Threshold voltage prediction apparatus and method for nonvolatile memory

Abstract

A threshold voltage prediction apparatus according to the present invention comprises: a memory storing a program for providing operation simulation for a nonvolatile memory apparatus; and a processor that executes the program stored in the memory, wherein the program: receives device characteristics including an arrangement structure of a plurality of devices and initial operation parameters to generate a virtual nonvolatile memory; sets a variable parameter for determining the degree of charge loss for each of a plurality of compact models based on a charge loss model and applies the variable parameter to each device; performs Monte-Carlo simulation for the virtual nonvolatile memory on the basis of the plurality of compact models applied to respective devices and a physical mechanism causing charge loss; and outputs a change in threshold voltage of each device, which is predicted on the basis of results of the simulation.

Description

Device and method for predicting threshold voltage of nonvolatile memory

[0001] The present invention relates to a device and method for predicting a threshold voltage of a nonvolatile memory.

[0002] Non-volatile memory, such as flash memory, can store information even when power is not supplied. This memory stores information by storing charges and controlling the threshold voltage distribution of transistors. In flash memory, threshold voltages are not uniform across each transistor, but rather distributed, as shown in Figure 1, depending on manufacturing process variables and random variables. Accurately predicting and controlling the threshold voltage distribution is crucial to minimizing errors when reading stored information.

[0003] However, as multi-bit technology for increasing bit density develops, the free space between distributions decreases, and errors occur when distributions overlap due to the reduced free space, so an accurate understanding and interpretation of the space between distributions is required.

[0004] Charge loss is a major factor in generating errors. Charge loss, also known as retention, primarily refers to the change in threshold voltage reduction caused by charge loss. This charge loss tends to increase over time, and the magnitude of the physically occurring components varies depending on the device structure, pattern, and threshold voltage. However, previous studies have mainly simulated only the average of a single transistor or multiple transistors or the movement of transistors at specific points, making it difficult to address distributed issues that arise during actual operation.

[0005] Conventional Monte Carlo simulation techniques for simulating charge loss in flash memory, a non-volatile memory, from the perspective of threshold voltage distribution simply apply probability density distributions based on measured data results as input values, making it impossible to alter internal physical parameters. Furthermore, existing analyses and modeling of charge loss elements have been conducted on the average of multiple devices or at a specific p-level within a certain distribution, presenting limitations in analyzing the impact of charge loss from the perspective of threshold voltage distribution.

[0006] Therefore, technologies to solve these problems are required.

[0007] The present invention, in order to solve the above-mentioned problem, provides a threshold voltage prediction device and method for predicting a threshold voltage change of a non-volatile memory by performing a Monte-Carlo simulation on a virtual non-volatile memory based on a plurality of variable parameters affecting a threshold voltage change and a physical mechanism causing charge loss in a compact model.

[0008] However, the technical tasks that this embodiment seeks to accomplish are not limited to the technical tasks described above, and other technical tasks may exist.

[0009] As a technical means for solving the above-described technical problem, a threshold voltage prediction device according to the present invention comprises: a memory storing a program that provides an operation simulation for a nonvolatile memory device; and a processor executing the program stored in the memory, wherein the program receives characteristics of a device including a layout structure and initial operation parameters of a plurality of devices to generate a virtual nonvolatile memory, sets variable parameters that determine a degree of charge loss for each of a plurality of compact models based on a charge loss model, and applies the variable parameters to each device, performs a Monte-Carlo simulation on the virtual nonvolatile memory based on the plurality of compact models applied to each device and a physical mechanism causing charge loss, and outputs a threshold voltage change of each device predicted based on the simulation results.

[0010] In addition, a threshold voltage prediction method for predicting a threshold voltage of a nonvolatile memory using a threshold voltage prediction device according to an embodiment of the present invention includes the steps of: receiving characteristics of a device including a layout structure and initial operating parameters of a plurality of devices to generate a virtual nonvolatile memory; setting variable parameters for determining a degree of charge loss for each of a plurality of compact models based on a charge loss model and applying the set of variable parameters to each device; performing a Monte-Carlo simulation on the virtual nonvolatile memory based on the plurality of compact models applied to each device and a physical mechanism causing charge loss; and predicting a threshold voltage change of each device based on a simulation result.

[0011] According to the above-described problem solving means of the present invention, a threshold voltage distribution is predicted by combining a physical mechanism affecting charge loss, multiple variable parameters of a compact model, and statistical probabilistic Monte Carlo simulation, so that simulation results close to actual data can be derived.

[0012] Additionally, you can change the variable parameter values ​​of the compact model to check and analyze various results accordingly.

[0013] Additionally, since it outputs predicted threshold voltage changes based on a mathematical model, it can be applied to new semiconductors, unlike existing methods.

[0014] Figure 1 is an example diagram for explaining the threshold voltage error of a conventional non-volatile memory.

[0015] FIG. 2 is a conceptual diagram schematically illustrating the configuration of a threshold voltage prediction device according to one embodiment of the present invention.

[0016] Figure 3 is an example of an energy band diagram for a non-volatile memory device.

[0017] Figure 4 is an example graph showing a change in threshold voltage according to a change in variable parameters of a compact model.

[0018] Figure 5 is a graph showing an example of threshold voltage changes according to a physical mechanism and a final threshold voltage change that is a composite of the threshold voltage changes.

[0019] Figure 6 is an example of a graph showing threshold voltage changes for a given element.

[0020] Figure 7 is an example of a graph showing the tail tendency results of the threshold voltage distribution.

[0021] Figure 8 is an example of a graph showing the bit error rate.

[0022] Figure 9 is an example of a graph showing the expected lifespan of a device.

[0023] FIG. 10 is a flowchart illustrating a threshold voltage prediction method according to an embodiment of the present invention.

[0024] Hereinafter, the present invention will be described in detail with reference to the attached drawings. However, the present invention can be implemented in various different forms and is not limited to the embodiments described herein. In addition, the attached drawings are only intended to facilitate easy understanding of the embodiments disclosed in the present specification, and the technical ideas disclosed in the present specification are not limited by the attached drawings. In order to clearly explain the present invention in the drawings, parts that are not related to the description are omitted, and the size, shape, and shape of each component shown in the drawings can be variously modified. The same / similar drawing reference numerals are assigned to the same / similar parts throughout the specification.

[0025] The suffixes "module" and "part" used in the following description for components are assigned or used interchangeably solely for the convenience of writing the specification, and do not in themselves have distinct meanings or roles. Furthermore, in describing the embodiments disclosed herein, detailed descriptions of related known technologies have been omitted if they are deemed to obscure the gist of the embodiments disclosed herein.

[0026] Throughout the specification, when a part is said to be "connected (connected, in contact with, or coupled)" to another part, this includes not only cases where it is "directly connected (connected, in contact with, or coupled)" but also cases where it is "indirectly connected (connected, in contact with, or coupled)" with another member in between. Furthermore, when a part is said to "include (have or provide)" a certain component, this does not mean that it excludes other components, but rather that it may "include (have or provide)" other components, unless otherwise specifically stated.

[0027] As used herein, ordinal terms such as "first," "second," etc., are used solely to distinguish one component from another and do not limit the order or relationship between the components. For example, the first component of the present invention may be referred to as the "second component," and similarly, the second component may also be referred to as the "first component."

[0028] FIG. 2 is a conceptual diagram schematically illustrating a threshold voltage prediction device according to one embodiment of the present invention.

[0029] Referring to FIG. 2, a threshold voltage prediction device (100) according to an embodiment of the present invention is described. The voltage prediction device (100) predicts a threshold voltage change according to charge loss for each element through an operation simulation for a non-volatile memory based on a physical mechanism that causes charge loss. To perform such an operation, the threshold voltage prediction device (100) includes a memory (110) and a processor (120).

[0030] The memory (110) stores a program that provides an operation simulation for a non-volatile memory. The program receives the arrangement structure of a plurality of elements, the characteristics of the elements, and the initial operation parameters to create a virtual non-volatile memory, sets variable parameters that determine the degree of charge loss for each of a plurality of compact models based on a charge loss model, and applies the variable parameters to each element, performs a Monte-Carlo simulation for the virtual non-volatile memory based on the plurality of compact models applied to each element and the physical mechanism causing the charge loss, and predicts a threshold voltage change of each element based on the simulation results.

[0031] Meanwhile, the memory (110) should be interpreted as a general term for a non-volatile storage device that maintains stored information even when no power is supplied and a volatile storage device that requires power to maintain the stored information. The memory (110) may perform a function of temporarily or permanently storing data processed by the processor (120). The memory (110) may include a magnetic storage media or a flash storage media in addition to a volatile storage device that requires power to maintain the stored information, but the scope of the present invention is not limited thereto.

[0032] And, the processor (120) executes a program stored in the memory (110), performs a Monte-Carlo simulation on a virtual non-volatile memory based on a plurality of compact models applied to each of a plurality of elements of the non-volatile memory and a physical mechanism causing charge loss, and predicts a threshold voltage change of each element based on the simulation results. In the present embodiment, the processor (120) may be implemented in the form of a microprocessor, a central processing unit (CPU), a processor core, a multiprocessor, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), etc., but the scope of the present invention is not limited thereto.

[0033] In addition, the processor (120) provides a function of executing a threshold voltage prediction program that predicts a threshold voltage change of a non-volatile memory stored in the memory (110) and controlling the hardware of the threshold voltage prediction device (100) according to the execution of the threshold voltage prediction program. That is, the processor (120) can perform hardware control functions such as a necessary file system, memory allocation, network, basic library, timer, device control (display, media, input device, 3D, etc.), and other utilities according to the execution of the program.

[0034] Next, the operation of the threshold voltage prediction program to predict threshold voltage changes for each of multiple elements of nonvolatile memory is described.

[0035] The threshold voltage prediction program receives the layout structure of multiple devices, the characteristics of the devices, and the initial operating parameters, and models a virtual non-volatile memory based on the layout structure of the multiple devices, the characteristics of the devices, and the initial operating parameters, and sets variable parameters that determine the degree of charge loss for each of the multiple compact models and applies them to each device. For reference, the present invention targets devices such as 3D NAND flash, and the layout structure of the devices is a multi-layer stacked structure arranged on a substrate, and more specifically, a structure vertically stacked on the substrate can be exemplified. Accordingly, the layout structure can include characteristic information such as the distance between cells, the number of stacked layers, and the cell arrangement of each layer. In this way, multiple compact models can be prepared by considering various types of layout structures. Thereafter, a Monte-Carlo simulation is performed on the virtual non-volatile memory based on the multiple compact models applied to each device and the physical mechanism that causes the charge loss, and a threshold voltage change prediction result of each device is output based on the simulation results.

[0036] A compact model based on the charge loss model can be defined through mathematical equation 1.

[0037]

[0038] In mathematical equation 1, is the change in threshold voltage due to charge loss by the physical mechanism k ( ) represents, represents the time constant, represents the shape parameter of the charge loss model, represents the charge loss time. Here, the variable parameter is , and There can be one or more of them. Here, each variable parameter can have different weights depending on the physical mechanism k.

[0039] Additionally, the physical mechanism may be one or more of detrapping, trap assisted tunneling (TAT), interface trap recovery, and lateral migration.

[0040] Figure 3 is an example of a schematic energy band diagram for a 3D NAND flash memory device. Referring to Figure 3, the physical mechanisms that cause charge loss are described. Detrapping refers to the unintentional trapping of electrons in the tunneling oxide during a program operation that writes data to non-volatile memory, and trap-assisted tunneling (TAT) refers to the easier tunneling of electrons trapped in the charge trap layer (CTL) toward the channel with the help of a single trap in the tunneling oxide. In addition, interface trap recovery refers to the recovery of an interface trap created by a program and erase (P / E) cycle, and lateral migration refers to the direction of charge movement in the charge trap layer (CTL).

[0041] Next, we describe the operation of the threshold voltage prediction program performing a simulation on a virtual non-volatile memory.

[0042] To explain the overall flow of the simulation performed by the threshold voltage prediction program, the simulation applies the same operations that occur in the operation of an actual nonvolatile memory to the modeled virtual nonvolatile memory, such as programming at each programming level in the erase operation that erases data recorded in the nonvolatile memory, and then increasing the threshold voltage by surrounding cells (Z-interference), decreasing the threshold voltage by charge loss in the charge-trap layer (Retention), and increasing the threshold voltage by the number of cycles (Cycling) after programming.

[0043] Specifically describing the simulation operation, the threshold voltage prediction program models a virtual non-volatile memory based on the arrangement structure of multiple devices, generates a threshold voltage distribution for erasing before a program operation for writing data to the virtual non-volatile memory based on the characteristics and initial operating parameters of the devices, and sets the initial operating parameters for each device by randomly matching the thickness of each layer of the structure according to the process variables.

[0044] Thereafter, the threshold voltage prediction program determines the simulation operation order of multiple compact models in the same order as the operation of an actual memory device, performs Monte Carlo simulation for each compact model based on the variable parameters and physical mechanisms of each compact model, and can predict threshold voltage changes based on the simulation results.

[0045] Fig. 4 is a graph exemplarily illustrating threshold voltage changes according to changes in variable parameters of a compact model. Fig. 4 shows that threshold voltage changes according to changes in each variable parameter for any one of multiple physical mechanisms causing charge loss. As shown in Fig. 4, the compact model can predict threshold voltage changes according to changes in each variable parameter for each physical mechanism, and comprehensively reflect this to predict threshold voltage changes for multiple physical mechanisms.

[0046] Figure 5 shows the simulation results of a compact model for each physical mechanism according to temperature changes, the combined simulation results, and actual measured values. It can be confirmed that the threshold voltage change of the virtual non-volatile memory predicted by the threshold voltage prediction program through Monte Carlo simulation matches the actual measured threshold voltage change.

[0047] The present invention utilizes a compact model based on TCAD simulation or measurement results, and employs Monte Carlo simulation to apply it in a distributed manner. Therefore, the parameters of the Monte Carlo simulation input random variables or the parameters of the simulation model can be modified, allowing for the verification and analysis of the results of various output distributions.

[0048] Fig. 6 is a graph showing a threshold voltage change for a given element, and Fig. 7 is a graph showing the results of the tail tendency of the threshold voltage distribution. The threshold voltage prediction program can output simulation results as shown in Figs. 6 and 7. Fig. 6 is a graph showing a threshold voltage change of one element among a plurality of elements, and a graph like Fig. 6 can be output for each element. In addition, Fig. 7 shows the results of the tail tendency of the threshold voltage distribution over time for analyzing the simulation expansion and charge loss model results in units of pages, which are units of program / read.

[0049] Additionally, the threshold voltage prediction program can analyze error factors, including bit error rate and expected lifespan of the device, based on the threshold voltage change of each device predicted through simulation results.

[0050] Figure 8 is a graph showing the bit error rate, and Figure 9 is a graph showing the expected lifespan. The threshold voltage prediction program can calculate the bit error rate as shown in Figure 8 and output it as a graph, and can calculate the expected lifespan for multiple elements as shown in Figure 9 and output it visually.

[0051] Referring back to FIG. 2, the communication module (130) may include a device including hardware and software required to transmit and receive signals, such as control signals or data signals, via wired or wireless connections with other network devices in order to perform data communication with external devices and signal data. The database (140) may store various data for the operation of the threshold voltage prediction program. For example, data required for the operation of the threshold voltage prediction program, such as information on a plurality of physical mechanisms, a compact model, and a module for performing Monte Carlo simulations, may be stored.

[0052] FIG. 10 is a flowchart illustrating a threshold voltage prediction method according to an embodiment of the present invention.

[0053] Referring to FIGS. 2 and 10, a threshold voltage prediction method (S100) using a threshold voltage prediction device (100) according to an embodiment of the present invention will be described. The threshold voltage prediction device (100) receives the characteristics of a device including the arrangement structure of a plurality of devices and initial operating parameters to generate a virtual non-volatile memory (step S110), and sets variable parameters that determine the degree of charge loss for each of a plurality of compact models based on a charge loss model and applies the set parameters to each device (step S120). Thereafter, the threshold voltage prediction device (100) performs a Monte-Carlo simulation for the virtual non-volatile memory based on the plurality of compact models applied to each device and the physical mechanism causing the charge loss, and outputs the threshold voltage change of each device predicted based on the simulation results (step S140).

[0054] A compact model based on the charge loss model can be defined through mathematical equation 2.

[0055]

[0056] Mathematical expression 2 is substantially identical to Mathematical expression 1 above. In Mathematical expression 2, is the change in threshold voltage due to charge loss by the physical mechanism k ( ) represents, represents the time constant, represents the shape parameter of the charge loss model, represents the charge loss time. Here, the variable parameter is , and There can be one or more of them. Here, each variable parameter can have different weights depending on the physical mechanism k.

[0057] Additionally, the physical mechanism may be one or more of detrapping, trap assisted tunneling (TAT), interface trap recovery, and lateral migration.

[0058] Next, a process (step S130) in which a threshold voltage prediction device (100) performs a simulation for a virtual non-volatile memory is described.

[0059] A threshold voltage prediction device (100) models a virtual non-volatile memory based on a layout structure of a plurality of elements, generates a threshold voltage distribution for erasing before the virtual non-volatile memory is programmed based on the characteristics and initial operating parameters of the elements, and sets initial operating parameters for each element by randomly matching the thickness of each layer of the structure according to process variables.

[0060] Thereafter, the threshold voltage prediction device (100) determines the simulation operation order of a plurality of compact models in the same order as the operation of an actual memory device, performs Monte Carlo simulation for each compact model based on the variable parameters and physical mechanisms of each compact model, and can predict the threshold voltage change of the virtual non-volatile memory based on the simulation results.

[0061] Fig. 4 is a graph exemplarily illustrating threshold voltage changes according to changes in variable parameters of a compact model. Fig. 4 shows that threshold voltage changes according to changes in each variable parameter for any one of multiple physical mechanisms causing charge loss. As shown in Fig. 4, the compact model can predict threshold voltage changes according to changes in each variable parameter for each physical mechanism, and comprehensively reflect this to predict threshold voltage changes for multiple physical mechanisms.

[0062] Figure 5 shows the simulation results of a compact model for each physical mechanism according to temperature changes, the combined simulation results, and actual measured values. It can be confirmed that the threshold voltage change of the virtual non-volatile memory predicted by the threshold voltage prediction program through Monte Carlo simulation matches the actual measured threshold voltage change.

[0063] Next, the process of outputting the predicted threshold voltage change of each element based on the simulation results (step S140) is described.

[0064] The threshold voltage prediction device (100) can visually output simulation results as shown in FIGS. 6 and 7. FIG. 6 is a graph showing a threshold voltage change of one of a plurality of devices, and a graph like FIG. 6 can be output for each device. In addition, FIG. 7 shows the tail trend results of the threshold voltage distribution over time for analyzing the simulation expansion into a page unit, which is a program / read unit, and the charge loss model results.

[0065] Additionally, in the process of outputting the threshold voltage change of each element (step S140), the threshold voltage prediction device (100) can analyze error factors including the bit error rate and the expected lifespan of the elements based on the threshold voltage change of each element predicted through the simulation results. The threshold voltage prediction device (100) can calculate the bit error rate and output it as a graph, as in FIG. 8, and can calculate the expected lifespan for a plurality of elements and output it visually, as in FIG. 9.

[0066] The present invention may also be implemented in the form of a non-transitory storage medium containing computer-executable instructions, such as program modules executed by a computer. Computer-readable media may be any available media that can be accessed by a computer, and include both volatile and nonvolatile media, removable and non-removable media. Computer-readable media may also include computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data.

[0067] Additionally, although the methods and systems of the present invention have been described with respect to specific embodiments, some or all of their components or operations may be implemented using a computer system having a general-purpose hardware architecture.

[0068] Those skilled in the art will appreciate that the present invention can be readily modified into other specific forms without altering the technical spirit or essential characteristics of the present invention based on the above description. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. The scope of the present invention is defined by the following claims, and all changes or modifications derived from the meaning and scope of the claims and their equivalents should be construed as being included within the scope of the present invention.

[0069] The scope of the present invention is indicated by the claims described below rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention.

[0070] The form for carrying out the invention is substantially the same as the best form for carrying out the invention.

[0071] The present invention can be applied to a technology for analyzing charge loss factors of NAND Flash, and thus has industrial applicability.

Claims

1. In a threshold voltage prediction device for predicting a threshold voltage change of a non-volatile memory device, A memory storing a program that provides a simulation of the operation of a non-volatile memory device; and Including a processor for executing a program stored in the above memory, The above program is, A threshold voltage prediction device that receives characteristics of a device including a layout structure and initial operating parameters of a plurality of devices to generate a virtual non-volatile memory, sets variable parameters that determine the degree of charge loss for each of a plurality of compact models based on a charge loss model and applies the parameters to each device, performs a Monte-Carlo simulation on the virtual non-volatile memory based on the plurality of compact models applied to each device and a physical mechanism causing charge loss, and outputs a threshold voltage change of each device predicted based on the simulation results.

2. In paragraph 1, The above compact model is, It is defined by mathematical formula 1, [Mathematical Formula 1] Above is the change in threshold voltage due to charge loss by the physical mechanism k ( ) represents, represents the time constant, represents the shape parameter of the charge loss model, A threshold voltage prediction device, which represents the charge loss time.

3. In paragraph 2, The above variable parameters are, Above , above and above A threshold voltage prediction device, which is one or more of:

4. In paragraph 1, The above physical mechanism is, A threshold voltage prediction device, wherein at least one of detrapping, trap assisted tunneling (TAT), interface trap recovery, and lateral migration is performed.

5. In paragraph 1, The above program is, A threshold voltage prediction device that analyzes error factors, including bit error rate and expected lifespan of a device, based on threshold voltage changes of each device predicted through simulation results.

6. A method for predicting a threshold voltage change of a non-volatile memory device using a threshold voltage prediction device, A step of generating a virtual non-volatile memory by receiving characteristics of a device including a layout structure of a plurality of devices and initial operating parameters; A step of setting variable parameters that determine the degree of charge loss for each of a plurality of compact models based on a charge loss model and applying them to each element; A step of performing a Monte-Carlo simulation for the virtual non-volatile memory based on the plurality of compact models applied to each element and the physical mechanism causing charge loss; and A threshold voltage prediction method, comprising a step of predicting a threshold voltage change of each element based on simulation results. do, 7. In paragraph 6, The above compact model is, It is defined by mathematical formula 1, [Mathematical Formula 1] Above is the change in threshold voltage due to charge loss by the physical mechanism k ( ) represents, represents the time constant, represents the shape parameter of the charge loss model, A threshold voltage prediction method, which represents the charge loss time.

8. In paragraph 7, The above variable parameters are, Above , above and above A method for predicting threshold voltage, wherein one or more of the following are present.

9. In paragraph 6, The above physical mechanism is, A threshold voltage prediction method, wherein at least one of detrapping, trap assisted tunneling (TAT), interface trap recovery, and lateral migration is used.

10. In paragraph 6, The step of predicting the above threshold voltage change is: A threshold voltage prediction method, which analyzes error factors including bit error rate and expected lifespan of a device based on threshold voltage changes of each device predicted through simulation results.

11. A computer-readable storage medium having recorded thereon a program for performing a threshold voltage prediction method according to any one of claims 6 to 9.

Images