RESUMEN
Frequency dependence in phonon surface scattering is a debated topic in fundamental phonon physics. Recent experiments and theory suggest such a phenomenon, but an independent agreement between the two remains elusive. We report low-temperature dependence of thermal conductivity in silicon nanowires fabricated using a two-step, metal-assisted chemical etch. By reducing etch rates down to 0.5 nm/s from the typical >100 nm/s, we report controllable roughening of nanowire surfaces and selectively focus on moderate roughness scales rather than the extreme scales investigated previously. This critically enables direct comparison with perturbation-based spectral scattering theory. Using experimentally characterized surface roughness, we show that a multiple scattering theory provides excellent agreement and explanation of the observed low-temperature dependence of rough surface nanowires. The theory does not employ any fitting parameters. A 5-10 nm roughness correlation length is typical in metal-assisted chemical etching and resonantly scatters dominant phonons in silicon, leading to the observed ~T(1.6-2.4) behavior. Our work provides fundamental and quantitative insight into spectral phonon scattering from rough surfaces. This advances applications of nanowires in thermoelectric energy conversion.
Asunto(s)
Nanocables/química , Fonones , Silicio/química , Metales/química , Tamaño de la Partícula , Propiedades de Superficie , Conductividad TérmicaRESUMEN
SETTING: Tuberculosis (TB) is the most common cause of pericarditis worldwide and carries a high mortality, even with effective anti-tuberculosis treatment. In the light of a randomized control trial in 2014, the American Thoracic Society and the Centers for Disease Control and Prevention/Infectious Diseases Society of America recently revised their recommendations against the routine use of adjunctive corticosteroids. OBJECTIVE: To evaluate the strength of evidence that resulted in this reversal of the guideline recommendations on the use of adjunctive corticosteroids in TB pericarditis by a meta-analysis, followed by a sensitivity analysis. DESIGN: Systematic review and meta-analysis of published randomized control trials. RESULTS: We identified five randomized control trials that met the eligibility criteria. Combining the results of the included trials, there was no overall mortality benefit from adjunctive corticosteroids (a random-effects model yielded a non-significant relative risk of 0.66 and 95%CI of 0.35-1.27). A sensitivity analysis further confirmed that the results of the meta-analysis were robust. CONCLUSION: Routine addition of oral corticosteroids to standard anti-tuberculosis treatment does not reduce mortality among patients with TB pericarditis.