Lecture 4 Electric Field Effect on Flames: Ionic wind and Joule ...

Lecture 4 Electric Field Effect on Flames: Ionic wind and Joule heating

Yiguang Ju

Princeton University

Princeton Combustion Summer School2021.6.21

+ - + - + -

Flameweak plasmane,ni~1012/cm3

+-

+-

+-

DC fieldElectron heating:σE2

Ionic wind~10 m/s

-

+

Brande (1814)

Calcote (1963)

+ - + - + -

AC fieldElectron heating

CH + O = CHO+ + e

2

Flame is a weakly ionized plasma. It produces ionic wind, electron heating under an electric field

CHO+ + H2O = H3O+ + CO

f

aL

RT

ES

2exp

neDCcombustion

neACcombustion

nemwcombustion

ttt

ttt

ttt

~Timescales:

Ionic wind and Joule heating by an electric field

)( UVvm iini

[1] Robinson, M., 1962, “A History of the Ionic Wind,” American Journal of Physics, Vol. 30, pp. 366-372.

Flow induced by ion collision with neutral molecules in a flame and corona with an electric field. The ionic wind velocity can be 10 m/s which significantly modifies the near electrode flow field.

Momentum transfer between:

Electron-molecule collision energy transfer: )(2

32

geBeiei

i

TTknkE

Joule heating:

Methane, Φ = 1.0, Air Flow = 30 slm, Fuel Flow = 3.2 slm Flow velocity ≈ 1.0 m/s, Voltage: 0

V vs. t 2000 V, The anode-cathode gap was kept constant at 40mm, Ganguly et al., 2008

3

2E

frequencycollision molecule neutral-ion :inv

Ionic wind

Calcote, 3rd Symposium on Combust. Flame, Explosion Phenomena(1948)

Carleton and Weinberg, Nature 330 (1987)

Lawton, Mayo, Weinberg, Proc. Roy. Soc. A 303 (1968)

F = qE

Min Suk Cha

Ionic wind: Mechanism• In flames, most of ions are positive ions.• Electron mobility is high (smaller mass than ions), its motion

is reduced by the motion of positive ions.• Therefore, ionic wind by negative ions and electrons is

smaller than that by the positive ions.

e

an

od

e

ca

tho

de

++

++

++

++

++

++

++

++

++

++

++

++

+

+

++

++

++

+

+

+

+

+

+

+ +

+

+ +

+

+

+

+

+ +

+

e

electron H3O+

Heavier positive charge carrier

e

an

od

e

ca

tho

de

++

++

++

++

++

++

++

++

++

++

++

++

+

+

++

++

++

+

+

+

+

+

+

+ +

+

+ +

+

+

+

+

+ +

+

--

--

--

--

--

-

--

--

--

-

--

--

--

--

--

--

- --

-

--

--

-

-

-

--

-

---

-

-

-

-

-

O2- H3O

+

Comparable positive/negative charge carrier

U U

2/1

VU

q

q

)(C/mdensity Charge:

)(kg/mdensity gas:

(V) change Voltage :V

(m/s) velocity windionic sticCharacteri:

Charge :q

field Electric :

3

3

q

qU

EF=qE

E

q

Axisymmetric jet flames: transverse DC fields

Park and Cha, Combust. Flame, submitted (2017)

EEl

ect

rod

e

Ele

ctro

de

L = 50 mm

80

mm

Lase

r

L / 2 mm

30

mm

6 mm

CH4 or

CH4 – air mixture

High

Voltage

Premixed Non-Premixed

16kV-16kV 16kV-16kV

Most ions are positive.

Axisymmetric jet flames: transverse DC fields

Premixed Nonpremixed

Vertical Horizontal Vertical Horizontal

Fig. 2a Fig. 2b

0 k

V

−8 k

V

−16

kV

Gro

un

d

H = 2 mm

6 mm

10 mm

14 mm

Nozzle

1.4

0 0 −20 20

[mm]

25 (a) Vertical

Nozzle

0.6

0

0 −20 20

[mm]

12.5

−12.5

0

(b) Horizontal

E

Min Suk Cha

Counterflow nonpremixed flames: DC fields

Park and Cha et al., Combust. Flame, 168 (2016)

E = V/d

Fuel + N2

O2 + N2

o Drastic change in flow field

o Formation of the dark zone

0 kV

− 2.4 kV

(a)

(b)

(c)

(d)

− 1.6 kV

− 0.5 kV

Cathode

Anode

o Flame acts as a source of flow

o Double stagnation planes

Positive ion

Negative ion

Electron

Neutral molecule

Movement of ion (+)

Movement of ion (−)

Bulk flow

+

−

Counterflow nonpremixed flames: AC fields

No field 2 kV, 10 Hz

2 kV, 100 Hz 2 kV, 1000 Hz

Park and Cha et al.Under preparation

Effect of AC frequency on flow filed

Propagating edge flames in counterflow: DC

Propagation of nonpremixed edge flames through DC fields

(c)

Stagnation Plane

162 kV

(d)

Stagnation Plane

(a)

Ud [cm/s] = 126–2 kV

Propagating direction

Field line

(b)

(e) Propagating direction

Field line

131+2 kV

(f)

Tran and Cha , Combust. Flame, 173 (2016)

fuel/inert

O2/inert

DF

RPF

LPF propagating

direction YF

GND

HV

Field direction

Propagating edge flames in counterflow: DC

o Ionic wind and

secondary flow

modification is the

most important

factor

o Reduced

displacement

speeds

o Rather unaffected

propagation speeds

Combust. Flame 173:114(2016)

E = V/dUd :displacement speedUu: unburned gas velocityUedge =Ud – Uu

Propagating edge flames in counterflow: AC

o Wavy motion is

closely related with

fAC.

o Ionic wind and

secondary flow

modification is the

most important

factor

o Reduced

displacement

speeds

o Rather unaffected

propagation speeds

Proc. Combust. Inst. 36: (2017)

kV

Stagnation Plane

Ud [cm/s] = 162

2 kV, 50 Hz 135

2 kV, 100 Hz 135

2 kV, 2000 Hz 129

2 kV, 1 Hz 131Propagating direction

Field line

50

100

150

200

250

(a)

CH4/O

2/N

2 = 1/2/5.5

Uedge

Ud

0 kV 2 kV1 kV

50

100

150

200

250

(b)

C3H

8/O

2/N

2 = 1/5/14.3

[cm

/s]

1 10 100 100050

100

150

200

250

(c)CH4/O

2/CO

2 = 1/2/2.3

Applied fAC

[Hz]

Applied fAC [Hz]

Sp

ee

ds [

cm

/s]

1 10 100 1000

Tran and Cha Proc. Combust. Inst., 36 (2017)

Fig. 2.29 Image sequence of a propane/air flame with an equivalence ratio of 1.2. The applied dc

voltage was slowly increased (left to right), leading to the flame blowing off the burner [65].

DC electric field on flame stability

• Ionic wind

• Corona effect

• Instability

• Electron heating

Fuel: Methane

Φ = 1.0

Air Flow = 30 slm

Fuel Flow = 3.2 slm

Flow velocity ≈ 1.0 m/s

0 V - 2000 V (DC)

Why?

kgEqn

bu

bu

ei

])([

Instability Growth rate:

Wisman, D., Ryan, M., Carter, C. and Ganguly, B.,

2008. In 46th AIAA Aerospace Sciences Meeting and

Exhibit (p. 1400).

Pri

nce

ton

Un

iver

sity

Combustion lab.

Reduction of emission via DBD electric field and discharge

on a diffusion flame – E-field makes a flame shorter through ionic wind effect

– As soon as a discharge lights up

• No yellow luminosity

– Reduction of soot particles

• Onset of PAHs is suppressed by DBD

0 kV 4 kV 6 kV 8 kV 9 kV 11 kV 14 kV

brush corona weak streamer strong streamer

Soot suppression / Enhanced reaction rate

PAH

PLIF

PAH

PLIF

Cha et al., Combust. Flame 141:438, 2005)

Pri

nce

ton

Un

iver

sity

Combustion lab.

• Microwave frequency is 2.45 GHz

• Three stub tuner to tune the cavity

• Actual Q (5-1000)

Mass flow rate = 5744 st. cm3/min

Exit velocity = 54 cm/s, Equivalence ratio = 0.70

Effect on microwave electric field on flame speed enhancement

Zaidi, S., Stockman, E., Qin, X., Zhao, Z., Macheret, S., Ju, Y., Miles, R., Sullivan, D. and Kline, J., 2006, In 44th AIAA

Aerospace Sciences Meeting and Exhibit (p. 1217).

Pri

nce

ton

Un

iver

sity

Combustion lab.

Estimated vs Experimental Results for Laminar Flame Speed Enhancement

0.0 0.2 0.4 0.6 0.80.0

0.2

0.4

0.6

0.8

1.0

qe0-x

/qrt

ne

Ele

ctr

on

nu

mb

er

de

ns

ity

(1

01

01

/cm

3)

X (cm)Fra

cti

on

of

ele

tro

n h

ea

tin

g a

nd

no

rma

lize

d h

ea

t re

las

e

qr/q

r,max

CH4-air

=1.0

0

10

20

30

40

50

MW Off

Sref = 29.6 cm/s

MW On

Sref = 35.7 cm/s

1.1

mm

In flames, microwave field

mainly heats the electrons

and raises flame temperature

Pri

nce

ton

Un

iver

sity

Combustion lab.

Planar FRS Measurements 30 W Pulsed Microwave Enhancement

30 kW-peak, 1 ms, 1000 Hz Pulses

17

No MW Pulsed MW

• Observed Flame Speed Effect

– Increase of 0.25 L/D units ~6% flame

speed enhancement

• 80 K in post flame temperature (CW

saw ~150 K)

• Energy deposited within a few mm of

reaction zone

Flame Shifted

Coordinates

CH4/air, f = 0.76

80 K 12 W energy deposition 50% magnetron coupling efficiency

Temperature [K]

Stockman, E.S., Zaidi, S.H., Miles, R.B., Carter, C.D. and Ryan, M.D., 2009. Combustion and Flame, 156(7), pp.1453-1461.

Pri

nce

ton

Un

iver

sity

Combustion lab.

(-) (+)

Breakdown/

Arc Initiation

Non-Equilibrium

Critical

Point

Near

Equil.

R1

2

3

4

(a)

Combustion enhancement by a gliding arc (Joule heating and kinetic effect)

(Yardimici et al. 1999).

Pri

nce

ton

Un

iver

sity

Combustion lab.

Short-cut

Left: A short-cut event recorded at 20 kHz framing rate using an exposure time of 13.9 μs. The short-cut current path is

indicated by the arrow in the frame of t = 50 μs. Right: Three typical single-shot OH PLIF images of a gliding arc using an

exposure time of 2 µs, at two flow rates (a) 17.5 SLM, (b) 42 SLM. The typical thickness of the OH distribution is labelled

in the images with unit of centimeters (Courtesy from Dr. Z.S. Li at Lund University)

1. Sun, Z.W., Zhu, J.J., Li, Z.S., Aldén, M., Leipold, F., Salewski, M., Kusano, Y., Optics Express. 2013, 21 (5) 6028-6044.

2. Zhu, J., Sun, Z., Li, Z., Ehn, A., Aldén, M., Salewski, M., Leipold, F., Kusano, Y., Dynamics, Journal of Physics D: Applied Physics. 2014, 47 (29) 295203.

Short cut event and OH measurements in a gliding arc (air)

The combination of thermal heating and radical production (high electron density and high electron energy) of a

non-thermal gliding arc can enhance ignition and flame stabilization in both thermal and kinetic ways.

• Low frequency electric field generates ionic wind flowing to both electrodes from a flame due to positive and negative charge carriers.

• Ionic wind can reduce soot/NOx formation due to the change of mixing and flame temperature.

• Ionic wind may induce flame instability due to the force field.

• Ionic wind also modifies flame speed and reduces flame temperature due to increased heat losses from the flame zone.

Summary: Electric field effect on flames1. Ionic wind

2. Joule heating

• Electric field generates Joule heating in the flame zone and at the downstream of the flame.

• The electron Joule heating can enhance flame speed via the increase of flame temperature.

• Microwave Joule heating in flames is not energy efficient because much of the energy absorbed by in the burned gas.

3. Radical production by strong electric field

• When the electric field is above the breakdown threshold, a gliding arc or corona can produce radicals to enhance ignition via kinetic pathway.

• A gliding arc has high temperature and high electronic energy and density, which lead to both thermal and non-thermal enhancement effects on flames.

It is necessary to understand the kinetic effect of non-thermal plasma at high E/N on combustion

Lecture 5 Chemistry and Kinetic Studies of Plasma-Assisted Combustion

Yiguang Ju, Princeton University

• Important chain-initiation and branching reactions in combustion• Electron impact reactions• Electronically excited species (radical production and fast heating)• Vibrationally excited species (slow heating)• O3/NOx

• Plasma chemistry and timescales• Impact of plasma chemistry on combustion


1. Important combustion reactions

O+RH → R+OH

R → R’’+2OH

O+RH → R’’+ 3OH

Plasma assisted combustion:

e+O2 → e+ 2O

Plasma provides new reaction pathways to accelerate chain reaction processes

RH+ O2 → R+HO2 High Temperature (>1100 K) slowRH+HO2 → R+H2O2 High pressure/low temperature (>550 K) slowR+HO2 → RO+OH High pressure/low temperature (>550 K) slow

1. Chain initiation and propagation reactions

R → RO2→QOOH → O2QOOH →R’’+2OH Low temperature (300-800K) Slower

H2O2 → 2OH Intermediate temperature (800-1100 K) Slow

H+ O2 → O+OH High Temperature (>1100 K) Fast2. important branching reactions at different temperatures

e+O2 → e+ O2(a1Δg) H+O2(a

1Δg) → OH+O Faster

Faster

Major kinetic pathways of Plasma Assisted Combustion & Chemical Processing

e-

N2(A)N2(B)N2(C)N+N

O2N2(*)N2(*) + 2 O

O2 O+O(1D)

Fuel (RH)RH(v)R + H

R + R’

R’ + 2H

Hot flame

Cool flame

N-heptane/He

O2/He

O2 HO2(

1Δ) O + OH

N2(*)N2(v’)

O2N2(v’’) + 2 O

RHR + OH

RCO + H2

R’OH + R

O2 + O O3

R + O2(1Δ) = ?

N2(*) + Fuel=?

O3 + Fuel=?

O(1D) + Fuel=?Plasma

Excitation

Dissociation

Excitation

Ju & Sun, PECS, 2015

• Electron impact cross-sections/branching ratio• Electronically excited sates: reaction rates• Vibrationally excited sates: reaction rates• Long life time active species: O3, NOx, …

NOx+ Fuels = ?

O2NOx

e + RH = ?

4

Measurements of exited species and radical production in plasma

O is produced by

N2(B,C)+O2=2O+N2

Stancu, G.D., Kaddouri, F., Lacoste, D.A. and Laux, C.O., 2010.

Journal of Physics D: Applied Physics, 43(12), p.124002.

A schematic of the key reaction pathways for high

pressure fuel oxidation of at different temperatures

(blue arrow: Below 700K; yellow arrow: 700-1050

K; red: above 1050K). Green: plasma activated

pathway

Fuel(RH)

+OH

R+O2RO2

QOOH

O2QOOH

HO2

H2O2

2OH

Small

alkeneC2H3/CH2O

H/HCO+O2+(M)

Plasmae, R*, N2*, O2*,O*RH(v), R(v), N2(v), O2(v), HO2(v)

+O2

+O2

CO/CO2

R*, O

(1D

)

O2(v), O2(a1Δg)

RO2*

III

Interaction of plasma chemistry with reaction kinetics of

large alkanes w/wo in plasma assisted combustion

Reaction rate Transition state theory

DCABBAkk

21

*

)*

exp(**

**)(

Tk

EE

qq

q

h

TkTk

B

BA

BA

ABB

How does plasma affect elementary rate constant?

e.g. at 800 K

O2 (a1Δg) + H = OH+O Fast

O2 + H = OH+O Slow

CH3 +O2(v) → CH2O+OH Fast

CH3 +O2 → CH2O+OH Slow

2. Plasma chemistry and timescales of kinetic processes

Electron Kineticstfp

EEDF

Ionization

Excitation / Quenching

Ion/Molecule Kinetics

tfp

ttr

trot

telec

Recombination

Combustion Processes tfp

ttr

trot

tvib

Ion-Ion, Ion-Molecular

Radicals

Molecules

10-14 10-12 10-10 10-8 10-6 10-4 10-2 s

Courtesy of Andrey Starikovskiy

Fig. 1.5 Schematic of timescales and key kinetic

pathways at different stages of plasma assisted

ignition and combustion.

Ju and Sun, PECS, 2015

Potential Energy Curves of O2

O2(B3Su-), 8.4 eV

smax = 1.0 A2 (9.4 eV)

O2(3Pg), 5.6 eVsmax = 0.16 A2 (12 eV)

O2(A3Su+), 4.5 eV

smax = 0.18 A2 (6.6 eV)

O2 (b1Σg+) at 1.6 eV

O2 (a1Δg) at 0.98 eV

r, nm

E, e

V

DE ~ 1.5 eV

DE ~ 1 eV

O2 (a1Δg)

Electron impact reaction is a function of electron energy distribution (E/N)

Electron impact reaction cross sections-O2

1. effective 2. rotational excitation3-6. O2(v1) - O2 (v4)

7. O2(a1) 8. O2(b1) 9. O2(A3Su+), 4.5 eV

10. O+O 11. O+O(1D) 12. O+O(1S)13. O2

+

Potential Energy Curves of N2

N2(A3Su+), 6.2 eV

smax = 0.08 A2 (10 eV)

N2(B3Pg), 7.35 eVsmax = 0.20 A2 (12 eV)

N2(C3Pu), 11.03 eVsmax = 0.98 A2 (14 eV)

r, nm

E, e

V

Threshold energy diagram

Electron impact reaction is a function of electron energy distribution

Energy Transfer of non-equilibrium excitation in Plasma Discharge

1 10 100 10001E-3

0,01

0,1

1

H2(rot)

O2(4.5 eV)

O2(dis)

H2(el)

Rot+tr

O2(a+b)

O2(v)

H2(v)

ion

N2(el)

N2(v)

N2:O

2:H

2 = 4:1:2

En

erg

y lo

ss f

racti

on

E/N, Td

Physics of Nonequilibrium Systems Laboratory

O2 (a1Δg)

Influence of Electronic and Vibrational Excitation on Combustion Kinetics

N2 + e = N2(C3) + eN2(C3) + O2 = N2 + O + OO2 + e = O + O + e

N2 + e = N2(v) + eN2(v) + HO2 = N2 + HO2(v)HO2(v) = O2 + H1 10 100 1000

1E-3

0,01

0,1

1

H2(rot)

O2(4.5 eV)

O2(dis)

H2(el)

Rot+tr

O2(a+b)

O2(v)

H2(v)

ion

N2(el)

N2(v)

N2:O

2:H

2 = 4:1:2

En

erg

y lo

ss f

racti

on

E/N, Td


Influence of Vibrational Excitation on Low-Temperature Kinetics: H2O2 Decomposition

Measured and calculated OH decay time. P = 1 atm. a) 3%H2 + air; b) 0.3%C4H10 + air.


Effect of “Hot” Atoms on Active Species Production in High-Voltage Pulsed Discharges

0,1 1

1

10

Cro

ss s

ecti

on

, 10

-16 c

m2

E, eV

[15]

[33]

[23]

This work

Potential energy curves and hot atoms formation

Momentum transfer cross section for the H-H2 scattering

Cross sections for scattering of H atoms with H2, O2, CH4 and N2

0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 4,5 5,0 5,5 6,0 6,5 7,0

100

101

Cro

ss

se

cti

on

, 1

0-1

6 c

m2

Energy, eV

H-H2 (el)

H-O2 (el)

H-CH4 (el)

H-N2 (el)

H+O2=O+OH (new)

H+CH4=CH

3+H

2 (new-1)

H+CH4=CH

3+H

2 (new-2)

(a)

PRINCETON

University

Nonequilibrium distributions of neutral species are formed in different physical situations.In laboratory experiments and in the terrestrial atmosphere, there are numerous collisionalprocesses in which translationally energetic (superthermal) atoms with energies muchabove thermal energies are produced.

Direct electron-impact dissociatione + O2 → e + O2* O2* → 2O(3P,1D) + 1.3 eVe + H2 → e + H2* H2* → 2H(1S) + 4.5 eVe + CH4 → e + CH4* CH4* → CH3 + H + 3.5 eV

Effect of “Hot” Atoms on Active Species Production in High-Voltage Pulsed Discharges

0 10 20 30 40 50

10-1

100

CH4:O

2=1:2

CH4:O

2:N

2=1:2:8

H2:O

2=1:9

H2:O

2=2:1

En

erg

y, e

V

Number of collisions

(a)

Average energy of H atoms in various gaseous mixtures

PRINCETON

University

0.4 0.5 0.6 0.7 0.80.9 11 2 3 4 5 6 7 810

-3

10-2

10-1

1

Sp

ecie

s a

mo

un

t p

er

on

e h

ot

ato

m

Initial H atom energy, eV

H CH3

H2 OH

O

HO2H2O

H(h

) + O

(1D)

H(h

)

O(1

D)

Gro

und

Sta

te

No

Rad

icals

0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

Pro

du

ctio

n, p

pm

/10

-5 e

V/m

ole

c

CH4-2O2 mixture

T=300 K; P=1 atm

[H2]

[H2O]

[H2O2]

[CH2O]

[CH3OH]

[CH3O2H]

H(h

) + O

(1D)

H(h

)

O(1

D)

Gro

und

Sta

te

No

Rad

icals

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

Pro

du

ctio

n, p

pm

/10

-5 e

V/m

ole

c

2H2-O2 mixture

T=300 K; P=1 atm

[H2O]

[H2O2]

Analysis of the effect of formation of "hot" atoms with excessive translational energy shows the important role of these processes in formation of active radicals.

The density of radicals produced in discharge plasma can be several times higher than that produced in the absence of high-energy atoms.

The effect plays a fundamental role in the formation of the initial distribution of active species in combustible mixtures and can greatly influence the kinetics of ignition and oxidation at low gas temperatures.

Species produced during energy degradation of one “hot” H atom

Plasma-assisted oxidationin H2-O2 mixture

Plasma-assisted oxidationin CH4-O2 mixture

Fast Gas Heating

Slow Gas Heating

Electron-ion recombination

Ion-ion recombination

Collisional energy transfer Electronically-excited species

Hot atom and molecule

e + O2+ → O + O* + ΔE

N2(A,B,C,a) + O2 2O + DE

O2* → 2O(3P,1D) + DE

O2- + O2

+ + M→ 2O2 + M + ΔE

Vibrational energy relaxation N2(v) + M = N2(v-1) + M +DE

0,0 0,1 0,2 0,3 0,4 0,520

30

40

50

ne0

=1015

cm-3 1 atm

ne0

=1014

cm-3 1 atm

ne0

=1015

cm-3 300 Tor

ne0

=1014

cm-3 300 Tor

Fra

cti

on

al p

ow

er,

%

Mole fraction of O2

E/N = 103 Td

Fractional Electron Power Transferred Into Heat in N2:O2 Mixtures

High oxygen, faster gas heating!

Gas heating at high E/N

Princeton Plasma Combustion KineticsMajor Pathways

Ar O2 N2 H2 CxHyOz

Ar+ O2+ N2

+ H2+ CxHyOz

+,…, CxH1Oz+

Ar, N2, O2

H2, CxHyOz

Ar2+, N4

+, O4+

, N2O2+, NH2

+, H3+, HO2

+, H3O+; O-, O2-, O3

-, O4-; CxHyOz

+,…, CxH1Oz+

Charge transfer, negative and complex ions formation

O2, CxHyOz N2, O2, CxHyOz O2, CxHyOz

Electron-ion recombination

O2+

, O4+, CxHyOz

+

Ion-ion recombination

O2- + N2

+; O2- + CxHyOz

+

Molecule-ion reactions

O2- + H; O- + H2

electron detachment Electronically-

excited particles formation

Low-Temperature ReactionsFast Gas Heating

“Hot” atoms and molecules formation Ionic chains

CxHyOz

H- transfer

Oh, Hh, Nh, O2h, H2

h

Ionization by electron impact. k = f(E/N)

O(1D), O(1S), N(2D), H(n=2)Andrey Starikovskiy

Princeton Plasma Combustion KineticsMajor Pathways

Ar O2 N2 H2 CxHyOz

N2(vib)

Energy transfer to buffer

Formation of vibrationaly-excited products

VT relaxation

Slow Gas Heating

Reactions of vibrationaly excited molecules

Vibrational levels excitation by electron impact. k = f(E/N)

H2(vib)

H2(v) + O → H + OH(v) H2(v) + OH → H2O + H

N2(v) + O; N2(v) + H2

N2(v) + H2O; N2(v) + CxHy

N2(v) + HO2 → N2 + HO2(v) OH(v) + N2 → OH + N2(v)

Reactions of vibrationaly excited molecules

HO2(v) → H + O2 OH(v) + H2 → H2O + H

Energy transfer to reagents

Electron impact ionization/dissociation/excitation

e +O2 =O++O+2e (R1a)

e +O2 =O+O(1D) (R1b)

e +O2 =O2(1Δg)+e (R1c)

e +O2 =O2(v)+e (R1d)

Electron ion recombination, attachment, charge transfer

e+O2+ =O+O(1D) (R2a)

O2+ +O2

- =2O2 (R2b)

e+O2 +M = O2- +M (R2c)

H2O+N2+ =H2O

++N2 (R2d)

Dissociation and energy transfer by ions and excited species

N2(A,B,C)+O2 =O+O(1D)+N2 (R3a)

O(1D)+H2 = OH+H (R3b)

H+ O2(1Δg)= O+OH (R3c)

N++O2= O++NO (R3d)

CH3+HO2(v)=CH2O+OH (R3e)

N2(v=5) +N2 = N2(v=3) + N2 (R3f)

N2(v) + HO2 → N2 + HO2(v) (R3g)

Typical plasma reactions for radical production and heating

Recombination/fast heatingRecombination/fast heatingAttachmentCharge Transfer

Radical productionNon-equilibrium excitation

Slow heating

>10 eV

~10 eV~1 eV0.2-2 eV

Radical productionFast heating

Pressure

Time

What are the major species produced by plasma?

•Long lifetime species?

•Short lifetime plasma generated species?

NO, O3, O2(a1Δg)

O, N2 (A,B,C)*

19

Fig. 3.5: Rate constants (a) and reaction flux (b) for reactions for dissociation by electron

impact at electric field values equal to 200 Td and 500 Td and chain branching reactions.

Ju and Sun, PECS, 2015.

Question: When will electron impact dissociation process become important in combustion?

3. Impact of plasma chemistry on combustion

Comparison of the reaction rates of electron impact and O2(a1Δg) for radical production

A. M. Starik, B. I. Loukhovitski, A. S. Sharipov and N. S. Titova, 2016, Phil. Trans. R. Soc. A 373: 20140341

S M Starikovskaia, J. Phys. D: Appl. Phys. 47 (2014) 353001

Important radical production channels

(Ground)

(Ground)

(Ground)

Multi-channel insertion reactions between O(1D) and fuel molecules

Brian M. Hays and Susanna L. Widicus Weaver, dx.doi.org/10.1021/jp400753r | J. Phys. Chem. A 2013, 117, 7142−7148

Data Source: NIST Kinetics https://kinetics.nist.gov/kinetics/index.jsp

O + H2 → OH + H

k ≈ 6.0×1011 cm3 ·mol-1·s-1 (1200 K)

O(1D) + H2 → OH + H

k ≈ 1.0 × 1013 cm3 ·mol-1·s-1 (300 K)

O(1D) + CH3OH CH2OH+OH Φ=0.55 ± 0.10

HOCHO+2H Φ=0.45 ± 0.10

kCH3OH+O(1D) = (3.0 ± 0.3) × 1014 cm3 mole-1 s-1

Matsumi, Y, et al., The Journal of Physical Chemistry (1994)

Few data are available for oxygenated fuels

23

Figure 1: Ethylene ozonolysis reaction pathway

with major intermediate species (primary

ozonide, POZ; secondary ozonide: SOZ).

O3 Enhanced Low Temperature Fuel Oxidation:

Ozonolysis via C=C double bond

9 9.5 10 10.5

Sig

na

l In

ten

sit

y /

a.u

.

Photon Energy (eV)

POZ IE

SOZ IE

KHP conformers

Lowest energy conformer

Hydroxymethyl Formate IE

0E+0

1E-6

2E-6

3E-6

4E-6

5E-6

300 400 500 600 700 800

Mo

le F

rac

tio

n

Temperature (K)

C2H4O3

0

0.0E+0

2.5E-5

5.0E-5

7.5E-5

1.0E-4

1.3E-4

1.5E-4

300 400 500 600 700 800

Mo

le F

rac

tio

n

Temperature (K)

CH3OHC2H4OCH4O2C2H6O2

0

0.0E+0

5.0E-6

1.0E-5

1.5E-5

2.0E-5

2.5E-5

3.0E-5

3.5E-5

300 400 500 600 700 800

Mo

le F

rac

tio

n

Temperature (K)

C2H6OC2H4O2C2H2O

0

SOZ

Rousso, Aric; Hansen, Nils; Jasper, Ahren; Ju, Yiguang, Journal of Physical Chemistry A, 2018, 122(43),

Ethanolketene

Hydroperoxde

2424

Criegee Adduct Network for Aerosol Formation

11 Distinct chains were

observed in this work

First additions (Blue) were all

identified and quantified

using PIE scans and calculated

ionization energies

CI adducts responsible for

most of the large oxidation

reactions in this system at

atmospheric temperatures

Rousso et al., Phys. Chem. Chem. Phys., 2019, Accepted.

25

N-dodecane/NOx Coupling Reactions Validation

N-dodecane/NOx coupling reactions [1]

RH + NO2 = R + HONO

R + NO2 = RO + NO

O2QOOH + NO = OH + 2CH2O + CH2 + Alkene + NO2

RO2 + NO = RO + NO2

NOx sub-model: HP-Mech [2]

[1] J.M. Anderlohr et al. Combust. Flame. 156 (2009) 505–521

[2]. http://engine.princeton.edu/mechanism/HP-Mech.html

[3 G. Kukkadapu et al. 10th U.S. National Combustion Meeting

(2017), paper 2RK-0500]

400 600 800 1000-0.002

-0.001

0.000

0.001

0.002

0.003

0 ppm

300 ppm

LLNL/NOx

Fu

el

mo

le f

rac

tio

n

Temperature [K]

nC12

H26

= 1.3

Exp.

N-dodecane model: Detailed LLNL model [3]

N-dodecane/NOx coupling reactions validated by JSR experiment

Experimental and simulated results

LLNL/NOx coupled model

http://engine.princeton.edu/mechanism/HP-Mech.html

2021 AIAA SciTech Forum26

Plasma effect on NO formation

NO production in 0.80 N2/0.20 O2 and 0.01 C12H26/0.19 O2/0.80 N2

• NO increases with frequencies in N2/O2

• NO decreases dramatically with 1% n-dodecane.

• RO2 + NO = RO + NO2

• NO pathways in N2/O2

N + O2 = NO + O

N(2D) + O2 → NO + O/O(1D)

NO2 + O = O2 + NO

• Strong NOx/fuel kinetic coupling in PAC

Ammonia/air flames with and without

plasma: NOx emissions without and

with plasma ( φ= 0.94)

W. Sun et al,, Combustion and Flame 228 (2021) 430–432

X. Mao et al., Proc. Combut. Inst. 38, 2021

PAC: how does plasma change the branching reactions in combustion?

t1 t2

Hot ignitionLow temperature ignition

0.0 0.1 0.2300

600

900

1200

1500

Tem

pera

ture

(K

)

Time (sec)

R+O2=RO2

HCO+O2=CO+HO2

2HO2=H2O2+O2

H2O2=2OH

H+O2=O+OHO+H2=H+OH

RO2→QOOH →R’+OHO2QOOH →R’’+2OH

Thermal effectKinetic effect

500-850 K

850-1100 K

>1100 K

High Temp.

Intermediate Temp.

Low Temp.

Large molecules Fuel fragments Small molecules

CH2O+X=HCO+XH

Schematic of kinetic and thermal enhancement pathways of plasma assisted combustion for liquid fuels at high, intermediate, and low temperature, respectively

Y. Ju and W. Sun, Prog. Energy Combust. Sci., 2015

28

Plasma activated LTC at much shorter time, lower pressure….

We can create cool flames even at1 atm or below?

Plasma activated Cool Flames :A new way to burn with plasma

Residence time

Tem

pera

ture Plasma

generated

LTC

Ignition

Extinction

HTC

LTC

t2 t1

t2<< t1

Fuel (RH)

+(OH, HO2)

R

nOH

aldehyde C2H3/CH2O

H/HCO+O2+(M)

+M

+O2

CO2

CO

OQ’O

PlasmaO(1D), O, R, O3

O2(1Δ), N2(v), …

alkene +

+O2

RO2

QOOH

O2QOOH

+O2

KOOH +O2

O2P(OOH)2

H2O2

+O2

HO2

+HO2

Observation of plasma activated self-sustaining Cool Flames

Heated N2 @ 550 K

N2 @ 300 K

Stagnation

plane

Oxidizer @ 300 K

with plasma discharge

Fuel/N2 @ 550 K

Fig. 1 Schematic of experimental setup

(a) Hot diffusion flame (b) Cool diffusion flame

Fig. 2 Hot and cool n-heptane

diffusion flames at the same condition

Tf~1900 K Tf~650 K

400

800

1200

1600

2000

2400

0.1 1 10 100 1000 10000

Ma

xim

um

te

mp

era

ture

Tm

ax

[K]

Strain rate a [s-1]

nC7H16/N2 vs O2 or O2/O3

in counterflow burnerXf = 0.05,Tf = 550 K, and To = 300 K

Extinction limit ofconventional hot diffusion flame

(HFE)

Extinction limit ofcool diffusion flame

(CFE)

without O3

with O3

HF branch

CF branchHTI

LTI

Extinction/instability

Transition to hot flame

Won, S.H., Jiang, B., Diévart, P., Sohn, C.H.

and Ju, Y., 2015. Proceedings of the

Combustion Institute, 35(1), pp.881-888.

6. Diagnostics of plasma physics and chemistry in PAC

1. Measurements of plasma properties and kinetic processes

2. Diagnostics methods of plasma produced active species and non-equilibrium

3. Multispecies diagnostics and kinetic model development of plasma assisted combustion

Yiguang Ju, Princeton University


Why do we need measurements of E, Te, and ne?

• Kinetic enhancement and induced chemistry from plasma begins with energy transfer from electrons

• Development of predictive models requires validation of these plasma parameters

• Few studies measure all three parameters in the same discharge

2

Electrons(ne, Te)

Gas molecules

Ions, radicals, excited species

E

6.1.1 Measurements of Plasma Properties: electron density and temperature

Thomson scattering

𝑃𝑠 ∝ 1 − sin2 𝜃 cos2 𝜙0

𝑁𝑠 =𝐸𝐿

ℎ𝜈0Δ𝐿 𝑛𝑒

𝑑𝜎𝑒

𝑑Ω𝜂

𝜃

𝜙0

𝐸𝑖0

𝑘𝑠

𝑘0

𝑧𝑥

𝑦

• H. Van der Meiden, "Thomson scattering on low and high temperature plasmas", Ph.D, Technische Universiteit Eindhoven, 2011.• A. Roettgen, "Vibrational Energy Distribution, Electron Density and Electron Temperature Behavior in Nanosecond Pulse Discharge Plasmas by Raman and

Thomson Scattering", Ph.D, The Ohio State University, 2015.

k0: Laser beam direction, ks: Scattering signal wave vector 𝐸𝑖0: Polarized electric field, scattering is rotationally symmetric about 𝐸𝑖0. x-z plane: the plane of observation θ: the scattering angle relative to the laser beam. 𝜙0: angle between observation plane and the polarization angle. 𝐴𝑒 and 𝐴𝑁2

: integrated intensities of the Thomson and Raman spectra𝑑𝜎𝑒

𝑑Ωand

𝑑𝜎𝑁2

𝑑Ωthe Thomson and N2 Raman scattering cross sections

𝑓𝐽=6: the fraction of N2 molecules in the J = 6 rotational state

Δ𝜆 1 𝑒: the half 1/e width of the Gaussian broadening profile

EL: laser energy, η: optical efficiency, ΔL: length of observed scattering segment.

𝑛𝑒 =

𝐴𝑒𝐴𝑁2

𝑑𝜎𝑁2

𝑑Ω𝑅𝑎𝑚𝑎𝑛,532𝑛𝑚

𝑑𝜎𝑒𝑑Ω

𝑛𝑁2𝑓𝐽=6

Δ𝜆 1 𝑒 =2𝜆0

𝑐sin

𝜃

2

2𝑘𝐵𝑇𝑒𝑚𝑒

12

𝑇𝑒 =𝑐2𝑚𝑒

8𝑘𝐵 sin2 𝜃 2

Δ𝜆 1 𝑒

𝜆0

2

Power of scattering:

Number of photo-electrons:

Roettgen (2015): use Rotational Raman Scattering for calibration using the J = 6 → 8 transition of N2 at P = 100 Torr

Thomson Scattering Experimental Setup and Calibration

Timothy Chen, Princeton, 2017

A. van Gessel, E. Carbone, P. Bruggeman and J. van der Mullen, Plasma Sources Science and Technology, vol. 21, no. 1, p. 015003, 2012.

B. Chen TY, Rousso AC, Wu S, Goldberg BM, Van Der Meiden H, Ju Y, Kolemen E. Journal of Physics D: Applied Physics. 2019 Feb 27;52(18):18LT02. Decoupling Raman and Thomson Signals

Te~Δ𝜆2ne~𝐴𝑟𝑒𝑎

Thomson scattering in ns-pulsed Ar DBD75 Torr pure Ar, 50 Hz pulse frequency, 30 ns pulse width, 90 ms residence time

0 20 40 60 80 100 120 140 160 180 200

0E+00

1E+12

2E+12

3E+12

4E+12

Electron Number Density

Ele

ctr

on N

um

be

r D

en

sity (

cm

-3)

Time (ns)

5

Approximate detection limit

Electron density below detection limit before 30 ns

0 20 40 60 80 100 120 140 160 180 200

0

1

2

3

4

5

Electron Temperature

Ele

ctr

on T

em

pe

ratu

re (

eV

)

Time (ns)

Electron temperature does not decay below 0.5 eV.

A. Chen TY, Rousso AC, Wu S, Goldberg BM, Van Der Meiden H, Ju Y, Kolemen E. Journal of Physics D: Applied Physics. 2019 Feb 27;52(18):18LT02.

CH4 addition to He decreases ne and Te

0 20 40 60 80 100 1200.0

1.0x1012

2.0x1012

3.0x1012

4.0x1012

5.0x1012

0% CH4 sim. 0% CH4 exp.



Ele

ctr

on

De

nsity (

cm

-3)

Time (ns)

6

0 10 20 30 40 50 60 70 800

5

10

15

20

25

Ele

ctr

on T

em

pera

ture

(eV

) 0% CH4 sim.

1% CH4 sim.

2% CH4 sim.

0% CH4 exp.

1% CH4 exp.

2% CH4 exp.

Time (ns)

•More CH4 → lower overall ne and Te•Nonlinear decay in ne and Te with CH4 addition•Both trends captured by model

Chen, T.Y., Taneja, T.S., Rousso, A.C., Yang, S., Kolemen, E. and Ju, Y., 2021. Proceedings of the Combustion Institute, 38(4), pp.6533-6540.

•Electron density: area under Thomson scattering spectrum

•Electron temperature: spectral linewidth

•Gaussian scattering lineshape: Maxwellian EEDF

•Raman scattering rotational transitions in N2 used for absolute calibration

Filtered Thomson Scattering:

ne , Te, and EEDF inference

0.0E+00

1.0E+04

2.0E+04

3.0E+04

4.0E+045.0E+04

6.0E+04

7.0E+04

8.0E+04

525 530 535 540

Wavelength (nm)

Gaussian

Fit

Rayleigh

scattering blocked

528 532 5360

10000

20000

30000

40000

50000

60000

Inte

nsity [a.u

.]

Wavelength [nm]

N2 Raman scattering

10 mm

He, 200 Torr

Courtesy of Prof. Igor V. Adamovich

Thomson Scattering SpectraNs pulse discharge in H2-He and O2-He, P=100 Torr

10% O2-He

ne= 1.7·1013 cm-3, Te= 1.6 eV, T=350 K

524 528 532 536 540

0

4000

8000

12000

16000

Inte

nsity [C

oun

ts]

Wavelength [nm]

Thomson signal

Gaussian fit

526 528 530 532 534 536 5380

5000

10000

15000

20000

25000

30000

Inte

nsity [a

.u.]

Wavelength [nm]

Synthetic spectrum

Experiment

5% H2-He,

ne = 1.5∙1014 cm-3, Te = 2.0 eV

Electron Density and Electron TemperatureNs pulse discharge in O2-He

•“Double maxima” in ne, Te : two discharge pulses ≈ 400 ns apart

•Electron temperature in the afterglow Te ≈ 0.3 eV (controlled by superelastic collisions)

•Modeling predictions in good agreement with data

•Measurements in air are more challenging (strong interference from N2, O2 Raman scattering)

0 200 400 600 8000

1

2

3

4 Experimental n

e

Predicted ne

Experimental Te

Predicted Te

Time [ns]

ne [10

14 c

m-3]

0

1

2

3

4

Te [eV

]

10% O2-He

6.1.2 Electric Field Induced Second Harmonic Generation (E-FISH)

•3rd order nonlinear optical process:

𝑃2𝜔 =3

2𝑁𝜒𝑖,𝑗,𝑘,𝑙

3 −2𝜔, 0 ,𝜔, 𝜔 𝐸𝑗𝑎𝑝𝑝𝑙𝑖𝑒𝑑

𝐸𝑘𝜔 𝐸𝑙

𝜔

(𝑃2𝜔)2 ~ 𝐼2𝜔 = 𝐴 ∗ 𝑁2𝐸𝑎𝑝𝑝𝑙𝑖𝑒𝑑2 𝐼𝜔

2

•Note 𝐼2𝜔 is proportional to 𝐸𝑎𝑝𝑝𝑙𝑖𝑒𝑑2 and 𝐼𝜔

2 and A is calibrated by using known applied voltage

•Cannot distinguish between negative and positive polarity by E-FISH alone

Dogariu et al. Phys. Rev. Applied 7, 024024 (2017) 10

E-FISH Experimental Setup

• Discharge: 75 Torr pure Ar, 50 Hz pulse frequency, 20 ns pulse width, 90 msresidence time

• Laser: Ti:Sapphire, 50 fs, 800 ± 15 nm, 6 mJ

• Second harmonic isolated by a prism and a dichroic mirror

11

T.Y. Chen et al. AIAA SciTech (2020)

Combined E-FISH and Thomson scattering in ns-pulsed Ar DBD

0 20 40 60 80 100 120 140 160 180 200

0E+00

1E+12

2E+12

3E+12

4E+12

Electron Number Density

Electric Field

Ele

ctr

on

Num

be

r D

ensity (

cm

-3)

Time (ns)

0

1000

2000

3000

4000

5000

Ele

ctr

ic F

ield

(V

/cm

)

12

Peak at 30 ns in electric field could be from secondary discharge leading to detectable electron densities

0 20 40 60 80 100 120 140 160 180 200

0

1

2

3

4

5

Electron Temperature

Electric Field

Ele

ctr

on

Te

mp

era

ture

(e

V)

Time (ns)

0

1000

2000

3000

4000

5000

Ele

ctr

ic F

ield

(V

/cm

)

Rises in electric field appear to correlate with electron temperature rises but improved signal-to-noise is required

Electric Field Measurements in 2-D Ns Pulse

Discharge in Atmospheric Air

• Ns pulse discharge between a high-voltage electrode and a thin quartz plate

• Discharge gap 0.6 - 1.0 mm, two-dimensional geometry, diffuse plasma

• Time-resolved electric field measured at multiple locations in the discharge gap

-100 -50 0 50 100 150 200 250-7.5

-5.0

-2.5

0.0

2.5

Time [ns]

-7.5

-5.0

-2.5

0.0

2.5

0

1

2

3 Voltage [kV]

Current [A]

Coupled energy [mJ]

Laser beam

locations

Front view, 100 ns gate

Side view, 2 ns gateTop view, 2 ns gate

• Surface ionization wave plasma ~ 200 μm thick, wave speed ~ 0.03 mm/ns

• Electric field measured by picosecond four-wave mixing (calibration by electrostatic field)

• Time resolution 2 ns, spatial resolution across laser beam ~ 100 μm

• Objective: electric field mapping in ns pulse discharges in high-pressure fuel-air mixtures

“Curtain Plasma” Images, Negative Polarity Pulse

Laser beam

locations

Electric Field Vector Components

in a Surface Ionization Wave Discharge

• Initial field offset (at t < 0): charge accumulation on dielectric from previous pulse

• Field follows applied voltage rise, increases until “forward breakdown”

• After breakdown, field reduced due to charge accumulation on dielectric

• Field is reversed after applied voltage starts decreasing

• Away from HV electrode, field peaks later (Ey before Ex): surface ionization wave

• Measurements in a hydrogen-air diffusion flame underway

Laser beam

locations-100 0 100 200 300 400-6

-4

-2

0

2

4

6

8 Voltage

Current

Absolute field

Actual field

Time [ns]

-U [

kV

], I [

A]

-30

-20

-10

0

10

20

30

40

Ele

ctr

ic f

ield

[kV

/cm

]

HV electrode

Forward

breakdown

Reverse

breakdown

-100 -50 0 50 100 150 200

0

5

10

15

20

25

30

- Ex

Ey

(Ex

2 + E

y

2)

1/2

Ele

ctr

ic f

ield

[kV

/cm

]

Time [ns]150 μm from surface

4s

4p

3d

ionization15.75eV

13.3eV

14.7eV

4d

3s23p6↔ 3s23p54S1

10121 llL

0,1,2,... SLSLJ

2

3

1

3

0

3 PP,P :Term

02/12/121 SSS

1,... SLSLJ

1

1P :Term

3s23p6↔ 3s23p54p1

1,02

1

2

1

number) quantum(any 2,1,011

21

21

ssS

llL

3

3

2

3

1

3

2

1

2

3

1

3

0

3

1

1

1

3

0

1 D,D,D,D ,P,P,P,P ,S,S :Term

3,2,1 2,1,0, 1, ,0,1,2,... SLSLJ

S=0 S=1

L-S coupling

,S,P,D ,S ,P,P,P ,D,D,D 0

1

1

1

2

1

1

3

2

3

1

3

0

3

3

3

2

3

1

3

3s23p54p1

6.1.3 Plasma property measurements using H2/Ar emission lines

17

Stark broadening of hydrogen lines and Ar optical emission line-ratio method

Schematic diagram of the experimental setup. The spatially resolved optical

measurement system is shown on the left bottom. On the right bottom is a zoom-in

figure showing the stainless steel needle tips and the discharge gap.Xi-Ming Zhu, James L Walsh, Wen-Cong Chen1 and Yi-Kang Pu, J.

Phys. D: Appl. Phys. 45 (2012) 295201 (11pp)

18

Experimentally measured electron densities in a high-pressure

nanosecond pulsed microplasma (Ar/Ne = 700/30 Torr, discharge

current lasts for about 100 ns, pulse period 1 ms). In the legend on

the right top, ‘line ratio’ refers to the line-ratio method and ‘Stark

broadening’ refers to the Stark broadening method using Ar 696.5

nm, Hα and Hβ lines with a single-Voigt fitting procedure. In the

legend on the left bottom, ‘centre’ and ‘edge’ denote ne,centre and

ne,edge obtained with double-Voigt fitting. The solid line shows a

function, ne = 6 × 1018 × exp(−(t/0.15)0.22), where ne and tare inunits of cm−3 and ns, respectively.

Uncertainties in the ne measurement (%) using the

Stark roadening method with Ar 696.5 nm line, Hα

line and Hβ line (for ne > 1016 cm−3) and that using the line-ratio method.

Xi-Ming Zhu, James L Walsh, Wen-Cong Chen1 and Yi-Kang Pu, J.

Phys. D: Appl. Phys. 45 (2012) 295201 (11pp)

What is hybrid fs/ps CARS?•Use broadband femtosecond pulse for simultaneous acquisition of many Raman transitions and picosecond probe for spectral resolution

•Several advantages:• 𝐼𝐶𝐴𝑅𝑆 ∝ I𝑓𝑠

2 ∗ Ips → ultrafast pulses provide high signal and straightforward spatially-resolved one-dimensional (1-D) imaging

• No need to scan Stokes frequency and only needs 2 beams

• Delayed probe pulse avoids non-resonant background

• Can make near surface measurements (within 50 to 100 𝜇𝑚)

•Would like to make spatially-resolved 1-D measurements in pulsed plasma

𝝎𝒑𝒖𝒎𝒑

𝝎𝑺𝒕𝒐𝒌𝒆𝒔𝝎𝒑𝒓𝒐𝒃𝒆

t (ps)

𝝉𝒑𝒓𝒐𝒃𝒆

Rotational or vibrational energy level

19

fs laserps probe signal

6.2.1 Measurements of temperature, vibrational level populations and fast heating using fs/ps

Measurements of temperature, vibrational level populations

and fast heating using picosecond CARS

• Compression waves formed by “rapid” heating, on sub-acoustic time scale, τacoustic ~ r / a ~ 2 μs

• What processes control other features of temperature rise (e.g. “slow” heating”)?

t= 1-10 μs (frames are 1 μs apart)

Air, P=100 Torr

10

mm

2 mm

ns pulse discharge and afterglow: Air vs. nitrogen, P=100 Torr

• Strong vibrational excitation in the discharge, N2(v=0-8)

• Tv(N2) rise in early afterglow: V-V exchange, N2(v) + N2(v=0) → N2(v-1) + N2(v=1)

• Tv(N2) decay in late afterglow: V-T relaxation, N2(v) + O → N2(v-1) + O , radial diffusion

• “Rapid” heating: quenching of N2 electronic states, N2(C,B,A,a) + O2 → N2(X) + O + O

• “Slow” heating: V-T relaxation, N2(X,v) + O → N2(X,v-1) + O

• “Rapid” heating: pressure overshoot , compression wave formation

• NO formation: dominated by reactions of N2 electronic states, N2* + O → NO + N

Comparison with modeling predictions in air:

vibrational kinetics and temperature rise

A. Montello, Z. Yin, D. Burnette, I.V. Adamovich, and W.R Lempert, Journal of Physics D: Applied Physics 46 (2013) 464002

Single-shot measurement rotational and rovibrational energy distributions

by Hybrid fs/ps coherent anti-Stokes Raman scattering (CARS) spectroscopy

(Four wave mixing and fs broadband dual pumping)

fs ps

ωp1: Rovibrational Raman transition (Q-branch, Δv=+1, ΔJ =0)

ωp2: pure rotational Raman transition (S-branch, Δv=0, ΔJ =+2)

ωprobe: frequency-narrowed ps probe pulse

Dedic, C.E., Meyer, T.R. and Michael, J.B., 2017. Single-shot ultrafast coherent anti-Stokes Raman scattering of vibrational/rotational

nonequilibrium. Optica, 4(5), pp.563-570.

The He∕N2 dielectric barrier discharge

First level vibrational temperature can be measured from the pure rotational CARS spectrum

Increasing vibrational energy level red-shifts the rotational energy level F(v,J)

𝐹 𝑣, 𝐽 = 𝐵𝑣 𝐽 𝐽 + 1 − 𝐷𝑣 𝐽2 𝐽 + 1 2

𝐵𝑣 = 𝐵𝑒 − 𝛼 𝑣 +1

2+ 𝛾𝑒 𝑣 +

1

2

2, 𝐷𝑣 = 𝐷𝑒 + 𝛽𝑒 𝑣 +

1

2 23

Simulated rotational N2 CARS spectraTrot= 400K Tvib = 4000 K Trot= 500K, varied Tvib

More details available inChen T.Y. et al. Opt. Lett. (2020)

Spatially and temporally resolved fs/ps CARS measurements of rotation-vibration non-equilibrium in a CH4/N2 ns-pulsed discharge

Discharge geometry and experimental setup

24

Generates stable thin filament (d ~ 1mm)

• 40% CH4, 60% N2, 60 Torr• 4 kV, 500 ns pulse width, 20

Hz, 8 mm gap, 220 Ω resistor• fs laser: <7 fs, 0.6 mJ

(bandwidth up to CH4 𝜈1 Q-branch at 2916 cm-1)

• ps laser: 65 ps, 7 mJ• Want to measure rotational N2

CARS and vibrational CH4 CARS• All data is time-resolved and

1-D with 40 𝜇𝑚 resolution starting within 150 𝜇𝑚 of cathode

Rotational CARS can easily be extended to 1-D with only two beams for time and spatially resolved measurements

25

• No need for frequency-converting optics (i.e. OPA) for separate Stokes beam [1]• Bandwidth of 100 fs Ti:Sapphire regenerative amplifier enough to resolve

rotations (<200 cm-1)

20 𝜇𝑠 after voltage pulse, 40% CH4/60% N2, 60 Torr

[1] Dedic, C.E. et al., Optica, (2017)

V-T relaxation and peaks in Tvib are immediately obvious from time evolutions of Tvib and Trot

Vibration to rotation/translation relaxation

Vibration to rotation/translation relaxation

26

Tvib is strongly peaked ~ 1 mm from the cathode

Non-equilibrium energy transfer between N2rotational and vibrational energy states.

27

CH4 number density is almost flat with near constant Trot

Time interval with high Tvib

6.2.2 TALIF measurements

• Use two photons to access an electronic level that spontaneously emits a fluorescence photon

• Can access states that would require a vacuum UV photon for single photon LIF

• Fluorescence only generated at laser focus → localized images and high spatial resolution• ITALIF ∝ 𝐼𝑙𝑎𝑠𝑒𝑟

2

• Species that can be measured: O [1], H [2], N [3], CO [4], NH3 [5]

• High sensitivity: O detection limit ~ 1011 –1012 cm-3

Ground state

Electronically excited state

Lower excited state

225.6 nm

225.6 nm

Oxygen atom

2p 3P

3s 3S

2p4 3P

844.6 nm

[1] K. Niemi et al. Plasma Sources Sci. Technol. 2005[2] L. Cherigier et al. J. Appl. Phys. 1999[3] F. Gaboriau et al. J. Phys. D. 2009[4] M.A. Damen et al. Plasma Sources Sci. Technol. 2019[5] D. Zhang et al. Int. J. Hydrogen Energ. 2019 28

fs TALIF can do 2-D imaging of radicals

29

2-D imaging of O-radical in an O2/He atmospheric plasma jet

J.B. Schmidt et al. Plasma Sources Sci. Technol. 2017

Currently have fs laser and filters to do 2-D TALIF

6.2.3 O2(a1Δg) measurements by ICOS cavity

O2 (a1Δg) at 0.98 eV

O2 (b1Σg+) at 1.6 eV

O2 (a1Δg) + H = OH+O fast

O2 + H = OH +O slow

O2(a1Δg) measurements Lifted flame experimental system

microwave

power supply

T3

T2 T1 fuel

oxidizer

φ=1camera

O2

Ar

ignition

system

P1

P2

Lifted Flame

NO

ICOS Cavity

FTIR

C3H8 or C2H4

O3

Absorption

Cell

vacuum

pump

3-way

valve

vacuum

pumpIn

ten

sity

254 nmWavelength

w/o O3

w/ O3

Hg light

emissionlower light

intensity

254nm

Detector

Notch

Filter

Flow

Hg Light

O3

O3

O3

O3

O3

O3

O3

O3 O

3

Off-Axis ICOS Cavity

Diode LaserComputer

PD

LensMirror

LensMirror

Flow Flow

NOx

O3

O2(a1Δg)

6636.16 6636.20 6636.24

Frequency [cm-1]

Cro

ss-Sectio

n

(x1

0-2

3) [cm

2]

Ca

vit

y E

nh

an

ced

Ab

sorp

tio

n (

GA

)

Q(12) Experimental measurement

Q(12) Curve fit

Paramagnetic (radical) species

Absorption

Dispersion

HO2 energy levelsZeeman splitting

6.2.4 HO2/OH using mid-IR Faraday Rotational Spectroscopy

Laser Lock-In Amplifier

+Bfield

0( ) sin 2RMS RMSV GP

Bremfield et al., 2013, JPC letters, 2013; Kurimoto et al. 2014

Experimental results: HO2/OH measurements

Implication:RO2→QOOH→O2QOOH uncertaintyHCO+O2=HO2+CO reaction uncertainty and HCO formation pathway?

Signal

DME flow reactormodel validation

Sensitivity OHHO2

3 detection limit = 20 ppbv / 𝐻𝑧 detection limit 1 ppmv / 𝐻𝑧

6.3. Experimental measurements of Plasma chemistry and Kinetic Processes

4.5 mm

NRP spark discharge

grounded electrode

Preheated air at 1000 K

NRP discharge in air at 1000 K, 1 atm:

• 10-ns pulse• 5.7 kV• 10 kHz• Gap: 4 mm• 670 mJ/pulse

• Measured quantities:

• O atoms: TALIF with absolute calibration (Xe)

• N2 (A): CRDS

• N2 (B) and N2 (C): OES

• Temperature: OES

• Electron density: Hb Stark broadening

Courtesy of Prof. Christophe Laux

-10 0 10 20 30 40 5010

12

1013

1014

1015

1016

1017

0

1

2

3

4

5

6

0

50

100

150

200

250

300

Time (ns)

N2(A)

N2(C)

Ab

so

lute

de

nsitie

s [

cm

-3]

N2(B)

02.0x10

17

4.0x1017

6.0x1017

8.0x1017

1.0x1018

1.2x1018

O (3P) density

1500

2000

2500 from N

2(C-B)

from N2(B-A)

Te

mp

era

ture

[K

]

Temperature

0

10

20

30

40

Cu

rre

nt [A

]

Vo

lta

ge

(V

)

V Iconduction

E/N

[T

d]

Measurements of V, I, temperature, densities

hheating =21±5%

hdiss. = 35±5%

Ultrafast heating:900 K in 20 ns

Ultrafast

dissociation of O2

Rusterholtz et al, J. Phys.D, 46, 464010, Dec 2013

Summary of processes involved in flame stabilization by NRP discharges

e-

N2(A)N2(B)N2(C)

O2N2(X)

N2(X) + 2 O + E

T

Thermal effects

O22 O

Chemical effects: RH + O R + OH

Oxidation

5 μs after pulse

(Xu et al., APL. 99, 121502, 2011)

2-step mechanism (Popov, 2001):

N2 + e → N2* + e (N2* = N2 A, B, C, …) Thresholds: 6.2, 7.4, 11.0 eV

N2* + O2 → N2 + O + O + T T = 1.0, 2.2, 5.9 eV

Measurements of Chemical Processes in Plasma Assisted Ignition and Combustion

0.30 0.31 0.32 0.33 0.34 0.35 0.36

150

225

300

375

450

525

600

no plasma

with plasma (f=5 kHz)

with plasma (f=20 kHz)Ex

tin

ctio

n s

tra

in r

ate

(1

/s)

Fuel mole fraction Xf

1.0E-7 1.0E-6 1.0E-5 1.0E-4 1.0E-3 1.0E-2

0.0E+0

1.0E-5

2.0E-5

3.0E-5

4.0E-5

5.0E-5

Time, seconds

O atom mole fraction

Air

Air-ethylene, =0.5

J. Uddi et al. 2009

Atomic O production

O (3P)

W. Sun et. Al. 2010

O atom measurements by using TALIF

O atom formation in a plasma discharge of air and air-C2H4 mixture in a flow reactor

O atom formation in a ns plasmadischarge of methane/aircounterflow flames

Extension of extinction limit byplasma discharge

OH measurements in a flow reactor:Plasma chemical reactions result in ignition

H2 – air, ϕ=0.3T0=500 K, P=100 torr

C2H4 – air, ϕ=0.3T0=500 K, P=100 torr

Pulse #10 Pulse #100

End View

Long burst: plasma assisted ignition, Tignition ≈ 700 K < Tauto-ignition ≈ 900 K

H2-air, ϕ=0.4

50 pulses

Short burst: OH transient rise and decay

Top burner

Bottom burner (fuel)

OH

fluorescence

at Q1(6)

Direct image

OH PLIF measurements in Dimethyl ether (DME) Ignition

OH density vs. fuel mole fraction XO = 0.55, P = 72 Torr, f = 24 kHz, for DME (a = 250 1/s) and CH4 (a = 400 1/s,) as the fuel,

respectively (solid square symbols: increasing XF, open square symbols: decreasing XF)

What is the role of plasma before ignition of DME?

Flame

S-shaped ignition and extinction curvesDME vs. CH4

A schematic of the key reaction pathways for high pressure fuel

oxidation of at different temperatures

(blue arrow: Below 700K; yellow arrow: 700-1050 K;

red arrow: above 1050K).

Fuel(RH)

+OH

R+O2RO2

QOOH

O2QOOH

HO2

H2O2

2OH

Small

alkeneC2H3/CH2O

H/HCO+O2+(M)

Plasmae, R*, N2*, O2*R(*), R(v), N2(v), O2(v)

+O2

+O2

CO/CO2

e +O2=O+O(1D) +e

H+O2(1Δg) =O+OH

O(1D)+RH =OH+R

N2(A,B,C)+O2=O+O+N2

N2(v)+HO2 =OH+O+N2

R(v,*)+O2=RO+OH

=???

O3+M =O+O2+M

Slow

6.4 Multispecies diagnostics and kinetic modeling

e +Ar =Ar*+e

e +O2 =O+O(1D)

e +O2 =e + O2(v)

e +O2 =O2-

e +C7H16 =H+R

......

Question to aks:

How does the key plasma reactions affects n-heptane dissociation and oxidation

in the first 10 ms with efifferent excitation processes involving Ar and O2?

Ar* +C7H16 =?

O(1D) +C7H16 =?

O2(v) +C7H16 =?

O2(v) +C7H15 =RO2(v)

......

42

6. 4.1 Measurements of elementary rates of O(1D) reaction with fuel: ethanol

C2H5OH + O(1D) → CH3CHOH + OH (a)

CH2CH2OH + OH (b)

H abstraction CH3CH2O + OH (c)

O(1D) insertion CH3O + CH2OH (d)

HO2/OH time history using FRS

43

C2H5OH + O(1D) → CH3CHOH + OH 0.44 (a)

CH2CH2OH + OH (b)

H abstraction CH3CH2O + OH (c)

O(1D) insertion CH3O + CH2OH 0.42 (d)

0.14

Reaction Branching ratio

(b)Measured OH/HO2 time history and branching ratios

Hongtao Zhong, Chao Yan,Chu C. Teng, Timothy Y. Chen, Gerard Wysocki, and Yiguang Ju, IJCK, 2021

Results:O(1D) insertion is about 42%

OH HO2

Comparison of thermal (1, 10 atm) and plasma (1 atm)

propane fuel consumption as a function of temperature.

Nicholas Tsolas, Jong Guen Lee and Richard A. Yetter, 2015, Phil. Trans. R. Soc. A 373: 20140344.

6.4.2 Flow reactor studies of plasma assistedlow temperature reaction

45

6.4.3 O3 assisted low temperature fuel oxidation

Extreme Low Temperature Combustion (ELTC): O3 + Methyl Hexanoate

PI-MBMS+JSR (Sandia collaboration)

ELTCC7H14O2

•O3 induces extreme low temperature oxidation for a long carbon chain fuel below 550 K.

Rousso, A.C., Jasper, A.W., Ju, Y. and Hansen, N., The Journal of Physical

Chemistry A 2020 124 (48), 9897-9914

46

400 600 800 1000-0.002

-0.001

0.000

0.001

0.002

0.003

0 ppm

300 ppm

LLNL/NOx

Fu

el

mo

le f

rac

tio

n

Temperature [K]

nC12

H26

= 1.3

Exp.

n-Dodecane/NOx kinetic coupling

RO2 + NO = RO + NO2

RO2 → QOOH → O2QOOH → 2OHX

NO+HO2 = OH+NO2

CH2O+NO2=HCO+HONOHONO + M = OH + NO + M

6.4.4 Effects of NOx on low temperature fuel oxidation: n-dodecane

JSR/1atm

Zhou, M., Yehia, O.R., Reuter, C.B., Burger, C.M., Murakami, Y., Zhao, H. and Ju, Y., 2020. Proc Combustion Institute, Vol 38.

•NOx inhibits low temperature fueloxidation, but suppresses NTC effectand enhances intermediatetemperature fuel oxidation

Ge Etalon

Reactor

Collimating

Lenses

Mirror

Flip Mirror

Quartz

Wall

Macor

Wall

Vacuum

Chamber

6.4.5 Multispecies measurements for kinetic model development

47

• Reactor

• Pressure: 60 Torr

• Initial Temperature: 300 K

• Flow speed: 40 cm/s

• Reactor size: 45 x 14 x 152 mm3

• Nanosecond repetitively pulsed discharge: FID GmbH FPG 30-50MC4

• Peak Voltage: ~7 kV

• Pulse Duration: 12 ns FWHM

• Continuous mode: 0 – 5 kHz

• Pulse burst mode: 150 pulses, 30 kHz

• Quartz double dielectric barrier: 1.6 mm thickness

• Gap distance: 14 mm

Mini-Herriott cell showing 24 pass

configuration

Experimental Apparatus

Vacuum Chamber

Detector

Electrode Connection

Alignment Laser

QCL Laser

N2 Purge Box

Laser inlet purge tube

To Vacuum

Observation Window

Direct and ICCD Images of Plasma Discharge in a Reactor

Stoichiometric mixtures: C2H4/O2 with 75% AR, 60 Torr, Vmax= 6 kV

•Direct Image: 1 kHz, 3.6 mJ/pulse, 2 s exposure time.•ICCD images: Gate time = 100 ns, Gain = 250

1000 Hz

2000 Hz

3000 Hz

1000 HzDirect

ICCD

Cathode

Anode


• Nanosecond-pulsed discharge

• Mixtures:

Pyrolysis：0.01 C12H26/0.99 N2

Oxidation: 0.01 C12H26/0.19 O2/0.80 N2

• Pressure: 30 Torr

• Flow rate: 0.5 m/s

Methods Species

Time-dependent

(TDLAS)

n-dodecane, CH4, C2H2, H2O, CH2O

Temperature

Steady-Sate

(GC)

H2, CH4, C2H2, C2H4, C2H6,

C3H4, C3H6, C4H8, C5H10

H2O, CO, CO2, CH2O

Steady-Sate

(DM-FRS) NO Experimental setup

Plasma assisted low temperature n-dodecane oxidation

Species diagnostics


6.5 Kinetic model development

Plasma-assisted n-dodecane/air combustion mechanism

N and NOx with fuel and fuel radicals[3-5]

n-dodecane mechanism: reduced from Cai’s model[1]

by Princeton CHEM-RC

C0-C2 chemistry: HP-Mech[2]

[1] L. Cai et al., Combust. Flame 173 (2016) 468–482.

[2] Princeton HP-Mech. http://engine. princeton.edu/mechanism/HP-Mech.html, 2017.

[3] H. Zhao et al., Combust. Flame 197 (2018) 78–87.

[4] P. Gokulakrishnan et al. , J. Eng. Gas Turb. Power 140 (4) (2018).

[5] J. Anderlohr et al., Combust. Flame 156 (2) (2009) 505–521.

Plasma kinetics

O2(𝑎1Δg), O2(b

1Σg+), O2

*, O(1D), O(1S),

N2(A), N2(B), N2(a’), N2(C), N(2D), N2+, O2

+, e

• 198 species

• Plasma skeletal mechanism: 239 reactions

Combustion reduced mechanism: 1128 reactions

Combustion kinetics:

Species Reactions

N2(A)

(6.17-7.8 eV)

(up to two C-C bonds)

N2(A) + C12H26 → N2 + C12H25-1 + H

N2 + C10H21-1 + C2H5

N2 + C8H17-1 + cC3H6 + CH3

…

N2(a’)

(8.4-8.89 eV)

(up to two C-C

or C-H bonds)

N2(a’) + C12H26 → N2 + C10H21-1 + C2H4 + H

N2 + C10H21-1 + CH3 + CH2

…

O(1D) O(1D) + C12H26 → C12H25-1 + OH

C10H21-1 + CH3 + CH2O

…

O(1S) O(1S) + C12H26 → C12H25-1 + OH

C10H21-1 + CH3 + CH2O

…

Excited species + fuel dissociation reactions


Time-dependent voltage and temperature

30 kHz, 600 pulses in pyrolysis case, 450 pulses in oxidation case

• The peak E/N in this study is ~ 400-600 Td

Applied voltage in the simulation


Plasma effect on NO formation

NO production in 0.80 N2/0.20 O2 and 0.01 C12H26/0.19 O2/0.80 N2

• NO increases with frequencies in N2/O2

• NO decreases dramatically with 1% n-dodecane.

• RO2 + NO = RO + NO2

• NO pathways in N2/O2

N + O2 = NO + O

N(2D) + O2 → NO + O/O(1D)

NO2 + O = O2 + NO

• Strong NOx/fuel kinetic coupling in PAC

Ammonia/air flames with and without

plasma: NOx emissions without and

with plasma ( φ= 0.94)

W. Sun et al,, Combustion and Flame 228 (2021) 430–432

X. Mao et al., Proc. Combut. Inst. 38, 2021


• In the oxidation condition, under-prediction of fuel oxidation, H2O, and CH2O formation

n-dodecane density CH4 and C2H2 density in the pyrolysis case H2O and CH2O density in the oxidation case

Kinetic model validation: species time histories in

n-dodecane pyrolysis/oxidation

• In the pyrolysis case, the model agrees well with the experimental data


Kinetic model validation: Species in steady state measurements

• In oxidation, primary products are

well-predicted

• C0–C5 hydrocarbons are under-

predicted

• The uncertainty of electron impact

reactions with n-dodecane at fuel rich

conditions

e + C12H26 → e + 2pC4H9 + 2C2H4

….

0.01 C12H26/0.99 N20.01 C12H26/0.19 O2/0.80 N2

• In pyrolysis, predicted well with the

measurements

Andrey Starikovskiy and Nikolay Aleksandrov, AIAA paper-2017-1977

Cross-sections database available for electron-molecule collisions

A new reaction pathway of plasma assisted low temperature combustionvia excited RO2*(v)

Vibrational and electronically excited O2(v) collides with fuel radical (R) forming highly energized RO2* in which the vibrational energies are quickly redistributed due to strong coupling between different vibration states. These RO2*, comparing to those formed by ground state oxygen with R, carry much higher internal energies that enable them overcome the barrier TS1, TS2, and TS3 much easier to produce the bimolecular product HO2+Alkene and OH+Ether. Therefore, the ignition processes/species profiles of the mixture can be significantly different from the ground state system.

R+O2

RO2OH + RO (cyc-Ether)

HO2+alkene

TS1

QOOH

TS2

TS3

R+O2*(v,e)

EnergizedRO2*(v)

• Increase the rate• Change branching ratios (TSi)•Modify pressure dependenceR(v)*+O2

CH2O + OH + R”

Non-equilibrium plasma kinetics

Rousso, Aric, Suo Yang, Joseph Lefkowitz, Wenting Sun, and Yiguang Ju. " Proceedings of the Combustion Institute, Vo.36, 2017, Pages 4105–4112

• Time-resolved, spatially-resolved, in-situ laser diagnostics of electric field, electron density, and electron temperature, excited and radical species greatly enhanced the understanding of plasma kinetic and chemical process in PAC.

• Production of O and O(1D), O2(singlet), N2(*), and N2(v) by the plasma is the major processes in kinetically enhancement of combustion.

• Fast and slow heating in PAC is important, but the energy transfer processes are very complicated.

• Plasma activated low temperature combustion pathways and enable cool flame formation, but existing mechanisms have large uncertainties, especially for large hydrocarbons.

• Electron impact reaction cross-sections of large alkanes and reaction rates involving O(1D) and non-equilibrium excitations are poorly known.

• The effect of vibrational species excitation on PAC is still poorly known.

Summary

Date post:	19-Mar-2023
Category:	Documents
Upload:	khangminh22
View:	2 times
Download:	0 times

Lecture 4 Electric Field Effect on Flames: Ionic wind and Joule ...

Documents