TiO 2 nanoparticles were obtained from industrial TiOSO 4 by hydrolysis method. SnO 2/TiO 2 and SnO 2-TiO 2 composite powders were prepared by stepwise precipitation method and coating method, respectively. The phase ...TiO 2 nanoparticles were obtained from industrial TiOSO 4 by hydrolysis method. SnO 2/TiO 2 and SnO 2-TiO 2 composite powders were prepared by stepwise precipitation method and coating method, respectively. The phase transformation of TiO 2 and the effect of composite mode of SnO 2 on phase transformation of TiO 2 have been investigated by TG-DTA and XRD. The phase transform of pure TiO 2 from anatase to rutile begins at 750 ℃ and the presence of SnO 2 markedly reduces the transform temperature: for coated SnO 2-TiO 2 composite with ω(SnO 2)=20% it was 400 ℃. The SnO 2/TiO 2 composite prepared by precipitation method and followed by calcination at 400 ℃ for 30 min possesses 55% rutile TiO 2. The formation of SnO 2-TiO 2 solid- solution occurrs mainly due to the substitution of Ti 4+ crystal lattice sites by Sn 4+ ions of SnO 2.展开更多
文摘TiO 2 nanoparticles were obtained from industrial TiOSO 4 by hydrolysis method. SnO 2/TiO 2 and SnO 2-TiO 2 composite powders were prepared by stepwise precipitation method and coating method, respectively. The phase transformation of TiO 2 and the effect of composite mode of SnO 2 on phase transformation of TiO 2 have been investigated by TG-DTA and XRD. The phase transform of pure TiO 2 from anatase to rutile begins at 750 ℃ and the presence of SnO 2 markedly reduces the transform temperature: for coated SnO 2-TiO 2 composite with ω(SnO 2)=20% it was 400 ℃. The SnO 2/TiO 2 composite prepared by precipitation method and followed by calcination at 400 ℃ for 30 min possesses 55% rutile TiO 2. The formation of SnO 2-TiO 2 solid- solution occurrs mainly due to the substitution of Ti 4+ crystal lattice sites by Sn 4+ ions of SnO 2.
文摘准确模拟日光温室内环境的变化过程是实现温室环境精准调控的前提。该研究以3个生长季的日光温室室内实时气象观测资料为基础,利用Elman神经网络建模的方法,对日光温室室内1.5 m气温、0.5 m气温和CO2浓度进行逐时模拟,对日光温室室内平均湿度、平均温度、最高温度和最低温度进行逐日模拟,建立基于Elman神经网络的日光温室室内环境逐时及逐日模拟模型,利用独立的气象观测资料对模型进行验证,并基于逐步回归方法和BP神经网络方法结果进行对比分析。结果表明:1)基于Elman神经网络的日光温室室内环境(1.5m气温、0.5m气温和CO2浓度)逐时模拟值与实测值的均方根误差(Root Mean Square Error,RMSE)分别为2.14℃、1.33℃和55.32μmol/mol,归一化均方根误差(Normalized Root Mean Square Error,NRMSE)分别为10.01%、5.87%和10.70%,基于Elman神经网络的日光温室室内环境逐时模拟效果和稳定性最优。2)基于Elman神经网络的日光温室室内环境(日均空气湿度、日均气温、日最高气温和日最低气温)逐日模拟值与实测值的RMSE分别为0.59%、0.88℃、2.02℃和0.98℃,NRMSE分别为0.79%、4.44%、7.02%和6.66%,基于Elman神经网络的日光温室室内环境逐日模拟效果和稳定性最优。研究结果可以准确模拟日光温室室内逐时及逐日环境,也可以为环境模型与作物模型相互耦合提供技术支撑。
