ENEA C.R. Casaccia, via Anguillarese 301, Rome Italy .

XIV A.I.VE.LA. National Meeting

Experimental study of turbulence-flame front interactions by means of PIV-LIF technique.

Troiani G., Marrocco M.

ENEA C.R. Casaccia, via Anguillarese 301, Rome Italy.

Experimental evidences

Multi-Scale Interactions between turbulence and flame front

• Turbulent velocity fluctuations increase the mass consumption rate, hence the turbulent burning velocity (ST) well above its laminar value (SL).

• Increasing turbulence beyond a certain level increases the mass consumption rate very little: leveling off of ST .

• At higher turbulent fluctuations possible flame extinctions (flame quenching).

Multi-Scale Interactions between turbulence and flame front

Joint PIV-LIF analysis of turbulent flames

• Scales larger than flame front: flame wrinkling,

flame front surface increase,

effects on the turbulent burning velocity.

• Scales smaller than flame front (high Karlovitz effects): smallest eddies penetration into the thermal thickness flame front,

thermal gradient misaligned to the radical species concentration gradient,

thickening of flame front,

increase of laminar burning velocity, quenching [1,2] .

from Flamelets to Extended-Flamelets concept. [3]

1. Ronney, Yakhot (1992) Combst. Sci. and Tech. 86.2. Gülder et al. (2000) Combustion and Flame 120.3. Poinsot, Veynante, Candel (1991) J. Fluid. Mech. 228.

2

LKa

LASER INDUCED FLUORESCENCE (LIF)

• Radical OH is excited by a =282 nm radiation >> fluorescence emission at =309 nm.

• Flame front position educed by OH concentration distribution

LASER Nd:Yag DYE LASER SHG

=532 nm =564 nm =282 nm =282 nm

=564 nm

x (mm)

T(K

)

OH

0 1 2 3 4

500

750

1000

1250

1500

1750

2000

2250

2500

0

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

Experimental apparatus

LASER Nd:Yag

DYE LASER SHG

532 nm

BEAM - SPLITTER

565.8554 nm

282.9277 nm

PRISMA

DANTEC SYSTEM

DELAY GENERATOR

CCD PIV

ICCD LIF

OTTICHE

PC LIF PC PIV

MULTI I/O BOX

FILTRO

FILTRO

FASCIO LASER PRIMARIO

FASCIO LASER SECONDARIO

CAMERA DI COMBUSTIONE

LAMA LASER PRIMARY LASER BEAM

SECONDARY LASER BEAM

LASER SHEET

COMBUSTION CHAMBER

FILTER

FILTER

x (mm)

y(m

m)

-40 -20 0 20

0

10

20

30

40

50

60

70

x (mm)

y(m

m)

-40 -20 0 20

0

10

20

30

40

50

60

70

x (mm)

y(m

m)

-40 -20 0 20

0

10

20

30

40

50

60

70

x (mm)

y(m

m)

-40 -20 0 20

0

10

20

30

40

50

60

70

a) Φ =0.64

d) Φ =1.48b) Φ =0.83

c) Φ =1.31

Flames at different equivalence ratios (Re=103 )

Flame topology changes due to variations in the turbulent burning velocity

CH4+Air

Flame wrinkling

Hot reacting flow island formation

Large scale turbulence-flame front interaction

Flame stretching

Small scale turbulence-flame front interactions

In this case the ratio ST/SL decreases until quenching is approached.

The flame front experiences thickening but it is still a continuous interface between reactants and products: Extended-Flamelets assumption[3].

3. Poinsot, Veynante, Candel (1991) J. Fluid. Mech. 228.

When turbulent scales are smaller than flame thickness (high Karlovitz number), some small eddies can penetrate into the flame front and modify its diffusive properties, increasing flame thickness and laminar burning velocity.

Mass conservation

Turbulence enhances flame surface wrinkling

L

T

L

T

S

'u

A

A

S

S

Turbulent burning velocity (ST)

Characteristics

• leveling off of ST at high u’.

• Flame extinction, Quenching, at higher u’ (symbol “x”).

• Quenching depends also from Karlovitz number which is not taken into account in figure 1.

Burning velocity (ST) depends from both small scale and large scale turbulence -flame front interactions.

Turbulent burning velocity (ST)

can be directly measured or predicted by modeling.

1. Mean velocity, upstream the turbulent flame brush, can be considered equal to the burning velocity

2. Models for turbulent burning velocity must take into account:

a. Flame surface increase by wrinkling

b. Flame thickening

c. Laminar flame velocity modifications

(fractal dimension)

Action of small scales

Flame front fractal dimension

u' /ul

D

4 6 8 10 12 14

2.2

2.3

2.4

2.5

2.6

2.7

2.8

Measured fractal dimension appears constant at varying turbulence intensity as confirmed in [2]

2. Gülder et al. (2000) Combustion and Flame 120.

LT

T SA

AS

D2

o

ITT A

AS

If the flame front can be considered a prefractal:

The flame surface area measurement scales as a power law of the resolution adopted for the measurement.

εo: order of the integral scale.

εI: order of the flame thickness.

2D

75.025.0T

LL

T ReAS

uf

S

S

Gouldin (1987)

A

outer cutoff

slope = 2-D

Inner cutoff

A

AT

c) a) b) d)

Future development

• Measurement of fractal dimension at different turbulence intensity

• Direct measurement of burning rates.

• Measurement of inner cutoff and evaluation of possible scaling law in the form Ka-p .

• Measurement and analysis of small scales turbulence interacting within the flame front.

• LASER measurement of both temperature (CARS) and species concentration distribution inside the flame front.

• Assessment of new models for the prediction of the turbulent burning velocity.