2022 - Flames With Plasmas

JID: PROCI
ARTICLE IN PRESS [mNS;July 19, 2022;18:14]
Available online at www.sciencedirect.com
Proceedings of the Combustion Institute 000 (2022) 1–24

www.elsevier.com/locate/proci
Flames with plasmas

Deanna A. Lacoste a,b
a Mechanical Engineering Program, Physical Science and Engineering Division, King Abdullah University of Science and
Technology (KAUST), Thuwal 23955-6900, Saudi Arabia
b Clean Combustion Research Center, King Abdullah University of Science and Technology, Saudi Arabia
Received 29 January 2022; accepted 25 June 2022

Available online xxx
Abstract
Non-equilibrium plasma discharges can create a favorable environment for combustion processes, at low en-
ergy cost. “How favorable?” and “How low?” are two open questions that can be answered only by gaining
a deep understanding of the coupling between flames and plasmas. This paper reviews recent progress in
plasma-assisted combustion with a special focus on flame dynamics. First, different non-equilibrium plasma
discharges are presented and their effects on flame properties are discussed. Then, an additional level of com-
plexity is introduced by considering the effect of such plasmas on the coupling between flames and acoustic
waves. While initial results suggested that the best plasma actuation would combine thermal and chemical
effects, a shift towards favoring a chemical impact is emerging. Reasons for this shift are explained and future
research directions are finally proposed.
© 2022 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
Keywords: Plasma-assisted combustion; NRP discharges; Thermoacoustic instabilities; Lean blow-off limit;
Non-equilibrium plasma
1. Introduction chemistry, the temperature, or the transport prop-

erties will affect the flame burning rate, structure,
A flame can be defined as a thin exothermic re- and temperature. This is the concept upon which
action front [1]. Its properties depend on the chem- plasma-assisted combustion (PAC) has been devel-
ical composition of the medium, the thermody- oped.
namic conditions, and the flow. By changing any When an electric field is applied to a gas, the
of these, it is possible to modify the combustion charged particles are accelerated. If the electric field
rate by several orders of magnitude. In extreme is strong enough and last for a sufficiently long
cases, the flame can be extinguished, for example time, by collision, the electrons will excite, disso-
by quenching, blowout, or when the lean flamma- ciate, and ionize some of the gas neutral particles,
bility limit is reached, or it can turn into another generating more electrons that will be accelerated
regime of combustion, such as a detonation [2] or by the electric field, etc. Without going into the de-
MILD combustion [3]. In between, changing the tails of the physics involved, this process of elec-
tric field-induced gas ionization is usually called a
breakdown [4]. If after the breakdown a flow of
E-mail address: [email protected] electrons (electric current) is established, then the
https://doi.org/10.1016/j.proci.2022.06.025
1540-7489 © 2022 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
Please cite this article as: D.A. Lacoste, Flames with plasmas, Proceedings of the Combustion Institute, https://doi.org/
10.1016/j.proci.2022.06.025
JID: PROCI
2 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx
entire process is called an electric discharge or a dis- sures (atmospheric and above), due to the fast rate
charge. At ambient conditions, the density of elec- of collisions between particles, thermal equilib-
trons in air is typically of 103 cm−3 . After break- rium is usually reached in a very short time. There-
down, the electron density can reach any value before, the generation of non-equilibrium plasmas re-
tween 109 and 1019 cm−3 . For these densities of elec- quires specific devices. Some of them are presented
trons, one can refer to a plasma of air. in Section 2. For applications in combustion, these
A plasma can be defined as a quasineutral gas plasmas are very interesting because they can cre-
of charged and neutral particles which exhibits col- ate reactive chemical species and heat the gas in a
lective behavior [5]. A system is quasineutral if its controlled manner. Depending on the targeted ef-
characteristic dimension is much larger than the fect, non-equilibrium plasmas can be tuned to pro-
Debye length, λD , which can be expressed as: mote the production of different chemical species
coupled or not with gas heating.
λD = 69(Te /ne )1/2 (m), (1)
Plasmas can affect a flame by three main ways:
where Te is the electron temperature in K and ne is (i) by increasing the gas temperature, (ii) by chang-
the electron density in m−3 . In this definition, “col- ing the chemical composition, and (iii) by changing
lective behavior” means that the motions of parti- the flow. These effects are generally coupled but,
cles depend not only on local conditions (by colli- depending on the electric field applied, it is possi-
sions) but on the state of the plasma in remote re- ble to promote one or the other. For thermal plas-
gions (by electric or magnetic fields). Therefore, it mas, heating of the gas is the dominant effect but,
depends on the electron concentration and energy. as a consequence of a strong and fast increase in
With this definition, ambient air, which is a weakly temperature, turbulence as well as acoustic waves
ionized gas, is not a plasma, while an hydrocarbon- can be generated [12,13]. With non-equilibrium
air premixed flame at atmospheric pressure, with plasmas, the chemical effect can be promoted. Ex-
a typical electron density ne = 1010 cm−3 [6] and amples of chemical species directly produced by
an electron temperature Te = 2200 K, is a plasma electron impact in hydrocarbon-air discharges are
[7]. Thus, flames can be directly affected by elec- atomic oxygen, O, atomic hydrogen, H, excited oxy-
tric fields. For example, by applying an AC electric gen, O∗2 , excited nitrogen, N∗2 , etc. [14]. The way to
field to a flame, it is possible to change its shape by control the production of these species will be in-
ionic wind [8,9] or to reduce the soot formation [10]. troduced in Section 2. These species are extremely
Study of the electrical properties of flames, as well reactive and they can accelerate the combustion
as the flame response to external electric fields are processes. For a stronger acceleration, the chemical
interesting parts of the PAC research. Nevertheless, effect can be coupled with a thermal effect, and/or
for concision purpose, only interactions between an increase in mixing and transport, e.g., by ionic
flames and plasmas produced by electric discharges wind, turbulence, or shock waves.
will be discussed in this paper. Flames with plasmas have been studied with
Electric discharges turn gas into plasma [4]. different objectives in mind: from improving igni-
Depending on the conditions, this plasma can be tion to anchoring turbulent flames, from increas-
in thermal equilibrium (all particles follow Boltz- ing burning rates to extending lean flammability
mann distributions at the same temperature), or in limits. These studies have in common a focus on
thermal non-equilibrium (different particles have the plasma efficiency. The plasma energy or power
different temperatures). Starting from a stoichio- should be as low as possible compared to the ex-
metric mixture of air and fuel at standard condi- pected gain. For example, in a conventional spark
tions, the typical temperature of a thermal plasma ignition engine for road transportation, the energy
is 6000–10,000 K. This is three to five times higher of a spark discharge is of the order of 100 mJ,
than the temperature of a flame burning the same which corresponds to roughly 10−4 % of the ther-
mixture. In addition, the electrical energy necessary mal energy released by combustion of the fuel in
to sustain such a plasma is high and the amount of the cylinder. This is negligible and trying to de-
thermal nitric oxide, NO, produced is also high. For crease it further is pointless. Therefore, research
these reasons, thermal plasmas have strong limita- in plasma-assisted ignition has focused on condi-
tions for applications in combustion, even though tions for which ignition is difficult, e.g., for very lean
they have been used for more than a century as or highly diluted mixtures. Then, the gain will be
ignition sources in engines [11]. Comparisons be- in fuel saving thanks to leaner conditions and/or
tween conventional ignition sources and ignition avoidance of misfire ignition. For combustion sys-
systems based on non-equilibrium plasmas will be tems with a permanent flame such as boilers or
discussed in Section 3.1. gas turbine engines, the plasma power should be
Non-equilibrium plasmas, also called non- as low as possible compared to the flame thermal
thermal plasmas, produced by electric discharges power. If the focus is on extension of the lean blow-
have usually high-temperature electrons (e.g., Te = off limit, the plasma power should be compared to
50, 000 K), while the neutral particles (molecules the power equivalent of the fuel being saved. Re-
and atoms) are at a much lower temperature, typ- cent PAC studies in these fields are discussed in
ically between 300 and 4000 K. At elevated pres- Section 3.
10.1016/j.proci.2022.06.025
JID: PROCI
D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 3
Finally, the coupling of flames and plasmas electron-impact reaction rates both strongly de-
have been considered as a potential solution to pend on E/N. The reduced electric field controls
mitigate thermoacoustic instabilities in combustion the modes into which the electron energy is dissi-
systems. Thermoacoustic instabilities can happen pated [30]. For example, Fig. 1 presents the frac-
when the fluctuations of heat release rate from a tion of dissipated energy by electrons during a dis-
flame couple with an acoustic mode of the com- charge in air [28]. At low reduced electric field,
bustion system [15]. These instabilities can cause E/N < 5 Td, the main fraction of dissipated en-
strong thermal and pressure fluctuations that can ergy per electron goes into vibrational excitation
threaten the integrity of the engine [16]. Control of oxygen. For 5 Td < E/N < 100 Td vibrational
of thermoacoustic instabilities have been inten- excitation of nitrogen is the main pathway, while
sively studied, using both passive and active ap- at E/N > 100 Td, electronic excitation of nitrogen
proaches [17,18]. Recently, active approaches us- dominates. Thus, the chemistry induced by the dis-
ing non-equilibrium plasma discharges to control charge can be tuned by controlling the reduced elec-
flame dynamics have been successfully attempted tric field. While the gas density depends on the ther-
[19–27]. Section 4 presents recent findings in this modynamic conditions of the gas, the electric field
field and propose a discussion to get closer to real depends on the applied voltage and the geometry
engine conditions. and gap distance of the electrodes.
The main objectives of this paper are the follow- Finally, the last important parameter to con-
ing: sider for a source of non-equilibrium plasma is
the repetition rate. Except for single shot applica-
• If you are a PAC curious or beginner, to help tions such as ignition, PAC relies on long-duration
you understand the important points of this non-equilibrium plasmas. The repetition rate re-
research (Sections 1–3). quired for a given application depends on the ther-
• If you already have experience in PAC, to modynamic and flow conditions. Indeed, to main-
present the current state of the art and some tain a sufficient average level of chemical species
of the remaining open questions (Sections 3 and/or gas temperature in the discharge, the re-
and 4). combination rates as well as the residence time
• If you are an expert in PAC, to challenge in the inter-electrode area must be taken into
some of your ideas and to encourage you to account.
try to prove me wrong (Sections 3.3, 4.2, and
4.3). 2.1. Dielectric barrier discharges (DBD)
Dielectric barrier discharges (DBD) are pro-

2. Non-equilibrium plasma sources duced by applying en electric field, generally AC
or pulsed, between two electrodes separated by a
Different sources can be used to produce non- gas and one or more dielectric layers [31]. At at-
equilibrium plasmas by electric discharges at atmo- mospheric pressure, if the electric field is strong
spheric pressure and above. Some of the most com- enough, breakdown occurs in the gas, and an ion-
monly used in PAC studies are presented in this sec- ized channel joining one electrode and the dielectric
tion. surface forms. Electrons flowing in the discharge
As briefly discussed in the introduction, an elec- accumulate on the dielectric surface, building up
trical discharge can be decomposed in two phases, a potential that finally suppress the local electric
the breakdown and the subsequent flow of elec- field. Thus, the discharge stops before the plasma
trons in the ionized channel. If nothing limits the could reach thermodynamic equilibrium. Multiple
electric current after breakdown, the plasma will discharges are produced in parallel, at other loca-
rapidly turn into a thermal plasma. To avoid tran- tions of the electrodes/dielectric surface, and DBD
sition to a thermal plasma, there are two simple so- discharges at elevated pressures are generally in-
lutions. The first one is to stop the discharge shortly homogeneous. When the electric field is reversed
after the breakdown and the second option is to or stopped, electrons accumulated on the dielectric
control the flow of electrons, i.e., the current. Thus, surface are removed by the reverse electric field or
applying short duration electric fields, i.e., typically by recombination. In some cases, they may con-
shorter than 100 ns, or using devices to avoid strong tribute to the breakdown of the next half period.
currents are the two main ways of producing non- Fig. 2 presents examples of DBD geometries used
equilibrium plasmas by electric discharges. For ap- in PAC as well as an example of measured voltage
plications in PAC, other parameters must however and current, obtained in a DBD for an AC applied
be considered. voltage of 7 kV at 4 kHz.
In weakly ionized plasma chemistry, i.e., for At atmospheric pressure, DBD are a com-
ne << N , where N is the gas number density, the mon and cheap way of producing non-equilibrium
reduced electric field, E/N, where E is the elec- plasma with a negligible thermal impact. They
tric field, and the electron density, ne , are the have been used for more than a century to gener-
two key parameters [29]. Electron temperature and ate ozone from air. Different types of high-voltage
10.1016/j.proci.2022.06.025
JID: PROCI
Fig. 1. Fraction of dissipated energy by electrons for a discharge in air as a function of the reduced electric field, E/N. tr
+ rot, sum of translational and rotational excitation; v, vibrational excitation; el, electronic excitation; ion, ionization; dis,
dissociation; a + b, sum of electronic excitation of singlet ‘a’ and ‘b’ states of molecular oxygen. Data from [28].
Fig. 2. (a) Example of DBD configuration used for plasma-assisted ignition in [32]. (b) Example of DBD configuration
used for swirl flame stabilization in [33]. (c) Example of current and voltage signals measured during swirl flame stabiliza-
tion, taken from [33].
sources can be utilized, from the most basic AC DBD produced by AC or DC pulsed high-voltage,
or pulsed DC systems, to nanosecond repetitively in addition to the chemical actuation of the gas,
pulsed ones. For applications to combustion, the transport effects can also be observed, from the
typical range of frequency used is from 1 to 20 kHz, so-called ionic wind [34]. This effect can be sup-
but lower or higher frequencies can also work. For pressed by using nanosecond pulsed high-voltage
10.1016/j.proci.2022.06.025
JID: PROCI
sources. In this case, the acronym ns-DBD is some- ers, i.e., tens of kilowatts for a single pair of elec-
times used. Another advantage of ns-DBD is that trodes. They produce a mix between equilibrium
the reduced electric field at which the discharge oc- and non-equilibrium plasmas, combining strong
curs can be controlled, while in traditional AC- gas heating with transport effects due to gas expan-
voltage DBD, discharges can happen at various sion and shock/acoustic waves, with chemical actu-
E/N, therefore the chemical production is less con- ation of the gas. For all these reasons, they are a
trolled. Fig. 2 c shows an example of AC-voltage very interesting source of non-equilibrium plasma
DBD where discharges, corresponding to the cur- for combustion applications. However, they must
rent peaks, happen at voltages from 2 to 10 kV [33]. be set aside for applications where a thermal effect
These multiple discharges happening during a cy- is beneficial.
cle are one of the interesting features of DBD. In-
deed, if large electrodes can be used, the plasma
volume produced by the multiple discharges can 2.3. Microwave discharges
be relatively large. Finally, the best advantage of
DBD is also their main limitation: the thermal ef- Discharges produced by high-frequency electro-
fect is minimal. For combustion applications where magnetic fields are referred to as microwave dis-
an increase in gas temperature is desirable, DBD charges [38]. Electrons in the gas are accelerated by
might not be suitable sources of non-equilibrium the microwave electric field, typically at a frequency
plasmas. of 2.45 GHz, and collide with the neutral parti-
cles. At the beginning of the applied field, these
2.2. Gliding arcs collisions are elastic and they result in an increase
of the gas temperature. After several collisions, the
A gliding arc occurs when a plasma discharge electrons get enough energy to produce ionization.
is generated between two or more diverging elec- The gas is then partially ionized and becomes a
trodes placed in a fast gas flow [35]. When a strong plasma. In a microwave discharge, only the electron
enough electric field is applied, breakdown hap- are accelerated by the electric field, as ions are too
pens at the shortest distance between the electrodes. heavy to follow the oscillating electric field. The mi-
Rapidly, e.g., within 1 μs in ambient air, the dis- crowave energy transmitted to the electrons has to
charge turns into an arc, i.e., a thermal plasma with be high enough to initiate the plasma. To concen-
a temperature controlled by the maximum power trate this energy, different methods have been de-
of the power supply (typically in the range 6000– veloped.
10,000 K). This arc is placed in a flow and it is A resonant cavity [39] can be used to generate
convected along the diverging electrodes at a speed a standing wave that is amplified by the waves pro-
that is usually comparable to the bulk flow veloc- duced by the microwave source. The resonance fre-
ity. During this movement, the length of the dis- quency of the cavity strongly depends on its geom-
charge and hence its dissipated power increase up etry and has to match the frequency of the source.
to the moment when the discharge power reaches While resonant cavity microwave sources are mini-
the maximum value available from the power sup- mizing the losses, this geometrical constraint is one
ply source. When the distance between the elec- of the limitation for their application in PAC. A sec-
trodes reaches a critical value, heat losses from the ond method is to work with a wave guide-to-coaxial
plasma begin to exceed the energy supplied by the line transition [40]. The microwave plasma is pro-
source and it is not possible to sustain the plasma duced on top of a nozzle. It has the shape of a con-
in the state of thermodynamic equilibrium. Conse- verging cone. It is sometimes called a free expand-
quently, the translational temperature of the neu- ing torch. Several designs of microwave plasma
tral molecules decreases while the plasma conduc- torch are available [38]. In PAC, this technique is in-
tivity is maintained by a high value of the elec- teresting because it is electrode-less. A third method
tron temperature. This new mode of the gliding consists of a planar microstrip line on a dielectric
arc is non-equilibrium. When the gap between the substrate and a copper ground plate. Surface waves
electrodes becomes too large, there is a decay of propagate in a gas channel along the interface be-
the non-equilibrium discharge, followed by a new tween the plasma and the dielectric surface [41].
breakdown at the shortest distance between the This technique produces low-power microplasmas.
electrodes and the cycle repeats. Fig. 3 presents ex- Microwave discharges have been studied since
amples of gliding arc electrode configurations used the 1970’s. They are well understood but the physics
in flame stabilization studies [36,37], along with an involved is more complicated than for other types
example of voltage and current signals measured of non-equilibrium plasma discharges. In addi-
in a gliding arc in air, powered by an AC voltage at tion, their thermal effect is usually significant, even
12.5 kHz [37]. though pulsed microwave systems can be used to
Gliding arcs have been studied since the control it [42,43]. On the other hand, they have ad-
early1990s. They operate at elevated pressures and vantages such as the ability to impact large volumes
they are easy and cheap to produce with DC or AC [44,45], a low cost, or electrode-less configurations
high voltage sources. They can dissipate large pow- [46], that make them interesting sources for PAC.
10.1016/j.proci.2022.06.025
JID: PROCI
Fig. 3. (a) Example of gliding arc electrode configuration used for jet flame stabilization in [36]. (b) Example of gliding
arc electrode configuration used for swirl flame stabilization in [37]. (c) Example of current and voltage signals measured
in air for an AC gliding arc at 12.5 kHz, taken from [37].
2.4. Nanosecond repetitively pulsed (NRP) sixth pulse, high current peaks and low voltage volt-
discharges age peaks, corresponding to a steady state regime,
can be observed. In this example, the building-up
Nanosecond repetitively pulsed (NRP) dis- process took five pulses before the synergistic ef-
charges are produced by applying nanosecond fect between pulses reached an almost steady state
high-voltage pulses at high repetition rate, typically regime.
between 5 and 200 kHz [47]. The voltage pulses NRP discharges have been studied since the be-
have a duration between 2 and 100 ns and a very ginning of the2000s. In air, their chemistry is rel-
sharp rising slope, between 1 and 10 ns. These fea- atively well known [49–51]. During the pulse, the
tures constrain the breakdown to occur at elec- main part of the electron energy goes into (1) dis-
tric fields much higher than for continuous applied sociation of oxygen, (2) excitation of the electronic
voltages. After breakdown, a strong current con- states of nitrogen, N2 (A, B, C, a’...), and (3) exci-
trolled by the power supply flows in the ionized tation of the vibrational states of nitrogen in its
channel. The electric field drops and the tempera- ground electronic state, N2 (X). The distribution be-
ture of the neutral particles increases very rapidly. tween these different pathways depends on the re-
Nonetheless, the discharge is stopped long before duced electric field (see Fig. 1). During and after
the plasma could reach thermodynamic equilib- a pulse, electronically excited nitrogen reacts with
rium. Afterward, the charged and excited parti- molecular oxygen very rapidly, i.e., within a few
cles undergo a post discharge chemistry. Before the tens of nanoseconds. This process is exothermic,
state of equilibrium is reached, i.e., charged par- therefore leading to a fast heating of the gas. This
ticles, excited particles, atoms, etc. remain at rela- model, first proposed by Popov in 2001 [49], and
tively high concentration in the interelectrode area, verified experimentally [50], is known as the two-
a new pulse is applied. This synergistic effect be- step mechanism of atomic oxygen production and
tween pulses is one of the key features of NRP gas heating:
discharges because it allows to maintain a high N2 (X ) + e− −→ N2 (A, B, C, a . . . ) + e− (2)
concentration of active species at low energy cost.
Fig. 4 presents an example of a train of 10 NRP
discharges in air [48]. During the first five pulses the N2 (A, B, C, a . . . ) + O2 −→ N2 (X ) + 2O + heat
current is low and the voltage is high, while after the
(3)
10.1016/j.proci.2022.06.025
JID: PROCI
Fig. 4. Examples of voltage and current measured for a train of 10 NRP discharges at 30 kHz in air. Taken from [48].
After a pulse, the vibrational levels of nitrogen sion and shock/acoustic waves [56,57]. Finally, for
molecules eventually reach equilibrium with the a fast repetition rate and a low applied voltage,
translational mode, slowly releasing heat [52]. In the reduced electric field can be low enough to fa-
parallel, O atoms start to react and, depending on vor vibrational excitation of nitrogen and a slow
the conditions, produce O3 , N2 O, NO...More de- gas heating. Other interesting features of NRP dis-
tails on fast gas heating and chemistry of NRP discharges are: they are easy to synchronize with opti-
charges in air and fuel-air mixtures can be found in cal diagnostics, the electrode geometry can be com-
a recent review paper by Popov and Starikovskaia pletely arbitrary, and the discharge power can be as
[51]. high as a few kilowatts. One of the limitations of
For combustion applications, NRP discharges this source of non-equilibrium plasma is the cost of
are extremely interesting because by playing with the power supply that is usually significantly higher
the reduced electric field, pulse duration, and rep- than AC or DC power supplies for DBD and glid-
etition frequency, the electron energy pathway can ing arcs.
be controlled. For example, for low applied voltages
and relatively low pulse repetition frequencies (5–
30 kHz), the glow regime of NRP discharges [53] is 2.5. Location of the plasma actuator
promoted, with mainly a chemical effect. When
the applied voltage is further increased, the spark Once a non-equilibrium plasma source has been
regime of NRP discharges [54,55] can be obtained chosen, an important decision is where to install the
with a strong fast gas heating, inducing gas expan- plasma actuator in the combustion system. There
10.1016/j.proci.2022.06.025
JID: PROCI
Fig. 5. Examples of location of the non-equilibrium plasma discharges for stabilization of a premixed jet flame over a bluff-
body. a) Discharges located in the recirculation zone, repetition rate 10 kHz. b) Discharges located in the jet, repetition
rate 50 kHz.
are several possible options and the best one de- was chosen in [20,64], with relatively low discharge
pends on both the application and the system. frequencies (e.g., 10 kHz). Another possible option
For ignition, an obvious guideline is to install is to install the plasma actuator on the side of the
the plasma actuator in a zone where the fuel-air bluff-body, to enhance the reactivity of the flow of
mixture is in the flammability range. In addition, fresh mixture and increase the burning velocity of
the flow properties as well as the presence of walls the flame (Fig. 5b). This strategy requires a faster
have to be considered. The ignition kernel should repetition rate (typically 50 kHz or higher), because
have enough time to develop and should not loose the flow velocity is much higher than in the recircu-
too much heat to a surface or by stretching [58–60]. lation zone. Thus, the best option would depend on
On the other hand, flow and turbulence can also the capabilities of the plasma source.
enhance the mixing and increase the ignition ker- Only a few studies have compared the effect of
nel wrinkling, helping its fast development [61,62]. the plasma location on the expected result (see for
In industrial systems, a good starting point is to example [69]). More work is required to optimize
choose the same location than the traditional spark this parameter for various configurations. Having a
ignition system because it usually has been opti- systematic approach, maybe coupled with machine
mized. In laboratory setups or new designs, a good learning tools, is necessary to further improve the
knowledge of the flow field and gas composition is effectiveness of plasma actuators.
necessary to select the best ignition location [63].
Finally, other constraints such as optical access
might also be important to consider [43].
For flame actuation, e.g., increase of the burning 3. Flame enhancement
velocity, decrease of the lean blow-off limit, control
of thermoacoustic instabilities, etc., several strate- Flame enhancement by non-equilibrium
gies have proven to be efficient. From localized dis- plasma, also called plasma-assisted combus-
charges in the burned gases [27,64], in the fresh tion (PAC), has been studied for a few decades
gases [21], or across the flame [65,66], to distributed and showed very interesting results. Several re-
discharges in the fresh mixture [36,67], in the fuel view papers [14,70–72] have summarized both
stream [68] or in the flame front [9], flame actua- the achievements and the mechanisms of plasma
tion by non-equilibrium plasma has been achieved. action. In this section, some of the key results on
Here again, a good knowledge of the flow field and ignition, burning velocity, and lean blow-off and
gas composition is important but it is not enough. stability limits are summarized. A short section on
Indeed, it is crucial to understand the flame dynam- modeling and simulations is also proposed.
ics or stabilization mechanisms.
For example, consider a premixed jet flame sta- 3.1. Ignition
bilized over a bluff-body. If the purpose of the
plasma actuation is to anchored the flame over the Studies of ignition by non-thermal plasma dis-
bluff-body while decreasing the burning velocity charges can be grouped in three large families,
(by changing the temperature or the equivalence from the most fundamentals, using shock tubes
ratio), or increasing the flow speed, then, a good and rapid compression machines (RCM), to the
strategy is to reinforce the stabilization effect of most applied using piston engines or scramjets. In
the bluff-body by applying localized discharges in between, constant volume combustion chambers
its recirculation zone, i.e., in the slow burned gases (CVCC) and continuous-flow burners have been
on top of it (Fig. 5a). For example, this location used to compare various ignition sources.
10.1016/j.proci.2022.06.025
JID: PROCI
Fig. 6. Measured and calculated ignition delay times in a mixture of C2 H6 :O2 :Ar=2:7:81 as a function of 1000/T . Shock
tube facility equipped with a volumetric nanosecond plasma system. Taken from [73].
To determine the plasma effect on ignition de- of ignition mechanisms by non-equilibrium plasma
lay time, shock tubes and rapid compression ma- discharges including all potential effects (chemi-
chines have been widely utilized. By comparing the cal, thermal, and transport). These systems are the
ignition delay times obtained with and without dis- most commonly used in plasma-assisted ignition
charges, it is possible to quantify the plasma effect research and a large range of conditions has al-
on the combustion chemistry. Coupled with chem- ready been covered. Different initial pressures, flow
ical modeling and simulations, these studies helped conditions, fuels, mixture composition, and plasma
in understanding the key chemical mechanisms of sources have been considered [13,48,71,72,74]. Typ-
plasma action on ignition. For example, Kosarev ically, plasma sources are compared with con-
et al. [73] studied in a shock tube the effect of a ventional ignition systems, which produce ther-
volumetric nanosecond plasma discharge on the mal plasmas. Lean or diluted flammability limits
ignition delay time of ethane-oxygen-argon mix- and propagation speed of the ignition kernel are
tures (2:7:81) at temperatures from 1200 to 1500 K. two common criteria of comparison. Additionally,
Fig. 6 presents a comparison between their experi- spectroscopic measurements and visualization can
mental and modeling results. Two main conclusions be performed and provide information about the
can be drawn from this figure. First, over the en- dominant plasma effect. However, a precise un-
tire range of temperatures, the use of nanosecond derstanding of the plasma action is usually not
volumetric discharges decreased by more than an achieved because, in these experiments, it is diffi-
order of magnitude the ignition delay time (empty cult to decouple the different effects. The vast ma-
black squares for autoignition delay time and solid jority of these studies can be regarded as proof-of-
black stars for ignition delay time with plasma). concepts, showing that for a similar energy depo-
Second, they obtained a very good agreement be- sition, non-equilibrium plasmas are more efficient
tween experimental results and simulations consid- ignition sources than conventional ignition systems
ering dissociation of oxygen coupled with gas heat- producing thermal plasmas. For example, in Xu
ing (solid black stars and green symbols). Shock et al. [75], ignition of lean propane-air mixtures by
tubes and RCM are used in PAC research to vali- NRP discharges was compared to a conventional
date chemical models but also to compare different ignition system, in a CVCC at initial pressures up
non-equilibrium plasma strategies. Unfortunately, to 10 bar (see Fig. 7). For similar energy deposition
the number of plasma-assisted ignition studies fo- (about 55 mJ) and ignition duration (about 3 ms),
cusing on ignition delay time is low and only a few the flame developed faster in the case of NRP ig-
fuels and plasma sources have been investigated. nition, compared to conventional ignition. These
More research is needed to validate chemical mod- results were attributed to an increase in the flame
els of PAC and to better understand the chemical wrinkling, i.e., to local changes in the flow proper-
effect of non-equilibrium plasmas on the combus- ties.
tion of fuels with large carbon content. Finally, a few studies of plasma-assisted ignition
The second family of experiments used for ig- by non-equilibrium plasmas in real engine condi-
nition studies comprises CVCC and continuous- tions [76–79], or in scramjets [60,80,81] have been
flow burners. These setups allow the investigation reported. Here, the challenges come from pressures
10.1016/j.proci.2022.06.025
JID: PROCI
Fig. 7. Schlieren images of lean propane-air flames at 10 bar ignited by a conventional car ignition system (57 mJ; 16 W;
3.5 ms) and 82 NRP discharges at 30 kHz (55 mJ; 22 W; 2.7 ms). Taken from [75].
and turbulence levels higher than in lab scale ex- This process can remove a part of the energy from
perimental setups. In addition, implementation of the ignition area, reducing the ignition efficiency of
non-equilibrium plasma sources and diagnostics the discharges. In addition, for multiple discharges
is also more challenging. These papers confirmed such as in bursts of NRP discharges, the generated
that, under engine conditions as well, compared to flow becomes the dominant process controlling ig-
thermal plasmas, plasma sources combining chem- nition, i.e., favoring or being detrimental to the de-
ical, thermal, and transport effects could improve velopment of the ignition kernel [13,83,84].
ignition.
For ignition of fuel-air mixtures, three effects
of non-equilibrium plasma discharges have proven 3.2. Burning velocity
to be important: (i) the decrease of ignition de-
lay time due to the production of active chemical While successful plasma-assisted ignition is rel-
species, (ii) the acceleration of chain branching and atively easy to achieve, enhancing the burning ve-
chain prolongation reactions due to fast tempera- locity by non-thermal plasma is more challenging.
ture increase, and (iii) the hydrodynamic perturba- To accelerate the burning velocity of a premixed
tions and resulting mixing, induced by gas heating. flame three options are possible. First, by increas-
This last effect is not always present and not always ing the reactivity and/or the temperature of the gas
beneficial for ignition. For example, in Castela et al. mixture, the laminar burning velocity can be in-
[82], a simplified model of the discharges for igni- creased. Second, by changing the flow properties or
tion of turbulent flows is proposed. This model dis- introducing inhomogeneities (temperature or mix-
tinguishes two main electron pathways: (1) the ex- ture composition), an increase of the flame sur-
citation and the subsequent relaxation of the elec- face area can be achieved. Finally, the overall burn-
tronic states of nitrogen, leading to a fast gas heating velocity of a flame can be increased if the oc-
ing associated with molecular oxygen dissociation currence of flame local extinction (e.g., by stretch
within the discharge characteristic time; and (2) the or curvature effects), is decreased. Non-equilibrium
excitation and relaxation of vibrational states of ni- plasma can be efficiently used for each of these op-
trogen, causing a much slower gas heating. In this tions [85–89]. However, in many configurations at
model, a fraction α of the electron energy goes into atmospheric pressure and above, the flame volume
dissociation of oxygen, a fraction β goes into fast is much larger than the discharge volume. A local
gas heating, and a fraction γ goes into slow gas plasma can struggle in accelerating a large flame in
heating, with α + β + γ = 100 % of the energy of such a way that, globally, the burning velocity is ac-
the discharge. They showed that if β is too large, celerated. Anyway, solutions exist and it is some-
a strong gas expansion associated with the gen- times possible to acceleration large flames with lo-
eration of shock and/or acoustic waves happens. calized discharges. Here, two examples are detailed.
10.1016/j.proci.2022.06.025
JID: PROCI
Fig. 8. Distance between the burner plate and the center of gravity of the flame, ZCOG , as a function of the plasma to
flame power ratio, Pratio, at 3 bar, for different applied voltages. Adapted from [90].
For configurations such as bluff-body stabilized could be easily accelerated with NRP discharges
turbulent flames or turbulent swirl flames, non- produced in the middle of the tube [91], while flame
equilibrium plasma discharges applied close to the acceleration with NRP discharges close to the tube
stabilization area can decrease the flame height and wall was more difficult to achieve [69]. Indeed, for
volume, i.e., decrease the overall burning velocity. discharges located in the boundary layer, only local
For example, Di Sabatino et al. [90], showed that ignition events and ignition kernel interactions with
the height of the flame’s center of gravity, ZCOG the leading shock upstream of the main flame could
(based on imaging of the flame OH∗ chemilumi- significantly accelerate the flame burning velocity.
nescence, see details in [90]), of a swirl-stabilized This result is interesting because it illustrates that
methane-air flame could be decreased by up to 25% the mechanisms of plasma action for flame accel-
by applying NRP discharges with a plasma power eration can be quite complicated, especially in envi-
of about 1.2% of the flame thermal power. Fig. 8 ronments where plasma, flame, and acoustic/shock
compares the ZCOG for the reference flame without waves interact. These interactions will be further
plasma (dash line), at 3 bar, with the ZCOG of the discussed in Section 4.
same flame subjected to NRP discharges, for ap-
plied voltages ranging from 12 to 16 kV, and for ra- 3.3. Lean stability and blow-off limits
tios of plasma power to flame thermal power from
0.25 to 2%. As the bottom part of the flame was not Studies showing that non-equilibrium plasmas
noticeably shifted inside the injection tube, these re- can extend the lean stability limit and/or the lean
sults illustrate that with NRP discharges, the flame blow-off limit of a flame are among the most com-
was more compact. In addition, as the flow and gas mon in PAC. The general idea is to install a plasma
mixture composition were kept constant, a more source in a burner (typically swirl stabilized, bluff-
compact flame was then the consequence of a faster body stabilized, or turbulent jet flames), or in the
burning rate. However, reasons for this effect were cavity of a scramjet, and to compare the lean sta-
not discussed. bility and/or blow-off limits of the flame with and
Another example of burning velocity acceler- without plasma, for various bulk flows. Different
ation by non-equilibrium plasma discharges can plasma actuation strategies have proven to be suc-
be found in Gray and Lacoste [69,91]. In these cessful. The three most popular are (a) with the dis-
studies, the objective was to accelerate a turbulent charges localized in the fresh gases [21], (b) with
stoichiometric hydrogen-air flame propagating in the discharges localized in the burned gases [64],
a tube until the onset of a detonation, i.e., until and (c) with the discharges distributed in the fresh
a deflagration-to-detonation transition (DDT) oc- gases [67]. Various plasma sources have been uti-
curred. Non-equilibrium plasma produced by NRP lized, for example NRP discharges [64], DBD [33],
discharges were used in burst mode (less than 100 or gliding arcs [36,60,92]. Extension of the lean
discharges) and phased with the arrival of the tur- stability and blow-off limits have been observed
bulent flame in the interelectrode area. The flame for a large variety of conditions, including differ-
ent fuels (methane, propane, kerosene, etc.), pres-
10.1016/j.proci.2022.06.025
JID: PROCI
sures [25,93], and supersonic flows [60,80,81,94]. quency of 30 kHz, could save up to 650 W of ther-
Except in some specific cases, noticeable effect of mal power, while maintaining a 16-kW flame. For
the plasma was observed for a plasma power of the same conditions, NRP spark discharges could
about 1% of the flame thermal power. save 600 W only. The authors explained that the less
When the purpose of the plasma actuation is to efficient effect of the NRP spark discharges at ele-
extend the lean stability or blow-off limit, the repe- vated pressures could be due to the generation of
tition rate of the discharges has to be adapted to the strong shock waves close to the flame stabilization
flow rate (see Section 2.5). The residence time in the area [90]. However, they did not explain why NRP
discharge area should be long enough for the gas to glow discharges became more efficient with increas-
be properly ”treated” by the plasma. For example, ing the pressure.
for discharges localized in the recirculation zone of Based on the model developed by Castela et al.
a bluff-body [20,64], i.e., in the burned gases, the [82] (see Section 3.1), and the effect of pressure on
velocity of the flow is usually low and the minimal the reduced electric field, an explanation can be
repetition rate at which the discharges had a strong proposed. For identical applied voltage and elec-
effect was 10 kHz. Higher repetition rates (up to trodes, an increase in pressure increases the gas
80 kHz), for a similar plasma power, did not im- density, N, and decreases the reduced electric field,
prove further the plasma effect. For discharges lo- E/N. As showed in Fig. 1, a decrease in E/N cor-
calized or distributed in the fresh gases, this resi- responds to a shift of the dissipated energy by elec-
dence time matching is even more important. For trons from electronic excitation to vibrational exci-
example in Di Sabatino et al. [90], it was found that tation of nitrogen. According to Castela’s model,
for high pulse repetition frequencies the plasma ef- the energy of the NRP glow discharges can be
fect on the flame stabilization was the strongest. split into two parts: electronic excitation of nitro-
However, the flow disturbances produced by the gen, responsible for chemical dissociation and fast
discharges were also mentioned [90,95] and, simi- gas heating, and vibrational excitation of nitro-
larly to what was obtained in plasma-assisted igni- gen, responsible for slow gas heating. Increasing the
tion studies, these could be detrimental to the flame pressure will then favor the slow gas heating. At
stabilization. pressures above ambient, this slow heating due to
In general, studies on the extension of the lean vibrational-translational relaxation can happen in
stability and blow-off limits by non-equilibrium tens of microseconds [82]. With a discharge repeti-
plasma discharges mainly consider strong dis- tion rate of a few tens of kilohertz, this relaxation
charges, i.e., discharges with a strong fast gas process can then be responsible for an almost con-
heating, that could be considered as local igni- stant gas heating in the interelectrode area. Thus,
tion sources [20,36,60,64,92,95,96]. With these dis- the main part of the discharge energy is deposited
charges, it is easy to reach a ratio of plasma power close to the flame stabilization area. On the other
to flame thermal power of about 1%, and therefore hand, at lower pressure, a larger fraction of the
to obtain a strong effect on the flame. However, one discharge energy goes into fast gas heating (higher
of the strong limitations of these non-equilibrium E/N), generating gas expansion and transport ef-
discharges is the production of large quantities of fects that can be detrimental to the local plasma
nitric oxide, NO [33,67]. On the same setups, it actuation, as observed in [83]. Temperature mea-
would be possible to test weaker discharges, such surements as well as numerical simulations applied
as NRP glow discharges [53] or DBD, but if the to this configuration would be valuable tools to val-
plasma power is too low, the effect on combustion idate this explanation.
is marginal. However, recently a few studies showed In Kim et al. [33], the lean blow-off limit of
a strong enhancement of the lean blow-off limit of a methane-air swirl flame was extended by using
methane-air swirl flames by discharges with a low DBD generated by an AC power supply, with an
thermal effect [33,93]. applied voltage up to 7 kV and a frequency up to
Di Sabatino and Lacoste [93] showed that at 4 kHz (see Fig. 10b). This study was performed at
pressures higher than 2 bar, NRP glow discharges ambient pressure. For the entire range of bulk flow
(with a low thermal effect) were more efficient than velocities studied (from 4 to 10 m/s), non-thermal
NRP spark discharges (with a strong thermal ef- plasma produced by DBD could extend the lean
fect) in extending the lean stability and blow-off blow-off limit. This effect was attributed to a chem-
limits of methane-air swirl flames, as shown in ical effect with the production of ozone, O3 , by the
Fig. 9. The ratio of plasma power to flame ther- low temperature discharges. The plasma effect was
mal power was kept constant for all pressures at compared with the effect of ozone addition, and
0.32% for NRP glow discharges (NRP Glows in they found that the DBD had a similar effect than
Figs. 9 and 10a) and 0.7% for NRP spark dis- an addition of 10,000 ppm of O3 on the extension
charges (NRP Sparks in Figs. 9 and 10a). For ex- of the lean blow-off limit. However, a small ther-
ample, they showed that at 4 bar (Fig. 10a), the mal effect was not completely ruled out, as the con-
use of NRP glow discharges obtained with an ap- centration of NO in the burned gases increased by
plied voltage of 13 kV and a pulse repetition fre- up to 20 ppm when the DBD were used.
10.1016/j.proci.2022.06.025
JID: PROCI
Fig. 9. Stability and lean blow-off limits of a methane-air swirl flame with and without NRP discharges as a function of
pressure. Error bars represent standard deviation in reproducibility of results. Taken from [93].
Fig. 10. a) Direct visualization of methane-air swirl flames at 4 bar, without plasma, with NRP glow discharges, and with
NRP spark discharges. Adapted from [93]. b) Direct visualization of methane-air swirl flames at atmospheric pressure
without plasma and with DBD. Adapted from [33].
Thus, the use of discharges favoring either a 3.4. Modeling of flames with plasmas
chemical effect or a slow gas heating might be
the future of the lean blow-off limit extension by Modeling and simulations of PAC are less
plasma. In order to have the maximal effect on common than experimental studies but they have
the flame, with a low plasma power and low NO played a major role for the understanding of the
emissions, additional efforts on the optimization of behavior of flames with plasmas. These studies can
the plasma location and an increase of the plasma be grouped into two subsets (1) the development of
volume and the repetition rate of the discharges chemical kinetic models, and (2) the development
are necessary. These parameters are conditions de- of computational fluid dynamics (CFD) codes in-
pendant, therefore, ideally, these studies should be cluding plasma models.
done in configurations close to the targeted appli- There are two challenges in the development of
cations. chemical kinetic models for plasma-assisted com-
10.1016/j.proci.2022.06.025
JID: PROCI
Fig. 11. Comparisons of the temporal evolution of the atomic oxygen mole fraction in air (black) and methane-air (blue)
following a 100-ns discharge. Experimental data (symbols) from [104] and simulations from [104] (dotted lines) and from
[103] (solid and dashed lines). Taken from [103].
bustion problems. First, the plasma temporal scale The 0-D simulations were performed with a plasma
is at least three orders or magnitude smaller than model including 71 species and 680 reactions [103].
the combustion temporal scale. Second, while the Note that, in some PAC configurations (see for
reaction rates of combustion are Arrhenius’ func- example [27,64]), the discharges are generated in
tions of the gas temperature, plasma reaction rates the burned gases. Chemical kinetic models of plas-
are usually functions of the reduced electric-field. mas of nitrogen, oxygen, water, and carbon dioxide
A common way to overcome these challenges is mixtures have yet to be developed and validated.
to consider that the discharge and the combus- Multi-dimensional CFD calculations of PAC
tion process happen in a sequential way [97]. Dur- problems usually do not include detailed plasma
ing the discharge phase, only the plasma chem- chemistry. Reduced models with a minimal number
istry is considered. At the end of the discharge, the of charged and excited particles [97,98,111–113] or
plasma composition and the thermodynamic con- phenomenological plasma models [82,83] are of-
ditions are then used as the initial conditions for ten preferred. The vast majority of these numer-
the combustion simulation. Other coupling strate- ical studies were 1-D or 2-D simulations, focus-
gies can also be found, such as a frozen electric-field ing on ignition problems for which the localisa-
approach in [98], or a 0-D calculation for the dis- tion of the discharge was a priori known, some-
charge followed by a 1-D calculation for the com- times including multiple discharges, representative
bustion [99]. of NRP discharges. To the best of my knowl-
While the chemistry of air-plasmas is relatively edge, CFD simulations of flame stabilization by
well known [100,101], the development of chemi- distributed discharges such as in [9,33,93,114] have
cal kinetic models for plasmas of gas mixtures rel- not be achieved, yet. Further work in plasma mod-
evant for combustion environments is still ongoing eling and in flame-plasma coupling is still required
[51,102,103]. One of the difficulties in the develop- to allow these multi-physics simulations.
ment of these models is to include only the species
that are essential for the plasma and the combus-
tion chemistry, in order to keep the size of the mod- 4. Application to flame dynamics
els compatible with multi-dimensional calculations.
The identification of these species is done by com- In combustion systems where a flame is con-
bining a modeling and an experimental approach tinuously burning, problems of flame motion or
[104–110]. For example, Fig. 11 presents the com- fluctuation can arise. Flame dynamics refer here
parison of the evolution of atomic O mole fraction to these undesired or uncontrolled changes in the
following a 100-ns discharge in air (black) and in flame stabilization area, surface or shape. Differ-
a methane-air mixture (blue) at low pressure [104]. ent instability mechanisms can be responsible for
10.1016/j.proci.2022.06.025
JID: PROCI
these changes. One of the most studied is the cou- system. Except for a few studies reporting on the
pling with acoustic waves, also known as thermoa- use of non-equilibrium plasma discharges as an
coustic coupling or thermaocoustic instabilities. In acoustic source [123], scientists have been mainly
gas turbine engines, problem of thermoacoustic in- considering non-equilibrium plasmas to modify the
stabilities is of utmost importance. The susceptibil- flame response. The flame response to acoustic
ity of flames to upstream perturbations and their perturbations depends on the amplitude and fre-
coupling with resonant modes of the system has quency of the acoustic waves. For perturbations of
been one of the biggest challenges in the develop- small amplitude, the flame response is linear and
ment of modern high-efficiency, low-emission gas only depends on the frequency. Thus, it can be stud-
turbine engines in recent decades [16]. The occur- ied in a systematic manner by using the flame trans-
rence of flame instabilities is generally considered fer function (FTF) formalism [124].
as one of the highest risk items in new engine de- When a flame is subjected to velocity perturba-
velopment [115]. While analysis and simulation of tions of the flow, u , the heat release rate fluctuates
thermoacoustic stability has made great progress in by an amount Q˙ . The FTF is defined as the ratio
recent years [116], fully predictive modeling is still of the relative heat release rate fluctuations and the
challenging. relative velocity fluctuations, in the frequency do-
In ground-based gas turbines, thermoacoustic main:
instabilities can be hindered by passive control sys-
tems. Passive systems use clever geometrical ar- Q˙ˆ (ω)/Q˙¯
F (ω) = , (4)
rangements of the combustion chamber elements uˆ (ω)/ū
to avoid thermoacoustic coupling. They most often
rely on the use of well-tuned resonators to damp where ω is the frequency, Q˙ is the heat release rate,
specific dangerous oscillation modes of the sys- u is the flow velocity at the flame location, Xˆ is the
tem, which have been identified during the design Fourier transform of the quantity X , and X̄ is the
and prototyping phases. Passive techniques how- mean value of X . It can also be expressed in terms
ever have severe drawbacks: they generate large ad- of a gain, G, and a phase, φ:
ditional expenses during the engine development F (ω) = G(ω)eiφ(ω) , (5)
and they lead to an additional weight of the engine,
which severely penalizes these solutions for aero- where G reflects the magnitude of the flame re-
engine gas turbines. For power generation appli- sponse and φ defines the time lag between the ve-
cations, the cooling and purge mass flow required locity and the heat release rate fluctuations, at the
for the damping resonators has an adverse effect base of the flame.
on the system efficiency. Therefore, alternative so- To measure the velocity at the flame base, op-
lutions such as active control strategies are desir- tical techniques such as laser Doppler velocime-
able. Their concept is to alter the flame dynamics try (LDV) can be used. However, other techniques
or the system’s acoustics by actively detuning the such as multi-microphone method [125] or hot-wire
coherent response between the flame and pressure anemometry may be preferred. Measurements of
oscillations using sensors and actuators. heat release rates can be more challenging. How-
Non-equilibrium plasma actuators have shown ever, in the case of perfectly premixed flames, it has
promising potential for active control of ther- been shown that fluctuations of heat release rate are
moacoustic instabilities. Recent experimental stud- proportional to fluctuations of chemiluminescence
ies have demonstrated the capability of non- of either the excited methyl radical, CH∗ , or the hy-
equilibrium plasma discharges to control flame dy- droxyl radical, OH∗ [126,127]. Thus, a camera or a
namics [20–22,25,67,95,114,117–121], without the photomultiplier tube (PMT), equipped with an ad-
drawbacks of traditional actuators such as loud- equate band pass filter can be utilized.
speakers or fuel valves [122]. Two groups of stud- In 2013, two papers [114,117] demonstrated that
ies can be differentiated: studies on the effect of non-equilibrium plasma discharges could signifi-
non-equilibrium plasma discharges on the flame cantly modify the response of methane-air flames
response to acoustic perturbations, and studies to acoustic perturbations. Volkov et al. [117],
demonstrating a mitigation of thermoacoustic cou- showed that a DC corona discharge could increase
pling. The key results are presented in the next sec- the FTF gain of a laminar flat flame (stabilized
tions, followed by a discussion about what is miss- over a porous plate) for frequencies higher than
ing before considering non-equilibrium plasma dis- 150 Hz. They attributed this result to a trans-
charges as an actuator to actively control flame dy- port effect by ionic wind. A few months later, La-
namics in gas turbine engines. coste et al. [114] demonstrated that NRP spark dis-
charges could significantly decrease the FTF gain
4.1. Flame transfer functions of a 4-kW swirl flame, for a range of frequencies
from 8 to 350 Hz. They did not provide an explana-
To stop a thermoacoustic instability, the two tion for this result but they mentioned that it could
main options are to alter the flame response to the be attributed to a thermal effect, a chemical effect,
perturbations or to modify the acoustic field of the or a hydrodynamic effect, due to gas expansion and
10.1016/j.proci.2022.06.025
JID: PROCI
acoustic waves generation following fast gas heat- These results were obtained for a laminar flame
ing. These two first papers are interesting because of about 100 W. Could they be extrapolated to tur-
they show that different plasma actuation strategies bulent high-pressure flames? Recently, Deng et al.
can significantly change the flame dynamics. The [120] investigated the effect of DBD produced by
next step was then to identify which plasma is the an AC voltage source on the transfer function of a
most efficient. turbulent swirl flame at atmospheric pressure. They
To compare different plasma strategies, it is im- showed that below the breakdown voltage (4 kV
portant to select non-equilibrium plasma sources in their case), for which only an ionic wind ef-
promoting different effects. In 2017, Lacoste et al. fect can be expected, the FTF was not affected by
[9] compared the response of a laminar methane- the electric field. This result could have been ex-
air flame to acoustic forcing, forcing by an AC elec- pected because the ion density in a flame is too
tric field, and forcing by NRP glow discharges. The low to generate strong ionic winds. If the ion den-
AC electric field was chosen because its main effect sity can be increased, for example by using DBD
was to generate ionic wind from the flame, while in the flame or in the flow upstream of the flame,
NRP glow discharges were selected due to their then the flow generated by ionic wind might be
minimal thermal and transport effects, therefore strong enough to modify the turbulent flame re-
promoting a chemical actuation of the gas. New sponse to acoustic perturbations. When they in-
definitions of FTF for forcing by electric fields, creased the applied voltage and DBD were pro-
FE (ω), and plasma, FP (ω), were proposed: duced, with a coupled transport and chemical effect
(see Section 2.1), the FTF gain was significantly de-
Qˆ˙ (ω)/Q¯˙ creased. However, with their configuration, it is not
FE (ω) = = GE (ω)eiφE (ω) , (6) possible to distinguish if this result was obtained
Vˆ (ω)/V̄
mainly due to a chemical effect, or to a coupled
where V is the applied voltage and V̄ is a refer- chemical-ionic wind effect. Regarding the effect of
ence voltage (1 V). Here also, the FTF can be ex- non-equilibrium plasma discharges on the transfer
pressed in terms of a gain GE and a phase φE . functions of flames at elevated pressures, to the best
Note that the denominator of this transfer func- of my knowledge, there is no published results yet.
tion being independant of the average flame ther- Thus, while it is established that non-
mal power, this definition does not allow quantita- equilibrium plasma can mitigate the response
tive comparisons between gains obtained for forc- of swirl turbulent flames to acoustic perturbations,
ing by acoustic waves and electric field. For plasma and demonstration has been made that chemical
forcing: actuation and ionic wind could have a strong
effect on the flame dynamics, the question of the
Qˆ˙ (ω)/Q¯˙ best plasma actuation for practical applications
FP (ω) = = GP (ω)eiφP (ω) , (7) remains open.
sinc(1/2)PP (ω)/Q¯˙
4.2. Control of thermoacoustic instabilities
where PP is the average plasma power deposited
during half a period of the forcing. The term A few papers report on the active control
sinc(1/2) comes from the Fourier expansion of the of thermoacoustic instabilities by non-equilibrium
square-wave plasma forcing (more details in [9]). plasma discharges [19,21,22,25,114,121]. In all
The terms GP and φP are the gain and the phase cases, the thermoacoustic coupling happened be-
of the FTF, respectively. The amplitude of the forc- tween longitudinal acoustic waves and premixed
ing is defined as PP /Q¯˙ , thus, quantitative compar- flames, and the non-equilibrium plasmas were pro-
isons between the FTF gains for forcing by acoustic duced by NRP discharges. Two types of control
waves and NRP glow discharges are possible. strategies have been attempted: by continuously ap-
The authors showed that, qualitatively, the plying the non-equilibrium plasma discharges un-
flame was similarly responsive to acoustic, AC elec- til the flame stabilizes, or by forcing the flame with
tric field, and plasma forcing, for the entire range of the plasma, with an appropriate phasing with re-
frequencies studied, as presented in Fig. 12. They spect to the acoustic fluctuation, i.e., with a feed-
explained this result by the fact that electric field back loop.
forcing was actually inducing a change in the local In 2013, Moeck et al. [19] showed that for a
flow velocity, which is similar to velocity fluctua- 43-kW swirl flame of natural gas and air at at-
tions induced by acoustic waves. When NRP glow mospheric pressure, a thermoacoustic instability at
discharges were applied, the flow field was not af- 150 Hz could be completely suppressed by apply-
fected but the local reactivity of the gas was in- ing NRP spark discharges at 50 kHz, continuously.
creased close to the tip of the flame. This change For these conditions, the plasma power was about
in the local reactivity could be modeled as a local 315 W, i.e., about 0.7% of the flame thermal power.
increase in the flame burning velocity, having a sim- Later on, these results were confirmed by similar re-
ilar kinematic impact that a change in the local flow sults [21,22,114]. Recently, Kim and Cohen [25] ex-
velocity [88]. tended these results to a pressure of 5 bar. This
10.1016/j.proci.2022.06.025
JID: PROCI
Fig. 12. Comparison of the transfer functions of a methane-air V-flame to forcing by acoustic waves, AC electric field,
and NRP glow discharges. Adapted from [9].
pressure was chosen because it is representative

of micro-gas turbine engines, or of a typical low-
power regime of aero-engine combustors. The au-
thors noticed that to obtain a similar plasma ef-
fect when the bulk flow velocity and/or the pressure
were increased, the electric power deposited in the
NRP spark discharges should be linearly increased.
To decrease the overall plasma power deposited,
i.e., to increase the plasma actuation efficiency, in-
stead of applying continuously the discharges, they
can be gated at a chosen frequency. The idea is
that NRP discharges produce a forcing that coun-
teract the acoustic perturbation. For example, for
a thermoacoustic instability at 120 Hz, i.e., with a
period of 8.3 ms, NRP discharges at 30 kHz can
be applied during 4.15 ms, which corresponds to
126 pulses, then no discharges are applied during
4.15 ms, before applying again 126 pulses at 30 kHz,
and so on. This is what was done in [19], for an at-
mospheric pressure natural gas and air swirl flame Fig. 13. Amplitude of the pressure fluctuations, P, for
of 41 kW. The authors showed that, depending on different pulse repetition frequencies, PRF, as a function
the phasing between the acoustic oscillations and of the delay, τ between the plasma forcing and the mea-
the plasma forcing, the amplitude of the peak pres- sured acoustic wave. The NRP glow discharges were ob-
sure induced by the thermoacoustic coupling could tained for an applied voltage of 6.8 kV. The duty cycle of
be either decreased or increased. This was the first the forcing was 50%, and the frequency of both the forc-
evidence that non-equilibrium plasma discharges ing and the acoustic wave was 170 Hz. Taken from [121].
could be used in a closed-loop control scheme to
mitigate thermoacoustic instabilities. Again, in this
study, as the spark regime of NRP discharges was Section 2.4). Fig. 13 presents the pressure fluctua-
utilized, the authors could not conclude on the tions of a 230-W wall stabilized laminar flame of
main mechanism of plasma action. methane and air, without and with plasma forc-
Recently, a similar approach was used with NRP ing. Without plasma forcing, the flame experiences
glow discharges [121]. Recall that for the glow a thermoacoustic instability at 170 Hz, with an am-
regime of NRP discharges, the fast gas heating plitude of the pressure fluctuations, P, of 0.3 Pa.
is minimal, there is no transport effect, and the For NRP glow discharges continuously applied at
chemical actuation of the gas is promoted (see 30 kHz, the pressure fluctuations could be com-
10.1016/j.proci.2022.06.025
JID: PROCI
pletely suppressed, but the plasma power was about Depending on the use (propulsion or power gener-
1.6% of the flame thermal power. Applied in a ation), the combustor pressure and temperature in
closed-loop control scheme with a duty cycle of gas turbine engines can be very different, typically
50%, the thermoacoustic instability could be sup- from 3 to 45 bar and from 500 to 800 K [129]. The
pressed for a plasma power of 0.8%. Even if this gas density is crucial for the properties of electrical
was obtained for a tiny flame, this result is promis- discharges. In addition to the obvious effect on the
ing for applications at elevated pressures. Indeed, as reduced electric field controlling the plasma chem-
discussed for lean flame stabilization in Section 3.3, istry, the gas density also affects the dimensions of
NRP glow discharges seems to be more efficient at the discharges. At low pressures, discharges are usu-
elevated pressure than NRP spark discharges. ally more diffuse than at higher pressures but the
pressure effect is highly non linear [4]. For exam-
4.3. Towards engine conditions ple, Xu et al. [75] showed that in air, the initial ra-
dius of NRP spark discharges of 6.2-mJ energy per
Applications of plasma-assisted flame dynam- pulse decreased from 220 to 150 μm when the pres-
ics mainly target combustion problems in gas tur- sure was increased from 1 to 5 bar, while between
bine engines. Thus, even though non-equilibrium 10 and 20 bar, the radius remained almost constant
plasma demonstrated good potentials for flame sta- at about 120 μm. This decrease in the discharge
bilization, the conditions at which the results were volume with pressure can significantly increase the
obtained are so far from real engine conditions strength of the shock waves from the discharge area
that it is not possible to conclude about the rel- (as discussed in [90]). Therefore, the main mecha-
evance of a plasma actuator. Over this decade, nism of plasma effect could transition from ther-
we learned that at pressures up to 5 bar, NRP mal to transport. Similarly, while the glow regime
discharges can stabilize turbulent methane-air or of NRP discharges was disregarded for PAC at at-
propane-air flames of a few tens of kilowatts, by mospheric pressure, mainly because (1) the spark
applying a plasma power of about 1% of the flame regime is very efficient and (2) it is difficult to ob-
thermal power. We learned that separately, differ- tain enough plasma power without transitioning to
ent effects of the plasma can be efficient in flame the spark regime, this regime is really promising at
stabilization but we do not know which effect, or elevated pressure.
which combination of effects, is the most efficient The question of pollutant emissions in non-
for flame stabilization. To gain confidence that a equilibrium plasma discharges will also have to be
plasma actuator can work in a real gas turbine en- taken into account during the design of industrial
gine, further work is needed. plasma actuators. In fuel-air mixtures, similarly to
First, the question of the efficiency of plasma what is happening in combustion, the production
actuation on other modes of thermoacoustic in- of NO by the discharges is significant when the gas
stabilities is still open. As highlighted by Poinsot temperature is high (typically higher than 1800 K
in [116], the vast majority of studies on flame dy- [1,105]). This local production of NO in the dis-
namics focus on longitudinal acoustic waves, while charges can be large enough to drastically increase
in practice transverse acoustic waves or azimuthal the global NO concentration in the burned gases of
acoustic waves can be as, or even more, detrimen- the combustion system [19,25,130–132]. For exam-
tal. These other acoustic modes can be significantly ple, in [19], NRP spark discharges at 20 kHz could
different than the longitudinal one, with, for ex- mitigate the thermoacoustic instability of a 43 kW
ample, standing or spinning waves, or featuring flame, but the concentration of NO in the exhaust
much higher frequencies. How the discharges could gases was increased from 2 to 17 ppm [67]. Thus,
affect the flame-acoustic coupling in these condi- focusing on optimizing the impact on flames of dis-
tions is unknown, and the plasma-acoustic cou- charges with low gas heating, such as DBD or NRP
pling is unknown as well. This later coupling for glow discharges, could be a good strategy to avoid
non-thermal discharges has been barely studied in further issues of pollutant emissions.
atmospheric pressure air [123,128], but the results Finally, while gas turbine engines are burning
showed strong interactions between the acoustic a large variety of fuels, from low to high carbon
field and the plasma discharges. In addition, aside content such as methane, syngas, kerosene, heavy
from acoustic perturbations, turbulent flames are fuel oil, or coal, PAC studies using non-equilibrium
also responsive to a broad range of disturbances, plasma discharges have mainly utilized methane,
e.g., flow, composition, temperature, or enthalpy. propane and ethylene, with only a few studies deal-
Recently, Sun et al. [26] showed that gliding arc ing with liquid fuels (n-heptane [27,133], dode-
discharges could enhance the lean blow-off limit cane [27], iso-octane [133], or kerosene [134]). The
of methane-air swirl flames perturbed by low fre- interactions between plasma discharges and liquid
quency flow velocity steps but further work is re- droplets can change the effect of plasma on the
quired to demonstrate plasma actuation for more flame dynamics. In addition, as mentioned in [25],
complex or coupled disturbances. the plasma effect is fuel dependant with, for a same
Another important parameter is the gas density. plasma power, a stronger effect on methane flames
10.1016/j.proci.2022.06.025
JID: PROCI
than on propane flames. Note that plasma-assisted • While the effects of pressure, mixture com-
coal combustion has been intensively studied but position, and flow are significant on combus-
mainly with thermal plasma (see for example [135– tion processes, their effects are as important
137]). It would be very interesting to have more to the physics of plasma and relatively poorly
information about plasma-liquid fuel interactions documented for conditions relevant to com-
as well as on the chemistry of plasmas of mix- bustion applications. Fundamentals studies
tures of large hydrocarbons and air. Finally, with of plasma at pressures higher than atmo-
the recent shift of research interest towards carbon- spheric as well as in different flows and gas
free fuels, recent papers on plasma-assisted ammo- mixtures are still lacking. So are PAC studies
nia flames have been published [138–140]. This re- in engine relevant conditions.
search is much needed for an efficient transition to
zero-carbon emissions.
Declaration of Competing Interest

5. Conclusion
The authors declare that they have no known
This paper first reviews some of the most com- competing financial interests or personal relation-
monly used non-equilibrium plasma sources for ships that could have appeared to influence the
plasma-assisted combustion (PAC) studies: DBD, work reported in this paper.
gliding arcs, microwaves, and NRP discharges.
Then a review of the results obtained for ignition,
acceleration of the burning velocity, and enhance-
ment of the lean blow-off limits is proposed, fol- Acknowledgments
lowed by a more detailed discussion on the utiliza-
tion of non-equilibrium plasma for the control of This paper is dedicated to Christophe O. Laux
flame dynamics, targeting application in gas tur- of CentraleSupelec and Jonas P. Moeck of NTNU,
bine engines. Main conclusions are: who both had an instrumental impact on my re-
• While ignition by non-thermal plasma dis- search in plasma-assisted combustion. Christophe
charges have been intensively studied in first introduced me to non-equilibrium plasma
CVCC and in continuous flow burners, there physics in 2003. He had the greatest influence on my
is a lack of experimental data validating thinking about plasma-assisted combustion and
chemical models of PAC and under condi- about how to seek for simplicity in complex prob-
tions relevant to engines, i.e., with commer- lems. I was fortunate to be one of the many young
cial fuels and at thermodynamic and flow researchers who could learn from Christophe about
conditions far from ambient. spectroscopy and plasma physics, benefiting from
• Fast gas heating obtained for discharges in his unshakeable positiveness and his passion for
air containing mixtures, for reduced electric beautiful science. In 2011, I became friend with
fields in the range 100 to 1000 Td, can be Jonas and we decided to study the impact of non-
favorable or detrimental to combustion pro- equilibrium plasma discharges on flame dynam-
cesses. It induces gas expansion and gener- ics. Thanks to Jonas, a new aspect of combus-
ates shock/acoustic waves, responsible for lo- tion opened up to me, providing enough challenges
cal flow disturbances and removal of a part and questions to keep me busy for the following
of the discharge energy from the plasma lo- decade. Jonas was and still is my reference for
cation. flame-acoustic coupling and thermoacoustic insta-
• The efficiency of plasma action is strongly bilities.
setup dependant. As the discharges usually I would like to sincerely thank all the persons
have a volume significantly smaller than the who, over these two decades, have worked with
flame volume, their location is crucial for the me in plasma-assisted combustion, with a special
combustion enhancement. In addition, in- thank to Ammar Alkhalifa, Anne Bourdon, Min
creasing their volume is highly desirable and Suk Cha, Davide Del Cont-Bernard, Francesco
strategies promoting large volumes of non- Di Sabatino, Andreas Ehn, Joshua A. T. Gray,
equilibrium plasmas in combustion environ- Thibault F. Guiberti, Sylvain Heitz, Guillaume
ments should be further developed. Pilla, João Vargas, Yuan Xiong, and Da A. Xu.
• Flame dynamics can be controlled by I also would like to acknowledge the funding
non-equilibrium plasma discharges obtained sources of the research presented in this paper: In
from various sources. However, which energy France, the Centre National de la Rechereche Sci-
pathway (thermal, chemical, or transport) entifique (CNRS), and the Agence Nationale de
should be promoted remains unclear. More la Recherche (ANR), in Germany, the Deutsche
studies comparing each of these effects, Forschungsgemeinschaft (DFG), and in Saudi
separately and combined are necessary to Arabia, the King Abdullah University of Science
design an efficient plasma actuator. and Technology (KAUST).
10.1016/j.proci.2022.06.025
JID: PROCI
Supplementary material [19] J.P. Moeck, D.A. Lacoste, C.O. Laux, C.O.
Paschereit, Control of combustion dynamics in
Supplementary material associated with this ar- a swirl-stabilized combustor with nanosecond
ticle can be found, in the online version, at doi:10. repetitively pulsed discharges, in: Proceedings of
the 51st AIAA ASM, 7 - 10 Jan. 2013, Grapevine,
1016/j.proci.2022.06.025
Texas, volume AIAA 2013-0565, 2013.
[20] D. Lacoste, D. Xu, J. Moeck, C. Laux, Dynamic re-
sponse of a weakly turbulent lean-premixed flame
References to nanosecond repetitively pulsed discharges, Proc.
Combust. Inst. 34 (2) (2013) 3259–3266.
[1] C.K. Law, Combustion Physics, Cambridge Uni- [21] W. Kim, J. Snyder, J. Cohen, Plasma assisted com-
versity Press, NY, USA, NY, USA, 2006. bustor dynamics control, Proc. Combust. Inst. 35
[2] J.H.S. Lee, The Detonation Phenomenon, Cam- (2015) 3479–3486.
bridge University Press, NY, USA, 2008. [22] F.G. del Campo, D.E. Weibel, J.S. Santosh,
[3] A. Cavaliere, M. de Joannon, Mild combustion, A.F. Ghoniem, Plasma-assisted control of com-
Prog. Energy Combust. Sci. 30 (2004) 329–366. bustion instabilities in low-emissions combustors
[4] Y.P. Raizer, Gas Discharge Physics, Springer-Ver- at realistic conditions, in: Proceedings of the AIAA
lag, Berlin Heidelberg, Germany, 1991. Propulsion and Energy Forum, Indianapolis, IN,
[5] F.F. Chen, Introduction to Plasma Physics and 2019, pp. 2019–3950.
Controlled Fusion, Springer International Publish- [23] M. Zhang, J. Li, W. Cheng, T. Li, Active con-
ing, Switzerland, 2016. trol of thermoacoustic instability using microsec-
[6] A.B. Fialkov, Investigations on ions in flames, Prog. ond plasma discharge, J. Appl. Phys. 127 (3) (2020)
Energy Combust. Sci. 23 (1997) 399–528. 033301.
[7] J. Cancian, B.A.V. Bennett, M.B. Colket, [24] V.P. Blanchard, N. Minesi, S. Stepanyan, G.-D.
M.D. Smooke, Prediction of electron and ion Stancu, C.O. Laux, Dynamics of a lean flame stabi-
concentrations in low-pressure premixed acetylene lized by nanosecond discharges, in: Proceedings of
and ethylene flames, Combust. Theory Model. 17 the AIAA Scitech 2021 Forum, 2021, p. 1700.
(2) (2013) 294–315. [25] W. Kim, J. Cohen, Plasma-assisted combustor
[8] M.V. Tran, M.S. Cha, Time evolution of propagat- dynamics control at realistic gas turbine condi-
ing nonpremixed flames in a counterflow, annular tions, Combust. Sci. Technol. 193 (5) (2021) 869–
slot burner under ac electric fields, Proc. Combust. 888.
Inst. 36 (1) (2017) 1421–1430. [26] J. Sun, Y. Tang, S. Li, Plasma-assisted stabilization
[9] D.A. Lacoste, Y. Xiong, J.P. Moeck, S.H. Chung, of premixed swirl flames by gliding arc discharges,
W.L. Roberts, M.S. Cha, Transfer functions of Proc. Combust. Inst. 38 (4) (2021) 6733–6741.
laminar premixed flames subjected to forcing by [27] G. Vignat, N. Minesi, P.R. Soundararajan,
acoustic waves, ac electric fields, and non-thermal D. Durox, A. Renaud, V. Blanchard, C.O. Laux,
plasma discharges, Proc. Combust. Inst. 36 (2017) S. Candel, Improvement of lean blow out perfor-
4183–4192. mance of spray and premixed swirled flames using
[10] Y. Xiong, M.S. Cha, S.H. Chung, Ac electric field nanosecond repetitively pulsed discharges, Proc.
induced vortex in laminar coflow diffusion flames, Combust. Inst. 38 (2021) 6559–6566.
Proc. Combust. Inst. 35 (3) (2015) 3513–3520. [28] A. Starikovskiy, N. Aleksandrov, A. Rakitin, Plas-
[11] E.J.J. Lenoir, Gas-engine, US Patent July 13 (1886) ma-assisted ignition and deflagration-to-detona-
No.345,596. tion transition, Philos. Trans. R. Soc. A 370 (2012)
[12] R. Maly, M. Vogel, Initiation and propagation of 740–773.
flame fronts in lean ch4-air mixtures by the three [29] J. Dutton, A survey of electron swarm data, J. Phys.
modes of the ignition spark, SIC 7 (1976) 821–831. Chem. Ref. Data 4 (1975) 577–856.
[13] M.T. Nguyen, S.S. Shy, Y. Chen, B.L. Lin, [30] G.J.M. Hagelaar, L.C. Pitchford, Solving the Boltz-
S.Y. Huang, C.C. Liu, Conventional spark ver- mann equation to obtain electron transport coeffi-
sus nanosecond repetitively pulsed discharge for a cients and rate coefficients for fluid models, Plasma
turbulence facilitated ignition phenomenon, Proc. Sources Sci. Technol. 14 (2005) 722–733.
Combust. Inst. 36 (2) (2017) 2801–2808. [31] U. Kogelschatz, Dielectric-barrier discharges: their
[14] Y. Ju, W. Sun, Plasma assisted combustion: dynam- history, discharge physics, and industrial applica-
ics and chemistry, Prog. Energy Combust. Sci. 48 tions, Plasma Chem. Plasma Process. 23 (1) (2003)
(2015) 21–83. 1–46.
[15] S. Candel, Combustion dynamics and control: [32] M. Boumehdi, S. Stepanyan, P. Desgroux, G. Van-
progress and challenges, Proc. Combust. Inst. 29 (1) hove, S. Starikovskaia, Ignition of methane- and
(2002) 1–28. n-butane-containing mixtures at high pressures by
[16] T.C. Lieuwen, V. Yang, Combustion Instabilities in pulsed nanosecond discharge, Combust. Flame 162
Gas Turbine Engines, 210, Progress in Astronautics (4) (2015) 1336–1349.
and Aeronautics, 2005. [33] G.T. Kim, C.S. Yoo, S.H. Chung, J. Park, Effects
[17] J.P. Hathout, A.M. Annaswamy, M. Fleifil, of non-thermal plasma on the lean blowout lim-
A.F. Ghoniem, A model-based active control its and CO/NO X emissions in swirl-stabilized tur-
design for thermoacoustic instability, Combust. bulent lean-premixed flames of methane/air, Com-
Sci. Technol. 132 (1–6) (1998) 99–138. bust. Flame 212 (2020) 403–414.
[18] W. Polifke, Modeling and analysis of premixed [34] D.G. Park, S.H. Chung, M.S. Cha, Visualization of
flame dynamics by means of distributed time delays, ionic wind in laminar jet flames, Combust. Flame
Prog. Energy Combust. Sci. 29 (1) (2002) 1–28. 184 (2017) 246–248.
10.1016/j.proci.2022.06.025
JID: PROCI
[35] A. Fridman, S. Nester, L.A. Kennedy, A. Saveliev, [51] N.A. Popov, S.M. Starikovskaia, Relaxation of
O. Mutaf-Yardimci, Gliding arc gas discharge, electronic excitation in nitrogen/oxygen and fuel/air
Prog. Energy Combust. Sci. 25 (2) (1999) 211–231. mixtures: fast gas heating in plasma-assisted igni-
[36] J. Gao, K. Chengdong, Z. Jiajian, A. Ehn, T. Hur- tion and flame stabilization, Prog. Energy Combust.
tig, T. Yong, C. Shuang, M. Aldén, L. Zhongshan, Sci. In press (2022).
Visualization of instantaneous structure and dy- [52] A. Flitti, S. Pancheshnyi, Gas heating in fast pulsed
namics of large-scale turbulent flames stabilized by discharges in N2–O2 mixtures, Eur. Phys. J. Appl.
a gliding arc discharge, Proc. Combust. Inst. 37 Phys. 45 (2) (2009) 21001.
(2019) 5629–5636. [53] D. Pai, G. Stancu, D. Lacoste, C. Laux, Nanosec-
[37] Y. Tang, D. Xie, B. Shi, N. Wang, S. Li, Flammabil- ond repetitively pulsed discharges in air at atmo-
ity enhancement of swirling ammonia/air combus- spheric pressure - the glow regime, Plasma Sources
tion using AC powered gliding arc discharges, Fuel Sci. Technol. 18 (2009) 045030.
313 (2022) 122674. [54] D. Pai, D. Lacoste, C. Laux, Nanosecond repeti-
[38] C. Tendero, C. Tixier, P. Tristant, J. Desmaison, tively pulsed discharges in air at atmospheric pres-
P. Leprince, Atmospheric pressure plasmas: a re- sure ’ the spark regime, Plasma Sources Sci. Tech-
view, Spectrochim. Acta Part B 61 (2006) 2–30. nol. 19 (2010) 065015.
[39] C.I.M. Beenakker, A cavity for microwave induced [55] D. Pai, D. Lacoste, C. Laux, Transitions between
plasmas operated in helium and argon at atmo- corona, glow, and spark regimes of nanosecond
spheric pressure, Spectrochim. Acta Part B 31 repetitively pulsed discharges in air at atmospheric
(1976) 483–486. pressure, J. Appl. Phys. 107 (2010) 093303.
[40] M. Moisan, G. Sauve, Z. Zakrzewski, J. Hubert, [56] D.A. Xu, D.A. Lacoste, D.L. Rusterholtz,
An atmospheric pressure waveguide-fed microwave P.-Q. Elias, G.D. Stancu, C.O. Laux, Experimental
plasma torch: the tia design, Plasma Sources Sci. study of the hydrodynamic expansion following
Technol. 3 (1994) 584–592. a nanosecond repetitively pulsed discharge in air,
[41] A.M. Bilgic, U. Engel, E. Voges, M. Kuckelheim, Appl. Phys. Lett. 99 (2011) 121502.
J.A.C. Broekaert, A new low-power microwave [57] D.A. Xu, M.N. Shneider, D.A. Lacoste, C.O. Laux,
plasma source using microstrip technology for Thermal and hydrodynamic effects of nanosecond
atomic emission spectrometry, Plasma Sources Sci. discharges in atmospheric pressure air, J. Phys. D:
Technol. 9 (2000) 1–4. Appl. Phys. 47 (2014) 235202.
[42] M.K. Le, A. Nishiyama, Y. Ikeda, Evaluation of [58] S.S. Shy, Y.W. Shiu, L.J. Jiang, C.C. Liu, S. Minaev,
a novel miniaturised microwave resonating igniter: Measurement and scaling of minimum ignition en-
the flat panel igniter, Proc. Combust. Inst. 37 (2019) ergy transition for spark ignition in intense isotropic
5613–5620. turbulence from 1 to 5 atm, Proc. Combust. Inst. 36
[43] S.A. Shcherbanev, M.A. Cherif, (2017) 1785–1791.
S.M. Starikovskaia, Y. Ikeda, Experimental [59] J.K. Lefkowitz, T. Ombrello, An exploration of in-
study of pulsed microwave discharges at pressures ter-pulse coupling in nanosecond pulsed high fre-
ranging over five orders of magnitude, Plasma quency discharge ignition, Combust. Flame 180
Sources Sci. Technol. 28 (2019) 045009. (2017) 136–147.
[44] K.W. Hemawan, C.L. Romel, S. Zuo, I.S. Wichman, [60] R. Feng, J. Zhu, Z. Wang, M. Sun, H. Wang, Z. Cai,
T.A. Grotjohn, J. Asmussen, Microwave plasma-as- B. An, L. Li, Ignition modes of a cavity-based
sisted premixed flame combustion, Appl. Phys. Lett. scramjet combustor by a gliding arc plasma, Energy
89 (2006) 141501. 214 (2021) 118875.
[45] A. Ehn, T. Hurtig, P. Petersson, J. Zhu, A. Lars- [61] L. Wu, J. Lane, N. Cernansky, D. Miller, A. Frid-
son, C. Fureby, J. Larfeldt, Z. Li, M. Aldén, Setup man, A. Starikovskiy, Plasma - assisted ignition
for microwave stimulation of a turbulent low-swirl below self-ignition threshold in methane, ethane,
flame, J. Phys. D: Appl. Phys. 49 (2016) 185601. propane and butane - air mixtures, Proc. Combust.
[46] V.M. Shibkov, L.V. Shibkova, V.G. Gromov, Inst. 33 (2) (2011) 3219–3224.
A.A. Karachev, R.S. Konstantinovskii, Influence [62] A. Saha, S. Yang, C.K. Law, On the competing roles
of surface microwave discharge on ignition of of turbulence and differential diffusion in facilitated
high-speed propane-air flows, High Temp. 49 (2) ignition, Proc. Combust. Inst. 37 (2019) 2383–2390.
(2011) 155–167. [63] J. Marrero-Santiago, F. Collin-Bastiani, E. Riber,
[47] C.H. Kruger, C.O. Laux, L. Yu, D.M. Packan, G. Cabot, B. Cuenot, B. Renou, On the mechanisms
L. Pierrot, Nonequilibrium discharges in air and ni- of flame kernel extinction or survival during aero-
trogen plasmas at atmospheric pressure, Pure Appl. nautical ignition sequences: experimental and nu-
Chem. 74 (3) (2002) 337–347. merical analysis, Combust. Flame 222 (2020) 70–84.
[48] S. Pancheshnyi, D. Lacoste, A. Bourdon, C. Laux, [64] G. Pilla, D. Galley, D. Lacoste, F. Lacas, D. Vey-
Ignition of propane-air mixtures by a repetitively nante, C. Laux, Stabilization of a turbulent pre-
pulsed nanosecond discharge, IEEE Trans. Plasma mixed flame using a nanosecond repetitively pulsed
Sci. 34 (1) (2006) 2478–2487. plasma, IEEE Trans. Plasma Sci. 34 (6) (2006)
[49] N. Popov, Investigation of the mechanism for rapid 2471–2477.
heating of nitrogen and air in gas discharges, [65] D. Del Cont-Bernard, M. Ruchkina, P. Ding,
Plasma Phys. Rep. 27 (10) (2001) 886–896. J. Bood, A. Ehn, D.A. Lacoste, Femtosecond
[50] D. Rusterholtz, D. Lacoste, G. Stancu, D. Pai, two-photon laser-induced fluorescence imaging of
C. Laux, Ultrafast heating and oxygen dissociation atomic hydrogen in a laminar methane–air flame
in atmospheric pressure air by nanosecond repeti- assisted by nanosecond repetitively pulsed dis-
tively pulsed discharges, J. Phys. D: Appl. Phys. 46 charges, Plasma Sources Sci. Technol. 29 (2020)
(2013) 464010. 065011.
10.1016/j.proci.2022.06.025
JID: PROCI
[66] D. Del Cont-Bernard, T.F. Guiberti, D.A. Lacoste, [81] S.B. Leonov, A. Houpt, S. Elliott, B. Hedlund,
Laser induced fluorescence investigation of the Ethylene ignition and flameholding by electrical dis-
chemical impact of nanosecond repetitively pulsed charge in supersonic combustor, J. Propuls. Power
glow discharges on a laminar methane-air flame, 34 (2018) 499–509.
Proc. Combust. Inst. 38 (2021) 6641–6649. [82] M. Castela, B. Fiorina, A. Coussement, O. Gicquel,
[67] D.A. Lacoste, J.P. Moeck, C.O. Paschereit, N. Darabiha, C.O. Laux, Modelling the impact of
C.O. Laux, Effect of plasma discharges on nitric non-equilibrium discharges on reactive mixtures for
oxide emissions in a premixed flame, J. Propuls. simulations of plasma-assisted ignition in turbulent
Power 29 (3) (2013) 748–751. flows, Combust. Flame 166 (2016) 133–147.
[68] C. Guerra-Garcia, M. Martinez-Sanchez, [83] M. Castela, S. Stepanyan, B. Fiorina, A. Cousse-
R.B. Miles, A. Starikovskiy, Localized pulsed ment, O. Gicquel, N. Darabiha, C.O. Laux, A 3-D
nanosecond discharges in a counterflow non- dns and experimental study of the effect of the
premixed flame environment, Plasma Sources Sci. recirculating flow pattern inside a reactive kernel
Technol. 24 (2015) 055010. produced by nanosecond plasma discharges in a
[69] J.A.T. Gray, D.A. Lacoste, Effect of the plasma lo- methane-air mixture, Proc. Combust. Inst 36 (3)
cation on the deflagration-to-detonation transition (2017) 4095–4103.
of a hydrogen–air flame enhanced by nanosecond [84] S. Lovascio, T. Ombrello, J. Hayashi, S. Stepanyan,
repetitively pulsed discharges, Proc. Combust. Inst. D. Xu, G. Stancu, C. Laux, Effects of pulsation
38 (2021) 3463–3472. frequency and energy deposition on ignition us-
[70] S. Starikovskaia, Plasma assisted ignition and ing nanosecond repetitively pulsed discharges, Proc.
combustion, J. Phys. D Appl. Phys. 39 (2006) Combust. Inst. 36 (2017) 4079–4086.
R265–R299. [85] K. Zaima, K. Sasaki, Control of burning velocity
[71] A. Starikovskiy, N. Aleksandrov, Plasma - assisted in premixed burner flame by high-energy electrons
ignition and combustion, Prog. Energy Combust. of dielectric barrier discharge, Jpn. J. Appl. Phys. 53
Sci. 39 (2013) 61–110. (2014) 066202.
[72] S. Starikovskaia, D.A. Lacoste, G. Colonna, Non-e- [86] M.D.G. Evans, J.M. Bergthorson, S. Coulombe,
quilibrium plasma for ignition and combustion en- Actuation of a lean-premixed flame by diffuse
hancement, Eur. Phys. J. D 75 (2021) 231. non-equilibrium nanosecond-pulsed plasma at at-
[73] I.N. Kosarev, N.L. Aleksandrov, S.V. Kindysheva, mospheric pressure, J. Appl. Phys. 122 (2017)
S.M. Starikovskaia, A.Y. Starikovskii, Kinetics of 173305.
ignition of saturated hydrocarbons by nonequilib- [87] A. Ehn, P. Petersson, J.J. Zhu, Z.S. Li, M. Aldén,
rium plasma: C2H6- to C5H12-containing mix- E.J.K. Nilsson, J. Larfeldt, A. Larsson, T. Hur-
tures, Combust. Flame 156 (2009) 221–233. tig, N. Zettervall, C. Fureby, Investigations of mi-
[74] R. Scarcelli, T. Wallner, S. Som, S. Biswas, I. Ekoto, crowave stimulation of a turbulent low-swirl flame,
D. Breden, A. Karpatne, L.L. Raja, Modeling Proc. Combust. Inst. 36 (2017) 4121–4128.
non-equilibrium discharge and validating tran- [88] D. Lacoste, J. Moeck, W. Roberts, S. Chung,
sient plasma characteristics at above-atmospheric M. Cha, Analysis of the step responses of laminar
pressure, Plasma Sources Sci. Technol. 27 (2018) premixed flames to forcing by non - thermal plasma,
124006. Proc. Combust. Inst. 36 (2017) 4145–4153.
[75] D.A. Xu, D.A. Lacoste, C.O. Laux, Ignition of qui- [89] A. Elkholy, Y. Shoshyn, S. Nijdam, J.A. van Oijen,
escent lean propane – air mixtures at high pres- E.M. van Veldhuizen, U. Ebert, L.P.H. de Goey,
sure by nanosecond repetitively pulsed discharges, Burning velocity measurement of lean methane-air
Plasma Chem. Plasma Process. 36 (2016) 309–327. flames in a new nanosecond dbd microplasma
[76] T. Shiraishi, T. Urushihara, M. Gundersen, A burner platform, Exp. Therm. Fluid Sci. 95 (2018)
trial of ignition innovation of gasoline engine by 18–26.
nanosecond pulsed low temperature plasma igni- [90] F. Di Sabatino, T.F. Guiberti, J.P. Moeck,
tion, J. Phys. D Appl. Phys. 42 (2009) 135208. W.L. Roberts, D.A. Lacoste, Actuation efficiency
[77] Y. Ikeda, M.K. A. Nishiyama, Microwave enhanced of nanosecond repetitively pulsed discharges for
ignition process for fuel mixture at elevated pressure plasma-assisted swirl flames at pressures up to
of 1mpa, in: Proceedings of the 47th AIAA ASM, 3 bar, J. Phys. D: Appl. Phys. 54 (2021) 075208.
5 - 8 Jan. 2009, Orlanda, FL, USA, volume AIAA [91] J.A.T. Gray, D.A. Lacoste, Enhancement of the
2009-223, 2009. transition to detonation of a turbulent hydro-
[78] T. Li, I. Adamovich, J. Sutton, Infuence of nanosec- gen–air flame by nanosecond repetitively pulsed
onds pulsed discharges on the composition of inter- plasma discharges, Combust. Flame 199 (2019)
mediate and final combustion products in the HCCI 258–266.
engine, Plasma Chem. Plasma Process. 39 (2019) [92] R.-Y. Ju, J.-H. Wang, H. Xia, Y.-M. Li, H.-B. Mu,
683–694. G.-J. Zhang, Z.-H. Huang, Effect of rotating glid-
[79] L. Merotto, M. Balmelli, W. Vera-Tudela, P. Soltic, ing arc plasma on lean blow-off limit and flame
Comparison of ignition and early flame propa- structure of bluff body and swirl-stabilized pre-
gation in methane/air mixtures using nanosecond mixed flames, IEEE Trans. Plasma Sci. 49 (2021)
repetitively pulsed discharge and inductive ignition 3554–3565.
in a pre-chamber setup under engine relevant con- [93] F. Di Sabatino, D.A. Lacoste, Enhancement of the
ditions, Combust. Flame 237 (2022) 111851. lean stability and blow-off limits of methane-air
[80] H. Do, M.A. Cappelli, M.G. Mungal, Plasma as- swirl flames at elevated pressures by nanosecond
sisted cavity flame ignition in supersonic flows, repetitively pulsed discharges, J. Phys. D Appl. Phys.
Combust. Flame 157 (2010) 1783–1794. 53 (35) (2020) 355201.
10.1016/j.proci.2022.06.025
JID: PROCI
[94] S.B. Leonov, Electrically driven supersonic com- [109] J.K. Lefkowitz, P. Guo, A. Rousso, Y. Ju, Species
bustion, Energies 11 (2018) 1733. and temperature measurements of methane oxida-
[95] W. Cui, Y. Ren, S. Li, Stabilization of premixed tion in a nanosecond repetitively pulsed discharge,
swirl flames under flow pulsations using microsec- Philos. Trans. R. Soc. A 373 (2015) 20140333.
ond pulsed plasmas, J. Propuls. Power 35 (2019) [110] Z. Eckert, N. Tsolas, K. Togai, A. Chernukho,
190–200. R.A. Yetter, I.V. Adamovich, Kinetics of plas-
[96] S. Barbosa, G. Pilla, D. Lacoste, P. Scouflaire, ma-assisted oxidation of highly diluted hydrocar-
S. Ducruix, C. Laux, D. Veynante, Influence of bon mixtures excited by a repetitive nanosecond
nanosecond repetitively pulsed discharges on the pulse discharge, J. Phys. D Appl. Phys. 51 (2018)
stability of a swirled propane/air burner represen- 374002.
tative of an aeronautical combustor, Philos. Trans. [111] S. Nagaraja, V. Yang, A ‘frozen electric-field’ ap-
R. Soc. A 373 (2015) 20140335. proach to simulate repetitively pulsed nanosecond
[97] F. Tholin, D. Lacoste, A. Bourdon, Influence of plasma discharges and ignition of hydrogen–air
fast-heating processes and o atom production by a mixtures, J. Phys. D Appl. Phys. 47 (2014) 385201.
nanosecond spark discharge on the ignition of a [112] T.A. Casey, J. Han, M. Belhi, P.G. Arias, F. Bisetti,
lean H2-air premixed flame, Combust. Flame 161 H.G. Im, J.-Y. Chen, Simulations of planar
(2014) 1235–1246. non-thermal plasma assisted ignition at atmo-
[98] S. Yang, S. Nagaraja, W. Sun, V. Yang, Multiscale spheric pressure, Proc. Combust. Inst 36 (2017)
modeling and general theory of non-equilibrium 4155–4163.
plasma-assisted ignition and combustion, J. Phys. D [113] N. Deak, A. Bellemans, F. Bisetti, Plasma-assisted
Appl. Phys. 50 (2017) 433001. ignition of methane/air and ethylene/air mixtures:
[99] Y. Wang, P. Guo, H. Chen, Z. Chen, Numeri- efficiency at low and high pressures, Proc. Combust.
cal modeling of ignition enhancement by repetitive Inst 38 (2021) 6551–6558.
nanosecond discharge in a hydrogen/air mixture ii: [114] D. Lacoste, J. Moeck, D. Durox, C. Laux,
forced ignition, J. Phys. D Appl. Phys. 54 (2021) T. Schuller, Effect of nanosecond repetitively
065502. pulsed discharges on the dynamics of a swirl - sta-
[100] M. Capitelli, C.M. Ferreira, B.F. Gordiets, A.I. Os- bilized lean premixed flame, J. Eng. Gas Turbines
ipov, Plasma Kinetics in Atmospheric Gases, Power 135 (2013) 101501.
Springer International Publishing, Switzerland, [115] T.C. Lieuwen, M. Chang, A. Amato, Stationary
2000. gas turbine combustion: technology needs and pol-
[101] I.A. Kossyi, A.Y. Kostinsky, A.A. Matveyev, icy considerations, Combust. Flame 160 (2013)
V.P. Silakov, Kinetic scheme of the non-equilib- 1311–1314.
rium discharge in nitrogen-oxygen mixtures, Plasma [116] T. Poinsot, Prediction and control of combustion
Sources Sci. Technol. 1 (3) (1992) 207–220. instabilities in real engines, Proc. Combust. Inst. 36
[102] X. Mao, A. Rousso, Q. Chen, Y. Ju, Numerical (2017) 1–28.
modeling of ignition enhancement of ch4 /o2 /he [117] E.N. Volkov, V.N. Kornilov, L.P.H. de Goey, Exper-
mixtures using a hybrid repetitive nanosecond imental evaluation of dc electric field effect on the
and dc discharge, Proc. Combust. Inst. 37 (2019) thermoacoustic behaviour of flat premixed flames,
5545–5552. Proc. Combust. Inst. 34 (2013) 955–962.
[103] L. Cheng, N. Barleon, B. Cuenot, O. Vermorel, [118] J. Choe, W. Sun, Blowoff hysteresis, flame morphol-
A. Bourdon, Plasma assisted combustion of ogy and the effect of plasma in a swirling flow, J.
methane-air mixtures: validation and reduction, Phys. D Appl. Phys. 51 (2018) 365201.
Combust. Flame 240 (2022) 111990. [119] Y. Tang, J. Zhuo, W. Cui, S. Li, Q. Yao,
[104] M. Uddi, N. Jiang, E. Mintusov, I. Adamovich, W. Non-premixed flame dynamics excited by flow
Lempert, Atomic oxygen measurements in air and fluctuations generated from dielectric-barrier-dis-
air/fuel nanosecond pulse discharges by two photon charge plasma, Combust. Flame 204 (2019) 58–67.
laser induced fluorescence, in: Proceedings of the [120] K. Deng, S. Zhao, C. Xue, J. Hu, Y. Zhong,
46st AIAA ASM, 8–12 Jan. 2008, Reno, NV, vol- Y. Zhong, Combustion instability of swirl premixed
ume AIAA 2008-1110, 2008. flame with dielectric barrier discharge plasma, Pro-
[105] M. Uddi, N. Jiang, I. Adamovich, W. Lempert, Ni- cesses 9 (8) (2021) 1405.
tric oxide density measurements in air and air/fuel [121] A.M. Alkhalifa, A. Alsalem, D.D. Cont-Bernard,
nanosecond pulse discharges by laser induced fluo- D.A. Lacoste, Active control of thermoacoustic
rescence, J. Phys. D Appl. Phys. 42 (2009) 075205. fluctuations by nanosecond repetitively pulsed glow
[106] I. Adamovich, I. Choi, N. Jiang, J.-H. Kim, S. Ke- discharges, Proc. Combust. Inst. 39(Submitted to).
shav, W. Lempert, E. Mintusov, M. Nishihara, [122] K.R. McManus, T. Poinsot, S. Candel, A review of
M. Samimy, M. Uddi, Plasma assisted ignition and active control of combustion instabilities, Prog. En-
high-speed flow control: non-thermal and thermal ergy Combust. Sci. 19 (1) (1993) 1–29.
effects, Plasma Sources Sci. Technol. 18 (034018) [123] O. Bölke, D.A. Lacoste, J.P. Moeck, Low-frequency
(2009) 13pp. sound generation by modulated repetitively pulsed
[107] T. Ombrello, S. Won, Y. Ju, S. Williams, Flame nanosecond plasma discharges, J. Phys. D Appl.
propagation enhancement by plasma excitation of Phys. 51 (2018) 305203.
oxygen. part i: effects of o3 , Combust. Flame 157 [124] S. Ducruix, D. Durox, S. Candel, Theoretical, and
(2010) 1906–1915. experimental determinations of the transfer func-
[108] T. Ombrello, S. Won, Y. Ju, S. Williams, Flame tion of a laminar premixed flame, Proc. Combust.
propagation enhancement by plasma excitation of Inst. 28 (1) (2000) 765–773.
oxygen. part i: effects of o2 (a1 δg ), Combust. Flame
157 (2010) 1916–1928.
10.1016/j.proci.2022.06.025
JID: PROCI
[125] B. Cosic, J.P. Moeck, C.O. Paschereit, Nonlinear in- [133] Y. Tang, Q. Yao, J. Zhuo, S. Li, Plasma-assisted
stability analysis for partially premixed swirl flames, pyrolysis and ignition of pre-vaporized n-heptane,
Combust. Sci. Technol. 186 (2014) 713–736. isooctane and n-decane, Fuel 289 (2021) 119899.
[126] I. Hurle, R. Price, T. Sugden, A. Thomas, Sound [134] R. Feng, J. Li, Y. Wu, M. Jia, D. Jin, Ignition and
emission from open turbulent premixed flames, blow-off process assisted by the rotating gliding arc
Proc. R. Soc. Lond. Ser. A 303 (1968) 409–427. plasma in a swirl combustor, Aerosp. Sci. Technol.
[127] R. Price, I. Hurle, T. Sugden, Optical studies of the 99 (2020) 105752.
generation of noise in turbulent flames, Symp. (Int.) [135] M. Sugimoto, K. Maruta, K. Takeda, O.P. Solo-
Combust. 12 (1) (1971) 1093–1102. nenko, M. Sakashita, M. Nakamura, Stabilization
[128] S.A. Heitz, J.P. Moeck, T. Schuller, D. Veynante, of pulverized coal combustion by plasma assist,
D.A. Lacoste, Influence of air flow parameters Thin Solid Films 407 (2002) 186–191.
on nanosecond repetitively pulsed discharges in [136] P.M. Kanilo, V.I. Kazantsev, N.I. Rasyuk,
a pin-annular electrode configuration, J. Phys. D K. Schünemann, D.M. Vavriv, Microwave plasma
Appl. Phys. 49 (2016) 155205. combustion of coal, Fuel 82 (2003) 187–193.
[129] A.H. Lefebvre, D.R. Ballal, Gas Turbine Combus- [137] M.A. Gorokhovski, Z. Jankoski, F.C. Lock-
tion - Alternative Fuels and Emissions, third, CRC wood, E.I. Karpenko, V.E. Messerle,
Press, Taylor and Francis Group, Boca Raton, FL, A.B. Ustimenko, Enhancement of pulver-
USA, 2010. ized coal combustion by plasma technology,
[130] W. Kim, H. Do, M.G. Mungal, M.A. Cappelli, In- Combust. Sci. Technol. 179 (2007) 2065–
vestigation of no production and flame structure 2090.
in plasma enhanced premixed combustion, Proc. [138] J. Choe, W. Sun, T. Ombrello, C. Carter, Plasma as-
Combust. Inst. 31 (2007) 3319–3326. sisted ammonia combustion: simultaneous nox re-
[131] R.A. Varella, J.C. Sagas, C.A. Martins, Effects of duction and flame enhancement, Combust. Flame
plasma assisted combustion on pollutant emissions 228 (2021) 430–432.
of a premixed flame of natural gas and air, Fuel 184 [139] G. Faingold, J.K. Lefkowitz, A numerical investiga-
(2016) 269–276. tion of NH3/O2/he ignition limits in a non-thermal
[132] Y. Xiong, O. Schulz, C. Bourquard, M. Weilen- plasma, Proc. Combust. Inst. 38 (2021) 6661–6669.
mann, N. Noiray, Plasma enhanced auto-ignition [140] M. Mlotek, M. Perron, K. Krawczyk, Ammonia de-
in a sequential combustor, Proc. Combust. Inst. 37 composition in a gliding discharge plasma, Energy
(2019) 5587–5594. Technol. 9 (2021) 2100677.
10.1016/j.proci.2022.06.025

2022 - Flames With Plasmas

Uploaded by

Copyright:

Available Formats

2022 - Flames With Plasmas

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2022 - Flames With Plasmas

Uploaded by

Copyright:

Available Formats

JID: PROCI

ARTICLE IN PRESS [mNS;July 19, 2022;18:14]

Available online at www.sciencedirect.com

Proceedings of the Combustion Institute 000 (2022) 1–24

Flames with plasmas

Received 29 January 2022; accepted 25 June 2022

1. Introduction chemistry, the temperature, or the transport prop-

2 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 3

Dielectric barrier discharges (DBD) are pro-

4 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 5

6 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 7

8 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 9

10 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 11

12 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 13

14 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 15

16 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 17

pressure was chosen because it is representative

18 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 19

Declaration of Competing Interest

20 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 21

22 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx 23

24 D.A. Lacoste / Proceedings of the Combustion Institute xxx (xxxx) xxx

You might also like