Sharif Digital Repository

The production of porous (SAFFIL) preforms and SAFFIL/SiCP hybrid preforms will be discussed. SAFFIL fibers and SiCP are dispersed into an aqueous medium that consists of colloidal silica binder, distilled water, dispersants and deflocculants using agitation, spontaneous sedimentation, followed by drying and final strengthening. The effect of different process parameters such as various deflocculants and surfactants on the mechanical properties, mixing time and volumetric change of these preforms will also be investigated  

Synthesis of dextran-graft-poly(hydroxyethyl methacrylate) gels as a new fluoride biosorbent was considered in this work. For this propose, the Taguchi experimental design method was used for optimizing the synthetic conditions of the gels to reach high level of fluoride absorbency. The effects of three main parameters including concentrations of monomer (hydroxyethyl methacrylate), crosslinking agent (ethylene glycol dimethacrylate), and initiator (ammonium persulfate) on the final properties of the prepared gels were investigated. The proposed mechanism for grafting and chemically crosslinking reactions was proved with equilibrium water absorption, Fourier-transformed infrared, scanning... 

Deep structures consisting of many layers of nonlinearities have a high potential of expressing complex relations if properly initialized. Autoencoders play a complementary role in training a deep structure by initializing each layer in a greedy unsupervised manner. Due to the high capacity presented by autoencoders, these structures need to be regularized. While mathematical regularizers (based on weight decay, sparsity, etc.) and structural ones (by way of, e.g., denoising and dropout) have been well studied in the literature, quite a few papers have addressed the problem of training autoencoder with non-smooth regularization. In this paper, we address the problem of training autoencoder... 

It has been shown that more accurate signal recovery can be achieved with low-coherence dictionaries in sparse signal processing. In this Letter, the authors extend the low-coherence attribute to receptive fields in convolutional neural networks. A new constrained formulation to train low-coherence convolutional neural network is presented and an efficient algorithm is proposed to train the network. The resulting formulation produces a direct link between the receptive fields of a layer through training procedure that can be used to extract more informative representations from the subsequent layers. Simulation results over three benchmark datasets confirm superiority of the proposed... 

Deep neural networks have presented superlative performance in many machine learning based perception and recognition tasks, where they have even outperformed human precision in some applications. However, it has been found that human perception system is much more robust to adversarial perturbation, as compared to these artificial networks. It has been shown that a deep architecture with a lower Lipschitz constant can generalize better and tolerate higher level of adversarial perturbation. Smooth regularization has been proposed to control the Lipschitz constant of a deep architecture and in this work, we show how a deep convolutional neural network (CNN), based on non-smooth regularization... 

Transformer networks, originally developed in the community of machine translation to eliminate sequential nature of recurrent neural networks, have shown impressive results in other natural language processing and machine vision tasks. Self-attention is the core module behind visual transformers which globally mixes the image information. This module drastically reduces the intrinsic inductive bias imposed by CNNs, such as locality, while encountering insufficient robustness against some adversarial attacks. In this paper we introduce K-sparse attention to preserve low inductive bias, while robustifying transformers against adversarial attacks. We show that standard transformers attend... 

27Al, 17O and 13C chemical shieldings of aluminum acetylacetonate complex, Al(acac)3, were calculated at some Density Functional Theory (DFT) levels of theory. In these calculations the X-ray structures of its different polymorphs were used. Using these calculated data observed discrepancies between the X-ray crystallography and solid state NMR experiment were explained in terms of the quality of the NMR data. In this survey we resorted to the simulated spectra using our calculated chemical shifts. Inorder toconfirm our conclusions, electric field gradient (EFG) tensors of the 27Al and 17O nuclei were calculated at the same levels of theory as used in the chemical shielding calculations. On... 

Previous studies on wireless sensor networks indicate one of the most important research topics is to consider both significance of the reduction in sensor nodes energy consumption and improving network lifetime. Regarding results of some researchers using sinks, mobility has a substantial impact on the performance of sensor networks concerning reducing total energy consumption, increasing the number of active nodes and solving the energy-hole problem. In this study, we consider hexagons beehives feature where sensor nodes are distributed across a hexagon randomly. This hexagonal is divided into equal clusters based on the radius of hexagonal. Also, the need for cluster heads to be close to... 

In this paper, we study the problem of estimating the direction of arrival (DOA) using a sparsely sampled uniform linear array (ULA). Based on an initial incomplete ULA measurements, our strategy is to choose a sparse subset of array elements for measuring the next snapshot. Then, we use a Hankel-structured matrix completion to interpolate for the missing ULA measurements. Finally, the source DOAs are estimated using a subspace method such as Prony on the fully recovered ULA. We theoretically provide a sufficient bound for the number of required samples (array elements) for perfect recovery. The numerical comparisons of the proposed method with existing techniques such as atomic-norm... 

A novel biopolymer-based superabsorbent hydrogel was synthesized through chemical crosslinking by graft copolymerization of partially neutralized acrylic acid onto the hydrolyzed collagen, in the presence of a crosslinking agent and a free radical initiator. The Taguchi method, a robust experimental design, was employed for the optimization of the synthesis reaction based on the swelling capacity of the hydrogels. This method was applied for the experiments and standard L16 orthogonal array with three factors and four levels were chosen. The critical parameters that have been selected for this study are crosslinker (N,N′-methylene bisacrylamide), initiator (potassium persulfate), and monomer... 

The system, Pd(OAc)2/imidazolium salts (L2), was found as an efficient catalyst in the Heck coupling reaction of olefins with aryl halides and Suzuki reactions of various aryl halides with aryl boronic acids under aerobic condition. This catalytic system demonstrates great tolerance to a wide range of groups on all substrates of aryl halides, alkenes and aryl boronic acids  

The belief propagation algorithm is a state of the art decoding technique for a variety of linear codes such as LDPC codes. The iterative structure of this algorithm is reminiscent of a neural network with multiple layers. Indeed, this similarity has been recently exploited to improve the decoding performance by tuning the weights of the equivalent neural network. In this paper, we introduce a new network architecture by increasing the number of variable-node layers, while keeping the check-node layers unchanged. The changes are applied in a manner that the decoding performance of the network becomes independent of the transmitted codeword; hence, a training stage with only the all-zero... 

In this work, we present the design and fabrication of a novel nanocomposite based on noble metal and metal oxide nanoparticles dispersed on highly porous carbon obtained via the pyrolysis of an inorganic complex and metal-organic frameworks. This nanocomposite is prepared by a two-step procedure: first, the composite support of nanoporous carbon (NPC) is obtained by the direct carbonization of the Cr-benzene dicarboxylic ligand (BDC) MOF in an Argon atmosphere at 500 °C (Cr2O3-NPC). A mixture containing Cr2O3-NPC and [PtCl(SnCl3)(SMe2)2] is then prepared, and underflow of Argon is heated to 380 °C. Finally, Pt-SnO2 nanoparticles are loaded on the Cr2O3-NPC support, and the obtained... 

The effects of hydrophilic and hydrophobic nanosilica (SiO2) on the morphology, mean diameter distribution of fibers, mechanical and piezoelectric properties of poly (vinylidene fluoride) (PVDF) nanofibers were studies. We prepared Nanofibers by the electrospinning of PVDF solutions containing 1.5 wt.% both hydrophilic and hydrophobic nano-SiO2 loadings. Morphology and diameter distribution of the electrospun nanofibers were studied using field emission scanning electron microscopy (FE-SEM) analysis. Tensile test was used to study the effect of both types of nanosilica on the tensile strength, young's modulus and strain at break. Piezoelectric characterization of the electrospun fibers were... 

In this comment, it is shown that there are some non-negligible big mistakes in the analyses and stability proof of the proposed controller in the quoted paper, which makes the main results of this paper to be incorrect. The main unavoidable mistakes in the stability analysis of the main theorem (Theorem 1) are stated and some remarks are also mentioned to fix some of them. © 2022 Chinese Automatic Control Society and John Wiley & Sons Australia, Ltd  

In this study, crosslinked and non-crosslinked carboxymethylcellulose (CMC) membranes were prepared with different concentrations of polymer. Then, the permeability of pure CO2, N2, and CH4 was measured through these membranes in dry state to investigate the influence of polymer concentration and applied feed pressure on permeability and permselectivity. The permeability of CO2 through membranes was higher than the other gases. A comparison of permeabilities revealed that the permeability of N2, CO2, and CH4 increased on an average of 33, 40 and 20 percent, respectivly, by increasing the feed pressure from 6 to 10... 

In this comment, it is shown that there are some non-negligible big mistakes in the analyses and stability proof of the proposed controller in the quoted paper, which makes the main results of this paper to be incorrect. The main unavoidable mistakes in the stability analysis of the main theorem (Theorem 1) are stated and some remarks are also mentioned to fix some of them. © 2022 Chinese Automatic Control Society and John Wiley & Sons Australia, Ltd  

A simple ultrasound-assisted dispersive micro-solid phase extraction method based on graphene-oxide (GO) nanosorbents (UAD-μ-SPE), followed by high performance liquid chromatography-ultraviolet detection (HPLC-UV) was developed for the simultaneous determination of theophylline, theobromine and caffeine in different infusion tea samples. GO was synthesized by Hummers method and characterized with field emission-scanning electron microscopy (FE-SEM), X-ray diffraction (XRD), and Fourier transform infrared (FT-IR) spectroscopy. The GO sorbent was dispersed in the sample solution with the assistance of ultrasound to enhance adsorption and desorption performance and rates. The main parameters... 

Molybdenum disulfide (MoS2), a two-dimensional transition metal has a 2D layered structure and has recently attracted attention due to its novel catalytic properties. In this study, MoS2 has been successfully intercalated using chemical and physical intercalation techniques, while enhancing its surface properties. The final intercalated MoS2 is of many interests because of its low-dimensional and potential properties in in-situ catalysis. In this research, we report different methods to intercalate the layers of MoS2 successfully using acid-treatment, ultrasonication, oxidation and thermal shocking. The other goal of this study is to form S[dbnd]O bonds mainly because of expected enhanced... 

In this work, a nanocomposite containing manganese dioxide (MnO2) modified reduced graphene oxide (rGO) supported bimetallic palladium-nickel (Pd-Ni) catalyst is prepared by electrodeposition method. The nanocomposite modifier film is prepared by forming a thin layer of graphene oxide (GO) via drop-casting of GO nanosheet dispersion on glassy carbon electrode (GCE), followed by electrochemical reduction of the film to provide rGO/GCE. Then, a two-step potential procedure is applied to deposit MnO2 nanoparticles on rGO/GCE. At the optimum deposition conditions, MnO2 nanoparticles with a thickness of 30-50 nm homogeneously covered the rGO surface (MnO2/rGO/GCE). Finally, the bimetallic Pd-Ni...