Application of Modified Lignocellulosic Biomass for Sorption of Anionic Dye Reactive Black 5 in an Air-Lift and Column Reactor

Abstract

The study presents research on the use of modified lignocellulosic biomass as a waste sorbent for the removal of anionic dyes from aqueous solutions. The sawdust used as sorption material was subjected to an acid-base modification and further functionalised by introducing amino groups into the biomass structure. Dynamic sorption experiments were carried out in two reactor types (airlift and column) with two sorbents: sawdust treated under acid-base conditions (S-AB) and sawdust aminated with epichlorohydrin after acid-base treatment and preactivation (S-AB-EA). The anionic dye Reactive Black 5 (RB5) was used as a sorbate. The experiments were carried out at two flow rates (0.1 and 0.5 dm³/h) and two feed concentrations (10 and 50 mg/dm³), maintaining the pH of the solution at 3, as determined in previous studies. The experimental data allowed the maximum sorption capacities of the tested sorbents to be determined under dynamic conditions and were described using the Thomas, Adams-Bohart and Yoon-Nelson models. The results showed that the flow rate, the dye concentration and the reactor type strongly influence the efficiency of dye removal. The highest capacity, 73.89 mg/g, was achieved in the airlift reactor for aminated sawdust and preactivation with epichlorohydrin (S-AB-EA) at a feed concentration of 50 mg/dm³ and a flow rate of 0.1 dm³/h.

1. Introduction

Industrial wastewater contains a wide variety of organic and inorganic compounds generated as by-products or residues of various technological processes. Its treatment poses a significant challenge for wastewater treatment engineers, particularly in the context of increasingly stringent environmental regulations that limit the permissible concentrations of pollutants discharged into surface waters.

Globally, approximately one million tonnes of dyes are produced annually, with anionic reactive dyes dominating textile applications. During dyeing, only a fraction of these dyes binds to fibres, while the rest is released into wastewater at concentrations up to 100 mg/L [1,2,3].

Conventional biological and physicochemical treatment methods, such as activated sludge, membrane bioreactors, coagulation–flocculation, or membrane filtration, are often ineffective for dye removal because dye molecules are chemically stable, resistant to microbial degradation, and possess complex aromatic structures [4]. Among the various approaches for dye removal, an interesting and promising method is the use of ionic liquids, which have been shown to effectively remove anionic dyes from contaminated water, with the added advantage of enabling subsequent recycling of the ionic liquid [5]. As a result, significant amounts of dyes persist through treatment and are discharged into surface waters.

The presence of dyes in aquatic environments poses severe ecological and health risks. Even at concentrations as low as 1 ppm, dyes impart intense colour, reduce light penetration, inhibit photosynthesis, lower dissolved oxygen levels, and disturb aquatic ecosystems. Moreover, many dyes and their degradation products exhibit carcinogenic and mutagenic properties, threatening both aquatic life and human health [6,7,8].

Given these limitations and environmental hazards, there is an urgent need for the development of efficient, sustainable, and economically viable wastewater treatment methods [9]. Adsorption has emerged as one of the most promising approaches due to its high efficiency, operational simplicity, and adaptability to various wastewater types [10,11]. However, the high cost of commercial activated carbon motivates the exploration of low-cost, eco-friendly alternatives derived from natural or waste materials. In this context, lignocellulosic waste such as sawdust represents an abundant and renewable resource that, when chemically modified (e.g., aminated), can achieve high adsorption capacities for anionic dyes while supporting circular economy principles and sustainable wastewater management.

Activated carbon is the most commonly used adsorbent for the removal of a large number of contaminants and is the most widely used adsorbent for dye removal [12]. Its large surface area and well-developed porosity provide high sorption efficiency [13]. However, the high production and regeneration costs of activated carbon limit its large-scale industrial application, motivating the search for cheaper alternative materials. Numerous studies describe the use of natural or waste-derived sorbents—such as rice husks, papaya seeds, rapeseed stalks, sunflower shells, compost, or feathers—as materials capable of adsorbing dyes [14,15,16,17,18,19,20,21]. These materials typically contain functional groups, including hydroxyl (-OH) and carboxyl (-COOH) groups, which enable the binding of cationic dyes. However, their acidic character limits their affinity for anionic dyes, which predominate in textile wastewater [22].

Sawdust, a lignocellulosic by-product of the wood industry, is an abundant, inexpensive, and environmentally friendly material with a structure favourable for sorption processes. Raw sawdust, rich in cellulose, hemicellulose, and lignin, contains numerous hydroxyl groups that can be chemically modified to improve its adsorption properties [23]. Chemical modification of natural sorbents further increases their adsorption capacity. Alkaline treatment with NaOH enlarges the specific surface area and introduces a negative charge that favours cationic dye adsorption, whereas acidic modification introduces acidic groups that enhance sorption of anionic dyes [24,25,26,27]. Amination, in turn, introduces amine groups, significantly improving the removal efficiency of anionic dyes. The degree of functionalisation and resulting adsorption capacity depend on pH, temperature, reagent dosage, and activation method; therefore, optimising these parameters is crucial [28,29].

One of the most promising approaches is amination, which involves introducing amine groups (-NH₂) into the sawdust structure. This modification imparts a basic character to the material, enhancing electrostatic interactions with anionic dye molecules [30,31,32]. Aminated sawdust exhibits structural similarities to chitosan—a natural polymer well known for its high adsorption capacity towards anionic dyes [24,33,34]—but, unlike chitosan, it is significantly cheaper and more readily available on an industrial scale.

Recent studies have highlighted the increasing interest in using aminated lignocellulosic materials as efficient adsorbents for removing synthetic dyes from aqueous solutions. Amination of lignocellulosic biomass increases the number of active adsorption sites, significantly improving dye removal efficiency. For example, aminated phenolated lignin has been shown to effectively remove anionic dyes under dynamic conditions [35], while reusable lignin-based hydrogel biocomposites have demonstrated high sorption capacity for various organic dyes [36]. Additionally, thermally modified lignocellulosic waste has been successfully applied in air-lift reactors for the removal of toxic dyes [37]. The effectiveness of the amination process depends on several factors, including the type of aminating agent, activation method, and reaction conditions (pH, temperature, reagent concentration). Epichlorohydrin is commonly used as an activating agent, as it reacts with polysaccharides and facilitates the introduction of amine groups. Other aminating agents reported in the literature include melamine, ethylenediamine, and triethylenetetramine, which differ in functional group density and structure, thereby influencing the adsorption capacity of the resulting sorbent [24,28].

Two main types of systems are widely used for removing dyes and other contaminants from industrial wastewater: packed-bed columns filled with sorbent and air-lift reactors. Packed-bed columns enable contact between the liquid and porous sorbent under continuous flow conditions, providing high adsorption efficiency with a relatively simple design [38,39]. The performance of these reactors depends on factors such as bed height, particle size, flow rate, and the chemical properties of the sorbent [40]. However, potential channeling of the flow and uneven sorbent utilisation can shorten the column’s operational life [41]. An alternative is the air-lift reactor, in which gas (usually air) is introduced to induce liquid circulation between the riser and downcomer zones. This improves mixing, mass transfer, and uniform contaminant distribution while minimising the formation of dead zones [42,43].

The research presented in this study addresses the urgent environmental challenge of removing anionic dyes from industrial wastewater. Demonstrating that waste-derived, aminated lignocellulosic materials can achieve high adsorption capacities opens the possibility of implementing a sustainable and economically feasible solution for large-scale wastewater treatment. The results obtained are significant for the development of new sorbents and the optimisation of technological processes, bridging the gap between laboratory studies and industrial applications.

2. Materials and Methods

2.1. Dyes

For this study, an anionic dye, Reactive Black 5, and a cationic dye, Basic Violet 10, both widely used in the textile industry, were employed. Both dyes were supplied by the “Boruta” S.A. Dye Production Plant (Zgierz, Poland). Detailed information on their chemical structures, molecular weights, and relevant physicochemical properties is provided in

Table 1. Characteristics of the dyes used in this study.

Dye Name	Reactive Black 5—(RB5)	Basic Violet 10—(BV10)
Structural formula
Molar weight	991 g/mol	479 g/mol
λmax	600 [nm]	554 [nm]
Type of dye	anionic–reactive	cationic
Use	Dyeing of wool, cotton, viscose, polyamide fibers.	Dyeing paper, leather, cotton. Paint production.

.

2.2. Beech Sawdust

The beech sawdust used in this study was supplied by the company “Kaczkan” and is a waste product from wooden floor production. For the experiments, a sawdust fraction with a particle size of 0.5 to 0.6 mm was selected. The moisture content of the beech sawdust was 7%. The fibre fractions were determined using the Van Soest chemical analysis method [44], which involves the selective separation of different fibre components under specific conditions using surfactants. The characteristics of the raw material, expressed as a percentage of dry matter (% w/w), are presented in

Table 2. Characteristics of beech sawdust.

Component	Dry Matter Content
Cellulose	41.0%
Hemicellulose	27.9%
Lignin	26.7%
Ash	0.1%
Extracts and other ingredients	4.3%

.

2.3. Preparation of Sorbents Used in the Study: S-AB and S-AB-EA

The sorbents used in this study were prepared following the procedure described in the publication “Effect of Beech Sawdust Conditions Modification on the Efficiency of Sorption of Anionic and Cationic Dyes” [24].

The same sorbent preparation method was used in this study to ensure comparability of results and to relate the obtained data to previous literature reports. A detailed description of the process, including modification conditions (types of reagents used, treatment time, and temperature), is provided in the cited publication and forms the basis of the experimental procedure used here.

2.4. FTIR Analysis

The functional groups present on the sorbent surface and the types of chemical bonds in the sorbents were characterised using Fourier Transform Infrared (FTIR) spectroscopy, employing the attenuated total reflectance (ATR) technique with a single-reflection attachment. Prior to analysis, the lignocellulosic sorbents were dehydrated using a hydraulic press. The prepared sample was then pressed against the ATR crystal with a controlled force. After each measurement, the diamond ATR crystal was cleaned with acetone (CH₃COCH₃), the background spectrum was recorded, and the next sample was placed on the crystal. FTIR spectra were collected over the range 4000–600 cm⁻¹ with a resolution of 2 cm⁻¹.

2.5. Studies on Sorption Under Flow Conditions

The efficiency of dye sorption under flow conditions was investigated using two types of reactors: an air-lift reactor and a packed-bed column. Two sorbents, S-AB and S-AB-EA, were used in the studies of dye sorption under flow conditions. The sorption efficiency of the selected sorbents was tested for the dye Reactive Black 5. Experiments were conducted at a constant sorbent dose and for two flow rates and two influent dye concentrations. The dye concentration of the feed solution was 10 and 50 mg/dm³.

The flow of the dye solution through the reactor was set at 0.1 V/h and 0.5 V/h for the air-lift reactor, and 2.5 V/h and 12 V/h for the packed-bed column (where V denotes the reactor volume), corresponding to 0.1 dm³/h and 0.5 dm³/h, respectively. Samples (10 cm³) for dye concentration analysis at the reactor effluent were taken directly from pocket settler compartments. The sampling frequency was 12 times per day (every 2 h). Experiments were carried out until the effluent dye concentration reached the same value as the influent concentration (C = C₀). The technological assumptions of the flow-through sorption studies are presented in

Table 3. Operational parameters for flow-through sorption experiments.

Parameter	Unit	Value
Solution flow rate	dm3/h	0.1
		0.5
Sorbent dosage	g s.m.	5
Dye concentration in the influent	mg/dm3	10
		50

.

2.5.1. Sorption Studies in an Air-Lift Reactor

The air-lift reactor had a total volume of 1 dm³. It had a circular cross-section (Φ = 0.06 m), a height of 0.43 m, and was made of plexiglass. Inside the reactor, a vertical baffle 0.25 m long was installed centrally, 0.05 m below the outlet nozzle. In the lower part of the reactor, two nozzles were installed to introduce air and the dye solution. The supplied air ensured circulation of the solution and sorbent around the baffle. In the upper part of the reactor, a pocket settler with an outlet nozzle was installed. Air was supplied to the reactor using an aeration pump at an airflow rate of 50 dm³/h. This gas flow prevented sedimentation of the sorbent in the lower part of the reactor and ensured its uniform distribution throughout the reactor volume. The dye solution was introduced into the reactor using a peristaltic pump (Figure 1).

2.5.2. Sorption Studies in a Packed Column Reactor

The packed column reactor had a total volume of 0.04 dm³. It had a circular cross-section (Φ = 0.025 m), a height of 0.18 m, and was made of plexiglass. In the lower part of the reactor, a nozzle allowed the introduction of the dye solution. In the upper part, a pocket settler with an outlet nozzle was installed. The dye solution was supplied to the reactor using a peristaltic pump (Figure 2).

2.6. Determination of Dye Concentration in Reactors

During the study, dye concentrations at the reactor outflow were determined using solution samples collected at 12 intervals per day (every 2 h). Dye concentrations were measured spectrophotometrically using a UV-3100 PC spectrophotometer (VWR Spectrophotometer, VWR International LLC., Mississauga, ON, Canada). The amount of dye adsorbed was calculated according to the following equation:

$${\mathrm{Q}}_{\mathrm{s}}={\displaystyle \frac{{{\mathrm{C}}_0-\mathrm{C}}_{\mathrm{s}}}{\mathrm{m}}}$$

(1)

where Q_s—mass of sorbed dye [mg/g]; C₀—initial dye concentration [mg/dm³]; C_s—dye concentration after sorption [mg/dm³]; m—concentration of the sorbent [mg/dm³]

The maximum sorption capacity of chitosan sorbents was determined based on mathematical models and calculated using the Statistica 12 programme.

2.7. Calculation Methods

The experimental results were analysed using the Thomas, Bohart–Adams, and Yoon–Nelson models.

Thomas model:

$${\displaystyle \frac{C}{C_0}}={\displaystyle \frac{1}{1+\mathit{exp}\left[\frac{k_{Th}}{Q}\hspace{0.33em}\cdot \left(q\hspace{0.33em}\cdot m-C_0\hspace{0.33em}\cdot V\right)\right]}}$$

(2)

where C—concentration at the outflow [mg/dm³]; C₀—initial concentration of the adsorbate [mg/dm³]; k_Th—rate constant [cm³/mg-min]; Q—maximum capacity [mg/g]; m—mass of the adsorbent in the reactor [g]; q—flow rate in the reactor [cm³/min]; V—volume of solution passed through the column [cm³].

Bohart–Adams model:

$${\displaystyle \frac{C}{C_0}}={\displaystyle \frac{1}{1+\mathit{exp}\left[\left(k_{BA}\cdot q_{BA}\cdot \frac{H}{v}\right)-k_{BA}\cdot C_0\cdot t\right]}}$$

(3)

where C—concentration of the dye solution at the outflow [mmol/dm³]; C₀—initial concentration of the dye solution [mmol/dm³]; k_BA—rate constant [dm³/mmol·h]; q_BA—adsorption efficiency [mmol/dm³]; H—height of the column [m]; v—linear flow velocity [m/h]; t—operating time of the reactor [h].

Yoon-Nelson model:

$${\displaystyle \frac{C}{C_0}}={\displaystyle \frac{c_0\hspace{0.33em}\cdot \mathit{exp}\left[k_{YN}\hspace{0.33em}\cdot \left(t-\tau \right)\right]}{1+\mathit{exp}\left[k_{YN}\hspace{0.33em}\cdot \left(t-\tau \right)\hspace{0.33em}\cdot C_0\right]}}$$

(4)

where C—concentration of the dye solution at the outflow [mmol/dm³]; C_o—initial concentration of the dye solution [mmol/dm³]; k_YN—rate constant [1/h]; Q—maximum capacity [mmol/g]; m—mass of the adsorbent in the reactor [g]; τ—time after which 50% of the dye was removed (C/C₀ = 0.5 h) [h]; q—flow rate in the reactor [cm³/min].

The experimental results were analysed using the Thomas, Bohart-Adams, and Yoon-Nelson models. The goodness of fit between the model predictions and the experimental data was evaluated using the coefficient of determination (R²). This coefficient indicates how well the model represents the observed data, with higher R² values reflecting better agreement between the experimental results and the model.

$$R^2={\displaystyle \frac{{\sum }_{i=1}^n={{(Q}_i-Q\left(C_i)\right)}^2}{\sum _{i=1}^n{\left(Q_i-\overline{Q}\right)}^2}}$$

(5)

where Q—${\displaystyle \frac{C_0-C}{m}}$; C₀—initial concentration [mg/dm³]; m—sorbent concentration in the tested sample [g/dm³].

2.8. Analytical Methods

Determination of Dye Concentration in Solution

The concentration of dye remaining in the aqueous solution was determined spectrophotometrically using a Genesys 20 spectrophotometer (Thermo Scientific, Waltham, MA, USA). Standard calibration curves were prepared for each dye tested. The absorbance of the solutions was measured at the specific wavelength corresponding to the maximum absorbance of each dye, with 600 nm used for Reactive Black 5 (RB5).

3. Results

This study aimed to evaluate the performance of the most effective sorbent, S-AB-A, for the removal of anionic dyes, using Reactive Black 5 as a model compound, under continuous-flow conditions in both air-lift and packed-column reactors. Experiments with the unmodified S-AB sorbent were conducted to enable a comparative assessment of the increase in sorption capacity resulting from the applied modifications.

To investigate the effect of sawdust modifications on their structural properties, FTIR analyses were performed for all sorbents: S—unmodified sawdust; S-AB—sawdust treated with acid and alkali; S-AB-E—sawdust further activated with epichlorohydrin; S-AB-A—sawdust activated with a 25% aqueous ammonia solution; and S-AB-E-A—sawdust subjected to acid and alkali treatment and subsequently activated with epichlorohydrin and a 25% ammonia solution (Figure 3).

FTIR Analysis of Sorbents FTIR analysis was conducted to investigate the structural changes in sawdust resulting from various chemical modifications. The spectra of all sorbents—unmodified sawdust (S), acid- and base-treated sawdust (S-AB), epichlorohydrin-activated sawdust (S-AB-E), ammonia-activated sawdust (S-AB-A), and sawdust sequentially treated with acid/base, epichlorohydrin, and ammonia (S-AB-E-A)—showed characteristic bands corresponding to functional groups present in the polysaccharide structure. In all spectra, a prominent peak was observed between 3600 and 3300 cm⁻¹, associated with O–H stretching vibrations in the cellulose ring and side chains (CH–OH and CH₂–OH) [45]. Absorption in the 3000–2800 cm⁻¹ range corresponds to C–H stretching vibrations [46,47]. A peak around 1743 cm⁻¹, assigned to C=O stretching, was significantly reduced in the spectrum of S-AB compared to unmodified sawdust (S) and disappeared entirely in the spectra of S-AB-E, S-AB-A, and S-AB-E-A. This reduction or disappearance indicates chemical reactions between the carbonyl groups and modifying agents [48,49]. Several bands were observed in the 1500–1200 cm⁻¹ range, corresponding to deformations of primary and secondary O–H groups, while stretching vibrations of C–O bonds appeared between 1200 and 1000 cm⁻¹ [47]. Notably, in the spectra of sawdust treated with ammonia (S-AB-A) and aminated after sequential activation (S-AB-E-A), a new peak appeared at 1640–1660 cm⁻¹, characteristic of N–H bending vibrations. This peak confirms the successful amination process, indicating the introduction of amino groups into the polysaccharide structure [47,50,51]. Overall, the FTIR analysis confirmed that each modification step altered the functional groups of sawdust, increasing its potential for dye adsorption by introducing reactive sites for chemical interactions.

Sorption studies were conducted for two dye concentrations, 10 and 50 mg/dm³, and two solution flow rates through the reactors, 0.1 dm³/h and 0.5 dm³/h, at pH 3, which was the pH of the RB5 dye solution used in the flow-through experiments.

Table 4. Constants determined from mathematical models.

Type of Reactor	Type of Sorbent	Flow [dm3/h]	Dye Concentration [mg/dm3]	Sorbent Sorption Depletion Time [h] (Ce = C0)	Thomas’s Model			Adams-Bohart Model			Yoon-Nelson Model
					kTh ml/min∙mg	q [mg/g]	R2	kAB mg/min	q [mg/g]	R2	kYN [mg/min]	q [mg/g]	R2
Reaktor air-lift	S-AB	0.1	10	72	0.2411	2.8	0.9666	0.0002	2.78	0.9660	0.0024	2.4	0.9520
			50	64	0.0578	11.2	0.9780	0.0001	11.2	0.9780	0.0029	10.0	0.9708
		0.5	10	62	0.9686	3.6	0.9809	0.0010	3.6	0.9809	0.0099	3.3	0.9758
			50	20	0.2672	12.7	0.9731	0.0003	12.7	0.9731	0.0128	13.9	0.9694
	S-AB-EA	0.1	10	646	0.0223	69.3	0.9946	0.0000	69.3	0.9946	0.0002	69.3	0.9946
			50	304	0.0136	73.9	0.9885	0.0000	73.9	0.9885	0.0007	70.1	0.9855
		0.5	10	214	0.0766	42.0	0.9890	0.0001	42.0	0.9890	0.0008	40.1	0.9872
			50	134	0.0565	40.0	0.9426	0.0001	39.9	0.9426	0.0027	35.1	0.9374
Reaktor column	S-AB	0.1	10	18	6.6419	0.3	0.9789	0.0066	0.3	0.9789	0.0431	0.25	0.9751
			50	10	1.2469	1.6	0.9975	0.0012	1.6	0.9975	0.0368	1.2	0.9915
		0.5	10	12	8.9545	1.5	0.9978	0.0089	1.5	0.9978	0.0519	1.1	0.9958
			50	14	2.3210	7.5	0.9995	0.0024	7.6	0.9995	0.0677	5.2	0.9990
	S-AB-EA	0.1	10	438	0.0307	37.4	0.9814	0.0000	37.4	0.9814	0.0003	36.02	0.9786
			50	218	0.0191	38.8	0.9564	0.0000	38.7	0.9564	0.0010	33.9	0.9436
		0.5	10	178	0.0833	33.7	0.9403	0.0000	33.7	0.9403	0.0008	30.0	0.9317
			50	76	0.0738	35.6	0.9554	0.0000	35.6	0.9554	0.0036	30.0	0.9444

presents the sorbent constants and maximum sorption capacities determined using mathematical models, the goodness of fit of the models to the experimental data (R²), and the total operating time of the reactor depending on the flow rate and dye solution concentration, assuming that the dye concentration in the reactor outflow was equal to the influent dye concentration (Ce = C₀).

According to the literature, the Thomas, Yoon–Nelson, and Bohart–Adams models are primarily used to determine the sorption capacities of sorbents in fixed-bed column reactors [52,53]. However, based on the results presented in Table 4, it can be concluded that these models can also be applied to describe data obtained from loop reactors, as for all three models and all experimental series, both for the air-lift and column reactors, very good fits to the experimental data were achieved (R² values ranging from 0.9317 to 0.9995). For both reactors, the best fit was obtained using the Thomas and Bohart–Adams models, with identical values for both models. Literature also confirms that the Thomas and Bohart–Adams models are mathematically equivalent, and their parameters are interchangeable [54]. Due to its greater popularity and the ease of comparing results with those of other authors, the obtained data are further discussed based on the Thomas model.

A clear advantage of the modified sorbent (S-AB-EA) over the unmodified sorbent (S-AB) is observed. In both reactor types, regardless of dye concentration and flow rate, the modified sorbent demonstrates a significantly higher maximum sorption capacity (q). For example, in the air-lift reactor at a flow rate of 0.1 dm³/h and a dye concentration of 10 mg/dm³, q for S-AB is 2.79 mg/g, while for S-AB-EA it reaches 69.26 mg/g. A similar trend is observed in the column reactor.

For the S-AB sorbent in the air-lift reactor, increasing the dye concentration from 10 to 50 mg/dm³ significantly reduces the time to sorbent saturation—from 72 h to 64 h at a flow rate of 0.1 dm³/h, and from 62 h to 20 h at 0.5 dm³/h. This indicates that higher dye concentrations cause faster saturation of the unmodified sorbent. The modified sorbent S-AB-EA shows a similar trend, but its operational time is significantly longer.

Higher flow rates result in shorter sorbent saturation times for all sorbents and both reactor types. In the air-lift reactor at 50 mg/dm³, the operational time of S-AB decreases from 64 h (0.1 dm³/h) to 20 h (0.5 dm³/h), while for S-AB-EA it drops from 304 h to 134 h. A similar effect is observed in the column reactor. This indicates that higher flow rates reduce the contact time between the sorbent and the solution, decreasing sorption efficiency.

The high adsorption efficiency of anionic dyes at low pH is primarily due to electrostatic attraction between the positively charged surface of the aminated sorbent and the negatively charged dye molecules. Under acidic conditions, amine groups are readily protonated according to the reaction: –NH₂ + H₃O⁺ → –NH₃⁺ + H₂O The resulting –NH₃⁺ groups interact electrostatically with the sulfonate groups (–SO₃⁻) of the anionic dyes (such as Reactive Black 5). This attraction is the dominant adsorption mechanism at pH = 3, accounting for the high sorption capacities observed experimentally. In addition to electrostatic interactions, hydrogen bonding may also contribute to dye binding. Such bonds can form between the hydrogen atoms of protonated amine or hydroxyl groups on the sorbent and the oxygen or nitrogen atoms of the dye molecules. However, under strongly acidic conditions, hydrogen bonding plays a secondary role compared to electrostatic forces. As the pH increases, the degree of protonation of amine groups decreases, reducing the number of positive sites available for anionic dye binding. Simultaneously, deprotonation of hydroxyl and carboxyl groups leads to the formation of negatively charged sites (–O⁻, –COO⁻), which electrostatically repel anionic dye molecules and lower adsorption efficiency. Regarding stability, the amine groups introduced through epichlorohydrin-assisted amination are chemically bonded to the lignocellulosic matrix through stable covalent linkages. These bonds are expected to remain intact during moderate regeneration cycles (e.g., rinsing with dilute acid or ethanol). However, repeated exposure to strong alkaline solutions (e.g., NaOH) could lead to partial hydrolysis or desorption of amine functionalities, gradually reducing adsorption capacity. Therefore, while the functional groups are expected to be stable under mild regeneration conditions, further systematic studies on desorption and reusability would be beneficial to confirm the long-term chemical stability and operational performance of the modified sorbents.

4. Discussion

The air-lift reactor provides a longer operational time for S-AB at low concentrations and low flow rates compared to the column reactor. However, for modified sorbents, both reactors achieve very high sorption capacities, although the air-lift reactor allows slightly longer operation at low flow rates.

The studies showed that a higher influent dye concentration leads to a decrease in the Thomas model rate constant (k_Th) and an increase in the amount of dye adsorbed on the sorbent (q) (Table 4). This effect may be due to the fact that the driving force of sorption is the concentration difference between the dye on the sorbent and the dye in the solution [55,56,57]. A higher driving force caused by the higher dye concentration results in better reactor performance. At higher dye concentrations, the probability of collisions between dye molecules and adsorption sites increases [58]. Conversely, as the flow rate increases, the k_Th constant rises, which is attributed to a reduction in mass transfer resistance [59,60]. A similar trend in the Thomas model rate constant was observed in literature data on chromium sorption using modified montmorillonite clay [61]. The increase in sorbent capacity with increasing initial dye concentration is also confirmed by Hameed [62], who reported that the initial concentration affects the overcoming of all mass transfer resistances between the aqueous and solid phases. Therefore, the higher the initial dye concentration, the greater the efficiency of the sorption process. Similar relationships were observed in studies on the sorption of Methyl Orange and Methyl Violet onto activated carbon derived from Phragmites australis ([63,64]) and Methylene Blue onto tea leaves [62]. The breakthrough time (Ce = C₀) was longer in all cases for the air-lift reactor compared to the column reactor (Table 4). For the tested sorbents and conditions, the longest period of high and stable sorption efficiency under flow conditions was obtained for S-AB-EA with an RB5 dye concentration of 10 mg/dm³ and a flow rate of 0.1 dm³/h. Under these conditions, the reactor reached Ce = C₀ after 646 h. The shortest reactor operation time—10 h—was observed for the column reactor with S-AB at a flow rate of 0.1 dm³/h and a dye concentration of 50 mg/dm³. Overall, the air-lift reactor achieved significantly longer operation time for all experimental series. Both the achieved sorption capacities and reactor operation times indicate the higher efficiency of the loop reactor compared to the fixed-bed column reactor. The results demonstrate that the flow rate and influent dye concentration significantly affect reactor operation time; increasing either parameter shortens the operation time. Based on the results, it was confirmed that higher flow rates and initial dye concentrations lead to a reduction in reactor operation time. A similar relationship was observed by Filipkowska and Waraksa [65], who studied dye sorption on chitosan under dynamic conditions. They obtained the longest reactor operation time (approximately 35 h) for the series with the lowest flow rate, which was 1 V/h (V—reactor volume) [65]. The higher efficiency of air-lift reactors compared to fixed-bed column reactors is also supported by literature data. This is due to the effective utilisation of the sorbent in the air-lift system, where the dye is evenly distributed throughout the reactor volume. In contrast, in fixed-bed column reactors, there is a significant risk of uneven dye distribution, leading to zones with higher or lower wastewater flow [66].

The relevance of kinetic and breakthrough models (such as Thomas, Yoon–Nelson, and Bohart–Adams) for scaling up industrial applications lies in their ability to predict sorption performance under continuous flow conditions. These models provide quantitative relationships between operational parameters (flow rate, influent concentration, sorbent mass) and key performance indicators, such as: 1. Sorption capacity (q_m)—the maximum amount of dye adsorbed per gram of sorbent. 2. Breakthrough curves—indicating the point at which the effluent concentration approaches the influent concentration (Ce → C₀). 3. Rate constants—describing the speed of sorption, which affects reactor residence time and efficiency. By fitting experimental data to these models, engineers can predict reactor performance at larger scales without conducting full-scale trials, optimise reactor design and operating conditions (such as flow rate, bed height, and sorbent dosage) to achieve desired removal efficiencies, estimate service times and replacement intervals for the sorbent to reduce downtime and operational costs, and compare different reactor configurations (e.g., air-lift vs. fixed-bed) under standardised conditions. In this study, the very good fit of the models (R² = 0.93–0.99) to both air-lift and packed-bed data indicates that these models can reliably guide industrial-scale implementation of aminated sawdust sorbents for continuous dye removal, facilitating process design, scaling, and performance prediction.

The results demonstrate the high efficiency of aminated lignocellulosic adsorbents in dynamic dye removal systems. These results may also have environmental implications. Using lignocellulosic waste and textile wastewater in dye removal processes is an effective and simple technological approach that reduces the negative environmental impact of industrial wastewater [67]. This method also aligns with the principles of the circular economy, using wastewater for water purification and the recovery of valuable resources [68]. Integrating innovative adsorption materials with well-designed reactor systems not only increases process efficiency but also provides a practical example of applying sustainable development principles in the industrial water treatment sector.

5. Conclusions

The removal of the anionic dye RB5 under flow conditions was significantly more efficient on aminated beech sawdust (S-AB-EA) than on unmodified S-AB. The results clearly demonstrate that, in addition to the type of sorbent used, both the wastewater flow rate and the influent dye concentration strongly influence reactor efficiency and operating time. The highest sorption capacity was obtained in the air-lift reactor using S-AB-EA at an initial dye concentration of 50 mg/dm³ and a flow rate of 0.1 dm³/h. Under these conditions, the maximum sorption capacity for S-AB-EA reached 73.89 mg/g, a value comparable to that achieved under static conditions, whereas for S-AB it was only 11.18 mg/g. The longest reactor operating time, defined as the point at which C₀ = Ce, was 646 h and was recorded for an initial concentration of 10 mg/dm³ and a flow rate of 0.1 dm³/h in the air-lift reactor. The study showed that the same mass of aminated sorbent placed in the air-lift reactor was capable of treating dye solutions with up to 25 times greater efficiency and for a period nine times longer than the fixed-bed reactor filled with S_AB. Increasing the wastewater flow rate or influent dye concentration shortened the reactor operating time, while a higher influent dye concentration increased the sorption capacity of the sorbent. Based on both the sorption capacities and operating times obtained, the air-lift reactor proved significantly more effective than the fixed-bed column reactor.