ILASS Americas 29th Annual Conference on Liquid Atomization and Spray Systems, Atlanta, GA, May 2017

High-Speed X-ray Fluorescence Measurements in an Impinging Jet Spray

B. R. Halls* and J. R. Gord Aerospace Systems Directorate Air Force Research Laboratory

WPAFB, OH 45433 USA

A. Douglawi, N. Rahman, and T. R. Meyer School of Mechanical Engineering

Purdue University West Lafayette, IN 47907 USA

A. L. Kastengren

X-Ray Science Division Argonne National Laboratory

Lemont, IL 60439 USA

Abstract A high-speed x-ray fluorescence measurement technique is developed to quantify the path-integrated liquid–liquid mixing in an impinging jet spray. Mixing is inferred from the amount of each liquid present along the x-ray beam. Measurements are made at the Sector 7 bending magnet beamline at the Advanced Photon Source located at Ar-gonne National Laboratory. The x-ray beam is spatially focused to 5×6 µm2, and spectrally filtered to 15±0.2 keV. The 15 keV beam excited sodium bromide doped into the spray and PIN diodes collected the fluorescence signal (captured orthogonally to the x-ray beam to minimize scatter) and the transmission signal. Two Reynolds numbers are investigated and measurements are performed with the x-ray beam perpendicular to the liquid sheet and parallel to the liquid sheet. Two measurements, for each condition, are performed with the tracer doped in one jet to deter-mine the mixing and doped in both jets for calibration and normalization. The accuracy and precision of the meas-urements are quantified and the limitations of the technique are discussed. This manuscript has been cleared for pub-lic release by the Air Force Research Laboratory.

*Corresponding author: hallsbenjamin@gmail.com

ILASS Americas 29th Annual Conference on Liquid Atomization and Spray Systems, Atlanta, GA, May 2017

Introduction Spray breakup and mixing processes are an integral

part of several applications including chemical pro-cessing and propellant injection in combustion devices [1–3]. In the case of liquid fuelled combustion previous work has shown that measurements in the absence of combustion, but at high temperatures and high pres-sures are still relevant to understanding the fundamental processes [4].

Several mechanical and laser-based diagnostics have been developed and employed in the measurement of sprays with great success [5–13]. However, laser diagnostics are susceptible to multiple scattering events and disruptions from refraction in large liquid struc-tures. The applicability of these techniques and the ef-fects of attenuation are discussed by [5]. Outlined in a review paper [6], Linne discusses several techniques specifically tailored for the near-field dense region of the spray and suggests that x-ray diagnostics may be the only way to directly measure mass distribution within a spray.

Of interest to the current work is the liquid mixing in an impinging jet spray; previous measurements have employed mechanical patternation, optical and laser-based diagnostics, and time-averaged x-ray fluores-cence. Using mechanical patternation, Rupe described the mixing by a ‘mixing factor’ based on the mass frac-tion of each fluid at a given location [7]. These meas-urements also discussed various mixing regimes, such as reflective mixing, transmitive mixing, and an inter-mediate regime termed well mixed [7,8]. Light trans-mission measurements using dyed fluids of various colors of light to determine the mixture ratio in each droplet have been performed [9]. Laser induced fluo-rescence was utilized to make mixing measurements based on the emission from a tracer in one of the jets [10,11]. Gas generation, from the mixing liquids, has been shown to enhance the mixing of propellants so long as it is not a dominating factor seen by [12], where the optimal performance is similar but not exactly Rupe’s conclusions on mixing performance. Exciplex fluorescence was applied in [13], where they concluded the combustion process was mixing limited due to the short chemical time scales.

These measurements were all made significantly below the impingement point, where gas initiation, and chemical reactions would originate in impinging jets. X-ray techniques have proven capable of directly measuring the mass distribution in all regions of the impinging jet spray [14].

Unlike visible light, which is strongly scattered from aerosols, droplets, and other liquid structures, the interaction of low-energy (~15 keV) x-rays with sprays is primarily through absorption and weak scattering. This greatly simplifies the analysis, as the attenuation signal can be related to the liquid mass density in the

path of the x-rays with minimal sensitivity to the drop-let size, shape, and number density. This gives a greater scope for quantitative measurements of liquid mass distribution with x-ray techniques than with most other optical techniques. Much of the recent work in advanc-ing x-ray spray measurements has taken place at the Advanced Photon Source (APS) located at the Argonne National Laboratory [15–22]. Synchrotron x-ray sources such as the APS provide a highly collimated, tunable x-ray beam with sufficient flux for radiography to be performed over a wide range of liquid path lengths. This has yielded highly accurate and precise attenuation measurements under a variety of spray con-ditions. Related to this work, time-averaged mixing has been investigated in gaseous jets [20], liquid jets [21], and liquid–gas flows [22].

The current study employs high-speed simultane-ous radiography and x-ray fluorescence to measure the mass distribution and liquid mixing in an impinging jet spray, respectively. Mixing was inferred from the measurement of the total mass and the measurement of the mass from a single jet that was doped with NaBr that fluoresced in the x-ray regime.

Experimental Setup

The liquid mixing studies were performed at the 7-BM beamline of the APS, a bending magnet beamline dedicated to time-resolved fluid dynamics measure-ments. The x-ray source creates a high-flux, polychro-matic, nearly collimated x-ray beam. The beamline consists of two enclosures, 7-BM-A and 7-BM-B. En-closure A contains slits and a double multilayer mono-chromator to condition the beam size and to select a narrow range of x-ray energies, respectively. The mon-ochromator can tune the photon energy from 5.1 to 15 keV (∆E/E = 1.4%). Enclosure B houses Kirkpatrick-Baez focusing mirrors [23], the spray was affixed to a two-dimensional translation stage, and the x-ray detec-tors to measure attenuation, fluorescence and to normal-ize the beam intensity. A schematic of the enclosures and experiment is shown in Fig. 1. A thorough descrip-tion of the beamline can also be found in [15]. Point-wise raster scanning data were collected while the spray was traversed across the incident beam, along the minor axis, and at several downstream locations, along the z-axis (normal to major and minor axes).

The experiments were conducted at a photon ener-gy of 15±0.2 keV, with the x-ray beam was focused to 5 µm (vertical) × 6 µm (horizontal) at full width at half maximum (FWHM). Two PIN diodes were used to measure the attenuation from both liquids and the fluo-rescence from one liquid. A third diamond-based transmission diode monitored the incoming x-ray inten-sity. The attenuation PIN diode was constructed of 300-µm-thick silicon and operated without bias voltage de-tected the transmitted x-ray intensity for attenuation

3

measurements. The PIN diode signal was normalized by the signal from a diamond-based transmission diode.

Figure 1. Schematic of APS 7-BM enclosures for x-ray energy selection, beam focusing, and detection of at-tenuation and fluorescence (above view).

The relative x-ray attenuation along the path of the beam, and hence the amount of liquid along that path, was then determined using Beer’s Law. For a mono-chromatic x-ray beam, Beer’s Law can be written as:

(1)

where I and I0 are the attenuated and non-attenuated x-ray intensities, respectively, measured at the PIN diode, α is the attenuation coefficient, and l is the path length of the x-ray beam through the fluid. In the case where the liquid is not contiguous along the path of x-ray beam, then l is interpreted as an equivalent path length (EPL) of the liquid intersecting the x-ray beam. The attenuation coefficient, α, is the product of the attenua-tion cross-section and number density of the fluid, which can be considered constant in the case of identi-cal water jets. The small change in attenuation due to the fluorescence tracers was accounted for and verified through calibration. The fluorescence PIN diode was placed 33° to the x-ray beam, and a selenium filter, with a Kα absorption edge at 12.7 keV, was used to block a majority of the scattered x-rays at 15 keV and transmit the Br Kα fluorescence x-rays at 11.9 keV.

Mixing between the two fluids was inferred from the mass of each fluid and the mass of the doped fluid. The mass of the doped fluid (2.5% NaBr by mass) is linear with the fluorescence signal and the EPL was determined through calibration.

A two-step process was necessary to determine the EPL of the undoped fluid because the attenuation measurement relies on knowledge of the attenuation coefficient of the mixture and, therefore, the local do-pant concentration, which is initially unknown. The first step assumes the fluid is composed of the doped solution, and a guessed EPL is determined. Then the guessed EPL is compared to the fluorescence EPL and the attenuation coefficient is updated. The second step uses the new attenuation coefficient and determines the total EPL. The fluorescence EPL is then subtracted from the total EPL and the result is the undoped fluid EPL.

The measurement spatial resolution was limited by the selected raster-scan point spacing of 80 µm. Noise in the attenuation and fluorescence signals was deter-mined by the signal’s standard deviation when no spray was present. The noise values equate to EPL values of 20 µm and 12 µm for the attenuation and fluorescence signals respectively.

The injector used in all the studies was a like-doublet impinging jet injector. The injector had an in-cluded angle of 60°, an orifice diameter of 0.5 mm, with the distance between the injector and impingement point defined as the free jet length of 8.5 mm, and a Re of 5000. Figure 2 displays a schematic of the impinging jet injector tips. A pressure fed system with two 6 L reservoirs and remotely operated valves fed the injector. The jet velocity was equalized and adjusted with two rotometers.

Figure 2. Schematic of the injector orifice tips; en-closed angle, 2θ, free jet length, FJ, and jet impact loca-tion of z = 0 mm, where the free jet centerlines cross. Results and Discussion

Data were collected at two locations in the spray: in the center of the doped jet and where the two jets meet along the centerline. The data collected in the doped jet are used for calibration because there is a

( )( )

( ), ,

0

,,

x y z dx lI y ze e

I y zα α− −∫= =

z=0mm FJ

2θ

z

minor

4

single known fluid so the EPL based on the attenuation measurement will equal the EPL based on the fluores-cence measurement. Figure 3 shows the data from the doped jet.

Figure 3. Line plot displaying 100 µs time segment of high-speed EPL data from the center of the doped jet.

The doped EPL shows the turbulent fluctuations in the liquid jet, and the neat EPL shows the difference between the fluorescence measurement and attenuation measurement, because no neat fluid was present in this location. The measurement uncertainty based on the RMS difference between EPLs was 7%.

Data collected just below where the jets are meet are shown in Fig. 4. The data show, on average, equal amounts of each liquid, since the location is at the cen-terline of the liquid sheet. However on an instantaneous basis the amount of each liquid is drastically different. Both fluids oscillate between ~0–1 mm EPL, and a few different events are shown. The first event is seen be-tween 0–50 µs where each fluid is oscillating back and forth where the local jet momentums are unmatched. The jet momentums are more closely matched between 50–75 µs where the amount of each liquid is closer to being equal. These fluctuations are likely linked to sheet dynamics and mixing behavior. The oscillatory motion is an indication of the two fluids passing through each other in transmitive mixing. The times of equal momentum likely see a growth in the size of the liquid sheet area as the two jets imping equally and spread along the major axis.

The main challenge when interpreting the data is that the signal is integrated along the x-ray beam in the spray. This integrating effect only allows the larger fluctuations to be measured. Comparisons with imaging data and time-averaged data are necessary to draw complete conclusions about the mixing dynamics.

Figure 4. Line plot displaying 100 µs time segment of high-speed EPL data from 0.25 mm below the jet im-pingement point. Conclusions

Time-resolved measurements of liquid mixing were made in an impinging jet spray. Straightforward data analysis and small uncertainty values resulted in quantitative mixing data showing the time evolution of transmitive mixing between the two jets. Acknowledgments

Benjamin R. Halls was funded under a National Research Council Post-doctoral Research Associateship Award at the Air Force Research Laboratory, Aero-space Systems Directorate, Wright-Patterson AFB. The measurements were performed at the 7-BM beamline of the Advanced Photon Source, Argonne National Labor-atory, supported by the U.S. Department of Energy un-der Contract No. DE-AC02-06CH11357. This manu-script has been cleared for public release by the Air Force Research Laboratory (No. 88ABW-2017-1221). References 1. Lefebvre, A., Atomization and Sprays, Hemisphere

Publishing Corp, New York 1989. 2. Ashgriz, N., Handbook of Atomization and Sprays,

Springer, 2011. 3. Meyer, T. R., Brear, M., Jin, S. H., Gord, J. R.,

Formation and Diagnostics of Sprays in Combus-tion, Handbook of Combustion, Wiley, New York, 291–322, 2010.

4. Nurick, W.H., and Clapp, S.D., J. Spacecraft, 6(11), 1969.

5. Sick, V., Stojkovic, B., Appl. Optics, 40(15) 2435-2442, 2001.

6. Linne, M. A., Prog. Energy Combustion Science, 39(5) 403-440, 2013.

5

7. Rupe, J. H., The Liquid-Phase Mixing of a Pair of Impinging Streams, JPL Progress Rept. 20-195, 1953.

8. Ashgriz, N., Brocklehurst, W., Talley, D., J. Prop. Power, 17(3), 2001.

9. Somogyi, D., Feiler, C.E., “Study of the mixture-ration distribution in the drops of sprays produced by impinging liquid streams”, NASA-TR, 1959.

10. McDonell, V., Phi, V., Samuelsen, S. Shahnam, M., Nejad, A., Carlson, R.A., Guernsey, “Structure of Sprays Generated by Unlike Doublet Injectors”, AIAA, JPC, June, 1999.

11. Yuan, T. Huang, B., Atomization and Sprays, 22(5), 391–408 2012.

12. Houseman, J., “Combustion effects in sprays”, JPL-TN, 1968.

13. Feikema, D.A., Smith, J.E., “Combustion and flow visualization of hypergolic combustion and gelled mixing behavior”, ARO-TR, 1997.

14. Halls, B.R., Heindel, T.J., Kastengren, A.L., Mey-er, T.R., I. J. Multiphase Flow 59, 2014.

15. Kastengren, A.L. and Powell, C.F., Exp. Fluids 55(3), 1686, 2014.

16. Powell, C.F., Yue, Y., Poola, R., and Wang, J., J. Synchrotron Radiation (Fast Communications), 7(6) 356–360, 2000.

17. Lin, K.-C., M. Ryan, M., Carter, C., Sandy, A. Na-rayanan, S., Ilavsky, J., and Wang, J., Nucl. In-strum. Methods Phys. Res. A 649(1), 219–221, 2011.

18. Lin, K.-C., Rajnicek, C., McCall, J., Carter, C., and Fezzaa, K., Nucl. Instrum. Methods Phys. Res. A 649(1), 194–196, 2011.

19. Schumaker, S.A., Kastengren, A.L., Lightfoot, M.D.A., Danczyk, S.A., and Powell, C.F., A study of gas-centered swirl coaxial injectors using X-ray radiography, 12th ICLASS, 2012.

20. Kastengren, A. L., Powell, C. F., Dufresne, E. M. & Walko, D. A., J. Synchrotron Rad., 18, 811-815, 2011.

21. Halls, B.R., Meyer, T.R., A.L. Kastengren, Optics Express, 2015.

22. Radke, C. D., McManamen, J. P., Kastengren, A. L., Halls, B. R., Meyer, T. R., Op. Lett., 2015.

23. Eng, P.J., Newville, M., Rivers, M.L., Sutton, S.R., P. Soc. Photo-Opt Ins., 3449, no. 145, 145-156, 1998.

24. Hubbell, J.H. and Seltzer, S.M., Tables of X-Ray Mass Attenuation Coefficients and Mass Energy-Absorption Coefficients (version 1.4). [Online] Available: http://physics.nist.gov/xaamdi [2013, 02 20]. NIST, Gaithersburg, MD (2004).

25. Beckhoff, B., B. Kanngiesser, N. Langhoff, R. Wedell, and H. Wolff, Handbook of Practical X-ray Fluorescence Analysis. Springer, Berlin, 2006.

26. Thompson, A.C., Kirz, J., Attwood, D.T., Gul-likson, E.M., Howells, M.R., Kortright, J.B., Liu, Y., Robinson, A.L., Underwood, J.H., Kim, K-J., Lindau, I., Pianetta, P., Winick, H., Williams, G.P., Scofield, J.H., X-ray Data Booklet, Lawrence Berkeley National Lab, 2009.