11D615-4

Thfoin et al.

the use of magnets in front of the spectrometer removes this disturbing signal. V. CONCLUSION

Hard X-ray spectra recorded by TCS-like spectrometers are overlaid by a background noise image. GEANT4 calculations show that this disturbing signal is mainly generated by diffusion of high energy electrons inside the diagnostic. This was confirmed by comparing experimental shots with and without a couple of magnets in front of the spectrometer. Concerning the noise generated by very hard X-rays, using more efficient shielding materials and increasing the shielding thickness will improve the diagnostic SNR. The new spectrometer SPECTIX (Spectromètre PETAL à Cristal en TransmIssion X), in construction for the LMJ/PETAL facility within the PETAL+ project,12, 13 will feature this improved shielding.

Rev. Sci. Instrum. 85, 11D615 (2014)

ACKNOWLEDGMENTS

This work was supported by the French Agency for Research (ANR) through Equipex PETAL+ coordinated by the University of Bordeaux and project TERRE ANR-2011BS04-014. 1 Y.

Cauchois, J. Phys. Rad. 3(7), 320 (1932). DuMond, Ergenisse der exaktenWissenschaften 28, 232 (1955). 3 A. L. Meadocroft, Rev. Sci. Instrum. 79, 113102 (2008). 4 L. T. Hudson, Rev. Sci. Instrum. 73, 2270 (2002). 5 J. F. Seely, High Energy Dens. Phys. 3, 263 (2007). 6 C. I. Szabo, Eur. Phys. J. Spec. Top. 169, 243 (2009). 7 J. F. Seely, Rev. Sci. Instrum. 81, 10E301 (2010). 8 J. Allison et al., IEEE Trans. Nucl. Sci. 53(1), 270 (2006). 9 J. F. Seely et al., High Energy Density Phys. 7, 150 (2011). 10 K. U. Akli, J. Instrum. 5, P07008 (2010). 11 C. D. Chen, Rev. Sci. Instrum. 79, 10E305 (2008). 12 D. Batani et al., Acta Polytec. 53(2), 103–109 (2013). 13 D. Batani et al., Phys. Scr. 2014, 014016 (2014). 2 J.

REVIEW OF SCIENTIFIC INSTRUMENTS 85, 11D616 (2014)

Mass-ablation-rate measurements in direct-drive cryogenic implosions using x-ray self-emission imagesa) A. K. Davis,b) D. T. Michel, S. X. Hu, R. S. Craxton, R. Epstein, V. N. Goncharov, I. V. Igumenshchev, T. C. Sangster, and D. H. Froula Laboratory for Laser Energetics, University of Rochester, Rochester, New York 14636, USA

(Presented 2 June 2014; received 1 June 2014; accepted 2 July 2014; published online 5 August 2014) A technique to measure the mass ablation rate in direct-drive inertial confinement fusion implosions using a pinhole x-ray framing camera is presented. In target designs consisting of two layers of different materials, two x-ray self-emission peaks from the coronal plasma were measured once the laser burned through the higher-Z outer layer. The location of the inner peak is related to the position of the ablation front and the location of the outer peak corresponds to the position of the interface of the two layers in the plasma. The emergence of the second peak was used to measure the burnthrough time of the outer layer, giving the average mass ablation rate of the material and instantaneous mass remaining. By varying the thickness of the outer layer, the mass ablation rate can be obtained as a function of time. Simulations were used to validate the methods and verify that the measurement techniques are not sensitive to perturbation growth at the ablation surface. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4890256] In cryogenic direct-drive inertial confinement fusion experiments, spherical targets composed of a plastic outer shell and an inner deuterium-tritium DT ice layer are uniformly illuminated by high-intensity laser beams.1 The ablation of the outer shell drives the remaining mass inward through the rocket effect and compresses the DT fuel with the goal of reaching thermonuclear ignition. Accurate measurements of the mass ablation rate are important in determining the total energytransferred to the fuel (Ek = 1/2Mr v 2 ) where ˙ is the total remaining mass, Mi is the Mr = Mi − mdtdS ˙ is the mass ablation rate, and v is the total initial mass, m velocity of the shell.2 Previous experiments have measured the mass of an imploding shell by x-ray backlighting the target3, 4 or using buried high-Z tracer layers.5, 6 The radiography experiments indicate a 10% measurement accuracy limited by Abel inversion, Rayleigh-Taylor perturbations of the shell, and the opacity modeling, while the high-Z tracer experiments were limited because of the sensitivity of the method to perturbations of the shell. In this paper, we present a technique where the soft xray emission from a two-layer target was measured with an x-ray framing camera (XRFC) and used to determine the time when the outer layer was ablated [Fig. 1]. The micron-level accuracy of this method results from using emission profile features that do not require Abel inversion to analyze and are insensitive to perturbations of the ablation front. The trajectories of the ablation front and the interface between the two layers were measured from the self-emission profiles and used to determine the time when the laser burns a) Contributed paper, published as part of the Proceedings of the 20th

Topical Conference on High-Temperature Plasma Diagnostics, Atlanta, Georgia, USA, June 2014. b) Electronic mail: [email protected]

0034-6748/2014/85(11)/11D616/3/$30.00

through the outer layer. This provides measurements of the average mass ablation rate and the instantaneous mass of the shell. The determined average mass ablation rate can be refined by varying the outer layer thickness over many shots. Simulations were used to verify that the measurement techniques are not sensitive to perturbation growth at the ablation surface. This technique was demonstrated in cryogenic experiments conducted on the OMEGA Laser System.8 The outer diameter of the targets was 865 μm, with a 7.4 ± 0.1-μmthick outer layer of CD (Z = 3.5) and a 52.2 ± 0.5-μm-thick inner layer of DT ice (Z = 1), where C is carbon, D deuterium, and T tritium. The targets were uniformly illuminated by 60 laser beams at a wavelength of 351 nm. Three 100-pslong pickets followed by a 1.4-ns step pulse [Fig. 2(a)] were used with a total energy of 27 kJ.1 An array of sixteen 20-μm pinholes was used to produce 16 x-ray images of the target on a four-strip XRFC.9, 10 A two-layer filter consisting of 0.3 μm of Al and 1 μm of polypropylene was placed in front of the microchannel plate on the XRFC to block emission below ∼700 eV. A magnification of 6 × was used, with an image-integration time of 50 ps and an interstrip timing of 250 ± 5 ps. Intensity lineouts were taken along chords through the center of each image and azimuthally averaged over 5◦ , which was chosen to match the 20-μm diagnostic resolution. The average of all of these lineouts is shown for each of the images in Fig. 1 to give the best representation of the overall emission profile. The lineouts show intensity peaks that indicate the locations of the ablation front and CD/DT interface in the plasma. The measurement of the peak location was determined from each lineout with a standard variation of σp = 2 μm, where the subscript /p/ denotes the peak location. These positions are averaged as N ∼ 60 independent measurements to give a resulting measurement accuracy for the

85, 11D616-1

© 2014 AIP Publishing LLC

Davis et al.

11D616-2

Rev. Sci. Instrum. 85, 11D616 (2014)

–400 –200 0 200 400

140 100 60 20

Intensity (counts)

y (µm)

I (counts)

(a)

(b) 100 60 20

100 60 20

Intensity (counts)

(c)

60 (d)

Intensity (counts)

y (µm)

I (counts)

–400 –200 0 200 400

100 (f)

40 20 0

y (µm)

I (counts)

–400 –200 0 200 400

(e) 100 60 20 –400

0 400 x (µm)

60 20 100 200 300 400 Radius (µm)

E23151J1

FIG. 1. (a), (c), (e) Self-emission images and (b), (d), (f) lineouts averaged over the entire image measured [(a), (b)] ∼200 ps before the CD burnthrough, [(c), (d)] during the CD burnthrough, and [(e), (f)] when the DT emission dominates the CD emission. The position of the measured intensity peak (green line) and inner gradient (blue line) are plotted on the images.

√ average peak location of σp / N < 1μm. The inner gradient of the inner peak, which also proves to be a valuable metric, was similarly located with a standard variation of σi = 5 μm √ and σi / N < 1μm, where /i/ denotes the inner gradient. Figure 2(b) shows the position of the ablation front determined from self-emission images where the maximum intensity is located near the ablation surface of the target [Figs. 1(b) and 1(f)]. Assuming spherically symmetric emission from the

400

(b)

Radius (µm)

I III

IV

2000

II

300

0.8 200

(a)

0.4 0.0

1600

0

2000

2400

Time (ps) FIG. 2. (a) Laser pulse shape. (b) Measurements of the location of the ablation front (closed circles for peak and open circles for gradient measurements) and the CD/DT interface (closed diamonds for shifted peak intensities and open diamonds for outer gradient points) are shown as a function of time, along with the third-order polynomial fits for the trajectories (blue dotted line for the ablation front and green for the CD/DT interface). The CD burnthrough time, defined as the departure of the fits, is marked (black dotted line).

target, peaks in the line-integrated intensity occur at the radii of maximum emission in the equatorial plane of the target due to the limb effect. A local maximum in emission occurs in the plasma just outside the ablation surface as a result of the combination of high particle density and temperature in this region. The resulting intensity peak gives the ablation surface location without requiring knowledge of the exact value of the emission, so no Abel inversion is necessary. At early times (Region I), the primarily DT target shell is optically thin to the CD emission, and the ablation front is determined from the peak in the emission profile. At late times (Region II), the shell is optically thick to the DT emission, providing a sharper inner intensity gradient to the peak due to the absorption of the x-rays emitted behind the target, and the ablation front is determined from the mid-intensity point on the inner gradient.7 As the laser burns through the outer CD layer (Region III), a transition between the CD-only [Fig. 1(b)] and DT-dominated [Fig. 1(f)] emission profiles is observed [Fig. 1(d)]. During this time, the DT begins to ablate, but the higher-Z CD emission is brighter than that of the DT and obscures the ablation front. As the CD moves outward from the ablation surface to lower densities, the relative difference between the CD and DT emission is reduced. Once the CD emission reaches the same level as DT emission, the emission around the ablation surface is dominated by DT and the ablation front radius is well-defined by the inner gradient location (Region II). Figure 2(b) shows that after the laser burns through the outer layer of the target, the CD/DT interface moves away from the ablation surface. The highest emission from the CD is located at the interface between the ablated CD and DT, as this is where the CD density is greatest and has the longest line-integrated emission. Thus, after the CD burnthrough, an outer intensity peak emerges in the emission images that marks the CD/DT interface location (Region III). At late times, when the separation between the CD interface peak and the DT ablation front peak is significant (Region IV), the CD emission peak grows broader because of the growth of perturbations at the interface [Fig. 1(f)]. In Region IV, a robust criterion was developed to determine the position of the CD emission peak and thereby identify the interface location [Fig. 3]. The maximum difference in measured intensity over a distance of 30 μm along the outer edge of the CD peak in a radial intensity lineout is determined, and a line with the slope given by this difference is calculated back toward the center of the target. The interface position is determined to be at the point where the measured intensity deviates from the value of this fit by 10%. The accuracy of this method is determined √using the same technique as discussed above to give σo / N < 1μm, where /o/ denotes the outer peak. The CD burnthrough time is marked by the departure of the CD/DT interface from the ablation front. After the positions of the ablation front and material interface were measured from each image, they were plotted with the time at which the image was taken to determine the trajectory of each. Third-order polynomial fits were calculated for the two data sets to accurately determine the separation of the trajectories over the entire time interval considered. (The fit for the interface included the initial ablation-front points, since before

Review of Scientific Instruments is copyrighted by the American Institute of Physics (AIP). Redistribution of journal material is subject to the AIP online journal license and/or AIP copyright. For more information, see http://ojps.aip.org/rsio/rsicr.jsp