Non-Volatile Memory Cell Screening via Weak Programming and Baking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing data retention screening process for non-volatile memory cells is inefficient, as it fails to detect defective cells that exhibit slow leakage, leading to increased testing time and costs, and some defective cells remain undetected due to the two-baking process.
Innovation Solution
A method involving erasing memory cells, weakly programming them to a modified erased state, performing initial and secondary read operations with specific margin read current thresholds, and baking the cells to detect defective cells with read currents below these thresholds, optimizing the screening process by setting margin thresholds closer to the programmed state for improved detection efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a two-baking process is used for data retention screening, then data retention can be verified, but testing time and costs increase
Solution Approach 1:
The patent applies a preliminary weak programming operation before the baking process to place memory cells in a modified erased state. This preliminary action prepares the cells so that during the subsequent baking process, any charge leakage will cause the read current to drop below a threshold, enabling detection of defective cells. This approach allows data retention verification to be achieved more efficiently than traditional two-baking processes.
2Ease of operation
If traditional screening thresholds are used, then standard read operations can be performed, but defective cells with slow leakage remain undetected
Solution Approach 1:
The patent changes the read current threshold parameter from the standard read reference current to a lower threshold that accounts for charge leakage during baking. By setting this adjusted threshold, the screening process can detect defective cells with slow leakage that would otherwise remain undetected during standard read operations, thereby improving measurement precision for defective cell detection.
3Productivity
If margin read current thresholds are set closer to the programmed state, then detection efficiency improves, but false negatives may increase
Solution Approach 1:
The patent applies a preliminary weak programming operation to place memory cells in a modified erased state before baking. This preliminary action ensures that during the subsequent read operation after baking, the read current of good cells remains above the adjusted threshold, while defective cells with charge leakage fall below the threshold. This approach improves detection efficiency while maintaining reliability by reducing false negatives.
4Loss of time
If the screening process is simplified, then testing time is reduced, but detection accuracy of slow leakage defects decreases
Solution Approach 1:
The patent changes the read current threshold parameter to a lower value that is specifically designed to detect slow leakage defects. This single threshold adjustment, combined with the weak programming pre-treatment, enables the simplified screening process to maintain high detection accuracy for slow leakage defects while reducing overall screening time compared to traditional multi-step processes.
Data Source
AI summary
A method for screening memory cells includes erasing the memory cells, weakly programming the memory cells to a modified erased state, performing a first read operation on the memory cells after the erasing and the weakly programming, screening any of the memory cells that exhibit a read current during the first read operation below a margin read current threshold M1, baking the memory cells after the first read operation, performing a second read operation on the memory cells after the baking, and screening any of the memory cells that exhibit a read current during the second read operation below the margin read current threshold M1.


