Accepted Manuscript

High-quality bio-oil from one-pot catalytic hydrocracking of kraft lignin over

supported noble metal catalysts in isopropanol system

Jing Yang, Liang Zhao, Shaotong Liu, Yuanyuan Wang, Liyi Dai

PII: S0960-8524(16)30516-8
DOI: http://dx.doi.org/10.1016/j.biortech.2016.04.029
Reference: BITE 16388

To appear in: Bioresource Technology

Received Date: 25 January 2016

Revised Date: 5 April 2016
Accepted Date: 6 April 2016

Please cite this article as: Yang, J., Zhao, L., Liu, S., Wang, Y., Dai, L., High-quality bio-oil from one-pot catalytic
hydrocracking of kraft lignin over supported noble metal catalysts in isopropanol system, Bioresource
Technology (2016), doi: http://dx.doi.org/10.1016/j.biortech.2016.04.029

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
High-quality bio-oil from one-pot catalytic hydrocracking of kraft lignin over

supported noble metal catalysts in isopropanol system

Jing Yang, Liang Zhao, Shaotong Liu, Yuanyuan Wang*, Liyi Dai*

Shanghai Key Laboratory of Green Chemistry and Chemical Processes, College of

Chemistry and Molecular Engineering, East China Normal University, 500 Dongchuan

Road, Shanghai 200241, China

*Corresponding authors: Liyi Dai, Yuanyuan Wang

E-mail: [email protected]; [email protected]

Tel.:+8621-54340133

1
ABSTRACT

Catalytic hydrocracking of kraft lignin was carried out in isopropanol system and

an orthogonal array design (OAD) was employed to optimize the experimental

conditions. GC-MS/FID, elemental analysis, GPC and 1H-13C HSQC NMR were carried

out for entire investigation of the liquid products. The results indicated that the

hydrocracking process was thermally controlled and catalysts showed significant

influences on the product distributions. Comparing with Pd/C, Pt/C and Ru/C, Rh/C

inhibited the self-condensation of isopropanol and reduced the formation of

oxygenic-chain compounds. The excellent catalytic activity for phenols conversion was

obtained over Rh/C. The routes of oxygenic-chain compounds formation and phenol

conversion were proposed in detail. The least oxygenic-chain compounds formation, the

highest phenols conversion (93.4%), the lowest O/C ratio (0.094) and the highest HHV

(37.969 MJ/kg) provided the possibility of the high quality bio-oil obtained over Rh/C

in isopropanol medium.

Keywords: kraft lignin; isopropanol; hydrocracking; orthogonal array design; nobel

metal catalysts;

1. Introduction

Diminishing fossil energy resources, increasing demands for fuels and inevitable

environmental questions have driven much of interest for seeking a series of promising

alternatives to the traditional fossil fuels. As a kind of nonedible biomass, lignin is

becoming a significant feedstock for fuels and high-value chemicals. However, lignin is

the least susceptible to various chemical technologies and the usability of it only

occupies approximately 2% in the conventional paper industries due to the complicated

three-dimensional amorphous polymer structure, (Gosselink et al., 2004). Furthermore,

2
the presence of excess oxygen in lignin decreased stability and heating values of

resulting oils, which makes it difficult to be used as fuels directly (Zhao et al., 2013;

Ruddy et al., 2014).

In order to address these above-mentioned problems, an efficient hydrocracking

technology for lignin need to be introduced urgently (Li et al., 2015; Xu et al., 2014). In

recent decades, a large number of literatures focus on lignin degradation in water or

alcohols system (Schutyser et al., 2015). Water is the most “green” system and

completely miscible with light gases, hydrocarbons and aromatic compounds (Nguyen

et al., 2014), but in which obtained oil with large amount content of oxygen require an

excess hydrogen supply for O-removal process (Mortensen et al., 2011; Saidi et al.,

2014). Alcohols have been discovered more deeply than water due to capability of

hydrogen-donating, which receives high-yield and high-calories liquid products. For

example, the lignin fragmentations are promoted and residues are suppressed in

methanol and ethanol mediums over various hydrodeoxygenation-functional catalysts

(Cheng et al., 2012; Singh et al., 2014). Narani et al. (2015) reported that kraft lignin

could be transformed into alkylphenolics over supported NiW and NiMo catalysts in

supercritical methanol. Huang et al. (2015) proposed that ethanol was used not only as

formaldehyde scavenger to prevent the repolymerization of lignin decomposition

products, but also as the reactant reacted with lignin side-chains to prevent the

condensation reaction between larger lignin fragments. Moreover, high-value

small-molecular chemicals had been obtained by catalytic cracking of kraft lignin in

ethanol system (Ma et al., 2014).

Similarly, isopropanol has the ability of hydrogen-donating to realize the

conversion of lignin during the dehydrogenation process of the transformation to

3
acetone (Wang and Rinaldi, 2012a, Ferrini and Rinaldi, 2014). What’s more, this

alcohol could be used as an H-transfer initiator to realize hydrogenation of phenol rings

to aromatic hydrocarbon by Wang and Rinaldi (2012b, 2013). Except that, Kloekhorst

et al. had reported that a high lignin oil yield associated with a low char yield were

observed in isopropanol system compared with methanol and ethanol (Kloekhorst et al,

2015). However, it is disappointing that the lowest aromatic compounds and the highest

oxygenic-chain compounds (aliphatic acid/ester, aliphatic amide, linear/branched

ketones/alcohols/ether) were obtained, which affected the quality of oil seriously.

Herein, it is important to figure out the formation pathway of oxygenic-chain

compounds and seek efficient hydrodeoxygenation (HDO) catalysts for decreasing the

oxygenic-chain compounds ingredients and promoting the self-hydrocracking of lignin

in isopropanol medium.

In early studies, the reaction of HDO were mainly catalyzed by the supported

sulfide Co-Mo catalysts that was prone to deactivate (Corma et al., 2007). The

application of noble metal catalysts to lignin hydrocracking process were reported

widely in the past few years (Bengoecheaa et al., 2015; Wildschut et al., 2009; Chen et

al., 2014). Comparing to Co-Mo catalysts, supported noble metal catalysts acted out

higher HDO activity (Lin et al., 2011). In our work, the commonly noble metal (Pd, Pt,

Ru, Rh) supported on active carbon were chosen to catalyze the lignin hydrocracking in

isopropanol system and performance comparison were conducted systematically among

HDO catalysts. Active carbon and non-catalyzed experiments were performed as blank

experiments compared with the supported catalysts. An orthogonal array design (OAD)

was employed to optimize the experimental conditions.

2. Materials and Methods

4
2.1. Materials and catalysts

The kraft lignin used in this work was provided by Sinopharm Chemical Reagent

Co. Ltd. The samples were dried overnight at 100 ºC before use. The activated carbon

(AC) and 5 wt % metal loadings for all catalysts used in this study were purchased from

Aladdin: Pt/C, Pd/C, Ru/C and Rh/C. AR reagent grade solvents and chemicals

including isopropanol, dichloromethane, diisobutyl ketone, 4-methyl-3-penten-2-one,

methyl cyclohexane, cyclohexane alcohol, 3,3,5-trimethylcyclohexanol,

1,2,4-trimethylbenzene, guaiacol, p-cresol, n-dodecane were also purchased from

Aladdin. All reagents were used as received.

2.2. Lignin hydrocracking

Lignin hydrocracking performed in a high-pressure and corrosion-resistant

autoclave with 30 ml volume. The reactor equipped with a temperature controller and a

pressure sensor. In order to ensure the same system pressures in the same temperature,

12 mL isopropanol was selected and never changed. In the typical run, kraft lignin,

catalysts and isopropanol were loaded into the reactor. Then, the sealed autoclave was

heated to the specified temperatures and kept for the desired residence time. After the

heating, the reactor was immersed into a cold water bath for about 5 min to quench the

reaction and then the reactor was opened and the gas was released. The reaction

mixture was extracted several times with 100 ml dichloromethane to collect all the

product fraction. The extract was next filtrated and the filtrate was rotary evaporated at

a reduced pressure (35 ºC and 1 bar for at least 30 min) to remove the dichloromethane

and the unreacted isopropanol. The liquid products were obtained and weighted. The

brown colored semi-liquid product was defined as bio-oil. The filtrate residue was

weighted after dried overnight in 120 ºC (Fig. 1). Due to the volatility of some light

5
ends, the bio-oil yields somewhat decreased during the solvent vaporization process.

The lignin conversion (X) were determined by the amount of the char that is filtrate

residue without catalyst (eq 1). The bio-oil yield (Y) was measured as the mass of

remaining oil divided by the weight of kraft lignin loaded into the reactor (eq 2). The

phenols conversion (W) were calculated using eq 3. However, it could not excluded

that the peak area of phenols, aromaic compounds, cycloalkanes,

cyclohexanols/cyclohexanones could not represent the entire phenols precursor obtained

from the liquefaction of lignin.

weight of residues − weight of the catalyst

 wt % = 100 − ∗ 100 1
weight of the parent lignin

weight of semi liquid

!"#$# wt % = ∗ 100 2
weight of the parent lignin

()*+#, wt %

= 100

peak area of phenols

−
peak area of phenols + aromatic hydrocarbones + cycloalkanes + cyclohexanols + cyclohexanones

∗ 100 3

Considering the uncertainties of the experimental data, all tests were carried in

duplicate. Recorded results were the mean values of the independent trials. Finally, the

repeatability of results under the optimal conditions was verified by repeated

experiments in several times.

2.3. Orthogonal array design (OAD)

As we known, various operating variables such as temperature (pressure), time,

feedstock ratio, various catalysts and catalyst concentration had different influences on

the yields and the quality of biofuels (Chang et al., 2015). For seeking the optimal

condition, a L25(56) OAD was conducted in our work. The design involving five factors

6
(temperature, time, lignin concentration, catalyst, catalyst concentration) at five levels

(270-350 ºC, 1-5 h, 1-5 wt %, five catalysts, 10-50 wt %) was shown in Table S1 and

the experimental order was presented in Table S2. In this work, lignin concentration

was the weight percentage of parent lignin against isopropanol (12 mL) and

catalyst concentration was the weight percentage of catalyst against parent

lignin.

Each factor contributing to the products yield/heating values was evaluated with

direct observation analysis of the OAD. Taking the effect of temperature on yield as

example, the average value of product yield at the five levels of temperature was

calculated so that the optimal temperature could be received (Fig. 2a). The other factors

could be evaluated in the same way. Notably, each reported oil yield was the average of

five measurements, for each of which the parameter of interest was kept constant and

the other parameters were changed. Therefore, the actual yield was no lower than that of

the maximum point in each graph under the optimal conditions. Although few certain

data were scattered or deviate unexpectedly in our method, it allowed us to identify the

optimal conditions of the reaction.

2.4. Product analysis

Liquid products dissolved in dichloromethane were analyzed qualitatively with an

Agilent 5975 C GC-MS system for the compound identification and further analyzed

quantitatively using Agilent 5975 C GC equipped with a FID with the same HP-5 MS

capillary column (30 m * 0.25 mm *0.25 µm). The oven temperature program was set

from an initial temperature of 45 ºC to 150 ºC at 5 ºC /min and maintain 150 ºC for 3

min; Then, raised to final temperature 280 ºC at 8 ºC /min and maintain the final

temperature for 2 min. The solvent delay was set as 3 min and the total runtime was

7
about 65 min. Identification of the compounds was achieved by comparing the mass

spectra obtained with the system’s mass spectral library (NIST II). In addition, the

identified compounds were quantitated on the basis of the p-cresol as an internal

standard.

The compositions of C, H, N in raw material (Kraft lignin) and liquid products

were measured by an elemental analyzer. Oxygen content was calculated by difference.

The higher heating value (HHV) of oil was calculated by the Dulong’s formula:

Heating value (MJ/kg) = 0.335*[C] +1.423*[H] -0.154*[O] -0.145*[N].

Where C, H, N is the wt % composition of each element.

Gel permeation chromatography (GPC) analyses were conducted by using Malvern

apparatus and a RI detector. The bio-oil (~5mg) was dissolved with 1 mL THF.

Analyses were performed at 35 ºC using THF as eluent with a flow rate of 1 ml min-1.

All the samples were filtered using 0.45 µm filter membrane prior to injection.
1
H-13C heteronuclear signal quantum coherence (HSQC) NMR spectrometry was

recorded using a Bruker 500 MHz spectrometer. Approximately 50 mg of bio-oil was

dissolved in 0.6 mL dimethylsulfoxide- d6 (DMSO). Spectra was acquired using a

standard pulse sequence HSQC programme with a spectral width of 240 ppm, 16 scans,

and 256 increments. Data processing was carried out using the MestReNova software.

3. Results and discussion

3.1. Influence of reaction conditions

3.1.1. Effect of operating variables on the yield of bio-oil

The initial experiments were performed to study the influence of different

operating variables such as temperature (270-350 ºC), time (1-5 h), lignin concentration

(1-3 wt %), catalysts (Pd/C, Pt/C, Ru/C, Rh/C, C) and catalyst concentration (10-50

8
wt %) on the product yields in isopropanol system. Due to fixed volume (30 mL) of

autoclave, isopropanol have different saturation vapor pressure at different

temperature. What’s more, the decomposition degree of lignin was influenced

by the pressure and which was listed in the Table S2.

The yield of both bio-oil and char under different reaction conditions

were listed in Table S2 in detail. According to the direct observation analysis

of OAD, the tendencies of product yields along with factors changing were

listed in Fig. 2. Temperature was considered as the most important factor on the

product yield. The yield of oil increased and the yield of char decreased with the

temperature increasing (Fig. 2a). When the temperature raised from 270 to 330 ºC, the

yield of oil improved sustainably from 20.7% to 47.5%. Whereas, it increased sharply if

continued to raise the temperature to 350 ºC, which suggested that a larger ingredient

variations occurred at a relatively high reaction temperature (exceed 330 ºC). No

downward tendency figured that lignin hydrocracking in isopropanol was thermally

controlled processes. Non-catalytic lignin liquefaction presented the similar oil yield

trend.

Catalysts were considered as another important influential parameter on the

products yields and the performance of different catalysts was summarized in Fig. 2b.

All catalysts except active carbon could improve the oil formation and Ru/C revealed

highest catalytic active. Similar results could also be found in other reported studies

(Patil et al., 2011).

Reaction time regarded as the third factor influenced on oil yield was investigated

from 1 h to 5 h (Fig. 2c). The oil yield increased continuously with the extension of

reaction time, but the upward trend slowed down over 4 h. Meanwhile, the char yield

9
indicated adverse trend and the minimum was received at 4 h, which meant the optimal

reaction time was 4 h.

Impact of lignin and catalysts concentration was studied from 1 to 3 wt % and 10

to 50 wt %, respectively (Fig. 2d and Fig. 2e). According to statistical results, these two

factors exhibited the minimal influence on the products yields. Comprehensive

consideration of bio-oil and char yields, 1 wt % lignin concentration and 40 wt %

catalysts concentration were decided as the optimal reaction condition.

The actual product yields were examined at 330 and 350 ºC based on the optimal

condition (0.1g kraft lignin, 0.04g catalysts, 12mL isopropanol and 4 h reaction time)

and results were listed in Table 1. At 350 ºC, the lignin conversion reached the

maximum (89.8%) over Ru/C. The reaction catalyzed by Pt/C, Ru/C and Rh/C received

the yield higher than 100%, which indicated that isopropanol reacted with lignin

through alkylation reaction as a reactant.

3.1.2. Effect of operating variables on elemental composition of bio-oil

The effects of operating variables on the heating values of liquid products were

studied through the same direct observation analysis methods and results were displayed

in Fig. 3. The detailed elemental composition of liquid products obtained from different

runs were summarized in Table S3. Similar to the tendency of operating variables

impact on products yields, temperature (Fig. 3a) was considered as the most important

factor for the products HHV, followed by the noble metal catalysts (Fig. 3b). HHV of

liquid products was soaring with the rise of temperature. In general, reaction times (Fig.

3c), lignin concentration (Fig. 3d) and catalyst concentration (Fig. 3e) had little effect

on the elemental composition of the oil. The best operating variable for yield and HHV

was 4 h, 1 wt % lignin concentration and 40 wt % catalyst concentration.

10
 Influences of different catalysts on the elemental composition of oil were tested at

330 and 350 ºC (Table 1). From 330 to 350 ºC, HHV of oil catalyzed by active carbon

exhibited an adverse result with non-catalyst, which suggested that the former had little

merit in hydrodeoxygenation. A plausible explanation was the deactivation of the

catalyst because the isopropanol decomposed to solid carbon (coke) that filled up the

pores of active carbon (Peng et al., 2014). However, the oil catalyzed by noble metal

catalysts received a much better result in O/C, H/C and HHV at the same reaction

conditions, which definitely suggested that the noble metal catalysts promoted

hydrodeoxygenation satisfactorily. In addition, Rh/C was inferior to Pt/C in

hydrogenation, but deoxygenation ability of Rh/C was superior to other catalysts

obviously. The maximum HHV 37.969 MJ/kg was obtained over Rh/C at 350 ºC after 4

h of reaction time, which was much higher than the literature published using ethanol or

water as medium (Patil et al., 2011).

3.1.3. Effect of catalysts on the distribution of molecular weight (Mw) of bio-oil

For further study the component of liquid products effected by catalysts, the gel

permeation chromatograms (GPC) was carried out to characterize the molecular weight

distribution of oil (Fig. S1). The broad peak of none at early elution volume suggested

the presence of a large number of components with a broad molecular weight

distribution originated from incomplete degradation of lignin. Furthermore, the

distribution of Mw continually centralized and moved away from large molecular weight

over the noble metal catalysts, and Rh/C catalyzed oil showed the lowest apparent Mw

values. That indicated that most of component of oil could be further analyzed by gas

chromatography (GC).

3.2. Molecular composition of bio-oil

11
3.2.1 Effect of catalysts on the distribution of molecular composition of bio-oil

The effects of catalysts on the molecular composition of oil were investigated by

GC. According to qualitative analysis of the compounds by GC-MS, the detailed mass

fractions of different catalyzed oils under 330/350 ºC were listed in Table S4,

which only contained 45 dominant constituents. Except that, the organic silicon

fraction from the column was removed and normalization method was applied to ensure

the total fraction percent up to 100%. The identified compounds were divided into

six categories: aliphatic hydrocarbon, oxygenic-chain compounds, phenols,

aromatic hydrocarbons, cycloalkanes and cyclohexanols/cyclohexanones. The

component classification summarized by the peak area of different compounds

was displayed in Table 2.

Phenols accounted for 75% of liquid products from non-catalyst at 330 ºC.

What’s more, the content of phenols decreased and aromatic compounds

appeared with increasing temperature. Meanwhile, the content of aliphatic

hydrocarbon slightly dropped from 16% to 11% and the content of

oxygenic-chain compounds dramatically raised from 9% to 31%. The content

changing tendency along with temperature of various categories from liquid

products catalyzed by active carbon was consistent with non-catalyst.

With the introduction of noble metal catalysts, phenols of oil dramatically

dropped and cyclohexanols/cyclohexanones appeared abundantly. Supported

noble metal catalysts have activity in deoxygenation process of lignin-derived phenolic

monomers into aromatic hydrocarbons which could transferred into cycloalkanes via

hydrogenation of benzene rings and in the hydrogenation of phenolic ring (Chen et al.,

2014; Hong et al., 2014) into staurted/unsaturated oxygenic hexatomic ring compounds.

12
The sum of aromatic hydrocarbons and cycloalkanes represented the catalytic

deoxygenation activity for phenolic monomers and the content of cyclohexanols and

cyclohexanones showed the catalytic hydrogenation activity for aromatic ring of various

noble metal catalysts, both of which meant the improvement of HHV of liquid products.

So, the least phenols and the most of aromatic hydrocarbons, cycloalkanes,

cyclohexanols and cyclohexanones were considered as the evaluation criterion of

activity of catalysts. Increasing temperature contributed to the phenols conversion

which reached the maximum (93.4 %) over Rh/C catalysts at 350 ºC, followed by Ru/C,

Pt/C and Pd/C. Similarly, the sum of aromatic hydrocarbons, cycloalkanes,

cyclohexanols and cyclohexanones reached the maximum (74.6 wt %) over Rh/C. In

conclusion, Rh/C was an effective hydrodeoxygenation catalyst for the conversion of

phenols.

Notably, rising temperature showed a significant impact on the content of

oxygenic-chain compounds in oil. With the introduction of noble metal catalysts,

oxygenic-chain compounds was much higher than non-catalyst at 330 ºC and

the highest came from Rh/C 47%. When the temperature reached 350 ºC, it

dropped obviously. It is the most important for catalysts to reduce the content of

oxygenic-chain compounds because of the harm of it to the oil quality. On

balance, Rh/C exhibited high catalytic activity for hydrocracking of lignin into

high-calories oil.

With these insights in hand, the most significant 15 species of the oil catalyzed by

Rh/C was quantitatively analyzed in optimal conditions through GC-FID by the use of

an internal standard, which accounted for 95% of the total integrated GC peak area for

all 45 detected products derived from lignin. The actual yields are listed in Table 3. The

13
overall yield of the 15 most abundant liquid products reached 114%.

3.2.2 1H-13C HSQC NMR of bio-oil over Rh/C

Through the above analysis, two conclusions could be drawn as follow: Firstly,

Rh/C showed an excellent activity of hydrodeoxygenation for the conversion of phenols.

Secondly, Rh/C preferably reduced the content of oxygenic-chain compounds in the oil.
1
H-13C heteronuclear signal quantum coherence (HSQC) NMR spectrometry of

product oil over Rh/C under the optimal conditions was applied to shed light on the

structural characteristics of oil and the spectrum was displayed in Fig. 4. Comparing

with the NMR spectrum of kraft lignin reported by other literature, the signals

belonging to the inter-unit linkages ( β-O-4, β-β, β-5, α-O-4 ) in the δ 3-5.5 ppm (1H)

and δ 50-90 ppm (13C) regions are lacking (Narani et al., 2015), which indicated that

interunit linkages of lignin cracked during the depolymerization process over Rh/C. No

cross-peaks assigned to methylene bridge isomers (δH 29-40 ppm and δC 3.5-4.0 ppm)

appeared in the spectrum, which confirmed the absence of phenol oligomerization over

Rh/C. Furthermore, no methoxy group signals remained practically in the NMR

spectrogram, which represented the success of demethoxylation of guaiacyl subunits.

Besides, signals corresponding to the aromatic oxygenic compounds (phenols, guaiacols

and alkylphenols) were observed between δH 6-8 ppm and δC 100-140 ppm. All signals

of aliphatic hydrocarbon, cyclohexane and cyclohexanols/cyclohexanones could

be observed in the HSQC spectrum. All results illustrated that lignin could be

depolymerized in a greater degree over Rh/C in isopropanol.

3.3. Possible catalytic process

According to the above results, a possible catalyzed hydrocracking pathway in

isopropanol was proposed in Fig. 5.

14
3.3.1 Formation process of oxygenic-chain compounds

The quantitative analysis of the liquid products reflected that the content of

oxygenic-chain compounds couldn’t be underestimated. It is essential to investigate the

route of the oxygenic-chain compounds formation that severely damaged the quality of

oil. Based on the literature reports, the thermal decomposition of isopropanol could be

converted into propene by dehydration, acetone by dehydrogenation and acetaldehyde

by demethanation (Ma et al., 2015). The ketone/enols were likely obtained from acetone

dimerization formed by pyrolysis of isopropanol through an aldol condensation reaction

(Trenwith, 1975). Therefore, the reactivity of isopropanol should be verified by heating

isopropanol under the optimal conditions firstly (Table 4). The quantified components

of the liquid products from isopropanol were listed in Table S5. Isopropanol produced a

large number of aliphatic hydrocarbon (C12-C26), small amount of 9-octadeceneamide

(C18) and few aliphatic acid/ester compounds (C 16 -C 18 ) which accounted for 71%,

17% and 11% of all products in 330 ºC, respectively. The produce of these high

molecular weight chemicals suggested that the polymerization occurred among propene,

acetone and acetaldehyde. When the temperature reached 350 ºC, aliphatic acid/ester

compounds increased accompanying with aliphatic hydrocarbon decreasing. The

overall yield of liquid products from isopropanol increased obviously with the

temperature increasing, which indicated the instability of isopropanol under the high

temperature environments. Notably, no linear/branched ketones/alcohols/ether (C6-C12)

produced during the whole heating process.

Secondly, the stability of solvent (isopropanol) over the different catalysts was

tested in 350 ºC and the results were listed in Table S5. In the reaction of active carbon,

the ratio of amide remained approximately 20%. The difference was that small amount

15
of ketones/alcohols/ether emerged with the decline of aliphatic acid/ester compounds.

However, more alcohols/ketones were received over noble metal catalysts, which

showed that the dehydrogenation of isopropanol was promoted by noble metal catalysts.

Aliphatic hydrocarbon and acid/ester compounds declined obviously, which

reflected that the polymerization among propene, acetone and acetaldehyde were

inhibited by noble metal catalysts. In addition, all noble metal catalysts inhibited the

formation of amide efficiently. Nevertheless, the overall yield of liquid products from

isopropanol decreased more or less in the presence of noble metal catalyst. Both the

minimum of liquid yield and the least oxygenic-chain compounds were observed in

Rh/C, which showed that isopropanol was the most stable and favorable to HHV of oil

over Rh/C.

Comparing with the presence of lignin, less aliphatic hydrocarbon and more

oxygenic-chain compounds were obtained. The former demonstrated that decomposed

lignin contributed to inhibit the polymerization of low molecular products (propene,

acetone and acetaldehyde) that could involve in the formation of cyclic compounds as

alkyl-substituents during the catalytic hydrocracking of kraft lignin. The latter suggested

that a part of oxygenic-chain compounds could be received by the cracking of lignin.

3.3.2 Hydrodeoxygenation process of phenols

Phenols occupied the major part of bio-oil and originated from the cracking of

kraft lignin mainly. Typical hydrogenated compounds including cyclohexanols and

cyclohexanones appeared in the noble metal catalyzed oils. The conversion of phenols

into cyclohexanols /cyclohexanones reached the maximum over Ru/C, but none

cycloalkanes could be detected in corresponding oils. A possibility explaination was

that the catalytic hydrogenation process of phenols halted on

16
cyclohexanols/cyclohexanones and further dehydration of cyclohexanol were not

implemented over Ru/C, which led to the higher oxygen content oil. No enough acidic

condition provided by the isopropanol system might be an issue (Zhao et al., 2009).

Phenols, as a typical primary product emerged in product oil, deoxygenation

pathways could be realized by the transalkylation of the methoxy group or

demethylation at the beginning and followed by the dehydration steps into aromatic

hydrocarbons over the noble metal catalysts. That process required more harsh reaction

condition, which was consistent with the effect of temperature on the molecular

composition of bio-oil. Therefore, the degree of deoxygenation was lower than the

hydrogenation of phenols. For the oil catalyzed by Rh/C, aromatic hydrocarbons

conversion via the deoxygenation of phenols reached 20.3% in 350 ºC, which was 10.5 %

higher than Ru/C. It was suggested that the deoxygenation activity of the front through

the removal of phenolic hydroxyl groups was relatively powerful. Except that,

cycloalkanes could be obtained by two pathways: dehydration of cyclohexanols and

hydrogenation of aromatic hydrocarbons. Both of these facilitated the improvement of

oil heating values. From Table 2, the total of cycloalkanes from Rh/C was higher than

Ru/C as well.

Integrated all the data, Rh/C was an efficient catalyst for obtaining high-calories

bio-oil as a result of the effective reduction of oxygenic-chain compounds, the workable

hydrogenation of phenolic rings and the feasible deoxygenation of phenolic hydroxyl

group.

4. Conclusions

Efficient catalytic hydrocracking of kraft lignin to yield high-quality bio-oil was

demonstrated in isopropanol system. The optimal condition was received by the OAD

17
analysis and hydrocracking process was thermally controlled. An extensive analysis of

the liquid products was performed using GC-MS/FID, elemental analysis, GPC and
1
H-13C HSQC NMR. Different from Pd/C and Pt/C catalysts, Ru/C and Rh/C inhibited

the formation of oxygenic-chain compounds along with the temperature increasing. The

maximum phenol conversion of 93.4% and the maximum HHV of 37.969 MJ/kg were

achieved over Rh/C. The routes of oxygenic-chain compounds formation and phenol

conversion were proposed in detail.

Acknowledgments

This work was finally supported by the key project of Shanghai Science and

Technology Committee (No. 14231200300).

References

1. Bengoecheaa, M.O., Hertzberg, A., Mileti′cb, N., Ariasb, P.L., Bartha, T., 2015.

Simultaneous catalytic de-polymerization and hydrodeoxygenation of lignin in

water/formic acid media with Rh/Al2O3, Ru/Al2O3 and Pd/Al2O3 as bifunctional

catalysts. J. Anal. Appl. Pyrolysis. 113, 713–722.

2. Corma, A., Lborra S., Velty, A., 2007. Chemical routes for the transformation of

biomass into chemicals. Chem. Rev. 107, 2411-2502.

3. Cheng, S.N., Wilks, C., Yuan, Z.S., Leitch, M., Xu, C., 2012. Hydrothermal

degradation of alkali lignin to bio-phenolic compounds in sub/supercritical ethanol and

water–ethanol co-solvent. Polym. Degrad. Stab. 97, 839-848.

4. Chen, W., Luo, Z.Y., Yu, C.J., Yang, Y., Li, G.X., Zhang, J.X., 2014. Catalytic

conversion of guaiacol in ethanol for bio-oil upgrading to stable oxygenated organics.

Fuel Process. Technol. 126, 420-428.

5. Chang, Z.F., Duan, P.G., Xu, Y.P., 2015. Catalytic hydropyrolysis of microalgae:

18
influence of operating variables on the formation and composition of bio-oil. Bioresour

Technol. 184, 349-354.

6. Ferrini, P., Rinaldi, R., 2014. Catalytic biorefining of plant biomass to non-pyrolytic

lignin bio-oil and carbohydrates through hydrogen transfer reactions. Angew. Chem. Int.

Ed. 53, 8634-8639.

7. Gosselink, R.J.A., de Jong, E., Guran, B., Abächerli, A., 2004. Co-ordination

network for lignin—standardisation, production and applications adapted to market

requirements (EUROLIGNIN). Ind. Crops Prod. 20, 121-129.

8. Hong, Y.K., Lee, D.W., Eom H.J., Lee, K.Y., 2014. The catalytic activity of

Pd/WOx/γ-Al2O3 for hydrodeoxygenation of guaiacol. Appl. Catal. B. 150–151,

438-445.

9. Huang, X., Koranyi, T.I., Boot, M.D., Hensen, E.J.M., 2014. Catalytic

depolymerization of lignin in supercritical ethanol. ChemSusChem 7, 2276-2288.

10. Huang, X., Korányi, T.I., Boot, M.D., Boot, M.D., Hensen, E.J.M., 2015. Ethanol as

capping agent and formaldehyde scavenger for efficient depolymerization of lignin to

aromatics. Green Chem. 17, 4941-4950.

11. Kloekhorst, A., Shen, Y., Yie, Y., Fang, M., Heeres, H.J., 2015. Catalytic

hydrodeoxygenation and hydrocracking of Alcell® lignin in alcohol/formic acid

mixtures using a Ru/C catalyst. Biomass Bioenerg. 80, 147-161.

12. Lin, Y., Li, C., Wan, H., Lee, H., Liu, C., 2011. Catalytic hydrodeoxygenation of

guaiacol on Rh-based and sulfided CoMo and NiMo catalysts. Energy Fuels 25,

890–896.

13. Li, C.Z., Zhao, X., Wang A.Q., Huber, G.W., Zhang, T., 2015. Catalytic

transformation of lignin for the production of chemicals and fuels. Chem. Rev. 115,

19
11559-11624.

14. Mortensen, P.M., Grunwaldt, J.D., Jensen, P.A., Knudsen, K.G., Jensen, A.D., 2011.

A review of catalytic upgrading of bio-oil to engine fuels. Appl. Catal. A. 407, 1–19.

15. Ma, R., Hao, W., Ma, X., Tian, Y., Li, Y., 2014. Catalytic ethanolysis of kraft lignin

into high-value small-molecular chemicals over a nanostructured alpha-molybdenum

carbide catalyst. Angew. Chem. Int. Ed. 53, 7310-7315.

16. Ma, R., Cui, K., Yang, L., Ma, X. L., Li, Y.D., 2015. Selective catalytic conversion

of guaiacol to phenols over a molybdenum carbide catalyst. Chem. Commun. 51,

10299-10301.

17. Nguyen, T.D., Maschietti, M., Mand, L.E., Vamling, L., Olausson, L., Andersson,

S.I., Theliander, H., 2014. The effect of temperature on the catalytic conversion of kraft

lignin using near-critical water. Bioresour. Technol. 170, 196-203.

18. Narani, A., Chowdari, R.K., Cannilla, C., Bonura, G., Frusteri, F., Heeres, H.J.,

Barta, K., 2015. Efficient catalytic hydrotreatment of kraft lignin to alkylphenolics

using supported NiW and NiMo catalysts in supercritical methanol. Green Chem. 17,

5046-5057.

19. Patil, P.T., Armbruster, U., Richter, M., Martin, A., 2011. Heterogeneously

Catalyzed Hydroprocessing of Organosolv Lignin in Sub- and Supercritical Solvents.

Energy Fuels 25, 4713-4722.

20. Peng, G., Steib, M., Gramm, F., Ludwig, C., Vogel, F., 2014. Synthesis factors

affecting the catalytic performance and stability of Ru/C catalysts for supercritical water

gasification. Catal. Sci. Technol. 4, 3329-3339.

21. Ruddy, D.A., Schaidle, J.A., Ferrell Iii, J.R., Wang, J., Moens, L., Hensley, J.E.,

2014. Recent advances in heterogeneous catalysts for bio-oil upgrading via “ex

20
situ catalytic fast pyrolysis”: catalyst development through the study of model

compounds. Green Chem. 16, 454-490.

22. Saidi, M., Samimi, F., Karimipourfard, D., Nimmanwudipong, T., Gates, B.C.,

Rahimpour, M.R., 2014. Upgrading of lignin-derived bio-oils by catalytic

hydrodeoxygenation. Energy Environ. Sci. 7, 103–129.

23. Singh, R., Prakash, A., Dhiman, S.K., Balagurumurthy, B., Arora, A.K., Puri, S.K.,

Bhaskar, T., 2014. Hydrothermal conversion of lignin to substituted phenols and

aromatic ethers. Bioresour Technol. 165, 319-322.

24. Schutyser, W., Bosch, S.V.D., Renders, T., Boe, D., Koelewijn, S.F., Dewaele, A.,

Ennaert, T., Verkinderen, O., Goderis, B., Courtinc C.M., Sels, B.F., 2015. Influence of

bio-based solvents on the catalytic reductive fractionation of birch wood. Green Chem.

17, 5035-5045.

25. Trenwith, A.B., 1975. Thermal decomposition of isopropanol. J. Chem. Soc.,

Faraday Trans.1 71, 2045-2050.

26. Wildschut, J., Mahfud, F.H., Venderbosch, R.H., Heeres, H.J., 2009.

Hydrotreatment of fast pyrolysis oil using heterogeneous noble-metal catalysts. Ind.

Eng. Chem. Res. 48, 10324–10334.

27. Wang, X., Rinaldi, R., 2012a. Solvent effects on the hydrogenolysis of diphenyl

ether with raney nickel and their implications for the conversion of lignin.

Chemsuschem. 5, 1455-1466.

28. Wang, X., Rinaldi, R., 2012b. Exploiting H-transfer reactions with RANEY® Ni for

upgrade of phenolic and aromatic biorefinery feeds under unusual, low-severity

conditions. Energy Environ. Sci. 5, 8244-8260.

29. Wang, X., Rinaldi, R., 2013. A route for lignin and bio-oil conversion:

21
dehydroxylation of phenols into arenes by catalytic tandem reactions. Angew. Chem.

Int. Ed. 52, 11499-11503.

30. Warner, G, Hansen, T.S., Riisager, A., Beach, E.S., Barta, K., Anastas, P.T., 2014.

Depolymerization of organosolv lignin using doped porous metal oxides in supercritical

methanol. Bioresour. Technol. 161, 78-83.

31. Xu, C.P., Arancon, R.A.D., Labidi, J., Luque, R., 2014. Lignin depolymerisation

strategies: towards valuable chemicals and fuels. Chem. Soc. Rev. 43, 7485-7500.

32. Zhao, C., Kou, Y., Lemonidou, A.A., Li, X.B., Lercher, J.A., 2009. Highly selective

catalytic conversion of phenolic bio-oil to alkanes. Angew. Chem. Int. Ed. 48,

3987-3990.

33. Zhao, Y., Xu, Q., Pan, T., Zuo, Y., Fu Y., Guo Q.X., 2013. Depolymerization of

lignin by catalytic oxidation with aqueous polyoxometalates. Appl. Catal. A. 467,

504-508.

22
Figure Captions

Fig. 1. Workup procedure of reaction product mixture from catalytic hydrocracking of

kraft lignin.

Fig. 2. Effect of operating variables on the yield of bio-oil and char.

Fig. 3. Effect of operating variables on the HHV of bio-oil.

Fig. 4. 1H-13C HSQC NMR spectra of DCM soluble fraction obtained after catalytic

treatment using Rh/C catalyst. Reaction condition: 0.1 g kraft lignin, 12 mL isopropanol,

0.04 g catalyst, 4 h, 350 ºC.

Fig. 5. Possible catalytic process over noble metal catalysts.

23
Table 1. Catalytic activities of runs in isopropanol system, elemental composition, HHV of products, and bio-oil yields at 330 and 350 ºC.

Reaction condition: 0.1 g kraft lignin, 12 mL isopropanol, 0.04 g catalyst, 4 h.

Y Bio-oil Xlignin
C H N O O/C H/C HHV (MJ/kg)
(%) (%)

Kraft Lignin 47.680 5.111 0.088 47.122 0.741 1.284 15.976 — —

330 ºC

No Catalyst 75.180 7.664 0.158 16.998 0.170 1.223 33.450 37.4 54.8

Pd/C 75.920 7.986 0.161 15.933 0.157 1.262 34.320 48.9 69.1

Pt/C 76.890 8.964 0.195 13.952 0.136 1.399 36.336 72.2 79.3

Ru/C 77.610 8.373 0.146 13.871 0.134 1.295 35.757 83.1 86.9

Rh/C 80.780 8.178 0.101 10.941 0.102 1.215 36.999 80.0 80.8

C 74.205 7.609 0.295 17.892 0.181 1.230 32.888 39.3 59.5

350 ºC

No Catalyst 78.700 7.295 0.144 13.861 0.132 1.112 34.590 47.1 65.8

Pd/C 79.470 8.190 0.213 12.127 0.114 1.237 36.378 70.5 82.6

Pt/C 78.700 9.275 0.158 11.867 0.113 1.414 37.712 >100 84.3

Ru/C 78.820 8.911 0.135 12.134 0.115 1.357 37.197 >100 89.8

Rh/C 80.838 8.775 0.228 10.162 0.094 1.303 37.969 >100 86.4

C 77.390 7.419 0.258 14.934 0.145 1.150 34.145 60.6 72.9

24
Table 2. Component classification based on peak area of bio-oil in isopropanol system at 330 and 350 ºC. Reaction condition: 0.1 g kraft

lignin, 12 mL isopropanol, 0.04 g catalyst, 4 h.

No catalyst C Pd/C Pt/C Ru/C Rh/C

330 ºC

Aliphatic hydrocarbon 14.7 19.1 8.1 11.9 0.7 1.3

Oxygenic-chain compounds 8.5 6.7 31.7 16.3 32.1 46.8

Phenols 67.8 62.6 33.2 11.1 10.0 6.5

Aromatic hydrocarbons 0 2.8 2.8 11.2 6.6 3.9

Cycloalkanes 0 0 0 2.6 6.7 8.8

Cyclohexanols, Cyclohexanones 0 2.0 14.9 42.1 41.0 31.9

350 ºC

Aliphatic hydrocarbon 10.2 22.8 6.5 5.2 8.7 0.5

Oxygenic-chain compounds 28.9 15.0 36.0 25.6 11.5 18.2

Phenols 49.7 51.9 19.8 7.4 6.2 5.3

Aromatic hydrocarbons 4.2 2.4 4.2 18.0 7.7 16.2

Cycloalkanes 0 0 8.5 11.8 0.7 7.0

Cyclohexanols, Cyclohexanones 0 0 25.2 26.6 63.8 51.4

25
Table 3. Quantified the main components of Rh/C catalyzed bio-oil and their mass yields (mg per 0.1 g lignin) at 350 ºC for 4 h.

No. Compounds Yield No. Compounds Yield No. Compounds Yield

1 4-Heptanone, 2-methyl- 2.9 6 Cyclohexanone, 3,3,5-trimethyl- 19.6 11 Isophorone 24.4

Benzene,
2 Benzene, 1,2,3-trimethyl- 9.4 7 Cyclohexanol, 3,3,5-trimethyl- 16.3 12 4.6
1,4-dimethyl-2-(2-methylpropyl)-

3 5-Tridecanol 2.7 8 2-Cyclohexen-1-ol, 3,5,5-trimethyl 2.3 13 Heptyl isobutyl ketone 5.7

Cyclohexane,
4 4.6 9 2-Cyclohexen-1-one, 3,5-dimethyl- 1.7 14 Phenol, 3,4-dimethyl- 5.2
1,1,3,5-tetramethyl-,cis-

2-Butenoic acid, 2-methyl-,

5 8.1 10 Phorone 2.7 15 Cyclohexene, 3,5,5-trimethyl- 3.6
2-propenyl ester, (E)-

26
Table 4. Quantified components classification (mg) of the liquid products of the reaction of isopropanol (12 mL) and the different catalysts

(0.04 g) without kraft lignin at 350 ºC for 4 h.

No Catalysts C Pd/C Pt/C Ru/C Rh/C

Aliphatic hydrocarbon (C12 -C26) 23.3 (26.6a) 24.9 14.1 11.2 15.9 10.8

Aliphatic acid/ester (C 16-C18) 18.6 (4.3 a) 11.1 10.4 4.9 0.9 1.4

9-Octadeceneamide (C18) 9.7（6.5 a） 10.3 1.1 1.0 0.9 1.1

Linear/branched ketones/alcohols/ether (C6-C 12) 0 (0 a) 6.1 22.5 18.7 15.4 15.2

a
Blank test (without lignin) was conducted in 330 ºC.

27
Fig. 1. Workup procedure of reaction product mixture from catalytic hydrocracking of kraft lignin.

28
29
Fig. 2. Effect of operating variables on the yield of bio-oil and char using direct observation analysis method of the OAD.

30
a b
34.636

33.500
Oil HHV (MJ/Kg)

Oil HHV (MJ/Kg)

32.51 32.722
32.319
31.830 31.787

30.530 30.368

29.210

270 290 310 330 350 Pd/C Pt/C Ru/C Rh/C C

Temperature/ ºC Catalyst

c d

Oil HHV (MJ/Kg)

Oil HHV (MJ/Kg)

32.219 32.434
32.088 32.125 32.132 31.886
31.616 31.660 31.596 31.658

1 2 3 4 5 1 1.5 2 2.5 3
Time/ h Lignin concentration/ wt %

31
e
Oil HHV (MJ/Kg)

32.152 32.071 32.089

31.738 31.657

10 20 30 40 50
Catalyst concentration/ wt %

Fig. 3. Effect of operating variables on the HHV of bio-oil using direct observation analysis method of the OAD.

32
Fig. 4. 1H-13C HSQC NMR spectra of DCM soluble fraction obtained after catalytic treatment using Rh/C catalyst. Reaction condition: 0.1

g kraft lignin, 12 mL isopropanol, 0.04 g catalyst, 4 h, 350 ºC.

33
Fig. 5. Possible catalytic process over noble metal catalysts.

34
HIGHLIGHTS

1. High-quality bio-oil was obtained by using isopropanol as the solvent.

2. This reaction was thermally controlled after the orthogonal array design analysis.

3. The aim of reducing oxygenic-chain compounds was realized over Rh/C.

4. Maximum phenol conversion (93.4%) and HHV (37.969 MJ/kg) were observed over Rh/C.

5. A possible catalytic hydrocracking pathway was proposed.

35