Investigation of free radicals and carbon structures in chars generated from pyrolysis of antibiotic fermentation residue

Abstract

The effect of pyrolysis on the chemical characteristics of Antibiotic Fermentation Residue (AFR) was studied in this paper. Electron Spin Resonance (ESR) spectrometry, X-ray photoelectron spectroscopy (XPS) and solid state ¹³C nuclear magnetic resonance (¹³C NMR) spectroscopy were applied to investigate the variation of free radicals and carbon structures. Results indicated that the free radical concentration (N_g) of AFR/chars changed dramatically during the pyrolysis process, first increasing and then decreasing. N_g reached 1.24 × 10¹⁹ spins per g in a N₂ atmosphere and 1.11 × 10¹⁹ spins per g in a CO₂ atmosphere at 320 °C. The g-values monotonically decreased, demonstrating that the chemical structures of free radicals changed during pyrolysis. Bond cleavage in methoxyl groups, aliphatic C–O bonds, aliphatic C–C bonds and C=O groups was enhanced with increasing pyrolysis temperature. These carbon structures converted to aromatic C–O bonds and aromatic C–C bonds through polymerization and condensation reactions. Compared with the CO₂ atmosphere, the N₂ atmosphere was more conducive to the production of free radicals and aromatization of carbon structures in chars during pyrolysis process.

---

1. Introduction

China is one of the largest producers of antibiotics in the world. 80% of cephalosporin antibiotics, 75% of penicillin industrial salt and 90% of streptomycin antibiotics are produced in China. As a kind of organic solid waste produced from the antibiotic ferment process, antibiotic fermentation residue (AFR) is mainly composed by mycelia, remaining substrate, intermediate metabolite and residual antibiotic,¹ and rich in lipid, polysaccharide and protein.² The residual antibiotic in AFR can easily accumulate in animals and human beings. Therefore AFR is identified as hazardous waste (HW02 Toxicity) on the China Hazardous Waste List.

Pyrolysis and incineration are effective ways for the treatment and resource recovery of biomass.^3–5 Jiang et al.⁶ investigated the pyrolysis and incineration characteristics by thermogravimetric analysis (TGA), and indicated that AFR addition would not affect the incineration system significantly. Du et al.^7,8 studied the thermal behavior and reaction characteristics of AFR mixed with coal and found interactions between AFR and coal in pyrolysis.

Intense free radicals reaction are happened during the pyrolysis process of biomass or fossil fuel, which could have significant effect on the types and quantities of pyrolysis products.⁹ Plenty of free radicals are created along with the homolytic cleavage of covalent bonds, and then formed new chemical bonds by combining with each other.^10–13 These two conversions occur sequentially and concurrently, and the products may undergo further pyrolysis to yield additional radical fragments, resulting in more coupling reactions.¹⁴ However, it is found that the lifetime of radicals varies greatly and could be as short as nanoseconds, which is too short to be detected.^14,15 Therefore, the radicals detected are only those with a long lifetime due to their poor mobility or inaccessibility by steric hindrance.^16,17 Liu et al.¹⁸ studied the variation of free radicals during coal pyrolysis, and the results indicated that free radicals concentration increased with increasing of temperature. Wang et al.¹⁹ pyrolyzed oil shale and characterized the properties of free radicals. The free radicals concentration in the products, such as semicoke, thermal bitumen and shale oil, greatly increased after pyrolysis. And g-values and linewidths of the products were also affected by temperature. He et al.¹⁶ pyrolysed walnut shell and corncob, and found that bio-tars also contained large amounts of radicals. The characteristics variation of free radicals revealed that the carbon structures of biomass changed during the pyrolysis process.^20–22 Qiao et al.²⁰ pyrolyzed chitin biomass and found that almost all the hydrocarbon functional groups vanished and converted into aromatic functional groups. Ben et al.²³ examined the carbon structures of kraft lignin and pine wood in slow pyrolysis process. The bonds breaking of aliphatic carbon function groups were enhanced and plenty of polyaromatic hydrocarbons with high molecular weight formed.

Although many works focused on the chemical characteristics of fossil fuels and biomass in pyrolysis process, less attention were paid to that of AFR, especially in free radicals and carbon structure characteristics change. In this study, Electron Spin Resonance (ESR) spectrometer was applied to investigate the free radicals characteristics of AFR, such as radicals concentration, g-value and linewidth. The X-ray photoelectron spectroscopy (XPS) and solid state ¹³C nuclear magnetic resonance (¹³C NMR) spectroscopy were employed to understand the carbon structures changes in AFR pyrolysis process. The pyrolysis experiments were carried out in two kinds of inert atmosphere (N₂ and CO₂ atmosphere), and the relationships between free radicals, carbon functional groups and temperature were evaluated.

2. Materials and methods

2.1. Sample preparation

The antibiotic fermentation residue (AFR) used in this study was produced from production process of terramycin. 50 g of AFR was dried in electricity heat drum wind drying oven (DGH-9023A, Pinggang, China) at 105 °C for 5 hours. The AFR sample was then grounded to fine particle, and its equivalent mean particle size was 46.68 μm analyzed by Malvern Laser Mastersizer (Mastersizer 3000, U.K.). The proximate and ultimate analysis of AFR were shown in Table 1.

Table 1 The proximate and ultimate analysis of AFR

Proximate analysis (wt%)		Ultimate analysis (wt%)
Moisture	2.57	C	44.54
Volatile	64.82	H	4.76
Ash	12.16	O	26.91
Fixed carbon	20.45	N	8.42
		S	0.64

---

2.2. Pyrolysis tests

The pyrolysis experiment was carried out in a tube furnace pyrolysis reactor (OTF-1200X-S50-SL, KJ Group, China) under the inert atmosphere of N₂ and CO₂, with a flow rate of 1000 mL min⁻¹. About 5 g AFR were heated from room temperature (25 °C) to 160, 320, 480 and 600 °C at the heating rate of 10 °C min⁻¹. After holding at the target temperature for 30 min, the reactor was cooled to room temperature at a cooling rate of 5 °C min⁻¹, and the chars were obtained to characterize their free radicals and carbon structures.

2.3. Electron Spin Resonance (ESR) spectrometer

The ESR was applied to investigated the free radicals characteristics of AFR samples and chars. The ESR experiments were carried out with a JES-FA 200 spectrometer (JEOL, Japan). ESR scan parameters during all the experiments were as follows: modulation frequency, 100 kHz, X-band; modulation amplitude, 100 Hz; microwave frequency, about 9070 MHz; microwave power, 0.0997 mW; central magnetic field, 324.254 mT; sweep width, 10 mT; time constant, 0.1 s and sweep time, 1 min. All observations were accomplished at room temperature (25 °C) and atmospheric pressure. 1,1-Diphenyl-2-picrylhydrazyl (DPPH) was used as standard sample to calculate the radical concentration in AFR and chars.¹⁶ And g-value and linewidth were extracted through the ESR analysis by JEOL computer software CW-ESR.

The free radicals concentration (N_g, spins per g) in samples were calculated by the developed standard curve of DPPH. The amount of free radicals in DPPH (R_g) with a certain quality was obtained by the following equation:

where m_(DPPH) is the mass of the mixture of DPPH and anhydrite calcium sulfate, g; ω_(DPPH) is the percentage of DPPH, wt%; M_(DPPH) is the molecular mass of DPPH (394.32 g mol⁻¹). The ESR spectra of DPPH samples are shown in Fig. S1.† The relationship of N_g(DPPH) versus S_(DPPH) (area of ESR spectra curve) based on numerical analysis is given by the following:

S_(DPPH) = 1834.05R_(DPPH) × 10⁻¹⁵ − 400.01 (2)

The N_g in samples can be calculated by eqn (3). S_(sample) is the area of the sample absorption curve; m_(sample) is the mass of mixture of sample.

2.4. X-ray photoelectron spectroscopy (XPS)

The XPS experiments were carried out with an AXIS ULTRA DLD system (Kratos, Japan). C 1s spectra were measured with the following scan parameters: source of X-ray, monochromatic Al Kα (150 W, hν = 1486.6 eV); pass energy, 20 eV; analysis area, 800 × 800 μm; resolution, 0.05 eV.

2.5. Solid state ¹³C nuclear magnetic resonance (¹³C NMR) spectroscopy

The solid state ¹³C NMR was employed to understand the carbon structures of AFR sample and chars. All the experiments were performed in a ¹³C solid state CP/MAS/TOSS NMR spectrometer (Advance 400 MHz, Bruker, Germany) with a frequency of 100.63 MHz and rotor spun of 5 kHz. The contact time and pulse repetition delay of ¹³C-¹H CP were 1 ms and 4 s respectively.

3. Results and discussion

3.1. The variation of free radicals characteristics during pyrolysis process

3.1.1 ESR spectra analysis and free radicals concentration changes. The ESR spectra of AFR during pyrolysis process were shown in Fig. 1. It was observed that all of the spectra were symmetrical curves without hyperfine structure. The peak intensities, wave linewidths and locations of ESR spectra during pyrolysis process changed significantly both in N₂ and CO₂ atmosphere, which indicated that pyrolysis temperature affected the formation, distribution and species of free radicals dramatically. The covalent bonds in AFR broke when thermochemical reaction happened during pyrolysis process, leading to the formation and coupling of free radicals. These results were consistent with Wang et al.,¹⁹ who also observed significant changes of ESR spectra in oil shale pyrolysis.

Fig. 1 The ESR spectra of AFR during pyrolysis process. (a) N₂ atmosphere, (b) CO₂ atmosphere.

The changes in the free radical concentration (N_g) during AFR pyrolysis are shown in Fig. 2. Dramatic changes of N_g were observed during pyrolysis process. A moderate increasing of free radicals were obtained over the range 25–160 °C. Desorption of moisture and release of light gases took place in this temperature range. Cleavage of chemical bonds, which characterizes the decomposition of organic matter in the AFR, was at its starting stage, with only a trace amount of free radicals being produced. When the AFR samples were heated to 320 °C, N_g increased sharply. A significant proportion of organic components were decomposed. The weak bonds in these components such as hydroxyl, carbonyl, alkyl, ether, amino and some side chain groups were ruptured.^20,24 A large amount of active free radicals formed in chars with the releasing of gaseous species including CO, CO₂, NO_x, HCN, NH₃ and hydrocarbon gases.²⁰ In the subsequent bonding process, a portion of active free radicals vanished due to condensation reaction among themselves, and others were reserved through hydrogen extraction or combining with stable fragments in the solid residues. As a result, plenty of stable free radicals formed. N_g of the stable free radicals at 320 °C reached 1.24 × 10¹⁹ spins per g in N₂ atmosphere and 1.11 × 10¹⁹ spins per g in CO₂ atmosphere. The free radicals in the AFR reduced dramatically in the range 320–600 °C. The carbonization degree of AFR improved significantly with the forming of multi-ring aromatic compounds during aromatization reaction. Meanwhile, the coupling of unpaired electrons were taken place in the reaction, and N_g decreased. As a result, the free radicals dropped to a lower level in concentration, about 4 × 10¹⁸ spins per g (both in N₂ and CO₂ atmosphere) at 600 °C.

Fig. 2 Variation of N_g with AFR pyrolysis temperature.

Basically, N_g of AFR in N₂ atmosphere were larger than in CO₂ atmosphere during pyrolysis process. Probably because large amounts of CO₂ was adsorbed on chars and involved in cross-linking reactions, reduced the particle swelling and inhibited the volatile release.²⁵ On the other hand, N₂ had lower thermal capacities and higher mass diffusivities compared to CO₂. More thermal energy could be exerted on the AFR particles in the N₂ atmosphere.²⁶ Therefore more organic components were thermal decomposed with the releasing of volatile matter. Bonds cracking were susceptible to occur in N₂ atmosphere,^18,19 forming more free radicals in the chars.

3.1.2 Analysis of g-value. The g-value reflects the interaction between spin motion and orbital motion of electrons in paramagnetic center. The g-value is determined by the chemical environments of unpaired electrons, and has been used to characterize interior-molecular structures. Base on the different g-values, the radicals could be classified as σ-type oxygen-containing radicals (2.0008–2.0014), graphite radicals (2.0015–2.0025), π radicals on aliphatic hydrocarbons (2.0025–2.0026), π radicals on aromatic hydrocarbons (2.0025–2.00291), π-type oxygen-containing radicals (2.0035–2.00469), nitrogen-containing radicals (2.0031) and sulfur-containing radicals (2.0080–2.0081).²⁷ The g-values of AFR and chars during pyrolysis in different atmosphere (N₂ and CO₂) were shown in Fig. 3.

Fig. 3 Variation of the g-values with AFR pyrolysis temperature.

The g-values revealed the superposition of various free radicals types. The g-value of AFR sample was 2.00339 at 25 °C, implying that most of the free radicals might be π radicals on aliphatic/aromatic hydrocarbons, π-type oxygen-containing radicals and nitrogen-containing radicals considering the chemical component in AFR (Table 1).^19,28 Meanwhile, some unpaired electrons with spins are localized on heteroatom sites, such as S (0.64% in the sample mass), which had strong spin–orbit coupling.²⁹ This could be another reason for AFR sample to reach a high g-value. With the increasing of temperature, the g-value decreased monotonically. The decrease in the g-value was minor below 160 °C. There were slight bonds cracking in AFR with the desorption of combined water and light gases. Few changes of free radicals content and species were taken place. As a result, the g-values of chars were similar to original sample at 160 °C. When AFR samples were heated up to 600 °C, g-values of chars decreased sharply. On the one hand, the quantities of certain atoms such as O, N, S decreased with the releasing of volatile matter, resulting in the decrease of π-type oxygen-containing radicals, nitrogen-containing radicals and sulfur-containing radicals. On the other hand, with the increase of pyrolysis temperature, the aromatization of chars became more significant, and more simple aromatic and even multi-ring aromatic structures were formed. Large multi-ring aromatic components increased until a structure resembling graphite is approached, as reported by Petrakis.²⁷ The formation of these graphite-like radicals were considered to be another reason for the decrease of g-values in chars.

The species of free radicals showed similar variation trends both in N₂ atmosphere and CO₂ atmosphere. This was because that the pyrolysis reactivity in N₂ atmosphere was similar to CO₂ atmosphere at low temperatures (lower than 700 °C).^30–32 When the pyrolysis temperature was higher than 320 °C, the g-values in N₂ atmosphere were smaller than in CO₂ atmosphere. One possible reason was that the reactions of bonds cleavage were more active in N₂ atmosphere than CO₂ atmosphere. The deoxygenation and dehydrogenation mechanisms on chars were obviously exhibited, and the reactions in N₂ were more pronounced than in CO₂.³³ Thus, more oxygen-containing components and hydrocarbon components decomposed, resulting in a lower g-value in N₂ atmosphere.

3.1.3 Analysis of linewidth. The linewidth reflects the relaxation time of spinning electrons, which is used to measure the interactions between unpaired electrons and their chemical environment. The variation of linewidths during AFR pyrolysis process were shown in Fig. 4. All the linewidths fell in the range of 0.5–0.8 mT, in the same order of magnitude as the results of Wang et al.¹⁹ obtained from pyrolysis of oil shale. When the AFR was heated from 25 °C up to 480 °C, the linewidths increased monotonically in the N₂ and CO₂ atmosphere. The condensation reaction in the remaining solid became intensified with the releasing of volatile matter. A large number of simple aromatic structures produced and the dipolar spin–spin interaction between unpaired electrons became stronger, which was responsible for the linewidths broadening.²⁸ It was observed that linewidths were broader in N₂ atmosphere compared with in CO₂ atmosphere. This might due to less simple aromatic structures formed in CO₂ atmosphere compared with N₂ atmosphere.³⁴ And the dipolar spin–spin interaction between unpaired electrons became weaker, led to the linewidth narrower in CO₂ atmosphere.

Fig. 4 Variation of the linewidths with AFR pyrolysis temperature.

When AFR were gradually heated up to 600 °C, the structure of aromatic rings in chars was mainly peri-condensed during the carbonization process, and large amount of multi-ring aromatic components or even larger aromatic ring clusters were created. The spin-lattice interactions of paramagnetic centers in chars became stronger with the increase of multi-ring aromatic units.³⁵ Thus, the relaxation time of unpaired electrons increased, which made the linewidths narrower. The linewidths of chars in N₂ atmosphere was narrower at 600 °C than in CO₂ atmosphere. That was because the chemisorption of CO₂ on the nascent char surface could promote the thermal cracking of aromatic networks. The multi-ring aromatic structures, which could narrow linewidth through enhance the spin–lattice interactions, formed in CO₂ atmosphere were less than N₂ atmosphere.^36,37

3.2. Analysis of C element by XPS

The chemical nature of the C element on surface of AFR and chars during pyrolysis process were analyzed by XPS (C 1s). XPS spectra and deconvolution results were shown in Fig. 5 and 6. The deconvoluted peaks at 284.75, 285.3, 286.3, 287.5, 289.0 ± 0.2 eV represented C₁ (C–C or C–H bonds), C₂ (carbon bound to nitrogen, C–N), C₃ (carbon connect with hydroxyl or ether, C–O/C–OH), C₄ (carbon bound to oxygen with two bonds, C=O/O–C–O), C₅ (carbon bound to oxygen with three bonds, O–C=O), respectively.^2,38,39 C_O was the sum of C₃, C₄ and C₅, which meant the total groups of carbon bounding to oxygen in samples.

Fig. 5 XPS spectra for the C 1s of AFR during pyrolysis process. (a) N₂ atmosphere (b) CO₂ atmosphere.

Fig. 6 Carbon functional groups on surface of AFR and chars.

The ratio of C₁–C₅ dramatic changed dramatically at the temperature range 320–600 °C after a steady phase (25–160 °C). C₁ increased substantially to about 78% at 600 °C from 57% at room temperature (both in N₂ and CO₂ atmosphere). It indicated that the aromatization reactions occurred during pyrolysis, and led to the increase of the C–C bonds in aromatic ring system. Nitrogen containing components in AFR, such as amine and protein, decomposed during pyrolysis process and released NH₃, HCN and NO_x.⁴⁰ Thus, the percentage of C₂ in chars decreased from 7% to 2%. In addition, the percentage of C_O decreased from about 36% to 20% with the increase of temperature, which was mainly attributed to the bond cleavage reaction of C₄. The proportion of C₃ changed slightly, from 13% to 15%. The proportion of C₅ was at a low level with a percentage range of 0–3%, and had little impact on the percentage of C_O. Unpaired electrons located in carbon functional groups mentioned above were the main source of free radicals in raw AFR. The variation of C₁, C₂ and C_O during pyrolysis process would have significant effect on the characteristics of free radicals in chars, which were consistent with the variation of g-values.

3.3. Conversion of carbon structures analyzed by solid state ¹³C NMR spectra

3.3.1 Analysis of ¹³C NMR spectra. The ¹³C NMR spectra of AFR sample (25 °C) and chars (pyrolyzed under 320 °C, 480 °C, 600 °C, respectively) were measured to characterize the evolution of carbon structures during pyrolysis process in detail (Fig. 7).

Fig. 7 The ¹³C NMR spectra of AFR and chars. (a) N₂ atmosphere, (b) CO₂ atmosphere.

The carbon structures showed the same variation in N₂ and CO₂ atmosphere. There were distinct characteristic peaks of aliphatic carbon functional groups (chemical shift range 0–95.8 ppm) and carbonyl or carboxyl groups (166.5–215.0 ppm) in AFR sample when at room temperature (25 °C), which might be attributed to that the main components of AFR were crude lipid, protein and polysaccharide. With increasing of temperature, the peak intensity of aliphatic carbon functional groups and carbonyl groups decreased significantly, and almost disappeared at 600 °C. In contrast, the characteristic peaks of aromatic carbon (95.8–166.5 ppm) continuously increased with increasing pyrolysis temperature. This indicated that the species of carbon functional groups had obvious changes during pyrolysis process.

3.3.2 Conversion of carbon structures during pyrolysis process. The ¹³C NMR spectra of AFR and chars in this study were analyzed by curve fitting resolution.²⁸ The spectra were well fitted by Gaussian and Lorentzian lines with good correlation factor (R² ≥ 0.99). According to previous studies,^41–43 carbon structures in biomass mainly include aliphatic C–C bonds (0–55.2 ppm), methoxyl groups (55.2–60.8 ppm), aliphatic C–O bonds (60.8–95.8 ppm), aromatic C–H bonds (95.8–125.0 ppm), aromatic C–C bonds (125.0–142.0 ppm), aromatic C–O bonds (142.0–166.5 ppm) and C=O bonds (166.5–215.0 ppm). Based on the chemical shift ranges of carbon functional groups, the percentage of carbon functional groups in AFR and chars were summarized by calculating the share of certain fitted curves area in total fitted curves area, as shown in Fig. 8.

Fig. 8 Variation of carbon structures in AFR and chars. (a) N₂ atmosphere, (b) CO₂ atmosphere.

The total percentage of aliphatic C–C bonds, methoxyl groups and aliphatic C–O bonds in AFR (25 °C) was as high as 51.88%. The results above indicated that aliphatic structure was dominant in raw AFR carbon structures. In addition, the percentage of C=O bonds was 24.67%, which might mainly contributed by ester groups (–CO–OR) and amide groups (–CO–NH–) from lipid and protein in AFR. During pyrolysis, the percentage of different carbon functional groups changed obviously. The dominant carbon functional groups mentioned above decreased with the enhancing of pyrolysis degree. Meanwhile, the aromatic carbon functional groups, such as aromatic C–C bonds and aromatic C–O bonds, formed rapidly.

With pyrolysis temperature increasing from 25 °C to 320 °C, the percentage of methoxyl groups decreased from 10.65% to 2.58% (N₂ atmosphere) and 4.08% (CO₂ atmosphere), while the percentage of aliphatic C–O bonds decreased from 10.14% to 4.91% (N₂ atmosphere) and 5.31% (CO₂ atmosphere). This indicated that side chain groups, such as methoxyl groups and aliphatic C–O bonds, were susceptible to cracking by thermal decomposition, and the reaction was more active in N₂ atmosphere. These results were consistent with the ESR analysis in Section 3.1.1. With the further increasing of pyrolysis temperature, the percentage of methoxyl groups and aliphatic C–O bonds continued to decrease. Especially methoxyl groups, almost disappeared at 600 °C. By contrast, the significant decrease of the aliphatic C–C bonds and C=O bonds percentage happened when temperature higher than 480 °C, that was because the bond-breaking reactions of aliphatic C–C bonds and C=O bonds needed more energy.

The percentage of aromatic C–C bonds and aromatic C–O bonds increased with the increasing pyrolysis temperature, while the ratio of aromatic C–H bonds showed a contrary tendency. The increasing of aromatic C–C bonds and aromatic C–O bonds could be mainly attributed to the pyrolysis condensation reaction of aliphatic carbon functional groups, methoxyl groups and C=O bonds.²⁰ In addition, the bonds cracking in aliphatic/aromatic hydroxyl groups produced oxygen-containing free radicals and π radicals on hydrocarbons, and subsequently formed aromatic C–O bonds during pyrolysis.^44,45 The decrease of aromatic C–H bonds were due to its consumption in radical reactions which happened between aromatic C–H bonds and methoxyl groups. Meanwhile, more aromatic C–C bonds and aromatic C–O bonds were created. When AFR were pyrolyzed at 600 °C, the vast majority of carbon structures in chars converted into aromatic carbon functional groups which had high carbonization degree. Ben et al.^23,46 studied the conversion of carbon functional groups in kraft lignin pyrolysis process, and obtained the similar results. The ratio of aromatic carbon functional groups reached 94.14% in N₂ atmosphere and 90.13% in CO₂ atmosphere at 600 °C. This indicated that N₂ atmosphere was more beneficial to the reaction of AFR pyrolysis, resulting in a higher aromatization level of carbon structure in chars.

4. Conclusions

Pyrolysis process had significant influence on the AFR chemical properties such as free radicals and carbon structures. More free radicals and aromatic carbon functional groups were created in N₂ atmosphere compared with CO₂ atmosphere.

As the pyrolysis temperature increased, the degree of aromatization of carbon structures in the chars increased due to intensified condensation reaction, causing dramatic changes in the composition of free radicals.

The contributions of methoxyl groups, aliphatic C–O bonds, aliphatic C–C bonds and C=O bonds decreased substantially in the free radicals composition, while that of aromatic C–C bonds and C–O bonds increased significantly with increasing pyrolysis temperature.

Conflict of interest

No potential conflict of interest was reported by the authors.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant No. 51508553), Beijing Training Project For The Leading Talents in S & T (Grant No. LJ201620), and China Postdoctoral Science Foundation Funded Project (Grant No. 2016M591266).

References

1. C. Li, G. Zhang, Z. Zhang, D. Ma, L. Wang and G. Xu, Chem. Eng. J., 2015, 279, 530–537.
2. D. Ma, G. Zhang, P. Zhao, C. Areeprasert, Y. Shen, K. Yoshikawa and G. Xu, Chem. Eng. J., 2015, 273, 147–155.
3. Y. Wang, S. De and N. Yan, Chem. Commun., 2016, 52, 6210–6224.
4. N. Yan and X. Chen, Nature, 2015, 524, 155–157.
5. Y. Gao, X. C. Chen, J. Zhang and N. Yan, ChemPlusChem, 2015, 80, 1556–1564.
6. X. Jiang, Y. Feng, G. Lv, Y. Du, D. Qin, X. Li, Y. Chi, J. Yan and X. Liu, Environ. Sci. Technol., 2012, 46, 13539–13544.
7. Y. Du, X. Jiang, X. Ma, X. Liu, G. Lv, Y. Jin, F. Wang, Y. Chi and J. Yan, Energy Fuels, 2013, 27, 6295–6303.
8. Y. Du, X. Jiang, G. Lv, X. Ma, Y. Jin, F. Wang, Y. Chi and J. Yan, Energy Convers. Manage., 2014, 88, 459–463.
9. J. Zhan, R. Wu, X. Liu, S. Gao and G. Xu, Fuel, 2014, 134, 283–292.
10. C. Baehrle, V. Custodis, G. Jeschke, J. A. van Bokhoven and F. Vogel, ChemSusChem, 2014, 7, 2022–2029.
11. K. H. Kim, X. Bai, S. Cady, P. Gable and R. C. Brown, ChemSusChem, 2015, 8, 894–900.
12. J. Kibet, L. Khachatryan and B. Dellinger, Environ. Sci. Technol., 2012, 46, 12994–13001.
13. C. Amen-Chen, H. Pakdel and C. Roy, Bioresour. Technol., 2001, 79, 277–299.
14. M. Liu, J. Yang, Z. Liu, W. He, Q. Liu, Y. Li and Y. Yang, Energy Fuels, 2015, 29, 5773–5780.
15. R. V. Hedegaard, L. Liu and L. H. Skibsted, Food Chem., 2015, 167, 185–190.
16. W. He, Q. Liu, L. Shi, Z. Liu, D. Ci, C. Lievens, X. Guo and M. Liu, Bioresour. Technol., 2014, 156, 372–375.
17. W. He, Z. Liu, Q. Liu, D. Ci, C. Lievens and X. Guo, Fuel, 2014, 134, 375–380.
18. J. Liu, X. Jiang, J. Shen and H. Zhang, Powder Technol., 2015, 272, 64–74.
19. W. Wang, Y. Ma, S. Li, J. Shi and J. Teng, Energy Fuels, 2016, 30, 830–834.
20. Y. Qiao, S. Chen, Y. Liu, H. Sun, S. Jia, J. Shi, C. M. Pedersen, Y. Wang and X. Hou, Carbohydr. Polym., 2015, 133, 163–170.
21. Y. Wang, C. M. Pedersen, T. Deng, Y. Qiao and X. Hou, Bioresour. Technol., 2013, 143, 384–390.
22. L. Sanchez-Silva, D. Lopez-Gonzalez, J. Villasenor, P. Sanchez and J. L. Valverde, Bioresour. Technol., 2012, 109, 163–172.
23. H. Ben and A. J. Ragauskas, Bioresour. Technol., 2013, 147, 577–584.
24. M. L. Poutsma, Energy Fuels, 1990, 4, 113–131.
25. C. R. Shaddix and A. Molina, Proc. Combust. Inst., 2009, 32, 2091–2098.
26. A. Molina and C. R. Shaddix, Proc. Combust. Inst., 2007, 31, 1905–1912.
27. L. Petrakis and D. W. Grandy, Anal. Chem., 1978, 50, 303–308.
28. J. Liu, X. Jiang, J. Shen and H. Zhang, Adv. Powder Technol., 2014, 25, 916–925.
29. M. Korkmaz and T. Özbey, Fuel, 1991, 70, 789–793.
30. K. Li, C. Zhu, L. Zhang and X. Zhu, Bioresour. Technol., 2016, 209, 142–147.
31. C. A. Wang, X. Zhang, Y. Liu and D. Che, Appl. Energy, 2012, 97, 264–273.
32. R. K. Rathnam, L. K. Elliott, T. F. Wall, Y. Liu and B. Moghtaderi, Fuel Process. Technol., 2009, 90, 797–802.
33. W. Chen and B. Lin, Energy, 2016, 94, 569–578.
34. Y. Bai, Y. Wang, S. Zhu, F. Li and K. Xie, Energy, 2014, 74, 464–470.
35. B. Pilawa, A. B. Więckowski, H. Wachowska and M. Kozłowski, Fuel, 1998, 77, 1561–1567.
36. L. Chang, Z. Xie, K. C. Xie, K. C. Pratt, J. I. Hayashi, T. Chiba and C. Z. Li, Fuel, 2003, 82, 1159–1166.
37. S. P. Gao, J. T. Zhao, Z. Q. Wang, J. F. Wang, Y. T. Fang and J. J. Huang, J. Fuel Chem. Technol., 2013, 41, 257–264.
38. B. Q. Liao, H. J. Lin, S. P. Langevin, W. J. Gao and G. G. Leppard, Water Res., 2011, 45, 509–520.
39. B. Pilawa, A. B. Wieckowski, R. Pietrzak and H. Wachowska, Fuel, 2002, 81, 1925–1931.
40. X. Zhu, S. Yang, L. Wang, Y. Liu, F. Qian, W. Yao, S. Zhang and J. Chen, Environ. Pollut., 2016, 211, 20–27.
41. M. Nagy, M. Kosa, H. Theliander and A. J. Ragauskas, Green Chem., 2010, 12, 31–34.
42. P. Sannigrahi, A. J. Ragauskas and S. J. Miller, Energy Fuels, 2009, 24, 683–689.
43. R. Samuel, Y. Pu, B. Raman and A. J. Ragauskas, Appl. Biochem. Biotechnol., 2010, 162, 62–74.
44. W. J. DeSisto, N. I. Hill, S. H. Beis, S. Mukkamala, J. Joseph, C. Baker, T. Ong, E. A. Stemmler, M. C. Wheeler, B. G. Frederick and A. van Heiningen, Energy Fuels, 2010, 24, 2642–2651.
45. J. M. Younker, B. Ariana and A. C. Buchanan, ChemPhysChem, 2011, 12, 3556–3565.
46. H. Ben and A. J. Ragauskas, Energy Fuels, 2011, 25, 5791–5801.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra22192k

---