Chemosphere 90 (2013) 1501–1507

Contents lists available at SciVerse ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

Bench-scale gasification of cedar wood – Part II: Effect of Operational conditions

on contaminant release
Salah H. Aljbour a,⇑, Katsuya Kawamoto b
a
Department of Chemical Engineering, College of Engineering, Mutah University, Karak 61710, Jordan
b
Center for Material Cycles and Waste Management Research, National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, Ibaraki 305-8506, Japan

h i g h l i g h t s

" The paper presents experimental results on the evolution profile of tar during cedar wood gasification.
" Other contaminants release (H2S, COS, NH3, HCN and HCl) is also discussed.
" Tests have been conducted under various operating conditions in a bench-scale facility.

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

Article history: Here, we present the evolution profile of tar in the product gas during cedar biomass gasification. We also
Available online 12 September 2012 discuss the evolution of other contaminants (H2S, COS, NH3, HCN, and HCl). The cedar wood was gasified
under various operating conditions in a bench-scale externally heated updraft gasifier; this was followed
Keywords: by thermal reforming.
Ammonia Tar levels in the product gas were significantly affected by the operating conditions used. At a gasifi-
Chlorine cation temperature of 923 K, there was no clear relation between the evolution of phenolic tar in the
Biomass
product gas as a function of residence time. The evolution of PAH tar at a low gasification temperature
Gasification
Sulfur
was lower than the evolution of phenolic tar. With increasing temperature, the proportion of PAH tar con-
Tar tent became significant. At a gasification temperature of 1223 K, increasing the residence time reduced
the content of PAH tar owing to a catalytic effect associated with ash generation at high temperatures.
Increasing the steam-to-carbon (S/C) ratio under thermal conditions had a slight effect on PAH conver-
sion. However, increasing the equivalence ratio (ER) effectively reduced the tar levels.
The conversion of fuel-sulfur and fuel-nitrogen to volatile-sulfur and volatile-nitrogen, respectively,
increased with increasing S/C ratio and ER. The evolutions of COS and HCN gases were much smaller than
the evolution of H2S and NH3. The evolution of HCl in the product gas decreased slightly with increasing
ER. Increasing the S/C ratio decreased the HCl levels in the product gas. The effect of temperature on con-
taminant levels could not be fully understood due to limited availability of experimental data at various
temperatures. We also compare our findings with data in the literature.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction During the gasification process, tar is generated as a co-product

Biomass can deliver practically everything that fossil fuels pro-
vide, whether it be used to produce fuel or chemical feedstock (Le-
duc et al., 2009). Gasification of biomass is viewed as a promising
system for producing renewable hydrogen; such a process is ben-
eficial for the exploitation of biomass resources and for developing
a highly efficient, clean method of large-scale hydrogen production
(Wang et al., 2008).
⇑ Corresponding author.
E-mail address: saljbour@yahoo.com (S.H. Aljbour).
along with the product gas. Tar remains vaporized until the gas
carrying it is cooled. This results in the condensation of tar, thereby
making the product gas unsuitable for use in processing equipment
that has a low tolerance for tar (Devi et al., 2003). In addition to tar,
other minor and trace contaminants, including sulfur, nitrogen,
and chlorine compounds are present in the product gas. Formation
of these pollutants is unavoidable during the gasification process,
and is related to a large extent to the presence of S, N and Cl in
the solid fuels used (Wang et al., 2010). The presence of these pol-
lutants in the product gas is undesirable and limits the end-use
applications. Therefore, their levels in the product gas should be

0045-6535/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.chemosphere.2012.08.030
1502 S.H. Aljbour, K. Kawamoto / Chemosphere 90 (2013) 1501–1507
monitored and their release should be controlled to meet the oper-
ational requirements of each end-use application.
The present study was conducted within the framework of R&D
activities for gasification and reforming technologies for energy
and chemicals recovery from biomass resources. Gasification of
Japanese cedar wood has been investigated under various operat-
ing conditions. In our previous paper (Aljbour and Kawamoto,
2012), we presented the results of a parametric study on the effect
of operational conditions on product gas characteristics. Here, we
present our results on the evolution profiles of the main tar con-
stituents and other contaminants formed during cedar wood
gasification.
2. Experimental
2.1. Experimental apparatus
A detailed description of the experimental apparatus is given
elsewhere (Aljbour and Kawamoto, 2012).
2.2. Sampling procedure
We quantified tar levels and NH3, HCN, HCl, H2S and COS con-
centrations in the product gas. Following is a detailed description
of the methods and apparatuses used in sampling. The sampling
layout is given in Supplemental Material (SM), Fig. SM-1 (Aljbour
and Kawamoto, 2012).
2.2.1. Tar sampling and analysis
Tar was sampled by using the Solid Phase Adsorption (SPA)
method (Brage et al., 1997), whereby the tar is adsorbed onto a
porous sorbent column (silica-bonded amino phase). This method
is believed to capture tar compounds ranging from benzene to cor-
onen. We used 3-mL SPA columns (SampliQ Amino, Agilent Tech-
nologies). The SPA column contains 200 mg of amino phase
supported on silica. A needle was fixed to the tip of the SPA col-
umn. Before the column was used, the inlet-end flange was cut
off and the column was fitted to a gas-tight syringe (100 mL). Tar
samples were taken by inserting the syringe needle into the prod-
uct gas line via the appropriate septum then manually pulling
100 mL of the product gas through the column in 1 min. With
manual sampling of the product gas at a rate of 100 mL min1
there was a risk of analyte breakthrough from the SPA column,
thereby leading to uncertainty in the tar capture percentage. We
therefore used a gas sampling pump (GSP-250FT, GASTEC, Yamato)
to withdraw the product gas from the sampling line through a rub-
ber septum. The constant flow rate function in this pump automat-
ically controlled any changes in flow rate caused by variations in
product gas composition. For each sample, 100 mL of product gas
were withdrawn at a rate of 50 mL min1 for 2 min. The gas flow
rate was also measured by a digital gas flow rate meter (ADM
1000, Agilent Technologies). For high-tar experiments (gasification
at low temperature), two SPA columns in series were used to en-
sure that there was no tar breakthrough. Before the sampling, each
SPA column was conditioned with 0.5 mL dichloromethane or iso-
propanol with positive pressure from a syringe. After sampling, the
column was disconnected and placed in a carefully capped test
tube.
During each sampling, two samples were taken sequentially.
The first one was intended for phenolic tar collection, whereas
the other one was for PAH tar. The captured tar is recovered from
the SPA column by elution with 3 mL dichloromethane (for PAH
tar) and 3 mL isopropanol (for phenolic tar). The collected solution
was subjected to both quantitative and qualitative analysis using a
GC/MS system (GCMS-QP2010 Plus, Shimadzu) equipped with a
capillary column (CP-Sil PONA CB). The identity of the target ana-
lytes was confirmed on the basis of comparison with external stan-
dards. Three standard solutions were used to produce calibration
curves for 10 individual PAH tar species: naphthalene, acenaph-
thylene, fluorene, phenanthrene, anthracene, 2-phenylnaphtha-
lene, pyrene, fluoranthene, benz(b)anthrathene, and triphenylene.
Another three standard solutions were used to produce calibration
curves for six individual phenolic tar species: phenol, o-cresol, m-
cresol, pyrocatecol, hydroquinone, and 4-methylcatechol. In addi-
tion, calibration curves for styrene and propynyl benzene also were
produced.
2.2.2. NH3 and HCN sampling and analysis
A gas stream was withdrawn from valve three through silicon
tubing leading to an impinger bottle to capture NH3. Another gas
stream was withdrawn from valve four through silicon tubing
leading to an impinger bottle to capture HCN. Glass wool was
packed in the silicon tubing and maintained at a temperature of
373 K to ensure complete removal of dust. The impinger bottles
were immersed in ice baths. The solution for trapping NH3 con-
sisted of 75 mL of 5 g L1 boric acid solution; 75 mL of 0.1 N NaOH
solution was used for HCN collection. The sample flow through the
impinger was maintained at 0.5 L min1, and a total of 20 L of dry
gas was processed for each sample. Two impingers in series were
used in preliminary tests, but the results showed that the first imp-
inger bottle captured almost all of the analyte.
The trapping solutions were analyzed in accordance with Japa-
nese Industrial Standards JIS K 0099 (2004) and JIS K 0109 (1998)
for NH3 and HCN respectively. In these standards, NH3 was
quantified by indophenols blue absorption spectrophotometry
(wavelength = 640 nm) while HCN was quantified by first
conversion to cyanogen chloride followed by color reaction with
a pyridine-pyrazolone mixture (wavelength = 638 nm).
2.2.3. HCl sampling and analysis
A gas stream was withdrawn from valve five through silicon
tubing leading to an impinger bottle to capture HCl. Glass wool
was packed in the silicon tubing and maintained at a temperature
of 373 K to ensure complete removal of dust. The impinger bottle
was immersed in an ice bath and filled with 75 mL of deionized
water. The sample flow through the impinger was maintained at
0.5 L min1, and a total of 20 L of dry gas was processed for each
sample. The trapped HCl in solution was quantified by ion chroma-
tography in accordance with the Japanese Industrial Standards JIS
K 0107 (2002).
2.2.4. H2S and COS sampling and analysis
Samples of the product gas were periodically collected from the
downstream line of the tar capture impinger system, but before the
moisture trapping impingers, using Tedlar sampling bags (AS ONE).
H2S and COS in the product gas were determined by GC with flame
photometric detection (GC-14B, Shimadzu) equipped with Sunpac-
S column (Shinwa Chemical Industries). Standard calibration
curves for H2S and COS were prepared by using a certified gas mix-
ture with 100 and 50 ppm of H2S and COS respectively (Takachiho
Chem.). Lower concentration calibration standards were produced
by adjusting the gas injection volume.
3. Results and discussion
3.1. Tar formation and decomposition characteristics
3.1.1. Effect of residence time
We examined the effect of residence time on tar conversion un-
der gasification-reforming temperatures of 923 and 1223 K. Fig. 1a
 S.H. Aljbour, K. Kawamoto / Chemosphere 90 (2013) 1501–1507 1503

One could attribute the reduction in the PAH content in the

product gas (Fig. 1b) to enhancement of the thermal destruction
of PAHs at 1223 K with increased residence time. Another possible
explanation is that catalytic cracking of the PAH tar may take place
at this temperature. Char is believed to have a good catalytic activ-
ity for tar removal. The catalytic activity of char for tar reduction
can be related to its pore size distribution, internal surface area,
and mineral content. Abu El-Rub (2008) has conducted a system-
atic study on tar removal using biomass char as in situ catalyst.
It was reported that at a bed temperature of 1173 K, 90 g Nm3 ini-
tial naphthalene concentration and 0.3 s residence time, the bio-
mass char and the biomass ash catalysts (each added on the top
of a silica sand bed of the same volume), respectively promoted
94% and 74% naphthalene reduction. In our study, most of the car-
bon in the char collected during gasification at a temperature of
1220 K had been gasified. Our carbon elemental analysis of char
showed a carbon conversion more than 90%. The ash content in
the gasified char likely promotes catalytic tar conversion. Obvi-
ously, increasing the gas residence time will increase the contact
time of the tar-containing gas and the metal constituents of char,
thereby allowing for catalytic tar destruction.

3.1.2. Effect of temperature

The temperature at which gasification takes place is believed to
affect the quality and quantity of tar in the product gas. Brage et al.
(2000) have studied the tar evolution profile obtained from birch
Fig. 1. Tar content vs. residence time under ER = 0.2, S/C = 0 and (a) T = 923 K, (b)
wood gasification in a pressurized fluidized bed. When the temper-
T = 1223 K.
ature was raised from 973 to 1173 K, they observed complete
reduction of phenol content and 50% decrease in toluene content
but a considerable increase in benzene (14–24 g Nm3) and naph-
thalene (2–8 g Nm3) contents. Li et al. (2004) have indicated that
shows the concentration of total phenolic tar and the total PAH tar
the tar concentration primarily depends on the operating temper-
as a function of the residence time at gasification-reforming tem-
ature. They measured tar yield during biomass gasification in a cir-
perature of 923 K.
culating fluidized bed and found a sharp decrease, from 15.2 at
At a temperature of 923 K, phenolic tar accounted for the great-
973 K to 0.42 g Nm3 at 1080 K.
est proportion of the total tar. PAH tar evolution tended to be smal-
Here, we characterized the tar contents in the product gas at
ler than phenolic tar evolution at this low temperature. PAH tar
different gasification temperatures. The phenolic tar constituents
content in the product gas changed only very slightly with increas-
of the product gas decreased from 20 at 923 K to 13 at 1023 K, then
ing the residence time: the total PAH tar content in the product gas
to 0.1 g Nm3 at 1123 K. We examined the tar concentration of the
was around 3 g Nm3 for the range of residence times from 14 to
product gas as a function of the gasification temperature (Fig. 2a).
36 s. The same trend can be noticed for naphthalene which ac-
The naphthalene concentration in the product gas increased from 2
counted for the greatest proportion of the total PAH tar content.
at 923 K to 7 at 1023 K, and then decreased to 0.7 g Nm3 at
Naphthalene concentration generally dominated the other PAH
1223 K. Other tar constituents such as propynyl benzene, ace-
constituents in the total PAH tar.
naphtheleyne, fluorine, phenanthrene and anthracene increased
The relationship between the total phenolic tar content of the
very slightly with increasing the temperature. In contrast, styrene
product gas and the residence time was unclear. However, with re-
concentration decreased from 3 at 923 K to 0.1 g Nm3 at 1223 K.
spect to phenol concentration in the product gas, increasing the
The initial increase in the naphthalene concentration might have
residence time from 14 to 36 s linearly increased the phenol con-
been associated with the decrease in the phenolic tar fraction with
centration from 4 to 13 g Nm3. This might have been due to the
increasing the temperature. This behavior is in agreement with
transformation of other tar species into phenols at longer residence
experimental observations that phenols are decarbonylated into
times.
radicals of the cyclopentadiene type presumed to be intermediated
Fig. 1b shows the concentration of the total PAH tar as a func-
in the formation of light PAHs such as indene and naphthalene
tion of the residence time at a gasification-reforming temperature
(Brage et al., 1996). However, further increase in temperature pro-
of 1223 K. In contrast to the behavior shown in Fig. 1a, the PAH tar
motes naphthalene conversion. This might be associated with the
content in the product gas was reduced by increasing the residence
effect of char in promoting catalytic tar conversion at high temper-
time at a temperature of 1223 K. The PAH concentration declined
atures, together with the availability of enough residence time for
from 4 to 2 g Nm3 with increasing the residence time from 8.5
such conversion.
to 14 s. Naphthalene concentration in the product gas declined
The obvious decline in the concentration of styrene as com-
from 2 to 0.8 g Nm3 with increasing the residence time from 8.5
pared with the other PAH constituents with increasing tempera-
to 14 s. This trend is in agreement with experimental trends in lit-
ture might be attributable to inherent differences in thermal
erature. For example, Houben (2004) performed thermal tar crack-
stability.
ing experiments in a tubular reactor in a temperature range of
1173–1423 K and residence times of 1 and 12 s. Raising the tem-
perature as well as the residence time reduced the tar concentra- 3.1.3. Effect of steam-to-carbon ratio (S/C)
tion. At a temperature of 1423 K, a residence time of 4 s was Steam can be used to convert tar into an array of smaller and
needed to diminish the tar concentration to 0.2 g Nm3. lighter hydrocarbons. The cracking reaction takes place in steam
1504 S.H. Aljbour, K. Kawamoto / Chemosphere 90 (2013) 1501–1507

3.1.4. Effect of equivalence ratio (ER)

We studied the effect of ER on tar conversion under different
oxygen-addition conditions. Fig. 2c shows the PAH concentration
in the product gas as a function of the ER at a temperature of
1123 K and S/C = 1. Increasing the ER from 0 to 0.3 linearly de-
creased the PAH concentration in the product gas from 7 to
4 g Nm3. Within this range of ERs, naphthalene concentration in
the product gas linearly decreased from 5 to 3 g Nm3. This linear-
ity in the reduction of tar level as a function of the ER was also no-
ticed for the other PAH tar constituents. The effect of oxygen
addition in promoting the reduction of tar levels in the product
gas can be attributed to partial oxidation reactions of the tar com-
pounds. These reactions promote CO and H2 production. However,
since the activated kinetic energy of the gas components is nor-
mally lower than that of the raw materials and tar, gaseous prod-
ucts are first oxidized (Kawamoto et al., 2009). This decreases the
concentrations of the combustible gases (Aljbour and Kawamoto,
2012). Therefore, an optimized value of ER should be applied to
maximize the yield of the desired gas products and to minimize
the tar yield.

3.2. Contaminants formation characteristics

3.2.1. H2S and COS formation

We examined the percentage conversion of fuel sulfur into H2S
and COS gases as a function of the ER and the S/C ratio (Fig. 3a and
b, respectively). Increasing the ER at a temperature of 1123 K line-
arly increased the conversion of fuel sulfur into sulfur gases, with
H2S being the dominant species. At ER = 0, 81 wt.% of the fuel-
sulfur has been liberated into the product gas as H2S and 2 wt.%
as COS. At ER = 0.3, 92 wt.% of the fuel-sulfur has been liberated

Fig. 2. The evolution of tar {(a) effect of temperature at ER = 0.2 and S/C = 0, (b)
Effect of S/C at ER = 0.2 and T = 1123 K and (c) Effect of ER at S/C = 1 and T = 1123 K}.

gasification, where steam cracks the tar, producing simpler and

lighter molecules like H2 and CO and CO2.
We varied the S/C ratio at a gasification-thermal reforming tem-
perature of 1123 K, and we examined the concentration of PAH tar
as a function of the S/C ratio (Fig. 2b). The PAH concentration was
almost constant at 5 g Nm3 when the S/C was increased from 0 to
2. Naphthalene concentration was almost constant at 3 g Nm3
when the S/C was increased from 0 to 2. These findings suggest
that addition of steam had no appreciable effect in promoting
PAH conversion. This finding is consistent with literature data that
steam has only a small influence on the conversion of aromatics
(Jess, 1996). Steam reforming of tar usually enhanced by using Fig. 3. Contaminant release in the product gas {(a) effect of ER at S/C = 1 and
the appropriate reforming catalysts. T = 1123 K, and (b) effect of S/C at ER = 0.2 and T = 1123 K}.
 S.H. Aljbour, K. Kawamoto / Chemosphere 90 (2013) 1501–1507
into the product gas as H2S and 4 wt.% as COS. The corresponding
concentrations of the sulfur species in the product gas within the
above range of ERs were 35–39 ppmv for H2S and <2 ppmv for COS.
The fate of the remaining fuel-sulfur was unknown: we did not
know whether it was retained in the solid matrix or liberated to
form other S-gases. We did not analyze the sulfur in the solid frac-
tion. Cui et al. (2010) reported that H2S, COS and C4H4S are the sul-
fur species in the product gas during biomass steam gasification in
a bench-scale fluidized bed gasifier. They found that H2S and COS
concentrations were more stable than C4H4S over the test period.
Meng et al. (2011) studied the evolution of sulfur species in the
product gas during biomass gasification in a steam-oxygen blown
circulating fluidized bed gasifier. Using a biomass fuel with
0.76 wt.% sulfur, they reported fairly high concentrations of H2S,
COS and trace levels of CH3HS in the product gas.
The linearity of the relationship between the conversion of fuel-
sulfur to H2S and the ER (see Fig. 3) is applicable within the oper-
ating conditions we used, but any change in the process conditions
could give a different trend. For example, the presence of excessive
amount of oxygen in the system might lead to the formation of
more oxidized sulfur compounds, thereby, reducing the concentra-
tion of H2S (Gulyurtlu et al., 2008).
Notably, the final H2S concentrations in the product gas might
not exactly reflect the actual conversion of the fuel-sulfur to H2S.
Investigations on coal pyrolysis and hydro-gasification (Attar and
Dupuis, 1979) have concluded that part of the sulfur is retained
in the solid as alkaline sulfide owing to the reaction of H2S with
the alkaline minerals, thus reducing the sulfur existing as H2S.
Chen et al. (1999) investigated the effects of mineral matter on
products and sulfur distributions in hydropyrolysis of coal. They
found that basic mineral matter in coal can trap the sulfur pro-
duced. The presence of minerals can also reduce COS and CH3SH
released in pyrolysis and hydropyrolysis.
Similarly, we found that increasing the S/C at a temperature of
1123 K linearly increased the conversion of the fuel-sulfur into
H2S. However, the presence of steam decreased the amount of sul-
fur present as COS in the product gas. At S/C = 0, 88 wt.% of the
fuel-sulfur liberated into the product gas as H2S and 7 wt.% as
COS (Fig. 3c). At S/C = 2, 98 wt.% of the fuel-sulfur liberated into
the product gas as H2S and 2 wt.% as COS. This trend explains the
relationship between the H2S and COS gases based on a hydrolysis
reaction as follows:
COS þ H2 O ! H2 S þ CO2
In addition, increasing the S/C promotes hydrogen production
and results in the hydrogenation of COS, as follows:
COS þ H2 ! H2 S þ CO
The percentage conversion of the fuel-sulfur into H2S was stable
at 88 wt.% within the temperatures range of 1023–1123 K. How-
ever the percentage conversion of the fuel-sulfur into COS de-
creased from 12 at 1023 K to 7 wt.% at 1123 K. Meng et al. (2011)
observed that with increasing temperature, the concentration of
H2S remained fairly stable, whereas the concentrations of COS
and CH3HS decreased slightly. On the basis of equilibrium calcula-
tions to predict the behavior of H2S during gasification of a wide
range of biomass fuels, Kuramochi et al. (2005) reported that at
high temperatures (1023–1123 K), the H2S concentration during
biomass gasification was constant and proportional to the sulfur
content of feedstock. However, at low temperatures they found a
high possibility of metal sulfide formation.
3.2.2. NH3 and HCN formation
We also investigated the percentage conversion of fuel nitrogen
to NH3 and HCN gases as a function of the ER and the S/C ratio
(Fig. 3a and b, respectively). At 1123 K, increasing the ER increased
1505
the conversion of the fuel-nitrogen to NH3. For example, the con-
version of the fuel-nitrogen to NH3 increased from 25 at ER = 0 to
45 wt.% at ER = 0.3. Within this range of ERs, the conversion of
the fuel-nitrogen to HCN was less than 0.1 wt.%. The corresponding
concentrations of the nitrogenous species in the product gas over
the same ER range were 353–505 ppmv for NH3 and 2–6 ppmv
for HCN. The enhanced evolution of NH3 with increasing the ER
could be attributed to the destruction of the fuel and char matrix
as a result of oxygen gasification. This would have led to the evo-
lution of the fuel-nitrogen into more volatile-nitrogen. Zhou et al.
(2000) investigated the evolution of nitrogenous species as a func-
tion of ER during sawdust gasification in a fluidized bed gasifier. At
1073 K, they observed that a change in ER from 0.25 to 0.37 has
slightly increased the NH3 concentration from 310 to 350 ppmv.
This suggests that increasing the ER has a modest influence on
the concentration of NH3.
There are two possibilities for the fate of the remaining fuel-
nitrogen. The first is that nitrogen retained in the solid matrix.
The other is that it evolves into other nitrogenous gases in the
product gas. The volatile matter content of cedar wood that we
used here was 75.2 wt.% (Aljbour and Kawamoto, 2012). It is well
known that fuel with high volatile matter content is more reactive
and therefore can be converted readily into gas, producing less
char. This char is characterized by a high reactivity and it is contin-
uously consumed during gasification (Brage et al., 2000). Therefore,
in our experiment it was likely that a large fraction of the fuel
nitrogen became volatile and evolved into other nitrogenous gases
besides NH3 and HCN. Zhou et al. (2000) emphasized that NH3 and
N2 are the dominant species evolved from fuel-nitrogen during
biomass gasification. In our study, however, nitrogen is used as a
carrier gas and it is technically difficult to infer the concentration
of the evolved nitrogen from the mass balance.
Under ER = 0.2 and a temperature of 1123 K, increasing the S/C
ratio remarkably increased the conversion of fuel-nitrogen to NH3
(Fig. 3b); the conversion of fuel-nitrogen to NH3 increased from 6
at S/C = 0 to 55 wt.% at S/C = 2. The highest conversion of the
fuel-nitrogen to HCN over this S/C range was 0.2 wt.%. The corre-
sponding concentrations of NH3 in the product gas over this S/C
range were 72–686 ppmv. The enhanced evolution of NH3 in the
presence of steam could be related to occurrence of reforming
reactions with the volatile-nitrogen. In a comparative study of coal
and biomass pyrolysis and gasification, Tian et al. (2007) concluded
that thermal cracking-reforming of volatile-nitrogen constitutes an
ð1Þ
additional route of NH3 formation during the gasification of cane
trash in steam.
Under ER = 0.2 and S/C = 1, increasing the temperature de-
creased the conversion of fuel-nitrogen to NH3. The conversion of
ð2Þ
fuel-nitrogen to NH3 decreased from 65 at 1023 K to 34 wt.% at
1123 K. The corresponding concentrations of NH3 in the product
gas over this S/C range were 1023–404 ppmv. This decrease in
NH3 concentration with increasing temperature is in agreement
with the results of other researchers. For example, Leppälahti
(1993) reported that in pressurized high-temperature fluidized
bed and entrained-bed gasifiers, when the temperature increases,
the conversion usually decreases, so that above 1372 K, the conver-
sion is always less than 20%. Similarly Zhou et al. (2000), observed
that NH3 concentration decreased from 950 at 973 K to 400 ppmv
at 1173 K.
3.2.3. HCl formation
We also examined the concentration of HCl as a function of
the ER and S/C ratio (see Figs. SM-1 and SM-2, respectively).
There was a weak dependency between the concentration of
HCl and the ER at 1123 K and S/C = 1 (Fig. SM-1). The concentra-
tion of HCl decreased slightly upon increasing the ER from 0 to
0.3. Over this range of ER, the concentration of HCl was in the
1506 S.H. Aljbour, K. Kawamoto / Chemosphere 90 (2013) 1501–1507
range 112–80 ppmv. These results agree with those obtained by
Dias and Gulyurtlu (2008): they found that increasing the amount
of oxygen supplied to the gasifier had no significant effect on the
concentration of HCl in the gas-phase product under conditions of
70% refused-derived fuel (RDF) and 30% coal gasification in a bub-
bling fluidized bed gasifier at 1123 K. However, we found that a
further increase in the S/C ratio reduced the HCl concentration:
at S/C = 2, the HCl concentration decreased to 32 ppmv (Fig. SM-
2). HCl evolution into the gaseous phase during gasification is
highly affected by the inorganic content of the fuel. The presence
of inorganic impurities can hinder the release of chlorine to gas-
eous phase as HCl; a large fraction of the original chlorine is thus
retained in tar, char, ash and slag after gasification (Takeda et al.,
2009). On the basis of equilibrium calculations, Kuramochi et al.
(2005) demonstrated the effect of KCl formation on HCl emission.
They concluded that chlorine atoms in the feedstock almost form
solid KCl at low temperatures up to 823–873 K. At higher temper-
atures, the solid KCl is partially transformed to vapor phase as KCl
gas and the rest is evolved as HCl and NaCl gases. In addition, at
high temperatures, as the concentration of alkali chlorides in-
creases, the HCl emission decreases. We can therefore expect
the presence of steam to enhance the release of potassium from
the solid matrix and thus increase the likelihood of reaction with
chlorine to form KCl. Accordingly, reduced HCl emission is
achieved.
The concentration of HCl in the product gas at ER = 0.2 and
S/C = 1 was almost stable at 83 ppmv within the temperature
range of 1023–1123 K. However, previous investigations have
shown different trends with temperature change. Dias and
Gulyurtlu (2008) observed a slight decrease in the concentration
of HCl in the gas phase, from 95 to 65 ppmv when increasing
the temperature from 993 to 1173 K. van Paasen et al. (2006)
conducted gasification tests on several non-woody biomass fuels
in a bubbling fluidized bed gasifier with ER = 0.22. In gasification
tests of a mixed biomass fuel (sewage sludge and white pellets),
they found an increase of HCl concentration in the product gas
from 36 to 54 g Nm3 with increasing the gasification tempera-
ture from 1023 to 1123 K.
Equilibrium calculations (Kuramochi et al., 2005), indicated
that increasing the temperature will increase the formation of
metallic chlorides such as KCl, thereby, reducing the levels of
HCl in the gas phase. However, the concentrations of HCl in the
product gas within the temperature range of 1023–1123 K did
not change. This suggests that the formation of KCl was hindered
by other, parallel, reactions. The presence of metals such as
aluminum and silicon might have led to the formation of potas-
sium aluminum silicate products rather than the formation of
KCl, thus making the levels of the HCl in the product gas higher
than expected.
4. Conclusions
We investigated the effect of operational conditions (residence
time, ER, S/C ratio and gasification temperature) on tar levels and
H2S, COS, NH3, HCN and HCl concentrations in the product gas dur-
ing cedar wood gasification. There were substantial variations in
the release of contaminants in the product gas. Over the range of
operating conditions tested, the tar content in the product gas ran-
ged from 33.2 to <1 g Nm3. Higher temperatures, along with suf-
ficient contact time can assist in PAH reduction. Steam can slightly
increase PAH conversion, and increasing the ERs can greatly reduce
PAH levels in the product gas.
The conversion of fuel-sulfur to volatile-sulfur gases increases
with increasing the ERs and S/C ratio. Most of fuel-sulfur is
converted to H 2 S gas. The conversion of fuel-nitrogen to
volatile-nitrogen gases increases with increasing the ERs and S/C
ratio. The evolution rate of NH3 is higher than that of HCN. The
evolution of HCl is slightly increased by the presence of oxygen,
but increasing the S/C ratio beyond 1 tends to decrease the concen-
tration of HCl in the product gas.
Acknowledgments
The authors thank Mr. Kimura Makoto and Mr. Motoki Tosh-
iyuki of the National Institute for Environmental Studies for their
assistance in conducting the experiments.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.chemosphere.
2012.08.030.
References
Abu El-Rub, Z.Y.K., 2008. Biomass Char as an in situ catalyst for tar removal in
gasification systems. PhD thesis. Twente University, Twente, The Netherlands.
Aljbour, S.H., Kawamoto, K., 2012. Bench-scale gasification of cedar wood. part I:
Effect of operational conditions on the product gas characteristics.
Chemosphere http://dx.doi.org/10.1016/j.chemosphere.2012.08.029.
Attar, A., Dupuis, F., 1979. The rate and the fundamental mechanisms of the reaction
of hydrogen sulfide with the basic minerals in coal. Ind. Eng. Chem. Process Des.
Dev. 18, 607–618.
Brage, C., Yu, Q., Chen, G., Sjöström, K., 1997. Use of amino phase adsorbent for
biomass tar sampling and separation. Fuel 76, 137–142.
Brage, C., Yu, Q., Chen, G., Sjöström, K., 2000. Tar evolution profiles obtained from
gasification of biomass and coal. Biomass Bioenergy 18, 87–91.
Brage, C., Yu, Q., Sjöström, K., 1996. Characteristics of evolution of tar from wood
pyrolysis in a fixed-bed reactor. Fuel 75, 213–219.
Chen, H., Li, B., Zhang, B., 1999. Effects of mineral matter on products and sulfur
distributions in hydropyrolysis. Fuel 78, 713–719.
Cui, H., Turn, S.Q., Keffer, V., Evans, D., Tran, T., Foley, M., 2010. Contaminant
estimates and removal in product gas from biomass steam gasification. Energy
Fuels 24, 1222–1233.
Devi, L., Ptasinski, K.J., Janssen, F.J.J.G., 2003. A review of the primary measures for
tar elimination in biomass gasification processes. Biomass Bioenergy 24, 125–
140.
Dias, M., Gulyurtlu, I., 2008. H2S and HCl formation during RDF and Coal co-
gasification: a comparison between the predictions and experimental results.
Proceedings of the Biomass Gasification Technologies Workshop MRC Gebze
Campus-Türkiye.
Gulyurtlu, I., Pinto, F., Lopes, H., André, R.N., Dias, M., Cabrita, I., 2008. Prediction of
H2S and HCl formation during RDF and coal co-gasification in fluidized bed. In:
Proceedings of the 16th European Biomass Conference and Exhibition, Valencia,
Spain.
Houben, M.P., 2004. Analysis of Tar removal in a partial oxidation burner. PhD
thesis. Eindhoven University of Technology, Eindhoven, The Netherlands.
Jess, A., 1996. Mechanisms and kinetics of thermal reactions of aromatic
hydrocarbons from pyrolysis of solid fuels. Fuel 75, 1441–1448.
Kawamoto, K., Wu, W., Kuramochi, H., 2009. Development of gasification and
reforming technology using catalyst at lower temperature for effective energy
recovery: hydrogen recovery using waste wood. J. Environ. Eng. 4, 409–
421.
Kuramochi, H., Wu, W., Kawamoto, K., 2005. Prediction of the behaviors of H2S and
HCl during gasification of selected residual biomass fuels by equilibrium
calculation. Fuel 84, 377–387.
Leduc, S., Schmid, E., Obersteiner, M., Riahi, K., 2009. Methanol production by
gasification using a geographically explicit model. Biomass Bioenergy 33, 745–
751.
Leppälahti, J., 1993. Formation and behaviour of nitrogen compounds in an IGCC
process. Biores. Technol. 46, 65–70.
Li, X.T., Grace, J.R., Lim, C.J., Watkinson, A.P., Chen, H.P., Kim, J.R., 2004.
Biomass gasification in a circulating fluidized bed. Biomass Bioenergy 26,
171–193.
Meng, X., de Jong, W., Fu, N., Verkooijen, A.H.M., 2011. Biomass gasification in a
100 kWth steam-oxygen blown circulating fluidized bed gasifier: effects of
operational conditions on product gas distribution and tar formation. Biomass
Bioenergy 35, 2910–2924.
Takeda, M., Ueda, A., Hashimoto, H., Yamada, T., Suzuki, N., Sato, M., Tsubouchi, N.,
Nakazato, Y., Ohtsuka, Y., 2009. Fate of the chlorine and fluorine in a sub-
bituminous coal during pyrolysis and gasification. Fuel 85, 235–242.
Tian, F.-J., Yu, J., McKenzie, L.J., Hayashi, J.-i., Li, C.-Z., 2007. Conversion of fuel-N into
HCN and NH3 during the pyrolysis and gasification in steam: a comparative
study of coal and biomass. Energy Fuel 21, 517–521.
 S.H. Aljbour, K. Kawamoto / Chemosphere 90 (2013) 1501–1507 1507

van Paasen, S.V.B., Cieplik, M.K., Phokawat, N.P., 2006. Gasification of Non-woody
Biomass: Economic and Technical Perspectives of Chlorine and Sulphur
Removal from Product Gas, (Non-confidential version). Energy Research
Center of the Netherlands, Petten, The Netherlands, ECN-C-06-032.
Wang, L., Weller, C.L., Jones, D.D., Hanna, M.A., 2008. Contemporary issues in
thermal gasification of biomass and its application to electricity and fuel
production. Biomass Bioenergy 32, 573–581.
Wang, X., Si, J., Tan, H., Ma, L., Pourkashanian, M., Xu, T., 2010. Nitrogen, sulfur, and
chlorine transformations during the pyrolysis of straw. Energy Fuel 24, 5215–
5221.
Zhou, J., Masutani, S.M., Ishimura, D.M., Turn, S.Q., Kinoshita, C.M., 2000. Release of
fuel-bound nitrogen during biomass gasification. Ind. Eng. Chem. Res. 39, 626–
634.