13th Int Symp on Applications of Laser Techniques to Fluid Mechanics
Lisbon, Portugal, 26-29 June, 2006
Diagnostics of Reactive Zone in Premixed Flames
via Acetone-OH Simultaneous PLIF
Yuji Nakamura1, Satoshi Manome2, Hiroshi Yamashita3
1: Div. Mech. Space Eng., Hokkaido University, Sapporo, Japan, yuji-mg@eng.hokudai.ac.jp
2: Toho Gas Co. Ltd., Nagoya, Japan, manome@tohogas.co.jp
3: Dept. Mech. Sci. Eng., Nagoya University, Nagoya, Japan, yamashita@mech.nagoya-u.ac.jp
Abstract Imaging scheme based on Acetone-OH simultaneous PLIF (Planer Laser-Induced Fluorescence)
is applied for diagnostics of reactive zone over the premixed flames and the reactive zone imaging are
demonstrated for turbulent premixed flame conditions. In the present study, visualization of reactive zone
is brought by simultaneous measurement of unburned zone through acetone (seeded into the fuel flow) and
burned zone through OH (naturally generated by combustion), instead of looking for minor species formed
only at thin reaction zone (like CH, CHO, C2). One huge advantage of the present scheme is to use only
one laser and one detector combination to access multiple flame quantities (e.g. shape of flame surface,
flame thickness, relative flame intensity) in wide range of equivalence ratio of the mixture, mostly in lean
conditions. Applied excitation line is 283.22nm, which corresponds to OH Q1(7) (1,0) band absorption line
and is within the acetone absorption band to excite acetone and OH simultaneously. Instantaneous 2-D
reactive zone is imaged by single ICCD camera with band-pass filter which transmits both acetone and OH
fluorescence spectra. Demonstrations of reactive zone imaging show a distinct dark zone (non-signal zone)
sandwiched by acetone (unburned zone) and OH (burned zone) fluorescence regimes in all images beyond
the extinction conditions. In the previous study, we have found that the two boundaries of the observed
dark zone correspond to the fuel decomposition edge (unburned side) and ignition point (burned side),
respectively, and the dark zone is equal to the important radical pool (e.g. H2O2, HO2) to initiate the
subsequent chain branching reactions. Pocket structures (unburned pocket in hot region and burned pocket
in cold region) are properly captured in turbulent flame conditions. Local extinction phenomena are also
captured as “distorted” dark zone (no obvious sandwiched structure is found, but clear acetone front are
there). Capabilities on the present scheme for further flame study are addressed.
1. Introduction
Laser diagnostics on the chemically reactive flow (combusting flow) have been brought many
useful and important fundamentals to us. Since combustion takes place in quite narrow region (an
order of millimeter) and most of chemical reaction is so fast, fine spatial as well as time resolutions
are required for their diagnostics. Planar Laser-Induced Fluorescence (PLIF) scheme developed in
early 80’s made that possible and took us to the higher stage in flame study and still does (e.g.
Eckbreth 1988, Daily 1997). Recently, many laser applications have been proposed and utilized
on the flame studies; for example, Particle Image Velocimetry (PIV) for flow visualization (e.g.
Chigier 1991, Adrian 1991), PLIF for imaging of radical distributions (e.g. Eckbreth 1981, 1988,
Hanson 1986, Daily 1997), Raman/Rayleigh scattering for temperature measurement (e.g. Lapp
1974, Harvey 1981, Penner et al. 1985). Multiple quantities measurements have been a standard
now to obtain the detailed instantaneous flame status (Frank and Barlow 1998, Mansour et al. 1998,
Plessing et al. 2000, Watson et al. 2002, Tanahashi et al. 2005). For example, a visualization of
locally stretched flamelet structure and its time evolution in turbulent flame could enable us to
access the turbulent flame nature experimentally (ex. Chen and Bilger 2002) and direct comparisons
with corresponding numerical results give us further important mechanisms related to the local
flamelet characteristics (ex. Echekki and Chen 1996, Chen et al. 1999). Although the
developments have been contributed to gain our fundamental understanding of flames, but severe
-1-
13th Int Symp on Applications of Laser Techniques to Fluid Mechanics
Lisbon, Portugal, 26-29 June, 2006
cost problems always are followed. In general, one laser (with detector) system can derive only
one quantity. In the same manner, if user needs to measure two quantities, two laser systems are
required. On the other hand, if there is no limitation, we always prefer to measure many quantities
as much as possible simultaneously (but it is, of course, impossible). In this sense, our
requirement and system cost are in trade-off relation. It is, therefore, our hope to develop the
diagnostics scheme to visualize “multiple” quantities with much simpler but still effective way.
To meet such requirement, we have recently developed the scheme based on “acetone-OH
simultaneous PLIF concept” for premixed combustion diagnostics (Nakamura et al. 2006). A
basis of this concept is to image fuel-seeded acetone (as tracer of “unburned” zone) and
combustion-generated OH (as tracer of “burned” zone) simultaneously, instead of looking for minor
species generated only at thin reaction zone (e.g. CH, CHO, C2). Since their fluorescence zones
would not be overlapped as long as the proper combustion takes place, thin dark zone, i.e. nonfluorescence signal zone (call “reactive zone” in the present study), appears in the observed plane.
One huge advantage of this scheme is to use only one laser and one detector combination to access
multiple flame quantities (e.g. shape of flame surface, reactive zone thickness, relative flame
intensity) in wide range of equivalence ratio of the mixture (mostly in lean conditions). Moreover,
the fluorescence signals of both species (seeded acetone and combustion-generated OH) are not
very sensitive to the mixture condition as compared to CH (or C2, CHO), thus, it is suggested that
this scheme may properly work in extra-ordinary conditions, including locally-extinction flames.
In this paper, brief description for the presently-applied acetone-OH simultaneous PLIF scheme for
reactive zone imaging over premixed flames are made and demonstrations of reactive zone imaging
are performed turbulent premixed flames with/without local extinction.
2. Approach
2.1 Previous Acetone-OH PLIF Studies
Acetone (CH3COCH3) seeding into fuel flow for imaging of unburned regime has been often
employed for the test of combustion efficiency in the combustors (ex. Wolff et al. 1993). Since
acetone is decomposed when it is exposed into the high temperature field, no fluorescence from
acetone is observed in the hot region (including combustion zone). Not likely simple di-atomic
molecules (e.g. OH), chemical structure of acetone is rather complex, resulting that it has wide
range of absorption spectra (so as to fluorescence spectra). Oppositely, as is well-known, OH
radical exists only in hot region (not in unburned zone) and absorbs only specific lines.
Attempts have been made for acetone-PLIF on flame diagnostics studies with various excitation
lines: for example 266nm (e.g. Bryant et al. 2000, 2001) or 308nm (e.g. Lozano et al. 1992, Thurber
et al. 1998). Fluorescence spectra and its temperature dependency have been precisely
investigated in the past literatures (e.g. Lozano et al. 1992, Bryant et al. 2000). Bryant et al.
(2000) have performed acetone-PLIF and OH-PLIF separately for the acetone-seeded, jet diffusion
flames to study time-averaged flame structure in supersonic and subsonic reactive flow. For
instantaneous measurement purpose, Hanson’s group (Seitzman et al. 1994, Yip et al. 1994) have
investigated the mixing structure in reactive mixing layer flow through simultaneous acetone-OH
PLIF. Later, Tamura et al. (1998, 2000) applied the same concept to study incompleteness of the
combustion by looking up the unburned and burned regimes in the combustor. In their schemes,
fluorescence signals from acetone and OH have been separately captured by two ICCDs
simultaneously and superimposing two obtained images taken by each detector gives
“instantaneous“ structure of unburned and burned zone. Their technique is applicable for
instantaneous flame diagnostics, however, required system cost becomes much higher as compared
to a conventional PLIF set (one laser and one detector combination). In addition, complicated
-2-
13th Int Symp on Applications of Laser Techniques to Fluid Mechanics
Lisbon, Portugal, 26-29 June, 2006
system timing controller and high quality of image-superimposing technique are essential to
properly work. This requirement is quite severe since non-fluorescence zone observed between
acetone and OH is small enough (sub millimeter). Indeed, previous studies have been done to
look for the mixing structure over diffusion flames and little attempt have been made for premixed
flames. Only exception is Tamura et al.’s work (2000) but their interest is on the unburned status
over the combustor and not for the detail structure (like reactive zone) in the flame.
2.2 Acetone-OH PLIF for Premixed Combustion Diagnostics
Normalized LIF signal intensity [-]
-3-
Temperature [K]
Since the detail of presently-applied acetone-OH simultaneous concept is found in elsewhere
(Nakamura et al. 2006), only brief description is made here. Since acetone (CH3COCH3) seeded
in the fuel flow decomposes over 1000K, it does act as the tracer of unburned zone. Oppositely,
combustion-generated OH only exists under the high temperature condition, thus it does act as the
tracer of burned zone. Figure 1 shows 1-D flame structure of Bunsen-type, acetone-seeded,
laminar premixed flame (main fuel: methane, total equivalence ratio (φ): 0.85) (Nakamura et al.
2006). Total equivalence ratio defined
FLAME ZONE
as acetone-methane-air mixture is
UNBURNED ZONE
BURNED
ZONE
introduced throughout this study since
1800
acetone must be taken account as the
1
T
1500
combustible fuels.
Note that the
0.8
OH
acetone
1200
distributions of acetone, OH and
0.6
temperature are separately measured by
900
0.4
acetone-PLIF (excitation line: 266nm),
600
0.2
OH-PLIF (excitation line: 283nm) and
0
300
thermocouple, respectively. This figure
1 mm
transverse direction coodinate [mm]
clearly indicates that acetone signal is
mostly
disappeared
before
the
Fig.1 1-D flame structure of acetone-seeded methane-air
temperature increased point and OH is
premixed flame (total equivalence ratio, φ, is 0.85)
produced in the post-flame zone. More
(referred from Nakamura et al (2006))
importantly, acetone and OH signals are
not overlapped and the location of sudden
temperature rise (where the sharp heat release is expected) is included in the narrow regime where
neither acetone nor OH exists.
Considering the absorption character of acetone, it has wide band structure of its absorption
range (225nm-320nm at room temperature: note that the range does depend on field temperature
(Thurber et al. 1998)). One should be noted is that major OH excitation line (~280nm) is within
this range. Thus, technically speaking, only single line (which corresponds to OH absorption line)
is required to excite both acetone and OH simultaneously to emit “mixed” fluorescence signals
from the targeted premixed flame. Corresponding fluorescence image must include a dark zone
(i.e. non-signal zone) sandwiched by acetone and OH fluorescence regimes, where corresponds to
“flame (reactive) zone” (see Fig.1).
Previously we have proposed this visualization scheme, and the imaging and analyses of the
dark zone over various equivalence ratios (0.64<φ<1.15) in laminar, premixed flames have been
successfully done (Nakamura et al. 2006). Followings are the important results given by that work.
Typical 2-D PLIF images for laminar-premixed flames (see Fig.2) showed clear dark zone in the
fluorescence signals (bright zone) and its thickness varied depending on the mixture conditions.
According to the spectroscopy analyses of fluorescence signals, no major structure was found other
than acetone and OH under the imposed conditions, suggesting that the observed dark zone is made
by acetone and OH boundaries. The inverse of the thickness was found to have good correlation
with its burning velocity and the accuracy of the correlation was confirmed by the numerical
13th Int Symp on Applications of Laser Techniques to Fluid Mechanics
Lisbon, Portugal, 26-29 June, 2006
500
0
0
-3
[x10 ]
8.0
6.0
4.0
0.5
2.0
4
3
2
1
3.Experiment
-4-
0
-1
-2
-3
-4
1.0
0.5
0.5
0
0
2
1.5
1
1
0.5
0.5
0
0
4
3
2
1
0
-1
-2
-3
-4
3
0
0
XCH3 [-]
1
1.5
1.5
1
HRR [GW/m ]
Fluorescence signal [a.u.]
1000
0.5
1.5
[-]
Fluorescence signal [a.u.]
1500
0
[x10-3]
2.0
HO2, H2O2
1
1
0
X
1.5
2000
Temperature [K]
1.5
Fluorescence signal [a.u.]
0
0
2
0.5
Fluorescence signal [a.u.]
0.5
[-]
0.5
3
CH4
1.0
4
1
X
1.5
1
1.5
Fluorescence signal [a.u.]
[x10-3]
2.0
1.5
XHO2, H2O2 [-]
Fluorescence signal [a.u.]
predictions. Figure 3 showed the
comparisons
of
observed
fluorescence structure and predicted
flame structure including radical
profiles for φ=0.76.
It was
Visualized domain
indicated that the acetone decay
profile was well-matched to the that
FLAME
of methane and observed and
20mm
predicted OH profile was quite
Burner
similar each other. Prior to the
(CH4+acetone+Air) mixture
OH peak, intermediate radicals like
Fig.2 2-D acetone-OH simultaneous PLIF images (referred
HO2 and H2O2 were produced.
from Nakamura et al. 2006)
HO2 is known as the product of the
important reaction of H+O2 in
(d)
(b)
HO2
lower temperature regime and goes
to H2O2, then finally forms two OH
H2O2
radicals (this process is so called
chain branching reaction in lower
temperature).
In this sense,
(d)
(c)
HO2
disappearance of HO2 and H2O2
(thus starting point of OH
H2O2
production) means an ignition point.
After all, observed dark zone
corresponds to the zone from “fuel
(f)
(e)
decomposition front” to “ignition
point” and somewhat similar to the
preheated zone. In the present
study, we call this zone as “reactive
Normal coordinate to flame surface [mm]
Normal coordinate to flame surface [mm]
zone” hereafter.
There are several advantages on
Fig.3 Comparisons of observed fluorescence structure
the present scheme; one huge
(symbols) and 1-D theoretical flame structures (lines)
advantage is “simplicity” in the
(referred from Nakamura et al. 2006)
system.
As mentioned, this
system is based on “one laser and
one detector combination”, so that system cost is relatively inexpensive (same as conventional PLIF
set). Nonetheless, this scheme could bring multiple important flame quantities (ex. shape of flame
surface by looking at acetone or OH boundaries, reactive zone thickness by looking at the dark zone,
relative flame intensity by looking at the relative OH intensities). Other advantages are
“applicability” and “usefulness”. Recall that both species (seeded acetone and combustiongenerated OH) are not sensitive in the mixture condition as compared to CH (or C2, CHO), thus,
this scheme may properly work in combustion in wide range of mixture conditions, including nearextinction conditions. In addition, the local scalar gradient can be estimated properly through the
clear imaging of corresponding zone. Recent modeling studies (e.g. Chen and Bilger 2002, 2004)
have shown that the gradient of local scalar quantities has strong correlation to the local flamelet
structure. Combined with this concept, the present scheme may give further precise radical status
in the flame which is hard to access experimentally.
13th Int Symp on Applications of Laser Techniques to Fluid Mechanics
Lisbon, Portugal, 26-29 June, 2006
3.1 Experimental Apparatus
Figure 4 shows the experimental apparatus applied in the present study. Laser system consists
of the combination of Nd:YAG (GCR-230, Spectra Physics Inc., 10Hz of frequency with 5ns pulse)
and Dye laser (HD-300, Lumonics Inc.). Frequency doubled line of YAG laser (532nm) is turned
to 566.44nm by Dye laser, then converted to 283.22nm with BBO crystal. Selected absorption
line of OH is A2Σ+<− X2Π transition corresponding to (1,0) band of Q1(7) and this is also absorbed
by acetone, as mentioned previously. The wavelength-tuned laser beam is transformed to thin
laser sheet (its thickness is less than 1mm) by lens system, then exposed into the targeted flame.
Laser sheet is expanded by cylindrical lens and only center part is selected to use for better
uniformity of the laser profile. Laser power directly exposed into flames is approximately 5-10mJ.
Fluorescence signals from acetone-OH are detected by CCD camera with Image Intensifier
(C4272 & C4274, Hamamatsu Photonics Corp.). By employing relay lens system, magnified
image is properly introduced into the imaging plate. Imaging range is 36.5mmx29.0mm and
corresponding pixel number is 640x480, respectively. Minimum scaling factor is 0.06mm/pixel.
Image center is set 20 (or 60)mm and 80mm above the burner surface for side and top part imaging,
respectively (see Fig.5). Band-pass filter (BG-12, Schott Inc.) is mounted in front of camera
system to capture their fluorescence signals properly. In order to minimize the background noise
and obtain clear image of fluorescence signal, exposure time duration in Image Intensifier is set to
200ns, which is minimum in the present camera system. System timing control is led by pulse
generator (DG535, SRS Inc.).
Premixed flames (Bunsen type) provided by acetone-methane-air mixture are used in the present
study. Burner is placed vertically in open atmosphere and gases are uniformly ejected upward to
form premixed flame over the burner. Cold air and fuel gas flows (~300K) are introduced into the
mixing chamber through the gas supply system. Nozzle inner diameter is set to 27 mm and burner
rim thickness is 1mm. Turbulent flame is induced by perforated plate (Yamamoto et al. 2003)
mounted at the burner exit to generate the turbulence artificially. Small pilot flame is set around
the burner exit to avoid sudden blow-off. Fuel supply rate for pilot flame is much smaller than
that for the main flow (less than 1% of volumetric flow rate), thus, the presence of pilot flame has
little affect on the main flame behavior.
PC &
Monitor
YAG
Laser
Pulse
Generator
premixed flame
Laser sheet
Observed
domain
532nm
Dye
Laser
Laser sheet
ICCD camera
Relay Lens
Band-pass
Filter
Burner
& Flames
Observed
domain
20mm (or 60mm)
566.44nm
Spherical
Lens
Burner
Burner
BBO crystal
Perforated plate
283.22nm
Cylindrical Lens
CH4+Air+acetone
Harmonic
separator
(a) FLAME SIDE
Fig.4 Schematics of experimental apparatus
(b) FLAME TOP
Fig.5 Visualized location for PLIF
3.2 Experimental Conditions
Two series of total equivalence ratio, φ, of acetone-seeded methane-air premixed gas is
-5-
13th Int Symp on Applications of Laser Techniques to Fluid Mechanics
Lisbon, Portugal, 26-29 June, 2006
considered as experimental parameter (φ=0.54, 0.84) in the present study; one is near extinction
condition and the other is normal condition, respectively. Total volumetric flow rate of the
mixture is fixed throughout the present study and corresponding averaged flow speed is about
175cm/s. Imposed turbulent properties introduced by the perforated plate are measured by hotwire velocimetry (Yamamoto et al. 2003) under frozen flow and corresponding turbulent intensity
and (integral) scale are 0.129m/s and 3.5mm, respectively. According to the Peter’s diagram, the
turbulent flame considered in the present study is categorized in laminar flamelet regime (Peters
1999).
Acetone seeding is controlled by bubbling method, which is similar to Lozano et al. (1992).
Bubbling device is equipped in the gas supply system. Gas flow is separated into two routes; one
goes directly to the burner and the other goes to the burner through the bubbling device. Acetone
seeding rate can be well-controlled by changing the introduced volumetric flow rate into the device.
To minimize the acetone-seeding effects on the main combustion characteristics, imposed acetone
seeding rate remains to be constant at 5.7cm3/s throughout the present study irrespective of the
imposed mixture condition. For the mixture of φ=0.84, 8.0vol% of imposed methane corresponds
to the seeded acetone amount.
4.Results and Discussion
4.1 Diagnostics for Turbulent Flames
Figure 6 shows instantaneous simultaneous acetone-OH fluorescence images of turbulent premixed
flames with φ=0.84. [top] and [side] mean the observed locations (80mm and 20mm above the
burner surface, respectively). So as to the laminar flame cases, distinct dark zone is clearly
observed in all images along the flame surface. Well-known cusp formation (in (a)) and “pocket
structures” (in (b) and (c)) are properly captured. Pocket structures (ex. Chen et al. 1999) are more
frequently found around the flame tip (in [top] images), where the turbulent fluctuation is
pronounced. Two distinctive “pocket” structures are found; one is “burned (OH) pocket“ observed
in the cold regime (see (b1) and (b2)) and the other is “unburned (acetone) pocket“ formed in the
hot regime (see (c1)). Moreover, uncommon “dark spot” in the unburned area (in (a2) and (b1)) as
well as “OH spot” in the burned (hot) area (in (c2)) are also notified. They might be the result of
either stronger distortion by the turbulent flow or complex 3-D flame configuration appeared in the
(a1)
cusp
dark spot (b1)
burned
pocket
[side]
OH
[side]
(c1)
[side]
(a2)
dark spot
unburned
pocket
(b2)
cusp
burned
pocket
[top]
[top]
(c2)
OH spot
[top]
Fig.6 Selected 2-D PLIF images for turbulent premixed flames. Imposed mixture condition is φ=0.84.
[top] and [side] mean the imaging location as described in Fig.5.
-6-
13th Int Symp on Applications of Laser Techniques to Fluid Mechanics
Lisbon, Portugal, 26-29 June, 2006
observable plane (projecting flame goes across to the imaged plane normally) (ex. Chen et al. 1999,
Nada et al. 2004). Once the unburned pocket is formed, separated unburned gas is consumed as
the combustion progresses and finally locally-inhomogeneous burned spot would be expected there.
We have captured such instance in (c2): bright OH spot indicates inhomogeneous burned spot and
its brightness might be related to the lifetime of OH. In the same manner, the burned pocket
appeared in the unburned zone would finally become dark spot as seen in (a2) or (b1); the local
temperature at dark spot would be high enough to decompose acetone but low enough to generate
OH. Overall, it is ensured that this scheme has capability to capture the inhomogeneous
temperature field and complex flame structure.
4.2 Access to Locally-Extinction Phenomena
To demonstrate the locally-extinction phenomena by this scheme, imaging demonstrations with
extremely lean condition (φ=0.54) are performed for the same turbulent condition. Figure 7 show
the typical instantaneous 2-D PLIF images under the corresponding condition. Note that height of
the 2-D image center is 60mm above the burner surface and one side of the fluctuated flames are
magnified. 1-D distributions (height direction: y) of curvature defined by acetone boundary and
maximum OH fluorescence intensity in the same height are also shown. Note that a positive sign
of the curvature means that flame is convex toward the unburned zone. In all 2-D PLIF images,
dark zone is clearly observed locally, indicating that the proper combustion might be taken place
there. Observed thickness of the dark zone is summarized in Figure 8 in probability form. A
definition of the dark zone is the distance from the location at 50% of (initial) acetone fluorescence
to the location at 50% of maximum OH fluorescence. Statistically averaged thickness is 2.21mm
local
extinction
burned
unburned
5mm
Fig.7 2-D instantaneous acetone-OH PLIF images for near extinction condition (φ=0.54). 1-D
distributions in y-direction of local curvature defined by acetone boundary (open square) and
maximum OH fluorescence intensity along x-direction (closed circle) are also shown. Arrows in
the figure show the location of local extinction.
-7-
13th Int Symp on Applications of Laser Techniques to Fluid Mechanics
Lisbon, Portugal, 26-29 June, 2006
OH fluorescence intensity [a.u.]
and fluctuation range is about +- 0.6 mm.
0.2
Fluctuation could be due to the difference of local
burning velocity or 3-D flame structure effect as
mentioned previously.
60mm
More importantly, OH fluorescence zone is
0.1
locally disappeared and no more “sandwiched”
structure is sustained. Since acetone decomposes
at lower than flame temperature (~2000K), acetone
signal disappears when acetone attaches to either
0.0
flame front or “non-combustible” hot gas region
1.0
2.0
3.0
4.0
5.0
6.0
7.0
(e.g. burned gas). In this sense, boundary of the
observed dark zone thickness [mm]
acetone signal does not always act as the flame
(reaction layer) surface, more likely it shows the
Fig.8 PDF of observed dark zone thickness,
“cold surface”, i.e. isotherm of the decomposed
defined as the distance from location of acetone
50% to the location of OH 50% (from its
temperature of acetone (<1000K).
On the
maximum)
contrary, OH could exist either active reaction
zone or high temperature zone (>1600K), therefore,
a disappearance of OH signal in the image implies
1.4
that no significant radical reactions are notified.
Thus it is acceptable to define the “non1.2
sandwiched” structure reflected to the local
1
extinction point. It is worthwhile to note that the
local extinction is pronounced at the peak of
0.8
negative the curvature (see in Fig.7). Correlation
0.6
between the curvature calculated by cold surface
(acetone fluorescence boundary) and maximum
0.4
OH fluorescence intensity is summarized in Figure
0.2
9. Thevenin (2005) revealed that the flame
thickness has positive correlation to the curvature,
0
whereas Chen and Bilger (2002) showed that the
-0.6
-0.4
-0.2
0
0.2
0.4
gradient of reaction progress variable (i.e. inverse
of the flame thickness) has usually negative
local curvature of cold surface [1/mm]
correlation to OH mole fraction but become
Fig.9 Correlation of the local curvature and
slightly positive when the condition closes to the
OH fluorescence intensity in various observed
extinction. Present trend (in Fig.9) seems to follow
flamelets
their observations. Observed extinction behavior
might be attributed by local flame stretch, however,
the reason is remained as unknown unless local flow structure is revealed. Further study is needed
to understand the observed trend here.
One should be noted here is that the local, instantaneous curvature can be always determined by
cold surface (acetone fluorescence boundary) even at the extinction point. This advantage does
not have any other multi-line laser diagnostics scheme (e.g. CH-OH PLIF (Tanahashi et al. 2005)
since neither OH nor CH are observed at the extinction point). Any information at extinction
status could lead better understandings of the extinction mechanism and useful for the model
validation purpose. Current proposed scheme can bring us such important flame status without
any additional costly hardware.
Concluding Remarks
-8-
13th Int Symp on Applications of Laser Techniques to Fluid Mechanics
Lisbon, Portugal, 26-29 June, 2006
We have applied simultaneous acetone-OH PLIF scheme for diagnostics of reactive zone in
premixed flames and various imaging patterns have been observed through this scheme. Using
one laser and one detector combination, multiple quantities (such as flame shape, flame thickness,
combustion intensities etc.) have been given by this scheme. It is revealed that complex turbulent
flame characters including uncommon “pocket” structures are properly captured by this scheme.
Local extinction phenomena would be also properly captured and curvature-extinction correlation
defined by acetone (cold) front can be derived by this scheme. Usefulness of out proposed
imaging concept through simultaneous acetone-OH PLIF to look for flame insight is addressed.
Acknowledgments
This work is partially supported by Tanigawa-Netsugijyutsu Foundation and Yoshida Foundation
for Science and Technology. Technical advises and suggestions provided by Professor T. Niimi
(Nagoya Univ.) and Dr. R. Bryant (NIST) are quite helpful. Supports in numerical part by Mr. N.
Hayashi (Nagoya Univ.) are greatly appreciated. YN express sincere thanks for their contributions.
References
Adrian RJ (1991) Particle-imaging techniques for experimental fluid mechanics. Ann Rev Fluid Mech 23:261-304
Bryant RA, Donbar JM, Driscoll JF (2000) Acetone laser induced fluorescence for low pressure/low temperature flow
visualization. Exp Fluids 28(5):471-476.
Bryant RA, Driscoll JF (2001) Structure of supersonic flames imaged using OH/Acetone planar laser-induced
fluorescence. AIAA J 39(9):1735-1741.
Chen JH, Echekki T, Kollmann W (1999) The mechanics of two-dimensional pocket formation in lean premixed
methane-air flames with implications to turbulent combustion. Combust Flame 116:15-48
Chen YC, Bilger RW (2002) Experimental investigation of three-dimensional flame front structure in premixed
turbulent combustion –I: Hydrocarbon/air Bunsen flames. Combust Flame 131:400-435
Chen YC, Bilger RW (2004) Experimental investigation of three-dimensional flame-front structure in premixed
turbulent combustion – II. Lean hydrogen/air Bunsen flames. Combust Flame 138:155-174
Chigier N (ed.) (1991) Combustion measurements. NY: Hemisphere Publishing Corp.
Daily JW (1997) Laser induced fluorescence spectroscopy in flames. Prog Energy Combust Sci 23: 133-199
Echekki T, Chen JH (1996) Unsteady strain rate and curvature effects in turbulent premixed methane-air flames.
Combust Flame 106:184-202
Eckbreth A (1981) Recent advances in laser diagnostics for temperature and species concentration in combustion. Proc
Combust Inst 18:1471-1488
Eckbreth A C (1988) Laser diagnostics for dombustion temperature and species. Cambridge: Abacus Press
Frank JH, Barlow RS (1998) Simultaneous Rayleigh, Raman, and LIF measurements in turbulent premixed methane-air
flames. Proc Combust Inst 27:759-765
Hanson RK (1986) Combustion diagnostics: Planar imaging techniques. Proc Combust Inst 21:1677-1691
Harvey AB (ed.) (1981) Chemical applications of nonlinear raman spectroscopy. NY:Academic Press
Lapp M (1974) Flame temperature from vibrational Raman scattering, in laser Raman gas diagnostics. Lapp M and
Penny CM (eds), NY:Plenum Press
Lozano A, Yip B, Hanson RK (1992) Acetone: a tracer for concentration measurement in gaseous flows by planar laserinduced fluorescence. Exp Fluids 13:369-376
Mansour MS, Peters N, Chen YC (1998) Investigation of scalar mixing in the thin reaction zones regime using a
-9-
13th Int Symp on Applications of Laser Techniques to Fluid Mechanics
Lisbon, Portugal, 26-29 June, 2006
simultaneous CH-LIF/Rayleigh laser technique. Proc Combust Inst 27:767-773
Nakamura Y, Manome S, Hayashi N, Yamashita H (2006) Simultaneous acetone-OH PLIF concept for combustion
diagnostics on premixed flames. Exp Fluids, submitted
Penner SS, Wang CP, Bahadori MY (1985) Laser diagnostics applied to combustion system. Proc Combust Inst
20:1149-1176
Peters N (1999) The turbulent burning velocity for large-scale and small-scale turbulence. J Fluid Mech 384:107-132
Plessing T, Kortschik C, Peters N, Mansour MS, Cheng RK (2000) Measurements of the turbulent burning velocity and
the structure of premixed flames on a low-swirl burner. Proc Combust Inst 28:359-366
Seitzman JM, Miller MF, Island TC, Hansom RK (1994) Double-pulse imaging using simultaneous OH/Acetone PLIF
for studying the evolution of high-speed, reaction mixing layer. Proc Combust Inst 25(2):1743-1750
Tamura M, Sakurai T, Tai H (1998) Simultaneous laser-induced fluorescence imaging of unburnt and reacting areas in
combustion fields using a KrF Excimer laser. Opt Review 5(2):119-123
Tamura M, Sakurai T, Tai H (2000) Visualization of crevice flow in an engine using laser-induced fluorescence. Opt
Review 7(2):170-176
Tanahashi M, Murakami S, Choi GM, Fukuchi Y, Miyauchi T (2005) Simultaneous CH-OH PLIF and stereoscopic PIV
measurements of turbulent premixed flames. Proc Combust Inst 30:1665-1672
Thevenin D (2005) Three-dimensional direct numerical simulations and structure of expanding turbulent methane
flames. Proc Combust Inst 30:629-637
Thurber MC, Grisch F, Kirby BJ, Votsmeier M, Hanson RK (1998) Measurements and modeling of acetone laserinduced fluorescence with implications for temperature-imaging diagnostics. Appl Opt 37:4963-4978
Yamamoto N, Nakamura Y, Zhao D, Yamashita H (2003) Measurement of flame structure of turbulent premixed flame
by OH-PLIF. Fluid & Heat Eng Res 38:19-28 (in Japanese)
Yip B, Miller MF, Lozano A, Hanson RK (1994) A combined OH/acetone planar laser-induced fluorescence imaging
technique for visualizing combusting flows. Exp Fluids 17:330-336
Watson KA, Lyons KM, Carter CD, Donbar JM (2002) Simultaneous two-shot CH planar laser-induced fluorescence
and particle imaging velocimerty measurements in lifted CH4/Air diffusion flames. Proc Combust Inst 29:1905-1912
Wolff D, Schluter H, Beushausen V, Anderesen P (1993) Quantitative-determination of fuel air mixture distributions in
an internal-combustion engine using PLIF of acetone. An Int J Phys Chem (Berichte der Bunsen-Gesellschaft)
97(12):1738-1741
- 10 -