We investigated the effect of hydrophobic extract concentration on the safety of using hand dishwashing liquids (HDL). A series of formulations was prepared, differing in the concentration of the hydrophobic chamomile extract obtained in supercritical CO2 conditions (from 0 to 0.7 %). We found that an increase in the concentration of the extract led to a decrease in the zein number, and reduced changes in the pH level of bovine serum albumin solution (i.e., two parameters determining the irritant activity of the formulations). It was also found that the additives reduced transepidermal water loss and improved the skin hydration level. Based on the findings of the study, a mechanism has been proposed, according to which hydrophobic plant extracts form aggregates in the volume phase of the washing bath. The surface of the aggregates is the adsorption area for surfactant monomers responsible for the irritant effect. Increasing the addition of the extract was shown to reduce the negative impact of the formulations on the skin of the hands, thus contributing to a greater safety of use of HDL.
This paper describes a method that allowed counting of both the total culturable and antagonistic microorganisms in a given source such as compost. Fusarium solani, used as the test fungus, was spreadplated on quarter-strength (1/4) potato dextrose agar (PDA), its surface was exposed in a laminar flow for 4 h and then another layer (2-3 mm thick) of 1/4 PDA was poured over it, on which an appropriate dilution of a compost sample was spread-plated. Microorganisms in the compost samples appeared first, and were counted as total culturable organisms. Plates were further incubated until F. solani grew through the upper layer of PDA (generally in 4-8 days) and covered the whole plate including most of the microbial colonies, except for a few which had a halo around them. These were counted as antagonistic, and they were isolated and purified for further studies. The population of bacteria in the six specific compost samples (called Biodynamic or BD preparations by organic farmers) ranged from 3.45 log 10 (in BD502) to 8.59 log 10 (in BD504) per gram of materials. The population of antagonistic bacteria was counted for three of the six compost samples, and ranged from 3.24 log 10 (in BD502) to 6.90 log 10 (in BD500). Of the 67 bacterial isolates showing a halo that were assembled from different sources, 17 suppressed at least 1 of the 4 plant pathogenic fungi against which these were evaluated using the dual culture method.
Modern search systems use several large ranker models with transformer architectures. These models require large computational resources and are not suitable for usage on devices with limited computational resources. Knowledge distillation is a popular compression technique that can reduce resource needs of such models, where a large teacher model transfers knowledge to a small student model. To drastically reduce memory requirements and energy consumption, we propose two extensions for a popular sentence-transformer distillation procedure: generation of an optimal size vocabulary and dimensionality reduction of the embedding dimension of teachers prior to distillation. We evaluate these extensions on two different types of ranker models. This results in extremely compressed student models whose analysis on a test dataset shows the significance and utility of our proposed extensions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.