Solvent effect on HZSM-5 catalyzed solvolytic depolymerization of industrial waste lignin to phenols: superiority of the water–methanol system over methanol

Abstract

Kraft lignin from industrial black liquor was subjected to one-pot solvolytic depolymerization and hydrodeoxygenation over zeolite HZSM-5 as catalyst and NaOH as co-catalyst at 220 °C. The effect of methanol and a water–methanol (1 : 1) mixture as solvent on the lignin depolymerization products was studied. The products were characterized by infrared spectroscopy and numerical indices based on IR spectra of the product mixture were defined to study the functional group transformations occurring in the reaction. Compared to NaOH in pure methanol, NaOH in water–methanol was found to be efficient in suppressing char formation and enhancing product quality and quantity. Alkyl substituted phenols were found to be the major product (14.1 wt%). Other than phenols, formation of long chain aliphatic compounds was also observed.

---

Introduction

Lignocellulosic biomass is potentially the most suitable feedstock for the upcoming large-scale industrial biorefineries. This is due to the optimum availability of raw materials (e.g. straw, grass, wood, paper waste, etc.), novel greener delignification techniques^1a and renewability as compared to petroleum based resources. Another important factor in utilization of lignocellulose as feedstock is its lower cost. The lignocellulose materials consist of three primary chemical fractions or precursors: (1) hemicellulose/polyoses – a sugar polymer predominantly having pentoses; (2) cellulose – a glucose polymer; and (3) lignin – a polymer of phenols.^1b Lignin constitutes 15–30% by weight of the lignocellulosic biomass and 40% by energy. It is a complex amorphous polymer of phenyl propane type of subunits linked by ether (C–O–C) and C–C linkages. The complex non-uniform aromatic nature of lignin makes it difficult to be isolated from lignocellulose without modification and also to convert into useful products.^2,3 It is waste by-product of pulp and paper industries and is burnt in the furnace of chemical recovery boilers. Lignin shows significant challenges in conversion to low-molecular weight products. The refractory behaviour of lignin towards depolymerization is mainly due to the repolymerization reactions of the phenolic subunits.⁴ These repolymerization reactions lead to the formation of char.

Pyrolysis reaction at high temperature is one of the most studied processes for lignin degradation to low-molecular weight products.^5,6 However, these processes are energy demanding to high operational temperatures (400–800 °C) and also produce high amount of char (up to 40% by weight of lignin). Depolymerization of lignin in different solvents mainly alcohols show better alternative to reduce the char formation besides lowering the operational temperature. Formic acid as hydrogen donating solvent with ethanol as co-solvent has been used for lignin depolymerization at 380 °C and subsequent hydrodeoxygenation to alkyl phenols and C₈–C₁₀ aliphatics with reduced char formation (5% by weight of lignin).^7,8 Supercritical ethanol has been utilized to overcome the low lignin conversion and char formation in hydroprocessing of lignin in hot compressed water (300 °C). Hydroprocessing in supercritical ethanol (260 °C) using 5%Ru/γ-Al₂O₃ as catalyst has shown a lignin conversion of ca. 98% and yield of liquid products up to 92%. Liquid products were mainly composed of substituted phenols, such as guaiacols and syringols.⁹ Depolymerization of lignin has also been studied in water^10,11 and ethanol–water mixture¹² as a solvent at 400 °C using phenol/o-cresol as capping agent to prevent repolymerisation reactions within lignin. The results show formation of substituted alkyl phenols as products and lowering of average molecular weight of lignin to 450–475 g mol⁻¹ with no charring. However, the polydispersity index (>2) shows broader range of molecular distribution in products indicating the presence of considerable amount of oligomers in product mixture. Methanol has been used as solvent and source of reductive equivalents of hydrogen in one-pot lignin depolymerization and subsequent hydrodeoxygenation/hydrogenation at 300 °C over Cu-doped PMO (porous metal oxide) as catalyst. The results show conversion of lignin to monomeric substituted cyclohexyl derivatives with greatly reduced oxygen content and negligible aromatics without char formation.^13–15 The above results indicate that the solvent based lignin depolymerization strategies provide greater opportunity to curtail the challenges faced in lignin conversion processes.

Porous metal oxides and zeolites have been used in various hydrodeoxygenation reactions. Zeolite HZSM-5 is a high silica zeolite usually used in industrial petro-cracking reactions. HZSM-5 has been used as a hydrodeoxygenation catalyst in pyrolysis of lignin¹⁶ and biomass.¹⁷ Depolymerization of Alcell lignin is solubilized in acetone at 500–600 °C using HZSM-5 catalyst has produced high yields of gasoline range hydrocarbons such as benzene, toluene and xylene BTX. These high octane numbers hydrocarbons are normally produced by the catalytic reforming of naphthalenes.¹⁸ Lignin model compound have also been studied for hydrodeoxygenation over HZSM-5 and hydrogenation over Pd/C in one-pot to successfully produce cycloalkanes at 200 °C and 50 atm H₂ (ref. 19).

In this work, we have used methanol and water–methanol mixture as solvent for one-pot depolymerization–hydrodeoxygenation of lignin in presence of HZSM-5. The work aims to study the effect of methanol and its combination with water on the extent of lignin depolymerisation, preventing char formation and analyzing the product mixture for the extent of oxygen removal.

Experimental

Chemicals

Silica sol (LUDOX HS-30, 30 wt% suspension in H₂O, Aldrich), tetrapropyl-ammonium bromide (TPABr, 98%, Aldrich), ammonium sulphate ((NH₄)₂SO₄, BioXtra, ≥99.0%, Sigma-Aldrich), alumina (Al₂O₃, activated, neutral, Brockmann Activity I, Fluka), ethyl acetate (Lichrosolv, Merck), tetrahydrofuran (Lichrosolv, Merck), methanol (Emparta, Merck), NaOH (≥97%, Merck) and H₂SO₄ (97%, Merck) were used as received. Kraft lignin was isolated from industrial black liquor (BILT paper industry, India). Black liquor was acidified with 50% H₂SO₄ with vigorous stirring, precipitated lignin was washed thoroughly with distilled water and dried in oven at 80 °C for 1 h. Lignin was further purified using 1,4-dioxane. Purification with 1,4-dioxane ensures that lignin is free of carbohydrate impurities.²⁰ Solvent was recovered and pure lignin was used in the experiments. Proximate analysis of lignin showed 3.96 wt% moisture, 9.60 wt% ash, 45.10 wt% volatile matter and 42.8 wt% fixed carbon. Elemental analysis showed 59.3 wt% carbon, 7.1 wt% hydrogen, 0.1 wt% nitrogen, 0.24 wt% sulphur and 33.26 wt% oxygen (by difference). The weight average molecular weight (M_w) of kraft lignin²¹ is reported to be ∼5000 g mol⁻¹ with polydispersity index 5.9.

Catalyst synthesis

Zeolite HZSM-5 was synthesized according to the procedure previously described.²² The synthesis was done at 95 °C using silica sol as silica source, Al₂O₃ as aluminium source and TPABr as template. The zeolite was initially calcined at the rate of 4 °C min⁻¹ up to 100 °C, further it was increased to 10 °C min⁻¹ to 500 °C and maintained at 500 °C for 3 hours to decompose organic template. The calcined product was subjected to ammonium exchange with 1 M (NH₄)₂SO₄ solution (10 mL g⁻¹) for 10–15 minutes at ambient temperature. It was washed, dried and again calcined at 500 °C for 3 hours to give HZSM-5.

Catalyst characterization

The XRD patterns of samples were recorded on a PANalytical X'Pert PRO through a 2θ range from 5° to 60° (step size 0.017°, step time 20 s) using CuKα radiation (1.54 Å) at 40 kV and 100 mA. Scherrer's equation was used to calculate the crystallite size (1):where Γ is the mean crystallite size of the ordered (crystalline) domains, which may be smaller or equal to the grain size; K is a dimensionless shape factor, the shape factor has a typical value of about 0.89; λ is the X-ray wavelength; β is the line broadening at half the maximum intensity (FWHM), after subtracting the instrumental line broadening, in radians and θ is the Bragg angle. Lattice parameters were calculated with X'Pert High Score software. The surface morphology of the catalyst was studied by scanning electron microscopy on a FEI Quanta 450 FEG apparatus equipped with energy dispersive spectrometer (EDAX). Surface area, pore volume and average pore size were characterized and determined by N₂ adsorption–desorption (Micromeritics ASAP 2010, Norcross, GA, USA).

Catalytic reactions

Lignin (0.5 g), solvent (20 mL) and catalyst (0.1 g) were placed in a 25 mL stainless steel SS316 reactor. The reactor was sealed after purging it with argon 3–5 times to expel air. The reaction was performed at 220 °C without stirring. When the desired reaction time was reached, the reactor was quenched in water. The products from the reactor were washed with 20 mL water and centrifuged at 2500 rpm for 15 minutes to separate water soluble and water insoluble products. The solvent based separation of products was done with ethyl acetate (EtOAc) and tetrahydrofuran (THF). Initial lignin is soluble in THF and insoluble in EtOAc and water (as shown in Fig. S1, ESI†). The depolymerization products are expected to be soluble in either water or EtOAc. The unconverted lignin and high molecular weight fragments of lignin were separated as THF soluble products in product separation process as shown in Scheme 1.

Scheme 1 Product separation procedure.

The yields of the products were measured by weight after solvent evaporation in vacuum rotary evaporator. The yields were defined as shown in eqn (2)–(5):

G(wt%) = 100 − (W + E + T + C) (7)

where W = water soluble products, E = EtOAc soluble products, T = THF soluble products, C = char, G = gaseous products.

Product characterization

The FTIR spectra of various extractives after removal of respective solvent were recorded on Shimadzu IRAffinity-1 with DLATGS (deuterated, L-alanine doped triglycine sulfate) detector and DRS (diffuse reflectance spectroscopy) accessory. The main measurement features were a spectral range from 4000 to 400 cm⁻¹, 45 scans, and a resolution of 4 cm⁻¹.

The water soluble and EtOAc soluble products were qualitatively and quantitatively analyzed on a GC-MS-FID instrument (Thermo Scientific, Trace-1310/Thermo ISQ MS detector) equipped with an TRACE TR-5MS capillary column (30 m × 0.25 mm × 0.1 μm) and helium as a carrier gas. The column was initially kept at 50 °C for 2 min, then was heated at a rate of 8 °C min⁻¹ to 290 °C, and maintained for another 5 min. In MS, ion source temperature was kept at 230 °C and operated at full scan mode with mass range 40–500 amu. About 0.5 μL of sample was injected with an autosampler in a split mode with column flow 1.2 mL min⁻¹ and split flow 30 mL min⁻¹. Phenanthrene was used as an internal standard. Products were identified by comparison with pure compounds.

Sample (2 mg mL⁻¹) was dissolved in acetonitrile–methanol (1 : 2) and analyzed by full-scan electrospray ionisation mass spectrometry (m/z range from 60 to 2000 with 1 scan per s) on Waters, Micromass Q-TOF micro with Waters Alliance 2795 separation module (Waters Corporation, Milford, MA). The system consisted of Waters 1525μ binary HPLC pump with mobile phase degassing unit, a quadrupole-time of flight mass spectrometer and MassLynx™ 4.0 software. Samples of 20 μL were injected by the autosampler and led into the mass spectrometer by 70 cm of PEEK tubing (i.d. of 0.18 mm) without separation on a chromatographic column. Each sample was analyzed 5 times. The mobile phase consisted of 80 : 20 methanol–water, using a flow rate of 0.3 mL min⁻¹. Both positive and negative electrospray ionization was used to detect different compounds. In all cases, identical methods of extractions and analysis conditions were maintained.

Results and discussion

Catalyst characterization

The zeolite was synthesized by an energy conserving procedure at low temperature (95 °C) without using autoclave. The synthesized zeolite catalyst HZSM-5 was characterized by XRD, SEM, EDS and BET-BJH analysis. The XRD diffractogram of HZSM-5 is shown in Fig. 1. The peaks at 2θ values 7.94, 8.78, 23.07, 23.95 and 45.06 corresponding to 011, 020, 051, 033 and 0 10 0 crystal planes confirms the typical MFI (Mordenite Framework Inverted) structure of the synthesized zeolite. This was also confirmed with ICDD ref. code 00-044-0003. The average crystallite size (coherently diffracting domains) was calculated using the Scherrer's equation to be 11.78 nm. Table 1 shows the calculated lattice (orthorhombic) parameters of synthesized HZSM-5. The calculated lattice parameters were very close to the standard lattice parameters (as verified with standard ICDD data). The SEM micrographs showed that the morphology of the crystals was nearly cubical with four roughened edges (Fig. 2A). The size of the crystals varied from 2 μm to 10 μm. The EDS spectra, as shown in Fig. 2B, confirms the presence Si, Al and O elements with Si/Al ratio of 42. Surface area of the zeolite obtained using the BET method based on adsorption isotherms was 271.28 ± 3.60 m² g⁻¹. The BJH adsorption cumulative surface area and total pore volume of pores was calculated to be 100.96 m² g⁻¹ and 0.31 cm³ g⁻¹ respectively (Table 2). The average pore diameter of the zeolite was 4.6318 nm that indicated towards mesoporosity of the zeolite.

Fig. 1 XRD diffractogram of HZSM-5.

Table 1 Lattice parameters of HZSM-5

HZSM-5 (ICDD^a ref. code 00-044-0003)		HZSM-5 (synthesised)
Parameter	Value	Parameter	Value
a International centre for diffraction data.
a (Å)	20.104	a (Å)	20.094
b (Å)	19.897	b (Å)	19.891
c (Å)	13.395	c (Å)	13.393
Cell volume (10^6 pm^3)	5358.12	Cell volume (10^6 pm^3)	5353.61

---

Fig. 2 SEM micrographs of HZSM-5 at (A) 10 000×, (B) 5000×, (C) 1500× and (D) EDS spectra of HZSM-5.

Table 2 N₂ adsorption–desorption studies of HZSM-5

Property	Value
BET surface area	271.28 m^2 g^−1
Cumulative surface area of pores	100.96 m^2 g^−1
External surface area	170.32 m^2 g^−1
Total pore volume	0.31 cm^3 g^−1
Average pore diameter	4.63 nm

---

Lignin depolymerization studies

Lignin depolymerisation was studied in methanol and water–methanol mixture as solvents. The reactions were performed for different times as shown in Fig. 3. The lignin conversion in methanol as a solvent gradually decreased with reaction time from 82.8% at 1 h to 68.7% at 15 h (Fig. 3). As we aim at lignin conversion to monomeric products that are expected to be separated as water and EtOAc soluble products, hence yields of only these products are shown in Table 3. Yields of THF soluble and gaseous products can be found in Table S1, ESI.† The lignin conversion in water–methanol mixture without NaOH was much lower as compared to methanol (Table 3, entry 1 and 7). However, adding 2 mmol NaOH to water–methanol mixture increased the lignin conversion (Fig. 3). Maximum conversion of 98.5% was obtained at 7 h of reaction time in water–methanol mixture with NaOH. Hence, we focused our attention to the reaction time of 7 and 15 h for further comparisons. These comparisons are shown in Table 3. Without NaOH and zeolite (HZSM-5) catalyst, lignin conversion in methanol is 73.6% and 68.7% at 7 h and 15 h respectively. There was also considerable amount of char formation (Table 3, entry 1 and 2). Addition of NaOH did not improve the conversion or product yield and there was significant increase in the char formation at 15 h of reaction time. This may be due to NaOH catalyzing repolymerisation of the depolymerised products.²³ However, when HZSM-5 is used as a catalyst, the lignin conversion is improved to 85.1% and 80.9% at 7 h and 15 h of reaction time respectively. The yield of EtOAc soluble products was also increased to 38.5% and 26.5% respectively with significant lowering in the amount of char formation (Table 3, entry 5 and 6). This may be due to hydrodeoxygenation of products by HZSM-5 that minimizes the repolymerisation reactions.

Fig. 3 Lignin conversion without catalyst at different reaction time with methanol [black] and 2 mmol NaOH in water–methanol (1 : 1) [grey].

Table 3 Lignin conversion, product yields and char yields with and without catalyst^a

Entry	Solvent	Reaction time (h)	Catalyst (g)	NaOH (mmol)	% Lignin conversion	Difference in conversion^b	% Water soluble products^c	% EtOAc soluble products^c	% Char
a Reaction condition: lignin = 0.5 g, solvent = 20 mL, temperature = 220 °C.b Difference in lignin conversion at 15 h and 7 h (negative sign indicates decrease in conversion).c By weight after separation and solvent removal according the Scheme 1.
1	Methanol	7	—	—	73.6	−4.9	10.3	16.4	18.9
2	Methanol	15	—	—	68.7		9.5	19.3	16.5
3	Methanol	7	—	2	68.5	−25.3	2.7	18.6	19.1
4	Methanol	15	—	2	43.2		3.6	19.0	41.6
5	Methanol	7	0.1	—	85.1	−4.2	4.5	38.5	12.2
6	Methanol	15	0.1	—	80.9		4.6	26.5	16.5
7	Methanol–water (1 : 1)	7	—	—	56.9	—	9.3	43.6	19.3
8	Methanol–water (1 : 1)	7	—	2	96.3	−9.9	16	44.1	<1
9	Methanol–water (1 : 1)	15	—	2	86.4		10.9	42.8	<1
10 (ref. 22)	Methanol–water (1 : 1)	7	0.1	2	98.5	−10.4	16	44.9	<1
11	Methanol–water (1 : 1)	15	0.1	2	88.7		13.4	42	<1

---

Without NaOH and HZSM-5, shifting the solvent from methanol to water–methanol mixture did not improve the lignin conversion or a reduction in char formation but there was an increase in the yield of EtOAc soluble products (Table 3, entry 7). The lignin conversion and yield of EtOAc soluble products were sharply increased to 96.1% and 44.1% respectively at 7 h of reaction time in presence of NaOH. There was remarkably reduced formation of char to <1% (Table 3, entry 8 and 9). This is due the NaOH catalyzed hydrolysis of lignin coupled with the nucleophilic effect of water (as water is good nucleophile) that prevent the carbocations formed during cleavage of ether linkages to repolymerize with other lignin depolymerization products. Adding NaOH with zeolite HZSM-5 as catalyst, the lignin conversion in water–methanol mixture was slightly increased to maximum of 98.5% at 7 h. There was little increase in the yield of products with very low formation of char (Table 3, entry 10 and 11). Although addition of HZSM-5 in the presence of NaOH did not show any significant change in conversion and product yield, HZSM-5 was found to improve the product composition and monomeric product yield in EtOAc soluble products as noticed in the gas chromatographic analysis. EtOAc soluble products (in reaction where only NaOH and no HZSM-5 was used as catalyst) were not detectable in GC analysis. This suggested the presence of oligomeric products.^9,22 Extending the reaction time from 7 h to 15 h resulted in decrease in the lignin conversion in all the cases as can be seen in the entries provided in Table 3.

Functional groups changes during lignin conversion

FTIR analysis is used to assess the expected structural changes occurring in a mixture during depolymerisation experiments. The effect of solvent and catalyst conditions on the products recovered as EtOAc soluble products was studied by FTIR spectroscopy. The FTIR spectrum of lignin shows peaks at 2965 cm⁻¹, 2879 cm⁻¹ and 2837 cm⁻¹ due to symmetric and asymmetric stretching of C–H bonds with C–H deformation vibrations at 1453 cm⁻¹ (–CH₂–) and 1370 cm⁻¹ (–CH₃). Aromatic C–H vibrations are indicated by the peaks at 1596 cm⁻¹, 1512 cm⁻¹, 1425 cm⁻¹ and 1032 cm⁻¹. The peak at 1716 cm⁻¹ indicates C=O stretching vibrations. Broad vibrational bands at ∼3200 cm⁻¹ are due hydrogen bonded O–H stretching vibrations. The peaks at 1100–1350 cm⁻¹ are stretching vibrations of C–O and C–O–C (ether linkages) in syringyl and guaiacyl units (Fig. 4A).^24,25

Fig. 4 FTIR spectra of (A) lignin and EtOAc soluble products at reaction time (B) 7 h and (C) 15 h, FTIR ratios of functional group indices based on peak integration of EtOAc soluble products at reaction time (D) 7 h and (E) 15 h (solvent and catalyst conditions: M – only methanol, MN – methanol + NaOH, MH – methanol + HZSM-5, WMN – water–methanol (1 : 1) + NaOH, WMNH – water–methanol (1 : 1) + NaOH + HZSM-5).

The FTIR spectra of the EtOAc soluble products at 7 h of reaction time in different solvent–catalyst condition are shown Fig. 4B. There was a significant increase in the intensity of peaks at 2940 cm⁻¹, 2856 cm⁻¹, 1452 cm⁻¹ and 1362 cm⁻¹ along with decreased intensity of the peaks at 1598 cm⁻¹, 1512 cm⁻¹ and 1425 cm⁻¹. This indicated the hydrogenation of the aromatic rings of lignin due to hydrogen transfer from methanol. The appearance of stronger bands in the region ∼3200 cm⁻¹ indicated increase in the hydroxyl group content in the products as a result of depolymerization of lignin via cleavage of ether linkages. As compared to only methanol as solvent and no catalyst (M) condition, greater extent of depolymerization and hydrogenation of lignin was seen with NaOH as catalyst and methanol as solvent (MN). HZSM-5 as catalyst with methanol as solvent (MH) did not show any significant improvement in depolymerization and hydrogenation of lignin. However, water–methanol (1 : 1) as solvent with NaOH and HZSM-5 as catalyst (WMN and WMNH) showed better results as compared to M, MN, and MH conditions. There was complete disappearance of peak at 1512 cm⁻¹ and 1425 cm⁻¹ with substantially decreased intensity of peak at 1598 cm⁻¹. This was due to the fact that hydrolysis reactions greatly enhanced in presence of water.

Lignin conversion in high temperature water at autogenous pressure provided greater hydrolysis of aromatic ether linkages in lignin due to increased ionic product of water at high temperature.^26,27 Low concentration of NaOH increased the solubility of lignin in the reaction mixture making the lignin molecules thoroughly dispersed in the reaction mixture to reduce the possibility of recombination reactions, besides NaOH also promoted alkaline hydrolysis of lignin into smaller units.⁹ Methanol is used as a reactive solvent to provide the reductive equivalents of hydrogen^13,30 as shown in following reactions:

Reforming step: CH₃OH → CO + 2H₂ (6)

Water gas shift reaction: CO + H₂O → CO₂ + H₂ (7)

Also, methanol is also expected to inhibit the recombination reaction by acting as a capping agent. HZSM-5 has strong hydrodeoxygenation activity for the conversion of oxygenates to hydrocarbons.^28,29

FTIR spectra of the EtOAc soluble products at 15 h of reaction time in different solvent–catalyst condition are shown in Fig. 4C. There is no significant change in the peak positions in case of M and MN. In case of MH, there was disappearance of peak at 1719 cm⁻¹ and increase in the band intensity at ∼3200–3500 cm⁻¹ indicating hydrogenation of C=O functional groups to alcohols. The broad peak at ∼3295 cm⁻¹ and 1592 cm⁻¹ indicates the presence of phenolic structures in the product mixture. When the solvent is shifted to water–methanol (1 : 1) mixture (WMN and WMNH), decreased intensity peaks at ∼3295 cm⁻¹ and 3512 cm⁻¹ indicated removal of hydroxyl groups (hydrodeoxygenation). There was enhanced intensity of peak at 1592 cm⁻¹ indicating increase in the aromatic content in the product mixture. This was also verified by calculating numerical indices for each FTIR spectra to classify the chemical transformations taking place in the complex reaction product mixtures. These indices are given below in eqn (8)–(10):

Al = {A_Al/(A_Al + A_Ar + A_O)} × 100 (8)

Ar = {A_Ar/(A_Al + A_Ar + A_O)} × 100 (9)

O = {A_O/(A_Al + A_Ar + A_O)} × 100 (10)

where A_Al represents the integrated area of peaks of aliphatic C–H vibrations, A_Ar represents the integrated area of peaks of aromatic C–H vibrations and A_O represents the integrated area of addition of peaks of C=O and C–O vibrations. Thus, Al represents fraction of aliphatic functional groups present in the product mixture, Ar represents fraction of aromatic functional groups present in the product mixture and O represents fraction of oxygen containing functional groups present in the product mixture. We have plotted the ratios Al/Ar and Al/O against the solvent–catalyst condition studies. The Al/Ar and Al/O ratios increased from M to WMNH condition (Fig. 4D). When water–methanol (1 : 1) mixture was used as solvent with NaOH and HZSM-5 as catalyst (WMNH), maximum Al/Ar ratio (11.06) was found at 7 h of reaction time as compared to initial lignin (Al/Ar = 0.86). However, the Al/Ar ratio decreased to 4.43 at 15 h of reaction time (Fig. 5E). This indicated rearomatization of the products due to extended reaction time. The Al/O ratio also increased from M to WMNH conditions indicating removal of oxygen from products at 7 h of reaction time. The maximum Al/O ratio was 1.47 in WMNH condition as compared to 0.42 of initial lignin (Fig. 5D). These results indicated that methanol–water mixture as solvent supports more hydrogenation and hydrodeoxygenation activity for lignin conversion process in presence of catalyst.

Fig. 5 (A) Quantitative yield of monomeric products (B) yields of acyclic and cyclic products (for reaction condition see Table 3, entry 5 and 10).

Product analysis

Based on FTIR results, the water soluble and EtOAc soluble products from 7 h reaction time (Table 3, entry 5 and 10) were quantitatively analyzed with GCMS for comparison. The yields of products identified as water soluble and EtOAc soluble are shown combined in Fig. 5. The products from the reaction where methanol was used as solvent showed formation of 4.1 wt% of phenols like phenol, 2-methoxy- (0.31 wt%), phenol, 2,6-dimethoxy- (0.83 wt%), phenol, 2-methoxy-4-(1-propenyl)- (0.28 wt%), 4-hydroxy-2-methoxybenaldehyde (0.2 wt%), etc. Apart from phenols, 0.5 wt% acyclic aliphatic products like 1-hexanol, 2-ethyl-, acetic acid, 2-methylpropyl ester, 1-pentanol, 2-ethyl-4-methyl-, 16-hydroxy, 9-octadecanoic acid, methyl ester, etc. are also found. The amount of cyclic aliphatic products was negligible.

The reaction performed using water–methanol (1 : 1) as solvent without NaOH was not selected for GCMS analysis due to very low conversion and char formation (Table 3, entry 7). The products from reaction having water–methanol as solvent with NaOH (Table 3, entry 10) were constituted by 14.1 wt% phenols, 2.8 wt% cyclic aliphatic compounds and 3.7 wt% acyclic aliphatic compounds as shown in Fig. 5. The major phenols identified are durohydroquinone (3.1 wt%), phenol, 2,4,5-trimethyl (1.7 wt%), 1,4-benzenediol, 2,3,5-trimethyl (0.8 wt%) and phenol, 5-methoxy-2,3-dimethyl (0.9 wt%). Acyclic compounds comprised of 2.4 wt% deoxygenated acyclics (hydrocarbons) and 1.3 wt% of oxygenated acyclics. Major hydrocarbons identified were tetradecane (0.4 wt%), dodecane (0.3 wt%), nonadecane (0.15 wt%), tetradecane, 2,6,10-trimethyl (0.07 wt%), etc. The total ion chromatograms (GCMS) and detailed list of the products identified are shown in ESI.† The products from reaction in water–methanol as solvent (Table 3, entry 7) were analyzed by electrospray ionization mass spectroscopy (ESI-MS). The positive and negative mode spectra from ESI-MS are shown in Fig. 6. The spectra showed that the mass distribution of products ranges from ca. 60–800 m/z with number average (M_n) and weight average (M_w) molecular weight of 380.2 m/z and 448.0 m/z respectively.³⁰ The poly-dispersity index was found to be 1.2 (near to unity) which indicated narrow range of mass distribution and is much lower as compared to initial lignin (∼5–6). The PDI of lignin is definitely much higher than that of depolymerization and conversion products. Reduction in PDI was a result of effective conversion of lignin to low molecular weight compounds and consequent extraction in water and ethyl acetate. The products obtained showed formation of long chain alcohols, ketones and ester. In addition, acyclic C₁₂–C₁₉ hydrocarbons are formed that are reduced products of many of the open chain structures formed in pyrolytic degradation of lignin^5,6,31 and solvolytic depolymerisations.^{13,14,17,19,21}

Fig. 6 ESI-MS spectra of products (A) positive mode (B) negative mode.

Catalyst recovery

The pore size of the HZSM-5 catalyst is much smaller than the lignin molecules. When only HZSM-5 is used as catalyst, the pores of the zeolite may be blocked by lignin and lignin fragment and thus, deactivating the catalyst. Adding NaOH in methanol showed increase in char formation, hence NaOH was not utilized as co-catalyst with HZSM-5 in methanol. When NaOH is used as co-catalyst in presence of water, it results in fragmentation of lignin into smaller units. This is due OH⁻¹/H₂O has greater hydrolysing effect than OH⁻¹/MeOH.^9,12 The lignin fragments smaller than pore size of catalyst can diffuse into the pores for further hydrodeoxygenation/hydrogenation. The lignin fragments larger than pore size undergo hydrodeoxygenation/hydrogenation on catalyst surface and may further fragment into smaller units allowing their diffusion into the pores. The recovery of spent catalyst was not feasible without recalcination in the reactions where char formation was observed (Table 3, entry 5 and 6). Catalyst recovery was feasible in the reaction where very little char formation was observed (Table 3, entry 10 and 11). Amount of catalyst recovered from the process was ∼95–97 wt%. The XRD diffractogram of recovered catalyst from reaction (Table 3, entry 10) showed that it remained stable in the reaction. The XRD diffractogram of recovered catalyst is shown in Fig. 7. The catalyst reuse showed that the catalyst retained its activity up to 3 cycles.²²

Fig. 7 XRD spectra of HZSM-5 after and before reaction.

Conclusions

We have demonstrated that NaOH in water–methanol (1 : 1) mixture is an effective solvent for one-pot greener approach towards lignin depolymerization and conversion into value added compounds in comparison to NaOH in pure methanol. Methanol, besides being used as a solvent, is also aptly used as source of reductive equivalents of hydrogen instead of using molecular hydrogen. This study shows that alkaline water plays a significant role in suppression of char formation during lignin depolymerisation. HZSM-5 was synthesized at low temperature (95 °C). It catalyzes the hydrodeoxygenation and propyl side chain removal in the aromatic moieties of the depolymerised lignin. Alkyl substituted phenols are the major products formed. Phenol yield is increased manifold in water–methanol (14.1 wt%) as solvent in comparison to only methanol (phenol yield – 4.1 wt%) as solvent. This process generates value added phenols from industrial waste lignin which are versatile precursors of phenolic resins, polymers and pharmaceutical drugs.

Acknowledgements

First author Sunit K. Singh is thankful to The Director, VNIT, Nagpur for providing necessary facilities and financial support. We are thankful to SAIF, Chandigarh for providing XRD and ESI-MS facility and SAIF, Chennai for SEM-EDS facility.

References

1. (a) S. P. Magalhaes da Silva, A. M. da Costa Lopes, L. B. Roseiro and R. Bogel-Łukasik, RSC Adv., 2013, 3, 16040; (b) H. Xie and N. Gathergood, The Role of Green Chemistry in Biomass Processing and Conversion, JohnWiley & Sons, Inc., Hoboken, New Jersey, 2013.
2. D. Dimmel, Lignin and Lignans: Advances in Chemistry, ed. C. Heitner, D. R. Dimmel and J. A. Schmidt, CRC Press-Taylor & Francis Group, 2010.
3. J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius and B. M. Weckhuysen, Chem. Rev., 2010, 110, 3552.
4. V. M. Roberts, V. Stein, T. Reiner, A. Lemonidou, X. Li and J. A. Lercher, Chem.–Eur. J., 2011, 17, 5939.
5. H. Ben and A. J. Ragauskas, Energy Fuels, 2011, 25, 4662.
6. H. Ben and A. J. Ragauskas, Energy Fuels, 2011, 25, 2322.
7. M. Kleinert and T. Barth, Energy Fuels, 2008, 22, 1371.
8. G. Gellerstedt, J. Li, I. Eide, M. Kleinert and T. Barth, Energy Fuels, 2008, 22, 4240.
9. P. T. Patil, U. Armbruster, M. Richter and A. Martin, Energy Fuels, 2011, 25, 4713.
10. K. Okuda, M. Umetsu, S. Takami and T. Adschiri, Fuel Process. Technol., 2004, 85, 803.
11. K. Okuda, S. Ohara, M. Umetsu, S. Takami and T. Adschiri, Bioresour. Technol., 2008, 99, 1846.
12. Z. Yuan, S. Cheng, M. Leitch and C. Xu, Bioresour. Technol., 2010, 101, 9308.
13. K. Barta, T. D. Matson, M. L. Fettig, S. L. Scott, A. V. Iretskii and P. C. Ford, Green Chem., 2010, 12, 1640.
14. K. Barta, G. R. Warner, E. S. Beach and P. T. Anastas, Green Chem., 2014, 16, 191.
15. G. Warner, T. S. Hansen, A. Riisager, E. S. Beach, K. Barta and P. T. Anastas, Bioresour. Technol., 2014, 161, 78.
16. M. A. Jackson, D. L. Compton and A. A. Boateng, J. Anal. Appl. Pyrolysis, 2009, 85, 226.
17. T. R. Carlson, T. P. Vispute and G. W. Huber, ChemSusChem, 2008, 1, 397.
18. R. W. Thring, S. P. R. Katikaneni and N. N. Bakhshi, Fuel Process. Technol., 2000, 62, 17.
19. C. Zhao and J. A. Lercher, ChemCatChem, 2012, 4, 64.
20. P. M. Froass, A. J. Ragauskas and J. E. Jiang, J. Wood Chem. Technol., 1996, 16, 347–365.
21. A. L. Jongerius, P. C. A. Bruijnincx and B. M. Weckhuysen, Green Chem., 2013, 15, 3049.
22. S. K. Singh and J. D. Ekhe, RSC Adv., 2014, 4, 27971.
23. J. E. Miller, L. Evans, A. Littlewolf and D. E. Trudell, Fuel, 1999, 78, 1363.
24. O. Faix, Fourier transformed infrared spectroscopy, in Methods in Lignin Chemistry, ed. S. Y. Lin and C. W. Dence, Springer-Verlag, Berlin-Heidelberg, 1992.
25. C. G. Boeriu, D. Bravo, R. J. A. Gosselink and J. E. G. van Dam, Ind. Crops Prod., 2004, 20, 205.
26. N. Akiya and P. E. Savage, Chem. Rev., 2002, 102, 2725.
27. A. A. Peterson, F. Vogel, R. P. Lachance, M. Froling, M. J. Antal and J. W. Tester, Energy Environ. Sci., 2008, 1, 32.
28. U. V. Mentzela and M. S. Holma, Appl. Catal., A, 2011, 396, 59.
29. Y. Wang, Y. Fang, T. He, H. Hu and J. Wu, Catal. Commun., 2011, 12, 1201.
30. Q. Song, F. Wang, J. Cai, Y. Wang, J. Zhang, W. Yu and J. Xu, Energy Environ. Sci., 2013, 6, 994.
31. A. V. Maldhure and J. D. Ekhe, J. Environ. Chem. Eng., 2013, 4, 844.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra10240a

---