Fundamental Mechanisms, Predictive Modeling, and Novel ... · Flat flame burner inside a six-arm...

Nonequilibrium Thermodynamics Laboratories The Ohio State University

Fundamental Mechanisms, Predictive Modeling, and Novel Aerospace Applicationsand Novel Aerospace Applications

of Plasma Assisted Combustion

Overview of OSU research plan

Walter Lempert, Igor Adamovich, J. William Rich, and Jeffrey SuttonWalter Lempert, Igor Adamovich, J. William Rich, and Jeffrey Sutton

MURI Kick Off MeetingMURI Kick-Off MeetingNovember 4, 2009

Thrust 1. Experimental studies of nonequilibrium air-fuel plasma kinetics using advanced non-intrusive diagnostics


p g g

Task 1: Low-to-Moderate (T=300-800 K) temperature, spatial and time-Task 1: Low to Moderate (T 300 800 K) temperature, spatial and timedependent radical species concentration and temperaturemeasurements in nanosecond pulse plasmas in a variety of fuel-airmixtures pressures (P=0.1 - 5 atm), and equivalence ratios (φ~0.1-3.0)p ( ), q (φ )

Goal: Generate an extensive set of experimental data on radical speciesp pconcentrations and temperature rise; elucidate kinetic mechanisms oflow-temperature plasma chemical fuel oxidation and ignition usingkinetic modeling. Bridge the gap between room-temperature data(low-pressure gas discharges) and high-temperature data (shock tubes)

Test Bed #1: High-temperature, high-pressure nanosecond pulse discharge cell


Quartz channelTube furnace

nanosecond pulse discharge cell

Quartz channel(open on both sides)

Tube furnace enclosure Flow outHigh-pressure cell

Optical access Flat mirror with a hole for LIF pump laser beam

Copper / alumina ceramic high voltage electrode block

Flow inOptical access

window (mirror)

High-pressure discharge cell inside a tube furnace (6 inch bore, up to T=12000 C)High pressure discharge cell inside a tube furnace (6 inch bore, up to T 1200 C)Premixed fuel-air flow (~1 m/s), preheated in the furnace, from 0.1 atm to a few atmRepetitive nanosecond pulse discharge plasma: 20-40 kV, 5-25 nsec, 10 Hz to 100 kHzOptical access (LIF, TALIF, CARS, CRDS) on the sidesFuels: hydrogen, methane, ethylene, propane, pentane, methanol & ethanol vapor

Repetitive nanosecond pulse plasma for kinetic studies: Air P=60 torr ν=40 kHz 40 msec burst 1 μsec gate


Air, P=60 torr, ν=40 kHz, 40 msec burst, 1 μsec gate

20Voltage, kV

Gate 1 Voltage, kV

10

20(plasma):

1 µsec Gate 2 (flame): 18 µsec

• Some filamentarystructure in pulses#1 and #210

20

-10

0 • Uniform airplasma duringsubsequent pulses,at P 40 100 torr

-10

0

Air, P=40 torr

0 10 20 30 40 50 60 70Time μsec

-20

at P=40-100 torr

0 50 100 150 200 250

-20

Time nsec

grounded

floating

Time, μsecTime, nsec

Repetitive nanosecond pulse plasma for kinetic studies: Ethylene-air, P=40 torr, φ=1, ν=40 kHz


Ethylene air, P 40 torr, φ 1, ν 40 kHz

• Nearly uniform plasmaduring entire burst (exceptpulses #1 and #2)p )

• Ignition does not occur,likely due to rapid wallcoolingg

• Pressure is low – can thisexperiment be done at higherpressures?p

Repetitive nanosecond pulse plasma for kinetic studies: Ethylene-air, P=60 torr, φ=1, ν=40 kHz


• Uniform plasma during first few

Ethylene air, P 60 torr, φ 1, ν 40 kHz

p gtens of pulses (except pulses #1and #2)

• Well-defined filaments form inpulse #100, persist for severalhundred pulses

• After ignition occurs, flame fillsg ,entire discharge volume, andplasma becomes uniform again

• Filamentation likely due toyionization / heating instability

• This is unacceptable: need to keep thel if d i ti b tplasma uniform during entire burst

• We know that preheating will improveplasma uniformity

• Sustaining plasma in a heated cell will allowmeasurements at higher pressures

Time-resolved species concentrations: O and H atoms(Two-Photon Absorption LIF with Xe and Kr calibration)


Previous results: O atoms in air,O atom mole fraction7 0E 5

O atom mole fraction

(Two Photon Absorption LIF with Xe and Kr calibration)

Previous results: O atoms in air,methane-air, and ethylene-air atP=60 torr (single-pulse and burstmode, initially at T=300 K)3.0E-5

4.0E-5

5.0E-5Air

Air-methane, Φ=1.0

4.0E-5

5.0E-5

6.0E-5

7.0E-5Air

Air-ethylene, Φ=0.5

, y )

Objective: measure time-resolvedO and H atoms in nsec pulse0.0E+0

1.0E-5

2.0E-5

0.0E+0

1.0E-5

2.0E-5

3.0E-5

O and H atoms in nsec pulsedischarge plasmas in H2-air andCxHy air mixtures, at P ~ 0.1 - 1atm, T=300-800 K

0.0E+0 1.0E-3 2.0E-3 3.0E-3 4.0E-3

Time, seconds1.0E-7 1.0E-6 1.0E-5 1.0E-4 1.0E-3 1.0E-2

Time, seconds

[O], cm-30 20 40 60 80 100

Number of pulses ,

Outcome: kinetic mechanism oflow-temperature plasma fuel1E+15

1E+16

P=60 torr, ν=100 kHz

0 20 40 60 80 100

low temperature plasma fueldissociation and oxidation(specifically rates of O atomgeneration in the plasma and O

1E+14

air

air/CH4, φ=1.0

air/C2H4, φ=0.5

g patom reactions with fuel species)

0.0 0.2 0.4 0.6 0.8 1.01E+13

Time, msec

Time-resolved species concentrations: OH(LIF with Hencken adiabatic burner calibration)


(LIF with Hencken adiabatic burner calibration)

6 0E 5

O atom mole fraction7 0E 5


Work currently underway: OH inmethane-air and ethylene-air atP=60 torr (single-pulse and burst

d i i i ll T 300 K)3 0E-5

4.0E-5

5.0E-5

6.0E-5

Air

Air-methane, Φ=1.0

4.0E-5

5.0E-5

6.0E-5

7.0E-5Air


mode, initially at T=300 K)

Objective: measure time resolved0.0E+0

1.0E-5

2.0E-5

3.0E 5

OH0.0E+0

1.0E-5

2.0E-5

3.0E-5

OH

Objective: measure time-resolvedOH in nsec pulse dischargeplasmas in H2-air and CxHy airmixtures at P ~ 0 1 - 1 atm

0.0E+0 1.0E-3 2.0E-3 3.0E-3 4.0E-3


Time, seconds

2 0E 4OH mole fraction

mixtures, at P ~ 0.1 - 1 atm,T=300-800 K

1.5E-4

2.0E-4


OH, φ=0.1

Ethylene-airAll fuel oxidized

Outcome: kinetic mechanism oflow-temperature plasma fueloxidation (specifically rates of H

5.0E-5

1.0E-4 OH, φ=1.0

atom abstraction from fuelspecies)0 5 10 15 20 25

0.0E+0

Time, msec

Time-resolved species concentrations: NO(LIF with calibration using known NO-N2 mixture)


(LIF with calibration using known NO N2 mixture)

Mole fractions

Previous results : NO in air, methane-air and ethylene-air at P=60 torr(single-pulse, initially at T=300 K).St t f th t ki ti d l t

1.0E-5

1.0E-4

O

NO

O

NO

State-of-the-art kinetic models cannotexplain time-resolved data. Possibleeffect of N2(X,v) + O reaction.

1.0E-6

N

O3

N2(A)

O2(b)

NO2

Objective: measure time-resolved NOin nsec pulse discharge plasmas in H2-

1E-6 1E-5 1E-4 1E-3 1E-2 1E-1 1E+01.0E-7

Time, seconds

N2(X,v)

NO mole fraction in nsec pulse discharge plasmas in H2air and CxHy air mixtures, at P ~ 0.1 - 1atm, T=300-800 K4.0E-6

5.0E-6

NO mole fraction

Air

CH4-air, φ=0.5

C2H4-air, φ=1.0

Outcome: kinetic mechanism of low-temperature plasma fuel oxidation

1.0E-6

2.0E-6

3.0E-6

(specifically O2 dissociation vs. NOformation in N2* reactions)1.0E-6 1.0E-5 1.0E-4 1.0E-3 1.0E-2

Time, seconds

Time-resolved, spatially resolved temperature(purely rotational CARS)


(purely rotational CARS)

Previous results: time-resolvedtemperature in air andethylene-air at P=40 torr (burst

d i iti ll t T 300 K)0.6

0.8

1Ethylene-AirAir

mode, initially at T=300 K).Evidence of significantadditional heat release in fuel-air compared to air

0.2

0.4

air, compared to air

Objective: measuretemperature in nsec pulsedi h l i H i d

050 100 150 200 250 300

Raman Shift (1/cm)

T, 0Cdischarge plasmas in H2-air andCxHy air mixtures, at P ~ 0.1 - 1atm, T=300-800 K

500

600

700

800


Air

Air, model

Outcome: kinetic mechanism oflow-temperature plasmachemical energy release in

100

200

300

400

,

Fuel-air, φ=0.1

Fuel-air, φ=0.1, model

Fuel-air, φ=1.0


exothermic fuel oxidationreactions with radicals0 5 10 15

0

Time, msec

Test Bed #2: Flat flame McKenna burner with nanosecond pulse discharge


with nanosecond pulse discharge

Flame

15-15mm

HV electrode

Wire mesh HVE{~1-5 mm

Burner surface

{McKenna

burnerMeasurement locations (resolution < 200 μm)

Flat flame burner inside a six-arm cross vacuum chamber (8 inch bore)Premixed fuel-air flow (~0.1-1.0 m/s) with N2 co-flow, P=10-40 torrRepetitive nanosecond pulse discharge plasma: 20-40 kV, 5-25 nsec, 10 Hz to 100 kHzOptical access (LIF, TALIF, CRDS) on two perpendicular axesOptical access (LIF, TALIF, CRDS) on two perpendicular axesFuels: hydrogen, methane, ethylene, propane, pentane, methanol & ethanol vapor

Laboratory for Advanced Fluid Dynamics and Combustion Research

Interaction of plasma and flame chemistry: spatially resolved species concentrations and temperature


spatially resolved species concentrations and temperature

Steady laminar low pressure flat flames allow spatially resolvedSteady, laminar, low-pressure flat flames allow spatially-resolvedmeasurements of temperature and species concentrations

Minimize transport influence; isolate kinetic effects

Can investigate full range of temperature conditions (from below 500 K to2000 K) by adjusting measurement position (i.e. height above burner)

Typical spatial scale ~5-20 mm, spatial resolution <200 µm

Straightforward integration of nsec discharge plasma into a low-pressureflame facility and study of plasma effects (i.e. measurements with plasmaflame facility and study of plasma effects (i.e. measurements with plasma“off” and “on”)

Steady laminar 30 Torr 1 D flame


Steady, laminar, 30 Torr, 1-D flame

Previous low-pressure flame results (LIF):P=10 40 torr; CH C H C H C H ; φ=0 6 1 4


P=10-40 torr; CH4, C2H6, C3H8, C4H10; φ=0.6 -1.4

oscilloscope computer

Flame temperature from rotational structureof OH A-X (1,0) band near 282 nm

1

PMTPhotodiode

Vacuumchamber

Nd:YAG355/532 nm

Dye Laserλ/2

0.4

0.6

0.8

Rel

ativ

e LI

F Si

gnal

Q21

R1

R2

5.5 8.5

5.5

8.5

10.5

12.52.5

5.5

pinhole355/532 nm

Spectral features used for profiles of flame species:

CH A X (0 0) t 435 NO A X (0 0) t 226

0

0.2

281.0 281.2 281.4 281.6 281.8 282.0 282.2 282.4

R

Excitation Wavelength (nm)

CH A-X (0,0) at 435 nm NO A-X (0,0) at 226 nm

-22

-21

))

0 75

1.00

P1(7.5)CH LIF

0 75

1.00

P2(20.5)

NO LIF

-24

-23

ln(I/

B(2J

+1

-1/kBT

0.25

0.50

0.75

0.25

0.50

0.75


-250 1000 2000 3000 4000

Rotational Energy (cm-1)

0.00434.0 434.5 435.0 435.5


0.00226.2 226.3 226.4 226.5 226.6


Previous low-pressure flame results (LIF):P=10 40 torr; CH C H C H C H ; φ=0 6 1 4

Nonequilibrium Thermodynamics Laboratories The Ohio State University0.015

φ = 1 28

CH4 C2H6 C3H8 C4H10

P=10-40 torr; CH4, C2H6, C3H8, C4H10; φ=0.6 -1.4

0.000

0.005

0.010

XO

H

Current Exp.Calc.

φ = 1.28

Spatially-resolved measurementsof radicals to understand high-temperature flame chemistry,

0.005

0.010

0.015

XO

H

φ = 1.07

p yhelp kinetic model development

Kinetic modeling: GRI-Mech 3.0OH

0.0 0.5 1.0 1.5 2.0Height Above Burner (cm)

0.0000.0 0.5 1.0 1.5 2.0

Height Above Burner (cm)0.0 0.5 1.0 1.5 2.0

Height Above Burner (cm)0.0 0.5 1.0 1.5 2.0

Height Above Burner (cm)

40Current Exp.

CH4 C2H6 C3H8 C4H10

OH

CH

0

10

20

30

X CH (p

pm) Calc.

φ = 1.28We will look at the regionupstream of the flame where

li b t l

CH

5

10

15

X CH (p

pm)

φ = 1.07

coupling between plasmakinetics and flame chemistry ismost important



0

X




Low-pressure flame / plasma measurements (LIF CRDS)


Objective: Examine coupling of plasma and combustion kinetics in a 1 D low

(LIF, CRDS)

Objective: Examine coupling of plasma and combustion kinetics in a 1-D low-pressure flame. Use spatially-resolved species concentration and temperaturemeasurements by LIF (OH, H, O, and CH) and CRDS (HO2, HCO, CH3) tostudy the effect of quasi-steady (RF) and repetitively pulsed nsec dischargestudy the effect of quasi steady (RF) and repetitively pulsed nsec dischargeplasmas on low-temperature chemistry and coupling with the flame zone

Outcome: Kinetic mechanism of low-temperature plasma chemical fuelp poxidation and energy release, and its effect on flame speed and burn rate.Specifically, boundary between “low-T” and “high-T” chemistry bymeasuring HO2 radical concentration, at the conditions when O2 iselectronically excited

O2 + H → OH + O (high temperatures)

O2 + H +M → HO2 + M (low temperatures)

CRDS diagnostics will be used in both “test bed” experiments, (I) high-T,


CRDS diagnostics will be used in both test bed experiments, (I) high T,high-P nsec discharge plasma cell, and (II) low-P flame / plasma cell

Thrust 2. Kinetic model development and validation


Task 8: Development and validation of a predictive kinetic model of non-equilibrium plasma fuel oxidation and ignition, usingequilibrium plasma fuel oxidation and ignition, usingexperimental results of Thrust 1

Goal: Identify key mechanisms, reaction, and rates of plasma chemicalfuel oxidation processes for a wide range of fuels, pressures,temperatures and equivalence ratios This is absolutely essentialtemperatures, and equivalence ratios. This is absolutely essentialto predictive capability of the model.

Current state of the art: hydrocarbon-air, low-temperature plasma chemistry kinetic model


low temperature plasma chemistry kinetic model

• Air plasma model: equations for ground state species (N N O O O NO• Air plasma model: equations for ground state species (N, N2, O, O2, O3, NO,NO2, N2O), charged species (electrons and ions), and excited species(N2(A3Σ), N2(B3Π), N2(C3Π), N2(a'1Σ), O2(a1Δ), O2(b1Σ), O2(c1Σ), N(2D),N(2P), O(1D)) produced in the plasma.

• Two-term expansion Boltzmann equation for plasma electrons

• Fuel-air plasma: model combined with GRI Mech 3.0 CxHy oxidationp x ymechanisms, supplemented with fuel dissociation by electron impact and inreactions with electronically excited nitrogen

• Peak E/N adjusted for pulse energy to be same as predicted by thePeak E/N adjusted for pulse energy to be same as predicted by thenanosecond pulse discharge model

We have absolutely no reason to trust the model predictions: GRI Mech 3.0y p(or any other combustion mechanism) is not designed to work at lowtemperatures (starting at T=300 K)

Confidence in the model can be provided only by detailed kineticmeasurements such as discussed in Thrust 1 plan

Here is what we know so far: dominant radical and energy release processes in C2H4-air predicted by the model


Fuel energy releaseO atom generation

p 2 4 p y

O + CH2CHO = H + CH2 + CO2H + O2 + M = HO2 + MO + HO2 = OH + O2OH + HO2 = O2 + H2OOH + C H C H + H O

N2 + e- = N2(A3Σ) + e-

N2 + e- = N2(B3Π) + e-

N2 + e- = N2(C3Π) + e-

N2 + e- = N2(a'1Σ) + e-

OH + C2H4 = C2H3 + H2OHO2 + CH3 = OH + CH3OCH3O + O2 = HO2 + CH2OO2 + CH2CHO = OH + HCO + HCOHCO + O2 = HO2 + CO

O2 + e- = O(3P) + O(3P,1D) + e-

N2(C3Π) + O2 = N2 (a'1Σ) + O2N2(a'1Σ) + O2 = N2 (B3Π) + O2N2(B3Π) + O2 = N2 (A3Σ) + O2

3 HCO + O2 HO2 + COHO2 + HO2 = O2 + H2O2CH2 + O2 = H + H + CO2

N2(A3Σ) + O2 = N2 + O + OFuel dissociation

C2H4 + e- = products + e-N2(A3Σ) + C2H4 = N2 + C2H3 + H

1 5E 3

Pulse energy balance, J

N2(B3Π) + C2H4 = N2 + C2H3 + HN2(C3Π) + C2H4 = N2 + C2H3 + HN2(a'1Σ) + C2H4 = N2 + C2H3 + H

O atom decay 1.0E-3

1.5E-3Input energy

Heat, air

Heat, C2H4-air

O + C2H4 = CH3 + HCOO + C2H4 = H + CH2CHOC2H3 + O2 = HCO + CH2OC2H3 + O2 = O + CH2CHO

5.0E-4

O + O2 + M = O3 + MO + O3 = O2 + O2

1.0E-7 1.0E-6 1.0E-5 1.0E-4 1.0E-3 1.0E-2 1.0E-10.0E+0

Time, seconds

Model validation summary: so far so good…but no surprise if the model fails at some point


O atom mole fraction O atom mole fraction [O], cm-3 Number of pulses

… but no surprise if the model fails at some point

4.0E-5

5.0E-5


Air

Air-methane, Φ=1.05.0E-5

6.0E-5

7.0E-5

O atom mole fractionAir


1E+15

1E+16

[ ],


0 20 40 60 80 100

1.0E-5

2.0E-5

3.0E-5

1.0E-5

2.0E-5

3.0E-5

4.0E-5

1E+14

air

air/CH4, φ=1.0

air/C2H4, φ=0.5

0.0E+0 1.0E-3 2.0E-3 3.0E-3 4.0E-30.0E+0


0.0E+0

Time, seconds0.0 0.2 0.4 0.6 0.8 1.0

1E+13

Time, msec

T, 0C

500

600

700

800


Air

Air, model

Need a lot more data fromThrust 1 for extensive modelvalidation

100

200

300

400

,

Fuel-air, φ=0.1


Fuel-air, φ=1.0


validation

Outcome: a self-consistentlow-temperature fuel-air

l h i l h i0 5 10 15

0

Time, msec

plasma chemical mechanism

Thrust 3. Experimental and modeling studies of fundamental nonequilibrium discharge processes


of fundamental nonequilibrium discharge processes

Task 10: Characterization and modeling of nsec pulse dischargesTask 10: Characterization and modeling of nsec pulse discharges

Goal: Prediction of E/N and electron density in the plasma individualGoal: Prediction of E/N and electron density in the plasma, individualpulse energy coupled to the plasma, and their scaling withpressure, temperature, pulse waveform, and mixturecompositioncomposition

Two-pronged approach to plasma assisted ignition modeling


to plasma assisted ignition modeling

Predictive modeling of energy release rate and ignition delay time in low-Predictive modeling of energy release rate and ignition delay time in lowtemperature, repetitive nanosecond pulse fuel-air plasmas requires:

• E/N in the plasma, individual pulse energy coupled to the plasma, and their scaling withpressure temperature pulse waveform and mixture compositionpressure, temperature, pulse waveform, and mixture composition

• Air plasma and fuel-air plasma chemistry: reactions among ground state species, excitedspecies and radicals generated in the plasma, and their effect on energy release rate

These two problems require separate analysis:

• Nsec pulse plasma / sheath models cannot incorporate detailed reactive plasmachemistry: too many species ( 100) and reactions ( 1 000)chemistry: too many species (~100) and reactions (~1,000)

• Detailed plasma chemistry models (quasi-neutral) cannot incorporate repetitive, nsectime scale sheath dynamics and plasma shielding

Approach:

• Predict plasma E/N and coupled pulse energy using nsec pulse plasma / sheath model

• Incorporate results into fuel-air plasma chemistry model

Previous results: Repetitive nsec discharge pulse energy measurements


Repetitive nsec discharge pulse energy measurements

1

2

3[×106]

5.0

10.0

15.0

wer

[W]

ergy

[mJ]

⊗Flow into the page

Pulsed electrodes

-1

0

-5.0

0Pow

Ene

30

40 100

Flow into the page

Alumina ceramic plates

0

10

20

30

0

50

Vol

tage

[kV

]

Cur

rent

[A]

-50 0 50 100 150-10 -50

Time[ns]

Nitrogen, P=350 torr, ν=100 kHzNitrogen, P=300 torr, ν=100 kHz

Nitrogen, P 350 torr, ν 100 kHz 0.3 seconds after start (pulse # 30,000)

Pulse energy 11 mJ/pulse

Nitrogen, P=650 torr, ν=100 kHz Discharge power 110 W

What are the electric field and the electron density?

Previous results: Analytic nsec pulse discharge plasma / sheath model


L

Analytic nsec pulse discharge plasma / sheath model

• Equations for electron and ion number density

plasma

th d d

Equations for electron and ion number density• Poisson equation for the electric field• Plane-to-plane discharge geometry• Voltage pulse: Gaussian fit to experimental waveform

l

sheath

cathode anode• Voltage pulse: Gaussian fit to experimental waveform• Dielectric plate charging / plasma shielding

Analytic solution: time-dependent electron density and electric

l,ε

lsE

field in the plasma, coupled pulse energyExcellent agreement with numerical solution, experimental data

Plasma electric field, kV/cmnumerical model

analytic solution

14161820 applied voltage

1.5E+12

2.0E+12

Electron density, cm-3

shielded plasma

Vapp/(L+2l/ε)

468

101214

breakdown

5.0E+11

1.0E+12

numerical model

analytic solution

70 75 80 85 90 95 100Time, nsec

024

shielded plasma

86 88 90 92 940.0E+0

Time, nsec

breakdown

Previous results: Analytic nsec pulse discharge plasma / sheath model


Power density, kW/cm3 Coupled pulse energy, mJ

Analytic nsec pulse discharge plasma / sheath model

y,

numerical model

analytic solution40

50

breakdown1.5

2.0

p p gy,

P*=100 torrvalue inferred from

TALIF O atom measurements

20

30

Vapp/(L+2l/ε)

breakdown

hi ld d l

1.0P*=60 torr

P*=80 torr

80 90 100 110 1200

10shielded plasma

10 15 20 25 300.0

0.5P*=40 torr

80 90 100 110 120Time, nsec

10 15 20 25 30

Peak voltage, kV

⎥⎥⎤

⎢⎢⎡

+⎟⎟⎞

⎜⎜⎛

≈+= kl dftb kt t lVVCQQQ π21

2

02

• Coupled pulse energy scales with the number density, can be increased by increasing peak voltage reducing pulse duration

⎥⎦

⎢⎣

+⎟⎠

⎜⎝

+pulseRCpeak

peakloadafterbreaktotal VVCQQQ

τν2

peak voltage, reducing pulse duration

• Excellent agreement with numerical solution, experimental data

Electric field and electron density measurements: CARS, Thomson scattering


CARS, Thomson scattering

Rectangular cross section quartz channel

Quartz window at a Brewster angle

Quartz channel with a MgF2 window at a Brewster angleg Brewster angle

Flow Plasma

Entire test section mounted on a translation stage forspatially resolved measurements.

Objective: measure time- and space-resolved electric field and electron density innsec pulse discharge plasmas using psec CARS and Thomson scattering;comparison with the modelp

Outcome: predictive capability for electron impact kinetic processes in the plasma

Thrust 4. Studies of diffusion and transport of active species in representative 2-D reacting flow geometries


species in representative 2 D reacting flow geometries

Task 12: Ignition and flameholding in noneq ilibri m plasma ca it flo sTask 12: Ignition and flameholding in nonequilibrium plasma cavity flowsat low static temperatures

Goal: Determine viable approaches to flameholding in high-speed flowsusing low-temperature plasmas. We simply cannot process theentire flow with the plasma!

Previous results: cavity ignition in premixed ethylene-air flows by nsec plasma (25 kV, 20 nsec, 40 kHz)


flows by nsec plasma (25 kV, 20 nsec, 40 kHz)

Optical access windowsPressure tap

To vacuum and FTIR

p

Main flow

Injection flow

High voltage electrode block

Ceramic plates

Fuel-air, 200 torr, 50 m/s

Pulse #140 Pulse #180 Pulse #350Air, 200 torr, 50 m/s

#140 to #141 #180 to #181 #350 to #351

Pulse #100 Pulse #800

Fuel-air, 150 torr, 25 m/sPulse #400

Air, 150 torr, 25 m/s Diffuse plasma in air, filamentation in fuel-air during ignition, diffuse plasma after ignition

Previous results: cavity ignition and flameholding in premixed and non-premixed ethylene-air flows by nsec plasma

Nonequilibrium Thermodynamics Laboratories The Ohio State UniversityOptical access windows

Pressure tap

premixed and non premixed ethylene air flows by nsec plasma

Intensity (arbitrary units) OH emission

To vacuum and FTIR

p

Main flow

y ( y )Non-premixed flow, 100 m/sec

High voltage electrode block

Ceramic plates

0 0 0 2 0 4 0 6 0 8 1 0

Premixed flow, 100 m/sec

0.0 0.2 0.4 0.6 0.8 1.0

Time (sec)

• Ignition and stable flameholding in bothIgnition and stable flameholding in both premixed and non-premixed flows up to 100 m/sec (global φ=1 in both cases)

• 80 90% burned fuel fraction• 80-90% burned fuel fraction

• Plasma power ~100 W, combustion energy release 35 kW

Fuel-air, 175 torr, 85 m/s • After ignition, plasma needs to be “on” at all times (flame blow-off without plasma)

Ignition and flameholding in nonequilibrium plasma cavity flows


in nonequilibrium plasma cavity flows

Objectives:

• F rther st dies of ca it ignition and flameholding b repetiti e nsec• Further studies of cavity ignition and flameholding by repetitive nsecpulse plasmas in fuel injection flows (hydrogen and hydrocarbons)

• High frame rate (10-20 kHz) NO and OH PLIF imaging of ignitiong ( ) g g gprocess using burst mode laser

• Increasing flow velocity beyond 100 m/sec, operating at low globali l ti ( 0 1 0 2)equivalence ratios (φ=0.1-0.2)

• Comparison with kinetic modeling calculations using reducedplasma chemical ignition mechanism. Plasma flameholdingplasma chemical ignition mechanism. Plasma flameholdingmechanism after ignition – thermal or not?

Outcome: Demonstration of true predictive capability of the model

Fundamental Mechanisms, Predictive Modeling, and Novel ... · Flat flame burner inside a six-arm...

Documents

Transcript of Fundamental Mechanisms, Predictive Modeling, and Novel ... · Flat flame burner inside a six-arm...