Mobile QR Code QR CODE

  1. Department of Electrical and Computer Engineering and the Inter-university Semiconductor Research Center (ISRC), Seoul National University, Gwanak-gu, Seoul 08826, Republic of Korea
  2. Flash Process Architecture Team, Flash Development Center, Samsung Electronics Company, Hwaseong, Gyeonggi 18448, Republic of Korea



3D NAND flash memory, channel hole profile, horizontal scaling, program efficiency, device variation, high temperature operation

I. INTRODUCTION

The development of large-scale and high-density non-volatile memory devices is continuously driven by the rapid growth of data-intensive applications such as neuromorphic hardware, content-addressable memory (CAM), edge-computing, as well as by increasing demand for conventional solid-state devices (SSDs) [1- 7]. To meet these demands, vertical stacking has emerged as a key scaling paradigm in 3D NAND flash memory, enabling dramatic improvements in bit density by increasing the number of word-line (WL) stacks. Recent 3D NAND flash technologies have now exceeded 500 stacked WLs and continue advancing toward even higher integration [8].

However, as the vertical height of memory strings increases, maintaining structural and material uniformity becomes increasingly difficult. High-aspect-ratio (HAR) channel hole etch processes challenge the ability to form vertical profiles with consistent critical dimensions (CD), and nonidealities such as tapered channel holes or elliptical cross-sections become more prevalent, as shown in Figs. 1(a) and 1(b) [9, 10]. Concurrently, conformal deposition of the oxide/nitride/oxide (ONO) stacks for charge-trapping into deep channel holes is increasingly limited by aspect ratio constraints, often resulting in ONO thickness variation along the vertical direction or around the azimuthal perimeter [7, 11]. These fabrication-induced variations degrade gate controllability and introduce cell-to-cell variation, particularly in the threshold voltage ($V_{th}$), subthreshold swing ($SS$), and bit-line (BL) current ($I_{BL}$) after program operation.

In addition, the surge in edge computing and artificial intelligence (AI) accelerator applications has led to more thermally demanding operating environments, in which the operating temperature is close to 373 K [12, 13]. As I/O circuits scale in performance, 3D NAND flash memory is exposed to elevated temperatures. Since the $V_{th}$ intrinsically decreases with temperature owing to the reduction of the Fermi potential and the associated depletion charge, the apparent read $V_{th}$ of a programmed cell shifts to lower values at high temperatures, even without any additional charge loss in terms of data retention, as depicted in Fig. 1(c) [12]. This temperature-dependent $V_{th}$ shift must be considered when defining reference voltages and sensing conditions, particularly for aggressively scaled multi-level operation, such as quad-level cell (QLC) and penta-level cell (PLC) [14].

Fig. 1. (a) Channel hole profile along the WL location and cell structure of 3D NAND flash memory. The physical device parameters are depicted. (b) Plan view of 3D NAND flash memory cells with fabrication variation: conventional hole profile, elliptic hole profile, thick ONO stack, and thin ONO stack. (c) The threshold voltage (Vth) distributions of 3D NAND flash utilizing triple-level cell (TLC) operation with various read temperatures.

../../Resources/ieie/JSTS.2026.26.5.399/fig1.png

In this work, we investigate the impact of channel hole diameter ($d_{hole}$) scaling and fabrication-induced channel hole profile variations on 3D NAND flash device performance. Using technology computer-aided design (TCAD) simulations, we examine the effects of channel tapering, ellipticity, and ONO stack variation on $V_{th}$, $SS$, and $I_{BL}$ in both the pristine and programmed states with practical device dimensions. Furthermore, we evaluate how elevated operating temperatures interact with these structural nonidealities to degrade $V_{th}$ stability. The findings provide critical insights into design and process considerations necessary to enable future 3D NAND flash scaling beyond 500 WL stacks.

II. SIMULATION CONDITION

A single string of 3D NAND flash memory with four WLs is simulated as shown in Fig. 1(a). The 3D device simulation was performed to analyze electrical characteristics according to the structural properties. The following physical models are included in the device simulation: nonlocal tunneling, the Shockley-Read-Hall recombination, and drift-diffusion transport [15]. The simulation parameters were carefully calibrated with reference to experimental data from fabricated 3D NAND flash memory devices based on our previous works [16, 17]. Detailed specifications for the simulation are summarized in Table 1.

The fabrication-induced variations arising from channel hole etch and ONO stack deposition are simulated using the variation of device parameters. First, $d_{hole}$ is reduced toward the bottom WLs due to the vertical tapering angle [18, 19]. Thus, the $d_{hole}$ range of 130 nm to 70 nm is simulated which reflects typical dimensions of commercial 3D NAND flash technologies [7]. Second, the elliptic hole profile is considered with various ellipticity ($a/b > 1$) of an ellipse to account for non-idealities of channel profiles in the HAR hole etch process [9]. Last, ONO thickness ($t_{ONO}$), ranging from 15 to 20 nm, incorporates a fluctuation of ~ 10 % during the deposition process [18]. The fluctuation of $t_{ONO}$ accounts for both chip-to-chip and WL-to-WL device variations due to the difficulty of uniform deposition along deep and narrow channel holes [20].

The influence of high temperature on read characteristics is evaluated over a range of 300K to 373K, where 300K corresponds to the programming temperature condition.

Table 1. Simulation parameters for 3D NAND flash memory.

Parameters

Values

Channel hole diameters ($d_{hole}$)

70, 90, 110, 130 nm

Gate length ($L_{gate}$)

24 nm

Inter-cell length ($L_{inter}$)

20 nm

Thickness of ONO stack ($t_{ONO}$)

15, 16, 17, 19, 20 nm

Channel thickness ($t_{ch}$)

7 nm

Gate work function

4.17 eV

Electron trap energy level in Si3N4

$-1.0\text{ eV}$ from $E_C$

Hole trap energy level in Si3N4

$-2.5\text{ eV}$ from $E_C$

Effective tunneling mass in SiO2

Electron: $0.23m_0$ / Hole: $0.29m_0$

III. RESULTS AND DISCUSSIONS

1. Influence of Channel Hole Diameter on 3D NAND Flash Memory

To investigate the impact of the channel hole geometry, device characteristics were analyzed by varying $d_{hole}$ with fixed parameters of $t_{ONO} = 17\text{ nm}$, $t_{ch} = 7\text{ nm}$, and $L_g = 24\text{ nm}$. Fig. 2(a) shows the $I_{BL}$-$V_{WL}$ curves for both pristine and programmed states under the same bias condition. As shown in Fig. 2(b), reducing $d_{hole}$ leads to a monotonic increase in $V_{th}$ and a steeper $SS$, which is consistent with the well-known electrostatic advantage of macaroni-type gate-all-around (GAA) structures. Thus, reducing the $d_{hole}$ strengthens the gate controllability over the channel potential, thereby suppressing short-channel effects (SCEs) such as drain-induced barrier lowering (DIBL) and $V_{th}$ roll-off [21, 22].

Fig. 2. (a) IBL-VWL characteristics of the 3D NAND flash memory cells with various channel hole diameter (dhole). 20 V program bias and 8 µs program time are applied to pristine cells. (b) The trend of threshold voltage (Vth) and subthreshold swing (SS) with various dhole. Vth is extracted using constant current method from a BL current (IBL) of 1 nA. (c) Radial electric field distribution during program operation. (d) IBL according to the overdrive voltage (Vov).

../../Resources/ieie/JSTS.2026.26.5.399/fig2.png

The program characteristics are also strongly affected by the $d_{hole}$. The programmed $V_{th}$ increases as $d_{hole}$ shrinks, primarily driven by the progressive widening of the program window ($\Delta V_{th}$) between the pristine and programmed states, as shown in Fig. 2(b). This phenomenon is attributed to the electric field curvature effect inherent to cylindrical structures [23, 24]. Fig. 2(c) shows the radial electric-field distribution across the ONO stack during program operation. As the curvature radius decreases, the electric field across the tunneling oxide increases significantly, whereas the field in the blocking oxide is slightly relaxed. This geometry-induced field enhancement exponentially boosts the Fowler-Nordheim (FN) tunneling from channel to CTL while mitigating the back tunneling from CTL to gate, thereby maximizing programming efficiency.

On the other hand, $I_{BL}$ exhibits an opposite trend. Fig. 2(d) shows $I_{BL}$ as a function of the $d_{hole}$ for various overdrive voltages ($V_{ov} = V_{WL} - V_{th\text{, programmed}}$). At a fixed $V_{ov} = 1.5\text{ V}$, $I_{BL}$ increases almost linearly with $d_{hole}$. This behavior arises because the thin macaroni body operates under quasi-volume inversion, where the inversion layer extends across the entire silicon shell, rendering the effective conduction width proportional to the channel circumference [21, 25]. This reflects a fundamental trade-off: smaller channels provide superior program efficiency at the expense of reduced drive current. However, under the low-$V_{ov}$ conditions relevant to multi-level operation, the current difference between large and small diameters becomes less pronounced because the improved $SS$ of the smaller-diameter cells compensates for the reduced perimeter. Thus, properly optimized small-diameter channels can simultaneously secure a wide program window and sufficient read current within a limited memory window, which is favorable for high-density multi-level 3D NAND flash memory.

2. Influence of Fabrication-Induced Variations on 3D NAND Flash Memory

Structural non-idealities originating from HAR etching are represented by an elliptic channel hole, whose distortion is quantified by the ellipticity [26]. Fig. 3(a) depicts that the programmed $V_{th}$ decreases as the ellipticity increases, whereas the $I_{BL}$ monotonically increases as shown in Fig. 3(b). When the circular channel is distorted into an ellipse, the local radius of curvature varies along the perimeter: the regions near the major axis have a smaller curvature radius, whereas those near the minor axis exhibit a larger curvature radius. Under such conditions, the gate electric field is more weakly coupled to the channel in the minor-axis sectors, which induces weak program efficiency. Thus, the programmed $V_{th}$ is determined by the weakly-controlled minor-axis region, and increasing ellipticity further enlarges the local curvature radius, leading to a reduction of the global $V_{th}$ [17, 27]. In addition, this $V_{th}$ degradation is more sensitive to the smaller $d_{hole}$. Since the electric field in the tunneling oxide is inversely proportional to the curvature radius, the sensitivity of the field to geometric distortion is significantly higher for smaller dimensions. Thus, even for the same ellipticity, the increase in the curvature radius at the minor axis induces a much steeper reduction of the electric field in the tunneling oxide for a smaller $d_{hole}$, resulting in a more severe deterioration of the programmed $V_{th}$.

Fig. 3. (a) Vth and (b) IBL of the programmed 3D NAND flash memory cells according to the elliptic hole profile distortion. The ellipticity (a/b) is adjusted by altering only the length of the major axis while maintaining the length of the minor axis.

../../Resources/ieie/JSTS.2026.26.5.399/fig3.png

Fig. 4. (a) Vth and (b) IBL of the programmed 3D NAND flash memory cells with various thickness of ONO stacks (tONO). The relative thickness ratio of the oxide/nitride/oxide layers is kept constant, and only the total ONO stack thickness is varied.

../../Resources/ieie/JSTS.2026.26.5.399/fig4.png

In contrast, the trend of $I_{BL}$ is predominantly governed by the channel cross-section rather than by the curvature-induced variation of the local $V_{th}$. Under read conditions with a sufficiently large $V_{ov}$ the entire channel perimeter is driven into inversion. Thus, increasing ellipticity which enlarges the cross-sectional area at a fixed length of minor axis, leads to a monotonic increase in $I_{BL}$.

Fig. 4 illustrates the impact of $t_{ONO}$ variation induced during the deposition process. A thicker ONO stack reduces the tunneling electric field and increases the tunneling distance during programming, which lowers the injected charge density in the CTL and reduces the programmed $V_{th}$. The sensitivity of $V_{th}$ to $t_{ONO}$ becomes weaker as $d_{hole}$ decreases, since the strong radial confinement in small-diameter GAA channels concentrates the electric field in the tunneling oxide and partially compensates for variations of the physical oxide thickness. In contrast, the dependence of $I_{BL}$ on $t_{ONO}$ is relatively weak (Fig. 4(b)); a slight reduction of current with increasing $t_{ONO}$ is attributed to the reduction of the cross-sectional area of the silicon channel.

3. Temperature Dependency of the Read Operation

The effect of operating temperature on the programmed state is illustrated in Fig. 5. As the read temperature increases from 300 K to 373 K, the programmed $V_{th}$ gradually shifts to lower values for all channel diameters (Fig. 5(a)). This trend originates from the intrinsic temperature dependence of $V_{th}$ in MOSFETs: with increasing temperature, the intrinsic carrier concentration rises, and the Fermi potential decreases, leading to a reduction in $V_{th}$ even in the absence of additional charge loss in the CTL [11]. In poly-Si channels, the $V_{th}$ degradation becomes more severe because the potential barrier at grain boundaries is also lowered as temperature increases, which further facilitates carrier transport [13].

Fig. 5. (a) Vth and (b) IBL of the programmed 3D NAND flash memory cells according to the read temperature. The program operation is performed at 300 K.

../../Resources/ieie/JSTS.2026.26.5.399/fig5.png

The $I_{BL}$ exhibits a more pronounced temperature dependence. Fig. 5(b) shows that $I_{BL}$ decreases monotonically with increasing read temperature when a fixed $V_{ov}$ is maintained. It is owing to the degradation of $SS$ which is proportional to the temperature. In the subthreshold and near-threshold region relevant to multi-level sensing, the current further decreases because both the thermal voltage and the $SS$ increase with temperature, reducing the exponential factor in the $I_{BL}$ expression. Thus, high-temperature operation simultaneously lowers the apparent $V_{th}$ and suppresses the read current, effectively narrowing the sensing margin. These results highlight the necessity of reference voltage calibration and compensation schemes in deeply scaled 3D NAND flash, particularly when small-diameter channels are adopted for high-density multi-level operation.

IV. CONCLUSIONS

This study investigated the electrical characteristics of 3D NAND flash memory under the influence of geometric and thermal variations. It was confirmed that reducing the $d_{hole}$ enhances gate controllability and program efficiency owing to the electrostatic curvature effect, leading to an increased programmed $V_{th}$ accompanied by a reduction in $I_{BL}$. However, small-diameter channels exhibited increased sensitivity to elliptic distortion while suppressing the impact of ONO thickness variation. Furthermore, high-temperature operation was found to degrade the $SS$, simultaneously lowering the apparent $V_{th}$ and $I_{BL}$. These findings indicate that precise control of structural profiles and temperature-aware sensing is required for reliable 3D NAND flash scaling.

ACKNOWLEDGMENTS

This work was supported in part by the Samsung Electronics, in part by the MSIT under Grant RS-2025-02314443 (Development of Flash Memory-based AI Processing Unit for On-Device AI), Grant IITP-2023-RS-2023-00256081 (Artificial Intelligence Semiconductor Support Program), in part by Korea Institute for Advancement of Technology (KIAT) grant funded by the Korea Government (MOTIE) (RS-2025-02214408, HRD Program for Industrial Innovation), and in part by the BK21 FOUR program of the Education and Research Program for Future ICT Pioneers, Seoul National University in 2026.

REFERENCES

1 
J. S. Woo, K. M. Choi, G. T. Park, J. H. Chang, and W. Y. Choi, “Fully CMOS-Compatible 3-T Embedded NOR Flash Memory Achieving 28 ns Long-Term Potentiation/Long-Term Depression for High-Speed On-line Training Accelerators”, Advanced Intelligent Systems, vol. 8, no. 2, 2025.DOI
2 
G. T. Park, J. W. Lee, J. S. Woo, and W. Y. Choi, “XNOR Operation of Binary Neural Networks Using Nanoelectromechanical Memory Switches”, IEEE Transactions on Electron Devices, vol. 71, no. 12, pp. 7955-7962, 2024.DOI
3 
J. W. Lee, C. Kim, G. T. Park, S. H. Baek, M. S. Shin, and W. Y. Choi, “Ternary Content-Addressable Memory Using One Capacitor and One Nanoelectromechanical Memory Switch for Data-Intensive Applications”, Advanced Intelligent Systems, 2025.DOI
4 
M. Ryu, J. S. Woo, Y. Kim, J. H. Jeon, S. I. Cho, and W. Y. Choi, “Time-Domain Content-Addressable Memory Based on Single Ambipolar Ferroelectric Memcapacitor for High-Density and Highly-Precise Distance Function Computation”, Advanced Electronic Materials, vol. 11, no. 20, p. e00421, 2025.DOI
5 
J. Yeom, J. S. Woo, J. W. Lee, C. Park, J. Park, M. Ryu, and W. Y. Choi, “Implementation of Ternary Neural Networks Based on Ferroelectric Capacitors”, IEEE Electron Device Letters, vol. 47, no. 1, pp. 73-76, 2026.DOI
6 
J. H. Chang, J. S. Woo, S.-K. Sung, K.-W. Song, and W. Y. Choi, “Large-Scale and Highly Reliable Hopfield Neural Networks Using Vertical NAND Flash Memory for the In-Memory Associative Computing”, Advanced Intelligent Systems, 2025.DOI
7 
R. Lee, S. Park, Y. Seo, J. Kim, J. Jung, J. Yeo, Y. Ahn, H. J. Kim, Y. Kim, J. Lee, D. Lee, J. Park, W. Kim, B. Lee, K. Noh, S. Hong, and S. H. Hur, “Highly Scalable and Reliable Cell Characteristics for 1Tb 9th Generation 3D-NAND Flash Memory”, IEEE Symposium on VLSI Technology and Circuits, pp. 1-3, 2025.DOI
8 
S.-S. Park, J.-D. Lyu, M. Kim, J. Lee, Y. Song, C.-H. Yu, M. Hirano, Y. Kwon, J.-H. Park, H.-J. Kim, D. Lee, D. Seo, B. Go, S. Jeon, Y. Kim, D.-H. Kim, Y. Jo, H. Yoon, J. Park, I. Kim, S. Kim, H. Lee, J.-H. Yu, S.-L. Kim, H.-S. Ku, J. Seo, J. Byun, S.-H. Yun, K. Kang, S.-B. Kim, Y. Lee, Y. Lee, K. Kang, H.-J. Lee, Y.-H. Ryu, H. Kim, W. Kim, H. Choi, J. Jeon, A. Park, R. Song, J.-H. Kim, J.-S. Kim, H.-S. Lee, M.-K. Lee, J.-I. Son, J. Cho, M. Kim, J.-W. Im, J. Park, H.-J. Kwon, Y. Choi, C. Yoon, S. Lee, K. Song, and S.-H. Hur, “30.1 A 28Gb/mm2 4XX-layer 1Tb 3b/cell WF-bonding 3D-NAND flash with 5.6Gb/s/pin IOs”, IEEE International Solid-State Circuits Conference (ISSCC), pp. 504-505, 2025.DOI
9 
S. Park, J. Lee, J. Jang, J. K. Lim, H. Kim, J. J. Shim, M.-T. Yu, J.-K. Kang, S. J. Ahn, and J. Song, “Highly-Reliable Cell Characteristics with 128-Layer Single-Stack 3D-NAND Flash Memory”, IEEE Symposium on VLSI Technology and Circuits, pp. 1-2, 2021.Google Search
10 
U. M. Bhatt, S. K. Manhas, A. Kumar, M. Pakala, and E. Yieh, “Mitigating the Impact of Channel Tapering in Vertical Channel 3-D NAND”, IEEE Trans. Electron Devices, vol. 67, no. 3, pp. 929-936, 2020.DOI
11 
K. Parat, and A. Goda, “Scaling Trends in NAND Flash”, IEEE International Electron Devices Meeting (IEDM), pp. 2.1.1-2.1.4, 2018.DOI
12 
D. Resnati, A. Goda, G. Nicosia, C. Miccoli, A. S. Spinelli, and C. M. Compagnoni, “Temperature effects in NAND flash memories: A comparison between 2-D and 3-D arrays”, IEEE Electron Device Letters, vol. 38, no. 4, pp. 461-464, 2017.DOI
13 
D. Resnati, A. Mannara, G. Nicosia, G. M. Paolucci, P. Tessariol, A. S. Spinelli, A. L. Lacaita, and C. M. Compagnoni, “Characterization and modeling of temperature effects in 3-D NAND Flash arrays–Part I: polysilicon-induced variability”, IEEE Transactions on Electron Devices, vol. 65, no. 8, pp. 3199-3206, 2018.DOI
14 
Y. Luo, S. Ghose, Y. Cai, E. F. Haratsch, and O. Mutlu, “HeatWatch: Improving 3D NAND flash memory device reliability by exploiting self-recovery and temperature awareness”, Proceeding of IEEE International Symposium on High Performance Computer Architecture (HPCA), pp. 504-517, 2018.DOI
15 
Synopsys, Sentaurus Device User Guide, Version P-2019.12. Mountain View, CA: Synopsys, Inc., 2019.Google Search
16 
J. H. Chang, J. S. Woo, S.-K. Sung, K.-W. Song, and W. Y. Choi, “Ternary-State Vertical NAND Flash Memory for the Improvement of the Density and Accuracy of Quantized Neural Networks”, IEEE Transactions on Electron Devices, vol. 71, no. 6, pp. 3985-3988, 2024.DOI
17 
J. H. Chang, J. H. Uhm, H. S. Kwon, E. Kwon, and W. Y. Choi, “Influence of channel hole remaining ratio on hemi-cylindrical vertical NAND flash memory”, IEEE Electron Device Letters, vol. 43, no. 9, pp. 1432-1435, 2022.DOI
18 
J. H. Nam, J. H. Chang, K. M. Kim, J. W. Lee, D. E. Yang, S.-K. Sung, J. H. Kim, and W. Y. Choi, “Investigation of Hot Carrier Injection During GIDL Erase Operation in 3-D NAND Flash Memory”, IEEE Transactions on Electron Devices, vol. 74, no. 3, pp. 1326-1332, 2026.DOI
19 
Y. Kihara, M. Tomura, W. Sakamoto, M. Honda, and M. Kojima, “Beyond 10 µm depth ultra-high speed etch process with 84% lower carbon footprint for memory channel hole of 3D NAND flash over 400 layers”, Proceedings of IEEE Symposium on VLSI Technology and Circuits, pp. 1-2, 2023.DOI
20 
C. M. Compagnoni, and A. S. Spinelli, “Reliability of NAND Flash Arrays: A Review of What the 2-D-to-3-D Transition Meant”, IEEE Transactions on Electron Devices, vol. 66, no. 11, pp. 4504-4516, 2019.DOI
21 
S.-H. Oh, D. Monroe, and J. M. Hergenrother, “Analytic description of short-channel effects in fully-depleted double-gate and cylindrical, surrounding-gate MOSFETs”, IEEE Electron Device Letters, vol. 21, no. 9, pp. 445-447, 2000.DOI
22 
G. M. Paolucci, A. S. Spinelli, C. Monzio Compagnoni, and P. Tessariol, “A Semi-Analytical Model for Macaroni MOSFETs With Application to Vertical Flash Memories”, IEEE Transactions on Electron Devices, vol. 63, no. 5, pp. 1871-1876, 2016.DOI
23 
H.-T. Lue, Y.-W. Jiang, M.-F. Hung, Y.-R. Su, L.-Y. Liang, C.-W. Hu, C.-J. Chiu, K.-C. Wang, C.-Y. Lu, T.-H. Yeh, P.-Y. Du, R. Lo, W.-C. Chen, T.-H. Hsu, C. Huang, G.-R. Lee, and C.-P. Chen, “A novel double-density hemi-cylindrical (HC) structure to produce more than double memory density enhancement for 3D NAND flash”, IEEE International Electron Devices Meeting (IEDM), pp. 28.2.1-28.2.4, 2019.DOI
24 
J. H. Uhm, J. H. Chang, J. Kim, E. Kwon, and W. Y. Choi, “Azimuthal charge redistribution of hemi-cylindrical vertical NAND flash memory”, IEEE Electron Device Letters, vol. 44, no. 6, pp. 931-934, 2023.DOI
25 
F. Balestra, S. Cristoloveanu, M. Benachir, J. Brini, and T. Elewa, “Double-gate silicon-on-insulator transistor with volume inversion: A new device with greatly enhanced performance”, IEEE Electron Device Letters, vol. 8, no. 9, pp. 410-412, 1987.DOI
26 
M. Higuchi, A. Fayrushin, A. G. Mauri, H. Liu, and Y. Fukuzumi, “Modeling of the impact of elliptical shapes on main Read Window Budget mechanisms in 3D NAND”, IEEE International Memory Workshop (IMW), pp. 1-4, 2025.DOI
27 
J. Lee, Y. Kim, Y. Shin, S. Park, D. Kang, and M. Kang, “Analysis of cell current with abnormal channel profile in 3D NAND flash memory”, Journal of Semiconductor Technology and Science, vol. 24, no. 2, pp. 138-143, 2024.DOI
Jin Ho Chang
../../Resources/ieie/JSTS.2026.26.5.399/au1.png

Jin Ho Chang received the B.S. and M.S. degrees from the Department of Electronic Engineering, Sogang University, Seoul, South Korea, in 2021 and 2023, respectively. He is currently working toward a Ph.D. degree with the Department of Electrical and Computer Engineering, Seoul National University (SNU), Seoul, South Korea. He is also with the Inter-University Semiconductor Research Center (ISRC), SNU. His current research interests include NAND flash memory, brain-inspired computing devices, and nanoscale emerging devices.

Seon Ho Lee
../../Resources/ieie/JSTS.2026.26.5.399/au2.png

Seon Ho Lee received the B.S. degree from the Department of Electrical Engineering from Korea University, Seoul, Korea. He is currently working toward an integrated Ph. D. degree in the Department of Electrical and Computer Engineering, Seoul National University (SNU), Seoul, South Korea. He is also with the Inter-University Semiconductor Research Center (ISRC), SNU. His current research interests include NAND flash memory.

Jaemin Yeom
../../Resources/ieie/JSTS.2026.26.5.399/au3.png

Jaemin Yeom received the B.S. degree from the Department of Electrical and Computer Engineering from Seoul National University (SNU), Seoul, South Korea, in 2023. He is currently working toward the Ph.D. degree with the Department of Electrical and Computer Engineering, SNU, Seoul, South Korea. He is also with the Inter-University Semiconductor Research Center (ISRC), SNU. His research interests include ferroelectric, neuromorphic devices, and NAND flash memory.

Jae Hyun Nam
../../Resources/ieie/JSTS.2026.26.5.399/au4.png

Jae Hyun Nam received the B.S. degree from the Department of Electrical and Electronics Engineering from Chung-Ang University, South Korea in 2023. He is currently working toward an M.S. degree in the Department of Electrical and Computer Engineering, Seoul National University (SNU), Seoul, South Korea. He is also with the Inter-University Semiconductor Research Center (ISRC), SNU. His current research interests include, NAND Flash Memory.

Kyung Moon Kim
../../Resources/ieie/JSTS.2026.26.5.399/au5.png

Kyung Moon Kim received the B.S. degree from the Department of Electronic and Electrical Engineering from Sungkyunkwan University, South Korea in 2015. In 2016, he joined Samsung Electronics Company, Ltd., Gyunggi-Do, South Korea. He is currently working toward an M.S. degree in the Department of Electrical and Computer Engineering, Seoul National University (SNU), Seoul, Republic of Korea. He is also with the Inter-University Semiconductor Research Center (ISRC), SNU. His current research interests include NAND Flash Memory.

Jisun Baek
../../Resources/ieie/JSTS.2026.26.5.399/au6.png

Jisun Baek received the B.S. degree from the Department of Advanced Materials Engineering for Information and Electronics from Kyung Hee University, South Korea, in 2015, and the M.S. degree from the Department of Materials Science and Engineering from Yonsei University, South Korea, in 2017. In 2017, she joined Samsung Electronics Company, Ltd., Gyeonggi-do, South Korea. She is currently working toward a Ph.D. degree in the Department of Electrical and Computer Engineering, Seoul National University (SNU), South Korea. Her current research interests include NAND flash memory.

Da Eun Yang
../../Resources/ieie/JSTS.2026.26.5.399/au7.png

Da Eun Yang received the B.S. degree in the Department of Electronic and Electrical Engineering from Sungkyunkwan University, South Korea in 2023. She is currently working toward an M.S. degree in the Department of Electrical and Computer Engineering, Seoul National University (SNU), Seoul, South Korea. She is also with the Inter-University Semiconductor Research Center (ISRC), SNU. Her current research interests include NAND Flash Memory.

Suk-Kang Sung
../../Resources/ieie/JSTS.2026.26.5.399/au8.png

Suk-Kang Sung received the B.S., M.S and Ph.D degrees in electrical engineering from Seoul National University, Seoul, Korea. Since 2004, he has been with Samsung Electronics, where he has made significant contributions to the field of nonvolatile memory technologies. From 2004 to 2009, he was engaged in pioneering research on nanoscale CMOS devices including FinFETs, as well as nonvolatile memory technologies such as Charge Trap Flash (CTF) and nanocrystal memory, at the Semiconductor Research Center. In 2009, he transitioned to the Flash Process Architecture Team in the Memory Division of Samsung Electronics, where he developed several generations of planar floating-gate NAND and vertical CTF 3D-NAND products. He holds approximately 70 U.S. patents and has authored and co-authored over 60 technical journal and conference papers. Currently, he serves as a VP of Technology (Master) at Samsung Electronics, responsible for flash process architecture design and technological innovations. His current research interests focus on the integration of flash memory architectures to meet the demands of high-performance computing in the AI era, as well as the exploration of new nonvolatile memory device technologies.

Jun Hyoung Kim
../../Resources/ieie/JSTS.2026.26.5.399/au9.png

Jun Hyoung Kim received the B.S., M.S., and Ph.D. degrees from the Department of Electronic Engineering from University of Tokyo, Japan in 2007, 2009, and 2016, respectively. Since 2017, he has been with Samsung Electronics Company, Ltd., Kyungki-Do, Korea. and researched several generation Vertical 3D-NAND products. From 2022, he joined the Advanced Flash Technology Team. His current research interests are future nonvolatile memories including SONOS and nanocrystal memory and emerging device, NAND architecture for high performance, and highly manufacturable integration process.

Byung Chul Lee
../../Resources/ieie/JSTS.2026.26.5.399/au10.png

Byung Chul Lee received the B.S. and Ph.D. degrees from the Department of Electrical Engineering from Korea University, Seoul, South Korea, in 2013, 2021, respectively. In 2021, he Joined Samsung Electronics Company, Ltd., Gyeonggi-do, South Korea.

Woo Young Choi
../../Resources/ieie/JSTS.2026.26.5.399/au11.png

Woo Young Choi received the B.S., M.S., and Ph.D. degrees from the School of Electrical Engineering, Seoul National University, Seoul, South Korea, in 2000, 2002, and 2006, respectively. From 2006 to 2008, he held a postdoctoral position with the Department of Electrical Engineering and Computer Sciences, University of California at Berkeley, Berkeley, CA, USA. From 2008 to 2022, he was a professor at the Department of Electronic Engineering, Sogang University, Seoul. Since 2022, he has been a Faculty Member with Seoul National University, Seoul, where he is currently a Professor with the Department of Electrical and Computer Engineering.