=Paper=
{{Paper
|id=None
|storemode=property
|title=None
|pdfUrl=https://ceur-ws.org/Vol-865/LIIworkshop2012.pdf
|volume=Vol-865
}}
==None==
https://ceur-ws.org/Vol-865/LIIworkshop2012.pdf
Fifth International Workshop on Laser Induced Incandescence
Palais des Congrès, Le Touquet, France
May 8-11, 2012
The 5th International LII Workshop provides a forum for open high-level
discussion on the understanding of LII diagnostics and to foster relationships in joint
research. Laser-induced incandescence (LII) has proven to be a powerful tool for
particle concentration and particle size measurements in combustion, particle
synthesis, as well as in environmental applications. However, different experimental
approaches and data evaluation techniques exist which, while demonstrating the
complexity of the physical processes involved in LII and the need for further research,
has also somewhat hampered the acceptance by industry of LII as a measurement
standard. In order to strengthen the community and to explore the development of a
series of best practices for LII modeling, experiments, and data interpretation, a
series of workshops was initiated in Duisburg, Germany (2005), followed by
workshops in Bad Herrenalb, Germany (2006), Ottawa, Canada (2008) and Varenna,
Italy (2010).
The 5th International LII Workshop will be held May 9–11, 2012 in Le Touquet,
France
Organizers
Pascale Desgroux, Sébastien Batut, Eric Therssen ; PC2A, CNRS/Université Lille1,
Jérôme Yon, CORIA/INSA-Université Rouen
Advisory committee
Zeyad Alwahabi (Adelaide University); Per-Erik Bengtsson (Lund Institute of Technology);
Henning Bockhorn (Karlsruhe Institute of Technology); Silvana De Iuliis (CNR-IENI, Milano);
Pascale Desgroux (PC2A, Université Lille 1, Lille); Klaus-Peter Geigle (DLR, Stuttgart);
Douglas A. Greenhalgh (Glasgow Caledonian); Fengshan Liu (NRC Canada, Ottawa); Hope
Michelsen (Sandia National Laboratories); Christof Schulz (IVG, University of Duisburg -
Essen); Greg Smallwood (NRC Canada, Ottawa); Rainer Suntz (Karlsruhe Institute of
Technology); Kevin Thomson (NRC Canada, Ottawa); Stefan Will (Universität Erlangen); He
Xu (Beijing Institute of Technology)
We acknowledge the financial support of :
Program
Tuesday, May 08
Arrival, Check-in, Registration 17:30-19:30 at Palais des Congrès
General information
Wednesday, May 09
8:00 Registration
8:45 Welcome
Session 1: Experiment and modelling for LII fundamental knowledge
Experimental
9:00 Pulsed laser heating of differently aged soot probed using LII and LES
Nils-Erik Olofsson, Jonathan Johnsson, Henrik Bladh, Per-Erik Bengtsson
Lund University
9:20 Experimental setup to study soot sublimation as typically occurring in high fluence LII
Klaus Peter Geigle, Gregor Gebel, Markus Köhler
German Aerospace Centre (DLR), Stuttgart
Modelling
9:40 Influence of soot aggregate structure on particle sizing using LII
Jonathan Johnsson, Henrik Bladh, Nils-Erik Olofsson, Per-Erik Bengtsson
Lund University
10:00 Extending Time-Resolved LII to Metal Nanoparticles: Simulating the Thermal
Accommodation Coefficient
K. J. Daun, J. T. Titantah, M. Karttunen and T. A. Sipkens
University of Waterloo, University of Western Ontario
Coffee break: 10:20-11:00 (registration continuation)
Session 2: ELS fundamentals
11:00 A method for inferring the soot size distribution by Static Light Scattering :
Application to the CAST soot generator
Jérôme Yon, Chloé Caumont-Prim, Alexis Coppalle, Kuan Fang Ren
CORIA, University and INSA Rouen
11:20 Recent applications of the WALS-technique
Hergen Oltmann, Stefan Will
Universität Bremen, Universität Erlangen-Nürnberg
11:40 Poster advertising I
12:30-14:00: Lunch
Session 3: Combined methods for better knowledge of soot size
14:00 Soot primary particle sizing in turbulent flames via combined LII and Elastic Light Scattering
Brian Crosland, Matthew Johnson, Kevin Thomson
Carleton University, NRC Ottawa
14:20 Defining a measurement strategy for 2D soot particle size imaging through detailed LII signal-
decay analysis
Emre Cenker, Gilles Bruneaux, Thomas Dreier, Christof Schulz
IFP En, Rueil-Malmaison, IVG and CENIDE, University of Duisburg-Essen
14:40 Poster advertising II (follow up)
Free afternoon around 15:00
Poster session (with refreshment)
20:00- 22:30
Thursday, May 10
8:50 Opening of the workshop discussion (objectives)
9:00 Discussion on LII modelling
10:30-11:00: coffee break
Session 4: Experiments for particles properties fundamental knowledge
11:00 Effects of particle coatings on laser induced incandescence
Ray Bambha, Paul Schrader, Mark Dansson, Hope Michelsen
SANDIA
11:20 Determination of the dimensionless extinction coefficient for soot generated by a PMMA
flame
Damien Hebert, Alexis Coppalle, Jérôme Yon and Martine Talbaut
CORIA, University and INSA Rouen
Session 5: Combined methods
11:40 A combined laser induced incandescence, aerosol mass spectrometry, and scanning mobility
particle sizing study of non-premixed ethylene flames
Scott Skeen, Paul Schrader, Kevin Wilson, Nils Hansen, Hope Michelsen
SANDIA, Lawrence Berkeley National Lab
12:00 Soot particles detection by LIBS and LII analysis
Francesca Migliorini, Silvia Maffi, Silvana De Iuliis, Giorgio Zizak
CNR-IENI, Milano
12:30-14:00: lunch
14:00 Discussion on Experimental LII
Discussion about the LII workshop follow up
Poster session (and refreshments)
16:00-18:00
19:00 banquet
Friday, May 11
Session 6: applications
8:30 LII and one-wavelength Aethalometer measurements of particulate matter in different
environments
Silvana De Iuliis, Silvia Maffi, Francesca Migliorini, Giorgio Zizak
CNR-IENI, Milano
8:50 Time-resolved Laser induced incandescence measurement for a combustion field of the 0.5
kg-coal/h pulverized coal jet burner
Jun Hayashi, Nozomu Hashimoto, Noriaki Nakatsuka, Hirofumi Tsuji, Hiroaki Watanabe,
Hisao Makino, Fumiteru Akamatsu
Osaka University
9:15 Discussion on combined and emerging approaches
10:45-11:15: Coffee break
11:15 Summary and conclusion of the workshop, hot topics, discussion about a next
workshop
12:15 buffet
List of oral presentations
(alphabetic order first author)
Effects of particle coatings on laser induced incandescence
Ray Bambha, Paul Schrader, Mark Dansson, Hope Michelsen
SANDIA
Defining a measurement strategy for 2D soot particle size imaging through detailed LII
signal-decay analysis
Emre. Cenker, Gilles Bruneaux, Thomas Dreier, Christof Schulz
IFP En, Rueil-Malmaison, IVG and CENIDE, University of Duisburg-Essen
Soot primary particle sizing in turbulent flames via combined LII and Elastic Light Scattering
Brian Crosland, Matthew Johnson, Kevin Thomson
Carleton University, NRC Ottawa
Extending Time-Resolved LII to Metal Nanoparticles: Simulating the Thermal
Accommodation Coefficient
K. J. Daun, J. T. Titantah, M. Karttunen and T. A. Sipkens
University of Waterloo, University of Western Ontario
LII and one-wavelength Aethalometer measurements of particulate matter in different
environments
Silvana De Iuliis, Silvia Maffi, Francesca Migliorini, Giorgio Zizak
CNR-IENI, Milano
An experimental setup to study soot sublimation as typically occurring in high fluence LII
Klaus Peter Geigle, Gregor Gebel, Markus Köhler
German Aerospace Centre (DLR), Stuttgart
Time-resolved Laser induced incandescence measurement for a combustion field of the 0.5
kg-coal/h pulverized coal jet burner
Jun Hayashi, Nozomu Hashimoto, Noriaki Nakatsuka, Hirofumi Tsuji, Hiroaki Watanabe,
Hisao Makino, Fumiteru Akamatsu
Osaka University
Determination of the dimensionless extinction coefficient for soot generated by a PMMA
flame
Damien Hebert, Alexis Coppalle, Jérôme Yon and Martine Talbaut
CORIA, University and INSA Rouen
Influence of soot aggregate structure on particle sizing using laser-induced incandescence
Jonathan Johnsson, Henrik Bladh, Nils-Erik Olofsson, Per-Erik Bengtsson
University of LUND
Soot particles detection by LIBS and LII analysis
Francesca Migliorini, Silvia Maffi, Silvana De Iuliis, Giorgio Zizak
CNR-IENI, Milano
Pulsed laser heating of differently aged soot probed using LII and LES
Nils-Erik Olofsson, Jonathan Johnsson, Henrik Bladh, Per-Erik Bengtsson
Lund University
Recent applications of the WALS-technique
Hergen Oltmann, Stefan Will
Universität Bremen, Universität Erlangen-Nürnberg
A combined laser induced incandescence, aerosol mass spectrometry, and scanning mobility
particle sizing study of non-premixed ethylene flames
Scott Skeen, Paul Schrader, Kevin Wilson, Nils Hansen, Hope Michelsen
SANDIA, Lawrence Berkeley National Lab.
A method for inferring the soot size distribution by Static Light Scattering :
Application to the CAST soot generator
Jérôme Yon, Chloé Caumont-Prim, Alexis Coppalle, Kuan Fang Ren
CORIA, University and INSA Rouen
Effects of particle coatings on laser induced incandescence
Ray Bambha, Paul Schrader, Mark Dansson, Hope Michelsen
Sandia National Laboratories, Combustion Research Facility, Livermore, CA 94551 U.S.A.
Primary author email: rpbambh@sandia.gov
In exhaust plumes and under some combustor conditions soot particles are often coated
with unburned fuel, sulfuric acid, water, ash, and other combustion by-products.1,2 Diesel
particles, for example, can be comprised of as much as 50% volatile compounds.3 These
coatings can have an effect on particle optical properties and can thus have an influence on
optical diagnostics applied to coated particles. The effects of particle coatings therefore need to
be fully understood in order to apply optical diagnostics under a wide range of conditions.
We have compared time-resolved laser induced incandescence (LII) measurements on
uncoated soot generated in a coflow diffusion flame with LII measurements on heavily coated
soot generated in a fuel-rich premixed flame. Soot was extracted and cooled from both flames,
and a thermodenuder was used to vary the coating on the particles extracted from the premixed
flame. A scanning mobility particle sizer (SMPS) was used to monitor aggregate sizes from the
two flames, and transmission electron micrography (TEM) was used to characterize particle
morphologies. The results demonstrate striking differences in LII temporal evolution and
dependence on laser fluence between coated and uncoated particles. These results can be
understood in the context of particle energy balance during heating and cooling and are
consistent with predictions based on an LII model that includes a heavy organic coating.
References
(1) Kittelson, D. B. J. Aerosol Sci. 1998, 29, 575-588.
(2) Lighty, J. S.; Veranth, J. M.; Sarofim, A. F. J. Air Waste Manage. Assoc. 2000, 50, 1565-
1618.
(3) Witze, P. O.; Gershenzon, M.; Michelsen, H. A. Proc. SAE 2005, SAE Paper no. 2005-
01-3791.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Defining a measurement strategy for 2D soot particle size
imaging through detailed LII signal-decay analysis
E. Cenker1,2, G. Bruneaux1, T. Dreier2, C. Schulz2
1
IFP En, Rueil-Malmaison, France, emre.cenker@ifpen.fr
2
IVG and CENIDE, University of Duisburg-Essen, Duisburg, Germany
A combination of two-color soot pyrometry imaging, two-color time-resolved LII
(TiRe-LII), Laser Extinction Method (LEM) and Transmission Electron Microscopy
(TEM) of soot samples is used to define a strategy for two-dimensional imaging of
soot particle size distributions. TiRe-LII is carried out both by single point
measurements and 2D imaging, where LII signal-decay is determined for each pixel
through time-gate-sweeping of the camera gate relative to the laser pulse.
Experiments are carried out on an atmospheric laminar ethylene/air diffusion flame
from a Santoro burner with a 1064-nm laser sheet operated in the low-fluence
regime. For flame temperature measurements, two-color pyrometry images are
tomographically inverted. The resulting temperature fields are used as input for the
evaluation of the primary particle size from the local LII decay curves using the
LIISim model.
It was found that the LII signal shows different decay properties at different delays
after laser heating. As the particles cool down towards ambient temperature,
calculated decay constants increase. For closer inspection, the TiRe-LII signal is
divided into several 100-ns long segments and individual particle sizes are calculated
for each segment by two-color LIISim curve-fitting. As the time-window is shifted
further away from the laser, larger particle sizes are calculated. For a delay of 700 ns
between two segments, the calculated particle size difference is greater than 12 nm.
This variation is attributed to the polydisperse nature of the particle size distribution in
the region of interest where small particles cool down to ambient temperature within
a few hundred nanoseconds and their contribution to the detected LII signal fades
out.
The dependence of the predicted particle sizes on the boundary conditions imposed
for the simulation, such as ambient temperature, agglomeration, and accommodation
coefficients are also quantitatively investigated. For validation of the evaluated
particle sizes and uncertainty analysis, particles are sampled at different locations in
the flame above the burner head via thermophoretic sampling on TEM grids. Primary
particle sizes and dispersion are derived from TEM micrographs.
For the next step, these techniques will be applied in a high-pressure burner and a
high-pressure spray vessel. In light of the time-segmented decay analysis, an
optimized gate positioning for 2D-LII and a comprehensive simulation model strategy
will be determined.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
! " #
! " # "$
" "% ! & ! "$
" " & $ !#
"" ! & ' ($" ! $ &
& " &" ) " % "$
"!& " ! ! $! $!
* ! & "
% * & %$ "$"
&""%$ ! + & $!
( ! & " #
%$ % & " !&$ & %
,- % % & !%( &!
&&& $ " !&&& + " & "$"
! ./ !! %( &
% &./
012 " 3 4506 " & " !" &
" % & & 0 & " %
% & & !178 &
% * $ " &% ! # " &
! ,- &" " % 9 9 "
"! " %$& & " & &
& ,- & !278 %$ " %
" ! 9
, %#& !% $ $#
#": % !% &: ! $ "
!7#27 & ! ;7#77 ! " 71"" <
&&% % % ,- & $
& & % % $
%$ $ ! :% " ! !
% ! $! 9 ! &
" $ %$"
./ &5= $ >? 77; ?" + & !&&& % " 4 "
" ! # & & * "## # @1 A @2B;#@72
./C -D& <E -=F+(6 BBA C ! " %$ # &
&#: * $ !! ! "## # 2G 22 GA;1#G@
! " #
Extending Time-Resolved LII to Metal Nanoparticles:
Simulating the Thermal Accommodation Coefficient
K. J. Dauna*, J. T. Titantahb, M. Karttunena and T. A. Sipkensa
a
University of Waterloo, Waterloo ON Canada
b
University of Western Ontario, London ON Canada
*Corresponding Author, kjdaun@uwaterloo.ca
There is growing interest in adapting time-resolved laser-induced incandescence
(TiRe-LII) to size metal nanoparticles, owing to their emerging applications in
materials science. Extending TiRe-LII to new aerosols requires a model for the heat
transfer between the laser-energized nanoparticles and the surrounding gas.
Unfortunately, the thermal accommodation coefficient, α, which defines the energy
transferred when a gas molecule scatters from the particle surface, is rarely
available. This parameter can sometimes be obtained from LII measurements made
on a reference aerosol sized using electron micrography, but this process is
notoriously time-consuming, and thermophoretic sampling of metal nanoparticles is
often problematic. These challenges have precluded interpretation of data from
several pioneering TiRe-LII studies on metal nanoparticles, including one by
Murakami et al. [1] that intended to determine how the bath gas influences the growth
of molybdenum nanoparticles formed through laser-induced photolysis of Mo(CO)6.
Alternatively, it is sometimes possible to estimate α using molecular dynamics (MD).
In this technique, a pairwise potential between the gas
molecule and metal atoms is derived from ab initio
(generalized gradient approximations of density functional
theory, GGA-DFT) calculations of the gas/surface
potential. The potentials then differentiated to obtain
forces, and Newton’s equations of motion are time-
integrated to obtain atomic trajectories during a
gas/surface scattering event. Finally, α is found through
MD simulation of an argon
Monte Carlo integration over all incident gas molecular
molecule scattering from a laser- trajectories.
energized iron nanoparticle
This approach was initially used to characterize α between
soot and various gases, and is presently being extended to metal nanoparticles.
Preliminary results show that MD-derived Preliminary thermal accommodation coefficients for
accommodation coefficients are highly metal nanoparticles
sensitive to the potential well depth. αMD αexp
Ni/Ar 0.20±0.02
Unfortunately, a well-known limitation of
Fe/He 0.07±0.01 0.01 [2]
GGA-DFT is that they cannot describe the
long-range electron correlations Fe/Ar 0.04±0.01 0.1 [2], 0.13 [3]
responsible for van der Waals (vdW) Mo/He 0.006±0.002
forces, which contribute to the potential Mo/Ar 0.04±0.01
well. While the Ni/Ar interaction is dominated by a strong Casimir force, vdW forces
are thought to play a major role in other systems. Accordingly, true accommodation
coefficients are probably larger compared to ones found using ab initio derived gas-
surface potentials with no vdW correction. Current research is focused on identifying
an appropriate heuristic correction that can account for the dispersive forces.
[1] Y. Murakami, T. Sugatani, Y. Nosaka, J. Phys. Chem., 109 (2005) 8994.
[2] A. Eremin, E. Gurentsov, C. Schulz, J. Phys D: App. Phys, 41 (2008) 055203.
[3] B. F. Kock, C. Kayan, J. Knipping, H. R. Orthner, P. Roth, Proc. Comb. Inst., 30 (2005)
1689.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
LII and one-wavelength Aethalometer measurements of
particulate matter in different environments
Silvana De Iuliis, Silvia Maffi, Francesca Migliorini, Giorgio Zizak
CNR-IENI, Istituto per l’Energetica e le Interfasi, via Cozzi 53, 20125 Milano Italy,
deiuliis@ieni.cnr.it
Laser-Induced Incandescence (LII) technique is a powerful tool to measure
concentration and size of soot particulate. In this work LII measurements are
performed in different experimental conditions and compared with the ones derived
by using a commercial aethalometer. This instrument allows to obtain the on-time
concentration of optically absorbing aerosol particles by measuring the attenuation of
800 nm wavelength light through a quartz fiber filter. The filter is blackened over time
with the aerosol picked up inside the instrument at controlled flows. Measurements
are carried out with one second time-resolution. Absolute measurements in the ng/m3
range are derived for the particulate concentration. As for Laser-Induced
Incandescence, soot particles are sampled in a test cell, consisting of a pyrex tube.
The IR beam of a Nd:YAG laser (6 Hz, 200 mJ/cm2) is properly aligned within the
tube. The LII signal is detected at two wavelengths (530 nm and 700 nm) with PMT
modules coupled with interference filters. A fast digital oscilloscope, triggered by the
laser Q-switch pulse, is used for data acquisition and storage.
The two sets of measurements are carried out at the exhaust of a soot
generator (fuelled by methane) and of a diesel engine as well as in ambient air
conditions (office and laboratories). In this way, a wide range of soot load and
particulate of different nature are investigated.
Fig. 1: LII peak (left) and aethalometer measurements (right) versus time.
As an example, in Fig. 1 measurements carried out in ambient air are shown versus
time. Open symbols refer to the values of the LII peak at 500 nm wavelength
collected about every 10 minutes. The concentration values obtained with the
aethalometer are reported in closed symbols. The two sets of measurements are
quite well overlapped confirming that the two techniques are sensitive to the same
soot particulate and that the developed LII apparatus exhibits the high sensitivity
necessary for environmental measurements.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
An experimental setup to study soot sublimation as typically
occurring in high fluence LII
Klaus Peter Geigle, Gregor Gebel, Markus Köhler
Institute of Combustion Technology, German Aerospace Centre (DLR),
Pfaffenwaldring 38-40, Stuttgart, Germany, klauspeter.geigle@dlr.de
Particle sizing with LII requires modeling of the temporal behavior of the laser-
induced emission. While most models are well validated in the low fluence regime,
agreement for high fluences is identified to be insufficient, specifically when compa-
ring different models for the short time window of soot sublimation (see Fig. 15 in [1]).
As a consequence, particle sizes are typically deduced in a low fluence time-
resolved LII experiment, with the temporal fit window starting after sublimation is as-
sumed negligible. This approach becomes inconvenient with increasing pressure
when the decay rates decrease significantly towards the duration of the exciting laser
pulse. Modeling of the full LII profile is then desirable requiring best possible model-
ing of all sub-processes involved.
Sublimation of the soot surface due to a rapid temperature increase causes a
rapid vapor expansion of approximately 3-4 orders of magnitude within few nano-
seconds. This correlates with the assumption of a supersonic expansion once va-
porization becomes effective (see eq. 70 in [2]) and the related audible sound. To
confirm this assumption, attempts can be made to detect and characterize the re-
sulting blast wave. An example is identified in [3] where the expansion causes beam
steering of a monitor laser beam passing a pulsed laser heated soot volume.
Our experiment makes use of an experimental setup used to study plasma
ignition of sprays [4,5]. A green laser pulse is focused into a premixed sooting flame
and the resulting effect is monitored with a Schlieren setup involving an intensified
CCD camera for detection. Because the expansion of the created wave produces a
very weak gradient in our current setup, we had to use very high laser fluences,
clearly beyond typical LII applications. However, the expansion occurring at LII em-
ploying “high fluences” is expected to follow a similar behavior as that detected for
fluences close to plasma generation in flames. The wave expansion in the flame is
somewhat faster than speed of sound at the respective flame temperatures while ex-
trapolation to the wave origin is not possible at the required accuracy.
Based on this first approach we present ideas to optimize the experimental
setup for future experiments then suited to validate the assumptions currently em-
ployed in calculating the sublimation term in LII models.
[1] H.A. Michelsen et al., Appl. Phys. B 87,
503-521 (2007).
[2] H.A. Michelsen, J. Chem. Phys. 118,
7012-7045 (2003).
[3] K.A. Thomson et al., Proc. Combustion
Institute Canadian Section 2011 Conference, Pa-
per CICS11-39, Manitoba, Canada, 2011.
[4] G.C. Gebel et al., Proc. Fifth European
Combustion Meeting, Paper 061, Cardiff, 2011.
[5] G.C. Gebel et al., Proc. ASME Turbo Expo
2012, Paper GT2012-68963, Copenhagen, Den-
mark, 2012. Fig. 1: Exemplary picture from time
series visualizing an expansion wave
generated by a high fluence laser
pulse, detection delayed by 1.25 µs.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
! "
#$
#% &
!" #
$ % $ &$ '($$%) $*&$ + , * $ -.$
+ $%/)&+ -+0 1 2 .3 4 #5-51
! " " #"$%!&""'
()*+! , ) "$%!&""
-! --
.
. "$%!&"") . . "$%!&""
) .
! - . !
. "$%!&"")
+ - / ) "$%!&""
+ . + - + ! 012 #3 $ " )
4 3*)(' . ) -
5 5 &"" + "$%!&"" 6*( )
" + 7 &"" + +
- + . .
+ .) "$%!&"" 8* +
! + .
) ! + + *((
)
+ - - &"" + .
# +.! +' ! + .
) " +)
)" .
+ )
!
&#' (
%
$
%
!"#
+) +) 2
/ &"" + .
! " #
Determination of the dimensionless extinction coefficient
for soot generated by a PMMA flame
Damien Hebert 1, Alexis Coppalle 1, Jérôme Yon 1 and Martine Talbaut 1
1
Laboratoire CORIA - UMR 6614, CNRS - Université et INSA de Rouen, France
hebert@coria.fr
Experimental soot concentration data in flames are useful for the validation of
soot production and radiation models. Among the experimental methods available in
order to determine soot volume fractions in fluid flows, Laser Induced Incandescence
(LII) is a powerful method allowing to determine local soot volume fractions. Now, this
technic is mature enough to be applied to more and more complex situations
including in the field of fire studies, as in the present study. LII has been applied to
the determination of soot volume fraction fields in the flame of a vertical PMMA slab.
Indeed, during solid material combustion, the fuel pyrolysis is a key phenomenon,
which depends on the heat transfer between the flame and the unburnt material [1].
So, as a source of radiation, soot particles play an important role in the solid material
combustion.
An important step of the LII signal analysis is the calibration, in order to
determine the relationship between the LII signal and the soot volume fraction [2].
The most-used approach for calibration consists in measuring the light extinction
coefficient Kext, which depends on soot volume fraction fv. Kext is also a function of
soot particle morphology and of the optical index of the soot matter, which varies with
the wavelength. But much of the optical index or Kext data for soot have been
determined from measurements in gaseous or liquid fueled flames but few for the
solid combustion. Additionally, Kext has been usually determined ex-situ for soot
samples at ambient temperature [3]. Therefore one can wonder if the soot optical
properties at standard or flame temperatures can be considered similar. The present
work focuses on this question.
In this context, the spectral value of Kext has been measured by using in-situ
extinction measurements with a white laser beam (Leukos) crossing the flat flame of
a PMMA slab. With the same experimental setup, Kext is also determined in a
gaseous fueled flame generated by a bronze porous burner and fed with a mixing of
methane and ethylene. So the determined Kext coefficients are relevant of the soot
optical properties at high temperatures. In order to observe the influence of the soot
temperature on the spectral variation of Kext, it has been also determined by ex-situ
measurements after sampling of soot in the flame [4]. This sampling has allowed
additional measurements, the agglomerate soot particle size using a Scanning
Mobility Particle Sizer (SMPS) and the mass concentration with a Tapered Element
Oscillating Microbalance (TEOM).
The experimental results are analyzed to compare the spectral variations of
Kext in the 400-1100 nm range. Finally, different evaluations of the dimensionless
extinction coefficient Ke=Kext/fv are proposed corresponding to soot generated by
gaseous or solid combustion at standard or flame temperatures. A quantitative
comparison with the values found in the literature is presented.
[1] A.C. Fernandez-Pello and T. Hirano, Combust. Sci. Technol. 32 (1983) 1-31.
[2] C. Schulz et al, Appl. Phys. B 83 (2006) 333-354.
[3] G.W. Mulholland and M.Y. Choi, Symp. Int. Combust. 27 (1998) 1515.
[4] J. Yon et al, Appl. Phys. B 104 (2011) 253-271.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Influence of soot aggregate structure on particle sizing using
laser-induced incandescence
Jonathan Johnsson1, Henrik Bladh2, Nils-Erik Olofsson3, Per-Erik Bengtsson4
Combustion Physics, Lund University, Box 118, SE-22363, Lund, Sweden
1 2
E-mail: jonathan.johnsson@forbrf.lth.se, E-mail: henrik.bladh@forbrf.lth.se,
3 4
E-mail: nils-erik.olofsson@forbrf.lth.se, E-mail: per-erik.bengtsson@forbrf.lth.se
Soot aggregates formed in combustion processes can be described as
random fractal structures. For theoretical studies of the physical properties of such
aggregates, they have often been modelled as spherical primary particles in point
contact. However, transmission electron microscopy (TEM) images show that the
primary particles in general are more connected than in a single point; there is a
certain amount of bridging between the primary particles. The results of particle
sizing using laser-induced incandescence (LII) is crucially dependent on the heat
conduction rate from the aggregate, which, in turn, depends on the amount of
bridging.
In this work, aggregates with bridging are modelled using overlapping
spheres, see Fig. 1, and it is shown how such aggregates can be built with specific
fractal parameters. Aggregates with and without bridging are constructed, and it is
investigated how the bridging influences the heat conduction rate in the free-
molecular regime. It is shown that bridging has a significant influence on the shielding
parameters that are inferred from the heat conduction results, Fig. 2. These results
are used together with an LII model to show how LII particle sizing is affected by the
difference in bridging.
Figure 1. Example aggregate with 100 primary Figure 2. Example of shielding values, Ș, for
particles and 25 % bridging (kf = 2.3 and aggregates with point contact and with bridging. Np
Df = 1.8). denotes the number of primary particles per
aggregate and the heat accommodation coefficient is
here set to ĮT = 1.0.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Soot particles detection by LIBS and LII analysis
Francesca Migliorini, Silvia Maffi, Silvana De Iuliis, Giorgio Zizak
CNR-IENI, Istituto per l’Energetica e le Interfasi, via Cozzi 53, 20125 Milano Italy,
deiuliis@ieni.cnr.it
Laser-Induced Breakdown spectroscopy (LIBS) is an atomic emission
spectroscopy technique that has been used for elemental analysis of solid, gaseous
and aerosol samples. The LIBS technique involves a pulsed laser beam focused
onto the sample to create a microplasma. The resulting optical breakdown
decomposes and excites all species within the plasma volume. The light emission is
characterized by a continuum spectrum (Bremsstrahlung) containing discrete atomic
emission lines. Both the continuum spectrum and the atomic lines decay with time. In
general, the continuum spectrum decay faster than the atomic lines allowing the
possibility of detecting atomic lines with a good signal-to-noise ratio by adjusting the
delay and the integration time of the detector gate.
In this work the applicability of the LIBS technique to the detection of
carbonaceous particulate in a combustion environment is investigated. In particular, a
comparison of the carbon atom concentration derived with LIBS and
LIImeasurements is performed. Soot particles produced by an ethylene-fueled soot
generator are sampled and pumped for LIBS analysis into an optically equipped
sample chamber, the outlet of which is then piped in the LII measuring equipment. As
for LIBS the IR beam of a Nd:YAG laser was focused to create the plasma and the
relative spectral emission was collected onto a fiber bundle coupled to a
spectrograph-ICCD unit. As for LII, an home-made portable instrument has been
used. The LII signal is detected at two wavelengths (530 nm and 700 nm). A fast
digital oscilloscope is used for data acquisition and storage.
Since in LIBS technique the emission line is attributed to a particular atomic
element whatever is the initial molecular species containing that element, particular
care has to be taken in applying the technique to a combustion environment. In fact,
in the case of carbonaceous particles, the elemental analysis does not allow to
discriminate the contribution of soot particle and gas-phase species to the carbon
atoms measured. The aim of the work is to develop a new methodology to select the
contribution of soot particles carbon atoms in the LIBS signal.
In order to discriminate the contribution from soot particles in LIBS signals, a
proper choice of the laser operating condition is performed. The laser energy is
reduced to a value such that in a pure gas phase environment no breakdown is
produced and the LIBS signal is induced by the presence of particles in the probe
volume. However, even in these conditions carbon atoms coming from gas species
are activated as well. To discriminate the two contributions to the LIBS signal a
comparison is carried out between measurements below and above the breakdown
threshold. The results confirm that the LIBS technique, applied with the developed
procedure, is able to detect soot particles measuring soot concentration in agreement
with LII measurements.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
!" # $# %&&$! $
' )
!" ($! %&&$! $ !" * ($ + %&&$! $
! "
#
$ "
" %&
'() "
"" *" + ,-./ (012
345 %&
(012
6 7 "
8 #.*
9 ( "
% #.* "
9 '() "
: 9 5
& #$%
! "
#$% & #$%
'() / ! - ; 7 ! $ ! < # =.
= . > 22 5003
! " #
Recent applications of the WALS-technique
Hergen Oltmann1, Stefan Will2
1
Technische Thermodynamik, Universität Bremen, Germany
2
Lehrstuhl für Technische Thermodynamik, Universität Erlangen-Nürnberg, Germany
email: stefan.will@ltt.uni-erlangen.de
Nanoparticles produced in combustion processes often exhibit complex fractal
structures. While laser-induced incandescence (LII) is a proven technique for the
determination of primary particle size no information about aggregate sizes can be
obtained. To gather information about aggregate size and fractal dimension elastic
light scattering (ELS) [1] is an often used in situ method.
The wide-angle light scattering (WALS) approach [2] extends classical ELS-concepts
by using a combination of an ellipsoidal mirror and an intensified CCD-camera. The
ellipsoidal mirror redirects the light scattered within a plane onto the CCD-chip (cf.
Fig. 1), which makes it possible to almost instantaneously record a complete
scattering diagram over an angular range of approx. 10° to 170° with an angular
resolution ∆θ of typically 0.6°.
The basic performance of the approach was demonstrated previously by
measurements on soot particles in laminar premixed flames [2]. This contribution
highlights various recent developments and applications of the technique. These
include measurements in a turbulent diffusion flame [3], employing a pulsed laser
and underlining the favourable applicability to unsteady processes. Also
measurements with a particular high resolution of ∆θ = 0.3° were performed which
allow for a detailed investigation of selected angular regions. To simultaneously
measure the vv- and hh-scattering components polarization foils were mounted in
front of the ellipsoidal mirror. Radii of gyration obtained for soot particles in a
premixed ethene flame show good agreement with former results. Furthermore
investigations on silica particles produced in a diffusion flame were carried through
(cf. Fig. 2) for various relative velocities between the precursor flow (nitrogen flow
saturated with hexamethydisiloxane) and the methane/oxygen flow of the supporting
flame. Recorded scattering diagrams indicate a change in the structure of the silica
particles for the different velocities.
Fig. 1: Experimental setup Fig. 2: Measurement on silica particles in a diffusion flame
[1] C. M. Sorensen, Aerosol Sci. Technol. 35, 648-687 (2001)
[2] H. Oltmann, J. Reimann, S. Will, Combust. Flame 157, 516-522 (2010)
[3] H. Oltmann, J. Reimann, S. Will, Appl. Phys. B 106, 171-183 (2012)
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
A combined laser induced incandescence, aerosol mass
spectrometry, and scanning mobility particle sizing study of
non-premixed ethylene flames
Scott Skeen1, Paul Schrader1, Kevin Wilson2, Nils Hansen1, Hope Michelsen1
1
Sandia National Laboratories, Combustion Research Facility, Livermore, CA 94551 U.S.A.
2
Lawrence Berkeley National Laboratory, Berkeley, CA 94720 U.S.A.
Primary author email: hamiche@sandia.gov
Investigations into the chemical composition of soot particles often rely on samples extracted
from the flame environment for subsequent analysis by mass spectrometry. These mass spectra have
contributed to the general consensus that polycyclic aromatic hydrocarbons (PAHs) are involved in soot
1,2
nucleation and surface growth processes. In the pioneering work of Dobbins et al., soot and soot
precursor particles were extracted along the centerline of a non-premixed coflow flame by rapid-insertion
thermophoretic sampling and subsequently analyzed by laser microprobe mass spectrometry (LMMS).
Low in the flame, where TEM images suggested the particles were liquid-like, the mass spectrum was
dominated by species between 200 and 300 amu in size. Higher in the flame, where TEM images
indicated that carbonaceous aggregates were being formed, PAHs were no longer observed in the mass
spectra. The laser-desorption ionization experiments of Bouvier et al.,3 Lemaire et al.,4 and Faccinetto et
5 1,2 4
al. complemented and expanded the findings of Dobbins et al. Lemaire et al. showed that the fuel
5
composition strongly influences the soot composition. Faccenetto et al. developed a new sampling
method enabling some distinction between PAHs in the gas phase and those adsorbed onto the soot
particles. Intrusive sampling techniques such as those used in the studies referenced above perturb the
flame. In some instances such techniques may permit the agglomeration of existing particulates,
condensation of low vapor pressure species onto the surface of soot nuclei, and nucleation of new
clusters that could later be erroneously associated with nascent soot. Recently, we have coupled a non-
premixed, opposed-flow flame system to an aerosol mass spectrometer to investigate the chemical
composition of soot particles extracted from different regions of the flame. The present work provides
insight into the effects of our intrusive sampling method on the observed soot composition and size
distributions by combining laser-induced incandescence (LII) measurements with intrusive particle
diagnostics.
In this work, we performed time-resolved LII measurements in conjunction with flame-sampling
aerosol mass spectrometry (AMS) and scanning mobility particle sizing (SMPS) measurements to (1)
investigate differences in particle size and composition as a function of temperature and position in the
flame and (2) investigate processes occurring within the sampling system during sample extraction. We
probed three non-premixed, opposed-flow, ethylene flames at conditions ranging from nearly sooting to
moderately sooting. In situ time-resolved LII measurements revealed differences in the temporal
response as a function of laser fluence and position in the flame. These differences may result from
varying absorption coefficients due to particle composition and/or differences in the extent of particle
surface coatings. Particles extracted from different regions of the flame also yielded varying temporal LII
profiles, size distributions (as determined by SMPS), and chemical compositions (as determined by AMS).
The use of a thermal denuder prior to the ex situ LII, SMPS, and AMS instruments also provided
information on the extent of PAH condensation onto existing particles and the propensity for liquid- or tar-
like droplet nucleation within the sampling system.
References
(1) Dobbins, R. A.; Fletcher, R. A.; Chang, H. C. Combustion and Flame 1998, 115, 285-298.
(2) Dobbins, R. A.; Fletcher, R. A.; Lu, W. Combustion and Flame 1995, 100, 301-309.
(3) Bouvier, Y.; Mihesan, C.; Ziskind, M.; Therssen, E.; Focsa, C.; Pauwels, J. F.; Desgroux, P. Proc. Combust. Inst. 2007, 31, 841-849.
(4) Lemaire, R.; Faccinetto, A.; Therssen, E.; Ziskind, M.; Focsa, C.; Desgroux, P. Proc. of the Combustion Institute 2009, 32, 737-744.
(5) Faccinetto, A.; Desgroux, P.; Ziskind, M.; Therssen, E.; Focsa, C. Combustion and Flame 2011, 158, 227-239.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
A method for inferring the soot size distribution
by Static Light Scattering :
Application to the CAST soot generator
Jérôme Yon1, Chloé Caumont-Prim1, Alexis Coppalle1, Kuan Fang Ren1
Laboratoire CORIA - UMR 6614, CNRS - Université et INSA de Rouen, France
yon@coria.fr
The soot size distribution is often determined by using ex-situ granulometers,
after sampling of the particles. But the quenching of aggregation process during the
sampling is difficult to control and raises the question of representativeness of the
results (Ouf et al. 2010 [1]). For this reason, optical measurements are more suitable.
Thanks to the approximation of Rayleigh – Debye – Gans for Fractal Aggregates
which is well suitable for such particles, (RDG-FA theory - Dobbins et al., 1991 [2]),
measurement of static light scattering (SLS) can be interpreted in order to determine
a size parameter called gyration radius. The inversion of experimental data by this
theory to infer the gyration radius of monodisperse aggregates has been recently
validated (Caumont et al. 2010 [3]).
The SLS technique consists in measuring the signal scattered by the particles
after interaction with a polarized laser light (532 nm in our case) at different scattering
angles. Since a long time, some authors have proposed to determine a
representative gyration radius of the polydisperse population with this optical
technique (Dobbins & Megaridis 1991 [2], Köylü 1994 [4]). More recently, inversion
methods have been proposed by coupling scattering and extinction measurements
(Koylu & Faeth (1996) [4] and Iyer et al. (2007) [5]). But these methods rely on the
knowledge of soot optical index which is unfortunately not accurately known.
Moreover Burr et al [6] by using Bayes’ theorem showed mathematically that the
inverse problem is ill-posed..
This work presents a new inversion method for the determination of soot size
distribution in flames by measuring scattered light at different angles. It consists in
determining for each studied angle, by using the RDG-FA theory, a gyration radius
Rg*(ș) of a monodisperse population which has the same optical behavior as the real
polydisperse population. The so determined Rg*( ș) function informs us about
polydispersity of soots. The method has been validated by comparing obtained
results with size distribution determined by Transmission Electron Microscopy (TEM).
The method is now applied to characterize soot size distribution generated by the
Jing CAST apparatus that is suspected to become a standard for soot generation
including for LII calibration. Results are compared with soot size distribution
determined by DMS500 apparatus.
[1] Ouf, F.X., Yon, J., Ausset, P., Coppalle, A. Maillé, M. Aerosol Sci. Technol., 44, (11),
1005-1017. (2010).
[2] Dobbins, R.A., Megaridis, CM, Applied Optics, 30 (33), 4747-4754 (1991).
ième
[3] Caumont, C., Yon, J., Ren, K.-F., Coppalle, A. Proc. 26 Congrès Français sur les
Aérosols, Paris, France (2011).
[4] Köylü, Ü, Ö, Combustion and flame, 109, 488-500(1997).
[5] Iyer S. S., Litzinger, T. A., Lee, S-Y, Santoro R. J., Combustion and flame, 149 (1-2), 206-
216 (2007).
[6] D.W. Burr, K.J. Daun, O. Link, K.A. Thomson, G.J. Smallwood, Journal of Quantitative
Spectroscopy and Radiative Transfer, Vol. 112, Issue 6, pp. 1099-1107, (2010)
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
List of posters
(alphabetic order first author)
Sizing of soot aggregates by two-dimensional multi-angle light scattering (2D-MALS)
Michael Altenhoff, Jannis Reisky, Stefan Will
Universität Erlangen-Nürnberg, Germany Universität Bremen
Study of the wavelength dependence of the soot absorption function using the two-excitation
wavelength Laser Induced Incandescence: application to fluorescent species detection.
S. Bejaoui, R. Lemaire, E. Therssen, P. Desgroux
PC2A, University of Lille 1, Ecole des Mines de Douai
Comparison between Modeled and Measured Time-Resolved LII Signals and Soot Temperatures in a
Laminar Premixed Flame
S. Bejaoui, S. Batut, E. Therssen, P. Desgroux, F. Liu, K. A. Thomson, G. J. Smallwood
PC2A, University of Lille 1 NRC Ottawa,
Continuous Wave LII in an Atmospheric Pressure Kerosene Flame
John D. Black and Paul Wright
University of Manchester
Applicability of Wright’s correction to Fuchs boundary sphere method for TiRe-LII calculations
K.J. Daun, S.C. Huberman
University of Waterloo
Influence of temporal laser pulse length and shape on the time resolved laser induced incandescence
signal
M. Ditaranto, N.E. Haugen, C. Meraner, I. Saanum
SINTEF, Trondheim
Combined LII and LIF with multiple excitation wavelengths for diagnostics of soot and PAHs in
laminar flames
J. Dunn, I.S. Burns
University of Strathclyde
Experimental study of particle vaporization under pulse laser heating by LII and laser light extinction
Alexander Eremin, Evgeny Gurentsov, Ekaterina Mikheyeva, Konstantin Priemchenko
Joint Institute for High Temperature, Russian Academy of Sciences
Experimental investigation of the influence of inert gas on soot formation
A.Flügel, S. Beer, S. Will, J. Kiefer, A. Leipertz
University of Erlangen-Nürnberg, SAOT Erlangen, University of Aberdeen
Combination of high spatial resolution LII and LOSA measurements for determination of soot volume
fraction and PAH concentration in laminar diffusion flames
M. Leschowski, K. Thomson, D. Clavel, D. Snelling, C. Schulz, G. Smallwood
IVG and CENIDE, University Duisburg-Essen, NRC Ottawa
Combination of various particles measurement techniques for validation in laminar high pressure
flames
M. Leschowski, T. Dreier, C. Schulz
IVG and CENIDE, University Duisburg-Essen
Effect of laser pulse duration on laser-induced incandescence soot
F. Liu, G.J. Smallwood
NRC Ottawa
Measurements of Soot Volume Fraction under Conditions Relevant to Engine Exhaust Using Four-
Colour LII and Different Laser Energies
Fengshan Liu, Xu He, Hongmei Li, Fushui Liu, Gregory J. Smallwood
NRC Ottawa, Beijing Institute of Technology
Relationship between LII Signal and Soot Volume Fraction – Effect of Primary Particle Diameter
Polydispersity
Fengshan Liu, Gregory J. Smallwood
NRC Ottawa
Novel soot volume fraction measurement through ratio-pyrometry and absolute light calibration
Bin Ma and Marshall B. Long
Yale University
Modeling laser-induced incandescence of soot integrating spatial and temporal dependences of
parameters involved in energy and mass balances
Mohammed Mobtil, Romain Lemaire
Ecole des Mines de Douai
Evolution of the LII signals of soot particles measured in low pressure methane flames
T. Mouton, P. Desgroux, X. Mercier
PC2A, University of Lille1
Influence of LII on Soot Optical Properties in Reference Flames
K. Thomson, K-.P. Geigle, D. Snelling, F.Liu, M. Köhler, G. Smallwood
NRC Ottawa, DLR Stuttgart
Evaluation of particle sizes of iron-oxide nano-particles in a low-pressure flame-synthesis reactor by
simultaneous application of TiRe-LII and PMS
B. Tribalet, A. Faccinetto, T. Dreier, C. Schulz
IVG and CENIDE, University Duisburg-Essen
Comparison of different techniques for measurement of soot and PM emission from Diesel engine
Richard Viskup
Institute for Design and Control of Mechatronical Systems,Johannes Kepler University
Sizing of soot aggregates by
two-dimensional multi-angle light scattering (2D-MALS)
Michael Altenhoff1, Jannis Reisky2, Stefan Will1
1
Lehrstuhl für Technische Thermodynamik, Universität Erlangen-Nürnberg, Germany
2
Technische Thermodynamik, Universität Bremen, Germany
stefan.will@ltt.uni-erlangen.de
For the understanding of soot formation in combustion processes
comprehensive information about local size properties of complex soot aggregates is
desired. Elastic light scattering (ELS) is a well-established optical technique which
allows for the in situ determination of aggregate size and fractal dimension of soot
particles in flames [1]. Reimann et al. [2] used a two-dimensional combination of ELS
and laser-induced incandescence (LII) for the determination of various parameters of
soot particles in a premixed flame from a porous flat flame burner (McKenna type).
Although the general approach was successful both the measuring range in terms of
aggregate size and the information obtained were limited because of the use of a
fixed scattering angle of 90°.
In continuation and extension of this approach we performed two-dimensional
ELS-measurements under various scattering angles thus allowing for a simultaneous
acquisition of particle parameters at various heights above burner (HAB).
Measurements on a premixed ethene flame from a McKenna type burner with an
equivalence ratio of 2.7 were carried out by irradiating a laser-light-section and
detecting the scattered light using an intensified CCD camera (cf. Fig 1). The
detection angle varied equidistantly in the scattering vector q from 17° to 163°, and
the evaluation of obtained data was carried out for each pixel line from 10 mm to
20 mm HAB for three different areas: the flame axis only, the area determined by the
depth of field and the maximum evaluable region. The obtained radii of gyration show
good agreement with former results.
Fig. 1: Experimental setup
[1] C.M. Sorensen, Aerosol Sci. Technol. 35: 648-687 (2001)
[2] J. Reimann, S.-A. Kuhlmann, S. Will, Appl. Phys. B 96: 583-592 (2009)
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Study of the wavelength dependence of the soot absorption
function using the two-excitation wavelength Laser Induced
Incandescence: application to fluorescent species detection.
S. Bejaoui1, R. Lemaire2, E. Therssen1, P. Desgroux1
1
Laboratoire PC2A, UMR CNRS 8522, F-59655 Villeneuve d’Ascq, France
2
EMDouai, EI, F-59500 Douai, France
salma.bejaoui@ed.univ-lille1.fr
In this work, wavelength dependence of the absorption function of soot was
experimentally studied. We used a technique developed in our team which consists to heat
similarly the soot particles using two different laser excitation wavelengths. Thus, using two
lasers with the same temporal and spatial irradiance profiles, it is possible to find
combinations of the both lasers energies, below sublimation regime activation, insuring that
soot particles absorb the same energies, reach the same temperature and emit the same
Laser Induced Incandescence (LII) radiation [1]. Laser at 1064 nm is always chosen as a
reference excitation and compared with a UV-visible wavelength (λi) such as 266 nm, 355
nm, 532 nm. In this way we can deduce the relative evolution of the absorption function
E (m, λi )
versus wavelength.
E (m,1064nm)
Experiments are investigated in a turbulent diffusion flame of pulverised diesel and in
a premixed methane flame stabilized on a McKenna burner. It is found that up to 700 nm the
emission signal due to PAH /soot precursor LIF interferes with the LII one. Interestingly no
LIF emission could be identified above 700 nm. Therefore this spectral region appears very
attractive to collect soot incandescence in flames containing PAH.
The two-excitation wavelength LII method has been checked for the first time using a narrow
spectral detection set above 700 nm, by using several combinations of UV-visible and IR
radiations. Soot particles heating was controlled either looking at the Planck function above
700 nm or controlling the decay rate of the LII temporal signals. Once similar heating is
reached using any UV-vis radiation and the 1064 nm one, the method is used to get either
the ratio of soot absorption functions, or the LIF spectra of soot precursors even in the
presence of soot.
[1] Therssen E., Bouvier Y., Schoemaker-Moreau C., Mercier X., Desgroux P., Ziskind M., Focsa C. ,
Appl. Phys. B 89, 417-427 (2007).
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Comparison between Modeled and Measured Time-Resolved
LII Signals and Soot Temperatures in a Laminar Premixed
Flame
S. Bejaoui, S. Batut, E. Therssen, P. Desgroux
Laboratoire PC2A, UMR CNRS 8522, F-59655 Villeneuve d’Ascq, France
salma.bejaoui@ed.univ-lille1.fr
F. Liu, K. A. Thomson, G. J. Smallwood
National Research Council, Building M-9, 1200 Montreal Road, Ottawa, ON., Canada
LII experiments were conducted in a laminar premixed flame established with
a McKenna burner at atmospheric pressure burning a mixture of methane, oxygen,
and nitrogen with an equivalence ratio of 2.15 and with a flame stabilizer of stainless
steel plate placed at 20 mm above the burner. A Nd:YAG laser at 1064 nm with a
repetition rate of 10 Hz was used to produce a top-hat laser beam with a 6 ns pulse
duration (FWHM) as the excitation source. Time-resolved LII signals were measured
at 610 nm (20 nm FWHM) using a PMT at different locations in the flame and at
different laser fluences. Soot temperature measurements were also conducted
through recording LII spectra with a spectrograph at different locations along the
flame centerline and different laser fluences. TEM analyses of soot sampled at HAB
= 12 and 15 mm were also carried out to provide primary particle diameter
distribution and average number of primary particles in an aggregate for LII model
calculations.
Preliminary model calculations suggest that the base model LII developed at
NRC was able to reproduce the experimental resolved LII signals accurately in the
low-fluence regime; however, large discrepancies between the model and the
experimental results occur at high fluences. To understand the role of physical and
chemical processes that were not incorporated into the NRC LII model, annealing
and photodesorption were implemented and their effects on the LII model results
were investigated. This study is aimed at improving the NRC LII model at high laser
fluences through a detailed comparison between the experimental LII results and the
modeled LII results.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Continuous Wave LII in an Atmospheric Pressure
Kerosene Flame
John D. Black and Paul Wright
School of Electrical and Electronic Engineering, University of Manchester, UK
John.black-2@manchester.ac.uk
Fibre and diode lasers with sufficient power to heat soot particles to
incandescent temperatures are readily available at lower cost than the nanosecond
pulsed lasers traditionally used in LII. There are less stringent safety restrictions on
the use of CW lasers and they can be delivered with excellent beam quality through
standard optical fibres, making them more suitable for LII in practical environments.
Using the collimated beam from a diode laser at 803 nm in the power range 5 – 30
W, LII was easily observable in a highly sooting kerosene flame (Fv ~10-5). However,
the laser causes major changes in the combustion, increasing soot burn out rates
and transferring heat to other regions of the flame.
In contrast to short pulse LII, soot
particles experience laser heating and cooling by
heat transfer at rates comparable with their
reaction rate. Their residence time in the beam
and other processes such as photophoresis and
optical trapping also have to be considered.
Hence, modeling is much more complicated than
for short pulse LII, and the processes are not well
understood.
Visible emission spectra were collected
using a traversable fibre optic probe from a
magnified projected image of the flame shown in
Figure 1. There is a good match between
predicted emission spectra based on the Figure 1: LII in a quasi-2-D
kerosene lamp flame with 28 W
blackbody curve and observed spectra from the
1 mm diameter cw laser beam
flame in the wavelength range 590 – 790 nm. photographed through a BG3 filter
From these spectra estimated soot temperature
in the absence of the laser is 2150 K, rising to 2600 K in the region of a 28.5 W laser
beam. Temperature rise is linear in laser power. Local soot temperature is increased
both above and below the beam when the laser is present. Above the laser beam,
light emitted at 700 nm decreases quadratically with distance from the height of the
centre of the laser beam to the edge of the visible flame, although the soot particle
surface temperature remains at ~2350 K in this upper part of the flame. The intensity
of light emitted at 700 nm at the centre of the laser beam at varying laser power is in
good agreement with a prediction based on blackbody radiation. This indicates that
the mechanism of increased light emission is particle heating (LII) and not creation of
additional soot by laser stimulated reactions in the flame.
Although cw LII is at a very early stage of development, the potential for
combustion diagnostics – soot concentration, temperature, velocity by flow tagging,
etc. – is obvious. The observations described here should provide a basis for future
investigation of the processes involved.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Applicability of Wright’s Correction to Fuchs’ Boundary
Sphere Method for TiRe-LII Calculations
K. J. Daun*, and S. C. Huberman
University of Waterloo, Waterloo ON Canada
*Corresponding Author, kjdaun@uwaterloo.ca
When LII experiments are performed on high pressure aerosols, transition-regime
heat conduction from the laser-energized particles is usually calculated using Fuchs’
boundary sphere method. In this technique the Knudsen layer is represented by a
collisionless boundary sphere enveloping the particle, which in turn is surrounded by
a continuum gas. The analysis proceeds by equating the
$(#)" Q heat transfer through the two domains and then solving
l for the unknown boundary sphere temperature, T .
=E[$&l3#']"
#" This calculation requires specification of the spherical
P
shell thickness, , which is usually chosen as the mean
free path at T , ! "="!(T ). Filippov and Rosner [1]
a
instead advocate a more complex equation that
O
accounts for particle curvature and the directional
distribution of incident molecules, originally proposed by
Fuchs [2] and derived by Wright [3] to model evaporating
droplets. If a colliding molecule has travelled a distance
l from its most recent collision at an angle # relative to
the surface normal, the corresponding radial distance is $(l,#)"="(l2+a2+2lacos#)1/2%"a.
By integrating over all incident angles, the expected value of $(l, #) for a given l is
) & '& ' *
3
a3 , &1 + l a ' 1 + l 2 a2 1 + l 2 a
(2 5
2 l2
$ & l ' . 1 $ & l, # ' P# & # ' d# . 2 & ' % 2-
52
% + 1 + l 2 a2 (1)
l , 5 3 15 a -
0
/ 0
where P#(#) = 2cos#sin#. Filippov and Rosner [1] set l"="! in Eq. (1) to find , while
Wright [2] also considers the distribution of incident paths, Pl(l)= 1/! exp(%l/! ),
2 2
1
. 1 $ & l ' Pl & l ' dl . 1 $ & l ' exp & % l ! ' dl (2)
0 0
!
which can be solved numerically.
We use Direct Simulation Monte Carlo to investigate this phenomenon under
typical LII conditions. The Knudsen layer thickness is found by sampling the radial
distance that incident gas molecules travel
before they collide with the surface. The 0.90
Wright [3]
DSMC results reveal that particle curvature
Filippov and Rosner [1]
increases the Knudsen layer thickness 0.85
DSMC
compared to a flat surface ( /! "="2/3), an
effect captured by both Wright’s equation 0.80
/!
[3] and Filippov and Rosner’s [1]
approximation. This correction has a 0.75
negligible influence on transition regime
heat transfer rates, however, especially 0.70 4! = 2/3
considering other uncertainties involved in
the calculation, so it can be safely excluded 0.65 -1 0
10 10
when analysing TiRe-LII data. Kn
[1] A. V. Filippov, D. E. Rosner, IJHMT 43 (2000) 127.
[2] N. A. Fuchs, Soviet Phys. Tech. Phys. 3 (1958) 140.
[3] P. Wright, Discussions of the Faraday Society 30 (1960) 100.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
!
! " #"$ %" !
"#&'( ) * " "# " ! " +,"-
./((01 2 #"$.3 -/('(1) !
%! 4 " " # "#!
5('((( !" ! ) /56"4
" ! % # !" !
!6 . " 1) " 4 $ 6
" 4 * .'("/( 1 7
!"# "! " % ! ! " $ "
'( 6 !# 899 5:/ "#
3 ! !6 "# ## "#! 6
4 6 - ! " ! # "!
) # 6 "6 4 " # ' "#
! $ 6 ! * #" "#!
6"6 !) $ #" 5( ! %!
! " ! ! 6 !
" "6 4 " "# # 5( .
4 1)"6 4" %! #"" !
# - " " . " " $ 16 "#
4!"+" . 2+ 1 #"
## "# # %! ;</ ! " $
!" # ! "/(( . 4 1" ""
# 6 " 6" "# $ ) "#
! " " " "# !" - " "6 4" 4 $
" ! ## = "" "#" ! 6
#" " " 4 ! " " # 6 " "
"! ! " " " !! ! # "
# #"""##" > /" 6"
! " " ! ""6 4 " #
#"! " &5 ) 6 4" " 6""!
* ! !" !"# ) ! " 4 $
" # : " % ## "# # #""
%" !" # " " # ## "#
# " " # ' /
? /(( <'((;@ '((;3 -0(( @ ) " 4 ! -
'(( <85;@3 5( <:(; ? /(( @ '(( @3 $ " !6 "
5(
! " #
Combined LII and LIF with multiple excitation wavelengths for
diagnostics of soot and PAH in laminar flames
Jaclyn Dunn 1, Iain Burns 1
1
Department of Chemical & Process Engineering, University of Strathclyde, Glasgow, UK
e-mail:iain.burns@strath.ac.uk
Through the use of two complimentary laser techniques, laser induced
incandescence and laser induced fluorescence, the formation of soot from polycyclic
aromatic hydrocarbons (PAH) has been studied. Information on soot volume fraction
(from signal peak intensity) and particle cooling rate (from the signal decay time) is
obtained from incandescence signal, while fluorescence measurements offer
information on PAH present. These techniques were used to study a premixed flat
flame of ethylene and air at a range of equivalence ratios. This project involves the
use of three different excitation wavelengths (1064 nm, 532 nm, 290 nm). The
fluences were adjusted so that the soot particles are heated to the same temperature
by each excitation wavelength, resulting in equal incandescence intensity. Since no
fluorescence is detected for 1064 nm excitation the contributions from incandescence
and fluorescence can be separated. This is achieved by subtracting any signal
obtained for 1064 nm excitation from the signals obtained for excitation at shorter
wavelengths, leaving the remaining signal to be attributed to fluorescence. A
monochromator has been used to resolve the signals, thus generating a time-
sequence of emission spectra. This approach is helpful in identifying the
contributions of LII and LIF to the signals detected.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Experimental study of particle vaporization under pulse laser
heating by LII and laser light extinction
Alexander Eremin, Evgeny Gurentsov, Ekaterina Mikheyeva, Konstantin
Priemchenko
Joint Institute for High Temperature Russian Academy of Sciences, 125412 Izorskaya
st.13(2), Moscow, Russia, e-mail: gurentsov@ihed.ras.ru
Particle vaporization is one of the uncertainties in LII measurements of
particle size or volume fraction. The common recommended threshold of energy
fluence for soot vaporization is in the range of 0.1-0.2 J/cm2. However, for non-soot
particles or for carbonaceous particles different from soot, the vaporization threshold
might be unknown. One of the reasons is the decreasing of the particle refractive
index function E(m) with particle size [1]. Due to this effect the particle heat up
temperature couldn’t reach the vaporization threshold at low fluences. Besides that,
the vaporization temperature could decrease with particle size. The knowledge about
particle vaporization process is useful not only for correct LII measurements, but also
for the determination of particle thermodynamics properties changing due to size
effect [2]. Direct characterization of vaporization process of soot was performed by
simultaneous scattering and LII registration [3], by emission spectroscopy [4] and by
two pulse lasers [5] in flames. The goal of this work is the application of laser light
extinction measurements and Ti-Re LII for observation and analysis of vaporization
process of small carbon and iron particles.
The carbon particles were formed in pyrolisis of 1% C6H6+Ar behind reflected
shock wave. Two-color Ti-Re LII technique was applied for particle heat up
temperature and size measurements at fluencies around 0.4 J/cm2. He-Ne laser
beam was adjusted coaxially to the YAG (1064 nm) laser beam and allowed to
observe the decreasing of a volume of condensed phase due to vaporization.
Additionally, the real particle temperature equilibrated with bath gas during pyrolisis
process was measured by emission-absorption spectroscopy in visible range of
spectra. The measured gas-particle temperature was less than frozen temperature
behind shock wave due to endothermic effect of C6H6 decomposition. The
vaporization temperature of small growing carbon particles with mean diameters of 2-
14 nm was found to be in the range of 2900-3100 K in contrast to soot vaporization
temperature 4000 K [6].
Growing iron particles of different sizes (2-11 nm), synthesized in the laser
photolysis reactor [1], were heated by YAG laser pulse with fluences of 0.025-0.7
J/cm2. The same technique as for carbon particles was used for condensed phase
loss, temperature and size measurements. The essential difference of iron particles
vaporization temperature (2100-2700 K) from the bulk one (about 3050 K) in
dependence on particle size and pressure of a bath gas was found.
The dispersion of vaporization temperature observed in both experiments is
probably caused by with the variation of particle properties formed at different
conditions. The related value of evaporated fraction of condensed phase and particle
vaporization temperature are analyzed in dependence on experimental conditions.
[1] Eremin A., Gurentsov E., Popova ȿ., Priemchenko K. Appl. Phys. B, 2011, 104(2), 285.
[2] Xiong S., Qi W., Cheng Y., Huang B., Wang M., PCCP, 2011, 13, 10652.
[3] Yoder G.D., Diwakar P.S., Hahn D.W., Appl. Opt., 2005, 44(20), 4211.
[4] Goulay F., Schrader P.E., Nemes L., Dansson M.A., Michelsen H.A., Proc. Comb. Inst., 2009, 30,
963.
[5] Park J.K., Lee S.Y., Santoro R., Int. J. Automative Tech., 2002, 3(3), 95.
[6] De Iuliis S., Migliorini F., Cignoli F., Zizak G., Appl. Phys. B, 2006, 83, 397.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
! "
# $ % &'($ )
*
) + " ,- +", . ) +/ 0
% &'(* )
1
+ "! 2 ! 3 4 5 6
"! "3*7 1 + 4
! " ! ! #$ % !%
! !$! % #% #&
# # % # % $ !&
#$ $ % $# % #$ !
! # ! # ! #&
' () "#$ $$$ * ''+ ( ! % !
# ! ! $ # #% #$
$$ !% $ & $" #
( $"( ( # #$ ! (
$ ! (& # ! !!
! $"( ( ,- !! # . #/0 * ## $+
. / 1 $2%& # ( # ( $
& ! ( ( ) $ 3& ! 4
( 2 ! 1 0 ! 4 1&0 ! 4 - & #
( ! "!& !# $ $!# (
! ( ! # !#& "#$ $$$ (
% 5#$%"# # 6789 $ (
!# # !"2 '' & 1&0 ! 4
#$ 2#! $$ - : 2# $
6 ;< #$ 2# !# !% , :&
$! # 1&0 ! 4 # ! $% &
' $ !% $ #
$ $ ( $ $ $ %
$ $% ( &
! " #
$ %& '% ()'* $ + ( ,+-,*
./ $ -& , 01 02 -& '%
!
" # " #
$ "" %&'( # " $ $
" ! !
" ! " # %)' !
$ * * # "
# # + $ $"
# ! " " $ *
" #!
( " #
" # " ! $"
+ " !
" # # " ! $
," " ! -./01
# $ $ !
" " 0 ! #
, 234 !
" " !! !
$ " "!
# 5 ( "
6 " " $ &47
" ! # ! " "! 6 "
! # " " " $ ! $
6$ " ! " $ "" 87
" # # $
" +
& 9 ( : ; ; < 9 9 $ +
1 = .
* =0""5!9 >>>?>48 )77@
) ; ;: ! 9(: ;A =
-B ( ! $
$" = $ , %& )774
! " #
$ % &' % ! ( ) $
) * * +, )-
$ %. /% 01/2 $ ( 0 &(3&2
4' $ 3. & 56 57 3. /%
! !!! "#$% !& ! ! & '
! '! ( ) ! ! ' "**%
& + ' ! ,'! '! !
,'! '! ' ,, ! ()
*! ' '&'' !,' ' !+- '
,' . !! ! ' / ' ! 0' ! #$'!
!!!' ,,"1234 333 % !'!'
! ' &! !! ! ,!'!' "5' $
, 6% ' !' ! !' 1234733
'"8' !*!'! ,!'+!% !,!' '!
9: ,', .! !!',
0' ! ,! '! ** ";<**% '! '=+ &
> ?@$ ' "A + B!' 0<% ! '
!! !112 :23 ! !,!! , ','&
! 33 C ! 33 C
;<**'! ',' ' ' ! &' ;$A
! '! !' ! ;<** ' ,' ! ! !'
! #$! D '! ! 0' ! #$ ! ! !!!
!!&',! !,'''"8$;%&',! !!' !!'
! , ' ! , ! $ '! ! E0 !'
'!#$ !' + !+ '! ! ';<** !
F !!;<**! '! !! ' ,' + !
! ,' ! ! !!' ''! #$ ! ! ' &! !
E0 &! '! ! = ! ! !' &! ,++ '
! + ''& "8$;% ! ,''' ! $ ! ;<**
#$ ,' ' + ' '! ! '!;$A !
!' &! '! '! ! '! 8$;&!'!
F !! '! !' ! '! & ! ;<**
'! ! ,' '+,'! !' ' !!,'!
$ <5 < 5 G
!! !!!! '!!'! '! G
$,,#,! 971 :3D"336%
A0 ; ; F AA <
! G' ?'!!' ''!
=! G$,, 8+ A !" DDD4D21"339%
! " #
Measurements of Soot Volume Fraction under Conditions
Relevant to Engine Exhaust Using Four-Colour LII and
Different Laser Energies
Fengshan Liu1, Xu He2, Hongmei Li2, Fushui Liu2, Gregory J. Smallwood1
1
National Research Council, Building M-9, 1200 Montreal Road, Ottawa, ON., Canada
2
Beijing Institute of Technology, Beijing 100081, China
e-mail: Fengshan.Liu@nrc-cnrc.gc.ca
Available experimental evidence shows that the apparent soot volume
fraction (SVF) determined from two-colour LII (at 450 and 780 nm) for soot initially at
near ambient room temperature displays a laser fluence dependence: it first
increases with increasing the laser fluence until it reaches a maximum, after which it
decreases with further increasing in laser fluence, due to sublimation. It is suggested
that the low-fluence SVF anomaly is attributed to changes in soot emissivity in the
350 to 500 nm spectral range as a result of evaporation of condensed volatile organic
compounds from soot particle surfaces due to laser heating. However, there is
currently a lack of direct evidence to confirm the cause of the low fluence anomaly
and a lack of detection strategy on how to avoid the low fluence SVF anomaly. The
current approach using the two-colour LII technique is to operate the laser fluence
around 2.1 mJ/mm2 with a 1064 nm laser, see Fig. 1.
In an attempt to address these two questions experimental measurements of
SVF were conducted in a soot aerosol at near room temperatures at different laser
fluence. A Nd:YAG laser of 6 ns FWHM operated at 532 nm was used to excite the
soot particles. The resultant LII signals were detected at four spectral bands (with
spectral widths varying between 12 to 40 nm) centered at 400, 631, 780, and 840
nm. The experimental results from this work provide useful insights into the cause of
the low fluence SVF anomaly and an effective strategy to avoid this anomaly through
detection of LII signals at longer wavelengths.
ϭϮϬ ϱϬϬϬ
$FFHSWDEOH5DQJH
0DVV PJP
)OXHQFH P-PP
ϭϭϬ 7HPSHUDWXUH . ϰϴϬϬ
ϭϬϬ ϰϲϬϬ
'ƌĂǀŝŵĞƚƌŝĐ сϵϱŵŐͬŵϯ
^ŽŽƚŽŶĐĞŶƚƌĂƚŝŽŶ ŵŐͬŵϯ
ϵϬ ϰϰϬϬ
dĞŵƉĞƌĂƚƵƌĞ<
ϴϬ ϰϮϬϬ
ϳϬ 2SWLPXP5DQJH ϰϬϬϬ
0DVV
PJP
ϲϬ )OXHQFH ϯϴϬϬ
P-PP
7HPSHUDWXUH
ϱϬ . ϯϲϬϬ
ϰϬ >Žǁ&ůƵĞŶĐĞ͗
DŽĚĞƌĂƚĞ&ůƵĞŶĐĞ
,ŝŐŚ&ůƵĞŶĐĞ͗ ϯϰϬϬ
^s&ŶŽŵĂůLJ ^ƵďůŝŵĂƚŝŽŶ
ϯϬ ϯϮϬϬ
ϮϬ ϯϬϬϬ
ϭ͘ϬϬ ϭ͘Ϯϱ ϭ͘ϱϬ ϭ͘ϳϱ Ϯ͘ϬϬ Ϯ͘Ϯϱ Ϯ͘ϱϬ Ϯ͘ϳϱ ϯ͘ϬϬ
ϭϬϲϰͲŶŵ>ĂƐĞƌ&ůƵĞŶĐĞŵ:ͬŵŵϮ
Fig. 1 Variation of soot concentration and temperature with laser fluence.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Effect of Laser Pulse Duration on Laser-Induced
Incandescence of Soot
Fengshan Liu, Gregory J. Smallwood
National Research Council, Building M-9, 1200 Montreal Road, Ottawa, ON., Canada, e-mail:
Fengshan.Liu@nrc-cnrc.gc.ca; Greg.Smallwood@nrc-cnrc.gc.ca
The phenomenon of laser-induced incandescence (LII) has been utilized to
develop versatile diagnostic techniques for measurements of combustion-generated
soot concentration and primary particle size in many combustion applications. Many
aspects in the implementation of LII have been investigated, such as the excitation
laser wavelength and spatial profile, signal detection timing (prompt or delayed) and
temporal width, and signal detection wavelengths. One potentially important
parameter in LII practice that has received little attention in the LII community is the
laser pulse duration. This is perhaps due to the fact that almost all the LII
experiments conducted in the last two decades employed Q-switched, flashlamp
pumped Nd:YAG lasers as the light source, which all provide a similar pulse duration
in a narrow range between 5 and 10 ns FWHM. Few LII experiments were conducted
using a much long laser pulse (microsecond) or much a shorter one (picoseconds).
Picosecond laser pulses were shown theoretically to provide advantages over
nanosecond ones in the determination of primary particle size distribution using low-
fluence LII in flames at elevated pressures.
In this study the effect of laser pulse duration on the temperature histories of
primary soot particle of different sizes was numerically investigated in both low- and
high-fluence regimes under conditions of a typical atmospheric pressure laminar
diffusion flame. The laser pulse durations considered in this study vary from 100 ps
to 6 ns FWHM. Such laser pulse durations correspond to those of a typical
Ti:Sapphire pulsed laser operated at 780 nm. Such laser is capable of generating
laser pulses with duration from picoseconds to nanoseconds, depending on if the
regenerative amplifier is seeded by the femtosecond oscillator.
Numerical results show that at a fixed laser fluence in the low-fluence regime
with decreasing the laser pulse duration the effect of heat conduction on the peak
soot temperature is suppressed. Under the conditions considered here heat
conduction lowers the peak soot temperature of a 30 nm soot particle by about 70 K
when the laser pulse is 6 ns FWHM. As the laser pulse duration decreases, the
differences in the peak temperatures reached by particles of different sizes become
smaller, due to the effective separation of the volumetric heating process and the
surface dependent heat conduction cooling. In high-fluence regime, the peak soot
temperature increases more significantly with decreasing the laser pulse duration.
There seem no advantages using a shorter laser pulse than the commonly employed
6 ns FWHM one in the high-fluence regime.
In summary, a shorter laser pulse offers advantages over the typical 5 to 10
ns duration one in some low-fluence LII applications where it is desirable to suppress
the effect of heat conduction, such as particle size determination in high pressures
and evaluation of E(m). However, it does not seem to offer advantages in high-
fluence regime.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Relationship between LII Signal and Soot Volume Fraction –
Effect of Primary Particle Diameter Polydispersity
Fengshan Liu, Gregory J. Smallwood
National Research Council, Building M-9, 1200 Montreal Road, Ottawa, ON., Canada, e-mail:
Fengshan.Liu@nrc-cnrc.gc.ca; Greg.Smallwood@nrc-cnrc.gc.ca
LII has been increasingly applied to measure the concentration and size of
nano-particles, such as combustion generated soot in flames, in-cylinder, and engine
exhaust, black carbon in the atmosphere, and other synthesized nano-particles. The
basis of LII concentration measurement is the assumption that the LII signal is
proportional to the particle volume fraction, while the principle of LII particle sizing is
the particle size dependence of heat conduction cooling after the laser pulse. In this
work, the discussion is focused on measurement of volume fraction of combustion
generated soot using LII.
Two approaches have been developed to conduct quantitative LII
measurements of the soot concentration. The conventional method detects the LII
signal at one wavelength in the visible and seeks a calibration constant using a
known source of particle concentration, i.e., SLII = C×fv. The second one is the more
recently developed two-color LII or auto-compensating LII. In this method, absolute
LII intensities are detected at two wavelengths in the visible spectrum to infer the
soot temperature based on the pyrometry principle. This method does not require a
known particle source to arrive at a calibration constant. However, it requires the
knowledge of both relative and absolute values of E(m) at the two detection
wavelengths, which represents the main uncertainty of the two-color LII method. The
calibration constant C in the conventional LII method in general is not constant under
conditions other than those of the calibration. Since soot temperature is determined
in the two-color LII method, the two-color LII method can be viewed as a special
version of the conventional LII method in which the calibration constant is obtain in
situ.
It is shown numerically that for a polydisperse primary soot particles the soot
temperature derived in two-color LII is biased towards the temperature of those larger
and hotter particles. As smaller particles cool faster than larger ones, smaller
particles gradually ‘disappear’, leading to a decrease soot volume fraction
determined by the two-color LII. Based on numerically results, the relationship
between LII signal and soot volume fraction can be summarized as:
(1) It is linear in the two-color LII during and shortly after the laser pulse in the
low-fluence regime
(2) It is linear in the two-color LII only briefly around the peak of the laser
pulse in the high-fluence regime
(3) It is in general non-linear in the conventional LII method
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Novel soot volume fraction measurement through ratio-
pyrometry and absolute light calibration
Bin Ma and Marshall B. Long
Yale University, 15 Prospect St, New Haven, CT, USA
bin.ma@yale.edu
Novel ratio pyrometry and absolute light calibration methods have been
developed to obtain soot temperature and volume fraction in axisymmetric flames. A
consumer digital single lens reflex camera has been fully characterized and utilized
as a pyrometer. The incandescence from soot was imaged at the three wavelengths
of the camera’s color filter array (CFA). Temperatures were calculated by two-color
ratio pyrometry using a lookup table approach. While temperatures can be extracted
from color ratios, soot volume fraction requires an absolute light calibration of the
detector. The absolute light intensity calibration was provided by a flame-heated S-
type thermocouple. The spectral emissivity of S-type thermocouple wires (Pt and Pt-
10% Rh) was measured in the visible range. The measured spectral emissivity,
temperature, and diameter of the heated thermocouple wires allow them to serve as
a light source with spectral radiance that can be calculated by Planck’s law. Soot
volume fraction measurements were carried out on four different flames with varying
levels of soot loading. The results have been compared with previous LII results and
excellent agreement has been achieved.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Modeling laser-induced incandescence of soot integrating
spatial and temporal dependences of parameters involved in
energy and mass balances
Mohammed Mobtil1,2, Romain Lemaire1,2
1
Université Lille Nord de France, F-59000 Lille, France
2
EMDouai, EI, F-59500 Douai, France
romain.lemaire@mines-douai.fr
Laser-Induced Incandescence (LII) has become a widespread used technique
for soot volume fraction and primary particle size determination in flames and
exhaust gases. The correct interpretation of experimentally measured LII signals
implies a detailed understanding of the physical mechanisms that control the LII
phenomenon. It also needs, amongst other things, to thoroughly take into account
the experimental parameters involved in the excitation process (especially the spatial
and temporal profiles of the laser energy). Different models have been proposed
recently in the literature to predict the temporal behavior of LII signals1. Nevertheless,
except some recent works from Bladh et al.2, only few works took into account the
temporal and spatial characteristics of laser excitation sources presenting 2D
inhomogeneous distributions.
In the present work, the experimentally monitored characteristics of an
unfocused near-Gaussian laser beam have been considered as input data in our
model (see the figure below). The spatial discretization of the mass- and energy-
balance equations (based on the absorption (Qabs), radiation (Qrad), sublimation (Qsub)
and conduction (Qcond) terms) has been carried out using the finite element method
while the temporal discretization of these equations has been achieved following the
Crank-Nicolson scheme. By this way, we obtain a matrix shape equations system in
which the particles temperature and diameter (Tp and Dp, respectively) are the two
unknowns. Such a 3D problem being non-linear, we solved it by using the Newton
iterative method to obtain the evolution of Tp and Dp as a function of the time and of
the space in each mesh of the excitation volume. The temperature dependence of
parameters such as physical properties of soot has also been taken into account in
the different terms used to obtain the mass- and energy -balance equations which
allows determining the evolution of these properties for each time and spatial
position.
A mesh sensitivity study has
Spatial distribution of the laser energy
been carried out and the temporal
evolution of Qabs, Qrad, Qsub, Qcond, Tp
and Dp as a function of the space will
be presented in this work which is
still in progress. The spatial LII time
decays that have been calculated by
entering Tp and Dp into the Planck
function integrated over a given
range of wavelengths will be
presented and potentially confronted
with experimental data obtained
using such a laser profile.
1
H.A. Michelsen, F. Liu, B.F. Kock, H. Bladh, A. Boiarciuc, M. Charwath, T. Dreier, R. Hadef, M. Hofmann, J.
Reimann, S. Will, P.-E. Bengtsson, H. Bockhorn, F. Foucher, K.-P. Geigle, C. Mounaïm-Rousselle, C. Schulz, R.
Stirn, B. Tribalet, R. Suntz - Modeling laser-induced incandescence of soot: a summary and comparison of LII
models - Applied Physics B, 87, 503-521 (2007)
2
H. Bladh, J. Johnsson, P.-E. Bengtsson - On the dependence of the laser-induced incandescence (LII) signal on
soot volume fraction for variations in particle size - Applied Physics B, 90, 109-125 (2008)
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
EVOLUTION OF THE LASER INDUCED INCANDESCENCE
SIGNALS OF SOOT PARTICULES MEASURED IN LOW-
PRESSURE METHANE FLAMES
T. Mouton*, P. Desgroux, X. Mercier
Physicochimie des Processus de Combustion et de l’Atmosphère (PC2A), UMR CNRS 8522,
Université Lille 1 Sciences et Technologies. 59655 Villeneuve d’Ascq Cedex, France
thomas1.mouton@ed.univ-lille1.fr
The understanding of soot formation mechanisms in flames, and more
specifically the nucleation process, is still under debate. To deal with this crucial step,
low pressure laminar flames are particularly well suited because of the large reaction
zone, offering the possibility of examining the early soot formation zone. The number
of soot particles is however much lower than at atmospheric pressure, requiring the
use of sensitive techniques such as Laser Induced Incandescence (LII).
In this work, we have used the LII technique to probe soot particles formed in
various low-pressure premixed methane/oxygen/nitrogen flames stabilised for
different equivalent ratio (ĭ = 2.32, 2.05, 1.95) and pressure (P = 18.66 kPa (140
torr) and 26.66 kPa (200 torr)). Heating of the particles has been achieved by using
the 1064 nm excitation wavelength of a YAG laser, the energy profile of which has
been shaped as top hat. Temporal LII signals were measured by a photomultiplier
whereas we complementary used an intensified CCD camera coupled to
spectrometer in order to record the associated emission spectra. Measurements
have been done for different heights above the burner (HAB), included the very
beginning (nucleation step) of the soot formation processes in the flames.
By this way, we observed significant and surprising differences, mainly
concerning the evolution of the temporal signal according the flame height, between
the lowest equivalent ratio (ĭ=1.95 and 2.05) and the reference flame (ĭ=2.32).
While this last flame is characterised by the increase of the LII decay with HAB,
corresponding to the increase of the particles size as expected, no such evolution is
observed for the two other ones. In these conditions, the temporal LII decays remain
constant for all the heights above the burner, therefore questioning about the nature
of the formed species. As a consequence, examination has been focused on those
flames including fluence curves, measurement of relative volume fraction profiles and
spectral analysis.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Influence of LII on Soot Optical Properties
in Reference Flames
K. Thomson1, K-.P. Geigle2, D. Snelling1, F.Liu1, M. Köhler2, G. Smallwood1
1
National Research Council, Ottawa, Canada, kevin.thomson@nrc-cnrc.gc.ca
2
German Aerospace Centre (DLR), Stuttgart, Germany
When analyzing LII emission signals, it is typically assumed that the optical
properties of soot are not affected by this rapid heating. However, from the literature
it is known that the laser irradiances typical of ‘plateau LII’ can lead to significant
modification of soot particles and even the formation of new particles [1,2]. When
more moderate laser fluences are used, morphological changes are not observable
via high resolution transmission electron microscopy; however, there is still evidence
that the heating permanently influences the optical properties of the soot [3,4].
Variation of soot optical properties during or immediately after laser heating would
have impacts on the interpretation of LII signal which should be accounted for in the
theory in order to accurately use the emission data.
To study the optical properties of laser heated soot, we have monitored the
extinction coefficient of soot aerosols within the standard Gülder and McKenna
burners as a function of time while simultaneously heating the aerosol with laser
pulses typical of LII. Extinction coefficient measurements were made at wavelengths
of 405, 488, 632, and 804 nm and for a range of 1064 nm pulsed laser fluences.
We present a rich database of normalized extinction measurements which
give clues into the complex consequences of rapid laser heating of soot aerosols.
Normalized extinction coefficients show an enhancement of the propensity of soot to
absorb light over the time interval of the laser heating. A partial relaxation of this
enhancement is evident on the soot cooling time frame suggesting that the
enhancement is in part due to a temperature based phenomena such as particle
expansion or temperature dependent optical properties. A sustained residual
enhancement is observed in the McKenna soot data, indicative of a permanent
change to the soot optical properties, possibly due to graphitization. The variation of
the normalized extinction coefficient in the McKenna burner diminishes with
decreasing probe wavelength. This relates to the presence of non-soot material
which absorbs light in the UV wavelengths, but is not heated by the 1064 nm laser.
For the higher soot concentrations of the Gulder burner, heat transfer to the gas
phase leads to a gas temperature change and expansion which decreases the
attenuation propensity of the medium. The soot is more efficiently heated than the
McKenna soot, with sublimation initiated at lower fluences and greater sublimation at
a given fluence. This suggests a higher refractive index absorption function, E(mȜ)
for the Gulder soot. Normalized extinction coefficient measurements at 405 nm in
the Gulder flame at very high fluences demonstrate that the materials vaporized from
soot reform into species which are capable of absorbing 405 nm radiation, thus
masking the sublimation effect on normalized extinction coefficient.
Both reversible and non-reversible changes to soot’s ability to attenuate light
have been demonstrated in McKenna and Gulder flame soot. These variations
should be further quantified and incorporated into LII emission interpretation theory.
[1] R. L. Vander Wal and M. Y. Choi, Carbon 37, 231-239 (1999).
[2] H. A. Michelsen, A. V. Tivanski, M. K. Gilles, L. H. van Poppel, M. A. Dansson, and P.
R. Buseck, Appl Optics 46, 959-977 (2007).
[3] R.L. Vander Wal, T.M. Ticich, and a.B. Stephens, Applied Physics B: Lasers and
Optics 67, 115-123 (1998).
[4] S. De Iuliis, F. Cignoli, S. Maffi, and G. Zizak, Appl. Phys. B 104, 321-330 (2011).
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Evaluation of particle sizes of iron-oxide nano-particles in a
low-pressure flame-synthesis reactor by simultaneous
application of TiRe-LII and PMS
B. Tribalet, A. Faccinetto, T. Dreier, C. Schulz
Institute for Combustion and Gasdynamics (IVG), and Center for Nanointegration (CENIDE),
University of Duisburg-Essen, 47048 Duisburg, Germany, thomas.dreier@uni-due.de
Laser-induced incandescence (LII) has become a common method for in-situ
analysis of particle size and visualization of particle volume fractions predominantly
for soot diagnostics in a wide range of applications. Besides lower signal strength
due to less strongly absorbing material and lower heat-up temperatures, one of the
main challenges when applying LII to non-carbon nanoparticles is the poor data base
of relevant particle thermophysical properties, e.g., heat conduction, accommodation
coefficients, vaporization enthalpy and high-temperature chemistry for describing
particle cooling due to convection, vaporization, and other effects. In the present
work the measured laser-induced emission signals from flame-synthesized iron oxide
(Fe2O3) nanoparticles were evaluated in terms of particle sizing by using a modified
version of the TiRe-LII model developed by Kock et al. [1].
Iron oxide nanoparticles were synthesized in a rich, premixed H2/O2/Ar low-
pressure (30 mbar) flat flame in a low-pressure flame reactor that was doped with
ppm-levels of Fe(CO)5 as precursor material. By moving the burner relative to the
fixed measurement location (determined by either the laser beam or a sampling
nozzle for the particle mass spectrometer (PMS), respectively) the particle residence
time in the reactor can be varied. The particles were heated by a frequency-doubled
Nd:YAG laser and time-resolved LII-signal traces were recorded perpendicular to the
beam axis by a two-color detection unit equipped with narrow band-pass filters with
center-wavelengths at 500 and 700 nm, respectively, in front of two high-speed
photomultipliers with integrated amplifiers. Additional to the time-resolved
measurements, LII signals were detected spectrally-resolved using a spectrometer
with an intensified CCD camera. The PMS with a molecular-beam sampling system
was attached to the burner chamber for simultaneous particle sizing.
To determine a phenomenological evaporation heat flux term in the energy
balance, temperature decay curves obtained by two-color pyrometry were fitted
through variation of a parameterized form of the particle evaporation term. With these
evaporation parameters, LII-signal traces were evaluated in terms of particle size.
The obtained size parameters were verified by corresponding PMS measurements
for the same flame conditions. In addition, it was possible to calculate the energy
accommodation coefficient αT of the present particle material at several experimental
conditions. Emission spectra taken right after laser heating did not vary significantly
in shape as a function of laser fluence.
The combination of TiRe-LII and online molecular beam particle sampling with
subsequent particle mass spectrometry in low-pressure flames is a promising
approach for fundamental research on the characteristics of LII of various
nanoparticle materials.
Preference: Poster Presentation
1. B. F. Kock, B. Tribalet, C. Schulz, and P. Roth, "Two-color time-resolved LII
applied to soot particle sizing in the cylinder of a Diesel engine," Combust. &
Flame 147, 79–92 (2006).
th
5 international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France
Comparison of different techniques for measurement of soot
and PM emission from Diesel engine
Richard Viskup
Institute for Design and Control of Mechatronical Systems
Johannes Kepler University
Altenberger Straße 69, 4040 Linz, Austria
Richard.Viskup@jku.at
Here we present the comparison studies between different techniques for
measurement of soot and particulate matter (PM) emissions from passenger car
Diesel engine. The compared techniques include a filter paper type smoke meter,
photo-acoustic spectrometer, opacimeter, differential mobility spectrometer and laser
induced incandescence. We mainly focus our study to static and dynamic transient
measurements tests from the location position closer to the actual combustion event
- downstream of the turbine, position characterised by the higher temperature and
higher pressure of the emission gas, than the standard measurement position, in the
tailpipe of the exhaust manifold. The main task is to reveal the most accurate and
sensitive method for fast soot and PM emission measurement for this particular
measurement position. The issue of accuracy reliability of measured emission
response and understanding of variances in measured soot emission due to different
applied techniques can help to minimise the soot and a particulate matter emissions
from diesel engines and to meet the future European emission standards.
5th international workshop on Laser-Induced Incandescence
May 9-11, 2012, Palais des Congrès, Le Touquet, France