Professional Documents
Culture Documents
CriticalFuelPropertiesandBehavior
Jay Keller,
Sandia National Laboratories
Keynote Lecture presented at:
Workshop on
Compressed Natural Gas and Hydrogen Fuels:
Lessons Learned for the Safe Deployment of Vehicles
6/15-19/2009; 1
Sandia National
Laboratories
Hydrogen Behavior
Myth Busting
Jay Keller,
Sandia National Laboratories
Topical Lecture
Progress in Hydrogen Safety: International Short
Course Series
June 15-19, 2009
6/15-19/2009; 2
Sandia National
Laboratories
HydrogenMyths
D Hydrogen Molecular Diffusivity is 3.8 times that of CH4
Therefore it diffuses rapidly and mitigates any hazard
6/15-19/2009; 3
Sandia National
Laboratories
HydrogenMyths
D Hydrogen hazards can be compared favorably to
experiences with other hydrocarbon fuels
Less dangerous than gasoline, methane
Sandia National
Laboratories
HydrogenMyths
D Hydrogen Molecular Diffusivity is 3.8 times that of CH4
Therefore it diffuses rapidly and mitigates any hazard
6/15-19/2009; 5
Sandia National
Laboratories
MomentumDominatedJetsare
withintheIgnitionRegion
Unignited Jet Separation Distance Length Scales
Pressure = ~20 MPa (~3000 psig)
Hole
Diameter
1.5875 mm (1/16 inch)
0.794 mm (1/32 inch)
Flowrate
(2,463 ft3/min)*
2.430x10-2 Kg/sec
(615.9 ft3/min)*
6.075x10-3 Kg/sec
(154.1 ft3/min)*
Xmax - Distance to
4% mole fraction
Start of
Intermediate Region (Buoyancy)
H2 Mole Fraction
Sandia National
Laboratories
SmallUnignitedReleases:
MomentumDominatedRegime
Data for round turbulent jets
1/CL
D In momentum-dominated
regime, the centerline decay
rate follows a 1/ CL
dependence for all gases.
D The mole fraction centerline
decay rate increases with
increasing molecular weight.
Sandia National
Laboratories
HydrogenMyths
D Hydrogen Molecular Diffusivity is 3.8 times that of CH4
Therefore it diffuses rapidly and mitigates any hazard
6/15-19/2009; 8
Sandia National
Laboratories
Buoyancyeffectsare
characterizedbyFroudenumber
Fr=152
Fr=152
Fr=152
D Time-averaged H2 mole
fraction distributions.
0.8
Mole Fraction
Fr=99
Fr=99
Fr=99
0.6
D Froude number is a
measure of strength of
momentum force
relative to the buoyant
force
0.4
Fr=268
Fr=268
Fr=268
6/15-19/2009; 9
0.2
Influenceofbuoyancyis
quantifiedbytheFroudenumber
D Jets from choked flows (Mach 1.0) are
typically momentum-dominated
(Pupstream/Pdownstream>~2).
D Lower source pressures or very large
pressure losses through cracks lead to
subsonic, buoyancy-dominated plumes.
Simulation of H2 Leak
Frden = 1000
Y/D
g
0.08 m.f.
0.07 m.f.
X/D
Simulation of H2 Leak
Frden = 100
Y/D
Transition
Buoyancy
Dominated
X/D
Ricou and Spalding entrainment law (J.
Fluid Mechanics, 11, 1961)
Sandia National
Laboratories
SmallUnignitedReleases:
BuoyancyEffects
1/CL
6/15-19/2009; 11
Fr
Fr
Choked&UnchokedFlows
at20SCFM
Tank Pressure = 3000 psig, Hole Dia. = 0.297 mm
Exit Mach Number = 1.0 (Choked Flow)
H2 Mole Fraction
Fr ~ O(104)
R(m)
0.2
- 0.2
0
0.5
1.0
X(m)
1.5
0.5
6/15-19/2009; 12
1.0
1.5
2.0
X(m)
Red 10.4%
Orange 8.5%
Green 5.1%
Blue 2.6%
*(Chen and Rodi, 1980)
Sandia National
Laboratories
HydrogenMyths
D Hydrogen Molecular Diffusivity is 3.8 times that of CH4
Therefore it diffuses rapidly and mitigates any hazard
6/15-19/2009; 13
Sandia National
Laboratories
FlammabilityLimits
forH2
Upward Flame Propagation
Tube Dimensions,
cm
Diameter Length
7.5
150
5.3
150
5.3
150
5.3
150
5.0
150
5.0
150
4.8
150
4.5
80
4.5
80
Firing
end
Limits, percent
6/15-19/2009; 14
Reference
Lower
Higher
Closed
4.15
75.0
Half-saturated
356
Open
4.19
74.0
Dried
94
4.12
74.2
94
4.17
74.8
94
Closed
4.15
74.5
Half-saturated
356
Tube Dimensions,
Limits, percent
Firing
Water Vapor
Reference
Open
4.00
72.0end Dried
133 Content
cm
Diameter
4.00Length 73.8
Lower
38
Higher
Closed
4.10
----
56
7.5
150
Closed
6.5
----Half-saturated
356
3.90
----
57
5.0
150
6.7
----
356
2.5
150
7.15
----
356
2.5
150
Open
6.2
----Saturated
271
2.5
----- Capacity,
cc
----71.4
------273 Referen
Firing
Limits, percent
Water Vapor
Reference
0.9
150
6.7
65.7
Saturated
276
end
Lower
Higher Content
Downward Flame
Propagation
Tube Dimensions,
cm
Diameter Length
21.0
31
8.0
37
7.5
150
7.0
150
6.2
33
6.0
120
Water Vapor
Content
Firing
end
Open
Closed
Open
statedVapor Closed
9.2
Limits, percent Not Water
Reference
Not Content
stated
8.5
1,000
8.7
Lower
Higher
810
63 5.0
9.3
---Saturated
350
8.9
68.8
Half-saturated 324 4.6
35
356 9.4
8.8
74.5
----74.5
Saturated
115
8.5
---Partly dried
95
9.45
---
325
9/11/07
---67.5
75.5
73.5
70.3
64.8
Saturated
271
82
95
349
368
297
Sandia National
Laboratories
FlammabilityLimits
forH2
Upward Flame Propagation
Tube Dimensions,
cm
Diameter Length
7.5
150
5.3
150
5.3
150
5.3
150
5.0
150
5.0
150
4.8
150
4.5
80
4.5
80
Firing
end
Limits, percent
Water Vapor
Content
Reference
Lower
Higher
Closed
4.15
75.0
Half-saturated
356
Open
4.19
74.0
Dried
94
4.12
74.2
94
4.17
74.8
94
Closed
4.15
74.5
Half-saturated
356
Tube Dimensions,
Limits, percent
Firing
Water Vapor
Reference
Open
4.00
72.0end Dried
133 Content
cm
Diameter
4.00Length 73.8
Lower
38
Higher
Closed
4.10
----
56
7.5
150
Closed
6.5
----Half-saturated
356
3.90
----
57
5.0
150
6.7
----
356
2.5
150
7.15
----
356
2.5
150
Open
6.2
----Saturated
271
2.5
----- Capacity,
cc
----71.4
------273 Referen
Firing
Limits, percent
Water Vapor
Reference
0.9
150
6.7
65.7
Saturated
276
end
Lower
Higher Content
Flame Propagation
D78 investigationsHorizontal
of hydrogen
flammability
limits were identified between 1920 and 1950.
DHydrogen flammability limits are well
Propagation in a Spherical Vessel
established.
Downward Flame
Propagation
Tube Dimensions,
cm
Diameter Length
21.0
31
8.0
37
7.5
150
7.0
150
6.2
33
6.0
120
6/15-19/2009; 15
Firing
end
Open
Closed
Open
statedVapor Closed
9.2
Limits, percent Not Water
Reference
Not Content
stated
8.5
1,000
8.7
Lower
Higher
810
63 5.0
9.3
---Saturated
350
8.9
68.8
Half-saturated 324 4.6
35
356 9.4
8.8
74.5
----74.5
Saturated
115
8.5
---Partly dried
95
9.45
---
325
9/11/07
---67.5
75.5
73.5
70.3
64.8
Saturated
271
82
95
349
368
297
Sandia National
Laboratories
WhatisaReasonableFlame
StabilizationLimit?
Volume
Fraction
Sandia National
Laboratories
HydrogenMyths
D Hydrogen Molecular Diffusivity is 3.8 times that of CH4
Therefore it diffuses rapidly and mitigates any hazard
6/15-19/2009; 17
Sandia National
Laboratories
Hydrogenjetsandflamesare
similartootherflammablegases
D Fraction of chemical energy
converted to thermal radiation
D Radiation heat flux distribution
D Jet length
6/15-19/2009; 18
Sandia National
Laboratories
H2 FlameRadiation
D Orange emission
due to excited H2O
vapor
D Blue continuum
due to emission
from OH + H =>
H2O + h
D UV emission due
to OH*
D IR emission due to
H2O vibrationrotation bands
Sandia National
Laboratories
Hydrogenjetsandflamesare
similartootherflammablegases
D Fraction of chemical energy
converted to thermal radiation
D Radiation heat flux distribution
D Jet length
6/15-19/2009; 20
Sandia National
Laboratories
ThermalRadiationfrom
HydrogenFlames
D Previous radiation data for
nonsooting CO/H2 and CH4 flames
correlate well with flame residence
time.
D Sandias H2 flame data is a factor of
two lower than the hydrocarbon
flame data.
Sandia National
Laboratories
HydrogenMyths
D Hydrogen hazards can be compared favorably to
experiences with other hydrocarbon fuels
Less dangerous than gasoline, methane
Sandia National
Laboratories
ComparisonsofNGandH2
Behaviors
Comparison of Blow-Off Velocities
for Hydrogen and Natural Gas
Sandia National
Laboratories
SmallUnignitedReleases:
MomentumDominatedRegime
Data for round turbulent jets
1/CL
D In momentum-dominated
regime, the centerline decay
rate follows a 1/CL
dependence for all gases.
D The mole fraction centerline
decay rate increases with
increasing molecular weight.
Sandia National
Laboratories
Unignitedjetconcentration
decaydistancesforNGandH2.
Distance on Jet Centerline to Lower Flammability Limit
for Natural Gas and Hydrogen
Tank Pressure
Hole Diameter
3.175 mm (1/8 inch)
1.587 mm (1/16 inch)
Distance to 5% Mole
Fraction Natural Gas
1.19 m (3.90 ft)
0.59 m (1.93 ft)
6/15-19/2009; 25
Sandia National
Laboratories
60
40
20
20
40
60
80
100
120
140
160
Time(s)
Features: strong transient overextent of the hydrogen cloud (larger than at steady
0.5 m above ground the max LFL extents for both gases become almost equal
26
SmallUnignitedReleases:
IgnitableGasEnvelope
H2 Jet at Re=2,384; Fr = 268
0.6
0.4
Mole Fraction
0.8
D H2 flammability
limits: LFL
4.0%; RFR 75%
D CH4
flammability
limits: LFL
5.2%; RFR 15%
0.2
6/15-19/2009; 27
Sandia National
Laboratories
HydrogenMyths
D Hydrogen hazards can be compared favorably to
experiences with other hydrocarbon fuels
Less dangerous than gasoline, methane
Sandia National
Laboratories
LeanPremixedCombustion
forNOxControl
Sandia National
Laboratories
EffectofHydrogenEnrichment
onFlameStability
D Data points indicate
minimum equivalence
ratio at which a stable
flame can be maintained
nH2
D Hydrogen addition
significantly extends lean
flame stability
D Expected NOx levels less
than 3 ppm can be
achieved with 20 %
hydrogen addition
6/15-19/2009; 30
Sandia National
Laboratories
PresentDayH2ICEs:Emissions
DNOx is the only non-trivial
engine-out emission pollutant
10
NOx, ppm
10
10
US Federal Tier II
NOx < 0.07 g/mile
CARB, SULEV
NOx < 0.02 g/mile
10
10
0.25
0.5
0.75
1.25
1.5
6/15-19/2009; 31
***James Heffel, University of California, Riverside, College of Engineering Center for Environmental
Research and Technology (CE-CERT); Personal Communication; Under the technical guidance and
contract to Sandia National Laboratories, funding from the Hydrogen Program Office; OPT
Sandia National
Laboratories
HydrogenMyths
D Hydrogen hazards can be compared favorably to
experiences with other hydrocarbon fuels
Less dangerous than gasoline, methane
Sandia National
Laboratories
6/15-19/2009; 33
Alan Welch, David Mumford, Sandeep Munshi, Westport Innovations Inc., James Holbery, Pacific
Northwest National Laboratory, Brad Boyer, Matthew Younkins, Howard Jung, Ford Motor Co. Challenges
in Developing Hydrogen Direct Injection Technology for Internal Combustion Engines, SAE 2009-01-2379
Sandia National
Laboratories
HydrogenMyths
D Hydrogen hazards can be compared favorably to
experiences with other hydrocarbon fuels
Less dangerous than gasoline, methane
Sandia National
Laboratories
ProposedMechanismsfor
SpontaneousIgnition
D 81 ignitions of H2 releases have been reported
(MHIDAS database). In 11 cases the ignition
source was identified (flame, electric, hot
surface). In the remaining 70 no ignition source
could be identified.
D Proposed causes include the following:
Joule-Thomson
Static charge buildup in the flow
Shock heating that leads to ignition of H2/air mixtures
Catalytic reaction with materials present in the flow (iron oxide)
Friction heating of particulates / hot surface ignition
Sandia National
Laboratories
ProposedMechanismsfor
SpontaneousIgnition
D 81 ignitions of H2 releases have been reported
(MHIDAS database). In 11 cases the ignition
source was identified (flame, electric, hot
surface). In the remaining 70 no ignition source
could be identified.
D Proposed causes include the following:
6/15-19/2009; 36
Joule-Thomson
Static charge buildup in the flow
Shock heating that leads to ignition of H2/air mixtures
Catalytic reaction with materials present in the flow (iron oxide)
Friction heating of particulates / hot surface ignition
Sandia National
Laboratories
JouleThomsonEffect
High-pressure H2 Jet
P1
Patm
Sandia National
Laboratories
Acknowledgements
The author wishes to recognize the DOE Hydrogen Codes and Standards
program element and Antonio Ruiz, the Technical Program Manager, for
supporting most of the work reported on here.
The author also wishes to recognize the following people for their
contribution to the science discussed in this presentation who are not
otherwise recognized in the reference list.
Benard, Pierre, Universite du Quebec a Trios-Rivieres
Evens, Greg; Sandia National Laboratories
Groethe, Mark; SRI International
Houf, Bill; Sandia National Laboratories
Merilo, Eric; SRI International
Moen, Chris; Sandia National Laboratories
Schefer, Bob; Sandia National Laboratories
Tchouvelev, Andrei, Tchouvelev & Associates
White, Chris, Sandia National Laboratories
6/15-19/2009; 38
Sandia National
Laboratories
Publicationlist
(1)
(2)
(3)
(4)
(5)
(6)
11.3 m
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
6/15-19/2009; 39
Houf and Schefer, Predicting Radiative Heat Fluxes and Flammability Envelopes from Unintended Releases
of Hydrogen, accepted for publication Int. Jour. of Hydrogen Energy, Feb. 2006.
Schefer, Houf, San Marchi, Chernicoff, and Englom, Characterization of Leaks from Compressed Hydrogen
Dispensing Systems and Related Components, Int. Jour. of Hydrogen Energy, Vol. 31, Aug. 2006.
Molina, Schefer, and Houf, Radiative Fraction and Optical Thickness in Large-Scale Hydrogen Jet Flames,
Proceedings of the Combustion Institute, April, 2006.
Houf and Schefer, Rad. Heat Flux & Flam. Env. Pred. from Unintended Rel. of H2, Proc. 13th
Int. Heat Tran. Conf., Aug., 2006.
Schefer, Houf, Williams, Bourne, and Colton, Characterization of High-Pressure, Under-Expanded Hydrogen-Jet
Flames, submitted to Int. Jour. of Hydrogen Energy, 2006.
Houf and Schefer, Predicting Radiative Heat Fluxes and Flammability Envelopes from Unintended Releases of
Hydrogen, 16th NHA Meeting, Washington, DC, March 2005.
Schefer, R. W., Houf, W. G., Bourne, B. and Colton, J., Turbulent Hydrogen-Jet Flame Characterization, Int. Jour. of
Hydrogen Energy, 2005.
Schefer, R. W., Houf, W. G., Bourne, B. and Colton, J., Experimental Measurements to Characterize the Thermal and
Radiation Properties of an Open-flame Hydrogen Plume, 15th NHA Meeting, April 26-30, 2004, Los Angeles, CA.
Schefer R. W., Combustion Basics, in National Fire Protection Association (NFPA) Guide to Gas Safety, 2004.
P. Bnard (2007), Chapter 3 Hydrogen Release and Dispersion - Release of hydrogen - section a.1, , Biennial Report
on Hydrogen Safety, HySafe.
B. Angers, A. Hourri, P. Bnard, P. Tessier and J. Perrin (2005), Simulations of Hydrogen Releases from a Storage
Tank: Dispersion and Consequences of Ignition. International Conference on Safety 2005, Sept 8-10, 2005, Pisa, Italy.
A.V. Tchouvelev, P. Bnard, V. Agranat and Z. Cheng (2005), Determination of Clearance Distances for Venting of
Hydrogen Storage. International Conference on Safety 2005, Sept 8-10, 2005, Pisa, Italy (NRCAN, AUTO 21).
Tchouvelev A., P. Bnard, D. R. Hay, V. Mustafa, A. Hourri, Z. Cheng, Matthew P. Large, Quantitative Risk
Comparison of Hydrogen and CNG Refuelling Options, Final Technical Report to Natural Resources Canada for the
Codes and Standards Workshop of the CTFCA, August 2006 (194 pages).
Bnard, P., Tchouvelev, A., Hourri, A., Chen, Z., Angers, B. High Pressure Hydrogen Jets in a Presence of a Surface.
Proceedings of International Conference on Hydrogen Safety, San Sebastian, Spain, September 2007.
Tchouvelev, A.V., Howard, G.W. and Agranat, V.M. Comparison of Standards Requirements with CFD Simulations
for Determining of Sizes of Hazardous Locations in Hydrogen Energy Station. Proceedings of the 15th World
Hydrogen Energy Conference, Yokohama, June 2004.
Sandia National
Laboratories
Somepeoplejustdonotgetit!
DH2
is not toxic,
it is environmentally benign
we just borrow it -- (2H20 + E -> 2H2 + O2; then
2H2+O2 -> 2H2O + E)
SCIENCE
9/11/07
Sandia National
Laboratories
PresentationEnd
6/15-19/2009; 41
9/11/07
Sandia National
Laboratories