Influence of solvent on upgrading of phenolic compounds in pyrolysis bio-oil

Abstract

Solvents play a pivotal role in many chemical reactions as well as in upgrading of pyrolysis bio-oil. Herein we report a systematic study of the effects of solvents such as water, methanol, ethanol, propanol, butanol, acetone, ethyl acetate, tetrahydrofuran and hexane on hydrogenation of a lignin-derived component of pyrolysis bio-oil, phenol, over activated carbon supported palladium catalysts. The solvents were classified into four groups so as to better understand the effect of them. For hydrogen bond donor–hydrogen bond acceptor (HBD–HBA) solvents and hydrogen bond acceptor (HBA) solvents, the conversion of phenol decreased with increasing polarity/polarizability π*. Phenol was converted completely in hexane or water at 250 °C. However, in methanol or ethanol, it was converted partially. Synergistic action of multiple factors had an effect on the hydrogenation of phenol. To convert phenol better, water and hexane as solvents were excellent in upgrading of pyrolysis bio-oil.

---

1. Introduction

Pyrolysis bio-oil is considered to be a promising alternative liquid transportation fuel. It contains many components such as water, acids, alcohols, ethers, ketones, esters, aldehydes, phenols, saccharides, hydrocarbons, etc.¹ Bio-oil cannot be directly used as a transportation fuel without prior upgrading. Usually hydrotreating is used in upgrading pyrolysis bio-oil to improve the quality.^2–5

In hydrogenation reactions the solvent plays a pivotal role and have dramatic effect on rate and selectivity.^6,7 Zhang et al.⁸ investigated hydrogenation of cinnamaldehyde using Pd/C catalyst, and reported that the main products were hydrocinnamaldehyde, 3-phenyl propanol, and propyl benzene in non-polar solvents, while in polar solvents besides the products above additional cinnamaldehyde diacetal and hydrocinnamaldehyde diacetal were formed. Bennett et al.⁹ studied hydrogenation of 2-pentyne over Pd/Al₂O₃ and reported that the reaction in heptane was faster than in isopropanol due to increased substrate solubility. And the rate was intermediate in the solvent of 50/50 heptane/isopropanol, but both cis/trans pentene ratio and pentene/pentane ratio were the biggest. Akpa et al.⁷ studied solvent effects on hydrogenation of 2-butanone over Ru/SiO₂ catalyst and showed that the hydrogenation reaction had dramatically higher initial rate in water than in isopropyl alcohol, methanol, and heptane. Hronec et al.¹⁰ studied hydrogenation of furfural and furfuryl alcohol over Pt, Ru and Ni catalysts and reported that in water main product was cyclopentanone by the rearrangement of furfural and furfuryl alcohol, while in alcohols, the furan ring rearrangement did not proceed.

Selective hydrogenation of phenol to cyclohexanone, an important chemical feedstock, attracted more attention for its unique efficiency in gas phase^11–13 or liquid phase,^14,15 especially in green solvent water.^16–18 However, the role of solvent is not extensively studied and is not clear.

Herein we report a systematic study of solvent effects on the hydrogenation of phenol, a typical lignin-derived component of bio-oil, over activated carbon supported palladium catalyst. Because phenolic compounds are present with relatively big amount, the hydrogenation of phenol as a model is important guidance for upgrading pyrolysis bio-oil. Taking into account components of pyrolysis bio-oil, in this work we selected four groups of solvents such as water, hydrogen bond donor–hydrogen bond acceptor (HBD–HBA) solvents (methanol, ethanol, propanol and butanol), hydrogen bond acceptor (HBA) solvents (acetone, ethyl acetate and tetrahydrofuran), and hydrocarbon solvents (hexane). The relationships of reactants, products, solvents and catalysts are also investigated to understand the catalytic process. Results described herein may give researchers some references in the selection of solvents appropriate for upgrading of pyrolysis bio-oil and producing valuable chemicals.

2. Experimental

2.1. Materials

Pd/C (5%) catalyst was purchased from Aladdin. All of the analytical reagent grade chemicals were purchased from commercial suppliers: phenol (Aladdin); methanol, ethanol, propanol, ethyl acetate and tetrahydrofuran (Chengdu Kelong Chemical Reagent Company); butanol, acetone and hexane (Sinopharm Chemical Reagent Co., Ltd). H₂ (99.999%) was supplied by Hangzhou Jingong Specialty Gases Co., Ltd.

2.2. Phenol hydrogenation

The hydrogenation of phenol over Pd/C was carried out in various solvents in a 25 mL stainless steel autoclave. In a typical procedure for hydrogenation, phenol (0.250 g, 0.00266 mol), Pd/C (0.010 g) and solvent (10 mL) were added into the autoclave. Then, the autoclave was pressured to 1.0 MPa with hydrogen after being purged with hydrogen for 10 times to remove the air, and was heated to react at object temperature for 1 h with stirring at 800 rpm. After cooling to ambient temperature in ice water, products were analyzed on a gas chromatography (GC, Agilent 6820) with a flame ionization detector (FID) and a DB-WAX capillary column (30 m × 0.32 mm × 0.25 μm). Identification of the products was performed on a gas chromatography-mass spectrometry (GC-MS, Agilent 6890/5973).

2.3. Cyclohexanone hydrogenation

The hydrogenation of cyclohexanone on Pd/C was carried out as the hydrogenation of phenol.

2.4. Infrared (IR) spectroscopy

Fourier transform infrared (FTIR) spectra were recorded on a spectrometer (Bruker Vertex 70) in the range of 4000–650 cm⁻¹ with a spectral resolution of 4 cm⁻¹. Prior to measurement, a droplet of the sample was placed between two CaF₂ thin wafers. Spectra were recorded in air with an average of 64 scans.

3. Results and discussion

3.1. Phenol hydrogenation at different conditions

The hydrogenation of phenol over Pd/C at 120 °C and 150 °C are summarized in Table 1. Obviously, higher temperature led to higher conversion of phenol. For example, the conversion was 100% in hexane at 150 °C while it was 76.7% at 120 °C. At lower temperature the cyclohexanol was formed only in water and hexane. In order to study the reaction mechanism, higher reaction temperature and hydrogen pressure and longer reaction time were employed, and results were summarized in Table 2. At 250 °C phenol was converted completely in water, ethyl acetate, tetrahydrofuran and hexane, meanwhile only part of phenol was converted to cyclohexanone and cyclohexanol except in water, tetrahydrofuran and hexane. From the results in Tables 1 and 2 it was derived that phenol was hydrogenated to cyclohexanone firstly and then to cyclohexanol by the subsequent hydrogenation. Scheme 1 shows the reaction pathway for hydrogenation of phenol on Pd/C at different temperature and it is in accord with the mechanism of hydrogenation on supported Pd catalysts.^13,19

Table 1 Hydrogenation conversion of phenol over Pd/C at different temperature

Solvent	Phenol conversion at different temperature/%
	100 °C	120 °C	150 °C
Water	17.4	46.5	87.0
Methanol	2.8	3.8	6.2
Ethanol	3.7	5.8	9.7
Propanol	10.2	17.5	35.3
Butanol	11.2	18.8	38.5
Acetone	—	4.9	6.8
Ethyl acetate	—	31.2	64.8
Tetrahydrofuran	—	8.4	12.1
Hexane	—	76.7	100

---

Table 2 Solvent effects on the hydrogenation of phenol over Pd/C at 250 °C

Solvent	Phenol conversion/%	Selectivity C=O/%	Selectivity C–OH/%	Selectivity others/%
Water	100	5.7	94.3	0
Methanol	28.6	77.2	11.9	10.9
Ethanol	69.3	63.2	27.8	9.0
Propanol	96.0	57.6	34.2	8.2
Butanol	97.4	48.6	43.9	7.5
Acetone	79.6	38.6	57.9	3.5
Ethyl acetate	100	1.8	68.2	30.0
Tetrahydrofuran	100	8.0	92.0	0
Hexane	100	15.1	84.9	0

---

Scheme 1 Reaction pathway for hydrogenation of phenol on Pd/C at different temperature.

3.2. The effect of solvents

Solvents play a crucial role in many chemical reactions. Because of complicated physical and chemical properties of solvents, it is very difficult to evaluate all of solvents by single criterion. The solvents used here are hence classified into four groups: (1) water; (2) hydrogen bond donor–hydrogen bond acceptor (HBD–HBA) solvents, such as methanol (MeOH), ethanol (EtOH), propanol (PrOH) and butanol (BuOH); (3) hydrogen bond acceptor (HBA) solvents, such as acetone (Ace), ethyl acetate (EA) and tetrahydrofuran (THF); (4) hydrocarbon solvents, such as hexane (Hex). Some parameters of solvents such as HBD ability α,²⁰ HBA ability β,²⁰ polarity/polarizability π* (ref. 20) and polarity E^T_N (ref. 21) are listed in Table 3.

Table 3 Some parameters of solvents and phenol

Solvent	HBD ability α	HBA ability β	Polarity/polarizability π*	Polarity E^TN
Water	1.17	0.47	1.09	1.000
Methanol	0.98	0.66	0.60	0.762
Ethanol	0.86	0.75	0.54	0.654
Propanol	0.84	0.90	0.52	0.617
Butanol	0.84	0.84	0.47	0.586
Acetone	0.08	0.43	0.71	0.355
Ethyl acetate	0	0.45	0.55	0.228
Tetrahydrofuran	0	0.55	0.58	0.207
Hexane	0	0.00	−0.04	0.009
Phenol	1.65	0.30	0.72	—

---

From Table 3 it can be seen that water shows the highest polarity and displays the highest HBD ability. Hexane shows the lowest polarity and display no HBD ability and HBA ability. The HBD–HBA solvents show higher polarity and display higher HBD ability. The HBA solvents show lower polarity and HBA ability than that of HBD–HBA solvents, and display almost no HBD ability. These properties of solvents are correlated with the catalytic activity in the hydrogenation of phenol.

3.2.1 Water. Results summarized in Tables 1 and 2 show that solvents play a crucial role in hydrogenation of phenol. Water is an excellent solvent for this reaction.

The solvent may affect the adsorption strength of reactants and products on the catalyst surface. Because of the immiscibility with water cyclohexanone formed during hydrogenation of phenol may not quickly desorb from the catalyst surface and hence may be further hydrogenated to cyclohexanol. On the contrary, cyclohexanone could desorb easily for good solubility in other solvents, as a result further hydrogenation to cyclohexanol is hindered. Density functional theory (DFT) calculation results show that water can reduce the binding energy (20–25%) between phloroglucinol molecule and Pt(111) and Pd(111) surfaces in comparison to that in gas phase.²² It is plausible that the binding energy between phenol and metal surface decreases because of the formation of hydrogen bond between phenol and water. Moreover, water can lower the activation barrier as well as increase the proton diffusion coefficient.^7,23 As a consequence, water promotes the hydrogenation of phenol.

3.2.2 HBD–HBA solvents. As summarized in Table 1, the conversion is highest in butanol, followed by propanol, ethanol and methanol among HBD–HBA solvents at 150 °C. The same order is also present at 120 °C. Obviously methanol and ethanol inhibit hydrogenation of both phenol and cyclohexanone, also as seen in Table 2. That may result from strong interactions between methanol or ethanol and phenol and cyclohexanone.

Hydrogen solubility in the solvent may play a crucial role in catalytic reaction.⁶ The solubility of hydrogen in alcohols increases with increasing carbon chain length.²⁴ This is one of reasons that the conversion of phenol increases from methanol to butanol. Hydrogen solubility increases with increasing temperature and pressure in alcohols.²⁵ Correspondingly, the conversion of phenol is higher with increasing temperature and pressure in alcohols, as shown in Tables 1 and 2. The result that methanol effectively inhibits the hydrogenation of aromatic compounds is in accord with a previous work.²⁶ On the other hand, the competitive adsorption between phenol and alcohol molecule, especially methanol and ethanol, on the same active site may cause the drop in phenol conversion.¹³

The binding energy of phenol–methanol (265 ± 8 meV) is higher than that of phenol–water (243 ± 5 meV).²⁷ This means the interaction of phenol–methanol is stronger than that of phenol–water. Therefore hydrogenation of phenol is more difficult in methanol than in water.

To better discern inhibiting effect of methanol, the hydrogenation of phenol on Pd/C was performed in methanol at hydrogen pressure 4.0 MPa at 250 °C for 4 h with stirring, and then the catalyst was re-used in water for the same reaction under the same condition. The conversion was 44.8% in water. However the conversion was 100% if catalyst was used firstly in water.

The inhibiting effect of ethanol was also examined by hydrogenation of phenol in mixed solvents of ethanol with water or hexane, as shown in Fig. 1. In the mixed solvents, an increasing percentage of ethanol in water or hexane led to a decrease in phenol conversion and cyclohexanol selectivity, whereas the selectivity of cyclohexanone and other by-products yield increased.

Fig. 1 Hydrogenation conversion of phenol versus ethanol volume percent.

On the other hand, the hydrogenation rate of phenol may also depend on the activation itself. Karthick et al.²⁸ calculated the excitation energies from HOMO (highest occupied molecular orbital) to LUMO (lowest unoccupied molecular orbital) for 2,4,6-tris(dimethylaminomethyl)phenol in water, ethanol, and methanol by TD-DFT (time-dependent density functional theory) method with B3LYP/6-311++G (d, p) basis set using PCM (polarizable continuum model), and the values of excitation energies were 3.8562, 3.8596 and 3.8612 eV, respectively. Accordingly, it is plausible that the excitation energies of phenol in water, ethanol and methanol have the same order as water < ethanol < methanol. As a result, the order of reaction is water > ethanol > methanol, this is in accord with the activity order of hydrogenation as listed in Tables 1 and 2.

3.2.3 HBA solvents. As given in Table 1, among HBA solvents, the conversion is higher in ethyl acetate (64.8%) but very low in tetrahydrofuran (12.1%) and acetone (6.8%) at 150 °C. A similar result with the same order is also present at 120 °C.

For HBA solvents, the solvent molecule can supply lone pair electrons by oxygen atoms. The interaction between metal and lone pair electrons blocks the active site.²⁹ As a result, the hydrogenation activity decreases by competitive adsorption between reactants and solvents on the metal surface. At the same time, the interaction between HBA solvents and phenol (HBD substance with α = 1.65)²⁰ suppresses the adsorption of phenol onto the metal surface. Therefore the activity decreases in HBA solvents.

3.2.4 Hydrocarbon solvents. In hexane the conversion is highest among all of solvents and can come to 100% at 150 °C and 250 °C.

Phenol is immiscible with hexane. The interaction between phenol and hexane is obviously weak. Meanwhile the adsorption of hexane molecule on metal active site is also weak. As a result, the hydrogenation would not be inhibited in hexane.

3.3. Correlation of conversion with solvent parameters

To explain the dramatic influence of solvents, the relationship between the hydrogenation conversion of phenol and solvatochromic parameters such as HBD capability α, HBA capability β, polarity/polarizability π* and polarity E^T_N are investigated. For each group of solvents, plots of hydrogenation conversion of phenol as a function of the empirical parameter of solvent polarity/polarizability π* and HBD ability α are shown in Fig. 2 and 3, respectively. In HBD–HBA and HBA solvents, the conversion is decreasing with the increase of polarity/polarizability π* and HBD capability α.

Fig. 2 Hydrogenation conversion of phenol versus solvent polarity/polarizability π*.

Fig. 3 Hydrogenation conversion of phenol versus hydrogen bond donor (HBD) ability α.

3.4. Cyclohexanone hydrogenation

To investigate solvent effects on products, the hydrogenation of cyclohexanone in various solvents was carried out at 150 °C, and results were shown in Fig. 4. Obvious variation is seen in different solvents, and conversions are low in all of solvents except water. This means the hydrogenation is more difficult for cyclohexanone relative to phenol. The results indicate the discussion above is reasonable.

Fig. 4 Solvent effects on the conversion of cyclohexanone hydrogenation at 150 °C.

Fourier transform infrared (FTIR) spectra are used to discern the interaction between cyclohexanone and solvents. FTIR spectra of cyclohexanone in various solvents are shown in Fig. 5. For pure cyclohexanone C=O bond stretching vibration band is at 1715 cm⁻¹. In water and HBD–HBA solvents C=O stretching shifts to lower wavenumbers because of the formation of hydrogen bonding. On the contrary, there are no evident shifts for the C=O band when cyclohexanone is in HBA and hydrocarbon solvents. It is reasonable that the interaction between cyclohexanone and solvents is weaker for HBA and hydrocarbon solvents in contrast to water and HBD–HBA solvents.

Fig. 5 FTIR spectra of cyclohexanone in various solvents.

3.5. Influencing mechanism of reaction

From the discussion above, it can be concluded that chemical reactions are influenced by multiple factors, as shown in Scheme 2. If the promotion factor is bigger than the inhibition one, the reaction will accelerate. On the contrary, the reaction will decelerate.

Scheme 2 The relationship between reaction and interaction of solvent, catalyst, reactant and product.

Various kinds of interactions in hydrogenation of phenol are listed in Table 4. The promotion is biggest in hydrocarbon solvents followed by water. On the contrary, in HBD–HBA and HBA solvents the inhibition is major. As a result, the solvents order of conversion is hydrocarbon solvents > water > HBD–HBA solvents ∼ HBA solvents.

Table 4 Various kinds of interactions in hydrogenation of phenol

Solvent	Interaction (C = catalyst, P = product, R = reactant, S = solvent)						Sum
	S–R	S–C	S–P	C–R	C–P	P–R
Water	+	+	−	+	−	−	3+3−
HBD–HBA	−	−	+	+	−	−	2+4−
HBA	−	−	+	+	−	−	2+4−
Hydrocarbon	+	+	+	+	−	−	4+2−

---

4. Conclusion

Chemical reactions are influenced by multiple factors. When promotion interaction dominates, the reaction can occur easily and vice versa. The solvatochromic parameters of solvents such as HBD capability α, HBA capability β, polarity/polarizability π* and polarity E^T_N may have a correlation with the hydrogenation conversion of phenol over activated carbon supported palladium catalysts. Water and hydrocarbons are the excellent solvents for the hydrogenation of phenol. In HBD–HBA and HBA solvents, the phenol conversion was decreasing with increasing solvent polarity/polarizability π* and HBD capability α.

Acknowledgements

We thank the financial support of National Key Basic Research Program of China (2013CB228104).

Notes and references

1. X. Zheng and H. Lou, Chin. J. Catal., 2009, 30, 765.
2. J. Peng, P. Chen, H. Lou and X. Zheng, Energy Fuels, 2008, 22, 3489.
3. J. Peng, P. Chen, H. Lou and X. Zheng, Bioresour. Technol., 2009, 100, 3415.
4. W. Li, C. Pan, Q. Zhang, Z. Liu, J. Peng, P. Chen, H. Lou and X. Zheng, Bioresour. Technol., 2011, 102, 4884.
5. W. Li, C. Pan, L. Sheng, Z. Liu, P. Chen, H. Lou and X. Zheng, Bioresour. Technol., 2011, 102, 9223.
6. Z. Guo, Y. T. Chen, L. S. Li, X. M. Wang, G. L. Haller and Y. H. Yang, J. Catal., 2010, 276, 314.
7. B. S. Akpa, C. D'Agostino, L. F. Gladden, K. Hindle, H. Manyar, J. McGregor, R. Li, M. Neurock, N. Sinha, E. H. Stitt, D. Weber, J. A. Zeitler and D. W. Rooney, J. Catal., 2012, 289, 30.
8. L. Q. Zhang, J. M. Winterbottom, A. P. Boyes and S. Raymahasay, J. Chem. Technol. Biotechnol., 1998, 72, 264.
9. J. A. Bennett, R. P. Fishwick, R. Spence, J. Wood, J. M. Winterbottom, S. D. Jackson and E. H. Stitt, Appl. Catal., A, 2009, 364, 57.
10. M. Hronec, K. Fulajtarova and T. Liptaj, Appl. Catal., A, 2012, 437, 104.
11. Y. Z. Chen, C. W. Liaw and L. I. Lee, Appl. Catal., A, 1999, 177, 1.
12. S. Scire, S. Minico and C. Crisafulli, Appl. Catal., A, 2002, 235, 21.
13. S. Velu, M. P. Kapoor, S. Inagaki and K. Suzuki, Appl. Catal., A, 2003, 245, 317.
14. J. Liu, H. Li and H. Li, Chin. J. Catal., 2007, 28, 312.
15. H. Liu, T. Jiang, B. Han, S. Liang and Y. Zhou, Science, 2009, 326, 1250.
16. Y. Wang, J. Yao, H. Li, D. Su and M. Antonietti, J. Am. Chem. Soc., 2011, 133, 2362.
17. Y. Perez, M. Fajardo and A. Corma, Catal. Commun., 2011, 12, 1071.
18. A. Chen, Y. Li, J. Chen, G. Zhao, L. Ma and Y. Yu, ChemPlusChem, 2013, 78, 1370.
19. G. Neri, A. M. Visco, A. Donato, C. Milone, M. Malentacchi and G. Gubitosa, Appl. Catal., A, 1994, 110, 49.
20. Y. Marcus, Chem. Soc. Rev., 1993, 22, 409.
21. C. Reichardt, Chem. Rev., 1994, 94, 2319.
22. J. Yang, P. J. Dauenhauer and A. Ramasubramaniam, J. Comput. Chem., 2013, 34, 60.
23. X. Wu, J. Liu, D. Di Tommaso, J. A. Iggo, C. R. A. Catlow, J. Bacsa and J. Xiao, Chem.–Eur. J., 2008, 14, 7699.
24. M. S. Wainwright, T. Ahn, D. L. Trimm and N. W. Cant, J. Chem. Eng. Data, 1987, 32, 22.
25. J. V. H. d'Angelo and A. Z. Francesconi, J. Chem. Eng. Data, 2001, 46, 671.
26. X. Wang and R. Rinaldi, ChemSusChem, 2012, 5, 1455.
27. A. Courty, M. Mons, I. Dimicoli, F. Piuzzi, V. Brenner and P. Millie, J. Phys. Chem. A, 1998, 102, 4890.
28. T. Karthick, V. Balachandran, S. Perumal and A. Lakshmi, Spectrochim. Acta, Part A, 2013, 113, 52.
29. H. Takagi, T. Isoda, K. Kusakabe and S. Morooka, Energy Fuels, 1999, 13, 1191.

---