Fuel Processing Technology 194 (2019) 106104

Contents lists available at ScienceDirect

Fuel Processing Technology

journal homepage: www.elsevier.com/locate/fuproc

Research article

Effect of porous silica on the removal of tar components generated from T

waste biomass during catalytic reforming
⁎
Jun Kobayashia, , Katsuya Kawamotob, Noriyuki Kobayashic
a
Department of Mechanical Engineering, Kogakuin University, Tokyo, Japan
b
Graduate School of Environmental and Life Science, Okayama University, Okayama, Japan
c
Department of Materials Science and Engineering, Nagoya University, Aichi, Japan

A R T I C LE I N FO A B S T R A C T

Keywords: Thermal gasification of waste biomass requires a method of tar removal from gasified gases, since the tar pro-
Waste biomass duced is at risk of obstruction of the piping of the gasification processes and failure of the subsequent power
Gasification generation processes. Using laboratory-scale fluidized bed gasifier and catalytic reformers, the characteristics of
Tar removal tar removal was investigated. A commercial Ni-based catalyst and several kinds of porous silica with different
Porous silica
surface areas were applied to the reforming reaction. Moreover, the feasibility of regeneration by air oxidation of
Reforming catalyst
both the catalyst and the porous silica was evaluated. In gasification and reforming experiments using woody
waste, application of porous silica led to lower levels of highly polymerized polycyclic aromatic hydrocarbons
(PAHs). The greater the specific surface area of the porous silica, the lower the concentration of PAHs. Carbon
deposited on the surface of the porous silica and the mass of the carbon deposit reached a maximum at a surface
area of 195 m2/g. When both the catalyst and the porous silica regeneration by air oxidation was performed,
hydrogen concentration was found to level off from the third reforming operation and the amount of total tar did
not increase after the second reforming operation. Our results showed that the catalyst and the porous silica is
likely to have good reforming activity even repeatedly regeneration condition.

1. Introduction conversion processes. In order to establish a highly efficient gasification

process that produces convenient and clean fuel gases, we focused on
Flammable waste which includes biomass is regarded as a renew- steam reforming using catalytic reaction at relatively low temperature.
able energy resource in the same way as woody and agricultural bio- It is reported that the method results in low sensible heat loss and is
mass. Gasification technology is one of the methods for effective utili- useful not only for generating H2 rich gas but also for tar decomposi-
zation of the waste biomass energy, since it will be possible to improve tion. ([19]; Swami et al., 2006; [20]) In recent studies on the applica-
the handling as fuel and the flammability in comparison with conven- tion of reforming catalysts for promote of tar removal and H2 produc-
tional waste incinerators. The gasification of waste, moreover, has po- tion, various catalysts have been suggested. Noble metals such as Pt,
tential for highly efficient energy production, since the fuel gases can be Rh, and Pd [21,22], base metals such as Ni ([23,24]: [25]) and Co
applied to high-efficiency power generation such as gas turbine com- [26,27] and mineral ores such as limonite, dolomite and olivine
bined cycle or fuel cell. [1–3] In order to promote the development of [20,28,29] were evaluated. It has been shown that noble metal catalysts
the energy conversion, there is a need for a suitable gasification method have high activity in the decomposition of tar, but their high cost rules
to make effective use of the gasified gas. them out as a first choice. Catalytic supports such as alumina, silica,
Many types of gasifiers, such as fluidized-bed [4–6], down draft [7], ceria, zirconia and alkaline earth metal compounds have also been in-
and rotary kiln [8] types have also been proposed, and extensive re- vestigated to promote the activity of base metal catalyst. [30,31] Our
search and experiments continue to be conducted. Moreover, many research work has also shown CaO and alumina composite, used as a
different methods of biomass gasification using steam [9,10], air catalytic support, is able to prevent Ni active sites from coking. [32,33].
[11,12], oxygen [13] or CO2 [14] as gasifying agents, from low tem- To develop steam reforming using base metal catalysts such as Ni
peratures below 973 K [15,16] to high temperatures over 1273 K under relatively low temperature condition, however, further tar re-
[17,18] have been investigated to develop the effective energy moval will be a significant issue. In case of high temperature

⁎
Corresponding author at: 2665-1 Nakano, Hachioji, Tokyo 192-0015, Japan.
E-mail address: jkoba@cc.kogakuin.ac.jp (J. Kobayashi).

https://doi.org/10.1016/j.fuproc.2019.05.027
Received 20 November 2018; Received in revised form 17 May 2019; Accepted 18 May 2019
0378-3820/ © 2019 Elsevier B.V. All rights reserved.
J. Kobayashi, et al. Fuel Processing Technology 194 (2019) 106104

gasification, in which most of tar will be decomposed, there is not only catalyst bed. All reactors are preheated with electric heaters. Three
above-mentioned heat loss but also possibility of erosion on furnace thermocouples have been placed in the gasifier at distances of 50, 350,
walls by molten of alkali metal salts included in the biomass. Therefore, and 650 mm above the distributor, to measure the temperature of the
achieving practical gasification requires sufficient ability of tar removal fluidized bed, internally circulating particles and area near the outlet,
under relatively low temperature condition. As one of the methods respectively. Two thermocouples more have been installed in each re-
independent on metal catalysts, tar capture and removal using porous former: one placed in the catalyst bed and the other placed about
alumina (gamma alumina) has been studied, and several important 20 mm below the bottom of the catalyst bed, to measure the tempera-
results have been achieved. [34–37] Tar removal performance using ture of the catalyst bed and the reformed gas, respectively. In this
porous alumina would prove to depend on Lewis acid sites rather than system, feedstock is continuously supplied from the upper part of the
its physical pore structure. From these studies, it is necessary to eval- gasifier.
uate the function of porous material not only as the catalytic supports To perform the reforming process and catalyst regeneration si-
but also as the materials contributed to tar removal. Small pores, which multaneously, two reformers are connected to the gasifier through
are almost between 0.7 and 50 nm of pore diameter, will be suitable to several valves that are used for switching between reforming and re-
capture large hydrocarbon molecules contained in tar. [38,39] More- generation. All pipelines are kept at around 523 K by means of in-
over, a hydrophobic surface may have a strong affinity for organic sulating jackets. The oxidizing agents are steam and simulated air
compounds. [40] In above-mentioned previous works, several types of (mixed nitrogen and oxygen), preheated to around 550–600 K. This
activated carbon and char were mainly applied to tar removal, so we mixture is supplied from the bottom of the gasifier. When the gaseous
suppose that porous silica, which is an inorganic hydrophobic ad- mixture passes through the distributor, the feedstock and silica sand are
sorbent and mainly has mesopores, is highly effective in attracting tar fluidized and mixed in the gasifier. This mixture is constantly heated by
and has no effect from oxidizing agents. Moreover, it is thought that the the electric heater, since the thermal decomposition and volatilization
surface of porous silica will be hardly affected by poisoning chemical of the feedstock are endothermic. The produced gases and tar in the
compounds since the surface has few acidic sites. gasifier pass through the cyclone, where unreacted char and fly ash are
The main purpose of this study is to investigate the influence of removed, and then move on to the reformer. The tar and low hydro-
porous silica on the removal of tar by combining reforming catalyst. carbons are reformed by steam in the reformer, to be converted into
This study will demonstrate that the characteristics of the pore struc- gases such as H2, CO and CO2.
ture, such as the pore radius and specific surface area, of the porous
silica will in some way influence the decomposition and/or removal of 2.2. Feedstock and packing material in the reformer
tar components. The feasibility of regeneration of both the catalyst and
the porous silica are also evaluated. The ability of the hydrophobic and thermally stable porous silica to
promote tar removal and decomposition under steam reforming con-
2. Experimental ditions was investigated. In this study, five types of porous silica par-
ticles (Fuji Silysia Chemical Ltd.) with different porosities were used as
2.1. Apparatus adsorbents for tar to assist its decomposition and/or prevent coking on
the catalyst. Their specific surface areas, total pore volumes and mean
A laboratory-scale gasification unit shown in the schematic diagram pore diameters are shown in Table 1. The larger the surface area, the
of Fig. 1 was applied to the experiments. The system essentially consists smaller the mean pore radius. Here, the particle size of all porous silica
of two kinds of reactors in series: a primary reactor (gasifier) followed was around 1–3 mm in diameter. A comparison of these materials will
by two secondary reactors (reformers) made of stainless steel. The ga- reveal how the fine pore structure of porous silica works.
sifier is designed as a fluidized bed-type (ID 37–75 mm, H 1360 mm) As reported in our previous work [32,33], Ni-based commercial
and silica sand (mean particle diameter: 0.6 mm) is used as the fluidized reforming catalyst (ReforMax 330 LDP; Süd-Chemie Catalysts Japan,
bed material The reformers are designed as packed bed types (ID Inc.) was also used in the experiments. This catalyst contains Ni at 11 wt
48 mm, H 1620 mm) and a packed bed of porous silica is placed on the % and CaO at 13 wt%, with the balance being Al2O3. Its shape is a 10-

Fig. 1. Schematic diagram of experimental apparatus.

2
J. Kobayashi, et al. Fuel Processing Technology 194 (2019) 106104

Table 1 Table 4
Structural characteristics of porous silica. Experimental conditionsa (evaluation of catalyst regeneration by air oxidation).
Sample no. S1 S2 S3 S4 S5 Run no. 6 7 8 9 10 11 12 13

2 −1
Specific surface area [m ·g ] 71 107 195 280 430 Reforming reactor 1st 2nd 1st 2nd 1st 2nd 1st 2nd
Total pore volume [ml·g−1] 0.98 0.98 0.99 0.96 0.61 Feed rate [kg·h−1] 0.13 0.13 0.16 0.15 0.14 0.11 0.14 0.15
Mean pore diameter [nm] 48 31 16 13 5.9 Gasifier temp. [K] 1030 1030 1040 1040 1040 1030 1040 1030
Reformer temp. [K] 990 980 1040 1020 1040 1010 1020 1020
S/C [mol/mol] 2.0 2.0 1.6 1.7 1.9 2.4 1.9 1.7
ER [−] 0.14 0.14 0.11 0.12 0.13 0.17 0.13 0.12
Table 2
Maximum reformer 1110 1050 1070 1050 1100 1050 – –
Proximate and ultimate analyses of woody biomass.
temp. during air
Volatiles (wt% dry) 80.0 H (wt% dry) 5.9 oxidation [K]
Fixed carbon (wt% dry) 18.0 C (wt% dry) 51.4
a
Ash (wt% dry) 2.0 O (wt% dry) 40.7 The porous silica used in these experiments is S3.
Moisture (wt%) 9.0 N (wt% dry) < 0.1
HHV (MJ/kg dry) 20.5 Cl (wt% dry) < 0.01 reformer shown in Tables 3 and 4 is average of actual measured values
LHV (MJ/kg dry) 17.2 S (wt% dry) < 0.02
during each experiment. Gasification gas obtained by the gasifier has
been introduced into either of the reformers for about 2 h. The pro-
hole ring, with a diameter of 19 mm and height of 16 mm. The catalyst duced gas was passed through scrubbing bottles placed in an ice bath.
was pulverized and classified into 2–5-mm-diameter pieces for use in Tar and steam were condensed and deposited in the bottles. The con-
the experiments. densed tar components were extracted using n-hexane and analyses of
Woody waste from construction and demolition work in Japan was the extract as shown below were performed. After the tar was removed,
used as feedstock. The wood species perhaps is cypress or cedar but the the gas was passed through a series of gas analyzers. H2, O2, N2, CO,
details are unknown. That was pulverized by rotary cutter mill and CO2, CH4, C2H2, C2H4 and C2H6 were detected every 5 min using a
classified into 1–2-mm particles for experimental preparation. Micro-GC 3000 (Agilent Technologies). Qualitative and quantitative
Proximate and ultimate analyses of the feedstock are shown in Table 2. analyses of the removal tar containing PAHs were executed using
The composition of ash was low and that of volatile matter was very GC–MS (QP2010 Plus: Shimadzu, Ltd.), HPLC (LC-8020: Toso, Ltd.) and
high. Moreover, the feedstock contained > 40 wt% oxygen due to the FID-GC (GC-2014: Shimadzu, Ltd.). The deposited carbon content on
presence of cellulose and xylose. the catalyst and porous silica after the experiments were also measured
using a carbon determination unit (NC-22: Sumika Chemical Analysis
Service, Ltd.), and the nitrogen adsorption method at 77 K (Quadrasorb
2.3. Procedure
SI: Quantachrome Instruments) and X-ray diffraction (MultiFrex: Ri-
gaku Co.) was applied to pore structure and crystal structure analyses of
The experimental conditions for evaluation of porous silica and
these materials, respectively. In repeated experiments (Runs 6–13), the
regeneration of reformer are shown in Tables 3 and 4, respectively.
catalyst and porous silica in the other reformer was regenerated using
Here, the amount of steam injection was expressed as the parameter for
air at 1 l/min to remove tar and/or carbon deposits on these materials.
the S/C ratio (Molar ratio of steam/carbon in the feedstock [mol/mol]),
After regeneration was performed for about 2 h, the air in the reformer
while oxygen was expressed as ER (Equivalence ratio: ratio of actual
was purged with nitrogen.
oxygen supplied to total oxygen required to accomplish the complete
burning [−]).Since the feed rate fluctuated even when the rotary speed
of screw was constant, the mean feed rate (designated as 0.12 kg/h) was 3. Results and discussion
calculated after the experiments according to the amount of feedstock
actually supplied. The experiment of Run 0 was performed with catalyst 3.1. Effect of porous silica on catalytic reforming
only to compare with the results using both porous silica and catalysts
as shown in Run 1–5. When evaluating the regeneration of both catalyst Fig. 2 shows the average concentrations of the main components
and porous silica in Run 6–13, the amount of these materials was al- and light hydrocarbons below C2 in produced gases. Here, the con-
most same and 100 ± 1 g. centrations of the main components such as CO, CO2 and relatively low
Before starting the experiment, the gasifier and reformers were molecular weight hydrocarbon were fairly constant over time. In con-
preheated in a nitrogen atmosphere. After the temperature of these trast, the compositions of H2 and CH4 slightly decreased and increased
reactors reached the set value and was maintained for several hours, with lapse of time, respectively. Therefore, average composition for
feedstock was fed from the hopper through a screw feeder to the ga- each component during the experimental periods is shown. The main
sifier. Water for steam gasification was supplied with simulated air to a components showed almost the same concentration and were in-
preheater located upstream of the gasifier. These were heated and dependent of the specific surface area of the porous silica and its use or
evaporated through the preheater. Temperature of the gasifier and the non-use. Equilibrium compositions under similar conditions (1123 K,

Table 3
Experimental conditions (effect of pore structure).
Run no. 0 1 2 3 4 5

Feed rate [kg·h−1] 0.12 0.11 0.10 0.12 0.08 0.12

Gasifier temp. [K] 1030 1030 1030 1040 1020 1040
Reformer temp. [K] 1120 1130 1130 1120 1130 1120
S/C [mol/mol] 2.4 2.4 2.7 2.2 3.3 2.4
ER [−] 0.16 0.16 0.18 0.15 0.22 0.16
Reformer Porous silica Sample no. – S1 S2 S3 S4 S5
Weight [g] 0a 100 100 99 100 100
Catalyst Weight [g] 50 50 50 50 50 50

a
Non-porous alumina beads were used to a packed bed material instead of the porous silica.

3
J. Kobayashi, et al. Fuel Processing Technology 194 (2019) 106104

Fig. 2. Average composition of produced gas using catalyst and 5 kinds of the porous silica.

ER = 0.15, S/C = 2.6) based on our calculations are 46 vol% of H2,

21 vol% of CO2, 13 vol% of CO and < 0.1 vol% of CH4, and are roughly
in accordance with the experimental results in Fig. 2. This means that
the gasification and reforming reaction of woody waste closely ap-
proaches the reaction equilibrium, whether the porous silica is used or
not. On the other hand, hydrocarbon concentrations in the produced
gases, except for methane which can be synthesize through CO me-
thanation [41], were a little lower when using the porous silica with
catalyst, but varied according to the surface area. Under this set of
experimental conditions, the hydrocarbon concentrations tended to
decrease with increase in the surface area of the porous silica. The
reasons for this will be discussed in further detail in relation to carbon
deposits on the porous silica.
As an example, Fig. 3 shows each GC–MS chart of tar with or
without porous silica. It was found that a certain amount of dibenzo-
furan, fluorene and other PAHs were included in the tar when porous
silica was not used to the reformer. In contrast, it was suggested that the
amount of these PAHs remarkably reduced using porous silica com-
pared with intensities of GC–MS chart between Fig. 3(a) and (b).
However, certain amounts of naphthalene and anthracene/phenan-
threne have remained in the extract. Therefore, naphthalene, anthra-
cene, phenanthrene and pyrene are regarded as main indicators of the
tar removal. The quantitative evaluation results of these PAHs are
shown as follows.
Fig. 4 shows the concentrations of PAHs and total tar in the pro-
duced gases. These concentrations are the cumulative average values
and are indicated in the form of mass of material per unit volume of
nitrogen-free produced gas. Here, the concentrations in Runs 2 and 4
are relatively low because their ER and S/C values are higher than the
others. Concentrations of relatively heavy PAHs with > 3 benzene rings
were lower when using the porous silica, although the concentrations of
naphthalene using the porous silica was often higher than that without Fig. 3. GC–MS chart of collected tar with Run No. 0 and 3.
the porous silica. On the contrary, the concentrations of heavy PAHs
markedly decreased with increasing run number, i.e., surface area. On
materials in the reformer were observed after the gasification experi-
the other hand, when the porous silica was used, the concentrations of
ments. The carbon contents of the porous silica and the catalyst are
total tar decreased as compared with the case where the porous silica
shown in Fig. 5(a). This figure also indicates the concentration of total
was not used. The values reached a minimum when surface area was
tar as shown in Fig. 4. Here, units of carbon deposits are converted to
approximately 100–300 m2/g.
the same unit of tar concentration, which is mass of carbon par total
Previous work by co-author [42] reported that tar conversion using
amount of gasified gas without nitrogen, to enable comparison of
only the same catalyst as this study was about 84% although the ex-
carbon deposits with the total quantity of tar. Initially, there was almost
perimental conditions were slightly different. Assuming the relationship
no influence of the porous silica on the carbon deposit on the catalyst,
is maintained, tar conversion using both the catalyst and suitable
and the specific surface area little affected that on the catalyst. On the
porous silica such as S2-S4 would be calculated to be 95% or more.
other hand, the amount of carbon deposit on the porous silica changed
Carbon deposits on the silica surface were confirmed when the bed

4
J. Kobayashi, et al. Fuel Processing Technology 194 (2019) 106104

Fig. 4. Mean concentrations of PAHs and total tar in produced gases using catalyst and 5 kinds of the porous silica.

Fig. 5. Effect of carbon deposits on specific surface areas of the porous silica and change of porosity after experiments.

according to its surface area and reached a maximum at around 200 the volume of deposited carbons is calculated using graphite carbon
m2/g. This trend is almost the opposite to the behavior of total tar density (2.26 g/ml), and the amount is about 1/100 of the pore volume
concentration. It is supposed that the tar decomposes on the porous or less. Therefore, it is considered that adsorbed tar and/or deposited
silica and a part of that deposits as carbon-based material. Moreover it carbons did not fill the pore but blocked many pore entrance on the
is thought that above mentioned light hydrocarbon concentration de- outer surface. Consequently, the amount of carbon deposit on the
creases with decrease in tar concentration in the gas phase. Both the porous silica decreased. That explains the reason why mean pore radius
amount of carbon deposit and tar concentration using the silica with the hardly changed but total pore volume decreased after the experiments.
largest surface were lower than these with the smallest surface area. Considering the size of benzene circle (about 0.5 nm), PAHs are rela-
Fig. 5(b), on the other hand, shows that the surface area of the porous tively large molecules and it seems likely to close the entrance of re-
silica after the experiments reduced when the original surface area latively small pores, which have large surface area.
exceeded 200 m2/g. This tendency becomes more pronounced with Based on these results and conclusions, it was decided that the
increase in the original surface area and the behaviors of total pore porous silica with 195 m2/g (Sample No. S3) should be used in the
volume is also similar as shown in Fig. 5(d). On the contrary, mean pore following experimental investigation on the repeated use of catalytic
radius of the porous silica scarcely changed as shown in Fig. 5(c). Here, materials.

5
J. Kobayashi, et al. Fuel Processing Technology 194 (2019) 106104

Fig. 6. Variation in produced gas composition with time and the effect of regeneration on the composition in produced gases (1st reformer).

3.2. Regeneration of catalyst and catalytic activity of that at Runs 0–5, i.e., about 4000 h−1 (based on wet gas and ambient
temperature). The gasified gas therefore had sufficient residence time to
The effect of regeneration on repeated reforming reactions was in- be reformed.
vestigated using the abovementioned porous silica. Figs. 6 and 7 show The performance of the steam reforming reaction slightly declined:
the variation in the produced gas compositions with time from Runs 6 the concentrations of hydrocarbons increased with increasing number
to 13. The chief ingredients and low hydrocarbons in the produced of regeneration cycles, as seen in Figs. 6(b) and 7(b). However, time
gases are shown in these figures. course of these concentrations became plateau and continuous decrease
Produced gas compositions at Run 6 were comparable to the results of the performance had been prevented by the regeneration. Im-
for Run 3, but the concentration of H2 was higher and that of CH4 was mediately after starting regeneration, temperature of the catalyst rose
lower. After the first regeneration cycle (Run 8), the concentration of up to 1100 K and then returned to preset temperature around 1000 K
H2 decreased and that of CH4 increased with time. After the second and within 30 min. Since high temperature period in the regeneration could
third regenerations, the concentration of H2 continued to decrease cause sintering of dispersed nickel fine particles on the catalyst surface,
slightly, but the concentrations of both gases showed little further it is thought that the catalyst regeneration using air oxidation results in
change. Thus, it was found that the performance of the catalyst and the a little lower performance of the catalyst. In fact, compared between X-
porous silica leveled off at the second regeneration or later, although ray diffraction patterns before and after experiment, there were no
the performance was lower than with fresh materials. peaks based on nickel crystal in fresh catalyst although the catalyst after
The equilibrium concentration of H2 under 1030 K (about 50 vol%) regeneration experiments clearly had the peaks as shown in Fig. 8.
is higher than that under 1120 K, which is one of the reasons that H2 However, half width of the peaks, which is proportional to the size of
concentration at Run 6 was higher than that at Run 3. On the other nickel crystal, was almost same comparing used catalyst at Run No. 3
hand, the space velocity in the catalyst bed during Runs 6–12 was half with that at Run No. 12. The result indicates that the nickel crystal size

6
J. Kobayashi, et al. Fuel Processing Technology 194 (2019) 106104

Fig. 7. Variation in produced gas composition with time and the effect of regeneration on the composition in produced gases (2nd reformer).

in the used catalyst is largely maintained even if catalyst regeneration is

repeated. Sehested has reported that an increase in the nickel particle
size decreases the activity of a steam-reforming catalyst. [43] There-
fore, it is thought that the activity at the regeneration condition in this
work dose not decline much although that is slightly lower than the
fresh ones.
Figs. 9 and 10 show the effects of regeneration on the emission
behavior of PAHs and total tar in the produced gases. These con-
centrations were a little lower than those of Run 3. It was thought that
using twice the amount of the catalyst in Run 3 caused these results,
although the reforming temperature of Runs 6–13 was about 100 K
below that of Run 3. The concentrations of all PAHs and total tar
roughly increased after both the catalyst and the porous silica was re-
generated. However, increase in these concentrations has not been
continued and removal performance for the tar has been maintained
Fig. 8. X-ray diffraction patterns of reforming catalyst (Fresh: unused catalyst, after the second regeneration. Comparing between the results of Run
Run No. 3 and Run No. 12: used catalyst without air oxidation after the op- No. 10 and 12, the concentrations of all PAHs and total tar decreased by
eration). the regeneration although both gasification conditions such as S/C and
ER was almost similar. The results of Run No. 9 and 13 also indicated
similar tendency. Here, the results of Run No. 11 were excluded from

7
J. Kobayashi, et al. Fuel Processing Technology 194 (2019) 106104

Fig. 9. Mean concentrations of PAHs and total tar in produced gases while applying regeneration to the catalyst and the porous silica (1st reformer).

Fig. 10. Mean concentrations of PAHs and total tar in produced gases while applying regeneration to the catalyst and the porous silica (2nd reformer).

the comparison because both S/C and ER in the experimental condition using the porous silica, with its value reaching a minimum at a surface
were higher than the others. From these tendencies, it seems that per- area of around 100–300 m2/g.
formance of tar removal using porous silica would be maintained or 3) The porous silica had almost no influence on the carbon deposits
promoted by the regeneration treatment. on the catalyst. However, the amount of carbon deposited on the porous
From these results for the regeneration experiments, continuous silica reached a maximum when its surface area was around 200 m2/g.
reduction of the performances of both reforming and tar removal was This tendency was almost opposite to the concentration of total tar,
not confirmed although these performances slightly decline by re- suggesting that the tar in the gasified gas decomposes and is deposited
peating utilization and regeneration. as carbon.
4) The results of our regeneration experiments showed that the
4. Conclusions performance of the catalyst and the porous silica almost leveled off after
the second regeneration, although the performance was inferior to the
Gasification and reforming of woody waste using nickel catalyst and fresh ones.
porous silica were carried out under various conditions to clarify the
influence of porous silica on tar removal. The applicability of the re- Acknowledgements
generation operation was also investigated. The following conclusions
were obtained. A part of this work was carried out by the joint research program of
1) Main components such as H2, CO, CO2 and CH4 were in- the EcoTopia Science Institute, Nagoya University.
dependent of the specific surface area of the porous silica as well as its
use or non-use. On the other hand, hydrocarbon concentrations in the References
produced gases were slightly reduced when using the porous silica with
the catalyst and tended to decrease with increase in the surface area of [1] H. Ge, H. Zhang, W. Guo, T. Song, L. Shen, System simulation and experimental
verification: biomass-based integrated gasification combined cycle (BIGCC) cou-
the porous silica. pling with chemical looping gasification (CLG) for power generation, Fuel 241
2) The concentrations of relatively heavy PAHs were reduced when (2019) 118–128.
using the porous silica, although there was no clear difference in [2] E. Shayan, V. Zare, I. Mirzaee, On the use of different gasification agents in a
biomass fueled SOFC by integrated gasifier: a comparative exergo-economic eva-
naphthalene concentration between use and non-use of porous silica,
luation and optimization, Energy 171 (2019) 1126–1138.
and its concentrations fluctuated. Moreover, the concentrations of [3] W. Wu, K. Kawamoto, H. Kuramochi, Hydrogen-rich synthesis gas production from
heavy PAHs markedly decreased with increased surface area of the waste wood via gasification and reforming technology for fuel cell application, J.
porous silica. On the other hand, the concentration of total tar fell when Mater. Cycles Waste Manag. 8 (2006) 70–77.

8
J. Kobayashi, et al. Fuel Processing Technology 194 (2019) 106104

[4] J. Corella, A. Sanz, Modeling circulating fluidized bed biomass gasifiers. A pseudo- [24] T. Kimura, T. Miyazawa, J. Nishikawa, S. Kado, K. Okumura, T. Miyao, S. Naito,
rigorous model for stationary state, Fuel Process. Technol. 86 (2005) 1021–1053. K. Kunimori, K. Tomishige, Development of Ni catalysts for tar removal by steam
[5] M. Di Marcello, G.A. Tsalidis, G. Spinelli, W. de Jong, J.H.A. Kiel, Pilot scale steam- gasification of biomass, Appl. Catal. B Environ. 68 (2006) 160–170.
oxygen CFB gasification of commercial torrefied wood pellets. The effect of torre- [25] R. Zhang, Y. Wang, R.C. Brown, Steam reforming of tar compounds over Ni/olivine
faction on the gasification performance, Biomass Bioenergy 105 (2017) 411–420. catalysts doped with CeO2, Energy Convers. Manag. 48 (2007) 68–77.
[6] X.T. Li, J.R. Grace, C.J. Lim, A.P. Watkinson, H.P. Chen, J.R. Kim, Biomass gasifi- [26] M.A. Adnan, A. Hidayat, O.O. Ajumobi, S. Adamu, O. Muraza, M.M. Hossain,
cation in a circulating fluidized bed, Biomass Bioenergy 26 (2004) 171–193. Fluidizable Fe-Co/Ce-ZrO2 Catalysts for Steam Reforming of Toluene as a Tar
[7] P. Raman, N.K. Ram, Design improvements and performance testing of a biomass Surrogate in Biomass Gasification, Energy Fuel 32 (2018) 12833–12842.
gasifier based electric power generation system, Biomass Bioenergy 56 (2013) [27] T. Furusawa, A. Tsutsumi, Development of cobalt catalysts for the steam reforming
555–571. of naphthalene as a model compound of tar derived from biomass gasification,
[8] F. Fantozzi, B. D'Alessandro, U. Desideri, Integrated pyrolysis regenerated plant Appl. Catal. A 278 (2005) 195–205.
(IPRP): an efficient and scalable concept for gas turbine based energy conversion [28] S. Koppatz, C. Pfeifer, H. Hofbauer, Comparison of the performance behaviour of
from biomass and waste, J. Eng. Gas Turbines Power Trans. ASME 127 (2005) silica sand and olivine in a dual fluidised bed reactor system for steam gasification
348–357. of biomass at pilot plant scale, Chem. Eng. J. 175 (2011) 468–483.
[9] L. Wei, S. Xu, L. Zhang, C. Liu, H. Zhu, S. Liu, Steam gasification of biomass for [29] A. Soomro, S. Chen, S. Ma, W. Xiang, Catalytic activities of nickel, dolomite, and
hydrogen-rich gas in a free-fall reactor, Int. J. Hydrog. Energy 32 (2007) 24–31. olivine for tar removal and H2-enriched gas production in biomass gasification
[10] Y. Xiao, S. Xu, Y. Song, Y. Shan, C. Wang, G. Wang, Biomass steam gasification for process, Energy Environ. 29 (2018) 839–867.
hydrogen-rich gas production in a decoupled dual loop gasification system, Fuel [30] Asadullah, M., Ito, S.I., Kunimori, K., Yamada, M. and Tomishige, K., 2002b
Process. Technol. 165 (2017) 54–61. Biomass gasification to hydrogen and syngas at low temperature: novel catalytic
[11] B. Bronson, P. Gogolek, P. Mehrani, F. Preto, Experimental investigation of the system using fluidized-bed reactor, J. Catal., 208, 255–259.
effect of physical pre-treatment on air-blown fluidized bed biomass gasification, [31] D. Li, M. Tamura, Y. Nakagawa, K. Tomishige, Metal catalysts for steam reforming
Biomass Bioenergy 88 (2016) 77–88. of tar derived from the gasification of lignocellulosic biomass, Bioresour. Technol.
[12] F. Guo, Y. Dong, T. Zhang, L. Dong, C. Guo, Z. Rao, Experimental study on herb 178 (2015) 53–64.
residue gasification in an air-blown circulating fluidized bed gasifier, Ind. Eng. [32] J. Kobayashi, K. Kawamoto, Catalyst durability in steam reforming of thermally
Chem. Res. 53 (2014) 13264–13273. decomposed waste wood, J. Mater. Cycles Waste Manag. 12 (1) (2010) 10–16.
[13] P. Lv, Z. Yuan, L. Ma, C. Wu, Y. Chen, J. Zhu, Hydrogen-rich gas production from [33] J. Kobayashi, K. Kawamoto, R. Fukushima, S. Tanaka, Woody biomass and RPF
biomass air and oxygen/steam gasification in a downdraft gasifier, Renew. Energy gasification using reforming catalyst and calcium oxide, Chemosphere 83 (9)
32 (2007) 2173–2185. (2011) 1273–1278.
[14] B. Bhui, P. Vairakannu, Experimental and kinetic studies on in-situ CO2 gasification [34] K. Matsuoka, T. Shinbori, K. Kuramoto, T. Nanba, A. Morita, H. Hatano, Y. Suzuki,
based chemical looping combustion of low ash coal using Fe2O3 as the oxygen Mechanism of woody biomass pyrolysis and gasification in a fluidized bed of porous
carrier, J. CO2 Utiliz. 29 (2019) 103–116. alumina particles, Energy Fuel 20 (2006) 1315–1320.
[15] M. Asadullah, S.I. Ito, K. Kunimori, M. Yamada, K. Tomishige, Energy efficient [35] K. Matsuoka, K. Kuramoto, T. Murakami, Y. Suzuki, Steam gasification of woody
production of hydrogen and syngas from biomass: Development of low-temperature biomass in a circulating dual bubbling fluidized bed system, Energy Fuel 22 (2008)
catalytic process for cellulose gasification, Environ. Sci. Tech. 36 (2002) 1980–1985.
4476–4481. [36] T. Murakami, T. Yang, M. Asai, Y. Suzuki, Development of fluidized bed gasifier
[16] X. Xiao, X. Meng, D.D. Le, T. Takarada, Two-stage steam gasification of waste with triple-beds and dual circulation, Adv. Powder Technol. 22 (2011) 433–438.
biomass in fluidized bed at low temperature: Parametric investigations and per- [37] T. Namioka, K. Yoshikawa, H. Hatano, Y. Suzuki, High tar reduction with porous
formance optimization, Bioresour. Technol. 102 (2011) 1975–1981. particles for low temperature biomass gasification: effects of porous particles on tar
[17] N. Kobayashi, M. Tanaka, G. Piao, J. Kobayashi, S. Hatano, Y. Itaya, S. Mori, High and gas yields during sawdust pyrolysis, J. Chem. Eng. Japan 36 (2003) 1440–1448.
temperature air-blown woody biomass gasification model for the estimation of an [38] G. Ravenni, O.H. Elhami, J. Ahrenfeldt, U.B. Henriksen, Y. Neubauer, Adsorption
entrained down-flow gasifier, Waste Manag. 29 (2009) 245–251. and decomposition of tar model compounds over the surface of gasification char
[18] W. Yang, A. Ponzio, C. Lucas, W. Blasiak, Performance analysis of a fixed-bed and active carbon within the temperature range 250–800 °C, Appl. Energy 241
biomass gasifier using high-temperature air, Fuel Process. Technol. 87 (2006) (2019) 139–151.
235–245. [39] Y. Shen, N. Zhang, Facile synthesis of porous carbons from silica-rich rice husk char
[19] B. Lu, Y. Ju, T. Abe, K. Kawamoto, Hydrogen-enriched producer gas production and for volatile organic compounds (VOCs) sorption, Bioresour. Technol. 282 (2019)
chemical conversion to usable gas product through biomass gasification using NiO 294–300.
nanoparticles dispersed on SBA-15, J. Nanosci. Nanotechnol. 17 (2017) 6190–6197. [40] Z. Chen, Z. Hu, J. Wang, X. Wang, X. Niu, Y. Wang, Y. Shen, W. Teng, J. Fan,
[20] X.Y. Zhao, J. Ren, J.P. Cao, F. Wei, C. Zhu, X. Fan, Y.P. Zhao, X.-Y. Wei, Catalytic W.X. Zhang, Synthesis of mesoporous silica-carbon microspheres via self-assembly
Reforming of Volatiles from Biomass Pyrolysis for Hydrogen-Rich Gas Production and in-situ carbonization for efficient adsorption of Di-n-butyl phthalate, Chem.
over Limonite Ore, Energy Fuel 31 (2017) 4054–4060. Eng. J. 369 (2019) 854–862.
[21] K. Polychronopoulou, K. Giannakopoulos, A.M. Efstathiou, Tailoring MgO-based [41] S. Takenaka, T. Shimizu, K. Ostuka, Complete removal of carbon monoxide in hy-
supported Rh catalysts for purification of gas streams from phenol, Appl. Catal. B drogen-rich gas stream through methanation over supported metal catalysts, Int. J.
Environ. (2012) 111–112 (360-375). Hydrog. Energy 29 (2004) 1065–1073.
[22] S.M. Swami, M.A. Abraham, Integrated catalytic process for conversion of biomass [42] K. Kawamoto, W. Wu, H. Kuramochi, Development of gasification and reforming
to hydrogen, Energy Fuel 20 (2006) 2616–2622. technology using catalyst at lower temperature for effective energy recovery: hy-
[23] M. Artetxe, J. Alvarez, M.A. Nahil, M. Olazar, P.T. Williams, Steam reforming of drogen recovery using waste wood, J. Environ. Eng. 4 (2009) 409–421.
different biomass tar model compounds over Ni/Al2O3 catalysts, Energy Convers. [43] J. Sehested, Four challenges for nickel steam-reforming catalysts, Catal. Today 111
Manag. 136 (2017) 119–126. (2006) 103–110.