Available online at www.sciencedirect.

com
Proceedings
of the
Combustion
Institute
Proceedings of the Combustion Institute 33 (2011) 3225–3232
www.elsevier.com/locate/proci

Pure rotational CARS studies of thermal energy release

and ignition in nanosecond repetitively
pulsed hydrogen-air plasmas
Y. Zuzeek, S. Bowman, I. Choi, I.V. Adamovich, W.R. Lempert *
Michael A. Chaszeyka Nonequilibrium Thermodynamics Laboratories, Department of Mechanical
Engineering, The Ohio State University, Columbus, OH 43210, USA

Available online 1 September 2010

Abstract

Pure rotational CARS thermometry, complemented by UV emission measurements and ICCD imaging,
is used to study kinetics of low temperature plasma assisted fuel oxidation and ignition in a repetitive nano-
second pulse discharge in hydrogen-air mixtures, with number of pulses in a 40 kHz burst varying from a
few to a few hundred. Time-resolved OH emission, coupled with gated ICCD images of the plasma and the
flame, demonstrate that volumetric ignition of H2–air mixtures occurs in a spatially uniform plasma. The
results are shown to agree well with predictions of a new hydrogen-air plasma chemistry model, which
incorporates non-equilibrium plasma processes, H2–air chemistry, non-empirical scaling of nanosecond
pulse energy coupled to the plasma, and quasi-one-dimensional conduction heat transfer. In particular,
the results demonstrate that the heating rate in low temperature hydrogen-air plasmas is much faster than
in air plasmas, primarily due to energy release from exothermic reactions of fuel with O and H atoms gen-
erated in the plasma. Kinetic sensitivity analysis is used to identify dominant plasma and chemical pro-
cesses of hydrogen oxidation, demonstrating that additional heat release in these reactions is a key
factor in ignition kinetics. Kinetic modeling calculations demonstrate that removal of the radical genera-
tion processes by the nanosecond pulsed plasma from the model completely blocks subsequent exothermic
chemical reactions, thus making ignition impossible.
Ó 2010 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Keywords: Plasma assisted combustion; CARS spectroscopy; Nanosecond pulse; Ignition

1. Introduction combustion (PAC) [1]. One particular approach

The last decade has seen greatly increased
activity in the discipline known as plasma assisted
*
Corresponding author. Address: Department of
Mechanical Engineering, The Ohio State University,
E445 Scott Laboratory, 201 W. 19th Ave., Columbus,
OH 43210, USA. Fax: +1 614 292 3163.
E-mail address: lempert.1@osu.edu (W.R. Lempert).
utilizes high peak voltage nanosecond pulse dura-
tion discharges, operated at high pulse repetition
rate, on the order of 10–100 kHz. Such discharges
have inherently high reduced electric fields, E/N,
up to several hundred Townsend (1 Td = 1017
V cm2), such that a significant fraction of the total
discharge energy goes into molecular dissociation
and population of excited electronic states.
Repetitively pulsed nanosecond discharges also
exhibit stability at significantly higher pressures

1540-7489/$ - see front matter Ó 2010 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
doi:10.1016/j.proci.2010.05.100
3226 Y. Zuzeek et al. / Proceedings of the Combustion Institute 33 (2011) 3225–3232
compared to other types of electric discharges,
due to the very low duty cycle operation [2,3].
Several groups have recently reported the use
of nanosecond pulse discharges for fundamental
PAC studies [4]. Bozhenkov et al. [5] have shown
more than one order of magnitude ignition delay
time reduction in H2/Ar/air mixtures preheated
in a shock tube. Other recent examples include
large increases in the critical co-flow velocity [6]
and NO production rate [7] in methane–air PAC
flames, and fuel lean flame stability limit extension
[8] in premixed propane–air flames, to / = 0.3.
Uddi et al. [9] have used two photon absorption
laser induced fluorescence (TALIF) to study low
temperature generation and loss kinetics of atomic
oxygen in air, methane–air, and ethylene–air mix-
tures at 60 Torr. The results were found to agree
well with predictions of a hydrocarbon–air plasma
chemistry model, supplemented by GRI Mech 3.0
[10] and the hydrocarbon oxidation mechanism
of Wang et al. [11]. Finally, Zuzeek et al. [12] have
recently presented time-resolved Coherent Anti-
Stokes Raman Spectroscopy (CARS) measure-
ments of heat release in air and ethylene–air mix-
tures. Results were found to agree well with
predictions of the plasma chemistry kinetic model
of Uddi et al. [9], modified to incorporate a new
model for energy coupling in repetitively pulsed
nanosecond plasmas [13], as well as quasi-one-
dimensional heat transfer.
This paper presents new experimental CARS
thermometry results for heat release in nanosec-
ond repetitively pulsed H2–air discharges [14].
Sensitivity analysis predicts that the initial heating
rate, at temperatures of approximately 300 K, is
controlled by the low temperature processes
O + HO2 ? OH + O2 and OH + H2 ? H2O +
H, where HO2 is formed by three body recombi-
nation of O and H2. At intermediate tempera-
tures, 500–600 K, OH formation from chain
branching processes increases, with rapid concur-
rent increase in heat release due to plasma chem-
ical reactions, leading to rapid temperature rise
and, in some cases, volumetric ignition.
2. Experimental
Experiments are conducted in a 220 mm
long  22 mm width  10 mm height quartz
channel/plasma flow reactor [9,12], at 40 Torr
pressure and flow velocity 0.8 m/s, controlled
by mass flow controllers. Rectangular copper
plate electrodes, 14 mm wide by 65 mm long and
rounded at the corners to reduce the electric field
nonuniformity, are flush mounted to the top and
bottom of the 1.75 mm thick quartz channel,
which acts as a dielectric barrier. A Chemical
Physics Technologies power supply produces
20 kV peak voltage, 25 ns duration pulses at
40 kHz repetition rate. For all measurements pre-
sented in this paper, the pulser was operated in a
repetitive burst mode, generating sequences of up
to 1000 pulses at pulse repetition rate of 40 kHz
and burst repetition rate of 10 Hz. This rate
matches the repetition rate of the CARS system,
and ensures that each gas sample in the flowing
discharge cell experiences only a single burst. To
produce breakdown in the discharge on the first
pulse, the test cell was irradiated by a deuterium
UV lamp, providing pre-ionization. The equiva-
lence ratio in air–fuel mixtures was varied from
/ = 0.05 to 1.0.
The pure rotational CARS apparatus, pat-
terned after that of Alden et al. [15], has been
described in detail previously and will be only
summarized here. A broad-band Ti:sapphire laser
[16] creates two, orthogonally polarized, 15 ns,
30 mJ/pulse, CARS “pump” beams at 780
nm, which are overlapped in a focused planar
phase matching geometry with a 532 nm “probe”
beam, using a 400 mm focal distance lens. Pure
rotational CARS is generated with vertical polar-
ization, which allows for discrimination from
stray horizontally polarized probe light using a
polarizer. As detailed in [12] this results in
detected CARS resonant and non-resonant inten-
sities which are a factor of 9/16 and 1/9, respec-
tively, of the values which would be obtained if
all polarizations were aligned parallel [17]. The
CARS beam is spectrally resolved (resolution
1–2 Å) with a ½ m spectrometer (1800 lines/
mm grating), and a gated ICCD camera. Typi-
cally, for burst durations of 10 ms or less, the
CARS signal was averaged up to 600 bursts
(60 s). For longer bursts, five such spectra were
added together.
Inference of rotational temperature and O2
mole fraction is performed using the Sandia
National Laboratories CARS code [18]. Prior to
fitting, experimental spectra are corrected to
account for the finite spectral width (400 cm1
– FWHM) of the broad-band Ti:sapphire pump
beams, as described in [12]. CARS spectra are
modeled as pure air since the H2 (S(0)) transition
falls outside the spectral region of nitrogen and
oxygen. Uncertainty in rotational–translational
temperatures, as discussed in Section 4, is gener-
ally found to be in the range 20–200 K.
In order to detect ignition and gauge discharge
uniformity, time-resolved emission measurements
and plasma imaging have been performed. Emis-
sion measurements detect time-resolved OH emis-
sion with a photomultiplier tube and bandpass
filter centered at 310 ± 2 nm, captured with a dig-
ital oscilloscope. Time response of emission mea-
surements (10 ls) was controlled using a 50 kX
termination resistor. Broad-band visualization
images are obtained with a gated ICCD camera
and UV lens.
 Y. Zuzeek et al. / Proceedings of the Combustion Institute 33 (2011) 3225–3232
3. Hydrogen-air nanosecond pulsed discharge and
plasma chemistry model
The hydrogen-air nanosecond pulsed dis-
charge model used in this work is patterned
after that described by Uddi et al. [9,19]. Briefly,
the present model incorporates a non-equilib-
rium air plasma chemistry model [20], expanded
to include hydrogen dissociation processes in the
plasma, and the hydrogen–oxygen chemistry
model (22 reactions among H, O, OH, H2, O2,
H2O, HO2, and H2O2) developed by Popov
[14].
Species concentration equations are coupled
with the two-term expansion Boltzmann equation
for the energy distribution function of plasma
electrons, with electron impact cross sections
taken from [21–23]. The full list of air plasma pro-
cesses and their rates is given in [19]. The list of
hydrogen–oxygen chemical reactions and their
rates is given in [14].
As in our previous work [12], the pulse shape
used by the plasma chemistry model is a Gaussian
fit to a negative polarity voltage pulse. However,
as described in a recent paper by Adamovich et
al. [13] due to strong shielding of the applied volt-
age caused by charge accumulation on the quartz
channel walls, the field in the plasma after break-
down is much lower than the applied field. Sum-
marizing the results of [13], the model predicts
that energy coupled to the plasma during an indi-
vidual nanosecond discharge pulse is controlled
by the capacitance of the dielectric layers, C,
and by the breakdown voltage, Vb, so that
qpulse  CV 2b /2, with the energy coupled per mole-
cule being nearly independent of the number den-
sity. The coupled pulse energy predicted by the
nanosecond pulse discharge model in room tem-
perature (300 K) air was found to agree well with
the value inferred from previous TALIF measure-
ments, 0.76 mJ/pulse (0.28 meV/molecule/pulse)
[9,19]. Gradual plasma temperature rise during a
pulse burst results in breakdown voltage reduc-
tion, thus decreasing the energy coupled to the
plasma approximately linearly with density. In
the present model, the pulse peak reduced electric
field, (E/N)peak  250 Td, was adjusted to keep
the coupled pulse energy per molecule constant
(0.28 meV) during the burst. At these conditions,
almost 50% of the coupled pulse energy in a
hydrogen-air mixture at / = 0.5 is spent on gener-
ation of O and H atoms.
As described in our previous work [12],
energy loss due to conduction to the quartz
channel walls, which is very significant in hydro-
gen-air mixtures, is described by incorporating a
conduction heat transfer term into the energy
equation [13]. Finally, it is noted that UV pre-
ionization produces very low initial electron
density, which has negligible effect on the model
predictions.
3227
4. Results and discussion
4.1. Discharge emission and imaging measurements
Figure 1 shows an example OH emission data
trace obtained in stoichiometric H2–air at 70 Torr
pressure and 40 kHz burst rate. As can be seen in
the upper left inset, which is an expanded time
scale spanning the first 150 ls, in the absence of
ignition a series of sharp intensity spikes spaced
25 ls apart (40 kHz rate) is observed, which fall
to near zero in the interval between successive
pulses. In the event of ignition, an OH intensity
“footprint” appears, such as that seen at 10
ms, which does not decay between the pulses,
and which serves as an ignition marker [2]. As pre-
sented in [24], in this manner ignition has been
observed, and ignition delay determined, in H2–
air mixtures over a range of equivalence ratios
from u = 0.3 to 1, and pressures from 50 to
100 Torr.
In order to confirm diffuse volumetric ignition,
and to justify the use of a quasi-one-dimensional
approximation for kinetic and heat transfer anal-
ysis, UV ICCD camera images have been
obtained (primarily N2 second positive band
and, in some cases, OH A ? X emission) as a
function of pressure, burst duration, and pulse
number within, and/or after, a burst. As a repre-
sentative example, Fig. 2 shows a collage of six
images obtained from a u = 1 H2–air mixture at
40 Torr, where the emission fills the entire volume
defined by the electrode area and channel height.
The discharge pulse repetition rate is 40 kHz and
the images shown are obtained during a burst of
pulses, with pulse numbers in the burst ranging
from 1 to 700 (17.5 ms). Note that the individual
images shown were obtained from different bursts
Emission (arb. units)
0 5 10 15 20
Time, msec
Fig. 1. Time-resolved OH emission for stoichiometric
H2–air mixture at 70 Torr, excited by a nanosecond
pulse burst discharge (m = 40 kHz). Inset is expanded
time scale spanning first 150 ls.
3228 Y. Zuzeek et al. / Proceedings of the Combustion Institute 33 (2011) 3225–3232

Fig. 2. ICCD images of the plasma and the flame in H2–air excited by a m = 40 kHz discharge at / = 1 and P = 40 Torr.
Emission spans entire discharge volume.

and do not, therefore, illustrate discharge devel-
opment during a single burst.
The top four images in Fig. 2 were obtained
with a 2 ls intensifier gate, which was timed to
include the 25 ns discharge event. While some
filamentary structure is observed for the first
pulse, all images for pulse numbers exceeding
2–4 are observed to be uniform, including
images obtained for pulse numbers up to 1000
(not shown in Fig. 2). This result is in contrast
to our previous nanosecond pulse plasma images
obtained in ethylene–air mixtures, in which large
scale filaments were detected after 50–100
pulses, depending upon equivalence ratio [12,24].
While some evidence of filamentation was
observed after 100–200 pulses at pressures of
70 Torr or higher, these structures gradually dif-
fuse, resulting in uniform plasmas after 300
pulses (7.5 ms), i.e. before ignition occurred (see
Fig. 1) [24,25]. High thermal diffusivity of hydro-
gen, as well as its relatively high ionization poten-
tial (15.6 eV), compared to ethylene (10.5 eV), is
the most likely the explanation of the observed
improvement in discharge stability.
The bottom two images in Fig. 2 (labeled
“Flame”) were taken with an 18 ls gate and timed
to collect only emission that occurs after the dis-
charge emission has decayed (3 ls delay after the
discharge pulse). The observed uniform emission
(principally OH) is further evidence that volumet-
ric ignition has occurred in this case after 10 ms.
Figure 3 is similar to Fig. 2, except that the images
shown were obtained at 100–3000 ls (with a
100 ls gate) after the final pulse in a 15 ms
(600 pulse) burst for u = 1 at 70 Torr. Uniform
OH emission, again spanning the complete dis-
charge volume, is clearly evident, indicating con-
tinuous, combustion-induced, emission for a
period of several milliseconds after application
of the final discharge pulse.
4.2. Pure rotational CARS
CARS is a well known diagnostic technique for
high spatial and temporal resolution measurement
of temperature and major species distributions in
flames and plasmas, and is ideally suited as a
method to provide more quantitative data which

Fig. 3. ICCD images in / = 1 H2–air at 70 Torr, obtained at differing times after final pulse of 15 ms 40 kHz burst.
Emission spans entire discharge volume.
 Y. Zuzeek et al. / Proceedings of the Combustion Institute 33 (2011) 3225–3232
can be benchmarked against plasma chemical oxi-
dation kinetic modeling codes. Figure 4 shows a
comparison of experimental CARS temperatures
obtained in this work as a function of burst dura-
tion at 40 kHz discharge repetition rate in air and
in H2–air at / = 0.05, 0.5, and 1.0, along with pre-
dictions from the plasma kinetic modeling code.
In all cases, P = 40 Torr and the initial tempera-
ture is T = 300 K. Note that the raw CARS spec-
tra are similar in intensity and precision to those
of our previous work in ethylene–air mixtures
[12] and, therefore, will not be reproduced here.
Note also that the Sandia CARS code does not
provide any estimate of uncertainty in the fitting
parameters. To provide an estimate of experimen-
tal precision, ten spectra were obtained and fit for
each measurement condition presented in Fig. 4.
The statistical uncertainty (95% confidence inter-
val) in the mean, defined as 2.23 (“t” value) times
the experimental standard deviation, divided by
the square root of 10, is shown for representative
data in Fig. 4.
Focusing first on the experimental data in
Fig. 4, it can be seen that the initial heating rate,
for burst durations less than 3–4 ms (120–160
pulses), is approximately equal for each of the
mixtures to within the experimental uncertainty.
For burst durations exceeding approximately 5–
7 ms (200–280 pulses) clear trends become dis-
cernable. In particular, the rate of heating for mix-
tures at u = 0.5 and 1.0 can be clearly seen to
exceed that for pure air and the u = 0.05 mixture.
A somewhat closer examination of the data in
Fig. 4 shows that accelerated heating for the
u = 0.5 and 1.0 mixtures begins to occur at tem-
peratures of approximately 500–600 K. This will
be discussed in more detail in Section 5.
Number of pulses
T, K 0 200 400 600 800 1000 1200
1200 P=40 torr
φ=0 (air)
φ=0.05
900
φ=0.5
φ=1
600
300
0 5 10 15 20 25 30
Time, msec
Fig. 4. Comparison of CARS temperatures in a repet-
itively pulsed nanosecond discharge in air and in
hydrogen-air at P = 40 Torr with kinetic model predic-
tions. m = 40 kHz, burst repetition rate 10 Hz.
3229
Fig. 4 also shows that at burst durations in the
approximate range 15–20 ms (600–800 pulses), the
experimental CARS temperatures (as well as the
modeling predictions) for both u = 0.5 and 1 exhi-
bit a maximum, and then decay. Furthermore, as
shown in Fig. 5, the O2 mole fractions for the stoi-
chiometric mixture, obtained from the Sandia
CARS code (which floats this parameter) decrease
very rapidly in the region of 15–20 ms duration
bursts. Figure 5 also shows that the O2 mole frac-
tion decay agrees well with the kinetic model pre-
dictions, which also predicts large concurrent
decrease in H2 mole fraction. The observed tem-
perature maximum, coupled with the observed
rapid loss in O2, is a very strong indication that
ignition, accompanied by almost complete fuel
oxidation, has occurred in these mixtures, a result
which is consistent with the emission data and
images shown in Figs. 1–3.
While the experimental temperature rise in the
region of 15–20 ms is more gradual then that pre-
dicted, Figs. 4 and 5 show good overall agreement
between the model and the experiments. It is
believed, although not definitely determined, that
the burst to burst reproducibility of the ignition
event is not highly reproducible in time, leading
to averaging of the temperature profile in the
vicinity of the maximum. Nonetheless the overall
quantitative agreement demonstrates that the
plasma kinetic model adequately describes the
energy balance in the repetitively pulsed nanosec-
ond discharge, including the energy coupling to
the plasma, energy release during exothermic
plasma chemical reactions, and energy loss by
heat transfer to the test section walls. It is stressed
that this level of agreement is attained without
recourse to any adjustable parameters in the
model.
Mole fractions
P=40 torr, φ =1
0.3 O 2 , CARS spectra
O 2 , model
H 2 , model
0.2
0.1
0.0
0 5 10 15 20 25 30
Time, msec
Fig. 5. Experimental and calculated O2 mole fractions at
the conditions of Fig. 4. Also shown are calculated H2
mole fractions for the same conditions.
3230 Y. Zuzeek et al. / Proceedings of the Combustion Institute 33 (2011) 3225–3232
5. Kinetic sensitivity analysis
To provide insight into the chemical processes
which are primarily responsible for the observed
low temperature oxidation, heat release, and igni-
tion in hydrogen-air plasmas, a sensitivity analysis
has been performed, the principal results of which
will be summarized below. A more complete anal-
ysis can be found in [24,25].
At the lowest temperatures relevant to this
work, 300–600 K, a detailed analysis is given
in [25]. Summarizing it is found that at 300 K
plasma chemical oxidation proceeds, principally,
by the following mechanism. First, O and H are
formed during, and after, the discharge pulse
(on a time scale of 100–1000 ns) primarily via
electron impact and dissociating collisions with
electron metastable states of N2. Subsequent fuel
oxidation during the post-discharge stage (on a
time scale from 1 ls to 1 ms) occurs primarily
via a three process sequence,
H þ O2 þ M ! HO2 þ M ð1Þ
O þ HO2 ! OH þ O2 ð2Þ
OH þ H2 ! H þ H2 O ð3Þ
In other words, at this low temperature, OH for-
mation via the well known “high temperature”
chain processes,
Oð3 PÞ þ H2 ! H þ OH ð4Þ
H þ O2 ! O þ OH ð5Þ
is extremely slow, while processes (2) and (3) are
relatively fast, with activation barriers of
14.35 kJ/mol (1660 K) and essentially zero,
respectively [14].
As the initial temperature is increased, to 500–
600 K, the rates of processes (4) and (5), as well as
of the following three processes,
H þ HO2 ! H2 O þ O ð6Þ
H þ HO2 ! OH þ OH ð7Þ
H þ HO2 ! H2 þ O2 ð8Þ
increase rapidly, greatly accelerating the rate of
overall plasma chemical oxidation.
More quantitatively, modeling calculations
predict that at 300 K the total H2O produced
from a single discharge pulse is essentially identi-
cal to the atomic oxygen initially produced by
the discharge. This is clear from inspection of pro-
cesses (1)–(3), which do not produce any chain
propagation or branching. Indeed, the reaction
path (1)–(3) is equivalent to a net reaction
O + H2 ? H2O. However, modeling calculations
predict that at 600 K the rate of reactions (4)
and (7) increase rapidly, increasing the amount
of H2O produced by a single discharge pulse by
approximately one order of magnitude. The effect
of this on plasma chemical heat release can be
seen in Fig. 6, which shows model predictions
for total thermalized energy from a single dis-
charge pulse as a function of initial temperature.
At 300 K the additional thermalized energy
(beyond the 0.28 meV/molecule coupled directly
from the discharge) is minimal, approximately
10%. However, as the temperature is increased,
the additional thermal energy due to net exother-
mic plasma chemical reaction increases rapidly,
reaching a factor of approximately three greater
than the pulse input energy at 600 K. It is this
rapid increase in plasma chemical oxidation which
causes the observed rapid temperature rise of the
u = 0.5 and 1.0 mixtures, relative to u = 0.05
and pure air, observed in Fig. 4 after 5 ms
(200 pulses), and leads, ultimately, to the observed
ignition.
Figures 7 and 8 summarize the most significant
sensitivity analysis results for burst mode kinetics
and oxidation. Figure 7 plots time-resolved tem-
peratures predicted by the model at u = 1.0,
P = 40 Torr using both the full H2–air chemistry
model, and a reduced kinetic mechanism consist-
ing of reactions (1)–(8) plus the additional
process,
OH þ HO2 ! H2 O þ O2 ð9Þ
The full and reduced kinetic mechanisms result
in very similar predictions, both of which agree
well with the experimental data. Figure 8 shows
predictions of the full kinetic model for the tem-
poral evolution of key species for the same condi-
tions as Fig. 7. It can be seen that H, O, and OH
mole fractions are predicted to peak at 18 ms,
consistent with the observed maximum in the
experimental CARS temperatures observed in
Fig. 4. In addition, while not plotted here, it is
shown in [24] that the reduced kinetic model
Energy, meV/molecule
1.5
H2-air, P=40 torr
1.2 input
output, T=300 K
output, T=400 K
0.9 output, T=500 K
output, T=600 K
0.6
0.3
0.0
1E-8 1E-7 1E-6 1E-5 1E-4 1E-3 1E-2 1E-1 1E+0
Time, sec
Fig. 6. Thermalized energy per molecule vs. time after a
single-pulse nanosecond discharge. Hydrogen-air, / = 1,
P = 40 Torr, initial temperature T = 300–600 K. Energy
input by the discharge is 0.28 meV/molecule.
 Y. Zuzeek et al. / Proceedings of the Combustion Institute 33 (2011) 3225–3232 3231

Number of pulses charge input power is the same. It can be seen

T, K 0 200 400 600 800 1000 1200 clearly that removal of the radical generation pro-
1200
cesses in the pulsed discharge plasma completely
blocks subsequent exothermic chemical reactions
and makes ignition impossible.

900
6. Summary and conclusions

This paper presents, for the first time to our

600 knowledge, quantitative data on the kinetics of
φ=1 heat release in low temperature, non-equilibrium
full set plasma assisted fuel oxidation. Specifically, pure
300 reduced set rotational CARS is used to obtain time-resolved
no O and H generation by plasma rotational temperature and O2 mole fraction in a
repetitively pulsed discharge operated in burst
0 5 10 15 20 25 30 mode in air and in hydrogen-air mixtures at
Time, msec 40 Torr pressure and 40 kHz discharge repetition
rate, conditions in which UV ICCD camera
Fig. 7. Temperatures predicted in a nanosecond pulsed images show that the discharge remains stable
discharge burst (H2–air, / = 1, P = 40 Torr, and diffuse. In particular, it is found that for
m = 40 kHz) for full and reduced reaction sets, and with u = 0.5 and 1.0 mixtures at 40 Torr the CARS
O and H generation processes in the plasma removed. temperatures exhibit a clear maximum after
CARS temperatures shown for comparison.
600–800 discharge pulses, indicating that igni-
tion has occurred for this condition. Time-
resolved OH emission, coupled with gated ICCD
Species mole fractions images of the plasma and the flame, demonstrate
1.0E+0
H2 that volumetric ignition of H2–air mixtures occurs
H2O
in a spatially uniform plasma. The experimental
1.0E-1 results agree well with predictions from a new
low temperature H2–air plasma chemistry model
1.0E-2 which incorporates non-equilibrium plasma dis-
H charge processes, low temperature H2–air chemis-
try, non-empirical scaling of nanosecond
1.0E-3 O discharge pulse energy coupled to the plasma as
OH a function of plasma number density, and one-
1.0E-4 dimensional conduction heat transfer. Quantita-
tive agreement is attained without recourse to
1.0E-5 HO2 any adjustable parameters.
Sensitivity analysis indicates that at low initial
1.0E-6 temperatures, heating in excess of that resulting
from direct energy input from the pulser, is con-
0 5 10 15 20 25 30
trolled primarily by a sequence of reactions (1)–
Time, msec (3), all of which are relatively rapid at 300 K. At
higher initial temperatures, 500–600 K, the rate
Fig. 8. Predicted species mole fractions at the conditions of chemical oxidation in reactions (4) and (6)–(8)
of Fig. 7 for full reaction set. increases rapidly, producing chain branching and
additional heat release, approximately a factor
described directly above results in essentially of three greater than the pulse input energy at
indistinguishable results for predicted species 600 K. As the temperature increases, primarily
mole fractions. due to the plasma chemical heat release, ignition
Finally, to illustrate the critical effect of radical occurs by the well known chain mechanism of
generation by the nanosecond discharge, Fig. 7 reactions (4) and (5). Finally, kinetic modeling
also plots time-resolved temperatures in u = 1.0, calculations demonstrate clearly that removal of
P = 40 Torr H2–air predicted by the model using the radical generation processes in the pulsed dis-
only chemistry processes (reactions (1)–(9)), i.e. charge plasma completely blocks subsequent exo-
with all nanosecond discharge plasma processes, thermic chemical reactions and makes ignition
principally O and H formation via both electron impossible.
impact, and dissociating collisions with electron It is anticipated that the results of this work
metastable states of N2, removed. In both cases, will provide a foundation for future experimental
the coupled pulse energy per molecule, i.e. the dis- and computational studies of more complex
3232 Y. Zuzeek et al. / Proceedings of the Combustion Institute 33 (2011) 3225–3232
plasma assisted combustion environments and
geometries, such as those which might be
employed in future advanced concepts for high
speed ignition and flame holding.
Acknowledgements
The authors wish to acknowledge the sponsor-
ship of the US Air Force Office of Scientific Re-
search (Julian Tishkoff – Technical Monitor)
and the National Science Foundation (Phillip
Westmoreland – Technical Monitor) for support
of this research.
References
[1] S.M. Starikovskaia, J. Phys. D: Appl. Phys. 39
(2006) R265–R269.
[2] I.V. Adamovich, I. Choi, N. Jiang, et al., Plasma
Sources Sci. Technol. 18 (2009) 034018–034031.
[3] A. Hicks, S. Tirupathi, N. Jiang, et al., J. Phys. D:
Appl. Phys. 40 (2007) 1408–1415.
[4] A.Y. Starikovski, Proc. Combust. Inst. 30 (2005)
2405–2417.
[5] S.A. Bozhenkov, S.M. Starikovskaya, A. Yu.
Starikovskii, Combust. Flame 133 (2003) 133–146.
[6] W. Kim, H. Do, M.G. Mungal, M. Cappelli, IEEE
Trans. Plasma Sci. 34 (2006) 2545–2551.
[7] W. Kim, H. Do, M.G. Mungal, M. Cappelli, Proc.
Combust. Inst. 31 (2007) 3319–3326.
[8] G. Pilla, D. Galley, D.A. Lacoste, F. Lacas, D.
Veynante, C.O. Laux, IEEE Trans. Plasma Sci. 34
(2006) 2471–2477.
[9] M. Uddi, N. Jiang, E. Mintusov, I.V. Adamovich,
W.R. Lempert, Proc. Combust. Inst. 32 (2009)
929–936.
[10] http://www.me.berkeley.edu/gri_mech/version30/
text30.html, GRI-Mech 3.0.
[11] H. Wang, X. You, A.V. Joshi, S.G. Davis, A.
Laskin, F. Egolfopouls, C.K. Law, <http://ignis.us-
c.edu/USC_Mech_II.html>.
[12] Y. Zuzeek, I. Choi, M. Uddi, I.V. Adamovich,
W.R. Lempert, J. Phys. D: Appl. Phys. 43 (2010)
124001–1240012.
[13] I.V. Adamovich, M. Nishihara, I. Choi I, et al.,
Phys. Plasmas 16 (2009) 113505-1–113505-13.
[14] N.A. Popov, Plasma Phys. Rep. 34 (2008) 376–391.
[15] M. Alden, P.E. Bengtsson, H. Edner, Appl. Opt. 25
(1986) 4493–4500.
[16] N. Finkelstein, J. Gambogi, W.R. Lempert, et al.,
AIAA-94-0492, AIAA 32nd Aerospace Sciences
Meeting and Exhibit, Reno, NV, 1994.
[17] F. Vestin, M. Afselius, P.-E. Bengtsson, Proc.
Combust. Inst. 31 (2007) 833–840.
[18] Code provided by R. Palmer, Sandia National
Laboratories – Combustion Research Facility.
[19] M. Uddi, N. Jiang, I.V. Adamovich, W.R. Lem-
pert, J. Phys. D: Appl. Phys. 42 (2009) 075205–
075222.
[20] I.A. Kossyi, A.Y. Kostinsky, A.A. Matveyev, V.P.
Silakov, Plasma Sources Sci. Technol. 1 (1992) 207–
220.
[21] W. L. Morgan, J.P. Boeuf, L.C. Pitchford, BOLSIG
Boltzmann Solver, Kinema Software, Monument,
CO, 1996.
[22] Y. Itikawa, M. Hayashi, A. Ichimura, et al., J.
Phys. Chem. Ref. Data 16 (1986) 985–1010.
[23] Y. Itikawa, A. Ichimura, K. Onda, et al., J. Phys.
Chem. Ref. Data 18 (1989) 23–42.
[24] Y. Zuzeek, S. Bowman, I. Choi, I. Adamovich, W.
Lempert, AIAA-2010-648, 48th AIAA Aerospace
Sciences Meeting, 4–7 January, Orlando, FL, 2010.
[25] S. Bowman, I. Choi, I. Adamovich, W. Lempert,
AIAA-2010-1590, 48th AIAA Aerospace Sciences
Meeting, 4–7 January, 2010, Orlando, FL.