(A) study on the charge trap distribution characteristics in nitride and the retention characteristics of TaN-Al2O3-SiN-SiO2-Si memory structures

다국어 초록 (Multilingual Abstract)

The charge trap distribution characteristics in nitride and the retention characteristics of TaN-Al2O3-SiN-SiO2-Si (TANOS) are investigated to understand the device’s retention characteristics as the amount of threshold voltage shift by programming operation and charge trap distribution inside nitride vary. The model for charge trap distribution is composed of exponential distribution near conduction band and two Gaussian distribution in shallow and deep energy levels. The validity of this model is confirmed by comparing simulation results from simulator we developed by numerically solving equations which deal with electron’s state in nitride layer and experimental results for the device which is same as that for simulation. The first simulation for change of threshold voltage shift by programming operation reveals that the characteristic is deteriorated as the shift amount is increased because it makes more electrons located near conduction band. The other simulation for change of trap distribution inside nitride reveals that the more charge trapping dots are located near and above intrinsic Fermi-level between two Gaussian distributions, the more retention characteristics are improved. Especially, the change of trap energy depths of both Gaussian distributions affect very much on retention characteristics, so controlling over them is very important. These results can help anticipate the device’s retention characteristics for various programming amount and control optimum amount of threshold voltage shift for multi-level operation. Also, we can anticipate the device’s retention characteristics as charge trap distribution inside nitride varies, and which types of distribution change more critically affect on the characteristics.

목차 (Table of Contents)

• Contents i
• List of Figures iii
• List of Tables v
• Abstract vi
• 1. Introduction 1
• 2. Flash memory 3
• 2.1. History of flash memory 3
• 2.2. Types of flash memory 6
• 2.2.1. Floating gate type 6
• 2.2.2. Charge trap type - SONOS 8
• 2.2.3. Charge trap type - TANOS 10
• 2.3. Charge injection mechanisms 12
• 2.3.1. Channel hot-electron injection (CHEI) 12
• 2.3.2. Fowler-Nordheim (F-N) tunneling 17
• 2.3.3. Band-to-Band Hot-Hole Injection (BTB HHI) 22
• 2.4. Flash memory cell array 23
• 2.4.1. NOR architecture 23
• 2.4.2. NAND architecture 25
• 2.5. Multi-bit and multi-level cell 28
• 2.5.1. Multi-bit cell 28
• 2.5.2. Multi-level cell 29
• 2.6. Current issues on flash memory 30
• 2.6.1. Threshold voltage distribution 30
• 2.6.2. Data retention 31
• 3. Simulation model 33
• 3.1. Leakage paths of electrons trapped in nitride in TANOS 33
• 3.2. Numerical model for leakage 35
• 3.3. Simulation method 36
• 3.4. Nitride trap distribution model 37
• 3.5. Details for targeted device 39
• 4. Simulation results and discussion 40
• 4.1. Comparison with experimental results 40
• 4.1.1. Results from device using Si-rich silicon nitride 40
• 4.1.2. Results from device using stoichiometric silicon nitride 43
• 4.2. Dependency of retention characteristics on Vth difference between before and after being programmed 45
• 4.3. Dependency of retention characteristics on change of nitride trap distribution 47
• 4.3.1. Change of trap distributions in G1 47
• 4.3.2. Change of trap distributions in G2 51
• 5. Summary and conclusion 54
• References 56
• 국문요지 63
• Acknowledgment 64
• Declaration of ethical conduct in research 65
• 연구 윤리 서약서 66