4 ”S

4. ”Semi-intrusive”
ii t i ” techniques
t h i

I
I. Measurements
M t off soott - Laser
L
induced incandescence,, LII

II
II. Laser-induced
L i d d breakdown
b kd
spectroscopy, LIBS
I. Optical diagnostics of soot in flames

• Soot formation

• Soot measurements using

scattering/extinction

• Soot measurements using

g Laser-Induced
Incandescence, LII
 Soot in combustion processes

• For health and environment hazardous emissions

• Increased
I d radiation,
di ti more effective
ff ti heat
h t ttransfer.
f
Important in boilers, furnaces, camp fires,
candles etc. crucial component in fire spread
• Incomplete combustion, reduced engine efficiency
• Increased wear leading to reduced lifetimes of components
• Deposits in engines, turbines, furnaces
 Engine Emissions

From Kittleson (1998)

 S t formation
Soot f ti
GAS PHASE POLYCYCLIC AROMATIC HYDROCARBONS
Fuel + O2

...
C 3 H3

...
C 2 H3

...
CHO
....
Condensation
C d i P ti l Inception
Particle I ti
C O2 + H2 O

+ C 2 H2 PRIMARY SOOT PARTICLES

+ O 2 ,O H

Coagulation
Agglomeration

AGGREGATES

F. Mauss
 Soot characteristics

Primary soot particles

10 40nm
10-40nm
 Black body radiation

Black body radiation has a continuous spectrum described by

Planck’s radiation law
2hc2
I () 
1
hc / kT
(()) – emissivityy !
5
e 1
The radiation has a maximum that is shifted to shorter wavelengths
with increased temperature. This is described by Wien Wien’ss
displacement law:
maxT  2,898103 K  m

The total intensity of the radiation is increasing with temperature

according to Stefan-Boltzmann’s
Stefan Boltzmann’s law:
I  T 4

Henrik Bladh, Division of Combustion Physics

Planck emission from soot particles
 Soot extinction
Assuming d << 

I0/IT = exp(LKext)

N; number of soot particles

d; particle diameter
m; complex refractive index of soot particle

Conclusion:
Soot volume fraction measurable
Li off sight
Line i ht technique
t h i
No size information
 Soot scattering
Pol.
o ofo laser
ase light
g t

Soot scattering;
g;

Pol. of the
scattered light
 Soot scattering/extinction
g
Combined scattering /extinction
can be used for measurements of
N and d
 Laser-Induced Incandescence,, LII
Soot particles in a well-defined region are heated by means of
laser radiation
Heating of the soot particle leads to increased thermal
radiation
The increased radiation is detected

2000 K
4000
 Applications of laser‐induced incandescence

Flame studies In-cylinder engine Aero-engine exhaust

Joh
meas. meas.

hnsson et al. 2008 (pos

Desgro
Bla
adh et al. 20

oux et al. 20
006
ster)

008
Environmental Nanoparticle charact. Welding fumes
monitoring
Smalllwood et all. 2006

Gure
entsov et all. 2005

L
Lucas et al.. 2006
Courtesy: Henrik Bladh
 Increased black body radiation by LII
The red curve shows the black body radiation as a function of the wavelength
for T=1800 K, which is a typical flame temperature. The blue curve shows the
same radiation for T=4500 K, which is within the same temperature range as a
laser heated soot particle.

The left figure shows the real intensity difference for a large wavelength
interval, while the right figure shows the normalised signal strength for visible
light (normalization factor ~950).
13 10
x 10 x 10
2.5 2.5

2 2
4500K
4500K
Inttensity (W/m3)

Inttensity (W/m3)
1.5 1.5

1 1
1800K
1800K
0.5 0.5

0 0
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 400 450 500 550 600 650 700

Wavelength (nm) Wavelength (nm)

Henrik Bladh, Division of Combustion Physics

 Background and LII
Spectral
p behavior:

100
%
A spectral short-pass filter can
be used to suppress longer

Intensity (W/m )
wavelengths. The laser-heated
soot radiates more at shorter
wavelengths than the flame
does.

Wien’s displacement law!

0%
Wavelength (nm)

Henrik Bladh, Division of Combustion Physics

Soot temperature and LII signal
 LII - The laser wavelength

Which wavelength is suitable

for LII?

Soott absorbs
S b b wellll in
i a large
l
wavelength interval (UV -
visible – IR). There are,
however other species that
however,
can be excited with UV light
and cause disturbing
fluorescence
fluorescence. Therefore
wavelengths in the UV should
not be used. Most common is
the Nd:YAG laser at 1064 nm
(IR) and its second harmonic
at 532 nm.

Henrik Bladh, Division of Combustion Physics

 The radiation-particle
radiation particle interaction
A soot particle is exposed to
laser radiation.
Internal energy Planck radiation
The following
g p
processes are
intitiated
Absorption
 Heat transfer
Heat Transfer
 Vaporization
 Emission of Planck
radiation
V
Vaporisation
i ti
The energy balance equation:

 2 D3 E(m) 2ka (T  Tg )D2 Hv dM 1 dT

q(t )    D2   (D, )M b (T ,  )d  D3 s cs 0
   D  GMFP M v dt   6 dt

 
 PlanckRadiation
Absorption Heat Transfer Vaporization Internal Energy Change

Henrik Bladh, Division of Combustion Physics

LII as a function
f ti off laser
l fluence
fl

10

1
sity / a.u.
Intens

01
0.1

0.01

0.0 0.2 0.4 0.6 0.8 1.0

2
Fluence / J/cm
 LII dependence on the laser spatial profile
I
Region with no
The spatial
p profile of the laser
p vaporization
drastically affects the LII
signal.

The laser illuminates

particles with different
energies. This means that
contributions comes from
Region with
both parts (edges) where vaporization
there is no vaporization and
parts
t (middle)
( iddl ) where
h th
there
are vaporization.

I
Henrik Bladh, Division of Combustion Physics
 LII dependence on the laser spatial profile
LII experimentalists
p often refer to the LII fluence curve. A fluence curve
is the integrated LII signal plotted as a function of the laser fluence –
that is the laser pulse energy divided by the exposure area of the laser
beam. Below is shown modeled fluence curves for three different
spatial profiles.

I
Tophat

Gaussian sheet

Gaussian beam

Calibration of LII by scattering/extinction

Henrik Bladh, Division of Combustion Physics
 Time-resolved
Time resolved LII (TIRE-LII)
(TIRE LII)
Detection of the time-resolved LII signal may yield the primary
particle size. The decay time reflects the particle cooling and
small particles cool faster than large ones!
al intensity (a.u.)

10000

1000 Larger
LII signa

particles

100
Smaller
particles
0 100 200 300 400 500

Time (ns)
Henrik Bladh, Division of Combustion Physics
 Principle of 2
2-color
color LII
• Used to minimize uncertainties
in evaluated particle sizes from
LII signal decays as the rate of
absorbed laser energygy does not

nce [a.u.]
need to be known.
I2
• The temperature is calculated
using the relative irradiance at

Irradian
two wavelengths.
I1

λ1 λ2
W
Wavelength
l th [nm]
[ ]

Courtesy: Henrik Bladh

Modelling of time-resolved LII signals

Comparison
p of LII
models between
different research
groups
- Same experimental input
- Unconstrained models
- High fluence (0.7 J/cm2)

Michelsen et al
al., Modelling of
laser-induced incandescence of
soot: A summary and
comparison
p of LII models,,
submitted to Applied Physics B
 Experimental setup for two-color LII
Tophat spatial laser profile 
0.13 J/cm2
1064 nm

Tophat profile used for 
particle sizing as heating to 
diff
different temperatures 
tt t
increases the uncertainty!

From: Bladh, H., et al., Proc. Combust. Inst. 33, 641‐648 (2011)
 Modelling a soot particle in LII
Typical model Real-world examples
(microscopy)

Shape Shape
Sh
described • Diameter (D)
described
by: by: • Primary particle diameter (D)
• Number of primary particles
• Radius of gyration of aggregate
• Fractal parameters (”compactness”)
• etc.
t
• No particle looks the same as any
Jonathan Johnsson, Division of Combustion Physics, Lund University other!
 The effect of aggregation on LII signals

• Theory shows that LII signals should be affected by the level of soot
aggregation, i.e. decay rate/shape dependent on both primary
particle and aggregate size!
• First experimental evidence for this effect using LII on a cold soot
source: Soot ggenerator based on a quenched diffusion flame

Bladh et al. Appl. Phys. B, In press

 The effect of aggregation on LII signals

• First experimental evidence for this effect:

TEM data on p primaryy p

particle sizes: Similar LII signals:
g Different decayy rates
primary particle sizes!

Bladh et al. Appl. Phys. B, In press

 The effect of aggregation on LII signals

• Sampling using a 
pneumatic probe
• Large differences 
between TEM and LII 
sizes
– Deviation above 10 
mm HAB may be due 
to aggregation

– Uncertainties
Uncertainties in TEM 
in T M
sampling procedure 
and analysis

Aggregation model from 
Liu et al. 2006, Appl. Phys. B, 83, 383‐395 
Bladh, H., et al., Proc. Combust. Inst. 33, 641‐648 (2011)
 E(m) as function of height above burner (HAB)

• Evaluate difference between

gas temperature and
maximum soot temperature
for each height!
– Gas temperature from
rotational CARS
– Maximum temperature
from two-color
two color pyrometry

• Procedure as described by
Snelling et al. 2004,
Combust. Flame, 136,
180-190

• LII model to determine E(m)

Bladh, H., et al., Proc. Combust. Inst. 33, 641‐648 (2011)
 LII 2D measurements

Beam dump

Lens
Combination
Burner

M it
Monitor

CCD
Camera
Premixed flame

Diffusion flame
Engine
g Measurements
Laser entrance to the engine
Investigation
g though
g the piston
p
Investigation though the piston
LII signal Flame luminosity
1

0.9
3

0.8
2.5
07
0.7

2 0.6

0.5
15
1.5
0.4

1 0.3

0.2
0.5
0.1

Detection
D i ffrom b
below,
l through
h h the
h
piston
Left image shows soot volume fraction
in parts per million, ppm.
Calibration of Laser Induced Incandescence

The laser
Th l sheet
h t is
i reflected
fl t d into
i t the
th calibration
lib ti
burner using high reflective mirrors, and the
resulting incandescence is detected with the
camera Gain,
camera. Gain gate,
gate distance from burner,
burner laser
pulse energy et.c. is the same as during
measurements.
II. Laser-induced breakdown spectroscopy -
LIBS

J. Kiefer, J.W. Tröger, T. Seeger, A. Leipertz,

B Li,
B. Li Z
Z.S.
S Li and M M. Aldén
Aldén, ‘Laser
‘Laser-induced
induced
breakdown spectroscopy in gases using
ungated detection in combination with
polarization filtering and online background
correction’, Measurement science and
correction
technology 21, 065303 (2010).
 Polarization
suppression of laser
light - calibration

Calibration curves
 Detection limits

Argon gas with a

nominal purity of
99.996% with small
contents
t t off carbon
b
dioxide (<1 ppm),
oxygen (<4 ppm)
ppm),
water (<5 ppm) and
nitrogen (<10 ppm).

J. Kiefer, J.W. Tröger, T. Seeger, A. Leipertz, B. Li, Z.S. Li and M. Aldén, ‘Laser-induced
breakdown spectroscopy in gases using ungated detection in combination with polarization
filtering and online background correction’, Measurement science and technology 21,
065303 (2010).
 Application
pp I: Flames
Turbulent jet flame(CH4)

Turbulent jet flame(DME)

Atom ratio H/O and breakdown threshold

as a function of radial position in a non-
premixed methane flame
 Application II: Measurements of
N and
Na d K ffrom coal/biomass
l/bi
(a) (b) (c) (d) 20
K_coal_1
Na_coal_2
15

[Na,K] (ppm)
K wood 1
K_wood_1
Na_wood_2
10

0
0 5 10 15 20 25 30
Time (sec)

Release of sodium and potassium during

LIBS is
i operatedt d in
i (a)
( ) seeded
d d flame
fl (b) devolatilization with equivalence ratio of 1.3
devolatilization (c) char (d) ash cooking using LIBS.
phases