- Email: [email protected]
Optical diagnostics for high-speed flows
Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Progress in Aerospace Sciences journal homepage: www.elsevier.com/locate/paerosci
Optical diagnostics for high-speed flows Richard B. Miles 1,n Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ 08544, USA
art ic l e i nf o
a b s t r a c t
Article history: Received 25 June 2014 Accepted 11 September 2014
Since 2000 there has been a revolution in diagnostics of high-speed air flows. The foundations for this revolution were laid over the past few decades, but with the development of new short pulse and pulse burst laser technologies, higher laser powers and higher pulse energies, new high-speed cameras, better laser control and improved detection and laser delivery methodologies, many very effective new capabilities have been developed. Newly developed methods for molecular tagging velocimetry provide high fidelity visualization of transport properties and may be extended to simultaneous temperature measurements. Rapid field imaging with frequency tunable pulse burst lasers shows instantaneous flow structure and complex boundary and mixing interactions. Extending these pulse burst concepts to swept volumetric imaging is very promising for full volumetric data collection. Fast wavelength modulation spectroscopy follows real-time flow variation, and three-dimensional particle imaging extends particle imaging velocimetry to volumetric data acquisition. & 2014 Elsevier Ltd. All rights reserved.
Keywords: Diagnostics High speed flows Flow imaging
1. Introduction and retrospective The use of optical diagnostics for the study of high-speed flows dates back to the 1800s when shadowgraphs and schlieren yielded images of the bow shocks, Mach disks and other structures associated with high-speed projectiles and supersonic flows [21] (see Fig. 1). Notable progress in optical diagnostics through the twentieth century that did not involve laser technologies primarily focused on further development of schlieren and shadowgraph for high sensitivity and high-resolution imaging of flow structure. These approaches provided good resolution of large-scale structures but suffered from integration over the full optical path length, so details of turbulent boundary layers, curved and unsteady shocks and mixing structures were not well resolved. The introduction of an electron beam [30] overcame that problem since electron beams could be spatially collimated and electronically swept, providing luminous cross sections of shock and boundary layer structure. However, electron beams are limited to low density flows due to electron scattering, and they are very difficult to integrate into a test facility. Focusing schlieren [55] provided a method for imaging flow structure over a reduced integrated path length in higher density flows. Since the invention of the laser in 1960, the continuing evolution of optical flow diagnostics has been driven in large part by everincreasing laser and camera capabilities. The very first laser invented, the pulsed ruby laser, provided high energy and excellent coherence,
n
Tel.: þ 1 60 925 8 5131. 1 Robert Porter Patterson, Professor of Mechanical and Aerospace Engineering, Fellow AIAA.
which enabled the development of interferometric methods to measure flow field properties, such as flow velocity using seed particles [54] and the imaging of boundary layer structure using density variations [16]. It was not until after the invention of the frequency tunable laser in 1966 [50,48] that atomic and molecular spectroscopy could be utilized for diagnostics. The tunable dye lasers only operated efficiently in the visible portion of the spectrum where the air is highly transparent, so for these new applications, flow seeding became important. Initially seeding with sodium provided planar imaging of flow cross sections [33] and enhanced schlieren [4]. These advances were accomplished by tuning the laser either onto a resonance or near a resonance and utilizing the laser-induced fluorescence for imaging planar cross sections or the enhanced index of refraction for higher sensitivity schlieren. Tuning the lasers provided methods for imaging and measurement of velocity fields by taking advantage of the Doppler shift associated with the motion of the gas [58]. Due to the reactivity of sodium with air, these experiments were carried out in either helium or nitrogen flows. Imaging and interferometry in these early experiments were done with conventional hard film. Molecular iodine was subsequently used for laser-induced flow imaging [29] since it has spectral features throughout the visible and does not react with air. Later, tunable ultraviolet lasers became available through frequency up conversion of nanosecond laserdriven pulsed dye lasers, and nitric oxide [41] and acetone [25] became the preferred molecular species for seeding. CCD array and intensified CCD cameras became available and provided high sensitivity, time gating and convenient data processing capabilities. The possibility of using nonlinear optical methods for flow diagnostics became credible with higher energy, frequency tunable
http://dx.doi.org/10.1016/j.paerosci.2014.09.007 0376-0421/& 2014 Elsevier Ltd. All rights reserved.
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
2
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Fig. 1. Image of an underexpanded sonic jet taken with schlieren photography [26]
pulsed lasers and led to temperature and species measurements by Coherent Anti-stokes Raman Scattering (CARS), temperature and velocity measurements by Laser Induced Thermal Anemometry (LITA) and Molecular Tagging Velocimetry (MTV). The CARS approach [27,18] brings two or three pulsed laser beams together at a point and, through a resonant nonlinear interaction, generates a new laser beam whose intensity is determined by the properties of the gas at that point. CARS has the capability of measuring species concentrations as well as temperature. For that reason it is also of great interest for combustion studies. Originally CARS required that at least one of the lasers be scanned in frequency to acquire the data, and that meant that CARS could only be used for time averaged measurements. Broadband and later dual broadband CARS [14] solved that problem by replacing the scanned laser with a broadband laser and separating the multiple CARS signal frequencies that were simultaneously generated with a spectrometer. This allowed CARS to capture full spectral data over a limited band in a single pulse. LITA [9] was a similar local measurement approach. It used a pair of focused lasers to produce a localized thermal grating into the air through nonlinear mixing, and a probe laser to follow the motion of the acoustic waves created by that grating as they interfered with each other. This provided a local measure of the temperature though the speed of sound and a measure of the flow velocity through frequency offsets associated with the flow motion. MTV [20] introduced a line or array of lines into the flow and tracked them in time as they moved, providing a measure of both the velocity and the flow velocity structure. The first MTV concept implemented in unseeded air was Raman Excitation þLaser Induced Electronic Fluorescence (RELIEF) [34], which used three laser beams – two to drive the oxygen into the vibrational state through stimulated Raman excitation, and one to interrogate the displaced line or pattern by laser-induced fluorescence. Its great feature was that it did not require seeding of the air with other molecular species and produced negligible perturbation. It worked well because of the relatively long lifetime of the oxygen vibrational state (many microseconds even in humid air). It was limited by the complexity of the laser systems. Its success led to the development of other approaches including laser-induced Ozone Tagging Velocimetry (OTV) [44], which was also used as a tag in unseeded air. In this case, the ozone was created by a chemical reaction following laser-induced dissociation of molecular oxygen. The motion was tracked by subsequent laser-induced dissociation of the ozone and imaging of the fluorescence from the excited molecular oxygen fragment. Other MTV approaches for air developed before the 2000 used seed molecules and included biacetyl [17] and water vapor [5]
Single mode, frequency tunable lasers utilizing injection locking enabled the development of molecular, atomic and etalon filtered technologies, permitting strong suppression of background scattering [35], imaging of air temperature, velocity and density (Filtered Rayleigh Scattering [36]) and velocity imaging by Doppler shifted particle imaging through an iodine filter (Doppler Global Velocimetry [31]) as well as velocity and temperature imaging of Doppler shifted Rayleigh scattering through an etalon [49]. Single mode tunable diode lasers derived from the communications industry and augmented by wavelength modulation technology have also opened the door to diagnostic methods for air based on direct absorption spectroscopy using very weak near infrared lines in molecular oxygen [43]. This concept has been successfully implemented for density, velocity and temperature measurements based on the measurement of extinction, line shifts and line broadening. Particles have been used for centuries to observe flows, but the development of laser provided a method for quantitative measurement through instantaneous holographic imaging and other interferometric methods. Much early work focused on Laser Doppler Anemometry (LDA) [13] with continuous lasers for one or two component point measurements of flow velocity, in which two laser beams intersected at the sample point and the scattering of the particle as it moved through the interference pattern which was created provided the measure of velocity. With four crossing beams, two velocity components could be measured. The development of high power nanosecond pulsed lasers enabled imaging of time frozen particle fields and this led to particle imaging velocimetry (PIV) [1], where the two-dimensional velocity field was measured by the displacement of the particles captured with double pulsed laser systems. Digital PIV [56] was enabled by the development of highresolution CCD cameras and eliminated the need for hard film. Thus at the beginning of the twenty-first century many capabilities existed for optical diagnostics of high-speed flows. Since that time further development and implementation of these capabilities have occurred and laser technology has significantly advanced, enabling new approaches. In addition to achieving higher pulse energy, better reliability and higher efficiency lasers, optical fiber technologies, new cameras, frequency tunable pulse burst lasers and sub-picosecond lasers have opened up new possibilities for diagnostics. With these tools major advances have been made in high-speed imaging, molecular flow tagging, wavelength modulation spectroscopy, Particle Imaging Velocimetry, Coherent Antistokes Raman Scattering and Rayleigh scattering.
2. Imaging Laser Rayleigh scattering is the strongest non-resonant light scattering process available for air measurements, but the low scattering cross section of air molecules has made its use for highspeed diagnostics challenging and only recently practical with high energy pulsed lasers and high sensitivity, time gated cameras. It is best applied in free jet facilities where background scattering can be minimized. An important application of Rayleigh scattering in an free jet of air was the evaluation of the Mariah II/Radiatively Driven Hypersonic Wind Tunnel concept [39]. Those tests were undertaken for the validation of computational models of an electron beam heated hypersonic ground test facility and were conducted at Sandia National Laboratory using their 1 MW Hawk electron beam facility [28]. The configuration for the tests is shown in Figs. 2 and 3. The 1 MW electron beam is steered and focused into the nozzle from downstream using a carefully contoured magnetic field and the Rayleigh imaging is performed with a frequency doubled Nd:YAG laser focused to a thin sheet along the center line of the flow at the exit of the nozzle, providing
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎
3
Fig. 2. Setup for the Rayleigh experiments on the electron beam coupled MARIAH II/Radiatively Driven Hypersonic Wind Tunnel at Sandia National Laboratory.
Fig. 3. Time sequenced Rayleigh images of the density at the exit of the MARIAH II/RDHWT nozzle before and during the electron beam energy addition.
an image of the instantaneous density cross section. As the electron beam deposits energy into the core of the flow inside the nozzle, the total enthalpy is more than doubled. The enthalpy deposition leads to a reduced density in the core of the flow, and the ensuing density profile together with spatially localized temperature and velocity measurements provide a quantitative measure of the energy deposited. Fig. 4 shows the Rayleigh images of the evolution of that density profile starting before the electron beam is turned on and continuing until equilibrium is reached 300 μs later. The exit jet is an over expanded supersonic free jet at the beginning of the energy addition. For accurate measurements of the density, images such as these need to be carefully calibrated and background noise must be subtracted [3]. Filtered Rayleigh Scattering enhances the capability of laser Rayleigh scattering through the use of an atomic or molecular gas with a strongly absorbing sharp spectral line to filter the scattering. A cell with that gas is placed in front of the camera, and a narrow line width laser tuned in the vicinity of the strong absorption line is used to illuminate the flow. The filter suppresses background scattering, but permits light that has been frequency shifted by thermal, acoustic or convective motion to pass through. By scanning the laser, this filter can provide for measurement of temperature and pressure [38]. Single shot imaging of temperature fields can be acquired if the pressure is known [37]. The ability of Filtered Rayleigh Scattering to capture cross-sectional cuts of boundary layer structure in supersonic and hypersonic flows has produced detailed images of boundary layer behavior in the vicinity of shockinduced separation [6]. Image contrast is greatly enhanced by seeding the flow with about 1% of CO2 gas [45], which condenses, forming a nanoscale particle fog in the cold core of the flow. The
CO2 condensation highlights the outer portion of the boundary layer where the temperature rises to the sublimation temperature. This imaging capability is further enhanced by the pulse burst laser [22]. Because it is based on a master oscillator power amplifier design (MOPA) the laser is naturally single frequency (single mode) and frequency tunable over a limited range due to the diode pumped Nd:YAG master oscillator. The master oscillator operates as a continuous laser and only after pre-amplification is the laser beam temporally chopped into a pulse burst and passed through the power amplifiers. This feature makes it especially useful for applications such as Filtered Rayleigh Scattering that require narrow line width and tunability over a limited range. The very rapid response of the CO2 nanoparticle sublimation and condensation provides a clear set of images of the time evolving shock wave boundary layer structure as shown in Fig. 4, where the pulse burst laser has been used to acquire sequential images of the boundary layer driven fluctuations of a separated shock in the vicinity of a 151 ramp at 2 μs intervals. By tuning the laser wavelength relative to the iodine filter absorption edges, high and low velocity features can be highlighted as shown in the figure. Images were taken with a MHz rate camera with 30 frames of on board storage designed by Princeton Scientific Instruments. The initial design of the pulse burst laser had a limited time window of 100 μs or so over which the pulses could be generated based on the gain time of the flashlamp pumped power amplifiers. Recent work has significantly extended that range through precision-controlled diode pumping of the power amplifiers, and increased the overall energy of the pulse burst. These advances have enabled the dynamic imaging of lower frequency boundary layer instabilities and flow phenomena and extended the utility of the pulse burst laser to processes such as combustion ignition that occur over longer time intervals [51]. Extension of the spectral frequency range of the pulse burst laser has been achieved with the addition of optical parametric oscillators/amplifiers and frequency up conversion crystals, and now the capability for rapid planar laser-induced fluorescence imaging of nitric oxide [19,2] and other molecular species has been demonstrated. By combining the pulse burst laser with fast beam sweeping technologies, and lenslet array cameras, it is now being extended to full threedimensional data acquisition [52].
3. Advances in molecular tagging Molecular flow tagging has developed significantly since 2000. The Air Photolysis And Recombination Tracking (APART)
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
4
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Fig. 4. Filtered Rayleigh Scattering, 500 KHz rate images of a shock wave boundary layer interaction upstream of a 15 degree wedge. The presence of CO2 condensate nanoparticles in the low temperature core of the flow provides the contrast. Columns 2 and 3 are with the laser tuned to highlight high and low velocity features. (Flow is from right to left.)
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎
technique [10] uses a UV laser to dissociate oxygen and form nitric oxide, which is later imaged by laser-induced fluorescence. If nitric oxide is seeded into the flow or already present from combustion processes, then it can be used for molecular tagging in high-speed environments by taking advantage of the lifetime of the laserinduced fluorescence [11]. Due to quenching, the fluorescence lifetime is short, but at the low densities and high speeds associated with hypersonics, the lifetime is long enough to allow a time-gated camera to image the delayed fluorescence and thus the velocity profiles. The Vibrationally Excited NO Monitoring (VENOM) technique uses photodissociation of seeded NO2 to produce vibrationally excited NO, which is subsequently imaged by LIF after a delay [51]. The VENOM method also yields a measure of the temperature from the NO rotational spectrum. The availability of high pulse energy femtosecond lasers has enabled Femtosecond Laser Electronic Excitation Tagging (FLEET) [32]. In this case a 150 fsec, 2 mJ, Ti:sapphire laser is focused into air and dissociates nitrogen molecules throughout the focal zone by a high-order nonlinear interaction. The nitrogen atoms that are formed by this dissociation recombine over an 100 μs time interval, forming nitrogen molecules in the electronically excited B state. Those molecules fluoresce in the red and near-infrared portion of the spectrum, and that fluorescence can be imaged with a time gated, high sensitivity camera. This process is diagrammed in Fig. 5. Since the nitrogen atoms are only formed through the focal region of the excitation laser, they are initially formed along a straight line and act to tag that region. The timedelayed image shows the location to which each segment along that line has moved in the time interval between tagging and interrogation. An important aspect of FLEET is the continued luminosity of the tagged region for tens of microseconds. Thus by using a fast sequentially gated camera, the evolution of the
Fig. 5. Energy level diagram for molecular nitrogen showing the recombination path for nitrogen atoms leading to B to A first positive emission. Dissociation is by a highly nonlinear interaction driven by the 800 nm, 150 fsec laser.
5
tagged elements in the three-dimensional flow can be followed in real time. Fig. 6 shows images of FLEET lines written at the nozzle exit and imaged at one microsecond time delay intervals showing the flow at sequential heights above the exit of an over-expanded Mach 2.6 air jet. The shape of each line gives the instantaneous velocity profile at that location and the displacement from the original line gives a quantitative measure of the velocity. FLEET may also have the capability to measure temperature along the line by relying on the second positive ultraviolet and other spectral features from molecular nitrogen that are emitted at the time of tagging. This emission lasts only a few nanoseconds and apparently comes from molecules that are excited but not dissociated by the femtosecond laser pulse [15]
4. Advances in single point measurements The most important advances in the measurement of single point properties have involved the further development of CARS technologies. As noted earlier, CARS has the capability of capturing temperature and species information at the point where the lasers intersect. The CARS process can be separated into an initial step, which drives the selected molecules in the sample volume into a coherent oscillating state using a pair of lasers, and a probe that scatters coherently from the driven molecules in the volume, is frequency and phase shifted by this coherent process, and produces the CARS beam. For high-speed air applications, CARS is useful for measurements of temperature and nonequilibrium conditions at a point [47]. For combustion and SCRAM jet applications, its ability to sample species is of central importance. Much recent work has focused on methods to suppress background, improve single shot performance, and increase the sample acquisition rate. The most difficult background signal is from a similar third-order nonlinear process involving electronic resonances that occur simultaneously with CARS. The coherence associated with that background process is very short lived, so methods that use femtosecond lasers to take advantage of the longer coherence lifetime associated with the desired process have been developed. Very short-pulsed lasers have the great advantage of producing very high intensities with low energy pulses. The nonlinear process that leads to the CARS signal requires high intensity pulses, so with femtosecond lasers operation at high repetition rates becomes possible with practical laser systems. A very successful approach to background suppression and high-resolution signal generation uses femtosecond lasers to drive the coherence and a time delayed picosecond laser probe optimized in shape and delay to suppress the nonresonant background [42]. The femtosecond lasers couple to all the appropriate molecular states of interest and the bandwidth of the picosecond probe laser is broad enough to enable broadband CARS. This approach now promises to allow kHz rate measurements of temperature [40].
5. Advances in integrated path measurements
Fig. 6. FLEET lines written at the exit of a vertical Mach 2.6 overexpanded air jet. Lines are imaged at sequential 1 microsecond time intervals following the tagging of a straight line just above the exit.
Recent work has demonstrated the utility of the wavelength modulation spectroscopy in oxygen for measurements in wind tunnels [23,24], and this approach has the distinction of successful development for flight testing as part of the Hypersonic International Flight Research Experimentation (HIFiRE) 1 experimental package [8,7]. The wavelength modulation approach is seeing wide applications for combustion systems where the measurement of water vapor and carbon monoxide are also of interest. This approach uses diode lasers and is particularly attractive because of the low power requirements and efficient packaging associated with those laser systems.
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎
6
6. Advances in particle imaging velocimetry PIV has proven to be a very effective method for the measurement of velocities in a plane, and that success has motivated recent work extending it to the measurement of threedimensional velocity vectors, volume fields and to its application in hypersonic facilities. These efforts have been facilitated by the development of high pixel density cameras, the increasing pulse energy available for the illumination of larger volumes and the development of the pulse burst laser. Stereoscopic PIV [46] uses a pair of cameras to follow each particle's motion in three dimensions. This requires that the laser sheet be made relatively thick so that the particles are not lost as they move in the out-of-plane direction. The accuracy of the measurement is limited in the outof-plane direction viewing angle separation of the cameras. Another approach to capture three-dimensional PIV and simultaneous volumetric data is based on a repeatedly swept laser. The concept for this was demonstrated in water [12], but with new pulse burst laser and high-speed camera technologies, it has been extended to air flows [53]. In this case the laser is operated at 500,000 pulses per second and rapidly swept through the volume. Images are captured using a DRS Hadland Ultra68 intensified highspeed camera which yields 68 frames with 220 220 pixel resolution in 136 μs. By cycling twice, the cameras capture two time displaced volumetric data sets containing information on the displacement of all the particles within the scanned volume. Reduction of the data provides the full three-dimensional flow field velocity. Extension of PIV to hypersonic flows has also been a priority and a difficult task due to the requirement for very small particles in order to avoid problems with particle lag and scattering from walls, which obscures the PIV signal in just the region where the data are the most important. Recent success has been achieved ([57]) at Mach 7.4 through careful seeding and masking of wall scattering, and proper selection of data analysis algorithms.
7. Summary The field of high-speed diagnostics of air has added many new concepts and expanded previously existing approaches during the last decade or so, leading to the potential for detailed measurements of highly complex flows. Many of these new approaches have been enabled by new developments associated with laser and camera technologies. For example, these include the frequency tunable pulse burst lasers, high pulse energy femtosecond lasers, and multiple image storage fast camera systems. Other advances reflect continued development of methods that were previously proven, but that are now becoming more versatile and are being demonstrated as reliable instruments for flow field measurements. The first incorporation of an optical diagnostic into hypersonic flight occurred during this time, and doubtless that is just a taste of what we can expect over the next decade. The reduction of laser cost, size and weight makes transportable systems more available, providing new opportunities to move concepts that have been proven in small-scale laboratory settings out into the field. The scale of even very complex systems is being reduced to a size that may be practical for flight within the next five years or so.
Acknowledgments The Air Force Office of Scientific Research under Dr. John Schmisseur has supported the recent work at Princeton. Over the past several decades, the development of advanced laser diagnostics has been strongly supported by the Air Force Office of
Scientific Research. That support has led to the successful implementation of these diagnostics in laboratory facilities and has laid the foundation for many of the new advances reported here.
References [1] Adrian RJ. Particle-imaging techniques for experimental fluid mechanics. Annu Rev Fluid Mech 1991;23(1):261–304 (1991). [2] Bathel BF, Danehy PM, Inman JA, Jones SB, Stephen B, Ivey CB, et al. Velocity profile measurements in hypersonic flows using sequentially imaged fluorescence based molecular tagging. AIAA J 2011;49(9):1883–96 (Sep 2011). [3] Beck W, Stockman E, Zaidi S, Miles R. 2006, Rayleigh scattering measurements for obtaining spatially-resolved absolute gas densities in large-scale facilities, AIAA-2006-835, AIAA 44th Aerospace Sciences Meeting and Exhibit, Reno, Nevada, (January 9–12, 2006). [4] Bershader D, Prakash SG, Huhn F. 1976, Improved flow visualization by use of resonant refractivity, AIAA Paper 76-0071 (1976). [5] Boedeker LR. Velocity measurement by H2O photolysis and laser-induced fluorescence of OH. Opt Lett 1989;14(1989):473–5. [6] Bookey P, Wyckham C, Smits AJ. Experimental investigation of mach 3 shock wave turbulent boundary layer interactions. AIAA; 2005; 2005–4899. [7] Brown MS, Barhorst TF. Mass capture experiment onboard HIFiRE Flight 1. AIAA; 2011; 2011–359. [8] Chen S-J, Paige ME, Silver JA, Williams S, Barhorst T. Laser-based mass flow rate sensor onboard HIFiRE Flight. 1. AIAA; 2009; 2009–5065. [9] Cummings EB. Laser-induced thermal acoustics: simple accurate gas measurements. Opt Lett 1994;19(1994):1361. [10] Dam N, Klein-Douwel RJH, Sijtsema NM, ter Meulen JJ. Nitric oxide flow tagging in unseeded air. Opt Lett 2001;26:36–8 (2001). [11] Danehy PM, O'Byrne S, Houwing AFP, Fox JS, Smith DR. Flow-tagging velocimetry for hypersonic flows using fluorescence of nitric oxide. AIAA J. 2003;41(2):263–71 (2003). [12] Delo C, Smits AJ. Volumetric visualization of coherent structure in a low reynolds number turbulent boundary layer. Int J Fluid D 1997;1(3). [13] Durst F, Melling A, Whitelaw JH. Principles and practice of laser doppler anemometry. London: Academic Press; 0-12-225250-0 (1976). [14] Eckbreth AC, Anderson TJ. Dual broadband CARS for simultaneous, multilple species measurements. Appl Opt 1985;24(16):2731–6 (1985). [15] Edwards MR, Dogariu A, Miles RB. 2013, Simultaneous temperature and velocity measurement in unseeded air flows with FLEET, AIAA 2013-0043, 51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition, (2013). [16] Havener AG, Radley Jr. RJ. Turbulent boundary layer flow separation measurements using holographic interferometry. AIAA J 1974;12(8):1071–5 (August 1974). [17] Hiller B, Booman RA, Hassa C, Hanson RK. Velocity visualization in gas flows using laser-induced phosphorescence of biacetyl. Rev Sci Inst 1984;55 (1984):1964–7. [18] Huberwalchli P, Nibler JW. CARS spectroscopy of molecules in supersopnic free jets. J Chem Phys 1982;76(1):273–84 (1982). [19] Jiang N, Bruzzese J, Patton R, Sutton J, Yentsch R, Gaitonde DV, et al. NO PLIF imaging in the cubrc 48-inch shock tunnel. Exp Fluids 2012;53(6):1637–46 (Dec 2012). [20] Koochesfahani M. 1999, Molecular tagging velocimetry (MTV): progress and applications, Paper #AIAA 99-3786, Invited (1999). [21] Krehl POK. History of shock waves, explosions and impact: a chronological and biographical reference. Springer; 3540206787 (2008). [22] Lempert WR, Wu, P-F, Zhang B, Miles RB, Lowrance JL, Mastrocola V, et al. 1996, Pulse burst laser system for high-speed flow diagnostics, Paper #AIAA96-0179, AIAA 34th Aerospace Sciences Meeting, Reno, Nevada, (January 15– 18, 1996). [23] Lyle KH, Jeffries JB, Hanson RK. Diode-laser sensor for air-mass flux 1: design and wind tunnel validation. AIAA J 2007;45(9):2204–12 (2007). [24] Lyle KH, Kent H, Jeffries JB, Hanson RK. Diode-laser sensor for air-mass flux 2: nonuniform flow modeling and aeroengine tests. AIAA J 2007;45(9):2213–23 (2007). [25] Lozano A, Yip B, Hanson RK. Acetone – a tracer for concentration measurements in gaseous flows by planar laser-induced fluorescence. Exp Fluids 1992;13(6):369–76 (October 1992). [26] Mach E, Salcher P. Optische untersuchungen der luftstrahlen. Sitzungsber. Akad. Wiss. Wien 98 (Abth. IIa) 1889;1889:1303–9. [27] Maker PD, Terhune RW. Study of optical effects due to an induced polarization third order in the electric field strength physical review. Physical review 1965;137(3A):801–18 (1965). [28] Mansfield DK, Howard P, Luff JD, Miles R, Brown G, Girgis I, et al. 2004, Summary of the e-Beam coupled 1 MW radiatively-driven hypersonic wind tunnel experiments, AIAA-2004-1134, 42nd AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada (January 5–8, 2004). [29] McDaniel JC, Hiller B, Hanson RK. RK. Simultaneous multiple-point velocity measurements using laser-induced iodine fluorescence. Opt Lett 1983;8 (1):51–3 (1983). [30] McGuire B, Muntz EP, Mallin JR. Visualization technique for low density flows. IEEE T Aero Elec Sys 1967;AES-3(2):321–6 (March 1967).
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎ [31] Meyers JF. Development of doppler global velocimetry as a flow diagnostic tool. Meas Sci Technol 1995;6(6):769–83 (June 1995). [32] Michael J, Edwards MR, Dogariu A, Miles RB. Femtosecond laser electronic excitation tagging for quantitative velocity imaging in air. Appl Opt 2011;50 (5158):2011. [33] Miles RB. Resonant doppler velocimeter. Phys Fluids 1975;18:751 (June 1975). [34] Miles RB, Connors JJ, Markovitz EC, Howard PJ, Roth GJ. Instantaneous profiles and turbulence statistics of supersonic free shear layers by Raman excitation þlaser-induced electronic fluorescence (RELIEF) velocity tagging of oxygen. Exp Fluids 1989;8(1989):17–24. [35] Miles RB, Forkey JN, Lempert WR. 1992, Filtered Rayleigh scattering measurements in supersonic/hypersonic facilities, Paper #AIAA-92-3894, AIAA 17th Aerospace Ground Testing Conference, Nashville, TN (July 6–8, 1992). [36] Miles R, Forkey J, Finkelstein N, Lempert W. 1994, Precision whole-field velocity measurements with frequency-scanned filtered Rayleigh scattering, Proceedings of the 7th International Symposium on Applications of Laser Techniques to Fluid Mechanics, Lisbon, Portugal (July 11–14, 1994). [37] Miles RB, Yalin A, Tang Z, Zaidi S, Forkey J. Flow field imaging through sharpedged atomic and molecular notch filters. J Meas Sci Technol 2001;12:442–51 (April 2001). [38] Miles RB, Lempert W, Forkey J. Laser Rayleigh scattering. J Meas Sci Technol 2001;12:R33–51 (Invited, Published by the Institute of Physics Publishing, Bristol, England (May 2001). [39] Miles RB, Mansfield DK, Girgis I, Brown GL, Lipinski R, Lockner T. 2004, RDHWT/MARIAH II energy addition modeling and experiments review, AIAA2004-2480, Joint AIAA Aerodynamic Measurement Technology and Testing Conference and AIAA Ground Test Conference, Portland Oregon, (June 28-July 1, 2004). [40] Miller JD, Roy S, Slipcheno MN, Gord JR, Meyer TR. Single-shot, gas phase thermometry using pure rotational hybrid femtosecond/picosecond cohrent anti-stokes Raman scattering. Opt Express 2011;19(16):15627–40 (2011). [41] Paul PH, Lee MP, Hanson RK. Molecular velocity imaging of supersonic flows using pulsed planar laser-induced fluorescence of NO. Opt Lett 1989;14 (9):417–9 (May 1, 1989). [42] Pestov D, Murawski RK, Ariunbold GO, Wang X, Zhi M, Sokolov AV, et al. Optimizing the laser pulse configuration for coherent Raman spectroscopy. Science 2007;13:265–8 (April 2007).). [43] Philippe Louis C, Hanson Ronald K. Laser diode wavelength-modulation spectroscopy for simultaneous measurement of temperature, pressure, and velocity in shock-heated oxygen flows. Appl Opt 1993;32(30):6090–103.
7
[44] Pitz RW, Brown TM, Nandula SP, Skaggs PA, DeBarber PA, Brown MS, et al. Unseeded velocity measurement by ozone tagging velocimetry. Opt Lett 1996;21:755–7 (1996). [45] Poggie J, Erbland PJ, Smits AJ, Miles RB. Quantitative visualization of compressible turbulent shear flows using condensate-enhanced Rayleigh scattering. Exp Fluids 2004;37(3):438–54 ( September 2004). [46] Prasad AK. Stereoscopic particle image velocimetry. Exp Fluids 2000;29(103– 16):2000. [47] Roy S, Meyer TR, James R, Gord JR. Broadband coherent anti-stokes Raman scattering spectroscopy of nitrogen using a picosecond modeless dye laser. Opt Lett 2005;30(23):3222–4 (2005). [48] Schafer, Schmidt W, Volze J. Organic dye solution laser. Appl Phys Lett 1966;9 (1966):306–9. [49] Seasholtz RG, Buggele AE, Reeder MF. Flow measurements based on Rayleigh scattering and fabry perot interferometer. Opt Lasers Eng 1997;27(6):543–70 (August 1997). [50] Sorokin PP, Lankard JR. Stimulated emission observed from an organic dye, chloro-aluminum phthalocyanine. IBM J Res Dev 1966;10(1966):162–3. [51] Sanchez-Gonzalez R, Srinivasan R, Bowersox RDW, North SW. Simultaneous velocity and temperature measurements in gaseous flow fields using the VENOM technique. Opt Lett 2011;36(196-98):2011. [52] Thurow BS, Lynch K. Development of a high speed three dimensional flow visualization technique. AIAA J 2009;47(12):2857–65 (December 2009).). [53] Thurow BS, Lynch KP, Williams ST, Melnick MB. 2010 “3-D flow imaging using a MHz-rate pulse burst LASER system, 15th Int Symp on Applications of Laser Techniques to Fluid Mechanics Lisbon, Portugal, 05-08 (July, 2010). [54] Trolinger JD. Multiple exposure holography of time-varying three- dimensional fields. Appl Opt August 1968;7(8):1640 (1968). [55] Weinstein LM. Large field high brightness focusing schlieren system. AIAA J 1993;331(7):1250–5 (July 1993). [56] Westerweel J. Fundamentals of digital particle image velocimetry. J Meas Sci http://dx.doi.org/10.1088/0957-0233/8/12/002 Technol 1997;8(12):1379. (1997). [57] Williams OJ, Smits AJ. 2012, Application of PIV to the measurement of hypersonic turbulence, 16th Int Symp on Applications of Laser Techniques to Fluid Mechanics Lisbon, Portugal, 09-12 (July, 2012). [58] Zimmermann M, Miles RB. Hypersonic helium flow field measurements with the resonant doppler velocimeter. Appl Phys Lett 1980;37:885 (November 1980).
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
Contents lists available at ScienceDirect
Progress in Aerospace Sciences journal homepage: www.elsevier.com/locate/paerosci
Optical diagnostics for high-speed flows Richard B. Miles 1,n Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ 08544, USA
art ic l e i nf o
a b s t r a c t
Article history: Received 25 June 2014 Accepted 11 September 2014
Since 2000 there has been a revolution in diagnostics of high-speed air flows. The foundations for this revolution were laid over the past few decades, but with the development of new short pulse and pulse burst laser technologies, higher laser powers and higher pulse energies, new high-speed cameras, better laser control and improved detection and laser delivery methodologies, many very effective new capabilities have been developed. Newly developed methods for molecular tagging velocimetry provide high fidelity visualization of transport properties and may be extended to simultaneous temperature measurements. Rapid field imaging with frequency tunable pulse burst lasers shows instantaneous flow structure and complex boundary and mixing interactions. Extending these pulse burst concepts to swept volumetric imaging is very promising for full volumetric data collection. Fast wavelength modulation spectroscopy follows real-time flow variation, and three-dimensional particle imaging extends particle imaging velocimetry to volumetric data acquisition. & 2014 Elsevier Ltd. All rights reserved.
Keywords: Diagnostics High speed flows Flow imaging
1. Introduction and retrospective The use of optical diagnostics for the study of high-speed flows dates back to the 1800s when shadowgraphs and schlieren yielded images of the bow shocks, Mach disks and other structures associated with high-speed projectiles and supersonic flows [21] (see Fig. 1). Notable progress in optical diagnostics through the twentieth century that did not involve laser technologies primarily focused on further development of schlieren and shadowgraph for high sensitivity and high-resolution imaging of flow structure. These approaches provided good resolution of large-scale structures but suffered from integration over the full optical path length, so details of turbulent boundary layers, curved and unsteady shocks and mixing structures were not well resolved. The introduction of an electron beam [30] overcame that problem since electron beams could be spatially collimated and electronically swept, providing luminous cross sections of shock and boundary layer structure. However, electron beams are limited to low density flows due to electron scattering, and they are very difficult to integrate into a test facility. Focusing schlieren [55] provided a method for imaging flow structure over a reduced integrated path length in higher density flows. Since the invention of the laser in 1960, the continuing evolution of optical flow diagnostics has been driven in large part by everincreasing laser and camera capabilities. The very first laser invented, the pulsed ruby laser, provided high energy and excellent coherence,
n
Tel.: þ 1 60 925 8 5131. 1 Robert Porter Patterson, Professor of Mechanical and Aerospace Engineering, Fellow AIAA.
which enabled the development of interferometric methods to measure flow field properties, such as flow velocity using seed particles [54] and the imaging of boundary layer structure using density variations [16]. It was not until after the invention of the frequency tunable laser in 1966 [50,48] that atomic and molecular spectroscopy could be utilized for diagnostics. The tunable dye lasers only operated efficiently in the visible portion of the spectrum where the air is highly transparent, so for these new applications, flow seeding became important. Initially seeding with sodium provided planar imaging of flow cross sections [33] and enhanced schlieren [4]. These advances were accomplished by tuning the laser either onto a resonance or near a resonance and utilizing the laser-induced fluorescence for imaging planar cross sections or the enhanced index of refraction for higher sensitivity schlieren. Tuning the lasers provided methods for imaging and measurement of velocity fields by taking advantage of the Doppler shift associated with the motion of the gas [58]. Due to the reactivity of sodium with air, these experiments were carried out in either helium or nitrogen flows. Imaging and interferometry in these early experiments were done with conventional hard film. Molecular iodine was subsequently used for laser-induced flow imaging [29] since it has spectral features throughout the visible and does not react with air. Later, tunable ultraviolet lasers became available through frequency up conversion of nanosecond laserdriven pulsed dye lasers, and nitric oxide [41] and acetone [25] became the preferred molecular species for seeding. CCD array and intensified CCD cameras became available and provided high sensitivity, time gating and convenient data processing capabilities. The possibility of using nonlinear optical methods for flow diagnostics became credible with higher energy, frequency tunable
http://dx.doi.org/10.1016/j.paerosci.2014.09.007 0376-0421/& 2014 Elsevier Ltd. All rights reserved.
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
2
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Fig. 1. Image of an underexpanded sonic jet taken with schlieren photography [26]
pulsed lasers and led to temperature and species measurements by Coherent Anti-stokes Raman Scattering (CARS), temperature and velocity measurements by Laser Induced Thermal Anemometry (LITA) and Molecular Tagging Velocimetry (MTV). The CARS approach [27,18] brings two or three pulsed laser beams together at a point and, through a resonant nonlinear interaction, generates a new laser beam whose intensity is determined by the properties of the gas at that point. CARS has the capability of measuring species concentrations as well as temperature. For that reason it is also of great interest for combustion studies. Originally CARS required that at least one of the lasers be scanned in frequency to acquire the data, and that meant that CARS could only be used for time averaged measurements. Broadband and later dual broadband CARS [14] solved that problem by replacing the scanned laser with a broadband laser and separating the multiple CARS signal frequencies that were simultaneously generated with a spectrometer. This allowed CARS to capture full spectral data over a limited band in a single pulse. LITA [9] was a similar local measurement approach. It used a pair of focused lasers to produce a localized thermal grating into the air through nonlinear mixing, and a probe laser to follow the motion of the acoustic waves created by that grating as they interfered with each other. This provided a local measure of the temperature though the speed of sound and a measure of the flow velocity through frequency offsets associated with the flow motion. MTV [20] introduced a line or array of lines into the flow and tracked them in time as they moved, providing a measure of both the velocity and the flow velocity structure. The first MTV concept implemented in unseeded air was Raman Excitation þLaser Induced Electronic Fluorescence (RELIEF) [34], which used three laser beams – two to drive the oxygen into the vibrational state through stimulated Raman excitation, and one to interrogate the displaced line or pattern by laser-induced fluorescence. Its great feature was that it did not require seeding of the air with other molecular species and produced negligible perturbation. It worked well because of the relatively long lifetime of the oxygen vibrational state (many microseconds even in humid air). It was limited by the complexity of the laser systems. Its success led to the development of other approaches including laser-induced Ozone Tagging Velocimetry (OTV) [44], which was also used as a tag in unseeded air. In this case, the ozone was created by a chemical reaction following laser-induced dissociation of molecular oxygen. The motion was tracked by subsequent laser-induced dissociation of the ozone and imaging of the fluorescence from the excited molecular oxygen fragment. Other MTV approaches for air developed before the 2000 used seed molecules and included biacetyl [17] and water vapor [5]
Single mode, frequency tunable lasers utilizing injection locking enabled the development of molecular, atomic and etalon filtered technologies, permitting strong suppression of background scattering [35], imaging of air temperature, velocity and density (Filtered Rayleigh Scattering [36]) and velocity imaging by Doppler shifted particle imaging through an iodine filter (Doppler Global Velocimetry [31]) as well as velocity and temperature imaging of Doppler shifted Rayleigh scattering through an etalon [49]. Single mode tunable diode lasers derived from the communications industry and augmented by wavelength modulation technology have also opened the door to diagnostic methods for air based on direct absorption spectroscopy using very weak near infrared lines in molecular oxygen [43]. This concept has been successfully implemented for density, velocity and temperature measurements based on the measurement of extinction, line shifts and line broadening. Particles have been used for centuries to observe flows, but the development of laser provided a method for quantitative measurement through instantaneous holographic imaging and other interferometric methods. Much early work focused on Laser Doppler Anemometry (LDA) [13] with continuous lasers for one or two component point measurements of flow velocity, in which two laser beams intersected at the sample point and the scattering of the particle as it moved through the interference pattern which was created provided the measure of velocity. With four crossing beams, two velocity components could be measured. The development of high power nanosecond pulsed lasers enabled imaging of time frozen particle fields and this led to particle imaging velocimetry (PIV) [1], where the two-dimensional velocity field was measured by the displacement of the particles captured with double pulsed laser systems. Digital PIV [56] was enabled by the development of highresolution CCD cameras and eliminated the need for hard film. Thus at the beginning of the twenty-first century many capabilities existed for optical diagnostics of high-speed flows. Since that time further development and implementation of these capabilities have occurred and laser technology has significantly advanced, enabling new approaches. In addition to achieving higher pulse energy, better reliability and higher efficiency lasers, optical fiber technologies, new cameras, frequency tunable pulse burst lasers and sub-picosecond lasers have opened up new possibilities for diagnostics. With these tools major advances have been made in high-speed imaging, molecular flow tagging, wavelength modulation spectroscopy, Particle Imaging Velocimetry, Coherent Antistokes Raman Scattering and Rayleigh scattering.
2. Imaging Laser Rayleigh scattering is the strongest non-resonant light scattering process available for air measurements, but the low scattering cross section of air molecules has made its use for highspeed diagnostics challenging and only recently practical with high energy pulsed lasers and high sensitivity, time gated cameras. It is best applied in free jet facilities where background scattering can be minimized. An important application of Rayleigh scattering in an free jet of air was the evaluation of the Mariah II/Radiatively Driven Hypersonic Wind Tunnel concept [39]. Those tests were undertaken for the validation of computational models of an electron beam heated hypersonic ground test facility and were conducted at Sandia National Laboratory using their 1 MW Hawk electron beam facility [28]. The configuration for the tests is shown in Figs. 2 and 3. The 1 MW electron beam is steered and focused into the nozzle from downstream using a carefully contoured magnetic field and the Rayleigh imaging is performed with a frequency doubled Nd:YAG laser focused to a thin sheet along the center line of the flow at the exit of the nozzle, providing
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎
3
Fig. 2. Setup for the Rayleigh experiments on the electron beam coupled MARIAH II/Radiatively Driven Hypersonic Wind Tunnel at Sandia National Laboratory.
Fig. 3. Time sequenced Rayleigh images of the density at the exit of the MARIAH II/RDHWT nozzle before and during the electron beam energy addition.
an image of the instantaneous density cross section. As the electron beam deposits energy into the core of the flow inside the nozzle, the total enthalpy is more than doubled. The enthalpy deposition leads to a reduced density in the core of the flow, and the ensuing density profile together with spatially localized temperature and velocity measurements provide a quantitative measure of the energy deposited. Fig. 4 shows the Rayleigh images of the evolution of that density profile starting before the electron beam is turned on and continuing until equilibrium is reached 300 μs later. The exit jet is an over expanded supersonic free jet at the beginning of the energy addition. For accurate measurements of the density, images such as these need to be carefully calibrated and background noise must be subtracted [3]. Filtered Rayleigh Scattering enhances the capability of laser Rayleigh scattering through the use of an atomic or molecular gas with a strongly absorbing sharp spectral line to filter the scattering. A cell with that gas is placed in front of the camera, and a narrow line width laser tuned in the vicinity of the strong absorption line is used to illuminate the flow. The filter suppresses background scattering, but permits light that has been frequency shifted by thermal, acoustic or convective motion to pass through. By scanning the laser, this filter can provide for measurement of temperature and pressure [38]. Single shot imaging of temperature fields can be acquired if the pressure is known [37]. The ability of Filtered Rayleigh Scattering to capture cross-sectional cuts of boundary layer structure in supersonic and hypersonic flows has produced detailed images of boundary layer behavior in the vicinity of shockinduced separation [6]. Image contrast is greatly enhanced by seeding the flow with about 1% of CO2 gas [45], which condenses, forming a nanoscale particle fog in the cold core of the flow. The
CO2 condensation highlights the outer portion of the boundary layer where the temperature rises to the sublimation temperature. This imaging capability is further enhanced by the pulse burst laser [22]. Because it is based on a master oscillator power amplifier design (MOPA) the laser is naturally single frequency (single mode) and frequency tunable over a limited range due to the diode pumped Nd:YAG master oscillator. The master oscillator operates as a continuous laser and only after pre-amplification is the laser beam temporally chopped into a pulse burst and passed through the power amplifiers. This feature makes it especially useful for applications such as Filtered Rayleigh Scattering that require narrow line width and tunability over a limited range. The very rapid response of the CO2 nanoparticle sublimation and condensation provides a clear set of images of the time evolving shock wave boundary layer structure as shown in Fig. 4, where the pulse burst laser has been used to acquire sequential images of the boundary layer driven fluctuations of a separated shock in the vicinity of a 151 ramp at 2 μs intervals. By tuning the laser wavelength relative to the iodine filter absorption edges, high and low velocity features can be highlighted as shown in the figure. Images were taken with a MHz rate camera with 30 frames of on board storage designed by Princeton Scientific Instruments. The initial design of the pulse burst laser had a limited time window of 100 μs or so over which the pulses could be generated based on the gain time of the flashlamp pumped power amplifiers. Recent work has significantly extended that range through precision-controlled diode pumping of the power amplifiers, and increased the overall energy of the pulse burst. These advances have enabled the dynamic imaging of lower frequency boundary layer instabilities and flow phenomena and extended the utility of the pulse burst laser to processes such as combustion ignition that occur over longer time intervals [51]. Extension of the spectral frequency range of the pulse burst laser has been achieved with the addition of optical parametric oscillators/amplifiers and frequency up conversion crystals, and now the capability for rapid planar laser-induced fluorescence imaging of nitric oxide [19,2] and other molecular species has been demonstrated. By combining the pulse burst laser with fast beam sweeping technologies, and lenslet array cameras, it is now being extended to full threedimensional data acquisition [52].
3. Advances in molecular tagging Molecular flow tagging has developed significantly since 2000. The Air Photolysis And Recombination Tracking (APART)
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
4
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Fig. 4. Filtered Rayleigh Scattering, 500 KHz rate images of a shock wave boundary layer interaction upstream of a 15 degree wedge. The presence of CO2 condensate nanoparticles in the low temperature core of the flow provides the contrast. Columns 2 and 3 are with the laser tuned to highlight high and low velocity features. (Flow is from right to left.)
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎
technique [10] uses a UV laser to dissociate oxygen and form nitric oxide, which is later imaged by laser-induced fluorescence. If nitric oxide is seeded into the flow or already present from combustion processes, then it can be used for molecular tagging in high-speed environments by taking advantage of the lifetime of the laserinduced fluorescence [11]. Due to quenching, the fluorescence lifetime is short, but at the low densities and high speeds associated with hypersonics, the lifetime is long enough to allow a time-gated camera to image the delayed fluorescence and thus the velocity profiles. The Vibrationally Excited NO Monitoring (VENOM) technique uses photodissociation of seeded NO2 to produce vibrationally excited NO, which is subsequently imaged by LIF after a delay [51]. The VENOM method also yields a measure of the temperature from the NO rotational spectrum. The availability of high pulse energy femtosecond lasers has enabled Femtosecond Laser Electronic Excitation Tagging (FLEET) [32]. In this case a 150 fsec, 2 mJ, Ti:sapphire laser is focused into air and dissociates nitrogen molecules throughout the focal zone by a high-order nonlinear interaction. The nitrogen atoms that are formed by this dissociation recombine over an 100 μs time interval, forming nitrogen molecules in the electronically excited B state. Those molecules fluoresce in the red and near-infrared portion of the spectrum, and that fluorescence can be imaged with a time gated, high sensitivity camera. This process is diagrammed in Fig. 5. Since the nitrogen atoms are only formed through the focal region of the excitation laser, they are initially formed along a straight line and act to tag that region. The timedelayed image shows the location to which each segment along that line has moved in the time interval between tagging and interrogation. An important aspect of FLEET is the continued luminosity of the tagged region for tens of microseconds. Thus by using a fast sequentially gated camera, the evolution of the
Fig. 5. Energy level diagram for molecular nitrogen showing the recombination path for nitrogen atoms leading to B to A first positive emission. Dissociation is by a highly nonlinear interaction driven by the 800 nm, 150 fsec laser.
5
tagged elements in the three-dimensional flow can be followed in real time. Fig. 6 shows images of FLEET lines written at the nozzle exit and imaged at one microsecond time delay intervals showing the flow at sequential heights above the exit of an over-expanded Mach 2.6 air jet. The shape of each line gives the instantaneous velocity profile at that location and the displacement from the original line gives a quantitative measure of the velocity. FLEET may also have the capability to measure temperature along the line by relying on the second positive ultraviolet and other spectral features from molecular nitrogen that are emitted at the time of tagging. This emission lasts only a few nanoseconds and apparently comes from molecules that are excited but not dissociated by the femtosecond laser pulse [15]
4. Advances in single point measurements The most important advances in the measurement of single point properties have involved the further development of CARS technologies. As noted earlier, CARS has the capability of capturing temperature and species information at the point where the lasers intersect. The CARS process can be separated into an initial step, which drives the selected molecules in the sample volume into a coherent oscillating state using a pair of lasers, and a probe that scatters coherently from the driven molecules in the volume, is frequency and phase shifted by this coherent process, and produces the CARS beam. For high-speed air applications, CARS is useful for measurements of temperature and nonequilibrium conditions at a point [47]. For combustion and SCRAM jet applications, its ability to sample species is of central importance. Much recent work has focused on methods to suppress background, improve single shot performance, and increase the sample acquisition rate. The most difficult background signal is from a similar third-order nonlinear process involving electronic resonances that occur simultaneously with CARS. The coherence associated with that background process is very short lived, so methods that use femtosecond lasers to take advantage of the longer coherence lifetime associated with the desired process have been developed. Very short-pulsed lasers have the great advantage of producing very high intensities with low energy pulses. The nonlinear process that leads to the CARS signal requires high intensity pulses, so with femtosecond lasers operation at high repetition rates becomes possible with practical laser systems. A very successful approach to background suppression and high-resolution signal generation uses femtosecond lasers to drive the coherence and a time delayed picosecond laser probe optimized in shape and delay to suppress the nonresonant background [42]. The femtosecond lasers couple to all the appropriate molecular states of interest and the bandwidth of the picosecond probe laser is broad enough to enable broadband CARS. This approach now promises to allow kHz rate measurements of temperature [40].
5. Advances in integrated path measurements
Fig. 6. FLEET lines written at the exit of a vertical Mach 2.6 overexpanded air jet. Lines are imaged at sequential 1 microsecond time intervals following the tagging of a straight line just above the exit.
Recent work has demonstrated the utility of the wavelength modulation spectroscopy in oxygen for measurements in wind tunnels [23,24], and this approach has the distinction of successful development for flight testing as part of the Hypersonic International Flight Research Experimentation (HIFiRE) 1 experimental package [8,7]. The wavelength modulation approach is seeing wide applications for combustion systems where the measurement of water vapor and carbon monoxide are also of interest. This approach uses diode lasers and is particularly attractive because of the low power requirements and efficient packaging associated with those laser systems.
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎
6
6. Advances in particle imaging velocimetry PIV has proven to be a very effective method for the measurement of velocities in a plane, and that success has motivated recent work extending it to the measurement of threedimensional velocity vectors, volume fields and to its application in hypersonic facilities. These efforts have been facilitated by the development of high pixel density cameras, the increasing pulse energy available for the illumination of larger volumes and the development of the pulse burst laser. Stereoscopic PIV [46] uses a pair of cameras to follow each particle's motion in three dimensions. This requires that the laser sheet be made relatively thick so that the particles are not lost as they move in the out-of-plane direction. The accuracy of the measurement is limited in the outof-plane direction viewing angle separation of the cameras. Another approach to capture three-dimensional PIV and simultaneous volumetric data is based on a repeatedly swept laser. The concept for this was demonstrated in water [12], but with new pulse burst laser and high-speed camera technologies, it has been extended to air flows [53]. In this case the laser is operated at 500,000 pulses per second and rapidly swept through the volume. Images are captured using a DRS Hadland Ultra68 intensified highspeed camera which yields 68 frames with 220 220 pixel resolution in 136 μs. By cycling twice, the cameras capture two time displaced volumetric data sets containing information on the displacement of all the particles within the scanned volume. Reduction of the data provides the full three-dimensional flow field velocity. Extension of PIV to hypersonic flows has also been a priority and a difficult task due to the requirement for very small particles in order to avoid problems with particle lag and scattering from walls, which obscures the PIV signal in just the region where the data are the most important. Recent success has been achieved ([57]) at Mach 7.4 through careful seeding and masking of wall scattering, and proper selection of data analysis algorithms.
7. Summary The field of high-speed diagnostics of air has added many new concepts and expanded previously existing approaches during the last decade or so, leading to the potential for detailed measurements of highly complex flows. Many of these new approaches have been enabled by new developments associated with laser and camera technologies. For example, these include the frequency tunable pulse burst lasers, high pulse energy femtosecond lasers, and multiple image storage fast camera systems. Other advances reflect continued development of methods that were previously proven, but that are now becoming more versatile and are being demonstrated as reliable instruments for flow field measurements. The first incorporation of an optical diagnostic into hypersonic flight occurred during this time, and doubtless that is just a taste of what we can expect over the next decade. The reduction of laser cost, size and weight makes transportable systems more available, providing new opportunities to move concepts that have been proven in small-scale laboratory settings out into the field. The scale of even very complex systems is being reduced to a size that may be practical for flight within the next five years or so.
Acknowledgments The Air Force Office of Scientific Research under Dr. John Schmisseur has supported the recent work at Princeton. Over the past several decades, the development of advanced laser diagnostics has been strongly supported by the Air Force Office of
Scientific Research. That support has led to the successful implementation of these diagnostics in laboratory facilities and has laid the foundation for many of the new advances reported here.
References [1] Adrian RJ. Particle-imaging techniques for experimental fluid mechanics. Annu Rev Fluid Mech 1991;23(1):261–304 (1991). [2] Bathel BF, Danehy PM, Inman JA, Jones SB, Stephen B, Ivey CB, et al. Velocity profile measurements in hypersonic flows using sequentially imaged fluorescence based molecular tagging. AIAA J 2011;49(9):1883–96 (Sep 2011). [3] Beck W, Stockman E, Zaidi S, Miles R. 2006, Rayleigh scattering measurements for obtaining spatially-resolved absolute gas densities in large-scale facilities, AIAA-2006-835, AIAA 44th Aerospace Sciences Meeting and Exhibit, Reno, Nevada, (January 9–12, 2006). [4] Bershader D, Prakash SG, Huhn F. 1976, Improved flow visualization by use of resonant refractivity, AIAA Paper 76-0071 (1976). [5] Boedeker LR. Velocity measurement by H2O photolysis and laser-induced fluorescence of OH. Opt Lett 1989;14(1989):473–5. [6] Bookey P, Wyckham C, Smits AJ. Experimental investigation of mach 3 shock wave turbulent boundary layer interactions. AIAA; 2005; 2005–4899. [7] Brown MS, Barhorst TF. Mass capture experiment onboard HIFiRE Flight 1. AIAA; 2011; 2011–359. [8] Chen S-J, Paige ME, Silver JA, Williams S, Barhorst T. Laser-based mass flow rate sensor onboard HIFiRE Flight. 1. AIAA; 2009; 2009–5065. [9] Cummings EB. Laser-induced thermal acoustics: simple accurate gas measurements. Opt Lett 1994;19(1994):1361. [10] Dam N, Klein-Douwel RJH, Sijtsema NM, ter Meulen JJ. Nitric oxide flow tagging in unseeded air. Opt Lett 2001;26:36–8 (2001). [11] Danehy PM, O'Byrne S, Houwing AFP, Fox JS, Smith DR. Flow-tagging velocimetry for hypersonic flows using fluorescence of nitric oxide. AIAA J. 2003;41(2):263–71 (2003). [12] Delo C, Smits AJ. Volumetric visualization of coherent structure in a low reynolds number turbulent boundary layer. Int J Fluid D 1997;1(3). [13] Durst F, Melling A, Whitelaw JH. Principles and practice of laser doppler anemometry. London: Academic Press; 0-12-225250-0 (1976). [14] Eckbreth AC, Anderson TJ. Dual broadband CARS for simultaneous, multilple species measurements. Appl Opt 1985;24(16):2731–6 (1985). [15] Edwards MR, Dogariu A, Miles RB. 2013, Simultaneous temperature and velocity measurement in unseeded air flows with FLEET, AIAA 2013-0043, 51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition, (2013). [16] Havener AG, Radley Jr. RJ. Turbulent boundary layer flow separation measurements using holographic interferometry. AIAA J 1974;12(8):1071–5 (August 1974). [17] Hiller B, Booman RA, Hassa C, Hanson RK. Velocity visualization in gas flows using laser-induced phosphorescence of biacetyl. Rev Sci Inst 1984;55 (1984):1964–7. [18] Huberwalchli P, Nibler JW. CARS spectroscopy of molecules in supersopnic free jets. J Chem Phys 1982;76(1):273–84 (1982). [19] Jiang N, Bruzzese J, Patton R, Sutton J, Yentsch R, Gaitonde DV, et al. NO PLIF imaging in the cubrc 48-inch shock tunnel. Exp Fluids 2012;53(6):1637–46 (Dec 2012). [20] Koochesfahani M. 1999, Molecular tagging velocimetry (MTV): progress and applications, Paper #AIAA 99-3786, Invited (1999). [21] Krehl POK. History of shock waves, explosions and impact: a chronological and biographical reference. Springer; 3540206787 (2008). [22] Lempert WR, Wu, P-F, Zhang B, Miles RB, Lowrance JL, Mastrocola V, et al. 1996, Pulse burst laser system for high-speed flow diagnostics, Paper #AIAA96-0179, AIAA 34th Aerospace Sciences Meeting, Reno, Nevada, (January 15– 18, 1996). [23] Lyle KH, Jeffries JB, Hanson RK. Diode-laser sensor for air-mass flux 1: design and wind tunnel validation. AIAA J 2007;45(9):2204–12 (2007). [24] Lyle KH, Kent H, Jeffries JB, Hanson RK. Diode-laser sensor for air-mass flux 2: nonuniform flow modeling and aeroengine tests. AIAA J 2007;45(9):2213–23 (2007). [25] Lozano A, Yip B, Hanson RK. Acetone – a tracer for concentration measurements in gaseous flows by planar laser-induced fluorescence. Exp Fluids 1992;13(6):369–76 (October 1992). [26] Mach E, Salcher P. Optische untersuchungen der luftstrahlen. Sitzungsber. Akad. Wiss. Wien 98 (Abth. IIa) 1889;1889:1303–9. [27] Maker PD, Terhune RW. Study of optical effects due to an induced polarization third order in the electric field strength physical review. Physical review 1965;137(3A):801–18 (1965). [28] Mansfield DK, Howard P, Luff JD, Miles R, Brown G, Girgis I, et al. 2004, Summary of the e-Beam coupled 1 MW radiatively-driven hypersonic wind tunnel experiments, AIAA-2004-1134, 42nd AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada (January 5–8, 2004). [29] McDaniel JC, Hiller B, Hanson RK. RK. Simultaneous multiple-point velocity measurements using laser-induced iodine fluorescence. Opt Lett 1983;8 (1):51–3 (1983). [30] McGuire B, Muntz EP, Mallin JR. Visualization technique for low density flows. IEEE T Aero Elec Sys 1967;AES-3(2):321–6 (March 1967).
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i
R.B. Miles / Progress in Aerospace Sciences ∎ (∎∎∎∎) ∎∎∎–∎∎∎ [31] Meyers JF. Development of doppler global velocimetry as a flow diagnostic tool. Meas Sci Technol 1995;6(6):769–83 (June 1995). [32] Michael J, Edwards MR, Dogariu A, Miles RB. Femtosecond laser electronic excitation tagging for quantitative velocity imaging in air. Appl Opt 2011;50 (5158):2011. [33] Miles RB. Resonant doppler velocimeter. Phys Fluids 1975;18:751 (June 1975). [34] Miles RB, Connors JJ, Markovitz EC, Howard PJ, Roth GJ. Instantaneous profiles and turbulence statistics of supersonic free shear layers by Raman excitation þlaser-induced electronic fluorescence (RELIEF) velocity tagging of oxygen. Exp Fluids 1989;8(1989):17–24. [35] Miles RB, Forkey JN, Lempert WR. 1992, Filtered Rayleigh scattering measurements in supersonic/hypersonic facilities, Paper #AIAA-92-3894, AIAA 17th Aerospace Ground Testing Conference, Nashville, TN (July 6–8, 1992). [36] Miles R, Forkey J, Finkelstein N, Lempert W. 1994, Precision whole-field velocity measurements with frequency-scanned filtered Rayleigh scattering, Proceedings of the 7th International Symposium on Applications of Laser Techniques to Fluid Mechanics, Lisbon, Portugal (July 11–14, 1994). [37] Miles RB, Yalin A, Tang Z, Zaidi S, Forkey J. Flow field imaging through sharpedged atomic and molecular notch filters. J Meas Sci Technol 2001;12:442–51 (April 2001). [38] Miles RB, Lempert W, Forkey J. Laser Rayleigh scattering. J Meas Sci Technol 2001;12:R33–51 (Invited, Published by the Institute of Physics Publishing, Bristol, England (May 2001). [39] Miles RB, Mansfield DK, Girgis I, Brown GL, Lipinski R, Lockner T. 2004, RDHWT/MARIAH II energy addition modeling and experiments review, AIAA2004-2480, Joint AIAA Aerodynamic Measurement Technology and Testing Conference and AIAA Ground Test Conference, Portland Oregon, (June 28-July 1, 2004). [40] Miller JD, Roy S, Slipcheno MN, Gord JR, Meyer TR. Single-shot, gas phase thermometry using pure rotational hybrid femtosecond/picosecond cohrent anti-stokes Raman scattering. Opt Express 2011;19(16):15627–40 (2011). [41] Paul PH, Lee MP, Hanson RK. Molecular velocity imaging of supersonic flows using pulsed planar laser-induced fluorescence of NO. Opt Lett 1989;14 (9):417–9 (May 1, 1989). [42] Pestov D, Murawski RK, Ariunbold GO, Wang X, Zhi M, Sokolov AV, et al. Optimizing the laser pulse configuration for coherent Raman spectroscopy. Science 2007;13:265–8 (April 2007).). [43] Philippe Louis C, Hanson Ronald K. Laser diode wavelength-modulation spectroscopy for simultaneous measurement of temperature, pressure, and velocity in shock-heated oxygen flows. Appl Opt 1993;32(30):6090–103.
7
[44] Pitz RW, Brown TM, Nandula SP, Skaggs PA, DeBarber PA, Brown MS, et al. Unseeded velocity measurement by ozone tagging velocimetry. Opt Lett 1996;21:755–7 (1996). [45] Poggie J, Erbland PJ, Smits AJ, Miles RB. Quantitative visualization of compressible turbulent shear flows using condensate-enhanced Rayleigh scattering. Exp Fluids 2004;37(3):438–54 ( September 2004). [46] Prasad AK. Stereoscopic particle image velocimetry. Exp Fluids 2000;29(103– 16):2000. [47] Roy S, Meyer TR, James R, Gord JR. Broadband coherent anti-stokes Raman scattering spectroscopy of nitrogen using a picosecond modeless dye laser. Opt Lett 2005;30(23):3222–4 (2005). [48] Schafer, Schmidt W, Volze J. Organic dye solution laser. Appl Phys Lett 1966;9 (1966):306–9. [49] Seasholtz RG, Buggele AE, Reeder MF. Flow measurements based on Rayleigh scattering and fabry perot interferometer. Opt Lasers Eng 1997;27(6):543–70 (August 1997). [50] Sorokin PP, Lankard JR. Stimulated emission observed from an organic dye, chloro-aluminum phthalocyanine. IBM J Res Dev 1966;10(1966):162–3. [51] Sanchez-Gonzalez R, Srinivasan R, Bowersox RDW, North SW. Simultaneous velocity and temperature measurements in gaseous flow fields using the VENOM technique. Opt Lett 2011;36(196-98):2011. [52] Thurow BS, Lynch K. Development of a high speed three dimensional flow visualization technique. AIAA J 2009;47(12):2857–65 (December 2009).). [53] Thurow BS, Lynch KP, Williams ST, Melnick MB. 2010 “3-D flow imaging using a MHz-rate pulse burst LASER system, 15th Int Symp on Applications of Laser Techniques to Fluid Mechanics Lisbon, Portugal, 05-08 (July, 2010). [54] Trolinger JD. Multiple exposure holography of time-varying three- dimensional fields. Appl Opt August 1968;7(8):1640 (1968). [55] Weinstein LM. Large field high brightness focusing schlieren system. AIAA J 1993;331(7):1250–5 (July 1993). [56] Westerweel J. Fundamentals of digital particle image velocimetry. J Meas Sci http://dx.doi.org/10.1088/0957-0233/8/12/002 Technol 1997;8(12):1379. (1997). [57] Williams OJ, Smits AJ. 2012, Application of PIV to the measurement of hypersonic turbulence, 16th Int Symp on Applications of Laser Techniques to Fluid Mechanics Lisbon, Portugal, 09-12 (July, 2012). [58] Zimmermann M, Miles RB. Hypersonic helium flow field measurements with the resonant doppler velocimeter. Appl Phys Lett 1980;37:885 (November 1980).
Please cite this article as: Miles RB. Optical diagnostics for high-speed flows. Progress in Aerospace Sciences (2014), http://dx.doi.org/ 10.1016/j.paerosci.2014.09.007i