Journal ArticleDOI

# Detonation interaction with a diffuse interface and subsequent chemical reaction

27 Apr 2007-Shock Waves (Springer-Verlag)-Vol. 16, Iss: 6, pp 421-429

Abstract: We have investigated the interaction of a detonation with an interface separating a combustible from an oxidizing mixture. The ethylene-oxygen combustible mixture had a fuel-rich composition to promote secondary combustion with the oxidizer in the turbulent mixing zone that resulted from the interaction. Diffuse interfaces were created by the formation of a gravity current using a sliding valve that initially separated the test gas and combustible mixture. Opening the valve allowed a gravity current to develop before the detonation was initiated. By varying the delay between opening the valve and initiating the detonation it was possible to achieve a wide range of interface conditions. The interface orientation and thickness with respect to the detonation wave have a profound effect on the outcome of the interaction. Diffuse interfaces result in curved detonation waves with a transmitted shock and following turbulent mixing zone. The impulse was measured to quantify the degree of secondary combustion, which accounted for 1–5% of the total impulse. A model was developed that estimated the volume expansion of a fluid element due to combustion in the turbulent mixing zone and predicted the resulting impulse increment.
Topics: Detonation (67%), Combustion (51%)

Content maybe subject to copyright    Report

D.H. Lieberman · J.E. Shepherd
Detonation Interaction with a Diffuse Interface and Subsequent
Chemical Reaction
Received: date / Accepted: date
Abstract We have investigated the interaction of a deto-
nation with an interface separating a combustible from an
oxidizing mixture. The ethylene-oxygen combustible mix-
ture had a fuel-rich composition to promote secondary com-
bustion with the oxidizer in the turbulent mixing zone that
resulted from the interaction. Diffuse interfaces were cre-
ated by the formation of a gravity current using a sliding
valve that initially separated the test gas and combustible
mixture. Opening the valve allowed a gravity current to de-
velop before the detonation was initiated. By varying the de-
lay between opening the valve and initiating the detonation
D.H. Lieberman
Graduate Aeronautical Laboratories, California Institute of Technol-
ogy, Pasadena, CA USA
Tel.: +626-437-3192
Fax: +310-754-2799
E-mail: dlieberman@exponent.com Present address: Exponent Fail-
ure Analysis Associates, Los Angeles, CA USA
J.E. Shepherd
Graduate Aeronautical Laboratories, California Institute of Technol-
ogy, Pasadena, CA USA E-mail: joseph.e.shepherd@caltech.edu
it was possible to achieve a wide range of interface condi-
tions. The interface orientation and thickness with respect to
the detonation wave have a profound effect on the outcome
of the interaction. Diffuse interfaces result in curved detona-
tion waves with a transmitted shock and following turbulent
mixing zone. The impulse was measured to quantify the de-
gree of secondary combustion, which accounted for 1-5%
of the total impulse. A model was developed that estimated
the volume expansion of a ﬂuid element due to combustion
in the turbulent mixing zone and predicted the resulting im-
pulse increment.
Keywords Detonation · diffuse interface · secondary
combustion
1 Introduction
When a detonation wave propagating in a gaseous combustible
mixture reaches a concentration boundary, a complex inter-
action results between the detonation and interface between
the two gases. The details of this interaction are dependent
D. Lieberman and J. E. Shepherd "Detonation interaction with a Diffuse Interface and Subsequent
Chemical Reaction" Shock Waves 16(6):421-430, 2007. Preprint, see journal for final version http://
dx.doi.org/10.1007/s00193-007-0080-3

2 D.H. Lieberman, J.E. Shepherd
on the mixture compositions, the relative geometry of the
detonation and interface, and the characteristic thickness of
the interface.
In the present study, the interface is a composition gra-
dient between the combustible and non-combustible mix-
tures. The thickness of the interface is determined by mixing
caused by ﬂuid mechanical stirring and diffusion. We clas-
siﬁed the extent of the mixing at the interface by comparing
the cell size of the detonation with the interface thickness. A
sharp interface occurs when the detonation cell size is much
greater than the interface thickness.
1
A diffuse interface, the
case of the present study, occurs when the detonation cell
width in the combustible gas is much less than the interface
thickness.
The geometry of a detonation propagating in a composi-
tion gradient can be divided into two main categories. The
ﬁrst is when the detonation velocity is parallel to the direc-
tion of the gradient, and the second is when it is perpen-
dicular. In general, the composition gradient vector and the
detonation velocity vector are not parallel or perpendicular
but at some intermediate angle. The ﬁrst case has been stud-
ied by numerous researchers investigating the transmission
and successful re-initiation of a detonation across a gap of
non-combustible gas ([3], [8], [15], [16]). Other work [7] has
examined detonation transmission to an inert gas or other
combustible mixture and subsequent reﬂection off of a rigid
1
We have studied this case and the results will be presented in a
companion paper.
end-wall. The current study will focus to a large extent on
the perpendicular or nearly-perpendicular case.
When the detonation propagation direction is perpendic-
ular to the mixture gradient, a curved detonation wave results
that ultimately decouples into a shock wave and turbulent
mixing zone (TMZ), shown in Fig. 1b. This is due in part to
the dependence of the detonation velocity on the equivalence
ratio. For a detonation modeled as an ideal one-dimensional
discontinuity with no affect of curvature on reaction zone
structure, the normal component of the curved wave will cor-
respond to the local Chapman-Jouguet detonation velocity.
There has been little work done on detonation propagation
perpendicular to a continuous composition gradient. Exper-
iments [10] were carried out that measured concentration
gradients made by diffusion and used soot foils to charac-
terize the detonation propagation. Other work [4] examined
concentration gradients in a numerical study to investigate
the possibility of a ﬂame occurring in the incomplete com-
bustion products. Oblique detonations were observed [5] in
sharp interface experiments where various combustible mix-
tures were separated in parallel channels. It is possible for
combustion to occur in the turbulent mixing zone (TMZ) by
choosing a combustible mixture such that the combustion
products are incompletely oxidized. This allows further re-
actions to take place if the mixture downstream of the inter-
face contains an oxidizer [14]. The aim of the following dis-
cussion is to address the key physical issues that arise when
a detonation propagates in the direction normal to a concen-
tration gradient. Some of the main issues to be discussed are

Detonation Interaction with a Diffuse Interface and Subsequent Chemical Reaction 3
INTERFACE
6
#*
ENDWALL
([S
([S
([S
([S
([S
([S
`
Fig. 1 The interaction of a detonation with a diffuse interface is illustrated with supporting observations from experimental images. A detonation
wave (a) interacts with the diffuse interface and forms a curved wave (b). Upon exiting the combustible mixture, the detonation decouples
completely resulting in a transmitted shock and TMZ (c) and (d). When the shock reaches the endwall it reﬂects (e) and interacts for a second
time with the interface.
the general shape of the detonation wave, and the decou-
pling of the shock wave and reaction zone. The presence of
secondary combustion between the partially oxidized deto-
nation products and oxidizer in the TMZ is also examined.
2 Experimental setup
The experiments were carried out using the GALCIT Deto-
nation Tube (GDT) [1,2], which is 7.3 m long with an inside
diameter of 0.280 m. It is connected to a 0.762 m long square
test section with inside dimensions of 0.15 m by 0.15 m. A
wave cutting device extends 1 m into the end of the GDT to
cut out a square section of the circular detonation front be-
fore entering the test section. A sliding valve assembly sep-
arates the ethylene-oxygen combustible mixture in the GDT
from the oxidizer or inert diluent in the test section during
the experimental preparation.
Figure 2 is a view of the test section illustrating the loca-
tion of the end ﬂange of the GDT, the sliding valve assem-
bly, and the test section. Visualization for the experiments,
using a schlieren system [1], was made through an optical
viewport (BK7 or quartz windows) that could be arranged
in two separate positions. The ﬁrst position is located 0.275
m downstream of the sliding valve and is referred to as port
1. The second viewport position is located 0.56 m down-
stream of the sliding valve (see Fig. 2) and is referred to as

4 D.H. Lieberman, J.E. Shepherd
port 2. The locations of the pressure transducers and a quartz
window on the end wall used for ﬂuorescence imaging are
also shown. Fuel rich ethylene-oxygen mixtures occupied
the GDT, and either oxygen, nitrogen, or nitrous oxide occu-
pied the test section. The ethylene-oxygen equivalence ratio
Φ
= 2,2.5,3. The initial pressure and temperature for all ex-
periments were 15 kPa and 298 K, respectively. Detonations
'\$4
3LIDING6ALVE!SSEMBLY
0RESSURE4RANSDUCERS
0,)&7INDOW
1UARTZ7INDOW
4EST3ECTION
"+7INDOW
00RESSURE
4RANSDUCER
0
0
0
Fig. 2 A schematic of the test section with the sliding valve assembly
and the end ﬂange of the GDT.
were initiated in the GDT by discharging a 2
µ
F bank of
capacitors charged to 9 kV through a 0.16 mm copper wire.
Detonation velocities were measured to within 5% of the
Chapman-Jouguet speed. The cell sizes of the detonations
were measured using the soot foil technique [19] and were
in all cases below 5 mm.
The diffuse interface was made with a gravity current
(GC). The formation and propagation of the gravity current
was examined in the test section using acetone planar laser
induced ﬂuorescence (PLIF) [12,17], and in a half-scale wa-
ter channel [20] to understand the early time development of
the GC, see Fig 3. The thickness of the diffuse interface,
δ
c
,
was estimated from the thickness of the region of vorticity
obtained from digital particle image velocimetry (DPIV) [9,
20] measurements from the water channel experiments and
then re-scaled for the gas phase experiments. In the PLIF ex-
periments, the combustible mixture in the GDT was replaced
with an acetone-helium mixture of matched density.
The sliding valve, actuated by a falling mass, was de-
signed to completely isolate both the combustible mixture
and test gas, as well as to open sufﬁciently fast to control
the formation of the gravity current. The sliding valve was
measured to open in 170 ms with an uncertainty of 10 ms.
The mass needed to achieve this opening time was 55 kg.
3 Results and discussion
The general description of this problem, shown as a schematic
in Fig. 1, is of a detonation wave propagating through a dif-
fuse interface. The diffuse interface is composed of ethylene-
oxygen with
Φ
= 2.5 above a layer of pure oxygen. The
two layer situation was created by allowing a long time to
elapse between the creation of the gravity current and det-
onation initiation (see discussion below). The sequence be-
gins with a detonation wave (a) prior to encountering the
diffuse interface. A curved detonation wave (b) results dur-
ing the interaction leading to a decoupled transmitted shock
and TMZ. As the detonation exits the gravity current, what
remains is a transmitted shock followed by the TMZ (c). The

Detonation Interaction with a Diffuse Interface and Subsequent Chemical Reaction 5
Fig. 3 Time sequence illustrating the development of a gravity current
using dye visualization. The gravity current head in the dense ﬂuid (at
bottom of image) is propagating from right to left. The dark colored
ﬂuid is the salt water solution and the clear ﬂuid is de-ionized water.
The plate is completely withdrawn at 0.16 s.
shock wave and TMZ occupy a smaller axial distance than
the curved detonation wave with the distance between the
shock and TMZ (d) increasing with time. When the shock
reaches the endwall it reﬂects (e) and interacts with the inter-
face a second time. The reﬂected-shock-interface interaction
is not visible as it occurs in the space between the window
and the endwall. The complex interaction of the shock wave
with the lower experimental boundary Fig. 1b,c produces a
Mach reﬂection.
Gravity currents of various sizes were formed by vary-
ing the delay time, deﬁned from the time the sliding valve
reaches the open position to the time when a detonation is
initiated in the GDT. The range of behavior is shown in
Fig. 4 where four separate experiments are shown at dif-
ferent stages of the gravity current development. In these
experiments the
φ
= 2.5 ethylene-oxygen combustible mix-
ture is colored yellow for visibility and ﬂows below the less
dense nitrogen gas. Figure 4a is an image of a shock wave
followed by a TMZ. The delay time was 0 s, corresponding
to a planar interface, and consequently explains why there
is no gravity current visible on the overlay. This translates
to a leading shock wave that is for the most part perpen-
dicular to the top and bottom surface along with a reﬂected
trailing shock that is possibly due to slight imperfections in
the interface shape. As the gravity current grows, the wave
structure becomes more curved. Figure 4b shows the trans-
mitted shock, TMZ, and the location of the gravity current
before the combustible mixture was detonated with a delay
time of 1 s. The leading shock wave is curved with a Mach
stem at the top wall. Figure 4c shows the location of the
gravity current after a 2 s delay time and the presence of a
detonation propagating within 10% of the CJ velocity and
transverse waves visible just behind the detonation front at
the bottom of the ﬁgure. At a 3 s delay time, the curved det-
onation (Fig. 4d) looks similar to the detonation in Fig. 4c
except that the gravity current occupies half the height of the
test section thus changing the curvature of the leading wave.

##### Citations
More filters

Journal ArticleDOI
01 Mar 2016-Shock Waves
Abstract: The influence of transverse concentration gradients on detonation propagation in $$\hbox {H}_2$$ –air mixtures is investigated experimentally in a wide parameter range. Detonation fronts are characterized by means of high-speed shadowgraphy, OH* imaging, pressure measurements, and soot foils. Steep concentration gradients at low average $$\hbox {H}_2$$ concentrations lead to single-headed detonations. A maximum velocity deficit compared to the Chapman–Jouguet velocity of 9 % is observed. Significant amounts of mixture seem to be consumed by turbulent deflagration behind the leading detonation. Wall pressure measurements show high local pressure peaks due to strong transverse waves caused by the concentration gradients. Higher average $$\hbox {H}_2$$ concentrations or weaker gradients allow for multi-headed detonation propagation.

51 citations

Journal ArticleDOI
TL;DR: The propagation of detonations through several fuel–air mixtures with spatially varying fuel concentrations is examined numerically, finding that in the high-activation-energy mixture, detonations propagate more slowly as the magnitude of the mixture composition gradient is increased and can be quenched in a large enough gradient.
Abstract: The propagation of detonations through several fuelair mixtures with spatially varying fuel concentrations is examined numerically. The detonations propagate through two-dimensional channels, insid...

44 citations

01 Jan 2015-
Abstract: Explosion of H2-air mixtures portrays a major hazard in nuclear reactors during severe loss-of-coolant accidents. Spatial gradients in H2 concentration prevail in real-world scenarios. Mixture inhomogeneity can lead to significantly stronger explosions as compared to homogeneous mixtures. The present work identifies and quantifies the underlying physical mechanisms.

19 citations

### Cites background from "Detonation interaction with a diffu..."

• ...Lieberman and Shepherd [97] investigated detonation interaction with a diffuse interface between two mixture layers....

[...]

• ...Thank you Prof. Joseph E. Shepherd for inviting me to work in your group and for opening doors for the future....

[...]

27 Apr 2011-
Abstract: : Explosions in mixtures of natural gas (NG) and air have been of intense practical concern for coal mines for many years. Potentially explosive mixtures of NG and air can accumulate in the active ventilated areas or in unventilated sealed areas of these mines. If an ignition source, such as a simple spark, ignites the NG-air mixture and creates a flame, the initially slow-moving flame can become turbulent, accelerate rapidly, develop extremely intense pressure waves, and potentially generate enormous stress on coal mine seals. In this work, we attempt to answer the question: Given a large enough volume of flammable mixture of NG and air, can a weak spark ignition develop into a detonation? Large-scale numerical simulations, in conjunction with experimental work conducted at the National Institute for Occupational Safety and Health?s Gas Explosion Test Facility, were performed to address four specific problems: flame acceleration and deflagration-to-detonation transition (DDT) in obstructed channels containing a stoichiometric methane-air mixture, flame acceleration and DDT in fuel-lean and fuel-rich mixtures, effects of spatially varying fuel concentrations on detonations, and stochastic effects on flame acceleration and DDT.

17 citations

### Cites background from "Detonation interaction with a diffu..."

• ...For larger systems compared to the detonation cell size (such as those considered in experiments [89, 90]), where many triple points can form and interact, the influence of channel size would be less pronounced....

[...]

• ...Experiments have shown that the sizes of detonation cells vary spatially in relation to the variation in fuel concentration [89] and that the shape of the detonation wave is curved in these systems [89, 90]....

[...]

• ...The shape of the detonation front was similar for all three gradients considered, and did not exhibit the curved structure observed in experiments [89, 90]....

[...]

Journal ArticleDOI
Wenhu Han1, Cheng Wang1, Chung K. Law2Institutions (2)
Abstract: The role of a transversal concentration gradient in detonation propagation in a two-dimensional channel filled with an bump and thus the unreacted pocket behind the front, while the transverse wave causes mixing and burning of the residue fuel downstream. Nevertheless, for the steepened concentration gradient, a transverse detonation is present and consumes the fuel in the compressed and preheated zone by the leading shock; consequently, the detonation velocity deficit is not increased significantly for detonation with the single-head propagation mode close to the limit.

9 citations

##### References
More filters

Journal ArticleDOI
01 Oct 1957-Physics Today
Abstract: Elements of gasdynamics , Elements of gasdynamics , مرکز فناوری اطلاعات و اطلاع رسانی کشاورزی

1,467 citations

### "Detonation interaction with a diffu..." refers methods in this paper

• ...The increase in pressure is related to the velocity increase using the acoustic equation [13]...

[...]

Journal ArticleDOI
Abstract: The goal of the present study is to quantify and reduce, when possible, errors in two-dimensional digital particle image velocimetry (DPIV). Two major errors, namely the mean bias and root-mean-square (RMS) errors, have been studied. One fundamental source of these errors arises from the implementation of cross correlation (CC). Other major sources of these errors arise from the peak-finding scheme, which locates the correlation peak with a sub-pixel accuracy, and noise within the particle images. Two processing techniques are used to extract the particle displacements. First, a CC method utilizing the FFT algorithm for fast processing is implemented. Second, a particle image pattern matching (PIPM) technique, usually requiring a direct computation and therefore more time consuming, is used. Using DPIV on simulated images, both the mean-bias and RMS errors have been found to be of the order of 0.1 pixels for CC. The errors of PIPM are about an order of magnitude less than those of CC. In the present paper the authors introduce a peak-normalization method which reduces the error level of CC to that of PIPM without adding much computational effort. A peak-compensation technique is also introduced to make the mean-bias error negligible in comparison with the RMS error. Noise in an image suppresses the mean-bias error but, on the other hand, significantly amplifies the RMS error. A digital video signal usually has a lower noise level than that of an analogue one and therefore provides a smaller error in DPIV.

375 citations

### "Detonation interaction with a diffu..." refers methods in this paper

• ...The thickness of the diffuse interface, δc, was estimated from the thickness of the region of vorticity obtained from digital particle image velocimetry (DPIV) [9, 20] measurements from the water channel experiments and then re-scaled for the gas phase experiments....

[...]

29 Jul 1991-
Abstract: : Some experimental data on turbulent free-shear-layer growth, mixing, and chemical reactions are reviewed. The dependence of these phenomena on such fluid and flow parameters as Reynolds number, Schmidt number, and Mach number are discussed, with the aid of some direct consequences deducible from the large-scale organization of the flow as well as from some recent models. The mixing of two or more fluids that are entrained into a turbulent region is an important process from both a scientific and an applications vantage point. Species can be transported by turbulence to produce a more uniform distribution than some initial mean profile. This process is sometimes also referred to as mixing, without regard to whether the transported species are mixed on a molecular scale or not. If the issue of mixing arises in the context of chemical reactions and combustion, however, we recognize that only fluid mixed on a molecular scale can contribute to chemical product formation and associated heat release. The discussion in this paper will be limited to molecular mixing.

215 citations

01 Jan 1989-
Abstract: : Some experimental data on turbulent free-shear-layer growth, mixing, and chemical reactions are reviewed. The dependence of these phenomena on such fluid and flow parameters as Reynolds number, Schmidt number, and Mach number are discussed, with the aid of some direct consequences deducible from the large-scale organization of the flow as well as from some recent models. The mixing of two or more fluids that are entrained into a turbulent region is an important process from both a scientific and an applications vantage point. Species can be transported by turbulence to produce a more uniform distribution than some initial mean profile. This process is sometimes also referred to as mixing, without regard to whether the transported species are mixed on a molecular scale or not. If the issue of mixing arises in the context of chemical reactions and combustion, however, we recognize that only fluid mixed on a molecular scale can contribute to chemical product formation and associated heat release. The discussion in this paper will be limited to molecular mixing.

204 citations

Journal ArticleDOI
M. Vetter1, Bradford Sturtevant1Institutions (1)
01 Mar 1995-Shock Waves
Abstract: Results of flow visualization experiments of an impulsively accelerated plane interface between air and SF6 are reported. The shock tube used for the experiments has a larger test section than in previous experiments. The larger extent of uniform test flow relative to nonuniform boundary-layer flow permits unambiguous interpretation of flow-visualization photographs, and the influence of shock-wave/boundary-layer interactions is no longer dominant. The strong wall vortex observed in previous studies is not observed in these experiments. It is found that the thin membrane, which forms the initially plane interface, has a significant influence on the initial growth rate of the interface thickness. However, the measured growth rates after the first reflected shock are independent of membrane configuration and are in good agreement with analytical predictions.

203 citations

### "Detonation interaction with a diffu..." refers background in this paper

• ...It has been shown [18] that TMZ growth can increase by a factor of six after the re-shock event....

[...]

##### Network Information
###### Related Papers (5)
28 Jul 2007, Shock Waves

Kazuhiro Ishii, M. Kojima

01 Jan 1998

Mike Kuznetsov, V.I. Alekseev +3 more

01 Mar 2016, Shock Waves

L.R. Boeck, Frederik M. Berger +2 more

01 Apr 1991, Combustion and Flame

G.O. Thomas, P. Sutton +1 more

01 May 2013, Shock Waves

F. Ettner, Klaus G. Vollmer +1 more

##### Performance
###### Metrics
No. of citations received by the Paper in previous years
YearCitations
20221
20213
20204
20192
20181
20161