CPS considers block bid on persistent tribunal claimant
The Crown Prosecution Service is considering an application to restrict a prolific employment tribunal complainant after his latest failed attempt to bring proceedings over a legal trainee scheme.
A judgment published this week has detailed how Zakir Khan has made a string of claims, including two against the CPS, after applying for jobs and being rejected without interview. All of the claims were made under the Equality Act and alleged he suffered discrimination as a result of his disabilities.
Employment Judge Camp, sitting at the Birmingham tribunal, formally ended Khan's latest claim and described it as ‘totally without merit’. The CPS confirmed at the end of the preliminary hearing that an application may in future be made for an order preventing Khan from issuing any further employment tribunal proceedings without permission.
The tribunal heard that Khan had complained of disability discrimination in relation to the CPS legal trainee scheme. He said in his claim form that requirements for entry to the scheme – namely completion of the LPC – put in place barriers to stop him from applying. He explained that he did not hold the LPC because of the effects disabilities have had on him, and he lacked the funds to take the qualification.
It was revealed in the ruling that at the time the claim form was issued, Khan already had a ‘large number’ of similar claims in the system, including the two CPS cases. The first, and thus far the only one which has gone to trial, was against national firm Mills & Reeve, and was ‘wholly unsuccessful’.
The judge said Khan had confirmed earlier this month he did not intend to continue with the latest CPS proceedings, and this effectively constituted a withdrawal. Issuing an order for dismissal, the judge noted that ‘if anything, the present claim is even weaker than the previous one’, with Khan appearing not even to have contacted the CPS in relation to the trainee scheme.
Observation of persistent species temperature separation in inertial confinement fusion mixtures
Description of experiments and context
We report the results of novel separated reactant ICF experiments designed to help to understand the behavior of contaminant mass during ICF implosions, along with detailed and highly-resolved three-dimensional (3D) simulations that are in good agreement with the experimental data. Our experiments explore the limiting case where chunks of contaminant are present in the initial conditions, allowing the maximum possible time to atomically mix and equilibrate with the fuel. Our experiments involve the compression of capsules (Fig. 1b) containing deuterated open-cell foams whose pores are filled with hydrogen and tritium, and we control the initial conditions by varying the foam pore sizes. Contaminant and fuel are initially physically separated and heat to different temperatures due to this separation as well as their different specific heats. As the implosion progresses, hydrodynamic processes lead to atomic mixing of these materials, resulting in a complex distribution of materials containing regions of pure chunks of contaminant or fuel and atomically mixed regions. As a result of the process of atomic mixing, fuel and contaminant in atomically mixed regions locally achieve thermal equilibrium. Our experiments uniquely address the persistence of chunks and whether these chunks remain in thermal equilibrium with the fuel. Furthermore, our results indicate that the observed phenomena can be explained by appealing to well-known physical processes—hydrodynamic instabilities, which lead to mixing and local equilibration, combined with thermal conduction, which brings materials into thermal equilibrium over larger scales—and the time-scales over which they operate. These timescales have previously only been addressed theoretically.
Fig. 1: Initial configuration of experiments and simulations.
a Scanning electron microscope image of engineered foams44 showing different macro-pore sizes. b Capsule diagram; foam matrix is open cell, containing sub-micron pores that are also filled with HT. c Visualization of pores (solid spheres are macro-pores), colored by diameter, for 90 μm pore foam sample. d Visualization of initial conditions for 3D simulation with 90 μm pores; red: shell, green: foam pores (filled with HT), blue: glue spot for target mount. e Probability distribution function (PDF) of pore diameters.
We will demonstrate reactant temperature separation in experiments by contradiction. Assuming the reactants are in thermal equilibrium, TDT = TDD, and the DT∕DD yield ratio is determined by the extent to which the reactants are atomically mixed. However, our experimental data lie above this theoretical maximum, so this assumption cannot be correct. To further support this conclusion, reactant temperature separation is directly observed in simulations that are in agreement with the experimental observables. We have evaluated several alternative explanations, including kinetic effects, anomalously high ion temperatures, target mischaracterization, and complete atomic mixing, and these cannot plausibly explain the data.
The observation of species temperature separation in ICF mixing regions has important consequences for our understanding of implosions. Models used to estimate contaminant mass through enhanced X-ray bremsstrahlung emission assume thermal equilibrium between contaminant and DT2,3,4,5,6. Such models underpredict the amount of contaminant by overestimating the emission of the contaminant. Due to the difficulty of probing the relevant timescales and lengthscales, our understanding of ICF implosions relies on simulations, which nearly universally assume mixtures to be in thermal equilibrium through modeling choices. Only with significant time on today’s largest supercomputers is it possible to accurately simulate the inherently 3D hydrodynamics22 and coupled physics at the sub-micron scales where mixing occurs. Therefore, this has only been done a few times7,14,19,23,24. Instead, mixing is either ignored, enforced in initial conditions, or modeled with a dynamic mix model, e.g., Reynolds-Averaged Navier-Stokes (RANS). As presently implemented, these force mixtures into thermal equilibrium (i.e., they assume negligible equilibration timescales regardless of how chunky the mixture is). RANS models can be coupled to reaction rate equations25 as for reactive flows26,27, e.g., flames, detonations, and combustion, where equilibration timescales are much shorter than reaction timescales. This works in the absence of reactivity fluctuations28. However, our results indicate that fluctuations are present and persistent in ICF mixing layers, so that these methodologies are incomplete in the context of ICF. In ICF simulations, temperature gradients cause deviations between DT and DD reaction burn-weighted ion temperatures (BWTI), yet larger deviations are observed experimentally29. Thermal fluctuations caused by complicated material distributions with unequilibrated contaminants could contribute to this discrepancy. This could also explain why RANS simulations consistently require unphysical adjustments to reproduce the results of separated reactants experiments30,31. Additionally, thermal fluctuations could seed velocity gradients that broaden the neutron spectrum, resulting in artificially high measured ion temperatures32.
Another open question in ICF has been whether hydrodynamic approximation of the true system is sufficient to capture the evolution of contaminant in an ICF hot-spot. Our experiments, which involve complex and evolving material distributions, provide a test bed, and our high-resolution 3D simulations, which are necessary to test this, capture the experimental observables, providing evidence that the hydrodynamic approximation remains valid.
To test the impact of our observation, we performed 2D simulations9,33,34 of a high-yield implosion11 including the fill tube, a source of chunks of contaminant. In simulation, 160 ng of contaminant mass is injected into the hot spot, and its temperature remains 2.5 keV lower than that of the DT. However, applying experimental methods for calculating contaminant mass using continuum emission from ref. 5,6 to simulated X-ray self-emission produces an estimate of 80 ng. The discrepancy arises because the model assumes the contaminant is radiating at the temperature of the fuel, which is equivalent to approximately doubling the contaminant emissivity. This effect will also have implications for interpretation of the line emission from contaminant mass. For example, temperature fluctuations such at this will likely be associated with density fluctuations, which have been demonstrated to significantly affect emissivity35.
Our experiments, performed as part of Los Alamos National Laboratory’s (LANL) MARBLE campaign36, are separated reactants experiments in which deuterium resides in contaminant material instead of fuel, so DT reactions occur only as a consequence of atomic mixing between the contaminant and the fuel. Previous separated reactants experiments14,30,31,37,38,39,40,41 included deuterium in the shell, whereas our capsules contain open-cell deuterated polystyrene (CH0.5D0.5) foams42,43,44 whose pores are filled with hydrogen and tritium (HT). This represents the limiting case in which chunks of pre-mixed material exist in the hot spot before it forms and are given the maximum possible time to atomically mix and equilibrate with the fuel. Foams are engineered with fixed macro-pores sizes44 of 30, 50, and 90 μm, in addition to the sub-micron pores that exist throughout the open-cell foam matrix. The macro-pores seed hydrodynamic instabilities, primarily the Richtmyer-Meshkov instability45,46,47,48, which induce mixing between the foam and HT. The RMI is known to exhibit sensitivity to and memory of initial conditions both in fluid49 and plasma50 regimes. Larger pores are expected to produce mixing that is more “chunky” (and hence produce a lower DT∕DD yield ratio) whereas smaller pores produce more atomic mix (hence a higher DT∕DD ratio). Ten capsule implosions were directly driven at the Laboratory for Laser Energetics’ (LLE) OMEGA laser facility with a 1ns square pulse using 26kJ of laser energy; another 12 implosions were performed on a separate shot day that produced consistent results. In simulations, conditions in the burn region are similar to those observed in mixing regions in simulations of current high-yield implosions11.
Simulation details
Simulations used LANL’s xRAGE9,51, applying methodology for simulating direct drive implosions detailed in ref. 14. xRAGE features adaptive mesh refinement (AMR), making it ideal for resolving the foam pore structure. xRAGE has previously been used to successfully model experiments involving foams33,34,36,52,53,54,55,56,57. Simulations included shell thickness variations and the target mount. Example initial conditions are visualized in Fig. 1d. The use of 3D was necessitated by foam geometry and hydrodynamically unstable flows22. To resolve mixing processes, simulations used a maximum resolution of 0.25 μm. This resolves the Kolmogorov lengthscale58λK ≈ 0.5 μm for these experiments (using viscosity formula in ref. 59). Previous 3D simulations of separated reactants experiments14 and reduced-dimension grid resolution studies suggest that this resolution is sufficient. Due to the high carbon content of the implosion, it was not necessary to explicitly model diffusion and viscosity19,20,21. Instead, an implicit large eddy simulation (ILES) strategy60 was employed. Using this methodology, diffusion and viscosity are implicitly modeled by numerical filtering; this approach has been successfully used to model separated reactants ICF implosions14. Three simulations, using a total of 510 million CPU hours, were performed on Lawrence Livermore National Laboratory’s Sequoia supercomputer. The total cell count ranged from 0.3 to 3.0 billion. Axisymmetric 2D simulations showed substantial differences from 3D simulations, arising from differing surface area to volume ratios of the pores, leading to differences in pore collapse and shock dynamics, indicating the inadequacy of 2D.
We performed 1D simulations of all ten experiments including as-shot drives, HT fill pressures, and capsule geometries in order to evaluate our ability to model the shock and implosion trajectories. Due to the complex target design, variations in these parameters were large. Fill pressures varied from 6.2 to 8.6 atm, ablator thickness varied from 22.0 to 26.2 μm, and ablator thickness variances ranged from 0.5 to 8.7 μm. These parameters affect the shock speed through the foam as well as the implosion timescale. Therefore, the ability to capture the effects of varying these parameters is a strong constraint on simulations. By varying only these parameters according to measured values, our 1D simulations captured the measured bang time (time of peak neutron production) variations within experimental error for all but one of the ten shots. This is shown in Fig. 2, where error bars on the simulations indicate variation due to capsule ablator thickness variance.
Fig. 2: Comparison of simulated and experimental bang times.
Bang times (time of peak neutron production) for all ten capsule implosions compared to 1D simulation results. 1D simulations account for variations in HT fill pressure, as-shot laser power history, and capsule ablator thickness. The simulation error bars indicate the effect of measured ablator thickness variances, which ranged from 0.5 to 8.7 μm; the average ablator thickness varied from 22.0 to 26.2 μm. The simulated error bars were calculated based on the standard deviation of 1D simulations performed at the extremes of measured ablator thickness and ablator thickness variances. The experimental bang times were determined from a neutron temporal diagnostic and the error is based on the response time of this instrument61.
Several companion experiments were performed to evaluate simulations, including capsule implosions with deuterated methane, deuterated foams with and without hydrogen gas, and CH foams and deuterium gas. Cylindrical experiments were also performed to enable X-ray radiography of shock transit through various foams and pore collapse. For capsule implosions, simulations accurately captured experimental yield and bang time trends. For cylindrical experiments, simulations accurately captured shock and pore dynamics.
Comparison between simulations and experiments
The primary observable in experiments is the DT∕DD neutron yield ratio, which measures the amount of atomic mixing. Experimental results, including yield ratios and DT burn-weighted Ti (BWTI) from 10 shots, are presented along with results from 3D simulations in Fig. 3. Yield ratios are normalized by
$$\frac{\,{\rm{Average}}\, {\rm{1D}}\, {\rm{simulated}}\, {\rm{yield}}\, {\rm{ratio}}}{{\rm{Shot}}\, {\rm{1D}}\, {\rm{simulated}}\, {\rm{yield}}\, {\rm{ratio}}}$$
(2)
to account for variation of the laser pulse, fill pressure, and ablator thickness. The experimental results exhibit no significant dependence on pore size. Apparent temperatures from 3D simulations, including fluid motion corrections66, are plotted in Fig. 3b and are in good agreement with experiment. If we assume the reactants are in thermal equilibrium, observed yield ratios are consistently higher than the theoretical maximum. We calculate the maximum yield ratio (the uniform atomic mixed limit) assuming thermal equilibrium by dividing the DT reaction rate formula by the DD reaction rate formula and noting that atomic mix maximizes the former and minimizes the latter, hence maximizing the ratio. The resulting theoretical maximum is given by
$$\frac{2\ {n}_{\text{T}}\ {\overline{\sigma v}}_{\text{DT}}({T}_{\text{i}})}{{n}_{\text{D}}\ {\overline{\sigma v}}_{\text{DD}}({T}_{\text{i}})}=74\pm 18,$$
(3)
where nD and nT are the number of deuterons and tritons, respectively. This estimate is also consistent with 1D simulations where atomic mixing is enforced in the initial conditions. Our assumption that the reactants are in thermal equilibrium implies that the BWTI is the same for the DT and DD reactions. Therefore, we have evaluated this at the experimental DT BWTI, and this dominates the uncertainty. Nevertheless, it should be noted that our results do not depend on which temperature is used for this estimate: even making the extreme assumption that Ti = 11.6 KeV to maximize the reactivity ratio, the resulting estimate is 105. Experiments and simulations are consistent with DT reactions occurring at a higher temperature than DD reactions. In simulations, the reactant temperature separation can be directly observed (Fig. 4). This results in DT BWTIs that exceed the DD BWTIs by ≈0.25 keV, which is sufficient to raise the yield ratio by 30–40%. This is larger than the reaction BWTI separation that arises due to thermal gradients in a fully atomically mixed hot spot. For comparison, we include results that would have been produced by the 3D simulations if the reactants were in thermal equilibrium (labeled “Equilibrated Tion”). We calculated this by integrating the reaction rate equation spatially and temporally over the burn region using the instantaneous burn-weighted Ti to calculate the DT and DD reactivities rather than the local ion temperatures. The resulting yield ratios are below the theoretical maximum and decrease with increased pore size, dominated by reactant spatial distribution.
Fig. 3: Comparison between 3D simulations and experiments.
a DT∕DD reaction yield ratios corrected for as-shot variations. The black ‘x' symbols denote the experimental values with the solid black lines indicating the experimental measurement error, determined based on the method described in ref. 62. The black dashed line indicates the maximum theoretical yield ratio (Eq. (3)) for uniform atomic mix and the shaded region indicates uncertainty. This uncertainty is the maximum and minimum possible values calculated based on measurement uncertainties of foam densities, foam composition, fill pressure, and fill gas composition. Measurement errors are quantified in the Methods section and described in42,43,44,63,64,65. The red plus symbols indicate values that would have resulted from simulations if the reactants were in thermal equilibrium. b Comparison of experimental burn-weighted Ti, simulated thermal burn-weighted Ti from 3D simulations, and simulated apparent (including fluid motion corrections based on ref. 66) burn-weighted Ti from 3D simulations. A forward-fit approach using a relativistic energy distribution is used to analyzed the neutron time-of-flight signal to infer the neutron-average ion temperature, and the error bars are determined from the statistical uncertainty in this fit to the data67.
Fig. 4: Comparison of simulated species temperatures.
Mass-weighted ion temperatures for each reacting material in 3D simulations with varied pore size Data is included for simulations with 30 μm (black), 50 μm (blue), and 90 μm (red) pore sizes. The temperatures for the CH0.5D0.5 foam matrix use solid lines and the temperatures for the HT gas fill use dashed lines. The HT temperatures are consistently higher than the CH0.5D0.5 temperatures throughout the compression- and burn-phases of the implosion.
Simulated mass-weighted temperatures are in Fig. 4 (results for Te are similar). In all cases, the gas heats more than the foam matrix. The foam temperature peaks later than the HT temperature due to convective heating: vortical flows atomically mix hot HT with colder foam material, heating the latter. The level of separation between temperatures is proportional to pore sizes: simulations with smaller pores exhibit more atomic mixing, enabling local temperature equilibration. The 30 μm pore simulation exhibits the most atomic mix, as quantified by the spatial distribution of HT mass, whereas the 50 and 90 μm pore simulations demonstrate decreasing levels of atomic mix. This behavior is visualized in Fig. 5, where we visualize material distributions at bang time, where it is possible to observe the persistence of larger scale regions of pure HT in implosions with larger pores.
Fig. 5: Simulation visualizations.
Visualizations of the shock (red), pore interfaces (green), and hot spot boundary (blue) at bang time for the three different initial pore sizes. A quadrant has been cut out of the shock and pore data in order to aid in the observation of internal features.
MARBLE experiments underway at the NIF have demonstrated similar trends to the data presented above. With more laser power available to drive larger capsules, it is possible to obtain more implosion data, including X-ray imaging and BWTIs for both reactions. Results exhibit an increase in BWTI with increased pore size consistent with the results in this paper. Results also indicate a small increase in DT∕DD yield ratio with increased pore size, consistent with temperature separation between reactants.
Study highlights persistent unexplained gender differences in pay for physicians
When taking into account factors such as work-life balance, the pay difference between new male and female physicians is still largely unaccounted for, according to findings that were published Jan. 22 ahead of print and will also appear in the February issue of the journal Health Affairs.
The new study, "Differences in starting pay for male and female physicians persist; Explanations for the gender gap remain elusive," highlights continued disparity and persistent unexplained gender differences in pay for new physicians. Researchers looked at survey data from new physicians in New York who accepted positions in patient care from 1999-2017, and examined how the gender gap in total starting pay evolved and the extent to which preferences in work-life balance factors affect the gap.
Prior research, including my own, has documented a large gender gap in salary even after accounting for medical specialty, hours seeing patients, and practice setting among other factors. But we're first to incorporate physician preferences for work-life balance factors."
Anthony T. Lo Sasso, lead author, professor of economics in DePaul University's Driehaus College of Business
Specialty chosen consistently explained 40-55% of the total starting salary differences, while differences in number of job offers explained 2-9%, and hours of time spent in patient care explained 7%, according to the findings. However, despite women being much more likely than men to report that work-life balance factors were "very important", when these work-life balance variables were added to the researchers' model, the salary differences changed only negligibly (less than $1,000). In addition, work-life balance factors when added in from 2014-17, only explained less than 1% of the starting salary difference. Overall, 30-39% of the starting salary difference remains unexplained, the research showed.
Co-authors of the study are David Armstrong and Gaetano Forte from the Center for Health Workforce Studies at the University of Albany, and Susan E. Gerber, a faculty member in the departments of obstetrics and gynecology and medical education housed in the Feinberg School of Medicine at Northwestern University.
Among the findings:
The research also showed that women:
As next steps, researchers may include a series of questions in future surveys of new physicians that get at non-monetary job characteristics and whether their new job has those characteristics, which could help account for differences in the job beyond salary, explained Lo Sasso.
"To continue searching for an answer to this question, we believe it's important to dive deeper into job characteristics that are important and could plausibly differ by gender such as the role of negotiation, the amount of call, and the predictability of hours, among other factors," said Lo Sasso. "I'm primarily a quantitative researcher, but I'm not adverse to doing some qualitative research and talking to new physicians to learn about their job search experience.
"Until a firm conclusion has been reached on the reasons behind the gender pay gap in new physicians, it's important for employers to remain vigilant in ensuring pay equity. Greater transparency is needed, such as the salary difference between those physicians willing to do more on-call work and those who aren't. We also encourage residency programs to provide increased training on job-searching skills such as salary negotiation," Lo Sasso added.
Source:
Journal reference:
Lo Sasso, A. T., et al. (2020) Differences In Starting Pay For Male And Female Physicians Persist; Explanations For The Gender Gap Remain Elusive. Health Affairs. doi.org/10.1377/hlthaff.2019.00664.
No comments:
Post a Comment