Combining convolutional neural networks and in‐line near‐infrared spectroscopy for real‐time monitoring of the chromatographic elution process in commercial production of notoginseng total saponins

The chromatographic elution process is a key step in the production of notoginseng total saponins. Due to quality variability of loading samples and resin capacity decreasing over cycle time, saponins, especially the five main saponins of notoginseng total saponins, need to be monitored in real time during the elution process. In this study, convolutional neural networks, one of the most popular deep learning methods, were used to develop quantitative calibration models based on in‐line near‐infrared spectroscopy for notoginsenoside R₁, ginsenosides Rg₁, Re, Rb₁ and Rd, and their sum concentration, with root mean square error of prediction values of 0.87, 2.76, 0.60, 1.57, 0.28, and 4.99 mg/mL, respectively. Partial least squares calibration models were also developed for model performance comparison. Results show predicted concentration profiles outputted by both the convolutional neural network models and partial least squares models show agreements with the real trends defined by reference measurements, and can be used for elution process monitoring and endpoint determination. To the best of our knowledge, this is the first reported case study of combining convolutional neural networks and in‐line near‐infrared spectroscopy for monitoring of the chromatographic elution process in commercial production of botanical drug products.     

A Simple and Strong Electron‐Deficient 5,6‐Dicyano[2,1,3]benzothiadiazole‐Cored Donor‐Acceptor‐Donor Compound for Efficient Near Infrared Thermally Activated Delayed Fluorescence

Despite the success of thermally activated delayed fluorescent (TADF) materials in steering the next generation of organic light‐emitting diodes (OLEDs), effective near infrared (NIR) TADF emitters are still very rare. Here, we present a simple and extremely high electron‐deficient compound, 5,6‐dicyano[2,1,3]benzothiadiazole (CNBz), as a strong electron‐accepting unit to develop a sufficiently strong donor‐acceptor (D?A) interaction for NIR emission. End‐capping with the electron‐donating triphenylamine (TPA) unit created an effective D?A?D type system, giving rise to an efficient NIR TADF emissive molecule (λ_em=750 nm) with a very small ΔE_ST of 0.06 eV. The electroluminescent device using this NIR TADF emitter exhibited an excellent performance with a high maximum radiance of 10020 mW Sr^?1 m^?2, a maximum EQE of 6.57% and a peak wavelength of 712 nm.     

无衍射光的干涉实验与理论分析

分析了两束无衍射光的干涉场分布形式和干涉条纹轨迹。将一束单色光入射两小孔产生的两束相干光照射轴锥镜,在轴锥镜后将产生两束无衍射光。根据单束倾斜光入射轴锥镜的无衍射理论,分析出这两束无衍射光产生的干涉场为每束无衍射光的无衍射场的线性叠加。利用零阶贝塞尔函数的零点公式,推导出两束无衍射光的干涉条纹的轨迹为双曲线。计算结果表明,干涉场中两中心的间距与两孔实际的间距和干涉场距轴锥镜的距离成正比。实验结果与理论仿真相一致。     

基于麻雀搜索算法结合深度前馈神经网络的近红外模型转移方法研究

该文提出了一种基于麻雀搜索算法结合深度前馈神经网络（SSA-DFN）的近红外光谱模型转移方法。使用深度前馈神经网络拟合不同仪器采集到的光谱之间的非线性函数映射,并将麻雀搜索算法用于网络各层连接权值和阈值的初始化,通过种群中个体位置的迭代更新,求得连接权值和阈值的最优初始值;通过多次调整深度前馈神经网络模型的超参数,使网络拟合效果趋于最优,最终确定转移函数。为验证方法的有效性,分别从烟叶近红外光谱谱图、主成分投影和预测结果的角度,将SSA-DFN方法与分段直接校正算法（PDS）、典型相关性分析算法（CCA）转移前后的效果进行了对比。结果表明SSA-DFN方法转移后的从机光谱与原主机光谱重合度最高,转移后主、从机总糖、烟碱含量的预测结果差异不显著,预测平均误差从8.32%、9.15%分别降至4.65%、4.82%,预测均方根误差（RMSEP）和决定系数（R2）等指标均优于PDS和CCA,取得了最佳的转移效果,可满足企业需求。结果表明该方法是一种有效的模型转移方法。     

具有聚集诱导发光性质的近红外荧光染料

聚集诱导发光(AIE)现象的发现为解决传统有机荧光分子在高浓度和聚集形态下存在的荧光猝灭问题提供了最佳方案,并实现了在光电器件、化学传感、生物成像和靶向治疗等众多领域的广泛应用.随着对AIE 发光机理研究的不断深入,AIE 分子体系得到了极大的扩展.其中,一类具有给体-受体结构的AIE分子能够显著降低分子能隙,使发光分...     

近红外快速水份检测技术的应用

近红外检测技术是目前发展最快和最具有前景的分析技术之一,简要介绍了近红外检测技术的发展、优点和原理.利用该技术设计了豆类产品的水份检测仪器.并对近红外技术在水果、蔬菜、奶制品、制药和化工等方面的应用前景做了展望.     

F−(H2O)+CH3I配体交换反应动力学

本文介绍F^?(H₂O)+CH₃I→[FCH₃I]^?+H₂O在交叉分子束碰撞能量0.3∽2.6 eV的配体交换动力学成像结果. 产物的动能受到弱键结合配合物的稳定性的影响,大量水分子的内部激发不利于中间物有效的能量重新分配,随着碰撞增加,低动能受到抑制. 在0.3 eV时,内部亲核取代非常重要,为形成I^?和I^?(H₂O)的竞争性亲核取代途径提供了依据.     

Paraxial propagation of cosh-Airy vortex beams in chiral medium

Propagation dynamics of the cosh-Airy vortex(CAiV) beams in a chiral medium is investigated analytically with Huygens–Fresnel diffraction integral formula. The results show that the CAiV beams are split into the left circularly polarized vortex(LCPV) beams and the right circularly polarized vortex(RCPV) beams with different propagation trajectories in the chiral medium. We mainly investigate the effect of the cosh parameter on the propagation process of the CAiV beams.The propagation characteristics, including intensity distribution, propagation trajectory, peak intensity, main lobe's intensity, Poynting vector, and angular momentum are discussed in detail. We find that the cosh parameter affects the intensity distribution of the CAiV beams but not its propagation trajectory. As the cosh parameter increases, the distribution areas of the LCPV and RCPV beams become wider, and the side lobe's intensity and peak intensity become larger. Besides, the main lobe's intensity of the LCPV and RCPV beams increase with the increase of the cosh parameter at a farther propagation distance, which is confirmed by the variation trend of the Poynting vector. It is significant that we can vary the cosh parameter to control the intensity distribution, main lobe's intensity, and peak intensity of the CAiV beams without changing the propagation trajectory. Our results may provide some support for applications of the CAiV beams in optical micromanipulation.     

Pulsed,Crossed, Supersonic Molecular Beam Studies of Refractory Atom Reactions

Dynamics of refractory atom reactions have been studied with a crossed beam apparatus combining two pulsed, supersonic molecular beam sources, a pulsed UV laser for creating the refractory atoms in the gas phase by laser ablation, and a pulsed dye laser to probe the reaction products by laser-induced fluorescence. Examples of the A1(²P_j) + O₂(X³∑_g)→ A10(X²∑⁺) + O(³P_j), Mg(¹S_o) + N₂O(X¹∑⁺) → MgO(X¹∑⁺,a³Π) + N₂(X¹∑_g⁺) andC(³P_j) + NO(X²Π_r) → CN(X²∑⁺) + 0(³P_j) systems are given. Comparisons with the studies performed using the conventional steady-state beam approach are made.     

声光可调-近红外光谱技术分析烟草主要化学成分

建立了声光可调-近红外光谱方法(AOTF-NIR)检测烟草主要化学成分的方法。应用AOTF-NIR光谱仪测定了不同地区、不同等级烟草样品的近红外光谱,用Unscrambler(定量分析软件将光谱与对应的化学成分值相关联,建立了烟草中总糖、还原糖、总烟碱和钾的回归模型。用这些模型对未知样品进行了预测。总糖、还原糖、总烟碱和钾模型预测的平均相对标准偏差分别为2.71%、3.13%、4.04%和6.42%。