Topics in Catalysis

, Volume 60, Issue 9–11, pp 637–643 | Cite as

Decomposition of Lignin Using MO–MgAlOy Mixed Oxide Catalysts (M=Co, Ni and Cu) in Supercritical Ethanol

Original Paper
  • 229 Downloads

Abstract

In this study, we performed the depolymerization reaction of concentrated sulfuric acid hydrolysis lignin (CSAHL) in supercritical ethanol without supplying external hydrogen. Cu demonstrated the highest monoaromatic yield of 18.4 wt% among Co, Ni and Cu when incorporated into MgAlOy. With increasing the amount of Cu loading, it was found that the optimum Cu loading was 30 wt%, where the number of acid sites had the maximum. In case of CuO–MgAlOy sample with higher Cu loading than 30 wt%, the decrease in the acid sites and the significant sintering of Cu metal during reaction were observed, leading to the decline in the monoaromatic yield.

Keywords

Concentrated sulfuric acid hydrolysis lignin (CSAHL) Supercritical ethanol CuO(X)MgAlOy MO(30)MgAlOy In-situ hydrogen 

1 Introduction

Finding alternative resources for energy and chemicals becomes imperative as the conventional fossil fuel is rapidly depleted due to the increasing demand [1]. In this respect, lignin can be regarded as one of promising renewable resources for aromatic chemicals because it consists of methoxylated phenylpropane structures such as ρ-coumaryl, coniferyl and sinapyl alcohols [2, 3]. However, the complex and recalcitrant structure of lignin, linked together by C–O and C–C bonds, makes it hard to cleave into oligomers and/or monomers [4, 5]. Especially, the selective cleavage of C–O bonds in lignin remains a big challenge.

There are efforts for valorization of lignin by using various approaches such as hydrolysis, solvolysis, oxidation, pyrolysis and hydrogenolysis [6, 7]. Especially, hydrogenolysis is more attractive method than others to degrade lignin because it is more effective for breaking C–O bond in lignin. It is typically conducted not only in the presence of gas phase hydrogen but also in solvents such as formic acid, methanol and ethanol under supercritical condition which can easily donate hydrogen. Kleinert et al., Xu et al. and Huang et al. have stated that formic acid plays a role as hydrogen donors under supercritical condition [8, 9, 10]. Warner et al. also has reported on the hydrogen transfer from supercritical methanol during the lignin depolymerization [11]. Patil et al. and Brand et al. have claimed that supercritical ethanol can supply in-situ hydrogen and, at the same time, can act as a reactant [12, 13].

Meanwhile, there have been several reports on lignin depolymerization in presence of gas phase hydrogen by using heterogeneous catalysts. Song et al. has demonstrated that Ni catalyst can convert lignin into monomers via hydrogenolysis and solvolysis [7]. Cu and Ni supported catalysts have also been used to depolymerize and deoxygenate lignin [14, 15]. Porous metal oxides (PMOs) derived from hydrotalcite-like materials (CuO–MgAlOy) were found to be effective catalyst for lignin depolymerization because of no or less formation of char [16, 17]. However, to the best our knowledge, there have been few researches concerning the analysis of catalysts after lignin depolymerization in supercritical ethanol. For this purpose, we designed a catalyst basket in the reactor to easily separate post-reaction catalyst for analysis. In addition, we aimed at finding the best transition metal among Co, Ni and Cu and the optimum metal loading for lignin depolymerization under supercritical condition when incorporated into MgAlOy.

2 Experimental

2.1 Materials and Chemicals

The concentrated strong acid hydrolysis lignin (CSAHL) used in this research was supplied from GS Caltex. The CSAHL was acquired from oak wood (Quercus acuta, hardwood) sawdust (C: 46.5 wt%, H: 6.0 wt%, O: 45.4%) by using concentrated strong acid hydrolysis process. Thermogravimetric analysis (TGA) indicates that the ash content of CSAHL is about 6.8 wt%. For catalyst synthesis, Mg(NO3)2·6H2O, Al(NO3)3·9H2O, Cu(NO3)2·3H2O, Ni(NO3)2·6H2O, Co(NO3)2·6H2O, NaOH and Na2CO3 were purchased from Sigma–Aldrich. Ethanol (200 proof, ACS reagent, ≥99.5%, Sigma–Aldrich) was used as a reaction medium for lignin depolymerization under supercritical condition.

2.2 Catalyst Preparation

The catalysts, derived from hydrotalcite-like materials, were prepared by using co-precipitation method. Briefly, in order to synthesize MgAlOy, two different kinds of solutions were prepared. Solution A consisted of Mg(NO3)2·6H2O and Al(NO3)3·9H2O in distilled water, and solution B did NaOH and Na2CO3. Solution B was added into solution A until the pH reached 10 ± 0.05 by keeping the drop-wise rate (1 mL/min) at 65 °C. Synthesized material was calcined at 800 °C for 6 h in a muffle furnace to induce MgAlOy. The detailed procedures are described elsewhere [18, 19]. In case of MO-MgAlOy, solution A containing three precursors (M, Mg, Al) was used and the other procedures were exactly the same as described above. The product was calcined at 460 °C for 6 h in a muffle furnace to induce MO-MgAlOy. The final material was denoted to MO(X)MgAlOy [M=Co, Ni and/or Cu, X = 10, 20, 30, and 40 wt% based on metal (M)]. In case of Co and Ni, the amount of metal was fixed as 30 wt%. Typically, the molar ratio of M2+/M3+ in hydrotalcites (HTs) is 2/1; here M2+ originally stands for Mg ions and M3+ does Al ions. In this research, M2+ is partially substituted with Co, Ni or Cu ions, so that the molar ratio of M2+/M3+ remains constant as 2/1.

2.3 Characterization

The total inorganic content was measured by using a TA Instruments Q50 TGA in the temperature range of 50–800 °C at a rate of 10 oC/min in air flow (100 mL/min). Inductively coupled plasma-atomic emission spectroscopy (ICP-AES) was used (PerkinElmer/Optima-4300 DV) to confirm the metal contents (Mg, Al, Co, Ni and Cu) of prepared catalysts. The weight of the each sample was about 0.03 g, and the sample was dissolved in 5 mL aqua regia. Textural properties of the catalysts were measured using N2 physisorption apparatus (Micrometritics ASAP 2010) at constant temperature (77 K). All samples were pretreated at 250 °C for at least 4 h under vacuum condition prior to measurements. The powder X-ray diffraction (XRD) experiment was performed on a high power X-ray diffractometer (Rigaku Corp.) using Cu Kα as a radiation source at 40 kV and 30 mA. The scan step was fixed as 0.02° in the range of 10°–90°. Temperature programmed desorption of NH3 (NH3-TPD) was carried out for the as-calcined catalysts. After pretreatment at 400 °C for 1 h under a flow of He (50 sccm), 5% NH3 was introduced to 0.03 g of the sample at 50 °C for 0.5 h. After purging with He to remove physisorbed NH3, the temperature was increased from 50 to 430 °C at 10 oC/min under flowing He. The desorbed gas (NH3) was analyzed by a thermal conductivity detector (TCD) (BEL JAPAN INC.).

2.4 Catalytic Activity Measurements

Lignin depolymerization reaction was conducted in a 50 mL stainless-steel high pressure autoclave. The autoclave was filled with 0.5 g of lignin in 25 mL ethanol. 0.3 g of catalyst was loaded into the designed basket (stainless-steel, 500 mesh). 0.0013 g of fluoranthene was also added as an internal standard. The reactor was sealed and purged with nitrogen several times to remove oxygen. Before the reaction, 10 bar nitrogen at room temperature was loaded into the reactor. Then, the reaction mixture was heated to the desired temperature while stirring at 500 rpm for 40 min. After the reaction, the reactor was quickly quenched below 150 °C by using ice water to cool to room temperature. The yield of monomers was calculated by using Eq. (1):
$${\text{~~~~~~~Yield~of~monomers}}=\frac{{{\text{weight~~of~monomers~}}\left( {{\text{quantified~by~GC}} - {\text{FID}}} \right)}}{{{\text{weight~of~ash~free~lignin}}}}~~ \times 100$$
(1)

2.5 Product Analysis

The liquid phase product mixture was analyzed by an Agilent 6890 N GC equipped with a DB-5ms column (30 m × 0.25 mm × 0.25 μm) with a mass spectrometer to identify products. Meanwhile, the same liquid phase product was also injected into an Agilent 6890A GC together with a flame ionization detector (FID) with a DB-5 column (60 m × 0.25 mm × 0.25 μm) for the quantification of the products. The liquid product analysis was performed by using as-produced sample, in other words, without any dilution.

3 Results and Discussion

3.1 Effect of MOMgAlOy Mixed Oxide Catalysts (M=Co, Ni and Cu) on Lignin Depolymerization

3.1.1 Analysis of As-Calcined MOMgAlOy Samples

As shown in Table 1, the difference between the experimental values and the theoretical ones of M2+/M3+ and MO is no more than 12%. BET surface area of Co3O4(30)MgAlOy is about 198 m2/g and that of NiO(30)MgAlOy is about 194 m2/g, which are slightly larger than that of CuO(30)MgAlOy (170 m2/g). The XRD patterns of MO(30)MgAlOy are displayed in Fig. 1. The peaks arising from MgO phase are primarily shown in all catalysts and the broad XRD patterns of each transition metal oxides are displayed, indicating that Co, Ni or Cu oxides exist in the amorphous form or highly dispersed state.

Table 1

Results of ICP-AES and N2 physisorption of the catalysts

Entry

Catalyst

ICP-AES

N2 physisorption

As-calcined

Post-reaction

M2+,a/M3+,b

M wt. %c

SBET (m2/g)

SBET (m2/g)

1

MgAlOy

2.17

170

135

2

CuO(10)MgAlOy

1.78

11.1

171

47

3

CuO(20)MgAlOy

2.00

21.3

166

58

4

CuO(30)MgAlOy

1.81

34.5

170

100

5

CuO(40)MgAlOy

1.98

43.4

172

105

6

Co3O4(30)MgAlOy

1.89

31.9

198

7

NiO(30)MgAlOy

1.82

33.1

194

aM2+: Mg and Co, Ni or Cu ions

bM3+: Al3+

cM: Cu, Co and/or Ni

Fig. 1

XRD patterns of as-calcined MO(30)MgAlOy (M=Co, Ni and Cu)

Figure 2 displays the NH3-TPD profiles of MO(30)MgAlOy. According to the previous study, the NH3 desorbed below 300 °C is regarded as one from NH3 adsorbed on the weak acid sites of catalyst [20]. When the number of weak acid sites is counted by integrating the amount of NH3 desorbed up to 300 °C, the order is in the following: CuO(30)MgAlOy > NiO(30)MgAlOy > Co3O4(30)MgAlOy. This result is in accordance with that of Robinson who claimed that the catalyst containing Cu has a higher number of acid sites than the Ni-containing one, indicating that copper oxides provide new acid sites [21].

Fig. 2

NH3-TPD spectra of MO(30)MgAlOy (M=Co, Ni and Cu)

3.1.2 Lignin Depolymerization Over MOMgAlOy

Lignin depolymerization reaction was conducted in supercritical ethanol over both without catalyst (blank reaction) and MO(30)MgAlOy at 400 °C for 4 h. The total yield of monoaromatic compounds was obtained by adding up the yield of 19 different monoaromatic products such as γ-valerolactone, phenol, benzyl alcohol, o-cresol, ρ-tolualdehyde, guaiacol, 4-methyl benzyl alcohol, 2-ethyl phenol, 2-methyl benzyl alcohol, γ-heptalactone, 4-ethyl phenol, creosol, 2-propyl phenol, 4-ethyl guaiacol, syringol, 4-propyl guaiacol, 1,2,4-trimethoxy benzene, 3,4,5-trimethoxy toluene and ethyl vanillate. Most of the monoaromatic compounds are oxygen-containing aromatic rings resulting from the cleavage of the β-O-4 linkage or hydrogenolysis of lignin, indicating that the combination of catalyst and supercritical ethanol is effective for the lignin depolymerization without supplying external hydrogen. In addition, there is significant amount of furans which are not counted in the monoaromatic compounds [16]. Table 2 demonstrates the yield of monoaromatic compounds quantified from GC-FID analysis. In case of blank reaction, the yield of monoaromatic compounds is 1.0 wt%, acquired by adding up only 6 monomers such as benzyl alcohol, o-cresol, guaiacol, 4-methyl benzyl alcohol, 4-ethyl phenol, 2-propyl phenol among 19 monomers. The result indicates that the presence of the catalyst definitely affects improvement of monoaromatic yield. Meanwhile, the yield of monoaromatic compounds over CuO(30)MgAlOy is the highest as 18.4 wt%, followed by NiO(30)MgAlOy of 10.8 wt% and Co3O4(30)MgAlOy of 10.0 wt%. Figure 3a presents the distributions of selected monoaromatic compounds identified and quantified by GC-FID analysis. Creosol, ρ-tolualdehyde, phenol and syringol are the major components in all catalysts. The yield of monoaromatic compounds is well related to the number of acid sites as shown in Fig. 2, leading to the conclusion that CuO(30)MgAlOy demonstrated the higher monoaromatic yield due to the higher number of acid sites among Ni, Co and Cu.

Table 2

The yield of monoaromatic compounds over the catalysts

Entry

Catalyst

Ymonoaromatic compounds (wt%)

1

None

1.0

2

MgAlOy

5.7

3

CuO(10)MgAlOy

10.7

4

CuO(20)MgAlOy

14.5

5

CuO(30)MgAlOy

18.4

6

CuO(40)MgAlOy

15.5

7

Co3O4(30)MgAlOy

10.0

8

NiO(30)MgAlOy

10.8

Fig. 3

Distributions of selected monoaromatic compounds over a MO(30)MgAlOy (M=Co, Ni and Cu) and b MgAlOy and CuO(X)MgAlOy (X = 10, 20, 30 and 40 wt%)

3.2 Effect of Cu Loadings in CuO(X)MgAlOy on Lignin Depolymerization

3.2.1 Analysis of As-Calcined CuO(X)MgAlOy Catalysts

According to Table 1, there is not much difference between the experimental values obtained from ICP-AES and the theoretical ones of M2+/M3+ and MO, since the error is within 12%. Table 1 also shows BET surface area of MgAlOy, CuO(10)MgAlOy, CuO(20)MgAlOy, CuO(30)MgAlOy and CuO(40)MgAlOy catalysts. The samples have BET surface area of 166–172 m2/g implying that there is no remarkable change in BET surface area even though Cu loading is varying up to 40 wt%. Figure 4a demonstrates the XRD patterns of as-calcined catalysts. MgAlOy catalyst only shows the peaks arising from MgO phase. With increasing Cu loading, the peaks assigned to MgO phase become broad and small probably because of the decrease in MgO amount. Samples with 30 and 40 wt% of Cu loading show very broad CuO peak around at 36.9°, meaning that highly dispersed states of CuO exist in the samples although CuO phase is hardly seen up to 20 wt% of Cu loading. Interesting to note is that there is no evidence of the Al2O3 related phases in XRD, indicating that they exist in small crystallites or amorphous form so that they may not be detected by XRD [22].

Fig. 4

XRD patterns of the catalysts (a) as-calcined (b) after lignin depolymerization reaction

In order to measure the acid sites of the catalysts, NH3-TPD analysis was conducted as displayed in Fig. 5. The desorption feature of all catalysts are almost similar which have weak acid sites as evidenced by the NH3 desorption peak from 100 to 300 °C. Among the catalysts, the amount of weak acid sites is the highest over CuO(30)MgAlOy as 0.644 mmol/g followed by CuO(20)MgAlOy (0.422 mmol/g) ≈ CuO(40)MgAlOy (0.396 mmol/g) > CuO(10)MgAlOy (0.254 mmol/g). NH3 TPD result clearly indicates that CuO(30)MgAlOy has the maximum acid sites among the catalysts.

Fig. 5

NH3-TPD spectra of the catalysts

3.2.2 Catalytic Effect on Lignin Depolymerization

Lignin depolymerization reaction was conducted in supercritical ethanol to evaluate the effect of Cu loadings in the catalysts at 400 °C for 4 h. Table 2 displays the yield of monoaromatic compounds calculated from GC-FID result. MgAlOy demonstrates the lowest monoaromatic yield of 5.7 wt%. With increasing Cu loading up to 30 wt%, the yield of monoaromatic compounds gradually increased to 18.4 wt%. However, CuO(40)MgAlOy results in the lower yield (15.5 wt%) than CuO(30)MgAlOy, indicating that the higher Cu loading than 30 wt% does not help increasing the yield. Figure 3b depicts the distributions of selected monoaromatic compounds over MgAlOy and Cu-containing catalysts quantified by GC-FID. Among various monoaromatic compounds, guaiacol, ρ-tolualdehyde and phenol are the major compounds in all catalysts. Based on this result, it can be summarized that the monoaromatic yield has a close relationship with the number of acid sites, determined by the loading of copper metals.

3.2.3 Analysis of Post-Reaction CuO(X)MgAlOy Catalysts

First of all, we observed the color change of the catalyst after reaction (from green to brown) because of the reduction of CuO to Cu due to the in-situ hydrogen generation (not shown). N2 physisorption and XRD analysis were conducted for the post-reaction catalysts to understand the change in the physical properties after reaction. As presented in Table 1, BET surface area of MgAlOy decreased from 170 to 135 m2/g after reaction. Moreover, all CuO(X)MgAlOy catalysts show the significant decrease in BET surface area after reaction. Note that in case of CuO(30)MgAlOy and CuO(40)MgAlOy, the extent of decrement was less than the others, i.e. 170–100 and 172–105 m2/g, respectively. It can be summarized that the textural change of the CuO(X)MgAlOy catalysts occurred after reaction, resulting in the substantial decrease of BET surface area.

Figure 4b displays the XRD patterns of the post-reaction catalysts with various Cu loadings. The XRD patterns of all post-reaction catalysts clearly exhibit the highly crystalline of Cu, Mg(OH)2 and MgAl2O4 phases. First of all, MgO phase was transformed into Mg(OH)2 and MgAl2O4. We think that such change gives rise to the significant decrease in BET surface area after reaction. Metallic Cu phase at 43.3° and 50.5° is formed by the reduction of CuO due to in-situ hydrogen donated from supercritical ethanol. With increasing the Cu loading up to 30 wt%, there is a gradual increase in the peak intensity of Cu phase at 43.3°. However, at Cu loading of 40 wt%, more remarkable increase in Cu peak intensity (area: 298 vs. 490) was observed, resulting from the sintering during the reaction, which can account for decrease in monoaromatic yield.

4 Conclusions

We investigated the depolymerization of concentrated sulfuric acid hydrolysis lignin (CSAHL) over MOMgAlOy catalysts (M=Cu, Ni or Co) in supercritical ethanol without additional hydrogen supply to acquire high monoaromatic yield. According to our results, Cu was selected as the optimum metal among Co, Ni and Cu with the highest yield of monoaromatic compounds of 18.4 wt% when incorporated into MgAlOy. When it comes to the amount of Cu loading, the optimum Cu loading was 30 wt%, which showed the highest the number of acid sites. Meanwhile, it was found that as the Cu loading exceeded 30 wt%, there had been the decline of acid sites and the remarkable sintering of Cu metal during reaction, resulting in the decrement of monoaromatic yield.

Notes

Acknowledgements

This work was supported by the New and Renewable Energy Core Technology Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) granted financial resource from the Ministry of Trade, Industry and Energy, Republic of Korea (No. 20143030090940).

References

  1. 1.
    Huber GW, Iborra S, Corma A (2006) Chem Rev 106:4044–4098CrossRefGoogle Scholar
  2. 2.
    Beauchet R, Monteil RF, Lavoie JM (2012) Bioresource Technol 121:328–334CrossRefGoogle Scholar
  3. 3.
    Zakzeski J, Jongerius AL, Bruijnincx PCA, Weckhuysen BM (2012) Chemsuschem 5:1602–1609CrossRefGoogle Scholar
  4. 4.
    Barta K, Matson TD, Fettig ML, Scott SL, Iretskii AV, Ford PC (2010) Green Chem 12:1640–1647CrossRefGoogle Scholar
  5. 5.
    Jongerius AL, Bruijnincx PCA, Weckhuysen BM (2013) Green Chem 15:3049–3056CrossRefGoogle Scholar
  6. 6.
    Pineda A, Lee AF (2016) Appl Petrochem Res 6:243–256Google Scholar
  7. 7.
    Song Q, Wang F, Cai J, Wang Y, Zhang J, Yu W, Xu J (2013) Energ Environ Sci 6:994–1007CrossRefGoogle Scholar
  8. 8.
    Huang S, Mahmood N, Tymchyshyn M, Yuan Z, Xu C (2014) Bioresourc Technol 171:95–102CrossRefGoogle Scholar
  9. 9.
    Kleinert M, Barth T (2008) Chem Eng Technol 31:736–745CrossRefGoogle Scholar
  10. 10.
    Xu W, Miller SJ, Agrawal PK, Jones CW (2012) Chemsuschem 5:667–675CrossRefGoogle Scholar
  11. 11.
    Warner G, Hansen TS, Riisager A, Beach ES, Barta K, Anastas PT (2014) Bioresource Technol 161:78–83CrossRefGoogle Scholar
  12. 12.
    Brand S, Susanti RF, Kim SK, Lee HS, Kim J, Sang BI (2013) Energy 59:173–182CrossRefGoogle Scholar
  13. 13.
    Patil PT, Armbruster U, Richter M, Martin A (2011) Energ Fuel 25:4713–4722CrossRefGoogle Scholar
  14. 14.
    He J, Zhao C, Lercher JA (2012) J Am Chem Soc 134:20768–20775CrossRefGoogle Scholar
  15. 15.
    Strassberger Z, Alberts AH, Louwerse MJ, Tanase S, Rothenberg G (2013) Green Chem 15:768–774CrossRefGoogle Scholar
  16. 16.
    Huang XM, Koranyi TI, Boot MD, Hensen EJM (2014) Chemsuschem 7:2276–2288CrossRefGoogle Scholar
  17. 17.
    Macala GS, Matson TD, Johnson CL, Lewis RS, Iretskii AV, Ford PC (2009) Chemsuschem 2:215–217CrossRefGoogle Scholar
  18. 18.
    Cavani F, Trifirò F, Vaccari A (1991) Catal Today 11:173–301CrossRefGoogle Scholar
  19. 19.
    Matson TD, Barta K, Iretskii AV, Ford PC (2011) J Am Chem Soc 133:14090–14097CrossRefGoogle Scholar
  20. 20.
    Hu H, Cai S, Li H, Huang L, Shi L, Zhang D (2015) J Phys Chem C C 119:22924–22933CrossRefGoogle Scholar
  21. 21.
    Manfro RL, Pires TPMD, Ribeiro NFP, Souza MMVM (2013) Catal Sci Technol 3:1278–1287CrossRefGoogle Scholar
  22. 22.
    Fornasari G, Gusi S, Trifiro F, Vaccari A (1987) Ind Eng Chem Res 26:1500–1505CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2017

Authors and Affiliations

  1. 1.School of Chemical and Biological Engineering, Institute of Chemical ProcessesSeoul National UniversitySeoulRepublic of Korea

Personalised recommendations