From Experiment to CFD;Ongoing Work at the Combustion Chemistry Centre

1

Kenji Yasunaga

第48回燃焼シンポジウムワークショップ「ユニバーサル燃焼反応モデルの可能性を探る」スライド

Work areas

2

� Experiments

� Modelling and simulation

� Quantum chemistry

Experimental facilities

3

� Old

� Shock tube #1

� Low pressure, unheated, ...

� Rapid compression machine #1

� Twin piston, well understood, ...

� New

� Shock tube #2

� Hi-pressure, heated

Results from ST#1 (Montréal 2008)

4

� 2,5-dimethylfuran vs n-butanol vs ethanol� 0.75% fuel: 2.25%O2, balance Ar� Reflected shock pressure 1.0 atm

5.0 5.5 6.0 6.5 7.0 7.5

100

1000

Ignition delay time , τ / µs

10000 K / T

Unpublished 2,5-dimethylfuran

5

4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0

100

1000

T/K

0.75% dimethylfuran, 1 atm

φφφφ = 2.5 (Black)

φφφφ = 2.5 (Yasunaga)

φφφφ = 2.0 (Conroy)

φφφφ = 1.0 (Black)

φφφφ = 1.0 (Yasunaga)

φφφφ = 0.5 (Yasunaga)

ττ ττ / µµ µµ

s

104 K / T

2400 2200 2000 1800 1600 1400

Old RCM (Darren Healy)

6

� 5.1% CH4, 1.09%C2H6, 1.09% C3H8, 19.47% O2 and 73.25% diluent, φ = 0.5 in “air”.

� Combust. & Flame 2008, 155, 441; ibid. 451. 153, 316

� ST 8 bar� ST 30 bar� RCM 20 bar� RCM 20 bar� RCM 30 bar

0.6 0.8 1.0 1.2 1.4

0.01

0.1

1

10

100

ττ ττ / ms

1000 K / T

-20.0 0.0 20.0 40.0 60.0 80.0

0

10

20

30Pressure / atm

Time / ms

Low-T chemistry (Peter O’Toole)80:20 CH4:CH3OCH3 φ=0.5, Tc=715K O2/N2

7

Natural gas/DME blends [60%CH4:40% DME]

Left: φ=0.3 Right: φ=1.0

8

0.6 0.8 1.0 1.2 1.4 1.6

0.1

1

10

Ignition delay time / ms

1000 K / T

0.6 0.8 1.0 1.2 1.4 1.6

0.1

1

10

100

Ignition delay time / ms

1000 K / T

ST � 7 � 30 RCM � 7 � 15 � 30

CFD of unique machine (Judith Würmel)

Poor crevice Good crevice

9

New ST (Colin Tobin, Christine Conroy)

10

� Diameter: 6.3 cm Driver: 3 m Test: 5.73 m � Reflected shock p5 70 bar Initial T1 135˚C

New ST (Kenji Yasunaga)

11

0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5

0.1

1

10

100

Iso-octane oxidation, φ = 1.0 in 'air'Ignition delay time / m

s

1000 K / T

13 bar Fieweger

17 bar Fieweger

34 bar Fieweger

40 bar Fieweger

45 bar Fieweger

15 bar NUIG

34 bar NUIG

11.5 bar Sung

15.5 bar Sung

12 bar Minetti

15 bar Minetti

14 bar Vanhove

20 bar NUIG Yasunaga

Modelling (Zeynep Serinyel, Sinéad Burke)

12

� Natural gas C0 to C5 chemistry + biofuel blends� High-pressures & low temperatures

� High molecular weight HCs� Low volatility ⇔ heated STs, RCMs,

� Bio-fuels � Structural features R-C(O)OR’� Methyl formate HC(O)OCH3

� Higher alcohols � n-butanol, …

� Next-generation biofuels� 2,5-Dimethylfuran

Mechanism reduction [I. Gy. Zsély & Tibor Nagy]

Natural gas mechanism (NUIG C4_49) 230 species, 1327 rxns

13

5 10 15 20 25 30 35

0

10

20

30

40

50

60

70

80

90

100

110

Ignition delay

Pressure profile

Maximum relative error / %

Number of species in the reduced mechanisms

90% methane and 10% propane

Lean and stoichiometric

Gas turbine relevant conditions

Temperatures: 876–1,465 K

Pressures: 10–40 atm

Grouping of 30 experiments to

form 6 regimes

Quantum chemistryEnols C=C–OH

14

� Only one stable enol known� Detected in CH2=CH2 flames Taatjes 2005

� Astrochemists 2001� Mechanism: >C=C< + •OH� Ubiquitous ― O- and N-cpds.

� Exhaust gases? � Ethenol ⇒ shortfall in RCOOH in urban air� Archibald 2007 assumed 2:1 CH3CHO:CH2=CH(OH)

� In alcohols?� Iso-butanol 60:40 ethanal:ethenol ratio Qi et al. 2007

H• + CH3CHO � CH3•CHOH � CH2=CH(OH) + H•

� We compute a 3.1 ratio at 1,000K� J. Phys. Chem. A (2009) 113(27): 7834-7845.

Methyl formate decomposition

15

� Multilevel methods: CBS-QB3 and APNO, G3� Zero-point corrected electronic energies (0K)� Barriers to reaction & reaction enthalpy

� HCOOCH3 � CO + CH3OH285.5 ± 4.1 kJ/mol

� HCOOCH3 � 2CH2O321.2 ± 1.8 kJ/mol

� HCOOCH3 � CO2 + CH4

348.3 ± 3.5 kJ/mol

0.3 0.6 0.9 1.2 1.5 1.8 2.110

-26

10-21

10-16

10-11

10-6

10-1

104

109

CO + CH3OH

2 CH2O

CO2 + CH

4

k / s

-1

103 K / T

HCOOCH3 � CO + CH3OH [Wayne Metcalfe]Time-dependent solution of master equation

16

0.5 0.6 0.7 0.8 0.9 1.010

-2

100

102

104

106

108

High pressure limit

100 atm

10 atm

1 atm

0.1 atm

0.01 atm

k / s

-1

103 K / T

Kinetics: CH3COOH ⇒ CH4 + CO2

17

TS ~ TS for HCOOCH3 ⇒ CH4 + CO2

0.6 0.9 1.2 1.510

-9

10-7

10-5

10-3

10-1

101

103

105

Mackie et al.

Blake et al. ('68)

Blake et al. ('69)

Bamford et al.

This Study

k / s

-1

103 K / T

H-abstraction by •OH, HO2•, et cetera

18

� From hydrocarbons� Primary� Secondary� Tertiary

� How does O-atom influence abstraction?� Alcohols

� n-Butanol [HO2• J. Comp. Chem. in press]

� Ethers: DME, EME, iPME� Energetics: Coupled cluster in the basis set limit� Kinetics: RRKM, VRC-TST, Eckart tunneling

� Furans

•OH + Dimethyl ether [Chong-Wen Zhou]k / cm3 mol–1 s–1 = 2.74 T 3.94 exp(1,534/T)

19

1.867

1.0891.097

105.0

1.383

1.180

160.5

Reactant Complex (RC) TS1 (out) TS2 (in) Product Complex (PC)

1.293

1.199

105.1

2.037

100.7

2.450

DME + OHRC

Products

0.0

-5.1

-0.3

-24.8

-21.9

-4.1

0.84

-24.4

-22.1

G3MP2BH&H

3.0

2.8

CCSD(T)/CBS

-4.7

-0.13.5

-24.8

-22.5

TS2

TS1

PC

G3

0.5 1.0 1.5 2.01E11

1E12

1E13

1E14 Arif

Bonard

Wu

Ogura

CCSD(T)/CBS

G3

G3MP2BH&H

Curran

Cook

1000/T

k (cm3mol-1s-1)

EME + OH

PRC

-5.3

-2.0

-26.6

-23.1

2.7

-0.2

3.7

1.4

-15.8

-22.0

TS2b-2

TS2aTS2b-1

TS2c-2

TS2c-1

PC2a

P2a + H2O

P2b + H2O

P2c + H2O

-25.4PC2b

-20.0PC2c

-6.7

-1.5

-26.0

-23.2

-0.4

3.5

1.2

3.1

-24.6

-21.7

-19.2

-14.7

G3

G3MP2BH&H

0.0

•OH + CH3CH2OCH3

20

Pre-Reaction Complex (PRC) TS2a TS2b-1

TS2b-2 TS2c-1 TS2c-2

109.91.869

1.408

1.172 1.393

1.177

1.295

1.199

105.0

156.0

1.205

1.292

1.316

1.193 1.510

106.4

1.979 122.2 1.995

2.545

107.11.933

2.519

114.3

2.446

2.148

CH3CHOCH3 + H2O CH3CH2OCH2 + H2O CH2CH2OCH3 + H2O

k(total) / cm3 mol–1 s–1 = 20.9 T 3.61 exp(2,060/T)Branching CH3CH2OCH3 versus CH3CH2OCH3 CH2/OCH3 = 3.4 – 0.43 ln T +1,120/T

21

0.5 1.0 1.5 2.0

1E11

1E12

1E13

1E14

k(CH3)in

k(OCH3)in

1000 K/T

k / cm3mol-1s-1

k(CH3)out

k(OCH3)out

k(CH2)

k (total)

Alkyl furans

22

Quite different to toluene, xylenes, ..� Very strong ring-carbon hydrogen bonds

� 505 vs 472 (benzene)

� Very strong ring-carbon methyl bonds� 480 vs 427 (toluene)

� Weak Ring-CH2—H bonds� 360 vs 376 (toluene)

� Extremely weak ring-carbon-O—H bond� 268 vs 366 (HOO―H) or 206 (•OO―H)

� J Phys Chem A 2009, 113, 5128O O HH3C

Why 2,5-dimethylfuran?

23

Next-generation bio-fuel� Production

� From biomass [Nature 2007]

� Properties� RON 119, water insoluble

� Chemistry unknown

� ST pyrolysis (Lifshitz et al.)

� Rich flames by VUV-MBMS (Wu et al. C&F 2009, 156, 1365)

� Methyl radical to 2-methylnaphthalene(!)

� ST ignition delays (Black et al. 32nd Combustion Symposium)

� JSR oxidation (Dagaut & Togbé; CNRS Orléans)

•OH abstraction barriers

24

Pre-reaction complex

25

H-abstraction by •H, HO2•, O, O2, …

26

Zero-point corrected electronic energies

� From -CH3 ring-H

� OH 16 35

� H 29 88

� HO2 57 130

� O 12 65

� O2 148 260

TS < ∆∆∆∆E(rxn)

Initial decompositionsVasiliou et al. J Phys Chem A 2009, 113, 8540

27

Flow reactor 1,600K; acetylene, ketene, CO & propyne

272

289

288339

117

170

More

28

126

46

160

280

102

Not detected by Wu

Major projects (from September 2009)

29

� National� Science Foundation Ireland� Higher Education Authority� IRCSET

� International� FP7 EU Project: Gas Turbine Network

� Industrial� North American gas turbine manufacturer� French automotive manufacturer� Middle Eastern oil company

29

30

Henry Curran

John Simmie