Gamma Reaction History ablator areal density constraints upon correlated diagnostic modeling of NIF implosion experiments

Transcription

1 Gamma Reaction History ablator areal density constraints upon correlated diagnostic modeling of NIF implosion experiments C. Cerjan, 1, a) D. B. Sayre, 1 O. L. Landen, 1 J. A. Church, 1 W. Stoeffl, 1 E. M. Grafil, 2 H. W. Herrmann, 3 N. M. Hoffman, 3 and Y. Kim 3 1) Lawrence Livermore National Labortory 7 East Avenue, Livermore, CA 9455, USA 2) Colorado School of Mines, Golden, Colorado, 841, USA 3) Los Alamos National Laboratory, Los Alamos, New Mexico, 87545, USA (Dated: 1 March 215) The inelastic neutron scattering induced γ-ray signal from 12 C in an Inertial Confinement Fusion (ICF) capsule is demonstrated to be an effective and general diagnostic for shell ablator areal density. Experimental acquisition of the time-integrated signal at 4.4 MeV using threshold detection from four gas Čerenkov cells provides a direct measurement of the 12 C areal density near stagnation. Application of a threedimensional isobaric static model of data acquired in a recent high neutron yield National Ignition Facility (NIF) experimental campaign reveals two general trends: smaller remaining ablator mass at stagnation and higher shell density with increasing laser drive. PACS numbers: 1.163/ a) Electronic mail: cerjan1@llnl.gov 1

2 A comprehensive characterization of the gas, fuel and ablator assembly is essential in the analysis of Inertial Confinement Fusion implosion experiments since attaining laboratoryscale ignition relies crucially upon symmetric convergence of a sufficiently dense fuel assembly. Many implosion diagnostics are currently fielded at the National Ignition Facility 1 which monitor performance in a several hundred picosecond temporal window about peak neutron production, typically known as the stagnation phase of the implosion. For example, both time-resolved and time-integrated x-ray self-emission images are available for estimates of the symmetry and volume of the burning core, the time of peak energy production, the temporal burn width, and detection of ablator material mixed into the burning region by enhanced emissivity. Nuclear physics based diagnostics, such as the neutron Time-Of-Flight (ntof) and Magnetic Recoil Spectometer (MRS) detectors currently provide such essential quantities as the deuterium-tritium (DT) neutron yield as a function of neutron energy, the burn-weighted ion temperature, the total areal density of the assembled fuel, and residual velocity of the burning core. In broad terms, the x-ray based stagnation phase diagnostics provide only indirect information about the much colder fuel and remaining ablator assembly, typically derived from opacity variations 2. Diagnostic approaches that rely upon nuclear processes are more amenable to an analysis of both the hot core conditions and the colder assembled configuration. Neutron scattering phenomena are particularly valuable since non-burning or non-emitting material may be probed 3. For example, the elastic collisions of the DT-fusion neutrons with the compressed fuel and ablator assembly lowers the energy of the escaping neutrons that is then cleanly monitored in the 1-12 MeV spectral range by the ntof or MRS detectors. This energy range is chosen since it avoids the complications due to TT fusion neutron production from 1 to 1 MeV and other nuclear processes in this lower energy range. More detailed discussion of typical DT neutron spectra relevant to ICF diagnostics has appeared previously 4. However, this process is dominated by the scattering from the dense DT layer and is thus much less sensitive to the remaining hydrocarbon ablator. Earlier attempts to deduce an ablator areal density using different components of the neutron spectrum avoided the DT fuel layer contribution altogether by considering data from DT gas-filled implosion experiments, that is, capsules fielded without a DT-ice layer 5. Furthermore, both of these ablator areal density estimates required MCNPX modelling of idealized capsule conditions for their interpretation. Although the model choices and 2

3 procedure are reasonable, the derived values are thus less directly associated with the desired measurement. Nonetheless, these early attempts highlight the utility of nuclear diagnostics to obtain cold shell implosion performance metrics. Another NIF nuclear diagnostic capability, the Solid Radiochemistry Collector (SRC), is also sensitive to conditions in the compressed assembly 6. Briefly, neutron-activated debris from the exploding gold hohlraum is collected on witness plates within the chamber. Decay signals from two activation products, 198 Au and 196 Au, are obtained and their ratio measures the relative number of low energy neutrons to high energy (13-15 MeV) neutrons in the implosion spectrum. More specifically, the capture cross section 197 Au(n, γ) 198 Au is dominated by very low energy neutrons ( 1 MeV), hence the technique is quite sensitive to neutrons which have experienced many collisions with the compressed DT fuel and ablator assembly. Estimates of the remaining hydrocarbon areal density are obtained by fitting MCNPX neutron spectra from idealized capsule configurations to the experimental activation ratio. This procedure is quite similar to the ntof and MRS-based approaches noted above 5 but obviously may be used in the presence of a DT fuel layer. Recent improvements to the Gamma Reaction History (GRH) diagnostic 7 now provide a direct experimental determination of the areal density associated with the hydrocarbon ablator with or without a compressed DT fuel layer. Although the primary diagnostic function of the GRH measurements on NIF implosion experiments is to provide a time history of the core fusion process by monitoring DT-γ emission, it is also possible to extract the bright 4.4-MeV γ rays from the 12 C(n, n γ 4.4 ) reaction. This γ-ray signal is unaffected by any of the competing scattering processes in the DT layer that complicate the interpretation of the ntof, MRS, or SRC-derived neutron spectra. The calibrated time-integrated emission thus measures a spatially integrated hydrocarbon ablator areal density directly. In contrast, both the ntof and MRS neutron spectra will depend upon spatial variations in the shell density distribution along their respective lines of sight between the burning core and the detector. Since the γ-ray emission is approximately spatially isotropic, a single measurement inherently produces a shell-averaged areal density. The significance of this data in characterizing implosion performance arises primarily from its implications for mixing of ablator mass into the dense DT fuel layer region. More specifically, if the remaining ablator mass is approximately constant in a series of nearly identical implosion experiments then an increase in the measured ablator areal density could 3

4 imply entrainment of ablator material into the denser DT fuel layer. Since hydrodynamic instabilities can be a dominant failure mechanism 8, trends in the hydrocarbon areal density might supply important information about the growth of these instabilities especially at the DT fuel-ablator interface. High adiabat fuel and ablator configurations, though, are designed to minimize this interfacial fuel-ablator mixing and therefore improve implosion performance. Conversely, if the collective stagnation implosion diagnostics suggest minimal amounts of hot core or ablator mix, then improved implosion performance is consistent with mix being a major degradation mechanism. An analysis of the recent series of high foot implosion experiments conducted at the NIF provides an example of the utility of this approach 9. With neutron yields in excess of 1 15 signal detection in the GRH becomes statistically robust. Furthermore, the high foot experimental campaign proceeded systematically in the sense that successive implosions typically differed by an increase in hohlraum drive. In the next section, the GRH experimental apparatus and data analysis will be described and, in the third section, details of a simple model are provided that supplies a coherent framework in which to view the data, especially the increase in ablator density with increasing x-ray drive. The presentation concludes with a brief discussion of the relevance of the analyzed data to the high foot campaign and possible extensions of the GRH diagnostic. I. EXPERIMENTAL DETAILS AND DATA ACQUISTION The GRH diagnostic at NIF consists of four gas cells with selectable pressures to adjust the Čerenkov threshold on Compton-scattered electrons in the active detector volume1. For layered-dt implosions the chosen thresholds correspond to γ-ray energies of 1, 8, 4.5 and 2.9 MeV. The 1- and 8-MeV thresholds measure γ rays from a branch of DT fusion 11. Prompt γ rays emitted from neutron interactions with the hohlraum and thermo-mechanical package 12 surrounding the CH capsule dominate the 4.5-MeV threshold, while the 2.9-MeV signal divides somewhat evenly between 4.4-MeV γ rays from inelastic scatters of fusion neutrons off the CH capsule and the neutron-induced background from nearby material. It should be emphasized that each cell integrates the γ rays above its energy threshold, so the capability of a GRH threshold to distinguish between different γ-ray sources is limited by its 1-ps time resolution (FWHM). However, significant overlapping 4

5 of the sources occurs within the temporal resolution, and the neutron-induced background and T(d, γ) signals must be subtracted from the 2.9-MeV data in order to extract the 12 C(n, n γ 4.4 ) component. In work performed at the Omega Laser Facility, Hoffman et al. 13 have extracted carbon areal densities from GRH measurements of 12 C(n, n γ 4.4 ) reactions during direct drive implosions of CH capsules filled with DT gas. We now extend that work to include some of the indirect drive implosions of layered-dt capsules which have been performed at NIF. Following the procedure of Hoffman et al., we first calibrate GRH in situ by measuring the 12 C(n, n γ 4.4 ) signal produced by a graphite puck during exposure to a known fluence of 14-MeV neutrons from a DT exploding pusher. Exploding pushers shock heat a low density fuel to multi-kev temperatures, which results in capsule convergences or areal densities too small to effect the transport of DT neutrons outward from the burning plasmas. Thus, the DT neutron spectrum produced by exploding pushers is nearly mono-energetic and isotropic. The puck calibration consisted of two shots: one which fielded a thin graphite (1.8 g/cm 3 ) puck (approximately 3.2 cm in diameter by 1.1 cm thick with a.7 cm diameter openning in the middle) in a holder located 5.7 cm away from a DT exploding pusher and the other ran only the holder with a pusher. With measurements of the source-puck geometry and the mass of the puck, a ray trace program determined the puck to have an areal density of 25.8 ±.4 mg/cm 2 when weighted by its solid angle. Equivalently, the γ-ray signal produced by the puck would correspond to the exploding pusher having a carbon shell ρr of 23.7 mg/cm 2 (correcting for slight differences between puck and pusher locations relative to GRH, which lies 6 m away from the pusher, and neutron and γ-ray attenuations through the puck). Figure 1 shows how the puck signal was extracted from the calibration data taken with the 2.9-MeV threshold. Measurements of the puck signal and neutron activation yield, along with the known γ-ray angular distribution for the 12 C(n, n γ 4.4 ) reaction 14,15, determine a carbon ρr calibration constant with an uncertainty of 15%. The error bar results primarily from the quadrature sum of uncertainties from activation measurements (1%) and γ-ray angular distribution (1%). A separate calibration characterized the prompt neutron-induced background from the hohlraum and thermo-mechanical package used for layered-dt implosions. An exploding pusher with the same target holder as the implosions was measured with the 4.5- and 2.9-MeV thresholds, and a constant was determined to relate the 4.5- with the 2.9-MeV threshold signal. 5

6 Figure 2 shows a least squares fit to the four thresholded measurements from a layered- DT implosion. The fit decomposes the 2.9-MeV threshold into the 12 C(n, n γ 4.4 ) signal, a roughly equal intensity signal from the neutron-induced background, and a 5% contribution from the T(d, γ) reaction. The neutron-induced component in the 2.9-MeV threshold was constrained by its signal level in the 4.5-MeV threshold by using the aforementioned calibration constant. Measurements of T(d, γ) from the 1- and 8-MeV thresholds, the γ-ray spectrum from the reaction, and the response of the cell to the spectrum constrain its signal in the 2.9-MeV data. The model of the γ-ray signals used in the fit has been discussed 16. The carbon ρr is determined by integrating the fitted 4.4-MeV signal, dividing by the measured neutron yield, and multiplying by the calibration constant. For layered-dt implosions, which have increased convergence ratios, the DT areal density modifies the birth neutron spectrum by scattering processes, and a correction must be made to account for 12 C(n, n γ 4.4 ) reactions with neutron energies less than 14 MeV. Folding the downscattered spectrum with the 12 C(n, n γ 4.4 ) cross section determines a correction of 2% to the measured carbon ρr. Uncertainities from the downscatter correction and detector calibration limit the accuracy of the ρr measurements to ±2%. II. STATIC MODEL ANALYSIS The high foot campaign experiments retained the Si-doped hydrocarbon ablator as used in the National Ignition Campaign (NIC) but the laser drive was modified from four pulses to three with higher energy in the first (foot) pulse (hence the terminology high foot ). Physical estimates and radiation-hydrodynamic simulations suggested that this choice of laser drive would suppress fuel-ablator material mixing by reducing the compressibility of the DT fuel. This loss of compressibility decreases the likelihood of ignition since the necessary shell convergence cannot be attained, but increases the overall stability of the implosion 9. The experimental consequences were a ten-fold increase in neutron yield and very little ablator mix into the burning core compared with the best-performing implosions during the NIC. Consistent with decreased compressibility, the fuel areal density declined by 3%. In addition to the laser drive alteration, three different initial ablator outer radii were investigated: a thicker shell matching that used in the NIC, 195 µm, and shells with 2 and 3 µm thinner radii. The variation in hydrocarbon shell thickness was selected to systematically 6

7 increase the maximum implosion velocity without further increasing the laser drive power. Since these implosions produced higher neutron yield, there is interest in quantifying their implosion characteristics in detail. Numerically converged three-dimensional hydrodynamical simulations of the entire high foot campaign implosions are computationally prohibitive, so a more tractable approach was selected to examine the experimental data trends. A validated three-dimensional static model used in the earlier NIC campaign was applied to the interpretation of the experimental data 17. This model correlates the NIF implosion diagnostic information to deduce the thermodynamic properties of the burning core, the spatial density distribution of the DT fuel and the remaining hydrocarbon ablator in an attempt to generate a self-consistent description of the time-averaged implosion conditions. In essence, a three-dimensional volume is derived from the available x-ray and neutron images; assuming isobaric conditions at implosion stagnation, the hot core density and temperature variation are obtained; and the directional nuclear scattering and activation data is used to reconstruct the shell and ablator density distribution. The original DT fuel mass is also assumed to be conserved but the remaining hydrocarbon mass is of course variable. As noted above, however, the density and spatial distribution of the remaining hydrocarbon material is only weakly constrained by the ntof, MRS, and neutron imaging diagnostics since these signals are dominated by the compressed DT fuel layer. The GRH 12 C(n, n γ 4.4 ) yield is not and thus provides a more stringent constraint upon the density distribution of the remaining hydrocarbon ablator mass. The intent of the present investigation is to augment the existing three-dimensional model by incorporating this new data constraint. As a quantitative demonstration of the correlated, static three-dimensional fitting procedure, selected values from three NIF implosion experiments and their fits are listed in Table I. Shot N12321 was one of the best-performing NIC implosions while N14225 and N1477 are comparable shots from the recent high foot campaign, differing essentially only in the thickness of the fielded ablator shell. In this Table, the experimental values for the total yield, burn-weighted (ntof) ion temperature, and total shell areal density are given with their associated model fits. Once the fit burn-weighted thermodynamic conditions have been established, many other stagnation properties may be derived. Two such important quantities are also listed: the burning fuel areal density and hot core pressure. Again, introducing the GRH 12 C(n, n γ 4.4 ) time-integrated signal as a further constraint should complement 7

8 these hot core condition fits by providing additional detail about the DT-fuel hydrocarabon ablator interface. A representative reconstructed image is plotted in Figure 3 for shot N Fitting regions demarcate volumes of DT gas (region 1), the inner solid DT fuel layer (region 2), the outer DT fuel layer (region 3), and the hydrocarbon layer (region 4). Regions 1, 2, and 3 are allowed to have variable hydrocarbon mix depending upon the specific details of an individual implosion. The amount of hydrocarbon ablator present in regions 1 and 2 is constrained by x-ray emissivity measurements of hot core mix 2 whereas the density and spatial distribution of ablator material in regions 3 and 4 is only weakly constrained without the GRH data. TABLE I. Experimental and static three-dimensional fit characteristics for three selected shots: N12321, N14225 and N1477. The total neutron yield, ntof ion temperature, and total shell areal density are among the constraints applied in the fit. Deduced quantities include the burnweighted burning core areal density and pressure. Shot N12321 N12321 N14225 N14225 N1477 N1477 Exp Fit Exp Fit Exp Fit Y tot 4.83e14 5.e e e15 4.9e e15 T ion (kev ) ( ρr g ) tot cm 2 ( ρr g ) core cm 2 P (Gb) The particular concern of the work described here is the use of the time-integrated 4.4 MeV emission to fit the hydrocarbon areal density in the context of this model and thereby derive a better qualitative understanding of the stagnation conditions and data trends. It should be emphasized that the fitting procedure cannot provide a unique determination of the ablator areal density since independent measurements of the hydrocarbon density and its location are not available. The few constraints that exist arise from estimates of the remaining hydrocarbon ablator mass 18 and the ablator mass mixed into the burning core 2. In the case of the high foot series of implosions 9, very little ablator mix was observed in the burning core 19, implying that the remaining ablator material is either located in a compressed 8

9 shell around the DT fuel assembly or entrained in the fuel assembly, or both. The ablator mass at stagnation is typically estimated from so-called Convergent Ablator implosions which rely upon x-ray back-lighting to measure capsule velocity and remaining mass. These implosions used the same laser drive conditions as layered DT implosions but replaced the DT ice layer with a comparable mass of hydrocarbon. Small corrections for residual differences in laser and capsule mass are then included in the estimate of remaining mass for cryogenic DT layer implosions 2. This technique relies upon rocket curve scaling which has been validated by the agreement between experiment and simulation 21. As mentioned above, the cyrogenic implosions varied the laser energy, power, hohlraum lining material and ablator thickness to increase the incoming shell velocity which induced significant differences in the remaining mass and the GRH-derived ablator areal density. Consider first, for simplicity, a spherical implosion assembly consisting of the DT gas and DT fuel layer, with the remaining hydrocarbon ablator matched to the exterior of the DT fuel layer. The DT regions are assumed to be uncontaminated by ablator material and the exterior ablator region is described by a linear exponential form decreasing quadratically beyond a radius r and matched to a Gaussian form that describes the DT fuel density ( r ) 2 ρ (r) = ρ DT (r ) exp[ α (r r )]Θ (r r ) (1) r ( = ρ exp[ b(r r a ) 2 r ) 2 ] exp[ α (r r )]Θ (r r ), (2) r where r is the matching radius and α is the inverse decay length. An exponential form is physically reasonable since the hydrocarbon layer is releasing material without a boundary constraint after the shock passage that creates stagnation conditions while the quadratic decrease accounts for the spherical geometric contribution. The ablator density at r is required to match the compressed fuel Gaussian form which has a peak density of ρ, with a full width at half-maximum b, and radius at peak density of r a. The Gaussian form is independently determined by fitting the x-ray and nuclear diagnostics as in the static model, so the three parameters ρ, b, and r a are known. The remaining two parameters, α and r, will be determined using the GRH measured hydrocarbon areal density, ρ CH, and the estimate of the remaining ablator mass, m CH. The standard Heaviside step function, 9

10 Θ (s), is zero for s < and one for s. The angle-averaged areal density constraint yields ρ CH = 1 4π 2π dφ π dθ sin θ dr ρ (r) = ρ DT (r )r (1 + αr Ei( αr )), (3) where Ei (x) is the exponential integral function. Applying the remaining mass constraint produces a simple relationship between the matching radius and inverse decay length m CH = ρ DT (r ) 2π dφ π dθ sin θ dr r 2 Θ (r r ) exp[ α (r r )] = 4π r2 α ρ DT (r ), (4) since m CH is known experimentally. Replacing α in equation (3) produces an nonlinear equation for the remaining fit parameter, r. This nonlinear equation is readily solved by standard numerical techniques. In three-dimensional fits to selected implosions, more realistic conditions are introduced into the model. The density and radius at which the exponential model for the remaining ablator material begins is now a function of radial distance and spherical angle; hydrocarbon material may be introduced into the DT gas and fuel assembly; and other experimental constraints such as the burning core volume and fuel areal density are imposed. That is, the previous fitting procedure is applied to determine the burning core and assembled DT fuel layer properties 17 but the fit is now extended to include the remaining mass and GRHinferred ablator areal density values as constraints. Since the matching density and radius now depend upon the azimuthal and polar angles, the two constraint equations become and ρ CH = ρ CH (θ i, φ j ) (5) m CH 4π = m CH (θ i, φ j ) (6) at each angular mesh point (θ i, φ j ). Thus the constraints are applied along each angular ray. Estimates of the error ranges of the fit quantities are obtained from combinations of the extreme limits of the experimental constraints. An example of the angular variation in fitted quantities arising from this procedure is displayed for the high foot implosion N14511 in Figure 4. As noted above, the static threedimensional isobaric model correlates the implosion diagnostic information to reconstruct 1

11 the averaged thermodynamic conditions in the burning core region and the DT fuel layer. The derived density contours are plotted from an equatorial view. Pronounced asymmetry is clearly visible in both the burning core region and the associated DT fuel layer. This description may now be extended using the remaining hydrocarbon mass and areal density constraints. The radius at which the hydrocarbon mass is conjoined with the DT fuel layer is indicated by the white contour. Density profiles along selected rays are indicated by the dashed lines whose intersection with the initial radial position on the white contour is marked. The associated density profiles as a function of radial distance along the selected rays appear in the right-hand plot. This specific case clearly demonstrates the angular variation possible in the fitting methodology. Fits using the three-dimensional model to six of the high foot implosions were performed: two of the thicker (N131119, N14511), two thinner (N14225, N1452) and thinnest (N1477) capsule configurations (see Table II). Additionally, one of the better-performing NIC implosion experiments, N12321, was included for comparison. The original static model 17 was modified to replace the previous constant hydrocarbon ablator region (region 4 in Figure 3) with a quadratic exponentially decaying functional form. The use of this simple fitting form provides some initial qualitative insight into the differences among the chosen implosions. It is perhaps helpful to visualize the results of the extended fitting procedure by comparing identical cross sectional views of the stagnated assembly. The comparison appears in Figure 5 in which the equatorial plane cross sectional density contours for the analyzed implosion experiments are presented in chronological order. As in Figure 3, the white line contour separates the DT fuel layer from the remaining hydrocarbon ablator shell. Several clear trends may be noted. First, the better-performing low-adiabat implosion N12321 achieves the most favorable combination of high density mass distribution and relatively uniform assembly symmetry. The subsequent high-adiabat implosions do not attain this uniformly high density distribution. Second, all of the high-adiabat implosions appear to have cold fuel shape distortions and, in some cases, burning core asymmetries. Third, there are two implosions, N14511 and N1452, in which the peak hydrocarbon ablator layer density is several times larger than the peak densities in the other high-foot implosions. An examination of the three-dimensional matching densities and the associated widths in Table II for these two shots reveals substantially higher matching densities and much smaller 11

12 TABLE II. Experimental characteristics of the selected shot series including laser drive energy and hohlraum lining, capsule thickness, and peak implosion velocity. Spherically averaged peak densities, matching radii, and the e 1 ablator width (α 1 ) are derived from fitting the threedimensional static model using the experimental GRH hydrocarbon areal densities and estimated remaining ablator mass. These averaged fit values are designated as 3DF it. Error ranges for the experimental constraint values and the resultant fit quantities are indicated. Shot N12321 N N14225 N14511 N1452 N1477 Drive Energy (M J) U(Au) 1.98 Au Au U(Au) U(Au) 1.57 Au CH T hickness (µm) P eak V elocity ( µm) ns ( ρr mg ) CH 45 ±9 229 ± ± ±59 36 ± ±75 cm 2 m CH (µg) 319 ± ± ± ±4 14 ± ±34 3DF it r (µm) 53 ±5 57 ±2 63 ±3 54 ±3 5 ±5 55 ±4 ( 3DF it ρ g ) cm ±18 99 ± ± ±16 31 ± ±78 3DF it W idth (µm) 27 ±13 57 ±19 43 ±18 13 ±7 15 ±4 19 ±6 widths. According to this analysis, the ablator was thus very compressed conformally with the DT fuel layer. In all of the considered cases, though, the hydrocarbon ablator layer clearly remains outside the regions of highest DT fuel density corroborating the expected behavior of the high-adiabat design in this shot series. Even within the context of this oversimplified approach, conspicuous trends may be distinguished as a function of increasing implosion drive. For example, the lower velocity implosions, N and N14225 have qualitatively similar matching densities and radii whereas the higher velocity implosion shots N14511 and N1452 display much higher matching densities and significantly reduced matching radii. Broadly speaking, the increased final velocity of these two latter shots, due to a combination of higher laser energy, higher albedo of the Au-lined uranium wall, and thinner shell, compressed a smaller amount of residual ablator mass to an approximate three-fold increase in density at a 1% reduced radius. The thinnest shell implosion, N1477, differs somewhat different from both the two lower and two higher peak velocity implosions. In this case the smaller amount of initial ablator mass compensates for the decreased x-ray drive peak velocity leading to performance comparable 12

13 to the greater peak velocity implosions. A more complete description of velocity effects in the high foot series is consistent with these results 19. It is possible to isolate the qualitative effect of just decreasing the original shell thickness upon implosion performance by comparing N14225 and N1477. Both of these implosions were fielded with a standard gold hohlraum and laser drive energy, 1.6 MJ. Thus the primary configuration difference between these implosions was the thickness of the hydrocarbon ablator shell 177 µm for N14225 and 164 µm for N1477. Thinning the ablator shell produced a higher peak velocity, as was the intent, but the model fits help quantify the changes in the stagnated shell density. The remaining ablator material was compressed to a smaller radius, 63 to 55 µm; the density rose from 126 to 311 g/cm 3 ; and the extent of the remaining mass decreased from 43 to 19 µm. The increase in peak velocity apparently leads directly to a more compressed stagnated shell assembly albeit with higher low mode asymmetry. It was noted above that very little ablator mix into the burning core is detected as measured by the standard x-ray yield diagnostic 19. The GRH results are thus in qualitative agreement with this trend and the absence of DT fuel-ablator interfacial mix since no additional hydrocarbon material need be entrained in the DT fuel layer to satisfy the remaining mass and areal density constraints. Within the context of these simplified fitting techniques, the GRH-inferred ablator areal densities usefully complement the other stagnation diagnostic signatures and permit an extension of the three-dimensional static model to describe the time-averaged hydrocarbon density distribution at stagnation. III. CONCLUSION The GRH diagnostic capability on the NIF is routinely used to obtain the temporal duration of neutron production in cyrogenic DT-layered implosions. The analysis above suggests that further valuable information about ignition-relevant implosion experiments resides in the 4.4 MeV γ-ray signal. Some of the most important outstanding issues in the approach to achieving ignition concern the amount of DT fuel-ablator mix occurring and the overall compressibility of the assembled DT fuel and remaining ablator. As noted above, these issues might be directly diagnosed using the time-integrated 4.4 MeV γ-ray emission as captured by the GRH diagnostic. Furthermore, the spatial distribution of the remaining ablator 13

14 density along specific lines of sight is necessary to model the high-energy x-ray self-emission of the burning core 2, since hydrocarbon and dopant opacity effects must be included in that analysis. The GRH signal will provide important constraints on those models. Finally, the 4.4 MeV GRH data complements the existing the Solid Radiochemical Collection (SRC) diagnostic fielded on the NIF 6. The synergy between these complementary measurements is being actively investigated for consistency and to explore future radiochemical activation experiments on the NIF. ACKNOWLEDGMENTS This work was performed under the auspices of the U. S. Department of Energy by the Lawrence Livermore National Laboratory under Contract DE-AC52-7NA REFERENCES 1 M. J. Edwards, J. D. Lindl, B. K. Spears, S. V. Weber, L. J. Atherton, D. L. Bleuel, D. K. Bradley, D. A. Callahan, C. J. Cerjan, D. Clark, G. W. Collins, J. E. Fair, R. J. Fortner, S. H. Glenzer, S. W. Haan, B. A. Hammel, A. V. Hamza, S. P. Hatchett, N. Izumi, B. Jacoby, O. S. Jones, J. A. Koch, B. J. Kozioziemski, O. L. Landen, R. Lerche, B. M. MacGowan, A. J. MacKinnon, E. R. Mapoles, M. M. Marinak, M. Moran, E. I. Moses, D. H. Munro, D. H. Schneider, S. M. Sepke, D. A. Shaughnessy, P. T. Springer, R. Tommasini, L. Bernstein, W. Stoeffl, R. Betti, T. R. Boehly, T. C. Sangster, V. Yu. Glebov, P. W. McKenty, S. P. Regan, D. H. Edgell, J. P. Knauer, C. Stoeckl, D. R. Harding, S. Batha, G. Grim, H. W. Herrmann, G. Kryala, M. Wilke, D. C. Wilson, J. Frenje, R. Petrasso, K. Moreno, H. Huang, K. C. Chen, E. Giraldez, J. K. Kilkenny, M. Mauldin, N. Hein, M. Hoppe, A. Nikroo, and R. J. Leeper, Phys. Plasmas 18, 513(211). 2 T. Ma, P. K. Patel, P. T. Springer, M. H. Key, L. J. Atherton, L. R. Benedetti, D. K. Bradley, D. A. Callahan, P. M. Celliers, C. J. Cerjan, D. S. Clark, E. L. Dewald, S. N. Dixit, T. Döppner, D. H. Edgell, R. Epstein, S. Glenn, G. Grim, S. W. Haan, B. A. Hammel, D. Hicks, W. W. Hsing, O. S. Jones, S. F. Khan, J. D. Kilkenny, J. L. Kline, G. A. Kyrala, O. L. Landen, S. Le Pape, B. J. MacGowan, A. J. Mackinnon, A. G. MacPhee, N. B. Meezan, J. D. Moody, A. Pak, T. Parham, H.-S. Park, J. E. Ralph, S. P. Regan, 14

15 B. A. Remington, H. F. Robey, J. S. Ross, B. K. Spears, V. Smalyuk, L. J. Suter, R. Tommasini, R. P. Town, S. V. Weber, J. D. Lindl, M. J. Edwards, S. H. Glenzer, and E. I. Moses, Phys. Rev. Lett. 111, 854(213); S. P. Regan, R. Epstein, B. A. Hammel, L. J. Suter, J. Ralph, H. Scott, M. A. Barrios, D. K. Bradley, D. A. Callahan, C. Cerjan, G. W. Collins, S. N. Dixit, T. Doeppner, M. J. Edwards, D. R. Farley, S. Glenn, S. H. Glenzer, I. E. Golovkin, S. W. Haan, A. Hamza, D. G. Hicks, N. Izumi, J. D. Kilkenny, J. L. Kline, G. A. Kyrala, O. L. Landen, T. Ma, J. J. MacFarlane, R. C. Mancini, R. L. McCrory, N. B. Meezan, D. D. Meyerhofer, A. Nikroo, K. J. Peterson, T. C. Sangster, P. Springer, and R. P. J. Town, Phys. Plasmas 19, 5637(212). 3 M. Gatu Johnson, J. A. Frenje, D. T. Casey, C. K. Li, F. H. Séguin, R. Petrasso, R. Ashabranner, R. M. Bionta, D. L. Bleuel, E. J. Bond, J. A. Caggiano, A. Carpenter, C. J. Cerjan, T. J. Clancy, T. Döppner, M. J. Eckart, M. J. Edwards, S. Friedrich, S. H. Glenzer, S. W. Haan, E. P. Hartouni, R. Hatarik, S. P. Hatchett, O. S. Jones, G. Kyrala, R. A. Lerche, O. L. Landen, T. Ma, A. J. MacKinnon, M. A. McKernan, M. J. Moran, E. Moses, D. H. Munro, J. McNaney, H.-S. Park, J. Ralph, B. Remington, J. R. Rygg, S. M. Sepke, V. Smalyuk, B. Spears, P. T. Springer, C. B. Yeamans, M. Farrell, D. Jasion, J. D. Kilkenny, A. Nikroo, R. Paguio, J. P. Knauer, V. Yu Glebov, T. C. Sangster, R. Betti, C. Stoeckl, J. Magoon, M. J. Shoup III, G. P. Grim, J. Kline, G. L. Morgan, T. J. Murphy, R. J. Leeper, C. L. Ruiz, G. W. Cooper, and A. J. Nelson, Rev. Sci. Instrm. 83, 1D38(212); D. Bleuel, C. B. Yeamans, L. A. Bernstein, R. M. Bionta, J. A. Caggiano, D. T. Casey, G. W. Cooper, O. B. Drury, J. A. Frenje, C. A. Hagmann, R. Hatarik, J. P. Knauer, M. Gatu Johnson, K. M. Knittel, R. J. Leeper, J. M. McNaney, M. Moran, C. L. Ruiz, and D. H. G. Schneider, Rev. Sci. Instrm. 83, 1D313(212); V. Yu Glebov, T. C. Sangster, C. Stoeckl, J. P. Knauer, W. Theobald, K. L. Marshall, M. J. Shoup III, T. Buczek, M. Cruz, T. Duffy, M. Romanofsky, M. Fox, A. Pruyne, M. J. Moran, R. A. Lerche, J. McNaney, J. D. Kilkenny, M. J. Eckart, D. Schneider, D. Munro, W. Stoeffl, R. Zacharias, J. J. Haslam, T. Clancy, M. Yeoman, D. Warwas, C. J. Horsfield, J.-L. Bourgade, O. Landoas, L. Disdier, G. A. Chandler, and R. J. Leeper, Rev. Sci. Instrm. 81, 1D325(21). 4 J. A. Frenje, D. T. Casey, C. K. Li, J. R. Rygg, F. H. Séguin, R. D. Petrasso, V. Yu Glebov, D. D. Meyerhofer, T. C. Sangster, S. Hatchett, S. Haan, C. Cerjan, O. Landen, M. Moran, P. Song, D. C. Wilson, and R. J. Leeper, Rev. Sci. Instrm. 79, 1E52(28). 15

16 5 J. A. Caggiano, M. Eckart, C. Hagmann, R. Hatarik, J. McNaney, D. Sayre, V. Yu Glebov, J. Knauer M. Gatu Johnson, J. A. Frenje, C. K. Li, F. H. Séguin, R. D. Petrasso, R. M. Bionta, D. T. Casey, J. A. Caggiano, R. Hatarik, H. Y. Khater, D. B. Sayre, J. P. Knauer, T. C. Sangster, H. W. Herrmann, and J. D. Kilkenny, Rev. Sci. Instrm. 85, 11E J. M. Gostic, D. A. Shaughnessy, K. T. Moore, I. D. Hutcheon, P. M. Grant, and K. J. Moody, Rev. Sci. Instrum. 83, 1D94(212); C. Hagmann, D. A. Shaughnessy, K. J. Moody, P. M. Grant, N. Gharibyan, J. M. Gostic, P.T. Wooddy, P. C. Torretto, B. B. Bandong, R. Bionta, C. J. Cerjan, L. A. Bernstein, J. A. Caggiano, H. W. Herrmann, J.P. Knauer, D. B. Sayre, D. H. Schneider, E. A. Henry, and R. J. Fortner, Rev. Sci. Instrm. (submitted). 7 H. W. Herrmann, N. Hoffman, D. C. Wilson, W. Stoeffl, L. Dauffy, Y. H. Kim, A. McEvoy, C. S. Young, J. M. Mack, C. J. Horsfield, M. Rubery, E. K. Miller, and Z. A. Ali, Rev. Sci. Instrum. 81, 1D333(21); N. M. Hoffman, D. C. Wilson, H. W. Herrmann, and C. S. Young, Rev. Sci. Instrum. 81, 1D332 (21). 8 D. S. Clark, D. E. Hinkel, D. C. Eder, O. S. Jones, S. W. Haan, B. A. Hammel, M. M. Marinak, J. L. Milovich, H. F. Robey, L. J. Suter, and R. P. J. Town, Phys. Plasmas 2, 56318(213). 9 T. R. Dittrich, O. A. Hurricane, D. A. Callahan, E. L. Dewald, T. Döppner, D. E. Hinkel, L. F. Berzak Hopkins, S. Le Pape, T. Ma, J. L. Milovich, J. C. Moreno, P. K. Patel, H.-S. Park, B. A. Remington, and J. D. Salmonson, Phys. Rev. Lett. 112, 552(214); H.-S. Park, O. A. Hurricane, D. A. Callahan, D. T. Casey, E. L. Dewald, T. R. Dittrich, T. Döppner, D. E. Hinkel, L. F. Berzak Hopkins, S. Le Pape, T. Ma, P. K. Patel, B. A. Remington, H. F. Robey, and J. D. Salmonson, Phys. Rev. Lett. 112, 551(214); O. A. Hurricane, D. A. Callahan, D. T. Casey, E. L. Dewald, T. R. Dittrich, T. Döppner, M. A. Barrios Garcia, D. E. Hinkel, L. F. Berzak Hopkins, P. Kervin, J. L. Kline, S. Le Pape, T. Ma, A. G. MacPhee, J. L. Milovich, J. Moody, A. E. Pak, P. K. Patel, H.-S. Park, B. A. Remington, H. F. Robey, J. D. Salmonson, P. T. Springer, R. Tommasini, L. R. Benedetti, J. A. Caggiano, P. Celliers, C. Cerjan, R. Dylla-Spears, D. Edgell, M. J. Edwards, D. Fittinghoff, G. P. Grim, N. Guler, N. Izumi, J. A. Frenje, M. Gatu Johnson, S. Haan, R. Hatarik, H. Herrmann, S. Khan, J. Knauer, B. J. Kozioziemski, A. L. Kritcher, G. Kyrala, S. A. Maclaren, F. E. Merrill, P. Michel, J. Ralph, J. S. Ross, J. R. Rygg, M. B. Schneider, 16

17 B. K. Spears, K. Widmann, and C. B. Yeamans, Phys. Plasmas, 21, 56314(214); O. A. Hurricane, D. A. Callahan, D. T. Casey, P. M. Celliers, C. Cerjan, E. L. Dewald, T. R. Dittrich, T. Döppner, D. E. Hinkel, L. F. Berzak Hopkins, J. L. Kline, S. Le Pape, T. Ma, A. G. MacPhee, J. L. Milovich, A. Pak, H.-S. Park, P. K. Patel, B. A. Remington, J. D. Salmonson, P. T. Springer, and R. Tommasini, Nature, 56, 343(214). 1 H. W. Herrmann, C. S. Young, J. M. Mack, Y. Kim, A. McEvoy, S. Evans, T. Sedillo, S. Batha, M. Schmitt, D. C. Wilson, J. R. Langenbrunner, R. Malone, M. I. Kaufman, B. C. Cox, B. Frogget, E. K. Miller, Z. A. Ali, T. W. Tunnell, W. Stoeffl, C. J. Horsfield, and M. Rubery, Journal of Physics:Conference Series 244, 3247(21). 11 Y. Kim, J. M. Mack, H. W. Herrmann, C. S. Young, G. M. Hale, S. Caldwell, N. M. Hoffman, S. C. Evans, T. J. Sedillo, A. McEvoy, J. Langenbrunner, H. H. Hsu, M. A. Huff, S. Batha, C. J. Horsfield, M. S. Rubery, W. J. Garbett, W.Stoeffl, E. Grafil, L. Bernstein, J. A. Church, D. B. Sayre, M. J. Rosenberg, C. Waugh, H. G. Rinderknecht, M. Gatu Johnson, A. B. Zylstra, J. A. Frenje, D. T. Casey, R. D. Petrasso, E. Kirk Miller, V. Yu. Glebov, C. Stoeckl and T. C. Sangster, Phys. Rev. C85, 6161(212). 12 L. J. Atherton, Journal of Physics: Conference Series (28). 13 N. M. Hoffman, H. W. Herrmann, Y. H. Kim, H. H. Hsu, C. J. Horsfield, M. S. Rubery, E. K. Miller, E. Grafil, W. Stoeffl, J. A. Church, C. S. Young, J. M. Mack, D. C. Wilson, J. R. Langenbrunner, S. C. Evans, T. J. Sedillo, V. Yu. Glebov, and T. Duffy, Phys. Plasmas 2, 4275(213). 14 J. Benveniste, A. C. Mitchell, C. D. Schrader, and J. H. Zenger, Nucl. Phys. 19, 448 (196). 15 D. Spaargarten and C. C. Jonker, Nucl. Phys. 161, 354 (1971). 16 D. B. Sayre, L. A. Bernstein, J. A. Church, and W. Stoeffl, Rev. Sci. Instrum. 83, 1D95 (212). 17 C. Cerjan, P. T. Springer, and S. M. Sepke, Phys. Plasmas 2, 56319(213). 18 D. G. Hicks, N. B. Meezan, E. L. Dewald, A. J. Mackinnon, R. E. Olson, D. A. Callahan, T. Döppner, L. R. Benedetti, D. K. Bradley, P. M. Celliers, D. S. Clark, P. Di Nicola, S. N. Dixit, E. G. Dzenitis, J. E. Eggert, D. R. Farley, J. A. Frenje, S. M. Glenn, S. H. Glenzer, A. V. Hamza, R. F. Heeter, J. P. Holder, N. Izumi, D. H. Kalantar, S. F. Khan, J. L. Kline, J. J. Kroll, G. A. Kyrala, T. Ma, A. G. MacPhee, J. M. McNaney, J. D. Moody, M. J. Moran, B. R. Nathan, A. Nikroo, Y. P. Opachich, R. D. Petrasso, R. R. Prasad, J. E. 17

18 Ralph, H. F. Robey, H. G. Rinderknecht, J. R. Rygg, J. D. Salmonson, M. B. Schneider, N. Simanovskaia, B. K. Spears, R. Tommasini, K. Widmann, A. B. Zylstra, G. W. Collins, O. L. Landen, J. D. Kilkenny, W. W. Hsing, B. J. MacGowan, L. J. Atherton, and M. J. Edwards, Phys. Plasmas 19, 12272(212). 19 D. A. Callahan, O. A. Hurricane, D. E. Hinkel, T. Dppner, T. Ma, H-S. Park, L. F. Berzak Hopkins, E. Dewald, S. LePape, A. E. Pak, P. Patel, J. E. Ralph, B. K. Spears, O. Landen, M. A. Barrios Garcia, D. T. Casey, C. Cerjan, T. R. Dittrich, M. J. Edwards, S. W. Haan, A. Hamza, A. Kritcher, A. MacPhee, J. Milovich, R. Rygg, J. D. Salmonson, P. T. Springer, R. Tommasini, L. R. Benedetti, R. Bionta, E. Bond, D. Bradley, J. Caggiano, J. Field, D. Fittinghoff, R. Hatarik, S. Nagel, N. Izumi, S. Khan, R. P. J. Town, D. Sayre, J. L. Kline, G. Grim, F. Merrill, P. Volegov, C. Wilde, J. P. Knauer, and A. Nikroo, Phys. Plasmas (to appear). 2 O. L. Landen, J. Edwards, S. W. Haan, H. F. Robey, J. Milovich, B. K. Spears, S. V. Weber, D. S. Clark, J. D. Lindl, B. J. MacGowan, E. I. Moses, J. Atherton, P. A. Amendt, T. R. Boehly, D. K. Bradley, D. G. Braun, D. A. Callahan, P. M. Celliers, G. W. Collins, E. L. Dewald, L. Divol, J. A. Frenje, S. H. Glenzer, A. Hamza, B. A. Hammel, D. G. Hicks, N. Hoffman, N. Izumi, O. S. Jones, J. D. Kilkenny, R. K. Kirkwood, J. L. Kline, G. A. Kyrala, M. M. Marinak, N. Meezan, D. D. Meyerhofer, P. Michel, D. H. Munro, R. E. Olson, A. Nikroo, S. P. Regan, L. J. Suter, C. A. Thomas, and D. C. Wilson, Phys. Plasmas 18, 512(211). 21 Y. Saillard, Nucl. Fusion 46, 117(26); N. B. Meezan, A. J. MacKinnon, D. G. Hicks, E. L. Dewald, R. Tommasini, S. Le Pape, T. Döppner, T. Ma, D. R. Farley, D. H. Kalantar, P. Di Nicola, D. A. Callahan, H. F. Robey, C. A. Thomas, S. T. Prisbrey, O. S. Jones, J. L. Milovich, D. S. Clark, D. C. Eder, M. B. Schneider, K. Widmann, J. A. Koch, J. D. Salmonson, Y. P. Opachich, L. R. Benedetti, S. F. Khan, A. G. MacPhee, S. M. Glenn, D. K. Bradley, E. G. Dzenitis, B. R. Nathan, J. J. Kroll, A. V. Hamza, S. N. Dixit, L. J. Atherton, O. L. Landen, S. H. Glenzer, W. W. Hsing, L. J. Suter, M. J. Edwards, B. J. MacGowan, E. I. Moses, R. E. Olson, J. L. Kline, G. A. Kyrala, A. S. Moore, J. D. Kilkenny, A. Nikroo, K. Moreno, and D. E. Hoover, Phys. Plasmas 2, 56311(213). 18

19 IV. FIGURES 9 Puck and Holder 8 Relative time (ns) Entries 3 Mean RMS. Underflow Overflow Integral 1 puck Fusion + Capsule Holder Difference Puck region of interest Relative signal (arb.) FIG. 1. The holder only histogram has been scaled to the holder with puck by the ratio of measured neutron activation yields, and the green histogram shows the difference between the two. Error bars from oscilloscope noise are shown. (The difference histogram also implies the T(d, γ) peaks are inadequate for yield scaling due to variations in the neutron-induced signal from the capsule.) Neutron time-of-flight separates the fusion and capsule signals from the puck, which was 5.7 cm away from the exploding pusher target. The puck signal was calculated by integrating the difference histogram over the indicated region of interest. 19

20 N ch3 6 1 MeV fit Entries 5 Mean RMS.1418 Underflow Overflow 4 Integral Time (ns) N ch3 6 fit_ 8 MeV Glo C(n 5 Entries Mean n-in RMS 4 Underflo T(d Overflow LP Integral PMT signal (V) PMT signal (V) N ch MeV fit 2 Entries 11 Mean RMS.125 Underflow Overflow 4 Integral Time (ns) N ch MeV Entri 5 Mean RMS Unde 4 Over Integ PMT signal (V) PMT signal (V) FIG. 2. An example fit to a layered-dt implosion, N The data are oscilloscope traces of the thresholds over the burn duration (the digitizer noise is the size of the data points). The curves result from convolving functions describing the time dependence of the γ-ray sources with the impulse response function of the detector. Shapes for the DT and carbon signals are modeled with Gaussians, which are allowed to assume different parameters since the shell areal density may evolve over the burn duration. The neutron-induced background shape results from a convolution of the DT Gaussian with an MCNP simulation of the time profile of neutron interactions in the hohlraum and thermo-mechanical package. An x-ray background, produced by laser-plasma interactions (LPI), affects the photomultiplier tubes of the diagnostic and is most apparent in 1- and 8-MeV thresholds which have the lowest signal-to-noise ratio. Note that signal ringing found in the fitted curves arises from the photomultiplier tube used in the detector 16. 2

21 FIG. 3. Density contour plot of the three-dimensional static fit to N12321 with the separate material regions distinguished by the white line contours. Regions 1, 2 and 3 consist primarily of DT with possible admixtures of ablator material. Region 4 is pure hydrocarbon ablator material. FIG. 4. The left-hand panel contains an equatorial view of the density distrubtion obtained in a three-dimensional fit to shot N The dashed colored lines indicate the associated representative radial lineouts of the density distribution appearing in the right-hand panel. The circular marker indicates the matching radius, r, between the DT fuel layer and the hydrocarbon ablator along each angular ray. The black, red and green curves refer to θ =, π/2 and π radians respectively. FIG. 5. Equatorial views of the density distribution obtained in the fit to shots N12321, N131119, N14225, N14511, N1452, and N1477. The highlighted white contour marks the matching radius, r, at which the hydrocarbon layer begins. 21

22 puck Entries 3 Mean.799 RMS.6387 Underflow Overflow Integral Relative signal (arb.) Fusion + Capsule Puck and Holder Holder Difference Puck region of interest Relative time (ns)

23 N ch3 N ch3 PMT signal (V) 6 fit Entries 5 Mean RMS.1418 Underflow Overflow 4 Integral MeV PMT signal (V) MeV Global Fit C(n,n'γ) n-induced fit 3 Entries 55 Mean RMS.1362 Underflow T(d,γ) Overflow Integral LPI Time (ns) Time (ns) N ch3 N ch3 PMT signal (V) 6 fit 2 Entries 11 Mean RMS.125 Underflow Overflow Integral MeV PMT signal (V) MeV fit 1 Entries 11 Mean RMS.1362 Underflow Overflow Integral Time (ns) Time (ns)

24

25

26