detonation initiation and propagation in non-uniform ...yju/2nd-near-limit-flames... · combustion...

Detonation initiation and propagation in non-uniform gaseous mixture

2nd International Workshop on Near-Limit Flames Beijing, China, July 27-28, 2019

Peng Dai1, Zheng Chen2

1 Southern University of Science and Technology, China 2 College of Engineering, Peking University, China

2

Background

(Wang et al. 2015, 2017)

Autoignition in boosted SIEs Reactivity non-uniformity at high T & P End-gas autoignition/detonation Knock & Super knock Conventional knock Super knock

(Qi et al. 2015)

(Wang et al. 2015)

Combustion in SIEs

Normal combustion

Abnormal combustion

Turbulent flame

Autoignition

Detonation

3

Autoignition with reactivity non-uniformity

Reactivity gradient (Zel’dovich), SWACER (Lee) Ignition delay:

Reaction front speed, u vs. Sound speed, a

Detonation development: u~a

( ), , ,ig ig RT P Yτ τ φ= Ignites first Reaction front

87654321 igigigigigigigig ττττττττ <<<<<<<

11||

−

− ∇∂

∂=∇= T

Tu ig

ig

ττ

Ta ~

4

Analysis/simulation on hot spot Zel’dovich 1980; Kapilar & Dold 1989; He & Clavin 1992, 1994; Weber 1994; Short 1997; Khokhlov et al. 1997; Kapila et al. 2002; Libermann et al. 2002; Radulescu et al. 2013; Ju 2017; Im 2019… Hot spot/compositional stratification Different autoignition modes

Detonation & engine knock Bradley et al. 2002; Kalghatgi et al. 2012; Bates & Bradley 2016, 2017 Regime map in terms of ξ and ε ξ : non-dimensional T gradient ε : (r0/a)/τe time-scale ratio

Detonation peninsular & engine knock

Different modes of reaction front propagation

5

Kalghatgi & Bradley 2012, 2015; Bates & Bradley 2016, 2017

N2 K2

S

Detonation and engine knock

However, this detonation regime map was based on H2/CO. Can it be used for other fuels ? Is it affected by the initial temperature, pressure, equivalence ratio? What is the role of chemistry (low T chem. for large fuels) ? How about CO2 dilution, with NOx … ?

6

Objectives

To obtain detonation development regimes for fuels other than syngas To examine the effects of initial temperature, pressure and

equivalence ratio on detonation development To assess the influence of low-temperature chemistry, CO2 dilution,

and NOx addition on detonation development

inside NTC regime

outside NTC regime

7

Solver: A-SURF (Adaptive Simulation of Unsteady Reacting Flow) Unsteady, compressible flow Finite volume method MUSCL scheme for convection Central-difference for diffusion Dynamically adaptive mesh refinement

Mesh size = ∆x0/2Level , 10 levels Finest mesh size: 1 micro-meter

Point-implicit method for stiff ODEs Time step ~ 0.4 nano-second

Chemical kinetics Chemical mechanisms for n-heptane (44 species) and DME (55, 113 species) These mechanisms well predict low-temperature ignition.

X (cm)

Tem

pera

ture

(K)

Mes

hLe

vel

0 0.05 0.1 0.15 0.2 0.25260

280

300

320

340

360

380

400

420

-6

-4

-2

0

2

4

6expansion fan contcat line shock

shock

contact line

expansion fan

Numerical method

8

Solver: A-SURF (Adaptive Simulation of Unsteady Reacting Flow) Unsteady, compressible flow Finite volume method MUSCL scheme for convection Central-difference for diffusion Dynamically adaptive mesh refinement

Mesh size = ∆x0/2Level , 10 levels Finest mesh size: 1 micro-meter

Point-implicit method for stiff ODEs Time step ~ 0.4 nano-second

Chemical kinetics Chemical mechanisms for n-heptane (44 species) and DME (55, 113 species) These mechanisms well predict low-temperature ignition.

Numerical method

9

Different modes: (Zeldovich 1980, Bartenev & Gelfand, 2000, Bradley 2002…) I, Thermal explosion (homogeneous ignition): u=∞ (▽T=0) II, Supersonic reaction front: u >> a >> Su (small ▽T) III, Detonation development: u ~ a >> Su

IV, Subsonic reaction front: a > u >> Su

V, Deflagration/flame: a >> u ~ Su (large▽T)

0 0

c

dT dT adr dr

uξ = =

Different modes of reaction front propagation

0

e

ra

ετ

=

Excitation time

inside NTC regime

outside NTC regime

10

2015PCI: autoignition and detonation from a cold spot, nC7H16

2015CNF: multi-stage heat release, nC7H16

2017PCI: different initial temperatures, CH3OCH3

2017PCI: concentration gradient, nC7H16

2019PCI: effects of NOx, CH3OCH3

2019CNF: fuel lean and CO2 dilution, nC7H16

n-heptane, C7H16 DME, CH3OCH3

Our work

11







Our work

ε

ξ

0 5 10 15 200

20

40

60

80

100

I: supersonic autoignitivedeflagration

II: detonationIII: shock-detonationIV: shock-deflagration

1mm

1mm2mm

5mm x0=6.1mm

2mm 4.1mm x0=5mm

Circle:T0=900KDelta: T0=780K

0.5mm3mm

3mm0.5mm

I II

III

IV

12

Our work







13







Our work

ε

ξ

0 5 10 15 20 250

5

10

nC7H16, T0=802 K

DMET0=1035 K

DME, T0=982 K

DME, T0=802 K

B

A

n-heptane, C7H16

DME, CH3OCH3

14







Our work

15







Our work

16

Effects of NOx addition

intake Exhaust EGR valve NOX

CO2 H2O O2 N2 …

Exhaust gas recirculation (EGR)

T0 (K)

(dT/

dr) c

(K/m

m)

700 800 900 1000 1100 1200-5

0

5

cNO=0 ppmcNO=100 ppmcNO=500 ppm

time

(ms)

0

0.1

0.2

0.3

0.4

0.5

τ

τ1

φ=1.0, P0=40 atm

(a)

(b)

NO greatly promotes OH production and ignition 1st stage: HO2+NO=NO2+OH, DME+OH/H → CH3OCH2 → Low-T chemistry 2nd stage: NO2+CH3=NO+CH3O , NO2+H=NO+OH, HO2+NO=NO2+OH

Local NOx accumulation can induce autoignitive front propagation

(Dai and Chen, 2019 PCI)

( ) ( )( ) 1

NO NOcdc dr a d dcτ

−=

17


T0 (K)

(dT/

dr) c

(K/m

m)

700 800 900 1000 1100 1200-5

0

5

cNO=0 ppmcNO=100 ppmcNO=500 ppm

time

(ms)

0

0.1

0.2

0.3

0.4

0.5

τ

τ1

φ=1.0, P0=40 atm

(a)

(b)

NO greatly promotes OH production and ignition 1st stage: HO2+NO=NO2+OH , DME+OH/H → CH3OCH2 → Low-T chemistry 2nd stage: NO2+CH3=NO+CH3O , NO2+H=NO+OH, HO2+NO=NO2+OH



18


ε

ξ

0 5 10 150

1

2

3

4

5

6

7

8

9 NO concentration gradientc0=0 ppmc0=100 ppmc0=500 ppmc0=1000 ppm

temperature gradientc0=0 ppmc0=100 ppm

subsonic reactionfront propagation

supersonic reactionfront propagation

detonationdevelopment

(b)

NO greatly promotes OH production and ignition 1st stage: HO2+NO=NO2+OH , DME+OH/H → CH3OCH2 → Low-T chemistry 2nd stage: NO2+CH3=NO+CH3O , NO2+H=NO+OH, HO2+NO=NO2+OH



( ) ( )( ) ( )

0

0

, /2

, /2

i c r

NO NOi c r

dT dr dT drau dc dr dc dr

ξ= =

19


ε

ξ

0 5 10 150

1

2

3

4

5

6

7

8

9 NO concentration gradientc0=0 ppmc0=100 ppmc0=500 ppmc0=1000 ppm

temperature gradientc0=0 ppmc0=100 ppm




(b)

ε

ξ a

0 5 10 150

0.5

1

1.5

2

c0=0 ppmc0=100 ppmc0=500 ppmc0=1000 ppm

c0=0 ppmc0=100 ppm

temperature gradient



NO concentration gradient


ξa,l

ξa,u

(a)

In ξ-ε diagram, detonation regime strongly depends on NO addition level and spot type (NO spot, cold spot)

In ξa-ε diagram, detonation regimes are close to each other


0 /2r ra

average

aS

ξ ==( ) ( )( ) ( )

0

0

, /2

, /2

i c r

NO NOi c r

dT dr dT drau dc dr dc dr

ξ= =

20

Fuel lean or CO2 dilution

φ

time

(s)

(dT/

dr) c

(K/m

m)

0.4 0.6 0.8 110-7

10-6

10-5

10-4

10-3

-0.5

-0.45

-0.4

-0.35

-0.3

-0.25

-0.2

-0.15

(a) cCO2=0

τ

τe

(dT/dr)c

cCO2

time

(s)

(dT/

dr) c

(K/m

m)

0 0.1 0.2 0.3 0.4 0.510-7

10-6

10-5

10-4

10-3

-0.5

-0.45

-0.4

-0.35

-0.3

-0.25

-0.2

-0.15

-0.1τ

τe

(b) φ=1.0

(dT/dr)c

Fuel lean CO2 dilution

Both τ and τe increase with decreasing φ or increasing cCO2 τe is much more sensitive and it can be enlarged by 100 times Heat release can be significantly mitigated by fuel lean or CO2 dilution

(Dai, Chen, Gan, 2019 CNF)

21

Fuel lean or CO2 dilution (Dai, Chen, Gan, 2019 CNF)

φ

ξ a

0.5 0.6 0.7 0.8 0.9 10

0.5

1

1.5

2

2.5

3r0=5mmr0=3.5mmr0=2mm

(I)

(II)

(III)

cCO2

ξ a

0 0.05 0.1 0.15 0.2 0.250

0.5

1

1.5

2

2.5

3

r0=5mmr0=3.5mmr0=2mm

(I)

(II)

(III)

Fuel lean CO2 dilution

(I) supersonic reaction front propagation, (II) detonation development, and (III) subsonic reaction front propagation

Fuel lean or CO2 dilution great reduces the propensity of detonation development

22

ε

ξ

0 5 10 15 20 250

2

4

6

8

10

(b) (I)

(II)

(III)

DME/air, cold spot [32]φ=1.0, cNO=0φ=1.0, cNO=100ppm

n-heptane/air, hot spotφ=1.0, cCO2=0φ=0.7, cCO2=0φ=1.0, cCO2=0.13

ε

ξ a

0 5 10 15 20 250

0.5

1

1.5

2

2.5

(a) (I)

(II)

(III)

DME/air, cold spot [32]φ=1.0, cNO=0φ=1.0, cNO=100ppm

φ=1.0, cCO2=0

φ=1.0, cNO=0φ=1.0, cNO=0

n-heptane/air, hot spot

φ=0.7, cCO2=0φ=1.0, cCO2=0.13

0 0

c

dT dT a udr dr

ξ = =

In ξ-ε diagram, detonation regime strongly depends on equivalence ratio, NO addition and CO2 dilution

In ξa-ε diagram, detonation regimes are close to each other

Fuel lean or CO2 dilution (Dai, Chen, Gan, 2019 CNF)

0 /2r ra

average

aS

ξ ==

23

Summary

Detonation development regime depends on fuel type initial temperature/pressure/equivalence ratio/additives/dilution…

Proper parameters to quantify detonation development regime ?

Excitation time is an important parameter which is difficult to be measured in experiments or accurately predicted.

Chemistry (Low-T chemistry, NOx) greatly affects the detonation development due to reactivity gradient

0 0 ,c

dT dT a udr dr

ξ = =

0 /2 ... ?r ra

average

aS

ξ ==,/

e

aurτ

ξε 0=0 ,e

ra

ετ

=

Natural Science Foundation of China

24

Acknowledgments

Thank you for your attention!

detonation initiation and propagation in non-uniform ...yju/2nd-near-limit-flames... · combustion...

Documents