Fuel 183 (2016) 132138

Contents lists available at ScienceDirect

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

Full Length Article

Co-gasification of coal and biomass in a fixed bed reactor with separate

and mixed bed configurations
Yan Zhang , Yan Zheng
Key Laboratory of Ocean Energy Utilization and Energy Conservation of Ministry of Education, Dalian University of Technology, No. 2 Linggong Road, Ganjingzi District,
Dalian City 116024, China

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

Co-gasification of coal and biomass

was conducted in a fixed bed reactor.

Two separate and a mixed-bed

configurations were designed for

comparative study.

The mixed-bed configuration
produced a well-dispersed bio-ash
among coal char grains.

A stronger synergy effect was
observed in the mixed bed
configuration rather than in the
separate configuration.

The bio-char bed exhibited better tar
cracking performance than the coal
char bed.

a r t i c l e

i n f o

Article history:
Received 27 March 2016
Received in revised form 27 May 2016
Accepted 14 June 2016
Available online 21 June 2016
Keywords:
Co-gasification
Fixed-bed
Coal
Biomass
Synergy

a b s t r a c t
This study inspects the occurrence of synergy in the co-gasification of coal and biomass in a fixed bed
reactor. Two separate and a pre-mixed bed configurations with a congruent sample mass are designed
for comparative study. The intimate contact between bio-ash and coal char grains is a prerequisite for
the occurrence of the synergy effect in the co-gasification of coal and biomass. The separate bed configurations generally form either a coal-char/bio-ash or a bio-ash/coal-ash interface, whereas the pre-mixed
bed configuration produces well-dispersed bio-ashes that come into close contact with coal char grains.
As a result, a stronger synergy effect occurs in the mixed bed configuration rather than in the separate
configuration. A novel method is explored to collect and determine the heavy tar and water yields
generated during the initial pyrolysis step. Results show that bio-char bed has better tar cracking performance than coal char bed.
2016 Elsevier Ltd. All rights reserved.

1. Introduction
Co-gasification of coal and biomass in existing coal-based
systems is a promising approach to provide both environmental
and economic benefits, including considerable reduction in CO2
Corresponding author.
E-mail address: zhang.yan@dlut.edu.cn (Y. Zhang).
http://dx.doi.org/10.1016/j.fuel.2016.06.066
0016-2361/ 2016 Elsevier Ltd. All rights reserved.

emission, less waste disposal problems, and low fuel cost. However, the use of biomass when co-gasified with coal may simultaneously exist technical risk in terms of deterioration of the
system performance due to its inferior chemical and physical properties such as high moisture and volatile contents, low calorific
value and bulk density, and fibrous structure. Therefore, proper
choices of biomass/coal blends and system design and operation
are essential.

Y. Zhang, Y. Zheng / Fuel 183 (2016) 132138

Previous studies on co-gasification have been carried out

through different types of reactors such as thermogravimetric
analyses (TGA) [14], fixed-bed reactor [59], and fluidized-bed
reactor [10,11]. Much attention of these studies has been focused
on whether there is any possible interaction or synergy between
co-processed materials. If synergy occurs in coal/biomass blends,
it is expected that an attractive possibility of improving the overall
efficiency of the co-processing system is available, which can be in
turn to mitigating the aforementioned technical risk. However,
there is a contradiction on whether synergies occur even when
studies have been conducted with the same type of facilities. For
example, Fermoso et al. [6] and Howaniec et al. [5] reported
that an improvement in hydrogen production and cold gas
efficiency was achieved during the co-gasification of coal and
biomass in a fixed bed reactor. However, Collot et al. [12] found
some hints of synergy in the volatile yield in a fixed-bed reactor,
but they were too small to constitute clear evidence of synergy.
Contradictions on the existence of synergy between coal and
biomass suggest that further systematic researches on this issue
are needed.
Some scholars working on the topic of synergy have proposed
the use of weighted average performance parameters as the criteria for evaluating synergy, i.e., a nonlinear increase in gas yield,
char reactivity, or a nonlinear decrease in tar yield [24,13]. Three
or more gasification tests on individual biomass, coal, and their
blends with varying blending ratios should generally be conducted
at the same setting temperature and a fixed gas flow rate.
The synergy effect is then estimated from the deviation of
weighted average and experimental parameters. However, the
non-comparability of the weighted average and experimental performance parameters in some cases is an issue that has always
been overlooked. For example, nonidentity in both the total and
individual initial masses of coal and biomass may be linked to differences in both the external and internal diffusion effects of individual and co-gasification tests. A previous study revealed that the
sample mass dependence of char reactivity could be misdiagnosed
as synergy or inhibition as a result of the diffusion effect during the
co-gasification of coal and biomass [14]. Furthermore, additional
operational variables in terms of the gas-to-fuel ratio could not
be avoided even by using a fixed gas flow rate.
Taken tar yield as a topic of the research, quantitative
measurement of tar in the product gas is a great challenge. Some
uncertainty always exists in this measurement, because tar is not
well defined in literature. For example, it can refer to a sum of components with boiling points higher than 105 C [15], or it may
include all condensable organics with a molecular weight larger
than benzene [16]. A large variety of sampling and analysis methods have been developed to determine the concentration of tar in
the product gas, which makes uncertainty in tar definition and
the comparison of data among researchers. Most tar measurements are based on cold solvent trapping (CST) in impinger trains
with various organic solvents [1619]. This type of sampling has
drawbacks such as the long period for sampling, which provides
limited information about transient gasifier conditions [18]. In
addition, this methodology generally obtains a mixture of tar, solvents, and water. Evaporation or distillation must be taken for tar
determination.
This study investigates the co-gasification of Chinese
bituminous coal with forestry waste and the agricultural residue
in CO2 in a fixed bed mode. One mixed and two separate bed
configurations with congruent mass are designed for comparative
study. The study focuses on CO production, by which synergy
effects between the co-processed fuels can be evaluated. A novel
method to collect and determine the heavy tar and water yields
during the initial pyrolysis step is also explored.

133

2. Experimental
2.1. Materials and analyses
The bituminous coal used in this study was collected from the
Ningdong mine in the Ningxia Hui Autonomous Region, China. This
coal sample is denoted by NBC hereafter. Chinese redwood (CR)
and soybean stalk (SS) were selected as biomass samples. The
former is a woody waste, while the latter is an agricultural residue.
Both coal and biomass samples were ground and sieved into particle sizes of 0.250.71 mm. All the samples were used for experimental tests in an air-dried base.

2.2. Apparatus and procedures

The gasification tests for coal and biomass were performed in a
fixed bed reactor. The schematic construction of the reactor is
shown in Fig. 1. A quartz tube (400 mm high and 14 mm i.d.)
was set vertically inside an electric furnace. The quartz filter was
fixed in the quartz tube and located at the central heating zone
of the furnace. The sample was previously placed in a tubular
hopper (60 mm high and 14 mm i.d.) connected to the reaction
tube by a manually controlled valve. The valve allowed only purge
or reactant gas to pass through but completely stopped the sample
particles when it was closed. The reactor was pre-heated to 850 C.
After keeping the said temperature constant for 5 min, the valve
was opened, and the sample particles were dropped into the
reactor tube. The gasification of individual coal, biomass, or their
separate and mixed bed configurations was conducted isothermally at 850 C in a mixed CO2N2 stream (50 vol%). The total
gas flowrate was always 200 mL/min.
A total of 1.0 g of NBC coal and 0.5 g of CR or SS were used for
their individual gasification tests. For the study of the cogasification of coal and biomass, three sets of bed configurations
were designed: (1) 1.0 g of NBC was located above 0.5 g of the CR
or SS bed in the tubular hopper, as shown in inset (a) of Fig. 1. This
bed configuration is denoted by NBC/CR or NBC/SS. (2) 0.5 g of CR
or SS was situated above 1.0 g of the NBC bed in the tubular hopper, as exhibited in inset (b) of Fig. 1 and represented by CR/NBC
or SS/NBC. (3) 1.0 g of NBC was previously mixed with 0.5 g of
CR or SS and then packed into the tubular hopper, as displayed
in inset (c) of Fig. 1 and denoted by NBC-CR or NBC-SS. Preliminary
tests were done to check the configurations when the samples
were dropped from the tubular hopper into the reactor tube with
mixed gas. In the present study, the tubular hopper and the reactor
tube had the same inner diameter. It was confirmed that the
samples dropped into the reactor tube kept their original configurations well as in the tubular hopper. All experiments were
repeated 23 times and mean results were used to determine the
synergy effect among different configurations.
Certain efforts were made to collect the tar and water produced
in the initial pyrolysis/gasification step. The outlet of the reactor
tube was series-connected with two dry condenser tubes and a
chemical trap. The first condenser tube was an L-shaped quartz
tube packed with quartz wool to stop the heavy tars. The second
one was a straight Pyrex glass tube stored with polypropylene
wool to adsorb light tars. The chemical trap was a U-glass tube
in which 20 g of anhydrous CaCl2 was placed to adsorb water. All
condenser and trapping tubes were weighed before the experiments. The total yield of heavy and light tars was determined
through the weight deviation of the first and second trapping tubes
before and after gasification. The yield of water was evaluated from
the weight deviation of the U-glass tube before and after
gasification.

134

Y. Zhang, Y. Zheng / Fuel 183 (2016) 132138

Fig. 1. Schematic construction and bed configurations of the fixed bed reactor. Gas cylinder Gas flowmeter Tubular hopper Reactor tube Quartz wool Electric
furnace L-shaped quartz tube Pyrex glass tube U-glass tube Micro-gas chromatography Personal computer.

The product gases from the outlet of U-glass tube (water free)
were online analyzed by micro-gas chromatography (INFICON,
3000 Micro GC), which equipped with three different columns
and TCD detectors and took a sample every 4 min. CO2 and light
hydrocarbon species with carbon numbers from two to three (C2
to C3) were analyzed using a Pora Plot Q column connected to a
backflush inlet. Other gas species were analyzed using two MS5A columns. One was for H2 and the other was for N2, CO, and CH4.
2.3. Spectroscopic characterization of the surface of NBC and its cogasified chars
The surface state of coal and its chars was observed using a
scanning electron microscope (SEM) equipped with an energy dispersive X-ray spectrometer (EDX) (SEM HITACHI SU6600). The elemental composition of a sample is determined using characteristic
X-ray spectrum of the specimen being examined. The elemental
analysis was performed in a spot mode in which the beam is
localized on a given area manually chosen within the field of view.
Each original sample and its derived chars were characterized by
randomly selecting 3 or more particles of view. EDX spectra were
taken at an acceleration voltage of 20 kV and presented on the
same intensity scale.
3. Results and discussion
As far as the co-gasification is concerned, many researchers
investigated co-gasification characteristics of individual fuels and
their blends with varying blending ratios. The synergy effect was
frequently evaluated from the deviation between weighted average data of individual fuels and experimental results of the blends
[24,13]. However, variation in fuel blending ratios may simultaneously link to differences in both gas-to-fuel ratio and diffusion
effects of individual and co-gasification tests, which possibly make
a misdiagnosis on synergy or inhibition [14].
In this study, synergy effects are inspected by investigating
gasification characteristics of individual coal (1.0 g) and biomass

(0.5 g) as well as two separate and a pre-mixed bed configurations

with congruent sample masses as shown in Fig. 1. These experiments are designed based on the following two hypotheses. First,
the total sample mass of the three bed configurations (1.0 g of coal
and 0.5 g of biomass) is larger than that of individual coal bed
(1.0 g). If no synergy occurred between coal and biomass, the times
required for complete gasification of the separate and pre-mixed
bed configurations should be longer than that of individual coal
bed. Second, for a given gasification temperature and an identical
gas flow rate, the three bed configurations shown in Fig. 1 have
the same gas-to-fuel ratio. If no interactions occurred between coal
and biomass, the product yields (gas, tar, and water) and the times
required for complete gasification of the three bed configurations
should be always the same. Accordingly, conclusions on synergy
effects or interactions between coal and biomass in the following
sub-sections are drawn on the basis of testing the above
hypotheses.
3.1. Tar formation
In the present study, all the experiments were conducted by
quickly dropping the sample particles from the sample storage unit
into the hot zone of the reactor tube, where the samples were
stopped by a quartz filter. The pyrolysis and gasification reactions
simultaneously occurred under the mixed CO2/N2 gas flow:
Pyrolysis

Cx Hy Oz !Tar;H2 O; Char Cn Hm s ; Cn Hm g ; CO2 ; CO;H2 

Secondary decomposition

Cn Hm g ! C H2
Carbonization

Char Cn Hm s ! C H2
Gasification

C CO2 ! 2CO

1
2
3
4

Eq. (1) indicates the initial fast pyrolysis step, during which tar,
water, primary char ((CnHm)s), and gases are generated. Eq. (2)
denotes the secondary decomposition of hydrocarbon gases

135

Y. Zhang, Y. Zheng / Fuel 183 (2016) 132138

((CnHm)g), and Eq. (3) represents the further carbonization of the

primary char that remains in the reactor and produces graphitized
carbon and hydrogen. Eq. (4) describes the gasification step of
carbon with CO2, which is a rate-determining step that may control
the overall conversion process of coal and/or biomass.
According to Eqs. (1)(4), tar is primarily formed in the initial
pyrolysis step (Eq. (1)). A novel attempt of this study is to determine and evaluate tar formation. As described in Section 2.2, two
dry condenser tubes are used to collect tar products. The first is
an L-shaped quartz tube packed with quartz wool. The temperature of the product gas flowing past this unit is generally higher
than 200 C. Thus, only heavy tars are stopped here, which should
be water free. The second dry condenser unit is a straight Pyrex
glass tube in which polypropylene wool was placed to adsorb light
tars. The temperature of the product gas flowed through this zone
generally cooled down to 110120 C. Polypropylene wool is a
hydrophobic material that can adsorb some hydrophilic tar but
allows water vapor to pass through. The tar collected in this unit
is a relatively lighter fraction compared with that collected in the
L-shaped quartz tube. But it in principle belongs to heavy tar class.
In this way, the total yield of heavy tar is determined according to
the weight deviation of the first and second trapping units before
and after gasification. No solvents are employed for collection
and analysis. Quartz and polypropylene wools can be disposed of
upon completion of the gravimetric analysis.
Fig. 2 summarizes the three sets of experimental results for the
tar formation in the initial fast pyrolysis step. The first set of bar
charts presents the tar yields of individual NBC, CR, and SS. The
tar yields of CR and SS are 45 times greater than that of NBC.
The second set of bar charts indicates the tar yields obtained during the co-gasification of NBC and CR in different bed configurations. The results indicate that the tar yield of the three bed
configurations of NBC with CR increases in the order of CR/
NBC > NBC-CR > NBC/CR. Coincidentally, the third set of bar charts
in Fig. 2 shows that the tar yield of the three bed configurations of
NBC with SS increases in the order of SS/NBC > NBC-SS > NBC/SS.
Reproducibility confirmation has been performed by repeating
the experiments 23 times. All these results provided a verification
conclusion on the second hypothesis as proposed earlier in this
section, that is, interactions between coal and biomass occurred
in the three bed configurations.
Special explanations should be given for the different tar yields
of the NBC/CR (3.5 wt%) and CR/NBC (5.4 wt%) bed configurations.
Fig. 1 indicates that all the experiments were conducted in a downstream fixed bed mode. The tar components generated from various bed configurations in this mode undergo different thermal
histories. For example, in the case of CR/NBC, CR tar passes through

the NBC char bed. On the contrary, in the case of NBC/CR, NBC tar
passes through the CR char bed. The difference between the
heights of the NBC and CR beds is supposed to be a possible consequence of the difference between the tar yields of CR/NBC and
NBC/CR. Table 1 shows that the bulk densities of NBC, CR, and SS
are 0.732, 0.222, and 0.270, respectively. Significant differences
among the bulk densities of NBC, CR, and SS certainly cause their
different bed heights in the reactor tube. Indeed, the bed heights
of 1.0 g of NBC, 0.5 g of CR, and 0.5 g of SS in the reactor tube are
approximately 0.9, 1.6, and 1.3 cm, respectively. The different tar
formation behaviors of NBC/CR and CR/NBC suggest the different
tar cracking performances of the NBC and CR char beds. That is,
in the case of NBC/CR, CR char bed exhibits better performance
for NBC-tar cracking. On the contrary, in the case of CR/NBC, NBC
char bed exhibits worse performance for CR-tar cracking. The
abovementioned explanations are also suitable for NBC/SS and
SS/NBC in the third set of bar charts.
3.2. Water generation
Water is another liquid product that is mainly generated from
the initial fast pyrolysis step, as indicated in Eq. (1). Conventionally, water vapor in the product gas was simultaneously collected
with tar based on CST in impinger trains with various organic solvents [1619]. The water content in solvents was then subjected to
quantitative analysis by using a Karl Fischer moisture titrator [19].
This method is expensive and time-consuming. This study
employed a chemical trap to replace the CST and other cooling condensers to determine water generation during the initial pyrolysis.
A total of 20 g of anhydrous CaCl2 was packed in a U-glass tube
connected to a Pyrex glass tube in front of it. The product gas
flowed through this unit was generally cooled down to room temperature. All water vapor in the product gas was chemically
adsorbed by anhydrous CaCl2. After the experiments, the CaCl2 in
the U-glass tube was observed to be colorless to the naked eye.
In this way, a simple gravimetric analysis of the U-glass tube,
before and after the experiment, allows the mass of water to be
determined. The results obtained via this method are believed to
be more reliable for determining water production than those
obtained from the conventional water- or ice-cooled condenser
in a laboratory scale unit as used in this study.

Table 1
Properties of coal and biomass samples.

12

Tar [wt%]

10
8
6
4
2
0

Fig. 2. Tar yields of individual and co-processed fuels in the initial pyrolysis/gasification step.

Abbreviation

NBC

CR

SS

Proximate analysis (wt%, db)

Ash
Volatiles
Fixed carbon

9.7
36.8
53.5

3.5
75.1
21.4

5.3
75.9
18.8

Ultimate analysis (wt%, db)

Carbon
Hydrogen
Nitrogen
Oxygen & sulphura
Bulk density (g/cm3)

70.5
4.1
1.0
14.7
0.732

46.1
5.8
0.3
44.3
0.222

44.2
5.6
0.9
44.0
0.270

Ash composition (wt%, db)

SiO2
Al2O3
Fe2O3
CaO
MgO
Na2O
K2O
TiO2
P2O5
SO3

39.8
18.4
11.7
13.7
4.7
1.8
1.0
0.8
0.2
5.2

20.6
2.2
2.7
58.3
5.1
1.7
6.8
0.1
2.1
0.2

22.6
3.7
1.7
17.1
8.1
4.1
37.4
0.4
3.2
0.7

By difference.

Y. Zhang, Y. Zheng / Fuel 183 (2016) 132138

30

30

25

25

20
15

CO [Vol.%]

Water [wt%]

136

10

NBC/CR

20

15

1
0

10

(a)
0

30

Fig. 3 summarizes three sets of experimental results for water

generation during the initial pyrolysis step. The three sets of bar
charts presenting the water yield in Fig. 3 correspond to those of
the bar charts that exhibit the tar yield in Fig. 2. Compared with
the tar yields shown in Fig. 2, no evident differences in water yield
are observed among the three co-processing bed configurations.
3.3. Production of main gases
Several additional remarks should be addressed in this section
before discussing the formation behavior of the product gasses:
(1) Online analysis of the product gases was conducted through
micro-gas chromatography, which was set to take a sample gas
every 4 min. The first group of the analysis data of the product
gases was collected when the samples were dropped into the reaction tube for 5 min. (2) Among the light hydrocarbon gas species
(CnHm)g, only a trace amount of CH4 was detected under the experimental conditions used in this study, and its content in the product gas was below 0.1 vol%. Thus, the formation behaviors of
CH4 and other light hydrocarbon gas species are not discussed in
this study.
Fig. 4 shows the CO formation profiles of NBC with CR (Fig. 4a)
or SS (Fig. 4b) in three different bed configurations at 850 C, along
with those of their individuals for comparison. CO is the primary
gas product during CO2 gasification process. As indicated in Fig. 4a,
the CO concentration in the product gas is found to monotonously
decrease with the gasification time in all experimental runs. The
CO concentration approaching zero implies that char gasification
is almost complete. In this study, all the gasification tests were
conducted in a mixed CO2N2 stream (50 vol%). The total gas flowrate was always 200 mL/min. N2 is an inert gas during gasification,
which acts as balance gas for quantification of other product gases.
Given that the three bed configurations are the congruent in sample mass and the same in gas-to-fuel ratio, the CO concentration in
the product gas can be taken as a direct evaluation of the gasification process or synergy. That is, the sharper the decrease of CO concentration in the product gas, the faster the gasification reaction.
Likewise, the faster the time of CO concentration approaching zero,
the stronger the synergy degree occurred between coal and
biomass.
In Fig. 4a, the most notable difference in the times required for
complete gasification was observed in 0.5 g of individual CR and
1.0 g of individual NBC. The former took only 13 min, whereas
the latter required as long as 137 min. By contrast, the total reaction times of the three different bed configurations with congruent
masses did not exhibit any remarkable difference. We now prove
our hypotheses again based on this figure. First, the sample masses
in CR/NBC, NBC/CR, or NBC-CR beds (0.5 g of CR and 1.0 g of NBC)

20

40

NBC/SS

25

CO [Vol.%]

Fig. 3. Water yields of individual and co-processed fuels in the initial pyrolysis/gasification step.

NBC-CR

NBC

CR

CR/NBC

60

100 110 120 130 140

80

Time [min]

SS/NBC

100

NBC-SS

120

NBC

140

SS

20
15
10
5
0

(b)
0

20

40

60

80

100

120

140

Time [min]
Fig. 4. CO production during combined pyrolysis/gasification of NBC with (a) CR
and (b) SS.

were larger than that of individual NBC bed (1.0 g). If no synergy
occurred between NBC and CR, the times required for complete
gasification of the three co-gasified samples should be longer than
that of individual NBC. The observed results did not follow this
hypothesis. As can be seen from the inset in Fig. 4a, the complete
gasification times of the three co-gasified samples with a total
mass of 1.5 g were either comparable to or slightly faster than that
of individual NBC with a total mass of 1.0 g. This finding proves
that a certain synergy occurred during the co-gasification of NBC
and CR. Second, for the CR-related three bed configurations, CR/
NBC, NBC/CR and NBC-CR had congruent masses (1.0 g of NBC
and 0.5 g of CR) and the same gas-to-fuel ratio. If no interactions
occurred between NBC and CR, the complete gasification times of
the three tests should be always the same. Unfortunately, the
results were not consistent with this hypothesis. The inset in
Fig. 4a shows that the complete gasification time of CR/NBC
(137 min) was slightly longer than those of NBC/CR and NBC-CR
(129 min). This means that interactions or synergy effects in
NBC/CR and NBC-CR were slightly stronger than those in CR/NBC.
Fig. 4b shows evident changes in the CO formation profiles of
the SS-related samples. The gasification reactions of SS/NBC,
NBC/SS, and NBC-SS completed at 129, 117, and 93 min, respectively. The time of SS/NBC (129 min) was slightly shorter than
those of individual NBC (137 min) and the same configuration of
CR/NBC (137 min in Fig. 4a). By contrast, the times of NBC/SS
(117 min) and NBC-SS (93 min) were significantly shorter than
that of individual NBC as well as those of CR-related bed configurations (129137 min in Fig. 4a). On proving the hypotheses
described above, the following conclusions can be drawn: (1) the
occurrence of synergy between NBC and SS show strong
dependence on bed configurations. The mixed bed configuration
(NBC-SS) achieves the largest synergy effect among the three bed
configurations studied; and (2) by the comparison of
Fig. 4a and b, the synergy degree of NBC-SS is clearly greater than
that of corresponded NBC-CR. That is, the complete gasification
times of NBC-SS (93 min) and NBC-CR (129 min) are 44 min and
8 min faster than that of individual NBC (137 min), respectively.

Y. Zhang, Y. Zheng / Fuel 183 (2016) 132138

It is interesting to know why NBC/SS exhibits excellent

performance in comparison with SS/NBC, as shown in Fig. 4b. In
our previous paper [20], synergy is confirmed to start with the
formation of biomass ash during the co-gasification of coal and
biomass, where the potassium species in biomass ash have a catalytic role in promoting the gasification reactivity of coal char.
The difference in the catalytic behavior of SS ash in NBC/SS and
SS/NBC bed configurations is supposed to be the principle response
of their different gasification performances. First, in the case of the
NBC/SS separate bed configuration, CO2 reacts with NBC, and the
product gas then passes through the SS bed. Fig. 4b shows that
the gasification rate of individual SS is considerably faster than that
of individual NBC. This means that the SS bed can be completely
gasified in a short period of time to result in an SS ash layer under
the NBC bed, where an NBC-char/SS-ash interface forms. SS-ash is
rich in potassium (Table 1), which promotes the gasification
reaction of NBC chars on the interface. Second, the pathway of
the reactant gas in the SS/NBC bed configuration is in contrast to
that in NBC/SS. Although the masses of individual NBC and SS in
the NBC/SS and SS/NBC separate bed configurations are always
the same, the complete gasification time of SS/NBC (133 min) is
evidently longer than that of NBC/SS (117 min), as shown in Fig. 4b.
When a 50 vol% CO2 gas stream first flows through the SS bed, CO2
reacts with SS to form an SS-ash layer on top of the NBC bed. With
the progress of gasification, the NBC chars at the upper side of the
NBC bed react with CO2 to form a NBC-ash layer, where an
SS-ash/NBC-ash interface is formed. In this case, the SS-ash layer
no longer catalyzes the gasification of NBC chars.
This section focuses on the NBC-SS mixed bed configuration.
Fig. 4b shows that the said configuration exhibits a remarkable
enhancement in gasification time compared with NBC/SS,
SS/NBC, and individual NBC bed configurations. The reaction time
of NBC-SS is shortened for approximately 44 min compared with
the time of individual NBC and SS/NBC. Given the confirmed catalytic effect of SS ash stated in the previous paper [20] and in this
study, this enhancement scenario is possibly attributable to the
well-dispersed SS ash among the NBC grains. This result strongly
suggests that the pre-mixed bed configuration is the best means
NBC/CR

H2 [Vol.%]

1.5

CR/NBC

NBC-CR

NBC

137

of achieving synergy between potassium-rich biomass and coal.

In addition, this study provides further support to the synergy
effects between coal and biomass during co-gasification, as
observed by the congruent-mass thermogravimetric analysis
(TGA) method [20].
Fig. 5 shows the H2 formation profiles of NBC with CR (Fig. 5a)
or SS (Fig. 5b) in three different bed configurations. H2 should be
the byproduct gas produced when CO2 is used as a gasifying agent.
Fig. 5a and b shows that the H2 contents of individual CR and SS
approach zero after a gasification time of 9 min. By contrast,
individual NBC and the three co-gasification bed configurations
show similar H2 formation profiles, where the H2 contents sharply
decrease before the reaction time reaches 13 min and then slightly
decrease to almost zero (less than 0.1 vol%) after the gasification
time reaches approximately 100 min. The sharp decrease in H2 in
the first stage reflects the progress of both the primary initial
pyrolysis of the original fuels (Eq. (1)) and the secondary decomposition of light hydrocarbons (Eq. (2)). Results indicate that the
initial pyrolysis and secondary decomposition reactions are almost
complete after 13 min. By contrast, the production of H2 in the
second stage mainly originates from the carbonization of the
primary coal char (Eq. (3)). This scenario suggests that char carbonization and carbon gasification simultaneously occur during
the CO2 gasification of coal and biomass. No evident difference is
observed between the H2 formation profiles of the CR- and SSrelated bed configurations.

CR

(a)

0.5

0
0

20

40

60

80

100

Time [min]

H2 [Vol.%]

NBC/SS

SS/BC

BC-SS

BC

SS

1.5
1

(b)

0.5
0
0

20

40

60

80

100

Time [min]
Fig. 5. H2 production during combined pyrolysis/gasification of NBC with (a) CR
and (b) SS.

Fig. 6. SEM-EDS spectra of (a) original NBC and its chars obtained after cogasification with (b) CR and (c) SS.

138

Y. Zhang, Y. Zheng / Fuel 183 (2016) 132138

3.4. Inspection of the synergy performance in NBC-CR and NBC-SS

References

This section inspects the different synergy behaviors of NBC-CR

and NBC-SS. A comparison of Figs. 4a and 5a show that the synergy
degree of NBC-SS is clearly more pronounced than that of NBC-CR.
The complete gasification times of NBC-SS (93 min) and NBC-CR
(129 min) are 44 min and 8 min faster than that of individual
NBC (137 min), respectively. To give a reasonable explanation for
this scenario, SEM-EDX is used to analyze the surface elemental
composition of original NBC and its chars obtained after cogasification with CR and SS. Even though EDX technique is lack
of quantitative accuracy, the relative amounts of the corresponding
elements can be inferred from relative peak heights in EDX spectra
[21]. For acquiring a visual comparison, all EDX spectra are taken at
the same conditions and presented on the same intensity scale.
Fig. 6 shows the typical comparative results. Fig. 6a indicates that
no potassium peak exists in the spectrum of the original NBC particles. By contrast, potassium peaks are observed in the spectra of
the NBC chars obtained after the gasification of NBC-CR (Fig. 6b)
and NBC-SS (Fig. 6c), where the potassium peak intensity of NBC
chars from NBC-SS is considerably stronger than that from
NBC-CR. The results suggest some hints explaining why NBC-SS
exhibits a considerably stronger synergy than NBC-CR. First, the
appearance of potassium peaks in the SEM-EDS spectra of the
NBC chars from NBC-CR and NBC-SS means that the migration of
potassium from CR and SS to the surface of NBC char definitely
occurs during co-gasification. Second, a review of the data of the
ash components given in Table 1 shows that SS is rich in
potassium, whereas CR is more abundant in calcium. According
to the comparison of the potassium peak intensities in
Fig. 6b and c, potassium-rich SS gives more potassium migration
to the surface of NBC chars than calcium-rich CR. As a result,
potassium-rich SS exhibits stronger synergy than Ca-rich CR.

[1] Brown RC, Liu Q, Norton G. Catalytic effects observed during the co-gasification
of coal and switchgrass. Biomass Bioenergy 2000;18:499506.
[2] Krerkkaiwan S, Fushimi C, Tsutsumi A, Kuchonthara P. Synergetic effect during
co-pyrolysis/gasification of biomass and sub-bituminous coal. Fuel Process
Technol 2013;115:118.
[3] Habibi R, Kopyscinski J, Masnadi MS, Lam J, Grace JR, Mims CA, et al. Cogasification of biomass and non-biomass feedstocks: synergistic and inhibition
effects of switchgrass mixed with sub-bituminous coal and fluid coke during
CO2 gasification. Energy Fuels 2013;27:494500.
[4] Ding L, Zhang YQ, Wang ZQ, Huang JJ, Fang YT. Interaction and its induced
inhibiting or synergistic effects during co-gasification of coal char and biomass
char. Bioresour Technol 2014;173:1120.
[5] Howaniec N, Smolinski A. Steam co-gasification of coal and biomass-Synergy
in reactivity of fuel blends chars. Int J Hydrogen Energy 2013;38:1615260.
[6] Fermoso J, Arias B, Gil MV, Plaza MG, Pevida C, Pis JJ, et al. Co-gasification of
different rank coals with biomass and petroleum coke in a high-pressure
reactor for H2-rich gas production. Bioresour Technol 2010;2010(101):32305.
[7] Rizkiana J, Guan G, Widayatno BW, Hao X, Huang W, Tsutsumi A, et al. Effect of
biomass type on the performance of co-gasification of low rank coal with
biomass at relatively low temperatures. Fuel 2014;134:4149.
[8] Jeong HJ, Park SS, Hwang J. Co-gasification of coalbiomass blended char with
CO2 at temperatures of 9001100 C. Fuel 2014;116:46570.
[9] Zheng XY, Chen C, Ying Z, Wang B. Experimental study on gasification
performance of bamboo and PE from municipal solid waste in a bench-scale
fixed bed reactor. Energy Convers Manage 2016;117:3939.
[10] Saw WL, Pang SS. Co-gasification of blended lignite and wood pellets in a 100
kW dual fluidised bed steam gasifier: the influence of lignite ratio on producer
gas composition and tar content. Fuel 2013;112:11724.
[11] Sjstrm K, Chen G, Yu Q, Brage C, Rosn C. Promoted reactivity of char in cogasification of biomass and coal: synergies in the thermochemical process.
Fuel 1999;78:118994.
[12] Collot AG, Zhuo Y, Dugwell DR, Kandiyoti R. Co-pyrolysis and co-gasification of
coal and biomass in bench-scale fixed bed and fluidized bed reactors. Fuel
1999;78:66779.
[13] Wu ZQ, Wang SZ, Zhao J, Chen L, Meng H. Synergistic effect on thermal
behavior during co-pyrolysis of lignocellulosic biomass model components
blend with bituminous coal. Bioresour Technol 2014;169:2208.
[14] Zhang Y, Fan D, Zheng Y. Comparative study on combined copyrolysis/gasification of bituminous coal and walnut shell by conventional
and congruent-mass TGA methods. Bioresour Technol 2016;199:3825.
[15] Xu M, Brown RC, Norton G, Smeenk J. Comparison of a solvent-free tar
quantification method to the international energy agencys tar measurement
protocol. Energy Fuels 2005;19:250913.
[16] CEN/TS 15439:2006. Biomass gasification tar and particles in product gasessampling and analysis. Brussels: European Committee for Standardization;
2006.
[17] van Paasen SVB, Kiel JHA, Neeft JPA, Knoef HAM, Buffinga GJ, Zielke U, et al.
Guideline for sampling and analysis of tar and particles in biomass producer
gases, 2002. ECN-C-02-090. Final report documenting the Guideline, R&D work
and dissemination.
[18] Carpenter DL, Deutch SP, French RJ. Quantitative measurement of biomass
gasifier tars using a molecular-beam mass spectrometer: comparison with
traditional impinger sampling. Energy Fuels 2007;21:303643.
[19] Zhang Y, Kajitani S, Ashizawa M, Miura K. Peculiarities of rapid pyrolysis of
biomass covering medium- and high-temperature ranges. Energy Fuels
2006;20:270512.
[20] Zhang Y, Zheng Y, Yang MJ, Song YC. Effect of fuel origin on synergy during cogasification of biomass and coal in CO2. Bioresour Technol 2016;200:78994.
[21] Kutchko BG, Kim AG. Fly ash characterization by SEMEDS. Fuel
2006;85:253744.

4. Conclusions
The co-gasification of coal and biomass was conducted in a
fixed bed mode using two separate and a pre-mixed bed configurations with a congruent sample mass. The pre-mixed bed configuration produced a well-dispersed bio-ash among coal char grains,
which indicated stronger synergy than the separate ones. A new
method was also explored for tar collection and determination,
by which the performance of bio-char and coal char beds for tar
cracking was evaluated. By investigating the tar formation during
the initial pyrolysis step, the bio-char bed was found to demonstrate better tar cracking performance than the coal char bed.
Acknowledgements
The authors are grateful to the financial supports from the
National Natural Science Foundation of China (Grant No.
51376031).