Mechanochemical degradation of lignin and wood by solvent-free grinding in a reactive medium

Abstract

A mechanochemical approach for the cleavage of β-O-4-linkages in lignin is reported. The method is transition metal- and solvent-free, requires only inexpensive reagents, and tolerates the presence of water and standard reagent impurities.

---

1. Introduction

Recently, mechanochemistry has attracted much attention because it allows promotion of reactions under solvent-free conditions.¹ Even solids react quickly and quantitatively in the absence of a solvent, avoiding (or at least reducing) the use of those often environmentally problematic, or even hazardous, fossil-derived liquids. Mechanochemical synthesis is mainly done in ball mills, which leads to an efficient mixing of all reagents and provides a significant energy input.² Compared to other techniques, such as ultrasound or microwave, the energy efficiency of ball milling is higher.^3,4 When hard material is ground in a ball mill the particle size decreases. On a large scale, crushing ores is well established using this technique.⁵ With respect to molecular crystals mechanochemical reactions have been suggested to be driven by an unusual alternation of the electron distribution by molecular deformation.⁶

Depleting fossil carbon sources have generated strong pressure on the oil-based industries, and a shift to renewable resources is considered essential for sustaining industrial growth.^7–9 The usage of feedstock-dedicated plants is controversial, because of the low efficiency of land usage, collateral effects on the nourishment supply and increased greenhouse gas emissions.^8–10 In the light of these disadvantages, the exploitation of whole-plant material or waste biomass appears desirable. It provides the biopolymers cellulose, hemicellulose, and lignin in large quantities (cellulose: 25–50%, hemicellulose: 25–30%, lignin: 15–30% of dry biomass),^11,12 and consequently, those compounds are cheap representing attractive sources for platform chemicals.^13,14 Unfortunately, their poor solubility, complexity and structural diversity hamper their use as raw materials for new valuable carbon sources. Of particular interest in this context is lignin. It is a three-dimensional amorphous material consisting of methoxylated phenyl propane units, which are interlinked by various binding motifs. Among them, the β-O-4-linkage is the most abundant [∼60% in hardwood, ∼45% in softwood¹⁴]. Although ubiquitously in nature and known for decades, the technical use of lignin is limited by its poor solubility and structural complexity.¹⁵ Degradation or solvation processes generally require harsh reaction conditions, which follow a mechanical pretreatment.^14,16–18 For unmodified wood, these issues are even more severe.⁹ Understanding wood and lignin processing on a molecular level appears essential for improving their degradation efficiencies.^14,19,20

For the determination and analysis of lignin transformation pathways it is common to start with lignin model compounds, which contain the most significant binding motifs of the natural biopolymer.^14,21,22 Along those lines, numerous reports have recently focused on selective oxidative or reductive degradation processes catalysed by transition metals.^23,24 However, these methods generally require high catalyst loadings, and they suffer from incompatibilities with air, moisture and typical impurities present in the natural products. A simple and efficient method has recently been described by Hartwig, who reported the cleavage of lignin model compound 1a. Treatment of 1a in xylene with an excess of sodium tert-butoxide at 100 °C led to phenol 2a together with various unidentified by-products (Scheme 1).²⁴

Scheme 1

Searching for alternative approaches, we wondered about mechanochemical strategies for the degradation of lignin. In general, those have been known for a long time, but commonly they were applied for particle size reduction as a pretreatment for a chemical depolymerisation process performed under reaction conditions involving solvents in combination with elevated pressure and/or high temperature.^12,16,17 Because strongly basic solutions have also been applied in lignin cleavage reactions,^20,25,26 we wondered about a combination approach and questioned if solvent-free ball milling in the presence of a (solid) base could be used for lignin degradation (Scheme 2). To the best of our knowledge, this has never been studied although it appeared as an obvious advantageous option. Here, we describe the initial fine-tuning of the ball milling process using lignin model compounds and demonstrate the applicability of the method for the degradation of both technical organosolv lignin and beech wood. The latter two transformations are particularly interesting because they illustrate the potential of the process to overcome the major difficulties in lignin and wood conversions associated with their inherent insolubilities in common solvents.

Scheme 2

2. Results and discussion

For the initial reactivity search and the subsequent optimisation of the reaction conditions, dilignol erythro-1a was chosen as a model substrate. It is structurally well defined and can readily be accessed in a diastereomerically homogeneous form following established routes developed in our laboratories.²² Unfortunately, our first attempts to degrade erythro-1a by grinding with sodium carbonate (as base and grinding auxiliary) in a planetary ball mill remained unsuccessful, and no conversion was observed after 2 hours (Table 1, entry 1). Assuming that a stronger base was needed, metal hydroxides were applied [in combination with sodium sulfate as a grinding auxiliary (Table 1, entries 2–5)²⁷]. Under those conditions, dilignol erythro-1a reacted, and the formation of 2-methoxy phenol (2a) and a phenyl propane fragment resulting from C–O-bond cleavage was observed. Whereas with lithium hydroxide and potassium hydroxide the yield of 2a was low (only 16% and 29%, respectively), the use of sodium hydroxide gave 2a in remarkable 62% yield.

Table 1 Optimisation of the reaction conditions^a

Entry	No. of balls	Freq. (Hz)	Base (equiv.)	Time (min)	Yield^b (%)
a Use of 0.2 mmol of 1a and 1 g of Na2SO4 (as grinding auxiliary). b Determined by GC using 4-methoxyphenol as an internal standard. c No additional auxiliary. d Used as a 60% suspension in mineral oil.
1	30	13.3	Na2CO3 (9.5)^c	120	0
2	30	13.3	LiOH (3.5)	120	16
3	30	13.3	KOH (3.5)	120	29
4	30	13.3	NaOH (3.5)	120	62
5	30	13.3	Ca(OH)2 (3.5)	120	0
6	30	13.3	NaH (3.5)^d	120	47
7	30	13.3	NaOt-Bu (3.5)	120	70
8	30	13.3	NaOH (1.0)	120	27
9	30	13.3	NaOH (2.0)	120	40
10	30	13.3	NaOH (3.0)	120	52
11	30	13.3	NaOH (5.0)	120	71
12	30	13.3	NaOH (10.0)	120	76
13	30	13.3	NaOH (3.5)	5	2
14	30	13.3	NaOH (3.5)	15	16
15	30	13.3	NaOH (3.5)	60	43
16	30	13.3	NaOH (3.5)	200	70
17	30	13.3	NaOH (3.5)	720	82
18	30	13.3	NaOH (10.0)	720	94
19	15	13.3	NaOH (3.5)	120	37
20	45	13.3	NaOH (3.5)	120	54
21	60	13.3	NaOH (3.5)	120	52
22	30	10.0	NaOH (3.5)	120	23
23	30	16.7	NaOH (3.5)	120	56

---

Calcium hydroxide was ineffective. Also sodium hydride and sodium tert-butoxide could be applied leading to 2a in 47% and 70% yield, respectively (Table 1, entries 6 and 7). However, considering the future perspective of lignin and wood conversions, the latter strong bases appeared less attractive for potential scale-ups. Thus for further studies, sodium hydroxide was selected as a base.

Until then, 3.5 equiv. of the base in combination with a standard quantity of the grinding auxiliary had been applied. With less sodium hydroxide the yield of 2a was lower (Table 1, entries 8–10). Although increasing the amount of the base to 5 and 10 equiv. improved the yield of 2a (up to 76%), it was decided to continue the study with the original base quantity (3.5 equiv.). Longer reaction times led to higher yields (Table 1, entries 13–17), and after 12 h, 2a was obtained in 82% yield. Also here, increasing the base loading to 10 equiv. had a positive effect and 2a could be isolated in 94% yield (Table 1, entry 18).

Finally, the protocol was optimised by improving the mechanical grinding properties of the ball milling process. Three points, which were expected to have an influence on the degradation process, were addressed: first, the grinding bowl topology, second, the number of balls, and third, the grinding speed. The best result was achieved when the reaction was performed in the grinding bowl with a diameter of 4.8 cm using 30 balls and a rotational speed of 13.3 Hz. Use of a grinding bowl with a smaller diameter (2.6 cm), applying more or less balls (Table 1, entries 19–21), or reducing the rotational speed (to 10.0 Hz; Table 1, entry 22) gave 2a in lower yields. At 16.7 Hz more by-products were formed (Table 1, entry 23). Apparently, the reaction required both mechanical stress/pressure (represented by the kinetic energy in the system) and a specific temperature range (resulting from friction). The former was achieved by using an appropriate bowl size, which guaranteed a high motional freedom of the grinding balls. The temperature range was reached by fine-tuning of the rotational speed, which promoted particle interactions and frictions leading to a process temperature of ca. 50 °C.

An agglutination of molten reactants could be prevented by using a grinding auxiliary. Sodium sulfate proved the best. Whereas sodium chloride could also be used as a grinding auxiliary, the presence of calcium chloride inhibited the reaction. Silica, basic aluminum oxide and molecular sieves led to more unselective side reactions.

Next, the degradation method was applied to other lignin model compounds such as dilignols 1a–e, g and the less complex monolignols 1f and 1h (Table 2). Because product 2a was volatile and unstable on silica, the yields were initially determined by GC analysis for reactions on a 0.2 mmol scale. The results showed that neither steric hindrance nor the presence of multi-methoxy groups significantly affected the cleavage reactions. The solid dilignols erythro-1a and threo-1b and monolignols 1f and 1h gave higher yields than resin-like threo-1a, threo-1c, and erythro-1d, e, g. Performing reactions on a 0.8 mmol scale (Table 2, entries 1, 8 and 9) allowed us to isolate the products, and in this manner, 2a was obtained in 67% yield from erythro-1a (Table 2, entry 1). Starting from erythro-1g and 1h product 2c was isolated in 60% and 90% yield, respectively (Table 2, entries 8 and 9).

Table 2 C–O–Bond cleavage of β-O-4-model compounds 1a–h^a

Entry	Model compound	Product	Yield^b (%)
a Use of 0.2 mmol of 1, 3.5 equiv. of NaOH, 1 g of Na2SO4, 30 balls, 720 min, 13.3 Hz. b Determined by GC using 4-methoxyphenol as an internal standard. The values given in parentheses refer to the amount of isolated phenol obtained from reactions performed on a 0.8 mmol scale.
1			82 (67)
2			57
3			91
4			57
5			67
6			57
7			84
8			(60)
9			(90)

---

The data presented in Table 2 allow a first mechanistic interpretation, which we intended to validate by studying specifically designed model compounds 3–8. Compared to the originally applied mono- and dilignol derivatives 1, various potential deprotonation sites were blocked by the presence of methyl groups in those derivatives. Assuming that the degradation process began with a base-mediated deprotonation of the acidic alcoholic hydroxyl groups, the results obtained with methyl ethers 3–5 were particularly important. As shown by the data in Fig. 2, the yield of 2a was significantly reduced to 4% and 16%, respectively, when the primary hydroxyl group was methylated as in 3 and 4. In contrast, methylation of the secondary hydroxyl group had only a minor impact as revealed by the degradation of 5, which gave 2a in 50% yield. The impact of C-methylations remained inconclusive. The high yield of 2a (56%) obtained in the cleavage of substrate 8 showed that a substitution at the benzylic position had only a minor effect on the degradation efficiency. In contrast, additional methyl groups in the side chain (as in compounds 6 and 7) significantly reduced the yield of 2a (to 15% and 30%, respectively). In those cases, however, the data interpretation is more complex because the reduced cleavage rate could also be due to a conformational change induced by the buttressing dimethyl-substitution at the former methylene group. In summary, all data suggested that the initial deprotonation at the primary hydroxyl group was followed by intramolecular cleavage of the β-ether bond leading to two products: first, an unstable epoxide 9 that underwent subsequent transformations affording numerous unidentified (by)products, and second, phenolate 2a, which could be isolated as a single component in pure form. This mechanistic proposal was in agreement with the one suggested for the classical kraft pulp process.¹⁸ Although the protons at the hydroxyl-bearing carbons of the lignol derivatives 1 were not essential for the cleavage, they had a significant influence on the reaction rate indicating the occurrence of subsequent deprotonation steps, which could trigger further inter- or intramolecular reactions.²⁸

Next, the degradation of more complex systems was attempted. For this purpose, two samples of organosolv lignin (A and B) from different sources were applied as starting materials. The structural changes and the degradation progress under standardized reaction conditions (100 mg of lignin, 100 mg of sodium hydroxide, 2.5 g of sodium sulfate, 30 balls, 12 h at 13.3 Hz) were determined by 2D NMR (HSQC) analysis of the samples before and after the grinding. The spectra interpretation was based on the excellent report by Sun et al.¹⁷ The signal intensities before and after the treatment of the compounds were normed relative to the content of aromatic protons (δ_F1 = 100–120 ppm, δ_F2 = 7.5–6.0 ppm). Those signals were locally isolated and their intensity was expected to remain unchanged during the cleavage reaction. For the quantification the following isolated signals were used: A-α, A-β, B-α, B-β, C-α, C-β (Fig. 1a–f).

Fig. 1 HSQC NMR spectra of lignin samples A and B; (1a) organosolv lignin A; (1b) organosolv lignin A after mechanochemical treatment; (1c) organosolv lignin B; (1d) organosolv lignin B after mechanochemical treatment; carbohydrate section of sample B before (1e) and after (1f) mechanochemical treatment; all spectra were recorded in DMSO-d₆. For additional information see ESI.†

Organosolv lignin sample A had been extracted from beech wood in a neutral ethanol/water organosolv process. It was then washed three times with hot water to remove the cellulose and hemicellulose residues. Consequently, the NMR spectrum showed that the carbohydrate content was very low (Fig. 1a). The ratio of binding motifs (β-O-4 : resinol : phenylcoumaran = 3.05 : 1.00 : 0.50) revealed that the β-O-4 linkage was the most abundant one in this sample. After treatment under standard reaction conditions (see above for details), the original ratio was changed to 0.80 : 1.00 : 0.04 (Fig. 1b). Hence, 76% of the β-O-4 binding motifs present in the untreated organosolv lignin A were cleaved during the process. The resinol binding motifs appeared to remain unaffected by the treatment.

Also the second organosolv lignin (sample B) originated from beech wood. It was obtained in the same manner as sample A, but it lacked the subsequent purification steps. Consequently, the NMR spectrum showed a significant poly-/oligomeric carbohydrate (cellulose, hemicellulose) content (Fig. 1c and 1e). Before the mechanochemical treatment the ratio of the binding motifs β-O-4 : resinol : phenylcoumaran was 2.83 : 1.00 : 0.33 (Fig. 1c), which was similar to the one of sample A. After the grinding, the ratio was 1.21 : 1.00 : 0.12 (Fig. 1d) indicating that 55% of the β-O-4 linkages had been cleaved. Interestingly, also the glycosidic bonds (CH–O–CH) had been broken, as revealed in Fig. 1f. Apparently, the grinding process also led to depolymerization of the carbohydrate content of sample B.

Encouraged by the finding that the linkages of both lignin and carbohydrates were cleaved, the applicability of the mechanochemical process for the degradation of naturally occurring dried beech wood was tested. Because the milled wood sample was insoluble in DMSO, it could not be analyzed by NMR before the mechanochemical treatment. After the grinding, however, the residue easily dissolved in DMSO or a mixture of THF and water. Analysis by 2D NMR (HSQC) revealed the typical lignin/carbohydrate fingerprint described by Sun.¹⁸ The ratio of the binding motifs for β-O-4 : resinol : phenylcoumaran was 1.73 : 1.00 : 0.00 (Fig. 2a), which was similar to the one of the treated organosolv lignin samples A and B. Although a high carbohydrate content was found (Fig. 2a), NMR signals of typical glycoside bindings were lacking (Fig. 2b). Apparently, also in this wood sample, the cellulose and hemicellulose had been degraded to low-molecular weight carbohydrates. Overall these observations supported our initial hypothesis that base-assisted grinding could also be applied for the degradation of natural, untreated wood.

Fig. 2 HSQC NMR spectra of milled beech wood. (2a) Section containing β-O-4 binding, resinol motifs, phenylcoumaran motifs and carbohydrates; (2b) carbohydrate section; all spectra were recorded in DMSO-d₆. Assignments according to Fig. 1; for additional information see ESI.†

A schematic representation of the suggested course of the mechanochemical degradation of hemicellulose and lignin in the original wood structure based on the reported spectroscopic and mechanistic results is shown in Scheme 3.

Scheme 3

3. Conclusions

We have developed a base-assisted ball milling process and demonstrated its potential as a mechanochemical technique for the degradation of lignin and wood. Bonds in lignin are cleaved to provide fragments of lower molecular weight, and cellulose and hemicellulose are depolymerised to yield monomeric carbohydrates. Studying model compounds led to an understanding of the underlying cleavage processes, which in lignin predominantly involve the breaking of β-O-4 linkages. NMR spectroscopy indicated that also cellulose and hemicellulose in organosolv lignin and beech wood were degraded. The findings are in good agreement with those of solvent-based procedures,^16,18 and they expand the existing technology to a new dimension of biomass degradation. Under transition metal- and solvent-free conditions bonds in both lignin and carbohydrates are cleaved using inexpensive, readily available reagents. Currently, the applied base quantity is still too large to be applied on an industrial scale, but we envision that it can be reduced by subsequent reaction process optimisations.

4. Experimental section

4.1. Synthetic procedures

Substrates 1a–e were prepared by following a reported procedure.²² An alternative protocol was used for the synthesis of 1f.^23b The syntheses and structural characterisations of dilignol 1g, monolignol 1h and dilignol derivatives 3–8 are described in the ESI.†

4.2. Procedure for the cleavage of lignin model compounds in the ball mill

4.2.1. Optimized conditions for GC analysis. A grinding bowl (20 mL, ZrO₂) was charged with NaOH (28 mg, 0.7 mmol, 3.5 equiv.), substrate (0.2 mmol, 1.0 equiv.), Na₂SO₄ (1 g) and 30 balls (5 mm, ZrO₂). Then, the mixture was ground in a planetary ball mill (12 h, 13.3 Hz). Subsequently it was neutralized with aqueous acetic acid (1 M), and then 4-methoxy phenol (2 mmol, 24.8 mg) was added. After the slurry was ground for 1 min (13.3 Hz), it was diluted with water (3 mL) and ethyl acetate (6 mL) and mixed well afterwards. Then, 3 mL of the organic layer were distributed over five GC vials and diluted with ethyl acetate to a total of 2 mL each.

4.2.2. For the isolation of phenols 2a and 2c. Two grinding bowls (20 mL, ZrO₂) were charged with NaOH (56 mg, 1.4 mmol, 3.5 equiv.), substrate (0.4 mmol, 1.0 equiv.), Na₂SO₄ (2 g) and 30 balls (5 mm, ZrO₂) each. Then, the mixtures were ground in a planetary ball mill (12 h, 13.3 Hz) and subsequently neutralized with aqueous acetic acid (1 M). After the slurry was ground for 1 min (13.3 Hz) it was dissolved in water (10 mL) and CH₂Cl₂ (50 mL). The layers were separated, the organic layer was dried over MgSO₄, then filtered and concentrated to almost dryness in a vacuum (300 mbar, 15 °C). The crude mixture was separated by flash chromatography (100% CH₂Cl₂ to 100% EtOAc) to afford the phenolic product and the remaining starting material.

4.3. Degradation of organosolv lignin samples A and B in the ball mill

4.3.1. Origin of organosolv lignin samples. Organosolv lignin sample A was supplied by the “Institut für Holztechnologie und Holzbiologie (HTB)”, Hamburg (Dr. Jürgen Puls). It was derived from an organosolv process using beech wood and ethanol–water = 50/50 without an acid catalyst. The lignin was precipitated and washed with water.

Organosolv lignin sample B was supplied by the “Aachener Verfahrenstechnik (AVT)”, Aachen (Serafin Stiefel, Simon Roth) in collaboration with the “Max Planck Institute”, Mühlheim. It was derived from an organosolv process using beech wood and ethanol–water = 50/50 without an acid catalyst. The lignin was precipitated with water and used without further purification.

4.3.2. Procedure for the degradation of organosolv lignin samples A and B in the ball mill. A grinding bowl (20 mL, ZrO₂) was charged with NaOH (100 mg), organosolv lignin (100 mg), Na₂SO₄ (2.5 g) and 30 balls (5 mm, ZrO₂). Then, the mixture was ground in a planetary ball mill (12 h, 13.3 Hz) and subsequently neutralized with aqueous acetic acid (1 M). After the slurry was ground for 1 min (13.3 Hz) it was dissolved in water (10 mL) and THF (20 mL). In a round bottom flask the solvent was removed in a vacuum and DMSO-d₆ (5 mL) was added. The mixture was stirred vigorously for 20 min at 60 °C and then filtered through a pad of silica gel to remove inorganic salts and water. The silica gel was washed with DMSO-d₆ (3 × 1 mL). Subsequently the mixture was concentrated to 0.7 mL and transferred into an NMR tube to be analyzed by NMR spectroscopy.

4.4. Procedure for the degradation of milled beech wood in the ball mill

Beech wood toothpicks were dried in a vacuum for one week at room temperature. Subsequently, they were cut and ground in a planetary ball mill. The grinding procedure constitutes as follows: the grinding bowl (20 mL, ZrO₂) was charged with the wood material and with 30 balls (5 mm, ZrO₂). The material was ground in 40 intervals, each 5 min grinding at 8.3 Hz, 5 min pause, to prevent a temperature increase.

A grinding bowl (20 mL, ZrO₂) was charged with NaOH (150 mg), milled beech wood (100 mg), Na₂SO₄ (2.5 g) and 30 balls (5 mm, ZrO₂). Then, the mixture was ground in a planetary ball mill (12 h, 13.3 Hz) and subsequently neutralized with aqueous acetic acid (1 M). After the slurry was ground for 1 min (13.3 Hz), it was dissolved in water (10 mL) and THF (20 mL). In a round bottom flask the solvent was removed in a vacuum and DMSO-d₆ (5 mL) was added. The mixture was stirred vigorously for 20 min at 60 °C and filtered through a pad of silica gel to remove inorganic salts and water. The silica gel was washed with DMSO-d₆ (3 × 1 mL). Subsequently, the mixture was concentrated to 0.7 mL and transferred to an NMR tube to be analyzed by NMR spectroscopy.

Acknowledgements

The DFG funding within the Cluster of Excellence “Tailor Made Fuels from Biomass” (TMFB) at RWTH Aachen University is greatly appreciated. We thank Dr. Jürgen Puls from the Institut für Holztechnologie und Holzbiologie (HTB), Hamburg, Dr. Roberto Rinaldi from Max Planck Institut für Kohlenforschung, Mülheim a.d.R., Prof. Dr.-Ing. Antje Spieß, Prof. Dr.-Ing. Matthias Wessling, Serafin Stiefel and Simon Roth from the Aachener Verfahrenstechnik (AVT), RWTH Aachen University, for the organosolv lignin samples A and B, respectively. For contributions to the optimization process, we acknowledge Marc Schmitz, and we thank Dr. Christoph Raeuber (RWTH Aachen University) for supporting the NMR investigations.

Notes and references

1. For very recent reviews, see: (a) S. L. James, C. J. Adams, C. Bolm, D. Braga, P. Collier, T. Friscic, F. Grepioni, K. D. M. Harris, G. Hyett, W. Jones, A. Krebs, J. Mack, L. Maini, A. G. Orpen, I. P. Parkin, W. Ch. Shearouse, J. W. Steed and D. C. Waddell, Chem. Soc. Rev., 2012, 41, 413–447; (b) T. Friscic, Chem. Soc. Rev., 2012, 41, 3493–3510.
2. For overviews on the use of ball milling in organic synthesis, see: (a) B. Rodríguez, A. Bruckmann, T. Rantanen and C. Bolm, Adv. Synth. Catal., 2007, 349, 2213–2233; (b) A. Bruckmann, A. Krebs and C. Bolm, Green Chem., 2008, 10, 1131–1141; (c) A. Stolle, T. Szuppa, S. E. S. Leonhardt and B. Ondrushka, Chem. Soc. Rev., 2011, 40, 2317–2329; (d) A. Stolle, B. Ondrushka, A. Krebs and C. Bolm, in Innovative Catalysis in Organic Synthesis, ed. P. G. Andersson, Wiley-VCH, Weinheim, Germany, 2012, pp. 327–349.
3. (a) R. Trotzki, M. M. Hoffmann and B. Ondruschka, Green Chem., 2008, 10, 767–772; (b) F. Schneider, T. Szuppa, A. Stolle, B. Ondruschka and H. Hopf, Green Chem., 2009, 11, 1894–1899; (c) R. Thorwirth, F. Bernhardt, A. Stolle, B. Ondruschka and J. Asghari, Chem.–Eur. J., 2010, 16, 13236–132421; (d) For a more general review, see: R. B. N. Baig and R. S. Varma, Chem. Soc. Rev., 2011, 40, 1559–1584.
4. For recent work on processing of lignocellulosic material under microwave irradiation, see: J. Xu, J. Jiang, C. Hse and T. F. Shupe, Green Chem., 2012, 14, 2821–2830.
5. Z. V. Todres, Organic Mechanochemistry and its Practical Applications, Taylor & Francis, Boca Raton, 2006.
6. (a) J. J. Gilman, Science, 1996, 274, 65; (b) H. Watanabe, E. Matsui, Y. Ishiyama and M. Senna, Tetrahedron Lett., 2007, 48, 8132–8137.
7. (a) R. A. Kerr, Science, 2011, 331, 1510–1511; (b) D. J. Murphy and C. A. S. Hall, Ann. N. Y. Acad. Sci., 2011, 1219, 52–72; (c) A. Demirbas, Energy Convers. Manage., 2009, 50, 2239–2249; (d) G. H. Vogel, Chem. Eng. Technol., 2008, 31, 730–735.
8. C. Somerville, H. Youngs, C. Taylor, S. C. Davis and S. P. Long, Science, 2010, 329, 790–792.
9. (a) F. Cherubini and A. H. Stroemman, Energy Fuels, 2010, 24, 2657–2666; (b) E. S. Jensen, M. B. Peoples, R. M. Boddey, P. M. Gresshoff, H. Hauggaard-Nielsen, B. J. R. Alves and M. J. Morrison, Agron. Sustain. Dev., 2012, 32, 329–364; (c) C. Zhao and J. A. Lercher, Angew. Chem., Int. Ed., 2012, 51, 5935–5940.
10. H. Michel, Angew. Chem., Int. Ed., 2012, 51, 2516–2518.
11. A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek, J. Cairney, C. A. Eckert, W. J. Frederick Jr., J. P. Hallett, D. J. Leak, C. L. Liotta, J. R. Mielenz, R. Murphy, R. Templer and T. Tschaplinski, Science, 2006, 311, 484–489.
12. G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106, 4044–4098.
13. (a) J. B. Binder and R. T. Raines, J. Am. Chem. Soc., 2009, 131, 1979–1985; (b) F. Cherubini, Energy Convers. Manage., 2010, 51, 1412–1421; (c) J. C. Serrano–Ruiz, R. Luque and A. Sepúlveda–Escribano, Chem. Soc. Rev., 2011, 40, 5266–5281; (d) Y.-T. Cheng, J. Jae, J. Shi, W. Fan and G. W. Huber, Angew. Chem., Int. Ed., 2012, 51, 1387–1390.
14. J. Zakzeski, P. C. Bruijnincx, A. L. Jongerius and B. M. Weckhuysen, Chem. Rev., 2010, 110, 3552–3599.
15. M. E. Himmel, S.-Y. Ding, D. K. Johnson, W. S. Adney, M. R. Nimlos, J. W. Brady and T. D. Foust, Science, 2007, 315, 804–807.
16. (a) For the treatment of wood, see: A. Björkman, Nature, 1954, 174, 1057–1058; (b) For mechanochemical reactions of lignin model compounds, see: Z.-H. Wu, M. Sumimoto and H. Tanak, Holzforschung, 1994, 48, 395–399.
17. T.-Q. Yuan, S.-N. Sun, F. Xu and R.-C. Sun, J. Agric. Food Chem., 2011, 59, 10604–10614.
18. F. S. Chakar and A. Ragauskas, J. Ind. Crop. Prod., 2004, 20, 131–141.
19. (a) C. Somerville, S. Bauer, G. Brininstool, M. Facette, T. Hamann, J. Milne, E. Osborne, A. Paredez, S. Persson, T. Raab, S. Vorwerk and H. Youngs, Science, 2004, 306, 2206–2211; (b) H. Nimz, Angew. Chem., Int. Ed., 1974, 13, 313–321.
20. S. Rabe, M. Nachtegaal, T. Ulrich and F. Vogel, Angew. Chem., Int. Ed., 2010, 49, 6434–6437.
21. (a) S. Kawai, K. Okita, K. Sugishita, A. Tanaka and H. Ohashi, J. Wood Sci., 1999, 45, 440–443; (b) T. Kishimoto, Y. Uraki and M. Ubukata, Org. Biomol. Chem., 2008, 6, 2982–2987.
22. J. Buendia, J. Mottweiler and C. Bolm, Chem.–Eur. J., 2011, 17, 13877–13882.
23. (a) S. Son and F. D. Toste, Angew. Chem., Int. Ed., 2010, 49, 3791–3794; (b) J. M. Nichols, L. M. Bishop, R. G. Bergman and J. A. Ellman, J. Am. Chem. Soc., 2010, 132, 12554–12555; (c) S. K. Hanson, R. Wu and L. A. Silks, Angew. Chem., Int. Ed., 2012, 51, 3410–3413; (d) B. Sedai, C. Díaz–Urrutia, R. T. Baker, R. Wu, L. A. Silks and S. K. Hanson, ACS Catal., 2011, 1, 794–804; (e) S. K. Hanson, R. T. Baker, J. C. Gordon, B. L. Scott, A. D. Sutton and D. L. Thorn, J. Am. Chem. Soc., 2009, 131, 428–429; (f) P. Kelley, S. Lin, G. Edouard, M. W. Day and T. Agapie, J. Am. Chem. Soc., 2012, 134, 5480–5483; (g) S. K. Hanson, R. T. Baker, J. C. Gordon, B. L. Scott, L. A. Silks and D. L. Thorn, J. Am. Chem. Soc., 2010, 132, 17804–17816.
24. A. G. Sergeev and J. F. Hartwig, Science, 2011, 332, 439–443.
25. (a) C. Hagström-Näsi and E. Sjöström, J. Wood Chem. Technol., 1988, 8, 299–311; (b) J. E. Miller, L. Evans, A. Littlewolf and D. E. Trudell, Fuel, 1999, 78, 1363–1366; (c) S. Karagöz, T. Bhaskar, A. Muto and Y. Sakata, Fuel, 2004, 83, 2293–2299; (d) A. Vigneault, D. K. Johnson and E. Chornet, Can. J. Chem. Engin., 2007, 85, 906–916; (e) Y. Chen, R. R. Sharma-Shivappa, D. Keshwani and C. Chen, Appl. Biochem. Biotechnol., 2007, 142, 276–290; (f) S. Nenkova, T. Vasileva and K. Stanulov, Chem. Nat. Compd., 2008, 44, 182–185; (g) U. Wongiriwan, Y. Noda, C. Song, P. Prasassarakich and Y. Yeboah, Energy Fuels, 2010, 24, 3232–3228; (h) V. M. Roberts, V. Stein, T. Reiner, A. Lemonidou, X. Li and J. A. Lercher, Chem.–Eur. J., 2011, 17, 5939–5948.
26. For base cleavage studies involving lignin model compounds, see: (a) G. Miksche, Acta. Chem. Scand., 1973, 27, 1355–1368; (b) K. Poppius, Acta. Chem. Scand., 1984, 38, 611–616.
27. The amount of grinding auxiliary cannot be reduced on a small scale, because the grinding beaker needs a minimum filling for optimal performance. We assume that less can be used upon scale-up or by altering the reactor design.
28. (a) E. L. Clennan and G. L. Pan, Org. Lett., 2003, 5, 4979–4982; (b) A. Nagaki, E. Takizawa and I. Yoshida, J. Am. Chem. Soc., 2009, 131, 1654–1655; (c) V. Capriati, S. Florio and R. Luisi, Chem. Rev., 2008, 108, 1918–1942; (d) Y. Ma, Heteroat. Chem., 2002, 13, 310–315.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c2gc36456e

---