Systematic Fuel Cavity Asymmetries in Directly Driven Inertial Confinement Fusion Implosions

We present narrow-band self-emission x-ray images from a titanium tracer layer placed at the fuel-shell interface in 60-laser-beam implosion experiments at the OMEGA facility. The images are acquired during deceleration with inferred convergences of $\sim 9–14.$ Novel here is that a systematically observed asymmetry of the emission is linked, using full sphere 3D implosion modeling, to performance-limiting low mode asymmetry of the drive.

Figure 1

(a) Self-emission of Ti dopant is used to selectively characterize the shell surface directly in contact with fuel during deceleration. (b) Chord-integrated 5.4–6.0 keV Ti emission from 1D simulation of capsules with initial $0.1\mu \mathrm{m}$ thick, 1% by atom Ti tracer layer. The image is calculated for 40 ps gate beginning 100 ps prior to nuclear bang time. Dashed vertical line indicates the position of the fuel-shell interface. (c) Data are presented in the Letter for views both orthogonal and opposite to mounting.

Figure 2

(a) Omega capsules are positioned at target chamber center using a $17\mu \mathrm{m}$ stalk attached to the capsule by a conical glue fillet (typically $d=80$ and $h=40\mu \mathrm{m}$) external to the ablator. (b) Gated Ti self-emission extracted from MMI positioned at view No. 1 [see Fig. 1], with line of sight orthogonal to mounting and mounting from image bottom (shot 78 495). (c) Image from target inserted from diametrically opposite port such that mounting is from image top (78 505). Images are self normalized. (d) Radially integrated signal variation with angle for images of (b) and (c). Angle increases counterclockwise around image center with 0° at the 3 o‘clock position. Integration at each angle extends from the 20% intensity contour midway to the image center, and is length normalized.

Figure 3

Data with imager view opposite to capsule mounting, thus minimizing its contribution to observed asymmetry. (a),(b) Ti emission images for shot 79 972 and 79 980, respectively, at $\sim 100\mathrm{ps}$ prior to bang time. (c) Representative image (79 980) $\sim 100\mathrm{ps}$ later, i.e., at bang time. (d) Spectrum extracted from MMI gated at bang time with best fit from model ($n_e=5.5\times 10^{24}e/{\mathrm{cm}}^3$ and $T_e=1350\mathrm{eV}$).

Figure 4

Results from full sphere 3D simulation with representative beam-to-beam drive asymmetry. (a) A cross-sectional profile of $n_e$ at nuclear bang time with Ti-doped CH regions indicated by black contours. An asymmetry of the density profile aligns to the bulk motion of the target. (b) Corresponding profile of $T_e$. (c) Calculated 5–6 keV self-emission 80 ps prior to and at (d) bang time.

Figure 5

(a) Comparison of radially integrated emission at bang time between view No. 2 measurements [see Fig. 1] and simulation image [Fig 4]. Signals are phased by aligning their maxima. Error bars on the measurement curve indicate standard deviation obtained from three shots. (b) Fourier series analysis of the signals. Mode 1 amplitude is between 20%–40% of signal and at least $4\times$ greater than other modes for both cases.