Fuel 331 (2023) 125838

Contents lists available at ScienceDirect

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

Full Length Article

Hydrogen Production through Water Vapors using Optimized Corona-DBD

Hybrid Plasma Micro-Reactor
Muhammad Younas, Sumeer Shafique, Abrar Faisal, Ainy Hafeez, Fahed Javed, Maria Mustafa,
Fahad Rehman *
Microfluidics Research Group, Department of Chemical Engineering, COMSATS University Islamabad, Lahore Campus, Pakistan

A R T I C L E I N F O A B S T R A C T

Keywords: Hydrogen as secondary energy carrier is one of the promising routes to meet renewable and sustainable energy
Hydrogen energy demand. Hydrogen production using water plasmolysis, so far, has been considered as energy intensive. How­
Plasmolysis ever, recent developments in the field of microplasmas has reduced energy requirement of water plasmolysis on
Corona-DBD hybrid plasma reactor
par with electrolysis. However, there’s still room for improvement as water vapors dissociation efficiencies are
Micro channel reactor
Steam decomposition
less than thermodynamic and kinetic limits. One of the ways to do that is to split microchannel in plasma into
sub-micron channels by packing beads inside the reactor. The current study investigates the plasmolysis of water
vapours (steam) and argon gas with (1) empty channel plasma reactor and (2) glass beads packed plasma reactor.
A custom-made Corona-DBD hybrid microplasma reactor has been designed and employed for water steam
plasmolysis. This study aims to investigate the feasibility of water vapors dissociating into hydrogen at a rela­
tively small interelectrode gap at atmospheric pressure. The minimum breakdown voltage has been found to be
3.7 kVpk-pk at 5 mA current. The maximum hydrogen production rate was observed 23.9 g/kWh at optimized
conditions for the glass beads packed plasma reactor. So, in current study glass beads packed plasma micro­
reactor yielded 23.9g of H2/kWh which is significantly higher than any previously published study

However, recent developments in the field of microplasmas, better

understaning of the kinetics of the process and robust materials of
1. Introduction contrustion, have enhanced the efficiency of plasmolysis significantly
[12,14]. Thermodynamic limit for plasma processes and other hydrogen
The continuous urbanization and growth of the world’s population producing technologies has been reported 25-30 g/kWh [12,15].
have led to a considerable increase in energy demand [1,2]. Generally, Most of the studies on hydrogen production has reported energy
renewable energy resources such as wind, hydropower, solar, and yield ≤ 20 g/kWh. The energy yield is often reported in g/kWh [12,15].
biomass provide energy through a centralized power system- a capital It highlights the gram of hydrogen produced per kWh power consumed.
intensive system [3,4]. However, the solution is to store surplus Rincon et al. used a microwave reactor to produce a 0.3007 g/kWh yield
renewable energy in an energy carrier, such as a hydrogen. As a sec­ of hydrogen with 0.93% ethanol (EtOH). A selectivity of 82.5%
ondary energy carrier, hydrogen has been on the scene for several de­ hydrogen with a power consumption of 300 kW at the frequency of 2.45
cades now [5]. Its green nature is considered conditional depeding upon GHz was reported using ethanol as feedstock and and argon as carrier
the source of power used to produce it. Current, commercially domi­ gas [16]. Kirkpatrick et al. reported a 0.25 g/kWh yield of hydrogen by
nating process, Steam Methane Reforming (SMR) is neither green nor using a macroplasam reactor with corona discharge [17]. Fridman et al.
sustainable [6]. Hydrogen could also be produced by thermochemical [15] used a microwave plasma reactor to dissociate water vapors for
cycles [7], photocatalysis [8], biohydrogen [9], electrolysis and plas­ hydrogen production that yielded 10 g/kWh. Rehman et al. [12] used
molysis [10]. Electrolysis produce 4-5% of globally produced hydrogen Corona-DBD hybrid plasma reactor to dissociate steam with a yield of 10
which is considered totally green as it prevents GHG’s emission [11]. g of H2/kWh. It was found that the hydrogen yield could be increased by
However, it is considered as capital as well as energy intensive [6,12]. adding Argon in the feed as discussed in later part of the manuscript,
Plasmolysis, an alternative hydrogen production technology to elec­ section 5.5. Burlica et al. has used non-thermal pulsed plasma gliding arc
trolysis using electricity, was considered energy intensive [13].

* Corresponding author.
E-mail address: [email protected] (F. Rehman).

https://doi.org/10.1016/j.fuel.2022.125838
Received 29 May 2022; Received in revised form 17 August 2022; Accepted 23 August 2022
Available online 12 September 2022
0016-2361/© 2022 Elsevier Ltd. All rights reserved.
M. Younas et al. Fuel 331 (2023) 125838

center of T-shaped borosilicate plasma reactor as shown in Fig. 1.

Nomenclature
2.2. Experimetal Design
SMR Steam Methane Reforming
GHG’s Green House Gases Experiments were performed to optimize the plasma reactor and
DBD Dielectric Barrier Discharge investigate the effect of glass beads packed reactor on hydrogen pro­
SS Stainless Steel duction while comparing them with single-channel empty plasma
AC Alternating Current reactor. The steam was generated from the digitially control vaporizer
DC Direct Current (Electrical engineering services (EES)). Argon gas flowrate was
RF Radio Frequency controlled by using a flow meter (LZM-8M). However, the steam was
Pk-Pk Peak to Peak generated using a fully automated vaporizer with digital control over
V-I Voltage - Current pressure, temperature and volumetric flow rate. The experiments were
Ar Argon conducted by varying only one parameters out of temperature, pressure
L/D Length to Diameter Ratio and volumetric flow and keeping the rest at constant value. The plasma
ETOH Ethanol reactor was ignited by injecting water vapors from one side of reactor
HV High Voltage while argon from the other side of reactor, at fixed sinusoidal frequency
30.17 kHz. There were only a few centimeters (~12) difference between
vaporizer and reactor. The pressure was 1 bar and temperature were
equal or more than 100 ◦ Cin the experiments. Initially, steam was
reactor to produce hydrogen maximum energy yield 13 g/kWh [18]. condensing when the pipeline connecting the reactor and vaporizer was
Chehade et al. has used the microwave reactor to produce 13.3 g/kWh of at room temperature. However, after few minutes, the condensation
hydrogen energy by using 900 W power to dissociate water [19]. The stopped as the experimental setup reached the steam temperature. The
hydrogen production of corona discharge can be increased by incorpo­ Argon gas was 99.99% and purchased from Pakistan Oxygen Limited
rating Dielectric Barrier Discharge (DBD) material allowing reactor to (POL). The plasma reactor packed with glass beads and optimized by
operate at higher power levels and produce higher hydrogen yield [12]. varying electrode length from 1-4.5 cm, voltage 4-5.5 kVpk-pk, steam
When reactor is packed with glass beads, it is subdivided into sub­ flowrate 60-240 mL/h, steam temperature 398.15-443.15 K and argon
micron channels, creating close contact points between beads and flowrate 40-80 mL/min to investigate their effects on hydrogen pro­
beads-electrodes. Due to shorter distances between the beads and be­ duction yield (g/kWh). The consumed power was measured according to
tween beads and electrodes, the electric field present in these sub- voltage varying from 4-5.5 kVpk-pk. Every experiment was run for 2
micron gaps is stronger as compared to the mean value in the reactor. minutes before taking measurements.
Addition of glass beads not only creates the sub-micron channels inside The products were collected from the reactor outlet using gas
the plasma microreactor, it increases the dielectric constant. The higher collection bag (Restek, multilayered bag with polypropylene combo
dielectric constant can reduce the electron number density. Hower, it is valve, capacity: 1 liter) designed to contain hydrogen. The sample
hypothesized, that the plasma microreactor packed with glass beads, collected in the bags were kept in deep freezer (-2◦ C) for 3 hours to
despite lower electron density, can result in better electric field distri­ condense unreacted steam. The bags were attached with the hydrogen
bution as well as ehance energy efficiency of hydrogen production analyzer (RKI Eagle 2 Model) fitted with a pump and auto sampler to
process. quantify the hydrogen.
This study is envisaged on (1) the feasibility of hydrogen production
at atmospheric pressure by water vapor plasmolysis at relatively smaller 2.3. Hydrogen analysis
interelectrode distance, focused on (2) optimization of operational pa­
rameters (voltage, feed flowrate, carrier gas flowrate, length of ground The standard solutions of hydrogen, produced from hydrogen
electrode, power consumption and temperature of steam/water vapors) generator, and argon gas of different
and (3) comparison of hydrogen energy yield using empty channel and concentrations (from 1 to 10 Vol%) were used to calibrate hydrogen
glass beads packed plasma microreactor. analyzer. The calibration was used
to measure hydrogen produced from Microplasma reactor.
2. Material and Methods
3. Electrical characterization
2.1. Experimental setup
The Fig. 2 depicts Voltage-Current (V-I) output signal. The experi­
A custom- made power supply (PVM 2000) containing control panel ment began with a voltage of 4 kVpk-pk, which was subsequently
was used to ignite plasma. The power supply can provide peak to peak increased to 5.5 kVpk-pk. The glass beads packed plasma microreactor
voltages up to 60 kVpk-pk and frequency from 20 kHz to 100 kHz. The was ignited by varying voltage from 4 to 5.5 kVpk-pk and sinusoidal
display unit contained two manual meters for voltage and amperage as frequency was kept 30.17 kHz. The current-voltage output signals were
well as one digital LCD to display frequency. The voltage, current and recorded at digital oscilloscope. There may occur instabilities to monitor
frequency was calibrated and quantified using a high voltage probe (PT output signals of power supply because of distorted signals. However,
52409), current probe (Tek P6041) and oscilloscope (UTD 2202CM). A the current-voltage waveform in Fig. 2 is evident that the source is
T-shaped borosilicate plasma reactor was designed and developed with providing stable signals at 5.5kV. The minimum breakdown voltage for
7 mm external diameter and 4.88 mm internal diameter. The stainless- a discharge gap of 1.28 mm and steam flowing at 1 atm was measured
steel (SS) rod of diameter 2.40 mm was used as the live electrode. The 3.7 kVpk-pk at 5 mA. The reason for a relatively low breakdown voltage of
designed reactor contained 2.48 mm flowing channel with 1.24mm water vapours has been discussed elsewhere [12,15]. The breakdown
radial gap inside. The glass beads diameter ranges 1-2 mm. The volume voltage for air was noted 30 kVpk-pk /cm at 1 atm pressure, therefore, the
of reactor was calculated using expression (1) to be 193.22mm3 breakdown of water may be expected much higher than air due to
density difference between air and water. This happens because of high
V = πr2 h (1)
electric field that promote formation of gas channel in water body [15].
The aluminum tape, which was used as ground electrode, was The process occurring on the electrode surfaces can explain this phe­
wrapped around the reactor. Stainless steel circular rod was fitted in the nomenon. The processes occurring at the electrode surface can be

2
M. Younas et al. Fuel 331 (2023) 125838

Figure 1. Experimental setup for hydrogen production from Corona-DBD hybrid plasmolysis

4. Results and discussion

25
Voltage
6
Current 20 4.1. Effect of voltage on hydrogen production

15
4 The effect of voltage on hydrogen production has been investigated
10 with 180 mL/h inlet steam flowrate and 3.6 L/h argon flowrate. The
2 plasma microreactor was ignited at applied sine frequency 30.17 kHz,
Current (mA)
Voltage (kV)

5
electrode length 3 cm, steam temperature 403.15 K by varying voltage
0 0 from 4 kVpk-pk to 5.5 kVpk-pk to see its effect for water vapors dissociation
into hydrogen production, because beyond 5.5 kVpk-pk arcing started.
-5
-2 The experiments were conducted with reactor packed with the beads.
-10 The Fig. 3 shows a gradual increase in energy yield by increasing
-4 voltage. The experiments were repeated three times and standard error
-15
was calculated using standard deviation. The maximum error was found
-6
-20 to be 4% at 5.2 kV. Energy yield is calculated by following expression
(2);
-25
0 50 100 150 200 250 300 350 400 450 500 550 600

Time (μS)

Figure 2. Voltage current output signals waveform at 5.5kV

divided into two processes [12,20]. First process is further subdivided

into two phases. A bubble forms in the first phase as a result of liquid
vaporization caused by local heating in the strong electric field regions
at the electrode tip. The breakdown happens in the second phase, the
electronic phase, owing to ionizing collisions of electrons as they travel
through the breakdown gap. Because the water is already provided in
the vapor phase, the first phase does not exist in the current study
making the current plasomlyis more fast and energy efficient. The
overall energy for water breakdown through plasmolysis is lowered
because water is provided in the vapor phase, which gives an additional
energy (latent heat of vaporization of water). However, second process
is for underwater electical discharge and arc formation that is not
applicable in current study. Although, the breakdown voltage is deter­
mined by experimentation is 3.7 kVpk-pk at 5 mA by igniting atmospheric
DBD plasma reactor, the optimized analysis is being performed at 5.5
kVpk-pk at 35 mA.

Figure 3. The effect of voltage on hydrogen energy yield (L/D: 4.3, Steam
temperature: 403.15 K, Steam F.R: 180 mL/h, Argon F.R: 3.6 L/h)

3
M. Younas et al. Fuel 331 (2023) 125838

H2 produced (g/min) increased more than 4 cm or 5.71 L/D ratio, electric field strength was
Hydrogen energy yield =
[Plasma power + Qadd + Condensation heat] kWh not enough strong to dissociate more water vapors into its constituents
(2) such as hydrogen and oxygen. This is because, by increasing the L/D
ratio, the charge is spread over larger surface consequently weakens the
This is evident from Fig. 3, that energy yield (g/kWh) shows electric field strength, taking place over the large area of reactor.
increasing relation with gradually increasing voltage up to 5.5 kVpk-pk. Moreover, an increase in overlapping electrode length decreases power
This significant increase could be attributed to increase electrical energy density. So, when electric field strength as well as power density is
density. The increase in electrical energy density results more energized retarded, recombination effect may prevail, which consequently de­
electrons. Consequently, an increase in electrons’ energy produces more creases hydrogen energy yield. Therefore, the decrease in hydrogen
possibilities for electron collisions to water vapors molecules concentration is attributed to electric field distribution to large area,
[14,21,22]. The more possibilities of highly energized electrons at decrease in power density, and predominant recombination effect
increased voltage results more dissociation of water vapors into its because of larger residence time.
constituent elements like hydrogen and oxygen [23]. Therefore, high
voltage 5.5 kVpk-pk was being opted as optimized voltage for plasma
microreactor to dissociate water vapors to produce hydrogen. 4.3. Effect of water vapors/steam concentration on hydrogen production

Steam flow rate plays a pivotal role in the production of hydrogen as

4.2. Effect of length of electrode on hydrogen production
shown in Fig. 5. The flow rate of water vapors was varied from 60 mL/h
to 240 mL/h. The glass beads packed bed microplasma reactor was
The length of plasma zone plays a vital role in producing hydrogen
ignited at voltage 5.5 kVpk-pk. All experiments were carried out at con­
gas. The larger electrode length will lead to a larger plasma zone. which
stant frequency of 30.17 kHz. The plasma was ignited with these pre­
favorably results in more decomposition of water vapors/steam for
scribed operational parameters as well as 5.71 L/D ratio. The maximum
hydrogen production. On the other hand, the shorter length does not
hydrogen energy yield was found to be 19.5 g/kWh when steam inlet
allow for adequate residence time to generate the required results. The
flowrate was 180 mL/h while beyond its energy yield becomes almost
overlapping region of electrodes was changed to vary the length to
steady. The maximum error was found to be 4% at 150 mL/h. The
diameter ratio of the plasma zone. The length to diameter ratio (L/D)
hydrogen conversion could be calculated by following expression (4).
could be calculated by utilizing the following expression (3).
( )
( ) Hydrogenproduced mol
L Lengthofoverlappingelectrode Hydrogen conversion (%) = ( hr )*100 (4)
Length to diameter ratio = (3) watervaporsflowrate mol
D Diameterofreactor hr

At the frequency of 30.17 kHz and water vapors/steam inlet flowrate It was observed that increasing steam flowrate directly increases the
180 mL/h, overlapping length was varied between 1 to 4.5cm with a 7 breakdown voltage. By increasing the flow rate of water vapors their
mm inner diameter of the reactor. Micro plasma reactor was ignited at concentration in argon (Ar) increases which may enhance the break­
5.5 kVpk-pk from power supply at these prescribed electrode lengths. The down voltage. Fig. 5 shows a steady incease in the energy yield from
Fig. 4 depicts maximum energy yield was obtained at 4 cm long plasma 6.70 to 19.46 g/kWh as the flow rate is increased from 60 mL/hr to 240
zone or 5.71 length to diameter ratio (L/D), is 19.46%. The maximum mL/h. The increase in hydrogen yield from 180mL/h to 240 mL/h is
error was found to be 4% at 5.7 L/D ratio. approximately 4.6%. However, this increase lies in the range of exper­
The increase in hydrogen production by increasing overlapping imental error, and it seems system has reached to a steady state.. When
electrode length or L/D ratio is attributed to corresponding increase in steam flowrate was increased up to 240 mL/h, breakdown voltage was
residence time of steam in reactor in an igniting active zone of plasma. observed 3.9 kV. However, increasing steam flowrate, concentration of
This increases collision probability of molecules as well as promote water vapors in argon increases which eventually results more water
chemical reactions for desired results [24]. Moreover, hydrogen energy vapor dissociation as well as hydrogen production. There is an another
yield decreased gradually when ground electrode length was increased aspect that by increasing water vapors flowrate, residence time de­
by 4 cm or 5.71 L/D ratio with a corresponding residence time of more creases. However, residence time does not suppress below 10-2 s, min­
than 3.9 minutes. It clearly shows that when electrode length is imum residence time for hydrogen production by Rehman et al. [25],

Figure 4. The effect of L/D ratio on hydrogen energy yield (Voltage: 5.5 kVpk- Figure 5. The effect of steam flowrate on hydrogen energy yield (Voltage: 5.5
pk, Steam temperature: 403.15 K, Steam F.R: 180 mL/h, Argon F.R: 3.6 L/h) kVpk-pk, Steam temperature: 403.15 K, L/D: 5.71, Argon F.R: 3.6 L/h)

4
M. Younas et al. Fuel 331 (2023) 125838

therefore there could not bee seen any downfall of energy yield by absence of argon. Afterwards, micro plasma reactor was ignited by using
increasing water vapors flowrate. There could be observed a steep slope mixture of argon gas and water vapors. The steam and argon were mixed
up to 180 mL/h steam flowrate for hydrogen energy yield, while beyond using Y-shape inline mixer, having two inlets for argon and steam and
its hydrogen energy yield becomes stable. It seems that steady threshold one outlet for the mixture. The produced gas was collected and analyzed
limit has been achieved by system at flowrate 180 mL/h, while further by using hydrogen analyzer, in both scenarios. Hydrogen yield was
increase in steam flowrate couldn’t result in an immense increase in found to be doubled when argon was used because of the Penning effect.
hydrogen energy yield. Consequently, for this specified micro plasma The collisions of the argon excited species (Ar (3P)) with the water
reactor inlet steam flow rate up to level 180 mL/h has been opted as molecule are responsible for this finding. The produced Ar (3P) with
optimized parameter along with voltage 5.5 kVpk-pk. argon atoms by impact collision strike with water molecules which re­
sults water molecules dissociation. Consequently, the dissociation by Ar
(3P) results extra OH radicals’ production. The mechanism for extra OH
4.4. Effect of Steam Temperature on Hydrogen Production
radicals’ production has been demonstrated by Dey et al. [27].
Ar + e → Ar (3P) + e (1)
The effect of temperature was investigated by igniting glass beads
Ar (3P) + H2O → H + OH + Ar (2)
packed plasma microreactor at varying inlet steam temperature between
This was in agreement with discussion by Burlica et al. [18] as well as
388.15K to 448.15K. To investigate the effect of feeding water vapor
Rehman et al. [12].
temperature on hydrogen production, water vapor flowrate, argon
Micro plasma was ignited by varying argon flowrate 1.2 L/h to 4.2 L/
flowrate, L/D ratio were kept constant for all experimentation. The
h, while keeping other parameters constant as mentioned before, which
samples were collected and analyzed at sinusoidal wave frequency 30.17
consequently increases hydrogen energy yield shown in Fig. 7. The
kHz as well as 5.5 kVpk-pk applied plasma voltage. Fig. 6 depicts the
maximum error was found to be 4% at 3.6 L/h.
direct relation of feeding steam temperature increase with hydrogen
This is evident from Fig. 7 that hydrogen energy yield shows
production. At the highest feeding input water vapor temperature of
increasing relation with gradually increasing argon flowrate up to 3.6 L/
433.15 K, the maximum hydrogen energy yield was found 23.9 g/kWh.
h. However, beyond 3.6 L/h argon flowrate, hydrogen energy yield does
The maximum error was found to be 2.5% at 403 K.
not show a steep slope but steady. This significant increase could be
This significant increase could be attributed to an increase in water
attributed to extra OH radicals formation [12] and high electron energy
vapors energy and reactor heating temperature. The reactor heating
ionization that consequently result in more water vapors dissociation
temperature could add more energy that enhances electron energy
into their constituent elements like hydrogen and oxygen [18]. It may be
preferably [26]. So, increase in electron energy and water vapors energy
concluded that steady slope beyond 3.6 L/h argon flowrate shows that
results in more possibilities of water vapors-electron collision. The water
the packed-bed DBD plasma microreactor has achieved threshold limit
vapors dissociation into its constituent elements like hydrogen and ox­
and further increase in argon flowrate wouldn’t result tremendous in­
ygen could be as more as water vapors-electron collision possibilities.
crease in hydrogen energy yield. Hence, argon flowrate increase beyond
This could be observed from Fig. 6 that hydrogen energy yield increases
3.6 L/h will shift current hydrogen production system into an uneco­
up to 433.15 K, while beyond its hydrogen production becomes almost
nomical fashion.
steady. It seems almost threshold hydrogen energy yield may be
achieved.
4.6. Comparison of hydrogen energy yield for empty channel and glass
4.5. Effect of Argon flowrate on hydrogen production beads packed plasma reactor

The experimental results for effect of argon on the production of The comparison between empty channel reactor and glass beads
hydrogen was investigated by passing argon gas through glass beads packed reactor was investigated to find optimum hydrogen production
packed plasma microreactor at varying flowrates. The flowrate of argon yield. The optimized conditions found were 180 mL/h steam flowrate,
was varied from 1.2 to 4.2 L/h at stable constant sinusoidal 30.17 kHz 3.6 L/h argon flowrate, 5.71 L/D ratio, and 433.15 K steam temperature.
frequency, 5.71 L/D ratio, 180 mL/h steam flowrate at 433.15 K tem­ The empty channel plasma reactor followed by glass beads packed
perature. The micro reactor was initially ignited solely with steam in the reactor was ignited at varying voltage ranges 4-5.5 kVpk-pk. The empty

Figure 6. The effect of steam temperature on hydrogen energy yield (Voltage: Figure 7. The effect of argon flowrate on energy yield (Voltage: 5.5 kVpk-pk,
5.5 kVpk-pk, Steam flowrate: 180 mL/h, L/D: 5.71, Argon F.R: 3.6 L/h) Steam flowrate: 180 mL/h, L/D: 5.71, Steam temperature: 433.15 K)

5
M. Younas et al. Fuel 331 (2023) 125838

channel plasma reactor could sustain up to 4.6 kVpk-pk, befor converting

H2producedcontaininguseableenergyinkWh
of homogenous plasma into arc discharge. The Fig. 8 depicts the direct Hydrogenenergyefficiency=
[Plasmapower+Qadd+Condensationheat]kWh
relation for hydrogen energy yield at varying voltages for empty channel
reactor as well as reactor packed with glass beads. The maximum error (6)
was found to be 3% at 5.2 kV. In above expression, Qadd is amount of heat added to convert water
The maximum hydrogen energy yield at maximum voltage 5.5 kVpk- into steam and condensation heat is amount of heat required to
is obtained 23.9 g/kWh. However, empty channel plasma reactor yiel­ condense output gases stream before entering into hydrogen analyzer
ded 15.8 g/kWh at 4.6 kVpk-pk and above its empty channel plasma for hydrogen quantification. The maximum energy efficiencies 52.5%
reactor could not sustain. and 79.4% were observed, when plasma ignited at prescribed optimum
Generally, the Lissajous method is used to calculate the power parameters, for empty channel (at 64 W) and glass beads packed plasma
consumed by electrical discharges [28] which is approximatey < 30- reactors (at 100 W), respectively. A direct relation could be observed
60% of the total AC power used to sustain the plasmas [29]. The between power consumption (W) and energy efficiency (%) for empty
remaining power is squandered in the form of heat and light etc., channel and glass beads packed plasma reactor in Fig. 9.
However, one has to pay for the total amount of power utilized in the The maximum error was found to be 4% at 80 watts. It was observed
whole process- total AC power. However, total power could be calcu­ that reactor heating temperature increases at higher voltages. The
lated by using expression below [30]; reactor heating is considered an important factor which increases the
hydrogen mole fraction in the dissociated elements. As the name
P = VI cosϕ (5)
microreactor suggests, the volumes of microreactor are very small.
Where, V is AC voltage amplitude measured using high voltage Power levels of 30-100 W can heat up (Joule Heating) a microreactor
probe, I is the current measured by using current probe and Ф is the significantly and affects the kinetics and dynamics of overall process. El-
angle between voltage and current. Voltage is measured by the high Shafie et al. found that increasing the reactor heating temperature re­
voltage probe, current is measured by the current probe and frequency duces the ambient heat transfer rate, plasma dissociation became more
by the Oscilloscope. The phase angle was calculated from digital oscil­ potent which eventually hydrogen production [31,32]. Similarly,
loscope. The electrical power utilized in plasma in DBDs is not the same packed bed plasma microreactor ignition at higher power may enhance
as the total power drawn from the mains. This is due to the fact that in plasma reactor heating level that could also play a role for increased
capacitive systems, the voltage and current are not in phase. The phase hydrogen energy yield. The Joule heating or Ohmic heating can be
angle can cause the voltage to lag behind the current. Therefore, the calculated by Q = I2 Rt [33]. Where, I is the current, Q is the heat, t is the
discussed method provides an accurate method to calculate the power current flow time and R is the resistance. However, due to lack of
[28]. So, the power was calculated using equation (5) by varying the availability of instruments, it was not possible to calculate it.
voltage at a fixed frequency optimized by previous studies [14]. However, the difference in hydrogen energy yield and energy effi­
The varied plasma power is used to ignite DBD plasma reactor to ciency between empty channel and glass beads packed plasma reactor
optimize the power consumption, resulting in maximum hydrogen ef­ increases progressively. The diameter of the glass beads was in the range
ficiency (%). However, the power was varied by changing the voltage at of 1-2 mm with an average surface area of 28.27 mm2. The temperature
a fixed frequency. The power was varied from ~40 to ~100 watts to of the reactor and beads is expected to increase. However, the increase in
proceed plasma firing at stable constant argon flowrate 3.6 L/h, steam temperature of the reactor or beads was not measured as the main focus
flowrate 180 mL/h, steam temperature 433.15 K and L/D ratio 5.71. The of the study was investigate hydrogen yield under different reactor
plasma was ignited with prescribed conditions as well as at sinusoidal configuration and operational conditions. Therefore, while comparing
frequency 30.17 kHz. The energy efficiency of plasma was investigated empty channel and glass beads packed plasma reactor, it could be
by igniting plasma at optimized conditions by using empty channel and observed that at voltages above 4.6 kVpk-pk or 64 W, glass beads packed
glass beads packed plasma reactor. The energy efficiency was calculated reactor shows steep slope while empty channel plasma reactor couldn’t
by following expression (6). sustain to ignite. However, Fig. 10 shows discharge patterns for empty
channel and packed bed Corona-DBD hybrid plasma microreactor.
The empty channel plasma reactor was found to be less efficient due

Figure 8. The effect of voltage on hydrogen energy yield for empty channel Figure 9. The effect of voltage on hydrogen energy yield for empty channel
reactor and glass beads packed reactor (Argon F.R: 3.6 L/h: Steam flowrate: 180 reactor and glass beads packed reactor (Argon F.R: 3.6 L/h: Steam flowrate: 180
mL/h, L/D: 5.71, Steam temperature: 433.15 K) mL/h, L/D: 5.71, Steam temperature: 433.15 K)

6
M. Younas et al. Fuel 331 (2023) 125838

steam dissociation into its constituent elements like hydrogen and ox­
ygen. Thus, maximum hydrogen production energy yield and energy
efficiency could be observed for glass beads packed plasma reactor.
The optimum conditions for different configurations are expected to
be different. However, the purpose of this comparison is to evaluate the
effect of packing glass beads in plasma microreactor on hydrogen yield.
The comparison also highlights a significant difference in voltage at
which arcing occurs under both conditions. The maximum errors of 3
and 4% were found in the experiments as shown in Figs. 8 & 9,
respectively.

4.7. Energy yield for hydrogen production

Water vapors were dissociated into their constituent elements by

igniting plasma at optimized parameters. The produced gases were
collected in gas collection bag which was cooled in a deep freezer (-2◦ C).
Figure 10. The discharge patterns for empty channel and packed bed Corona- Furthermore, hydrogen gas was quantified with hydrogen analyzer.
DBD hybrid plasma microreactor Plasma was ignited with empty channel and glass beads packing which
resulted 15.8 and 23.9 g/kWh, respectively. Fig. 11 represents the en­
to absence of any dielectric material. It also causes arcing to occur at ergy yield of current work compared to different hydrogen production
lower voltage up to 4.6 kVpk-pk or 64 W power. The empty channel technologies.
plasma microreactors are attributed to lower mean electron energy, The present research will make use of waste heat, which is abundant
lower electric field homogeneity and increased electron density. How­ in chemical engineering firms and power plants. Most of the time, the
ever, packing plasma reactor with the beads divides the channel into exhaust temperature from most Industrial operations and power plants
several submicron channels and make close points between packed is being released at temperatures below 370 ◦ C [12]. It is considered
material-electrode and packed material itself. Eventually, the electric uneconomical to recover heat by using heat recovery technologies.
field becomes strong because of contact points of glass beads. The However, it is caused environmental pollution by releasing heat into
dielectric constant (4-5) of glass beads, packed in plasma reactor makes environment [37]. It would be highly feasible to utilize waste heat to
the electric field stronger and uniform. The packed bed microreactor convert water into vapors, which ultimately could be converted into
results in an increase of electron energy by increasing electric field hydrogen by plasmolysis [12].
strength, lowers electron density as well as increases electric field ho­ Plasma microreactor can be successfully upscaled by multiplexing
mogeneity [34-36], which eventually results in more water vapors/ them as shown by Hafeez et al. [36,38]. A small hydrogen generator
could be developed using the similar principles. At larger scale, the

Figure 11. Comparison of current work with different hydrogen-producing technologies energy yield

7
M. Younas et al. Fuel 331 (2023) 125838

generator can be developed having number of multiplexed units. How­ References

ever, it would require extensive R&D.
[1] Javed F, Aslam M, Rashid N, Shamair Z, Khan AL, Yasin M, et al. Microalgae-based
biofuels, resource recovery and wastewater treatment: a pathway towards
5. Conclusion sustainable biorefinery. 2019;255:115826.
[2] Ali S, Fazal T, Javed F, Hafeez A, Akhtar M, Haider B, et al. Investigating biodiesel
Hydrogen as a secondary energy carrier could play a critical role in production strategies as a sustainable energy resource for Pakistan. 2020;259:
120729.
providing an environmentally friendly option to cover the growing [3] Dehghani-Sanij A, Tharumalingam E, Dusseault M, Fraser RJR, Reviews SE. Study
global demand for energy. Hydrogen production highly depends on of energy storage systems and environmental challenges of batteries. 2019;104:
fossil fuels which are gradually depleting. The world is looking forward 192–208.
[4] Nikolaidis P, Poullikkas AJR. reviews se. A comparative overview of hydrogen
to produce hydrogen by utilizing renewable energy resources in cost production processes. 2017;67:597–611.
effective fashion. In current research, a T-shaped borosilicate plasma [5] Holladay JD, Hu J, King DL. Wang YJCt. An overview of hydrogen production
reactor has been designed and developed with 7 mm external diameter technologies. 2009;139(4):244–60.
[6] Rehman F. Simultaneous Generation and Separation of Hydrogen by Plasmolysis
as well as 4.88 mm internal diameter. The circular stainless-steel (SS)
Department of Chemical Engineeirng PhD. Sheffield: The University of Sheffield 2015.
rod diameter 2.40 mm has been used as a live electrode. The designed [7] Rosen MAJE. Advances in hydrogen production by thermochemical water
reactor contained 2.48 mm flowing channel with 1.24 radial gap inside. decomposition: a review. 2010;35(2):1068–76.
Afterwards, the production of hydrogen with plasmolysis has been [8] Colón GJACAG. Towards the hydrogen production by photocatalysis. 2016;518:48-
59.
investigated for two cases (1) empty channel plasma reactor (2) and [9] Fazal T, ur Rehman MS, Mushtaq A, Hafeez A, Javed F, Aslam M, et al.
Glass beads packed plasma reactor. To show the stability of signals, Simultaneous production of bioelectricity and biogas from chicken droppings and
voltage and current waveform has been given. The plasma reactor has dairy industry wastewater employing bioelectrochemical system. 2019;256:
115902.
been ignited to investigate minimum breakdown voltage which resulted [10] Younas M, Shafique S, Hafeez A, Javed F, Rehman FJF. An overview of hydrogen
3.7 kVpk-pk at corresponding current 5 Ma ◦ . Initially, glass beads production: current status, potential, and challenges. 2022;316:123317.
packed plasma reactor has been optimized by varying different param­ [11] Dawood F, Anda M. Shafiullah GJIJoHE. Hydrogen production for energy: An
overview. 2020;45(7):3847–69.
eters as voltage, L/D ratio, steam flow rate, steam temperature, and [12] Rehman F, Abdul Majeed WS, Zimmerman WB. Hydrogen production from water
argon flowrate. Optimum voltage, L/D ratio, steam flow rate, steam vapor plasmolysis using DBD-Corona hybrid reactor. Energy & fuels 2013;27(5):
temperature, and argon flowrate have been investigated with glass 2748-61.
[13] Givotov V, Fridman A, Krotov M, Krasheninnikov E, Patrushev B, Rusanov V, et al.
beads packed plasma reactor 5.5 kVpk-pk, 5.71, 180 mL/h, 433.15 K Plasmochemical methods of hydrogen production. 1981;6(5):441–9.
and 3.6 L/h, respectively. The maximum thermodynamics limit (kinetic [14] Rehman F, Liu Y, Zimmerman WBJijohe. The role of chemical kinetics in using O3
limit) for different hydrogen producing technologies has been given 25- generation as proxy for hydrogen production from water vapour plasmolysis. 2016;
41(15):6180-92.
30 g/kWh [12]. However, the maximum hydrogen yield for empty
[15] Fridman A. Plasma chemistry. Cambridge University Press; 2008.
channel and glass beads packed plasma reactors have been investigated [16] Rincón R, Marinas A, Muñoz J, Melero C, Calzada MD. Experimental research on
15.8 g/kWh (at 4.6 kVpk-pk or 64 W power) and 23.9 g/kWh (5.5 kVpk-pk ethanol-chemistry decomposition routes in a microwave plasma torch for hydrogen
or 100 W power), respectively. Furthermore, an investigation has been production. Chemical Engineering Journal 2016;284:1117–26.
[17] Kirkpatrick MJ, Locke BR. Hydrogen, oxygen, and hydrogen peroxide formation in
made for energy efficiency by varying power from 40-100 watts which aqueous phase pulsed corona electrical discharge. Industrial and Engineering
resulted maximum 79% % or 0.79 energy efficiency for glass beads Chemistry Research 2005;44(12):4243–8.
packed plasma reactor. While, empty channel plasma reactor could [18] Burlica R, Shih K-Y, Locke BJI, Research EC. Formation of H2 and H2O2 in a water-
spray gliding arc nonthermal plasma reactor. 2010;49(14):6342-9.
sustain up to 4.6 kVpk-pk or 64 W power, which resulted 52.45% energy [19] Chehade G, Lytle S, Ishaq H, Dincer I. Hydrogen production by microwave based
efficiency. Electrolysis gives 4% of total hydrogen production world­ plasma dissociation of water. Fuel 2020;264(August 2019):116831.
wide. Water electrolysis faces the challenges of hydrogen production in [20] Akiyama HJITod, Insulation e. Streamer discharges in liquids and their
applications. 2000;7(5):646-53.
a cost-competitive way, because it is considered an energy intensive [21] Ding C, Yuan D, Wang Z, He Y, Kumar S, Zhu Y, et al. Ozone production influenced
hydrogen production pathway. The maximum hydrogen production by increasing gas pressure in multichannel dielectric barrier discharge for positive
yield for electrolysis is 20 g/kWh [12]. However, the current study and negative pulse modes. 2018;40(3):228–36.
[22] Hafeez A, Shamair Z, Shezad N, Javed F, Fazal T, ur Rehman S, et al. Solar powered
investigated maximum hydrogen production from glass beads packed decentralized water systems: A cleaner solution of the industrial wastewater
plasma reactor yielded 23.9 g/kWh, which is about 16.31 % more treatment and clean drinking water supply challenges. 2020:125717.
hydrogen production yield while comparing it with electrolysis with [23] Varne M, Dey G. Das TNJIJoHE. Evaluation of optimum conditions for hydrogen
generation in argon-water vapor dielectric barrier discharge. 2016;41(48):
benefits of less power consumption and reduced equipment size. The
22769–74.
current study envisages that hydrogen production through cutting edge [24] Bahri M, Haghighat F, Rohani S, Kazemian HJCEJ. Impact of design parameters on
technology like plasmolysis could be used for rural areas through the performance of non-thermal plasma air purification system. 2016;302:204–12.
decentralized power system. [25] Rehman F, Liu Y, Zimmerman WBJ. The role of chemical kinetics in using O3
generation as proxy for hydrogen production from water vapour plasmolysis.
International Journal of Hydrogen Energy 2016;41(15):6180–92.
Declaration of Competing Interest [26] El-Shafie M, Kambara S, Hayakawa YJES. Engineering. Comprehensive assessment
of hydrogen production in argon-water vapors plasmolysis. 2021;9(2):267–83.
[27] Dey G, Zode S, Namboodiri VJPoP. Application of plasma for efficient H2
The authors declare that they have no known competing financial production: A realism of copper electrode in single dielectric barrier discharge
interests or personal relationships that could have appeared to influence reactor. 2018;25(10):103508.
the work reported in this paper. [28] Hafeez A, Javed F, Fazal T, Shezad N, ur Rehman MS, Rehman FJCE, et al.
Intensification of ozone generation and degradation of azo dye in non-thermal
hybrid corona-DBD plasma micro-reactor. 2021;159:108205.
Data availability [29] Duan Y, Huang C. Yu QJRosi. Cold plasma brush generated at atmospheric
pressure. 2007;78(1):015104.
[30] Hollenstein C, Guittienne P. Howling AJPSS, Technology. Resonant RF network
Data will be made available on request. antennas for large-area and large-volume inductively coupled plasma sources.
2013;22(5):055021.
Acknowledgment [31] El-Shafie M, Kambara S, Hayakawa YJIJoHE. Experimental analysis of plasma and
heating effect on H2 permeation behavior through Pd–Cu40% membranes in 1mm
gap length plate reactor. 2020;45(49):26310-20.
Authors would like to acknowledge the Higher Education Commis­ [32] El-Shafie M, Kambara S, Hayakawa Y. Miura TJIJoHE. Preliminary results of
sion (HEC) for funding this research under the National Research Pro­ hydrogen production from water vapor decomposition using DBD plasma in a
gram for Universities (NRPU) Project Number 7924/17. PMCR reactor. 2019;44(36):20239–48.
[33] Cai Z, Li S, Zhang Q, Wang C, Jin Z, Fu M, et al. Derivation and Verification of the
Relationship between Ablation Index and Baseline Impedance. 2021;2021.

8
M. Younas et al. Fuel 331 (2023) 125838

[34] Ding H-X, Zhu A-M, Yang X-F, Li C-H. Xu YJJoPDAP. Removal of formaldehyde [36] Hafeez A. Developing Multiplexed Hybrid Plasma Micro-Reactor for Wastewater
from gas streams via packed-bed dielectric barrier discharge plasmas. 2005;38(23): Sterilization. 2021.
4160. [37] Wei D, Lu X, Lu Z, Gu JJEc, Management. Performance analysis and optimization
[35] Kim H-H, Oh S-M, Ogata A, Futamura SJACBE. Decomposition of gas-phase of organic Rankine cycle (ORC) for waste heat recovery. 2007;48(4):1113-9.
benzene using plasma-driven catalyst (PDC) reactor packed with Ag/TiO2 catalyst. [38] Hafeez A, Shezad N, Javed F, Fazal T, ur Rehman MS, Rehman FJCE, et al.
2005;56(3):213–20. Developing multiplexed plasma micro-reactor for ozone intensification and
wastewater treatment. 2021;162:108337.