High-resolution Penumbral Imaging on the NIF

Size: px

Start display at page:

Download "High-resolution Penumbral Imaging on the NIF"

Harvey Lindsey
5 years ago
Views:

1 High-resolution Penumbral Imaging on the NIF October 6, 21 Benjamin Bachmann T. Hilsabeck (GA), J. Field, A. MacPhee, N. Masters, C. Reed (GA), T. Pardini, B. Spears, L. BenedeB, S. Nagel, N. Izumi, V. Smalyuk, D. Bradley, J. Kilkenny LLNL-PRES-XXXXXX This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC2-7NA Lawrence Livermore National Security, LLC

2 Motivation high-res ICF-hotspot simulations show complex behavior Simulated 3D hotspot Simulated 2D hotspot image Y (µm) 1 kev Tion iso-surface X (µm) Brian Spears & Dave Munro 2

$stagnation of spherical shock wave D 1 Penumbra Spatial expansion and structure of self emitting plasma Gated x-ray detector r I D 2 Limiting factors: Physical: - Diffraction (typically few µm in$

3 Principle of Penumbral Imaging D : hot spot diameter D 1 : variation in aperture / transition region D 2 : resulting length of penumbra D 3 : detector resolution Pinhole Target D Self emission at stagnation of spherical shock wave D 1 Penumbra Spatial expansion and structure of self emitting plasma Gated x-ray detector r I D 2 Limiting factors: Physical: - Diffraction (typically few µm in detector plane) Technical: - Magnification (1x PDIM, 1x 9-78, 8x DIXI IP, 64x DIXI) - Detector resolution: IP: 9µm, GXD: ~4µm, DIXI: ~27µm - transparency of tapered edge of Aperture => broadening of Penumbra (typically few µm in detector plane) - Dynamic detector range, QE, photon yield and noise B. Bachmann, Rev. Sci. Instrum. 8, 11D614 (214) D. Ress etal, Science 241, no. 4868, pp (1988) 3

4 Proof of Principle First: Simulated 1µm Pinhole Imaging Our imaginary setup: - Pinhole radius: 1µm - Magnification: 1x - Detector resolution: GXD (9x9µm pixel size) 2 source distribution (in target plane) SOURCE PHANTOM 4

5 Proof of Principle First: Simulated 1µm Pinhole Imaging Our imaginary setup: - Pinhole radius: 1µm - Magnification: 1x - Detector resolution: GXD (9x9µm pixel size) source distribution (in target plane) y (mm) Resulting pinhole image (detector plane) x (mm) That is what we would see with pinhole imaging SOURCE PHANTOM Pinhole Image (raytrace) 4

6 Proof of Principle Now: Simulated Penumbral Imaging Our imaginary setup: - Pinhole radius: 14µm - Magnification: 1x - Detector resolution: GXD (9x9µm pixel size) - Smallest distinguishable object in source distribution: 6nm = resulting best possible Penumbral Image Resolution with this setup (limited by detector resolution) source distribution (in target plane) y (mm) resulting Penumbral Image (in detector plane) x (mm) SOURCE PHANTOM Penumbral Image

7 Proof of Principle Simulated Penumbral Imaging Our imaginary setup: - Pinhole radius: 14µm - Magnification: 1x - Detector resolution: GXD (9x9µm pixel size) - Smallest distinguishable object in source distribution: 6nm = resulting best possible Penumbral Image Resolution with this setup (limited by detector resolution) Inverse Radon transform 2 1 source distribution (in target plane). 1 resulting Penumbral Image (in detector plane) reconstructed source distribution (in target plane) y (mm) x (mm) SOURCE PHANTOM Penumbral Image RECONSTRUCTED

8 Proof of Principle reconstruction visualization Reconstruction by inverse Radon transformation source distribution (in target plane) y (mm) resulting Penumbral Image (in detector plane) x (mm) SOURCE PHANTOM Penumbral Image RECONSTRUCTION

9 Proof of Principle GXD point spread function added Still working on including a more realistic point spread function and noise model for the GXD s For now: 4µm FWHM 2D-Gauss: source distribution (in target plane) resulting Penumbral Image (in detector plane) reconstructed source distribution (in target plane) y (mm) x (mm) SOURCE PHANTOM Penumbral Image RECONSTRUCTED 6

10 Proof of Principle additional noise added We expect noise to scale like sqrt(signal): Signal: 6 counts Noise at max signal: 2 counts source distribution (in target plane) resulting Penumbral Image (in detector plane) reconstructed source distribution (in target plane) y (mm) x (mm) SOURCE PHANTOM Penumbral Image RECONSTRUCTED 6

11 Proof of Principle additional noise added We expect noise to scale like sqrt(signal): Signal: 6 counts Noise at max signal: 2 counts 1µm radius Pinhole image y (mm) x (mm) source distribution (in target plane) resulting Penumbral Image (in detector plane) reconstructed source distribution (in target plane) y (mm) x (mm) SOURCE PHANTOM Penumbral Image RECONSTRUCTED 6

12 High-resolution imaging of complex hotspot Simulated hotspot emission with mix Brian Spears & Dave Munro 7

but more work to do 4 source distribution (in target plane) resulting Penumbral Image (in detector plane) 3 2 1

13 Reconstruction of a simulated complex hotspot Penumbral Image Setup: - Magnification: 31x - Pinhole diameter: 22µm - Noise: sqrt(signal) - GXD-Point Spread Function: 4µm FWHM Gauss - Low dynamic range in source Promising, but more work to do 4 source distribution (in target plane) resulting Penumbral Image (in detector plane) y (mm) SOURCE Simulated complex hotspot x (mm) Penumbral Image RECONSTRUCTION 8

14 Can we use DIXI for Penumbral Imaging? Setup: Source-aperture distance: 1mm Aperture-Detector distance: 64mm Detector resolution: ~27um Source for DIXI DIXI 1µm pinhole image 7µm Taking into account all known image degradation effects Simulated Penumbral Image by Terry Hilsabeck 9

15 Simulated Penumbral Image with DIXI setup Setup: Aperture diameter: 1µm Source-aperture distance: 1mm Aperture-Detector distance: 64mm Detector resolution: ~27um Simulated Penumbral Image by Terry Hilsabeck 1

16 Reconstruction of simulated Penumbral Image with DIXI setup Setup: Aperture diameter: 1µm Source-aperture distance: 1mm Aperture-Detector distance: 64mm Detector resolution: ~27um Simulated Penumbral Image by Terry Hilsabeck 1

17 Reconstruction of simulated Penumbral Image with DIXI setup Setup: Aperture diameter: 1µm Source-aperture distance: 1mm Aperture-Detector distance: 64mm Detector resolution: ~27um Limited mainly by pinhole size reconstructed source (in target plane) Simulated Penumbral Image by Terry Hilsabeck Limited mainly by detector resolution (4µm in target plane) 11

GA can manufacture high-quality coded apertures that meet the criteria for Penumbral Imaging High edge-quality s-shaped slit laser-cut by GA Circular large

18 GA can manufacture high-quality coded apertures that meet the criteria for Penumbral Imaging High edge-quality s-shaped slit laser-cut by GA Circular large area Penumbral Pinhole produced by GA (EDM + laser finished) SEM image of a tilted Penumbral Imaging slit pinhole 2 um 1mm For small objects For large objects 12

We successfully demonstrated Penumbral Imaging on the NIF on IP (1x Magnification) setup: - Pinhole diameter: 2mm - Magnification: 1x - Detector: Image plate Spatial resolution:

19 We successfully demonstrated Penumbral Imaging on the NIF on IP (1x Magnification) setup: - Pinhole diameter: 2mm - Magnification: 1x - Detector: Image plate Spatial resolution: Limited by: Image Plate Resolution: 9µm/1=6µm M up => Resolution up Penumbral Image (in detector plane) 2 1 y (mm) x (mm) Penumbral Image Play Video Reconstruction 13

reconstructed source (in target plane) 1 setup: - Pinhole diameter: 2mm - Magnification: 1x - Detector: Image plate -

20 We successfully demonstrated Penumbral Imaging on the NIF on IP (1x Magnification) setup: - Pinhole diameter: 2µm - Magnification: 1x - Detector: Image plate - Spatial resolution: 27µm pinhole image (in target plane) 1 Jets? reconstructed source (in target plane) 1 setup: - Pinhole diameter: 2mm - Magnification: 1x - Detector: Image plate - Spatial resolution: 6µm vs Pinhole Image Reconstructed Penumbral Image (additional blurring yields pinhole image) 14

21 We successfully fielded Penumbral Imaging on the NIF on IP (8x Magnification) Gbar solid sphere target, on Image Plate of DIXI LoS: Penumbral-images Pinhole array Image Plate: Magnification: 8 Raw-Image plate (equatorial view) µm radius (1/e) Highest resolution hotspot image on the NIF to date: 1.µm resolution (limited by IP resolution) vertical position (um) Reconstructed hotspot image horizontal position (um) Reconstruction 1

22 Summary - 1.µm resolution Penumbral Imaging has been successfully fielded on the NIF - Penumbral imaging has potential of significantly improving hotspot imaging in ICF implosions - Manufacturing challenges of quality apertures have been overcome - Many possible applications Reconstructed Penumbral Image N vertical position (um) horizontal position (um) Penumbral Imaging Pinhole Imaging 2 um vs. y (mm) x (mm) we just scratched the surface source distribution (in target plane) reconstructed source (in target plane)

24 Backup 24

Proof of Principle times more noise added (than one would expect) Noise=*sqrt(Signal): Signal: 6 counts Noise at max signal: 12 counts 2 1 source distribution (in target plane) 2.

25 Proof of Principle times more noise added (than one would expect) Noise=*sqrt(Signal): Signal: 6 counts Noise at max signal: 12 counts 2 1 source distribution (in target plane) 2. 2 resulting Penumbral Image (in detector plane) reconstructed source distribution (in target plane) y (mm) x (mm) SOURCE PHANTOM Penumbral Image RECONSTRUCTED 2

Proof of Principle skewing the penumbral image Skewing operation: Resample

This results in a - Shift - Scale - Rotation - Anisotropy of the penumbral

(in target plane) 2 1 2. 2 2 1 1 1 y (mm) 1. 1.. 1 1. 1 1 1 2 1 2 2 1 1 2 SOURCE PHANTOM 2.

26 Proof of Principle skewing the penumbral image Skewing operation: Resample Grid-coordinates: A= 1+u1 u2 u3 1+u4 B= u u6 x* y* x =A y +B; original skewed This results in a - Shift - Scale - Rotation - Anisotropy of the penumbral image source distribution (in target plane) reconstructed source distribution (in target plane) y (mm) SOURCE PHANTOM x (mm) Differences Original-Skewed RECONSTRUCTED 26

The role of the edge While the previously shown deconvolution with a contact radiograph gives approximate information of how the x-rays pass through the

a 3D ray tracing can capture the whole process: Shown to be in first 2um depth Penumbral Images: the width of the Penumbral Images depends on the size of

27 The role of the edge While the previously shown deconvolution with a contact radiograph gives approximate information of how the x-rays pass through the slit, e.g. a 3D ray tracing can capture the whole process: Shown to be in first 2um depth Penumbral Images: the width of the Penumbral Images depends on the size of ~3um Wide at top the Hotspot as well as the x-ray attenuation along the tapered edge Estimated Dimensions of S Slot in Test 4 => Knowledge of actual edge dimensions allows to account for these effects! 4-um Taper ~22 um Exit Partly transmission through material X-rays from Hotspot with size D D 27

28 Height of Cross Section [um] Test 4 White Light Interferometer Cross Section (Trial 4) Thickness [um]

Development of dual MCP x-ray imager for 40 ~ 200 kev region

Development of dual MCP x-ray imager for 40 ~ 200 kev region National ICF Diagnostics Working Group Meeting - October 6-8, 2015 N. Izumi, G. N. Hall, A. C. Carpenter, F. V. Allen, J. G. Cruz, B. Felker,