Effect of Molecular Organization on the Properties of Fractionated Lignin-Based Thiol–Ene Thermoset Materials

In this study, the combination of sequential solvent fractionation of technical Kraft lignin was followed by allylation of most OH functionalities to give highly functional thermoset resins. All lignin fractions were highly functionalized on the phenolic (≥95%) and carboxylic acid OH (≥85%) and to a significant extent on the aliphatic OH moieties (between 43 and 75%). The resins were subsequently cross-linked using thiol–ene chemistry. The high amount of allyl functionalities resulted in a high cross-link density. Dynamic mechanical analysis measurements showed that the thioether content, directly related to the allyl content, strongly affects the performance of these thermosets with a glass transition temperature (Tg) between 81 and 95 °C and with a storage modulus between 1.9 and 3.8 GPa for all thermosets. The lignin fractions and lignin-based thermosets’ morphology, at the nanoscale, was studied by wide-angle X-ray scattering measurements. Two π–π stacking interactions were observed: sandwich (≈4.1–4.7 Å) and T-shaped (≈5.5–7.2 Å). The introduction of allyl functionalities weakens the T-shaped π–π stacking interactions. A new signal corresponding to a distance of ≈3.5 Å was observed in lignin-based thermosets, which was attributed to a thioether organized structure. At the same time, a lignin superstructure was observed with a distance/size corresponding to 7.9–17.5 Å in all samples.

Initial

Procedures Initial Purification of Technical Lignin Powder
The washing step was performed in order to remove various impurities such as inorganics. Approximately 35 g of lignin was dispersed in 700 mL of deionized water and magnetically stirred at 60 °C for 2 h. Subsequently, the lignin was filtered using a sintered funnel (pore size 4). The procedure was repeated until the pH of the filtrate was above 5.5 (2-3 times). The washed lignin (SW-Initial) was dried in a vacuum oven at 50 °C overnight. The recovery yield was calculated to be 93 ± 2%. The washing step was performed ten times.
Lignin Extraction Using Sequential Solvent Fractionation Approach 15 ± 2 g of washed lignin was mixed with 150 ± 20 mL of EtOAc and magnetically stirred at room temperature for 2 h. The insoluble lignin was separated from the mixture by filtration through a filter paper (grade 3 Munktell) and subsequently dried in a vacuum oven at 50 °C overnight. The EtOAc solution was then rotary evaporated to recover the EtOAc soluble fraction. The EtOAc soluble lignin was then dried in a vacuum oven at 50 °C overnight and subsequently re-dissolved in acetone, precipitated in 200 mL of deionized water and finally freeze-dried. A fine, brown powder was obtained as the final product. The same procedure was then applied to the insoluble fraction using an extraction sequence of firstly EtOH, secondly MeOH, and finally acetone as extracting solvents. The lignin to solvent ratio was 1:10 (g/mL) in all cases. Five different fractions were obtained and named accordingly: SW-EtOAc, SW-EtOH, SW-MeOH, SW-Acetone, and SW-Insoluble. The last fraction was denoted as SW-Insoluble because it was not soluble in the four previous mentioned solvents (EtOAc, EtOH, MeOH, and acetone), however it is soluble in DMSO, which enables its characterization. 1, 2 The extraction step was performed four times. The fractionation yields are reported in Table S2.

Lignin Allylation
The initial lignin (SW-Initial) and all five fractions (1 g of each, ≈6.5 mmol of the different OH groups/g lignin) were respectively placed in different 10 mL pressure vials. Diallyl carbonate (3 eq. per hydroxyl group, ≈2.8 mg) and tetrabutylammonium bromide (1 eq. per hydroxyl group, 2.1 mg) were added and the pressure vials were then sealed. For each lignin fraction, the exact content of all different OH groups was calculated, and the DAC and TBAB content was adjusted accordingly. The mixture was magnetically stirred at 120 °C for 5 h. The reaction mixture was then left to cool to room temperature and then dissolved in 50 mL of EtOAc. The organic phase was extracted with deionized water to remove TBAB. The catalyst was then recovered from the aqueous mixture by rotary evaporating water and finally dried in a vacuum oven to give a pale-yellow solid (with a yield of 85%). The organic phase was dried over MgSO 4 , filtered, and then concentrated to a total volume of approximately 5 mL using a rotary evaporator. The concentrated solution was precipitated in 200 mL ice-cold cyclohexane under vigorous magnetic stirring. Chemically modified lignin was filtered by using filter paper (grade 3, Munktell) and washed several times with cyclohexane. Allylated lignin was then dried in a vacuum oven at 50 °C overnight to give a brightyellow powder. 3 The samples were denoted DAC-SW-Initial, DAC-SW-Solvent (solvent = EtOAc, EtOH, MeOH, Acetone), and DAC-SW-Insoluble. The lignin allylation was performed at least two times for each fraction.

Decarboxylation of Allylated Lignin Fractions
Approximately 100 mg of allylated lignin (with ≈6.5 mmol of the different OH groups before allylation/g lignin) was dissolved in 1 mL of tetrahydrofuran (THF) in a vial. For each lignin fraction, the exact content of all different OH groups before allylation was considered, and the LiOH content was adjusted accordingly (for 1 mmol of the different OH it was added approximatively 100 mg of LiOH). Lithium hydroxide (65.0 mg, 2.7 mmol) was dissolved in 1 mL of H 2 O, mixed with the dissolved lignin, and then stirred at room temperature for 1 week. The product was then precipitated by adding 1 M HCl (100 mL) and filtered. The product was then washed with 1 M HCl (50 mL), water (100 mL), and dried in a vacuum oven at 50 °C overnight. 3 The decarboxylated lignin was obtained as a yellow powder in yields of 67 ± 14 wt%. The obtained samples were denoted D-DAC-SW-Initial, D-DAC-SW-Solvent (solvent = EtOAc, EtOH, MeOH, Acetone), and D-DAC-SW-Insoluble.

Preparation of Lignin-Based Thermosets
For each free-standing sample, (40 ± 5) mg of allylated lignin was mixed with the thiol cross-linker in a vial. The equivalent ratio between the reactive groups (ene:thiol) was kept at 1:1. 150 mL of EtOAc were then added, to ensure a better mixing. The obtained mixtures were poured into soft silicone molds and left in the fume hood for several hours in order to enable the solvent to evaporate and form a smooth film. The molds were then transferred into an oven and cured at 125 °C for 33 h (the resin prepared from DAC-SW-EtOAc was cured at 125 °C for 50 h). The obtained thermosets were removed from the mold directly after they were taken out of the oven. 4 All formulations and thermosets were denoted in the following way: T3-DAC-SW-Initial, T3-DAC-SW-Solvent (solvent = EtOAc, EtOH, MeOH, Acetone), T3-DAC-SW-Insoluble. At least five samples were prepared from each allylated lignin fraction.

Characterization Techniques Size Exclusion Chromatography (SEC)
Size exclusion chromatography was used to determine the lignin number-average molar mass (M n ), weight-average molar mass (M w ), and dispersity (Đ). Prior to measurement, all samples were completely dissolved in the elution solvent (5 mg/mL) and filtered through a syringe filter of 0.45 μm pore size. The elution solvent was S7 DMSO with 0.5% (w/w) lithium bromide (LiBr) at a flow rate of 0.5 mL/min. The measurements were done by using SECcurity 1260 infinity GPC System equipped with a refractive index (RI) detector, a PSS GRAM precolumn, two PSS GRAM separation columns (particle size 10 µm and pore size 100 and 10000 Å, with dimensions 300 mm × 8 mm), at 60 °C. Pullulan standards were used as an internal standard with a molar mass range of 342-708000 g/mol. The SEC data should thus be considered as trends, rather than absolute values because of the lack of calibration standards with high structural similarity to technical lignin molecules. All data represent an average of two to six measurements.

Proton and Phosphorus Nuclear Magnetic Resonance Spectroscopy ( 1 H and 31 P NMR)
All spectra were recorded on a Bruker Avance III HD 400 MHz instrument with a BBFO probe equipment with a Z-gradient coil and processed as described previously. 2 Integration regions used for lignin in 31 P NMR spectrum were: aliphatic OH (149.5-145.5 ppm), C 5 -substituted OH (144.7-140.1 ppm), guaiacyl OH (140.1-138.8 ppm), p-hydroxyphenyl OH (138.8-137.0 ppm), carboxylic acid OH (136.0-133.6 ppm). 5 All individual data represent an average of two to four measurements.

Fourier Transform Infrared Spectroscopy (FTIR) and Real Time FTIR (RT-FTIR)
PerkinElmer Spectrum 100 instrument was used to analyze the chemical composition of the reagents and products. Attenuated total reflection (ATR) mode with a diamond crystal was used. The instrument was equipped with a temperature control unit (Specac, Heated Golden Gate Controller). All FTIR spectra were acquired at room temperature. For each sample, 16 scans were done, with a resolution of 4 cm -1 between 4000 and 600 cm -1 . For RT-FTIR, a new scan was acquired every minute with a resolution of 4 cm -1 . Thiol conversion was calculated by monitoring the absorption signal of S-H bond at 2607-2533 cm -1 . RT-FTIR studies were performed at 125 °C. FTIR spectra were analyzed with Spectrum software and RT-FTIR spectra with TimeBase ® software from Perkin-Elmer. FTIR spectra were baseline corrected and normalized to the 1509 cm -1 signal corresponding to the aromatic C=C stretching. 6

Differential Scanning Calorimetry (DSC)
Mettler Toledo DSC1 was used to analyze the thermal transitions of lignin samples. 10 mg of each sample were placed in a 100 μL aluminum crucible and covered with a pierced lid. The samples were then heated to 105 °C and held for 20 min, then cooled down to 20 °C and held for 10 min and finally heated up to 200 °C. The heating/cooling rate was 10 °C/min. The second heating midpoint was used to determine the glass transition temperature (T g ). The DAC-SW-EtOAc sample was cooled to -80 °C. All individual data represent an average of three to six measurements.

Thermogravimetric Analysis (TGA)
Mettler Toledo TGA/DSC1 instrument was used to study the thermal behavior of the samples. 10 mg of each sample were placed in a ceramic crucible for the analysis. For the first 10 minutes, samples were kept at 30 °C, afterwards they were heated up S8 to 800 °C, with a rate of 5 °C/min (in a nitrogen atmosphere, flow rate of 50 mL/min) and then kept at 800 °C for 10 min.

Dynamic Mechanical Analysis (DMA)
DMA Q800 instrument was used to investigate the viscoelastic properties of the thermosets in tensile mode. Samples were cooled to -100 °C and held for 5 min (using liquid nitrogen). The samples were then heated to 250 °C with a heating rate of 3 °C/min. The preload force of 0.01 N (force track 125%), with a strain of 0.1% at 1 Hz was used to measure the free sanding samples (12 mm × 5.2 mm × 0.15 mm). Storage modulus (E′), loss modulus (E′′), and tan δ were investigated. The maximum of tan δ was used to calculate the T g of the thermosets. All individual data represent an average of two to three measurements.

Wide-Angle X-ray Scattering (WAXS)
Using wide-angle X-ray scattering the molecular structure of materials can be studied. All X-ray experiments were performed at the beamline P03 at PETRA III (Hamburg, Germany). The details on the experimental set-up and procedure for data reduction as well as analysis can be found on page SI20. The powder samples (SW-Lignin, DAC-SW-Lignin), and the films (T3-DAC-SW-Lignin) were mounted in custom-made 3D-printed sample holders. The pellet pressing die and the hydraulic laboratory press were used to create pellets of lignin powder samples, with a diameter of 13 mm and a thickness of 1 mm. The geometry of the lignin based-thermosets was 12 mm × 5.2 mm × 0.15 mm.

Uniaxial Tensile Testing
Instron 5944 instrument (equipped with a 500 N cell) was used to determine the tensile properties of the lignin-based thermosets. The tensile grip distance was 15 mm and the crosshead speed was set to 5 mm/min. Rectangular samples (with approximate dimensions of 15 mm × 5.2 mm × 0.15 mm) were conditioned for 1 week in an environment controlled room where also the measurements were made (temperature of 22 °C and relative humidity of 50 ± 2%). All individual data represent an average of three to nine measurements. Figure S1. DMSO SEC overlay of SW-Lignin fractions. Pullulan standards were used as an internal standard.        Figure S5. FTIR spectra of SW-EtOH before and after allylation. Figure S6. FTIR spectra of SW-Lignin fractions before allylation (up) and after allylation (bottom). The region between 2650-1900 cm -1 was omitted due to absorbance of diamond ATR crystal in this region. S14 Figure S7. FTIR spectra of DAC, with a zoom of the region 1800-600 cm −1 .  * Decarboxylation reaction was performed in order to determine if hydroxyl groups were etherified or carboxyallylated. From previous studies it was shown that the phenolic OH are etherified and that carboxylic acid OH are most likely to be converted to the corresponding allyl esters. 3, 7 For a better visualization, it was assumed that all aliphatic OH were fully functionalized (carboxyallylated or etherified). Table S6. The amount of aliphatic OH before allylation (representing 100%, first column) and the total amount of allylated aliphatic OH (second column). The third and fourth columns show the amount/percentage of etherified and carboxyallylated aliphatic OH based on the amount of aliphatic OH before allylation. All data were determined by 31 P NMR.

Morphological Characterization Using X-ray Scattering
All X-ray experiments were conducted at beamline P03 of PETRA III (DESY, Hamburg, Germany) [8][9][10] . During the experiments, a fixed X-ray wavelength of 1.048 Å (E = 11.83 keV) and a fixed sample-detector-distance (SDD) of 21 cm was used. The beam was focused on the sample position with a size of 25 × 25 μm 2 . The detector was a LAMBDA 9M unit (X-spectrum, Hamburg, Germany) with a pixel size of 55 μm.
To prevent beam damage on the samples (due to the extremely high photon flux at a 3 rd generation synchrotron facility like PETRA III), they were scanned at 100 individual positions over a region of 2 × 2 mm 2 with an illumination of 1 s each. For the subsequent data processing, the images which showed the biggest similarity to each other, were summed up ( Figure S18). To further analyze the scattering patterns, the 2D detector images were reduced to 1D scattering profiles ( Figure S19-21). The data reduction procedure followed the well-known numerical recipes for azimuthal integration and normalization from literature. 11 The resulting scattering profiles show the radially averaged intensities over the so-called scattering vector q (q = (4π/λ)sinθ ), where is the X-ray wavelength and is the scattering angle, measured from the point where the primary beam hits the detector. After data reduction, the scattering background arising from air scattering was subtracted from each scattering profile. For modelling, the scattering profiles were normalized and a simple combination of Gaussian functions and a constant background was used. From their respective peak positions, the so-called D-spacing (D = 2π/q) was calculated, which gives indications about the real-space distances and sizes associated with the visible features in the scattering profiles. Figure S18. Summed WAXS pattern for one lignin sample. The white areas mark detector gaps or missing modules. The beam stop is visible as a black area shielding the detector from the primary X-ray beam during the experiments. For data processing, the beamstop, detector gaps, and faulty pixel areas were masked.

S24
Scheme S3. Schematic description of proposed changes in scattering distances, for lignin fractions after allylation. This representation is simplified to a one aromatic ring were L represents all possible lignin structers.