Radical-quinone chain suppression in catechyl lignin via isocyanate modification for permanently whitened and dispersible optical additives

Abstract

Lignin is an attractive bio-based feedstock for optical plastics, yet its use is limited by the intrinsic dark color arising from chromophores, radical-driven quinone chemistry, and aggregation-induced scattering. Here, structurally uniform catechyl lignin (C-lignin) is used to develop a mild, effective isocyanate modification strategy (50 °C, 5 h) that forms urethane linkages without breaking the benzodioxane backbone. The benzyl isocyanate modified C-lignin (BI-CL) achieves high lightness (L* = 92.87) with long-term color stability (>360 days). Quantitative phosphorus-31 nuclear magnetic resonance spectroscopy (³¹P NMR) confirms high hydroxyl consumption (up to 97.3%), while UV-vis and electron paramagnetic resonance (EPR) collectively show suppression of the visible absorption tail and phenoxy radicals. As a low-absorption, dispersible filler, BI-CL enables matrix-dependent optical engineering in low-density polyethylene with high transmittance and preserved thermal processability, providing an ideal route to sustainable optical films and packaging.

---

1 Introduction

Among the rapid growth of sustainable materials and high-performance polymers, lignin has attracted intense interest as a bio-based feedstock owing to its high aromatic content, rich functionality, and broad availability.^1–5 Nevertheless, the practical use of lignin in optical and advanced polymeric materials remains severely constrained by its dark color and the pronounced structural heterogeneity, e.g., traditional guaiacyl/syringyl/p-hydroxyphenyl (G/S/H) lignin as shown in Fig. 1a.^6–8 During extraction and processing, lignin forms chromophoric motifs, such as quinones, quinone methides, and extended conjugated or condensed structures,^9–11 often accompanied by radical chemistry, which together amplify visible-light absorption and scattering (Fig. S1). By contrast, catechyl lignin (C-lignin) consists of caffeyl alcohol units and has a structurally uniform, benzodioxane-rich backbone. It also has a comparatively narrow molecular weight dispersity,^12–15 offering a unique opportunity for molecularly precise chromophore deactivation and interfacial compatibility engineering.¹⁶ C-lignin, typically obtained from castor (Ricinus communis) seed coats with a productivity of 1.85 million t per year,¹⁷ can be a promising feedstock for plastic composite fillers. However, its decolorization and performance remain scarcely explored. As reported so far, most decolorization efforts have centered on conventional G/S/H lignins via acetylation,^18,19 alkylation,²⁰ esterification,¹¹ and grafting strategies.¹² While these modifications can improve lightness and compatibility, they often involve persistent trade-offs. For example, they may require the use of reactive or toxic reagents (e.g., acryloyl chloride and α-bromoisobutyryl bromide),¹⁸ or involve harsh processing conditions (e.g., 100 °C, 24 h)²¹ to achieve a high level of grafting.²⁰ There may also be a compromise between the lightness and tensile strength when the whitening lignin is used as a filler in composites.¹¹

Fig. 1 (a) Representative structures of conventional G/S/H lignin and catechyl lignin (C-lignin). (b) This work: isocyanate-enabled modification of C-lignin for efficient whitening. Right panel: some comparisons on lightness with our work and other papers.^11,14,22,23

Here, we report a mild isocyanate modification strategy for C-lignin whitening (50 °C, 5 h). This approach, demonstrated here for structurally uniform C-lignin and representative aliphatic and aromatic isocyanates, effectively converts lignin hydroxyl groups into urethane linkages while preserving the benzodioxane-rich backbone, transforming intrinsically colored C-lignin into a lighter, whiteness-stable modified product (Fig. 1b). Remarkably, the resulting C-lignin reaches a high lightness (L* = 92.87)^11,14,22,23 and has a long-term color retention period (>360 days). In addition to whitening, urethane grafting enhances the compatibility of polymer matrices, resulting in bio-based composite films with strong mechanical properties and tunable optical responses. Overall, these results show that structurally uniform C-lignin is a promising platform for optical and engineering materials via a mild chemical modification.

2 Experimental section

2.1 Materials and chemicals

The castor shells were sourced from Hebei Province, China. The C-lignin used in the article was extracted after pretreatment with dimethyl isosorbide (DMI)/H₂O under the influence of H₂SO₄. For a detailed process, please refer to the previous article.²⁴ Ethanol was purchased from Shanghai Titan Scientific, and ultrapure water was obtained from Heal Force (SMART-N). Benzyl isocyanate (BI, cas:3173-56-6), heptyl isocyanate (HEI, cas:4747-81-3), phenethyl isocyanate (PEI, cas:1943-82-4), and hexyl isocyanate (HI, cas:2525-62-4) were purchased from Macklin. 2,6-Dimethylphenyl isocyanate (2,6-DMPIC, cas:28556-81-2), and cyclohexyl isocyanate (ICCH, cas:3173-53-3) were from Energy Chemical. 200-mesh polyethylene (PE), high-density polyethylene (HDPE), LDPE, and PCL were obtained from Dongguan, Guangdong Province. The specifications, models, and procurement channels were identical to those in our previous study.^49,50

2.2 C-lignin modified by isocyanate derivatives

Initially, to assess the feasibility of the reaction between C-lignin and isocyanates, we performed a preliminary isocyanate modification experiment using a C-lignin model compound (M) under the conditions of lignin-to-isocyanate = 1 : 1.5 and lignin-to-solvent = 1 : 100 with six isocyanates (BI, HEI, PEI, HI, 2,6-DMPIC, ICCH). The synthesis procedure for the M is detailed in a previous paper.⁵¹ BI was selected as the modifying agent to optimize the reaction conditions. A total of 0.1 g of CL was dissolved in a water/ethanol mixture (v/v = 1 : 1). The effects of various solid-to-liquid ratios (CL : Sol = 1 : 60, 1 : 80, 1 : 100, 1 : 120) and different BI dosages (CL : BI = 1 : 1.5, 1 : 3, 1 : 4.5, 1 : 6) were investigated. Based on the optimized conditions, the effects of different isocyanate derivatives on the modification of CL were further explored. Specifically, 0.1 g of CL was dissolved in 5 mL of water and 5 mL of ethanol (CL : Sol = 1 : 100), followed by the addition of isocyanate derivatives at a molar ratio of 1 : 4.5 (CL : isocyanate), including BI (0.501 mL), HEI (0.617 mL), PEI (0.560 mL), HI (0.590 mL), 2,6-DMPIC (0.564 mL), and ICCH (0.518 mL). The mixtures were stirred at 50 °C for 5 h. The reaction produced a pale-yellow solution with a small amount of insoluble residue. After filtration and ethanol washing, the filtrate was concentrated by rotary evaporation, followed by the addition of deionized water. The precipitation was conducted at 5 °C for 12 h, and the resulting suspensions were filtered again and dried under vacuum for 3 h to yield the modified C-lignin samples, named HEI-CL, BI-CL, PEI-CL, HI-CL, 2,6-DMPIC-CL, and ICCH-CL. We introduce two yield-related parameters herein to evaluate the efficiency of isocyanate modification. BI-CL/BI (g g⁻¹ BI) is the mass ratio of BI-modified C-lignin (BI-CL) to BI.

2.3 Isocyanated lignin characterization

The ¹H NMR, ¹³C NMR, 2D-HSQC NMR spectra were recorded on a Bruker Avance 600 MHz spectrometer, with samples dissolved in dimethyl sulfoxide-d₆ (DMSO-d₆) (30 mg in 0.5 mL). The solvent peak at δ_C/δ_H 39.5/2.49 ppm was used as the internal reference. High-resolution mass spectrometry (HRMS) was performed using a UPLC-TOFMS system equipped with an electrospray ionization source operating in positive mode (ESI⁺). LC-MS was performed on a UPLC–PDA–SQD2 system (Waters) using an ACQUITY UPLC HSS T3 column (1.8 µm, 2.1 × 100 mm). FTIR spectra were obtained using a Nicolet iS50 FTIR spectrometer (ThermoFisher Scientific, China) with a 4000–400 cm⁻¹ wavenumber range. For quantitative ³¹P NMR, 25 mg of C-lignin was derivatized with 2-chloro-4,4,5,5-tetramethyl-1,3,2-dioxaphospholane (TMDP) in a pyridine/CDCl₃ (1.5 : 1, v/v) solution, using chromium acetylacetonate and endo-N-hydroxy-5-norbornene-2,3-dicarboximide (NHND) as the relaxation agent and internal standard, respectively.where η_OH, OH_CL and OH_mod represent hydroxyl removal efficiency, the total hydroxyl contents of pristine C-lignin and modified C-lignin, respectively, as determined by quantitative ³¹P NMR. Measurements were performed with an inverse-gated decoupling pulse sequence and 64 scans.

The CIELAB color space (L*, a*, b*) of modified lignin powders was measured using a CS-5960GX spectrophotometer. L*, a*, and b* represent the lightness of the color, the position between red and green, and the position between yellow and blue, respectively.

ΔE represents the total color difference between two samples in the CIELAB color space. A higher ΔE value indicates a greater overall color difference. ΔL*, Δa*, and Δb* denote the differences in L*, a*, and b* values between the modified sample and the reference sample, respectively.

APC was performed by dissolving the modified lignin (10 mg) in tetrahydrofuran (THF) (ca. 2 mg mL⁻¹) and filtering through a PTFE filter (0.45 µm). The average molecular weight was determined using a Shimadzu LC-20AD equipped with a PLgel 10 µm Mixed-B 7.5 mm I.D. Column and UV detection at 254 nm and 50 °C, using THF as the solvent (1 mL min⁻¹). The average molecular weight was calibrated with polystyrene standards.⁴⁹ SEM images of the modified lignin samples were captured using a ZEISS Sigma 360 field emission scanning electron microscope (Zeiss, Germany) at 2.0 kV. Thermogravimetric analysis (TGA) was carried out using a TGA/DSC 3+/1600 HT system (Mettler-Toledo, Switzerland) with a temperature range from 30 °C to 800 °C and a heating rate of 10 °C min⁻¹ in a nitrogen atmosphere. DSC was performed using a DSC 3+/700 (Mettler-Toledo, Switzerland) instrument, with a range of 30 °C to 200 °C and a heating rate of 10 °C min⁻¹ in a nitrogen atmosphere. The surface wettability of lignin samples before and after modification was evaluated by static water contact angle measurements using a contact angle goniometer (Dataphysics OCA 25, Germany) at room temperature. DLS measurements were conducted to determine the hydrodynamic diameter and PDI of lignin particles before and after modification using a BI-200SM/NanoBrook ZetaPALS/BI-DNDC (UK). Samples were vacuum-dried, packed in 3 mm i.d. quartz tubes (∼3 cm bed height), and irradiated with a 300 W xenon lamp (200–2500 nm). EPR spectra were collected on a pulsed/CW EPR spectrometer (benchtop EPR). Equal masses of dried lignin samples were loaded for comparison. Steady-state PL emission spectra were acquired on a Fluorolog-QM spectrometer (Horiba) using 1 × 1 cm four-window quartz cuvettes. Lignin solutions in THF (0.1 g mL⁻¹) were measured under identical conditions. Excitation wavelengths were selected for each solution; emission was scanned from 300 to 600 nm at 1200 nm min⁻¹. UV-vis absorption spectra were recorded on a UV-2600 spectrophotometer from 200 to 800 nm (1 nm interval, medium scan speed) using lignin solutions in THF (0.1 g mL⁻¹).

2.4 Preparation and characterization of lignin plastic composite films

Four different types of plastics, including 200 mesh PE, HDPE, LDPE, and PCL, were blended with C-lignin at a 95 : 5 (w/w) ratio, with a total blend mass of 1 g. The mixture was hot-pressed at 180 °C and 5 MPa for 15 minutes to form uniform, transparent plastic films. The lignin and LDPE blends were prepared with additional concentrations of 2 wt%, 10 wt%, 15 wt%, 20 wt%, and 25 wt%, and were labeled as LDPE_0.98CL_0.02, LDPE_0.9CL_0.1, LDPE_0.85CL_0.15, LDPE_0.8CL_0.2, and LDPE_0.75CL_0.25, respectively. The TG and DSC were consistent with those described above. Tensile properties were investigated using an electronic universal testing machine (TY-8000A) according to GB/T 1040.3-2006. The grip distance was 50 mm, and the test speed was 100 mm min⁻¹. The transmittance and haziness of the lignin-plastic composite films were measured using an Ultraviolet/Visible/Near-infrared (UV/Vis/NIR) spectrophotometer (UV3600 Plus, Shimadzu), with wavelengths ranging from 400 to 800 nm. The haziness was calculated using the following eqn (3):where T₁ is the incident light transmittance, T₂ is the total light transmitted by the sample, T₃ is the scattered light by the instrument, and T₄ is the light scattered by the instrument and sample.

3 Results and discussion

3.1 Modification on C-lignin by six isocyanates

To investigate the preservation of C-lignin backbone during the isocyanate carbamoylation, we first conducted the reaction using a benzodioxane-containing C-lignin model compound (M) (Fig. 2a). Two-dimensional heteronuclear single quantum coherence nuclear magnetic resonance spectroscopy (2D-HSQC NMR) shows that the characteristic benzodioxane correlations (I_α, I_β, and I_γ) are fully preserved after modification. Meanwhile, new cross-peaks corresponding to the isocyanate R substituents appear, consistent with urethane formation. Using benzyl isocyanate modified model compound (BI–M) as an example, the retained I_α/I_β/I_γ signals together with the emergence of benzyl resonances (peaks 1 and 2) provide direct evidence for carbamoylation (Fig. 2c). To further confirm the structural assignment of the carbamoylated model compound, additional spectroscopic analyses were performed for both the model compound M and the modified product BI–M (Fig. S2). As illustrated in Fig. S2a, the reaction between M and benzyl isocyanate leads to the formation of the urethane-linked product BI–M. The ¹H NMR spectra (Fig. S2b) show the expected aromatic and benzodioxane proton signals for M, while additional resonances appear after modification, which can be attributed to the introduced benzyl–phenyl moiety. Consistently, the ¹³C NMR spectrum of BI–M (Fig. S2c) displays a new signal corresponding to the urethane carbonyl carbon, supporting the formation of the carbamate linkage.

Fig. 2 (a) Carbamoylation of a benzodioxane C-lignin model (M) with six isocyanates: benzyl isocyanate (BI), heptyl isocyanate (HEI), phenethyl isocyanate (PEI), hexyl isocyanate (HI), 2,6-dimethylphenyl isocyanate (2,6-DMPIC), and cyclohexyl isocyanate (ICCH). (b) Corresponding modification of castor-derived C-lignin (CL) at CL : Sol = 1 : 100, CL : BI = 1 : 4. (c) 2D-HSQC NMR of BI–M and BI-CL. (d) FTIR spectra of pristine and modified CL. (e) Quantitative ³¹P NMR of hydroxyl groups before/after modification. (f) APC-derived molecular weight and dispersity. (g) DLS particle-size distributions of pristine and modified CL.

Further confirmation is provided by HRMS analysis (Fig. S2d), which shows a dominant ion peak at m/z 555.2138 corresponding to the [M + H]⁺ ion, matching well with the calculated exact mass of the proposed BI–M structure. The 2D-HSQC spectra (Fig. S2e and f) further demonstrate that the characteristic benzodioxane correlations are retained after modification, while additional cross-peaks associated with the benzyl substituent are observed. Although the chemical shift change of the benzodioxane γ-proton is relatively subtle and partially overlapped in the crowded aliphatic region, the combined ¹H NMR, ¹³C NMR, HRMS, and HSQC data clearly confirm the successful formation of the urethane linkage without disruption of the benzodioxane backbone. The same observations are obtained for the other five isocyanates, including heptyl isocyanate (HEI), phenethyl isocyanate (PEI), hexyl isocyanate (HI), 2,6-dimethylphenyl isocyanate (2,6-DMPIC), and cyclohexyl isocyanate (ICCH) (Fig. S3–S6 and Table S1), which demonstrates the success of the reaction while preserving the C-lignin backbone.

Encouraged by the C-lignin model results, we applied the strategy to native C-lignin isolated from castor seed coats²⁴ and modified it with six above-mentioned isocyanates with different structures (Fig. 2b and Table S2). Since the costs of isocyanates and solvents (ethanol/water, v/v,1 : 1) are not negligible, we optimized the reaction conditions to minimize their use. To ensure the complete dissolution of C-lignin at the beginning of the reaction, the C-lignin-to-solvent ratio (CL : Sol) has to be under 1 : 60. With the reaction ongoing, the unreacted lignin was precipitated and increased with the lower loading of solvent. After the reaction and filtration of unreacted lignin and byproducts, the modified lignin is harvested via precipitation using water (Fig. S7 and S8). Using BI as a representative reagent, we therefore optimized the CL : Sol ratio to be under 1 : 100 to minimize the amount of unreacted lignin and obtain the optimal product yield (Fig. S9a). At this solvent ratio, the BI-CL yield increases with enhanced BI dosage, resulting in a high yield (BI-CL/BI = 30.8 g g⁻¹) and excellent lightness (L* = 92.67) at CL : BI = 1 : 4.5, while marginal gains (yield +1.1% and L* +0.20) are obtained at CL : BI = 1 : 6 (Fig. S9b–d).

The HSQC spectrum of benzyl isocyanate modified C-lignin (BI-CL) again shows an intact benzodioxane framework, along with new signals associated with the incorporation of BI (Fig. 2c). Similar changes are observed for the other isocyanates (Fig. S10). Fourier transform infrared spectroscopy (FTIR) provides further evidence for urethane formation, with the appearance of N–H stretching bands (3320–3260 cm⁻¹), C=O vibrations (1620–1630 cm⁻¹), and NH–CO absorptions (1550–1570 cm⁻¹) across the modified samples (Fig. 2d and S11). Quantitative phosphorus-31 nuclear magnetic resonance spectroscopy (³¹P NMR) reveals substantial consumption of both catecholic and aliphatic hydroxyl groups (Fig. 2e and Table S3, S4), identifying hydroxyl groups as the primary reaction sites.

Notably, advanced polymer chromatography (APC) indicates a decrease in apparent molecular weight after modification, while dispersity (M_w/M_n, PDI) remains low (Fig. 2f and Table S5). This is because of two reasons. First, the initial post-modification filtration step removed high-molecular-weight (3401–4293 g mol⁻¹), recalcitrant lignin residues, yielding derivatives with lower molecular weight (M_w = 1817–2799 g mol⁻¹) and a narrower PDI (e.g., BI-CL = 1.03). Second, given the strong propensity of C-lignin to form intermolecular associations via hydrogen bonding, this change most likely reflects attenuated aggregation upon OH masking and hydrophobic shielding, which reduces hydrogen-bond-driven clustering.^20,25–27 Consistently, dynamic light scattering (DLS) reveals a pronounced reduction in aggregate size compared to unmodified C-lignin (Fig. 2g), supporting improved dispersibility and weakened self-association.^28,29 Overall, these characterization results show that isocyanate modification is effective across diverse R groups, preserves the benzodioxane-rich backbone, and mitigates intermolecular aggregation, providing a structural basis for the enhanced whitening and optical performance.

3.2 The isocyanate structure–property of whitening C-lignin

To understand the relationship between isocyanate structure and reaction efficiency and whitening performance, we compared six reagents with different R groups, varying degrees of hydrophobicity, and different levels of steric hindrance (Fig. 3a). A larger log P (the logarithm of the octanol/water partition coefficient), log P reflects higher relative hydrophobicity, while a larger steric parameter generally indicates greater steric demand of the substituent (Table S6). In general, reaction efficiency decreases as hydrophobicity and steric hindrance increase. BI, with moderate steric hindrance and higher polarity, can react with C-lignin hydroxyls efficiently, resulting in a hydroxyl-consumption of 97.3%. By contrast, the more hydrophobic and sterically demanding 2,6-DMPIC achieves only 84.9% hydroxyl consumption (Fig. 3b). Based on this observation, we propose that the level of contact between isocyanate and C-lignin, and the overall extent of carbamoylation, may be influenced by solvent compatibility and steric accessibility.

Fig. 3 (a) Physicochemical descriptors of the six isocyanates. (b) Hydroxyl consumption during CL modification. (c) Powder color metrics (L* and ΔE vs. pristine CL). (d) Literature benchmarking of lignin whitening performance. (e) UV-vis absorption spectra of pristine and modified CL. (f) Powder photographs and SEM images before and after modification.

Lightness (L*) and the total color difference (ΔE) are two key indicators used to evaluate lignin whitening performance. Pristine C-lignin shows a low L* of 49.31 and a large ΔE of 43.40. After isocyanate modification, the resulting lignin derivatives exhibit substantially increased lightness and reduced ΔE. Under the same conditions (CL : Isocyanate = 1 : 4.5; CL : Sol = 1 : 100), the isocyanate structure had a strong impact on whitening (Fig. 3c), with BI-CL, PEI-CL, and ICCH-CL reaching high lightness (L* = 92.67, 89.24, and 76.97) and low color difference (ΔE = 8.94, 11.33, and 21.93, respectively) while maintaining excellent yields of modified products. In contrast, 2,6-DMPIC-CL showed much lower lightness (L* = 71.68, ΔE = 25.42), due to less efficient carbamoylation as indicated by ³¹P NMR (Fig. 2e). This suggests that there is potential to improve the lightness by enhancing the level of hydroxyl substitution. The C-lignin modified by BI obtains high L* and low ΔE, which is superior to that reported in previous studies (Fig. 3d and Tables S7, S8).8,11,14,22,23,30–42

UV-vis spectroscopy reveals a pronounced suppression of absorption across 200–700 nm after modification (Fig. 3e), quantitatively confirming that the broad visible wavelength range corresponding to the brown coloration is effectively attenuated. Changes in surface morphology correspond to the decolorization of modified lignin. The original lignin contains highly conjugated, unsaturated carbonyl groups that are exposed due to the dense aggregation, leading to its dark color.²⁵ Scanning electron microscope (SEM) images reveal that higher-lightness samples exhibit more compact and smoother surface features (Fig. 3f), indicating that carbamoylation concurrently affects particle aggregation and surface microstructure. In addition to chemical suppression of chromophores, the grafted R groups likely cap residual chromogenic domains, reducing their contribution to visible absorption.^11,14

3.3 The mechanism of C-lignin whitening

To further reveal the mechanism of C-lignin decolorization, we used BI-CL as a representative sample and correlated optical signatures with radical readouts (Fig. 4a–c). Pristine C-lignin shows the expected aromatic π–π* absorption at 200–300 nm, together with a pronounced tail from 300 to 600 nm (Fig. 4a).¹⁰ For lignin, absorption in the 300–400 nm region is commonly linked to oxidation-derived chromophores (e.g., conjugated carbonyl/quinonoid motifs), while the broad 400–600 nm tail is generally associated with quinone/semiquinone-related states and the emergence of larger effective π-conjugated domains,⁴³ often strengthened by aggregation or charge-transfer (CT)-like associated states.^9,44,45 Consistent with the UV spectrum, C-lignin exhibits a clear organic-radical electron paramagnetic resonance (EPR) signal (Fig. 4b) and a stronger, broadened photoluminescence (PL) profile (Fig. 4c). These results are indicative of heterogeneous conjugated emissive sites and CT-like states. After isocyanate modification, BI-CL exhibits (i) suppression of the 300–600 nm absorption tail, (ii) a marked decrease in EPR signal intensity, and (iii) weakened PL emission. Together with the extensive hydroxyl consumption confirmed by ³¹P NMR (Fig. 2e), these measurements demonstrate that carbamoylation substantially alters the chromophore- and radical-related electronic signatures of C-lignin (Table S9). To probe the electronic consequences of hydroxyl capping at the molecular level, UV-vis and EPR measurements (Fig. S12) were performed on the model compound and its benzyl isocyanate derivative. Both compounds display characteristic aromatic π–π* absorption in the 200–300 nm region. Compared to M, BI–M exhibits a slightly reduced absorption tail in the 350–600 nm region, suggesting diminished long-wavelength electronic contributions after urethane formation. Given the limited conjugation and absence of aggregation in a discrete model molecule, only moderate spectral differences are expected. EPR spectra of M and BI–M show weak radical signals near g ≈ 2.003 under ambient conditions. BI–M displays a slightly lower signal intensity than M, consistent with reduced availability of oxidizable hydroxyl sites. Although the magnitude of change is modest, the direction of variation aligns with the lignin-level observations. These results support the chemical feasibility of hydroxyl capping and demonstrate directionally consistent electronic modulation at the molecular level.

Fig. 4 (a) UV-vis (THF, 0.1 g mL⁻¹), (b) EPR (equal sample mass), and (c) PL spectra (THF, 0.1 g mL⁻¹) of CL and BI-CL. (d) Representative chromophores in CL formed during pretreatment. (e) Proposed whitening mechanism upon isocyanate grafting. (f) L* values and photographs of CL and BI-CL after 360 days at room temperature. (g) In situ UV-EPR tracking of radical evolution over 110 min.

Based on these observations, we therefore propose a chromophore-evolution pathway in C-lignin (Fig. 4d), catechol/phenolic motifs undergo one-electron oxidation (O₂/light/heat) to form phenoxy radicals, which interconvert with semiquinone radicals and o-quinones. Quinones contribute directly to visible absorption and can further couple/condense with neighboring aromatics, increasing effective conjugation and amplifying the 300–600 nm tail. Within this framework, isocyanate grafting is likely to suppress this cascade (Fig. 4e). Hydroxyl-to-urethane conversion reduces readily oxidizable electron-donating sites and weakens phenoxy/semiquinone/quinone redox cycling, while hydroxyl capping also disrupts hydrogen-bonding networks and lowers local polarity, disfavoring CT-assisted aggregation.⁴⁶ As a result, BI-CL shows strong suppression of the 300–600 nm tail (Fig. 4a), a pronounced decrease in EPR intensity (Fig. 4b), and markedly weakened PL emission (Fig. 4c). This is consistent with the passivation of chromophoric centers and persistent radicals. Importantly, the whitening is durable. BI-CL maintains an essentially constant L* after 360 days at room temperature (Fig. 4f). In situ EPR further shows time-dependent radical growth for pristine C-lignin under xenon lamp irradiation, whereas BI-CL exhibits negligible radical increase over 110 min of continuous illumination (Fig. 4g). These results indicate that isocyanate modification not only lowers the initial chromophore population but also suppresses light/oxygen-triggered radical formation and quinone-driven amplification that would otherwise deepen color over time. The nonchromophoric, hydrophobic BI substituents provide steric and interfacial shielding that limits O₂/moisture access to susceptible motifs, and this hydrophobic surface modification (the water contact angle of BI-CL = 147°) will restrict moisture and oxygen uptake at particle interfaces, thereby improving the color stability of the material during storage and aging (Fig. S13).

3.4 Optical films with modified C-lignin additives

To test their processability, we screened six isocyanate-modified C-lignins and identified BI-CL as the most thermally robust sample, with a glass transition temperature (T_g) of 172 °C (Fig. S14 and S15). Given its stable whitening and low radical activity, BI-CL was selected as a functional filler and incorporated into commodity plastics. Composite films were prepared by powder blending followed by hot pressing (Fig. 5a), using 5 wt% BI-CL as a low-loading benchmark. At 5 wt%, BI-CL has a negligible impact on matrix melting behavior (Fig. 5b and S16). The melting peaks remain essentially unchanged for poly(ε-caprolactone) (PCL, 56 °C), low-density polyethylene (LDPE, 105 °C), high-density polyethylene (HDPE, 133 °C), and polyethylene (PE, 134 °C), and the composite traces overlap closely with those of the neat polymers. Mechanically, BI-CL reinforces LDPE at low loading (Fig. 5c), and the maximum stress increases from 9.74 MPa (neat LDPE) to 13.67 MPa (LDPECL). Increasing the BI-CL content further leads to a gradual strength rollback, with the 25 wt% film approaching the strength of neat LDPE. This is consistent with aggregation and interfacial stress concentration at high filler levels (Fig. S17).^47,48

Fig. 5 (a) Hot-pressing workflow for film fabrication. (b) DSC thermograms of neat plastics and BI-CL composites. (c) Tensile stress–strain curves of plastics/BI-CL films. (d) Haze spectra (400–800 nm). (e) Visible transmittance spectra (400–800 nm). (f) Transmittance–haze comparison across films. (g) Photographs of neat and composite films on printed text (0 cm) and elevated by 0.8 cm to visualize transparency/haze.

Optical performance is strongly substrate-dependent. In LDPE, BI-CL delivers a rare transmittance-up and haze-down response (Fig. 5d and e), across 400–800 nm, LDPECL shows higher transmittance than neat LDPE, with an 2% increase around 550–600 nm, while haze decreases in the 550–700 nm range (e.g., 50–60% to 45–57%). Photographs confirm improved pattern readability at both 0 and 0.8 cm (Fig. 5g). In contrast, BI-CL increases haze in PE, HDPE, and PCL (Fig. 5e–g), yielding frosted films with haze maxima of 70–95% while maintaining usable transmittance (75–92%). These variations highlight substrate-dependent crystallization and interfacial microstructures that modulate refractive-index inhomogeneity and scattering. Overall, BI-CL enables two application windows on transparency enhancement in LDPE and privacy-oriented haze tuning in PE/HDPE/PCL (Fig. 5f and Table S10).

4 Conclusion

Isocyanate modification consumes C-lignin hydroxyl groups nearly completely, suppressing radical-driven oxidative darkening and blocking the conversion of caffeyl alcohol-derived motifs into quinones and condensed chromophores. Meanwhile, the introduced nonchromophoric hydrophobic moieties physically shield residual chromophoric structures. By integrating chemical transformation with hydrophobic interface engineering, we obtain a C-lignin derivative exhibiting high lightness with long-term whiteness retention (>360 days) and establish a sustainable, long-lived optical additive that is readily implementable in commodity polymers. This strategy could fulfill the gap between the sustainability of lignin and the optical requirements of high-performance plastics.

Author contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Conflicts of interest

The authors declare no competing financial interest.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d6ta00662k.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (22378333, 22408294), Zhejiang Provincial Department of Science and Technology (2023SDXHDX0006). We acknowledge Westlake Education Foundation, the Research Center for Industries of the Future at Westlake University, the foundation of Muyuan Laboratory and thank the Westlake Center for Micro/Nano Fabrication, the Instrumentation and Service Center for Molecular Sciences, and the Instrumentation and Service Center for Physical Sciences (ISCPS), Westlake University.

References

1. A. J. Ragauskas, G. T. Beckham, M. J. Biddy, R. Chandra, F. Chen, M. F. Davis, B. H. Davison, R. A. Dixon, P. Gilna, M. Keller, P. Langan, A. K. Naskar, J. N. Saddler, T. J. Tschaplinski, G. A. Tuskan and C. E. Wyman, Science, 2014, 344, 1246843.
2. M. Dike and S. K. Dishari, Nat. Chem., 2025, 17, 974.
3. N. Li, K. Yan, T. Rukkijakan, J. Liang, Y. Liu, Z. Wang, H. Nie, S. Muangmeesri, G. Castiella-Ona, X. Pan, Q. Zhou, G. Jiang, G. Zhou, J. Ralph, J. S. M. Samec and F. Wang, Nature, 2024, 630, 381–386.
4. X. Pan, X. Li, R. Wang, Z. Wang, Y. Ni and Q. Wang, Nano Lett., 2025, 25, 4526–4531.
5. H. Du, K. Liu, T. Xu, C. Xu, M. Lin, Z. Fang, S.-W. Kim, J.-Y. Seo, J. Chen, H. Ma, B. S. Hsiao, L. W. DeVetter, Z. Piao, C. Si, C. Chen, Q. Yang, S.-Y. Lee, Y. Yao and X. Pan, Chem. Rev., 2025, 125, 11666–11814.
6. H. Zhang, S. Fu and Y. Chen, Int. J. Biol. Macromol., 2020, 147, 607–615.
7. Z. Zhang, J. Xu, Z. Zhou, A. Li, S. Zhu, J. Li, W. Zhang, F. Zhang and K. Chen, Int. J. Biol. Macromol., 2025, 290, 138857.
8. Y. Wu, X. Chen, S. Ni, Y. Zhang, Y. Fu and M. Qin, Int. J. Biol. Macromol., 2025, 309, 142840.
9. T. Tribulová, F. Kacik, D. Evtuguin and I. Cabalová, Cellul. Chem. Technol., 2016, 50, 659–667.
10. Y. Zhang and M. Naebe, ACS Sustain. Chem. Eng., 2021, 9, 1427–1442.
11. K. Shikinaka and Y. Otsuka, Green Chem., 2022, 24, 3243–3249.
12. Y. Li, X. Meng, R. Meng, T. Cai, Y. Pu, Z.-M. Zhao and A. J. Ragauskas, RSC Adv., 2023, 13, 12750–12759.
13. S. Wang, Q. Shen, S. Su, J. Lin and G. Song, Trends Chem., 2022, 4, 948–961.
14. S. Su, Q. Shen, S. Wang and G. Song, Int. J. Biol. Macromol., 2023, 239, 124256.
15. S. Su, C. Wu, D. Zhang, P. Wang and L. Wang, Chem. Eng. J., 2025, 509, 161217.
16. T. Gui, L.-P. Xiao, S.-L. Zou, Y. Zhang, X. Fu, C.-H. Liu and R.-C. Sun, Chem. Eng. J., 2024, 491, 151748.
17. S. Wang, K. Zhang, H. Li, L.-P. Xiao and G. Song, Nat. Commun., 2021, 12, 416.
18. O. Gordobil, R. Delucis, I. Egüés and J. Labidi, Ind. Crops Prod., 2015, 72, 46–53.
19. H. Zhang, X. Liu, S. Fu and Y. Chen, Int. J. Biol. Macromol., 2019, 133, 86–92.
20. X. Jiang, Z. Tian, X. Ji, H. Ma, G. Yang, M. He, L. Dai, T. Xu and C. Si, Int. J. Biol. Macromol., 2022, 201, 400–410.
21. Y. Wu, Y. Qian, A. Zhang, H. Lou, D. Yang and X. Qiu, Ind. Eng. Chem. Res., 2020, 59, 17057–17068.
22. K. Shikinaka, A. Suzuki and Y. Otsuka, ChemistrySelect, 2024, 9, e202400046.
23. S. Gao, X. Chen, G. Tian, Y. Fu, M. Qin and Z. Wang, Int. J. Biol. Macromol., 2023, 253, 127107.
24. C. Wu, S. Su and L. Wang, Sep. Purif. Technol., 2025, 374, 133580.
25. Y. Li, S. Zhao, Y. Li, A. J. Ragauskas, X. Song and K. Li, Int. J. Biol. Macromol., 2022, 223, 1287–1296.
26. N. Pajer, C. Cestari, D. S. Argyropoulos and C. Crestini, npj Mater. Sustainability, 2024, 2, 31.
27. T. Xiang, L. Chen, X. Qiu, D. Yang and J. Wang, Chem. Eng. J., 2024, 497, 154356.
28. H. Zhang, S. Hu, H. Yang, F. Lu and F. Yue, Biomacromolecules, 2025, 26, 4838–4847.
29. J. H. Lee, T. M. Kim, I.-G. Choi and J. W. Choi, Nanomaterials, 2021, 11.
30. J. Zheng, L. Chen, X. Qiu, Y. Liu and Y. Qin, Bioresour. Technol., 2023, 385, 129458.
31. Z. Pan, Y. Li, Z. Zhang, F. Xu, S. Ramaswamy, A. Abdulkhani and X. Zhang, Ind. Crops Prod., 2022, 186, 115173.
32. X. Su, Y. Fu, Z. Shao, M. Qin, X. Li and F. Zhang, Ind. Crops Prod., 2022, 176, 114359.
33. H. Zhang, F. Chen, X. Liu and S. Fu, ACS Sustain. Chem. Eng., 2018, 6, 12532–12540.
34. S. C. Lee, T. M. T. Tran, J. W. Choi and K. Won, Int. J. Biol. Macromol., 2019, 122, 549–554.
35. S. C. Lee, E. Yoo, S. H. Lee and K. Won, Polymers, 2020, 12, 699.
36. H. Zhang, X. Liu, S. Fu and Y. Chen, Ind. Eng. Chem. Res., 2019, 58, 13858–13867.
37. Z. Li, L. Chen and X. Qiu, Ind. Crops Prod., 2024, 208, 117857.
38. C. Lu, J. Xu, J. Xie, S. Zhu, B. Wang, J. Li, F. Zhang and K. Chen, Int. J. Biol. Macromol., 2022, 222, 2512–2522.
39. S. Hua, S. Fu, B. Zhang, H. Liu and K. Xu, Int. J. Biol. Macromol., 2025, 320, 145822.
40. J. Zhang, Z. Tian, X.-X. Ji and F. Zhang, Int. J. Biol. Macromol., 2023, 231, 123244.
41. S. Wang and M.-F. Li, Int. J. Biol. Macromol., 2024, 282, 136723.
42. H. Kong, X. Chen, S. Ni, Y. Zhang, M. Qin, Y. Fu and C. Si, Ind. Crops Prod., 2025, 229, 120965.
43. Y. Fu, Y. Qian, A. Zhang, H. Lou, X. Ouyang, D. Yang and X. Qiu, ACS Sustain. Chem. Eng., 2022, 10, 5421–5429.
44. G. S. Furman and W. F. W. Lonsky, J. Wood Chem. Technol., 1988, 8, 191–208.
45. S. Barsberg, T. Elder and C. Felby, Chem. Mater., 2003, 15, 649–655.
46. K. H. Kim, T. Dutta, E. D. Walter, N. G. Isern, J. R. Cort, B. A. Simmons and S. Singh, ACS Sustain. Chem. Eng., 2017, 5, 3913–3919.
47. R. R. N. Sailaja and M. V. Deepthi, Mater. Des., 2010, 31, 4369–4379.
48. S.-H. Hong and S.-H. Hwang, Mater. Today Commun., 2021, 28, 102696.
49. S. Yang, Y. Wang, X. Kang, X. Yang, C. Zhang, L. Deng, X. Meng, X. Song and L. Wang, Ind. Crops Prod., 2023, 203, 117169.
50. S. Yang, X. Yang, X. Meng and L. Wang, Green Chem., 2022, 24, 4082–4094.
51. S. Su, F.-s. Cao, S. Wang, Q. Shen, G. Luo, Q. Lu and G. Song, Green Chem., 2023, 25, 8172–8180.

---