Lignin‐Derived Sustainable Nano‐Platforms: A Multifunctional Solution for an Efficient Dye Removal

Abstract

In contrast to conventional non‐biobased adsorbents, lignin emerges as a cost‐effective and environmentally benign alternative for water treatment. This study identifies unexpected and unpredicted multifunctional properties of lignin nanoparticles (LNPs). LNPs, which are prepared by simple physical processes, demonstrated for the first time to behave as multifunctional materials able to adsorb and photodegrade methylene blue (MB) in aqueous medium upon UV irradiation. Furthermore, the synthetic approach adopted to synthesize LNPs – and therefore their surface properties – strongly affects their performances. More specifically, LNPs obtained by solvent‐antisolvent nanoprecipitation (SLNPs) show the highest MB adsorption properties (98 % removal), reaching a maximum adsorption capacity of 43.0 mg g⁻¹, and the fastest adsorption kinetics with respect to other lignin‐based adsorbents. Conversely, hydrotropic LNPs (HLNPs) exhibit exceptional photocatalytic activity, resulting in 98 % MB degradation over 6 hours of UV irradiation, combined with the ability to be easily recycled and reused. The present effort paves the way for the use of LNPs as efficient multifunctional materials able to perform concurrently adsorption and photocatalytic degradation of dye pollutants, toward the creation of a sustainable biobased water treatment platform.

The study demonstrates the multifunctionality of lignin nanoparticles (LNPs) in water treatment, acting as efficient adsorbents and photocatalysts for the dye methylene blue (MB). The LNPs showed fast adsorption kinetics and remarkable photocatalytic activity under UV irradiation. This research highlights the potential of lignin nanoparticles for sustainable biobased water treatment, providing adsorption and photodegradation capabilities simultaneously.

Introduction

Water is essential to life. This was emphasized by the United Nations General Assembly in New York on 28th July 2010 in the Resolution 64/292, where the “right to safe and clean drinking water and sanitation” ^[1] is acknowledged as an extension of the “right to life” ^[2] proclaimed by the same Assembly in Paris on 10th December 1948. Water is the pillar which supports directly and indirectly every aspect of human economic activities, from primary to quinary sector. It is thus necessary that water resources be reliable and safe. Nevertheless, pollution and natural resource crises have occupied the highest positions in the World Economic Forum′s Global Risks Landscape in the last 3 years,[ ³ , ⁴ , ⁵ ] and water pollution represents one of the most pressing environmental challenges to tackle. ^[6]

In this context, the challenge to reduce pollution levels and prevent pollution‐related diseases can be addressed considering wastewater no more a problem, but a reliable alternative source of water. ^[7]

With the aim of water reuse and recycling in mind, the quest for technological advancements of wastewater treatment is thus a crucial issue to improve the quality of the purified water, which can afterwards be safely reused or reintroduced into the water cycle following the principles of circular bioeconomy. ^[8]

Biomass‐based nanomaterials play a key role in this transition. ^[9] Among the different types of biomass, lignocellulose represents the most abundant renewable raw material on Earth with an estimated production of ca. 10 billion tons each year. ^[10]

In the last years, researchers devoted their efforts on the development of lignocellulosic nanomaterials as a replacement for conventional nanomaterials that may cause significant environmental issues.[ ⁹ , ¹¹ , ¹² ] Various fabrication methods have been developed and a wide range of applications have been studied for such bionanomaterials.[ ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ ] However, the commercialization of lignocellulosic nanomaterials is expensive, necessitating a cost‐effective industrial research along with environmental‐friendly production processes.

To achieve this goal, lignin nanoparticles (LNPs) are particularly promising due to the ease of synthesis, the possibility to entrap a wide variety of chemical species and to release them in response of external stimuli. ^[17]

Therefore, LNPs are ideal candidates to tackle the challenge of effectively and affordably removing persistent micropollutants from water, particularly synthetic organic dyes such as methylene blue (MB). Current methods of water remediation include membrane filtration and biological treatment, but they often face criticism due to high costs, energy consumption, and generation of secondary pollutants.[ ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² ]

Adsorption is a commonly used approach due to its simplicity, availability of adsorbents, selectivity, scalability, and the possibility of regeneration and reusability. ^[23] Due to its effectiveness and large surface area (800–1000 m² g⁻¹ on average for water treatment applications), activated carbon is one of the most widely used adsorbents. ^[24] However, the high cost of regeneration and the environmental impact significantly limit its use. ^[25]

Photocatalysis is another effective method for degrading pollutants in wastewater, capturing researchers′ interest for its potential to combat persistent pollutants such as synthetic organic dyes.[ ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ ] Its key advantage is the generation of harmless by‐products (CO₂, H₂O and inorganic salts) upon pollutant degradation, thereby mitigating the impact on the environment. ^[32] However, these systems commonly rely on semiconductors such as TiO₂ or ZnO synthesized from chemical precursors lacking a biobased origin.[ ²⁶ , ²⁷ , ²⁹ , ³⁰ , ³¹ , ³³ ]

In recent years, the integration of adsorption and photocatalysis has gained interest.[ ²⁹ , ³¹ , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ ] This approach capitalizes on the unique advantages of each method, making them complementary and addressing their respective limitations. Adsorption efficiently binds pollutants, while photocatalysis uses light‐activated processes for degradation. This synergistic integration promises more effective and sustainable water treatment strategies.

Nevertheless, there are no instances in the existing literature where a single‐component material is used for both adsorption and photocatalysis. Typically, examples involve composite materials in which certain components are dedicated to adsorption and others to photocatalysis.[ ²⁹ , ³¹ , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ ] In exploring this pathway, our study focuses on lignin, a particularly intriguing material with numerous unknown properties.

With the aim of exploiting LNPs features for water remediation purposes, LNPs were prepared using two different methods, solvent‐antisolvent (SLNPs) and hydrotropic (HLNPs). ^[16] The unprecedented multifunctional properties of lignin nanoparticles as adsorbent and concurrently photocatalyst were investigated and proved via experiments of adsorption and subsequent UV light‐driven photocatalytic degradation of aqueous solutions of MB.

Results and Discussion

HLNPs and SLNPs were synthesized in 80 % and 60 %, yield respectively Table 1. The different observed yields can be explained considering that in the hydrotropic approach also lower molecular weight lignin molecules precipitate from the solution forming the outer shell of the LNPs. Conversely, in the solvent‐antisolvent strategy, the lower molecular weight lignin fraction remains in the aqueous solution, as previously reported, thus determining a lower overall yield. ^[16]

Table 1.

Size measured by DLS and LDA, and ζ‐potential of lignin species; yields of synthesis for LNPs.

	Size/PDI[a] [nm]/–	Dx 90[b] [nm]	ζ‐potential [mV]	Yield [%]
SLNP	165/0.199	440	−59.4	60
HLNP	295/0.480	460	−41.4	80
SKL	1070/0.505	1550	−33.8	–

[a] Polydispersity index. [b] Dx 90 indicate the size below which 90 % of all particles are found.

The samples were dimensionally analyzed by means of dynamic light scattering (DLS) and laser diffraction analysis (LDA) and the relevant results are summarized in Table 1 (see supporting information Figure S1 for the distributions). The average particle size obtained from the DLS analysis, for SLNPs, HLNPs, and softwood Kraft lignin (SKL) were 165, 295, and 1070 nm, respectively, revealing that SLNPs exhibited a 100‐nm smaller size than HLNPs, confirming the yield values. LDA results showed that 90 % of the particles were below 440, 460 and 1550 nm (Dx 90) for SLNP, HLNP and SKL, respectively. The LDA method highlighted the same order of magnitude in terms of particle size and a similar trend, i. e. HLNPs slightly bigger than SNLPs, in agreement with DLS results and previously reported data. ^[16]

The ζ‐potential values of LNPs and SKL are reported in Table 1. Both HLNPs and SLNPs exhibited a more negative surface charge than SKL, and SLNPs presented a more negative value compared to HLNPs, which can be justified by two factors. The first is the smaller size of SLNPs and thus their higher surface‐area‐to‐volume ratio, which increases the fraction of OH groups exposed on the surface. The second is related to the composition of the NPs: the nucleation and growth processes of LNPs are reported to be kinetically dependent on the molecular weight of lignin chains. First, the large molecular weight lignin fractions nucleate, which is followed by the adsorption of lower molecular weight fractions onto these nuclei. ^[38] The structural characterization of these two fractions revealed that large molecular weight chains are endowed with a relatively low concentration of condensed phenolic and guaiacyl units, while a relatively high concentration of both these units characterizes the low molecular weight chains. ^[39]

Previous GPC analysis on the samples under investigation indicate that the lowest molecular weight fraction of pristine lignin is not involved in the formation of both SLNPs and HLNPs preparation. In addition, SLNPs (M_n 4500 Da) are composed of a lower amount of low molecular weight lignins than HLNPs (M_n 2500 Da). ³¹P‐NMR analyses reported a very similar concentration of condensed phenolic units between HLNPs and SLNPs and a higher concentration of guaiacyl units in SLNPs with respect to HLNPs. ^[16] This results in a higher concentration of total phenolic OH, justifying the more negative zeta potential exhibited by SLNPs. Furthermore, the different nature of the solvents used in the preparation method of SLNPs and HLNPs and the nature of the interactions of each solvent with lignin may further impact the distribution of OH groups on the surface of the particles resulting in different ζ‐potential values. ^[16] As a consequence, the ζ‐potential values indicated increasing colloidal stability in the order SKL<HLNPs<SLNPs. ^[40] The observed trend can be rationalized in terms of balance between water‐lignin hydrogen‐bond interactions and lignin intra/inter‐chain π‐π interactions: pristine SKL lacks a core‐shell structure such as SLNPs and HLNPs, so that water acts predominantly as antisolvent favouring intra/inter‐chain π‐π interactions with formation of large agglomerates. Conversely, the presence of a shell composed of low molecular weight lignin in SLNPs and HLNPs increases the surface hydrophilicity of these species, favoring hydrogen‐bond interactions between lignin chains and water molecules. ^[41] As previously discussed, the higher concentration of phenolic OH groups in SLNPs accounts for the higher colloidal stability with respect to HLNPs.

Further evidence of the difference of the surface chemistry of SKL and HLNPs was obtained via X‐ray photoemission Spectroscopy (XPS) analysis. Only oxygen (O1s core level) and carbon (C1s core level) characterized the elemental composition of the samples. In addition, it was possible to discriminate the oxygen‐containing functional groups of SKL and HLNPs, to understand each sample‘s peculiarities (as shown in Figure 1 and quantified in Table S1). Indeed, HLNPs were characterized by a lower contribution relative to aromatic carbonyl and carboxyl (C(=O*)O) groups (peak 1) and aliphatic carboxyl (C(=O*)O), carbonyl, ether and hydroxyl groups (peak 2), while they exhibited aliphatic carboxyl (C(=O)O*) groups and aromatic carboxyl (C(=O)O*), aromatic ether and phenolic groups (peak 3) to a greater extent with respect to SKL. ^[42]

Figure 1.

X‐ray photoemission spectroscopy (XPS) results with deconvoluted spectra of O1s core levels of SKL (a) and HLNPs (b). Each peak is associated with specific surface functional groups highlighting the different surface chemistry of the investigated samples. Peak 1 is associated to aromatic carbonyl and carboxyl (C(=O*)O) groups, Peak 2 to aliphatic carboxyl (C(=O*)O)/carbonyl/ether/hydroxyl groups, and Peak 3 to aromatic carboxyl (C(=O)O*)/ether/hydroxyl groups and aliphatic carboxyl (C(=O)O*) groups.

These results suggest that a higher concentration of phenolic OH groups is exposed on HLNPs surface than in the case of SKL, which is in line with the different zeta potential values discussed above.

Adsorption Studies

The adsorption kinetic experiments were performed to determine the adsorption rate constants and the adsorption capacity of SKL, HLNPs and SLNPs (each at 1.5 mg mL⁻¹) using a 30 mg L⁻¹ MB solution (Figure S2). It is noteworthy that after 120 min SLNPs adsorbed MB almost completely (98 % removal), with SKL and HLNPs being the second and the third most effective (73 %, and 56 %), respectively (Figure 2a, Equation (2)). The different removal ability of SLNPs and HLNPs can be explained in terms of surface‐area‐to‐volume ratio: SLNPs are composed of smaller primary nanoparticles (hydrodynamic diameter=165 nm, Table 1) and therefore, they are endowed with the highest specific surface area, resulting in a better MB percentage removal. Interestingly, SKL exhibited better removal performance than HLNPs, which could be attributed to the higher porosity of SKL aggregates, which allows a higher number of phenolic groups to be available for MB adsorption with respect to the less porous HLNPs. ^[16]

Figure 2.

UV‐Vis absorption spectra of MB (30 mg L⁻¹) at t=0 (purple) and t=120 min of adsorption for SLNPs (green), HLNPs (red), and SKL (blue) (1.5 mg mL⁻¹) (a); kinetics of MB adsorption (30 mg L⁻¹) (b), and adsorption isotherms (c), for SLNPs (green squares), HLNPs (red dots), and SKL (blue triangles) (1.5 mg mL⁻¹) in 30 mg L⁻¹ aqueous solution of MB.

Figure 2b shows the adsorption kinetics for SKL, SLNPs and HLNPs (Equation (3)). After 120 min, the adsorption capacity of the three samples reached its maximum value, defined as the equilibrium adsorption capacity (q_e ). SLNPs presented the highest adsorption capacity (18.1 mg g⁻¹), while SKL and HLNPs displayed lower values, respectively equal to 13.6 and 10.6 mg g⁻¹. These data confirmed the trend observed for the percentages of removal.

The adsorption kinetics data were fitted using pseudo‐first‐order (PFO) and pseudo‐second‐order (PSO) kinetics models (Figure S3). As evidenced by the correlation coefficient values (adj. R²) reported in Table 2, the kinetic data of all samples are better fitted with the PSO model, ^[43] revealing that the process is kinetically limited by the MB adsorption step, governed by the electrostatic interaction between the positively charged MB and the negatively charged SKL/HLNPs/SLNPs. ^[44] As described in the archival literature, both hydrogen bonding and π‐π interactions can also occur during the adsorption process. ^[45] The kinetic constants obtained from the models are reported in Table 2.

Table 2.

Correlation coefficient values (Adj. R²) and kinetic constants obtained from fitting data with pseudo‐first‐order (PFO) and pseudo‐second‐order (PSO) models.

Sample	SLNP	HLNP	SKL
	PFO Model
Adj. R2	0.85	0.62	0.85
qe/mg g−1	4.69	6.51	6.28
k1/min−1	0.05	0.02	0.02

	PSO Model
Adj. R2	0.99	0.96	0.99
qe/mg g−1	16.22	1.41	3.21
k2/g mg−1 min−1	0.07	5.10	1.28

In order to obtain the adsorption isotherms and the equilibrium values, kinetic experiments were performed with different concentrations of MB solutions (5, 10, 20, 30, 60, and 100 mg L⁻¹) and the concentration of lignin species (SKL or LNPs) equal to 1.5 mg mL⁻¹. Figure 2c shows the adsorption isotherms of SLNPs, HLNPs, and SKL. The maximum adsorption capacity (q_m ) resulted to be equal to 43.0, 12.2, and 12.8 mg g⁻¹, respectively, highlighting the best adsorption properties toward MB of SLNPs, with SKL and HLNPs to follow.

The isotherm data were compared with Langmuir and Freundlich models to evaluate the adsorption mechanism. From Table 3 and Figure S4, it is evident that the Langmuir model is a better fit for all the obtained isotherms, in full agreement with previous research findings.[ ⁴⁵ , ⁴⁶ , ⁴⁷ , ⁴⁸ , ⁴⁹ ] The model suggests the formation of a monolayer of MB on the surface of SKL and of LNPs.[ ⁵⁰ , ⁵¹ ] Therefore, it was possible to assume that the electrostatic interaction between the deprotonated phenolic groups on the surface of the lignin particles and the cationic MB was the major driving force for the adsorption process.

Table 3.

Correlation coefficient values (Adj. R²) from fitting data with Langmuir and Freundlich adsorption models and Langmuir and Freundlich model parameters.

Sample	SLNP	HLNP	SKL
	Langmuir Model
Adj. R2	0.99	0.99	0.99
qm /mg g−1	42.99	12.25	12.75
KL /L mg−1	1.07	2.52	2.54

	Freundlich Model
Adj. R2	0.54	0.47	0.79
KF /L1/n mg(1−1/n) g−1	14.42	7.08	5.71
1/n	0.47	0.18	0.23

The maximum adsorption capacity (q_max ) of lignocellulosic‐based materials used for the adsorption of MB were compared with the experimental data obtained in this study (Table 4). The values determined in our experiments are comparable with organosolv lignin particles derived from rice straw. However, in our experimental conditions, the removal reached the adsorption‐desorption equilibrium in 120 min compared to the previously demonstrated value of 250 min. ^[46] Even though some studies achieved higher adsorption capacities, the adsorption time frame (tens of hours) was much more extended than the one in the present study.[ ⁴⁵ , ⁴⁷ , ⁴⁸ , ⁴⁹ ]

Table 4.

Adsorption capacities of lignocellulosic‐based materials for MB reported in the literature.

Adsorbent	qmax (mg g−1)	Adsorption Time	Reference
Lignin‐Silica Hybrid Composites	60	50 h	[45]
Organosolv Lignin from Rice Straw	40	250 min	[46]
Reticulated Formic Lignin from Sugar Cane Bagasse	14.6	10 days	[47]
Sulfuric Acid Lignin from Spent Coffee Ground	66.2	24 h	[48]
Chitosan/Nano‐Lignin Composite	74	50 h	[49]
SLNPs	43.0	120 min	This Study
HLNPs	12.2	120 min	This Study
SKL	12.8	120 min	This Study

UV Light Photocatalytic Activity Studies

The development of multifunctional materials capable of both adsorbing and degrading pollutants is considered an environmentally friendly approach, characterized by the fact that less chemicals need to be used or released and little to no residues are produced. In exploring this pathway, the study focuses on lignin, a particularly intriguing material with numerous unknown properties. The investigation aims to determine whether lignin, known for its versatility, can also act as a photocatalyst. Remarkably, the results show that both SKL and LNPs exhibit the ability to photocatalytically degrade adsorbed MB upon UV irradiation (Figure 3a and b), a discovery not previously documented in the literature. Notably, this research addresses a gap in existing knowledge, as there is no information on the use of lignin as a photocatalyst or its potential for dual functionality in both adsorption and photocatalytic degradation processes. This characteristic improves the sustainability of the process as it enables recycling and reuse of particles, thus minimizing waste generation.

Figure 3.

Variations of the UV‐Vis absorption spectra of MB relative to photodegradation experiments at t=0 and t=6 h of UV irradiation, for HLNPs (a) and SKL (b); photodegradation kinetics (c) and pseudo‐first‐order (PFO) photodegradation kinetic model fittings (d) for HLNPs (red dots) and SKL (blue triangles) (1.5 mg mL⁻¹) in aqueous solution of MB (initial concentration 30 mg L⁻¹) under UV irradiation.

Two control experiments were conducted to confirm that SKL and LNPs in fact act as photocatalysts. For the first control experiment (Figure S4a), a solution of 30 mg L⁻¹ MB in the absence of lignin was left under UV irradiation for 360 min to verify the effect of the UV irradiation (photolysis) on MB. The second control experiment (Figure S4b) was conducted to investigate the degradation of MB in the absence of UV irradiation, yet in the presence of lignin particles. A solution of 30 mg L⁻¹ MB and 1.5 mg mL⁻¹ HLNPs was left under stirring in absence of any light for 120 min to reach the adsorption‐desorption equilibrium. Afterwards, the solution was kept under stirring in dark conditions for 360 min. As shown in the figure, in both control experiments non‐significant variations of the concentration of MB were observed (9.2 % decrease and 4.6 % increase in absorbance for the first and second control experiment, respectively). In the case of photolysis of MB (Figure S4a) the positive variation is reasonably ascribed to evaporation, while in the case of the second control experiment (Figure S4b), the decrease could be attributed to a still ongoing adsorption process.

For SLNPs, no significant decrease in the absorbance of MB was observed following a 360‐min irradiation period, indicating the absence of photodegradation. This observation suggests that the adsorption process led to the nearly complete coverage of SLNP surfaces by MB molecules. Consequently, the saturation of SLNPs with MB caused the UV light to be absorbed by MB, preventing the photoexcitation of lignin, which is considered the process triggering the photodegradation of MB (see below).

Conversely, the photodegradation experiments carried out in the presence of HLNPs, highlighted a photodegradation of 98 % of the initial MB (Figure 3a) after 6 h of UV irradiation, while SKL was able to photodegrade 69 % (Figure 3b) of the initial MB (Equation (8)).

The photocatalytic degradation kinetic data (Figure 3c) were fitted using pseudo‐first‐order (PFO) model (Equation (9)). This model yielded an accurate fitting of the data (Figure 3d, adj. R² is 0.87 and 0.83 for HLNP and SKL, respectively), indicating that HLNPs photodegraded MB with a faster rate (0.010 min⁻¹) compared to SKL (0.003 min⁻¹) (Table S2). Furthermore, this result is in accordance with Langmuir‐Hinshelwood model, which describes systems with fast adsorption/desorption equilibrium and a subsequent slow reaction surface step. ^[52] The comparison of the obtained rate constants with the state‐of‐the‐art literature is cumbersome, due the different operational parameters such as UV light source, irradiance, MB initial concentration, and photocatalyst dose. Considering these limitations, the performance of HLNPs were of the same order of magnitude of the ones reported for different non‐biobased photocatalysts (e. g. 1 %Pd‐TiO₂, 0.026 min⁻¹; 80 %brookite‐20 %rutile TiO₂ 0.020 min⁻¹; 70 %CeO₂/g‐C₃N₄, 0.016 min⁻¹; nanostructured ZnO, 0.021 min⁻¹) (cf. Table S3).[ ⁵³ , ⁵⁴ , ⁵⁵ , ⁵⁶ ]

In addition, the preparation of HLNPs is intrinsically sustainable, since it makes use of an abundant and inexpensive waste product and does not involve the utilization of hazardous chemicals, which represents an advantage in terms of resource efficiency with respect to non‐biobased photocatalysts.

Scavenger Experiments

Scavenger experiments were performed to investigate the mechanism of photodegradation of methylene blue. Considering that, although the particles’ morphology and concentration of functional groups on their surface are different, the starting material is the same, it is highly probable that the mechanism of photodegradation remains consistent for both HLNPs and SKL. Therefore, in this section the experiments were performed on HLNPs, which showed higher photodegradation ability.

The necessity to use relatively high concentrations of radical scavengers (0.5 м) to inhibit this process suggested that HLNPs formed a considerable amount of free radical species under UV irradiation. As shown in Figure 4a and b, a 41 % inhibition of MB photodegradation was observed in the presence of AgNO₃ (e⁻ scavenger), while values of 31 % for 2‐propanol (OH⋅ scavenger), and 15 % for Na‐EDTA (h⁺ scavenger) were registered. These results indicate that photoelectrons have a major role in the photodegradation of methylene blue, i. e. its photoreduction occurs as the main degradation pathway via photoexcited lignin states. Moreover, OH radicals and photoholes also contribute to the photodegradation process via the oxidation of MB, although to a lesser extent.

Figure 4.

Photodegradation kinetics (a) and degree of degradation (b) of MB (initial concentration 30 mg L⁻¹) under UV irradiation in the presence of HLNPs at 1.5 mg mL⁻¹ and free radical scavengers; degree of adsorption and photodegradation of MB (initial concentration 30 mg L⁻¹) by HLNPs (1.5 mg mL⁻¹) measured in three cycles (c).

Recyclability Experiments

Recyclability tests were carried out to evaluate the potential reuse of lignin particles after washing and drying. However, these tests encountered challenges inherent to the intrinsic nature of lignin as a material, which posed significant obstacles. The use of organic solvents such as ethanol or methanol, characterized by high solubility toward MB, ^[57] was not practical as these solvents led to the partial dissolution of lignin, with consequent modification of the structure of the particles. Experiments with different ratios of these solvents mixed with water resulted in ineffective MB removal and/or lignin fractionation/dissolution. Furthermore, experiments with different concentrations of aqueous NaOH solution for MB desorption from the particles surface proved to be inefficient. Strong acidic conditions were considered unfavorable due to their potential to induce lignin aggregation and structural changes. On these premises, a combination of mild acidic conditions and ultrasound was chosen for effective MB removal during the washing process.

In the attempt to assess the reusability of the particles, three adsorption and photodegradation cycles were performed, washing the nanoparticles before the subsequent cycle. Interestingly, the degree of adsorption remained relatively stable during the three cycles (Figure 4c). The relatively small decrease in the third cycle was attributed to the accumulation of MB molecules and degradation products on the particle surface during each cycle, as the surface of the particles was only partially cleaned by the washing steps. Concurrently, a progressive decrease in the degree of degradation after each cycle was registered (Figure 4c). These results could as well be attributed to the incomplete desorption of the MB molecules and photodegradation products from the particle surface during each cycle. This accumulation resulted in a reduced photoexcitable surface area of lignin particles, which reduces photocatalytic degree of degradation and leaves less surface area available for subsequent adsorption cycles. Additionally, these results are consistent with the single‐layer adsorption of MB molecules on the particles surface (see Adsorption Studies).

Conclusions

The full utilization of biomass, the most suitable substitute for fossil‐based resources, plays a central role in the adoption of circular bioeconomy paradigms that limit the current dependence on non‐renewable resources and minimizes environmental and health impacts. Therefore, in the present study, nanomaterials from underutilized, yet largely available biomass residues, were prepared.

Specifically, LNPs were produced via easy and straightforward physical processes and were used for water remediation purposes. For the first time, the ability of LNPs to act as multifunctional materials for adsorption and UV‐light driven photocatalytic degradation of the dye MB in water was demonstrated.

SLNPs were found to be the most effective in terms of removal of MB from water (98 % removal), reaching within 120 min a maximum adsorption capacity of 43.0 mg g⁻¹, comparable to values reported for other lignin‐based adsorbents in the literature. Nevertheless, the relevant adsorption kinetics were much faster. The maximum adsorption capacity for HLNPs and SKL was 12.2 and 12.8 mg g⁻¹, respectively, also in this case with a much faster adsorption kinetics compared to similar adsorbents previously reported. Data fitting suggest that the electrostatic interaction between the phenolic groups on the surface of LNPs and SKL and MB is the major driving force for its adsorption.

The photocatalytic activity experiments showed that both SKL and HLNPs have the ability to degrade adsorbed methylene blue (MB) upon UV irradiation ‐ a ground‐breaking discovery that has not been documented before. HLNPs and SKL demonstrated excellent photocatalytic activity after 360 min of UV irradiation (98 % and 69 % MB degradation, respectively). The free radical scavenger experiments revealed that a combination of free radicals (photoelectrons, photoholes and OH radicals) is involved in the UV‐light‐driven degradation of MB.

Most importantly, the recyclability experiments proved the reusability of HLNPs up to three cycles, even if a decrease in the adsorption capacity and photocatalytic degradation degree was recorded.

Overall, this work demonstrated that lignin nanoparticles can act as a multifunctional material, performing concurrently adsorption and photodegradation of methylene blue dye, with a potential to substitute conventional photocatalysts as a sustainable biomass‐derived alternative via the reduction of the environmental impact of their preparation. These characteristics make lignin nanoparticles a promising constituent for a future sustainable biobased water treatment.

Experimental Section

Materials

Softwood Kraft lignin (SKL) was provided by Stora Enso. Sodium p‐toluenesulfonate (Na‐PTS, 95 %), methylene blue (MB, ≥82 %), hydrochloric acid (HCl, 37 %), ethylenediaminetetraacetic acid disodium salt dihydrate (Na‐EDTA, 99–101 %), silver nitrate (AgNO₃, ≥99 %), and 2‐propanol ((CH₃)₂CHOH, ≥99.5 %) were purchased from Sigma‐Aldrich. Ultrapure water produced by means of Milli‐Q® Reference system that generates Type I ultrapure water (18.2 MΩ cm resistivity @25 °C, ≤5 ppb total organic carbon (TOC)) was used in the experiments unless otherwise specified.

Synthesis and Characterization of LNPs

Synthesis via Hydrotropic Method (HLNPs)

The HLNPs were prepared according to a previously reported protocol with minor modifications, as illustrated in Scheme 1a. ^[9] Briefly, SKL (0.25 g) were added to Na‐PTS aqueous solution (50 mL of a 2 м). The mixture was sonicated at 40 °C for 40 min, and then filtered under vacuum to remove undissolved residues (29 %). The lignin/Na‐PTS solution was added dropwise to of ultrapure water (150 mL) while stirring, then the pH of the mixture was adjusted to 2 with HCl solution (2 м). The obtained suspension was centrifuged at 23269 g for 15 min, and the solid was washed and centrifuged 3 times with ultrapure water. The prepared HLNPs were freeze‐dried for 24 h to yield a brown powder.

Scheme 1.

Preparation of nanoparticles via hydrotropic method (HLNPs) (a), and solvent‐antisolvent method (SLNPs) (b). Na‐PTS: sodium p‐toluene sulfonate.

Synthesis via Solvent‐Antisolvent Method (SLNPs)

The SLNPs were synthesized according to the previously reported protocol with minor modifications, as illustrated in Scheme 1b. ^[9] First, a mixture of 70 : 30 ethanol/water (100 mL) was prepared, and then SKL was added in a 21 g L⁻¹ concentration (2.1 g). The mixture was sonicated at 40 °C for 40 min, and then filtered under vacuum to remove undissolved residues (19 %). The obtained lignin‐ethanol/water solution was added rapidly to ultrapure water (1 L) to achieve a 10‐fold dilution. The obtained suspension was centrifuged at 23269 g for 15 min to recover the SLNPs, which were subsequently freeze‐dried for 24 h to yield a brown powder.

Characterization of LNPs

Laser diffraction analysis (LDA) was used to investigate the size of larger particles and possible aggregates (up to 3.5 mm) for SKL and LNPs. A Malvern Mastersizer 3000, equipped with a Hydro EV wet dispersion unit, including an in‐line sonication probe for agglomerate dispersion, was used. The samples were suspended in water and the analysis was conducted at 25 °C. Spherical particles for LNPs, and general purpose for SKL, fine powder mode sensitivity, and narrow mode particle distribution were considered for the software elaboration of the results. For each sample, at least 10 measurements were recorded.

Dynamic light scattering (DLS) analyses of the SKL and LNPs were carried out by means of a Malvern Zetasizer Ultra instrument to investigate the size of smaller size populations (1‐1000 nm) and the surface charge (ζ‐potential). A folded capillary zeta cell was employed for ζ‐potential measurements.

The X‐ray Photoelectron Spectroscopy (XPS) analysis was performed in Ultra High Vacuum (UGV, <10–9 mBar) using SPECS PHOIBOS 150 XPS system equipped with monochromatic Al Kα (1486.6 eV) X‐ray source (XR50 MF) and high‐speed imaging 2D CMOS true counting detector. The samples were prepared over a custom‐made gold‐sputtered sample holder. For this study, a full survey and two different core levels, namely C1s and O1s were investigated for each sample, with a passing energy set to 10 eV. The data fitting was performed using Kalibri KolXPD software using the same criteria of previous studies on carbonaceous materials.[ ⁵⁸ , ⁵⁹ , ⁶⁰ , ⁶¹ , ⁶² , ⁶³ , ⁶⁴ ] The comparison of oxide species in O1s was based on a single fit with constrained Gaussian and Lorentzian widths.

Adsorption Experiments

To study the adsorption capacity of SKL and LNPs, and the relevant MB adsorption kinetics and equilibrium, experiments were performed in the absence of light. MB aqueous solutions were prepared in different concentrations (5, 10, 20, 30, 60, and 100 mg L⁻¹) with the concentration of lignin species (SKL or LNPs) equal to 1.5 mg mL⁻¹.

To study the kinetics of adsorption, the aqueous solutions of MB with SKL or LNPs suspended in it were put under stirring for 2 h. Aliquots of 2 mL were taken out at 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 80, 100, and 120 min. After 120 min, no significant differences in UV‐Vis spectra were observed, which indicates the establishment of adsorption‐desorption equilibrium. The samples were centrifuged immediately at 17698 g for 10 min and the supernatant was removed. The absorbance of each supernatant was measured using a UV‐Vis spectrophotometer (Agilent 8456 in the range of 200–800 nm) in disposable plastic cuvettes. The concentration of MB was calculated using the Beer‐Lambert law:

$${\displaystyle \mathbf{A}=ϵ\mathbf{c}\mathbf{l}}$$ (1)

where A is absorbance at 664 nm (MB absorption maximum), $ϵ$ is molar absorption coefficient of MB at 664 nm obtained from the relevant calibration curve (69800 L mol⁻¹ cm⁻¹), c is concentration (mol L⁻¹), and l is the optical path length (1 cm).

The removal ratio was calculated for each sample using the following equation:

$${\displaystyle \%\mathbf{R}\mathbf{e}\mathbf{m}\mathbf{o}\mathbf{v}\mathbf{a}\mathbf{l}=\frac{{\mathbf{A}}_0-{\mathbf{A}}_{\mathbf{t}}}{{\mathbf{A}}_0}\times 100}$$ (2)

where A₀ is the absorbance at t=0 and A_t is the absorbance after t min of adsorption recorded at 664 nm, respectively.

The adsorption capacity (q_t (mg g⁻¹)) of the adsorbent (SKL or LNPs) was calculated using the following equation:

$${\displaystyle {\mathbf{q}}_{\mathbf{t}}=\frac{({\mathbf{c}}_0-{\mathbf{c}}_{\mathbf{t}})}{\mathbf{m}}\times \mathbf{V}}$$ (3)

where c₀ is the concentration of MB in the supernatant at t=0, and c_t is the concentration of MB in the supernatant at time t (g L⁻¹), V is the total volume of MB solution (mL), and m is the mass of SKL or LNPs (g).

Adsorption Kinetics and Equilibrium Models

The adsorption kinetics data were fitted using pseudo‐first‐order (PFO) and pseudo‐second‐order (PSO) kinetics models. The PFO model (Equation (4)) is based on the assumption that the rate of pollutant uptake linearly decreases with the initial concentration of adsorptive, which is generally true for the initial phase of an adsorption process. The linear form of the model is as follows:

$${\displaystyle \mathbf{l}\mathbf{n}({\mathbf{q}}_{\mathbf{e}}-{\mathbf{q}}_{\mathbf{t}})=\mathbf{l}\mathbf{n}{\mathbf{q}}_{\mathbf{e}}-{\mathbf{k}}_1\mathbf{t}}$$ (4)

where q_e and q_t are the adsorption capacities (mg g⁻¹) of the adsorbent at the equilibrium and at time t, respectively, and k₁ is the pseudo‐first‐order rate constant (min⁻¹). ^[44]

The PSO kinetic model (Equation (5)) assumes that the rate‐limiting step is adsorption, i. e. the interaction between the surface of the adsorbent and the adsorptive. The linear form of the model is as follows:

$${\displaystyle \frac{\mathbf{t}}{{\mathbf{q}}_{\mathbf{t}}}=\frac{1}{{\mathbf{k}}_2{\mathbf{q}}_{\mathbf{e}}^2}+\frac{\mathbf{t}}{{\mathbf{q}}_{\mathbf{e}}}}$$ (5)

where q_e and q_t are the adsorption capacities (mg g⁻¹) of the adsorbent at the equilibrium and at time t, respectively, k₂ is the PSO rate constant (g mg⁻¹ min⁻¹). ^[44]

Langmuir and Freundlich adsorption models were used to evaluate the adsorption mechanism and the maximum adsorption capacity. Langmuir model (Equation (6)) assumes that energetically equivalent sites of the adsorbent interact with the adsorptive creating a monolayer of adsorbates, with no mutual lateral adsorbate interactions.

Langmuir model assumes the following linear form:

$${\displaystyle \frac{{\mathbf{c}}_{\mathbf{e}}}{{\mathbf{q}}_{\mathbf{e}}}=\frac{1}{{\mathbf{K}}_{\mathbf{L}}{\mathbf{q}}_{\mathbf{m}}}+\frac{{\mathbf{c}}_{\mathbf{e}}}{{\mathbf{q}}_{\mathbf{m}}}}$$ (6)

where c_e is the concentration of the adsorbate at equilibrium (mg L⁻¹), q_e is the adsorption capacity at equilibrium (mg g⁻¹), K_L is the Langmuir adsorption equilibrium constant (L g⁻¹), and q_m is the maximum adsorption capacity (mg g⁻¹). ^[65]

Conversely, Freundlich adsorption model (Equation (7)) is based on the assumption that adsorption occurs at energetically heterogeneous sites, with the formation of adsorbate multilayers. The Freundlich isotherm equation in its linearized form is as follows:

$${\displaystyle \mathbf{l}\mathbf{o}\mathbf{g}{\mathbf{q}}_{\mathbf{e}}=\mathbf{l}\mathbf{o}\mathbf{g}{\mathbf{K}}_{\mathbf{F}}+\frac{1}{\mathbf{n}}\mathbf{l}\mathbf{o}\mathbf{g}{\mathbf{c}}_{\mathbf{e}}}$$ (7)

where c_e is the concentration of the adsorbate at equilibrium (mg L⁻¹), q_e is the adsorption capacity at equilibrium (mg g⁻¹), K_F is the Freundlich adsorption equilibrium constant (L^1/n mg^(1−1/n) g⁻¹), and 1/n is the adsorption intensity. ^[65]

Photocatalytic Activity Studies

All samples were irradiated by UV light (365 nm, 80 W m⁻²) for 6 h. Three samples were prepared with the concentration of MB equal to 30 mg L⁻¹ and the concentration of SKL, HLNPs, and SLNPs equal to 1.5 mg mL⁻¹, respectively. The samples were left under stirring in absence of light for 120 min to reach the adsorption‐desorption equilibrium, then each sample was irradiated with UV light for 360 min. Every 30 min an aliquot of 2 mL was withdrawn, centrifuged, and the supernatant was analyzed spectrophotometrically to determine the concentration of MB.

As a first control experiment, a solution of 30 mg L⁻¹ MB in the absence of SKL or LNPs was irradiated with UV light for 360 min to verify the effect of the sole UV irradiation on MB (photolysis). A 2 mL aliquot was taken before irradiation and another one after 360 min, and the relevant MB concentration were assessed.

A second control experiment was conducted to investigate the degradation of MB in the absence of UV irradiation. A solution of 30 mg L⁻¹ MB and 1.5 mg mL⁻¹ HLNPs was left under stirring in dark conditions for 120 min to reach the adsorption‐desorption equilibrium and an aliquot of 2 mL was withdrawn and centrifuged to determine the concentration of MB in the supernatant. Afterwards, the solution was kept under stirring in dark conditions for 360 min and, at this time point, another aliquot of 2 mL was withdrawn and centrifuged to determine MB concentration.

The degree of degradation (%DD) of MB was calculated using the following expression:

$${\displaystyle \%\mathbf{D}\mathbf{D}=\frac{{\mathbf{A}}_0-{\mathbf{A}}_{\mathbf{t}}}{{\mathbf{A}}_0}\times 100}$$ (8)

where A₀ is the absorbance at t=0 and A_t is the absorbance after a defined time of irradiation $t$ , both recorded at the absorption maximum of MB at 664 nm.

Photocatalytic Degradation Kinetics

The photocatalytic degradation kinetics were fitted using pseudo‐first‐order (PFO) model (Equation (9)): 9

$${\displaystyle \mathbf{l}\mathbf{n}\frac{{\mathbf{c}}_{\mathbf{t}}}{{\mathbf{c}}_0}=-{\mathbf{k}}_1\mathbf{t}}$$ (9)

where c_t is the MB concentration after irradiation time t, c₀ is the initial concentration of MB after 120 min stirring in dark conditions, and k₁ is the pseudo‐first‐order rate constant (min⁻¹). ^[66]

Scavenger Experiments

Scavenger experiments were performed on HLNPs to identify the radical species involved in the photodegradation process. Na‐EDTA, AgNO₃, and 2‐propanol were selected as hole, electron, and hydroxyl radical scavengers, respectively. Samples, each containing scavengers at a concentration of 0.5 mol L⁻¹ were left in the dark to reach the adsorption‐desorption equilibrium. Following this equilibrium phase, the samples were subjected to UV light irradiation for a duration of 360 min. Subsequently, 2 mL aliquots were systematically withdrawn at 30‐min intervals, and their absorbance was quantified spectrophotometrically. The obtained absorbance values were transformed into corresponding concentrations through the application of Beer‐Lambert law (Equation (1)).

Recyclability Experiments

The recyclability experiments were performed on HLNPs to determine the adsorption capacity and the degree of degradation (Equation (3) and (8)). The experiments were performed as described in sections Adsorption Experiments and Photocatalytic Activity Studies, for four cycles. After each cycle, the particles were washed with water, centrifuged, and sonicated at 40 % amplitude (160 W) in 0.1 м HCl solution for 10 min. Finally, they were washed once more with water, centrifuged, and dried under vacuum overnight. The dry particles were weighed and suspended in a volume of MB solution to achieve a particle concentration of 1.5 mg mL⁻¹. 2 mL aliquots were withdrawn at 0, 120 (end of adsorption), and 480 min (end of photocatalytic degradation), and their concentration was determined as described in Adsorption Experiments section.

Supporting Information

Supporting Information is available from the Wiley Online Library or from the author.

Conflict of Interests

The authors declare no conflict of interest.

Supporting information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

Rahimihaghighi M., Gigli M., Ficca V. C. A., Placidi E., Sgarzi M., Crestini C., ChemSusChem 2024, 17, e202400841. 10.1002/cssc.202400841

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.