Modified Spruce Sawdust for Sorption of Hexavalent Chromium in Batch Systems and Fixed-Bed Columns

This study investigated the potential use of spruce sawdust that was pretreated with diethylene glycol and sulfuric acid for the removal of hexavalent chromium from wastewater. The sawdust pretreatment process was conducted at different temperatures and times. The adsorbent was characterized by quantitative saccharification, scanning electron microscopy, and Brunauer–Emmet–Teller surface area analysis. Adsorption capacity was studied for both batch and column processes. The experimental adsorption isotherms were simulated using seven isotherm models, including Freundlich and Langmuir models. By using the Langmuir isotherm model, the maximal Cr(VI) adsorption capacity of organosolv-pretreated spruce sawdust (qm) was 318.3 mg g−1. Furthermore, the kinetic data were fitted to Lagergren, pseudo-second-order, and intraparticle diffusion models, revealing that the adsorption of Cr(VI) onto spruce sawdust pretreated with diethylene glycol and sulfuric acid is best represented by the pseudo-second-order kinetic model. Three kinetic models, namely, the Bohart–Adams model, Thomas model, and modified dose–response (MDR) model, were used to fit the experimental data obtained from the column experiments and to resolve the characteristic parameters. The Thomas adsorption column capacity of the sawdust was increased from 2.44 to 31.1 mg g−1 upon pretreatment, thus, demonstrating that organosolv treatment enhances the adsorption capability of the material.


Introduction
The increased magnitude of industrial processes has led to overwhelming environmental pollution and enormous damage to specific ecosystems. Specifically, the high toxicity of heavy metals makes them significant environmental and public health threats. Heavy metals speak to a significant issue for the environment and for all the life forms, since they are not biodegradable and can gather in living tissues, reaching human organisms as well through the chain food [1]. The use of chromium is particularly worrying because it is being increasingly utilized in developing countries in a number of industries, such as leather tanning, electroplating, textile dyeing, and metal finishing [2].
The International Agency for Research on Cancer has concluded that the sixth oxidation state of chromium, namely Cr(VI), is carcinogenic to human beings. The World Health Organization has dictated that the highest allowable level for Cr(VI) in drinking water is 0.05 mg L −1 [3,4]. Furthermore, a number of countries are facing issues in relation to the presence of hexavalent chromium in natural ecosystems. For instance, in Greece, the Asopos River has been considered a "processed industrial waste receiver" since 1969. It was recently discovered that the Asopos and Thiva Basins are massively contaminated with Cr(VI) because of the excessive industrial activity around the area and unregulated discharge even though a Cr(VI) limit in industrial discharge is mandatory to acquire licensing [5]. Table 1 displays the results of pretreatment with 50% diethylene glycol/50% H 2 O and pretreatment with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 . The percentage of lignin in the materials increases with the intensity of the pretreatment. By changing the isothermal pretreatment time, the percentage of lignin in the materials increases. For the pretreatment with 50% diethylene glycol/50% H 2 O, there is a relative stability for the percentage of cellulose at 200 • C and 220 • C for an isothermal reaction time of 50 min. Additionally, for pretreatment with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 , the percentage of cellulose following isothermal reaction for 50 min at 220 • C is low. Table 1 shows that, for pretreatment with 50% diethylene glycol/50% H 2 O, more extreme pretreatment conditions lead to smaller percentages of xylan and mannan. Conversely, for pretreatment with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 , a steep decrease in mannan is observed, particularly for an isothermal reaction time of 0 min.

Organosolv Pretreatment
The solid residue yield (SRY) decreases with pretreatment temperature for both systems. The temperatures used were 160 • C, 180 • C, 200 • C, and 220 • C, which were reached after 42, 50, 62, and 80 min preheating, respectively. The SRY is lower for an isothermal reaction time of 50 min in relation to an isothermal reaction time of 0 min. Table 2 displays the specific surface areas of the materials for all pretreatment conditions. There is an increase in the specific surface area upon increasing the pretreatment temperature. For the pretreatment with 50% diethylene glycol/50% H 2 O, the largest specific surface area appears to be 4.078 m 2 g −1 , which is achieved by pretreatment at 220 • C for 50 min. For the pretreatment with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 , the largest specific surface area appears to be 11.335 m 2 g −1 , which was achieved by pretreatment at 220 • C for 50 min.  2.3. Microstructure Figure 1 shows the SEM micrographs of untreated spruce sawdust and those treated under acid-catalyzed organosolv conditions at 180 • C for 50 min at two different magnifications (×7500 and ×30,000). The surface of the pretreated material exhibits greater roughness than the untreated material. This is particularly evident at ×30,000 magnification.

Adsorption Isotherms
Therefore, in this study, seven isotherm models were applied to fit the experimental results ( Figure 2). The Freundlich [35] isotherm is given by the following equation.
where q is the amount adsorbed per unit mass of the adsorbent (mg g −1 ), C e is the equilibrium concentration of the adsorbate (mg L −1 ), and K F , n is the Freundlich constants related to adsorption capacity and intensity, respectively. Equation (1) in logarithmic form gives the following. The Langmuir isotherm [36] is given by the following equation.
where K L is the Langmuir constant related to the energy of adsorption (L mg −1 ) and q m is the amount of Cr(VI) adsorbed (mg g −1 ) when saturation is attained. The parameters K L and q m can be obtained either by plotting 1/q versus 1/C e or by non-linear regression analysis. The characteristics of the Langmuir isotherm can be described by a dimensionless constant called the 'equilibrium parameter' or 'separation factor' R L .
where C 0 is the initial concentration (mg L −1 ) and K L is the Langmuir constant (L mg −1 ). The value of R L indicates whether the type of the proposed isotherm is either unfavorable (R L > 1), linear (R L = 1), favorable (0 < R L < 1), or irreversible (R L = 0). The Sips (Langmuir-Freundlich) [37] isotherm equation is shown below.
where K L and q m are the Langmuir constants, and n is the Freundlich constant. The Radke-Prausnitz [38,39] isotherm equation is shown below.
where s is a new constant. Table 3 presents all the parameters of the Freundlich and Langmuir models using nonlinear regression analysis (NLRA) for all the samples with different pretreatment conditions (temperature, isothermal treatment time, and added acid). The standard error of estimate (SEE) was calculated in each case by the following expression.
where y i is the experimental value of the depended variable, y i,theor is the theoretical or estimated value of the depended variable, n is the number of experimental measurements, and p is the number of parameters (the difference [n' − p ] being the number of degrees of freedom). The fitting of the Freundlich adsorption model to the experimental data is very satisfactory (see Table 3 and Figure 2a,b).   Table 3 shows the Freundlich isotherm capacity coefficient K F of Cr(VI) adsorption on untreated and pretreated spruce sawdust. According to the same table with the K F measurement for Cr(VI) removal, it can be observed that pretreatment at 160 • C provides insufficient improvements to the adsorption capacity of the material. By contrast, more intense pretreatment, such as pretreatment at 220 • C for 0 min and pretreatment at 180 • C for 50 min, yield significant improvements in K F adsorption capacity.
In relation to industrial applications, the adsorption capacity q m is given in Table 3 according to the Langmuir model. On this basis, pretreatment with an organic solvent catalyzed by acid significantly improves q m . Pretreatment at 160 • C does not improve the adsorption capacity of our material for time 0. By contrast, for the more intense pretreatment temperatures of 180 • C, 200 • C, and 220 • C, we see a significant improvement with a maximum temperature of 220 • C for 0 min (q m = 257.2 mg g −1 ). Conversely, for pretreatment with organic solvent catalyzed by acid for an isothermal time of 50 min, improvement is observed for the two lowest pretreatment temperatures (160 • C and 180 • C) with maximum temperatures of 180 • C for 50 min, as shown in Table 3, where q m = 318.3 mg g −1 for the Langmuir model.
In the current study, the R L values were calculated in the range of zero to one for all initial Cr(VI) concentrations (C 0 ) and for all adsorbents considered. This indicates a sympathetic adsorption. By contrast, R L > 1 represents an adverse adsorption, and R L = 1 represents a linear adsorption. Furthermore, the adsorption is permanent if R L = 0. Table 4 presents the parameters of seven isothermal models in their nonlinear form, which is similar to previous pretreatments. The materials for which the parameters of the seven isothermal models are presented are untreated spruce sawdust, sawdust pretreated with 50% diethylene glycol/50% H 2 O at 220 • C for 50 min, and sawdust pretreated with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 at 180 • C for 50 min, which has the optimum adsorption capability. The fitting of the Sips model was better when compared to other six isotherms models but also Freundlich and Langmuir models were very satisfactory using only two parameters. Table 4. Estimated parameter values for the various isotherm models for Cr(VI) adsorption on untreated and organosolv pretreated spruce sawdust. In Table 5, we compare the Cr(VI) adsorption parameters K F , n, q m , and K L for pretreated and untreated spruce sawdust with other adsorbents derived from agricultural or waste materials, according to available literature data. The optimal q m value found in this work was higher than the corresponding values of most of the other materials in Table 5, while the K F was the highest. Table 5. Cr(VI) adsorption capacity for various lignocellulosic materials.

Adsorption Kinetics
Various kinetic models have been used to identify the reasonable mechanisms for solid/liquid adsorption frameworks. Among them, Lagergren's pseudo-first-order [47] and pseudo-second-order kinetic models [48] as well as the intraparticle diffusion kinetic model [49] are the three most common models. Thus, these three kinetic models were utilized to fit the experimental data for the adsorption of Cr(VI) on pretreated spruce sawdust.
The widely used Lagergren equation [47] or the pseudo-first order kinetic model is shown below.
where q and q t are the amounts of Cr(VI) adsorbed per unit mass of the adsorbent (in mg g −1 ) at equilibrium time ( t → ∞ ) and adsorption time t, respectively, while k is the pseudo-first order rate constant for the adsorption process (in min −1 ).
where C, C 0 , and C e are the concentrations of Cr(VI) in the bulk solution at time t, 0, and ∞, respectively, while m is the weight of the adsorbent used (in g), and V is the solution volume (in mL). Further modification of Equation (12) in logarithmic form gives the following.
The pseudo second order kinetic model [48] is as follows.
where k 2 (min −1 ) is the rate constant of second order adsorption. The possibility of intra-particle diffusion was explored by using the intra-particle diffusion model [49].
where q t is the amount of Cr(VI) adsorbed at time t, k p (mg g −1 min −0.5 ) is the intra-particle diffusion rate constant, and c (mg g −1 ) is a constant related to the thickness of boundary. A value of c close to zero indicates that diffusion is the only controlling step of the adsorption process. The most appropriate model was chosen in terms of both SEE and q e values. Tables 6 and 7 show the kinetic results. Figure 3 shows the second-order kinetics of Cr(VI) adsorption on untreated and pretreated spruce sawdust. Pretreatment with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 at a medium temperature (180 • C) leads to the highest adsorbance.
The SEE values for the second-order kinetic models are 0.133 for untreated spruce sawdust and 0.150 for spruce sawdust pretreated with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 at 180 • C for 50 min.  Table 6. Pseudo-first and pseudo-second order kinetic models parameters for Cr(VI) adsorption on untreated and organosolv pretreated spruce sawdust.

Pseudo-First-Order Model
Pseudo-Second-Order Model For pretreatment with 50% diethylene glycol/50% H 2 O, the NLRA estimates for k 2 of the second-order kinetic model return values from 0.0003-0.0008 g mg −1 min −1 , and the range of SEE error values is 0.079-0.181. However, for pretreatment with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 , the NLRA estimates for the k 2 of the second-order kinetic model return values from 0.0003-0.0021 g mg −1 min −1 , and the range of SEE error values is 0.106-0.178. Therefore, these SEE values are lower than those of the first-order kinetic model. According to the above, the pseudo-second-order kinetic model demonstrates that the equilibrium amount of adsorbed absorbate controls the number of binding sites [50].
Finally, the NLRA estimates for the k p of the intraparticle diffusion model for the pretreatment with 50% diethylene glycol/50% H 2 O, for 0 min, at 160 • C, 180 • C, 200 • C, and 220 • C, are 0.1133, 0.1011, 0.1203 and 0.1655, respectively. Therefore, according to the intraparticle diffusion model, a higher pretreatment temperature corresponds to higher values of k p . The SEE values of the intraparticle diffusion model are 0.122-0.203. The minimum value is for untreated spruce sawdust, and the maximum value is for spruce sawdust pretreated with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 at 180 • C for 50 min (see Figure 4). The intraparticle diffusion model involves a multistage adsorption process that includes the mass transfer of adsorbate molecules to the external surface of the adsorbent, their mass transfer to the internal surface of the adsorbent, and their sorption on the active sites of the adsorbent [51]. Assessing these three kinetics models demonstrated that the material improves substantially when it is pretreated with sulfuric acid, diethylene glycol, and water.
All SEE error values for the second-order kinetic model were found to be somewhat lower than those of the Lagergren and intraparticle diffusion models, indicating the marginally higher suitability of second-order kinetics in the adsorption of Cr(VI) by spruce sawdust pretreated with organic solvent.
According to the findings presented in Figure 5, the simulated q-values obtained via pseudo-first and pseudo-second order kinetic models were decreased by 31% and 13%, respectively (as an average), when compared to the experimental equilibrium uptakes achieved by isothermal experiments (equilibrium reached after seven days).

Adsorption Columns
For the adsorption column experiments, the 'bed depth service model' developed by Bohart and Adams [52] is commonly used as follows.
where C = effluent concentration (mg L −1 ), C i = influent concentration (mg L −1 ), K = adsorption rate coefficient (L mg −1 min −1 ), N = adsorption capacity coefficient (mg L −1 ), x = bed depth (cm), u = linear velocity (cm min −1 ), and t = adsorption time (min). The Thomas [53] model is one of the most widely used models in the column performance theory. The main difference between the Bohart-Adams and the Thomas model is the form of the sorption isotherm assumed. The latter assumes a Langmuir (favorable) isotherm. It has been shown that, when the sorption isotherm is highly favorable, the actual Thomas model reduces to the Bohart-Adams model. The expression by Thomas for an adsorption column is: where k T is the Thomas rate constant, q 0 is the sorption capacity of the adsorbent per unit mass of the adsorbent, M is the mass of adsorbent, and Q is the flow rate. It is evident that Equations (17) and (18) look strikingly similar even at first glance. They are mathematically identical and have interchangeable parameters [54,55]. The Modified Dose-Response (MDR) model was proposed by Yan et al. [55] because it minimizes the error resulting from the use of the Thomas model, especially at lower and higher time periods of the breakthrough curve. The MDR model is expressed as: where b mdr = (q 0 M)/C i and α mdr are the modified dose-response constants, and V = Q . t. Table 8 shows the parameters of the Bohart-Adams, Thomas, and MDR models for untreated and pretreated spruce sawdust for Cr(VI) removal. The pretreatment conditions of spruce sawdust are 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 for 160 • C, 180 • C, 200 • C, and 220 • C, and the isothermal reaction time is 50 min. The experiments were performed at a flow rate of 2 mL min −1 , the mass of the material was 18 g, and the Cr(VI) concentration was 60 mg L −1 . According to Table 8, the Bohar-Adams capacity N (mg L −1 ) for the untreated spruce sawdust is 662 mg L −1 for the material pretreated with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 at 180 • C for 50 min is 7814 mg L −1 , which is the maximum value of N. However, for the most severe pretreatment conditions (50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 at 220 • C for 50 min), the value of N decreases at 998 mg L −1 .
Furthermore, we observe that, according to the Thomas model, the adsorption capacity q 0 (mg g −1 ) for the untreated material is equal to 2.437 mg g −1 , while that for the material pretreated at 180 • C for 50 min is 31.08 mg g −1 . Consequently, the pretreated material at the optimal conditions has an adsorption capacity that is 13 times that of the untreated material. However, for the most severe pretreatment conditions, it is reduced to 4.452 mg g −1 .
Similarly, the MRD model shows that the material pretreated at 180 • C for an isothermal reaction time of 50 min has an adsorption capacity that is 14 times that of the untreated material. For the extreme pretreatment at 220 • C for 50 min, the adsorption capacity of the material is greater than that of the unprocessed material but was not optimal, as we expected. In general, the adsorption capacity q 0 (mg g −1 ) values estimated, according to the MDR model, are very similar to those of the Thomas model, resulting in the same conclusion with regard to the process optimization. Figure 6 shows the theoretical curves from the Thomas model for untreated and pretreated with the acid-catalyzed organic solvent spruce. The fitting of the Thomas (or the Bohart-Adams) model to the experimental data is not always better than that of the MDR model but it fits better than the high C/C i experimental values while the MDR model fits the very low initial C/C i values better. The Cr(VI) adsorption capacity according to the Thomas capacity parameter q 0 (mg L −1 ) of organosolv-pretreated spruce sawdust was compared with those of other adsorbents reported in the literature (Table 9). According to the Thomas model, spruce sawdust pretreated with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 at 180 • C for an isothermal reaction time of 50 min was q 0 = 31.08 mg g −1 . The Cr(VI) adsorption capacity for the organosolv-pretreated spruce sawdust is comparable than the reported values of some previously studied adsorbents. Therefore, it can be concluded that organosolv-pretreated spruce sawdust is competitive to other modified adsorbents lignocellulosic adsorbents.

Chemical-Physical Mechanisms of the Process
Adsorption on sawdust can be physical, ion exchange, and chemical [12]. Generally, three types of mechanisms are involved in an adsorption process: (i) film diffusion, which involves the movement of adsorbate molecules from the bulk of the solution toward the external surface of the adsorbent, (ii) particle diffusion, where the adsorbate molecules move in the interior of the adsorbent particles, and (iii) adsorption of the adsorbate molecules on the interior of the porous adsorbent [61,62]. The Thomas kinetic model satisfactorily described the column adsorption. This model is based on the mass transfer model, which postulates that adsorbate emigrates from the solution to a film around the particle and expands through the liquid film to the surface of the adsorbent. Subsequently, this step is followed by intraparticle diffusion and adsorption on the active site, assuming Langmuir isotherm for equilibrium, plug flow performance in the bed, and second-order reaction kinetics [12,63].
Moreover, to predict the rate determining the diffusion mechanism within the biosorption system above, applied an intraparticle diffusion model that proved to be very effective. Webber's pore diffusion model was applied on the kinetic data with the pore diffusion factor described by Equation (16), where k p is the intra-particle diffusion rate constant. In the case that intraparticle diffusion is the rate limiting step, the q versus t 0.5 will be linear with slope k p and the plot will pass through the origin, i.e., intercept c = 0. Otherwise, some other mechanism along with intraparticle diffusion is involved in the biosorption process, such as film diffusion. Based on the kinetics constants presented in Table 7 and the interpretation of the pretreated sawdust data in Figure 4, the sorption process consists of two steps. The linearization did not pass through the origin, indicating that intraparticle diffusion is not the rate-limiting step and implying that the biosorption is affected by more than one process. Additionally, in the same figure, the first part refers to the film diffusion and the second part correlates with the diffusion within the adsorbent. The initial segment in the plot specified an external mass transfer while the second segment is caused by intraparticle diffusion or pore diffusion [64,65]. With regard to the untreated sawdust, the second part does not appear within 24 h of our experiment (see Figure 4).
According to the literature, Cr(VI) can be adsorbed either by cellulose (modified or not, activated or not) [19,66] either by lignin (obtained from lignocellulosic biomass in combination with some modification and/or activation) [20,27,67,68]. According to the present experiments, there are optimal conditions for achieving maximum Cr(IV) adsorption capacity by material containing both cellulose and lignin. According to the presented data, there is no need to achieve maximum cellulose concentration or maximum lignin concentration (Table 1). Even the BET surface area (Table 2), maximization is not necessary or desired to obtain optimized adsorption material.
Studying the mechanism of Cr(VI) biosorption on the untreated and pretreated sawdust surface is important for understanding the enhanced removal of Cr(VI) from aqueous solution using pretreated sawdust. FTIR was used as a qualitative technique for assessing the chemical structure of sawdust [65].
The FTIR spectrum of the untreated spruce and of the pretreated (50% diethylene glycol/50% water/0.045 N H 2 SO 4 , 180 • C, 50 min) spruce sawdust is provided in Figure 7. The comparison of these FTIR spectrums shows that some peaks were shifted. Moreover, the transmittance was significantly increased with regard to the pretreated sawdust compared to the untreated material. The major peaks, the shift of the peaks, the assignment, and the corresponding component are presented in Table 10. In the case of the pretreated material, there is a peak shift at wavenumber 3347 cm −1 (decreased by 117 cm −1 compared to the peak at 3464 cm −1 for the untreated sawdust) representing the O-H stretching existing in cellulose, hemicelluloses, and lignin. Moreover, a peak shift at 2942 cm −1 (increased by 33 cm −1 compared to the 2909 cm −1 for the untreated material) indicates the presence of C-H stretching and a peak shift at 1700 cm −1 (decreased by 35 cm −1 compared to the 1735 cm −1 for untreated) indicates the C=O stretching. In addition, a peak shift at 1457 cm −1 (increased by 22 cm −1 compared to the 1435 cm −1 for the untreated) can be assigned to C-H deformation existing in lignin. The peak shift at 1043 cm −1 (increased by 25 cm −1 compared to the 1068cm −1 for the untreated) can be attributed to C-OH stretching vibration and C-O deformation. The appearance of a new peak at 1039 cm −1 can be attributed to the C-O stretching and aromatic C-H in plane deformation. Finally, the peak shifts at 855 and 851 cm −1 in the FTIR spectrum of pretreated spruce sawdust may be due to the C-O-C stretching and aromatic C-H out of plane bending, respectively. These peaks exist in cellulose/hemicelluloses and in lignin, respectively. In Figure 8, the transmittance of the FTIR spectra of the pretreated (50% diethylene glycol/50% water/0.045 N H 2 SO 4 , 180 • C, 50 min) spruce sawdust before and after Cr(VI) adsorption are given. The adsorption conditions were C 0 = 7 mg L −1 , t = 1450 min, and adsorbent = 1 g L −1 . The Transmittance was significantly increased after Cr(VI) adsorption on pretreated sawdust. The major peaks are given in Table 11. There is a peak shift at a wavenumber of 3410 cm −1 (increased by 63 cm −1 compared to the 3347 cm −1 before adsorption) representing the O-H stretching existing in cellulose, hemicelluloses, and lignin. Furthermore, a peak shift at 1684 cm −1 (decreased by 16 cm −1 compared to the 1700 cm −1 before adsorption) indicates the C=O stretching. In addition, a peak shift at 1076 cm −1 (increased by 8 cm −1 compared to the 1068 cm −1 before adsorption) can be attributed to the C-OH stretching vibration and C-O deformation. Finally, the peak shift at 850 cm −1 in the FTIR spectrum of pretreated spruce sawdust after Cr(VI) adsorption may be due to the C-O-C stretching. Shifts or changes of these peaks would indicate interactions between the Cr(VI) with functional groups on the solid surface as a result of biosorption or a chemical reaction. Consequently, the increases in the transmittance of these peaks indicated the oxidation of cellulose/lignin in the occurrence of the adsorbed Cr(VI). In addition, it can be seen that the whole region of different functional group bands on the sawdust surface were involved in the biosorption of Cr(VI). The FTIR spectrum confirms the changes in the functional groups and surface properties of spruce sawdust, as shown by the shift of some functional groups. Tables 10 and 11 show a number of major Transmittance peaks indicating the characteristic structure of the lignocellulosic complex before and after biosorption of Cr(VI), which is in agreement with those reported in the literature [65,[69][70][71][72][73].

Materials
As an example, the source for industrial applications, which was a local furniture construction company, was used to acquire the spruce sawdust. At the point when the material was obtained, the moisture content was 8.3% (w/w). The portion with particle sizes of 0.2-1 mm was isolated. Table 1 shows the contents of the raw material.

Organosolv Pretreatment
The organosolv pretreatment processes were performed in a 3.75 L batch reactor (PARR 4843, Parr Instrument Company, Moline, IL, USA). The isothermal treatment time was 0 or 50 min (excluding the nonisothermal preheating and the cooling periods). The reaction was either (i) autocatalyzed (i.e., catalyzed by the organic acids generated by the spruce sawdust itself due to autohydrolysis during treatment) or (ii) catalyzed by 0.045 N H 2 SO 4 at a liquid-solid ratio of 10:1. The liquid phase volume was 1000 mL diethylene glycol and 1000 mL water. The spruce sawdust loading was 200 g, and the stirring speed was 150 rpm. The temperatures used were 160 • C, 180 • C, 200 • C, and 220 • C, which were reached after the preheating periods of 42, 50, 62, and 80 min, respectively.

Characterization
For the quantitative saccharification of the untreated spruce sawdust and the organosolv reaction solid residues, the technique of Saeman et al. (1945) [74] was used. High-performance liquid chromatography (HPLC) (Agilent 1200, Agilent Technologies, Santa Clara, CA, USA) with an Aminex HPX-87H Column as well as a refractive index detector and 5 mM H 2 SO 4 in water as the mobile phase was utilized for the glucose, xylose, and arabinose analyses of the filtrates from the quantitative saccharification. Cellulose was evaluated as glucan, and Hemicellulosess were assessed as xylan and arabinan [75]. Lastly, according to the Tappi T222 om-88 method (Tappi 1997) [76], the acid-insoluble lignin (Klason lignin) was established.
BET theory rationalizes the physical adsorption of gas molecules on a solid surface and serves as a basis for measuring the specific surface area of a material. Our measurements were made according to DIN 66,132 (DIN 66132, 1975) [77] by using liquid nitrogen (N 2 ) with a Nova ® Surface Area Analyzer (Quantachrome Instruments, Boynton Beach, FL, USA) for approximately 20 h at 105 • C.
Fourier transform infrared (FTIR) spectra were obtained using a spectroscope (MAGNA-IR 750 Spectrometer, Serrie II, Nicolet, Madison, WI, USA). The sampling technique used herein was diffuse reflectance. The powder samples were scanned for a wavenumber of 650-3500 cm −1 .
Hexavalent chromium was monitored by the 1,5-diphenylcarbohydrazide strategy by utilizing solitary dry powder detailing. This reagent contains an acidic buffer joined with 1,5-diphenylcarbohydrazide, which provides a purple color when hexavalent chromium is present. The test outcomes were measured at 540 nm by using a HACH DR4000U UV-visible spectrophotometer (Loveland, CO, USA), i.e., HACH Method 8023. The highest concentration that can be estimated with this technique is 0.7 mg L −1 . Thus, we used a dilution factor of 10-1000. Each sample was measured in triplicate to eliminate dilution error. pH measurements were performed using a WTW MultiLab model 540 (Weilheim, Germany) computerized pH meter.

Adsorption Isotherm Studies
Adsorption isotherms were obtained from batch tests conducted in 500 mL Erlenmeyer flask using a JULABO's digital SW22 Shaking Water Bath (JULABO GmbH, Seelbach, Germany). The sorbent weight was 2 g (i.e., m/V = 4 g L −1 ), the temperature was 23 • C, and the initial Cr(VI) concentration varied from 15 to 700 mg L −1 . K 2 Cr 2 O 7 was used as the Cr(VI) source.
The pH of the solutions was set to two by utilizing diluted H 2 SO 4 . The jugs were fixed and mechanically tumbled for seven days. This timeframe was chosen following the initial investigations to accomplish the equilibrium conditions.
The utilization of adsorption isotherms is extremely helpful in investigating the interaction between the adsorbate and adsorbent in any framework. The parameters received from diverse models provide vital information on the surface properties and affinities of the adsorbent. There are several conditions for breaking down exploratory adsorption equilibrium information, and the best known surface adsorption models for single solute frameworks are the Langmuir and Freundlich models.

Kinetic Adsorption Studies
Batch adsorption rate tests were performed in a 4-L totally blended glass reactor fitted with a bent cutting-edge compost stirrer working at 600 rpm to maintain the lignocellulosic material in suspension. The sorbent weight was 4 g (i.e., m/V = 1 g L −1 ), and the underlying Cr(VI) fixation was 7 mg L −1 . The effect of contact time on the uptake of Cr(VI) was evaluated in batch tests. Furthermore, the pH was adjusted equal to 2 (the pH of the solutions was balanced utilizing diluted H 2 SO 4 ).

Adsorption Column Studies
Experiments with adsorption columns were performed on a 15 × 2.5 cm diameter column that contained 18 g of sorbent. The flow rate was set equal to 2 mL min −1 by using an HPLC pump (LaPrep P110, VWR International, Dublin, Ireland). The column experiments were conducted at 23 • C, the initial Cr(VI) concentration of the solution was 60 mg L −1 , and the pH of the solution was two. Frequently, a sample was collected, and the concentration of Cr(VI) was determined.

Conclusions
Spruce sawdust is a low-cost and efficient absorbent for the removal of Cr(VI) from aqueous solutions. The maximum Cr(VI) adsorption capacity of organosolv-pretreated spruce sawdust using the Langmuir isotherm model is 318.3 mg g −1 . The rate of adsorption was found to follow the pseudo-second-order kinetic model. It was observed that pretreatment with 50% diethylene glycol/50% H 2 O improves the adsorption capacity of sawdust by 84%, whereas pretreatment with 50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 enhances the adsorption capacity of sawdust by 89%.
According to adsorption kinetics, the material pretreated with organic solvent and acid catalyst (50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 ) at the optimal conditions, which were a temperature of 180 • C and pretreatment time of 50 min, adsorbed 3.7 times faster than the raw material (according to the intraparticle diffusion model).
Furthermore, our Thomas model showed that the material pretreated with organic solvent and acid catalyst (50% diethylene glycol/50% H 2 O/0.045 N H 2 SO 4 ) at a temperature of 180 • C and an isothermal reaction time of 50 min has an adsorption capacity that is 13 times that of the untreated material. Overall, our results prove that the organosolv pretreatment of spruce sawdust produces a promising adsorbent candidate for the removal of Cr(VI).
Adsorption columns are widely used in water/wastewater treatment systems and their reliability makes these systems suitable for all types of facilities, e.g., large industrial-scale treatment plants with a high capacity. Furthermore, some pilot scale experiments are necessary to check the industrial reliability of the presented process.