energies

Article
Effect of Injection Flow Rate on Product Gas Quality
in Underground Coal Gasification (UCG) Based on
Laboratory Scale Experiment: Development of
Co-Axial UCG System
Akihiro Hamanaka 1, *, Fa-qiang Su 2,3 , Ken-ichi Itakura 3,4 , Kazuhiro Takahashi 3,4 ,
Jun-ichi Kodama 5 and Gota Deguchi 6
1 Department of Earth Resources Engineering, Kyushu University, 744 Motooka, Nishi-ku,
Fukuoka 819-0395, Japan
2 School of Energy Science and Engineering, Henan Polytechnic University, 2001 Century Avenue,
Jiaozuo 454-003, Henan, China; sfqmuroran@gmail.com
3 Center of Environmental Science and Disaster Mitigation for Advanced Research, Muroran Institute of
Technology, 27-1 Mizumoto, Muroran 050-8585, Japan; itakura@mmm.muroran-it.ac.jp (K.I.);
ktakahashi@mmm.muroran-it.ac.jp (K.T.)
4 Department of Information and Electronic Engineering, Muroran Institute of Technology, 27-1 Mizumoto,
Muroran 050-8585, Japan
5 Division of Sustainable Resources Engineering, Hokkaido University, Kita 13 Nishi 8, Kita-ku,
Sapporo 060-8628, Japan; kodama@eng.hokudai.ac.jp
6 Underground Resources Innovation Network, Kita 47 Higashi 17, Higashi-ku, Sapporo 007-0847, Japan;
gota@mue.biglobe.ne.jp
* Correspondence: hamanaka@mine.kyushu-u.ac.jp; Tel.: +81-92-802-3331

Academic Editor: Mehrdad Massoudi

Received: 7 January 2017; Accepted: 13 February 2017; Published: 16 February 2017

Abstract: Underground coal gasification (UCG) is a technique to recover coal energy without mining
by converting coal into a valuable gas. Model UCG experiments on a laboratory scale were carried
out under a low flow rate (6~12 L/min) and a high flow rate (15~30 L/min) with a constant oxygen
concentration. During the experiments, the coal temperature was higher and the fracturing events
were more active under the high flow rate. Additionally, the gasification efficiency, which means
the conversion efficiency of the gasified coal to the product gas, was 71.22% in the low flow rate
and 82.42% in the high flow rate. These results suggest that the energy recovery rate with the
UCG process can be improved by the increase of the reaction temperature and the promotion of the
gasification area.

Keywords: underground coal gasification; gasification efficiency; co-axial UCG system; acoustic emission

1. Introduction
Underground coal gasification (UCG) is a technique to extract energy from coal in the form of heat
energy and combustible gases through chemical reactions in the underground gasifier. The product
gas has as a variety of uses: electricity supply with a gas turbine, hydrogen production, and the
other chemical feedstock [1–4]. UCG technology enables us to utilize coal resources that remain
unrecoverable underground due to either technological or economic reasons. The annual consumption
of coal in Japan was 177 million tons in 2014 while the annual domestic coal production is only
1.3 million tons [5]. This fact means that more than 99% of the coal used in Japan depends on
importation from overseas countries. Additionally, 25% of the domestic primary energy supply
consists of coal. Now, Japan has an underground and several surface coal mines in a limited region.

Energies 2017, 10, 238; doi:10.3390/en10020238 www.mdpi.com/journal/energies

Energies 2017, 10, 238 2 of 11
Energies 2017, 10, 238 2 of 11

supply consists of coal. Now, Japan has an underground and several surface coal mines in a limited
region.
Actually,Actually, the geological
the geological conditions conditions
of the coalofseam
the coal seamare
in Japan inquite
Japancomplicated
are quite complicated
as it has many as faults,
it has
many faults,
foldings, andfoldings, and steep
steep dipping. dipping.most
Accordingly, Accordingly,
coal mines mosthadcoal to bemines
closedhad to be of
because closed because of
the difficulty
the difficulty
mining, of mining,costs,
high operation high and
operation costs,costs.
high labor and high labor abundant
However, costs. However, unusedabundant
coal resourcesunused coal
remain
resources
underground. remain Such underground.
coal resourcesSuch coal resources
are estimated to be 30arebillion
estimated tons. toForbe 30 reason,
that billion UCG tons. hasForgreat
that
reason,
potential UCG has great
to recover vastpotential
amountstoofrecover
energyvast
fromamounts
these coal of resources.
energy from these coal resources.
Many benefits
benefits are are anticipated
anticipated from
from this this technology:
technology: utilizing unused utilizing unused
coal, lower coal, lower
capital/operating
capital/operating costs, noash,
costs, no surface disposal surface disposal
and the ash, and
possibility of thethe possibility of carbon
combination the combination
capture and of storage.
carbon
capture and storage. In the UCG process, the oxidants are injected from
In the UCG process, the oxidants are injected from an injection well in order to promote the gasification an injection well in order to
promote
reactions, the
and gasification
product gas reactions,
consists of and
CO, H product
2 , CH4 , CO gas2 ,consists
and other ofhydrocarbons
CO, H2, CHare 4, CO 2, and other
recovered from
hydrocarbons
a production well. are recovered from a zones
Typical reaction production
during well.
the Typical
UCG process reaction canzones duringroughly
be divided the UCG process
into three
can
zonesbe [6]:
divided
The roughly
oxidization intozone,
three the
zones [6]: The zone,
reduction oxidization
and the zone, the reduction
drying and pyrolysis zone,zoneand the drying
(Figure 1).
and pyrolysis
Oxidation zone of
is source (Figure
heat to1). Oxidation
promote is source ofprocess,
the gasification heat tomeaning promotethat thethe gasification
oxidation process,
reaction
meaning
makes thethat the oxidation
temperature reaction
of the makes
coal seam theReduction
rise. temperature is theof the main coal seam rise.
chemical Reduction
reaction in the is the
UCG
main chemical reaction in the UCG process. In this process, CO or
process. In this process, CO2 or H2 O(g) are reduced to CO and H2 as the main chemical reactions.
2 H 2 O (g) are reduced to CO and H2
as
Asthe main
these chemical
reactions arereactions.
endothermic As these reactions
reactions, the are endothermic
temperature reactions,
of the coal seam the temperature
is decreased of the
when
coal seam is decreased when promoting the reduction reactions. On
promoting the reduction reactions. On the other hand, these chemical reactions are promoted under the other hand, these chemical
reactions are promoted
high temperatures. under high
In summary, onetemperatures.
of keys to success In summary,
in an efficient one ofUCG keys process
to success in keep
is to an efficient
a high
UCG processinisthe
temperature to reaction
keep a highzone.temperature
In the drying in and
the pyrolysis
reaction zone. zone,In the drying
various kinds and pyrolysis
of gases zone,
are formed,
various kinds of gases are formed, not only CH
not only CH4 , CO, CO2 , H2 , but also other hydrocarbons. 4 , CO, CO 2 , H 2 , but also other hydrocarbons.

Figure
Figure 1.
1. Typical
Typical chemical
chemical reaction
reaction zone
zone during
during UCG
UCG process.
process.

The contents of the product gas can be roughly controlled with the temperature and the
The contents of the product gas can be roughly controlled with the temperature and the injection
injection materials in the surface coal gasification plant. It is, however, difficult to control the quality
materials in the surface coal gasification plant. It is, however, difficult to control the quality of the
of the product gas during the UCG process because the conditions of the underground reactor are
product gas during the UCG process because the conditions of the underground reactor are constantly
constantly changing due to changes in the temperature field and the expansion of the cavity [7,8].
changing due to changes in the temperature field and the expansion of the cavity [7,8]. The composition
The composition of the product gas changes depending on the injected oxidant used, the operating
of the product gas changes depending on the injected oxidant used, the operating pressure, the coal
pressure, the coal quality, and the mass and energy balance of the underground reactor [9]. The
quality, and the mass and energy balance of the underground reactor [9]. The calorific value of the
calorific value of the product gas recovered by UCG is usually3 low (3~4 MJ/Nm3) when air is injected
product gas recovered by UCG is usually low (3~4 MJ/Nm ) when air is injected as a gasification
as a gasification agent, meaning that the usage of the gas is limited because of its low calorific value.
agent, meaning that the usage of the gas is limited because of its low calorific value. On the other hand,
On the other hand, researchers have obtained results to improve the quality of the product gas by
researchers have obtained results to improve the quality of the product gas by using a mixture of air
using a mixture of air and oxygen due to an increase of the reaction temperature in the underground
and oxygen due to an increase of the reaction temperature in the underground reactor [10–12].
reactor [10–12].
Energies 2017, 10, 238 3 of 11

Energies 2017, 10, 238 3 of 11

We discuss the development of a co-axial UCG system, which is compact and flexible, that can be
adopted under complicated
We discuss geological
the development of aconditions
co-axial UCG[13–15].
system, A which
co-axial UCG system
is compact uses only
and flexible, that one
can well
be adopted under complicated geological conditions [13–15]. A co-axial UCG system
set with a double pipe: oxidants are injected from an inner pipe, and product gases generated in the uses only one
well set
coal seam with
are a double
collected pipe:an
from oxidants are injected
outer pipe (Figurefrom
2). an inner
This UCGpipe, and product
system gases generated
has superiority in of
in terms
the coal seam are collected from an outer pipe (Figure 2). This UCG system has superiority
applicability compared to the conventional one, but the recovered energy from the coal is relatively in terms
of applicability compared to the conventional one, but the recovered energy from the coal is
low because the gasification area in a co-axial system is limited around a well even though operating
relatively low because the gasification area in a co-axial system is limited around a well even though
this system saves costs [16,17].
operating this system saves costs [16,17].
FromFrom
these backgrounds,
these backgrounds, thetheobjective
objectiveof
of this studyisistotoclarify
this study clarifythethe effect
effect of the
of the injection
injection flow flow
conditions on the product gas quality including the gasification efficiency, which
conditions on the product gas quality including the gasification efficiency, which means the means the conversion
efficiency of theefficiency
conversion gasified ofcoal
the(chemical energy
gasified coal of product
(chemical energygas/chemical energy of gasified
of product gas/chemical energy ofcoal),
in order to develop
gasified coal), inaorder
co-axial UCG system
to develop with
a co-axial UCG high efficiency.
system with high efficiency.

Figure 2. Concept of co-axial UCG system.

Figure 2. Concept of co-axial UCG system.
2. Materials and Methods
2. Materials and Methods
A diagram of model UCG experiment conducted using a coal block is shown in Figure 3. The
Acoal blocks of
diagram used in this
model UCG study were rectangular
experiment conductedshapeusingthat theblock
a coal rangeisofshown
lengthinand width
Figure are coal
3. The
0.15~0.20 m, and the range of height is 0.20~0.25 m. Coal samples were supplied
blocks used in this study were rectangular shape that the range of length and width are 0.15~0.20 m, from Sanbi Mining
Co.,range
and the Ltd. In of
Hokkaido,
height is Japan. Typical m.
0.20~0.25 proximate and ultimate
Coal samples wereanalyses of thefrom
supplied coal are shown
Sanbi in Table
Mining 1. Ltd.,
Co.,
The samples were put into a drum can which had 0.28 diameter and 0.36 m height. The space
Hokkaido, Japan. Typical proximate and ultimate analyses of the coal are shown in Table 1.
between them was filled by heat-resistant cement in order to prevent heat release and gas leakage. A
The samples were put into a drum can which had 0.28 diameter and 0.36 m height. The space
co-axial well which is used for ignition, and production well was prepared with 35 mm diameter.
between them was filled by heat-resistant cement in order to prevent heat release and gas leakage.
Ignited charcoals were supplied to the bottom of a co-axial well in an ignition stage, then a mixture
A co-axial
of air well
and which
oxygeniswas usedsupplied
for ignition, and production
continuously well wasagent
as a gasification prepared withto35sustain
in order mm diameter.
the
Ignited charcoalsprocess.
gasification were supplied to the
In this study, thebottom of a co-axial
model UCG well were
experiments in ancarried
ignitionoutstage,
underthen a mixture of
the different
air and oxygen
flow rate towas supplied
estimate continuously
the effects of injectionas a gasification
flow, agent
the lower flow ratein(experiment
order to sustain
1: 6~12 the gasification
L/min) and
the higher
process. In this flow rate
study, the(experiment
model UCG 2: 15~30 L/min), on
experiments thecarried
were quality out
of product
under gas while the flow
the different oxygen rate to
concentration
estimate the effectswas kept as flow,
of injection stable the
(50%)
lowerbased
flowonrate
the(experiment
previous experiments [14]. During
1: 6~12 L/min) and thethehigher
gasification
flow rate process,
(experiment the injection
2: 15~30 L/min), flowon rate was arranged
the quality to keep
of product the optimal
gas while thermodynamic
the oxygen concentration
conditions for gasification reactions. Figure 4 shows the injection conditions applied for each
was kept as stable (50%) based on the previous experiments [14]. During the gasification process,
experiment. At the end of these experiments, CO2 or N2 gas which turned down coal temperature
the injection flow rate was arranged to keep the optimal thermodynamic conditions for gasification
was injected to extinguish the combustion. The total times to inject gasification agents in respective
reactions. Figure 4 shows the injection conditions applied for each experiment. At the end of these
experiments, CO2 or N2 gas which turned down coal temperature was injected to extinguish the
combustion. The total times to inject gasification agents in respective experiments were almost 7 h.
Energies 2017, 10, 238 4 of 11

Energies 2017, 10, 238 4 of 11

experiments were almost 7 h. After the process, a mixture of white cement and gypsum was filled
into Energies 2017, 10, 238
the post-gasification cavity to investigate the gasification area by means of a cross-section 4 of 11
study.
After the process, a mixture of white cement and gypsum was filled into the post-gasification cavity to
experiments were almost 7 h. After the process, a mixture of white cement and gypsum was filled
investigate
into the the gasification cavity
post-gasification area byto means of athe
investigate cross-section
gasification study.
area by means of a cross-section study.

Figure
Figure3.
Figure 3.3.AAAdiagram
diagram of
diagram ofthe
of UCG
theUCG
the model
UCGmodel
model experiment.
experiment.
experiment.

Table
Table 1.
1.1.Proximate
Table Proximateand
Proximate andultimate
and ultimate analyses
ultimateanalyses ofof
analyses the
the
of coal.
coal.
the coal.
Calorific Value
Calorific Value Proximate
ProximateAnalysis (wt.%)
Analysis (wt. %) Ultimate
Ultimate Analysis
Analysis (wt. (wt.
%) %)
Calorific Value
(MJ/kg) Proximate Analysis (wt. %) Ultimate Analysis (wt. %)
(MJ/kg) Moisture
Moisture Ash Volatiles
Ash Volatiles Fixed
FixedCarbon
Carbon C C H H N N S S O O
(MJ/kg) Moisture1.1 Ash 8.1Volatiles44.9 Fixed Carbon C 75.0 H 5.78 N S
31.4831.48 1.1 8.1 44.9 45.9
45.9 75.0 5.781.231.23 0.07 9.71O9.71
0.07
31.48 1.1 8.1 44.9 45.9 75.0 5.78 1.23 0.07 9.71

(a) (b)

(a)
Figure 4. Gasification (b)
agents during experiments: (a) Experiment 1; (b) Experiment 2.

TheFigure
flow 4.
Figure
Gasification
4.rate
Gasification
agents during
of the product gas experiments:
agents during was measured
experiments:
(a) Experiment
(a) using
1; (b) Experiment
an ultrasonic
Experiment
2.
flowmeter.
1; (b) Experiment 2. The
compositions of product gas (O2, N2, CO2, H2, CO, CH4, C2H4, C2H6, C3H6, and C3H8) were monitored
The 30
every flowminrate
usingof thechromatograph
a gas product gas(Micro was GC measured using an
3000A; INFICON Co.,ultrasonic flowmeter.
Ltd., East Syracuse, NY, The
The flow rate of the product gas was measured using an ultrasonic flowmeter. The compositions
compositions of product
USA). Meanwhile, gas (O2, N
temperature and2, CO 2, H2,emission
acoustic CO, CH(AE) 4, C2H 4, C2monitored
were H6, C3H6, to and C3H8) were
visualize monitored
inner part of
of product gas (O2 , and
N2 , obtain
CO2 , H 2 , data
CO, CH 4, C 2H 4, C 2H 6 , CINFICON
3 H6 ,by
and C3 H ) were
8Ltd., monitored every
every
the30coal
minsample
using a gas chromatograph
the (Micro
about GC 3000A;
fracturing activities Co.,type
using K East Syracuse, NY,
thermocouples
30 min using aChino
(SUS310S; gas chromatograph
Corp., Tokyo, (Micro
Japan) andGC 3000A; INFICON
piezoelectric Co.,transducers
Ltd., East Syracuse, NY, USA).
USA). Meanwhile, temperature and acoustic emission (AE)acceleration
were monitored (620 HT;
to visualize TEAC
inner part of
Meanwhile,
the coal temperature
Corp.,sample
Tokyo, Japan),
and andtheacoustic
respectively.
obtain data emission
Theabout (AE) were
layoutfracturing
of sensors are monitored
shown
activities Figure to
by using5. visualize
type K inner part of the
thermocouples
coal sampleChino
(SUS310S; and obtain
Corp.,the data about
Tokyo, Japan)fracturing activitiesacceleration
and piezoelectric by using type K thermocouples
transducers (SUS310S;
(620 HT; TEAC
Chino Corp., Tokyo, Japan) and piezoelectric acceleration transducers (620
Corp., Tokyo, Japan), respectively. The layout of sensors are shown Figure 5. HT; TEAC Corp., Tokyo,
Japan), respectively. The layout of sensors are shown Figure 5.
Energies 2017, 10, 238 5 of 11
Energies 2017, 10, 238 5 of 11

Energies 2017, 10, 238 5 of 11

(a) (b)
Figure 5. Layout of sensors: (a) Thermocouples; (b) Piezoelectric acceleration transducers.
Figure 5. Layout of sensors: (a) Thermocouples; (b) Piezoelectric acceleration transducers.
AE is a kind of phenomenon to emit low-level elastic wave from solids when they are stressed
or deformed. The number of AE occurrences are commonly increased just before solids is
AE is a kind of phenomenon to emit low-level elastic wave from solids when they are stressed or
destructed: the stress of rock reach to uniaxial compressive strength in uniaxial compressive test. AE
deformed.monitoring
The number of AE occurrences
is, therefore, (a)
used in the field areofcommonly
rock mechanics, increased (b)just
concrete, andbefore
mining solids
to predict is destructed:
the
the stress of rock reach
damage and the tofailure
Figure uniaxial
5. Layout compressive
of ofbrittle materials
sensors: strength
because
(a) Thermocouples; in reach
(b) they uniaxial
Piezoelectric compressive
structural
acceleration failure bytest.
transducers. AE monitoring
accumulating
microfracture
is, therefore, used in the [18–22].
fieldAs of many AE activities can
rock mechanics, be detected
concrete, andattributable
mining totopredict thermalthe stress during and the
damage
AE is a kind of phenomenon to emit low-level elastic wave from solids when they are stressed
UCG process
failure of brittleormaterials[23], it is also useful to estimate the progress of gasification process and special events [18–22].
deformed. because
The number they
of AEreach structural
occurrences failure by
are commonly accumulating
increased microfracture
just before solids is
such as collapsethe
destructed:
of stress
coal of in rock
the reach
cavity to
and extensive
uniaxial compressive
propagation
strength in
of gasification
uniaxial
zone.test.
compressive
InAEthis study,
As many AE activities
AE events and AE
can counts
be detected attributable to thermal stress during UCG processdata, [23],asit is also
monitoring is, therefore,are usedcalculated
in the fieldbyofmeans of data processing
rock mechanics, concrete, and ofmining
the raw AE signal
to predict the
useful to estimate
showndamage the
in Figure progress
and 6. theBoth of gasification
parameters
failure are counted
of brittle materials process
becausewhen and
theyAE special
signal
reach events
is higher
structural than
failure such
by a as collapse
threshold. Besides,
accumulating of coal in
the cavity AE
andevent
extensive
microfracture propagation
[18–22].
is not counted until the of
As many AE gasification
signal is damped.zone.
activities can be detected
The dead In time
this meaning
study, AE
attributable to for events
thermal stress and
during
that period AE counts are
is usually
UCG process [23],in it isbrittle
also useful to estimate theconcrete,
progress ofand gasification process and show
specialthe
events
calculated several
by means milliseconds
of data processing materials:
of the rock,
raw AE signal data, coal.
as AE
shown events
such as collapse of coal in the cavity and extensive propagation of gasification zone. In this study,
in Figure 6.number
Both of
parameters
cracks initiated inside a coal sample and AE counts reflect the magnitude of AE event. All AE
are counted when AE and
AE events signal is higher
AE counts than by
are calculated a means
threshold. Besides,of AE
of data processing the rawevent is not
AE signal data,counted
as until the
waveforms from the sensors were recorded using an oscilloscope (GR-7000; Keyence Corp.,
shown in Figure 6. Both parameters are counted when AE signal is higher than a threshold. Besides,
signal is damped.
Elmwood The NJ,
Park, dead USA)time with meaning time for that period is usually several milliseconds in brittle
AE event is not counted until sampling
the signal is damped. of 10The
µs.dead time meaning for that period is usually
materials: rock,several
concrete, and coal.
milliseconds AEmaterials:
in brittle eventsrock,
show the number
concrete, of cracks
and coal. AE initiated
events show inside
the number of a coal sample
cracks initiated inside a coal sample and AE counts reflect the magnitude
and AE counts reflect the magnitude of AE event. All AE waveforms from the sensors were recorded of AE event. All AE
waveforms from the sensors were recorded using an oscilloscope (GR-7000; Keyence Corp.,
using an oscilloscope
Elmwood(GR-7000; Keyence
Park, NJ, USA) Corp.,
with sampling time Elmwood
of 10 µs. Park, NJ, USA) with sampling time of 10 µs.

Figure 6. Definition of AE event and AE count.

Figure Figure 6. Definition of AE event and AE count.
6. Definition of AE event and AE count.

3. Results and Discussion

3.1. Temperature and AE

Temperature profiles for each experiment are plotted against the elapsed time in Figure 7.
The trend of the temperature increment was almost the same in both experiments. The temperatures
 Energies 2017, 10, 238 6 of 11
3. Results and Discussion
3. Results and Discussion
3.1. Temperature and AE
3.1. Temperature and
Temperature AE for each experiment are plotted against the elapsed time in Figure 7. The
profiles
Energies 2017, 10, 238 6 of 11
trend of the temperature
Temperature profiles increment was almost are
for each experiment theplotted
same inagainst
both experiments.
the elapsed time Thein temperatures
Figure 7. Thein
T11 and T12 located in the lower part of the co-axial well increased
trend of the temperature increment was almost the same in both experiments. The temperatures in the initial stage of in the
in T11 and
experiment.
T11 and T12 T12 located
Subsequently, in
located in the the lower
the lower part
temperatures of
part of the the co-axial
increased well
co-axialinwell increased
T13, increased
T21, and T22. in the
in theThese initial stage
results
initial of thethe
of
stage indicate
experiment.
experiment.
that Subsequently,
Subsequently,
the gasification the
area moved temperatures
the temperatures increased
in an upwardincreased in
direction inT13, T21,
T13,the
along and
T21, T22.
and T22.
co-axial These
well Theseresults
and results indicate
expanded indicate that
to the
the
widergasification
that the
area. area moved
gasification
Regarding areathe intemperature
movedan upward direction
in an upward along
direction
increment the co-axial
along
quantity, the well well
theco-axial
maximum and expanded
and
coalexpanded to the
temperature to wider
thein
area. Regarding
wider area. the temperature
Regarding the increment
temperature quantity,
increment the maximum
quantity, the
experiment 2 reached over 1200 °C, much higher than that in experiment 1. Besides, the highcoal
maximumtemperature
coal in experiment
temperature in 2
reached over 1200 ◦ C, much higher than that in experiment 1. Besides, the high temperature area
experiment area
temperature 2 reached
expanded over 1200
quickly°C, in
much higher than
experiment that inresults
2. These experiment
can be 1. Besides,
explained theby highthe
expanded
temperature
promotion quickly
area
of the inoxidation
experiment
expanded 2. These
quickly
reaction inresults
with the can
experiment be2.explained
increase These by thecan
of the results
oxygen promotion of the oxidation
be explained
inflow because by the
the coal
promotion
reaction
temperature of
thethe
withincreased oxidation
increase with of thereaction
the oxygenwith
oxidation the increase
inflow
reactionbecause of
thethe
and reacted coaloxygen inflow
temperature
with the because
coal increased
and oxygen. the coal
withThisthe
temperature
oxidation
finding leads increased
reaction with
and reacted
to a possibility the
to with oxidation
createthe reaction
coal and
a larger and
oxygen. This
gasification reacted with
finding
area with the coal and
leads temperature
a higher oxygen.
to a possibility forto This
create
a short
finding
aperiod
larger leadsdue
to a to
gasification
of time possibility
areathewith atohigher
promotion create athe
larger
oftemperaturegasification
oxidation for area period
a short
reaction. with a higher
of timetemperature for a short of
due to the promotion
theperiod of time
oxidation due to the promotion of the oxidation reaction.
reaction.

(a)
(a) (b)
(b)
Figure
Figure7.7.Temperature
Temperatureprofiles:
profiles:(a)
(a) Experiment 1; (b)
Experiment 1; (b) Experiment
Experiment2.2.
Figure 7. Temperature profiles: (a) Experiment 1; (b) Experiment 2.
Figure
Figure8 8shows showsthe theresults
resultsofofAE AE activities
activities for
for each
each experiment.
experiment. From From the the results
resultsofofthe theAE AE
event, Figure
it 8 shows
increased the
with results
the of AE
elapsed activities
time in for
both each experiment.
experiments, From
meaning
event, it increased with the elapsed time in both experiments, meaning that the number of the the results
that the of the
number AE of event,
the
itfracturing
increased
fracturing events
eventsincreased.
with theincreased.Based
elapsed Basedon
time onthe
in theresults
both results of
of the
experiments,the temperature
meaning profile,we
that
temperature profile, the wecan
numbercanconsider
considerthat
of the thatthe
fracturingthe
fracturing
events events
increased.
fracturing are
eventsBased aregenerated ininaawide
on the results
generated of the
wide range due
due to
temperature
range to theprofile,
expansion
we can
expansion of the
of the gasification
consider that the
gasification zone
zone ininthe
fracturingthe
later
events stage
later are ofofthe
generated
stage theexperiments.
in a wide range
experiments. Additionally,
due to theAE
Additionally, events
events in
expansion
AE initial stage
stagezone
of the gasification
initial of experiment
of experiment
in the later22stage wereof
were
higher
higher than
than those
those ofofexperiment
experiment 11 and
and the
the AE
AE counts
counts were the
the highest
highest
the experiments. Additionally, AE events in the initial stage of experiment 2 were higher than those in
in the
the initial
initial stage
stage ofof
of experiment
experiment
experiment 2,2,1when
whenthe
and theAE
the temperature
temperature
counts were changed
changed drastically.
drastically.
the highest in the This means
stagethat
means
initial that the
the fracturing
of experiment fracturing events
2, wheneventsthe
causedbybythe
caused
temperature thethermal
changed thermal stresswere
stress
drastically. were activated
Thisactivated
means that in experiment
in experiment
the fracturing2 after
after
events the
the ignition
ignition
caused by ofof the
the the coal.
coal.The
thermal The
stress
were activation
activation
activatedofoffracturing
fracturing
in experiment events
events creates
creates
2 after alarge
large number
the aignition number of cracks;
of the coal. as
as aa result,
The activation result, chemical
chemicalreactions
of fracturing reactions are
events creates are
promoted
apromoted
large number and
and are are attributable
attributable
of cracks; to increasing
to increasing
as a result, chemicalthe the reaction-specific
reaction-specific
reactions surface
surface
are promoted area.
andarea. Therefore,
Therefore, the
are attributable the reaction
reaction
to increasing
area areaof of coal
coal gasification
gasification can
can bebe expanded
expanded by
by utilizing
utilizing thermal
thermal stress
stress
the reaction-specific surface area. Therefore, the reaction area of coal gasification can be expandedleading
leading to
to activated
activated fracture
fracture by
events
events under
under thethehigh
hightemperature.
temperature.
utilizing thermal stress leading to activated fracture events under the high temperature.

(a) (b)
(a) (b)

Figure 8. Cont.
Energies 2017, 10, 238 7 of 11
Energies
Energies 2017,
2017, 10,
10, 238 7 7ofof1111

(c) (d)
(d)
Figure
Figure results of AE
AE activities: (a)
(a) AE events
events (experiment 1);
1); (b) AE
AE counts
Figure 8.8.Monitoring
Monitoring activities:
results of AE activities: AE(experiment
(a) AE events (experiment
1); (b) AE counts (b) counts
(experiment 1);
(experiment 1); (c) AE events (experiment
(experiment (experiment 2);
2); (d)
(d) AE
AE counts
counts (experiment
(experiment 2).
2).
(c) AE events (experiment 2); (d) AE counts (experiment 2).

3.2. Product
3.2. Product Gas Quality and Gasification Efficiency Efficiency
3.2. Product Gas Quality and Gasification Efficiency
Figure 99 shows
Figure shows the the monitoring
monitoring resultsresults of of the
the main
main compositions
compositionsand andthe thecalorific
calorificvalue
valueofofthe the
Figure 9 shows the monitoring results of the main compositions and the calorific value of the
product gas, which can be calculated with the concentration of the combustible gas contentssuch
product gas, which can be calculated with the concentration of the combustible gas contents suchasas
product gas, which can be calculated with the concentration of the combustible gas contents such as
CO, CH
CO, CH44,, H
H22,, and
and other
other hydrocarbons
hydrocarbons [24]. [24]. The
The calorific
calorific value
value ofof the
the product
productgas gasininexperiment
experiment22
CO, CH4 , H2 , and other hydrocarbons [24]. The calorific value of the product gas in experiment 2
decreased dramatically with a decrease of the combustible gas contents after 2 h elapsed,while
decreased dramatically with a decrease of the combustible gas contents after 2 h elapsed, whilethatthat
decreased dramatically with a decrease of the combustible gas contents after 2 h elapsed, while that of
of experiment 1 decreased slightly. These differences are attributed to the inhibition ofof the
of experiment 1 decreased slightly. These differences are attributed to the inhibition the
experiment 1 decreased slightly. These differences are attributed to the inhibition of the gasification
gasification reaction
gasification reaction withwith the
the heat-resistant
heat-resistant cement cement around
around aa coal coal block
block based
basedon onthetheresults
resultsofofaa
reaction with the
cross-section studyheat-resistant
after the cement around a coal block basedthat on the results of a cross-section study
cross-section study after the experiment
experiment (Figure(Figure 10), 10), meaning
meaning that the the gasification
gasificationarea areaexpanded
expanded
after the
rapidly in experiment
in the
the initial (Figure
initial stage
stage due 10), meaning that the gasification area expanded rapidly in the initial
rapidly due to to the
the excess
excess oxygen
oxygen inflow
inflow in in experiment
experiment 2.2. Additionally,
Additionally,the the
stage due to the excess oxygen inflow in experiment 2. Additionally, the calorific value in the end
calorific value in the end phase of experiment 1 was relatively lower than that in the beginningofof
calorific value in the end phase of experiment 1 was relatively lower than that in the beginning
phase of experiment
experiment 1 was relatively lower than that insomewhat
the beginning of experiment 2 even generated
though the
experiment 22 even
even though
though the
the injection
injection flowflow rate
rate was
was somewhatsimilar. similar.Molten
Moltenslag slagwaswas generated
injection
during the flow rate
the gasification was somewhat
gasification process
process due similar. Molten slag was generated during the gasification process
during due toto the
the ash
ash contents
contents of of the
thecoal.
coal.Therefore,
Therefore,the theslag
slagformation
formationmay may
due to the
prevent theash contents of
promotion of the
the gasification
coal. Therefore, reactionthe slag formation
because of the may prevent
limitation of thegas-solid
the promotion of the
contact.
prevent the promotion of the gasification reaction because of the limitation of the gas-solid contact.
gasification
From these reaction because
these considerations,
considerations, theof the limitation of the gas-solid contact. From these considerations,
From the effect
effect ofof the
the injection
injection flowflow rate
rate on
onthethequality
qualityof ofthe
theproduct
productgas gasand
and
the effect of the injection flow rate on the quality of the product gas and the gasification efficiency,
the gasification efficiency, which means the energy recovery rate from coal, is discussed by usingaa
the gasification efficiency, which means the energy recovery rate from coal, is discussed by using
which
part ofofmeans
the data
datathewhich
energywere
recovery
not rate from coal, is discussed bycement using a part of the data which were
part the which were not affected
affected by by thethe heat-resistant
heat-resistant cement and andthe theslag
slagformation.
formation.The The
not affected by the heat-resistant cement and the slag formation. The
range of data processing was 0.5~3.0 h for experiment 1, and 0.5~1.5 h for experiment 2. The flow rate range of data processing was
range of data processing was 0.5~3.0 h for experiment 1, and 0.5~1.5 h for experiment 2. The flow rate
0.5~3.0
in thesehranges
for experiment 1, and 0.5~1.5
for both experiments washstable
for experiment
(6 L/min for 2. experiment
The flow rate 1, 15 inL/min
these forranges for both
experiment
in these ranges for both experiments was stable (6 L/min for experiment 1, 15 L/min for experiment
experiments
2). Additionally, was thestable (6 amount
total L/min for of experiment
oxygen inflow 1, 15 L/min
was almostforthe
experiment 2). Additionally,
same in these ranges. the total
2). Additionally, the total amount of oxygen inflow was almost the same in these ranges.
amount of oxygen inflow was almost the same in these ranges.

(a) (b)
(a) (b)
Figure 9. Main compositions and the calorific value of a product gas: (a) Experiment 1; (b)
Figure 9. Main compositions and the calorific value of a product gas: (a) Experiment 1; (b)
9. Main
Experiment
Figure 2. compositions and the calorific value of a product gas: (a) Experiment 1; (b) Experiment 2.
Experiment 2.
Energies 2017, 10, 238 8 of 11
Energies 2017, 10, 238 8 of 11

(a) (b)
Figure 10. Cross-section study after the experiment: (a) Experiment 1; (b) Experiment 2.
Figure 10. Cross-section study after the experiment: (a) Experiment 1; (b) Experiment 2.

The typical product gas data in the range of data processing are listed in Table 2. The average
Thevalue
calorific typical product
and gasproduction
the total data in the flow
rangeofofthe
data processing
product gas inare listed in Table
experiment 2 were2. The average
higher than
calorific value and the total production flow of the product gas in experiment 2 were higher
those in experiment 1. These differences can be explained by the promotion of the reduction reaction than those
in experiment
in 1. These differences
which combustible can be
gas contents wereexplained by the
produced promotion
during the UCGof theprocess
reduction reactionthe
because in
which combustible
concentrations of Hgas contents were produced during the UCG process because the concentrations
2 and CO were higher, and that of CO2 was lower in experiment 2. This fact
of H 2 and CO were higher,
indicates that the gasification andreactions
that of CO
are2 was lower under
promoted in experiment 2. This factinindicates
high temperature that the
the gasification
gasification reactions are promoted under high temperature in the gasification regions
regions as a result of the increase of the oxygen inflow to activate the oxidation reaction. as a result of
the increase of the oxygen inflow to activate the oxidation reaction.
Table 2. Typical product gas data (0.5~3.0 h for experiment 1, 0.5~1.5 h for experiment 2).
Table 2. Typical product gas data (0.5~3.0 h for experiment 1, 0.5~1.5 h for experiment 2).
Average Calorific Total Product N2, O2 Free
Value Calorific)
Average(MJ/Nm
3 Flow
Total (m3)
Product H2 (%) CH4 N(%) CO (%)
2 , O2 Free CO2 (%)
3
Experiment 1 Value 6.39
(MJ/Nm ) Flow (m3 )
1.57 H29.34
2 (%) 5.40
CH 4 (%) 38.20
CO (%) 22.11
CO 2 (%)
Experiment
Experiment 21 8.11
6.39 2.26
1.57 33.65
29.34 5.19
5.40 45.23
38.20 17.61
22.11
Experiment 2 8.11 2.26 33.65 5.19 45.23 17.61
Balance computation is a useful method to discuss the amount of coal reacted in the UCG
process [25]. The amount of gasified coal is calculable by the balance of the C element, as shown in
Balance computation is a useful method to discuss the amount of coal reacted in the UCG
Table 3. The amount of carbon content in a tar is not included in the balance sheet. The amounts of C
process [25]. The amount of gasified coal is calculable by the balance of the C element, as shown in
reacted in experiments 1 and 2, respectively, were 0.37 kg and 0.57 kg, meaning that 0.50 kg and 0.76
Table 3. The amount of carbon content in a tar is not included in the balance sheet. The amounts of C
kg of coal are expected to have been gasified based on the ultimate analyses of the coal. Considering
reacted in experiments 1 and 2, respectively, were 0.37 kg and 0.57 kg, meaning that 0.50 kg and 0.76 kg
that the coal calorific value was 31.48 MJ/kg, the gasification efficiency, the energy recovery rate
of coal are expected to have been gasified based on the ultimate analyses of the coal. Considering that
from coal, is calculable using Equation (1).
the coal calorific value was 31.48 MJ/kg, the gasification efficiency, the energy recovery rate from coal,
is calculable using Equation (1). ET Wg
Rg =ET /Wg ×100 , (1)
Rg = Q × 100, (1)
Qcc
where
where R Rgg is
is the
the gasification
gasification efficiency (%), E
efficiency (%), ETT means
means the
the total
total energy
energy (MJ),
(MJ), W
Wgg represents
represents the
the gasified
gasified
coal (kg), and Q stands for the coal calorific value (MJ/kg).
coal (kg), and Qc stands for the coal calorific value (MJ/kg).
c

Table 3. Calculation of C balance (0.5~3.0 h for experiment 1, 0.5~1.5 h for experiment 2).

Experiment 1 Experiment 2
Total Amount of Balance of C Total Amount of Balance of C
Component
Product Gas Element Product Gas Element
mol mol kg mol mol kg
CH4 2.32 2.32 0.03 3.64 3.64 0.04
CO 15.87 15.87 0.19 30.11 30.11 0.36
CO2 11.45 11.45 0.14 11.82 11.82 0.14
C2H4 0.49 0.98 0.01 0.84 1.68 0.02
C2H6 0.12 0.24 0 0.15 0.30 0.00
Energies 2017, 10, 238 9 of 11

Table 3. Calculation of C balance (0.5~3.0 h for experiment 1, 0.5~1.5 h for experiment 2).

Experiment 1 Experiment 2
Total Amount Balance of Total Amount Balance of
Component
of Product Gas C Element of Product Gas C Element
mol mol kg mol mol kg
CH4 2.32 2.32 0.03 3.64 3.64 0.04
CO 15.87 15.87 0.19 30.11 30.11 0.36
CO2 11.45 11.45 0.14 11.82 11.82 0.14
C2 H4 0.49 0.98 0.01 0.84 1.68 0.02
C2 H6 0.12 0.24 0 0.15 0.30 0.00
C3 H6 0.06 0.18 0 0.00 0.00 0.00
C3 H8 0.02 0.06 0 0.00 0.00 0.00
Total 30.33 31.1 0.37 46.56 47.55 0.57

Table 4 presents the calculation results for the gasification efficiency. The energy of the product gas
can be calculated with the product flow rate and the calorific value of the product gas. The values for
the gasification efficiency in experiments 1 and 2, respectively, are 71.22% and 82.42%. A comparison of
the results shows that the UCG process with the higher oxygen inflow has a higher efficiency for energy
recovery from coal than that with the lower oxygen inflow, attributable to increasing the combustible
contents in the product gas. The product gas quality depends on the reduction reaction during the
UCG process, meaning that the temperature field is strongly affected. Therefore, it might be possible to
estimate the gasification efficiency by creating a proper numerical model. Additionally, it is pointed out
that the recovered energy from coal can be improved with the increase of the reaction temperature and
the expansion of the gasification area during the UCG process. In summary, an increase of the oxygen
inflow makes the temperature of the coal increase by the promotion of the oxidation reaction; then the
fracture events are activated. As a result, the quality of the product gas is improved, attributable to
expanding the gasification area with a high temperature.

Table 4. Calculation of gasification efficiency (0.5~3.0 h for experiment 1, 0.5~1.5 h for experiment 2).

Coal Calorific Energy of Amount of C Gasified Gasification

Value (MJ/kg) Product Gas (MJ) Element (kg) Coal (kg) Efficiency (%)
Experiment 1 11.21 0.37 0.50 71.22
31.48
Experiment 2 19.72 0.57 0.76 82.42

The product gas recovered with the UCG process is variable depending on the type of UCG
operation, the coal quality, and the underground conditions. According to the results of this study,
a key factor to improve the quality of the product gas is to expand the gasification reaction area with
an increase of the reaction temperature. The reaction temperature can be improved by the promotion
of the oxidation reaction while the expansion of the reaction area depends on the coal characteristics.
It means that not only injection conditions but also the improvement of the gasifier conditions to
promote the expansion of reaction area should be discussed to develop a highly efficient UCG system.

4. Conclusions
The injection condition is one of the key parameters to control the product gas quality in the UCG
process. This paper discussed the effect of the injection flow rate on the product gas quality by means
of model UCG experiments on a laboratory scale. The results showed that the gasification efficiency
can be improved with the increase of the reaction temperature and the expansion of the gasification
area by increasing the oxygen inflow. This finding suggests that a key issue for the improvement of the
gasification efficiency is to control the gasifier conditions. The proper injection conditions promote
Energies 2017, 10, 238 10 of 11

the oxidation reaction which increases the coal temperature, leading to activation of fracturing events
caused by thermal stress. Therefore, the control of fracturing events is another option that can be
considered for the improvement of the overall process.
To develop a more efficient UCG system, the improvement techniques to expand the gasification
area with a high temperature, such as the improvement of coal permeability and the effects of the
reactor pressure, are necessary to investigate in future studies.

Acknowledgments: This work was supported by the Japanese Society on UCG, Mikasa City, Center of
Environmental Science and Disaster Mitigation for Advanced Research of Muroran Institute of Technology,
JSPS KAKENHI Grant Number 15H02332, Sanbi Mining Co., Ltd., the Ministry of Education, Culture, Sports,
Science and Technology (MEXT), Japan. The authors gratefully acknowledge their support.
Author Contributions: A.H. and K.I. conceived and designed the experiments; F.S. performed the experiments;
K.T. carried out gas analysis and analyzed the data; J.K. and G.D. extensively supported to the experiments and
the data analyses with constructive advises; A.H. wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Khadse, A.; Qayyumi, M.; Mahajani, S.; Aghalayam, P. Underground coal gasification: A new clean coal
utilization technique for India. Energy 2007, 32, 2061–2071. [CrossRef]
2. Porada, S.; Czerski, G.; Dziok, T.; Grzywacz, P.; Makowska, D. Kinetics of steam gasification of bituminous
coals in terms of their use for underground coal gasification. Fuel Process. Technol. 2015, 130, 282–291.
[CrossRef]
3. Stańczyk, K.; Kapusta, K.; Wiatowski, M.; Świadrowski, J.; Smoliński, A.; Rogut, J.; Kotyrba, A. Experimental
simulation of hard coal underground gasification for hydrogen production. Fuel 2012, 91, 40–50. [CrossRef]
4. Yang, L.H.; Zhang, X.; Liu, S.Q.; Yu, L.; Zhang, W.L. Field test of large-scale hydrogen manufacturing from
underground coal gasification (UCG). Int. J. Hydrogen Energy 2008, 33, 1275–1285. [CrossRef]
5. Japan’s Energy White Paper 2016. Available online: http://www.enecho.meti.go.jp/about/whitepaper/
2016pdf/whitepaper2016pdf_2_1.pdf (accessed on 28 December 2016).
6. Yang, L.; Liang, J.; Yu, L. Clean coal technology—Study on the pilot project experiment of underground coal
gasification. Energy 2013, 28, 1445–1460. [CrossRef]
7. Nourozieh, H.; Kariznovi, M.; Chen, Z.; Abedi, J. Simulation Study of Underground Coal Gasification in
Alberta Reservoirs: Geological Structure and Process Modeling. Energy Fuels 2010, 24, 3540–3550. [CrossRef]
8. Luo, Y.; Coertzen, M.; Dumple, S. Comparison of UCG cavity growth with CFD model predictions.
In Proceedings of the Seventh International Conference on CFD in the Minerals and Process Industries,
Melbourne, Australia, 9–11 September 2009.
9. Bhutto, A.W.; Bazmi, A.A.; Zahedi, G. Underground coal gasification: From fundamental to applications.
Prog. Energy Combust. Sci. 2013, 39, 189–214. [CrossRef]
10. Kaćur, J.; Durdán, M.; Laciak, M.; Flegner, P. Impact analysis of the oxidant in the process of underground
coal gasification. Measurement 2014, 51, 147–155. [CrossRef]
11. Stańczyk, K.; Howaniec, N.; Smoliński, A.; Świadrowski, J.; Kapusta, K.; Wiatowski, M.; Grabowski, J.;
Rogut, J. Gasification of lignite and hard coal with air and oxygen enriched air in a pilot scale ex situ reactor
for underground gasification. Fuel 2011, 90, 1953–1962. [CrossRef]
12. Hongtao, L.; Feng, C.; Xia, P.; Kai, Y.; Shuqin, L. Method of oxygen-enriched two-stage underground coal
gasification. Min. Sci. Technol. (China) 2011, 21, 191–196. [CrossRef]
13. Su, F.Q.; Itakura, K.; Deguchi, G.; Ohga, K.; Goto, T. Laboratory Studies on Evaluation of Gasification Effect
for Conversion of Coal Resources in Underground Coal Gasification (UCG) reactors. Adv. Mater. Res. 2012,
600, 111–115. [CrossRef]
14. Su, F.Q.; Itakura, K.; Hamanaka, A.; Deguchi, G.; Ohga, K. Ex-situ UCG Model Experiments with Oxygen
Enriched Air in an Artificial Coal Seam. In Proceedings of the International Conference on Material Science
and Engineering Technology 2016, Phuket, Thailand, 14–16 October 2016.
15. Su, F.Q.; Itakura, K.; Hamanaka, A.; Deguchi, G.; Ohga, K. Evaluation of gasification zone and energy
recovery during UCG process with the Coaxial models. In Proceedings of the 5th Sustainable Energy and
Environmental Sciences, Singapore, 22–23 February 2016.
Energies 2017, 10, 238 11 of 11

16. Hamanaka, A.; Itakura, K.; Su, F.Q.; Takahashi, K.; Satoh, K.; Deguchi, G.; Kodama, J. Model Experiment for
Co-axial Underground Coal Gasification System Development. In Proceedings of the 24th World Mining
Congress (Sustainability in Mining), Reo de Janeiro, Brazil, 18–21 October 2016.
17. Hamanaka, A.; Itakura, K.; Su, F.Q.; Satoh, K.; Takahashi, K.; Deguchi, G.; Kodama, J.; Ohga, K. Development
of Compact Underground Coal Gasification (UCG) System as a Local Energy Resource. In Proceedings of
the International Symposium on Earth Science and Technology 2015, Fukuoka, Japan, 3–4 December 2015.
18. Lokner, D. The Role of Acoustic Emission in the Study of Rock Fracture. Int. J. Rock Mech. Min. Sci.
Geomech. Abstr. 1993, 30, 883–899. [CrossRef]
19. Tang, C.A.; Liu, H.; Lee, P.K.K.; Tsui, Y.; Tham, L.G. Numerical studies of the influence of microstructure on
rock failure in uniaxial compression—Part I: Effect of heterogeneity. Int. J. Rock Mech. Min. Sci. 2000, 37,
555–569. [CrossRef]
20. Li, Y.P.; Chen, L.Z.; Wang, Y.H. Experimental research on pre-cracked marble under compression. Int. J.
Solids Struct. 2005, 42, 2505–2516. [CrossRef]
21. Zhang, M.; Shimada, H.; Sasaoka, T.; Matsui, K.; Dou, L. Seismic energy distribution and hazard assessment
in underground coal mines using statistical energy analysis. Int. J. Rock Mech. Min. Sci. 2013, 64, 192–200.
[CrossRef]
22. Suzuki, T.; Ohtsu, M. Quantitative damage evaluation of structural concrete by a compression test based on
AE rate process analysis. Constr. Build. Mater. 2004, 18, 197–202. [CrossRef]
23. Su, F.Q.; Nakanowataru, T.; Itakura, K.; Ohga, K.; Deguchi, G. Evaluation of Structural Changes in the Coal
Specimen Heating Process and UCG Model Experiments for Developing Efficient UCG Systems. Energies
2013, 6, 2386–2406. [CrossRef]
24. Su, F.Q.; Itakura, K.; Deguchi, G.; Ohga, K.; Kaiho, M. Evaluation of Energy Recovery from Laboratory
Experiments and Small-scale Field Tests of Underground Coal Gasification (UCG). J. MMIJ 2015, 131, 203–218.
[CrossRef]
25. Wiatowski, M.; Kapusta, K.; Świadrowski, J.; Cybulski, K.; Ludwik-Pardala, M.; Grabowski, J.; Stańczyk, K.
Technological aspects of underground coal gasification in the Experimental “Barbara” Mine. Fuel 2015, 159,
454–462. [CrossRef]

© 2017 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).