Search

Physically Unclonable Functions (PUFs), as an alternative hardware-based security method, have been challenged by some modeling attacks. As is known to all, samples are significant in modeling attacks on PUFs, and thus, some efforts have been made to expand sample sets therein to improve modeling attacks. A closer examination, however, reveals that not all samples contribute to modeling attacks equally. Therefore, in this article, we introduce the concept of sample essentiality for describing the contribution of a sample in modeling attacks and point out that any sample without sample essentiality cannot enhance some modeling attacks on PUFs. As a by-product, we find theoretically and empirically that the samples expanded by the procedures proposed by Chatterjee et al. do not satisfy our sample essentiality. Furthermore, we propose the notion of essential sample sets for datasets and discuss its basic properties. Finally, we demonstrate that our results about sample essentiality can be used to reduce samples efficiently and benefit sample selection in modeling attacks on arbiter PUFs.

Adsorptive desulfurization (ADS) from fuels is considered as the most promising strategy due to the advantage of low energy consumption, to treat on sulfur which would induce serious air pollution. In this work, Sn-based DESs with urea were firstly synthesized and introduced into hexagonal boron nitride (h-BN) as both reaction solvents and metal sources. Adsorption experiments show that the as-prepared adsorbents present higher adsorption capacity for dibenzothiophene (DBT) from model oils as compared with the commercial BN. Especially, the Sn-BN-800 (SnO2 supported BN with OH groups substitution at B sites) shows the highest performance. By combining experimental and theoretical analysis, we propose that the performance enhancement arises from the π-complexation and hydrogen bonding interaction between SnO2 and DBT. In addition, the OH groups, which were substituted at the edge site of B, may also enhance the electron transfer from DBT to Sn-BN-800, leading to the improved adsorption capacity.

The selectivity for sulfur removal from oils is an important topic. In this work, the selectivity for different sulfur removal methods has been studied by conceptual density functional theory (CDFT) at the B3LYP/6-311++G(3df,2p) level of theory. In principle, the selectivity is directly related to the mechanisms of sulfur removal. It cannot be precisely elucidated until the mechanisms are totally known. However, current work shows that relationships can be constructed between CDFT and the selectivity. That is, for hydrodesulfurization, good descriptors will be ionization energy, hardness, and bond lengths of SC; for adsorptive desulfurization, the hardness is a good descriptor; for oxidative desulfurization, good descriptors are electron density and Fukui function. And for extractive desulfurization (nonmetal-based ionic liquids), electron affinity and electrophilicity may be good descriptors. In addition, structures and frontier orbitals of various sulfides have also been discussed. It is hoped that these relationships between CDFT and selectivity can give useful information to develop highly efficient sulfur removal methods for specific sulfides, like 4,6-dimethyldibenzothiophene, and 4-methyldibenzothiophene. © 2016 American Institute of Chemical Engineers AIChE J, 62: 2087–2100, 2016

The chemical hardness of adsorbents is an important physicochemical property in the process of adsorption based on the hard and soft acids and bases (HSAB) theory. Tuning chemical hardness of adsorbents modulated by their concomitants is a promising approach to enhance the adsorptive capacity in principle. In the present work, we report an efficient strategy that the adsorption capacity for aromatic sulfocompounds can be enhanced by tuning the chemical hardness. This strategy is first theoretically explored by introducing C element into the network of hexagonal boron nitride (h-BN) based on a series of model materials (model_xC, x = 1–5). Computational results show that the chemical hardness is reduced after gradually C-doping, which may lead to an enhancement of adsorption capacity according to the HSAB theory. Then, a series of C-doped h-BN materials (BCN-x, x = 10–50) were controlled synthesized. All of the as-prepared materials show better adsorption capacities (e.g., 27.43 mg g–1 for BCN-50) than pure h-BN. Experiment results show that the adsorption capacity correlates well with the C content in the BCN-x, which is consistent with the results predicted by theoretical calculation. This strategy may be helpful to rationally design highly efficient adsorbents in separation engineering and may be expanded to similar two-dimensional materials, where the π–π interaction is the dominant driven force.

Deep eutectic solvents (DESs), as ionic liquid analogues for green solvents, have gained increasing attentions in chemistry. In this work, three typical kinds of DESs (ChCl/Gly, ChCl/AcOH and ChCl/Urea) were successfully synthesized and characterized by Fourier transform infrared spectroscopy (FTIR) and Raman. Then comprehensive and systematical analyses were performed by the methods of density functional theory (DFT). Two methods (B3LYP/6-311++G(2d,p) and dispersion-corrected B3LYP-D3/6-311++G(2d,p)) were employed to investigate the structures, vibrational frequencies and assign their ownership of vibrational modes for the DESs, respectively. Nearly all the experimental characteristic peaks of IR and Raman were identified according to the calculated results. By linear fitting of the combined calculated vs experimental vibration frequencies, it can be found that both of the two methods are excellent to reproduce the experimental results. Besides, hydrogen bonds were proved to exist in DESs by IR spectrum, structure analysis and RDG analysis. This work was aimed at predicting and understanding the vibrational spectra of the three typical DESs based on DFT methods. Moreover, by comparing experimental and theoretical results, it provides us a deep understanding of the formation mechanisms of DESs.

The principle of kernel Extreme Learning Machine (ELM) is demonstrated. On this basis, a multi - scale wavelet kernel extreme learning machine is proposed. The multi-scale wavelet kernel is used as the kernel function of the extreme learning machine. The test shows that it is an achievable extreme learning machine. Experiments show that, using the multi-scale wavelet kernel extreme learning machine in the wind turbine bearing fault diagnosis has higher classification accuracy and speed than the support vector machine classification algorithm, and has excellent application value.