A switchable-hydrophilicity solvent system, consisting of a fatty acid-based natural deep eutectic solvent (NaDES), complemented by a bio-friendly dilute amine solution, has been introduced. The potential of the most benign switchable solvent system has been characterised in microalgae biorefining according to the recently proposed ‘Circular Extraction’ scheme.

/pdf/a-novel-switchable-hydrophilicity-natural-deep-eutectic-4e2zu5nmax.pdf

A novel switchable-hydrophilicity, natural deep eutectic solvent (NaDES)-based system for bio-safe biorefinery

CO2/N2-responsive oil-in-water emulsions using a novel switchable surfactant.

Hypothesis One of the major challenges in applying CO2-responsive surfactants concerns their tunable switchability and robustness under operating conditions. We hypothesize that combining monoethanolamine (MEA) with long-chain fatty acids (LCFAs) of variable chain lengths through electrostatic attraction could develop a series of CO2-responsive surfactants with tunable switching pH. Experiments The tunability of switching pH for this group of surfactants was demonstrated by in situ probing of the CO2-responsive characteristics at the oil/water interface using dynamic interfacial tension (IFT) measurements. Two protocols were applied to distinguish interfacial response and solution response. The key importance of interfacial response was demonstrated by two essential applications of CO2-responsive surfactants: demulsification of stable emulsions, and alternation of the interfacial properties of ultra-heavy crude oil-water interfaces. Findings The switching pH of the CO2-responsive surfactants was controlled by the hydrocarbon chain length of LCFAs. More importantly, their switching behaviour was found to be different at the interface and in the bulk solution, which is attributed to the enhanced molecular interactions at the interface. Since most applications require surfactants to be switched at the interface, it is thereby most appropriate to determine the switching pH through their interfacial responses.

CO2-responsive surfactants with tunable switching pH.

Switchable solvent-based micro-extraction of pesticides in food and environmental samples

CO2 responsive emulsions: Generation and potential applications

A fatty acid solvent of switchable miscibility

Spiking neural networks (SNNs) are bio-inspired neural networks with asynchronous discrete and sparse characteristics, which have increasingly manifested their superiority in low energy consumption. Recent research is devoted to utilizing spatio-temporal information to directly train SNNs by backpropagation. However, the binary and non-differentiable properties of spike activities force directly trained SNNs to suffer from serious gradient vanishing and network degradation, which greatly limits the performance of directly trained SNNs and prevents them from going deeper. In this paper, we propose a multi-level firing (MLF) method based on the existing spatio-temporal back propagation (STBP) method, and spiking dormant-suppressed residual network (spiking DS-ResNet). MLF enables more efficient gradient propagation and the incremental expression ability of the neurons. Spiking DS-ResNet can efficiently perform identity mapping of discrete spikes, as well as provide a more suitable connection for gradient propagation in deep SNNs. With the proposed method, our model achieves superior performances on a non-neuromorphic dataset and two neuromorphic datasets with much fewer trainable parameters and demonstrates the great ability to combat the gradient vanishing and degradation problem in deep SNNs.

/pdf/multi-level-firing-with-spiking-ds-resnet-enabling-better-2w8hhwnz.pdf

Multi-Level Firing with Spiking DS-ResNet: Enabling Better and Deeper Directly-Trained Spiking Neural Networks

Viscosity reduction of extra-heavy oil using toluene in water emulsions

Human brain can effectively integrate visual and auditory information. Dynamic Vision Sensor (DVS) and Dynamic Audio Sensor (DAS) are event-based sensors imitating the mechanism of human retina and cochlea. Since the sensors record the visual and auditory input as asynchronous discrete events, they are inherently suitable to cooperate with the spiking neural network (SNN). Existing works of SNNs for processing events mainly focus on unimodality, however, audiovisual multimodal SNNs are still limited. In this paper, we propose an end-to-end event-based multimodal spiking neural network. The network consists of visual and auditory unimodal subnetworks and a novel attention-based cross-modal subnetwork for fusion. The attention mechanism measures the significance of each modality and allocates the weights to two modalities. We evaluate our proposed multimodal network on an event-based audiovisual joint dataset (MNIST-DVS and N-TIDIGITS datasets). Experimental results show the performance improvement of this multimodal network and the effectiveness of our proposed attention mechanism.

Event-Based Multimodal Spiking Neural Network with Attention Mechanism

Screening of suitable deep eutectic solvents (DESs) as extractants is vitally important in an extraction process. In this study, a multiscale method combining conductor-like screening model for real solvents (COSMO-RS) calculation, experimental validation, and process simulation is presented. This method was applied to screen DESs for extracting m-cresol from cumene. First, the COSMO-RS model was performed to calculate the phase equilibrium of different ternary systems at different feed compositions, thereby prescreening DESs by investigating the effects of DES structures on the extraction performance. Then, the prescreened DESs were studied by extraction experiments to further validate their extraction performance. The extraction mechanism was investigated through FT-IR characterization. Afterward, continuous process simulation by Aspen Plus was employed to identify more promising DESs. The COSMO-RS calculation and experimental results showed that both choline chloride (ChCl)/ethylene glycol (EG) (1:2) and ChCl/glycerol (Gly) (1:2) demonstrated a high extraction performance, which were selected as two suitable DESs. Considering the mass purity and recovery ratio of m-cresol and cumene products in industrial applications, as well as the extractant dosage and equipment costs, ChCl/Gly (1:2) is considered a more promising DES in industrial application.

https://pubs.acs.org/doi/pdf/10.1021/acsomega.2c04234

Qian Liu

Papers

A fatty acid solvent of switchable miscibility

Multi-Level Firing with Spiking DS-ResNet: Enabling Better and Deeper Directly-Trained Spiking Neural Networks

Viscosity reduction of extra-heavy oil using toluene in water emulsions

Event-Based Multimodal Spiking Neural Network with Attention Mechanism

Multiscale Screening of Deep Eutectic Solvents for Efficient Extraction of m-Cresol from Model Coal Tar