MOS Interface Physics, Process and Characterization

Shengkai Wang, Xiaolei Wang

• Publisher: CRC Press
• Year: 2021
• ISBN:
  9781000455762
• Edition: 1
• Topic: Tecnología e ingeniería
• Subtopic: Ingeniería eléctrica y telecomunicaciones

The electronic device based on Metal Oxide Semiconductor (MOS) structure is the most important component of a large-scale integrated circuit, and is therefore a fundamental building block of the information society. Indeed, high quality MOS structure is the key to achieving high performance devices and integrated circuits. Meanwhile, the control of interface physics, process and characterization methods determine the quality of MOS structure.

This book tries to answer five key questions: Why are high-performance integrated circuits bonded together so closely with MOS structure? Which physical phenomena occur in MOS structure? How do these phenomena affect the performance of MOS structure? How can we observe and quantify these phenomena scientifically? How to control the above phenomena through process? Principles are explained based on common experimental phenomena, from sensibility to rationality, via abundant experimental examples focusing on MOS structure, including specific experimental steps with a strong level of operability.

This book will be an essential reference for engineers in semiconductor related fields and academics and postgraduates within the field of microelectronics.

CHAPTER 1

Physics of Interface

DOI: 10.1201/9781003216285-2

1.1 MOS INTERFACE

The metal–oxide–semiconductor (MOS) interface refers to the heterojunction interface in the gate structure of the MOS device, as shown in Figure 1.1. The gate structure of the MOS device is usually composed of a metal/oxide dielectric/semiconductor substrate, and the oxide dielectric can be a stack of one or several insulating dielectrics. The interface appears between different kinds of materials. Therefore, the interface can be considered as the boundary between any two different materials. It should be noted that in most cases, the interface does not refer to a surface of infinitesimal thickness, but refers to the transition area between two different materials, and it is a thin layer with a certain thickness, usually about 3 Å.

FIGURE 1.1 Schematic of modern metal–oxide–semiconductor (MOS) device.

The following types of interfaces in the MOS gate structure can appear: oxide/semiconductor interface, oxide/oxide interface and metal/oxide interface. Generally speaking, the characteristics of the oxide/semiconductor interface are the most important. It seriously affects the gate control capability of the MOS devices, the mobility of carriers on the semiconductor substrate and the reliability of the MOS gate structure. In addition, after the introduction of high dielectric constant gate dielectric (such as HfO₂) into the gate structure of silicon-based MOS devices, the HfO₂/SiO₂ interface also plays an important role, involving the shift of the device threshold voltage and the reliability of the gate structure.

1.2 THE PHYSICAL NATURE OF INTERFACE STATES AND BULK DEFECTS

The concept of interface states does not belong to the category of classical physics, but belong to the category of quantum mechanics and solid-state physics. The interface state refers to the real-space distribution of the electronic wave function near the interface, i.e., the electronic wave function attenuates to both sides of the interface. The electronic energy level corresponding to the interface state is usually located inside the band gap. The study of the interface states has been going on for nearly hundred years, but it is still not fully understood. The research on the interface state of silicon semiconductor is the most complete, and the understanding is the most profound. Here we take the interface state of silicon as an example. The physical origin of the interface state of silicon is often attributed to dangling bonds, or P_b centers. This concept is actually a visual explanation given from a chemical point of view. From a physical point of view, it needs to be considered from the energy band point of view. The interface state is actually not generated out of thin air, but the energy level in the conduction band or valence band of silicon is pulled into the forbidden band, and then becomes the interface state, as shown in Figure 1.2.

FIGURE 1.2 Origin of interface trap.

Due to the interrupt of periodicity at the SiO₂/Si interface, the solutions of Schrödinger’s equation with complex wave vectors become of physical relevance for energies within band gaps, resulting in gap states at the SiO₂/Si interface. These interfacial gap states are derived from the virtual gap states of the complex band structure of the silicon semiconductor, and they may arise from intrinsic, defect, or structure induced gap states. They consist of valence- and conduction-band states. The characteristics of these gap states change across the band gap from predominately donor- to acceptor-like closer to the valence band top and the conduction band bottom, respectively. The energy at which their characteristic changes is called their branch point, or most generally, charge neutrality level (CNL). This energy level shift comes from the change of the potential energy at the interface relative to the potential energy inside the silicon. The physical sources of these changes include dangling bonds and interface structure relaxation. It should be noted here that certain chemical bonds can also lead to the interface state energy level. For example, the Ga-O bond on the surface of InGaAS has the bond energy in the InGaAS forbidden band. Although there is no dangling bond, the bond energy position deviates from the conduction. Band or valence band can still lead to interface states.

The physical nature of bulk defects originates from atomic vacancies, interstitial atoms, replacement atoms, dislocations, structural changes, etc. Any deviation from the perfect structure of the material may cause defects. Corresponding to the gate structure of MOS devices, body defects often involve oxygen vacancies. Similar to the generation process of the interface state, the body defect energy level also pulls the electron energy level from the conduction or valence band into the forbidden band.

1.3 MOS INTERFACE PASSIVATION METHODS

The method of interface passivation comes from the physical nature of interface defects. At present, the most well-researched semiconductor substrates include silicon, germanium, silicon germanium and ...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Preface
• Authors
• Introduction
• Chapter 1 ▪ Physics of Interface
• Chapter 2 ▪ MOS Processes
• Chapter 3 ▪ MOS Characterizations
• Appendix I: Physical Constants
• Appendices II–V: Useful Data for MOS Interface in Periodic Table