Flexible thermoelectric (TE) cooling has emerged as a promising solution for the thermal management of high-power-density flexible electronics. However, the cooling efficiency of current devices is severely restricted by the inferior electrical transport properties of n -type flexible films compared to their p -type counterparts. To address this challenge, this study introduces a magnetic nanocomposite strategy to engineer n -type Bi 2 Te 3 -based flexible films. A series of n -type x Fe/Bi 2 Te 2.7 Se 0.3 /Epoxy flexible thermoelectric-magnetic nanocomposite (TMN) films were designed and fabricated via screen printing and hot-press curing. The introduced Fe-based nanoparticles (Fe/Fe 3 O 4 composite nanoparticles) act as a magnetic secondary phase, establishing nanoscale micro-magnetic fields. These fields induce spin-dependent scattering and weak localization effects, leading to a significant negative magnetoresistance. This process enhances the carrier scattering parameter, which is crucial for maintaining a high Seebeck coefficient. Concurrently, the electrical conductivity is markedly improved through a synergistic mechanism: interfacial charge transfer and the release of trapped electrons via in-situ oxidation of Fe nanoparticles increase the carrier concentration, while the Fe-based nanoparticles also promote strong (000 l ) orientation of BTS grains, preserving high carrier mobility. Consequently, the optimal film ( x = 0.3%) exhibited a significantly enhanced electrical conductivity of 5.5×10 4 S·m -1 and a power factor of 1.54 mW·K -2 ·m -1 at 300 K, increased 42% then that of the BTS matrix. The maximum zT reaches 0.93 at 390 K for N03, increased by 2.4 times compared as that of BTS matrix. The maximum cooling temperature difference of the single-leg device reaches 2.82 K, which is the maximum value reported for the single-leg devices. This work elucidates the critical role of TMN coupling in n -type matrices and provides a viable pathway for developing high-performance flexible TE materials for efficient solid-state cooling.
Ke et al. (2026) studied this question.