Metal Oxides for Non-volatile Memory: Materials, Technology and Applications

Front Cover
Metal Oxides for Non-Volatile Memory: Materials, Technology and Applications
Copyright
Contents
Contributors
Series editor biography
Preface to the series
Chapter 1: Introduction to non-volatile memory
1.1. Introduction and history
1.1.1. Outline of this work
1.2. Flash non-volatile memory
1.2.1. Programming- and erase-mechanism
1.2.2. NOR- and NAND-Flash
1.2.3. Performance and scaling issues
1.3. Novel concepts for non-volatile memories
1.3.1. Resistive switches
1.3.1.1. Memristive devices
1.3.1.2. Nanoionic effects
1.3.1.3. Switching kinetics and energy consumption
1.3.2. Magnetoresistive random access memories
1.3.2.1. Giant magnetoresistance
1.3.2.2. Tunnel magnetoresistance
1.3.2.3. Spin-transfer torque MRAM (STT-MRAM)
Acknowledgements
References
Chapter 2: Resistive switching in metal-oxide memristive materials and devices
2.1. Mechanisms of resistive switching in metal-oxide memristive materials and devices
2.1.1. Classification of resistive switching
2.1.2. Operation mechanisms of memristive devices based on TMO and SiOx
2.2. Local analysis of resistive switching of anionic type
2.2.1. Conductive atomic force microscopy
2.2.2. Investigations of individual filaments by CAFM
2.2.3. Investigations of resistive switching in YSZ films by CAFM
2.2.3.1. Sample preparation
2.2.3.2. CAFM measurements technique
2.2.3.3. Effect of annealing on the CAFM images of YSZ films
2.2.3.4. Imaging of individual filaments by CAFM
2.2.3.5. Quantum size effects in the electron transport via individual filaments
2.3. Multiscale simulation of resistive switching in metal-oxide memristive devices
2.3.1. Phenomenological approach
2.3.2. Atomistic approach
2.4. Conclusions
Acknowledgments
References
Chapter 3: Charge trapping NVMs with metal oxides in the memory stack
3.1. Introduction
3.2. History of charge trap memory devices
3.3. SONOS memory devices
3.3.1. Traps in the CT NV memory stack
3.3.2. Program and erase of CT memory device
3.4. CT memory cell reliability
3.4.1. SONOS endurance
3.4.2. Data retention
3.5. New materials for charge trap memory stack-Metal oxides
3.5.1. Why metal oxides in CT memory stack?
3.5.2. Charge trap memory stack with high-K metal oxides
3.5.2.1. High-K dielectric for blocking layer
3.5.2.2. High-K dielectric for tunneling layer
3.5.2.3. High-K dielectric for trap layer
3.5.3. TANOS with high-K metal oxide dielectric
3.5.3.1. Enhancement of TANOS stacks
3.5.3.2. Band gap engineered SONOS (BE-SONOS) with high-K dielectric
3.5.4. Metal oxides in FINFET based CT memory device
References
Chapter 4: Technology and neuromorphic functionality of magnetron-sputtered memristive devices
4.1. Features of magnetron sputtering
4.2. Performances and reproducibility of memristive devices
4.3. Functionality of memristors as elements for neuromorphic systems
4.4. Conclusions
Acknowledgments
References
Chapter 5: Metalorganic chemical vapor deposition of aluminum oxides: A paradigm on the processstructure-properties relati ...
5.1. Introduction
5.2. Process kinetic modeling and simulation of the MOCVD of metal oxides: The case of Al2O3 films
5.2.1. Model characteristics
5.2.2. Bibliographic analysis on the chemical mechanisms and kinetic laws
5.2.3. Kinetic model development and validation
5.2.4. Optimization of an original reactor by process simulation
5.3. Local coordination affects properties: The case of amorphous Al2O3 barrier coatings
5.3.1. Microstructure of amorphous alumina films
5.3.2. Application examples of barrier function of amorphous alumina films
5.4. Concluding remarks
Acknowledgements
References
Chapter 6: MOx materials by ALD method
6.1. Introduction
6.2. ALD fundamentals
6.2.1. Saturating self-limiting reactions
6.2.2. Deposition temperature
6.2.3. Precursors
6.3. ALD of oxides for memory devices
6.3.1. Atomic layer deposition of Al2O3
6.3.2. Atomic layer deposition of HfO2 and ZrO2
6.3.2.1. Chloride precursor-based ALD
6.3.2.2. Alkylamide precursor-based ALD
6.3.2.3. Cyclopentadienyl precursor-based ALD
6.3.2.4. Application of Hf-based oxides in memory storage devices
6.3.3. Atomic layer deposition of TiO2
6.3.4. Atomic layer deposition of Ta2O5
6.3.5. Atomic layer deposition of NiO
6.3.6. Atomic layer deposition of SiO2
6.4. Conclusions
References
Chapter 7: Nano-composite MOx materials for NVMs
7.1. Introduction
7.2. Experimental
7.2.1. Fabrication routes
7.2.1.1. Synthesis of nanocrystals embedded in metal oxide dielectrics by magnetron-sputtering
7.2.1.2. Synthesis of nanocrystals embedded in metal oxide dielectrics by ULE-IBS
7.2.2. Characterization techniques
7.2.2.1. Transmission electron microscopy
7.2.2.2. Atom probe tomography
7.2.2.3. XRD
7.2.2.4. FTIR
7.2.2.5. Electrical characterization
7.2.3. Synthesis of nanocrystals embedded in metal oxides by radio frequency magnetron sputtering
7.2.3.1. Synthesis of Ge-NCs in HfO2 host
Properties of HfGeOx layers versus annealing treatment
Evolution of trilayer structures with annealing treatment
Electrical properties of the structures
7.2.3.2. Synthesis of Si-NCs in HfSiOx films by magnetron sputtering
7.2.4. Ultra-low energy Si implantation into HfO2-based layers
7.2.5. ULE-II synthesis of Si- and Ge-NCs memories using SiN/HfO2/SiO2 stacks
7.2.6. ULE-II synthesis of Ge-NCs memories using Al2O3 as gate oxide
7.3. Conclusion
Acknowledgments
References
Chapter 8: MOx in ferroelectric memories
8.1. Introduction
8.2. Ferroelectricity-A material property
8.3. Negative capacitance in ferroelectrics
8.4. Ferroelectricity in hafnium oxide
8.4.1. Switching kinetics at nanoscale
8.4.2. Accumulative switching
8.5. Ferroelectric memories
8.5.1. 1T1C FeRAM implementation
8.5.2. Planar 1T FeFET implementation
8.5.3. 3D FeFET implementation
8.5.4. Ferroelectric tunnel junction
8.6. Summary and future prospects
References
Chapter 9: ``Metal oxides in magnetic memories´´: Current status and future perspectives
9.1. Introduction
9.1.1. Spintronics
9.1.2. Magnetic tunnel junction (MTJ) and tunneling magnetoresistance (TMR)
9.1.3. Spin-transfer torque
9.2. Magnetic random access memory (MRAM)
9.2.1. MRAM storage principle
9.2.2. MRAM read principle
9.2.3. Early MRAM concepts
9.2.4. State-of-the-art MRAMs
9.3. Metal oxides in MRAMs
9.3.1. Tunnel barrier and TMR magnitude
9.3.2. The importance of oxide on MTJ crystal structure
9.3.3. Tunnel barrier breakdown
9.4. Perspectives
References
Chapter 10: Correlated transition metal oxides and chalcogenides for Mott memories and neuromorphic applications
10.1. Introduction
10.2. Mott insulators and Mott transitions
10.2.1. Mott insulators: Definition
10.2.2. Insulator to metal transition in Mott insulators
10.2.3. Renowned examples of Mott and charge transfer insulators
10.2.3.1. The vanadium oxide (V1-xCrx)2O3
10.2.3.2. NiS2-xSex
10.2.3.3. AM4Q8 compounds
10.2.4. The famous example of VO2 that exhibit an IMT but not a Mott transition
10.3. Electric Mott transitions
10.3.1. Joule heating and thermally driven IMT
10.3.2. Electromigration and filling controlled IMT
10.3.3. Dielectric breakdown and bandwidth controlled IMT
10.4. Electric Mott transition by dielectric breakdown: Detailed mechanism
10.4.1. Phenomenology of the dielectric breakdown
10.4.2. The initial spark: Creation of hot electrons and electronic avalanche
10.4.3. Runaway process and creation of a conducting filamentary path
10.4.4. Non-volatile electric Mott transition: A consequence of lattice compression specific to Mott physics
10.5. Microelectronic applications of Mott insulators: Toward Mottronics
10.5.1. Mott memories
10.5.1.1. Mott memories based on a thermally driven mechanism
10.5.1.2. Mott memories based on filling-controlled mechanism
10.5.1.3. Mott memories based on bandwidth-controlled mechanism
10.5.1.3.1. Performances of Mott insulator based ReRAM devices
10.5.2. New components based on Mott insulators for neurocomputing
10.5.2.1. Artificial synapses
10.5.2.2. Artificial neurons
10.5.2.2.1. Neuristor based on NbO2
10.5.2.2.2. Leaky integrate and fire neuron based on Mott insulators
10.6. Conclusion
References
Chapter 11: The effect of external stimuli on the performance of memristive oxides
11.1. Introduction
11.2. Electrical field
11.3. Magnetic field
11.4. Thermochemical treatments
11.5. Strain
11.6. Radiation
11.7. Outlook
References
Chapter 12: Nonvolatile MOX RRAM assisted by graphene and 2D materials
12.1. MOX RRAM with graphene-based electrodes
12.1.1. Transparent and flexible graphene electrode
12.1.2. Low power dissipation graphene electrode
12.1.3. Scaling down of MOX RRAM by graphene edge electrode
12.1.4. Heat-resistant graphene electrode for high-temperature device
12.1.5. MOX RRAM with reduced graphene oxide (rGO) electrode
12.2. Modulating ion migration in MOX RRAM by 2D materials
12.2.1. Inducement of ion migration by sporadic 2D material fragments
12.2.2. Suppressing the over-growth of CF by 2D material barrier
12.2.3. Monitoring ion migration by 2D material barrier
12.2.4. Confining ion migration in MOX RRAM by 2D material defect engineering
12.3. MOX RRAM assisted by additional 2D intercalation layer
12.3.1. MOX RRAM with intrinsic 2D materials
12.3.2. Oxidized film of 2D materials for MOX RRAM
12.4. Conclusion
References
Chapter 13: Ubiquitous memristors on-chip in multi-level memory, in-memory computing, data converters, clock generation an ...
13.1. Introduction
13.2. Multi-level memory and in-memory arithmetic structures
13.3. ADC and DAC in-memory data converters
13.4. Memristor-based clock signal generators
13.5. Metastable memristive transmission lines
13.6. Conclusions
Acknowledgment
References
Chapter 14: Neuromorphic applications using MOx-based memristors
14.1. Introduction on neuromorphic computing
14.2. Recap of MOx-based memristor technology
14.2.1. Switching mechanisms
14.2.2. State retention and volatile effects
14.2.3. Arrangement of devices into crossbars
14.3. Advanced memristor functionalities useful for neuromorphic applications
14.3.1. Multilevel operation
14.3.2. Memristor plasticity
14.3.2.1. Analog plasticity
14.3.2.2. Stochastic plasticity
14.3.3. Computing with spike timing and spike rate
14.3.4. Negative differential resistance enabling oscillation
14.4. Overview of neuromorphic concepts and system prototypes
14.4.1. Acceleration of neural networks and machine learning algorithms
14.4.2. Spike-based brain-inspired architectures
14.4.2.1. Memristive synapses
14.4.2.2. Memristive neurons
14.4.2.3. Routing solutions for reconfigurable networks
14.4.2.4. Brain-inspired architectures and systems
14.4.3. Non-spiking analog computing
14.5. Conclusions and outlook
References
Index
Back Cover