Designing optical glasses by machine learning coupled with a genetic algorithm

Engineering new glass compositions have experienced a sturdy tendency to move forward from (educated) trial-and-error to data- and simulation-driven strategies. In this work, we developed a computer program that combines data-driven predictive models (in this case, neural networks) with a genetic algorithm to design glass compositions with desired combinations of properties. First, we induced predictive models for the glass transition temperature (T_g) using a dataset of 45,302 compositions with 39 different chemical elements, and for the refractive index (n_d) using a dataset of 41,225 compositions with 38 different chemical elements. Then, we searched for relevant glass compositions using a genetic algorithm informed by a design trend of glasses having high n_d (1.7 or more) and low T_g (500 °C or less). Two candidate compositions suggested by the combined algorithms were selected and produced in the laboratory. These compositions are significantly different from those in the datasets used to induce the predictive models, showing that the used method is indeed capable of exploration. Both glasses met the constraints of the work, which supports the proposed framework. Therefore, this new tool can be immediately used for accelerating the design of new glasses. These results are a stepping stone in the pathway of machine learning-guided design of novel glasses.     

Cleaning capacity of octenidine as root canal irrigant: A scanning electron microscopy study

The aim of this study was to assess the cleaning capacity of the octenidine hydrochloride (OCT) used as root canal irrigant by scanning electron microscopy (SEM) analysis. Sixty human unirradicular extracted teeth were randomly distributed in 6 groups (n = 10) according to irrigant solutions which were used during root canal preparation: G1, 0.1% OCT; G2, 2% chlorhexidine (CHX); G3, 2.5% sodium hypochlorite (NaOCl); G4, OCT + 17% ethylenediaminetetraacetic acid (EDTA); G5, 2.5% NaOCl + 17% EDTA and G6, distilled water. All specimens were instrumented with ProTaper system up to F4. Teeth were sectioned and prepared for SEM. The smear layer was evaluated using a 5‐score system and the data were analyzed by Kruskal–Wallis and Dunn (α = 0.05). In all root canal thirds there was no significant difference between OCT, CHX, NaOCl, and water groups (p > .05), and these groups showed higher smear layer values than NaOCl + EDTA and OCT + EDTA groups (p < .05). There was no significant difference between NaOCl + EDTA and OCT + EDTA groups (p > .05). It was concluded that OCT used as a single root canal irrigant presented poor cleaning capacity and could be used in association with a final irrigation with EDTA to obtain smear layer removal.     

Peracetic acid as a single endodontic irrigant: effects on microhardness,roughness and erosion of root canal dentin

The aim was to assess the effects of 1% peracetic acid (PAA) as a single endodontic irrigant on microhardness, roughness, and erosion of root canal dentin, compared with 2.5% sodium hypochlorite (NaOCl) and with 2.5% NaOCl combined with 17% EDTA. Forty human, single‐rooted tooth hemisections were submitted to Knoop microhardness test, before and after the following irrigation protocols: PAA = 1% PAA; NaOCl = 2.5% NaOCl; NaOCl‐EDTA‐NaOCl = 2.5% NaOCl +17% EDTA +2.5% NaOCl; and SS = saline. Another 40 roots were instrumented, irrigated with the same protocols, and sectioned longitudinally. The roughness analysis was performed on the mesial section using a confocal laser scanning microscope, whereas erosion was analyzed on each third of the distal section, using a scanning electron microscope. The data were analyzed using ANOVA and Tukey post‐tests, and Kruskal‐Wallis and Dunn post‐tests (α = .05). The PAA and NaOCl‐EDTA‐NaOCl groups showed no significant differences (p > .05); both promoted reduction in microhardness and increase in roughness, compared with the NaOCl and SS groups (p < .05). NaOCl‐EDTA‐NaOCl promoted higher erosion in the cervical and middle thirds than the other groups (p < .05); there was no difference among PAA, NaOCl, and SS (p > .05). There was also no difference among the groups regarding the apical third (p > .05). PAA used as a single endodontic irrigant caused reduction in root canal dentin microhardness and increase in roughness in a similar way to NaOCl‐EDTA‐NaOCl; however, PAA caused less erosion than NaOCl‐EDTA‐NaOCl.     

Limnological features of Funil Reservoir (R.J., Brazil) and indicator properties of rotifers and cladocerans of the zooplankton community

Funil Reservoir receives inflow from a highly industrialized region and acts as a natural sink to pollutants. Among the consequences of the uncontrolled nutrient loading is the constant presence and periodic heavy blooms of Microcystis aeruginosa . This study verified limnological features and zooplankton assemblage of the Funil Reservoir focusing on the environmental-indicator properties of rotifers and cladocerans. The summer bloom of M. aeruginosa caused reductions in water transparency, nitrate and orthophosphate concentrations and raises in chlorophyll a , pH, dissolved oxygen and chemical oxygen demand values. All zooplankton species presented spatial and temporal variations with the exception of the copepods, which were present in all samples. According to canonical correspondence analysis, ammonium was the variable most related to zooplankton variability and different Rotifera and Cladocera assemblages indicated distinct environment conditions. Rotifer taxa associated with increases of water temperature and chlorophyll a concentration were found with high densities during M. aeruginosa blooms. Species of rotifer and cladocerans are suggested as indicators that can be used to identify different physical and chemical gradients or eutrophic increases in Funil Reservoir.     

Crystallization of porcine insulin with carbon dioxide as acidifying agent

Recent studies on the use of volatile electrolytes such as CO₂ in protein precipitation showed that these agents are a promising alternative to the conventional acids. This use of volatile electrolytes prevents protein denaturation due to local pH extremes, and saline effluent generation is greatly reduced, as the volatile electrolyte may be separated and recovered from solution just by pressure release. In this work, insulin was successfully crystallized in the presence of zinc using CO₂ as acidifying agent. The crystals obtained were rhombohedral, a common shape for porcine insulin crystals that contain zinc in their structure, and their average size varied with the mixing applied.     

Analyzing the Trade-off Between Investing in Service Channels and Satisfying the Targeted User Service for Brazilian Internet Service Providers

The computer connection to the Internet is provided by firms known as internet service providers (ISPs). The simplest mode of physical connection is when the user connects to an ISP's service channel by an ordinary telephone line (dial-up). Finding an available channel may not be an easy task, especially during the peak hours of many Brazilian ISPs. This results in a problem for the ISPs, which is how to achieve the most appropriate trade-off between investing in capacity and satisfying the target user service level. This paper analyzes this trade-off based on a three-step approach: (i) determine user arrival and service processes in chosen periods, (ii) select an appropriate queueing model using some simplifying assumptions, and (iii) generate trade-off curves between system performance measures. To illustrate the application of this approach, some results derived from a case study performed at an ISP in Sao Paulo state are given.     

GaAs W-band IMPATT diodes for very low-noise oscillators

W-band single-drift flat-profile IMPATT diodes were fabricated from GaAs MBE material and tested in a full-height waveguide resonant cavity with resonant cap. A quasi-optical parabolic Fabry-Perot resonator was used to determine the FM noise of the GaAs IMPATT oscillator. With a minimum noise measure of 20 dB at power levels around 20 mW, IMPATT diode oscillators can compete well with oscillators using Gunn devices. The (N/C)/sub FM/=-82 dBc measured at 100 kHz frequency off-carrier and at Q/sub EX/=95 is comparable to the value obtained from Gunn devices. The maximum available output power of 270 mW, however, markedly exceeds that of Gunn oscillators.<>