P. Xu,H. Luo
Abstract:The latest field-effect transistors are entering the regime where quantum effects within the conduction channel can play a significant role because of the increasingly reduced dimensions. We investigate the effects of quantized states in conduction channels in transistors with dimensions close to those presently used. We use the standard configuration of Si-based metal-oxide-semiconductor field-effect transistors (MOSFETs), as a simplified model to provide an estimate of the effect of quantization with respect to the dimensions of the conduction channel. The study shows simulated results of drain currents for various combinations of dimensions, in which distinguishable current levels as a function of the applied gate bias can be obtained at room temperature. The same qualitative dependence on dimensions is expected to apply to the state-of-the-art transistor architectures with dimensions near this range, such as fin field-effect transistors (FinFETs) and gate-all-around field-effect transistors (GAAFETs). The results show that utilizing quantized states in the conduction channel for multinary computation has become a possibility with their present dimensions.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: as the size of field - effect transistors (FETs) continues to shrink, quantum effects begin to significantly affect device performance in the conductive channel. Specifically, the research focuses on how to utilize the quantized states in nano - scale Si - based metal - oxide - semiconductor field - effect transistors (MOSFETs) for multinary computation and achieve distinguishable current step characteristics at room temperature.
### Problem Background
1. **Influence of Quantum Effects**:
- As the transistor size decreases, especially when entering the nano - scale, quantum effects become non - negligible. These effects include the quantization of electrons or holes in the conductive channel.
2. **Limitations of Existing Technologies**:
- The current semiconductor industry mainly focuses on increasing the number of transistors per unit area and overcoming the complexity brought about by size reduction, such as short - channel effects. However, there is relatively little research on multinary computation, especially the possibility of implementing multivalued logic within a single transistor.
3. **Requirement for Multinary Computation**:
- Traditional binary computation relies on two states (0 and 1), while multinary computation can use more states to represent information, which may potentially improve computational efficiency and data density. For example, ternary computers were studied in the 1950s.
### Research Objectives
The main objective of this paper is to explore how to use the quantized states in the conductive channel for multinary computation when approaching the current state - of - the - art transistor size. Specifically:
1. **Simulation and Analysis**:
- Using the standard Si - based MOSFET structure as a simplified model, study the influence of the quantized states in the conductive channel at different sizes on the drain current.
2. **Experimental Verification**:
- Through simulation, a current curve with obvious step characteristics was obtained at room temperature, and these steps correspond to different gate - bias voltages.
3. **Application Prospects**:
- Explore whether these quantized states can implement multinary computation within the existing silicon - based technology framework without adding additional terminals or complexity.
### Key Formulas
The key formulas involved in the paper include:
- **Energy Quantization in Triangular Potential Well**:
\[
E_n=\frac{9\pi^2 n^2}{2 m^* L^2}
\]
where \(n = 1, 2, 3,\ldots\), \(m^*\) is the effective mass, and \(L\) is the conductive channel length.
- **Three - Dimensional Quantized Energy Levels**:
\[
E_{n_x, n_y, n_z}=\frac{\hbar^2\pi^2 n_x^2}{2 m^* l^2}+\frac{\hbar^2\pi^2 n_y^2}{2 m^* w^2}+\frac{9\pi^2 n_z^2}{2 m^* L^2}
\]
where \(n_x, n_y, n_z = 1, 2, 3,\ldots\), \(l\) is the channel length, \(w\) is the channel width, and \(L\) is the length along the gate direction.
- **Landauer Formula for Calculating Drain Current**:
\[
I=\frac{e}{L}\int v(E_n)[f(E_n - E_F^s)-f(E_n - E_F^d)]dE_n
\]
where \(L\) is the channel length, \(E_F^s\) and \(E_F^d\) are the quasi - Fermi levels in the source and drain regions respectively, \(v(E_n)\) is the velocity, and \(f\) is the Fermi - Dirac distribution function.
### Conclusion
This research demonstrates the feasibility of using quantized states for multinary computation at the nano - scale, and these effects are visible at room temperature. This provides a theoretical basis and technical support for the future development of more efficient computing architectures.