3.2.2 Adsorption isotherms
Figure 6 shows the adsorption isotherms with the initial concentration from 10 to 100 mg/L at 25°C. The Langmuir, Freundlich, Temkin, and D-R isotherm models are employed to analyze the fitted data (Fig. 7 and Table S5), which further investigates the distribution characteristics between adsorbents and adsorbates (Li et al. 2022 b; Sun et al. 2022; Xia et al. 2022).
Please insert Fig. 6 here.
The Langmuir model (Fig. 7a) postulates a monolayer and homogeneous surface of the adsorbent, while the Freundlich model (Fig. 7b) emphasizes heterogeneous multilayer adsorption on the adsorbent surface (Xia et al. 2019). The Temkin model (Fig. 7c) simulates the relationships between adsorbents and adsorbents. The Dubinin-Radushkevich (D-R) model (Fig. 7d) differentiates the chemisorption or physisorption of the process. The Langmuir model is more suitable for fitting the adsorption curves of BC, BCNP-Fe, and BCP-Fe (Table S5). The correlation coefficient is higher than that of the Freundlich model, indicating that monolayer adsorption predominates in the observed adsorption behavior. The adsorption process of BCHNP-Fe is well-fitted with both Langmuir and Freundlich models because monolayer and multilayer adsorption coexists.
Separation factor KL is all less than 1 in the Langmuir model due to favorable adsorption. 1/n in the Freundlich model is between 0 and 1, suggesting that the adsorption process is viable as well. βT in the Temkin model is determined by the variation of adsorption energy, and values below 1 indicate endothermic adsorption reactions (Liu et al. 2019; Sun et al. 2022). D-R models obtain adsorption energy E less than 8 kJ/mol due to the physisorption process.
Please insert Fig. 7 here.
3.2.3 Adsorption thermodynamics
The effects of different temperatures and the adsorption of hydroquinone are investigated by the standard Gibbs free energy change (\(\varDelta {G}^{^\circ }\)), enthalpy change (\(\varDelta {H}^{^\circ }\)), and entropy change (\({\varDelta S}^{^\circ }\)) (Li et al. 2022 a). Table S6 shows the relevant thermodynamic parameters at different temperatures. Negative \(\varDelta {G}^{^\circ }\) demonstrates the spontaneous adsorption process of BC, BCNP-Fe, BCHNP-Fe, and BCP-Fe on hydroquinone. Moreover, absolute \(\varDelta {G}^{^\circ }\) increases with the increased temperature, indicating a preference for adsorption at higher temperatures.
Positive \(\varDelta {H}^{^\circ }\) reveals that hydroquinone adsorption is endothermic due to the heightened temperature inducing a more vigorous intermolecular motion. It increases the interaction of hydroquinone molecules on the surfaces of BC, BCNP-Fe, BCHNP-Fe, and BCP-Fe. Meanwhile, \(\varDelta {H}^{^\circ }\) less than 40 kJ/mol shows predominantly physical adsorption with chemisorption. \({\varDelta S}^{^\circ }\)>0 indicates an increase in disorder or randomness at the solvation interface (Li et al. 2022 a; Wang et al. 2019; Yao et al. 2021).
3.2.5 Effect of pH
Solution pH is a critical element for the efficiency of removing hydroquinone. It affects the form of the adsorbed contaminant and the surface charge of the adsorbent (Liu et al. 2021; Xia et al. 2022; Yan et al. 2022).
The following equations show the ionization process of hydroquinone in aqueous solutions (Bianco-Martinez et al. 2009; Shi et al. 2018; Suresh et al. 2011).
where HQ(OH)2 is neutral hydroquinone; HQ(OH)(O)− is hydroquinate; HQ(O)2− is hydroquinate dianions (Shi et al. 2018; Suresh et al. 2011). pKa1 and pKa2 for hydroquinone are 9.9 and 11.6, respectively (Bianco-Martinez et al. 2009).
The work investigates the effect of initial pH (1–9) on the adsorption process. The adsorption capacity remains at 89 mg/g in acidic conditions for BCHNP-Fe (Fig S4). The availability of additional free H+ facilitates the formation of hydrogen bonds in BCHNP-Fe, the doping of N, P heteroatoms and the larger specific surface area and richer porous structure improving the adsorption of hydroquinone under acidic conditions (Shi et al. 2018; Suresh et al. 2011). However, transient increases at pH 2 are observed for BCNP-Fe, BCP-Fe, and BC potentially attributed to enhanced adsorption capacity from the zero-point charge of biochar.
The removal efficiency of BCNP-Fe, BCP-Fe, and BC for hydroquinone increases at pH 2–5, and hydroquinone exists in its neutral form. The influence of hydrogen bond formations between biochar and hydroquinone is not obvious, suggesting that van der Waals force dominates the adsorption process of hydroquinone (Li et al. 2013; Shi et al. 2018). When pH > 5, accumulated OH− of hydroquinone affects the π-π dispersion interaction between hydroquinone and occupies the adsorption sites of biochar, which reduces the adsorption capacity (Li et al. 2013; Shi et al. 2018; Suresh et al. 2011). Therefore, pH 5 is chosen as the optimum for adsorbing hydroquinone solutions.
3.2.6 Effect of carbon dosages
The adsorbent dosages of 50 mg/L hydroquinone solutions are 10, 20, 30, 40, 50, and 60 g, which are used to investigate the effect of adsorbent dosages on adsorption. The removal rates of BCNP-Fe and BCP-Fe remain after the addition of 30 g solutions (Fig S5). However, the carbon content of BCHNP-Fe has little effect on the removal rate due to the doping of BCHNP-Fe with N, P heteroatoms as well as larger specific surface area and richer porous structure. (Shi et al. 2018).
3.3 Removal mechanism
Further analysis was performed on the properties of biochar to investigate the mechanism of biochar on hydroquinone (Fig. 8). Two hydroxyl groups (− OH) in hydroquinone facilitated the formation of hydrogen bonds through the interaction between -OH groups in organic substances and oxygen-containing groups in carbonaceous compounds (Xu et al. 2022). Consequently, the -OH substitution on hydroquinone and the hydroxyl/carboxyl groups on the biochar surface formed hydrogen bonds.
Please insert Fig. 8 here.
The -NH2 bands on the BCNP-Fe, BCHNP-Fe, and BCP-Fe surface varied from 3,155, 2,969, and 2,927 cm− 1 to 3,173, 3,154, and 3,156 cm− 1 by comparing hydroquinone adsorption. The hydrogen bonds were formed between the -NH2 group and the phenol hydroxyl group (Fig. S1). Meanwhile, the hydroxyl group donating electrons contributed to high π-electron density and acted as a π-electron donor. Co-doping facilitates the provision of electron-donor and acceptor regions in biochar, with π-π interaction serving as primary driving force for adsorbing organic materials onto biochar (Mei et al. 2021; Shi et al. 2018; Xu et al. 2022). It could be also verified based on the ID/IG ratio of Fig. 2.
The C = C stretching vibration shifted from the front after the adsorption of hydroquinone according to the infrared spectrogram of biochar (Fig. S1). It was attributed to the π-π interaction between the benzene ring of hydroquinone and biochar. In Tables 1 and 2, the adsorption of hydroquinone by biochar was influenced by the BET specific surface area, total pore volume, and elemental analysis. Biochar had excellent hydrophilicity, large specific surface area, and total pore volume, which could remove hydroquinone (Xu et al. 2022). Mesoporous diffusion also played a crucial role in removing hydroquinone.π-π bonding between the aromatic ring of hydroquinone and the aromatic molecules on the biochar surface was reduced by the decreased aromaticity of biochar. Besides, the π-π bond interaction between biochar and hydroquinone was weakened. Fe-O bonds were shifted after the adsorption of hydroquinone, indicating that surface complexation played a role in adsorbing hydroquinone (Fig. S1) (Mei et al. 2021).
3.4 Desorption and reusability
Adsorption-desorption cycle experiments were performed utilizing warm water at 60°C as a resolving agent to evaluate the regeneration performance of BC, BCNP-Fe, BCHNP-Fe, and BCP-Fe. The removal reusability of BC, BCNP-Fe, and BCP-Fe gradually decreased after recycling five times (Fig. 9), with the adsorption capacity of BCHNP-Fe for hydroquinone remaining above 90%. The enhanced binding ability of hydroquinone to BCHNP-Fe, attributed to large amounts of hydroquinone on the BCHNP-Fe surface was eliminated during the desorption process, whereas for BC, BCNP-Fe, and BCP-Fe probably because of the occupation of larger adsorption sites by hydroquinone. The result indicates that BCHNP-Fe has excellent reusability properties for adsorbed hydroquinone (Shi et al. 2018).
Please insert Fig. 9 here.