Synthesis of zeolite LTN (“Linde Type N”) was investigated under insertion of a SiO2-rich filtration residue (FR) from waste water cleaning of the silane production. A new synthesis procedure was therefore developed...Synthesis of zeolite LTN (“Linde Type N”) was investigated under insertion of a SiO2-rich filtration residue (FR) from waste water cleaning of the silane production. A new synthesis procedure was therefore developed applying a flotation mechanism with the aim to grow LTN in form of thin membrane like sheets. Preparation starts with preactivation of FR by slurrying first in alkaline solution, followed by an addition of aluminate solution and citric acid. The latter was added as suitable chelating agent for the initiation of the flotation process. In the course of these experiments, we succeeded in synthesizing zeo-lite LTN with more or less zeolite SOD as byproduct in the form of a stable compact membrane-like layer at low temperature of 60℃. The crystallization was performed under isotherm static conditions in an open reaction system without addition of organic templates as structure directing agents (OSDA’s). FR was utilized as a total substitute of sodium silicate in all experiments and an expansive pre-treatment procedure like calcinations was not needed. Furthermore, membrane formation with LTN of usual synthesis needs chemically functionalized supports. In contrast self-supporting membranous LTN layers were grown for the first time in the present study.展开更多
Significant progress has been made in image inpainting methods in recent years.However,they are incapable of producing inpainting results with reasonable structures,rich detail,and sharpness at the same time.In this p...Significant progress has been made in image inpainting methods in recent years.However,they are incapable of producing inpainting results with reasonable structures,rich detail,and sharpness at the same time.In this paper,we propose the Pyramid-VAE-GAN network for image inpainting to address this limitation.Our network is built on a variational autoencoder(VAE)backbone that encodes high-level latent variables to represent complicated high-dimensional prior distributions of images.The prior assists in reconstructing reasonable structures when inpainting.We also adopt a pyramid structure in our model to maintain rich detail in low-level latent variables.To avoid the usual incompatibility of requiring both reasonable structures and rich detail,we propose a novel cross-layer latent variable transfer module.This transfers information about long-range structures contained in high-level latent variables to low-level latent variables representing more detailed information.We further use adversarial training to select the most reasonable results and to improve the sharpness of the images.Extensive experimental results on multiple datasets demonstrate the superiority of our method.Our code is available at https://github.com/thy960112/Pyramid-VAE-GAN.展开更多
文摘Synthesis of zeolite LTN (“Linde Type N”) was investigated under insertion of a SiO2-rich filtration residue (FR) from waste water cleaning of the silane production. A new synthesis procedure was therefore developed applying a flotation mechanism with the aim to grow LTN in form of thin membrane like sheets. Preparation starts with preactivation of FR by slurrying first in alkaline solution, followed by an addition of aluminate solution and citric acid. The latter was added as suitable chelating agent for the initiation of the flotation process. In the course of these experiments, we succeeded in synthesizing zeo-lite LTN with more or less zeolite SOD as byproduct in the form of a stable compact membrane-like layer at low temperature of 60℃. The crystallization was performed under isotherm static conditions in an open reaction system without addition of organic templates as structure directing agents (OSDA’s). FR was utilized as a total substitute of sodium silicate in all experiments and an expansive pre-treatment procedure like calcinations was not needed. Furthermore, membrane formation with LTN of usual synthesis needs chemically functionalized supports. In contrast self-supporting membranous LTN layers were grown for the first time in the present study.
基金The authors gratefully acknowledge the financial support of the National Natural Science Foundation of China(Grant No.61925603).
文摘Significant progress has been made in image inpainting methods in recent years.However,they are incapable of producing inpainting results with reasonable structures,rich detail,and sharpness at the same time.In this paper,we propose the Pyramid-VAE-GAN network for image inpainting to address this limitation.Our network is built on a variational autoencoder(VAE)backbone that encodes high-level latent variables to represent complicated high-dimensional prior distributions of images.The prior assists in reconstructing reasonable structures when inpainting.We also adopt a pyramid structure in our model to maintain rich detail in low-level latent variables.To avoid the usual incompatibility of requiring both reasonable structures and rich detail,we propose a novel cross-layer latent variable transfer module.This transfers information about long-range structures contained in high-level latent variables to low-level latent variables representing more detailed information.We further use adversarial training to select the most reasonable results and to improve the sharpness of the images.Extensive experimental results on multiple datasets demonstrate the superiority of our method.Our code is available at https://github.com/thy960112/Pyramid-VAE-GAN.
