L-Lysine-Modified Lignin for Polishing Alkaline Road-Marking Wash Water: High Uptake of Cationic Dyes with Acid-Enabled Regeneration

Abstract

Road-marking operations generate alkaline wash water with intense color and soluble cationic additives. A new biomass adsorption material (LML) was developed to address dye pollution in road-marking wash water effectively. Enzymatically hydrolyzed lignin was used as the raw material for the first time. L-lysine was modified to the structure of the lignin benzene ring using a simple one-step synthesis method, which endowed lignin with a large number of active carboxyl and amino functional groups to improve its adsorption capacity. The adsorption performance of LML for methylene blue in water was also investigated. The experimental results show that the LML has a high dye removal rate under alkaline conditions. The fitted adsorption model shows that the saturated adsorption capacity of LML for methylene blue (MB) is 129.4 mg g⁻¹ and malachite green (MG) is 244.9 mg g⁻¹, which is in line with the Langmuir isotherm adsorption model. The adsorption process is endothermic, which means that the adsorption capacity increases with increasing temperature. Kinetic studies showed that the adsorption process reached equilibrium within 120 min following a pseudo-second-order kinetic model. The cycle experiment shows that the removal efficiency of the adsorbent for dyes can still reach 90% after five cycles, indicating a good practical application value for the polishing of road-marking wash water.

1. Introduction

Road-marking activities, such as line application, removal of aged markings, and equipment rinsing, generate wash water with intense color, complex soluble organics, abundant suspended solids, and a typically alkaline reaction. Discharging without adequate polishing burdens sewer networks and receiving water, which underscores the need for simple, robust post-treatment prior to discharge or reuse. Recent Water studies have framed this need within resource-efficient, low-maintenance polishing strategies for diverse effluents [1,2,3,4].

A wide variety of technologies are available for color and dye removal, including adsorption, biological treatment, membrane processes, solidification, photolysis, chemical oxidation, and electrochemical oxidation. While effective in specific settings, many advanced options impose operational complexity, energy demand, or high lifecycle costs, and industrial sectors facing corrosive or colored streams have highlighted the value of modular polishing steps that allow recovery or regeneration to contain costs and minimize secondary wastes [5,6].

Among the available approaches, adsorption is particularly appealing owing to its relatively low cost, ease of operation, and high sorbent capacity [7,8]. Traditional adsorbents, including porous activated carbons, metal–organic frameworks, and inorganic silica, have demonstrated considerable effectiveness in dye removal. However, their large-scale application is often constrained by high production costs, challenges in regeneration, and labor-intensive processes [9,10]. These limitations underscore the growing interest in bio-based sorbents, which offer advantages of low cost, tunable surface functionality, and improved reusability [11,12].

Lignin, a major by-product of cellulose manufacturing and the second most abundant biopolymer in nature, bears aliphatic and phenolic hydroxyl and methoxy groups that can engage dyes through combined physical and chemical interactions [13,14,15,16]. However, native lignin often exhibits limited capacity and insufficient stability in complex aqueous matrices, highlighting the importance of molecular modifications to enrich high-affinity sites [17,18].

In this study, L-lysine was grafted onto a low-cost, readily functionalizable lignin backbone through a one-step synthesis to increase the density and diversity of active sites. We evaluate the modified lignin for polishing alkaline road-marking wash water using methylene blue and malachite green as representative cationic chromophores, and we establish isotherm, kinetic, and thermodynamic behavior alongside acid-enabled regeneration [19]. The study positions a regenerable, lignin-based adsorbent as a compact polishing step suited to decentralized treatment and potential on-site reuse. We note that actual road-marking wash water may contain a broader spectrum of pigments and particulates, application to such matrices will require further validation.

2. Materials and Methods

2.1. Materials and Instruments

The samples were characterized by a scanning electron microscope (SEM, S4800, Hitachi, Tokyo, Japan), physical property measurement system (PPMS-9, Quantum Design Instrument, San Diego, CA, USA), Fourier-transform infrared spectrometry (FT-IR360, Nicolet, Madison, WI, USA), Thermogravimetric analyzer (Pyris 1, Perkin Elmer, Waltham, MA, USA) and XRD diffractometer (XRD-6100, Shimadzu, Kyoto, Japan). X-ray photoelectron spectroscopy (XPS) data was gotten with a spectrometer. (PHI 5700 ESCA System, PE, Eden Prairie, MN, USA). The materials were uniformly dispersed with an ultrasonic apparatus (KQ5200E, Kunshan Instrument, Kunshan, China) was used to ensure uniform dispersion. A spectrophotometer (TU-1901, Persee, Beijing, China) was utilized to determine the absorbance of dyes in the solutions. The sorption experiments were carried out in a constant temperature shaking table (SHA-B, Shengtang, Tianjin, China).

Enzymatically hydrolyzed lignin was extracted from corn stover residues. Formaldehyde solution (37% w/w) was purchased from Tianjin Comio Chemicals. L-lysine, malachite green, and methylene blue were obtained from Aladdin (Shanghai, China, analytical grade). Analytical-grade sodium hydroxide and hydrochloric acid were obtained from Tianjin Guangfu. Deionized water was used throughout the experiments.

Real samples: Wash water was collected after roadmarking operations, from equipment rinsing and surface cleanup. The field indices recorded included pH, apparent color, chemical oxygen demand, turbidity, and suspended solids.

Simulated samples: Methylene blue and malachite green served as representative cationic dyes to establish reproducible controls under the preset concentration, temperature, kinetic, and isotherm conditions. A small amount of latex or pigment particulates was added to emulate operational turbidity.

2.2. Experimental Methods

2.2.1. Preparation of L-Lysine-Modified Lignin

LML was synthesized using a one-step grafting route. Dissolve 5.0 g of enzymatically hydrolyzed lignin in 90 mL of 0.1 mol L⁻¹ sodium hydroxide under mechanical stirring for 10 min. Add 2.0 g of L-lysine and 10 mL of formaldehyde, set the mixture to ${90 }^{\circ }\mathrm{C}$, and maintain for 5 h. After reaction, introduce 1.0 mol L⁻¹ hydrochloric acid to precipitate the product. The solid was collected by high-speed centrifugation and washed repeatedly with distilled water until the filtrate was clear and the pH was approximately 7. Dry the final precipitate at ${60 }^{\circ }\mathrm{C}$ under vacuum for 12 h to obtain LML.

2.2.2. Adsorption Experiments

The adsorption of the cationic dyes methylene blue and malachite green was evaluated as a function of initial concentration, temperature, and contact time. An accurately weighed 30 mg portion of the adsorbent was added to 30 mL of dye solution at the prescribed concentration and shaken in a thermostatic water-bath shaker for 24 h at the selected temperature to ensure equilibrium. The supernatant was filtered through a 0.22 μm membrane, and absorbance was recorded by UV–Vis spectrophotometry at 664 nm for methylene blue and 617 nm for malachite green. The equilibrium adsorption capacity Q was calculated as follows:

$$Q={\displaystyle \frac{(C_0-C_e)\times V}{m}}$$

(1)

Symbols are defined as follows: Q denotes the adsorption capacity of the adsorbent for the dye, mg g⁻¹; $C_0$ and $C_e$ denote the initial and equilibrium dye concentrations in solution, mg L⁻¹; m denotes the mass of adsorbent, g; and V denotes the solution volume, L.

The removal efficiency ($R_e$) was calculated from the change in dye concentration according to

$$R_e=\left({\displaystyle \frac{C_0-C_e}{C_0}}\right)\times 100\%$$

(2)

where $C_0$ (mg L⁻¹) and $C_e$ (mg L⁻¹) are the initial and equilibrium dye concentrations, respectively.

Isotherm, kinetic, and thermodynamic behaviour for MB and MG on LML was examined over initial dye concentrations of 40 mg L⁻¹ to 500 mg L⁻¹, contact times of 0 min to 180 min, and temperatures of 298 K, 308 K, and 318 K. Regeneration was performed by eluting the spent adsorbent with 0.1 mol L⁻¹ hydrochloric acid. All experiments have been repeated in triplicate, and the averages of results were taken (mean ± SD, n = 3).

2.2.3. Regeneration Experiment

In this subsection, we specify that 500 mL of textile-mill dye wastewater was treated in cyclic batch adsorption, and the spent adsorbent was regenerated with 0.1 mol L⁻¹ HCl, and the regenerated material was reused in the subsequent cycle. The accompanying discussion in the Results clarifies that the removal efficiency did not exhibit an appreciable decline across cycles, indicating that LML maintains effective adsorption performance under realistic wastewater conditions and routine acid regeneration.

3. Results and Discussion

3.1. Effect of Modification on Adsorption

As shown in Figure 1, a broad absorption band spanning 3400–2800 cm⁻¹ is assigned to the O–H/N–H envelope together with aliphatic –CH₂– vibrations associated with lysine moieties. Distinct bands at ∼1730, ∼1650, and ∼1560 cm⁻¹ are attributed to carboxyl C=O stretching and amide I/II vibrations of grafted lysine, while the band at ∼1410 cm⁻¹ corresponds to the symmetric stretch of COO⁻. In addition, the feature at 1260–1200 cm⁻¹ is assigned to C–N stretching. Taken together, and considering that the aromatic ring bands at ∼1600/1510 cm⁻¹ remain essentially unchanged, these results corroborate that L-lysine is introduced onto the aromatic rings of lignin via a Mannich-type reaction without disrupting the lignin backbone.

As shown in Figure 2, native lignin exhibits weak dye uptake, which is consistent with the limited number of accessible binding sites on the unmodified matrix. After L-lysine grafting, LML provides a higher density of carboxyl and amino sites that interact with cationic chromophores through electrostatic attraction and hydrogen bonding, leading to a clear increase in adsorption. The enhancement was observed for both methylene blue and malachite green, indicating that the modification increased the quantity and strength of the active sites rather than favoring a single dye structure. These results confirm that the molecular modification of lignin is effective for strengthening dye capture and improving polishing performance.

3.2. Effect of Adsorbent Dosage

Optimizing the adsorbent dose is essential to control the cost and effectively use the material. As shown in Figure 3, the removal efficiency increased with increasing dose, attributable to the greater number of accessible active sites at higher solid loadings. Above 30 mg per 30 mL, the improvement became marginal under otherwise identical conditions, indicating that further additions mainly introduced unused sites. Accordingly, 30 mg was selected as the working dose for subsequent tests to balance the performance with the sorbent consumption.

3.3. Effect of pH

As shown in Figure 4, the removal efficiency of LML toward the cationic dyes increases monotonically from 60.11% at pH 2 to 99.98% at pH 9 and then remains essentially constant between pH 9–12 (99.95% at pH 12). L-lysine grafting introduces additional carboxyl and amino functionalities onto the lignin framework. Under acidic conditions, protonation reduces the number of available adsorption sites, leading to lower removal efficiency. Under alkaline conditions, deprotonation enhances electrostatic attraction to the cationic dye, resulting in higher removal efficiencies.

3.4. Isotherm Adsorption

The initial dye concentration strongly influences the uptake of the sorbent. Batch tests were therefore conducted for methylene blue (MB) and malachite green (MG) at ${25 }^{\circ }\mathrm{C}$, ${35 }^{\circ }\mathrm{C}$, and ${45 }^{\circ }\mathrm{C}$ across a series of starting concentrations. In all cases, the equilibrium uptake increased with concentration and then approached a plateau as the surface sites became saturated, consistent with the monolayer coverage on a finite population of active sites.

To clarify the sorption mechanism, the equilibrium data were analyzed using both Langmuir Freundlich isotherm, Sips and Redlich–Peterson models:

The Langmuir model can be expressed as:

$$q_e={\displaystyle \frac{q_m{K}_L{C}_e}{1+K_L{C}_e}}$$

(3)

The Freundlich model can be expressed as:

$$q_e=K_F{C}_e^n$$

(4)

The Sips model can be expressed as:

$$q_e={\displaystyle \frac{q_m{K}_S{C}_e^s}{1+K_S{C}_e^s}}$$

(5)

The Redlich–Peterson model can be expressed as:

$$q_e={\displaystyle \frac{K_{R-P}{C}_e}{1+{\alpha }_{R-P}{C}_e^{\beta }}}$$

(6)

where $q_e$ is the equilibrium adsorption capacity (mg g⁻¹); $q_m$ is the monolayer maximum adsorption capacity (mg g⁻¹); $C_e$ is the equilibrium concentration (mg L⁻¹); $K_L$ is the Langmuir affinity constant (L mg⁻¹); $K_F$ is the Freundlich constant ((mg g⁻¹)/(mg L⁻¹)ⁿ); and n is the Freundlich exponent (dimensionless). $K_S$ is the Sips equilibrium binding constant ((L mg⁻¹)^s), and s is the Sips model parameter (dimensionless). $K_{R-P}$ and ${\alpha }_{R-P}$ are the Redlich–Peterson model constants, and $\beta$ is the corresponding model expo-nent (dimensionless).

Figure 5 and

Table 1. Isotherm model parameters of MB (Data are mean ± SD (n = 3), at 25 °C, SD ≤ 0.20).

Model	Parameters	25 °C	35 °C	45 °C
Langmuir	$K_L$ (L mg−1)	0.69	1.04	1.82
	$q_m$ (mg g−1)	129.4	162.6	220.0
	$R^2$	0.84	0.80	0.70
Freundlich	$K_F$ ((mg g−1)/(mg L−1)n)	72.43	92.74	130.5
	n	1.3	1.4	1.4
	$R^2$	0.84	0.90	0.95
Sips	$q_m$ (mg g−1)	154.3	218.7	320.0
	$K_S$	0.94	0.82	0.78
	s	0.37	0.31	0.24
	$R^2$	0.90	0.93	0.93
Redlich–Peterson	$K_{R-P}$	71.28	78.96	50.00
	${\alpha }_{R-P}$	0.48	0.89	0.10
	$\beta$	1.03	0.89	1.19
	$R^2$	0.84	0.93	0.35

show that the Langmuir model provided consistently superior fits for both methylene blue and malachite green, with the coefficient of determination R² exceeding 0.70 at all three temperatures. At ${25 }^{\circ }\mathrm{C}$, the Langmuir monolayer capacities were 129.4 mg g⁻¹ for methylene blue and 244.9 mg g⁻¹ for malachite green, and the fitted $q_{\max }$ values closely matched the observed plateaus across temperatures. The Freundlich exponent n ranged from 1.2 to 1.6, which lies within the commonly accepted favorable interval of 1 to 10 and indicates progressively enhanced adsorption with increasing concentration. Moreover, both $q_{\max }$ and the Langmuir constant $K_L$ increased with temperature for the two dyes, supporting an endothermic uptake and pointing to a thermally strengthened affinity between the dye molecules and the modified lignin surface.

The non-linear isotherm fits are presented in Figure 5, and the fitted parameters are listed in Table 1. Based on R², the adsorption of MB on LML at 25 °C and 35 °C is best captured by the Sips model. The Langmuir and Freundlich models describe the limiting homogeneous and heterogeneous cases, respectively, while Sips spans the low- to high-concentration regime and is applicable to solid–liquid systems irrespective of surface uniformity. The Redlich–Peterson equation likewise overcomes concentration-range limitations, approaching the Freundlich and Langmuir forms as $\beta$ → 0 and $\beta$ → 1, respectively. Because Sips better describes high-concentration behavior than the Freundlich model, its superior fit corroborates the porous structure of LML. In addition, the best fit at 35 °C is given by the Freundlich model, further suggesting that elevating the temperature intensifies heterogeneous adsorption, primarily by altering the uniformity of surface adsorption sites. Overall, the MB adsorption on LML is most suitably described by the Sips model, and increasing temperature tends to render the surface adsorption sites more heterogeneous. Model coefficients in Equations (3)–(6) were estimated from the non-linear (untransformed) forms using least-squares fitting. Inearized plots, when shown, are for visualization only.

The linearized isotherm fits are shown in Figure 5, and the corresponding fitting parameters are summarized in

Table 2. Isotherm model parameters of MG.

Model	Parameters	25 °C	35 °C	45 °C
Langmuir	$K_L$ (L mg−1)	1.16	0.95	4.84
	$q_m$ (mg g−1)	244.9	300.0	320.0
	$R^2$	0.90	0.84	0.76
Freundlich	$K_F$ ((mg g−1)/(mg L−1)n)	123.3	142.9	182.4
	n	1.5	1.6	1.2
	$R^2$	0.93	0.95	0.94
Sips	$q_m$ (mg g−1)	292.9	369.2	386.8
	$K_S$	0.70	0.65	1.07
	s	0.46	0.38	0.32
	$R^2$	0.99	0.99	0.98
Redlich–Peterson	$K_{R-P}$	774.2	300	320
	${\alpha }_{R-P}$	4.81	1.38	1.08
	$\beta$	0.91	0.07	0.02
	$R^2$	0.99	0.90	0.72

. Comparing the coefficients of determination (R²) for the four models indicates that the adsorption of MG on LML is best described by the Sips model.

3.5. Adsorption Kinetics

Contact time is a key factor in practical applications. As shown in Figure 6, the uptake of both methylene blue (MB) and malachite green (MG) on LML increased rapidly at the outset due to abundant vacant active sites and then increased more slowly and reached equilibrium at approximately 120 min. This behavior indicates a strong affinity and fast adsorption rate; the rapid stage is attributable to surface carboxyl, amine, and hydroxyl groups together with the lignin aromatic domains.

The overall kinetics were described using the pseudo-first-order (PFO) and pseudo-second-order (PSO) models:

$$\mathrm{PFO}:\ln (q_e-q_t)=\ln q_e-k_{1}t$$

(7)

$$\mathrm{PSO}:{\displaystyle \frac{t}{q_t}}={\displaystyle \frac{1}{k_2{q}_e^2}}+{\displaystyle \frac{t}{q_e}}$$

(8)

In these equations, $q_e$ and $q_t$ are the equilibrium uptake and the uptake at time, respectively, $k_1$ is the PFO rate constant, and $k_2$ is the PSO rate constant.

The fitted lines and the parameters obtained from these linearizations are presented in Figure 6 and

Table 3. Parameters of kinetic adsorption models.

Dye	Kinetic Model	Kinetic Parameter
Methylene blue	Pseudo-first-order	$q_e$ = 98.80 mg g−1 $k_1$ = 0.0536 min−1 R2 = 0.9710
	Pseudo-second-order	$q_e$ = 108.17 mg g−1 $k_2$ = 8.1949 g mg−1 min−1 R2 = 0.9811
Malachite green	Pseudo-first-order	$q_e$ = 201.64 mg g−1 $k_1$ = 0.0251 min−1 R2 = 0.9290
	Pseudo-second-order	$q_e$ = 243.53 mg g−1 $k_2$ = 1.1885 g mg−1 min−1 R2 = 0.9589

, respectively. For both dyes, the PSO linearization yields consistently higher R² values than the PFO linearization, and the PSO-derived $q_e$ agrees more closely with the experimental equilibrium uptake. These results indicate that adsorption of MB and MG on LML is better described by the PSO model, consistent with a chemisorption-controlled process. Moreover, the non-zero intercepts in Figure 6 suggest that intraparticle diffusion contributes but is not the sole rate-limiting step.

3.6. Thermodynamics of Adsorption

Thermodynamic analysis based on the temperature dependence of the equilibrium uptake was used to assess spontaneity and interfacial disorder. The standard Gibbs free energy change $\mathsf{\Delta }G$, enthalpy change $\mathsf{\Delta }H$, and entropy change $\mathsf{\Delta }S$ are obtained as follows:

$$\mathsf{\Delta }G=-RT\ln K_0$$

(9)

$$\ln K_0={\displaystyle \frac{\mathsf{\Delta }S}{R}}-{\displaystyle \frac{\mathsf{\Delta }H}{RT}}$$

(10)

where $K_0$ is the equilibrium constant, R is the universal gas constant (8.314 × 10⁻³ kJ mol⁻¹ K⁻¹); T ($\mathrm{K}$) is the absolute temperature in Kelvin, $\mathsf{\Delta }G$ (kJ mol⁻¹) is the Gibbs free energy change, $\mathsf{\Delta }H$ (kJ mol⁻¹) is the enthalpy change, and $\mathsf{\Delta }S$ (J mol⁻¹ K⁻¹) is the entropy change.

As shown in Figure 7, $\ln K_0$ varies linearly with $1/T$, indicating that the van’t Hoff approach is applicable. The fitted parameters are summarized in

Table 4. Thermodynamic parameters for dye adsorption on LML.

Dye	T (K)	ln(K0)	$\mathbf{\Delta }{\mathit{G}}^{\circ }$ (kJ mol−1)	$\mathbf{\Delta }{\mathit{H}}^{\circ }$ (kJ mol−1)	$\mathbf{\Delta }{\mathit{S}}^{\circ }$ (J mol−1 K−1)	R2
	298	13.28	−32.91
Methylene blue	308	13.54	−34.67	15.13	161.4	0.9683
	318	13.66	−36.13
	298	11.99	−29.70
Malachite green	308	12.21	−31.27	32.31	207.5	0.9252
	318	12.81	−33.87

. The negative $\mathsf{\Delta }G$ values at all temperatures confirmed that dye adsorption on the LML was spontaneous. A positive $\mathsf{\Delta }H$ indicates an endothermic process, and, thus, a higher temperature promotes uptake. The positive $\mathsf{\Delta }S$ further suggests increased randomness at the solid–liquid interface during adsorption.

3.7. Reusability of LML

Operational stability and reuse, quantified by capacity retention and regeneration efficiency, are increasingly emphasized for dye-sorbent deployment [20,21]. As shown in Figure 8, the removal efficiencies of methylene blue and malachite green by LML were 99.9% after the first regeneration cycle.

With increasing number of cycles, the efficiency declined slightly, reaching approximately 91% after five cycles, and then remained above 90% under identical conditions [22]. The modest loss in performance is plausibly associated with partial blocking or alteration of active sites during acid desorption, and with minor sorbent loss during handling [23]. Overall, the results indicate that LML can be regenerated with hydrochloric acid and reused for at least five cycles with limited performance decay, which also aligns with the modular polishing and potential acid recovery strategies.

Under the regeneration condition, the removal efficiency of LML remained essentially unchanged over repeated adsorption–desorption runs, suggesting stable adsorption sites and practical reusability under our operating window. We note that this conclusion is performance-based; comprehensive post-cycle structural characterization will be pursued in future work to further substantiate the material’s long-term integrity.

Compared with other studies, the adsorbent developed in this work exhibits superior adsorption performance (

Table 5. Comparison with other methods.

Adsorbent	Equilibrium Time	Adsorption Capacity	Refs
Pristine lignin	≈30 min	MG 31.2 mg g−1	[24]
Activated carbon from fig leaves	200 min	MG 51.79 mg g−1	[25]
Activated carbon from Rumex abyssinicus stems	40 min	MG 98.43 mg g−1	[26]
KOH-activated bamboo biochar (KBBC-900)	20 min	MB 67.71 ± 0.19 mg g−1	[27]
Natural zeolite	60 min	MB 21.19 mg g−1	[28]
LML	120 min	MG 244.9 mg g−1 MB 129.4 mg g−1	This study

).

4. Conclusions

This study targeted road-marking wash water and demonstrated that L-lysine-modified lignin (LML) efficiently removes cationic colorants [29]. L-lysine was grafted onto the lignin aromatic framework through a one-step synthesis, enriching the carboxyl and amino functional groups, thereby enhancing the adsorption capacity for dye purification [30,31]. LML showed strong performance toward methylene blue and malachite green [32,33], with Langmuir monolayer capacities of 129.4 mg g⁻¹ and 244.9 mg g⁻¹, respectively, and reached equilibrium within 120 min. Reusability tests also showed that LML can be reused over multiple cycles, indicating good regenerability [34,35]. Overall, the modified lignin adsorbent combines low cost, renewability, and suitability for field deployment, offering a practical route for the rapid polishing and potential reuse of wastewater generated during road-marking construction and maintenance [36].