Studies on the Removal of Phenol and Nitrophenols from Water by Activated Carbon Developed from Demineralized Kraft Lignin

: The present investigation deals with the importance of interaction and position of the nitro group in the adsorptive removal of 2-nitrophenol (2-NP), 4-nitrophenol (4-NP) and phenol by demineralized kraft lignin activated carbon (DKLAAC). The adsorption of phenol and NPs on DKLAAC was found to follow the order 2-NP > 4-NP > phenol. In this study, both solubility and donor–acceptor complex mechanism played an important role besides the porosity and surface area of the materials. Accordingly, the NP possessing the least solubility would noticeably exhibit a higher afﬁnity to be adsorbed at the solid–liquid interface. Thus the highly hydrophobic 2-NP was adsorbed to a greater extent followed by 4-NP and phenol. Moreover, the adsorption capacity as affected by contact time, initial phenol concentration, pH, and temperature was also investigated. The experimental adsorption capacity by DKLAAC was 2.09, 2.34, and 2.20 mmol · g − 1 for phenol, 2-NP, and 4-NP at 25 ◦ C, respectively, with the maximum amount being adsorbed within 40 min. The experimental data obtained for the removal of phenol and NPs were adequately ﬁtted by the Langmuir adsorption isotherm and pseudo-second order kinetic models. Additionally, the temperature study revealed the adsorption process to be endothermic and spontaneous with high afﬁnity between DKLAAC and phenols.


Introduction
The contamination of water sources due to organic pollutants has become a serious concern in the last few decades [1]. These pollutants are found to have noxious impacts on humans due to their toxicity and carcinogenicity [2] and therefore their presence in the water systems is undesirable. Among organic pollutants, phenol and its derivative compounds have detrimental effects on living beings even at low concentrations owing to their poisonous, carcinogenic, mutagenic, and teratogenic nature and, therefore, the US Environmental Protection Agency has classified them as priority pollutants [3].
Various removal techniques such as ozonation [4], biological degradation [5,6], reduction [7], and adsorption [8] have been employed by the researchers for the removal of phenols. Amidst these, adsorption employing different adsorbents such as clay [9], zeolites [10], metal organic frameworks [11], biomass [12], and activated carbon [8] have been used successfully by researchers. Adsorption is recognized as a versatile technique owing to its simplistic design and convenient operating technique [13][14][15][16] with activated carbons as the extensively utilized materials for the removal of phenol and its derivatives [8,17]. However, to understand the process better for the removal of molecules, the study of adsorbate-adsorbent interactions and the effect of groups is also necessary.
The chemical structures and presence of groups on the molecules are the determining factors which can play a crucial role in their removal. In order to demonstrate this, a study was carried out on the adsorption of phenol and nitrophenols by a microporous activated carbon (DKLAAC). The nitrophenols (NPs) have two functional groups, -OH, which acts as an activating group, and -NO 2 , which behaves as a deactivating group [18,19]. Among these, the nitro group (-NO 2 ), being highly electronegative, withdraws π-electrons from the aromatic ring and, subsequently, the electron density of the accessible π-orbital permits 2-NP and 4-NP to interact with π-electrons on the adsorbent surface [20]. Furthermore, the role of nitro groups present in phenol is also important in the adsorption process since nitro groups are hydrophobic in nature, resulting in low aqueous solubility [21]. The adherent property of the phenolic molecules from aqueous media towards the surface of hydrophobic adsorbents mainly arises due to their dislike for water [22]. Besides nitro groups, phenol and NPs have -OH groups, which are polar in nature and help phenol molecules to form hydrogen bonding with the available functional groups on the adsorbent. Taken together, it can be said that two main adsorption mechanisms may be operating in a process: one due to the hydrophobicity between adsorbent and adsorbate and the other owing to the hydrogen bonding. Therefore, it is significant to understand the method of binding of these molecules for adsorptive removal of phenol and NPs. Moreover, it can facilitate easy prediction of the removal of noxious phenol and NPs if we know the role and binding nature of the existing functional groups.
Hence, the present study aimed to investigate a systematic comparison on the basis of functional groups for the adsorption of phenol, 2-NP, and 4-NP onto DKLAAC prepared in an oxidizing atmosphere. It is expected that the -NO 2 group present on phenol can play an important role and additionally its position is expected to make a difference in adsorption among them. Further, in order to compare the efficiency of DKLAAC, its precursor DKL and a commercial activated carbon (CAC) were also studied and reported herein. Moreover, the effect of various process parameters including contact time, initial phenol concentration, pH, and temperature which can affect the phenol and NPs adsorption on DKLAAC were also discussed.

Materials and Methods
All the reagents utilized in the current work were of analytical reagent (AR) grade. The adsorbates, phenol, 2-NP, and 4-NP were supplied by Sisco Research Laboratories Pvt. Ltd., Mumbai, India (SRL). The CAC and raw kraft lignin (RKL) samples was supplied by Rankem (Gujarat, India) and LignoTech Iberica (Cantabria, Spain), respectively. Doubledistilled water was utilized to prepare the desired solutions.
The RKL was used as a precursor to prepare DKL and discussed in our previous work [23]. The activation was carried out by heating the DKL at 950 • C (rate of 8 • C min −1 ; time: 30 min) in an oxidizing atmosphere. The details of the process are given in our previous work [23].
The amount of adsorbates adsorbed per gram of the adsorbent was calculated by using the following Equation (1): where, C 0 (mol·L −1 ) is the initial concentration, C e (mol·L −1 ) is the equilibrium concentration in solution, V is the volume of solution (L), W is the mass (g) of adsorbent used, and q e (mol·L −1 ) is the amount of adsorbate adsorbed at equilibrium.

Effect of Contact Time and Initial Concentration
Equilibrium time plays a crucial role and is an important parameter that must be taken into the consideration to ensure the efficiency of the adsorbent. Therefore, to determine the time taken to reach the equilibrium for maximum adsorption of phenol, 2-NP, and 4-NP, the adsorption of adsorbates at a fixed concentration of 2 × 10 −3 mol·L −1 onto DKLAAC was studied ( Figure 1A) as a function of contact time. This adsorbate concentration (2 × 10 −3 mol·L −1 ) was chosen from the range of concentration used for developing adsorption isotherms. Results showed that initially the adsorption of phenol, 2-NP, and 4-NP was rapid and thereafter it reached equilibrium slowly. This is possibly owing to the ample vacant adsorption sites/pores and functional groups during the initial stage. Furthermore, less steric hindrance was felt by phenol and NPs while approaching the DKLAAC surface. However, a decrease in the adsorption was observed with time as compared with the initial stage owing to the reduced accessibility of vacant sites and enhancement in the steric hindrance by phenol and NPs molecules on DKLAAC [24,25]. Although the position of the -NO 2 group on phenol was thought to have an effect on the equilibrium, there was only a little difference observed in this case. It is clear from Figure 1A that the maximum amount of phenol, 2-NP, and 4-NP was removed within~40 min attaining equilibrium at~240 min [26]; therefore, a time span of 300 min was considered appropriate for all the studies to ensure equilibrium adsorption.  The change in initial phenol/NPs concentration also alters the amount of the adsorption which was investigated at different concentrations (1.5 × 10 −3 and 2.0 × 10 −3 mol·L −1 ). The results are presented in Figure 1B Figures S1 and S2). The increase in the adsorption of phenol and NPs was observed with increase in the initial concentration [27]. This was because on increasing the initial concentration, the adsorbate molecules deliver the required driving force to reduce the mass transfer resistance between the adsorbent and adsorbates by concentration gradient effect [28,29].

Effect of Solution pH on Adsorption
The effect of pH of adsorbate solutions on their adsorption by DKLAAC was studied, since the pH of the aqueous solution is a critical factor in adsorption process. In order to investigate this effect, the adsorption parameters such as initial concentration, temperature, and contact time were kept constant during adsorption and the initial pH of the phenol and NPs was adjusted in the range of 2-12 using HCl (Himedia, Mumbai, India) or NaOH (SRL, Mumbai, India). The results of effect of pH on adsorption of phenol, 2-NP, and 4-NP on DKLAAC are shown in Figure S3 (Supplementary Material). Results showed that in the pH range of 2-8 for phenol and in the range 2-7 for 2-NP and 4-NP, a very small variation in the adsorption capacity of DKLAAC was observed, which may be attributed to the dispersion forces with small contribution of electrostatic forces in acidic solution [30]. The other reason for this small or no change in the adsorption is the amphoteric nature of DKLAC which has a pH pzc of~7.5. Interestingly, on the other hand, at higher pH the amount adsorbed decreased due to the repulsion between the negatively charged carbon surface and the phenolate/nitrophenolates. Moreover, the low adsorption at high pH is probably due to the presence of OH − ions competing with the phenolates for the residual positively charged adsorption sites on the activated carbon surface [31]. Therefore all the experiments were carried out at pH 6-7 (neutral).

Effect of Temperature
In order to check whether the process is endothermic or exothermic and to compare the effect of temperature on removal of phenol and NPs on DKLAAC, experiments were performed at a temperature range of 25-45 • C (Figure 2A-C) and the results represented an increase in the amount of adsorption on DKLAAC from 2.09 to 2.27, 2.34 to 2.53, and 2.20 to 2.43 mmol·g −1 for phenol, 2-NP, and 4-NP, respectively with a rise in solution temperature from 25-45 • C. This can be explained on the basis of diffusion of phenol and NP molecules throughout the outer surface and inside the pores of DKLAAC as well as decreased viscosity of the adsorbate solution [32][33][34][35][36]. Moreover, phenol and NPs usually form the H-bonds which are broken at high temperature and are responsible for the low solubility of the phenols, resulting in their higher affinity towards adsorbent surface [37]. Owing to these reasons, an increase in adsorption was observed with a rise of temperature, signifying the adsorption process to be endothermic in nature [34].

Adsorption Isotherms
Generally, on increasing the concentration of adsorbates, the adsorption increases and attains a saturation point from which the binding interactions between adsorbate and adsorbent can be estimated by using the adsorption isotherms. To investigate the impact of the presence of functional groups on the molecules in the adsorption process, the adsorption study of phenol, 2-NP, and 4-NP was carried out as a function of equilibrium concentration and the isotherms obtained are presented in Figure 2A-C. The maximum adsorbed amount (experimental, q exp ) on DKLAAC produced from DKL as elucidated from the adsorption isotherms (Figure 2A-C) were found to be 2.09, 2.34, and 2.20 mmol·g −1 for phenol, 2-NP, and 4-NP at 25 • C, respectively. It can be inferred from Figure 2A-C that the adsorption of phenol and NPs on DKLAAC follows the order 2-NP > 4-NP > phenol.
Moreover, to compare the impact of the role/position of the nitro group on the adsorption efficiency of DKLAAC, DKL (precursor to DKLAAC) and CAC were studied and the adsorption isotherms obtained for DKLAAC and CAC are shown in Figure 3A-C respectively for phenol, 2-NP, and 4-NP. Similar to the prepared DKLAAC, the removal of phenol and NPs on CAC also followed the order 2-NP > 4-NP > phenol. However, the amount of phenols adsorbed on CAC (surface area: 770 m 2 ·g −1 ) was observed to be relatively lower (1.14, 1.37, and 1.25 mmol·g −1 for phenol, 2-NP, and 4-NP, respectively) than DKLAAC at same temperature (25 • C) due to a more porous nature and high surface area (surface area: 1305 m 2 ·g −1 ) [38]. On the contrary, a negligible adsorption of phenol, 2-NP, and 4-NP was found on the surface of DKL because of its low surface area and therefore the Figure for the same is not shown. Isotherm models, namely, Langmuir, Temkin, Freundlich, and Dubinin-Radushkevich (D-R) were used for the analysis of experimental adsorption data obtained for the removal of phenol and NPs (Table 1). The mathematical representation of the Langmuir equation (Equation (2)) [39] can be given as: where, C e and q e are the equilibrium concentration and adsorbed amount of phenols respectively, b and q max are the Langmuir constant and maximum adsorption capacity respectively. The plots of 1 q e vs. 1 C e for the adsorption of phenol, 2-NP, and 4-NP are drawn and shown only for 2-NP in Figure 4A, while the similar plots for phenol and 4-NP are provided in the supplementary material ( Figures S4 and S5). The values of Langmuir capacity (q max ) were obtained from the intercept, whereas the adsorption constant b was assessed from the slope of the drawn plots (Table 1, Figures 4A, S4 and S5). A close comparison of the experimentally observed values and theoretically obtained values by the Langmuir model (Table 1) clearly shows the close proximity of both, thereby demonstrating the suitability of Langmuir isotherm model. Moreover, the constant 'b' which reflects the equilibrium constant and affinity of the process was evaluated (Table 1) and the values obtained for phenol, 2-NP, and 4-NP were found to be 1.90 × 10 4 , 6.49 × 10 4 , and 5.43 × 10 4 L·mol −1 , respectively, clearly indicating that the DKLAAC has a maximum affinity for 2-NP and minimum affinity for phenol. In addition to the Langmuir model, the data of phenol and NPs were also studied using the Freundlich model (Equation (3)) [40], which is represented as: where q e denotes the adsorbed amount at the equilibrium, while K f and n represent the adsorption capacity and heterogeneity factor, respectively. The Freundlich constants for the removal of phenol, 2-NP and 4-NP on DKLAAC were elucidated by plotting logq e against logC e with the plots for 2-NP at 25 • C, 35 • C, and 45 • C being shown in Figure 4B, while that for phenol and 4-NP shown respectively in Figures S6 and S7. The Temkin model [41] that explains the adsorbate-adsorbent interactions was also studied and can be represented in a linear mathematical form (Equation (4)) as follows: where, K T (L·mg −1 ) and B T (dimensionless) are the constants related to the Temkin isotherm binding constant, and the heat of adsorption, respectively, and B T (Equation (5)) can be given by: where, T (K) is temperature, b T (J·mol −1 ) is the Temkin constant and R is the universal gas constant (R = 8.314 J·mol −1 ·K −1 ). A plot of q e versus lnC e is shown only for 2-NP ( Figure 5A), while that for phenol and 4-NP are shown in the Supplementary Material ( Figures S8 and S9). The corresponding isotherm parameters obtained are given in Table 1. The Dubinin-Radushkevich (D-R) model (Equations (6) and (7)) [42] was also used to illustrate the adsorption of phenols onto DKLAAC and can be given as: where ε is the Polanyi potential, q m (mg·g −1 ) is the D-R adsorption capacity, and β DR (mol 2 ·kJ −2 ) is a constant which can be obtained from the mean free energy E (kJ·mol −1 ) by using equation E = 1/ √ (2β DR ). The free energy parameter E helps to predict the nature of the process as either physisorption or chemisorption. The D-R model constants q m and β DR (Table 1) were obtained from the slope and intercept of the plots (lnq e vs. ε 2 ) shown for 2-NP in Figure 5B as well as for phenol and 4-NP respectively in Figures S10 and S11 of the Supplementary Material. The comparison of correlation coefficients among different isotherm models tested (Table 1) clearly reveals the best applicability of the Langmuir model compared with the other investigated models. Similar adsorption isotherm behavior was also reported for the removal of phenols by activated carbon prepared from eucalyptus seed [43] and Luffa cylindrica [44].
The studies show that the extent and adsorption behavior of phenol and NPs are not only limited to the surface area and porosity of the adsorbent, but also to a variety of forces operating on the adsorption surface. Nevertheless, the functional groups attached to adsorbate molecules may also play a vital role in the adsorption process and thus cannot be ignored [45]. Therefore, in the present investigation, the role of the nitro groups and their position on the adsorption process were investigated. The order of adsorption capacity of phenol and NPs on DKLAAC (2-NP > 4-NP > phenol) confirmed the impact of the presence of nitro group (an electron-withdrawing group). Furthermore, it is expected that these ring-deactivating groups with decreasing electron density can undergo complex formation via the donor-acceptor mechanism and facilitate adsorption [46,47]. Other researchers [48] have also agreed to the fact that that the phenol adsorption on the AC surface occurs through the donor-acceptor complex mechanism in which the aromatic ring of adsorbate and the carbonyl oxygen on carbon's surface act as an acceptor and donor, respectively. Besides this, the H-bonding with hydrophilic sites/groups on AC surface generated during the activation period may also be effective in the adsorptive removal of phenol and NPs.
Another important parameter that can potentially influence the adsorption is the high hydrophobicity of adsorbate, which usually shows high affinity towards the AC surface [48]. Interestingly, this factor is also true in our case where hydrophobicity assists in promoting the adsorption of NPs owing to their poor solubility. Accordingly, 2-NP, being highly hydrophobic in character with least solubility, showed the maximum adsorption value (Table 1). Therefore, it may be suggested that a combined effect of solubility and donor-acceptor complex mechanism plays an important role in our study besides the surface area and porous nature of DKLAAC. A comparison of the adsorption capacities of different adsorbents employed by the researchers in the literature with DKLAAC is given in Table 2. A relatively high adsorption capacity and large surface area of DKLAAC demonstrate the impact of air activation technique involved in the preparation of DKLAAC.

Thermodynamic Studies
The thermodynamic parameters viz. change in enthalpy (∆H • ), free energy (∆G • ), and entropy (∆S • ), being significant to understand the feasibility, nature of the process, and spontaneity, must be determined to evaluate the adsorption process utilizing the following Equations (8) and (9): where R (8.314 J·mol −1 ·K −1 ) is the universal gas constant and T and b are the temperature ( • C) and thermodynamic equilibrium constant, respectively. The ∆H • and ∆S • were calculated from the lnb vs. 1 T plot and the values determined for these parameters are summarized in Table 3. Table 3. Thermodynamic parameters for the removal of phenol and NPs on DKLAAC.

Phenols
Temperature The positive values of the ∆H • , 6.8, 6.9, and 7.1 kJ·mol −1 obtained for phenol, 2-NP, and 4-NP, respectively, indicate that the process follows an endothermic route, while its small magnitude specifies the physical adsorption (Table 3) [36]. Similar results were reported in the literature [22,59]. The observed endothermic behavior of the adsorption process may be due to the breaking of hydrogen bonds, resulting in enhanced adsorption at high temperatures as already discussed in the preceding paragraph. The negative ∆G • values (−24.4 to −29.8 kJ·mol −1 ) confirm the spontaneity of the adsorption process and the range of the values indicates the process to be physisorptive in nature [60]. Table 3 also reveals that the ∆G • values increased on raising the temperature, suggesting the adsorption to be favorable at elevated temperatures. Nevertheless, the positive values of ∆S • are indicative of high affinity for the removal of phenols by DKLAAC. The thermodynamic results obtained were compared with other reported values for various adsorbents (Table 4) and were found to be comparable.

Kinetic Studies
Kinetic investigations were carried out to discover more about the adsorption of phenol and NPs (differentiated on the basis of their position) on DKLAAC and the data were tested using various well-known models.
The pseudo-first order (PFO) equation by Lagergren [66] suggested that the kinetics of adsorption of the adsorbate from aqueous solution can be represented as shown below (Equation (10)): log q e − q t = logq e − k 1 2.303 t (10) where q e and q t are the adsorbed amount of phenols at equilibrium and at time t, respectively. k 1 is the Lagergren rate constant for the PFO equation. The model was applied and the values of q e (from slope) and k 1 (from intercept) were evaluated from the curves plotted between log (q e − q t ) and t ( Figure 6A). The obtained values are presented in Table 5.   Besides PFO, the pseudo-second order (PSO) [67] model was also utilized to analyze the kinetics of the process and can be described as shown below (Equation (11)): where q t and q e are the adsorbed amount of phenols at time t and equilibrium, respectively. k 2 is the equilibrium rate constant for the PSO equation. To evaluate the applicability of the PSO adsorption model, t/q t was plotted as a function of time ( Figure 6B), whereas the values obtained from the graphs are given in Table 5. Furthermore, the Elovich Equation [68] was also used for the analysis of the experimental adsorption data and is given as follows (Equation (12)): where, q t (mmol·g −1 ) is the amount adsorbed at time t (min), α (mmol·g −1 ·min −1 ) is the constant for initial adsorption, β (g·mmol −1 ) is the rate constant linked to the extent of surface coverage, and the plots obtained are presented in Figure 7A. The results obtained from these kinetic models are shown in Table 5. It was found that the PSO model has better R 2 values and therefore fits best to the adsorption data of phenol and NPs as compared with the PFO and Elovich models, which are similar to the findings reported for adsorption of phenols by Luffa cylindrica [44].
In order to investigate the rate-controlling mechanism and to understand the adsorption process more precisely, the equilibrium data were also validated using an intraparticle diffusion model (IPD) model [69]. Out of the four steps commonly involved in the adsorption process, the first and last step [70] are usually very fast and account for the migration of adsorbate (phenols) from the liquid to the adsorbent (DKLAAC) surface. The second and third steps account for the diffusion of adsorbate molecules (phenols) and are considered as rate-controlling steps either alone or in combination. The rate constants for using the IPD model can be calculated with the following expression (Equation (13)): where K id is the rate constant and C is a constant corresponding to the thickness of the boundary layer. Plots between q t and t 1/2 were made for the analysis of the adsorptive removal of phenol and NPs ( Figure 7B). The Figure 7B consists of two linear sections with film diffusion and IPD as the first and second steps, respectively. The intra particle rate constants K id (K id1 for the 1st step and K id2 for the 2nd step) and C (C 1 for the 1st step and C 2 for the 2nd step) values were evaluated from these linear parts of the plots and are shown in Table 5. As is clear from the Figure 7B, the straight line does not pass through the origin, and it can be collectively deduced that the adsorption of phenol, 2-NP, and 4-NP onto DKLAAC is a complex process involving a combination of IPD and surface sorption in the rate-limiting step of the adsorption process [44].

Conclusions
This work was carried out to investigate the impact of the nitro groups and to know the role of their position on the adsorption of phenol, 2-NP, and 4-NP on DKLAAC. The interaction and position of nitro groups present on the phenol play a crucial role in their adsorption and it was found that nitro groups (deactivating group) would increase the extent of adsorption via the electron donor-acceptor mechanism and the hydrophobicity of the adsorbate. Interestingly, the hydrophobicity factor promotes the adsorption of NPs as it was found that 2-NP possessing the least solubility (more hydrophobicity) adsorbed to a greater extent than 4-NP and phenol. The experimental adsorption capacities were found to be 2.09, 2.34, and 2.20 mmol·g −1 respectively for phenol, 2-NP, and 4-NP on DKLAAC, following the order 2-NP > 4-NP > phenol at 25 • C. The experimental data were best described using the Langmuir isotherm and an increase in the adsorption of phenols on raising the temperature from 25 • C to 45 • C indicated the process to be endothermic. The thermodynamic parameters could well-describe the spontaneity and favorability of the adsorption process. The negative values of ∆G • suggested that the adsorption process was governed by physisorption, while small positive ∆H • values indicated the endothermic and physisorption nature of the adsorption process. Moreover, the high affinity for removal of phenols by DKLAAC was revealed by the positive value of ∆S • . A pseudo second-order model was found to precisely describe the kinetic data.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/agronomy12102564/s1, Figure S1: Effect of initial concentration for the removal of phenol by demineralized kraft lignin activated carbon (DKLAAC) at different initial concentrations at 25 • C; Figure S2: Effect of initial concentration for the removal of 4-nitrophenol (4-NP) by DKLAAC at different initial concentrations at 25 • C; Figure S3: Effect of pH on the adsorption of Phenol, 2-NP and 4-NP on DKLAAC at 25 • C with initial phenol concentration at 1 × 10 −3 M; Figure S4 Data Availability Statement: All data generated or analyzed during this study are included in this article.