Supercritical fluid process has been used in several industrial fields as a novel technique to produce nanoparticles. Throttling effect may occur when supercritical CO₂ and ethanol passes through a coaxial annular nozzle together in supercritical antisolvent, exerting considerable negative effects on particle size and morphology, thus, it is imperative to study the throttling effect. A new experimental system was developed to study the effects of inlet temperature, inlet pressure and ethanol content on the throttling effect of supercritical CO₂ system using a 100 μm diameter coaxial annular nozzle in this paper. Supercritical CO₂ and desired amount of ethanol were mixed in the coaxial annular nozzle and the temperature and pressure at the inlet and outlet of the nozzle were recorded by the data acquisition system. The results show that high inlet temperature and ethanol content can acquire higher throttling temperature while high inlet pressure enhances the throttling effect, obtaining a lower throttling temperature. The initial density and phase state were confirmed to be the key factors to affect the throttling effect. In order to accurately predict the throttling effect, a back-propagation neural network model with the Correlation Coefficient of 0.99 531 and the Mean Retive Error ranging from 1.0841 % to 1.3209 % was proposed based on the experimental data, which demonstrated that it can be used as a powerful tool to predict the throttling effect of supercritical CO₂ containing ethanol system.

超临界流体工艺作为一种生产纳米颗粒的新技术已被用于多个工业领域。当超临界CO ₂和乙醇在超临界反溶剂中一起通过同轴环形喷嘴时，可能会发生节流效应，对粒径和形貌产生相当大的负面影响，因此，研究节流效应势在必行。本文开发了一个新的实验系统，研究入口温度、入口压力和乙醇含量对超临界 CO ₂系统节流效果的影响，该系统使用直径为 100 μm 的同轴环形喷嘴。超临界 CO ₂在同轴环形喷嘴中混合所需量的乙醇，数据采集系统记录喷嘴入口和出口处的温度和压力。结果表明，较高的入口温度和乙醇含量可以获得较高的节流温度，而较高的入口压力会增强节流效果，获得较低的节流温度。初始密度和相态被证实是影响节流效果的关键因素。为了准确预测节流效果，基于实验数据提出了相关系数为0.99 531、平均相对误差为1.0841 % ~ 1.3209 %的反向传播神经网络模型，证明该模型是可以使用的作为预测超临界 CO 节流效果的有力工具₂含乙醇制。