欢迎访问过程工程学报, 今天是

过程工程学报 ›› 2025, Vol. 25 ›› Issue (12): 1292-1299.DOI: 10.12034/j.issn.1009-606X.225088

• 研究论文 • 上一篇    下一篇

丙烯酰胺-丙烯酸共聚过程转化率的原位拉曼光谱监测方法

温佳娴1， 常诚2,3*， 顾雪萍2,3*， 冯连芳2,3， 张才亮2,3   

  1. 1. 浙江大学工程师学院，浙江 杭州 310015 2. 浙江大学化学工程与生物工程学院，化学工程与低碳技术全国重点实验室，浙江 杭州 310027 3. 浙江大学衢州研究院，浙江 衢州 324000
  • 收稿日期:2025-03-27 修回日期:2025-06-12 出版日期:2025-12-28 发布日期:2025-12-29
  • 通讯作者: 常诚 changcheng@zju.edu.cn
  • 基金资助:
    浙江省高端化学品技术创新中心

In situ Raman spectroscopic monitoring of conversion of acrylamide-acrylic acid copolymerization

Jiaxian WEN1,  Cheng CHANG2,3*,  Xueping GU2,3*,  Lianfang FENG2,3,  Cailiang ZHANG2,3   

  1. 1. College of Engineers, Zhejiang University, Hangzhou, Zhejiang 310015, China 2. State Key Laboratory of Chemical Engineering and Low-carbon Technology, College of Chemical & Biological Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China 3. Institute of Zhejiang University-Quzhou, Quzhou, Zhejiang 324000, China
  • Received:2025-03-27 Revised:2025-06-12 Online:2025-12-28 Published:2025-12-29
  • Supported by:
    Zhejiang Provincial Innovation Center of Advanced Chemicals Technology

RichHTML

PDF

摘要： 丙烯酰胺-丙烯酸自由基共聚反应具有快引发、快终止的特性。单体转化率作为聚合过程的关键参数，亟需高时效性监测手段。通过原位拉曼光谱技术可以实现在线监测，但单体转化率对光谱特征域的影响机制仍缺少研究。在引发温度为1~30℃及复合引发剂质量比(偶氮/氧化还原引发剂)为0∶100~90∶10的条件下研究单体转化率与拉曼光谱的对应关系。基于碳骨架摇摆振动、丙烯酰胺C-N键伸缩振动及单体CH2和C=C伸缩振动特征峰，建立基于峰面积的分析方法。通过解析单体转化率与峰面积的关联，构建双单体转化率机理模型。在线监测结果与核磁共振氢谱对比，采用1386~1517 cm-1和1517~1732 cm-1特征域建立的光谱模型对丙烯酰胺、丙烯酸转化率预测的R2分别为0.989和0.981，验证集平均相对误差为2.14%。该方法可用于丙烯酰胺-丙烯酸自由基共聚过程的转化率实时监测，为高黏体系的在线质量监测提供了解决方案。

Abstract: The aqueous copolymerization of acrylamide (AM) and acrylic acid (AA) is essential for synthesizing poly(AM-co-AA), an anionic polymer widely applied in enhanced oil recovery, mining, and wastewater treatment. However, rapid reaction kinetics, high viscosity, and sampling challenges hinder real-time monitoring of monomer conversion rates using conventional offline methods. This study proposes a non-destructive in situ Raman spectroscopic approach to track conversion rates dynamically. Characteristic Raman bands associated with structural changes during polymerization are identified: C-N stretching (1386~1517 cm-1) for AM consumption, C=C stretching (1517~1732 cm-1) for SA consumption, and CH2 bending (1363 cm-1) as an invariant reference for spectral normalization. Experiments were conducted under varied initiation temperatures (1~30℃) and composite initiator ratios (azo/redox initiators: 0∶100~90∶10). Real-time Raman spectra were processed to correct baseline drift and normalized using the CH2 bending peak. Key spectral regions were analyzed to quantify peak area changes, which were correlated with monomer consumption derived from offline NMR validation. A mechanistic model linking peak area changes to conversion rates was developed, with parameters regressed via multivariate least-squares fitting. The model demonstrated high accuracy, achieving average R2 values of 0.989 and 0.981 for predicting AM and SA conversions using 1386~1517 cm-1 and 1517~1732 cm-1 bands, respectively. Independent validation tests yielded an average relative error of 2.14%. This approach enables real-time, non-invasive monitoring of copolymerization kinetics, overcoming limitations of traditional destructive techniques. By integrating spectral analysis with mechanistic modeling, the method minimizes interference from overlapping peaks and environmental fluctuations. The results underscore Raman spectroscopy's potential for online quality control in industrial batch processes, ensuring consistent product performance through precise conversion rate tracking. The established framework provides a foundation for optimizing reaction conditions and enhancing production efficiency in polymer manufacturing.