See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/367090001

A temporally-resolved investigation on energy deposition from laser-induced

plasmas in combustion environments: The pre-breakdown region and
breakdown initiation

Article  in  Combustion and Flame · January 2023

DOI: 10.1016/j.combustflame.2022.112612

CITATIONS READS

0 43

4 authors, including:

Chai Shu Ziqing Zhao

Shanghai Jiao Tong University Shanghai Jiao Tong University
3 PUBLICATIONS   3 CITATIONS    4 PUBLICATIONS   7 CITATIONS   

SEE PROFILE SEE PROFILE

Wendong Wu
Shanghai Jiao Tong University
11 PUBLICATIONS   107 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Laser-induced breakdown spectroscopy View project

All content following this page was uploaded by Wendong Wu on 13 January 2023.

The user has requested enhancement of the downloaded file.

 Combustion and Flame 249 (2023) 112612

Contents lists available at ScienceDirect

Combustion and Flame

journal homepage: www.elsevier.com/locate/combustflame

A temporally-resolved investigation on energy deposition from

laser-induced plasmas in combustion environments: The
pre-breakdown region and breakdown initiation
Shu Chai a, Haimeng Peng a, Ziqing Zhao a, Wendong Wu a,b,∗
a
Institute of Thermal Energy Engineering, School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, PR China
b
Interdisciplinary Research Center for Engineering Science, School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, PR China

a r t i c l e i n f o a b s t r a c t

Article history: Combustion diagnostic techniques based on laser-induced plasmas have recently demonstrated their ca-
Received 1 August 2022 pabilities in the simultaneous measurement of mixture fraction and temperature. Their successful appli-
Revised 27 December 2022
cations rely on the interpretation of energy deposition behaviors during the plasma formation process.
Accepted 28 December 2022
In this work, the energy deposition behaviors of laser-induced breakdown were analyzed quantitatively
with high temporal resolution in methane-air mixture, heated air, and after-burn gas. Using a nanosec-
Keywords: ond laser with multiple longitudinal modes, a novel method was introduced to determine the breakdown
Laser-induced plasmas delay with 0.1 ns resolution, by matching the random temporal features of incident and transmitted laser
Energy deposition pulses shot-to-shot. The laser irradiance at the moment of breakdown and the pre-breakdown energy can
Combustion diagnostics
then be precisely measured. The influence of gas temperature and laser energy on the breakdown delay
LIBS
and pre-breakdown energy was experimentally investigated, demonstrating that the controlling factor of
LIBT
breakdown initiation is energy threshold instead of irradiance threshold under our experimental condi-
tions. The difference between pre-breakdown energy and breakdown threshold was observed and found
to be increasing with gas temperature. Based on the theoretical analysis on electron growth and loss, the
mechanisms behind such observations were analyzed and explained. The physical connections between
monitored quantities in recently developed thermometry methods, i.e., breakdown threshold, breakdown
delay, and deposited energy were clarified. The contributions from elongated breakdown delay in the pre-
breakdown region and reduced absorption efficiency in the post-breakdown region to the total deposited
energy were also discussed. Utilizing temporally resolved methods, this work constitutes a fundamen-
tal study of energy deposition behaviors, focusing on the breakdown initiation in typical combustion
environments, which will further promote the successful implementation of combustion diagnostic ap-
proaches based on laser-induced plasmas.
© 2023 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

1. Introduction induced breakdown spectroscopy (LIBS), has been applied in pre-

mixed [1] and non-premixed [2,3] flames, in practical hydrocar-
The diagnostic methods based on laser-induced plasmas (LIP) bon fueled systems (Liquefied Petroleum Gas [4], biodiesel [5]), in
have seen recent developments in combustion research. Laser- particle-laden reacting flows [6,7], pressurized systems [8,9], and
induced plasmas are formed when the beam from a pulsed laser turbulent flames [10], demonstrating its usefulness in a good range
is focused on a spot, and the local intensity of the electric field of fuels and combustion conditions.
is high enough to atomize and ionize the sample. The atomic It is always desirable that the flame temperature can be mea-
emissions lines from the cooling plasma can be utilized to mea- sured simultaneously with the equivalence ratio. A series of tech-
sure elemental composition; thus, the mixture fraction and equiva- niques have been recently developed, utilizing laser-induced plas-
lence ratio can be further obtained. This technique, known as laser- mas to measure temperature or density. The work by Lee et al.
[11] assumed that the nitrogen signal was inversely proportional
to the local temperature, which was suited for the premixed
∗
Corresponding author at: School of Mechanical Engineering, Room A311, Shang- flame. McGann et al. [12] measured gas density with Direct Spec-
hai Jiao Tong University, 800 Dongchuan Rd, Minhang District, Shanghai, 200240,
tra Matching (DSM) method in a supersonic combustor, which
PR China.
E-mail address: [email protected] (W. Wu). matched an unknown spectrum to an extensive reference spectrum

https://doi.org/10.1016/j.combustflame.2022.112612
0010-2180/© 2023 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
S. Chai, H. Peng, Z. Zhao et al. Combustion and Flame 249 (2023) 112612

database constructed under well-known gas conditions. Zhang ameters of the droplets seems to have more significant effect on
et al. [13] inferred the gas temperature in methane jets from the the deposited energy at lower incident energy. The energy deposi-
full width at half maximum (FWHM) of Hydrogen peak at 656 nm. tion studies in the laser ignition area were performed more under
The above-mentioned methods extracted temperature information elevated pressure [34,35] (increased density) instead of reduced
from the details of plasma radiated spectra. Relying on different pressure (reduced density), and usually in the pre-ignition mix-
physical principles, some other methods can be applied with low- ture [36,37] instead of the high temperature reacting flow. Such re-
cost ungated spectrometers. For example, two laser-induced plas- search shed a light on the energy deposition behaviors in combus-
mas can be generated as sound sources. The local gas temperature tion systems; however, the environment concerned in combustion
between these two plasmas can be determined through the mea- diagnostics is quite different. Thus, this investigation will be per-
surement of the sound speed between them [14,15]. However, the formed under high temperature (reduced gas density) instead of
spatial resolution of such an approach is limited by the spacing elevated pressure (increased gas density). Moreover, even though
between two breakdown spots. the temporal structure of laser pulse and its influence on the en-
The reduced gas density in the high-temperature environment ergy deposition has been recognized [38], temporally resolved en-
has a significant impact on the collision-dominated plasma for- ergy deposition behaviors have only been reported in the air [39],
mation process. Thus, the energy deposition behavior in this pro- and hardly in a quantitative way. The recent work by Williamson
cess is also a reflection of gas density or temperature. The tem- et al. [17] determined the breakdown delay by scattered laser radi-
perature dependency of the breakdown threshold was utilized by ation from plasma and related it to flame temperature. Except for
Kiefer et al. [16] to infer the temperature in a laminar methane- that, to the best of author’s knowledge, the temporally resolved
air flame. The results agreed well with coherent anti-Stokes Ra- energy deposition behaviors in combustion environments have not
man scattering (CARS) measurements. The breakdown threshold been reported.
was determined by incrementing the laser energy until the prob- In this work, we will investigate the energy deposition behav-
ability that breakdown occurs is 50%, which limits the temporal iors in the laser-induced plasma formation process within a broad
resolution of this approach. In a recent study by Williamson et al. temperature and composition region, ranging from combustible
[17], the breakdown threshold was estimated by measuring the mixture under room temperature, heated air, to high tempera-
point in time that breakdown initiates along the intensity gradi- ture after-burn gas. By quantitatively monitoring the energy de-
ent of the laser pulse. Thus, the flame temperature and elemen- position behaviors with a high temporal resolution, the influence
tal composition can be measured simultaneously despite the large of temperature and laser energy on the breakdown initiation will
uncertainty in determining the breakdown delay. Do et al. [18] es- be analyzed. The controlling factor and the mechanisms of break-
tablished a universal correlation between the deposited energy by down will be further investigated. Based on these findings, the
plasma and gas density, which was then successfully applied in physical connections between the measured quantities in differ-
temperature field measurement in a supersonic combustor. Fur- ent thermometry methods, i.e., breakdown threshold, breakdown
thermore, the residual energy of laser pulse, which is the differ- delay, and deposited energy, will be clarified. How the temporal
ence between the incident energy and deposited energy of plasma, structure of energy deposition behaviors translates to these mea-
can also be correlated with gas temperature [19]. Under this prin- sured quantities will be explained, thus establishing the common
ciple, the temperature was measured in heated gas mixtures with physical background for the thermometry methods based on the
compositional correction [20]. Our recent work found the acous- energy deposition behaviors of laser-induced plasmas, and further
tic signal released by laser-induced plasmas is an indication of the promoting their successful implementations.
deposited energy [21], and the acoustic energy has a strong depen-
dency on the flame temperature [22]. Therefore, the acoustic-based 2. Methods
laser-induced breakdown thermometry was introduced. The spatial
and temporal resolution of this approach were found to be compa- The optical setup is shown schematically in Fig. 1. A 532 nm
rable to that of LIBS, then the simultaneous measurement of mix- laser beam with 5 ns full width at half-maximum (FWHM), 0.5
ture fraction and flame temperature has been demonstrated in a mrad divergence, and 9 mm diameter was generated by a Nd:YAG
counter-flow diffusion flame [3]. The abovementioned approaches laser (Quantel Q-smart 850). Using a combination of half-wave
can be easily used in combination with compact spectrometers. plate and polarizer, the laser pulse energy can be varied without
It should be noted that, the measured quantities in these ap- changing other beam properties. The measured shot-to-shot laser
proaches, i.e., the breakdown threshold, the breakdown delay, the energy fluctuation was less than 3%. Using a pair of beam splitters,
residual laser energy, and the acoustic energy, are all lumped quan- the energy and temporal profile of incident pulses can be simul-
tities describing the energy deposition process which has a tem- taneously monitored shot-to-shot, by a pyroelectric energy meter
poral structure. Thus, to understand the physical connections be- (Ophir PE25BF-DIF-C) and a fast photodiode (Thorlabs DET025A,
tween these lumped quantities beyond mere data correlations, and rise/fall time < 0.15 ns), respectively. The beam was focused by a
to establish the physical background of these new thermometry f = 200 mm 1-inch UV fused silica plano-convex lens to generate
approaches, a systematic investigation on temporally resolved en- plasmas. The radius of beam waist in the focal point r f was mea-
ergy deposition behaviors is fundamentally important. sured by a beam profiler (Ophir LT665), yielding 61.8 μm full width
Studies on the energy deposition behaviors have been exten- at 1/e2 (FW/e2 ). The maximal laser energy was approximately 260
sively done in the laser ignition area, where the concerned quanti- mJ. Another plano-convex lens (f = 200 mm) was placed behind
ties are minimal ignition energy (MIE) [23–26], breakdown thresh- the focal point to collimate the diverging laser beam transmitted
old [23,25,27], and deposited energy [23,28,29]. A series of works through the plasma. The energy and temporal profile of transmit-
by Phuoc [30,31] analyzed how the deposited energy was parti- ted laser beam were monitored with a second set of pyroelectric
tioned into different forms, therefore the available energy for igni- energy meter (Ophir PE50BF-DIFH–C) and fast photodiode (Thor-
tion can be predicted. Gebel et al. [32] also studied the proportion labs DET025A). The instrument uncertainty of energy meters and
of deposited energy converted to the blast wave, reporting com- photodiodes approximates 3% and 0.5% at the confidence interval
parable results. Recently, the work by Oliveira et al. [33] showed (CI) of 95%, respectively. Because the energy loss by plasma scat-
that the absorbed energy is also related to the presence of droplets tering [40] and diffraction [41] can be neglected, the deposited en-
in sprays. The absorbed energy fluctuated more under higher inci- ergy within plasmas was approximated by the difference between
dent energy. In a polydisperse spray ignition, the Sauter mean di- the energy of incident and transmitted beam. The plasma thermal

2
S. Chai, H. Peng, Z. Zhao et al. Combustion and Flame 249 (2023) 112612

Fig. 1. The schematic of optical setup.

Table 1
Experimental conditions. The combustible mixture and after-burn gas were stoichiometric.

Gas compositions Volumetric flow rate of CH4 (L/min) Volumetric flow rate of air (L/min) Gas Temperature (K)

Combustible mixture 3 28.56 298±3

Air \ 30 298±3
\ 30 400±2
\ 30 600±4
\ 30 800±3
\ 30 1000±8
After-burn gas 1.05 10 1612±2
1.45 13.80 1693±2
2.11 20.83 1805±2
2.93 27.89 1911±5

radiation was collected in the back-scattering position by a third 600 K, 800 K, and 1000 K. The maximal fluctuation of gas tem-
photodiode covered by a 532 nm notch filter. The signals of all perature was approximately 8 K, which happens to the 10 0 0 K
photodiodes were fed into a digital oscilloscope (Teledyne LeCroy case. Nevertheless, the temperature fluctuations were less than 1%
WavePro 404HD, 4 GHz bandwidth and 10 GS/s sampling rate). The throughout the 400 K to 10 0 0 K range. The high temperature after-
instrument uncertainty of oscilloscope approximates 0.5% of full burn gas was generated using a McKenna flat-flame burner under
scale at 95% CI, which is dominated by DC vertical gain accuracy. atmospheric pressure. The 60 mm-diameter burner head is com-
The laser, energy meters, and oscilloscope were synchronized by a posed of sintered stainless steel and is water-cooled. By changing
digital delay generator (Stanford Research System DG645). the total flow rate of stoichiometrically mixed methane and air, the
The study was performed under different experimental condi- downstream gas temperature in the measurement point was ad-
tions, which include room-temperature synthetic air and methane- justed from 1612 K to 1911 K with a fluctuation less than 5 K. The
air mixture, electrically heated synthetic air (400 K-1000 K), and temperature in the after-burn gas was also measured by the type-
high temperature after-burn gas (1612 K-1911 K). Details of exper- S thermocouple. In this study, the flow rate of methane, nitrogen,
imental conditions were listed in Table 1. Under room tempera- and air were regulated by pre-calibrated mass-flow controllers (Al-
ture, methane and synthetic air were mixed stoichiometrically and icat Scientific) with fluctuations less than 0.3%, 0.3%, and 1%, re-
jetted by a 15 mm stainless-steel nozzle. To homogenize the gas spectively. The plasmas were generated at 5 mm above the exit
flow, a 5 mm-thick sintered porous brass with 30 μm pore size plane of the nozzle, gas heater, or Mckenna burner. The laser was
was flush-mounted at the exit of nozzle. The flow velocity was operated at 2 Hz to avoid the disturbance from the preceding laser
2.98 m/s to quickly blow off the flame ignited by breakdown, thus shot, and an ensemble of 200 shots was sampled in each experi-
ensuring an unburnt mixture for every shot. Because the break- mental condition.
down threshold of aerosol is orders of magnitude lower than gas, The temporal profiles of incident and transmitted laser pulses
two 8 mm stainless- steel tubes were installed adjunct to the noz- were measured using fast photodiodes. To evaluate the quantifica-
zle, providing shroud gas (Nitrogen) against the disturbance from tion performance of this method, the integrated area of photodi-
aerosols. The synthetic air was heated by a custom-built cylin- ode signal was plotted against the readouts of corresponding en-
drical gas heater. The heater uses resistance wires inside of an ergy meter. The data from the combustible mixture at 298 K and
alundum sheath. The diameter of sheath was 22 mm. A radiation after-burn gas at 1911 K are shown in Fig. 2. For both incident and
and conduction-corrected thermocouple (type-S) with 0.2 mm lead transmitted pulses, the determination coefficients (R2 ) of linear fit-
wires and 0.5 mm bead was set up adjacent to the focal point ting were over 0.99, and the intercepts of fitted line were close to
to continuously monitor the gas temperature. To obtain a uniform 0, which implied that the beam steering and defocusing had neg-
temperature distribution, the heated flow passed through a clus- ligible influence on the temporal profile measurement. Therefore,
ter of randomly stacked mini alundum tubes that acted as packing spatially integrated laser irradiance I (t ) at a time t can be obtained
near the heater exit. Using a voltage regulator to change the heat- by normalizing the photodiode signals using energy meter read-
ing power, the synthetic air flow (30 L/min) was heated to 400 K, outs, I (t ) = V (t )E/π r 2f S, where V (t ) is the intensity of photodiode

3
S. Chai, H. Peng, Z. Zhao et al. Combustion and Flame 249 (2023) 112612

Table 2
Description of the quantities. The subscript in denotes the quantity was determined from the incident laser pulse and tr denotes
from the transmitted laser pulse.

Name Variable Definition or equation

Laser energy E The readout of laser energy meter.

Deposited energy Ede Ein − Etr
Breakdown threshold Eth Ein at 50% breakdown probability
Intensity of photodiode signal V (t ) The vertical readout in the digital oscilloscope

Integrated area of photodiode signal S S = V (t )dt
V (t )E
Laser irradiance I (t ) π r2f S
Iin (t )−Itr (t )
Instantaneous absorption efficiency η (t ) Iin (t )
× 100%
Breakdown delay tbd η (tbd ) = 1%
Vin (tbd )Ein
Laser irradiance at breakdown Ibd π r2f Sin
 tbd
Pre-breakdown energy E pre π r2f 0 I (t )dt
Absorption efficiency in the post-breakdown region η Ede
Ein −E pre
× 100%

Fig. 2. The relations between the integrated area of photodiode signal and readouts Fig. 3. The temporal structure of incident and transmitted pulse normalized by en-
of energy meter. ergy meter readouts. The absorption profile was obtained by subtracting the trans-
mitted temporal profile from the incident temporal profile.

signals, E is the readout of energy meter, r f is the radius of beam The shot-to-shot matching between the random temporal struc-

waist in the focal point, and S = V (t )dt is the integrated area of ture of incident and transmitted pulse in the pre-breakdown re-
photodiode signal. All the quantities concerned in this work are gion demonstrates the validity of our measurement.
summarized in Table 2. An uncertainty analysis was performed, For the case in Fig. 3, there is an abrupt increase in laser en-
and the procedure of which was described in the Appendix. The ergy absorption starting around 5.7 ns, which indicates the initi-
reported error bars in the plots represent the expanded uncer- ation of breakdown. In this work, the time point where the ab-
tainty at 95% CI, which are combined from uncertainties of break- sorption efficiency η (t ) exceeds 1% was defined as the breakdown
down process and measurement instrument. initiation moment. The procedure to find this time point is to scan
the absorption efficiency profile backwards from the time point of
3. Results and discussion maximal absorption efficiency, then the first found point having
1% absorption efficiency was registered as the breakdown initia-
3.1. The temporal structure of energy deposition behaviors tion moment. The breakdown delay tbd was defined as the time
between the incident pulse arrival moment and the breakdown ini-
Typical temporal profiles of the incident and transmitted laser tiation moment. Therefore, the pulse width of laser can be sepa-
pulses for the heated air under 10 0 0 K are demonstrated in Fig. 3. rated into two regions by the breakdown initiation moment: pre-
The incident and transmitted energy was 56.79 mJ and 42.10 mJ, breakdown region and post-breakdown region. The total energy
respectively. The laser operates with multiple longitudinal modes, in the pre-breakdown region, namely pre-breakdown energy E pre ,
thus its temporal structure has multiple random peaks. As indi- can be obtained by integrating spatially averaged laser irradiance
t
cated by the black solid line, the temporal profile of incident laser E pre = π r 2f 0bd I (t )dt.
pulse clearly shows a random structure. The signals from differ- The plasma thermal radiation or elastically scattered laser by
ent photodiodes were aligned in the time domain to compensate plasma can be also used as the indication of breakdown initiation.
the differences in optical path length, cables, and electronics. The For these approaches, a careful correction on the differences in op-
temporal features of transmitted laser pulse normalized by en- tical path length, cables, and electronics must be performed to ac-
ergy meter readouts (yellow solid line) matched very well to the curately determine the breakdown delay. Comparatively, matching
corresponding incident laser pulse in the pre-breakdown region. the random features in the temporal profiles of incident and trans-

4
S. Chai, H. Peng, Z. Zhao et al. Combustion and Flame 249 (2023) 112612

time point of peak irradiance. However, we have observed many

events in which the breakdown initiated at the moment after the
peak irradiance. For example, in Fig. 3, the peak irradiance arrived
at 4 ns, while the breakdown was initiated around 5.7 ns. There-
fore, such results clearly demonstrated that, under our experimen-
tal conditions, for a typical nanosecond laser, the breakdown is not
controlled by the irradiance threshold.
To further investigate the controlling factor of breakdown, the
pre-breakdown energy E pre as a function of incident energy un-
der different temperatures is plotted in Fig. 6, where each data
point is an average of 200 shots. The widely accepted breakdown
threshold, which was determined by increasing the laser energy
incrementally until the probability that breakdown occurs is 50%
[17,21], is indicated by the red dash line at each temperature in
Fig. 6 as a benchmark. The thermal radiation from plasma was
used as the criterion of breakdown occurrence. At each tempera-
ture, 10 0 0 shots were collected, and afterward, a binomial logistic
regression was applied. For temperatures under 10 0 0 K (Fig. 6(a)
to 6(f)), the pre-breakdown energy was constant for different en-
ergy levels under the same temperature. For the synthetic air un-
Fig. 4. The breakdown delay as functions of incident energy in 298 K mixture, der 298 K (Fig. 6(b)), the pre-breakdown energy varied less than
600 K air, and 1911 K after-burn gas. 3.17 mJ, while the incident energy increased from 40 mJ to 260
mJ. In the after-burn gas (Fig. 6(g) to 6(j)), the pre-breakdown en-
ergy slightly decreased when the incident energy increased. For
mitted pulse is straightforward. The breakdown delay in three rep-
example, in the after-burn gas under 1911 K (Fig. 6(j)), the pre-
resentative gas conditions, 298 K mixture, 600 K air, and 1911 K
breakdown energy decreased from 43.24 mJ to 35.72 mJ, while the
after-burn gas, are shown in Fig. 4 with respect to incident energy.
incident energy increased more than 3 times from 80 mJ to 260
Each data point is an average of 200 shots. Power laws were used
mJ.
to fit the data, and R2 was 0.993, 0.985, and 0.962, respectively.
Therefore, the results obtained from temporally resolved mea-
The results agree with the literature [17,42].
surements indicated that, the initiation of breakdown is depen-
dent on pre-breakdown energy instead of instantaneous irradiance
3.2. The controlling factor of breakdown initiation under our experimental conditions, which cover a typical range
of laser energy in the nanosecond regime. Furthermore, there are
The initiation of breakdown is a result of free electron growth differences between the pre-breakdown energy and breakdown
at the focal volume where the initial free electrons can either be threshold determined by 50% probability method, while the dif-
generated by multiphoton ionization, or simply provided by im- ferences were more significant in the high-temperature after-burn
purities and heating. The initial free electrons can subsequently gas, which highlighted the temperature effect. In the following sec-
absorb laser energy via inverse bremsstrahlung and collide with tion, the theoretical background of breakdown initiation and the
neutrals to produce additional free electrons [43]. The repeated mechanism behind such observations will be further analyzed.
collision ionization results in a fast exponential increase of elec-
tron density, namely cascade ionization [44]. Other than provid- 3.3. The theoretical analysis of breakdown initiation
ing initial free electrons, multiphoton ionization is important only
for very short (picosecond or shorter) laser pulses and relatively As shown in Fig. 7, fitting the breakdown threshold by power
low gas density when the product of gas pressure and laser pulse law yields an R2 of 0.999, confirming that the energy absorption in
width, P τ p , is lower than 10−7 torr·s, as pointed out by Morgan the gas breakdown is dominated by inverse bremsstrahlung [46].
[45]. Competing with the electron growth is the loss due to re- The exponent of fitting, which is 0.348 in this study, is depen-
combination with cationic, attachment with heavy molecular, and dent on the competition between electron growth and loss [47].
diffusion out from the focal volume [43], which also play impor- If the electron loss is negligible, it will be close to unity [47] be-
tant roles in breakdown initiation. However, there is still ambigu- cause the rate of cascade ionization is proportional to gas density
ity in the combustion community regarding the controlling factor [48]. Therefore, an exponent of 0.348 indicates that the electron
of breakdown initiation, the use of instantaneous laser irradiance loss is significant. Among the electron loss mechanisms, recom-
(irradiance threshold), or total pulse energy (energy threshold) are bination was negligible because it plays a role only under high
mixed, even in the nanosecond laser regime. electron density, which much exceeded the typical values in the
Figure 5 shows the laser irradiance at the breakdown initia- pre-breakdown region [49]. Attachment should be taken into ac-
tion moment as a function of incident energy. Each data point was count since the gas contained electronegative molecules, i.e., oxy-
an average of 200 shots. The instantaneous laser irradiance at the gen, water, or methane. Due to the small radius of the focal spot
breakdown initiation moment was in the order of 1011 W/cm2 and (61.8 μm) and reduced gas density, the diffusion loss should be
increased with incident laser energy. For example, in the 298 K carefully considered [44]. The laser pulse width in the nanosecond
air, it increased more than 3 times from 2.05 × 1011 W/cm2 to regime is also sufficiently long to allow the free electrons to diffuse
6.84 × 1011 W/cm2 , as the incident energy increased from 40 mJ out from the focal volume, since the characteristic time of which
to 260 mJ. Measurements in the heated air and after-burn gas have is in the order of picosecond [50]. Including cascade ionization-
similar trends, with higher values in the same order of magnitude. dominated electron growth and electron loss by attachment and
Moreover, since we are using a laser with multiple longitudinal diffusion, the integral form of time-dependent electron density in
modes, which features multiple random peaks in its temporal pro- the pre-breakdown region can be expressed as [43]

file, if the breakdown initiation is primarily controlled by laser ir- ne f (t ) t

radiance, the initiation moment should always be before or at the ln = (vi −va − vd )dt (1)
n0 0

5
S. Chai, H. Peng, Z. Zhao et al. Combustion and Flame 249 (2023) 112612

Fig. 5. Instantaneous laser irradiance at the breakdown initiation moment as a function of incident energy in the combustible mixture and air (a), and after-burn gas (b).

where n0 is the initial free electrons density before the laser ar- tth − tbd is non-zero; thus, there is a difference between the
rives, ne f (t ) is the electron density at the time t, vi is the cascade pre-breakdown energy and breakdown threshold, which was rep-
ionization rate. va and vd is the electron loss rate due to the at- resented by the left hand of Eq. (5). The term Kd T 2 explains the
tachment and diffusion, respectively. temperature effect. In the high temperature after-burn gas, the
vi scales proportionally with gas density N and laser irradi- value of Kd T 2 is larger than that under room temperature; there-
ance I (t ), except for very high pressure (hundreds of atmospheres); fore, the difference between the pre-breakdown energy and break-
thus, it can be expressed as vi = S1 NI (t ) [48], where S1 is a con- down threshold became more significant. The larger Kd T 2 can also
stant. According to Chan [51], va = ha vm , where ha is the attach- explain the noticeable decrease of E pre when the incident energy
ment efficiency depending on the electron energy and vm is the increases in the after-burn gas. For a given temperature, Eth and
electron momentum transfer collision frequency. The average value tth are constants, when the incident energy increases, tbd decreases
of ha can be treated as a constant and vm is proportional to the gas due to the steeper gradient of laser profile. Since Eth − E pre is cor-
density, so the electron loss rate by attachment can be expressed related with tth − tbd by the coefficient Ka + Kd T 2 , the decrease in
as va = S2 N, where S2 is a constant. vd was inversely proportional tbd leads to a more pronounced decrease in E pre under high tem-
to gas density [45], then we have vd = S3 /N, where S3 is a constant. perature, which may not be noticeable under room temperature.
Inserting the above equations into Eq. (1) gives According to Eq. (5), E pre can be linearly correlated to tbd when
 the gas temperature is fixed, because Eth , tth , Ka , and Kd T 2 are all
ne f (t ) t
S3
ln = S1 NI (t ) − (S2 N+ )dt (2) constants. The linear correlation between E pre and tbd in the 1911 K
n0 0 N
after-burn gas is illustrated in Fig. 8(a), with an R2 of 0.964. E pre
Under the assumption of ideal gas, Eq. (2) becomes
S  and tbd as functions of incident energy are plotted together in
ne f (tbd ) S1 E pre 2 Fig. 8(b). The trends of E pre and tbd agreed well. Such experimental
ln = − + S3 T tbd (3)
n0 π r2f T T results validate our theoretical interpretations.
The important role of electron diffusion under higher temper-
The breakdown is initiated once the accumulated free electrons
atures can be further validated in Fig. 9, which shows the pre-
reach a critical density [45]. The electron density at the breakdown
breakdown energy as a function of incident energy in pure ni-
initiation moment is a function of gas temperature and is inde-
trogen under 298 K and 10 0 0 K. Nitrogen is chosen because the
pendent on the incident pulse energy [52]. Therefore, the left-hand
attachment loss is negligible for the inert gas, so the influence
side of Eq. (3) becomes a constant for any incident pulse energy, if
from diffusion can be isolated. The pre-breakdown energy in ni-
the gas temperature is fixed. If we consider the breakdown thresh-
trogen behaved similarly to the case in air. The difference between
old determined by the 50% probability method as the lower limit
the breakdown threshold and the pre-breakdown energy is larger
of incident energy that can induce a breakdown, ideally the break-
for 10 0 0 K. The more significant influence of electron diffusion on
down initiation moment tth should also be approaching the end of
breakdown initiation under high temperature agrees with previous
pulse width. Both Eth and tth are constants for fixed gas conditions.
experiments [53] and numerical simulations [54,55] under reduced
Then for the breakdown initiation moment, we have
S1 E pre
S  S1 Eth
S  gas pressure.
2 2 It should be noted that Eq. (5) is an idealized theoretical inter-
− + S3 T tbd = − + S3 T tth (4)
π r2f T T π r2f T T pretation. A more accurate analysis should be based on solving the
Eq. (4) gives time-dependent Boltzmann equation for the Electron Energy Dis-
S S3 2
 tribution Function (EEDF) with a set of rate equations controlling
Eth − E pre = π r 2f T (tth − tbd ) = (Ka + Kd T 2 )(tth − tbd )
2 the rate of change in excited state populations. Nevertheless, this
+
S1 S1 interpretation still agrees well with experimental observations.
(5) Another comment is that, the moment of breakdown initiation
where Ka and Kd are constants representing the influence intro- is a critical state where the maximal free electrons density just
duced by the electron attachment and diffusion loss on breakdown reaches the critical density. At tth , the rate of cascade ionization
initiation, respectively. and diffusion loss should be balanced. Therefore, tth should be in

6
S. Chai, H. Peng, Z. Zhao et al. Combustion and Flame 249 (2023) 112612

Fig. 6. The pre-breakdown energy as functions of incident energy under different gas temperatures. The red dash line indicates the breakdown threshold.

7
S. Chai, H. Peng, Z. Zhao et al. Combustion and Flame 249 (2023) 112612

Fig. 7. The breakdown threshold under different gas temperatures. Fig. 9. The pre-breakdown energy as a function of incident energy for nitrogen un-
der 298 K and 10 0 0 K.

the falling edge of incident pulse, but not at the end of pulse. tth
represents the upper extreme of tbd . breakdown, the breakdown delay cannot be treated as the time
point at which the accumulated laser energy reaches the break-
3.4. The physical connections between LIP-based thermometry down threshold. According to Eq. (5), the breakdown threshold and
methods breakdown delay can be correlated by
 tbd
Breakdown delay, breakdown threshold, and deposited energy Eth = I (t ) + Kd T 2 (tth − tbd ) (6)
are the measured quantities in recently developed LIP-based gas 0

thermometry. After the mechanisms of breakdown initiation were where the second term in the right-hand side represents the cor-
investigated, the physical connections between these thermometry rection due to diffusion. Under fixed gas compositions and optical
methods can be clarified. setup, Kd is fixed and tth is only dependent on gas temperature.
The breakdown delay reflects the breakdown initiation in the Therefore, the diffusion correction term Kd T 2 (tth − tbd ) is also only
time domain. Combining breakdown delay with the temporal dependent on the gas temperature, which ensures the breakdown
structure of incident laser pulse, the pre-breakdown energy can threshold and breakdown delay are correlated at every tempera-
be obtained. The pre-breakdown energy, which is the controlling ture, supporting the thermometry method recently proposed by
factor of breakdown initiation, equals the widely used breakdown Williamson et al. [17]. As a consequence, the two quantities are
threshold only when the electron loss can be neglected in the equivalent in gas temperature measurement.
collision-dominated breakdown process. Generally, electron diffu- Utilizing a temporally resolved laser energy absorption profile
sion loss is important under reduced gas density, which is the case (Fig. 3), the deposited energy can be obtained by integrating in
in high temperature environments. Therefore, under high tem- the post-breakdown region. The deposited energy versus gas tem-
peratures, using critical free electron density as the criterion for perature under an incident energy level of 80 mJ is shown in

Fig. 8. In the 1911 K after-burn gas, the pre-breakdown energy as a function of breakdown delay (a), and the pre-breakdown energy and breakdown delay as functions of
incident energy (b).

8
S. Chai, H. Peng, Z. Zhao et al. Combustion and Flame 249 (2023) 112612

Fig. 10. (a) The deposited energy versus temperature. (b) The breakdown delay and the averaged absorption efficiency in the post-breakdown region versus gas temperature.
The incident energy level was 80 mJ.

Fig. 10(a), and the breakdown delay and averaged absorption ef- of electrons was found to be significant under high temperature.
ficiency in the post-breakdown region are shown in Fig. 10(b). The Based on the findings above, the physical connections between
decrease of deposited energy when temperature increases is con- the breakdown delay, the breakdown threshold, and the deposited
tributed by two main factors. One factor is that the longer break- energy were clarified. The decreased deposited energy when gas
down delay under higher temperature, which narrows the dura- temperature increased was contributed not only by the elongated
tion of laser absorption given the laser pulse width is unchanged breakdown delay in the pre-breakdown region, but also by the re-
(black points, Fig. 10(b)). The other factor is the decrease of absorp- duced absorption efficiency in the post-breakdown region.
tion efficiency in the post-breakdown region, as shown by the red
points in Fig. 10(b). The decreased absorption efficiency is a result
of reduced collision frequency under higher temperature, which af- Declaration of Competing Interest
fects the energy absorption by the inverse bremsstrahlung process.
Therefore, the combined effect of elongated breakdown delay in The authors declare that they have no known competing finan-
the pre-breakdown region and reduced absorption efficiency in the cial interests or personal relationships that could have appeared to
post-breakdown region leads to the decrease of deposited energy influence the work reported in this paper.
under higher gas temperature. Under different incident energy lev-
els and focusing geometry, the influences of gas temperature on
the deposited energy via the breakdown delay and absorption effi- Acknowledgement
ciency will be investigated in our next work.
This work was funded by the National Natural Science Foun-
4. Summary and conclusion dation of China (No. 51906149), Natural Science Foundation of
Shanghai (Nos. 20ZR142850 0, 21DZ120530 0), The Oceanic In-
In this work, a high temporal resolution method was introduced terdisciplinary Program of Shanghai Jiao Tong University (No.
to quantitatively analyze the energy deposition behaviors of laser- SL2020ZD202).
induced plasmas in combustion environments. Using a nanosec-
ond laser with multiple longitudinal modes, the breakdown delay
was determined with 0.1 ns resolution by shot-to-shot matching Appendix. Uncertainty analysis
the random temporal structure in incident and transmitted laser
pulses. The instantaneous laser irradiance at the breakdown initi-  standard uncertainty us of E and V (t ) is evalu-
The combined
ation moment and the pre-breakdown energy were obtained over ated by us = SE 2 + b2s , where SE is the event-to-event uncertainty
a typical laser energy range from 40 mJ to 260 mJ, and in a broad represented by the standard deviation of the mean (Type A eval-
temperature region ranged from 298 K to 1911 K, covering com- uation), bs is the standard instrument uncertainty obtained from
bustible mixture under room temperature, heated air, and high the manufacturers’ specification (Type B evaluation). The combined
temperature after-burn gas. For a specific temperature, the instan- standard uncertainty of Eth is approximated by the standard instru-
taneous laser irradiance at the breakdown initiation moment in- ment uncertainty of incident energy at 50% breakdown probability.
creased significantly with laser energy, while the pre-breakdown Table A.1 listed the expanded uncertainties of dominant ele-
energy was nearly a constant, revealing the controlling factor of mental error sources in laser energy and photodiode signal mea-
the breakdown initiation was energy threshold instead of irradi- surements, at 95% confidence interval (CI). The standard uncer-
ance threshold in the nanosecond laser regime. Then the mech- tainties of each source were obtained by dividing a coverage fac-
anisms of breakdown initiation were investigated. The difference  standard instrument uncertainty V (t ) is evaluated by
tor of 2. The
2 2
between the breakdown threshold and the pre-breakdown energy bs [V (t )] = bs ( photodiode ) + bs (oscil l oscope ) .
was observed, and became more pronounced under high temper- The standard uncertainties of derived quantities were evaluated
atures. This experimental observation was theoretically explained using the Taylor series method (TSM) for propagation of uncertain-
by analysis based on electron growth and loss. The diffusion loss ties [56]. For Y = f (X1 , X2 , · · ·, XN ), the standard uncertainty can be

9
S. Chai, H. Peng, Z. Zhao et al. Combustion and Flame 249 (2023) 112612

Table A.1
Dominant elemental error sources of instrument uncertainties in laser energy and photodiode signal measurements. The coverage factors are 2 for the expanded uncertain-
ties at 95% confidence interval.

Quantity Instrument Dominant elemental error source Expanded uncertainty at 95% confidence interval Standard uncertainty

Laser energy Energy meter Calibration ± 3% ±1.5%

Intensity of Photodiode Linearity ± 0.5%a ±0.25%
Photodiode signal Digital oscilloscope DC vertical gain ± 0.5% full scale ±0.25% full scale
a
The typical value of a silicon photodiode.

combined as: [4] S.H. Lee, H.T. Hahn, J.J. Yoh, Towards a two-dimensional laser induced break-
 down spectroscopy mapping of liquefied petroleum gas and electrolytic
N  2 oxy-hydrogen flames, Spectrochim. Acta B 88 (2013) 63–68.

∂ f
N
∂f ∂f [5] K.E. Eseller, F.Y. Yueh, J.P. Singh, Laser-induced breakdown spectroscopy mea-
us (Y ) = us (Xi ) + 2 r (Xi , X j )us (Xi )us (X j ) surement in methane and biodiesel flames using an ungated detector, Appl.
i=1
∂ Xi
1≤i< j
∂ Xi ∂ X j Opt. 47 (2008) G144–G148.
[6] W. Zhu, X. Li, R. Sun, Y. Yan, M. Yuan, X. Ren, X. Meng, X. Yu, Dynamic be-
(A.1) haviors of the sodium, calcium and iron release during coal combustion using
multi-point LIBS, Combust. Flame 244 (2022) 112237.
where Xi refers to the input variables and r (Xi , X j ) is the correla- [7] Y. Yuan, S. Li, Q. Yao, Dynamic behavior of sodium release from pulverized coal
combustion by phase-selective laser-induced breakdown spectroscopy, Proc.
tion coefficient between two variables. In this work, only E and S Combust. Inst. 35 (2015) 2339–2346.
are linearly correlated, yielding r (E, S )≈1. Using TSM, the standard [8] Y. Wu, M. Gragston, Z. Zhang, P.S. Hsu, N. Jiang, A.K. Patnaik, S. Roy, J.R. Gord,
uncertainty of Ede was combined from Ein and Etr . High-pressure 1D fuel/air-ratio measurements with LIBS, Combust. Flame 198
(2018) 120–129.
Analytically, S is a summation of all photodiode readings over [9] L. Zimmer, S. Yoshida, Feasibility of laser-induced plasma spectroscopy for
the laser pulse. However, the irradiance profile of a multimode measurements of equivalence ratio in high-pressure conditions, Exp. Fluids 52
laser varies shot-to-shot. Analyzing the statistics of irradiance at a (2011) 891–904.
[10] M. Kotzagianni, R. Yuan, E. Mastorakos, S. Couris, Laser-induced breakdown
specific time can be very misleading. Thus, simply estimating the
spectroscopy measurements of mean mixture fraction in turbulent methane
uncertainty in S by combining V (t ) falls short of physical signifi- flames with a novel calibration scheme, Combust. Flame 167 (2016) 72–85.
cance. Instead, we evaluated a typical value of standard instrument [11] T.W. Lee, N. Hegde, Laser-induced breakdown spectroscopy for in situ diag-
uncertainty bs (S ), which was further combined with the event-to- nostics of combustion parameters including temperature, Combust. Flame 142
(2005) 314–316.
event uncertainty SE (S ) to evaluate us (S ). The instrument uncer- [12] B. McGann, C.D. Carter, T. Ombrello, H. Do, Direct spectrum matching of
tainty of S in a single shot was evaluated by the root sum square laser-induced breakdown for concentration and gas density measurements in
of instrument uncertainties of V (t ) over the pulse duration. 200 turbulent reacting flows, Combust. Flame 162 (2015) 4479–4485.
[13] Z.F. Zhang, T. Li, X.F. Xue, S. Huang, Simultaneous measurements of fuel con-
sampled pulses in each experimental condition gave 200 instru- centration and temperature in gas jets by laser induced breakdown spec-
ment uncertainties of S, the average of which was treated as a typ- troscopy, Spectrochim. Acta B 161 (2019) 105706.
ical value of bs (S ). Using TSM, the standard uncertainty us (Ibd ) was [14] S. Li, W. Weng, C. Kong, M. Aldén, Z. Li, Dual-laser-induced breakdown ther-
mometry via sound speed measurement: a new procedure for improved spa-
combined from Vin (tbd ), Ein , and Sin , while r f was treated as a con- tiotemporal resolution, Sensors-Basel 20 (2020) 2803.
stant. [15] J. Lee, C. Bong, S.H. Lee, S.K. Im, M.S. Bak, Laser-induced breakdown thermom-
tbd was determined as the first time point where η (t ) is over etry via time-of-arrival measurements of associated acoustic waves, Appl. Phys.
Lett. 113 (2018) 123504.
1%. The standard instrument uncertainty bs [η (tbd )] was combined [16] J. Kiefer, J.W. Troger, Z.S. Li, T. Seeger, M. Alden, A. Leipertz, Laser-induced
using TSM. The temporal resolution of digital oscilloscope is 0.1 ns. breakdown flame thermometry, Combust. Flame 159 (2012) 3576–3582.
Under our experimental conditions, the typical value of bs [η (tbd )] [17] A.P. Williamson, U. Thiele, J. Kiefer, Comparison of existing laser-induced
is about one order of magnitude lower than η (tbd ) − η (tbd − 0.1ns ) breakdown thermometry techniques along with a time-resolved breakdown
approach, Appl. Opt. 58 (2019) 3950–3956.
and η (tbd + 0.1ns ) − η (tbd ). Thus, bs [η (tbd )] will not introduce sig- [18] H. Do, C.D. Carter, Q. Liu, T.M. Ombrello, S. Hammack, T. Lee, K.-.Y. Hsu, Si-
nificant uncertainty in breakdown delay determination, which tes- multaneous gas density and fuel concentration measurements in a supersonic
combustor using laser induced breakdown, Proc. Combust. Inst. 35 (2015)
tifies the validity of our temporal profile matching approach.
2155–2162.
The standard instrument uncertainty of tbd due to the √ temporal [19] Z.H. Tian, M.R. Dong, S.S. Li, J.D. Lu, Spatially resolved laser-induced breakdown
resolution of digital oscilloscope is bs (tbd ) = 0.5 × 1/ 3 × 0.1ns ≈ spectroscopy in laminar premixed methane-air flames, Spectrochim. Acta B
0.029ns.Therefore, the standard uncertainty of tbd was evaluated by 136 (2017) 8–15.
 [20] S.S. Li, M.R. Dong, L.H. Cheng, F.S. Luo, W.H. Zhao, J.D. Lu, Temperature mea-
2 2
us (tbd ) = SE (tbd ) + bs (tbd ) . surement with compositional correction of gas mixture based on laser-induced
t
The standard uncertainty of 0bd Vin (t )dt was evaluated with the plasma, Appl. Opt. 59 (2020) 7638–7645.
[21] W. Wu, S. Chai, Y. Zhou, Monitoring the deposited energy in laser-induced
same procedure to that of S. Using TSM, the standard uncertainty
t plasmas with an acoustic, Appl. Opt. 61 (2022) 1734–1742.
of E pre was combined from Ein , 0bd Vin (t )dt , and Sin ; the standard [22] W. Wu, A. Adeosun, R.L. Axelbaum, A new method of flame temperature mea-
uncertainty of η was combined from Ede , Ein , and E pre . According surement utilizing the acoustic emissions from laser-induced plasmas, Proc.
Combust. Inst. 37 (2019) 1409–1415.
to the central limit theorem, the overall errors of each quantity [23] T.X. Phuoc, F.P. White, Laser-induced spark ignition of CH4/air mixtures, Com-
approximate the normal distribution, thus multiplying the standard bust. Flame 119 (1999) 203–216.
uncertainty of each quantity by a coverage factor of 2 yields the [24] Y. Sung, G. Charalampous, Y. Hardalupas, G. Choi, Laser ignition and flame
characteristics of pulsed methane jets in homogeneous isotropic turbulence
expanded uncertainty at 95% CI. without mean flow, Proc. Combust. Inst. 36 (2017) 1653–1660.
[25] Y. Kobayashi, S. Nakaya, M. Tsue, Laser-induced spark ignition for DME–air
References mixtures with low velocity, Proc. Combust. Inst. 37 (2019) 4127–4135.
[26] S. Jo, J.P. Gore, Laser ignition energy for turbulent premixed hydrogen air jets,
Combust. Flame 236 (2022) 111767.
[1] F. Ferioli, S. Buckley, Measurements of hydrocarbons using laser-induced break-
[27] N. Mokrani, P. Gillard, Laser induced breakdown in gas mixtures. Experimental
down spectroscopy, Combust. Flame 144 (2006) 435–447.
and statistical investigation on n-decane ignition: pressure, mixture composi-
[2] Y. Ren, A. Kreischer, F. Cameron, H. Pitsch, Quantitative measurement of mix-
tion and equivalence ratio effects, J. Hazard. Mater. 388 (2020) 119266.
ture fraction in counterflow diffusion flames by laser-induced breakdown
[28] B.A. Ochs, S. Menon, Laser ignition in supersonic channel flow, Combust. Flame
spectroscopy, Combust. Flame 241 (2022) 112–130.
214 (2020) 90–102.
[3] W.D. Wu, R.L. Axelbaum, Interpreting diffusion flame structure by simul-
[29] L. Wermer, J.K. Lefkowitz, T. Ombrello, S.-k. Im, Spark and flame kernel in-
taneous mixture fraction and temperature measurements using optical and
teraction with dual-pulse laser-induced spark ignition in a lean premixed
acoustic signals from laser-induced plasmas, Proc. Combust. Inst. 38 (2021)
methane–air flow, Energy 215 (2021) 119162.
1665–1674.

10
S. Chai, H. Peng, Z. Zhao et al. Combustion and Flame 249 (2023) 112612

[30] T.X. Phuoc, F.P. White, An optical and spectroscopic study of laser-induced [43] T.X. Phuoc, Laser-induced spark ignition fundamental and applications, Opt.
sparks to determine available ignition energy, Proc. Combust. Inst. 29 (2002) Laser Eng. 44 (2006) 351–397.
1621–1628. [44] R. Tambay, R. Thareja, Laser-induced breakdown studies of laboratory air at
[31] T.X. Phuoc, An experimental and numerical study of laser-induced spark in air, 0.266, 0.355, 0.532, and 1.06μm, J. Appl. Phys. 70 (1991) 2890–2892.
Opt. Laser Eng. 43 (2005) 113–129. [45] C.G. Morgan, Laser-induced breakdown of gases, Rep. Prog. Phys. 38 (1975)
[32] G.C. Gebel, T. Mosbach, W. Meier, M. Aigner, Optical and spectroscopic diag- 621–665.
nostics of laser-induced air breakdown and kerosene spray ignition, Combust. [46] T.X. Phuoc, Laser spark ignition: experimental determination of laser-in-
Flame 162 (2015) 1599–1613. duced breakdown thresholds of combustion gases, Opt. Commun. 175 (20 0 0)
[33] P.M. de Oliveira, M.P. Sitte, M. Kotzagianni, P.M. Allison, E. Mastorakos, A 419–423.
laser-induced breakdown spectroscopy method to assess the stochasticity of [47] P. Chylek, M.A. Jarzembski, V. Srivastava, R.G. Pinnick, Pressure dependence of
plasma-flame transition in sprays, Meas. Sci. Technol. 33 (2022). the laser-induced breakdown thresholds of gases and droplets, Appl. Opt. 29
[34] M. Weinrotter, H. Kopecek, E. Wintner, M. Lackner, F. Winter, Application of (1990) 2303–2306.
laser ignition to hydrogen–air mixtures at high pressures, Int. J. Hydrogen En- [48] M.N. Shneider, R.B. Miles, Laser induced avalanche ionization in gases or gas
ergy 30 (2005) 319–326. mixtures with resonantly enhanced multiphoton ionization or femtosecond
[35] G. Zhai, S. Xing, A.C. Yuen, P.R. Medwell, S. Kook, G.H. Yeoh, Q.N. Chan, Laser laser pulse pre-ionization, Phys. Plasmas 19 (2012) 083508.
ignition of iso-octane and n-heptane jets under compression-ignition condi- [49] D.C. Smith, Laser-induced gas breakdown and plasma shielding, Proc. SPIE 195
tions, Fuel 311 (2022) 122555. (1979) 171–181.
[36] J.-.L. Beduneau, B. Kim, L. Zimmer, Y. Ikeda, Measurements of minimum ig- [50] P.E. Nielsen, G.H. Canavan, Electron cascade theory in laser-induced breakdown
nition energy in premixed laminar methane/air flow by using laser induced of preionized gases, J. Appl. Phys. 44 (1973) 4224–4225.
spark, Combust. Flame 132 (2003) 653–665. [51] C.H. Chan, C.D. Moody, W.B. McKnight, Significant loss mechanisms in gas
[37] S. Nakaya, S. Iseki, X. Gu, Y. Kobayashi, M. Tsue, Flame kernel formation behav- breakdown at 10.6 μ, J. Appl. Phys. 44 (1973) 1179–1188.
iors in close dual-point laser breakdown spark ignition for lean methane/air [52] N. Bloembergen, Laser-induced electric breakdown in solids, IEEE J. Quantum
mixtures, Proc. Combust. Inst. 36 (2017) 3441–3449. Electron. 10 (1974) 375–386.
[38] M. Nishihara, J. Freund, N. Glumac, G. Elliott, Influence of mode-beating pulse [53] D. Rosen, G. Weyl, Laser-induced breakdown in nitrogen and the rare gases at
on laser-induced plasma, J. Phys. D: Appl. Phys. 51 (2018) 135601. 0.53 and 0.357μm, J. Phys. D: Appl. Phys. 20 (1987) 1264.
[39] C.V. Bindhu, S.S. Harilal, M.S. Tillack, F. Najmabadi, A.C. Gaeris, Energy ab- [54] Y.E.E.D. Gamal, K.A. Elsayed, M.A. Mahmoud, Numerical investigation of the
sorption and propagation in laser-created sparks, Appl. Spectmsc. 58 (2004) electron dynamic dependence on gas pressure in the breakdown of hydrogen
719–726. by KrF laser radiation, Opt. Laser Technol. 44 (2012) 2154–2160.
[40] E. Schwarz, S. Gross, B. Fischer, I. Muri, J. Tauer, H. Kofler, E. Wintner, Laser-in- [55] M.M. Omar, A.M. Aboulfotouh, Y.E.E. Gamal, On the study of threshold inten-
duced optical breakdown applied for laser spark ignition, Laser Part. Beams 28 sity dependence on the gain and loss processes in laser induced spark ignition
(2010) 109–119. of molecular hydrogen, AIP Conf. Proc. 1653 (2015) 020081.
[41] X. Jian, and Ma, R. Dennis, and Alexander, E. Dana, Laser spark ignition [56] H.W. Coleman, W.G. Steele, Experimentation, Validation, and Uncertainty Anal-
and combustion characteristics of methane-air mixtures, Combust. Flame 112 ysis for Engineers, John Wiley & Sons, 2018.
(1998) 492–506.
[42] S. Oh, S. Bae, C.D. Carter, H. Do, Laser pulse width control using in-
verse-Bremsstrahlung photon absorption, Opt. Lett. 44 (2019) 3721–3724.

11

View publication stats