CIESC Journal ›› 2024, Vol. 75 ›› Issue (2): 463-474.DOI: 10.11949/0438-1157.20230960

• Thermodynamics • Previous Articles     Next Articles

Prediction of vapor-liquid equilibrium data of uranium hexafluoride and fluoride and simulation of distillation process

Tong YANG1(), Huan WANG1,2, Chun DENG1()   

  1. 1.State Key Laboratory of Heavy Oil Processing, College of Chemical Engineering and Environment, China University of Petroleum, Beijing 102249, China
    2.CNNC No. 7 Research and Design Institute Co. , Ltd. , Taiyuan 030012, Shanxi, China
  • Received:2023-09-14 Revised:2024-01-09 Online:2024-04-10 Published:2024-02-25
  • Contact: Chun DENG

六氟化铀及氟化物汽液相平衡数据预测及精馏过程模拟

杨同1(), 王欢1,2, 邓春1()   

  1. 1.中国石油大学(北京)化学工程与环境学院,重质油全国重点实验室,北京 102249
    2.中核第七研究设计院有限公司,山西 太原 030012
  • 通讯作者: 邓春
  • 作者简介:杨同(1998—),男,博士研究生,1719408185@qq.com
  • 基金资助:
    国家自然科学基金项目(U2267226)

Abstract:

The purification of uranium hexafluoride by distillation technology is an important part of the uranium conversion process. The simulation of the distillation process can provide key support for process design and operation optimization. However, due to the lack of vapor-liquid equilibrium and physical properties data for uranium hexafluoride and fluoride, modelling and simulating the purification process of uranium hexafluoride became challenging. To address this issue, we utilized COSMOtherm and Turbomole software to predict vapor-liquid equilibrium data for UF6-TiF4 binary systems. We indirectly verified the accuracy of the COSMO-RS model by predicting known experimental data on vapor-liquid equilibrium for WF6-UF6 binary system. Additionally, Aspen Plus software's property constant estimation system was employed to estimate missing physical properties such as infinitely dilute aqueous Gibbs generating energy. The UF6 saturated vapor pressure experimental data and simulated values in the literature were compared, and the relative error was within 1.76%. Based on experimental data and predicted vapor-liquid phase equilibrium data in the literature, Aspen Plus software was used to regress the binary interaction parameters of the NRTL model. For separating uranium hexafluoride from fluoride, two purification schemes were designed: one involving direct separation sequence and another involving indirect separation sequence. Sensitivity analysis was conducted to optimize key parameters including plate number, feed position, and reflux ratio to minimize total annual cost (TAC) while maintaining desired purity levels in UF6 product output streams. The results indicated that TAC was lower for the indirect separation sequence purification scheme.

Key words: fluoride, physical property estimation, vapor-liquid equilibria, distillation

摘要:

精馏技术提纯六氟化铀是铀转化工艺的重要环节,该精馏工艺的模拟可为工艺设计及操作优化提供关键支撑。然而,由于六氟化铀及氟化物汽液相平衡数据及物性数据的缺乏,六氟化铀提纯精馏工艺的建模和模拟难以开展。为此,利用COSMOtherm和Turbomole软件,预测UF6与TiF4二元体系汽液相平衡数据。通过预测已知的WF6-UF6二元体系汽液相平衡实验数据以间接验证COSMO-RS模型预测的准确性。利用Aspen Plus软件的物性常数估算系统(property constant estimation system),估算出缺失的无限稀释水溶液Gibbs生成能等物性参数。文献中UF6饱和蒸气压实验数据与模拟值进行对比,相对误差在1.76%以内。根据文献中的实验数据和预测的汽液相平衡数据,利用Aspen Plus软件回归NRTL模型的二元交互作用参数。针对六氟化铀及氟化物的分离,设计直接分离序列和间接分离序列提纯UF6的两种方案。设定UF6产品纯度,以总年度费用(TAC)最小化为目标,采用灵敏度分析对塔板数、进料位置和回流比的关键参数进行优化,结果表明,间接分离序列提纯UF6方案的TAC更低。

关键词: 氟化物, 物性估算, 汽液平衡, 精馏

CLC Number: