Scholarlink

<p>Vapor–liquid critical properties of mixtures are key parameters in the petrochemical industry and supercritical technology. Experimental measurements and theoretical calculations are the primary methods for determining the critical parameters of mixtures. However, existing empirical correlations to quickly predict the critical temperatures and pressures of mixtures are limited by critical volume data for pure substances. In this work, improved methods of Li method and Kreglewski–Li (KL) method are proposed. Improved methods do not require critical volume data for pure substances, but replace it with acentric factors, normal boiling points, or critical temperatures of pure substances that are easier to obtain and more accurate. About 9,000 critical temperature and critical pressure data points for binary and ternary mixtures were collected to compare and evaluate the Li method, KL method, and improved methods. Notably, the improved methods are only applicable to the class I and II mixtures according to the classification of Van Konynenburg and Scott. Overall, compared with the original method, both Improvement 3 (critical volumes for pure substances of Li method and KL method are replaced with critical temperatures of pure substances) and Improvement 4 (critical volumes for pure substances of Li method and KL method are replaced with normal boiling points of pure substances) greatly improve the accuracy. Meanwhile, when predicting critical temperatures and critical pressures, Improvement 3 not only reduces the input thermophysical property parameters but also improves the prediction accuracy. Among the improved methods, Improvement 4 shows the highest prediction accuracy. The average absolute relative deviation (AARD) and average absolute deviation (AAD) of Improvement 4 for predicting the critical temperatures of binary and ternary mixtures are 1.88%, 7.83&#xa0;K, 1.60%, and 7.63&#xa0;K, respectively. The AARD and AAD for predicting the critical temperature of the binary mixtures composed of two pure substances with both acentric factors greater than 0.0955 by Improvement 4 are 1.56% and 7.33&#xa0;K. The AARD and AAD of Improvement 4 for predicting the critical pressures of binary and ternary mixtures are 4.34%, 0.30&#xa0;MPa, 3.70%, and 0.19&#xa0;MPa, respectively. The optimal model selection depends on the specific mixture type under consideration when using improved methods specifically.</p> Graphical abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction for Critical Temperature and Critical Pressure of Mixtures by Improved Empirical Correlations

  • 期刊论文

  • 发表日期：

    2025年4月26日

查看全文
  • Bo Tang,
  • Xueqiang Dong,
  • Yanxing Zhao,
  • Maoqiong Gong

摘要

Vapor–liquid critical properties of mixtures are key parameters in the petrochemical industry and supercritical technology. Experimental measurements and theoretical calculations are the primary methods for determining the critical parameters of mixtures. However, existing empirical correlations to quickly predict the critical temperatures and pressures of mixtures are limited by critical volume data for pure substances. In this work, improved methods of Li method and Kreglewski–Li (KL) method are proposed. Improved methods do not require critical volume data for pure substances, but replace it with acentric factors, normal boiling points, or critical temperatures of pure substances that are easier to obtain and more accurate. About 9,000 critical temperature and critical pressure data points for binary and ternary mixtures were collected to compare and evaluate the Li method, KL method, and improved methods. Notably, the improved methods are only applicable to the class I and II mixtures according to the classification of Van Konynenburg and Scott. Overall, compared with the original method, both Improvement 3 (critical volumes for pure substances of Li method and KL method are replaced with critical temperatures of pure substances) and Improvement 4 (critical volumes for pure substances of Li method and KL method are replaced with normal boiling points of pure substances) greatly improve the accuracy. Meanwhile, when predicting critical temperatures and critical pressures, Improvement 3 not only reduces the input thermophysical property parameters but also improves the prediction accuracy. Among the improved methods, Improvement 4 shows the highest prediction accuracy. The average absolute relative deviation (AARD) and average absolute deviation (AAD) of Improvement 4 for predicting the critical temperatures of binary and ternary mixtures are 1.88%, 7.83 K, 1.60%, and 7.63 K, respectively. The AARD and AAD for predicting the critical temperature of the binary mixtures composed of two pure substances with both acentric factors greater than 0.0955 by Improvement 4 are 1.56% and 7.33 K. The AARD and AAD of Improvement 4 for predicting the critical pressures of binary and ternary mixtures are 4.34%, 0.30 MPa, 3.70%, and 0.19 MPa, respectively. The optimal model selection depends on the specific mixture type under consideration when using improved methods specifically.

Graphical abstract