Fuel 162 (2015) 47–54

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Reforming of lignin-derived tars over char-based catalyst

using Py-GC/MS
Kezhen Qian, Ajay Kumar ⇑
Department of Biosystems and Agricultural Engineering, Oklahoma State University, Stillwater, OK, United States

h i g h l i g h t s

Catalytic reforming of lignin tar was performed in a Py-GC/MS.

Gasification-derived char was used as precursor of reforming catalyst.

Effects of temperature, pressure and atmosphere on reforming of tar were assessed.

Char-based catalyst effectively decreased the lignin tar.

a r t i c l e i n f o a b s t r a c t

Article history: Tar removal is one of the major challenges in implementation of biomass gasification technology. Syngas
Received 11 June 2015 tars causes formation of aerosols and soots, which plug filters, reactors and fuel lines. In this study, a
Received in revised form 17 August 2015 char-derived catalyst was tested in removal of tar produced from pyrolysis of kraft lignin in a pyroprobe
Accepted 21 August 2015
reactor. The effects of reaction temperature (700, 800 and 900 °C), water amount (5–10 ll), pressure
Available online 2 September 2015
(0.1–2.2 MPa) and atmosphere (inert and hydrogen) on catalytic conditioning of tar components were
assessed. The tar components were analyzed by GC/MS. Catechols were the most abundant tar compo-
Keywords:
nents followed with phenols and guaiacols during non-catalytic kraft lignin pyrolysis. Results indicated
Lignin
Char
that the char-based catalyst effectively decreased the contents of lignin tar. Reaction temperature, water
Activated carbon loading and reaction pressure significantly affected the tar removal. An increase in reaction temperature
Catalyst led to an increase in removal efficiency of most tar components except naphthalene. Excessive water
Tar removal loading (10 ll) decreased the tar removal efficiency of char-based catalyst. High pressure promoted
Py-GC/MS the catalytic conversion of lignin tar. Tar contents decreased significantly when hydrogen was used as
a reaction medium and thus promoted the conversion of lignin into non-condensable gas.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction technology at commercial scales for fuels, chemicals and power

production.
Increase in global greenhouse gas emissions and concerns about Biomass is mainly composed of cellulose, hemicellulose and lig-
global fossil fuels reserves have promoted the research in renew- nin. Many studies have shown that cellulose, hemicellulose and
able energy. Biomass gasification is one type of efficient renewable lignin produce different tar compounds [1,2]. Primary tars pro-
energy technology converting lignocellulosic solid feedstocks into duced from cellulose are furans and small molecule aldehydes
combustible gas. However, during gasification many contaminants [3]. Primary tars produced from hemicellulose are acetic acid [2]
are generated, such as NOx, SOx and tar. Particularly, the presence and those produced from lignin are furfurals and phenolics [4].
of considerable amount of tar in syngas leads to formation of aero- Approximately 20–40 wt.% of biomass, such as switchgrass, wheat
sols and soots due to repolymerization and plugs filters and fuel straw and eastern red is composed of lignin [5]. Lignin is a complex
lines due to tar condensation [1]. Therefore, tar removal is one of polymer of p-hydroxyphenyl, guaiacyl and syringyl alcohols. Three
the major challenges in implementation of biomass gasification species of hydroxycinnamyl alcohols (p-coumaryl alcohol, coni-
feryl alcohol and sinapyl alcohol) are considered as monoligol
monomers incorporated in lignin polymer structure in the form
⇑ Corresponding author at: 228 Agricultural Hall, Biosystems and Agricultural
Department, Biobased Products and Energy Center, Oklahoma State University,
of p-hydroxyphenyl, guaiacyl, and syringyl phenylpropanoid
Stillwater, OK 74078, United States. [1,6]. Since only the lignin fraction of the biomass is aromatic in
E-mail address: ajay.kumar@okstate.edu (A. Kumar). nature, lignin represents a potential precursor for formation of

http://dx.doi.org/10.1016/j.fuel.2015.08.064
0016-2361/Ó 2015 Elsevier Ltd. All rights reserved.
48 K. Qian, A. Kumar / Fuel 162 (2015) 47–54

polyaromatic hydrocarbon (PAH) in tar. The study of catalytic 2.2. Pyrolysis of kraft lignin using Py-GC/MS
lignin-derived tar reforming is critical for understanding strategies
to reduce syngas tar. Catalytic reforming was performed using a commercial micro-
The tar formation is affected by reaction conditions such as pyrolyzer (Pyroprobe model 5200/high pressure, CDS Analytical
pressure. The pressure will affect composition of syngas as well Inc., Oxford, PA). The pyrolyzer was connected with a gas chro-
tar. Knight [7] studied the effect of pressure on the biomass gasifi- matograph/mass spectrometer (GC/MS, Agilent 7893, Agilent,
cation products and found that increasing pressure (from 0.8 to Santa Clara, CA, USA). The probe of Pyroprobe consisted of a plat-
2.2 MPa) decreased oxygenated species. Specifically, particularly inum heating coil, which can be heated up to 1400 °C. The lignin,
phenols were almost completely eliminated; but the PAH fractions named as Indian AT, (Mead Westvaco, Richmond, VA, USA). Indian
increased. Research was also conducted on pressurized gasification AT is a purified form of kraft pine lignin. It is derived by hydrolysis
of coal [8]. Pressurized operation not only reduced the volatile evo- of kraft lignin, removing the sodium the hemicellulose [17]. About
lution during coal pyrolysis and increased char gasification rate by 0.5 mg of the lignin powder and 5 mg of catalyst were packed in a
influencing the physical structure of pyrolysis char, but also low- quartz tube (approximately 25 mm long and 1.9 mm inner diame-
ered the energy cost for compressing syngas prior to using in a ter), which was then held in the platinum heating coil. To make
gas turbine [8]. sure all tar volatiles passed through the catalyst layer, one layer
The tar formation is also affected by reaction atmosphere such of catalyst was kept on each side of the lignin powder. The catalyst
as hydrogen. Many studies have found that hydrogen atmosphere layer and lignin layer were separated by quartz wool. In order to
promotes hydrocracking and deoxygenation reactions of biomass simulate steam gasification, 5 ll water was injected into lignin
pyrolysis, thus decreasing the oxygen content of bio-oil while powder during non-catalytic pyrolysis. The sample was pyrolyzed
increasing the gas yield [9,10]. Thangalazhy-Gopakuma et al. [9] at a heating rate of 1500 °C/s in presence of different gases (He,
studied the hydrogen effect on formation of oxygenated com- 100% and H2 100%). To make sure the sample was completely pyr-
pounds during pine wood pyrolysis. They found that the presence olyzed, the sample was held at pyrolysis temperature for 20 s. To
of hydrogen significantly enhanced hydrodeoxygenation, which use helium as reactant gas, 40 ml/min helium was purged for about
rejected the bio-oil oxygen in the form of water [9]. As a result, 1 min to remove air. To use H2 as reactant gas, 40 ml/min helium
yield of higher molecular weight oxygenated compounds under was also purged for about 2 min to remove air and residual hydro-
H2 atmosphere was much lower than that under helium atmo- gen before switching to reactant gas (H2) with flow rate of 40 ml/
sphere [9]. Takarada et al. [11] studied coal pyrolysis in hydrogen min. The actual temperature of biomass inside the quartz tube
and helium in a fluidized bed. The yields of permanent gases in was, typically, about 50–100 °C lower than that of the filament
hydrogen were about five times than those in helium. The yields temperature [18,19].
of aromatic hydrocarbon (benzene, toluene and naphthalene,
etc.) also enhanced. 2.3. Tar composition analysis
Char is a byproduct of gasification and pyrolysis. Char’s variable
and tailorable surface properties (surface functional groups, poros- The reactant gas carried the pyrolysis vapors (tars) from the
ity and surface area) offer great potential value for its applications probe to a trap (adsorbent). The trap adsorbed the condensed
in renewable energy, agriculture and environmental remediation, vapors. Non-condensable gases escaped from the trap and were
etc. [12,13]. As for syngas cleaning, char and char derived catalysts not analyzed in this study. The adsorbed tar was desorbed by heat-
were proved to be effective alternatives to traditional catalysts for ing the trap to 300 °C and purging with helium. The gaseous tar was
tar removal [14,15]. However, limited studies have investigated then carried by helium gas and injected into the GC/MS through a
the reaction conditions, such as pressure and atmosphere on the transfer line for compositional analysis. The transfer line was
formation of lignin tar and the performance of char based catalyst heated at 300 °C to prevent tar condensation. A capillary column
in catalytic tar removal. (HP-5, 0.03 mm OD, 3 m length) was installed in the GC for separat-
The objective of this study was to study the effects of reaction ing the tar components. The injector of the GC was held at 250 °C.
conditions, such as pressure and atmosphere on non-catalytic The column temperature was maintained at 40 °C for 2 min and
and catalytic reforming of lignin tar. The catalytic and non- then increased to 280 °C with a heating rate of 5 °C /min. Helium
catalytic pyrolysis of kraft lignin was conducted at 700–900 °C of ultra-high purity (99.999%) was used as a carrier gas at a flowrate
and 0–300 psig in a pyroprobe reactor in presence of 5 or 10 ll of 1.25 mL/min.
water. Char-based catalyst was used for catalytic pyrolysis. Since The mass spectrometer was configured for electron impact ion-
hydrogen is the primary component of syngas and is found to have ization at 70 eV, with an interface temperature of 250 °C. Tar com-
significant effect on tar formation, the effect of hydrogen on tar pounds were identified by comparing the mass spectra with the
cracking was also investigated. NIST (National Institute of Standards and Technology) mass spec-
tral library and the retention time of the standard compounds.
The concentration of tar components were determined using an
2. Material and methods external standard method. 26 tar components were quantified
using 24 external standards including 10 aromatic hydrocarbons,
2.1. Catalyst preparation 8 phenols, 6 phenol-guaiacols, 1 furan and benzoic acid.
2-methyl-phenol and 4-ethylcatechol were quantified using corre-
Catalysts were prepared by loading nickel on char-derived acti- sponding standard from similar structure compounds [20] (P-cresol
vated carbon. The char was produced by downdraft gasification of for 2-methyl-phenol and methylcatechol 4-ethylcatechol).
red cedar [16]. The red cedar char was then mixed with KOH
(Fisher Scientific, Pittsburgh, PA, USA) and activated at 800 °C for 2.4. Experimental design and data analysis
1.5 h under nitrogen flow of 200 ml/min in a fixed-bed tubular
reactor. Activated carbon obtained was then impregnated with For non-catalytic pyrolysis of lignin, a full factorial design was
nickel nitrate solution (Sigma Aldrich, St. Louis, MO, USA) and then performed at five pressures of 0, 0.5, 0.8, 1.1 and 2.2 MPa (0, 50,
dried at 105 °C for 3 h. The dried catalyst was reduced in 100 ml/ 100, 150 and 300 psig) and three temperatures of 700, 800 and
min hydrogen flow at 350 °C for 3 h. Reduced catalyst was then 900 °C. For catalytic pyrolysis of lignin, a full factorial design was
kept in vacuum desiccator for further test. performed at three pressures of 0.1, 0.5 and 1.1 MPa (0, 50 and
 K. Qian, A. Kumar / Fuel 162 (2015) 47–54 49

150 psig) and three temperatures of 700, 800 and 900 °C. To study The composition of tar obtained from kraft lignin in this study
the effect of water loading on catalytic performance, 5 and 10 ll was significantly different as compared to that obtained from
were used at the three temperatures (700, 800 and 900 °C). The lig- pyrolysis of lignin reported in other studies [24,25]. In lignin
nin was also gasified under hydrogen atmosphere at 800 °C to tar [24,25], syringols and guaiacols were the most abundant
study the hydrogen effect on catalyst performance. components followed by phenols. Catechols, one of the sub-
The statistical analysis was performed using JMP software (SAS species of phenols, were scarcely detected. In this study, tars
Institute Inc. Cary, NC, USA). Results were analyzed at a = 0.05 sig- were dominated by phenols, especially catechols. This difference
nificance level. Polynomial surfaces were plotted using MATLAB may be caused by high reaction temperature (700–900 °C as
(Mathworks, Natick, MA, USA). The surfaces were generated compared to 400–600 °C in literature) and water used in this
approximating the tar contents to a quadratic polynomial of tem- study. Hu et al. [6] and Jiang et al. [23] found that high temper-
perature and pressure. The coefficients for the respective surfaces ature (>600 °C) promoted the demethoxylation and demethyla-
are given in Fig. 2. tion reactions of methoxyl groups, resulting in generation of
phenols, such as phenol, cresol and catechols (see Fig. S2,
3. Results and discussions Supplementary material). The temperature (700–900 °C) applied
in this study may have favored the demethoxylation of guaiacols
The composition of tar produced from pyrolysis of lignin is into phenol. On other hand, H donor provided by water could be
complex. Tar compounds range from simple single-ring aromatics stabilized by aromatic O radical generated from hemolytic crack-
to polycyclic aromatics. This paper, however, only analyzed and ing of aromatic O–CH3 resulting in catechols formation (see
discussed compounds with boiling point below 250 °C which is Fig. S1, Supplementary material). The lack of syringols in tar
detectable by GC/MS. 46 major tar components are listed in may be related to the source of lignin. The syringols were not
Table S1 (Supplementary material) with name and family group. observed in the tar produced from pyrolysis and hydrogenolysis
Out of 46 listed, 26 compounds were quantified using external of pine lignin either [4,26]. Thangalazhy [26] attributed the
standards (see Table S1). In the discussion that follows, the tar con- absence of syringols in tars to the missing sinapyl alcohol struc-
tent for a compound was defined as tures in pine wood.
  The variations in total contents of monoaromatic hydrocarbons,
mass of tar lg polyaromatic hydrocarbons, phenols, and guaiacols with respect to
T¼
mass of lignin mg lignin reaction temperature and pressure are depicted in Figs. 2 and 3.
Statistical analysis showed that temperature (p-value < 0.05) and
The removal percentage was defined as pressure (p-value < 0.05) significantly affected production of
T0  T monoaromatic hydrocarbons, polyaromatic hydrocarbons and phe-
Removal percentage ð%Þ ¼  100 nols. However, guaiacols was statistically significantly affected
T0
only by pressure (p-value < 0.05) but not by temperature
where T0 represents content of tar produced from lignin pyrolysis (p-value > 0.05). The interaction of temperature and pressure was
with no catalyst at a specific reaction temperature; and T represents significant on phenols (p-value < 0.05) but not statistically signifi-
content of tar produced from lignin with catalyst at the specific cant on monoaromatic hydrocarbons, polyaromatic hydrocarbons
reaction temperature. and guaiacols (p-value > 0.05).
Second order polynomial regression was applied to generate
3.1. Non-catalytic lignin gasification response surfaces of monocyclic aromatic hydrocarbons and poly-
cyclic aromatic hydrocarbons as functions of temperature and
Contents of aromatic and phenolic tar compounds obtained at pressure (Fig. 2). Similar shapes of surface plots of monoaromatic
700–900 °C pyroprobe temperature and 0.1 MPa (0 psig) are pre- hydrocarbons and polyaromatic hydrocarbons showed that pres-
sented in Fig. 1. The bar height and the error bar represent average sure and temperature have similar effects of monoaromatic and
and standard deviation of two replications, respectively. As seen polyaromatic hydrocarbons. At a given temperature both the con-
from Fig. 1, the most abundant tar components were phenols, such tents of monoaromatic and polyaromatic hydrocarbons reached a
as catechol, 4-methylcatechol and phenol. The second most peak at 1.1 MPa (150 psig) and with increasing temperature the
abundant tar components were guaiacols, including creosol, total contents further increased. The highest contents of monoaro-
2-methoxyphenol and 2-methoxy-4-vinylphenols. Polyaromatic matic hydrocarbons (7.2 lg/mg lignin) and polyaromatic hydrocar-
hydrocarbons were the least abundant in tar. Syringol groups bons (2 lg/mg lignin) were observed at 900 °C and 1.1 MPa (150
derived from sinapyl alcohol monomers were not detected in kraft psig).
lignin tar. High fractions of phenols and guaiacols originated from The optimum pressure for maximum polyaromatic hydrocar-
phenolic monomers present in the lignin polymer. Temperature bon yield at all temperatures was 1.1 MPa (150 psig). The effects
significantly affected the tar composition. Most of the phenolic of pressure on the polyaromatic hydrocarbon were explained by
components decreased when temperature increased from 700 to Mayerhofer et al. [27]. On one hand, increase in system pressure
900 °C, while the aromatic hydrocarbons increased with increasing caused the reaction equilibrium to shift to fewer molecules based
temperature. The decrease of phenolic components with increasing on equilibrium law. In order to have fewer molecules in the sys-
temperature was consistent with data from lignin pyrolysis in tem, polyaromatics were favored at high pressure as these have
other studies [21–23]. In Jiang’s study [23], the total yield of phe- the higher aromaticity. On the other hand, high pressure sup-
nols and yields of important phenolics (guaiacol, syringol, pressed the volatilization of tar and extended tar residence time,
alkylphenol and catechol) from Asian lignin reached maximum at which promoted polymerization reactions forming polyaromatic
600 °C followed by a decrease with further increase in temperature hydrocarbons. However, after the pressure reached a certain value
to 800 °C. Similarly, in another study of noncatalytic pyrolysis of (1.1 MPa in our study) leading to the maximum residence time
prarie cordgrass lignin [21], most of the guaiacols and syringols needed for polymerization, further increase in pressure may have
reached maximum at a pyroprobe temperature of 600 °C. The enhanced steam reforming of the polyaromatic hydrocarbons that
decrease in content of phenolic was probably due to demethoxyla- could also be catalyzed by lignin char. In summary, moderate pres-
tion, demethylation, and alkylation of phenolics occurred at high sure increase (0.1–1.1 MPa) promoted the formation of polyaro-
pyrolysis temperatures [23]. matic hydrocarbons; however, further increase in pressure (up to
50 K. Qian, A. Kumar / Fuel 162 (2015) 47–54

14

Tar content (µg/mg lignin)

12
10 700 °C 800 °C 900 °C

8
6
4
2
0

Fig. 1. Composition of tars produced from kraft lignin at various reaction temperatures.

Fig. 2. Aromatic hydrocarbon content in tar produced from kraft lignin pyrolysis at different temperatures and pressures (a) total content of monocyclic aromatic
hydrocarbon (benzene, toluene, ethylbenzene, xylene and styrene) (b) total content of polyaromatic hydrocarbon (naphthalene, 1,5-dimethynaphthalene, 2-methylnaph-
thalene and indene).

2.2 MPa) led to a decrease in polyaromatic hydrocarbons due to (0–150 psig), phenols (Fig. 3(a)) decreased with increase in pres-
consumption in the steam reforming reaction. sure as well as temperature. Up to 2.2 MPa (300 psig), the content
Second order polynomial regression failed to generate response of phenols further decreased but hardly changed with tempera-
surfaces of phenols and guaiacols (R2 obtained were low). Data was ture. The highest phenol content (20 lg/mg lignin) was observed
plotted without fitting. The trends of phenols and guaiacols with at 700 °C and 0 psig, and the lowest content (2.4 lg /mg lignin)
respect to reaction pressure and temperature (Fig. 3) differed from was observed at 700 °C and 2.2 MPa (300 psig). Similar to aromatic
the trends of aromatic hydrocarbons (Fig. 2). At 0.1–1.1 MPa hydrocarbons and phenols, pressure significantly affected contents
 K. Qian, A. Kumar / Fuel 162 (2015) 47–54 51

Fig. 3. Phenolic contents in tar produced from kraft lignin pyrolysis produced at different temperature and pressure (a) total content of phenols (2,6-dimethylphenol, 2,4-
dimethylphenol, 4-methylcatechol, phenol, 2-methylphenol, 4-ethylcatechol, catechol and p-cresol) (b) total content of guaiacols (creosol, 2-methoxyphenol, 2-methoxy-4-
vinylphenol, eugenol, tran-isoeugenol and vanillin).

of guaiacols. The plot of guaiacols (Fig. 3(b)) showed that guaiacols Fig. S2. The temperature levels were set at 700, 800 or 900 °C
reached a maximum at 1.1 MPa (50 psig) and decreased with fur- and volume of water injected on lignin was 5 ll (same as non-
ther increase in pressure from 0.5 to 2.2 MPa (50 to 300 psig). No catalytic lignin pyrolysis) or 10 ll.
clear trend was found for the effect of temperature on total guaia- The amount of water loading significantly affected catalyst per-
cols content. The highest guaiacol content (7.8 lg/mg lignin) was formance. With 10 ll water, the total phenolics contents reduced by
observed at 900 °C and 0.5 MPa (50 psig). 50% at 700 °C and by 30% at 800–900 °C, while with 5 ll water, total
In literature, most of pressurized pyrolysis and gasification phenolics’ contents reduced by more than 90% at all temperature
were conducted at pressure below 150 psig [7,27,28]. At reaction (see Fig. 5). The total aromatic hydrocarbons hardly reduced in cat-
pressure below 150 psig, our observations on the effect of temper- alytic conditioning of lignin tar with 10 ll water, while these
ature on tar contents were consistent with literature. Mayerhofer greatly reduced (50% on average) in presence of 5 ll water. As
et al. [27] reported that tar composition varied with temperature. shown in Fig. 4(b) and Fig. S2, the contents of most phenolics
Phenolic species (phenol and cresols) greatly decreased with reduced by less than 30% with 10 ll water; only 4-ethylcatechols,
increasing temperature whereas naphthalene increased with 2-methoxyphenol and 2-methoxy-4-vinylphenol reduced by more
increasing temperature. Mastral et al. [28] reported an increase than 50%. Most of the aromatic hydrocarbons did not reduce with
in polyaromatic compounds with increasing temperature during 10 ll water. With 5 ll water, most of the phenols were significantly
pyrolysis of polyethylene in a free-fall reactor at 800–1000 °C. removed and the maximum removal rates for catechol and 4-ethyl
However, effect of pressure on tar composition was inconclusive. catechol were 96% and 94%, respectively. A 70–80% of total guaia-
Mayerhofer et al. [27] found phenols and cresols were not affected cols was removed at 700–900 °C and 2-methoxy-4-vinylphenol
by increasing pressure (from 0.1 to 0.3 MPa) but polyaromatic was removed the most (80–90%). For aromatic hydrocarbons in
increased by almost 200%. Knight [7] found that increasing pres- the presence of 5 ll water, the highest removal rate (100%) was
sure (from 0.8 to 2.1 MPa) resulted in decrease of oxygenated spe- found for xylenes, styrene and indene, as their contents were below
cies. Phenols were almost completely eliminated, while the GC/MS detection limit and thus considered as zero. The lowest
polyaromatic increased. Berrueco et al. [29] performed pressurized removal rate was found for naphthalene that remained in tar. Over-
gasification of torrefied woody biomass in a lab-scale fluidized bed all, the catalyst performed better with 5 ll water than with 10 ll
at 0.1, 0.5 and 1 MPa. Their data showed an increase in tar content water. This can be attributed to the excess water (10 ll) that may
as pressure increased. have clogged the char pore and prevented tar vapor from accessing
active sites on the catalyst. The effect of temperature on lignin tar
3.2. Catalytic removal of tar produced from lignin pyrolysis removal was moderate as compared to effects of water. Increase
in temperature only increased removal of benzene, toluene, cate-
3.2.1. Effects of temperature and water amount chols and benzofuran for catalytic conditioning with 5 ll water.
Aromatic hydrocarbons, phenolics and total contents of lignin Overall, the char-derived catalysts were the most effective in
tar with and without catalyst are illustrated in Figs. 4 and 5 and removing phenolics, lesser effective in removing monoaromatic
52 K. Qian, A. Kumar / Fuel 162 (2015) 47–54

(a)
25
700 °C, no catalyst
700 °C, with catalyst, 10µl H2O
700 °C, with catalyst, 5µl H2O
20 800 °C, no catalyst
800 °C, with catalyst, 10µl H2O
Tar content (µg/mg lignin)

800 °C, with catalyst, 5µl H2O

900 °C, no catalyst
15
900 °C, with catalyst, 10µl H2O
900 °C, with catalyst, 5µl H2O

10

0
Phenols Catechols Guaiacols
(b)
Fig. 4. Effects of temperature and water loading on individual tar content in kraft lignin tar, (a) content of aromatics, (b) content of phenolics. Reaction temperature: 700, 800
and 900 °C, pressure: 0 psig, water amount: 5 or 10 ll. Phenols represented total content of 2,6-dimethylphenol, 2,4-dimethylphenol, phenol, 2-methylphenol, and p-cresol;
catechol represented total content of 4-methylcatechol, 4-ethylcatechol and catechol; guaiacols represented total content of creosol, 2-methoxyphenol, 2-methoxy-4-
vinylphenol, eugenol, tran-isoeugenol and vanillin.

hydrocarbons and the least effective in removing polyaromatics. At hydrocarbons and polyaromatic hydrocarbons [30–32]. Coll et al.
the lowest pyroprobe temperature (700 °C), the average removal of [32] studied steam reforming on five model tar compounds, includ-
phenolics was about 50%. Catechols were removed more than 85%, ing benzene, toluene, pyrene, anthrathene and naphthalene, and
which was the highest among all individual phenolics. However, found that naphthalene was the toughest and benzene was the
almost no aromatic hydrocarbons were removed at 700 °C. When easiest to reform. By reviewing steam reforming of phenol and ben-
the pyroprobe temperature was raised to 900 °C, more than zene at similar reaction conditions (similar nickel/aluminum cata-
90% of phenolics and 60% of monoaromatic hydrocarbons were lyst and similar reactor) [31,33,34], it can be observed that that
removed while removal efficiencies of naphthalene. 1,5- phenol was easier to reform than benzene or toluene. High phenols
dimethylnaphthalene and 2-methylnaphthalene were still less than conversion (more than 90%) can be achieved even at low tempera-
30%. The removal efficiencies of individual tar compounds can be ture (450 °C) by steam reforming over Ni/aluminum catalyst [31].
attributed to reactivity and stability of each compound. According However, catalytic conversion of benzene or toluene require at least
to literature, the reactivity of tar compounds in catalytic environ- 600 °C over Ni/aluminum in order to achieve conversion/removal
ment from the highest to lowest are phenolics, monoaromatic above 60% [33,34].
 K. Qian, A. Kumar / Fuel 162 (2015) 47–54 53

25 found that NiCl2 and ZSM-5 zeolite catalysts hardly improved

Total aromatics, with catalyst, 5µl H2O
Total aromatics, with catalyst, 10µl H2O decomposition of phenolic hydroxyl groups of lignin but signifi-
Total phenolics, with catalyst, 5µl H2O cantly improved the decomposition of aliphatic hydroxyl groups,
Tar content (µg/mg lignin)

20 Total phenolics, with catalyst, 10µl H2O carboxyl and aromatic-methoxyl groups. However, by studying
steam reforming of phenol using in-situ FT-IR, Garbarino et al.
15 [31] found that phenol steam reforming reaction occurred at the
expense of surface phenate species adsorbed on Ni centers. These
surface phenate species were active above 400 °C and directly
10 reformed with steam into CO and hydrogen with no intermediates
or by-products.

5
3.2.2. Effect of pressure
The effect of pressure on catalysts performance was studied
0
700 800 900
with 10 ll water. Aromatic hydrocarbons and phenolics at three
Temperature (°C) pressure settings of 0.1, 0.5 and 1.1 MPa (0, 50 and 150 psig) and
a pyroprobe temperature of 900 °C are illustrated in Fig. S3. The
Fig. 5. Effects of temperature and water loading on total tar contents produced pressure significantly affected the catalytic reforming of lignin
from kraft lignin. Reaction temperature: 700, 800 and 900 °C, pressure: 0 psig, tar over char-based catalyst. When pressure increased from 0.1
water amount: 5 or 10 ll.
to 0.5 MPa (0 to 50 psig), the removal of most aromatic hydrocar-
bons increased from nearly 0 to 30% except o-xylene and the
removal of most phenolics increased from 30% to 50% except
30.00
p-creosol. When the pressure further increased to 1.1 MPa (150
Total aromatics, no catalyst
Total tar coentent (µg/mg lignin)

Total phenolics, no catalyst psig), the removal of most aromatic hydrocarbons and phenolics
25.00
Total aromatics, with catalyst increased to more than 70%. Catechol, 2-methoxyvinylphenol,
Total phenolics, with catalyst 4-methylcatechol and o-xylene were not detected at 1.1 MPa
20.00
(150 psig).
The total aromatic hydrocarbons and phenolics with and with-
15.00 out catalyst are illustrated in Fig. 6. Increase in pressure increased
removal of total aromatic hydrocarbons and phenolics. The
10.00 removal efficiency of aromatic hydrocarbons and phenolics
increased from 0% to 75% and from 20% to 75%, respectively, when
5.00 pressure increased from 0.1 MPa (0 psig) to 1.1 MPa (150 psig). The
effects of pressure on removal efficiency could be related to pro-
0.00 longed residence (reaction) time. As discussed in Section 3.2.1, ris-
0 50 150
ing pressure suppresses the release of tar through the catalytic
Pressure (psig) layer and thus increases the reaction time of tar with the catalyst
Fig. 6. Total tar contents of catalytic conditioning of kraft lignin tar as a function of subsequently increasing tar reforming.
pressure at 900 °C in the presence of 10 ll water.

3.2.3. Effect of atmosphere

Aromatics hydrocarbons and phenolic substitutes obtained in
As shown in Fig. 4(a), benzene content increased in the presence the presence of H2 are illustrated in Fig. 7. It was obvious that most
of the catalyst probably because of catalyst promoted conversion of tar components were lower under hydrogen atmosphere than
phenols into benzene or oligomerization of hydrocarbon monomer. those under helium atmosphere. The major decreases (>50%) were
However, benzene was more likely converted from oligomerization observed in catechols, 2-methoxyphenol, 2-methoxy-4-
of hydrocarbon monomer than conversion of phenols as suggested vinylphenol and trans-isoeugenol and minor decreases (<20%)
by Ben et al. [35] and Garbarino et al. [31]. Ben Ragauskas [35] were observed in phenol, dimethylphenols and polyaromatic

10
Tar content (µg/mg lignin)

9
He, no catalyst
8
H2, no catalyst
7
6 H2, with catalyst
5
4
3
2
1
0

Fig. 7. Effect of H2 atmosphere on content of tars produced from kraft lignin at 800 °C.
54 K. Qian, A. Kumar / Fuel 162 (2015) 47–54

hydrocarbons. Monocyclic aromatic hydrocarbons such as benzene [4] Faix O, Meier D. Pyrolytic and hydrogenolytic degradation studies on
lignocellulosics, pulps and lignins. Holz Roh Werkst 1989;47:67–72.
and toluene showed slight increase. Decrease in tar contents in the
[5] Pasangulapati V, Ramachandriya KD, Kumar A, Wilkins MR, Jones CL, Huhnke
presence of hydrogen was possibly because hydrogen acted as a RL. Effects of cellulose, hemicellulose and lignin on thermochemical
gasification agent and thus promoted the conversion of lignin into conversion characteristics of the selected biomass. Bioresour Technol
non-condensable gas rather than tar. 2012;114:663–9.
[6] Hu J, Shen D, Xiao R, Wu S, Zhang H. Free-radical analysis on thermochemical
Tar contents were further reduced in presence of the catalyst. transformation of lignin to phenolic compounds. Energy Fuels
Dimethylphenols, 2-methoxyphenol, vanillin, ethylbenzene, 2012;27:285–93.
methylnaphthalene, 1, 5-dimethylnaphthalene, trans-isoeugenol [7] Knight RA. Experience with raw gas analysis from pressurized gasification of
biomass. Biomass Bioenergy 2000;18:67–77.
and xylenes were not detected. Most of the remaining tar com- [8] Berrueco C, Montané D, Matas Güell B, del Alamo G. Effect of temperature and
pounds were reduced by more than 50%. However, the removal dolomite on tar formation during gasification of torrefied biomass in a
of naphthalene was only about 20%. pressurized fluidized bed. Energy 2014;66:849–59.
[9] Thangalazhy-Gopakumar S, Adhikari S, Gupta RB, Tu M, Taylor S. Production of
hydrocarbon fuels from biomass using catalytic pyrolysis under helium and
4. Conclusions hydrogen environments. Bioresour Technol 2011;102:6742–9.
[10] Meesuk S, Cao J-P, Sato K, Ogawa Y, Takarada T. The effects of temperature
on product yields and composition of bio-oils in hydropyrolysis of rice husk
Pyrolysis of kraft lignin was performed in a pyroprobe reactor in using nickel-loaded brown coal char catalyst. J Anal Appl Pyrol 2012;94:
presence of water with a novel char-based catalyst. The effects of 238–45.
[11] Takarada T, Tonishi T, Fusegawa Y, Morishita K, Nakagawa N, Kato K.
reaction temperature, water loading, pressure and atmosphere on Hydropyrolysis of coal in a powder-particle fluidized bed. Fuel 1993;72:921–6.
tar were investigated by conducting quantitative analysis of tar. [12] Qian K, Kumar A, Zhang H, Bellmer D, Huhnke R. Recent advances in utilization
Catechols were found to be the most abundant tar produced from of biochar. Renew Sust Energy Rev 2015;42:1055–64.
[13] Xie T, Reddy KR, Wang C, Yargicoglu E, Spokas K. Characteristics and
lignin in absence of catalyst, followed by phenol and guaiacols.
applications of biochar for environmental remediation: a review. Crit Rev
Increase in pressure from 0.1 to 1.1 MPa (0 to 150 psig) led to Environ Sci Technol 2014;45:939–69.
increase in aromatic hydrocarbons. High pressure may have caused [14] Bhandari PN, Kumar A, Huhnke RL. Simultaneous removal of toluene (model
increase in tar residence time and resulted in polymerization reac- tar), NH3, and H2S, from biomass-generated producer gas using biochar-based
and mixed-metal oxide catalysts. Energy Fuels 2013;28:1918–25.
tions forming polyaromatic hydrocarbons. Further increase in [15] Abu El-Rub Z, Bramer EA, Brem G. Experimental comparison of biomass chars
pressure to 2.2 MPa (300 psig) led to a reduction of polyaromatic with other catalysts for tar reduction. Fuel 2008;87:2243–52.
hydrocarbons due to steam reforming reaction. [16] Qian K, Kumar A, Bellmer D, Yuan W, Wang D, Eastman MA. Physical
properties and reactivity of char obtained from downdraft gasification of
During catalytic reforming of lignin tar, the effect of tempera- sorghum and eastern red cedar. Fuel 2015;143:383–9.
ture on lignin tar removal was moderate. Increase in temperature [17] Schorr D, Diouf PN, Stevanovic T. Evaluation of industrial lignins for
only increased removal of benzene, toluene, catechols and benzo- biocomposites production. Ind Crop Prod 2014;52:65–73.
[18] Thangalazhy-Gopakumar S, Adhikari S, Gupta RB. Catalytic pyrolysis of
furan with 5 ll water. High water loading (10 ll) decreased the biomass over H + ZSM-5 under hydrogen pressure. Energy Fuels
removal efficiency probably because the excess water clogged 2012;26:5300–6.
the char pore and prevented tar vapor from accessing active sites [19] Zhang M, Resende FLP, Moutsoglou A, Raynie DE. Pyrolysis of lignin extracted
from prairie cordgrass, aspen, and Kraft lignin by Py-GC/MS and TGA/FTIR. J
on the catalyst. Higher pressure promoted the catalytic decompo- Anal Appl Pyrol 2012;98:65–71.
sition of lignin tar. When pressure increased from 0.1 to 1.1 MPa (0 [20] Zhang M, Resende FLP, Moutsoglou A. Catalytic fast pyrolysis of aspen lignin
to 150 psig), the removal of most aromatic hydrocarbons increased via Py-GC/MS. Fuel 2014;116:358–69.
[21] Zhang M, Moutsoglou A. Catalytic fast pyrolysis of prairie cordgrass lignin and
from nearly 0% to 70% and the removal of phenols increased from
quantification of products by pyrolysis–gas chromatography–mass
30% to 70%. Catechol, 2-methoxyvinylphenol, 4-methylcatechol spectrometry. Energy Fuels 2014;28:1066–73.
and o-xylene at 1.1 MPa (150 psig) reached nearly 100% removal. [22] Zhou S, Garcia-Perez M, Pecha B, Kersten SRA, McDonald AG, Westerhof RJM.
When pyrolysis was performed in hydrogen atmosphere, tar con- Effect of the fast pyrolysis temperature on the primary and secondary products
of lignin. Energy Fuels 2013;27:5867–77.
tents significantly decreased, as hydrogen acted as a gasification [23] Jiang G, Nowakowski DJ, Bridgwater AV. Effect of the temperature on the
agent, promoting the conversion of lignin into non-condensable composition of lignin pyrolysis products. Energy Fuels 2010;24:4470–5.
gas. In all cases, removal of phenolics was higher than that of aro- [24] Fahmi R, Bridgwater AV, Thain SC, Donnison IS, Morris PM, Yates N. Prediction
of Klason lignin and lignin thermal degradation products by Py-GC/MS in a
matic hydrocarbons. Naphthalene, 1,5-dimethylnaphthalene, 2- collection of Lolium and Festuca grasses. J Anal Appl Pyrol 2007;80:16–23.
methylnaphthalene were the toughest tar component to be decom- [25] Fu Y, Kato K, Ohtani H, Chen Y. Pyrolysis products of lignin in open and sealed
posed with the lowest removal efficiency. vessels studied by PY-GC-MS. J Wood Chem Technol 2013;34:1–7.
[26] Thangalazhy-Gopakumar S, Adhikari S, Gupta RB, Fernando SD. Influence of
pyrolysis operating conditions on bio-oil components: a microscale study in a
Acknowledgements pyroprobe. Energy Fuels 2011;25:1191–9.
[27] Mayerhofer M, Mitsakis P, Meng X, de Jong W, Spliethoff H, Gaderer M.
Influence of pressure, temperature and steam on tar and gas in allothermal
Authors acknowledge the support of South Central Sun Grant fluidized bed gasification. Fuel 2012;99:204–9.
Program and Oklahoma Agricultural Experiment Station for this [28] Mastral JF, Berrueco C, Ceamanos J. Pyrolysis of high-density polyethylene in
study. free-fall reactors in series. Energy Fuels 2006;20:1365–71.
[29] Berrueco C, Recari J, Güell BM, del Alamo G. Pressurized gasification of
torrefied woody biomass in a lab scale fluidized bed. Energy 2014;70:68–78.
Appendix A. Supplementary material [30] Abu El-Rub Z, Bramer EA, Brem G. Review of catalysts for tar elimination in
biomass gasification processes. Ind Eng Chem Res 2004;43:6911–9.
[31] Garbarino G, Sanchez Escribano V, Finocchio E, Busca G. Steam reforming of
Supplementary data associated with this article can be found, in phenol-ethanol mixture over 5% Ni/Al2O3. Appl Catal B – Environ 2012;113-
the online version, at http://dx.doi.org/10.1016/j.fuel.2015.08.064. 114:281–9.
[32] Coll R, Salvadó J, Farriol X, Montané D. Steam reforming model compounds of
biomass gasification tars: conversion at different operating conditions and
References tendency towards coke formation. Fuel Process Technol 2001;74:19–31.
[33] Mudinoor A, Bellmer D, Marin L, Kumar A, Huhnke R. Conversion of toluene
[1] Font Palma C. Modelling of tar formation and evolution for biomass (model tar) using selected steam reforming catalysts. T ASABE
gasification: a review. Appl Energy 2013;111:129–41. 2011;54:1819–27.
[2] Shen DK, Gu S, Bridgwater AV. Study on the pyrolytic behaviour of xylan-based [34] Tao J, Zhao L, Dong C, Lu Q, Du X, Dahlquist E. Catalytic steam reforming of
hemicellulose using TG–FTIR and Py–GC–FTIR. J Anal Appl Pyrol toluene as a model compound of biomass gasification tar using Ni-CeO2/SBA-
2010;87:199–206. 15 catalysts. Energies 2013;6:3284–96.
[3] Lv G, Wu S. Analytical pyrolysis studies of corn stalk and its three main [35] Ben H, Ragauskas AJ. Pyrolysis of kraft lignin with additives. Energy Fuels
components by TG–MS and Py-GC/MS. J Anal Appl Pyrol 2012;97:11–8. 2011;25:4662–8.