Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Combining automated microfluidic experimentation with machine learning for efficient polymerization design


Understanding polymerization reactions has challenges relating to the complexity of the systems, the hazards associated with the reagents, the environmental footprint of the operations and the highly nonlinear topologies of reaction spaces. In this work, we aim to present a new methodology for studying polymerization reactions using machine-learning-assisted automated microchemical reactors. A custom-designed rapidly prototyped microreactor is used in conjunction with automation and in situ infrared thermography for efficient, high-speed experimentation to map the reaction space of a zirconocene polymerization catalyst and obtain fundamental kinetic parameters. Chemical waste is decreased by two orders of magnitude and catalytic discovery is reduced from weeks to hours. Bayesian regularization backpropagation is used in conjunction with kinetic modelling to understand the reaction space and resultant technoeconomic topology. Here, we show that efficient microfluidic technology can be coupled with machine-learning algorithms to obtain high-fidelity datasets on a complex chemical reaction.

This is a preview of subscription content

Access options

Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Fig. 1: Overview of the chemistry.
Fig. 2: Reactor schematic and computational and experimental verification of performance.
Fig. 3: Process flow diagram for an automated thermographic microreactor system used in the understanding of metallocene catalysts.
Fig. 4: Flowcharts for experimental control and data handling.
Fig. 5: Experimental results for catalytic productivity and kinetic rate constants.
Fig. 6: ANN results for catalytic productivity over various concentrations and temperatures.

Data availability

The datasets generated and/or analysed during the current study are available in the Zenodo repository,[50].

Code availability

The code used for data analysis in the current study is publicly available in the Zenodo repository,[50]. The code used for experimental control is available in the Zenodo repository,[51]. This code is not publicly available due to inclusion of third party code but is available upon reasonable request from the corresponding author.


  1. Brintzinger, H. H., Fischer, D., Mülhaupt, R., Rieger, B. & Waymouth, R. M. Stereospecific olefin polymerization with chiral metallocene catalysts. Angew. Chem. Int. Ed. 34, 1143–1170 (1995).

    Google Scholar 

  2. Shamiri, A. et al. The influence of Ziegler–Natta and metallocene catalysts on polyolefin structure, properties and processing ability. Materials 7, 5069–5108 (2014).

    Google Scholar 

  3. Kaminsky, W. Highly active metallocene catalysts for olefin polymerization. J. Chem. Soc. Dalton Trans. 1998, 1413–1418 (1998).

    Google Scholar 

  4. Sinclair, K. B. Future trends in polyolefin materials. Macromol. Symp. 173, 237–261 (2001).

    Google Scholar 

  5. Plastics and Polymers Global Market Briefing 2018 (The Business Research Company, 2018).

  6. Sumerin, V. & Thorman, J. Ziegler–Natta catalyst and preparation thereof. US patent 10,118,977 (2018).

  7. Kesti, M. R., Coates, G. W. & Waymouth, R. M. Homogeneous Ziegler–Natta polymerization of functionalized monomers catalyzed by cationic group IV metallocenes. J. Am. Chem. Soc. 114, 9679–9680 (1992).

    Google Scholar 

  8. Society of the Plastics Industry, US Department of Energy Improving Energy Efficiency at US Plastics Manufacturing Plants 40 (US DOE, 2005).

  9. Worrell, E., Bernstein, L., Roy, J., Price, L. & Harnisch, J. Industrial energy efficiency and climate change mitigation. Energy Efficiency 2, 109–123 (2009).

    Google Scholar 

  10. Khripko, D., Schlüter, B. A., Rommel, B., Rosano, M. & Hesselbach, J. Energy demand and efficiency measures in polymer processing: comparison between temperate and Mediterranean operating plants. Int. J. Energy Environ. Eng. 7, 225–233 (2016).

    Google Scholar 

  11. Sinn, H. & Kaminsky, W. Ziegler–Natta catalysis. Adv. Organomet. Chem. 18, 99–149 (1980).

    Google Scholar 

  12. Chien, J. C. W. & Wang, B.-P. Metallocene–methylaluminoxane catalysts for olefin polymerizations. IV. Active site determinations and limitation of the 14CO radiolabeling technique. J. Polym. Sci. A 27, 1539–1557 (1989).

    Google Scholar 

  13. Kaminsky, W. (ed.) Metalorganic Catalysts for Synthesis and Polymerization (Springer, 2011).

  14. Rieger, B., Jany, G., Steimann, M. & Fawzi, R. Synthesis of ethylene bridged biscyclopentadiene ligand precursor compounds and some of their ansa-zirconocene derivatives chiral epoxides: a synthetic strategy of high variability. Z. Naturforsch. B Chem. Sci. 49, 451–458 (1994).

    Google Scholar 

  15. Kolthammer, B. W. S., Mangold, D. J. & Gifford, D. R. Polymerization kinetics of octene-1 catalyzed by metallocene methylaluminoxane investigated with attenuated total reflectance Fourier transform infrared (ATR-FT-IR) spectroscopy. J. Polym. Sci. A 30, 1017–1026 (1992).

    Google Scholar 

  16. Charpentier, P. A., Zhu, S., Hamielec, A. E. & Brook, M. A. Continuous solution polymerization of ethylene using metallocene catalyst system, zirconocene dichloride/methylaluminoxane/trimethylaluminum. Ind. Eng. Chem. Res. 36, 5074–5082 (1997).

    Google Scholar 

  17. D’Agnillo, L., Soares, J. B. P. & Penlidis, A. Effect of operating conditions on the molecular weight distribution of polyethylene synthesized by soluble metallocene/methylaluminoxane catalysts. Macromol. Chem. Phys. 199, 955–962 (1998).

    Google Scholar 

  18. Martínez, S., Cruz, V. L., Ramos, J. & Martínez-Salazar, J. Polymerization activity prediction of zirconocene single-site catalysts using 3D quantitative structure-activity relationship modeling. Organometallics 31, 1673–1679 (2012).

    Google Scholar 

  19. Moscato, B. M., Zhu, B. & Landis, C. R. GPC and ESI-MS analysis of labeled poly(1-hexene): rapid determination of initiated site counts during catalytic alkene polymerization reactions. J. Am. Chem. Soc. 132, 14352–14354 (2010).

    Google Scholar 

  20. Santos, L. S. & Metzger, J. O. Study of homogeneously catalyzed Ziegler–Natta polymerization of ethene by ESI-MS. Angew. Chem. Int. Ed. 45, 977–981 (2006).

    Google Scholar 

  21. Silveira, F., De Sá, D. S., Da Rocha, Z. N. & Dos Santos, J. H. Z. Metallocene combinations in ethylene polymerization: a cyclic and differential pulse voltammetry study. Macromol. React. Eng. 2, 253–264 (2008).

    Google Scholar 

  22. Moscato, B. M., Zhu, B. & Landis, C. R. Mechanistic investigations into the behavior of a labeled zirconocene polymerization catalyst. Organometallics 31, 2097–2107 (2012).

    Google Scholar 

  23. Gonzalez-Ruiz, R. A., Quevedo-Sanchez, B., Laurence, R. L., Henson, M. A. & Bryan Coughlin, E. Kinetic modeling of slurry propylene polymerization using rac-ET(Ind)2ZrCl2/MAO. AIChE J. 52, 1824–1835 (2006).

    Google Scholar 

  24. Christianson, M. D., Tan, E. H. P. & Landis, C. R. Stopped-flow NMR: determining the kinetics of [rac-(C2H4(1-indenyl)2)ZrMe][MeB(C6F5)3]-catalyze dpolymerization of 1-hexene by direct observation. J. Am. Chem. Soc. 132, 11461–11463 (2010).

    Google Scholar 

  25. Rubens, M., Vrijsen, J. H., Laun, J. & Junkers, T. Precise polymer synthesis by autonomous self-optimizing flow reactors. Angew. Chem. Int. Ed. 58, 3183–3187 (2019).

    Google Scholar 

  26. Knox, S. T. & Warren, N. J. Enabling technologies in polymer synthesis: accessing a new design space for advanced polymer materials. React. Chem. Eng. 5, 405–423 (2020).

    Google Scholar 

  27. Kaminsky, W. Zirconocene catalysts for olefin polymerization. Catal. Today 20, 257–271 (1994).

    Google Scholar 

  28. Jensen, K. F. Flow chemistry—microreaction technology comes of age. AIChE J. 63, 858–869 (2017).

    Google Scholar 

  29. Heider, P. L. et al. Development of a multi-step synthesis and workup sequence for an integrated, continuous manufacturing process of a pharmaceutical. Org. Process Res. Dev. 18, 402–409 (2014).

    Google Scholar 

  30. Hartman, R. L., Naber, J. R., Buchwald, S. L. & Jensen, K. F. Multistep microchemical synthesis enabled by microfluidic distillation. Angew. Chem. Int. Ed. 49, 899–903 (2010).

    Google Scholar 

  31. Kim, J. O. et al. A monolithic and flexible fluoropolymer film microreactor for organic synthesis applications. Lab Chip 14, 4270–4276 (2014).

    Google Scholar 

  32. Hu, C., Morris, J. E. & Hartman, R. L. Microfluidic investigation of the deposition of asphaltenes in porous media. Lab Chip 14, 2014–2022 (2014).

    Google Scholar 

  33. Rizkin, B. A., Popovic, F. G. & Hartman, R. L. Spectroscopic microreactors for heterogeneous catalysis. J. Vac. Sci. Technol. A 37, 050801 (2019).

    Google Scholar 

  34. Gromski, P. S., Granda, J. M. & Cronin, L. Universal chemical synthesis and discovery with ‘The Chemputer’. Trends Chem. 1–9 (2019);

  35. Coley, C. W. et al. A robotic platform for flow synthesis of organic compounds informed by AI planning. Science 365, eaax1566 (2019).

    Google Scholar 

  36. Theurkauff, G., Bondon, A., Dorcet, V., Carpentier, J. F. & Kirillov, E. Heterobi- and -trimetallic ion pairs of zirconocene-based isoselective olefin polymerization catalysts with AlMe3. Angew. Chem. Int. Ed. 54, 6343–6346 (2015).

    Google Scholar 

  37. Song, F., Cannon, R. D. & Bochmann, M. Zirconocene-catalyzed propene polymerization: a quenched-flow kinetic study. J. Am. Chem. Soc. 125, 7641–7653 (2003).

    Google Scholar 

  38. Christopher, J. N., Diamond, G. M., Jordan, R. F. & Petersen, J. L. Synthesis, structure and reactivity of rac-Me2Si(indenyl)2Zr(NMe2)2. Organometallics 15, 4038–4044 (1996).

    Google Scholar 

  39. Lenton, T. N. et al. Formation of trivalent zirconocene complexes from ansa-zirconocene-based olefin-polymerization precatalysts: an EPR- and NMR-spectroscopic study. J. Am. Chem. Soc. 135, 10710–10719 (2013).

    Google Scholar 

  40. Ning, Y., Cooney, M. J. & Chen, E. Y. X. Polymerization of MMA by oscillating zirconocene catalysts, diastereomeric zirconocene mixtures, and diastereospecific metallocene pairs. J. Organomet. Chem. 690, 6263–6270 (2005).

    Google Scholar 

  41. Bochmann, M., Cannon, R. D. & Song, F. Kinetic and mechanism of alkene polymerization. Kinet. Catal. 47, 160–169 (2006).

    Google Scholar 

  42. Song, F., Hannant, M. D., Cannon, R. D. & Bochmann, M. Zirconocene-catalysed propene polymerisation: kinetics, mechanism and the role of the anion. Macromol. Symp. 213, 173–185 (2004).

    Google Scholar 

  43. Su, Y., Song, Y. & Xiang, L. Continuous-flow microreactors for polymer synthesis: engineering principles and applications. Top. Curr. Chem. 376, 44 (2018).

    Google Scholar 

  44. Zhang, J. S., Zhang, C. Y., Liu, G. T. & Luo, G. S. Measuring enthalpy of fast exothermal reaction with infrared thermography in a microreactor. Chem. Eng. J. 295, 384–390 (2016).

    Google Scholar 

  45. Hany, C., Lebrun, H., Pradere, C., Toutain, J. & Batsale, J. C. Thermal analysis of chemical reaction with a continuous microfluidic calorimeter. Chem. Eng. J. 160, 814–822 (2010).

    Google Scholar 

  46. Pradere, C., Joanicot, M., Batsale, J.-C., Toutain, J. & Gourdon, C. Processing of temperature field in chemical microreactors with infrared thermography. Quant. Infrared Thermogr. J. 3, 117–135 (2007).

    Google Scholar 

  47. Terms, F. Heat of polymerization. Polym. Rev. 3, 339–356 (1969).

    Google Scholar 

  48. Rizkin, B. A. & Hartman, R. L. Supervised machine learning for prediction of zirconocene-catalyzed α-olefin polymerization. Chem. Eng. Sci. 210, 115224 (2019).

    Google Scholar 

  49. Iooss, B. & Lemaître, P. A review on global sensitivity analysis methods. Uncertain. Manag. Simulation-Optimization Complex Syst. Algorithms Appl. 59, 101–122 (2015).

    Google Scholar 

  50. Rizkin, B. A., Shkolnik, A. S., Ferraro N. J. & Hartman R. L. Combining automated microfluidic experimentation with machine learning for efficient polymerization design. Zenodo (2020).

  51. Rizkin, B. A., Shkolnik, A. S., Ferraro N. J. & Hartman R. L. Combining automated microfluidic experimentation with machine learning for efficient polymerization design, control code. Zenodo (2020).

Download references


This material is based on work supported by the National Science Foundation under grant no. CBET-1701393. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Author information

Authors and Affiliations



Conceptualization was provided by R.L.H., data curation by B.A.R., formal analysis by B.A.R. and A.S.S., funding acquisition by R.L.H., investigation by B.A.R. and A.S.S., methodology by B.A.R. and N.J.F., project administration by B.A.R. and R.L.H., resources by R.L.H., software by B.A.R., A.S.S. and N.J.F., supervision by R.L.H., validation by B.A.R. and A.S.S., visualization by B.A.R., A.S.S. and R.L.H., writing of the original draft by B.A.R. and A.S.S. and review and editing by R.L.H.

Corresponding author

Correspondence to Ryan L. Hartman.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Supplementary methods, Figs. 1–5, Tables 1 and 2, discussion and analysis.

Reporting Summary

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Rizkin, B.A., Shkolnik, A.S., Ferraro, N.J. et al. Combining automated microfluidic experimentation with machine learning for efficient polymerization design. Nat Mach Intell 2, 200–209 (2020).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing