Plasma-Assisted Combustion Technology using
Nanosecond Pulsed Power
Ryo Nakamura
Hidenori Akiyama, Takashi Sakugawa
Graduate School of Science and Technology
Kumamoto University
2-39-1 Kurokami, Chuo-ku, Kumamoto 860-8555, Japan
Institute of Pulsed Power Science
Kumamoto University
2-39-1 Kurokami, Chuo-ku, Kumamoto 860-8555, Japan
Abstract— A next generation type of ignition system has been
developed for gasoline combustion engines called “plasma-assisted
combustion technology” which both improves engine performance
by increasing lean burn flammability and reduces emissions by
applying non-equilibrium plasma to gasoline. However, detailed
effects of non-equilibrium plasma on gasoline are poorly
understood. This study investigates the reforming mechanism of
gasoline by non-equilibrium plasma using gas chromatography.
First, a nanosecond pulsed power generator using fast recovery
diodes was developed to generate non-equilibrium plasma. This
pulse generator supplied 9 ns rise-time, 17 ns FWHM, and 32.5 kV
amplitude pulses to a 1 kΩ load. Non-equilibrium plasma can be
produced at pressures between 0.1 MPa to 0.5 MPa, limited by the
vessel. Pulsed power from this pulsed generator was applied to a
spark plug used by conventional ignition systems in gasoline
engines. The gasoline was reformed by non-equilibrium plasma at
0.1 MPa; components of the reformed gasoline were then
investigated using gas chromatography. Results reveal that lower
hydrocarbons such as methane, ethylene, ethane and propylene,
were newly produced. Also, increasing pulse shot number was
found to increase generated amounts of these lower hydrocarbons.
We concluded that the lower hydrocarbons may be produced by
the active species generated by the non-equilibrium plasma. It is
well known that lower hydrocarbons more easily ignite than
higher hydrocarbons. Plasma-assisted combustion technology
thus holds potential to improve the engine performance. We are
currently investigating the credibility of our hypothesis and effects
of non-equilibrium plasma to generate lower hydrocarbons,
considering the features of applying pulses.
Keywords—Pulsed power; Nanosecond pulse; Non-equilibrium
plasma; Plasma assisted combustion; Gas chromatography
I. INTRODUCTION
The development of internal combustion technologies with
higher efficiency and eco-friendliness is required to mitigate
resource consumption and worldwide environmental problems.
Lean burn techniques and exhaust gas recirculation (EGR),
already in use, are considered ideal technologies to counter these
problems. However, misfiring and a lowering of combustion
speed both occur in ultra-lean regions, limiting these
technologies. In order to solve this, several new ignition
enhancement techniques have been proposed such as plasma jet
ignition [4] and laser ignition [5]. However, these techniques
remain impractical.
Recently, performance improvement of electronic
components has allowed development of a next-generation type
of ignition system using non-equilibrium plasma called
“plasma-assisted combustion technology”. This new ignition
system applies non-equilibrium plasma to gasoline before
ignition and reforms gasoline to more combustible gases. Many
researchers have reported that plasma-assisted combustion
could enhance ignition and flame stabilization by reducing
ignition delay and improving combustion speed [1] - [3]. As a
result, this technology enables extension of lean combustion
limits.
Yiguang Ju et al. summarized major plasma-assisted
combustion enhancement pathways as consisting of the four
following phenomena: 1) thermal effect via temperature rise; 2)
kinetic effect via plasma-generated electronically excited
molecules and active species; 3) diffusion transport
enhancement effect via fuel decomposition and low temperature
oxidation; and 4) convective transport enhancement due to
plasma generated ionic wind, hydrodynamic instability, and
flow motion via Coulomb and Lorentz forces [6]. In particular,
investigating chemical reactions of fuel decomposition and low
temperature oxidation by plasma species is essential to
understand effects of plasma-assisted combustion in detail.
Though many reports about the work of production radicals via
fuel decomposition and oxidation and their contribution to
promoting chain-branching and chain-termination reactions
exist, details of chemical reactions of fuel decomposition
remains poorly understood.
Thus, this study investigates the reforming mechanism of
gasoline using gas chromatography based on our hypothesis that
lower hydrocarbons should be generated from higher
hydrocarbons via fuel decomposition through the application of
non-equilibrium plasma. In order to generate non-equilibrium
plasma, we first developed an inductive energy storage (IES)
pulsed power generator using a semiconductor opening switch.
This generator could supply pulses short enough to generate
non-equilibrium plasma between the gaps of a conventional
spark plug. Using this generator, we next reformed gasoline in a
chamber at atmospheric pressure and analyzed components of
reformed gases using gas chromatography. Finally, we
investigated the relationship between pulse shot number and the
components of reformed gases.
II. EXPERIMENTAL METHOD
A. IES pulsed power generator
Three required characteristics of pulses generating nonequilibrium plasma are a very short pulse duration, short rising
time, and high electric field. Conventionally, to meet these
requirements, capacitive energy storage (CES) pulsed power
generators such as Marx generators, Blumleins, and magnetic
pulse compression (MPC) are widely used, but the size of these
generators are too large for vehicles because capacitive energy
density is approximately 100 times lower than inductive energy
density. A further problem is that generators using spark gap
switches require regular maintenance.
As such, in this study, we adopted an inductive energy
storage (IES) pulsed power generator as the power supply for
non-equilibrium plasma due to its ease at generating nanosecond
pulses. Moreover, such generators are sufficiently compact for
vehicles and are maintenance free as they use semiconductor
opening switches. An IES pulsed power generator is classified
as a type of voltage amplifier caused by a quick interruption of
a circuit current using the opening switch in the same manner as
a surge voltage (inductive voltage). Thus, opening switch
characteristics are most important for an IES pulsed power
generator. In this study, we adopted VMI fast recovery diodes
(FRD) K100UF as the opening switch, and SiC-MOSFET
(CREE C2M0025120D) as the primary closing switch.
Fig. 1. IES type pulsed power generator using FRD.
Fig. 2. Typical waveforms of non-equilibrium plasma at 0.1 MPa.
An electric circuit diagram of the generator is shown in Fig.
1. It consists of primary and secondary circuits via a saturable
transformer ST. The primary circuit consists of a capacitor C0
with capacitance 560nF, a switch (SiC-MOSFET), and a
saturable coil SI0, whose role is magnetic assistance to reduce
switching loss. The secondary circuit consists of a capacitor C1
with capacitance 1 nF, a resistor with resistance 1 kΩ, and an
FRD. The winding numbers of SI0 are 10, and the turn ratio of
the ST is 1:16.
The circuit operation is described as follows. First, the
capacitor C0 is charged in the range of 10 V to 1 kV using a DC
high-voltage power supply; the maximum input energy is 250mJ
under a charge of 1 KV. Arrival of an external triggering pulse
to the drivers of SiC-MOSFET triggers initiation; it then
discharges through the loop of C0-SI0-ST-C1-FRD, resulting in
an FRD forward current capacitor C1 is charged by the ST
voltage. Meanwhile, both the inductive energy corresponding to
the inductance of the secondary circuit and the carrier are stored
in the FRD. When the ST saturates at the time of maximum
capacitor C1 voltage, capacitor C1 discharges through the loop
C1-ST-FRD, resulting in an instant FRD reverse current. During
the reverse current, the FRD suddenly recovers, and the
inductive energy 𝐿 𝑑𝑖 ⁄𝑑𝑡 is supplied to the load.
Consequently, this IES pulsed power generator supplies a 9
ns rise-time, 17 ns FWHM, and 32.5 kV amplitude pulses using
a 1 kΩ resistor as load. The current was measured using a current
monitor (Pearson, Model6585), and voltage was calculated by
the inductance of the secondary circuit and the FRD reverse
current. Output waveforms of voltage and current when
generating non-equilibrium plasma using a conventional spark
plug with a gap distance of 9mm as load at 0.1 MPa are shown
in Fig. 2. At this time, measurements showed that the charged
voltage of capacitance C0 was 570V, the output voltage was 17.2
kV, rise-time was 12.4 ns, and the streamer current was 42.0 A.
Also, consumption energy of non-equilibrium plasma was
approximately 2.3 mJ/pulse.
B. Analyzing reformed gases using gas chromatography
To understand the mechanisms and effects of plasmaassisted combustion, it is essential to analyze the components of
reformed gases. Thus, in this study, we analyzed components
using gas chromatography.
Fig. 3 shows the experimental setup for our analysis. It
consists of an IES pulsed power generator, spark plug (NGK,
BP4ES), chamber, rotary pump RP, gas sampler and gas
chromatograph GC (SHIMADZU, Tracera). The IES pulsed
power generator is connected to the spark plug, which had been
modified in order to extend its gap distance and discharge area;
the spark plug is fixed to the flange of the chamber. The volume
of the chamber is 233mm2, and 100μl gasoline is injected into
the chamber. After injecting gasoline, nanosecond pulses were
supplied to the spark plug under atmospheric pressure. Pulse
frequency was 1.2 kHz and processing time was varied at 30s,
90s and 180s to investigate the relationship between the pulse
shot number and product component changes. Fig. 2 shows
typical waveforms of non-equilibrium plasma; an image of non-
III. EXPERIMENTAL RESULTS AND DISCUSSTION
Fig. 5 shows experimental results of gas chromatogram of
non-reformed gasoline, while Fig. 6 shows that of reformed
gasoline by applying non-equilibrium plasma 180 seconds for a
total pulse shot number of 216,000. As mentioned in the
previous section, parameters were fixed, with the charged
voltage of capacitor C0 is 570 V, pulse frequency 1.2 kHz, and
consumption energy per pulse 2.3 mJ.
Fig. 3.
Fig. 4.
Experimental setup.
Images of non-equilibrium plasma.
As shown, gasoline resulted in peaks of nitrogen (N2),
carbon dioxide (CO2), propane (C3H8), isobutane (iso-C4H10)
and normal butane (n-C4H10), while reformed gasoline resulted
in newly detected peaks of methane (CH4), ethylene (C2H4),
ethane (C2H6), propylene (C3H6). Fig. 7 shows the relationship
between product components and pulse shot number; processing
times were 30s, 90s and 180s, representing shot numbers
respectively 36,000, 108,000 and 216,000. The vertical axis
represents the magnification between the intensity of each
results and the intensity of results from the shot number of
36,000. Fig. 7 illustrates the tendency of lower hydrocarbons
such as methane, ethylene, ethane, propylene and propane to
increase with increasing shot number. Conversely, isobutane
and normal butane decreased with increasing shot number.
These results support our hypothesis that these lower
hydrocarbons are generated via decomposition of higher
hydrocarbons by the work of active species.
In fact, it has been confirmed by other research groups that
components such as ethylene (C2H4), formaldehyde (CH2O),
hydrogen (H2) and carbon monoxide (CO) are produced by
equilibrium plasma is shown in Fig. 4. The right photograph in
Fig. 4 was shot from coaxial direction.
Reformed gases were pulled into the gas sampler using a
rotary pump. When a bourdon tube pressure gauge showed 0.09
MPa, sample gases were pulled into the gas chromatograph. In
the gas chromatograph, sample gases pass the column (Shinwa
Chemical Industries, MICROPACKED ST) with carrier gas He
and were detected by dielectric-barrier discharge ionization
detector (BID), which can detect 100 times more sensitively
than a thermal conductivity detector (TCD) and two times more
than a flame ionization detector (FID). BID enabled detection of
even very small amounts of reformed products. After injecting
sample gas into gas chromatograph, gases remaining in the
chamber were evacuated by closing the gas pulling valve and
opening the scavenging valve.
The relationship between gases and retention time is mainly
determined by column temperature. This and other parameters
of the gas chromatograph such as carrier gas flow velocity and
the column pressure were controlled by a program called
“method file.” In this study, standard gas was first analyzed, and
the relationship between the components of standard gas and
retention time was investigated. Next, non-reformed gasoline
and reformed gasoline were analyzed using the same method file
as that which analyzed standard gas. In this way, we could know
what gases were detected.
Fig. 5. Chromatogram of gasoline.
Fig. 6. Chromatogram of reformed gasoline at 216,000 shots.
combustion speed in lean condition by preventing minimum
ignition energy.
IV. ISSUES
Fig. 7. Relationship between product components and shot number.
There are some unknown peaks in Fig. 6 at retention times
of 0.34 min, 0.61 min, 1.05 min, 3.75 min, 4.58 min and 5.31
min. Considering the official report from SHIMADZU
Corporation using a column identical with that of our research,
peaks at 0.34 min are expected to be hydrogen (H2), 0.61 min
oxygen (O2), and 1.05 min carbon monoxide (CO). However, a
few unknown peaks remain which we will endeavor to identify
as a future task. Also, whether the amounts of newly generated
lower hydrocarbons are sufficient to extend the lean combustion
limit was unable to be ascertained because exact concentrations
of products remain unable to be identified. Thus, we intend to
create a calibration curve and investigate the absolute value of
products by absolute calibration method.
reforming higher hydrocarbon fuels such as normal heptane (nC7H16) using ozone (O3) [9]. In the case of this study, it is
considering that the sources of lower hydrocarbons are normal
butane and isobutane or more high hydrocarbons which we
couldn’t detect now. Also, it is confirmed that ethylene (C2H4),
ethane (C2H6), propylene (C3H6) and hydrogen (H2) are
produced by decomposition of methane (CH4) and
recombination reactions (reactions (1) - (3)) to form higher
hydrocarbons [11] - [15].
Moreover, the main components of gasoline are generally
considered to be higher hydrocarbons with carbon numbers of 7
to 10. However, because of our method file or column, we were
unable to detect peaks of such higher hydrocarbons using gas
chromatography. We are considering another possibility: if
these higher hydrocarbons cannot be vaporized under these
conditions, we could expect further increasing generated
amounts of lower hydrocarbons by increasing such higher
hydrocarbons which are their source.
CH3* + CH3* → C2H6
CH3*
+ CH3 → C2H4 + H2
*
CH3* + C2H3* → C3H6
(1)
(2)
(3)
Considering these results, there is the possibility that the reaction
pathways of production of lower hydrocarbons in this study are
mainly divided into two individual routes, one is the direct
production by the decomposition of higher hydrocarbons and
another is the indirect production by the recombination of
radicals such as CH3 and C2H3.
Results of research on minimum ignition energy against
equivalence ratio conducted by Lewis and von Elbe show
minimum ignition energy of lower hydrocarbons is relatively
lower than that of higher hydrocarbons in lean condition [7]. For
example, butane has the lowest ignition energy among the
components of gasoline at stoichiometric ratio at 0.8 mJ. But its
energy increases to 4 mJ at equivalence ratio 𝜙 = 0.7, while the
minimum ignition energy of ethane and methane, which are
newly generated by non-equilibrium plasma, decreased to 1.5
mJ and 0.8 mJ, respectively. Considering this report and our
experimental results, plasma-assisted combustion could prevent
a sharp rising of minimum ignition energy by reforming higher
hydrocarbons in gasoline to lower hydrocarbons via fuel
decomposition by the work of active species, thus resulting in
prevention of misfire.
In addition, research results showed that increased ignition
energy against minimum ignition energy leads to greater initial
combustion speed in hydrocarbon-air mixtures [8]. Thus, under
identical discharge energy to ignite mixtures, plasma-assisted
combustion not only prevents misfire but also improves
V. CONCLUSIONS
In this study, we investigated reform mechanisms of plasmaassisted combustion using gas chromatography. First, we
developed an IES pulsed power generator capable of generating
non-equilibrium plasma using a conventional spark plug. In the
present study, an output voltage of 17.2 kV, rise-time of 12.4 ns,
streamer current of 42.0 A, and consumption energy of nonequilibrium plasma of 2.3 mJ/pulse were measured when the
charged voltage of capacitance C0 was 570V.
Next, we analyzed reformed gases using gas
chromatography. After applying non-equilibrium plasma to
gasoline, newly-generated lower hydrocarbons such as methane,
ethylene, ethane and propylene were observed. Also observed
was that lower hydrocarbons increased while isobutane and
normal butane decreased with increasing shot number.
Application of 216,000 shot pulses compared to 36,000 shots
increased the intensity of propylene by 1.82 times, propane by
2.04 times, ethylene by 2.57 times, ethane by 3.07 times, and
methane by 3.68 times. These results indicate that lower
hydrocarbons are generated via decomposition of higher
hydrocarbons by the work of active species. Also, considering
the experimental results reported by other research group, there
may be an indirect pathway which lower hydrocarbons are
generated via recombination of radicals. These newly-generated
lower hydrocarbons would contribute to improvement of lean
burn technology to prevent misfire and increase initial
combustion speed. Accordingly, significance of plasma-assisted
combustion for practical application was shown by our
experimental results.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
Y. Ju and W. Sun, “Plasma assisted combustion: Dynamics and
chemistry,” Prog Energy Combsut. Sci., vol. 48, pp. 21-83, 2015.
K. Tanoue, E. Hotta, and Y. Moriyoshi, “Enhancement of ignition
characteristics of lean premixed hydrocarbon-air mixtures by repetitive
pulse,” Int. J. engine Res., vol. 10, pp.399-407, 2009.
M. Watanabe, E. Hotta, K. Tanoue, K. Ushimaru, T. Kuboyama, and Y.
Moriyoshi, “5. Radical measurements and ignition characteristics of
repetitive nano-pulse discharges plasma,” J. Plasma Fusion Res., Vol.
84, No. 4, pp.229-233, 2013.
H. F. Behbahani, A. M. Warrisand, and F. J. Weinberg, “The destruction
of nitric oxide by nitrogen atoms from plasma jets: designing for
thermal stratification,” Combust. Sci. Technol., vol. 30, pp.289-302,
1983.
G. Herdin, J. Klausner, E. Winter, M. Weinrotter, J. Graf, and K. Iskra,
“Laser ignition: A new concept to use and increase the potentials of gas
engines,” ASME 2005 Int. Combust. Engine Div. Fall Tech. Conf.,
pp.673-681, 2005
Y. Ju, et al., “Plasma assisted low temperature combustion,” Plasma
Chem. Plasma Process., vol. 36, pp. 85-105, 2016.
B. Lewis and G. von Elbe, Combustion Flames and Explosions of Gases,
3rd Ed., New York: Academic Press, 1987.
E. Murase, O. Moriue, H. Hashimoto, and I. Matsuzaki, “A Study on
Equivalence-Ratio Dependence of Minimum Ignition Energy Based on
Initial Burning Velocity,” Trans. Japan Soc. Mech. Eng. Series B, vol.
79, No. 805, pp. 1839-1847, 2013.
S. H. Won, B. Jiang, P. Dievart, C. H. Sohn, and Y. Ju, “Self-sustaining
n-heptane cool diffusion flames activated by ozone,” Proc. Combust.
Inst., vol. 35, pp.881-888, 2015.
H. J. Curran, P. Gaffuri, W. J. Pitz, and C. K. Westbrook, “A
Comprehensive Modeling Study of n-Heptane Oxidation,” Combust.
Flame, vol. 114, pp.149-177, 1998.
Th. Hammer, Th. Kappes, and M. Bakdauf, “Plasma catalytic hybrid
processes: gas discharge initiation and plasma activation of catalytic
processes,” Catalysis Today, vol. 89, pp. 5-14, 2004.
A. Indarto, J. W. Choi, H. Lee, and H. K. Song, “A brief catalyst study
on direct methane conversion using a dielectric barrier discharge,” J.
Chin. Chem. Soc., vol. 54, pp.823-828, 2007
O. Khalifeh, A. Mosallanejad, H. Taghvaei, M. R. Rahimpour, and A.
Shariati, “Decomposition of methane to hydrogen using nanosecond
pulsed plasma reactor with different active volumes, voltages and
frequencies,” Appl. Energy, vol. 169, pp. 585-596, 2016
O. Khalifeh, H. Taghvaei, A. Mosallanejad, M. R. Rahimpour, and A.
Shariati, “Extra pure hydrogen production through methane
decomposition using nanosecond puled plasma and Pt-Re catalyst,”
Chem. Eng. J., vol. 294, pp. 132-145, 2016.
W. Sun, M. Uddi, S. H. Won, T. Ombrello, C. Carter, and Y. Ju,
“Kinetic effects of non-equilibrium plasma-assisted methane oxidation
on diffusion flame extinction limits,” Combust. Flame, vol. 159,
pp.221-229, 2012
S. Ahmed, A. Aitani, F. Rahman, A. Al-Dawood, and F. Al-Muhaish,
“Decomposition of hydrocarbons to hydrogen and carbon,” Appl. Catal.
A: Gen., vol. 359, pp.1-24, 2009.
T. Paulmier and L. Fulcheri, “Use of non-thermal plasma for
hydrocarbon reforming,” Chem. Eng. J., vol. 106, pp.59-71, 2005.
N. Rueangjitt, C. Akarawitoo, T. Sreethawong, and S. Chavadej,
“Reforming of CO2-containing natural gas using an AC gliding arc
system: effect of gas components in natural gas,” Plasma Chem. Plasma
Process., vol. 27, pp. 559-576, 2007.