Landauer-QFLPS model for mixed Schottky-Ohmic contact two-dimensional transistors

Zhao-Yi Yan,Zhan Hou,Kan-Hao Xue,Tian Lu,Ruiting Zhao,Junying Xue,Fan Wu,Minghao Shao,Jianlan Yan,Anzhi Yan,Zhenze Wang,Penghui Shen,Mingyue Zhao,Xiangshui Miao,Zhaoyang Lin,Houfang Liu,He Tian,Yi Yang,Tian-Ling Ren
2023-03-20
Abstract:Two-dimensional material-based field effect transistors (2DM-FETs) are playing a revolutionary role in electronic devices. However, after years of development, no device model can match the Pao-Sah model for standard silicon-based transistors in terms of physical accuracy and computational efficiency to support large-scale integrated circuit design. One remaining critical obstacle is the contacts of 2DM-FETs. In order to self-consistently include the contact effect in the current model, it is necessary to perform self-consistent calculations, which is a fatal flaw for applications that prioritize efficiency. Here, we report that the Landauer-QFLPS model effectively overcomes the above contradiction, where QFLPS means quasi-Fermi-level phase space theory. By connecting the physical pictures of the contact and the intrinsic channel part, we have successfully derived a drain-source current formula including the contact effect. To verify the model, we prepared transistors based on two typical 2DMs, black phosphorus (BP) and molybdenum disulfide (MoS2), the former having ambipolar transport and the latter showing electron-dominant unipolar transport. The proposed new formula could describe both 2DM-FETs with Schottky or Ohmic contacts. Moreover, compared with traditional methods, the proposed model has the advantages of accuracy and efficiency, especially in describing non-monotonic drain conductance characteristics, because the contact effect is self-consistently and compactly packaged as an exponential term. More importantly, we also examined the model at the circuit level. Here, we fabricated a three-bit threshold inverter quantizer circuit based on ambipolar-BP process and experimentally demonstrated that the model can accurately predict the circuit performance. This industry-benign 2DM-FET model is supposed to be very useful for the development of 2DM-FET-based integrated circuits.
Applied Physics,Mesoscale and Nanoscale Physics
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: in two - dimensional material field - effect transistors (2DM - FETs), how to establish a device model that can accurately describe the contact effect and maintain computational efficiency at the same time. Specifically, the existing models either sacrifice physical accuracy to improve computational efficiency or are too complicated to be applied to large - scale integrated circuit design. The paper proposes a model named Landauer - QFLPS, which effectively overcomes the above contradiction by combining the Landauer formula and the quasi - Fermi level phase - space theory (QFLPS). This model can self - consistently include the contact effect and is applicable to 2DM - FETs with Schottky or ohmic contacts, and also shows advantages in describing non - monotonic drain conductance characteristics. ### Paper Background Two - dimensional material field - effect transistors (2DM - FETs) have received extensive attention because of their potential to continue Moore's Law. Their atomic thickness and the interface without dangling bonds with the gate oxide make them highly tunable in FET device applications. However, despite years of development, no device model has been able to rival the Pao - Sah model of standard silicon - based transistors in terms of physical accuracy and computational efficiency, especially in supporting large - scale integrated circuit design. One of the key obstacles is the contact problem of 2DM - FETs. In order to self - consistently include the contact effect in the current model, self - consistent calculations are required, which is a fatal flaw in efficiency - seeking applications. Conversely, in order to improve efficiency, the self - consistency of the model usually needs to be abandoned. ### Paper Objectives The objective of the paper is to develop a 2DM - FETs model that can accurately describe the contact effect and maintain computational efficiency. Specifically, the paper proposes the Landauer - QFLPS model, which successfully derives the drain - source current formula including the contact effect by connecting the physical pictures of the contact part and the intrinsic channel part. The paper verifies the effectiveness of this model through experiments and shows its application potential in circuit - level simulations. ### Main Contributions 1. **Model Innovation**: Proposed the Landauer - QFLPS model, which effectively solves the self - consistent description problem of the contact effect by combining the Landauer formula and the quasi - Fermi level phase - space theory (QFLPS). 2. **Experimental Verification**: Transistors were fabricated using two typical 2DM materials (black phosphorus BP and molybdenum disulfide MoS₂) to verify the effectiveness of the model. 3. **Circuit - level Application**: A three - bit threshold inverter quantization circuit (ATIQ) was fabricated based on the ambipolar - BP process, and the model's ability to predict circuit performance was verified in the experiment. 4. **Wide Applicability**: The model is applicable not only to 2DM - FETs with ambipolar transmission but also to 2DM - FETs with unipolar transmission. ### Model Details - **Model Construction**: The channel region is divided into three parts: the source contact region, the intrinsic channel region, and the drain contact region. By introducing the quasi - Fermi level phase - space theory (QFLPS), the model can self - consistently describe the contact effect and channel transmission. - **Key Formulas**: The current formulas for the source contact region and the drain contact region are derived. These formulas combine the Landauer formula and the QFLPS model by introducing the barrier attenuation factor (BAF). - **Parameterization Strategy**: A simple but efficient model parameterization strategy is proposed, that is, the model parameters only depend on the gate - source voltage \( V_{GS} \) and are approximated by the linearized Gaussian wavelet function. ### Experimental Verification - **BP - FET Verification**: A batch of transistors were fabricated using black phosphorus (BP), their I - V characteristics were tested, and the accuracy of the model was verified. - **ATIQ Circuit Verification**: A three - bit threshold inverter quantization circuit (ATIQ) was fabricated, and the model's ability to predict circuit performance was verified in the experiment. - **MoS₂ - FET Verification**: Transistors were fabricated using molybdenum disulfide (MoS₂) to further verify the applicability of the model in the case of unipolar transmission. ### Conclusion The Landauer - QFLPS model proposed in the paper strikes a balance between physical accuracy and computational efficiency, can self - consistently describe the contact effect, and is applicable to different types and transmission modes of 2DM - FETs. This model has shown good performance in both experiments and circuit - level simulations and is expected to be used in the large - scale integration of 2DM - FETs.