Adsorption of phthalic acid esters (PAEs) on chemically aged biochars

2.1 Sorbates

Analytical grade reagents DMP (>99.5%), DEP (>99.5%) and DBP (>99.5%), were obtained from Beijing Chemicals Reagent Company (Beijing, China). Table 1 shows the selected physico-chemical properties of DMP, DEP and DBP. The selection of these three compounds was made in view of their different aqueous solubility, non-volatile nature, and a gradient in hydrophobicity (log K_ow) as well as molecular size.

Table 1:

Selected physico-chemical properties of the phthalic acid esters (PAEs).

Structure	Chemicals	Abb.	MFa	Ma	Csb	ρa	log Kowa	MLc	MWc	SAd	HDd	HAd	HBd
	Dimethyl phthalate	DMP	C10H10O4	194.2	5220	1.18	1.53	1.04	0.366	392.2	0	8	8
	Diethyl phthalate	DEP	C12H14O4	222.2	1080	1.12	2.39	1.19	0.405	414.0	0	8	8
	Dibutyl phthalate	DBP	C16H18O4	278.3	11.2	1.05	4.61	1.45	0.409	580.0	0	8	8

^aData from Ref. [22].

^bData from Ref. [23].

^cData from Ref. [24].

^dData from Ref. [25].

MF, Molecular formula; M, molar mass (g mol^-1); C_s, water solubility; ρ, density (g cm^-3); log K_ow, octanol/water partition constant; ML, molecular length (nm); MW, molecular width (nm); SA, solvent accessible surface area (Å); HD, hydrogen bonding donor; HA, hydrogen bonding acceptor; HB, hydrogen bond forming ability.

2.2 Preparation of biochar and oxidized biochars

For the preparation of biochars, peanut shell biomass was obtained from the local market, Kunming, Yunnan, China. Two kinds of biochars were obtained by the pyrolysis of peanut shell as reported by Keiluweit et al. [26]. The samples were air-dried for 2 days and subsequently oven-dried overnight at 70–80°C, and then biomass feedstock was ground by a high speed rotary machine to obtain milled biomass material. The milled peanut shell was placed in a ceramic pot, purged with N₂ and then covered tightly with a lid. The pyrolysis was carried out under continuous N₂ flow, at specific temperatures (300°C and 500°C), in a muffle furnace (Beijing Instrument and Equipment Co. Ltd., China) under oxygen limited atmosphere for 4 h, to avoid the calcination of the biomass in the presence of oxygen. These two temperatures were chosen to provide a clear distinction between the categories of biochars and their description of the sorption behavior. In order to produce more solid residue (biochar), the heating rate was controlled at 5°C min^-1 for slow pyrolysis. The biochar samples were washed with deionized water (four times) to remove the water-soluble inorganic materials and then they were oven-dried. The charred solids were passed through a 100 mm sieve to produce the final biochar samples. Peanut shell biochars (original) were referred to as B300 and B500, with numbers referring to the pyrolysis temperatures.

For oxidation, a mixture of concentrated HNO₃ and H₂SO₄ was selected to oxidize the obtained peanut-shell biochars. Biochars were oxidized with an HNO₃/H₂SO₄ [1:3 (v/v)] using a modified method [27]. Each biochar sample (5 g) was immersed in Teflon line autoclaves containing 400 ml of a HNO₃/H₂SO₄. Then, these autoclaves were put in an oven and the temperature was set at 70°C and maintained for 6 h, to sustain an oxidation environment. Since the reaction was highly exothermic, the samples were allowed to cool down at room temperature. To remove residual acid, oxidized biochars were first drained through a mesh sieve and then washed repeatedly with deionized water for 1 h until the pH was stabilized at around 4, and stored for further use. The oxidized biochar samples are denoted as B300-Ox and B500-Ox.

2.3 Biochar characterization

Biochar characteristics were analyzed with a CHN analyzer for bulk elemental composition (elemental analyzer, Elementar, MicroCube, Germany), X-ray photoelectron spectroscopy (XPS) for surface composition (PHI 5500, X-ray photoelectron spectrometer), Brunauer-Emmett-Teller SA, and functional groups by diffuse reflectance infrared Fourier-transformed (DRIFT) spectroscopy, (Varian 640-IR, USA). In brief, SA and pore volume (PV) measurements were obtained from N₂ at 77 K. For bulk elemental compositions, a 2 mg sample was added into the CHN analyzer. The temperatures of combustion and reduction tubes were 1150°C and 850°C, respectively. The combustion tube was used only at 1150°C for O mode analysis. For XPS analysis, the pressure ranges were set as fast entry chamber 2×10^-6 mbar, preparation chamber 4×10^-8 mbar and sample analysis chamber 4×10^-9 mbar. The setting for high transmission was used for the analysis at 90° electron take off angle for normal non-charging samples. The analyzer slit width was set as 0.8 mm. The MULTIPAK software was used for data acquisition and data analysis. The DRIFT spectra were recorded in the 4000–400 cm^-1 region, with a resolution of 4 cm^-1 for biochar (0.5%) prepared in KBr pellets. Table 2, and Figures 1 and 2 represent the bulk and surface characteristics, SA, PV, XPS and DRIFT spectra, respectively.

Figure 1:

X-ray photoelectron spectra (XPS) of biochars and oxidized biochars.

Figure 2:

Diffuse reflectance infrared Fourier-transformed (DRIFT) spectra of biochars and oxidized biochars.

2.4 Scanning electron microscopy (SEM)/Energy dispersive X-ray spectroscopy (EDX)

The degree of surface oxidation due to chemical treatment was evaluated with the help of scanning electron microscopy (SEM)/EDX data. The samples were mounted onto aluminum stubs using double-sided carbon tape, gold coated with a sputter coater. Samples were analyzed with a Tescan, VEGA3 SBH (Czech Republic), equipped with an EDX Thermo Fisher Noran System 7, Thermo Fisher Scientific (Waltham, MA, USA). For each analysis, the voltage was set at 20 keV, while the working distance was 9 mm and the dead time for X-ray acquisition was between 20% and 25%. A color code was assigned for the elements; blue for carbon and white for oxygen. Dot maps were acquired at 180× magnification and 600 frames each to optimize electron detection from the surface of the biochar samples.

2.5 Batch experiment

Using a batch equilibrium technique, adsorption isotherms of DMP, DEP and DBP were obtained. The sorbates (DMP, DEP and DBP) were dissolved in methanol to serve as stock solutions, and then diluted to the background solution containing 0.02 m NaCl (to maintain a constant ionic strength) and 200 mg l^-1 NaN₃ (bio-inhibitor). The volume percentage of methanol was kept below 0.10% (v/v) to minimize the co-solvent effect. Adsorption experiments were conducted in 4 ml glass vials with Teflon-lined screw caps. The aqueous:solid ratios (v:w) were 4000:1 for each sorbent. All vials head space was minimal (<0.30 ml). Using 0.1 m NaOH or 0.1 m HCl solution, the pH was adjusted at 7.0±0.2 to attain apparent equilibrium. The vials were kept in the dark and shaken for 7 days with the help of a rotary shaker at room temperature. During this time period, all sorbates (DMP, DEP and DBP) were stable and no apparent degradation was noticed. All vials were centrifuged at 2500 rpm for 10 min, and 2 ml of each supernatant was subjected to the determination of DMP, DEP and DBP concentrations by high performance liquid chromatography (Agilent Technologies, 1260 series). The same concentration series of DMP, DEP and DBP solutions (without any sorbent), were run using the same conditions as the controls, showing that the loss of the initially added amounts of PAEs was <3%. Each concentration point including control (without biochar) was run in duplicate. Hence, the amount of DMP, DEP and DBP adsorbed by biochars was calculated by mass difference.

2.6 Ultraviolet/visible (UV/vis) complexation

In order to check the π-acceptor strength order of PAEs, ultraviolet/visible (UV/vis) spectra of test solutes in chloroform were obtained at room temperature on a UV/vis spectrometer (UV2600, Shimadzu, Japan). Pyrene was used as a π-donor, whereas PAEs (DMP, DEP and DBP) served as π-acceptors. Each solution, i.e. DMP, DEP and DBP was mixed with pyrene at a fixed concentration of 0.01 m, and then the solutions (after shaking for 1 h), were quantified by the UV/vis spectrometer.

2.7 Sorbate quantification

The concentrations of DMP, DEP and DBP in the supernatants were quantified at 228 nm by high performance liquid chromatography equipped with a reversed-phase C18 column (5 μm, 4.6×150 mm) and a UV detector. The mobile phase was 25:75 (v:v) of deionized water and acetonitrile. The flow rate was 1 ml min^-1 and temperature of the column was 35°C.

2.8 Data analysis

Adsorption isotherm fitting: Two different models were used in this work to fit the adsorption isotherms. The equations are given as follows:

where S_e (mg g^-1) is the solid phase concentration and C_e (mg l^-1) represents the aqueous phase concentration. K_F is the Freundlich adsorption parameter and n is the nonlinearity factor. S_L⁰ (mg g^-1) is the adsorption capacity, and b (l mg^-1) is the adsorption affinity parameter for Langmuir model, respectively.

Since the numbers of parameters used in the two models were not the same, the coefficient of determination (radj2) could not be compared directly. The adjusted (radj2) was calculated and then compared using following formula:

where m is the number of data points and p is the number of parameters in the fitting equation.

3.1 Biochar characteristics

The bulk and surface elemental compositions, atomic ratios and SA of biochars are listed in Table 2. With the increase of pyrolysis temperature, the bulk polarity and aromaticity of biochar decreased as indexed by the reduced H/C and [(O+N)/C] atomic ratios. The increased pyrolysis temperature accounted for increased C and ash content, whereas O and H contents decreased, suggesting an accumulated carbonization extent following decarboxylation and dehydration during pyrolysis [26]. The carbon content in B500 was higher than that of B300, which was consistent with previous studies, suggesting higher carbonization [26], [28]. Oxidation greatly influenced the bulk composition of biochars with an increase in O content and a reduction in ash contents. The increased O/C contents were also discovered during aging of biochars. The pronounced reduction in C and ash content was observed for B500. Overall, there was rise in bulk polarity of oxidized samples according to [(O+N)/C] index, with no significant effect on aromaticity index H/C.

The surface properties of biochars also varied due to oxidation as observed by surface contents (C, O and N). The major element in biochars was C, followed by oxygen, suggesting enrichment of oxygen on the surface. Oxidation treatment also introduced some functional groups to biochar surfaces such as carboxyl and phenolic groups, which was indicative of changes in surface elemental composition. This is in agreement with DRIFT spectra as shown in Figure 2. Therefore, the surface composition of oxidized samples showed a significant change in surface functionality. These surface functional moieties are assumed to become strong adsorption sites for water or other chemicals [18]. The difference in surface and bulk composition among differently treated biochars indicates their heterogeneous arrangements and structures.

With an increase in pyrolysis temperature, the SA of biochars also changed. The alteration and transformation in the elemental contents gave rise to the increase in SA. A decrease of SA was noticed after oxidation (Table 2). This suggests that oxidation (with H₂SO₄:HNO₃) was effective and damaged biochars, thus reducing the adsorption of N₂ by active sites. The pyrolysis temperature is a key parameter controlling the characteristics of SA and PV. At lower temperature, pores were undeveloped, and their blockage might take place due to volatile or aliphatic components, thereby resulting in reduced SA, whereas, volatile and aliphatic carbon evaporated at higher temperature and destruction of more pores occurred along with the formation of amorphous carbon. The oxidation of biochars reduced the SA of biochars because oxidizing agents such as HNO₃ can demolish the pores, but such agents are helpful for the addition of oxygen containing surface functional groups to surface.

DRIFT spectroscopy was used to identify the functionalization of biochar before and after oxidation. The DRIFT spectra of organic functional groups appear to be increased with oxidation and hydroxyl, alkyl and aromatic carbonyl groups were clearly shown. The chemical oxidation abated the relative intensity of hydroxyl, alkyl and aromatic carbonyl groups. The aliphatic carbon (C-H stretching of alkyl structure) was no longer clear at around 2922–2854 cm^-1 and 1734 cm^-1. After oxidation, carboxyl group contents were increased around 1500–1450 cm^-1 and 1350–1200 cm^-1. The peaks (at 1260–1200 cm^-1 and 1070–1030 cm^-1) were attributed to the stretch vibration of the carboxylic acid functional group (-COOH) and asymmetric stretch of carbohydrates (-COO-), respectively.

3.2 SEM/EDX analysis of biochars

The surface morphologies of the biochars and their oxidized products are shown in Figure 3. The surfaces of oxidized biochars were clean and it is easy to deduce that the inorganic matter on the surface was swept away during oxidation as depicted by SEM images. Remarkably, some floccules were formed on the surface of biochars after HNO₃/H₂SO₄ oxidation, probably due to the formation of oxidized matter on the surface of biochars, and also in line with O/C ratios. According to the EDX spectra and images of oxidized biochars, the oxidation occurred on the surface of biochars, because the oxygen contents were introduced mostly on the surfaces of biochars as observed by DRIFT and XPS elemental spectra. The higher temperature biochars were more stable and the disordered C could not rearrange during decomposition allowing the non-uniform growth of crystallites with turbostratic C. Therefore, the extensive growth of graphitic sheets become terminated and pinned by structural defects, thus, becoming the source of micropores.

Figure 3:

Scanning electron microscope (SEM) images and corresponding SEM/EDX graphs.

3.3 PAEs adsorption on biochars

Adsorption isotherms of dialkyl phthalates (DMP, DEP and DBP) on different biochars are shown in Figures 2 and 3, and the fitting parameters of Freundlich and Langmuir models are summarized in Table 3, respectively. Figure 4 compares the adsorption isotherms of PAEs on biochars based on oxidation, and Figure 5 compares the adsorption isotherms based on pyrolysis temperature. The Freundlich isotherm (FM) model was used to describe the isotherms, and FM fitted the adsorption isotherms very well as shown by a higher radj2 (0.992–0.998, Table 2). For a given type of PAEs, the coefficients K_F and n of the FM were different for different biochars.

Figure 4:

Adsorption of dimethyl phthalate (DMP), diethyl phthalate (DEP) and dibutyl phthalate (DBP) on biochars and oxidized biochars, as a function of oxidation (300°C left and 500°C right hand). Open and closed symbols (○, •) represent original and oxidized biochars, respectively.

Figure 5:

Adsorption of dimethyl phthalate (DMP), diethyl phthalate (DEP) and dibutyl phthalate (DBP) on biochars and oxidized biochars, as a function of temperature (original on left and oxidized biochars on right hand). Open and closed symbols (○, •) represent 300°C and 500°C biochars, respectively.

Sorption capacities of the biochars were compared with sorption coefficients K_d. In general, the single-point adsorption coefficients (log K_d) (calculated from FM) followed the order of DBP>DMP>DEP. Significantly, the PAEs sorption was not proportional to any of the parameters that could be positively correlated with the SA (N₂-SA), aromaticity, ash content, or pyrolysis temperature (Table 2).

DBP adsorption was higher than that of DEP and DMP, suggesting that hydrophobicity played an important role. The DMP adsorption was higher than that of DEP, thus indicating an opposite trend with their hydrophobicity, but was according to their polarity index. The variation in biochars surface properties is believed to be responsible for the difference in adsorption. The interaction effects can be easily estimated and tested by analysis of variance (ANOVA). The purpose of the ANOVA is to investigate which adsorption parameters significantly affect the performance characteristics and the contribution of each parameter on the adsorption efficiency. In addition, the Fisher’s ratio can also be used to determine which parameters have a significant effect on the performance characteristics. The F>Prob in ANOVA confirms that the adsorption of PAEs on biochars were significant factors. The sum of the squares used to estimate factors affect and Fisher’s F ratios (defined as the ratio of mean square effect and the mean square error) are also represented in Table 4.

3.4 Adsorption mechanism

The log K_d values of DMP, DEP and DBP were different for the tested biochars, respectively, implying various adsorption mechanisms due to distinct physico-chemical properties of PAEs and biochar surface chemistry. The other different possible mechanisms which could be considered are: (i) hydrophobic interactions between PAEs and biochars; (ii) π-π electron donor-acceptor (EDA) interactions; (iii) hydrogen bonding between the PAEs molecule and the adsorbent O-containing functional groups; (iv) π-hydrogen bonding between the -OH groups on biochar surface and PAEs aromatic surfaces.

Hydrophobic interactions between PAEs molecules and biochars surface could play a key role in adsorption and was considered to be primary mechanism [15], [29]. With the rise in coverage area and hydrophobicity, the alkyl chain length of phthalates increases in the order of DMP<DEP<DBP. However, the sorption coefficient of DMP was greater than that of DEP, which shows an opposite trend with their hydrophobicity. This implies that the DMP polarity was also the additional factor in adsorption.

For the adsorption of chemicals having benzene rings, the EDA interactions have been proposed to be important mechanisms. Considering the molecular structure of DMP, DEP and DBP, these PAEs can serve as electron acceptors because of ester functional groups. As proposed in the EDA theory, the non-covalent interactions between a π-donor compound and a π-acceptor compound are much stronger than that between donor-donor or acceptor-acceptor pairs. The higher O contents reflect the presence of functional moieties on biochar surface. These hydrophilic groups on the surface of biochars may act as multiple drivers in sorption of organic chemicals. The polar functional moieties can create three-dimensional water clusters by joining to neighbor water molecules via strong H-bonding. Additionally, surface oxidation may increase H-bonding interactions. At the same time, these three-dimensional water clusters block the accessible path on biochar surface for sorption of organic molecules, since biochar primarily consists of disordered stacks of graphitic sheets, which tend to be highly polarizable.

The oxygen-containing groups on the edges of biochar could facilitate the π-polarity of aromatic rings on the surface of biochar, thus forming strong π-EDA interactions with aromatic compounds. In this study, we have noted that aging also tends to decrease the sorption capacity of PAEs. The sorption of PAEs on original biochars suggests that the π-EDA interactions were greater than the inhibiting effect of water clusters. However, in the case of oxidized biochars, the water clusters suppressed the sorption and they were greater than those of π-EDA interactions. Thus, the functional moieties were involved in adsorption drive by π-EDA formation, depending on the degree of substitution (by electron donation) of their biochars edges. Additionally, the same functional moieties could suppress the adsorption drive as happened in the case of oxidized biochars.

Structural considerations suggest that the aromatic ring in DMP should be more electron-deficient (i.e. stronger π-acceptors) than the ring structures in DEP and DBP. In order to check π-π EDA interaction strength, UV/vis region for three PAEs (π-acceptor) mixtures with pyrene (π-donor) was obtained. The π-π EDA complexes often indicate a charge transfer band (CT band) in the UV-vis region [18]. There was a weak absorbance for pyrene and it has been proved to be a stronger π-donor, however, there was no absorbance for any of the PAEs in the test domain. The CT band intensity revealed an increase with π-acceptor ability in the order of DMP-Pyr>DEP-Pyr>DBP-Pyr. Hence, the CT band showed the EDA interaction in the order of polarity strength for pyrene. Thus, π-π EDA interaction suggests the difference in sorption coefficient (log K_d), importantly, for the opposite trend in increased sorption of DMP than DEP.

In particular, hydrogen bonding may play an important role in hydrophobic interactions in the presence of oxygenated functional groups/carbonyl groups [30], [31]. The DRIFT spectra indicated that biochar contains O-containing functional groups, e.g. -OH and -COOH as shown in Figure 1. These O-containing polar functional groups can readily act as H-bond acceptor sites for O-atoms of PAEs ester group. Hydrogen bonding donor, hydrogen bonding acceptor, and hydrogen bond-forming ability for PAEs has the same values, i.e. 0, 8 and 8, respectively (Table 1). Thus, the hydrogen bonding and π-hydrogen bonding would contribute to the overall adsorption equally. Therefore, the H-bonding may not be helpful for complete understanding of the adsorption drive.

In essence, we conclude that the higher adsorption of DBP may be ascribed to the hydrophobicity, and water clusters formation inhibited the adsorption, due to the existence of functional moieties. The effective π-π EDA interaction is believed to be responsible for the higher adsorption of DMP as compared with DEP.