Flameholding Analyses in Supersonic Flow

A. Thakur and C. SegalUniversity of Florida, Gainesville, Florida

Abstract

Flameholding in supersonic flow depends on the local conditions in recirculation regions and the mass transfer into and out of these regions. Large gradients in gas composition and temperature exist in a recirculation region. Hence stability parameter correlations developed for premixed flames cannot be used to determine the blowout stability limits for non-premixed flames encountered in practical devices. In the present investigation carried out in cold-flow, mixture samples at different locations in the recirculation region and the shear layer formed behind a rearward-facing step have been extracted and analyzed by a mass spectrometer todetermine the local distribution of species concentration. The difference between the local mass fraction of injected gas within the recirculation region determined from mass spectroscopy and the global mass fraction based on the total mass of air and injected gas is identified. A simple geometrical configuration has been used with a rearward facing step followed by a constant area duct and the sampling has been focused on the recirculation region both in an axial and a transverse direction. Air arrives at the step at Mach 1.6 at 300K. Argon has been used to model a gas with molecular weight close to propane, a gaseous hydrocarbon that is the subject of a parallel study in chemically reacting flows.

Introduction

A substantial database of flame stability exists for premixed gases1,2,3 from which stability limits for rich and lean flames have been obtained for a number of fuels. The stability limit has usually been cast in terms of the flameholding boundary on an equivalence ratio vs. a stability parameter plane. These stability parameters depend on the flow velocity, temperature, size and shape of the flameholder and have received various formulations in different studies, from empirical formulations to expressions that reflect global Damkhöler numbers4.

In the case of non-premixed gases the determination of stability limits is less straightforward, primarily due to the non-homogeneity of the parameters in the recirculation region behind the flameholder. It is difficult to estimate the spatial species concentration and temperature distribution in the recirculation regions of these flows due to the presence of large gradients and the complex, 3-D, flow structure. These difficulties are compounded by the uncertainty in the shape of the recirculation region, which is a function of the amount of heat release which, in turn, is dictated by the local mixing and combustion efficiencies. Recent data5

for non-premixed gases in a supersonic combustor configuration have been produced based on estimates of mass exchanges in the recirculation region formed behind the flameholder. Fresh air mass entrainment has been obtained using empirical data. Correlated with the size of the recirculation region and the estimated residence time a local equivalence ratio and other parameters responsible for flame stability have been indicated. With the observation that the recirculation region remains subsonic, it is noted in ref. 5 that stability parameters suggested for subsonic, premixed gases can be applied for non-premixed flows as well. The underlying assumption is that mixing of the fuel injected into the recirculation region is complete and

12th AIAA International Space Planes and Hypersonic Systems and Technologies15 - 19 December 2003, Norfolk, Virginia

AIAA 2003-6909

Copyright © 2003 by University of Florida. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.

2

uniform, conditions that can be assumed for a range of operation, in particular at large equivalence ratios. For other conditions, at lean fuel-air mixtures, experimental data suggests that rich mixtures can exist in the recirculation region6 accompanied by (i) large gradients in temperature and (ii) rapid changes of the temperature distribution as the overall equivalence ratio changes which suggests similar gas composition time dependency.

Sampling or optically-based methods used as a means of mapping the fuel distribution in non-premixed supersonic flows have been applied extensively7,8,9 and have shown that high local concentrations can persist for a long distance due to the characteristics of mixing layers growth and the role of large scale roll-up layers. Similar mechanisms are also responsible for the mass transfer between a recirculation region and the flow that surrounds it, although a more complex situation arises due to the complicated streamlines geometry and the possible fuel injection or entrainment in this region. The boundary layer formed upstream of the step is pushed by the expansion towards the region behind the flameholder. A shear layer forming between the boundary layer and the recirculation region brings fresh air in the region and helps the transfer of burnt gases into the core airflow. Any species concentration within this shear layer can have any value, at a given instance, from zero to one10. If fuel is injected into the recirculation region, shear layers appear around the jet boundaries in which the flame is initiated, followed by mixing and heat exchange between the hot gases in the region. Thus a primary recirculation of gases exists that engulfs the recirculation region with additional smaller recirculations present. Even in a two-dimensional geometry, a complex 3-D flow pattern emerges. In this region, large local equivalence ratio can exist even when the global equivalence ratio indicates an, overall, lean mixture11.

Sampling in the recirculation region and subsequent mass spectroscopic analyses has been undertaken in this study to (i) evaluate the distribution of gases in the region and (ii) to relate the local concentration of the injected gas simulating fuel to the global fuel-air mass ratio and the other flow conditions at the leading edge of the recirculation region. Argon has been used as the injectant gas. Both axial and transverse sampling has been done in non-reacting Mach 1.6 flow with argon injected upstream of the recirculation region or directly into the base of the step.

Experimental Facility

The facility at University of Florida provides direct connect tests with a variable combustion chamber entrance Mach number of 1.6 – 3.6 and stagnation temperatures corresponding to Mach 3.0 - 4.8 flight. All the experiments presented here are performed with combustion chamber entrance Mach 1.6 and cold air. The facility has been described in detail elsewhere.12 Briefly, this is a continuously operating facility using a vitiated heater based on hydrogen combustion with oxygen replenishment, electronically controlled by a fuzzy logic controller13 to maintain a constant 0.21 oxygen mole fraction at all conditions and to maintain the constant stagnation temperature at the heater exit required for the experiment. A bellmouth with four-side contraction leads to the supersonic nozzle with compression on two sides and interchangeable nozzle blocks that cover the range of Mach 1.6 to 3.6. A constant area isolator is placed between the nozzles and the combustor section to protect the nozzle from upstream pressure rise due to combustion in the test section. The isolator includes two fuel injection holes on each side, of 1-mm dia. equally spaced in a cross section. These injectors are referred to in the following text as the “upstream” injection. The selection of these orifices’ location was

3

determined from considerations of fuel penetration and spreading and is described in detail in Cuesta et al14. Optical access is available to the isolator’s flow from three sides. The test section is symmetric with ample optical access through covering windows. The isolator cross-section is 2.5 x 2.5 cm2 upstream of the rectangular, rearward facing step having step height H = 1.25 cm, and follows with a constant cross-section area over its 26H length. The test section wall at the base of the step has five fuel injection holes on each side, of 0.5-mm dia. equally spaced in a cross section.

Cold flow at Mach 1.6 using argon as simulated fuel has been used since it has a molecular weight close to that of propane. The inert gas has been injected transverse into the air stream at one of the two locations: (i) at the base of the step, and (ii) upstream of the test section in the isolator. The injection locations are shown in Figure 1.

The physical location of the mass sampling ports in the recirculation region behind the step is shown in Figure 1. The coordinate system is also shown in the figure. The test section window wall covering the step has five mass sampling ports in the recirculation region along the axial x-direction, equally spaced from x/H = 0.2 to 3.2. These ports are 0.6 mm inner diameter steel tubes that end at the test section window wall and do not physically intrude into the recirculation region. In separate tests, other tubes are inserted from the window wall to verify the two-dimensionality of species distribution in the recirculation region. In this case, two groups of three stainless steel tubes each are placed at two locations and penetrate into the test section to sample species at three different depths, equally spaced in the transverse z-direction from z/W = 0.33 to 1.0. Here W = 1.25 cm is the test section half-width.

A schematic diagram of mass sampling from the recirculation region and subsequent analysis by a mass spectrometer is shown in Figure 2. The water-cooling jacket shown in the figure is provided to quench the reactions and freeze the species composition coming out of the combustion chamber for the combustion tests that will follow in a future experiment. The sampling ports coming out of the recirculation region are connected to a manifold that has a single outlet going to the mass spectrometer. The input of species to the manifold is regulated by a series of computer-controlled miniature solenoid valves that supply gas mixture from one sampling port at a time for analysis. Sampling from each port is preceded by injection of nitrogen in the manifold to purge the line and flush the species from the previous port, hence preventing mixing of samples from two adjacent ports. The species are analyzed by Stanford Research Systems RGA-300 mass spectrometer that uses electron impact to ionize the gas and a RF quadruple filter to sort species according to their mass-to-charge ratio. The mass spectrometer has an operating pressure range of 10-4 torr (1.3 x 10-7 atm). It can detect species up to a mass to charge ratio of 300 and has a resolution of 0.5 AMU @ 10% peak height.

Results

The mass sampling experiments are carried out under steady airflow conditions of P0air = 4.7 atm, T0air = 300 K and M0 = 1.6 at the isolator entrance. This air stagnation pressure ensures that the flow is supersonic throughout the isolator and the test section. The expansion at the step forms a shear layer and a recirculation region behind the step. Argon is injected at steady stagnation pressures and T0argon = 297 K. The stagnation pressures of air and argon are stabilized to within 0.1 atm of the desired value.

The composition of gas in the recirculation region is analyzed by the mass spectrometer in Partial pressure of species vs. Time mode. The species scanned are nitrogen (m/z = 28, 14),

4

oxygen (m/z = 32, 16) and argon (m/z = 40, 20). Sampling is done sequentially for 15sec at the purge port and for 25sec at each of the sampling ports. The spectrometer has a fast response time to the change in composition of gas while switching from one port to another. The local mass fraction of argon is determined from the partial pressures of nitrogen, oxygen and argon.

where m& are the mass flow rates and P are the partial pressures of the selected gases as measured by the mass spectrometer. Since air already contains 0.7% argon, this amount is subtracted to arrive at the correct argon mass fraction at each sampling port. Further, the time-averaged argon mass fraction at a port is obtained by averaging the mass fractions obtained over the sampling time period.

The global mass fraction of argon is determined from the total mass of argon injected and the total mass of air traveling through the test section. Both local and global mass fractions are indicated in the results below.

Mass sampling in axial direction

Mass sampling of the recirculation region species is done in the axial x-direction with argon injected upstream or at the base of the step. At both injection locations, argon is injected at two stagnation pressures.

Upstream injection

For upstream injection, argon is injected at a low stagnation pressure P0argon = 2.2 atm and a moderate stagnation pressure P0argon = 5.2 atm resulting in dynamic pressure ratios, qr, of 0.89 and 2.18, respectively. The global argon mass fractions for the two injection pressures are MassFrargon (%) = 0.14, 0.34 respectively. The axial distribution of argon mass fraction in the recirculation region for the two argon stagnation pressures is shown in Figure 3. A schematic diagram of the injection and sampling locations is indicated at the bottom of the figure. Each test is repeated four times. Since mass sampling is done under steady airflow and fuel injection conditions, the variation in the argon mass fraction distribution between different tests can be largely attributed to the inherently dynamic nature of the flow in the recirculation region. The argon mass fraction in the recirculation region for P0argon = 2.2 atm case is higher than P0argon = 5.2 atm case by 10% on the average. Hence more fuel injected upstream reaches the recirculation region if it is injected into the airflow boundary layer. Injection at higher dynamic pressure ratio results in part of the fuel going out into the core airflow and is less efficient in reaching the recirculation region.

A comparison of global vs. local argon mass fraction is shown in Table 1 below.

argon argonargon

argon argon

(%) 100* 100*air nitrogen oxygen

m PMassFr

m m P P P= =

+ + +&

& &

Table 1. Global and local argon mass fractions for upstream injectionqr local MassFrargon global MassFrargon local/global MassFrargon

0.89 1.82-2.75 0.14 13-202.18 1.70-2.36 0.34 5-7

5

At P0argon = 2.2 atm (qr = 0.89), global MassFrargon (%) = 0.14 and local MassFrargon (%) = 1.82 -2.75. At P0argon = 5.2 atm (qr = 2.18), global MassFrargon (%) = 0.34 and local MassFrargon (%) = 1.70 - 2.36. The local fuel mass fraction in the recirculation region is much higher than the estimated global value. This difference is quite significant especially for low fuel injection pressure P0argon = 2.2 atm, when the locally measured mass fraction of injectant is 13-20 times larger than based on global estimate. In this case the results show that while the global fuel mass fraction suggests a lean fuel-air mixture, the recirculation region in fact has a much richer fuel-air composition. Further, as the fuel stagnation pressure is decreased from 5.2 atm to 2.2 atm thus suggesting a leaner global fuel-air mixture, the composition in the recirculation region becomes richer in fuel. In combustion experiments, this may lead to flameout on the fuel rich side rather than the fuel lean side, as was found in ref. 6.

Base injection

For base injection, argon is injected at a moderate stagnation pressure P0argon = 5.4 atm and a high stagnation pressure P0argon = 13.3 atm. A dynamic pressure ratio in this case is difficult to determine without a definite knowledge of the local Mach number. The axial distribution of argon mass fraction in the recirculation region for the two argon stagnation pressures is shown in Figure 4. It is observed that increasing the fuel injection stagnation pressure substantially does not result in a corresponding increase in the fuel mass fraction in the recirculation region, as part of the fuel escapes into the core airflow without getting a chance to mix within the recirculation region.

Table 2 indicates the locally measured and the global argon mass fractions. At P0argon =

5.4 atm, global MassFrargon (%) = 0.25 and local MassFrargon (%) = 1.42 – 1.90. At P0argon = 13.3 atm, global MassFrargon (%) = 0.61 and local MassFrargon (%) = 1.78 – 2.35. As in the case of upstream injection, it is observed that the recirculation region has a much richer fuel-air composition than the suggested global fuel mass fraction. However, the differences between locally measured and globally estimated argon mass fractions are lower than for the upstream injection, since argon is injected directly into the recirculation region in the present case. Furthermore, the differences between high and low pressure injections are lower for base injection.

For both upstream and base injection, the argon mass fraction distribution in the recirculation region shows small gradients in the axial x-direction, with a decrease towards x/H = 1.5 and a slight increase towards x/H = 2.5. This indicates a uniform mixing of fuel and air in the recirculation region. This can be attributed to cold-flow conditions and this finding cannot be extrapolated to combustion experiments where sharp gradients in species concentration distribution may exist.

Table 2. Global and local argon mass fractions for base injectionP0 argon [atm] local MassFrargon global MassFrargon local/global MassFrargon

5.4 1.42-1.90 0.25 5-8 13.3 1.78-2.35 0.61 3-4

6

Mass sampling in transverse direction

Upstream injection

Mass sampling of the recirculation region species is done in the transverse z-direction for the same airflow and argon injection conditions as in axial sampling. The transverse sampling is done at two of the five ports investigated in axial sampling: @ x/H = 1.7 and @ x/H = 3.2. The purpose is to verify the two-dimensionality of species distribution in the recirculation region.

For upstream injection, the transverse distribution of argon mass fraction in the recirculation region is shown in Figure 5 for two argon stagnation pressures. At P0argon = 2.2 atm, @ x/H = 1.7 the figure shows a rather uniform distribution of argon in transverse direction, while some variation is observed at @ x/H = 3.2. At P0argon = 5.2 atm, the fuel distribution is quite uniform at both locations. When compared with the local mass fractions measured from the wall, at both locations, the inflow measurement averages up to 8% less than the wall measurement at low injection pressure and up to 10% for the higher injection pressure.

Base injection

For base injection, the transverse distribution of argon mass fraction in the recirculation region is shown in Figure 6 for two argon stagnation pressures. A similar trend is observed for the two injection pressures P0argon = 5.4 atm and 13.3 atm with lower measured mass fraction at z/W = 0.67 in comparison with z/W = 0.33 and 1.0 port depths. This trend is clear for both pressures, at both axial locations. With base injection, at low pressure the differences between wall measured and inflow averaged local mass fractions are larger, with up to 28% at low pressures and up to 20% for the higher pressure. It can be noted from figs. 5 and 6 that the dip at z/W = 0.67 is more pronounced for the base injection. This points out to a more complicated 3-d flow in this injection configuration due to the direct injection of argon in the base region.

Summary

A mass sampling of inert injectant in a supersonic flow evaluated the penetration of the gas in the recirculation region behind a rearward-facing step. Two injection locations were chosen with two injection pressures in each case. The results showed the following:- in both cases there is a significant difference between the locally measured injectant mass fractions and the globally estimated injectant/air mass fraction. - with upstream injection the difference is as large as 13-20 times at low injection dynamic pressure ratio and drops to 5-7 times for higher pressure ratio. At low injection pressure, most of the fuel remains in the boundary layer that carries it into the recirculation region. This indicates that in a non-premixed wall injection fueled supersonic flow, rich mixtures can exist in the recirculation region even when globally the mixture is lean.- the base injection brings the gas directly into the recirculation region and the differences between locally measured and globally evaluated mass fractions are lower than in the upstream injection case, an indication that more fuel remains in this region for base injection.- the transverse mass sampling shows a relatively uniform distribution with upstream injection.- the differences between wall measured mass fractions and inflow measurement show a difference of less than 8% for the upstream injection.

7

- in the case of base injection the transverse measurements indicate a dip in mass fraction at z/W = 0.67 in comparison with the measurements at z/W = 0.33 or 1.0 duct width, which is an indication of the more complex flow patterns due to this type of injection.- with base injection the differences between the wall measurement and the averaged inflow measurement can reach 20-28% depending on the injection pressure.

Acknowledgement

This work is part of the Institute for Future Space Transport activities, a NASA established URETI to support Reusable Launch Vehicles research.

References

1 Ozawa, R. I., “Survey of Basic Data on Flame Stabilization and Propagation for High Speed Combustion Systems”, The Marquart Co., TR AFAPL-TR-70-81, Jan. 1971.2 Huellmantel, L. W., Ziemer, R. W., Cambel, A. B., “Stabilization of Premixed Propane-Air Flames in Recessed Ducts”, Journal of Jet Propulsion, pp. 31-43, Jan. 1957.3 Ogorodnikov, D. A., Vinogradov, V. A., Shikhman, Y. M., Strokin, V. N., “Russian Research on Experimental Hydrogen-Fueled Dual-Mode Scramjet: Conception and Preflight Tests”, Journal of Propulsion and Power, Vol. 17, No. 5, pp. 1041-1048, 2001. 4 Ortwerth, P. J., Mathur, A. B., Segal, C., Mullagilli, S., “Combustion Stability Limits of Hydrogen in a Non-Premixed, Supersonic Flow”, Proceedings of ISABE 99, Paper 99-143, 1999. 5 Morrison, C. Q., Campbell, R. L., Edelman, R. B., “Hydrocarbon Fueled Dual-Mode Ramjet/Scramjet Concept Evaluation”, Proceedings of ISABE 97, Paper 97-7053, pp. 348-356, Sept. 1997.6 Owens, M. G., Tehranian, S., Segal, C., Vinogradov, V. A., “Flameholding Configurations for Kerosene Combustion in a Mach 1.8 Airflow”, Journal of Propulsion and Power, Vol. 14, no. 4, pp.456-461, July-Aug. 1998.7 Rogers, R. C., “A Study of the Mixing of Hydrogen Injected Normal to a Supersonic Airstream”, NASA Technical Note, TN D-6114, 1971.8 Cox, S. K., Fuller, R. P., Schetz, J. A., Walters, R. W., “Vortical Interactions Generated by an Injector Array to Enhance Mixing in Supersonic Flow”, 32nd Aerospace Sciences Meeting and Exhibit, Paper 94-0708, 1994.9 VanLerberghe, W. M., Santiago, J. C., Dutton, J. C., Lucht, R. P., “Mixing of a Sonic Transverse Jet Injected into a Supersonic Flow”, AIAA Journal, Vol. 38, No. 3, pp. 470-479, March 2000.10 Dimotakis, P. E., “Turbulent Free shear Layer Mixing and Combustion”, High-Speed Flight Propulsion Systems, (Murthy, S. N. B., Curran, E. T., eds.), Progress in Astronautics and Aeronautics, Vol. 137, 1991.11 Mitani, T., Takahashi, M., Tomioka, S., Hiraiwa, T., Tani, K., “Analyses and Application of Gas Sampling to Scramjet Engine Testing”, Journal of Propulsion and Power, Vol. 15, No. 4, pp. 572-577, July-August 1999. 12 Segal, C., Young, C. D., "Development of an Experimentally-Flexible Facility for Mixing-Combustion Interactions in Supersonic Flow", Transactions of ASME - Journal of Energy Resources Technology, Vol. 118, pp. 152-158, June 1996.13 Owens, M., Segal, C., “Development of a Hybrid-Fuzzy Air Temperature Controller for a Supersonic Combustion Test Facility”, Experiments in Fluids, Vol. 31, no. 1, pp. 26-33, 2001.

8

14 Cuesta, D. F., Segal, C., Goldman, A., Ortwerth, P. J., “Flame Stability and Heat Release in Mach 1.6 Flows”, 12th International Space Planes and Hypersonic Systems and Technologies Conference, Norfolk, VA, Paper 2003-6913, Dec. 2003.

9

Figure 1. Schematic diagram of the test section, which is symmetric about the centerline. Argon is injected from sonic, round orifices upstream from the step and in the base of the step. The step height, H = 1.25 cm and the duct half-width, W = 1.25 cm.

Figure 2. Mass sampling from the recirculation region behind the step for analysis by the mass spectrometer. The axial location of the sampling ports is indicated in the following figures.

10

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

Distance from step (x/H)

Arg

on

Lo

cal m

ass

frac

tio

n (

%)

(a) P0argon = 2.2 atm, global MassFrargon = 0.14%.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

Distance from step (x/H)

Arg

on

Lo

cal m

ass

frac

tion

(%)

(b) P0argon = 5.2 atm, global MassFrargon = 0.34%.

Figure 3. Argon mass distribution in axial direction for upstream injection. a) At low dynamic pressure ratio there is a larger scatter of measured mass fraction than at, b) higher dynamic pressure, when most of the injected gas is expected to penetrate the core flow.

11

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

Distance from step (x/H)

Arg

on

Lo

cal m

ass

frac

tio

n (

%)

(a) P0argon = 5.4 atm, global MassFrargon = 0.25%.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

Distance from step (x/H)

Arg

on

Lo

cal m

ass

frac

tio

n (

%)

(b) P0argon = 13.3 atm, global MassFrargon = 0.61%.

Figure 4. Argon mass distribution in axial direction for base injection shows a slight decrease towards x/H = 1.5 and increases again around x/H =2.5.

12

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Distance from w all (z/W )

Arg

on

Lo

cal m

ass

frac

tio

n (

%)

(a) x/H = 1.7

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Distance from wall (z/W)

Arg

on

Lo

cal m

ass

frac

tio

n (

%)

(b) x/H = 3.2

Figure 5. Transverse argon mass fraction distribution with upstream injection at low pressure, P0argon = 2.2 atm. Relatively uniform distribution is noted, with a slight dip at z/W = 0.67 duct depth for axial location x/H = 3.2.

13

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Distance from wall (z/W)

Arg

on

Lo

cal m

ass

frac

tio

n (

%)

(c) x/H = 1.7

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Distance from wall (z/W)

Arg

on

Lo

cal m

ass

frac

tio

n (

%)

(d) x/H = 3.2

Figure 5 cont’d. Increased pressure to P0argon = 5.2 atm. with upstream injection shows a uniform pattern across the duct at both axial locations, x/H = 1.7 and 3.2.

14

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Distance from wall (z/W)

Arg

on

Lo

cal m

ass

frac

tio

n (

%)

(a) x/H = 1.7

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Distance from wall (z/W)

Arg

on

Lo

cal m

ass

frac

tio

n (

%)

(b) x/H = 3.2

Figure 6. Transverse argon mass fraction distribution with base injection at P0argon = 5.4 atm. indicates a lower value at z/W = 0.67 for both axial locations.

15

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Distance from wall (z/W)

Arg

on

Lo

cal m

ass

frac

tio

n (

%)

(c) x/H = 1.7

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Distance from wall (z/W)

Arg

on

Lo

cal m

ass

frac

tio

n (

%)

(d) x/H = 3.2

Figure 6 cont’d. Base injection at P0argon = 13.3 atm. indicates the dip at z/W = 0.67.