Synaptic Characteristics Dependent on Programming-Pulse Properties of Low-Power Thin-Film Transistors with a Hf-ZnO Channel Layer
We present a study on the synaptic characteristics dependent on programming-pulse properties (i.e., a programming-pulse height (PPH) and the number of programming pulses) of a low-power thin-film transistor (Syn-TFT) with a Hf-ZnO channel layer. For this, the static and pulsed characteristics of the fabricated Syn-TFTs are experimentally monitored. When positive programming pulses are applied to the gate electrode, charges (e.g., electrons) within the channel layer are trapped toward the disordered gate-oxide stack (i.e., Al2O3/HfOx), which can lead to the increase of the threshold voltage, so the output current of the Syn-TFT is depressed. Here, for a higher PPH (i.e., the intensity of the synaptic stimulation), it is found that the output current rapidly arrives at a full depression as its minimum value for applying a small number of positive programming pulses (i.e., the number of synaptic stimulation times), thus a faster programming speed, resulting in a shorter retention time as a measure of the retention characteristics. This is because a higher PPH contributes to the increased amount of trapped electrons in the gate insulator. On the other hand, the programming speed for a lower PPH is found to be slower, applying the more number of synaptic stimulation times to approach the full depression. However, it is also observed that the effective retention time is longer due to the increase of the trap activation energy related to the deep trap state, which is proportional to the number of programming pulses, increasing the accumulating effects of presynaptic stimulations.
Authors
- Sungsik Lee (ORCID: https://orcid.org/0000-0002-5523-8476)
- Danyoung Cha (ORCID: https://orcid.org/0000-0002-4713-7579)
- Byun Sung-Hui (ORCID: https://orcid.org/0009-0007-3601-4788)
- Jeongseok Pi
Institutions
- Pusan National University (KR)
Publication Details
- Journal
- ACS Applied Electronic Materials
- Published
- 2024-12-28
- DOI
- https://doi.org/10.1021/acsaelm.4c01884
- Citations
- 7
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 1.46
Funders
- Pusan National University