Solution-processed polyimide memristors for high-performance and reconfigurable T-type radio-frequency routing
Radio-frequency (RF) switches are essential for reconfigurable wireless front-ends that require low-loss transmission, high isolation, and ideally zero-static-power operation. Memristive RF switches are promising because their programmed low-resistance state (LRS) and high-resistance state (HRS) can respectively provide low-loss RF ON transmission and low-parasitic RF OFF isolation. For flexible and integrated front-ends, polymer memristive switches are attractive because solution-processable and mechanically compliant dielectric layers can be formed over large areas. However, the connection between polymer transport physics, RF parasitics, mechanical compliance, and circuit-level routing remains insufficiently established. Here, we report solution-processed polyimide (PI) memristive devices based on an Au/PI/Cu architecture as transport-informed nonvolatile RF switches. Low-temperature measurements are consistent with Cu-rich metallic filament conduction in the LRS and Mott variable-range hopping in the HRS, linking the RF ON/OFF responses to distinct transport pathways. The devices exhibit stable bipolar switching for nearly 2000 cycles, projected 10-year retention at 120.4 °C, and RF responses verified up to 67 GHz with an ON resistance of 5.7 Ω and OFF capacitance of 2.7 fF. Time-domain measurements reveal nanosecond-scale RF reconfiguration, and power-handling tests show stable ON state transmission up to 31.6 dBm while the OFF state remains isolated without RF-induced self-switching. Flexible devices preserve RF functionality under static bending down to a radius of 15 mm and after 1000 bending cycles, while a monolithic T-type routing matrix demonstrates three-state signal reconfiguration with inter-port isolation below -20 dB. These results establish solution-processed PI as a multifunctional polymer platform for next-generation RF front-ends.
Authors
- Sungmoon Park (ORCID: https://orcid.org/0009-0000-7622-9213)
- Myungsoo Kim (ORCID: https://orcid.org/0000-0002-5169-6381)
- Changwoo Pyo (ORCID: https://orcid.org/0009-0008-9179-2301)
- Junyeong Jang (ORCID: https://orcid.org/0009-0001-9745-031X)
- Yoongi Cho (ORCID: https://orcid.org/0009-0009-8546-1933)
Institutions
- Ulsan National Institute of Science and Technology (KR)
Publication Details
- Journal
- Materials Horizons
- Published
- 2026-11-16
- DOI
- https://doi.org/10.1039/d6mh01250g
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Korea Institute for Advancement of Technology
- National Research Foundation of Korea
- Ministry of Science and ICT, South Korea