Photonic Displacement Interferometer

Transcription

1 Photonic Displacement Interferometer Thermomechanical Shock (TMS) and Early Thermostructural Response (TSR) Measurement Applications Scott C. Jones Sandia National Laboratories Radiation Materials Science, Dept Heterodyne Workshop LLNL July 26 1 Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy under contract DE-AC4-94AL85.

2 Introduction Laser interferometry use in thermomechanical response measurements goes back to 1968 (Ref 1) Surface displacement in e beam experiments to measure Gruneisen parameter in 197 (Refs 2-4) Development of Photonic Doppler Velocimeter (PDV) 24 (Ref 5) Adaptation of PDV as a displacement interferometer Examples from disks and cantilevered beams Issues concerning thin specimens Radiation Source Fluence Measurements Z Observation of material change ACKNOWLEDGMENTS Jim Greenwoll (SNL 1344), Todd Simmermacher (SNL 1523) Pat Rose, Mike Lynch (ITT), Doug Reeder, Ed Stretanski, Rich Woodring (Ktech) Capt. Tim Skaar (USAF/DTRA/SNL) 2

3 Fiber Optic Interferometer Probe Pulsed Radiation Shifted Unshifted x Photonic Displacement Interferometer Optical path length difference between light reflected by sample and probe results in an interference phase angle when recombined (arbitrary φ o ) Displacement of sample surface causes phase angle to evolve Recorded intensity I = A + Bcos(4πx/λ + φ o ), A, B approximately constant One fringe (Δφ = 2π) equivalent to.775 micron (λ/2) displacement Positive or negative, nonuniform motion displacement vs time accurate 3

4 Thermomechanical Shock Pressure ~ ργdose Surface Motion Stress wave propagation For our studies, velocities are nonuniform and small (peak values as small as a few m/s) Peak displacements: fraction of one micron to several microns Required VISAR delays (τ) for one-fringe precision are impractical (532 nm air delay) Validity of FT for extracting velocity unclear nonuniform motion, reversals in velocity 4

5 Attributes Immune to electromagnetic noise Non contact/no bond joints to transducer: - Multiple shots on sample, material change/damage Sample surface preparation: little to none (aided by IR wavelength) Probe spot size: > 35 microns (depends on surface quality) Enhances/extends validity of 1-D analysis Present system: 4 channel utilizing 3 mw total power (2W, 155 nm Erbium-fiber laser) DC 3.5 GHz detectors 8 channel, 2 Gs/sec, 5 MHz BW, 1 Mpoints/channel ( 4 msec) recording 5

6 6

7 SPHINX Electron Beam Test Environment SPHINX accelerator configured for electron beam mode (a TSR disc experiment is shown). Typical spectrum and time history is shown from shot ~ 2 MeV end-point ~ 9 ns FWHM Filtered (32 mil Al & 5 mil Ti) dosedepth profiles are shown for Al 661, Al 775, SS34L & Cu. 7

8 Noise Immunity Typical Signal Quality 434 steel, 3 mm (totally stopping) in SPHINX E beam Sanded, machined surface SX257 Steel SX257 Steel Signal (V).5 Center Corner Diode voltage (! ~ 1-6 ) Signal (V).5 Center Corner Diode voltage (! ~ 1-6 ) -.5 1x1-6 2x1-6 3x1-6 4x1-6 Time (sec) Time (sec) 8

9 Fringe Analysis.3 φ o Signal (Volts).2.1.6x1-5.8x1-5 1.x1-5 Time (sec).3 Signal (Volts).2.1 I = A + Bcos (4πx/λ + φ ) Normalize and offset: S = (I-A)/B = cos (4πx/λ + φ ) x = (λ/4π) (cos -1 S φ ) Reversal 6.5x1-6 7.x1-6 Time (sec) 9 Reversals not always obvious makes automation difficult without quadrature recording. Analysis routine is under development

10 Uncertainty Fringe Reconstruction Fringe Reconstruction 1 1 Normalized Signal Normalized Signal -1.6x1-5.7x1-5.8x1-5.9x1-5 Time x x x x1-6 Time 1 ½ Fringe = ¼ wavelength ~.38 µm Rule-of-thumb: Resolve.2.5 fringe Total phase (displacement) accumulates (+/-), but uncertainty resets every ½ fringe Uncertainty is inherently << 1/4 fringe, but φ and reversals near intensity extrema increase it Constancy of interference contrast determines phase resolution uncertainty Affected by surface quality, large displacements, recognition of superposed motions Quadrature recording will reduce these contributors Displacement ~ (N + n) fringes, Fractional uncertainty = δn/(n+n)

11 6.35 mm 661 Al Disk, SPHINX e-beam Low Fluence Center Channels 11 Displacement (cm) Displacement (cm) 2x1-4 1x1-4 2x1-4 1x1-4 -1x1-4 -2x1-4 -3x1-4.5x1-6 1.x x1-6 2.x1-6 Center Channels 1-D motion ends here or earlier x1-6 4x1-6 6x1-6 8x1-6 Three successive shots on SPHINX, monitoring rear surface Center channels on disk (range thick) 1 inch aperture on a 1 ¾ inch disk Captures information for obtaining deposition profile, fluence diagnostic Probe spot size ~ 5 microns enhances validity of 1-D analysis for nonuniform e-beam profiles Peak velocity ~ 7.5 m/sec would require VISAR delay of 35 nsec for one-fringe accuracy of peak (532 nm). Evident that TSR begins before first TMS response is complete (rear surface) Clean TSR measurement within loading region from t =.

12 VISAR Comparison Simulated VISAR results for measured displacement data: v (S(t) S(t-τ))/τ 532 nm air delay of > 3 ns required for one-fringe precision of peak velocity 155 nm delay of ~ 1 ns or greater required for 155 nm Long delays lead to large distortion displacement analysis during τ required anyway 12

13 6.35 mm 661 Al Disk High Fluence Displacement (cm) Reversal not recognized.2x1-5.4x1-5.6x1-5.8x1-5 1.x x1-5 Time (sec) Ch 1, center Ch 3, + 1 mm 13 Three channels on single shot Range thick sample, 25.4 mm aperture Response is not radially symmetric about center TMS response provides loading information across structure, input for response modeling TMS/TSR response for code/model V&V when combined with other diagnostics Reversals can occur at intensity extrema Superimposed motions mask symmetries that help identify reversals Quadrature (2 ch) recording makes reversals unambiguous Requirement for routine analysis of records > ~ 1 microseconds

14 Design Details of TSR Cantilever Beam Experiment SPHINX e-beam Clamp L = 62 mm Single Cantilevered Beam TSR/TMS L/1 L/2 V2 V1 V3 F1/F2 TC1 TC2 w = 12.7 mm Exposure Area (Dia. = 25.4 mm) L tot = mm WD = 28 mm ozoptics Pigtail Focuser Type: LPF-4 OD = 8 mm t = 2. mm (Al) t =.5 mm (34L SS) Fused silica window shields probes Courtesy ITT-AES 2 mm 14

15 .5 mm 34L SS Cantilevered Beam High Fluence SPHINX e-beam x1-4.8x1-4 Ch 1, center Ch 2, -1 mm (clamp) Ch 3, + 1 mm Optical Signal (V) Diode (!1-6 ).6x1-4.4x1-4 Ch 1 Ch 2 Ch 3 Diode x1-7 3.x x1-4 Cantilevered beam is thin (ρt = ~.4 g/cm 2 ).5x1-6 1.x x1-6 Time Jump in Ch 2 displacement is substantial and positive in this example Increase in optical intensity in channel 2 at beam time indicates radiation-induced darkening is not only likely effect occurring Because material is thin, material response to deposition at probed surface should begin instantaneously, complicating interpretation what is real motion, what is other optical effects? 15

16 Considerations for Thin Samples Probe Fused silica window Electronic excitation complex n c (x,t) window and gas, thermal Maxwell s equations Sample L 1 L o Radiation pulse Photoelastic n(x,t) u 2 (t) TMR changes path length u 1 (t) window OPL = L 1 + n(x,t)dx + L o - u 1 (t)dt - u 2 (t)dt + n(x,t)dx gas 16 In addition, any phase change on reflection from sample due to currents, excitations

17 Thermoelastic Calorimetry Application Pulsed Radiation Filter X-ray or e-beam fluence, or source X-ray yield may be obtained using: Source spectrum and flux history Rad transport Absorber model (Γ, c o, ν, Y, ) - elastic behavior best. Material choice 17 Potentially cheaper, faster result than from quartz TEC gauges, immune to pulsed power noise environment

18 PRS Fluence Measurement Z Z1594 SS wire array 2 mm Al 661 T6 absorber 2 probes agree to +/- 1.1% in peak displacement ~ 4 % uncertainty individual channel 53.6 kj yield, 2 ns pulse Yields (kj) Z1594 Z1595 Z1596 PDI PCD Z1596 SS wire array 2 mm Al 661 T6 absorber 25 kj model scaled to 21.9 kj total x-ray yield Low yield, long pulse (4.4 ns) 18

19 Material Change, Evolution Measurement of material response to pulsed radiation using bonded transducers precludes observing (efficiently) changes in response behavior or material modification is a change in observed response due to degradation or failure of the bond? Non-contact measurement of material response in a radiation source capable of repetitive pulsing (without refurbishment) allows observation of the material modification resulting from exposure This capability allows observational assessment of material or component survivability in pulsed radiation environments. 19

20 Material Evolution Poled PZT ceramic, rough ground surface High Fluence Pulse Displacement (cm) 6x1-5 4x1-5 1 st Shot 2 nd Shot 2x1-5 2 Repetitive shots on sample indicates material capability to withstand pulsed radiation deposition Incipient spall signature on 15 th shot Structure, dynamics of damage can be analyzed from record Time (nsec) Fluence sufficient to heat into paraelectric phase on part of deposition profile Gruneisen of depoled material much smaller in FE2 phase than for poled material

21 Material Evolution, PZT PZT ceramic, rough ground surface Repetitive shots on sample indicates material capability to withstand pulsed radiation deposition Incipient spall signature on 15 th shot Structure, dynamics of damage can be analyzed from record 21

22 Summary Very simple interferometer system for multipoint surface displacement measurements Noise immunity in pulsed power environment is enormous advantage Non contact, optical technique allows for monitoring changes in material or structural response due to radiation deposition damage thresholds 1-D measurements of material response possible in small diameter drive sources Early TSR results will provide interesting data for model comparison quadrature recording required for reliable, unambiguous fringe unwrap In-line phase shifter being investigated Laser frequency stability a question 22

23 References 1. R.B. Oswald, Jr., et al, Appl. Phys. Lett. 13, 279 (1968). 2. R.B.Oswald, Jr., et al, Appl. Phys. Lett. 16, 24 (197). 3. F.C. Perry, J. Appl. Phys. 41, 187 (197). 4. F.C. Perry, J. Appl. Phys. 41, 517 (197). 5. O.T. Strand, et al, Velocimetry Using Heterodyne Techniques, in 26 th International Congress on High Speed Photography and Photonics, Vol 558, D.L. Paisley, ed. (SPIE 24). 23