TW201506930A - Method and apparatus for adjusting drain bias of a memory cell with addressed and neighbor bits - Google Patents

Method and apparatus for adjusting drain bias of a memory cell with addressed and neighbor bits Download PDF

Info

Publication number
TW201506930A
TW201506930A TW103136995A TW103136995A TW201506930A TW 201506930 A TW201506930 A TW 201506930A TW 103136995 A TW103136995 A TW 103136995A TW 103136995 A TW103136995 A TW 103136995A TW 201506930 A TW201506930 A TW 201506930A
Authority
TW
Taiwan
Prior art keywords
data
bit
stylized
threshold voltage
current load
Prior art date
Application number
TW103136995A
Other languages
Chinese (zh)
Other versions
TWI541804B (en
Inventor
Han-Sung Chen
Chung-Kuang Chen
Chun-Hsiung Hung
Original Assignee
Macronix Int Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Macronix Int Co Ltd filed Critical Macronix Int Co Ltd
Priority to TW103136995A priority Critical patent/TWI541804B/en
Publication of TW201506930A publication Critical patent/TW201506930A/en
Application granted granted Critical
Publication of TWI541804B publication Critical patent/TWI541804B/en

Links

Landscapes

  • Read Only Memory (AREA)

Abstract

The storage layer such as a nitride layer of a nonvolatile memory cell has two storage parts storing separately addressable data, typically respectively proximate to the source terminal and the drain terminal. The applied drain voltage while sensing the data of one of the storage parts depends on data stored at the other storage part. If the data stored at the other storage part is represented by a threshold voltage exceeding a minimum threshold voltage, then the applied drain voltage is raised. This technology is useful in read operations and program verify operations to widen the threshold voltage window.

Description

具有定址及相鄰位元之記憶胞的汲極偏壓調整方法與裝置 Method and device for adjusting threshold of memory of cell with address and adjacent bit

本發明係關於非揮發記憶體,特別是關於具有定址及相鄰位元之記憶胞的汲極偏壓調整方法與裝置。 The present invention relates to non-volatile memory, and more particularly to a method and apparatus for adjusting the bias voltage of a memory cell having addressed and adjacent bits.

例如氮化矽唯讀記憶體(NROM)的電荷捕捉記憶胞是可以藉由例如是通道熱電子注入(CHE)的機制被程式化至此氮化矽儲存層中的不同區域部分內。一單一記憶胞可以分別儲存不同地址的資料於此氮化矽儲存層中靠近源極及靠近汲極的不同部分內。記憶胞的臨界電壓範圍是與可以儲存於此記憶胞中每一部分的不同資料值相關。舉例而言,在多階記憶胞的每一部分中,四個臨界電壓範圍可以代表儲存兩個位元的四個不同資料值。在一個三個階級記憶胞中,八個臨界電壓範圍可以代表儲存三個位元中的八個不同資料值。 For example, a charge trapping memory cell of a tantalum nitride read only memory (NROM) can be programmed into portions of different regions of the tantalum nitride storage layer by a mechanism such as channel hot electron injection (CHE). A single memory cell can store data of different addresses in different portions of the tantalum nitride storage layer near the source and near the drain. The threshold voltage range of the memory cell is related to different data values that can be stored in each part of the memory cell. For example, in each portion of a multi-level memory cell, the four threshold voltage ranges can represent four different data values that store two bits. In a three-class memory cell, eight threshold voltage ranges can represent eight different data values stored in three bits.

然而,因為儲存在氮化矽儲存層中的不同部分內之資料會彼此影響之第二位元效應,而限制了一記憶胞可用的臨界電壓區間。這些不同部分在此處稱為相同記憶胞的”定址位元”及”鄰近位元”。在此說明書中,”位元”並不限於儲存單一資料位元的資料,而是指可以儲存1、2、3或更多位元資料的電荷儲存層的不同實體位置。此定址位元是指在一例如是程式化或讀取的命令中可以被定址的實體資料位置,而鄰近位元則是指在一相同記憶胞中與定址位元鄰近的實體資料位置。 However, because the data stored in different portions of the tantalum nitride storage layer affect each other's second bit effect, it limits the threshold voltage interval available for a memory cell. These different portions are referred to herein as "addressed bits" and "adjacent bits" of the same memory cell. In this specification, a "bit" is not limited to storing a single data bit, but refers to a different physical location of a charge storage layer that can store 1, 2, 3 or more bits of data. The addressed bit refers to the location of the physical data that can be addressed in a command such as a stylized or read command, and the adjacent bit refers to the location of the physical data adjacent to the addressed bit in the same memory cell.

美國專利US6011725描述執行一反向讀取操作而讀取此記憶胞的 一定址部分,在此引為參考資料,在其中施加至源極和汲極的電壓極性是與在通道熱電子程式化操作時所施加至此記憶胞之定址位元的電壓極性相反。於此記憶胞之定址位元所執行的反向讀取操作必須”擊穿”位於相同記憶胞的鄰近位元底下的通道。當此記憶胞的鄰近位元所儲存的資料值與一高臨界電壓相關,則此反向讀取電流減少。在Lue Hang-Ting等人的論文”Studies of the reverse read method and second-bit effect of 2 bit/cell nitride-trapping device by quesi-two-dimensional model”,IEEE Transactions on Electron Devices,Vol.43,No.1,page 119,January 2006中描述如此之第二位元效應會縮小可用的臨界電壓區間,在此亦引為參考資料。 U.S. Patent No. 6,011,725 describes the implementation of a reverse read operation to read this memory cell. The address portion is hereby incorporated by reference, in which the polarity of the voltage applied to the source and the drain is opposite to the polarity of the voltage applied to the address bit of the memory cell during the hot metal stylization operation of the channel. The reverse read operation performed by the address bit of the memory cell must "break down" the channel underneath the adjacent bit of the same memory cell. When the data value stored by the neighboring cell of the memory cell is related to a high threshold voltage, the reverse read current is reduced. "Lyue Hang-Ting et al." "Studies of the reverse read method and second-bit effect of 2 bit/cell nitride-trapping device by quesi-two-dimensional model", IEEE Transactions on Electron Devices, Vol.43, No .1, page 119, January 2006 describes such a second bit effect that narrows the available threshold voltage range and is also cited here.

此第二位元效應可以藉由在此反向讀取操作時增加所施加的電壓大小來解決。然而,如此增加電壓大小又會遇到與在讀取定址位元時會錯誤地程式化鄰近位元的讀取干擾效應問題。 This second bit effect can be resolved by increasing the magnitude of the applied voltage during this reverse read operation. However, increasing the voltage level in this way encounters the problem of read disturb effects that would erroneously program adjacent bits when reading the addressed bits.

因此,希望能夠藉由改善第二位元效應來提供較寬的臨界電壓區間,而同時減少所伴隨之例如讀取干擾程式化等缺點。 Therefore, it is desirable to be able to provide a wider threshold voltage interval by improving the second bit effect while reducing the disadvantages associated with, for example, read disturb stylization.

此處所描述之技術係關於一記憶電路其具有一非揮發記憶胞及一控制電路。此非揮發記憶胞包括一第一電流負載終端(源極或汲極)、一第二電流負載終端(汲極或源極)及一閘極。此非揮發記憶胞具有多個儲存部分其可以分別儲存資料。一第一儲存部分鄰近該第一電流負載終端並且儲存第一資料。一第二儲存部分鄰近該第二電流負載終端並且儲存第二資料。該第一儲存部分及該第二儲存部分可以是一氮化矽儲存層中的不同部分。 The technique described herein pertains to a memory circuit having a non-volatile memory cell and a control circuit. The non-volatile memory cell includes a first current load terminal (source or drain), a second current load terminal (drain or source), and a gate. The non-volatile memory cell has a plurality of storage portions that can store data separately. A first storage portion is adjacent to the first current load terminal and stores the first data. A second storage portion is adjacent to the second current load terminal and stores the second data. The first storage portion and the second storage portion may be different portions of a tantalum nitride storage layer.

此控制電路施加一讀取編排偏壓至該非揮發記憶胞的該第一電流負載終端,該第二電流負載終端及該閘極,所施加的該讀取編排偏壓係讀取該第一資料及該第二資料的其中之一者,該讀取編排偏壓根據該第一資料及該第二資料的其中之另一者。舉例而言,該第一資料決定在該讀取編排偏壓中施加至該第一電流負載終端的一第一電壓資料以讀 取該第二資料。在該第二資料所執行的反向讀取操作,電子自該第二電流負載終端流動至該第一電流負載終端。為了減少第二位元效應,藉由增加施加至該第一電流負載終端的電壓來增加反向讀取電流。而為了避免讀取干擾(第一資料不欲之程式化),假如由該第一儲存部分的一第一臨界電壓所代表的該第一資料超過一最小臨界電壓,則增加施加至該第一電流負載終端的電壓。假如由該第一儲存部分的一第一臨界電壓所代表的該第一資料沒有超過一最小臨界電壓,則施加一較小的電壓至該第一電流負載終端。此技術的某些實施例中包括,一記憶體儲存資料位元由代表該第一資料的一臨界電壓是否超過一最小臨界電壓來決定,其中該控制電路自該記憶體讀取該資料位元以控制該程式化驗證編排偏壓中是否施加該第一電壓或該第二電壓至該第一電流負載終端以程式化驗證該第二資料。 The control circuit applies a read current bias to the first current load terminal of the non-volatile memory cell, the second current load terminal and the gate, and the read programming bias voltage is applied to read the first data And one of the second materials, the read programming bias is based on the other of the first data and the second data. For example, the first data determines a first voltage data applied to the first current load terminal in the read programming bias to read Take the second data. In the reverse read operation performed by the second data, electrons flow from the second current load terminal to the first current load terminal. In order to reduce the second bit effect, the reverse read current is increased by increasing the voltage applied to the first current load terminal. In order to avoid reading interference (the first data is not intended to be stylized), if the first data represented by a first threshold voltage of the first storage portion exceeds a minimum threshold voltage, an increase is applied to the first The voltage of the current load terminal. If the first data represented by a first threshold voltage of the first storage portion does not exceed a minimum threshold voltage, a smaller voltage is applied to the first current load terminal. Some embodiments of the technology include: a memory storage data bit is determined by whether a threshold voltage representing the first data exceeds a minimum threshold voltage, wherein the control circuit reads the data bit from the memory Controlling whether the first voltage or the second voltage is applied to the first current load terminal in the stylized verification programming bias to programmatically verify the second data.

類似地,該第二資料決定在該讀取編排偏壓操作中施加至該第二電流負載終端的一電壓以讀取該第一資料。如同對該第一電流負載終端所描述的,為了避免讀取干擾(第一資料不欲之程式化),假如由該第二儲存部分的一第二臨界電壓所代表的該第二資料超過一最小臨界電壓,則增加施加至該第二電流負載終端的電壓。假如由該第二儲存部分的一第二臨界電壓所代表的該第二資料沒有超過一最小臨界電壓,則施加一較小的電壓至該第二電流負載終端。 Similarly, the second data determines a voltage applied to the second current load terminal during the read programming bias operation to read the first data. As described for the first current load terminal, in order to avoid read disturb (the first data is not intended to be stylized), if the second data represented by a second threshold voltage of the second storage portion exceeds one The minimum threshold voltage increases the voltage applied to the second current load terminal. If the second data represented by a second threshold voltage of the second storage portion does not exceed a minimum threshold voltage, a smaller voltage is applied to the second current load terminal.

是否將該第一電流負載終端或是將該第二電流負載終端的電壓增加,此被增加電壓的電流終端稱為汲極電壓終端,因為汲極終端是一較高電壓的電流負載終端,其是電子流的目的地(與電流的方向相反)。此外,是否讀取在第一儲存部分的第一資料,或是讀取在第二儲存部分的第二資料,被讀取的儲存部分稱為定址位元,而沒有被讀取的儲存部分稱為鄰近位元。如同之前所描述的,”位元”在此說明書中是指一記憶胞中的一特定儲存位置且並不限於儲存單一資料位元的資料。每一個定址位元及鄰近位元可以儲存1、2、3或更多位元資料的電荷儲存層的不同實體位置。 Whether to increase the voltage of the first current load terminal or the second current load terminal, the current terminal of the increased voltage is called a drain voltage terminal, because the drain terminal is a higher voltage current load terminal, Is the destination of the electron flow (in the opposite direction of the current). In addition, whether the first data in the first storage portion is read, or the second data in the second storage portion is read, the stored portion that is read is referred to as an addressed bit, and the stored portion that is not read is called It is an adjacent bit. As previously described, a "bit" in this specification refers to a particular storage location in a memory cell and is not limited to storing a single data bit. Each of the addressed bits and adjacent bits can store different physical locations of the charge storage layer of 1, 2, 3 or more bits of data.

當汲極電壓被增加時,因為第二位元效應減少而會造成感測臨界電 壓範圍的偏移。因此,可以有多重的程式化臨界電壓範圍與相同的感測臨界電壓範圍對應。對一定址的位元,儲存在鄰近位元的資料決定特定程式化臨界電壓範圍所對應的感測臨界電壓範圍。此外,程式化臨界電壓範圍的整體數目超過感測臨界電壓範圍的整體數目。 When the drain voltage is increased, the sensing threshold is caused because the second bit effect is reduced. The offset of the pressure range. Therefore, multiple stylized threshold voltage ranges can be associated with the same sense threshold voltage range. For a given bit, the data stored in the adjacent bit determines the sensing threshold voltage range corresponding to the particular programmed threshold voltage range. In addition, the overall number of programmed threshold voltage ranges exceeds the overall number of sensing threshold voltage ranges.

如同之前所描述的,讀取編排偏壓係讀取儲存於記憶胞中該第一資料及該第二資料的其中之一者,且該讀取編排偏壓根據該第一資料及該第二資料的其中之另一者。此技術的某些實施例中,於施加該讀取編排偏壓之前,此資料決定該第一資料及該第二資料的其中之另一者。此技術的某些實施例中,一讀取操作決定該第一資料及該第二資料的其中之另一者。 As described above, the read programming bias reads one of the first data and the second data stored in the memory cell, and the read programming bias is based on the first data and the second The other of the information. In some embodiments of the technique, the data determines the other of the first data and the second data prior to applying the read programming bias. In some embodiments of the technology, a read operation determines the other of the first data and the second data.

此處亦描述此技術的其他實施例。這些實施例是關於將讀取利用程式化驗證取代。 Other embodiments of this technique are also described herein. These embodiments are concerned with replacing the read with stylized verification.

本發明之另一實施例是施加程式化驗證編排偏壓而不是加一讀取編排偏壓。一程式化驗證操作決定一記憶胞中是否進行一最小數量的程式化,其中一讀取操作決定一記憶胞的臨界電壓是否高於或低於與不同資料值相關兩個最小臨界電壓範圍間的一中間臨界電壓。 Another embodiment of the present invention is to apply a stylized verification programming bias instead of adding a read programming bias. A stylized verification operation determines whether a minimum number of stylizations are performed in a memory cell, wherein a read operation determines whether a threshold voltage of a memory cell is higher or lower than a minimum threshold voltage range associated with different data values. An intermediate threshold voltage.

此程式化驗證編排偏壓是施加來程式化驗證該記憶胞所儲存之該第一資料及該第二資料的其中之一者,且該程式化驗證編排偏壓根據該第一資料及該第二資料的其中之另一者。於施加程式化驗證編排偏壓之前,此資料決定該第一資料及該第二資料的其中之另一者。此技術的某些實施例中,一讀取操作決定該第一資料及該第二資料的其中之另一者。此技術的另一些實施例中,一程式化命令的輸入資料決定該第一資料及該第二資料的其中之另一者。 The stylized verification programming bias is applied to programmatically verify one of the first data and the second data stored by the memory cell, and the stylized verification programming bias is based on the first data and the first The other of the two materials. This data determines the other of the first data and the second data prior to applying the stylized verification programming bias. In some embodiments of the technology, a read operation determines the other of the first data and the second data. In still other embodiments of the technology, the input data of a stylized command determines the other of the first data and the second data.

此處亦描述此技術的其他實施例。這些實施例是關於將讀取利用程式化驗證取代。 Other embodiments of this technique are also described herein. These embodiments are concerned with replacing the read with stylized verification.

本發明之另一實施例是關於一種記憶方法,包括:施加一讀取編排偏壓至一非揮發記憶胞的一第一電流負載終端、一第二電流負載終端及一閘極,所施加的該讀取編排偏壓係讀取第一資 料及第二資料的其中之一者,該讀取編排偏壓根據該第一資料及該第二資料的其中之另一者,該第一資料儲存於一第一儲存部分鄰近該第一電流負載終端及該第二資料儲存於一第二儲存部分鄰近該第二電流負載終端。 Another embodiment of the present invention is directed to a memory method comprising: applying a read current bias to a first current load terminal of a non-volatile memory cell, a second current load terminal, and a gate applied The reading and programming biasing system reads the first capital And reading the programming bias according to the other of the first data and the second data, the first data being stored in a first storage portion adjacent to the first current load The terminal and the second data are stored in a second storage portion adjacent to the second current load terminal.

此處亦描述此技術的其他實施例。這些實施例是關於將讀取利用程式化驗證取代。 Other embodiments of this technique are also described herein. These embodiments are concerned with replacing the read with stylized verification.

本發明之另一實施例是關於一種記憶方法,包括:施加一程式化驗證編排偏壓至一非揮發記憶胞的一第一電流負載終端、一第二電流負載終端及一閘極,所施加的該程式化驗證編排偏壓係讀取第一資料及第二資料的其中之一者,該讀取編排偏壓根據該第一資料及該第二資料的其中之另一者,該第一資料儲存於一第一儲存部分鄰近該第一電流負載終端及該第二資料儲存於一第二儲存部分鄰近該第二電流負載終端。 Another embodiment of the present invention is directed to a memory method comprising: applying a stylized verification programming bias to a first current load terminal, a second current load terminal, and a gate of a non-volatile memory cell, The stylized verification programming biasing system reads one of the first data and the second data, and the reading programming bias is based on the other of the first data and the second data, the first The data is stored in a first storage portion adjacent to the first current load terminal and the second data is stored in a second storage portion adjacent to the second current load terminal.

此處亦描述此技術的其他實施例。這些實施例是關於將讀取利用程式化驗證取代。 Other embodiments of this technique are also described herein. These embodiments are concerned with replacing the read with stylized verification.

3450‧‧‧積體電路 3450‧‧‧Integrated circuit

3400‧‧‧非揮發記憶胞陣列 3400‧‧‧Non-volatile memory cell array

3401‧‧‧列解碼器 3401‧‧‧ column decoder

3402‧‧‧字元線 3402‧‧‧ character line

3403‧‧‧行解碼器 3403‧‧ ‧ row decoder

3404‧‧‧位元線 3404‧‧‧ bit line

3405‧‧‧匯流排 3405‧‧‧ Busbar

3407‧‧‧資料匯流排 3407‧‧‧ data bus

3406‧‧‧感測放大器/資料輸入結構/汲極線電路 3406‧‧‧Sense Amplifier/Data Input Structure/Dension Line Circuit

3409‧‧‧程式化驗證及讀取編排偏壓狀態機構根據鄰近位元決定汲極電壓及程式化VT 3409‧‧‧Stylized verification and read orchestration bias state mechanism determines the bucker voltage and stylized VT based on adjacent bits

3408‧‧‧偏壓調整供應電壓 3408‧‧‧ bias adjustment supply voltage

3411‧‧‧資料輸入線 3411‧‧‧ data input line

3415‧‧‧資料輸出線 3415‧‧‧ data output line

第1~2圖一個非揮發記憶胞的示意圖,其具有分別儲存資料的定址位元及鄰近位元。 Figures 1 to 2 are schematic diagrams of a non-volatile memory cell having address bits and neighboring bits for storing data, respectively.

第3~5圖是臨界電壓的圖示,顯示一"定址位元"之"鄰近位元"所造成之第二位元效應。 Figures 3 through 5 are graphical representations of the threshold voltage, showing the second bit effect caused by the "adjacent bits" of an "addressed bit".

第6~7圖是臨界電壓的圖示,顯示增加汲極電壓其可以減少一"定址位元"之"鄰近位元"所造成之第二位元效應。 Figures 6-7 are graphical representations of the threshold voltage, showing that increasing the gate voltage can reduce the second bit effect caused by the "adjacent bit" of an "addressed bit".

第8~23圖是一"定址位元"和一"鄰近位元"的臨界電壓圖示,顯示不同的"定址位元"和"鄰近位元"的資料排列,及許多種資料排列中的第二位元效應的減少。 Figures 8~23 are plots of threshold voltages for an "addressed bit" and an "adjacent bit", showing the arrangement of different "addressed bits" and "adjacent bits", and in many types of data. The second bit effect is reduced.

第24和25圖是臨界電壓的圖示,顯示不同的"定址位元" 和"鄰近位元"的資料排列之程式化臨界電壓範圍與感測臨界電壓範圍。 Figures 24 and 25 are graphical representations of threshold voltages showing different "addressed bits" Stylized threshold voltage range and sensing threshold voltage range for data arrangement with "adjacent bits".

第26圖顯示不同汲極偏壓的臨界電壓分佈,其根據一鄰近位元中的資料值會調整會不調整此汲極偏壓。 Figure 26 shows the threshold voltage distribution for different buckling biases, which will adjust the buckling bias based on the value of the data in an adjacent bit.

第27圖是汲極線驅動電路區塊的方塊示意圖,其具有由儲存在一靜態隨機存取記憶體SRAM中的資料所決定的汲極電壓。 Figure 27 is a block diagram of a drain line driver circuit block having a drain voltage determined by data stored in a static random access memory (SRAM).

第28圖是通常靜態隨機存取記憶體SRAM區塊,其自一汲極線驅動電路區塊產生信號以控制汲極電壓。 Figure 28 is a typical static random access memory SRAM block that generates a signal from a drain line driver circuit block to control the drain voltage.

第29圖是一汲極線驅動電路的一範例電路圖。 Figure 29 is a circuit diagram showing an example of a drain line driving circuit.

第30圖為根據鄰近位元的值而具有調整汲極偏壓之一讀取操作的流程圖。 Figure 30 is a flow chart showing one of the read operations of adjusting the gate bias based on the value of the adjacent bit.

第31~33圖為根據鄰近位元的值而具有調整汲極偏壓之程式化操作的流程圖。 Figures 31-33 are flow diagrams of the stylized operation of adjusting the gate bias based on the values of adjacent bits.

第34圖顯示根據本發明一實施例具有此處所描述之改良及記憶陣列的積體電路的簡化方塊示意圖。 Figure 34 shows a simplified block diagram of an integrated circuit having the improved and memory arrays described herein in accordance with an embodiment of the present invention.

此處所示的範例為一記憶胞中具有四個可能資料值。其他的範例可以是具有兩個可能資料值、具有八個可能資料值或是具有其他數目的可能資料值。 The example shown here has four possible data values in a memory cell. Other examples may be having two possible data values, having eight possible data values, or having other numbers of possible data values.

第1~2圖一個非揮發記憶胞的示意圖,其具有分別儲存資料的定址位元及鄰近位元。 Figures 1 to 2 are schematic diagrams of a non-volatile memory cell having address bits and neighboring bits for storing data, respectively.

第1圖顯示一非揮發記憶胞具有資料儲存於氮化矽儲存層的不同部分之中。此處將同一記憶胞中的這些不同部分稱為"定址位元"及"鄰近位元"。在此描述中,"位元"一詞是指電荷儲存層中不同的實體位置其可以儲存1、2、3或是更多位元的資 料。此記憶胞具有一閘極終端,兩個電流負載終端-汲極終端和源極終端。汲極終端電壓會根據靠近源極終端的資料值而變動。因為此特定電流負載終端是作為汲入電流的終端會變動,而造成是汲極終端的此特定電流負載終端也會隨著變動。在第1圖中,定址位元是在此記憶胞電荷儲存層的左側部分且靠近左邊的電流負載終端,而鄰近位元是在此記憶胞電荷儲存層的右側部分且靠近右邊的電流負載終端。此右邊和左邊的電流負載終端設定為較低電壓的源極和較高電壓的汲極,且電子流的方向是流動至汲極,顯示不同"定址位元"及"鄰近位元"的程式化和反向讀取操作中的情況。 Figure 1 shows a non-volatile memory cell with data stored in different parts of the tantalum nitride storage layer. These different parts of the same memory cell are referred to herein as "addressed bits" and "adjacent bits." In this description, the term "bit" refers to a different physical location in the charge storage layer that can store 1, 2, 3 or more bits. material. This memory cell has a gate terminal, two current load terminals - a drain terminal and a source terminal. The bucker terminal voltage will vary depending on the value of the data near the source terminal. Because this particular current load terminal will change as the terminal of the inrush current, the specific current load terminal of the bungee terminal will also change. In Figure 1, the address bit is the current load terminal on the left side of the memory cell charge storage layer and close to the left side, and the adjacent bit cell is the current load terminal on the right side of the memory cell charge storage layer and close to the right side. . The right and left current load terminals are set to a lower voltage source and a higher voltage drain, and the direction of the electron flow is to the drain, showing different "addressed bits" and "adjacent bits" And reverse read operations.

第2圖顯示一非揮發記憶胞具有資料儲存於氮化矽儲存層的不同部分之中,其類似於第1圖。但是與第1圖不同的是,定址位元是在此記憶胞電荷儲存層的右側部分且靠近右邊的電流負載終端,而鄰近位元是在此記憶胞電荷儲存層的左側部分且靠近左邊的電流負載終端。因為在第2圖中"定址位元"和"鄰近位元"的位置對調,其源極和汲極的位置以及電子流的方向也是和第1圖是相反的。 Figure 2 shows a non-volatile memory cell with data stored in different parts of the tantalum nitride storage layer, similar to Figure 1. However, unlike FIG. 1, the address bit is the current load terminal on the right side of the memory cell charge storage layer and close to the right side, and the adjacent bit cell is on the left side of the memory cell charge storage layer and close to the left side. Current load terminal. Since the positions of the "addressed bit" and the "adjacent bit" are reversed in Fig. 2, the positions of the source and the drain and the direction of the electron flow are also opposite to those of Fig. 1.

第3~5圖是臨界電壓的圖示,顯示一"定址位元"之"鄰近位元"所造成之第二位元效應。當感測此定址位元時一個低汲極電壓VBLR_1施加至汲極終端。此定址位元的感測臨界電壓(VT)分佈及此定址位元在此記憶胞中的實際位置顯示於圖中的實線。此鄰近位元的感測VT分佈及此鄰近位元在此記憶胞中的實際位置顯示於圖中的虛線。 Figures 3 through 5 are graphical representations of the threshold voltage, showing the second bit effect caused by the "adjacent bits" of an "addressed bit". A low drain voltage VBLR_1 is applied to the drain terminal when sensing the addressed bit. The sensed threshold voltage (VT) distribution of the addressed bit and the actual location of the addressed bit in the memory cell are shown in the solid line in the figure. The sensed VT distribution of this neighboring bit and the actual position of this neighboring cell in this memory cell are shown in the dashed line in the figure.

在第3圖中,定址位元具有一初始感測VT分佈。所附的圖式中顯示定址位元是在此記憶胞之電荷儲存層的左側部分。鄰近位元尚未被程式化至一高VT分佈,所以並未有第二位元效應。 In Figure 3, the addressed bit has an initial sensed VT distribution. The attached bitmap shows that the address bit is the left portion of the charge storage layer of the memory cell. Neighboring bits have not been programmed to a high VT distribution, so there is no second bit effect.

在第4圖中,鄰近位元被程式化至一高VT分佈。所附的圖 式中顯示鄰近位元是在此記憶胞之電荷儲存層的右側部分。 In Figure 4, the neighboring bits are programmed to a high VT distribution. Attached figure The adjacent bit is shown in the right part of the charge storage layer of the memory cell.

在第5圖中,定址位元自此初始感測VT分佈偏移至一個較高的感測VT分佈。此"感測VT分佈"的偏移並未反應在"程式化VT分佈"中的改變,因為並沒有電荷被程式化至此定址位元。而是,定址位元的感測VT分佈偏移是由第二位元效應所導致。定址位元的感測VT分佈之更高偏移,是因為鄰近位元之下的通道部分具有較高的臨界電壓,而減少感測電流及增加定址位元的感測VT分佈所致。 In Figure 5, the addressed bit is offset from this initial sensed VT distribution to a higher sensed VT distribution. The offset of this "sensing VT distribution" does not reflect the change in the "stylized VT distribution" because no charge is programmed into this addressed bit. Rather, the sensed VT distribution offset of the addressed bit is caused by the second bit effect. The higher offset of the sensed VT distribution of the addressed bit is due to the higher threshold voltage of the channel portion below the adjacent bit, which reduces the sense current and increases the sensed VT distribution of the addressed bit.

第6~7圖是臨界電壓的圖示,顯示增加汲極電壓其可以減少一"定址位元"之"鄰近位元"所造成之第二位元效應。此定址位元的感測VT分佈顯示於圖中的實線。此鄰近位元的感測VT分佈顯示於圖中的虛線。 Figures 6-7 are graphical representations of the threshold voltage, showing that increasing the gate voltage can reduce the second bit effect caused by the "adjacent bit" of an "addressed bit". The sensed VT distribution of this addressed bit is shown in the solid line in the figure. The sensed VT distribution of this neighboring bit is shown in the dashed line in the figure.

在第6圖中,鄰近位元的值被感測,以決定感測此定址位元的汲極電壓。施加鄰近位元的字元線讀取電壓以感測此鄰近位元。假如此鄰近位元的感測VT超過字元線讀取電壓的臨界電壓VT,則其結果是資料0,Nb_h(鄰近位元高準位)=1,Nb_l(鄰近位元低準位)=0。假如此鄰近位元的感測VT沒有超過字元線讀取電壓的臨界電壓VT,則其結果是資料1,Nb_h=0,Nb_l=1。在此範例中,鄰近位元的感測VT沒有超過字元線讀取電壓的臨界電壓VT,則其結果是資料1,Nb_h=0,Nb_l=1。 In Figure 6, the value of the adjacent bit is sensed to determine the sense of the drain voltage of the addressed bit. A word line read voltage is applied to adjacent bits to sense this neighboring bit. If the sensing VT of the neighboring bit exceeds the threshold voltage VT of the word line reading voltage, the result is data 0, Nb_h (adjacent bit high level) = 1, Nb_l (near level low level) = 0. If the sense VT of the adjacent bit does not exceed the threshold voltage VT of the word line read voltage, the result is data 1, Nb_h=0, Nb_l=1. In this example, the sense VT of the adjacent bit does not exceed the threshold voltage VT of the word line read voltage, and the result is data 1, Nb_h = 0, Nb_l = 1.

在第7圖中,因為第6圖的結果,施加高汲極電壓VBLR_h以感測此定址位元。此高汲極電壓傾向會增加感測電流,而且傾向反制第二位元效應。因為第二位元效應的減少,此定址位元的VT分佈偏移亦減少。此VT分佈偏移的減少使得可用的VT區間變寬。 In Fig. 7, because of the result of Fig. 6, a high drain voltage VBLR_h is applied to sense the addressed bit. This high buckling voltage tends to increase the sense current and tends to counteract the second bit effect. Because of the reduction in the second bit effect, the VT distribution offset of this addressed bit is also reduced. This reduction in VT distribution offset widens the available VT intervals.

第8~23圖是一"定址位元"和一"鄰近位元"的臨界電壓圖示,顯示不同的"定址位元"和"鄰近位元"的資料排列,及許多種資料排列中的第二位元效應的減少。分別在定址位元資料值和鄰 近位元資料值上方的"H"或"L"指示在感測一特定位元時是施加高汲極電壓VBLR_h或是低汲極電壓VBLR_l。假如此鄰近位元的感測VT沒有超過最小臨界值VGN,則施加低汲極電壓VBLR_l來感測定址位元資料值。假如此鄰近位元的感測VT超過最小臨界值VGN,則施加高汲極電壓VBLR_h來感測定址位元資料值。假如此定址位元的感測VT沒有超過最小臨界值VGN,則施加低汲極電壓VBLR_l來感測鄰近位元資料值。假如此定址位元的感測VT超過最小臨界值VGN,則施加高汲極電壓VBLR_h來感測鄰近位元資料值。圖中所示的最小臨界值是介於資料值"3"和"4"的程式化臨界分佈之間。其他的實施例可以施加其他的最小臨界值。 Figures 8~23 are plots of threshold voltages for an "addressed bit" and an "adjacent bit", showing the arrangement of different "addressed bits" and "adjacent bits", and in many types of data. The second bit effect is reduced. Separate bit data values and neighbors The "H" or "L" above the near-bit data value indicates whether a high-thaw voltage VBLR_h or a low-thaw voltage VBLR_1 is applied when sensing a particular bit. If the sensed VT of such neighboring bits does not exceed the minimum threshold value VGN, a low drain voltage VBLR_1 is applied to sense the address bit data value. If the sensed VT of such neighboring bits exceeds the minimum threshold value VGN, a high drain voltage VBLR_h is applied to sense the address bit data value. If the sensed VT of the addressed bit does not exceed the minimum threshold VGN, a low drain voltage VBLR_1 is applied to sense the adjacent bit data value. If the sensed VT of the addressed bit exceeds the minimum threshold VGN, a high drain voltage VBLR_h is applied to sense the adjacent bit data value. The minimum threshold shown in the figure is between the stylized critical distributions of the data values "3" and "4". Other embodiments may apply other minimum thresholds.

第8~11圖是當定址位元的資料值是"1"時(自最低到最高VT值1到4中)的臨界電壓的圖示。在第8圖中,鄰近位元具有資料值"1"。在第9圖中,鄰近位元具有資料值"2"。在第10圖中,鄰近位元具有資料值"3"。在第11圖中,鄰近位元具有資料值"4"。因為鄰近位元的感測VT超過最小臨界值VGN,所以施加高汲極電壓VBLR_h來感測定址位元資料值。因為減少的VT"0A"此定址位元的程式化VT"1"被感測到。雖然顯示不同的程式化VT分佈,此"1"及"0A"的VT可以被認為是相同的,因為兩者皆是小於用來感測最低VT資料值和任何較高VT資料值之間差值的VG1臨界值。 Figures 8-11 are graphical representations of the threshold voltages when the data value of the addressed bit is "1" (from the lowest to the highest VT value of 1 to 4). In Fig. 8, the adjacent bit has a data value of "1". In Fig. 9, the adjacent bit has a data value of "2". In Fig. 10, the adjacent bit has a data value of "3". In Fig. 11, the adjacent bit has a data value of "4". Since the sense VT of the adjacent bit exceeds the minimum threshold VGN, the high drain voltage VBLR_h is applied to sense the address bit data value. Because of the reduced VT "0A", the stylized VT "1" of this addressed bit is sensed. Although different stylized VT distributions are displayed, the VTs of "1" and "0A" can be considered identical because both are less than the difference between the lowest VT data value and any higher VT data values. The value of the VG1 threshold.

第12~15圖是當定址位元的資料值是"2"時(自最低到最高VT值1到4中)的臨界電壓的圖示。在第12圖中,鄰近位元具有資料值"1"。在第13圖中,鄰近位元具有資料值"2"。在第14圖中,鄰近位元具有資料值"3"。在第15圖中,鄰近位元具有資料值"4"。因為鄰近位元的感測VT超過最小臨界值VGN,所以施加高汲極電壓VBLR_h來感測定址位元資料值。此定址位元的程式化VT"2A"被感測到為減少的VT"2"。此定址位元 被程式化至VT"2A"使得感測到VT是"2"。 Figures 12 to 15 are graphs showing the threshold voltages when the data value of the addressed bit is "2" (from the lowest to the highest VT value of 1 to 4). In Fig. 12, the adjacent bit has a data value of "1". In Fig. 13, the adjacent bit has a data value of "2". In Fig. 14, the adjacent bit has a data value of "3". In Fig. 15, the adjacent bit has a data value of "4". Since the sense VT of the adjacent bit exceeds the minimum threshold VGN, the high drain voltage VBLR_h is applied to sense the address bit data value. The stylized VT "2A" of this addressed bit is sensed as a reduced VT "2". Addressing bit Stylized to VT "2A" makes it sense that VT is "2".

第16~19圖是當定址位元的資料值是"3"時(自最低到最高VT值1到4中)的臨界電壓的圖示。在第16圖中,鄰近位元具有資料值"1"。在第17圖中,鄰近位元具有資料值"2"。在第18圖中,鄰近位元具有資料值"3"。在第19圖中,鄰近位元具有資料值"4"。因為鄰近位元的感測VT超過最小臨界值VGN,所以施加高汲極電壓VBLR_h來感測定址位元資料值。此定址位元的程式化VT"3A"被感測到為減少的VT"3"。此定址位元被程式化至VT"3A"使得感測到VT是"3"。因為此定址位元"3A"也是超過最小臨界值VGN,鄰近位元也是感測到具有高汲極電壓VBLR_h。此鄰近位元VT被程式化至VT"4A"使得鄰近位元VT被感測到為減少的VT"4"。否則,假如鄰近位元VT被保持在VT"4"而沒有被程式化至VT"4A"的話,則此鄰近位元VT被感測到為較VT"4"減少的VT"4B"。 Figures 16 to 19 are graphical representations of the threshold voltages when the data value of the addressed bit is "3" (from the lowest to the highest VT value of 1 to 4). In Fig. 16, the adjacent bit has a data value of "1". In Fig. 17, the adjacent bit has a data value of "2". In Fig. 18, the adjacent bit has a data value of "3". In Fig. 19, the adjacent bit has a data value of "4". Since the sense VT of the adjacent bit exceeds the minimum threshold VGN, the high drain voltage VBLR_h is applied to sense the address bit data value. The stylized VT "3A" of this addressed bit is sensed as a reduced VT "3". This addressed bit is programmed to VT "3A" so that the sensed VT is "3". Since this address bit "3A" also exceeds the minimum threshold VGN, the neighboring bit is also sensed to have a high drain voltage VBLR_h. This neighboring bit VT is programmed to VT "4A" such that the neighboring bit VT is sensed as a reduced VT "4". Otherwise, if the neighboring bit VT is held at VT "4" without being programmed to VT "4A", then this neighboring bit VT is sensed as VT "4B" which is reduced by VT "4".

第20~23圖是當定址位元的資料值是"4"時(自最低到最高VT值1到4中)的臨界電壓的圖示。在第20圖中,鄰近位元具有資料值"1"。在第21圖中,鄰近位元具有資料值"2"。在第22圖中,鄰近位元具有資料值"3"。在第23圖中,鄰近位元具有資料值"4"。如第19圖中所示,其是與第22圖中相反的安排,其定址位元與鄰近位元對調,此鄰近位元VT的程式化VT是"3A",其被感測到為減少的VT"3"。因為鄰近位元的感測VT"3A"超過最小臨界值VGN,所以鄰近位元感測為具有高汲極電壓VBLR_h。此定址位元被程式化至VT"4A"使得定址位元VT被感測到為減少的VT"4"。假如定址位元VT被保持在VT"4"而沒有被程式化至VT"4A"的話,則此定址位元VT被感測到為較VT"4"減少的VT"4B"。在第23圖中,鄰近位元具有資料值"4"。因為鄰近位元超過最小臨界值VGN,施加高汲極電壓VBLR_h來感測定址位元。此定址位元被程式化至 VT"4A"使得定址位元VT被感測為VT"4"。因為定址位元"4A"也是超過最小臨界值VGN,鄰近位元也是感測到具有高汲極電壓VBLR_h。此鄰近位元被程式化至VT"4A"使得鄰近位元VT被感測為VT"4"。否則,假如鄰近位元VT被保持在VT"4"而沒有被程式化至VT"4A"的話,則此鄰近位元VT被感測到為較VT"4"減少的VT"4B"。 Figures 20 to 23 are graphical representations of the threshold voltages when the data value of the addressed bit is "4" (from the lowest to the highest VT value of 1 to 4). In Fig. 20, the adjacent bit has a data value of "1". In Fig. 21, the adjacent bit has a data value of "2". In Fig. 22, the adjacent bit has a data value of "3". In Fig. 23, the adjacent bit has a data value of "4". As shown in Fig. 19, which is the reverse of the arrangement of Fig. 22, the address bit is swapped with the adjacent bit, and the stylized VT of the adjacent bit VT is "3A", which is sensed to be reduced. VT "3". Since the sense VT "3A" of the adjacent bit exceeds the minimum threshold VGN, the neighboring bit is sensed to have a high drain voltage VBLR_h. This addressed bit is programmed to VT "4A" such that the addressed bit VT is sensed as a reduced VT "4". If the addressed bit VT is held at VT "4" without being programmed to VT "4A", then the addressed bit VT is sensed as a VT "4B" that is reduced by VT "4". In Fig. 23, the adjacent bit has a data value of "4". Since the neighboring bit exceeds the minimum threshold value VGN, the high drain voltage VBLR_h is applied to sense the address bit. This addressed bit is stylized to VT "4A" causes the addressed bit VT to be sensed as VT "4". Since the addressed bit "4A" also exceeds the minimum threshold VGN, the neighboring bit is also sensed to have a high drain voltage VBLR_h. This neighboring bit is programmed to VT "4A" such that the neighboring bit VT is sensed as VT "4". Otherwise, if the neighboring bit VT is held at VT "4" without being programmed to VT "4A", then this neighboring bit VT is sensed as VT "4B" which is reduced by VT "4".

第24和25圖是臨界電壓的圖示,顯示不同的"定址位元"和"鄰近位元"的資料排列之程式化臨界電壓範圍與感測臨界電壓範圍。"定址位元"是具有底線的而"鄰近位元"是不具有底線的。 Figures 24 and 25 are graphical representations of threshold voltages showing the stylized threshold voltage range and sensing threshold voltage range for different "addressed bits" and "adjacent bits" data arrangements. The "addressed bit" has a bottom line and the "adjacent bit" does not have a bottom line.

第24圖顯示由在程式化此記憶胞時所增加電荷定義之程式化VT分佈。因為於感測時一記憶胞底下所儲存的電荷通常不會改變(除了讀取干擾效應之外),此程式化VT分佈在讀取偏壓改變時通常也不會改變。 Figure 24 shows the stylized VT distribution defined by the added charge when staging this memory cell. Because the charge stored under a memory cell typically does not change during sensing (except for read disturb effects), this stylized VT distribution typically does not change when the read bias changes.

第25圖顯示由在感測VT分佈,其在感測所施加汲極電壓時變動。在第25圖中的感測VT直接決定儲存於一記憶胞中的感測資料值,且感測VT分佈的數目與由定址位元或是鄰近位元所代表的資料值的數目相等。 Figure 25 shows the variation in the sensed VT distribution as it senses the applied drain voltage. The sense VT in Fig. 25 directly determines the sensed data values stored in a memory cell, and the number of sensed VT distributions is equal to the number of data values represented by the addressed bits or adjacent bits.

如同第15、19、22和23圖中所討論的,此程式化VT分佈預期由高汲極電壓和感測VT分佈導致的VT偏移。所以第24圖的程式化VT分佈並不包括高汲極偏壓將VT分佈偏移至較小值的效應。而第25圖的感測VT分佈則包括高汲極偏壓將VT分佈偏移至較小值的效應。因此,第24圖中的<定址位元><鄰近位元>程式化VT分佈<2><4>偏移至第25圖中的較低感測VT分佈。一個類似的自程式化VT分佈偏移至較低VT大小的感測VT分佈也顯示於<定址位元><鄰近位元>結合中的<3><4>、<4><3>和<4><4>。 As discussed in Figures 15, 19, 22 and 23, this stylized VT distribution is expected to be a VT offset caused by a high drain voltage and a sense VT distribution. Therefore, the stylized VT distribution of Figure 24 does not include the effect of high-bias bias biasing the VT distribution to a smaller value. The sensed VT distribution of Figure 25 includes the effect of high-bend bias biasing the VT distribution to a smaller value. Therefore, the <addressed bit> <adjacent bit> stylized VT distribution <2><4> in Fig. 24 is shifted to the lower sensed VT distribution in Fig. 25. A similar self-programmed VT distribution offset to a lower VT size of the sensed VT distribution is also shown in <addressed bits> <adjacent bits> combined <3><4>, <4><3> and <4><4>.

將第24圖的程式化VT分佈與第25圖的感測VT分佈指示 一特定感測資料值可以由多重程式化VT範圍代表。此外,感測VT範圍的數目與可以由定址位元或是鄰近位元所儲存的資料值的數目相等,程式化VT範圍的數目則是超過可以由定址位元或是鄰近位元所儲存的資料值的數目。 The stylized VT distribution of Fig. 24 and the sensed VT distribution indication of Fig. 25 A particular sensed data value can be represented by multiple stylized VT ranges. In addition, the number of sensed VT ranges is equal to the number of data values that can be stored by the addressed bit or neighboring bits, and the number of programmed VT ranges is greater than that that can be stored by the addressed bit or neighboring bits. The number of data values.

第26圖顯示不同汲極偏壓的臨界電壓分佈,其根據一鄰近位元中的資料值會調整或不調整此汲極偏壓。 Figure 26 shows the threshold voltage distribution for different buckling biases that adjust or not adjust the buckling bias based on the data values in an adjacent bit.

此縱軸是以對數方式表示顯示此模擬測試記憶陣列中的位元數目具有一特定感測VT於橫軸上。虛線軌跡與沒有根據鄰近位元中的資料值調整汲極偏壓之感測程序對應。而實線軌跡則是與會根據鄰近位元中的資料值調整汲極偏壓之感測程序對應。此圖示顯示調整汲極偏壓會將VT區間變寬,特別是在兩個最低的VT分佈間,其具有一個變寬的約為0.5V之VT間距。 The vertical axis is a logarithmic representation showing that the number of bits in the analog test memory array has a particular sense VT on the horizontal axis. The dashed trace corresponds to a sensing procedure that does not adjust the drain bias based on the data values in adjacent bits. The solid line trajectory corresponds to a sensing program that adjusts the 偏压 bias according to the data value in the adjacent bit. This illustration shows that adjusting the 偏压 bias will widen the VT interval, especially between the two lowest VT distributions, with a widened VT spacing of approximately 0.5V.

第27圖是汲極線驅動電路區塊的方塊示意圖,其具有由儲存在一靜態隨機存取記憶體SRAM中的資料所決定的汲極電壓。自上到下,此區塊包含一鄰近位元靜態隨機存取記憶體SRAM區塊、SENAMP感測放大器、YMUX行多工器及ARRAY記憶陣列。 Figure 27 is a block diagram of a drain line driver circuit block having a drain voltage determined by data stored in a static random access memory (SRAM). From top to bottom, this block contains an adjacent bit SRAM block, SENAMP sense amplifier, YMUX line multiplexer and ARRAY memory array.

第28圖是通常靜態隨機存取記憶體SRAM區塊,其自一汲極線驅動電路區塊產生信號以控制汲極電壓。一SRAM記憶體如同第6圖般產生Nb_h、Nb_l信號。假如此鄰近位元的感測VT超過字元線讀取電壓的臨界電壓VT,使得高汲極電壓可以被施加以感測此定址位元,Nb_h(鄰近位元高準位)=1,Nb_l(鄰近位元低準位)=0。假如此鄰近位元的感測VT並沒有超過字元線讀取電壓的臨界電壓VT,使得低汲極電壓可以被施加以感測此定址位元,Nb_h(鄰近位元高準位)=0,Nb_l(鄰近位元低準位)=1。 Figure 28 is a typical static random access memory SRAM block that generates a signal from a drain line driver circuit block to control the drain voltage. An SRAM memory generates Nb_h and Nb_1 signals as shown in FIG. If the sensing VT of the neighboring bit exceeds the threshold voltage VT of the word line read voltage, the high drain voltage can be applied to sense the addressed bit, Nb_h (adjacent bit high level) = 1, Nb_l (adjacent bit low level) = 0. Suppose the sensing VT of the adjacent bit does not exceed the threshold voltage VT of the word line read voltage, so that the low drain voltage can be applied to sense the addressed bit, Nb_h (adjacent bit high level) = 0 , Nb_l (adjacent bit low level) = 1.

第29圖是一汲極線驅動電路的一範例電路圖。兩個平行的 反及閘串列與Vdd供應電壓和DL汲極線連接。左側的反及閘串列接收閘極電壓Vblr_h和Nb_h。右側的反及閘串列接收閘極電壓Vblr_l和Nb_l。假如Nb_h(鄰近位元高準位)=1,Nb_l(鄰近位元低準位)=0,則左側的反及閘串列開啟而右側的反及閘串列關閉。DL汲極線則具有Vblr_h的值(小於一電晶體VT)。假如Nb_h(鄰近位元高準位)=0,Nb_l(鄰近位元低準位)=1,則左側的反及閘串列關閉而右側的反及閘串列開啟。DL汲極線則具有Vblr_l的值(小於一電晶體VT)。 Figure 29 is a circuit diagram showing an example of a drain line driving circuit. Two parallel The gate sequence is connected to the Vdd supply voltage and the DL drain line. The reverse gate series on the left receives the gate voltages Vblr_h and Nb_h. The reverse gate sequence on the right side receives the gate voltages Vblr_l and Nb_l. If Nb_h (adjacent bit high level) = 1, Nb_l (adjacent bit low level) = 0, then the left and right gate series are turned on and the right side is closed. The DL drain line has a value of Vblr_h (less than a transistor VT). If Nb_h (adjacent bit high level) = 0, Nb_l (adjacent bit low level) = 1, the left side of the reverse gate sequence is closed and the right side of the reverse gate sequence is turned on. The DL drain line has a value of Vblr_l (less than a transistor VT).

第30圖為根據鄰近位元的值而具有調整汲極偏壓之一讀取操作的流程圖。在步驟12,接收一"讀取命令"及一記憶胞"定址位元的位址"。在步驟14,儲存定址位元的位址,且根據此定址位元,取得鄰近位元的位址。一個表格可以同時對定址位元的地址和鄰近位元的位址產生索引,允許定址位元重新映射至鄰近位元。在步驟16,使用鄰近位元的位址來感測鄰近位元。其預設值是使用低汲極電壓。如第6圖中所示,儲存此資料其指示鄰近位元值具有足夠高的VT以允許高汲極電壓來感測定址位元。如第28圖中所示的一範例記憶體儲存此資料。在步驟18,重新取得在步驟14中所儲存的定址位元的輸入位址。在步驟20,檢查鄰近位元SRAM且根據在步驟16中所儲存的資料,決定鄰近位元的感測VT是否超過臨界VT。在步驟22,假如鄰近位元的感測VT超過臨界VT,則汲極電壓是vblr_h。在步驟24,假如鄰近位元的感測VT沒有超過臨界VT,則汲極電壓是vblr_l。在步驟26,利用vbrl_h或是vblr_l汲極電壓來感測定址位元。在步驟28,結束此讀取命令。 Figure 30 is a flow chart showing one of the read operations of adjusting the gate bias based on the value of the adjacent bit. At step 12, a "read command" and a memory "address of the addressed bit" are received. In step 14, the address of the addressed bit is stored, and based on the addressed bit, the address of the neighboring bit is obtained. A table can simultaneously index the address of the addressed bit and the address of the adjacent bit, allowing the addressed bit to be remapped to the adjacent bit. At step 16, the neighboring bit is sensed using the address of the neighboring bit. The default value is to use a low drain voltage. As shown in Figure 6, storing this data indicates that the neighboring bit values have a sufficiently high VT to allow high drain voltages to sense the address bits. An example memory as shown in Figure 28 stores this material. At step 18, the input address of the addressed bit stored in step 14 is retrieved. At step 20, the neighboring bit SRAM is checked and based on the data stored in step 16, it is determined whether the sensed VT of the neighboring bit exceeds the critical VT. At step 22, if the sense VT of the neighboring bit exceeds the critical VT, the drain voltage is vblr_h. At step 24, if the sense VT of the adjacent bit does not exceed the critical VT, then the drain voltage is vblr_1. At step 26, the address bit is sensed using the vbrl_h or vblr_l drain voltage. At step 28, this read command is ended.

第31~33圖為根據鄰近位元的值而具有調整汲極偏壓之程式化操作的流程圖。在第31圖中,汲極電壓是根據感測資料初始地決定。在第32和33圖中,汲極電壓是根據與程式化命令一起之輸入資料初始地決定。 Figures 31-33 are flow diagrams of the stylized operation of adjusting the gate bias based on the values of adjacent bits. In Fig. 31, the drain voltage is initially determined based on the sensed data. In Figures 32 and 33, the drain voltage is initially determined based on the input data along with the stylized commands.

在第31圖中,在步驟30,接收具有程式化位址的"程式化命令"及"欲 被程式化的資料"。在步驟32,根據在步驟30所接收之欲被程式化的資料,欲被程式化至定址位元的內容被處理以決定對應的VT是否超過第6圖中所示的臨界VT,且其結果儲存至SRAM1,且欲被程式化至鄰近位元的內容被處理以決定對應的VT是否超過第6圖中所示的臨界VT,且其結果儲存至SRAM2。在步驟34,讀取鄰近位元,其預設值是使用低汲極電壓。如第6圖中所示,儲存此資料其指示鄰近位元值具有足夠高的VT以允許高汲極電壓來感測定址位元。如第28圖中所示的一範例記憶體儲存此資料。在步驟36,根據在步驟34中所儲存的資料,程式化驗證此定址位元具有高或低汲極偏壓。在步驟38,讀取定址位元,其預設值是使用低汲極電壓。儲存此資料其指示定址位元是否具有足夠高的VT以允許高汲極電壓來感測鄰近位元。在步驟40,根據在步驟38中所儲存的資料,程式化驗證此鄰近位元具有高或低汲極偏壓。在步驟42,決定在步驟36及40的程式化驗證操作是否通過。在步驟44,假如在步驟36及40的程式化驗證操作一者失敗,則程式化鄰近位元及/或定址位元,而且此流程自步驟34開始重複。在步驟46,假如在步驟36及40的程式化驗證操作通過的話,則結束此程式化命令。 In Figure 31, in step 30, a "stylized command" with a stylized address is received. Stylized data. In step 32, based on the data to be programmed in step 30, the content to be programmed into the addressed bit is processed to determine if the corresponding VT exceeds the one shown in Figure 6. The critical VT, and the result is stored in SRAM1, and the content to be programmed to the adjacent bit is processed to determine whether the corresponding VT exceeds the critical VT shown in Figure 6, and the result is stored in SRAM 2. 34, reading adjacent bits, the preset value is to use a low drain voltage. As shown in Figure 6, storing this data indicates that the adjacent bit value has a sufficiently high VT to allow high drain voltage sensing The address bit stores the data as an example memory as shown in Figure 28. At step 36, based on the data stored in step 34, the address bit is stylized to verify that the addressed bit has a high or low drain bias. The address bit is read at step 38, the default value of which is to use a low drain voltage. This data is stored which indicates whether the addressed bit has a sufficiently high VT to allow high drain voltage to sense adjacent bits. 40, according to the data stored in step 38, the program Verify that the neighboring bit has a high or low drain bias. In step 42, it is determined if the stylized verify operation in steps 36 and 40 has passed. In step 44, if the stylized verify operation in steps 36 and 40 fails. The program then programs the neighboring bits and/or the addressed bits, and the process repeats from step 34. At step 46, if the stylized verify operation in steps 36 and 40 is passed, the stylized command is terminated.

在第32圖中,在步驟50,接收具有程式化位址的"程式化命令"及"欲被程式化的資料"。在步驟52,根據在步驟50所接收之欲被程式化的資料,欲被程式化至定址位元的內容被處理以決定對應的VT是否超過第6圖中所示的臨界VT,且其結果儲存至SRAM1,且欲被程式化至鄰近位元的內容被處理以決定對應的VT是否超過第6圖中所示的臨界VT,且其結果儲存至SRAM2。在步驟54,檢查儲存於SRAM1及SRAM2中的資料以決定是否施加較高的汲極電壓來對鄰近位元進行程式化驗證。對SRAM1及SRAM2中的資料進行檢查是因為定址位元及鄰近位元兩者的內容會決定此VT,如第24~25圖中所示。在步驟56,使用根據SRAM1中地址位元的資料之汲極電壓,程式化驗證鄰近位元。在步驟58,檢查儲存於SRAM1及SRAM2中的資料以決定是否 施加較高的汲極電壓來對定址位元進行程式化驗證。對SRAM1及SRAM2中的資料進行檢查是因為定址位元及鄰近位元兩者的內容會決定此VT,如第24~25圖中所示。在步驟60,如第28圖中所示的一般儲存SRAM2中鄰近位元的資料於鄰近位元SRAM中,對定址位元設定程式化驗證的汲極電壓。在步驟62,決定在步驟56及60的程式化驗證操作是否通過。在步驟64,假如在步驟56及60的程式化驗證操作一者或兩者失敗,則程式化鄰近位元及/或定址位元,而且此流程自步驟54開始重複。在步驟66,假如在步驟56及60的程式化驗證操作通過的話,則結束此程式化命令。 In Fig. 32, at step 50, a "stylized command" having a stylized address and "material to be stylized" are received. At step 52, based on the data to be programmed in step 50, the content to be programmed to the addressed bit is processed to determine if the corresponding VT exceeds the critical VT shown in Figure 6, and the result The content stored to SRAM1 and intended to be programmed to adjacent bits is processed to determine if the corresponding VT exceeds the critical VT shown in Figure 6, and the result is stored in SRAM2. At step 54, the data stored in SRAM1 and SRAM2 is examined to determine if a higher drain voltage is applied to programmatically verify adjacent bits. The data in SRAM1 and SRAM2 is checked because the contents of both the addressed bit and the adjacent bit determine the VT, as shown in Figures 24-25. At step 56, the neighboring bit is programmatically verified using the drain voltage of the data according to the address bits in SRAM1. At step 58, the data stored in SRAM1 and SRAM2 is checked to determine whether A higher drain voltage is applied to programmatically verify the addressed bits. The data in SRAM1 and SRAM2 is checked because the contents of both the addressed bit and the adjacent bit determine the VT, as shown in Figures 24-25. At step 60, the data of the neighboring bits in the general storage SRAM 2 as shown in FIG. 28 is stored in the adjacent bit SRAM, and the gated voltage of the stylized verification is set for the addressed bit. At step 62, it is determined whether the stylized verification operation at steps 56 and 60 has passed. At step 64, if one or both of the stylized verification operations in steps 56 and 60 fail, the neighboring bits and/or addressing bits are programmed, and the process begins with step 54. At step 66, if the stylized verification operation in steps 56 and 60 is passed, the stylized command is terminated.

在第33圖中,在步驟70,接收具有程式化位址的"程式化命令"及"欲被程式化的資料"。在步驟72,根據在步驟70所接收之欲被程式化的資料,欲被程式化至定址位元的內容被處理以決定對應的VT是否超過第6圖中所示的臨界VT,且其結果儲存至SRAM1,且欲被程式化至鄰近位元的內容被處理以決定對應的VT是否超過第6圖中所示的臨界VT,且其結果儲存至SRAM2。在步驟74,檢查儲存於SRAM1及SRAM2中的資料以決定是否施加較高的汲極電壓來對定址位元及鄰近位元進行程式化驗證。對SRAM1及SRAM2中的資料進行檢查是因為定址位元及鄰近位元兩者的內容會決定此VT,如第24~25圖中所示。在步驟76,使用根據SRAM1中定址位元的資料之汲極電壓,程式化驗證鄰近位元,且使用根據SRAM2中鄰近位元的資料之汲極電壓,程式化驗證定址位元。在步驟78,決定在步驟76的程式化驗證操作是否通過。在步驟80,假如在步驟76的程式化驗證操作失敗,則程式化鄰近位元及/或定址位元,而且此流程自步驟74開始重複。在步驟82,假如在步驟76的程式化驗證操作通過的話,則結束此程式化命令。 In Fig. 33, at step 70, a "stylized command" having a stylized address and "material to be stylized" are received. At step 72, based on the data to be programmed in step 70, the content to be programmed to the addressed bit is processed to determine if the corresponding VT exceeds the critical VT shown in Figure 6, and the result The content stored to SRAM1 and intended to be programmed to adjacent bits is processed to determine if the corresponding VT exceeds the critical VT shown in Figure 6, and the result is stored in SRAM2. At step 74, the data stored in SRAM1 and SRAM2 is examined to determine whether a higher drain voltage is applied to programmatically verify the addressed bit and adjacent bits. The data in SRAM1 and SRAM2 is checked because the contents of both the addressed bit and the adjacent bit determine the VT, as shown in Figures 24-25. At step 76, the neighboring bit is programmatically verified using the drain voltage of the data in accordance with the addressed bit in SRAM1, and the addressed bit is programmed to verify using the drain voltage based on the data of the adjacent bit in SRAM2. At step 78, it is determined if the stylized verification operation at step 76 has passed. At step 80, if the stylized verification operation at step 76 fails, the neighboring bits and/or addressing bits are programmed, and the flow repeats from step 74. At step 82, if the stylized verification operation at step 76 is passed, the stylized command ends.

第34圖顯示根據本發明一實施例具有此處所描述之改良及記憶陣列的積體電路的簡化方塊示意圖。其中積體電路3450包括記憶陣列3400。一字元線(列)解碼器與區塊選擇解碼器3401與沿著記憶陣列3400列方向安排之複數條字元線3402 及串列選擇線耦接及電性溝通。一位元線(行)解碼器與驅動器3403與沿著記憶陣列3400行方向安排之複數條位元線3404耦接及電性溝通,以自該記憶陣列3400的記憶胞讀取資料及寫入資料。位址係由匯流排3405提供給字元線解碼器與區塊選擇解碼器3401及位元線解碼器3403。方塊3406中的感測放大器與資料輸入結構,包括如第28圖中所示的汲極線電路及SRAM記憶體,經由匯流排3407與位元線解碼器3403耦接。資料由積體電路3450上的輸入/輸出埠提供給資料輸入線3434輸入至方塊3406中的資料輸入結構。資料由方塊3406中的感測放大器,經由資料輸出線3415,提供至積體電路上的輸入/輸出埠,或者至積體電路3450其他內部/外部的資料源。程式化驗證及讀取編排偏壓狀態機構3409根據一相同記憶胞中定址位元的鄰近位元來決定此汲極電壓與程式化VT,及控制偏壓調整供應電壓3408的應用。 Figure 34 shows a simplified block diagram of an integrated circuit having the improved and memory arrays described herein in accordance with an embodiment of the present invention. The integrated circuit 3450 includes a memory array 3400. A word line (column) decoder and block selection decoder 3401 and a plurality of word lines 3402 arranged along the column direction of the memory array 3400 And serial selection line coupling and electrical communication. A one-line (row) decoder and driver 3403 is coupled and electrically communicated with a plurality of bit lines 3404 arranged along the row direction of the memory array 3400 to read data and write from the memory cells of the memory array 3400. data. The address is provided by the bus 3405 to the word line decoder and block select decoder 3401 and bit line decoder 3403. The sense amplifier and data input structure in block 3406, including the drain line circuit and SRAM memory as shown in FIG. 28, is coupled to the bit line decoder 3403 via bus bar 3407. The data is supplied to the data input line 3434 via the input/output port on the integrated circuit 3450 to the data input structure in block 3406. The data is provided by the sense amplifier in block 3406, via the data output line 3415, to the input/output ports on the integrated circuit, or to other internal/external data sources of the integrated circuit 3450. The stylized verification and read orchestration bias state mechanism 3409 determines the application of the gate voltage and the stylized VT and the control bias adjustment supply voltage 3408 based on neighboring bits of the addressed bit in the same memory cell.

本發明之較佳實施例與範例詳細揭露如上,惟應瞭解為上述範例僅作為範例,非用以限制專利之範圍。就熟知技藝之人而言,自可輕易依據下列申請專利範圍對相關技術進行修改與組合。 The preferred embodiments and examples of the present invention are disclosed in detail above, but it should be understood that the above examples are merely exemplary and are not intended to limit the scope of the patent. For those skilled in the art, the related art can be modified and combined easily according to the scope of the following patent application.

Claims (12)

一種積體電路,包含:一非揮發記憶胞,包括:一第一電流負載終端、一第二電流負載終端及一閘極;一第一儲存部分鄰近該第一電流負載終端,並且儲存第一資料;一第二儲存部分鄰近該第二電流負載終端,並且儲存第二資料;控制電路施加一程式化驗證編排偏壓至該第一電流負載終端、該第二電流負載終端及該閘極;所施加的該程式化驗證編排偏壓係根據該第一資料及該第二資料的其中之一者,以程式化驗證該第一資料及該第二資料的其中之另一者。 An integrated circuit comprising: a non-volatile memory cell, comprising: a first current load terminal, a second current load terminal and a gate; a first storage portion adjacent to the first current load terminal, and storing the first a second storage portion is adjacent to the second current load terminal and stores the second data; the control circuit applies a stylized verification programming bias to the first current load terminal, the second current load terminal and the gate; The stylized verification programming bias applied is based on one of the first data and the second data to programmatically verify the other of the first data and the second data. 如申請專利範圍第1項之積體電路,:其中該第一資料決定一施加至該程式化驗證編排偏壓中之該第一電流負載終端之第一電壓資料,且所施加之該程式化驗證編排偏壓係用以程式化驗證該第二資料;其中該第二資料決定一施加至該程式化驗證編排偏壓中之該第二電流負載終端之一第二電壓資料,且所施加之該程式化驗證編排偏壓係用以程式化驗證該第一資料。 The integrated circuit of claim 1, wherein the first data determines a first voltage data applied to the first current load terminal in the stylized verification programming bias, and the stylized Verifying that the programming bias is used to programmatically verify the second data; wherein the second data determines a second voltage data applied to the second current load terminal of the stylized verification programming bias, and the applied The stylized verification programming bias is used to programmatically verify the first data. 如申請專利範圍第1項之積體電路,其中響應由該第一儲存部分的一第一臨界電壓所代表的該第一資料超過一最小臨界電壓,在該程式化驗證編排偏壓中施加一第一電壓至該第一電流負載終端以程式化驗證該第二資料;響應由該第一儲存部分的一第二臨界電壓所代表的該第一資料沒有超過一最小臨界電壓,在該程式化驗證編排偏壓中施加一第二電壓至該第一電流負載終端以程式化驗證該第二資料;該第一電壓較該第二電壓大。 The integrated circuit of claim 1, wherein the first data represented by a first threshold voltage of the first storage portion exceeds a minimum threshold voltage, and one of the stylized verification programming biases is applied. The first voltage is applied to the first current load terminal to programmatically verify the second data; and the first data represented by a second threshold voltage of the first storage portion does not exceed a minimum threshold voltage, and the stylized A second voltage is applied to the first current load terminal to verify the second data; the first voltage is greater than the second voltage. 如申請專利範圍第3項之積體電路,更包括:一記憶體儲存資料位元由代表該第一資料的一臨界電壓是否超過一最小臨界電壓來決定,其中該控制電路自該記憶體讀取該資料位元以控制該讀取編排偏壓中是否要施加該第一電壓或該第二電壓至該第一電流負載終端來讀取該第二資料。 For example, the integrated circuit of claim 3 includes: a memory storage data bit is determined by whether a threshold voltage representing the first data exceeds a minimum threshold voltage, wherein the control circuit reads from the memory The data bit is taken to control whether the first voltage or the second voltage is applied to the first current load terminal to read the second data in the read programming bias. 如申請專利範圍第1項之積體電路,其中該第一資料及該第二資料的每一者是複數個資料值之一者,該複數個資料值具有一第一總數的資料值,該複數個資料值由複數個程式化臨界電壓範圍代表,該複數個程式化臨界電壓範圍具有一第二總數的程式化臨界電壓範圍,該第二總數大於該第一總數。 For example, in the integrated circuit of claim 1, wherein each of the first data and the second data is one of a plurality of data values, the plurality of data values have a first total data value, The plurality of data values are represented by a plurality of stylized threshold voltage ranges having a second total number of programmed threshold voltage ranges greater than the first total number. 如申請專利範圍第1項之積體電路,其中該第一資料及該第二資料的每一者是複數個資料值之一者,該複數個資料值由複數個程式化臨界電壓範圍代表,該複數個資料值之一資料值由該複數個程式化臨界電壓範圍中的多個程式化臨界電壓範圍代表。 For example, in the integrated circuit of claim 1, wherein each of the first data and the second data is one of a plurality of data values, the plurality of data values are represented by a plurality of stylized threshold voltage ranges, One of the plurality of data values is represented by a plurality of stylized threshold voltage ranges in the plurality of programmed threshold voltage ranges. 如申請專利範圍第1項之積體電路,其中該第一資料及該第二資料的每一者是複數個資料值之一者,該複數個資料值由複數個程式化臨界電壓範圍代表,該複數個資料值之一資料值由該複數個程式化臨界電壓範圍中的多個程式化臨界電壓範圍代表,該第一資料由根據該第二資料的該多個程式化臨界電壓範圍之一特定者代表。 For example, in the integrated circuit of claim 1, wherein each of the first data and the second data is one of a plurality of data values, the plurality of data values are represented by a plurality of stylized threshold voltage ranges, One of the plurality of data values is represented by a plurality of stylized threshold voltage ranges in the plurality of programmed threshold voltage ranges, the first data being one of the plurality of programmed threshold voltage ranges according to the second data Representative of a specific person. 如申請專利範圍第1項之積體電路,其中該第一儲存部分及該第二儲存部分是一氮化矽儲存層中的不同部分。 The integrated circuit of claim 1, wherein the first storage portion and the second storage portion are different portions of a tantalum nitride storage layer. 如申請專利範圍第1項之積體電路,其中該第一資料及該第二資料 的每一者是多重位元。 For example, the integrated circuit of claim 1 of the patent scope, wherein the first data and the second data Each of them is a multiple bit. 如申請專利範圍第1項之積體電路,其中於根據該第一資料及該第二資料之另一者施加該程式化驗證編排偏壓之前,一程式化命令的輸入資料決定該第一資料及該第二資料之另一者。 The integrated circuit of claim 1, wherein the input data of a stylized command determines the first data before applying the stylized verification programming bias according to the other of the first data and the second data And the other of the second information. 如申請專利範圍第1項之積體電路,其中於根據該第一資料及該第二資料之另一者施加該程式化驗證編排偏壓之前,一讀取操作決定該第一資料及該第二資料之另一者。 The integrated circuit of claim 1, wherein the reading operation determines the first data and the first before applying the stylized verification programming bias according to the other of the first data and the second data The other of the two materials. 一種記憶體操作方法,包括:施加一程式化驗證編排偏壓至一非揮發記憶胞的一第一電流負載終端、一第二電流負載終端及一閘極;所施加的該程式化驗證編排偏壓係根據該第一資料及該第二資料的其中之一者,以讀取第一資料及第二資料的其中之另一者;該第一資料儲存於一第一儲存部分鄰近該第一電流負載終端及該第二資料儲存於一第二儲存部分鄰近該第二電流負載終端。 A memory operation method includes: applying a stylized verification programming bias to a first current load terminal, a second current load terminal, and a gate of a non-volatile memory cell; applying the stylized verification programming offset Pressing one of the first data and the second data to read the other of the first data and the second data; the first data is stored in a first storage portion adjacent to the first The current load terminal and the second data are stored in a second storage portion adjacent to the second current load terminal.
TW103136995A 2012-03-21 2012-03-21 Method and apparatus for adjusting drain bias of a memory cell with addressed and neighbor bits TWI541804B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW103136995A TWI541804B (en) 2012-03-21 2012-03-21 Method and apparatus for adjusting drain bias of a memory cell with addressed and neighbor bits

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW103136995A TWI541804B (en) 2012-03-21 2012-03-21 Method and apparatus for adjusting drain bias of a memory cell with addressed and neighbor bits

Publications (2)

Publication Number Publication Date
TW201506930A true TW201506930A (en) 2015-02-16
TWI541804B TWI541804B (en) 2016-07-11

Family

ID=53019465

Family Applications (1)

Application Number Title Priority Date Filing Date
TW103136995A TWI541804B (en) 2012-03-21 2012-03-21 Method and apparatus for adjusting drain bias of a memory cell with addressed and neighbor bits

Country Status (1)

Country Link
TW (1) TWI541804B (en)

Also Published As

Publication number Publication date
TWI541804B (en) 2016-07-11

Similar Documents

Publication Publication Date Title
US8339861B2 (en) Method and apparatus of performing an erase operation on a memory integrated circuit
US8730737B2 (en) Method and system for minimizing number of programming pulses used to program rows of non-volatile memory cells
US7518909B2 (en) Non-volatile memory device adapted to reduce coupling effect between storage elements and related methods
KR101600539B1 (en) Programming method of nonvolatile memory device
US9245644B2 (en) Method and apparatus for reducing erase disturb of memory by using recovery bias
JP4469649B2 (en) Semiconductor flash memory
US7394698B1 (en) Method and apparatus for adjusting a read reference level under dynamic power conditions
US9117492B2 (en) Method and apparatus for adjusting drain bias of a memory cell with addressed and neighbor bits
US20130215687A1 (en) Method and Apparatus for Dynamic Sensing Window in Memory
US7830708B1 (en) Compensating for variations in memory cell programmed state distributions
US6754106B1 (en) Reference cell with various load circuits compensating for source side loading effects in a non-volatile memory
US8787078B2 (en) Method and apparatus for reducing read disturb in memory
US7830721B2 (en) Memory and reading method thereof
US8537630B2 (en) Memory program discharge circuit of bit lines with multiple discharge paths
US7933150B2 (en) Nonvolatile semiconductor memory device and programming method thereof
JP2009080884A (en) Nonvolatile semiconductor memory device
US7486568B2 (en) Method and apparatus for protection from over-erasing nonvolatile memory cells
TWI541804B (en) Method and apparatus for adjusting drain bias of a memory cell with addressed and neighbor bits
TWI473094B (en) Method and apparatus for adjusting drain bias of a memory cell with addressed and neighbor bits
CN103325413B (en) There is integrated circuit and the method for operating thereof of the storage unit of addressing and adjacent bit
US9396770B2 (en) Method and apparatus for adjusting drain bias of a memory cell with addressed and neighbor bits
CN102446556B (en) Memory and method using same
JP2007294008A (en) Semiconductor storage device