In this study, to improve the threshold voltage ( ${V}_{\text {th}}$ ) variability and cell performance in three-dimensional (3-D) NAND flash memory, we analyzed the electrical characteristics with respect to various channel thickness ( ${T}_{\textsf {ch}}$ ) and average grain size (GS) values. The 3-D random Voronoi grain patterns were applied to a polycrystalline silicon (poly-Si) channel to determine the actual grain shape using technology computer-aided design (TCAD). For statistical analysis, key electrical characteristics such as the threshold voltage ( ${V}_{\textsf {th}}$ ), subthreshold swing (SS), maximum transconductance ( ${g}_{m}$ ), and on-current ( $I_{\mathrm{\scriptscriptstyle ON}}$ ) were extracted from samples with different patterns of grain boundaries (GBs) at specific ${T}_{\text {ch}}$ and GS values. The standard deviation of ${V}_{\text {th}}$ ( $\sigma {V}_{\text {th}}$ ) increased with an increase in GS at ${T}_{\text {ch}} >22$ nm, and no increase trend was observed for $\sigma {V}_{\text {th}}$ at ${T}_{\text {ch}} < 22$ nm. The mean SS, ${g}_{m}$ , and $I_{\mathrm{\scriptscriptstyle ON}}$ related to the performance improved overall with an increase in GS at the same ${T}_{\textsf {ch}}$ value. Based on a comprehensive analysis of various 3-D grain patterns, optimal structures were proposed in terms of variability and/or performance. Furthermore, based on the results, we suggest suitable ${T}_{\textsf {ch}}$ and GS parameters for the given target of 3-D NAND flash devices.

A machine learning (ML) method was used to optimize the trap distribution of the charge trap nitride (CTN) to simultaneously improve its performance/reliability (P/R) characteristics, which are tradeoffs in 3-D NAND flash memories. Using an artificial neural network (ANN), we modeled the relationship between trap distributions and P/R characteristics. The ANN was trained using a large experimentally-calibrated technology computer-aided design (TCAD) simulation dataset. The gradient descent method was adapted to optimize the trap distribution, achieving the best P/R characteristics based on the well-trained ANN. Eventually, we found the best trap profile distributed in both space and energy. In particular, the energetic trap distribution had a larger impact on the P/R characteristics than that of the spatial trap distribution. Furthermore, in terms of the P/R characteristics, it was generally preferable to increase all inputs of the energetic trap distribution. However, the acceptor-like trap energy level ( E_TA ) and its standard deviation (σ _EA ) caused a tradeoff between P/R characteristics; therefore, ML was used to determine their optimal points. The proposed ML method allows the optimization of trap distribution to obtain the best P/R characteristics rapidly and quantitatively. Our findings could be used as a guideline for determining the physical properties of CTN in 3-D NAND flash cells.

Data retention (a time-variant characteristic of 3-D- NAND flash memory) is predicted through a bi-directional long short-term memory (LSTM) neural network (NN) model that learns sequential data obtained from chip measurements of a triple-level cell (TLC). The predicted results for all time points of each program (PGM) state are accurately predicted by the threshold voltage ( ${V}_{\text {th}}$ ) distribution. Thus, the predicted ${V}_{\text {th}}$ can be used to analyze the cause of retention failure. When the ${V}_{\text {th}}$ of the target cell is high or when that of the adjacent cell is small, the ${V}_{\text {th}}$ loss of the target cell is large. In addition, the ${V}_{\text {th}}$ loss increases as the ${V}_{\text {th}}$ of the adjacent cell decreases. Using a fully calibrated TCAD simulation, we verify the NN-based ${V}_{\text {th}}$ prediction by checking the change in the electron concentration in the nitride layer. Furthermore, the NN model predicts the ${V}_{\text {th}}$ for cells existing in other blocks, showing that they are consistent with the measured ${V}_{\text {th}}$ . The prediction times were 5 $\times \,\,10^{{5}}$ s, 5 $\times \,\,10^{{6}}$ s, and 2 $\times \,\,10^{{6}}$ s, but using machine learning (ML), we reduced the time required to predict the ${V}_{\text {th}}$ to only 2 s. Therefore, the proposed ML method enables fast, accurate, and effective predictive modeling of the time-variant ${V}_{\text {th}}$ of 3-D TLC NAND flash memory. Finally, the predicted ${V}_{\text {th}}$ can be included in the read retry table or included in the lookup table of the compensation circuit in NAND solutions. This can save a significant amount of time that would otherwise be spent on actual long-term measurements.

The enhancement of program efficiency is essential for fast NAND cell operation. However, it is difficult to simultaneously consider many factors, such as structural parameters and trap characteristics, having complex relationships. To overcome these problems, we proposed a neural network (NN)-applied optimization method. First, an optimal network structure was selected by comparing the network performance and learning time. The selected network accurately predicted the threshold voltages of the 21 states of a single NAND cell within a second. Next, an optimization method to improve program efficiency is suggested. The improved NAND cell structure is obtained using a trained NN and numerical method. Here, the optimization required only a few minutes for one optimization process and could consider all parameters simultaneously. Finally, the optimized NAND cell was evaluated using a technology computer-aided design (TCAD) simulation, and its program efficiency was verified. This study shows a specific example of machine learning applied to the semiconductor area, especially in NAND flash memory.

A machine-learning (ML) technique was used to optimize the energetic-trap distributions of nano-scaled charge trap nitride (CTN) in 3D NAND Flash to widen the threshold voltage (V_th) window, which is crucial for NAND operation. The energetic-trap distribution is a critical material property of the CTN that affects the V_th window between the erase and program V_th. An artificial neural network (ANN) was used to model the relationship between the energetic-trap distributions as an input parameter and the V_th window as an output parameter. A well-trained ANN was used with the gradient-descent method to determine the specific inputs that maximize the outputs. The trap densities (N_TD and N_TA) and their standard deviations (σ_TD and σ_TA) were found to most strongly impact the V_th window. As they increased, the V_th window increased because of the availability of a larger number of trap sites. Finally, when the ML-optimized energetic-trap distributions were simulated, the V_th window increased by 49% compared with the experimental value under the same bias condition. Therefore, the developed ML technique can be applied to optimize cell transistor processes by determining the material properties of the CTN in 3D NAND Flash.

Electrical characteristics of scaled 3-D nand cells for next-generation node development were predicted using transfer learning with limited data. The nand cell structure parameters were considered as the inputs, and outputs included key electrical characteristics, such as cell ${V}_{t}$ , the difference in ${V}_{t}$ between the initial and programming states ( $\Delta {V}_{t}$ ), subthreshold swing (SS), and on-current ( ${I}_{\text {ON}}$ ). A multilayer perceptron (MLP) model comprising four hidden layers and focusing on large nand cells (25 nm gate length) with 2000 data points served as a pre-trained model. The transfer model leveraged pre-trained knowledge to predict the electrical characteristics of smaller cells (19 nm gate length) with 500 data points without weight and bias training. Evaluation of test data exhibited remarkable accuracy with both the mean and standard deviation below 3%, proving the model’s effectiveness despite limited data. In addition, a comprehensive evaluation was conducted by comparing the performance of the model with variations in the dataset size and the presence of transfer learning, highlighting the effectiveness and advantages of transfer learning. Transfer learning could provide detailed structure information of the next node for engineers and expedite device development, resulting in significant time and cost savings.