Twisted and coiled polymer actuators (TCPAs) are nylon 'artificial muscle' fibers whose contractile actuation is now predictable by a two-phase microstructure model coupled to Kirchhoff rod theory (Wang et al., 2024) — but only for an idealized perfect, constant-helix-angle helix in quasistatic, rate-independent actuation. The open problem is to remove the model's four acknowledged limitations, presented below as three items because item (iii) absorbs the last two: (i) predict the dynamics of coil formation during manufacturing (localized twist-buckling instability, one-turn-at-a-time snap-up with self-contact, and the load–twist–geometry relation), (ii) predict the end torque required to hold the actuator during passive and active actuation, including its non-monotonic torque–displacement behavior, and (iii) incorporate viscoelastic rate effects, tie-molecule constraints, and swelling, including swelling-induced actuation that the centerline rod reduction cannot represent. Later literature (2024–2026) has delivered adjacent pieces — a 3D finite-element actuation model with fabrication-process physics, a thermo-viscoelastic-viscoplastic constitutive model for precursor fibers, and dynamic modeling of second-order coiled actuators — but each assumes away part of the idealization set, and the coiling-dynamics and holding-torque questions remain unaddressed.

Background

Twisted and coiled polymer actuators (TCPAs) are artificial-muscle fibers made from inexpensive polymer fishing line: the filament is twisted under tension until it buckles into a tight helix, then annealed so the coil shape is locked in. When the finished coil is heated (for example by Joule heating of a wrapped heating wire), it contracts by up to tens of percent while exerting large forces, giving muscle-like specific work (Haines et al., Science 2014). The physical mechanism is subtle: twisting and annealing a semicrystalline nylon fiber stores elastic energy and locks in residual stresses tied to its crystalline/amorphous microstructure, and heating changes this stored energy so the fiber tends to untwist, which geometrically converts into axial contraction of the helix. Understanding and predicting this mechanism matters for robotics, prosthetics, and wearable devices, where TCPAs are a leading low-cost actuator platform. Wang et al. (Int. J. Mech. Sci. 280:109440, 2024; arXiv:2404.00802) built the first analytically predictive model combining two ingredients: (a) a two-phase thermomechanical microstructure model, treating nylon as crystalline and amorphous phases that store and exchange strain energy in the spirit of the Takayanagi models (classical micromechanics schemes representing a semicrystalline polymer as coupled crystalline and amorphous blocks) used for two-way shape-memory polymers, calibrated by torsional actuation tests of twisted-and-annealed fibers and dynamic mechanical analysis (DMA, the oscillatory test that measures viscoelastic moduli); and (b) Kirchhoff rod theory — the classical theory of slender elastic rods, in which the rod is reduced to a centerline carrying bending and twisting energy — augmented with an additional term coupling bending and twisting. Their model reproduces the equilibrium coil geometry after fabrication, the axial stiffness in active (heated) and passive states, and the contraction under load, and it quantifies how the level of stored energy sets actuation performance. In its own limitations section the authors state precisely what the model cannot do, and this defines the present problem. First, it assumes a perfect helix of constant helix angle, so it cannot describe how the coil forms during manufacturing, where the twisted fiber undergoes a localized secondary instability in which single helical turns snap into self-contact one at a time as twist is inserted; the related question of how manufacturing load, twist density, and final coil geometry relate through this buckling instability has been studied for passive hyperelastic filaments (Liu, Liu, He and Liu, J. Mech. Phys. Solids 182:105478, 2024) but not for annealed nylon, whose microstructure evolves during annealing. Second, it predicts only tensile force, not torque: holding a helical rod at fixed geometry requires a combined force and torque (a terminal wrench), and for twisted helices the holding torque is known to vary non-monotonically with stretch (Đuričković, Goriely and Maddocks, Phys. Rev. Lett. 111:108103, 2013); no setup has measured this torque on a TCPA during contractile actuation. Third, it treats actuation as a quasistatic, rate-independent process, excluding viscoelasticity, the role of tie molecules (polymer chains that bridge crystalline lamellae and govern strength and recovery in semicrystalline fibers), and swelling, even though different heating rates give different responses and solvent uptake can itself drive actuation. Fourth, because the Kirchhoff reduction eliminates the fiber cross-section, keeping only the centerline, swelling-induced actuation cannot be represented directly; the authors suggest that thermal expansion or swelling could be reintroduced as an additional strain term, or the model extended to Cosserat rod theory (a rod theory retaining more of the cross-sectional mechanics) within a finite-element framework. The authors separately note (their Section 8.3) that next-generation actuator materials such as shape-memory polymers and liquid crystal elastomers still face a separable materials-engineering challenge — spun fibers must simultaneously satisfy a two-network microstructure, sufficient toughness for twist insertion, and spinnability — which is recorded here as context rather than as part of the modeling problem.

Problem Statement

The objective is to complete the predictive mechanics of nylon twisted and coiled polymer actuators by removing the four idealizations acknowledged by the two-phase microstructure plus Kirchhoff rod model of Wang et al. (2024) (the last two of which are addressed together in item (3) below), while retaining agreement with the equilibrium, stiffness, and contraction experiments that model already matches. Specifically, develop a predictive model (analytic, numerical, or both) that: (1) Manufacturing coiling — predicts the coil-formation process during fabrication: the onset of the localized secondary (twist-buckling) instability of the twisted nylon fiber, the sequential snap-up of single helical turns into self-contact as twist is inserted, and the resulting relation between manufacturing load, inserted twist density, and final coil geometry, accounting for the microstructure evolution that occurs during annealing; (2) Actuation torque — predicts the terminal wrench, in particular the end torque required to hold the TCPA in its helical configuration during passive stretching and during active (heated, contractile) actuation, including its non-monotonic dependence on displacement, and states the falsifiable torque–displacement predictions to be checked against dedicated measurements on a purpose-built setup once such measurements exist; (3) Rate- and solvent-dependent response — incorporates the viscoelastic (and, where relevant, viscoplastic) behavior of the precursor fiber and the role of tie molecules so that the different responses observed at different heating rates are captured rather than assumed quasistatic, and represents swelling (solvent uptake) so that swelling-induced actuation can be predicted despite the centerline reduction of rod theory. The source lists the corresponding limitations severally as directions for further work; each remains open, and together they define the research program of a predictive mechanics of the nylon TCPA from fabrication through rate-dependent actuation. The separable materials-engineering challenge of spinning new high-strength actuator fibers (shape-memory polymers, liquid crystal elastomers) satisfying two-network, toughness, and spinnability criteria is explicitly not part of this problem.

References

  1. Qiong Wang, Anan Ghrayeb, SeongHyeon Kim, Liuyang Cheng, Sameh Tawfick (2024), 'The mechanics and physics of twisted and coiled polymer actuators', International Journal of Mechanical Sciences 280:109440. DOI: 10.1016/j.ijmecsci.2024.109440; arXiv:2404.00802, https://arxiv.org/abs/2404.00802
  2. Carter S. Haines, Márcio D. Lima, Na Li, Geoffrey M. Spinks, Javad Foroughi, John D. W. Madden, Shi Hyeong Kim, Shaoli Fang, Mônica Jung de Andrade, Fatma Göktepe, Özer Göktepe, Seyed M. Mirvakili, Sina Naficy, Xavier Lepró, Jiyoung Oh, Mikhail E. Kozlov, Seon Jeong Kim, Xiuru Xu, Benjamin J. Swedlove, Gordon G. Wallace, Ray H. Baughman (2014), 'Artificial muscles from fishing line and sewing thread', Science 343(6173):868-872. DOI: 10.1126/science.1246906, https://doi.org/10.1126/science.1246906
  3. Bojan Đuričković, Alain Goriely, John H. Maddocks (2013), 'Twist and Stretch of Helices Explained via the Kirchhoff-Love Rod Model of Elastic Filaments', Physical Review Letters 111(10):108103. DOI: 10.1103/PhysRevLett.111.108103, https://doi.org/10.1103/PhysRevLett.111.108103
  4. Lei Liu, Hao Liu, Yuming He, Dabiao Liu (2024), 'Mechanics and topology of twisted hyperelastic filaments under prescribed elongations: Experiment, theory, and simulation', Journal of the Mechanics and Physics of Solids 182:105478. DOI: 10.1016/j.jmps.2023.105478, https://doi.org/10.1016/j.jmps.2023.105478
  5. Gurmeet Singh, Qiong Wang, Samuel Tsai, Sameh Tawfick, Umesh Gandhi, Veera Sundararaghavan (2026), 'Actuation mechanisms in twisted and coiled polymer actuators using finite element model', Materials Today Communications 54:115653. DOI: 10.1016/j.mtcomm.2026.115653; arXiv:2501.02710, https://arxiv.org/abs/2501.02710
  6. Juanjuan Guo, Zhiwen Gao, Jizeng Wang (2026), 'Thermo-viscoelastic-viscoplastic model for precursor fibers and its application to TCPAs', International Journal of Mechanical Sciences 313:111316. DOI: 10.1016/j.ijmecsci.2026.111316, https://doi.org/10.1016/j.ijmecsci.2026.111316
  7. Boyi Xu, Xiaojie Wang, Yitong Zhou (2026), 'Dynamic thermomechanical modeling of second order twisted and coiled actuators', European Journal of Mechanics - A/Solids 119:106171. DOI: 10.1016/j.euromechsol.2026.106171, https://doi.org/10.1016/j.euromechsol.2026.106171
  8. Nadja Schenk, Lukas Benecke, Chokri Cherif, Johannes Mersch (2026), 'Review of Thermally Activated Twisted Coiled Polymer Actuators: A Materials Science Perspective', Advanced Functional Materials 36:e76552. DOI: 10.1002/adfm.76552, https://doi.org/10.1002/adfm.76552
  9. Sean J. D. Lugger, Tom A. P. Engels, Ruth Cardinaels, Tom Bus, Dirk J. Mulder, Albert P. H. J. Schenning (2023), 'Melt-Extruded Thermoplastic Liquid Crystal Elastomer Rotating Fiber Actuators', Advanced Functional Materials 33:2306853. DOI: 10.1002/adfm.202306853, https://doi.org/10.1002/adfm.202306853
  10. Melvin Colorado Escobar, Timothy J. White (2024), 'Fast and Slow-Twitch Actuation via Twisted Liquid Crystal Elastomer Fibers', Advanced Materials 36:2401140. DOI: 10.1002/adma.202401140, https://doi.org/10.1002/adma.202401140
  11. Ali Abbas, Jianguo Zhao (2017), 'A physics based model for twisted and coiled actuator', 2017 IEEE International Conference on Robotics and Automation (ICRA), pp. 6121-6126. DOI: 10.1109/ICRA.2017.7989726, https://doi.org/10.1109/ICRA.2017.7989726