TW200805382A - System and method for operating non-volatile memory using temperature compensation of voltages of unselected word lines and select gates - Google Patents

System and method for operating non-volatile memory using temperature compensation of voltages of unselected word lines and select gates Download PDF

Info

Publication number
TW200805382A
TW200805382A TW96119150A TW96119150A TW200805382A TW 200805382 A TW200805382 A TW 200805382A TW 96119150 A TW96119150 A TW 96119150A TW 96119150 A TW96119150 A TW 96119150A TW 200805382 A TW200805382 A TW 200805382A
Authority
TW
Taiwan
Prior art keywords
voltage
volatile storage
word line
storage element
state
Prior art date
Application number
TW96119150A
Other languages
Chinese (zh)
Other versions
TWI336081B (en
Inventor
Nima Mokhlesi
deng-tao Zhao
Original Assignee
Sandisk Corp
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
Priority claimed from US11/424,812 external-priority patent/US7342831B2/en
Priority claimed from US11/424,800 external-priority patent/US7391650B2/en
Application filed by Sandisk Corp filed Critical Sandisk Corp
Publication of TW200805382A publication Critical patent/TW200805382A/en
Application granted granted Critical
Publication of TWI336081B publication Critical patent/TWI336081B/en

Links

Classifications

    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C16/00Erasable programmable read-only memories
    • G11C16/02Erasable programmable read-only memories electrically programmable
    • G11C16/04Erasable programmable read-only memories electrically programmable using variable threshold transistors, e.g. FAMOS
    • G11C16/0483Erasable programmable read-only memories electrically programmable using variable threshold transistors, e.g. FAMOS comprising cells having several storage transistors connected in series
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C11/00Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor
    • G11C11/56Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using storage elements with more than two stable states represented by steps, e.g. of voltage, current, phase, frequency
    • G11C11/5621Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using storage elements with more than two stable states represented by steps, e.g. of voltage, current, phase, frequency using charge storage in a floating gate
    • G11C11/5642Sensing or reading circuits; Data output circuits
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C16/00Erasable programmable read-only memories
    • G11C16/02Erasable programmable read-only memories electrically programmable
    • G11C16/06Auxiliary circuits, e.g. for writing into memory
    • G11C16/26Sensing or reading circuits; Data output circuits
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C16/00Erasable programmable read-only memories
    • G11C16/02Erasable programmable read-only memories electrically programmable
    • G11C16/06Auxiliary circuits, e.g. for writing into memory
    • G11C16/34Determination of programming status, e.g. threshold voltage, overprogramming or underprogramming, retention
    • G11C16/3418Disturbance prevention or evaluation; Refreshing of disturbed memory data
    • G11C16/3427Circuits or methods to prevent or reduce disturbance of the state of a memory cell when neighbouring cells are read or written
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C16/00Erasable programmable read-only memories
    • G11C16/02Erasable programmable read-only memories electrically programmable
    • G11C16/06Auxiliary circuits, e.g. for writing into memory
    • G11C16/34Determination of programming status, e.g. threshold voltage, overprogramming or underprogramming, retention
    • G11C16/3436Arrangements for verifying correct programming or erasure
    • G11C16/3454Arrangements for verifying correct programming or for detecting overprogrammed cells
    • G11C16/3459Circuits or methods to verify correct programming of nonvolatile memory cells
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C7/00Arrangements for writing information into, or reading information out from, a digital store
    • G11C7/04Arrangements for writing information into, or reading information out from, a digital store with means for avoiding disturbances due to temperature effects
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C16/00Erasable programmable read-only memories
    • G11C16/02Erasable programmable read-only memories electrically programmable
    • G11C16/06Auxiliary circuits, e.g. for writing into memory
    • G11C16/34Determination of programming status, e.g. threshold voltage, overprogramming or underprogramming, retention
    • G11C16/3418Disturbance prevention or evaluation; Refreshing of disturbed memory data
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C2211/00Indexing scheme relating to digital stores characterized by the use of particular electric or magnetic storage elements; Storage elements therefor
    • G11C2211/56Indexing scheme relating to G11C11/56 and sub-groups for features not covered by these groups
    • G11C2211/562Multilevel memory programming aspects
    • G11C2211/5621Multilevel programming verification

Abstract

Reading and verify operations are performed on non-volatile storage elements using temperature-compensated read voltages for unselected word lines, and/or for select gates such as drain or source side select gates of a NAND string. In one approach, while a read or verify voltage is applied to a selected word line, temperature-compensated read voltages are applied to unselected word lines and select gates. Word lines which directly neighbor the selected word line can receive a voltage which not temperature compensated, or which is temperature-compensated to a reduced degree. The read or verify voltage applied to the selected word line can also be temperature-compensated. The temperature compensation may also account for word line position.

Description

200805382 九、發明說明: 【發明所屬之技術領域】 本發明係關於非揮發性記憶體技術。 【先如技術】200805382 IX. DESCRIPTION OF THE INVENTION: TECHNICAL FIELD OF THE INVENTION The present invention relates to non-volatile memory technology. [First as technology]

半^體§己憶體愈來愈普遍地用於各種電子震置中。舉例 而言,非揮發性半導體記憶體可用於蜂巢式電話、數位相 機、個人數位助理、移動計算裝置、非移動計算裝置或其 他裝置中。電可擦可程式化唯讀記憶體(eeprom)及快閃 記憶體便是最受歡迎的非揮發性半導體記憶體之一。 EEPROM及快閃記憶體兩者皆使用一浮動閘⑯,該浮動 閘極定位於-半導體基板中之溝道區上方並與該溝道區絕 緣。該浮動閘極定位於源極區與沒極區之間。一控制間極 設置於該浮動閘極上方,並與該浮動問極絕緣。電晶體之 臨限電壓受浮動閘極上所保持之電荷量控制。亦即,必須 在導通該電晶體以容許其源極與汲極之間導電之前施加至 控制閘極之最小電壓量係由該浮動閘極上之電荷位準护 制。 工 當程式化-EEPROM或快閃記憶體裝置(例如__ nand快 閃記憶體裝置)時’通常對控制閘極施加一程式化電壓且 將位元線接地。來自溝道之電子注人至浮動閘極内。當電 子在浮動閘極中積聚時草黏 田 ^ ,予勳閘極變成帶負電荷且儲存元 件之臨限電壓升高,從而佶 句使储存70件處於程式化狀態。更 多關於程式化之資訊可在 仕 ‘ % 為 Source Side SelfHalf-body § recalls are increasingly used in a variety of electronic shocks. For example, non-volatile semiconductor memory can be used in cellular telephones, digital cameras, personal digital assistants, mobile computing devices, non-mobile computing devices, or other devices. Electrically erasable and programmable eeprom and flash memory are among the most popular non-volatile semiconductor memories. Both the EEPROM and the flash memory use a floating gate 16 that is positioned above and insulates the channel region in the semiconductor substrate. The floating gate is positioned between the source region and the non-polar region. A control electrode is disposed above the floating gate and insulated from the floating electrode. The threshold voltage of the transistor is controlled by the amount of charge held on the floating gate. That is, the minimum amount of voltage that must be applied to the control gate before the transistor is turned on to allow conduction between its source and drain is protected by the charge level on the floating gate. When stylized-EEPROM or flash memory devices (such as __ nand flash memory devices), a stylized voltage is typically applied to the control gate and the bit line is grounded. The electrons from the channel are injected into the floating gate. When the electrons accumulate in the floating gate, the grass sticks to the field, and the threshold is negatively charged and the threshold voltage of the storage element rises, thereby causing the storage of 70 pieces to be stylized. More information about stylization is available in ‘ % for Source Side Self

Boosting Technique for Non-Vni +·ι a.Boosting Technique for Non-Vni +·ι a.

Volatile Memory”之美國專利 121185.doc 200805382 M59,397中及標稱為”〜咖⑽Me·”&quot;之 美國專利6,917,542中找到,以上兩個專利皆以全文引用的 方式併入本文中。 某些EEPROM及快閃記憶體裝置具有一用於儲存兩個電 荷範圍之浮動閘極,且因此,可在兩種狀態(例如一擦除 狀態與-程式化狀態)之間程式化/擦除儲存元件。此種快 閃記憶體裝置有a夺稱作=進製快閃記憶體裝置。 多狀態快閃記憶體裝置係藉由識別由各禁止範圍分開之 多個不同之容許/有效程式化臨限電壓範圍所構建。每一 不同之臨限電壓範圍均對應於在該記,隨裝置_所編碼之 資料位元集合之一預定值。 在^削的非揮發性5己憶體裝置(例如Nand快閃記憶體裝 置)中,溫度變化會導致各種關於讀取及寫入資料之問 通屺丨思體裝置基於其所處環境而經受變化之溫度。舉例 而口,某些現有兄憶體裝置被認定為應在_4〇。與之 間使用。工業、軍事甚至消費者應用中之裝置可經歷顯著 之溫度變化。溫度影響諸多電晶體參數,其中最主要者為 臨限包壓。特定而言,溫度變化可導致讀取錯誤並使非揮 發性儲存元件之不同狀態之臨限電壓分佈變寬。當前,藉 由以下方式補償溫度變化:以一計及選擇儲存元件之臨限 電壓之溫度變化之方式,改變施加至選擇字線之讀取/驗 證電壓。此方法至多能解決儲存元件之臨限電壓分佈之平 均偏移問題,為了簡明起見,假定臨限電壓皆處於相同之 貧料狀態。然而,需要一種經改良之技術,以便進一步減 121I85.doc 200805382 小因溫度變化所致之每_狀態之臨限電屢分佈之擴展。 【發明内容】 本發明藉由提供-種用於操作非揮發性儲存器之系統與 方法來解决上述及其他問題,其中將經溫度補償之電遂施 加至非選擇非揮發性儲存元件及/或選擇閉極。本發明達 成各種益處’包括經改良之讀取及寫入效能。 於—K⑻例中,藉由下述方式來操作非揮發性儲存器: 將第一電壓(例如,一讀取或驗證電壓)施加至一選擇字 線t確定-與該選擇字線相關聯之第一非揮發性儲存元件 之程式化狀態。該第一非揮發性儲存元件設置於一組非揮 發性儲存7C件中。例如’該第一電塵可係一讀取電壓,其 =於讀取第—非揮發性儲存元件在其程式化後之程式化狀 悲。或者,該第一電麼可係一驗證電壓,其用於驗證該第 一非揮發性儲存元件是否達到了所期望之程式化狀態。例 如此種驗證電麼可施加於一系列此類脈動之個別程式化 脈動之間。而且’在施加該第一電壓之同時,將一經溫度 補之電Μ施加至-個或多個與該非揮發性儲存元件組相 關聯之非選擇字線。 於一種方法中,將相同之經溫度補償電壓施加至該等非 選擇字線中之每一者。於另一種方法中,將不同之經溫度 補償電壓施加至不同之非選擇字線。於再-種方法中,一 個或兩個直接鄰近該選擇字線之非選擇字線接收一未經溫 度補償或經溫度補償一減小之量(相對於施加至其他非選 擇字線之經溫度補償電壓而言)之電壓。亦可將一經溫度 121185.doc 200805382 補償之電壓施加至源極及/或沒極選擇雜,例如當所選 擇之非揮發性儲存元件處於一 NAND串中時。亦可對該第 一電壓進行溫度補償。 ^ 月她例中,藉由下述方式來操作非揮發性儲存 益·將-第-電壓施加至一選擇字線以確定一與該選擇字 線相關聯之第一非揮發性儲存元件之程式化狀態。該第一 非揮發性儲存元件設置於—組非揮發性儲存元件中。另 外,根據該選擇字線在複數個與該組非揮發性儲存元件相 :聯之字線中的一相對位置來對該第一電a進行溫度補 償。舉例而言,當該選擇字線相對於包括複數個字線之塊 之源極更接近於沒極時,可使用一更大之溫度補償量值。 於另-實施例中,藉由下述方式來操作非揮發性儲存 裔.將-第一 „施加至一選擇字線以確定一與該選擇字 線相關聯之第一非揮發性儲存元件之程式化狀態,其中該 第-非揮發性齡元件設置於_組非揮發性儲存元件中。 在施加該第一電遷之同時’將-經溫度補償之電麼施加至 至少-與該組非揮發性儲存元件相關聯之第—非選擇字 線。另外’在施加該第-電麼之同時,將一未經溫度補償 或經溫度補償-減小之量(相對於施加至㈣—非 線之經溫度補償電壓)之電遂施加至至少一與該組非揮發 性儲存元件相關聯之第:非選擇字線。於_種方法中 '少-第-非選擇字線不直接鄰近該選擇字線,而該至;: 一第一非選擇字線直接鄰近該選擇字線。 於再f加例中,精由下述方式來操作非揮發性儲存 121185.doc 200805382 器:將一第一電壓施加至一選擇字線以確定一與該選擇字 j相關^之第—非揮發性儲存元件之程式化狀態,其中該 $非揮發性儲存元件設置於一組非揮發性儲存元件中。 田/第非揮發性儲存元件不直接鄰近該選擇閘極時,在 施:I亥第一電壓之同時,將一第一經溫度補償電壓施加至 〃第非揮發性儲存元件相關聯之選擇閘極。當該第一 ^揮發性儲存元件直接鄰近該選擇閘極時,在施加該第一 私之同時’將一未經溫度補償或經溫度補償一減小之量 (相對於該第一經溫度補償電壓)之電壓施加至該選擇閘 〇該k擇閘極及該第一非揮發性儲存元件可設置於一 NAND串中,其中該選擇閘極處於該NAND串之-源極或 汲極側。 提供用於操作非揮發性儲存器及非揮發性儲存系統之對 &quot; &quot;亥等非揮發性儲存系統包括一組非揮發性儲存元 個或夕個如本文中所述用於操作該非揮發性健存 元件組之電路。 【實施方式】 ^月提供-種用於以一改良讀取及寫入效能之方式操 =發,错存器之系統與方法。經改良之效能係藉由將 :’皿又補饧之電壓施加至非選擇非揮發性儲 選擇閘極而達成。亘鹖〆忐π' ^ 括··減小之讀取干擾、程 八化狀恶之間的減水 ▲ 几铪、因較大程式化步長之使用而 ^ . 或猎由將狀態壓縮得更靠近在一起 而減小之操作窗口。 、 121185.doc -10- 200805382 一適用於構建本發明之記憶體系統之一實例使用NAND 快閃記憶體結構,其包括將多個電晶體串聯佈置於兩個選 擇閘極之間。該等串聯電晶體及該等選擇閘極稱作一 NAND串。圖1係一顯示一個NAND串之俯視圖。圖2係該 NAND串之一等效電路。圖1及2中所繪示之NAND串包括 串聯並夾於第一選擇閘極120與第二選擇閘極122之間的四Volatile Memory, U.S. Patent No. 1, 218, 185, filed to s. s. s. s. s. </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> <RTIgt; The EEPROM and flash memory device have a floating gate for storing two charge ranges, and thus, the memory element can be programmed/erased between two states (eg, an erased state and a -stylized state). Such a flash memory device has a device called a flash memory device. A multi-state flash memory device identifies a plurality of different allowable/effective stylized thresholds separated by prohibited ranges. The voltage range is constructed. Each of the different threshold voltage ranges corresponds to a predetermined value in the set of data bits encoded by the device. The non-volatile 5 memory device (such as Nand) In flash memory devices, temperature changes can cause various reading and writing data to be subjected to varying temperatures based on their environment. For example, some existing brothers It is considered to be used in conjunction with. Industrial, military, and even consumer applications can experience significant temperature changes. Temperature affects many transistor parameters, the most important of which is the threshold pressure. Temperature changes can cause read errors and widen the threshold voltage distribution of different states of the non-volatile storage element. Currently, temperature changes are compensated for by taking into account the temperature change of the threshold voltage of the storage element. In this way, the read/verify voltage applied to the selected word line is changed. This method can solve at most the average offset problem of the threshold voltage distribution of the storage element. For the sake of simplicity, it is assumed that the threshold voltages are all in the same lean state. However, there is a need for an improved technique to further reduce the expansion of the per-electrical distribution of each of the states due to temperature changes of the 121I85.doc 200805382. [Invention] The present invention provides for operation Non-volatile storage systems and methods to address these and other problems in which temperature compensated electrolysis is applied to non-selective non-volatile The present invention achieves various benefits 'including improved read and write performance. In the -K (8) example, the non-volatile memory is operated by: first voltage ( For example, a read or verify voltage is applied to a select word line t to determine a stylized state of the first non-volatile storage element associated with the selected word line. The first non-volatile storage element is disposed in a group Non-volatile storage in a 7C piece. For example, the first electric dust may be a read voltage, which is a stabilizing state after reading the first non-volatile storage element after its stylization. Or, the first The voltage can be a verification voltage that is used to verify that the first non-volatile storage element has reached the desired stylized state. For example, such verification power can be applied to a series of such programmed pulsations of such pulsations. between. Moreover, while applying the first voltage, a temperature-compensated electrode is applied to one or more non-selected word lines associated with the non-volatile storage element group. In one method, the same temperature compensated voltage is applied to each of the non-selected word lines. In another method, different temperature compensated voltages are applied to different non-selected word lines. In the re-method, one or two unselected word lines directly adjacent to the selected word line receive an amount that is not temperature compensated or temperature compensated by a decrease (relative to the temperature applied to other unselected word lines) The voltage of the compensation voltage). A voltage compensated by temperature 121185.doc 200805382 may also be applied to the source and/or the poleless selection, such as when the selected non-volatile storage element is in a NAND string. The first voltage can also be temperature compensated. ^ In her example, the non-volatile storage benefit is applied by applying a -first voltage to a selected word line to determine a program of the first non-volatile storage element associated with the selected word line. State. The first non-volatile storage element is disposed in the set of non-volatile storage elements. In addition, the first electrical a is temperature compensated according to a relative position of the selected word line in a plurality of word lines associated with the set of non-volatile storage elements. For example, a larger temperature compensation magnitude can be used when the selected word line is closer to the pole than the source of the block comprising the plurality of word lines. In another embodiment, the non-volatile storage is operated by applying -first to a selected word line to determine a first non-volatile storage element associated with the selected word line. a stylized state in which the first-non-volatile age component is disposed in the _ group of non-volatile storage elements. At the same time as the first electromigration is applied, 'the temperature-compensated electricity is applied to at least- The first-non-selected word line associated with the volatile storage element. In addition, 'when the first-th power is applied, the amount is reduced without temperature compensation or temperature compensation--relative to the application to (four)-non-line The temperature compensated voltage is applied to at least one of the non-selected word lines associated with the set of non-volatile storage elements. In the method, the 'less-first-non-selected word lines are not directly adjacent to the selection Word line, and the to;: A first unselected word line is directly adjacent to the selected word line. In the re-addition case, the non-volatile storage is operated by the following method. 121185.doc 200805382: a first Applying a voltage to a selected word line to determine a phase associated with the selected word j The stylized state of the non-volatile storage element, wherein the non-volatile storage element is disposed in a set of non-volatile storage elements. The field/non-volatile storage element is not directly adjacent to the selection gate Applying a first temperature-compensated voltage to the selected gate associated with the second non-volatile storage element while applying the first voltage of the first pass; when the first volatile storage element is directly adjacent to the select gate At the very extreme, applying a voltage that is not temperature compensated or temperature compensated by a decrease (relative to the first temperature compensated voltage) is applied to the selection gate while the first private is being applied The pole and the first non-volatile storage element can be disposed in a NAND string, wherein the select gate is on the source or drain side of the NAND string. Provided for operating non-volatile memory and non-volatile storage A system of non-volatile storage systems such as &quot;&quot;Hai includes a set of non-volatile storage elements or circuits for operating the non-volatile storage element group as described herein. [Embodiment] ^月Offer - used for A system and method for improving read and write performance in a method of operating, failing, and modifying. The improved performance is achieved by applying a voltage of ''reserved' to the non-selective non-volatile storage gate. Achieved. 亘鹖〆忐π' ^ Included · Reduced reading interference, the reduction between the eight evils ▲ Several 铪, due to the use of larger stylized steps ^ or Hunting by the state compression An operational window that is closer together and reduced. 121185.doc -10- 200805382 An example of a memory system suitable for use in constructing the present invention uses a NAND flash memory structure that includes a plurality of transistors arranged in series Between the two selected gates, the series transistors and the select gates are referred to as a NAND string. Figure 1 is a top view showing a NAND string. Figure 2 is an equivalent circuit of the NAND string. The NAND strings illustrated in Figures 1 and 2 include four connected in series and sandwiched between a first select gate 120 and a second select gate 122.

個電晶體100、102、104及106。選擇閘極120將NAND串連 接選通至位元線126。選擇閘極122將NAND串連接選通至 源極線128。選擇閘極120藉由對控制閘極12〇cg施加適宜 之電壓來加以控制。選擇閘極122藉由對控制閘極122C(} 施加適宜之電壓來加以控制。電晶體1〇〇、1〇2、1〇4及1〇6 中之每一者皆具有一控制閘極及一浮動閘極。電晶體1〇〇 具有控制閘極100CG及浮動閘極i〇〇FG。電晶體102包括控 制閘極102CG及浮動閘極102FG。電晶體1〇4包括控制閘極 104CG及浮動閘極i〇4FG。電晶體1〇6包括控制閘極1〇6Cg 及浮動閘極106FG。控制閘極i〇〇CG連接至(或係)字線 WL3,控制閘極102〇(}連接至字線WL2,控制閘極i〇4ce 連接至字線WL1,且控制閘極106CG連接至字線wl〇。在 一實施例中,電晶體100、102、104及1〇6皆為儲存元件, 亦稱作記憶體單元。在其他實施例中,儲存元件可包括多 個電晶體,或者可不同於圖⑷中所繪示。選擇閘極Η。 連接至選擇線SGD。選擇閘極122連接至選擇線SGS。 圖3提供上述NAND串之剖視圖。如圖3中所繪示, NAND串中之電晶體形成於p井區14〇中。每一電晶體皆包 121185.doc 200805382 括一由控制閘極(100CG、102CG、104CG及106CG)及一浮 動閘極(100FG、102FG、104FG及106FG)組成之堆疊閘極 結構。該等控制閘極及浮動閘極通常糟由沈積多晶秒層而 形成。該等浮動閘極形成於一氧化物膜或其他介電膜頂部 上之P-井表面上。控制閘極位於浮動閘極上方,其中一中 間多晶矽介電層將控制閘極與浮動閘極相分離。儲存元件 (100、102、104及106)之控制閘極形成字線。N+摻雜擴散 區13〇、I32、134、136及138共享於鄰近儲存元件之間,The transistors 100, 102, 104 and 106. Select gate 120 strobes the NAND string connections to bit line 126. Select gate 122 strobes the NAND string connection to source line 128. The selection gate 120 is controlled by applying a suitable voltage to the control gate 12?cg. The selection gate 122 is controlled by applying a suitable voltage to the control gate 122C (}. Each of the transistors 1〇〇, 1〇2, 1〇4, and 1〇6 has a control gate and A floating gate. The transistor 1 has a control gate 100CG and a floating gate i〇〇FG. The transistor 102 includes a control gate 102CG and a floating gate 102FG. The transistor 1〇4 includes a control gate 104CG and a floating gate. Gate i〇4FG. The transistor 1〇6 includes a control gate 1〇6Cg and a floating gate 106FG. The control gate i〇〇CG is connected to (or is) the word line WL3, and the control gate 102〇(} is connected to The word line WL2, the control gate i〇4ce is connected to the word line WL1, and the control gate 106CG is connected to the word line w1〇. In an embodiment, the transistors 100, 102, 104 and 1〇6 are all storage elements. Also referred to as a memory unit. In other embodiments, the storage element may comprise a plurality of transistors, or may be different from that illustrated in Figure 4. Select gate Η. Connect to select line SGD. Select gate 122 is connected to Select line SGS. Figure 3 provides a cross-sectional view of the above NAND string. As shown in Figure 3, the transistor in the NAND string is formed in p-well 14〇. Each transistor is packaged 121185.doc 200805382 includes a stacked gate structure consisting of a control gate (100CG, 102CG, 104CG, and 106CG) and a floating gate (100FG, 102FG, 104FG, and 106FG). The control gates and floating gates are typically formed by depositing polycrystalline seconds. The floating gates are formed on the surface of the P-well on top of an oxide film or other dielectric film. The control gate is floating. Above the gate, an intermediate polysilicon dielectric layer separates the control gate from the floating gate. The control gates of the storage elements (100, 102, 104, and 106) form word lines. N+ doped diffusion regions 13〇, I32 , 134, 136 and 138 are shared between adjacent storage elements,

藉此使該等儲存元件相互串聯連接以形成一 NAND串。該 等N+摻雜區形成儲存元件中之每一者之源極及汲極。舉例 而言,N+摻雜區130用作電晶體122之汲極及電晶體106之 源極,N+摻雜區132用作電晶體1〇6之汲極及電晶體1〇4之 源極,N+摻雜區134用作電晶體104之汲極及電晶體1〇2之 源極,N+摻雜區136用作電晶體1〇2之汲極及電晶體1〇〇之 源極,而N+摻雜區138用作電晶體1〇〇之汲極及電晶體12〇 之源極。N+摻雜區126連接至NAND串之位元線,同時n+ 摻雜區丨28連接至多個NAND串之一共用源極線。 注意,雖然圖I-3顯示NAND串中之四個儲存元件,但使 用四個電晶體僅作為一實例。 ^例用於本文中所述技術之一 NAND串可具有少於四個儲左 健存兀件或多於四個儲存元件。 舉例而言,某些NAND串將句枯s 士一 匕括8個、b個、32個或64個儲 存元件等。本文之論述並非 非侷限於—NAND串中任何特定 數量之儲存元件。 竹疋 每一儲存元件均可儲存 以類比形式或數位形式表 示之資 121185.doc 200805382 料。當儲存-個位元之數位資料時,將储存元件的可能之 臨限電塵範圍劃分成兩個範圍’該兩個範圍被指派給㈣ 資料”1”及”〇&quot;。於一 NAND型快閃記憶體之實例中,在擦 除儲存元件後該臨限電壓為負並被定義為邏輯&quot;i&quot;。而2 一程式化操作後臨限電壓為正並定義為邏輯&quot;0、當臨限 電壓為負並藉由向控制閘極施加0伏來嘗試一讀取時,儲 2元件將導通以指示正儲存邏輯丨。而當臨限電壓為正且 猎由向控制閘極施加〇伏來嘗試一讀取操作時,儲存元件 將不會導通,此指示儲存邏輯〇。 :儲存元件亦可儲存多種狀態’由此儲存多個數位資料 位兀。在健存多個貧料狀態之情形下,臨限電壓窗口被劃 分成多種狀態。例如,若使用四種狀態,則將有四個臨限 電壓範圍指派給資料值” U,,、” 1〇”、&quot;〇1,,及&quot;⑽&quot;。在一 \.,一 NAND型記憶體的_個實例中,在擦除操作後臨限電壓為 負並被定義為”11”。對狀態”1〇”、”〇1,,及”〇〇&quot;使用正臨限 包£。於某些實施方案中,係使用一格雷碼指派方案將資 料值(例如邏輯狀態)指派給臨限值範圍,以使若一浮動閘 極之臨限電壓錯誤地偏移至其相鄰物理狀態,則僅會影響 個位7L。㈣化至儲存元件巾之f料錢存元件之臨限 電壓範圍之間的具體關係相依於儲存元件所採用之資料編 碼方案。舉例而言’美國專利第6,222,762號及於測年6 月13日申凊且標稱為,,τ咖Cells F〇r A Me则7 SyStem”之美國專利申1青案第l〇/461,244號闡述各種用於多 狀態快閃儲存元件之資料編碼方案,二者皆以全文引用的 121185.doc -13- 200805382 方式併入本文中。 在以下美國專利/專利申請案中提供N A N D型快閃記憶體 及其操作之㈣實例’ ~有此等美國專利/專射請案皆 用的方式併人本文中:美國專利第5,谓,315號; 美國專利第5,774,397號;纟國專利第M46,935號;美國專 利第5,386,422號;美國專利第6,456,528號及美國專第 6,522,58〇號。除NAND快閃記憶體外,本發明亦可使用其 他類型之非揮發性記憶體。 適用於快閃EEPROM系統的另一種類型之儲存元件利用 一非導電性介電材料取代導電性浮動閘極以便以非易失性 方式儲存電荷。此種儲存元件闌述於一篇由Chan等人所著 的文旱&quot;A True Single-Transistor 〇xide-Nitride_Oxide EEPROM Device),,(IEEE Electr〇nDevice 乙扣⑽,第印&quot; 卷,Νο·3,1987年3月,第93-95頁)中。一由氧化矽 '氮化 矽及氧化矽(&quot;ΟΝΟ”)形成之三層介電質夾於一導電性控制 閘極與儲存元件溝道上方的一半導電性基板之表面之間。 忒儲存元件藉由將電子自儲存元件溝道注入至氮化物内而 程式化,其中電子陷獲並儲存於一有限區中。然後,所儲 存之電街以一可偵測方式改變儲存元件溝道之一部分之臨 限電壓。該儲存元件係藉由將熱電洞注入氮化物内來進行 擦除。亦參見由Nozaki等人所著之於”A EEPR〇m with MONOS Memory Cell for Semiconductor DiskThereby the storage elements are connected in series to each other to form a NAND string. The N+ doped regions form the source and drain of each of the storage elements. For example, the N+ doping region 130 serves as the drain of the transistor 122 and the source of the transistor 106, and the N+ doping region 132 serves as the drain of the transistor 1〇6 and the source of the transistor 1〇4. The N+ doping region 134 is used as the drain of the transistor 104 and the source of the transistor 1〇2, and the N+ doping region 136 is used as the drain of the transistor 1〇2 and the source of the transistor 1〇〇, and N+ Doped region 138 is used as the drain of the transistor 1 及 and the source of the transistor 12 。. N+ doped region 126 is connected to the bit line of the NAND string while n+ doped region 丨28 is connected to one of the plurality of NAND strings to share the source line. Note that although Figure I-3 shows four storage elements in a NAND string, the use of four transistors is only an example. An example NAND string for use in one of the techniques described herein may have fewer than four memory storage elements or more than four storage elements. For example, some NAND strings will include 8, b, 32, or 64 storage elements. The discussion herein is not limited to any particular number of storage elements in a NAND string. Bamboo stocks Each storage element can be stored in analogy or in digital form. 121185.doc 200805382 Material. When storing the digits of one bit, the possible threshold of the storage component is divided into two ranges 'The two ranges are assigned to (4) the data "1" and "〇". In a NAND type In the example of flash memory, the threshold voltage is negative after the storage element is erased and is defined as logic &quot;i&quot;. And after a stylized operation, the threshold voltage is positive and defined as logic &quot;0, When the threshold voltage is negative and a read is attempted by applying 0 volts to the control gate, the bank 2 component will conduct to indicate that the logic 正 is being stored. When the threshold voltage is positive and the hunter is applied to the control gate When squatting to try a read operation, the storage element will not be turned on. This indication stores the logic 〇. : The storage element can also store multiple states 'by storing multiple digital data bits. In this case, the threshold voltage window is divided into multiple states. For example, if four states are used, four threshold voltage ranges are assigned to the data values "U,,," 1〇", &quot;〇1, , and &quot;(10)&quot;. In an example of a NAND type memory, the threshold voltage is negative after the erase operation and is defined as "11". Use the positive threshold for the status "1〇", "〇1,, and"〇〇&quot;. In some embodiments, a Gray code assignment scheme is used to assign a data value (eg, a logic state) to a threshold range such that if a threshold voltage of a floating gate is erroneously shifted to its neighboring physical state , it will only affect the single digit 7L. (4) The threshold for the storage of the components of the storage device. The specific relationship between the voltage ranges depends on the data coding scheme used by the storage components. For example, U.S. Patent No. 6,222,762 and claimed on June 13th of the year of the test, and 标 CCells F〇r A Me, 7 SyStem, US Patent Application No. 1/〇, 461, No. 244 describes various data encoding schemes for multi-state flash memory components, both of which are incorporated herein by reference in their entirety. (4) Examples of flash memory and its operation' ~ There are ways to use these US patents/special shots in this article: US Patent No. 5, No. 315; US Patent No. 5,774,397; U.S. Patent No. 5,386,422; U.S. Patent No. 6,456,528; and U.S. Patent No. 6,522,58. In addition to NAND flash memory, other types of non-volatile memory can be used in the present invention. Another type of storage element of a flash EEPROM system utilizes a non-conductive dielectric material in place of a conductive floating gate to store charge in a non-volatile manner. Such a storage element is described in a by Chan et al.文旱&quot;A True S ingle-Transistor 〇xide-Nitride_Oxide EEPROM Device),, (IEEE Electr〇nDevice (10), ed. &quot; Volume, Νο. 3, March 1987, pp. 93-95). The three layers of dielectric formed by the ruthenium oxide and ruthenium oxide (&quot;ΟΝΟ) are sandwiched between a conductive control gate and a surface of the semiconductive substrate above the channel of the storage element. The germanium storage element is programmed by injecting electrons into the nitride from the channel of the storage element, wherein the electrons are trapped and stored in a limited area. The stored electrical street then changes the threshold voltage of a portion of the channel of the storage element in a detectable manner. The storage element is erased by injecting a thermal cavity into the nitride. See also "A EEPR〇m with MONOS Memory Cell for Semiconductor Disk" by Nozaki et al.

Application (IEEE Journal of Solid-State Circuits)第 26 卷,Νο·4,1991年4月,第497-501頁),其闡述了一種具有 121185.doc •14- 200805382 分裂閘極構造至類似儲存元件,其中一經摻雜之多晶矽閘 極延伸於儲存元件溝道之一部分上方以形成一分離選擇電 晶體。以上兩篇文章皆以全文引用的方式併入本文中。在 William D. Brown及 Joe E. Brewer戶斤編輯之’’Nonvolatile Semiconductor Memory Technology”(IEEE Press, 1998)1.2 • 部分中所提及之程式化技術亦於彼部分中描述為適用於介 - 電電荷陷獲裝置,該文章以引用的方式併入本文中。本發 明亦可使用本段所描述之儲存元件。因此,本文所描述之 技術亦適用於不同儲存元件之介電區之間的耦合。 用於在每一儲存元件中儲存兩個位元之另一種方法已由 Eitan等人闡述於’’NROM: A Novel Localized Trapping,2-Bit Nonvolatile Memory Cell)’’ (IEEE Electron Device Letters),第 21 卷,第 11 號,2000 年 11 月,第 543-545 頁) 中。一ΟΝΟ介電層延伸跨越源極擴散區與汲極擴散區之間 的溝道。一個資料位元之電荷局部化於毗鄰汲極之介電層 ^ 中,而另一資料位元之電荷則局部化於毗鄰源極之介電層 中。多狀態資料儲存係藉由分別讀取介電質内在空間上分 離之電荷儲存區之二進製狀態來實現。本發明亦可使用本 段所描述之儲存元件。 圖4圖解說明一例如圖1-3中所示彼等之NAND儲存元件 陣列之實例。沿每一行,一位元線206耦合至NAND串150 之汲極選擇閘極之汲極端子126。沿NAND串之每一列,一 源極線204可連接該等NAND串之源極選擇閘極的所有源極 端子128。在美國專利第5,570,3 15號、第5,774,397號及第 121185.doc -15- 200805382 6,046,935號中可找到一 ναν〇架構陣列及 ^ 統一部分之操作之實例。 、為冗憶體系 該儲存元件陣列被劃分成大量儲存 閃EEPROM糸战二 干鬼。通常對於快 糸統而έ,塊即為擦除單位。 包含可一刼換^ k η 、Ρ ’每一塊皆 除的表小數量之儲存元件。 分為若干個頁面。頁面係程式化單位二塊通常被劃 將個別頁面劃分成多個段,且胃 ^例中可 -V η ^^ 3有作為一基本程Application (IEEE Journal of Solid-State Circuits), Vol. 26, Νο. 4, April 1991, pp. 497-501), which describes a split gate structure with 121185.doc •14-200805382 to similar storage elements A doped polysilicon gate extends over a portion of the channel of the storage element to form a separate select transistor. Both of the above articles are incorporated herein by reference in their entirety. The stylization techniques mentioned in the section "'Nonvolatile Semiconductor Memory Technology" by William D. Brown and Joe E. Brewer (IEEE Press, 1998) 1.2 are also described in the section as applicable to dielectric-electric Charge trapping device, which is incorporated herein by reference. The present invention also uses the storage elements described in this paragraph. Thus, the techniques described herein are also applicable to the coupling between dielectric regions of different storage elements. Another method for storing two bits in each storage element has been described by Eitan et al. in 'NROM: A Novel Localized Trapping, 2-Bit Nonvolatile Memory Cell'' (IEEE Electron Device Letters), Volume 21, No. 11, November 2000, pp. 543-545. A dielectric layer extends across the channel between the source diffusion region and the drain diffusion region. The dielectric material of the other data bit is localized in the dielectric layer adjacent to the source. The multi-state data storage system reads the dielectric space by reading the dielectric space. The binary state of the separate charge storage regions is implemented. The storage elements described in this paragraph can also be used in the present invention. Figure 4 illustrates an example of an array of such NAND storage element arrays as shown in Figures 1-3. Each row, a bit line 206 is coupled to the drain terminal 126 of the drain select gate of the NAND string 150. Along each column of the NAND string, a source line 204 can be connected to the source select gate of the NAND string. Examples of the operation of a ναν〇 architecture array and a uniform portion can be found in U.S. Patent Nos. 5,570,315, 5,774,397, and 121,185, doc -15-200805382 6,046,935. Recall that the array of storage elements is divided into a large number of storage flash EEPROMs. Usually for fast έ, the block is the erase unit. It can be replaced by ^ k η , Ρ 'each block is divided A small number of storage elements are divided into several pages. The page is a stylized unit. The two blocks are usually divided into individual segments, and the stomach can be used as a basic process.

式化知作一次寫入之最少數量之儲存元 /φ φ ^ . I 一列儲存元 仵中通㊆儲存一個或多個資料頁面。一 多個^ 、可儲存一個或 羽&amp;。一扇區包括使用者資料及 i甬堂白紅分4去 自貝枓。開銷資料 I $匕括依據該扇區之使用者資料曾 ,ϋΟΡ、 _L #出的一糾錯碼 (ECC)。控制器(下文所述)之一部分 卜只可叶正程式化至今 陣列中時計算ECC,且亦在正自該陣 &quot; 卞〜,取資料時檢查 ECC。或者,將Ecc及/或其他開銷資料儲存於與其所從屬 之使用者資料不同之頁面甚至不同之塊中。 一使用者資料扇區通常為512個位元組 、、且其相當於磁碟 驅動器内一扇區之大小。開銷資料通常 ^ 附加之16-20 個位元組。大量頁面形成一個塊,例如自 曰Μ固頁面(舉例而 言)至多達32個、64個、128個或更多個wτ # 只曲不4。於某些 實施例中,一列NAND串包括一塊。 於一實施例中,藉由在源極線及位元線淫 尺,予勒之同時將ρ· 井升高至一擦除電壓(例如20伏)達一充 心听間週期並將 一選擇塊之字線接地來擦除記憶體儲存元杜 丁儿仟。由於電容性 耦合,非選擇字線、位元線、選擇線及0源極亦被升高至 121185.doc -16- 200805382 =電堡之一很小部分。由此將_強.電場施加 道化機制將浮動閑極之電子發射至基板側時,選擇儲存 件之資料被擦除。當電子自浮動閘極傳送至㈣^存几 選擇儲存元件之臨限電壓降低。可對整個記憶體陣列 獨的塊、或另一單元之儲存元件實施擦除。 圖5圖解說明一根據本發明之一實施例具有用於並行於 式化儲存元件之一頁面之讀取/寫入電路之記憶體 裝置別。記憶體裝置296可包括一個或多個記憶體晶粒 ⑽。記憶體晶粒298包括一二維儲存元件陣列3〇〇、控制 電路310、及項取/寫入電路泌。於某些實施例中,該儲 存疋件陣列可為三維。記憶體陣列3〇〇可由字線藉由列解 碼器330及由位元線藉由—行解碼器則來定址。讀取/寫 入電路365包括多個感測塊且允許並行讀取或程式化 儲存元件之一頁面。通常,一控制器350以-個或多個呓 憶體晶粒298形式包括於同一記憶體裝置296(例如—可抽 換式儲存卡)中。命令及資料藉由線32〇在主機與控制器 350之間且藉由線318在控制器與一個或多個記憶體晶粒 298之間傳送。 控制電路310與讀取/寫入電路365配合,以對記憶體陣 列300實施記憶體操作。控制電路31〇包括一狀態機312、 一晶片上位址解碼器314、一溫度補償控制315及一功率控 制模組316。下文將特別結合圖14來進一步闡述溫度補償 控制3 1 5。狀態機3 12提供對記憶體操作晶片級控制。晶片 121185.doc 200805382 亡位址解碼器314在主機或記憶體控制器所用位址與解碼 器330及360所用硬體位址之間提供—位址介面。功率控制 模組316控制在記憶體操作期間供給字線及位元線之功率 及電壓。 在某些實施方案中,可組合圖5之某些組件。於不同設 計中,可將除儲存元件陣列300以外的圖5之一個或多個組 件(單獨地或組合地)視為一管理電路。舉例而言,一個或 多個管理電路可包括控制電路31〇、狀態機312、解碼器 3 14/360、功率控制316、感測塊4〇〇、讀取/寫入電路%^ 控制為3 5 0專中的任一者或其一組合。 圖6圖解說明圖5中所示記憶體裝置296之另一佈置。各 種周邊電路對記憶體陣列300之存取係以對稱形式在該陣 歹J之對置側上實施,由此將每一側上之存取線及電路之密 度減半。因此,列解碼器分裂為列解碼器330A及 330B, 行解碼器分裂為行解碼器36〇A及36〇b。類似地,讀取/寫 入電路分裂為自陣列300底部連接至位元線之讀取/寫入電 路365A及自陣列300頂部連接至位元線之讀取/寫入電路 365B。以此方式,該等讀取/寫入模組之密度實質上減 半。圖6之裝置亦可包括一如上文關於圖5之裝置所述之控 制器。 圖7係一個別感測塊400之方塊圖,該感測塊被分區成一 稱作感測模組380之核心部分及一共用部分39〇。於一實施 例中,每一位元線將具有一單獨感測模組38〇及一組多個 感測模組380將具有一個共用部分390。於一實例中,一感 121185.doc •18· 200805382 測塊將包括一共用部分390及8個感測模組380。一群組中 之每一感測模組皆藉由一資料匯流排372與相關聯之共用 部分通信。關於其他細節,可參考2004年12月29日申請且 標稱為 ’’Non-Volatile Memory &amp; Method with Shared Processing for an Aggregate of Sense Amplifiers” 之美國 專利申請案ll/026,536,該申請案以全文引用的方式併入 本文中。The minimum number of storage elements known as a write once /φ φ ^ . I A list of storage elements 仵中通七 stores one or more data pages. One or more ^, can store one or feather &amp; One sector includes user data and i甬堂白红分4 is taken from Bellow. The overhead data I $ includes an error correction code (ECC) from the user data of the sector, ϋΟΡ, _L #. One part of the controller (described below) can only be programmed to calculate the ECC in the array, and it is also checking the ECC from the array &quot; 卞~. Alternatively, store Ecc and/or other overhead data in a different page than the user profile to which it belongs. A user data sector is typically 512 bytes and is equivalent to the size of a sector within the disk drive. The overhead data is usually ^16-20 bytes appended. A large number of pages form a block, for example from a tamping page (for example) up to 32, 64, 128 or more wτ # 曲曲不4. In some embodiments, a column of NAND strings includes one block. In one embodiment, the ρ·well is raised to an erase voltage (eg, 20 volts) by a source line and a bit line, and the ρ·well is raised to an audible period and a selection is made. The word line of the block is grounded to erase the memory storage element Du Dinger. Due to capacitive coupling, the unselected word lines, bit lines, select lines, and source 0 are also raised to 121185.doc -16- 200805382 = a small portion of the electric castle. Thus, when the _strong. electric field is applied to the doping mechanism to emit the electrons of the floating idler to the substrate side, the data of the selected storage device is erased. When the electrons are transferred from the floating gate to (4), the threshold voltage of the selected storage element is lowered. Erasing can be performed on a single block of the entire memory array or a storage element of another unit. Figure 5 illustrates a memory device having read/write circuits for parallelizing one of the pages of the storage element in accordance with one embodiment of the present invention. Memory device 296 can include one or more memory dies (10). The memory die 298 includes a two-dimensional array of storage elements 3, a control circuit 310, and a term fetch/write circuit. In some embodiments, the array of storage elements can be three dimensional. The memory array 3 can be addressed by the word line by the column decoder 330 and by the bit line by the line decoder. The read/write circuit 365 includes a plurality of sense blocks and allows one page of storage elements to be read or programmed in parallel. Typically, a controller 350 is included in the same memory device 296 (e.g., a removable memory card) in the form of one or more memory die 298. Commands and data are transferred between the host and controller 350 via line 32 and between the controller and one or more memory dies 298 via line 318. Control circuit 310 cooperates with read/write circuit 365 to perform memory operations on memory array 300. The control circuit 31 includes a state machine 312, an on-chip address decoder 314, a temperature compensation control 315, and a power control module 316. The temperature compensation control 3 15 will be further explained below in particular in conjunction with FIG. State machine 3 12 provides wafer level control of memory operations. Wafer 121185.doc 200805382 The dead address decoder 314 provides an address interface between the address used by the host or memory controller and the hardware address used by the decoders 330 and 360. Power control module 316 controls the power and voltage supplied to the word lines and bit lines during memory operation. In some embodiments, certain components of Figure 5 can be combined. In a different design, one or more of the components of Figure 5 (alone or in combination) other than storage element array 300 can be considered a management circuit. For example, one or more management circuits may include control circuit 31, state machine 312, decoder 3 14/360, power control 316, sense block 4, read/write circuit %^ control to 3 Any of the 50 or any combination thereof. FIG. 6 illustrates another arrangement of the memory device 296 shown in FIG. The access of the various peripheral circuits to the memory array 300 is performed in a symmetrical form on the opposite side of the array J, thereby halving the density of the access lines and circuits on each side. Therefore, the column decoder is split into column decoders 330A and 330B, and the row decoder is split into row decoders 36A and 36B. Similarly, the read/write circuit is split into a read/write circuit 365A connected from the bottom of the array 300 to the bit line and a read/write circuit 365B connected from the top of the array 300 to the bit line. In this way, the density of the read/write modules is substantially halved. The apparatus of Figure 6 can also include a controller as described above with respect to the apparatus of Figure 5. Figure 7 is a block diagram of an additional sensing block 400 that is partitioned into a core portion called a sensing module 380 and a shared portion 39A. In one embodiment, each bit line will have a single sensing module 38 and a plurality of sensing modules 380 will have a common portion 390. In one example, a sense 121185.doc •18· 200805382 test block will include a common portion 390 and eight sensing modules 380. Each of the sensing modules in a group communicates with an associated shared portion via a data bus 372. For further details, reference is made to U.S. Patent Application Serial No. </RTI> </RTI> </RTI> </RTI> <RTIgt; </RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> <RTIgt; The manner of reference is incorporated herein.

感測模組380包括感測電路370,該電路確定一所連接位 元線.中之導電電流是高於還是低於一預定臨限位準。感測 模組380亦包括一位元線鎖存器382,該鎖存器用於設定所 連接位元線上之一電壓狀態。舉例而言,鎖存於位元線鎖 存器382中之一預定狀態將會導致所連接位元線被拉至一 指定程式化禁止之狀態(例如Vdd)。 /、用邛刀390包括一處理器392、一組資料鎖存器394 一耦合於資料鎖存器394組與資料匯流排32〇之間的ι/〇 =外6。處理器392實施計算。舉例而言,其功能之一係 疋儲存於所感測儲存元件中之f料並將經確定之資料健 於貧㈣存m資料鎖存器394組用於儲存在讀取; :期間由處理器392確定之資料位元。其亦用於儲存在i 二化#作期間自資料匯流排320輸入之資料位元。輸入」 L料Λ元表示欲程式化至記憶體内之寫人資料。1/0介1 貝科鎖存器394與資料匯流排32()之間提供—介面。 之下項取或感測期間’該系統之操作處於狀態機3i2控, 6亥狀態機控制將不同之控制閘極電堡供給已定址j ^21185.doc -19- 200805382 儲存元件。當感測模組380步進穿過各種對應於由記憶體 所支援之各種記憶體狀態之預定控制閘極電壓時,其可在 4龟C之下跳閘且藉由匯流排3 7 2將一輪出自感測模 ㈣〇提供至處理器392。此時,處理器392藉由慮及感測 模組之跳閘事件及關於自狀態機藉由由輸入線393所施加 之㈣閘極電壓之資訊來確定結果記憶體狀態'然後其將 什异该記憶體狀態之-二進製編碼並將該結果f料位元儲 存至資料鎖存器394中。於核心部分之另-實施例中,位 元線鎖存器382具有雙重❹,即可充# —用於鎖存感測 核組380之輸出之鎖存器亦可充當一如上文所述之 鎖存器。 預』某些實施方案將包括多個處理器392。於一實施例 中’每-處理器392皆將包括一輸出線(其未緣示於圖 =使遠等輸出線中之每一者皆經連線&quot;或&quot;在一起。於某些 在連接至經連線之”或”線前反轉該等輸出線。 -成°在程式化驗證過程期間作出一程式化過程何時 確定,此乃因接收經連線&quot;或&quot;之狀態機可確定 …:程式化之位元何時達到所期望之位準。舉例而 州元皆達到其所期望之位準時,該位元之—邏 皆輸出二0二連線之”或&quot;線(或反轉資料υ°當所有位元 貝料〇(或反轉資料η拉 化過程。由於— 、、’、彳該狀態機知曉終止程式 狀離機/理器皆與8個感測模組通信,因此兮 狀悲、機需要對經連 u此口亥 器392添加邏輯以 行8次讀取,或者向處理 ,、、關聯位元線之結果以使狀態機僅 121185.doc -20- 200805382 需對經連線之”或”線進行一次讀取。類似地,藉由正確選 擇邏輯位準,全局狀·態機可偵測第一位元何時改變其狀態 並相應地改變演算法。 在程式化或驗證期間,欲程式化之資料自資料匯流排 320儲存於資料鎖存器394組中。在狀態機控制下,程式化 操作包含將一系列程式化電壓脈動施加至已定址儲存元件 之控制閘極上。在每一程式脈動之後進行讀回(驗證),以 確定該儲存元件是否已程式化成所期望之記憶體狀態。處 理器392監控相對於所期望之記憶體狀態讀回之記憶體狀 態。當兩者一致時,處理器222設定位元線鎖存器214,以 將名位兀線拉至一指定程式化禁止之狀態。由此禁止耦合 至忒位7G線之儲存元件進一步程式化,即使在程式化脈動 出現在其控制閘極上時亦如此。於其他實施例中,該處理 器最初載入位元線鎖存器382且該感測電路在驗證過程期 間將該鎖存器設定至一禁止值。 /㈣存H堆疊394包含—對應於感測模組之資料鎖存 器隹且於只施例中,每一感測模組380具有三個資料 鎖存裔。於某些實施方案(但並非所需)中,將資料鎖存哭 實施為—移位暫存11以便將其中所儲存之並行資料轉換成 :枓匯流排320之串行資料,反之亦然。在該較佳實施例 中,可將對應於m個儲在 # 存疋件之項取/寫入塊的所有資料鎖 存器鏈接在一起以形成一 攻塊移位寄存态,以便可藉由串行 傳运來輸入或輸出一資料 Λ ^ 貝枓鬼。特疋而言,採用r個讀取/寫 入权組之排,以使其資 义^、貝枓鎖存為組中的每一者依序將資料 121185.doc -21- 200805382 移入或移出資料匯流排,仿佛其係一用於整個讀取/寫入 塊之移位寄存器之一部分一般。 關於非揮發性記憶體裝置各種實施例之結構及/或操作 之附加資訊可在以下專利中找到:(1)2004年3月25日公開 的第2004/0057287號美國專利公開申請案&quot;Non-Volatile Memory And Method With Reduced Source Line Bias * Errors ; (2)2004年6月10日公開的第2004/0109357號美國公 開專利申請案 ’’Non-Volatile Memory And Method with Improved Sensing” ;(3)2004 年 12 月 16 日申請的標稱為 &quot;Improved Memory Sensing Circuit And Method For Low Voltage Operation”之第11/015,199號美國專利申請案,發 明者為Raul-Adrian Cernea; (4)2005年4月5曰申請的標稱為 ’’Compensating for Coupling During Read Operations of Non-Volatile Memory”之第11/099,133號美國專利申請案, 發明者為Jian Chen ;及(5)2005年12月28日申請的標稱為 ’’Reference Sense Amplifier For Non-Volatile Memory”之第 V 11/321,953號美國專利申請案,發明者為Siu Lung Chan及Sensing module 380 includes a sensing circuit 370 that determines whether the conductive current in a connected bit line is above or below a predetermined threshold level. Sensing module 380 also includes a bit line latch 382 for setting a voltage state on a connected bit line. For example, latching in a predetermined state in bit line latch 382 will cause the connected bit line to be pulled to a specified stylized inhibit state (e.g., Vdd). The boring tool 390 includes a processor 392, a set of data latches 394 coupled to the ι/〇 = outer 6 between the data latch 394 group and the data bus 32 〇. Processor 392 performs the calculations. For example, one of its functions is to store the material stored in the sensed storage element and to stabilize the determined data in the lean (four) memory data latch 394 group for storage in the read; 392 identified data bits. It is also used to store the data bits input from the data bus 320 during the i-times. The input "L" element indicates the writer's data to be programmed into the memory. The 1/0 interface 1 provides a interface between the Beckey latch 394 and the data bus 32 (). During the lower item fetch or sensing period, the operation of the system is in the state machine 3i2 control, and the 6-Hai state machine control supplies the different control gates to the stored address j^21185.doc -19-200805382 storage element. When the sensing module 380 steps through a variety of predetermined control gate voltages corresponding to various memory states supported by the memory, it can trip under 4 turtles C and a round by bus bar 3 7 2 The sensor module (4) is supplied to the processor 392. At this time, the processor 392 determines the result memory state by considering the trip event of the sensing module and the information about the (four) gate voltage applied by the state machine 393 from the state machine. Then it will be different. The binary state of the memory state is stored in the data latch 394. In another embodiment of the core portion, the bit line latch 382 has a double turn, that is, the latch for latching the output of the sense core group 380 can also function as described above. Latches. Some embodiments will include multiple processors 392. In one embodiment, 'per-processor 392 will include an output line (which is not shown in the figure = making each of the far output lines connected by &quot; or &quot; together. Inverting the output lines before connecting to the "or" line of the connection. - When a stylization process is made during the stylization verification process, which is due to the status of the received connection &quot; or &quot; The machine can determine... when the stylized bit reaches the desired level. For example, when the state is at the level it expects, the bit-logic of the bit outputs the 222 line or the line. (or reverse the data υ ° when all the bits of the material 〇 (or reverse the data η pull process. Because -,, ', 彳 the state machine knows to terminate the program-like off-machine / processor and 8 sense modules Group communication, so the sorrow, the machine needs to add logic to the connection 392 to read 8 times, or to process, ,, correlate the result of the bit line to make the state machine only 121185.doc -20 - 200805382 A read of the connected "or" line is required. Similarly, by correctly selecting the logical level, all The state machine can detect when the first bit changes its state and change the algorithm accordingly. During the stylization or verification, the data to be programmed is stored in the data latch 394 from the data bus 320. Under state machine control, the stylization operation involves applying a series of programmed voltage ripples to the control gate of the addressed storage element. After each program pulse is read back (verified) to determine if the storage element has been programmed into a The desired memory state. The processor 392 monitors the memory state read back relative to the desired memory state. When the two match, the processor 222 sets the bit line latch 214 to pull the name line. The state of the stylization inhibit is specified to a state that prohibits the storage element coupled to the 7G line from being further programmed, even when the programmed ripple occurs on its control gate. In other embodiments, the processor is initially The bit line latch 382 is loaded and the sense circuit sets the latch to a disable value during the verify process. /(4) The memory H stack 394 includes - corresponding to the sense module Data latches, and in the only embodiment, each sensing module 380 has three data latches. In some embodiments (but not required), the data latching cry is implemented as a shift The temporary storage 11 converts the parallel data stored therein into: serial data of the bus bar 320, and vice versa. In the preferred embodiment, the items corresponding to the m stored in the # memory can be taken. All data latches of the / write block are linked together to form a tap shift register state so that a data can be input or output by serial transport. In other words, r is used. A row of read/write rights groups, so that their assets, and the latches are latched into each group, and the data 121185.doc -21- 200805382 is sequentially moved into or out of the data bus, as if it were One part of the shift register for the entire read/write block is general. Additional information regarding the structure and/or operation of various embodiments of the non-volatile memory device can be found in the following patents: (1) U.S. Patent Application Serial No. 2004/0057287, issued March 25, 2004, &quot; -Volatile Memory And Method With Reduced Source Line Bias * Errors; (2) US Patent Application No. 2004/0109357, published on Jun. 10, 2004, 'Non-Volatile Memory And Method with Improved Sensing'; (3) U.S. Patent Application Serial No. 11/015,199, filed on Dec. 16, 2004, entitled &quot;Improved Memory Sensing Circuit And Method For Low Voltage Operation, by Raul-Adrian Cernea; (4) 2005 US Patent Application No. 11/099, 133, entitled "'Compensating for Coupling During Read Operations of Non-Volatile Memory", filed on April 5, the inventor is Jian Chen; and (5) December 28, 2005 U.S. Patent Application Serial No. U.S. Patent Application Serial No. Serial No. No. No. No. No. No. No. No. No.

Raul-Adrian Cernea。上文剛剛列出之所有五個專利文獻 皆以全文引用的方式併入本文中。 參見圖8,闡述儲存元件陣列300之一實例性結構。作為 一實例,闡述一分區成1,〇24個塊之NAND快閃EEPROM。 可同時擦除儲存於每一塊中之資料。在一實施例中,該塊 係同時受到擦除之儲存元件之最小單位。於每一塊中,於 此實例中,存在8,512個對應於位元線BL0、 121185.doc -22- 200805382 ^ BL8511之4亍。於—稱作所有位元線(ABL)架構之 λ把例中,可在讀取及程式化操作期間同時選擇—塊之所 有位元線。可同時程式化沿一共用字線並連接至任一位元 線之儲存元件。 、圖8顯示四個串聯連接以形成一 Nand串之儲存元件。雖 然圖中顯示每_NAND串中包括四㈣存元件,但也可使 用:於=少於四個儲存元件(例如,16個、32個、“個或 )°亥NAND串的-個端子藉由-汲極選擇閘極 ’、接至選擇閘極汲極線SGD)連接至一對應位元線,而 另個端子糟由-源極選擇閘極(其連接至選擇閉極源極 線SGS)連接至^源極。 —於稱作奇偶架構之另一實施例中,如圖9中所示,將位 儿線劃分成偶數位元線及奇數位元線。圖9圖解說明一將 .己體陣列組織成一奇偶記憶體架構之塊之實例。於一 奇數/偶數位元線架構中,同時程式化沿一共用字線並連 接=奇數位元線之儲存元件,而在另一時間程式化沿一共 用子線並連接至偶數位元線之儲存元件。可將資料程式化 不同之塊中並可自不同之塊同時讀取資料H塊 /中在此κ例中,存在8,512個劃分成偶數行及奇數行之 行。位元線亦書]分# 一J刀成偶數位元線(BLe)及奇數位元線 (BLo)。於此實例中 T 顯不四個串聯連接以形成一 NAND串 諸存疋件儘官圖中顯示在每一 NAND串中包括四個儲 存元:,然而,亦可使用多於或少於四個儲存元件。 在讀取及程式化提从 才呆作之一構造期間,同時選擇4,256個 121185.doc -23- 200805382Raul-Adrian Cernea. All five patent documents just listed above are incorporated herein by reference in their entirety. Referring to Figure 8, an exemplary structure of an array of storage elements 300 is illustrated. As an example, a NAND flash EEPROM with a partition of 1, 24 blocks is illustrated. The data stored in each block can be erased at the same time. In one embodiment, the block is simultaneously subjected to the smallest unit of the erased storage element. In each block, in this example, there are 8,512 corresponding to the bit line BL0, 121185.doc -22-200805382 ^ BL8511. In the λ example of all bit line (ABL) architectures, all bit lines of the block can be selected simultaneously during read and program operations. A storage element along a common word line and connected to any bit line can be programmed simultaneously. Figure 8 shows four storage elements connected in series to form a Nand string. Although the figure shows that four (four) memory elements are included in each _NAND string, it can also be used: - less than four storage elements (for example, 16, 32, "ones") The gate is selected from the drain gate, connected to the selected gate drain line SGD, and connected to a corresponding bit line, and the other terminal is connected to the source select gate (which is connected to the selected closed source line SGS). Connected to the ^ source. - In another embodiment called the parity structure, as shown in Figure 9, the bit lines are divided into even bit lines and odd bit lines. Figure 9 illustrates one will. An example of a block of a body array organized into an odd-even memory structure. In an odd-numbered/even-bit line architecture, a storage element along a common word line and connected to an odd bit line is simultaneously programmed, while in another time program A storage element along a common sub-line and connected to an even bit line. The data can be programmed into different blocks and the data can be read simultaneously from different blocks. In this κ case, there are 8,512 partitions. It is an even-numbered line and an odd-numbered line. The bit line is also a book] minute #一J刀成偶数线线(BLe) and odd Bit line (BLo). In this example, T is not connected in series to form a NAND string. The memory map includes four memory elements in each NAND string: however, it can also be used. More or less than four storage elements. During the construction of reading and stylization, select 4,256 121185.doc -23- 200805382

儲存兀件。所選擇之儲存元件具有相同之字線及相同種類 之位元線(例如,偶數位元線或奇數位元線)。因此,可同 時讀取或程式化532個資料位元組(其形成一邏輯頁面),而 :個記憶體塊可儲存至少8個邏輯頁面(四個字線,每一個 白具有可數邏輯頁面與偶數邏輯頁面)。對於多狀態儲存 兀件,當每一儲存元件儲存兩個資料位元,其中該兩個位 ,之每個白儲存於一不同頁面中時,一個塊儲存16個邏 輯頁面。亦可使用其他尺寸之塊及頁面。 對於ABL或奇偶架構,儲存元件可藉由將卜井升高至一 擦除電壓(例如’ 2G伏)並將—選擇塊之字線接地來加以擦 除源極線及位兀線浮動。可冑整個記憶體陣Μ、單獨塊 或該記憶體裝詈$ _立β八+ μ七_ α 。販衣罝之部分之儲存兀件之另一單元實施擦 除電子自儲存凡件之浮動閘極傳送至ρ-井區以使儲存元 件之Vth變為負。 八於讀取及驗證操作中,選擇閘極(SGD及SGS)連接至一 介於2.5伏至4.5伏範圍内之電壓且非選擇字線(例如當WL2 係選擇字線時’其為WLQ、WUawl3)升高至—讀取通過 電麼(通常為—介於4.5伏至6伏範圍内之㈣)以使電晶體 運作為傳遞閘極。選擇字線WL2連接至—電麼,該電遷之 位準係針對每—讀取及驗證操作而規^,以便確^相關儲 存疋件之VTH是高於還是低於此位準。舉例而言,於一針 對兩位準儲存元件之讀取操作中,可將選擇字線和接 地,以偵測VTH是否高於〇伏。於一針對兩位準儲存元件之 驗證操作中’可將選擇字線WL2連接至(舉例而言)〇8伏, 121185.doc •24- 200805382 以驗證vTH是否已達到至少心8伏。源極及井處於〇伏下。 將選擇位元線(假定為偶數位元線(BLe))預充電至一(舉例 而言)〇·7伏之位準。若Vth高於該字線上之讀取或驗證位 準’則與所關注儲存元件相關聯之位元線(BLe)之潛在位 準因非導電性儲存元件而維持高位準。另一方面,若該 Vth低於讀取或驗證位準,則相關位元線⑺[匀之潛在位準 會因導電性儲存元件使位元線放電而降至一例如低於〇.5 伏之低位準。因此,儲存元件之狀態係由一連接至位元線 之電壓比較感測放大器來偵測。 根據此項技術中之習知之技術實施上述擦除、讀取及驗 證操作。因此,熟習此項技術者可改變所解釋之許多細 即。亦可使用此項技術中習知之其他擦除、讀取及驗證技 術。 圖1 0圖解說明當每一儲存元件儲存兩個資料位元時,該 儲存元件陣列之實例性臨限電壓分佈。提供經擦除儲存元 件之一第一臨限電壓分佈E。亦繪示經程式化儲存元件之 ,個臨限電壓分佈A、BAC。於一實施例中,£分佈中之 L限電壓為負’而A、B&amp;C分佈中之臨限電壓為正。 每一不同臨限電壓範圍介對應於該組資料位元之預定 值。程式化至儲存元件巾之:諸與該料元件之臨限電壓 位準之間的具體關係相依於針對儲存元件所採用之資料編 碼方案。舉例而言’美國專利第6,222,762號及⑽4年⑽ ^日公開的美國專料開巾請㈣彻55_號闡述各 用於多狀恶快閃儲存元件之資料編碼方案’此兩個申請 121185.doc -25· 200805382 案皆以全文引用的方式併入本文中。於一實施例中, -格雷碼指派方案來為臨限電壓範圍指派資料值,以便若 二動閑極之臨限電壓錯誤地偏移至其相鄰物Save your mail. The selected storage elements have the same word line and the same type of bit line (e.g., even bit lines or odd bit lines). Therefore, 532 data bytes can be simultaneously read or programmed (which form a logical page), and: a memory block can store at least 8 logical pages (four word lines, each with a countable logical page) With even logical pages). For multi-state storage, when each storage element stores two data bits, each of which is stored in a different page, one block stores 16 logical pages. Blocks and pages of other sizes can also be used. For an ABL or parity architecture, the storage element can be used to erase the source line and the bit line floating by raising the well to an erase voltage (e.g., &lt; 2G volts) and grounding the word line of the -select block. You can install the entire memory array, a separate block, or the memory with $_立β八+ μ七_α. Another unit of the storage element of the clothing compartment is implemented to remove the floating gate of the electronic self-storing piece to the p-well zone to make the Vth of the storage element negative. In the read and verify operations, the select gates (SGD and SGS) are connected to a voltage between 2.5 volts and 4.5 volts and the unselected word lines (eg, when WL2 selects the word line), it is WLQ, WUawl3 ) Raise to - read through the electricity (usually - (4) in the range of 4.5 volts to 6 volts) to operate the transistor as a transfer gate. The selection word line WL2 is connected to the power, and the level of the electromigration is determined for each read and verify operation to determine whether the VTH of the associated memory element is above or below this level. For example, in a read operation of a pair of quasi-storage elements, the word line and ground can be selected to detect if the VTH is above the stagnation. The select word line WL2 can be connected to, for example, 〇8 volts, 121185.doc •24-200805382 in a verify operation for two quasi-storage elements to verify that the vTH has reached at least 8 volts. The source and well are under crouching. The selected bit line (assumed to be an even bit line (BLe)) is precharged to a level of, for example, 〇·7 volts. If Vth is above the read or verify level on the word line, then the potential level of the bit line (BLe) associated with the storage element of interest maintains a high level due to the non-conductive storage element. On the other hand, if the Vth is lower than the read or verify level, the associated bit line (7) [the potential level of the uniformity will be lowered by the conductive storage element to discharge the bit line to, for example, less than 〇5 volts. The low level. Therefore, the state of the storage element is detected by a voltage comparison sense amplifier connected to the bit line. The above erase, read and verify operations are performed in accordance with techniques well known in the art. Therefore, those skilled in the art can change many of the details explained. Other erasing, reading and verifying techniques known in the art can also be used. Figure 10 illustrates an exemplary threshold voltage distribution of the array of storage elements as each storage element stores two data bits. A first threshold voltage distribution E of one of the erased storage elements is provided. It also shows the threshold voltage distribution A and BAC of the stylized storage element. In one embodiment, the L-limit voltage in the £ distribution is negative and the threshold voltage in the A, B &amp; C distribution is positive. Each of the different threshold voltage ranges corresponds to a predetermined value of the set of data bits. Stylized to the storage component: The specific relationship between the threshold voltage levels of the component is dependent on the data encoding scheme used for the storage component. For example, 'U.S. Patent No. 6,222,762 and (10) 4 years (10) ^, the United States specializes in the opening of the towel, please (4) complete 55_ to explain the data encoding scheme for each of the multi-equivalent flash memory storage components' two applications 121185. Doc-25·200805382 is incorporated herein by reference in its entirety. In an embodiment, the Gray code assignment scheme assigns a data value to the threshold voltage range so that if the threshold voltage of the second active erroneously erroneously shifts to its neighbor

僅會影響一個位元。一個眚如仏价 J 個實例給臨限電壓範圍E (狀態e )指 給臨限電壓範圍A(狀態A)指派&quot;1〇&quot;,給臨 範圍B(狀態B)指派,,〇〇,,,$ ^ 並6 5品限電壓範圍C(狀態C)指派 。…、'而’在其他實施例中,不使用 四種狀態,但本發明亦可用於其他多態結構’包括== 括多於或少於四種狀態之結構。 提供三個讀取參考電MVra、Vrb及vrc以自儲存元㈣ 取資料。藉由測試-既定儲存元件之臨限電屢是高於還是 低於vmuw,該系統可確㈣儲存元件處於何種 狀態。 此外,還提供三個驗證參考„vva、Vvwvve。當將 儲存元件程式化至狀態八時,該系統將測試彼等儲^件 是具有一大於Vva還是一等於Vva之臨限電壓。當將儲存 元件程式化至狀態B時,該系統將測試該等儲存元件是具 有大於還是等於Vvb之臨限電屢。當將儲存元件程式化= 狀態C時,該系統將測試儲存元件是具有大於還是等於 Vvc之臨限電壓。 、 於稱作全序列程式化之實施例中,可將健存元件自擦 除狀態E直接程式化至程式化狀態A、8或匚中之任一者f 舉例而口 ’可首先擦除一欲程式化之儲存元件群體,以使 該群體中的所有儲存元件皆處於擦除狀態e。然後,將使 121185.doc -26- 200805382 用一系列諸如由圖13之控制閘極電壓序列所繪示之程式化 脈動將儲存元件直接程式化成狀態A、B或C。當某些儲存 元件正自狀態E程式化至狀態A時,其他儲存元件正自狀 態E程式化至狀態B及/或自狀態E程式化至狀態c。當在 WLn上自狀態E程式化至狀態c時,可最大化耦合至wLn-1 下方®比鄰浮動閘極之寄生耗合量,此乃因與自狀態E程式 化至狀態A或自狀態E程式化至狀態b時之電壓變化相比, WLn下方浮動閘極上之電荷量變化最大。當自狀態E程式 化至狀悲B時,耦合至毗鄰浮動閘極之耦合量減小但仍很 高。當自狀態E程式化至狀態A時,該耦合量更進一步減 小。因此,隨後讀取WLn-Ι之每一狀態所需之校正量將端 視WLn上之毗鄰儲存元件之狀態而異。 圖11圖解說明一程式化多狀態儲存元件之兩遍技術之實 例,该多狀態儲存元件儲存有兩個不同頁面(一下頁面及 上頁面)之資料。所繪示之四種狀態係:狀態Ε(1ι)、狀 ^AdO)、狀態B(〇〇)、狀態c(〇1)。對於狀態£,兩個頁面 白儲存 1 。對於狀態A,下頁面儲存一 ”〇”而上頁面儲 存一 ”1”。對於狀態B,兩個頁面皆儲存”〇,,。對於狀態c, 亇諸存1而上頁面儲存” 〇 ”。注意,雖然給該等狀態 之每者扎派了特定位元圖案,但亦可指派不同之位元圖 案。 ° 於,第_、鱼口 、 、私式化中,根據欲程式化至下邏輯頁面中之 位來6又疋該儲存元件之臨限電壓位準。若彼位元係—邏 軏T,則該臨限電壓會由於其處於因先前受到擦除而得 121185.doc -27- 200805382 到之適宜狀態中而不會發生改變。然而,若欲程式化之位 $為-邏輯”G”,則該儲存元件之臨限位準增加而成為狀 態A,如箭頭U00所示。此會終止該第一遍程式化。 於-第二遍程式化中,根據正程式化至上邏輯頁面中之 位元來設定該儲存元件之臨限電壓位準。若上邏輯頁面位 元欲儲存一邏輯&quot;Γ,,則不會發生程式化,此乃因該儲存 元件係端視對下頁面位元之程式化而處於狀態£或八 者皆攜帶-上頁面位元&quot;n之一者中。若上邏輯頁面位元 欲成為-邏輯,,〇&quot;’則該臨限電壓偏移。若該第一遍使儲 存元件保持處於擦除狀態E中,則於該第二階段中,該儲 存元件被程式化,以使該臨限電壓增加而處於狀態c中, 如箭頭1120所纷示。若作為第一遍程式化之結果該儲存元 件已被程式化成狀態A,則該儲存元件在該第二遍中被進 一步程式化,以使該臨限電壓增加而處於狀態6中,如箭 頭1110所繪示。第二遍之結果係、欲將該儲存元件程式化二 指定用來儲存上頁面之一邏輯&quot;〇&quot;而不改變下頁面之資料 之狀態。於圖10及圖u兩者中’輕合至晚鄰字線上之浮動 閘極之耦合量取決於最終狀態。 於一實施例中,若欲寫入足以填滿一整個頁面之資料, 則可設置-系統來實施完全序列寫人。若紋夠之資料寫 入一整個頁面,則該程式化過程可以所接收之資料來程式 化下頁面。當接收到後續資料時,系統則程式化上頁面。 於再一實施例中,該系統可開始以程式化下頁面之模式進 行寫入且若隨後接收到足以填滿—字線之儲存元件之全部 121185.doc -28 - 200805382 或大部之資料時,則轉換成完全序列程式化模式。此實施 例之更多細節揭示於發明者Sergy A· Gorobets及Yan Li於 2004年12月14日提出申請的標稱為&quot;Pipe lined Programming of Non-Volatile Memories Using Early Data’,之第 ll/〇13,125號美國專利申請案中,該申請案以全文引用的 方式併入本文中。Only one bit is affected. For example, a J example of a price for a threshold voltage range E (state e) means assigning a threshold voltage range A (state A) to &quot;1〇&quot;, assigning a range B (state B), 〇〇 ,,,$^ and 6 5 limit voltage range C (state C) assignment. In other embodiments, four states are not used, but the invention is also applicable to other polymorphic structures' including == structures that include more or less than four states. Three read reference MVra, Vrb and vrc are provided to retrieve data from the storage element (4). By testing - if the threshold of a given storage component is repeatedly above or below vmuw, the system can determine (4) what state the storage component is in. In addition, three verification references are provided, „vva, Vvwvve. When the storage elements are programmed to state eight, the system will test whether their storage has a threshold voltage greater than Vva or equal to Vva. When stored When the component is programmed to state B, the system will test whether the storage component has a power limit greater than or equal to Vvb. When the storage component is programmed = state C, the system will test whether the storage component has greater than or equal to The threshold voltage of Vvc. In the embodiment called full sequence programming, the self-erase state E of the memory component can be directly programmed to any of the stylized states A, 8 or f. 'You can first erase a population of staging storage elements so that all storage elements in the group are in the erased state e. Then, 121185.doc -26- 200805382 will be controlled by a series such as shown in Figure 13. The stylized pulsation depicted by the gate voltage sequence directly programs the storage element into state A, B, or C. When some storage elements are programmed from state E to state A, the other storage elements are programmed from state E to State B and/or from state E to state c. When staging from state E to state c on WLn, it maximizes the parasitic consumption of coupling to the adjacent floating gate of wLn-1. The amount of charge on the floating gate below WLn varies the most from the voltage change from state E to state A or from state E to state b. When staging from state E to sorrow B, coupling The coupling amount to the adjacent floating gate is reduced but still high. When the state E is programmed to state A, the coupling amount is further reduced. Therefore, the correction required for each state of WLn-Ι is subsequently read. The amount will vary depending on the state of the adjacent storage elements on WLn. Figure 11 illustrates an example of a two-pass technique for a stylized multi-state storage element that stores two different pages (the next page and the previous page) The four states are: state Ε (1ι), shape ^AdO), state B (〇〇), state c (〇1). For the state £, the two pages are stored 1 for white. State A, the next page stores a "〇" and the upper page stores a "1". For State B, both pages are stored "square ,,. For state c, save the file and save the page "”". Note that although a particular bit pattern is assigned to each of these states, a different bit pattern can be assigned. °, in the _, fish mouth, and privateization, according to the position to be programmed to the next logical page, the threshold voltage level of the storage element is 6 again. If the bit system is - T, the threshold voltage will not change because it is in the appropriate state due to the previous erasure of 121185.doc -27-200805382. However, if the bit to be programmed is $-logic "G", then the threshold level of the storage element is increased to become state A, as indicated by arrow U00. This will terminate the first pass of stylization. In the second pass stylization, the threshold voltage level of the storage element is set according to the bit in the normalized upper logical page. If the upper logical page bit wants to store a logical &quot;Γ, then no stylization will occur, because the storage component is viewed in the stylization of the next page bit and is in the state of £ or all of them are carried - One of the page bits &quot;n. If the upper logical page bit is to be - logical, 〇 &quot;' then the threshold voltage offset. If the first pass keeps the storage element in the erased state E, then in the second phase, the storage element is programmed to increase the threshold voltage to be in state c, as indicated by arrow 1120. . If the storage element has been programmed into state A as a result of the first programming, then the storage element is further programmed in the second pass to increase the threshold voltage to state 6 as indicated by arrow 1110. Drawn. The result of the second pass is that the storage element is to be programmed to store the logic of one of the pages &quot;〇&quot; without changing the state of the information on the next page. In Figure 10 and Figure u, the amount of coupling of the floating gate to the adjacent word line depends on the final state. In one embodiment, if a data sufficient to fill an entire page is to be written, then the system can be set to implement a full sequence of writers. If the enough data is written into an entire page, the stylization process can program the page with the received data. When the follow-up data is received, the system stylizes the page. In still another embodiment, the system can begin writing in a stylized page mode and if subsequently received enough to fill all of the 121185.doc -28 - 200805382 or most of the storage elements of the word line , then convert to full sequence stylized mode. Further details of this embodiment are disclosed in the "Pipe lined Programming of Non-Volatile Memories Using Early Data', filed on December 14, 2004 by the inventors Sergy A. Gorobets and Yan Li. In U.S. Patent Application Serial No. 13,125, the disclosure of which is incorporated herein in its entirety by reference.

圖12A-C揭示另一用於程式化非揮發性記憶體之過程, 對於任一特定儲存元件,其藉由在針對先前頁面寫入毗鄰 儲存元件後相關於一特定頁面寫入至彼特定儲存元件來減 小浮動閘極至浮動閘極之耦合效應。於一實例性實施方案 中,非揮發性儲存元件使用四種資料狀態來針對每一儲存 元件儲存兩個資料位元。舉例而言,假定狀態£係擦除狀 態,而狀態A、B及C係程式化狀態。狀態£儲存資料u。 狀態A儲存資料〇1。狀態B儲存資料1〇。狀態c儲存資料 〇〇。此係-非格雷編碼之實例,此乃因兩個位元皆在眺鄰 狀態A與B之間變化。亦可使用資料狀態至物理資料狀態 之其他編碼。每-儲存元件儲存兩個資料頁面。出於參^ 之目的,將此等資料頁面稱作上頁面及下頁面;然而亦可 賦予其其他標記。對於狀態A,上頁面儲存位元〇而下 儲存位元1 〇對於狀雜P u百 狀上頁面儲存位元1而下頁面儲在 位元〇。料狀態C,兩個頁面皆儲存位元資料I 該程式化過程係-兩步驟式過程。於第—步驟中 化下頁面。若下頁面欲保持處於資料】,則該儲存2 態保持處於狀態£。若嗲資 牛狀 右该貝科欲被程式化至〇,則該館存一 121185.doc -29- 200805382 件之臨限電壓升高,以將該儲存元件程式化至狀態B,。因 此,圖12A顯示儲存元件自狀態e至狀態B,之程式化。狀態 B’係一中間狀態B,因此,將驗證點繪示成Vvb,,Vvb,低 於 Vvb 〇 於一實施例中,在一儲存元件自狀態E程式化至狀態W 後,該NAND串中的其鄰近儲存元件(WLn+1)則將相關於 •其下頁面來程式化。舉例而言,重新參見圖2,在程式化 f.x 儲存元件1〇6之下頁面後,將程式化儲存元件104之下頁 面。在程式化儲存元件104後,若儲存元件1〇4具有一自狀 悲E升南至狀態B’之臨限電壓,則浮動閘極至浮動閘極之 耦合效應將升高儲存元件106之視在臨限電壓。此將具有 使狀悲B’之臨限電壓分佈變寬至圖12B之臨限電壓分佈 1250所繪示之臨限電壓分佈之效應。當程式化上頁面時, 該臨限電壓分佈之視在變寬將得以糾正。 圖12C、、、曰示耘式化上頁面之過程。若該儲存元件處於擦 ” 除狀態E且上頁面保持處於1,則該儲存元件將保持處於狀 態E。若該儲存元件處於狀態』,且其上頁面資料欲被程式 化至〇,則該儲存元件之臨限電壓將升高,以使該儲存元 #處於狀態A。若該儲存元件處於中間臨限電壓分佈⑽ 中,且上頁面資料欲保持處,則該健存元件將被程式 化至最、、、ς狀態B。若該儲存元件處於中間臨限電壓分佈 且上頁面資料欲變為資料〇,則該儲存元件之臨 限私壓將升向,以使該儲存元件處於狀態c。圖所 緣示之過程減小了浮動閘極至浮動問極之耗合效應,此乃 121185.doc -30- 200805382 因僅鄰近儲存元件之上頁面程式化對一既定儲存元件之視 在臨限電壓有影響。一替代狀態編碼之一實例係當上頁面 資料係1時,自分佈1250移至狀態C,且當上頁面資料為〇 時移至狀態B。 雖然圖12A-C提供一關於四種資料狀態及兩個資料頁面 之實例,但圖12A-C所教示之概念亦可應用於具有多於或 少於四種狀態及不同於兩個頁面之其他實施方案。 圖13顯示一電壓波形13〇〇,其包括一系列施加至一經選 擇用於程式化之字線之程式化脈動131〇、132〇、133〇、 1340、1350、…。於一實施例中,該等程式化脈動具有一 電壓Vpgm,該電壓始於12伏並針對每一連續程式化脈動 增加例如0.5伏之增量,直至達到為2〇伏之最大值為止。 在該等程式化脈動之間係驗證脈動組1312、1322、1332、 1342、1352、.…。於某些實施例中,每一資料正程式化成 之狀態皆可具有一驗證脈動。於其他實施例中,可具有更 多或更少驗證脈動。每一組中之驗證脈動皆可具有例如 Vva、Vvb及Vvc之幅值(圖1〇)。 於一實施例中,資料係沿一共用字線程式化至儲存元 件。因此,在施加該等程式脈動前,選擇該等字線中之一 者供進行程式化。該字線將稱作選擇字線。一塊中之剩餘 子線稱作非選擇字線。選擇字線可具有一個或兩個鄰近字 線。若選擇字線具有兩個鄰近字線,則汲極側上之鄰近字 線稱作汲極側鄰近字線且源極側上之鄰近字線稱作源極側 鄰近字線。舉例而言,若圖2iWL2係選擇字線,則 121185.doc -31- 200805382 為源極側鄰近字線而WL3係汲極側鄰近字線。 每儲存元件塊皆包括一組形成行之位元線及一組形成 列之字線。。於-實_巾,該等位元線係4彳分成奇數位 元線及偶數位元線。同時程式化沿一共用字線並連接至奇 數位兀線之儲存元件,而在另一時間程式化沿一共用字線 並連接至偶數位元線之儲存元件(”奇數/偶數程式化n)。在 另一實施例中,針對該塊中的所有位元線,沿一字線程式 化儲存元件(”所有位元線程式化”)。於其他實施例中,可 將位元線或塊分解成其他群組(例如左及右群組、多於兩 個群組等等)。 圖14圖解說明一臨限電壓隨溫度及字線位置之變化。線 1410表示溫度係數對字線位置之關係。線142〇表示臨限電 壓變化與溫度變化之比率(△ντΓΟ對字線位置,其中Vread 係對施加至非選擇字線之電壓之溫度補償。在此種情況 下,溫度相依性量值減小,雖然一字線位置相依性減小, 但该子線位置相依性相依性仍然存在。線丨43〇表示(AVt〆 C )對子線位置之關係’其中Vread係對施加至非選擇字線 之電壓之溫度補償而Vcgr係對施加至選擇字線之電壓之溫 度補償。在此種情況下,溫度相依性之量值相對於線142〇 進一步減小,而字線位置相依性仍然存在。線丨44〇表示 (△VT/ C )對字線位置之關係,其中Vread係對施加至非選擇 字線之電壓之溫度補償且因進一步字線位置相依性為 Vcgr ’故對施加至選擇字線之電壓實施為vCgr2温度補 償。在此種情況下,相對於線1430之情形,實質上移除了 121185.doc -32- 200805382 字線位置相依性。字線相依性亦可藉由〜㈣施加 擇字線。 &amp; 特定而5,已觀察到非揮發性儲存元件之臨限電壓隨溫 度增加而降低。相對於溫度變化之電壓變化可以一溫度= 數⑷來表示,其通常約為_2mvrc。該溫度係數相依於記 十思體裝置之各種特徵,例如摻雜、佈局等等。此外,預期 該溫度係數將隨記憶體尺寸之減小而在量值上增加。該溫 度係數可識別電壓或電流變化與溫度變化之比率。例如酿 對於一-4(TC至+85T:之操作範圍,臨限電壓可變化約(85_ (-40))x(-2)==250 mV。因此,可藉由根據溫度偏置施加至 一選擇字線之讀取或驗證電壓來改良一個或多個與該選擇 字線相關聯之選擇儲存元件之讀取或驗證操作準確度。此 外’當不使用相依於字線之溫度補償時,該溫度係數可根 據字線位置而異,如線14 1〇所指示。舉例而言,假定一塊 中具有32個字線,則線1410可在WL0(源極側字線)處具有 一約為-1.9 mV/°C之值且在WL31(汲極側字線)處具有一約 為-2.1 mV/°C之值。因此,在一可能之設計中,溫度係數 跨越字線之變化為0.2 mV。自一 70 nm ABL架構晶片獲得 之實驗性資料顯示基於字線位址之約15 %之平均頁面溫度 係數變化,其中WL3 1(其串聯電阻完全處於其源極側上)因 其源極側上之溫度感應串聯電阻變化而遭受更多損害,從 而比一亦經歷串聯電阻變化(但僅在其汲極侧處)之Wl〇頁 面,導致附加之體效應。 已知各種技術可用於將經溫度補償之讀取電壓提供至選 121185.doc -33- 200805382 擇字線。大部分此等技術不依賴於獲得一實際溫度量測, 但此方法亦可能。舉例而言,標稱為&quot;v〇ltage Circuitry Having Temperature C〇mpensati〇n&quot;之美國專利 6,801,454闡述一種根據溫度係數將讀取電壓輸出至一非揮 發性記憶體之電壓產生電路,該專利以引用的方式併入本 文中。該電路使用-帶隙電流,該電流包括一不相依於溫 度之部分及一隨溫度增加而增加的相依於溫度之部分。標 稱為&quot;N〇n-Volatile Mem〇ry Whh Temperature_c〇寧12A-C disclose another process for staging non-volatile memory for any particular storage element by writing to a particular page associated with a particular page after writing an adjacent storage element for a previous page. Components to reduce the coupling effect of the floating gate to the floating gate. In an exemplary embodiment, the non-volatile storage element uses four data states to store two data bits for each storage element. For example, assume that the state is an erased state, while states A, B, and C are stylized states. Status £ store data u. State A stores data 〇1. State B stores data 1〇. Status c stores data 〇〇. This is an example of a non-Gray code, since both bits vary between neighboring states A and B. Other codes for data status to physical data status can also be used. Each data element stores two data pages. For the purpose of reference, these data pages are referred to as the upper page and the lower page; however, other marks may be assigned thereto. For state A, the upper page stores the bit 〇 and the next bit stores the bit 1 〇 for the plethora of P 百 on the page to store the bit 1 and the next page is stored in the bit 〇. Material state C, both pages store bit data I This stylized process is a two-step process. In the first step, the page is turned down. If the next page is to remain in the data, then the storage 2 state remains in the state £. If the arsenal is to be stylized to the cockroach, the library will store a threshold voltage of 121185.doc -29- 200805382 to program the storage component to state B. Thus, Figure 12A shows the stylization of the storage element from state e to state B. State B' is an intermediate state B. Therefore, the verification point is depicted as Vvb, Vvb, which is lower than Vvb. In one embodiment, after a storage element is programmed from state E to state W, the NAND string is Its adjacent storage element (WLn+1) will be programmed in relation to its next page. For example, referring back to Figure 2, after the page below the stylized f.x storage element 1〇6, the lower page of the storage element 104 will be programmed. After the staging of the storage element 104, if the storage element 1〇4 has a threshold voltage from the state of the first to the state B', the coupling effect of the floating gate to the floating gate will increase the view of the storage element 106. At the threshold voltage. This will have the effect of widening the threshold voltage distribution of the sorrow B' to the threshold voltage distribution depicted by the threshold voltage distribution 1250 of Figure 12B. When the page is stylized, the apparent widening of the threshold voltage distribution will be corrected. Figure 12C, , and show the process of simplification of the upper page. If the storage element is in the erasing state E and the upper page remains at 1, the storage element will remain in the state E. If the storage element is in the state, and the page information on it is to be programmed to 〇, then the storage The threshold voltage of the component will rise so that the storage element # is in state A. If the storage component is in the intermediate threshold voltage distribution (10) and the upper page data is to be maintained, the health component will be programmed to The most, , and ς state B. If the storage element is in the middle threshold voltage distribution and the upper page data is to be changed to data 则, the threshold private pressure of the storage element will be raised to make the storage element in the state c. The process illustrated by the figure reduces the effect of the floating gate to the floating pole. This is 121185.doc -30- 200805382 because the page is stylized only on the storage element and the apparent storage limit of a given storage element The voltage has an effect. An example of an alternate state code is when the page data system 1 is moved from the distribution 1250 to the state C, and moves to the state B when the upper page data is 。. Although FIG. 12A-C provides a Information And examples of two data pages, but the concepts taught in Figures 12A-C can also be applied to other embodiments having more or less than four states and different from two pages. Figure 13 shows a voltage waveform 13〇〇 And comprising a series of stylized pulses 131, 132, 133, 1340, 1350, ... applied to a word line selected for stylization. In one embodiment, the programmed pulses have a voltage Vpgm The voltage begins at 12 volts and is incremented by, for example, 0.5 volts for each successive stylized pulsation until a maximum of 2 volts is reached. Between these programmed pulsations, the pulsing groups 1312, 1322 are verified. 1332, 1342, 1352, ... In some embodiments, each data can be programmed to have a verification pulse. In other embodiments, there may be more or less verification pulses. The verification pulse can have amplitudes such as Vva, Vvb, and Vvc (Fig. 1A). In one embodiment, the data is threaded along a common word to the storage element. Therefore, before applying the program pulse , select these word lines The word line will be referred to as a selected word line. The remaining sub-line in a block is called a non-selected word line. The selected word line can have one or two adjacent word lines. If the selected word line has two adjacent lines The word line, the adjacent word line on the drain side is referred to as the drain side adjacent word line and the adjacent word line on the source side is referred to as the source side adjacent word line. For example, if FIG. 2iWL2 selects the word line, 121185.doc -31- 200805382 is the source side adjacent word line and the WL3 system side is adjacent to the word line. Each storage element block includes a set of bit lines forming a row and a set of word lines forming a column. - real _ towel, the bit line system 4 彳 is divided into odd bit lines and even bit lines. At the same time, stylized along a common word line and connected to the storage elements of the odd bit line, and stylized at another time A storage element ("odd/even stylized n") along a common word line and connected to even bit lines. In another embodiment, the storage elements ("all bits are threaded") are threaded along a word for all of the bit lines in the block. In other embodiments, the bit lines or blocks may be decomposed into other groups (e.g., left and right groups, more than two groups, etc.). Figure 14 illustrates a threshold voltage as a function of temperature and word line position. Line 1410 represents the relationship of temperature coefficient to word line position. Line 142 〇 represents the ratio of the threshold voltage change to the temperature change (ΔντΓΟ versus word line position, where Vread is the temperature compensation for the voltage applied to the unselected word line. In this case, the temperature dependence is reduced. Although the dependence of the position of a word line is reduced, the dependency dependence of the position of the sub-line still exists. The line 丨43〇 indicates the relationship of (AVt〆C) to the position of the sub-line where the Vread pair is applied to the unselected word line. The voltage is compensated for temperature and Vcgr is the temperature compensation for the voltage applied to the selected word line. In this case, the magnitude of the temperature dependence is further reduced relative to line 142, and word line position dependencies still exist. Line 丨44〇 indicates the relationship of (ΔVT/C) to the position of the word line, where Vread is temperature compensated for the voltage applied to the unselected word line and is applied to the selected word due to further word line position dependence Vcgr ' The voltage of the line is implemented as vCgr2 temperature compensation. In this case, the position dependence of the word line position of 121185.doc -32-200805382 is substantially removed relative to the case of line 1430. The word line dependency can also be obtained by ~(4) Shi Select word line. &amp; Specific 5. It has been observed that the threshold voltage of non-volatile storage elements decreases with increasing temperature. The voltage change with respect to temperature change can be expressed as a temperature = number (4), which is usually about _2mvrc. The temperature coefficient is dependent on various features of the device, such as doping, layout, etc. Furthermore, it is expected that the temperature coefficient will increase in magnitude as the memory size decreases. The temperature coefficient identifies the voltage. Or the ratio of current change to temperature change. For example, for a range of -4 (TC to +85T:, the threshold voltage can vary by about (85_ (-40)) x (-2) == 250 mV. Therefore, The read or verify operation accuracy of one or more selected storage elements associated with the selected word line can be improved by reading or verifying the voltage applied to a selected word line based on the temperature bias. Further, when not used Depending on the temperature compensation of the word line, the temperature coefficient may vary depending on the word line position, as indicated by line 14 1 。. For example, assuming that there are 32 word lines in a block, line 1410 may be at WL0 (source) Side word line) has a value of approximately -1.9 mV The value of /°C and has a value of about -2.1 mV/°C at WL31 (the drain side word line). Therefore, in a possible design, the temperature coefficient varies by 0.2 mV across the word line. Experimental data obtained on a 70 nm ABL architecture wafer shows an average page temperature coefficient change of approximately 15% based on the word line address, where WL3 1 (its series resistance is completely on its source side) due to its source side Temperature-induced series resistance changes and suffers more damage, resulting in additional body effects than W1〇 pages that also experience series resistance changes (but only at their drain sides). Various techniques are known for temperature compensation. The read voltage is supplied to the selected word line 121185.doc -33- 200805382. Most of these techniques do not rely on obtaining an actual temperature measurement, but this method is also possible. For example, U.S. Patent No. 6,801,454, the disclosure of which is incorporated herein by reference in its entirety, the entire disclosure of the disclosure of the disclosure of the entire disclosure of The patent is incorporated herein by reference. The circuit uses a bandgap current that includes a portion that is independent of temperature and a temperature dependent portion that increases with increasing temperature. The nominal &quot;N〇n-Volatile Mem〇ry Whh Temperature_c〇宁

Data Read之美國專利6,56〇,152使用—種用於偏置施加至 資料儲存元件之源極或沒極之電麼之偏麼產生器電路,該 專利以引用的方式併入本文中。標稱為&quot;Muhi_state eeprom Read and Write Circuits and Techn咖es&quot;之美國專 利5,172,338闡述-種使用以與資料儲存單元相同之方式且 ^同-積體電路晶片上形成之參考儲存單元之溫度補償技 術’該專利以引用的方式併入本文中。該等參考健存單元 提供參考位準’可將選擇單元之所量測電流或電壓與該參 考=準相比較。提供溫度補償,此乃因溫度以與自資料儲 存早元讀取之值相同之方式影響參考位準。如本文中所 述’此等技術中之任-種技術以及任何其他已知技術皆可 用來為選擇字線、非選擇字線及/或選擇閘極之電壓提供 溫度補償。U.S. Patent No. 6,56, 152, the disclosure of which is incorporated herein by reference in its entirety in its entirety in its entirety in its entirety in its entirety in its entirety in its entirety in its entirety in its entirety in its entirety. U.S. Patent No. 5,172,338, the disclosure of which is incorporated herein by reference in its entirety in the same extent as the data storage unit and the temperature of the reference storage unit formed on the integrated circuit. Compensation Technology 'This patent is incorporated herein by reference. The reference health unit provides a reference level 'the measured current or voltage of the selected unit can be compared to the reference = quasi-phase. Temperature compensation is provided because the temperature affects the reference level in the same manner as the value read from the data storage early element. Any of these techniques, as well as any other known techniques, as described herein, can be used to provide temperature compensation for voltages of selected word lines, unselected word lines, and/or selected gates.

加 因此’藉助習用技術,藉由選擇字線施加至一個或多個 擇儲存元件之讀取或驗證電壓受到溫度補償。然而,施 至剩餘字線之電壓(其稱作—讀取電壓ν_)及施加至選 121185.doc •34- 200805382Therefore, by means of conventional techniques, the read or verify voltage applied to one or more of the selected storage elements by the selected word line is temperature compensated. However, the voltage applied to the remaining word lines (which is referred to as the read voltage ν_) is applied to the selected 121185.doc •34-200805382

擇閉極之電壓(其稱作選擇間極、源極之㈣或選擇閑 極、汲極之Vsgd)尚未受到溫度補償。人們一直認為僅對 選擇儲存元件進行溫度補償^矣。特定而言,人們_直認 為將非選擇儲存元件及選擇閘極過驅動収以超過其臨: 電壓以使溫度變化不明顯影響其導電性即可'然而,當電 晶體按比例縮小至更小之尺寸時,其特徵降格,且飽:電 流越來越多地偏離呈一平坦輪廓,如同由汲極電流⑹對 控制閘極電壓(Veg)之圖表中—小斜坡所表示的那樣。 為瞭解決此等問題,建議當前正被讀取的一儲存元件之 路徑中之Vread、Vsgd、Vsgs及任何其他必需的電晶體具 有施加至其閘極之經溫度補償偏壓,以使每一電晶體之導 通電流變得較少相依於溫度。藉由對此等所施加之電壓進 行溫度跟蹤,可進一步減小因溫度變化所致之每一狀態之 臨限分佈之擴展。此結果可以多種方式(其未必互斥)來加 以利用。舉例而言,可減*Vread。因此,可減少過驅動 量,亦即Vread超過儲存元件之最高程式化狀態之臨限電 壓之程度’從而減少因使用高Vread值所致之相關聯讀取 干擾。 此Vread之減小對於諸多不同之讀取/驗證技術有所幫 助。該減小對於採用多個讀取操作之讀取/驗證技術特別 重要。舉例而言,由Jian Chen於2005年4月5曰申請且標稱 為’’Compensating For Coupling During Read Operations OfThe voltage of the closed-pole (which is called the selection of the interpole, the source (four) or the selection of the idle, the Vsgd of the drain) has not been temperature compensated. It has been thought that temperature compensation is only performed on selected storage elements. In particular, people think that the non-selective storage element and the selection gate are overdriven to exceed their voltage: so that the temperature change does not significantly affect its conductivity. However, when the transistor is scaled down to smaller In the case of dimensions, the features are degraded, and the saturation: current is increasingly deviated from a flat profile, as indicated by the small slope of the graph of the gate voltage (Veg) from the drain current (6). In order to address these issues, it is recommended that Vread, Vsgd, Vsgs, and any other necessary transistors in the path of a storage element that is currently being read have a temperature compensated bias applied to their gates, such that each The on current of the transistor becomes less dependent on temperature. By performing temperature tracking on these applied voltages, the spread of the threshold distribution of each state due to temperature changes can be further reduced. This result can be exploited in a variety of ways (which are not necessarily mutually exclusive). For example, *Vread can be subtracted. Therefore, the overdrive amount, i.e., the degree of Vread exceeding the threshold voltage of the highest stylized state of the storage element, can be reduced, thereby reducing the associated read disturb caused by the use of a high Vread value. This reduction in Vread helps with many different read/verify techniques. This reduction is especially important for read/verify techniques that employ multiple read operations. For example, applied by Jian Chen on April 5, 2005 and is nominally '’Compensating For Coupling During Read Operations Of

Non-Volatile Memory”之共同待決美國專利申請案第 1 1/099,133號(檔案編號SAND-1040US0)闡述一種其中針對 121185.doc -35- 200805382 每一程式化狀態以不同之位準對選擇儲存元件實施多個讀 取操作,該專利申請案以引用的方式併入本文中。例如, 位準之間的增量可為50-100 mV。該技術反對字線至字線 之電容性柄合效應’其中當隨後程式化一鄰近儲存元件 (通常為一汲極側鄰居)時,一先前經程式化儲存元件之臨 限電壓偏移得更局。若該偏移足約大,則可造成一讀取夢 誤。當將該鄰近儲存元件程式化至一更高狀態(例如,狀 怨C)時’該麵合最高。為瞭解決此問題,根據在該選擇儲 存元件後程式化之鄰近字線上的鄰近儲存元件之狀態來為 每一程式化狀態選擇多個讀取操作中之一者。 在此技術之一變化形式中,如由圖13中之驗證脈動組所 示’針對選擇字線上的每一狀態使用一個讀取位準,同時 調節施加至鄰近字線之讀取電壓。此變化形式闡述於由 Nima Mokhlesi於2006年3月17日申請且標稱為,,ReadNon-Volatile Memory, co-pending U.S. Patent Application Serial No. 1 1/099,133 (file number SAND-1040US0) describes a type in which each of the stylized states for 121185.doc-35-200805382 is selected for storage at different levels. The element performs a plurality of read operations, which are incorporated herein by reference. For example, the increment between levels can be 50-100 mV. This technique opposes the capacitive shank of the word line to the word line. Effect 'When a subsequent storage element (usually a drain side neighbor) is subsequently programmed, the threshold voltage offset of a previously programmed storage element is further reduced. If the offset is about large, it can cause A read dream error. When the adjacent storage element is stylized to a higher state (for example, the blame C), the face is the highest. To solve this problem, according to the stylized proximity after the storage element is selected The state of the adjacent storage elements on the word line selects one of a plurality of read operations for each stylized state. In one variation of this technique, as indicated by the verification pulsation group in FIG. Everything on the line State using a read level, while adjusting a read voltage is applied to the adjacent word line. Forth in this variation on March 17, 2006 and filed by the Nima Mokhlesi nominally ,, Read

Operation For Non-Volatile Storage With Compensation For Coupling”之共同待決美國專利申請案第丨丨门料””號(檔案 編號SAND-1089US2)中,該專利申請案以引用的方式:二 本文中。在以上兩種情況之任一情況下,因用於讀取相同 資料量之讀取操作次數增加,受讀取干擾之影響增加。本 文中所提供之溫度補償技術可減輕此問題。 本文中所提供之溫度補償技術之再一優點在於,各種程 式化狀態(例#狀態E、A、BAC)之臨限電•分佈之間= 冗裕可隨著因溫度變化減小所致每—狀態之臨限電壓分佈 之擴展而增加。另一優點在於’可藉由以下方式增加程式 121185.doc -36- 200805382 化效能’例如’可藉由消耗各種程式化狀態之臨限電壓分 佈之間的增加之冗裕而在程式化脈動之樓梯級數中使用一 更大步長。另一優點在於,整個記憶體操作窗口(例如, 用於將資料儲存於儲存元件中之臨限電壓範圍)可因將程 式化狀L[縮得更靠近在—起而減小。由此不僅減少讀取 及寫入干擾,且亦增加寫入效能,此乃因達到一所期望之 程式化狀態所需之程式化脈動將因一較小之窗口而變得更 少〇 可更進一步地藉由提供一計及選擇字線在其他非選擇字 線(其與-組非揮發性儲存元件相關聯)中—相對位置的經 溫度補償電壓改良準確度。該準確度改良可藉由將線14二 與1430相比較而看到。可對選擇字線單獨地或結合非選擇 字線之溫度補償來實施此溫度補償。參見圖⑸。亦可為 非選擇字線提供一字線相依性。 ' …^丨〜不公收瓜又仃為之 時序圖中經溫度補償之電壓施加至所有非選擇字線並 施加至兩個選擇閘極。一般而言,在讀取及驗證操 間,選擇字線或其他控制線連接至—電壓,該電壓之—位 準係針對每―讀取及驗證操作所規定,以確定相關儲存= 件之-臨限電壓是否已達到此位準。在施加字線電壓後, 量測儲存元件之導電電流以確定該儲存元件是否已導通 若量測到該導電電流大於某—值,則假定該儲存元件 通且施加至字線之電壓大於儲在 储存兀件之臨限電壓。若量制 到該導電電流不大於此某-值,則心儲存元件未導通= 121185.doc -37- 200805382 施加至字線之電壓不大於儲存元件之臨限電壓。 存在許多種用於在讀取或驗證操作期間量測一儲存元件 之導電電流之方式。於一實例中,允許(或不允許)以包括 該儲存元件之NAND串使該位元線放電之速率來量測儲存 元件之導電電流。在一時間週期後量測該位元線上之電荷 以確定其是否已放電。於另一實施例中,選擇儲存元件之 導電允許電流在一位元線上流動或不流動,此係根據感測 放大器中一電容器是否因電流流動而充電來加以量測。上 文論述了兩個實例。 圖 15a顯示波形 SGD、WLunselected、WLn、SGS、選擇 BL、及始於一約為0伏之穩態電壓Vss之源極。SGD代表汲 極側選擇閘極之閘極。WLunselected代表非選擇字線。 WLn係經選擇用於讀取/驗證之字線。SGS係源極側選擇閘 極之閘極。選擇BL係經選擇用於讀取/驗證之位元線。源 極係儲存元件之源極線(參見圖4)。注意,所繪示之SGS及 選擇BL存在兩種變化形式。一組此等波形SGS(選項1)及選 擇BL(選項1)繪示一針對一儲存元件陣列之讀取/驗證操 作,該讀取/驗證操作藉由確定該位元線是否已放電來量 測一儲存元件之導電電流。另一組此等波形SGS(選項2)及 選擇BL(選項2)繪示一對一儲存元件陣列之讀取/驗證操 作,該讀取/驗證操作以使該感測放大器中一專用電容器 放電之速率來量測一儲存元件之導電電流。 首先,將參照SGS(選項1)及選擇BL(選項1)來闡述藉由 確定位元線是否已放電來量測一儲存元件之導電電流中所 121185.doc -38- 200805382 牵涉的感測電路及儲存元件陣列之行為。於時_處, SGD及SGS(選項2)分別升高至Vsgd·^ 如其中,,V 表示-經溫度補償之電Μ。Vsgd七及Vsg_藉由針對溫度 分別偏置Vsgd及Vsgs所獲得。例如,^以及%#約為3·5 伏。例如,可根據上述補償技術來施加溫度補償。使非選 擇字線升高至Vread-tc。Vread_tc係藉由針對溫度偏置Operation For Non-Volatile Storage With Compensation For Coupling, co-pending U.S. Patent Application Serial No. (SAND-1089US2), filed on In either case, the number of read operations used to read the same amount of data increases, and the effect of read disturb increases. The temperature compensation technique provided in this paper can alleviate this problem. A further advantage of the temperature compensation technique is that between the various stylized states (example #state E, A, BAC), the current limit and the distribution = redundancy can be reduced with the change of temperature. The increase in the voltage-limited distribution is increased. Another advantage is that the program can be increased by the following method: 121185.doc -36-200805382 performance can be increased, for example, by consuming a threshold voltage distribution of various stylized states. It is redundant and uses a larger step size in stylized pulsating staircase series. Another advantage is that the entire memory operation window (for example, for storing data in storage elements) The voltage limit range can be reduced by shortening the stylized shape L [shrinking closer to the start-up. This not only reduces read and write interference, but also increases write performance, because a desired program is achieved. The stylized pulsations required for the state will be reduced by a smaller window, and further by providing a count of selected word lines on other non-selected word lines (the NAND-non-volatile storage elements) Correlation) The temperature-compensated voltage improvement accuracy of the relative position. This accuracy improvement can be seen by comparing line 14 to 1430. The temperature of the selected word line can be selected individually or in combination with the unselected word line. Compensation is used to implement this temperature compensation. See Figure (5). A word line dependency can also be provided for the unselected word lines. ' ...^丨~Unfairly collected and temperature-compensated voltages are applied to all non-selectives in the timing diagram. The word line is applied to two select gates. In general, between the read and verify operations, the word line or other control line is connected to the voltage, which is for each read and verify operation. Prescribed to determine the relevant storage = Whether the threshold voltage has reached this level. After applying the word line voltage, measure the conduction current of the storage element to determine whether the storage element is turned on. If the measured current is greater than a certain value, it is assumed The voltage of the storage element is applied to the word line is greater than the threshold voltage stored in the storage element. If the conduction current is not greater than the value, the cardiac storage element is not turned on = 121185.doc -37- 200805382 The voltage applied to the word line is no greater than the threshold voltage of the storage element. There are a number of ways to measure the conduction current of a storage element during a read or verify operation. In one example, it is allowed (or not allowed) to The NAND string including the storage element measures the rate at which the bit line is discharged to measure the conduction current of the storage element. The charge on the bit line is measured after a period of time to determine if it has been discharged. In another embodiment, the selection of the conduction of the storage element allows current to flow or not flow on a single bit line, which is measured based on whether a capacitor in the sense amplifier is charged due to current flow. Two examples are discussed above. Figure 15a shows the waveforms SGD, WLunselected, WLn, SGS, select BL, and the source starting at a steady state voltage Vss of about 0 volts. SGD represents the gate of the gate selection gate. WLunselected represents a non-selected word line. WLn is selected for reading/verifying word lines. The gate of the gate side of the SGS system selects the gate of the gate. The bit line selected by the BL system for reading/verification is selected. The source line of the source storage element (see Figure 4). Note that there are two variations of the SGS and the selected BL. A set of such waveforms SGS (option 1) and select BL (option 1) illustrate a read/verify operation for an array of storage elements by determining whether the bit line has been discharged Measure the conduction current of a storage component. Another set of such waveforms SGS (option 2) and select BL (option 2) illustrate a read/verify operation of a one-to-one array of storage elements for discharging a dedicated capacitor in the sense amplifier The rate is used to measure the conduction current of a storage element. First, reference will be made to SGS (option 1) and selection BL (option 1) to describe the sensing circuit involved in measuring the conduction current of a storage element by determining whether the bit line has been discharged. 121185.doc -38 - 200805382 And the behavior of the array of storage elements. At time _, SGD and SGS (option 2) are raised to Vsgd·^ respectively, where V is the temperature compensated enthalpy. Vsgd VII and Vsg_ are obtained by biasing Vsgd and Vsgs for temperature respectively. For example, ^ and %# are about 3.5 volts. For example, temperature compensation can be applied in accordance with the compensation techniques described above. Raise the non-selected word line to Vread-tc. Vread_tc is biased against temperature

Vread所獲得。例如,Vread約為6伏。該選擇字線針對一 讀取操作升高至Vcgr-tc(控制閘極讀取電壓),例如圖1〇之 Vra、Vrb或Vrc,或者針對一驗證操作升高至一驗證位 準,例如圖10之Vva、Vvb或Vvc。在一種方法中,將選擇 BL(選項1)預充電至約〇 7伏。施加至非選擇字線之vread_ tc充當一過驅動電壓,此乃因其導致非選擇儲存元件導通 並充當傳遞閘極。施加至非選擇儲存元件之過驅動電壓等 於施加至該控制閘極之電壓超過臨限電壓之量。 如所提及,將Vread選擇成一充分高於儲存元件之最高 限電壓之位準以確保非選擇儲存元件處於導電或導通^ 態。舉例而言,狀態E、A、c之臨限電壓可分別假定 為-2伏、〇伏、2伏及4伏,而Vread在無溫度補償之情況下 可為6伏。在此種情況下,處於狀態E之儲存元件經心 (-2) = 8伏之過驅動,處於狀態A之儲存元件經6_〇 = 6伏之過 驅動,處於狀態B之儲存元件經6_2=4伏之過驅動,且處於 狀態C之儲存元件被經6-2=2伏之過驅動。雖然在每一種情 形下非選擇儲存元件皆處於導電狀態,但其導電性將基於 其過驅動之程度而變化。受過驅動越高,非選擇儲存元件 121185.doc -39- 200805382 之導電性越好’此乃因其源極沒極電阻更小而電流攜載能 力更大。類似地,受過驅動越低,非選擇元件之導電性越 差,此乃因其源極汲極電阻更大而電流攜載能力更小。因 此,與選擇儲存元件處在相同NAND串中之儲存元件將根 據其程式化狀態而具有不同之導電性,即使其皆處於一般 導電狀態中。因此,該選擇儲存元件之讀取位準將根據其 相應程式化狀態而受非選擇儲存元件之影響。 假定一溫度補償為-0.2伏 仇= 5.8 伏Obtained by Vread. For example, Vread is approximately 6 volts. The selected word line is boosted to Vcgr-tc (control gate read voltage) for a read operation, such as Vra, Vrb or Vrc of FIG. 1 or raised to a verify level for a verify operation, such as a map. 10 Vva, Vvb or Vvc. In one method, select BL (option 1) is precharged to approximately 伏7 volts. Vread_tc applied to the unselected word line acts as an overdrive voltage because it causes the non-selected storage element to conduct and act as a transfer gate. The overdrive voltage applied to the non-selective storage element is equal to the amount of voltage applied to the control gate that exceeds the threshold voltage. As mentioned, Vread is selected to a level well above the maximum voltage of the storage element to ensure that the non-selected storage element is in a conducting or conducting state. For example, the threshold voltages of states E, A, and c can be assumed to be -2 volts, volts, 2 volts, and 4 volts, respectively, while Vread can be 6 volts without temperature compensation. In this case, the storage element in state E is driven by the center (-2) = 8 volts, the storage element in state A is driven by 6_〇 = 6 volts, and the storage element in state B is 6_2. = 4 volts overdrive, and the storage element in state C is driven over 6-2 = 2 volts. Although the non-selective storage element is in a conductive state in each case, its conductivity will vary based on the degree of overdrive. The higher the drive, the better the conductivity of the non-selective storage component 121185.doc -39- 200805382' because of its smaller source-pole resistance and higher current carrying capacity. Similarly, the lower the drive, the worse the conductivity of the non-selected components due to their larger source-drain resistance and lower current carrying capacity. Therefore, the storage elements in the same NAND string as the selected storage elements will have different electrical conductivities depending on their stylized state, even if they are all in a general conductive state. Therefore, the read level of the selected storage element will be affected by the non-selected storage element according to its corresponding stylized state. Assume that a temperature compensation is -0.2 volts = 5.8 volts

如,出於與非選擇儲存元件類似之緣由,可對施加至選擇 閘極之電壓進行溫度補償,由此允許為Susy」伏之 Vsgd-tc或Vsgs_tc。非選擇字線及選擇閘極之溫度補償往 往會使對選擇字線臨限電壓之讀取更相依於溫度。由此, 母-與選擇儲存元件串聯之非選擇儲存元件皆對選擇儲存 元件之臨限電壓所獲得之讀取具有一小影響,例如,3 mV。雖然一個非選擇儲存元件對讀取之影響較小,但當 存在31個㈣擇字料,該等非選㈣存元件巾之每一者 之累積影響可合計達—顯著位準,例如% mv。非選擇字 線之溫度補償效應料具有更多字線之記㈣裝置且當使 用減小之過驅動電壓時更為明顯。 寺門t2處NAND串可控制位元線。同樣於時間^ 源極側選擇閉極係藉由SGS(選項υ升高至vs㈣而導 ==提供一路徑以耗散該位元線上之電荷。若經選擇用 於靖取之儲存元件之臨限電 WLn之睑也朽、隹 電姿大於Vcgr或施加至選擇字線 準,則該選擇儲存元件不會導通且該位元線 121185.doc 200805382 不放電,如線1450所繪示。若經選擇用於讀取之儲存元件 中之臨限電壓低於Vcgr_tc或低於施加至選擇字線wLn之驗 證位準,則經選擇用於讀取之儲存元件將導通(導電)且該 位元線電壓將耗散,如曲線1452所繪示。在時間12後及時 間t3别的某一點(其由特定實施方案確定)處,感測放大器 將確定該位元線是否已耗散一充足量。在^與〇之間,感 測放大器量測所估計之BL電壓。在時間〇處,所繪示之波 形將降低至Vss(或另一備用值或恢復值)。 下文將參照SGS(選項2)及選擇BL(選項2)來論述感測電 路及儲存元件陣列以充電感測放大器中一專用電容器充電 之速率量測儲存元件之導電電流之行為。在時間11處, SGD升高至Vsgd_tc,非選擇字線(WLunselected)升高至 Vread-tc,且選擇字線(WLn)針對一讀取操作升高至vcgr_ tc(例如Vra、Vrb或Vrc),或者針對一驗證操作升高至一驗 證位準(例如Vva、Vvb或Vvc)。在此種情況下,無論 NAND串正在做什麼,感測放大器使位元線電壓保持不 k,以便感測放大器在位元線”夾持”至彼電壓時量測流動 之電流。在時間tl後及時間t3前之某一點(其由特定實施方 案確疋)處,感測放大器將確定該感測放大器中之電容器 是否已耗散一充足量。在時間t3處,所繪示之波形將降低 至Vss(或另一備用值或恢復值)。注意,在其他實施例中, 可改變某些波形之定時。 圖15b繪示圖15a之時序圖,其中不同之經溫度補償電壓 係根據字線位置施加至選擇字線。如結合圖丨4所論述,在 121185.doc -41 - 200805382 一種方法中,當該字線之位置相對於源極更接近於没極 時’可將-較高量值之溫度補償(例如,負值更大)施加至 選擇字線。此由圖15b之時序圖加以例示,其巾施加至一For example, for reasons similar to non-selective storage elements, the voltage applied to the selected gate can be temperature compensated, thereby allowing Vsygd-tc or Vsgs_tc to be Susy. The temperature compensation of the unselected word lines and the selected gates tends to make the reading of the threshold voltage of the selected word line more dependent on the temperature. Thus, the mother-non-selective storage elements in series with the selected storage element have a small effect on the reading obtained by selecting the threshold voltage of the storage element, for example, 3 mV. Although a non-selective storage element has less effect on reading, when there are 31 (four) selection materials, the cumulative effect of each of the non-selected (four) storage component towels can be aggregated to a significant level, such as % mv . The temperature compensated effect material of the unselected word line has more word line (4) devices and is more pronounced when a reduced overdrive voltage is used. The NAND string at the gate t2 can control the bit line. Similarly, at the source side, the source is selected to be closed by SGS (option υ is raised to vs (four) and == provides a path to dissipate the charge on the bit line. If selected for storage of the component If the power-limiting WLn is also 隹, the 隹 隹 is greater than Vcgr or applied to the selected word line, the selected storage element will not be turned on and the bit line 121185.doc 200805382 will not be discharged, as shown by line 1450. If the threshold voltage in the storage element selected for reading is lower than Vcgr_tc or lower than the verify level applied to the selected word line wLn, then the storage element selected for reading will be conductive (conductive) and the bit line The voltage will be dissipated as depicted by curve 1452. At some point after time 12 and at time t3, which is determined by the particular implementation, the sense amplifier will determine if the bit line has been dissipated by a sufficient amount. Between ^ and ,, the sense amplifier measures the estimated BL voltage. At time ,, the plotted waveform will be reduced to Vss (or another alternate or recovered value). Reference will be made to SGS (option 2) And select BL (option 2) to discuss the sensing circuit and storage element array The rate at which a dedicated capacitor is charged in the charge sense amplifier measures the behavior of the conduction current of the storage element. At time 11, SGD rises to Vsgd_tc, the unselected word line (WLunselected) rises to Vread-tc, and the word line is selected (WLn) is raised to vcgr_tc (eg, Vra, Vrb, or Vrc) for a read operation, or raised to a verify level (eg, Vva, Vvb, or Vvc) for a verify operation. In this case, What the NAND string is doing, the sense amplifier keeps the bit line voltage not k, so that the sense amplifier measures the current flowing when the bit line is "clamped" to the voltage. After time t1 and before time t3 At a point (which is determined by the particular implementation), the sense amplifier will determine if the capacitor in the sense amplifier has dissipated a sufficient amount. At time t3, the waveform depicted will be reduced to Vss (or another Alternate or recovered value. Note that in other embodiments, the timing of certain waveforms may be changed. Figure 15b illustrates the timing diagram of Figure 15a, wherein different temperature compensated voltages are applied to the selected word line based on word line position. As combined As discussed in 丨4, in a method of 121185.doc -41 - 200805382, when the position of the word line is closer to the source than the source, the temperature compensation can be made to a higher value (for example, a negative value) Large) applied to the selected word line. This is illustrated by the timing diagram of Figure 15b, the towel is applied to a

論近於源極之選擇字線(例如WLG)之經溫度補償電壓由 -虛線顯示’而施加至更接近於沒極之選擇字線(例如 WL3i)之經溫度補償電壓由一實線顯示。當選擇字線 Μ與汲極t間時施加至其之經溫度補償電壓處於當選擇 .字線處於源極側或汲極側時施加至其之電壓的中間,例如 與距源極或㈣之距離成比例。可為施加至選擇閘極及非 選擇字線中之一者或多者之電壓提供一字線位置相依性。 可以類比方式修改圖16-18以提供一字線位置相依性。 圖16係一解釋讀取/驗證操作期間某些波形之行為之時 序圖’其中經溫度補償之電壓施加至除直接鄰近一選擇字 線之字線以外的所有非選擇字線,並施加至兩個選擇閘 極。波形SGD、SGS(選項1)及SGS(選項2)與圖15a中相 同。選擇BL及源極波形(其未繪示)亦與圖Ua中相同。注 忍’ 4示5己為WL0至WLn-2之波形代表施加至位於第一字線 WL0與字線WLn_2之間且包括第一字線WL0及字線WLn_2 之字線的經溫度補償之讀取電壓,字線WLn-2緊接著選擇 字線WLn之一源極側鄰近字線WLn-1。標記為WLn-2至 WL3 1之波形代表施加至字線WLn+2(其緊接著選擇字線 WLn之一汲極側鄰近字線WLn+Ι)與WL31(其直接鄰近汲極 側選擇閘極)之間且包括字線WLn+2及WL3 1之字線的經溫 度補償之讀取電壓。假定一 NAND串上存在三十二個儲存 121185.doc -42- 200805382 元件’但亦可使用不同之數量。對於此等非選擇字線,如 所述施加溫度補償。類似地,對於選擇字線WLn,施加經 溫度補儐之控制閘極讀取電壓Vcgr_tc。The temperature compensated voltage applied to the selected word line (e.g., WLG) near the source is shown by the dashed line and the temperature compensated voltage applied to the selected word line (e.g., WL3i) closer to the pole is displayed by a solid line. The temperature-compensated voltage applied thereto when the word line Μ and the drain t are selected is in the middle of the voltage applied thereto when the word line is on the source side or the drain side, for example, from the source or (4) The distance is proportional. A word line position dependency can be provided for the voltage applied to one or more of the select gate and the unselected word line. Figures 16-18 can be modified in an analogy manner to provide a word line position dependency. Figure 16 is a timing diagram illustrating the behavior of certain waveforms during a read/verify operation where the temperature compensated voltage is applied to all non-selected word lines except for the word line directly adjacent to a selected word line, and applied to both Select the gate. The waveforms SGD, SGS (option 1) and SGS (option 2) are the same as in Figure 15a. The selection of the BL and source waveforms (not shown) is also the same as in Figure Ua. The waveform representing WL0 to WLn-2 represents a temperature-compensated read applied to a word line between the first word line WL0 and the word line WLn_2 and including the first word line WL0 and the word line WLn_2. Taking the voltage, the word line WLn-2 is next to the source side of the selected word line WLn adjacent to the word line WLn-1. Waveforms labeled WLn-2 through WL3 1 represent application to word line WLn+2 (which is followed by one of the drain side adjacent word lines WLn+Ι of the selected word line WLn) and WL31 (which is directly adjacent to the drain side select gate) A temperature compensated read voltage between and including the word lines of word lines WLn+2 and WL3 1 . Suppose there are thirty-two storage 121185.doc -42- 200805382 components on a NAND string, but different numbers can be used. For such non-selected word lines, temperature compensation is applied as described. Similarly, for the selected word line WLn, a temperature-compensated control gate read voltage Vcgr_tc is applied.

對於直接鄰近該選擇字線之字線WLn_ i及WLn+丨中之任 何一者或兩者,所施加之讀取電壓未經溫度補償,或經溫 度補償-減小之量’例如’與施加至其他非選擇字線之溫 度補偵相t匕日月顯減小之量。一對特定記憶體裝置之最優 補償可藉由測試來確定。合意之情形係因選擇儲存元件與 郴近儲存元件之間的寄生電容通路而對字線丨及 WLn+i實施不同於其他字線之處理。亦即,一施加至鄰近 儲存元件之Vread之溫度補冑電壓心電容方式輕合至選 擇諸存元件《而將其臨限電壓偏移得更高。特別對於上 述針對每一程式化狀態採用多個讀取位準之讀取/驗證技 術,此可能成問題。此外,合意之情形係關於溫度補償以 彼此不同之方式來處理字線WLn-Ι及WLn+Ι。 圖17係一解釋在讀取/驗證操作期間某些波形之行為之 時序圖1中選擇字線直接鄰近—源極侧選擇閘極。波形 SGD、SGS(選項1}及⑽(選項2)與圖仏中相同。選擇肌 及源極波形(未繪示)亦與圖15a中㈣。此處,選擇字線 WL0直接㈣源極側選擇閘極。如所提及,對於某此讀 取/驗證技術’合意之情形係不將溫度補償用於施加至鄰 存元件之電晶體之讀取錢。此等鄰近電晶體在 括源極側選擇閉極而在另一側上包括與脱1相關 存疋件。因此’在—種可能之方法中,所施加之電 121185.doc -43- 200805382 壓未經溫度補償,或溫度補償一較施加至其他非選擇字線 及另一選擇閘極(其不直接鄰近選擇儲存元件之汲極側選 擇閘極)之補償為少之量。特定而言,Vsgs可施加至SGS, Vread-tc可施加至WL〇及WL2至WL3 i,VRad可施加至 WL1 ’且Vsgd-tc可施加至SGD。For either or both of the word lines WLn_i and WLn+丨 directly adjacent to the selected word line, the applied read voltage is not temperature compensated, or is temperature compensated-reduced by an amount 'eg' and applied to The temperature of other unselected word lines is reduced by the amount of time. The optimal compensation for a particular pair of memory devices can be determined by testing. The desirable situation is to treat the word lines 丨 and WLn+i differently than other word lines by selecting a parasitic capacitance path between the storage element and the near storage element. That is, a temperature applied to the Vread adjacent to the storage element is capacitively coupled to the selected storage element to shift its threshold voltage higher. This may be problematic especially for the above described read/verify techniques that employ multiple read levels for each stylized state. Further, it is desirable that the word lines WLn-Ι and WLn+Ι are processed in a manner different from each other with respect to temperature compensation. Figure 17 is a timing diagram illustrating the behavior of certain waveforms during a read/verify operation. Figure 1 shows the selected word line directly adjacent to the source side select gate. Waveforms SGD, SGS (options 1} and (10) (option 2) are the same as in Figure 选择. Selecting the muscle and source waveforms (not shown) is also the same as in Figure 15a (4). Here, select word line WL0 directly (four) source side The gate is selected. As mentioned, for some such read/verify techniques, it is desirable to not use temperature compensation for the reading of the transistor applied to the adjacent component. These adjacent transistors are included in the source. The side selects the closed pole and the other side includes the disconnecting element. Therefore, in the possible method, the applied electricity 121185.doc -43- 200805382 is not temperature compensated, or temperature compensated The compensation is less than the amount applied to other non-selected word lines and another select gate (which is not directly adjacent to the drain side select gate of the selected storage element). In particular, Vsgs can be applied to SGS, Vread-tc It can be applied to WL〇 and WL2 to WL3 i, VRad can be applied to WL1 ' and Vsgd-tc can be applied to SGD.

圖18係一解釋在讀取/驗證操作期間某些波形之行為之 時序圖,其中選擇字線直接鄰近一汲極侧選擇閘極。波形 SGD、SGS(選項1)及8&lt;38(選項2)與圖15a中相同。選擇bl 及源極波形(未繪示)亦與圖15a中相同。此處,選擇字線 WL3 1直接鄰近汲極侧選擇閘極。如所提及,對於某些讀 取/驗證技術’合意之情形係不將溫度補償用於施加至鄰 近選擇儲存it件之電晶體之讀取電壓。此等鄰近電晶體在 一側上包括汲極侧選擇閘極而在另一侧上包括wl3〇。因 此在種可能之方法中,所施加之電壓未經溫度補償, 或溫度補償一較施加至其他非選擇字線及另一選擇閘極 (不直接鄰近該選擇料元件之源極側選擇閘極)之補償為 少之量。特^而言,Vsgs七可施加至⑽,Ha可施 加至WL0至WL39, Vread可施加至饥3〇,且可施加 至SGD。因此,在圖17及圖18之方法中,可分別根據鄰近Figure 18 is a timing diagram illustrating the behavior of certain waveforms during a read/verify operation in which the selected word line is directly adjacent to a drain side select gate. The waveforms SGD, SGS (option 1) and 8 &lt; 38 (option 2) are the same as in Fig. 15a. The selection of bl and source waveforms (not shown) are also the same as in Figure 15a. Here, the selected word line WL3 1 is directly adjacent to the drain side selection gate. As mentioned, for certain read/verify techniques, it is desirable to not use temperature compensation for the read voltage applied to the transistor adjacent to the selected storage device. These adjacent transistors include a drain side select gate on one side and wl3 turns on the other side. Therefore, in a possible method, the applied voltage is not temperature compensated, or the temperature compensation is applied to other unselected word lines and another selected gate (not directly adjacent to the source side select gate of the selected material element) The compensation is a small amount. Specifically, Vsgs can be applied to (10), Ha can be applied to WL0 to WL39, Vread can be applied to hunger, and can be applied to SGD. Therefore, in the methods of FIGS. 17 and 18,

儲存元件是否經選擇或未煙14 s y L 〜干\不、,工選擇而將施加至一個或兩個選 擇閘極之電麼設定至不同之也進在 1 J &lt;位早,例如一未經溫度補償之 位準或經補償之位準。 圖”係一流程圖,其闡述一種用於程式化非揮發性記憶 體之方法之-實施例。於某些實施方案中,儲存元件於程 121185.doc -44 - 200805382 式化之前被擦除(以塊或其他單元為單位)。於步驟1900 中’控制器發出一,,資料載入,,命令而控制電路310接收輸 入。於步驟1905中,將指定頁面位址之位址資料自控制器 或主機輸入至解碼器314。於步驟1910中,將已定址頁面 之一程式化資料頁面輸入至一資料緩衝器以供程式化。將 該資料鎖存於適宜之鎖存器組中。於步驟1915中,控制器 向狀態機312發出一,,程式化&quot;命令。 在由”程式化”命令觸發後,使用施加至適宜字線的圖13 之步階式脈動 1310、1320、1330、1340、1350、.··將在步 驟1 91 0中鎖存之資料程式化至狀態機3 12控制之選擇儲存 几件中。於步驟1920中,將程式化電壓Vpgm初始化成開 始脈動(例如12伏或另一值)並將狀態機3 12維持之程式化計 數器PC初始化成〇。於步驟1925中,將第一Vpgm脈動施加 至選擇字線以開始程式化與選擇字線相關聯之儲存元件。 若邏輯”0”儲存於特定資料鎖存器中,此以指示程式化對 應之儲存元件,則將對應位元線接地。另一方面,若邏輯 ”1”儲存於該特定鎖存器中,此指示對應之儲存元件仍保 持在其當前資料狀態中,則將對應位元線連接至vdd以禁 止程式化。 於步驟1930中,如所論述,使用經適宜溫度補償之電壓 及未經溫度補償或經溫度補償一減小之量之電壓來驗證選 擇儲存元件之狀態。若偵測到一選擇儲存元件之目標臨限 包壓已達到該適宜位準,則將儲存於對應資料鎖存器中之 貧料改變為邏輯”1”。若偵測到該臨限電壓尚未達到該適 121185.doc -45- 200805382 宜^立準,目彳γ 、、]不改變儲存於該對應資料鎖存器中之資料。 此方式’ “、、而私式化一具有一儲存於其對應資料鎖存 之邏輯丨,1丨丨夕〇 σ甲 二^位70線。當所有資料鎖存器皆儲存邏輯,,” 、時’錄態機(藉由上述經連線”或&quot;類型之機制)知曉所有 二擇:绪f τΜ牛皆已程式化。於步驟⑺”中,檢查該等資料 〔=疋否均儲存有邏輯’’ 1,'。若如此,則該程式化過程 70成且口所有選擇儲存元件皆經程式化及驗證而係成功。 步驟Μ0中報告”通過”狀態。於一實施例中,如先前參 ^圖15.18所述,步驟193()之驗證包括將經温度補償之電 壓提供至一個或多個非選擇字線,並提供至-個或多個選 擇閘極。 ' ”若於步驟1935中確定並非所有資料鎖存器皆儲存邏輯 ”1”,則該程式化過程繼續。於步驟1945中,對照程式化 限制值PCmax檢查程式化計數器pc。程式化限制值之一實 例為二十,但亦可使用其他數量。若程式化計數器pc不小 ; 於PCmax,則程式化過程已失敗並於步驟195〇中報告,,失 敗”狀態。若程式化計數器PC小於PCmax,則Vpgm位準增 加步長並於步驟1955中增量程式化計數器pc。於步驟1955 後,該過程循環回至步驟1925以施加下一Vpgm脈動。 出於例證及說明之目的,上文已對本發明進行了詳細說 明。本文不意欲包羅無遺或將本發明限制於所揭示之精確 形式。根據上文之教示亦可作出許多種修改及改變。所述 實施例之選擇旨在最佳地解釋本發明之原理及其實際應 用,藉以使其他4(省此項技術者能夠以適合於所構想具體 121185.doc -46 - 200805382 應用之各種實施例形式及使用各種修改來最佳地利用本發 明。本發明之範嘴意欲由隨附申請專利範圍來界定。 【圖式簡單說明】 圖1係一 NAND串中之俯視圖。 圖2係NAND串之等效電路圖。 圖3係NAND串之剖視圖。 圖4係-NAND快閃儲存元件陣列之方塊圖。 ,, 圖5係一非揮發性記憶體系統之方塊圖。 ' 圖6係一非揮發性記憶體系統之方塊圖。 圖7係一繪示一感測塊之一實施例之方塊圖。 圖8圖解說明—記憶體陣列組織成所有位元線記憶體架 構之塊之組織形式之實例。 圖9圖解說明一記憶體陣列組織成奇偶記憶體架構之塊 之組織形式之實例。 圖10繪示一實例性臨限電壓分佈組。 圖11繪示一實例性臨限電壓分佈組。 “ 圖12A-c顯示各種臨限電壓分佈並闌述—用於程式化非 揮發性記憶體之過程。 圖13係-在程式化期間施加至非揮發性儲存元件之控制 閘極之實例性波形。 圖14圖解說明一臨限電壓隨溫度及字線位置之變化。 圖15a係一解釋在讀取/驗證操作期間某些波形之行為之 時序圖,#中將經溫度補償之電壓施加至所有非選擇字線 並施加至兩個選擇閘極。 121185.doc -47- 200805382 圖15b繪示圖15a之時序圖,其中根據字線位置將不同之 經溫度補償電壓施加至選擇字線。 圖16係一解釋在讀取/驗證操作期間某些波形之行為之 時序圖,其中將經溫度補償之電壓施加至除直接鄰近選擇 子線之子線以外的所有非選擇字線,並施加至兩個選擇閘 極。 圖17係一解釋在讀取/驗證期間某些波形之行為之時序 圖,其中選擇字線直接鄰近一源極側選擇閘極。 圖18係一解釋在讀取/驗證期間某些波形之行為之時序 圖’其中選擇字線直接鄰近一汲極側選擇閘極。 圖19係一流程圖,其闡述一用於程式化非揮發性記情體 之過程之一實施例。 【主要元件符號說明】 100 電晶體 100FG 浮動閘極 100CG 控制閘極 102 電晶體 102FG 浮動閘極 102CG 控制閘極 104 電晶體 104CG 控制閘極 104FG 浮動閘極 106 電晶體 106CG 控制閘極 121185.doc 200805382 106FG 120 120CG 122 122CG 126 128 130 132 134 136 138 140 150 204 206 286 298 300 310 312 314 315 316 浮動閘極 電晶體 控制閘極 電晶體 控制閘極 沒極端子 N+摻雜層 N+摻雜擴散區 N+摻雜擴散區 N+摻雜擴散區 N+掺雜擴散區 N+摻雜擴散區 p井區 NAND 串 源極線 位元線 記憶體裝置 記憶體晶粒 記憶體陣列 控制電路 狀態機 晶片上位址解碼器 溫度補償控制 功率控制模組 121185.doc -49- 200805382 318 線 320 資料匯流排 330 解碼器 330A 列解碼器 330B 列解碼器 350 控制器 360 解碼器 360A 行解碼器 360B 行解碼器 365 讀取/寫入電路 365A 讀取/寫入電路 365B 讀取/寫入電路 370 感測電路 372 匯流排 380 感測模組 382 位元線鎖存器 390 共用部分 392 處理器 393 輸入線 394 資料鎖存器 396 I/O介面 400 感測塊 1250 臨限電壓分佈 1300 波形 121185.doc -50- 200805382 1310 共用部分 1312 驗證脈動組 1320 共用部分 1322 驗證脈動組 1330 共用部分 1332 驗證脈動組 1340 共用部分 1342 驗證脈動組 1350 共用部分 1352 驗證脈動組 1410 線 1420 線 1430 線 1440 線 1450 線 A 狀態 B 狀態 Bf 狀態 C 狀態 E 狀恶 121185.doc - 51 -Whether the storage component is selected or not smoked 14 sy L ~ dry \ no, the work selection will be applied to one or two select gates, the power is set to be different, also in the 1 J &lt; bit early, for example, one Temperature compensated level or compensated level. Figure is a flow diagram illustrating an embodiment of a method for staging non-volatile memory. In some embodiments, the storage element is erased prior to characterization of 121185.doc -44 - 200805382 (in blocks or other units). In step 1900, the controller issues a data entry, and the control circuit 310 receives the input. In step 1905, the address of the specified page address is self-controlled. The host or host is input to the decoder 314. In step 1910, one of the programmed pages of the addressed page is input to a data buffer for stylization. The data is latched into a suitable set of latches. In step 1915, the controller issues a , stylized &quot;command to the state machine 312. After being triggered by the "stylized" command, the stepped pulsations 1310, 1320, 1330 of Figure 13 applied to the appropriate word line are used, 1340, 1350, . . . program the data latched in step 1 91 0 to the selection of the control of the state machine 312. In step 1920, the programmed voltage Vpgm is initialized to start pulsing (eg 12 Volt or another And initializing the stylized counter PC maintained by the state machine 3 to 〇. In step 1925, the first Vpgm pulse is applied to the selected word line to begin programming the storage elements associated with the selected word line. Stored in a specific data latch, which indicates the stylized corresponding storage element, grounds the corresponding bit line. On the other hand, if the logic "1" is stored in the specific latch, this indication corresponds to If the storage element remains in its current data state, the corresponding bit line is connected to vdd to disable stylization. In step 1930, as discussed, the appropriate temperature compensated voltage is used and either temperature compensated or temperature compensated A reduced amount of voltage is used to verify the state of the selected storage element. If it is detected that the target threshold of a selected storage element has reached the appropriate level, the lean material stored in the corresponding data latch is changed. It is logic "1". If it is detected that the threshold voltage has not yet reached the appropriate 121185.doc -45-200805382, the target γ, ,] does not change the data stored in the corresponding data latch. In this way, ',, and privately, one has a logical 储存 stored in its corresponding data latch, 1 丨丨 〇 σ 二 2 ^ 70 lines. When all data latches store logic, the "," recorder (by the above-mentioned "wired" or "type" mechanism) knows all the two choices: the thread f τ Μ has been programmed. In step (7), it is checked that the data [= 均 is stored with logic ''1'. If so, the stylization process 70 is successful and all the selected storage elements are programmed and verified. The "pass" state is reported in step 。 0. In one embodiment, as previously described in Figure 15.18, the verification of step 193() includes providing a temperature compensated voltage to one or more unselected word lines and providing To one or more selection gates. ' ” If it is determined in step 1935 that not all data latches store a logical "1", then the stylization process continues. In step 1945, the stylized counter pc is checked against the stylized limit value PCmax. An example of a stylized limit value is twenty, but other quantities can be used. If the stylized counter pc is not small; at PCmax, the stylization process has failed and is reported in step 195, "failed" state. If the stylized counter PC is less than PCmax, the Vpgm level is increased by the step size and in step 1955 The programmatic counter PC is incrementally incremented. After step 1955, the process loops back to step 1925 to apply the next Vpgm pulse. The present invention has been described above for purposes of illustration and description. This document is not intended to be exhaustive or The invention is limited to the precise forms disclosed, and many modifications and changes can be made in accordance with the teachings hereinabove. The selection of the embodiments is intended to best explain the principles of the invention and its application. (The skilled person in the art is able to make the best use of the present invention in various embodiment forms suitable for the specific application of the specific model 121185.doc -46 - 200805382 and using various modifications. The scope of the present invention is intended to be covered by the accompanying patent application. Figure 1 is a top view of a NAND string. Figure 2 is an equivalent circuit diagram of a NAND string. Figure 3 is a cross-sectional view of a NAND string. Figure 5 is a block diagram of a non-volatile memory system. Figure 6 is a block diagram of a non-volatile memory system. Figure 7 is a block diagram of a sensing block. Figure 8 illustrates an example of an organization of memory blocks organized into blocks of all bit line memory architectures. Figure 9 illustrates the organization of a memory array organized into blocks of parity memory architecture. An example of a form is shown in Figure 10. Figure 10 illustrates an exemplary threshold voltage distribution group.Figure 11 illustrates an exemplary threshold voltage distribution group. "Figure 12A-c shows various threshold voltage distributions and details - for stylized non- Process of Volatile Memory Figure 13 is an exemplary waveform of a control gate applied to a non-volatile storage element during stylization. Figure 14 illustrates a threshold voltage as a function of temperature and word line position. A timing diagram that explains the behavior of certain waveforms during a read/verify operation, in which a temperature compensated voltage is applied to all unselected word lines and applied to two select gates. 121185.doc -47- 200805382 Figure 15b shows Figure 15a is a timing diagram in which different temperature compensated voltages are applied to selected word lines based on word line locations. Figure 16 is a timing diagram illustrating the behavior of certain waveforms during read/verify operations, where temperature compensated The voltage is applied to all non-selected word lines except the sub-lines directly adjacent to the selected sub-line and applied to the two select gates. Figure 17 is a timing diagram illustrating the behavior of certain waveforms during read/verify, where The selected word line is directly adjacent to a source side select gate. Figure 18 is a timing diagram illustrating the behavior of certain waveforms during read/verify 'where the selected word line is directly adjacent to a drain side select gate. Figure 19 is a flow diagram illustrating one embodiment of a process for stylizing a non-volatile quotation body. [Main component symbol description] 100 transistor 100FG floating gate 100CG control gate 102 transistor 102FG floating gate 102CG control gate 104 transistor 104CG control gate 104FG floating gate 106 transistor 106CG control gate 121185.doc 200805382 106FG 120 120CG 122 122CG 126 128 130 132 134 136 138 140 150 204 206 286 298 300 310 312 314 315 316 Floating gate transistor control gate transistor control gate without terminal N+ doped layer N+ doped diffusion region N+ Doped diffusion region N+ doped diffusion region N+ doped diffusion region N+ doped diffusion region p well region NAND string source line bit line memory device memory grain memory array control circuit state machine on-chip address decoder temperature Compensation Control Power Control Module 121185.doc -49- 200805382 318 Line 320 Data Bus 330 Decoder 330A Column Decoder 330B Column Decoder 350 Controller 360 Decoder 360A Line Decoder 360B Line Decoder 365 Read/Write Circuit 365A Read/Write Circuit 365B Read/Write Circuit 370 Sensing Circuit 372 Bus Bar 380 Sensing Mode Group 382 Bit Line Latch 390 Common Part 392 Processor 393 Input Line 394 Data Latch 396 I/O Interface 400 Sensing Block 1250 Threshold Voltage Distribution 1300 Waveform 121185.doc -50- 200805382 1310 Common Part 1312 Verification Pulsation group 1320 Common portion 1322 Verify pulsation group 1330 Common portion 1332 Verify pulsation group 1340 Common portion 1342 Verify pulsation group 1350 Common portion 1352 Verify pulsation group 1410 Line 1420 Line 1430 Line 1440 Line 1450 Line A State B State Bf State C State E Shape Evil 121185.doc - 51 -

Claims (1)

200805382 十、申請專利範圍: 1· 一種非揮發性儲存系統,其包括: 一組非揮發性儲存元件;及 -個或多個電路’其藉由複數個控制線與該組非揮發 性儲存兀件通信,該一個或多個電路⑷將至少—第一電200805382 X. Patent Application Range: 1. A non-volatile storage system comprising: a set of non-volatile storage elements; and - one or more circuits 'by a plurality of control lines and the set of non-volatile storage ports Communication, the one or more circuits (4) will be at least - the first 2. [知加至冑擇控制線以確定至少一與該選擇控制線相 關聯之第-非揮發性儲存㈣之—程式化狀態,及⑻在 -其中施加該至少—第一電壓之時間之至少一部分期 間’將:經温度補償之電壓施加至至少—與該組非揮發 性儲存元件相關聯之第一非選擇控制線。 如凊求項1之非揮發性儲存系統,其中·· 該—個或多個電路將一系列該等第-電壓施加至該選 擇控制線以確定該至少一第一非揮發性儲存元件之該程 式化,態,該經溫度補償之電壓係在將該系列第一電壓 中之每一第—電壓施加至該選擇控制線時施加。 3. 如請求項丨之非揮發性儲存系統,其中: 該經溫度補償之電麼之一位準係根據該至少一第一非 選擇控制線在該複數個控制線中之一相對位置而設定。 4. 如請求項1之非揮發性儲存系統,其中·· 在一其中施加該至少一第一電壓之時間之至少一部分 期間,該-個或多個電路將一電壓施加至一直接鄰近該 選擇控制線之第二非選擇控制線,該㈣未經溫度補償 或經溫度補償-相對於施加至該第—非選擇控制線之該 經溫度補償電壓之減小量。 121185.doc 200805382 5.如請求項1之非揮發性儲存系統,其中: 二一,中施加該至少一第—電壓之時間之至少一部分 忒1固或多個電路將-經溫度補償之電壓施加至 包括該至少_笛_ 非揮叙性儲存元件的一 NAND串之至 夕個選擇閘極之一控制閘極。 6_如請求項1之非揮發性儲存系統,其中·· 將與該至少—第—非選擇控 7. 之非揮發性儲存㈣維持在—導電狀態中。 °月,項1之非揮發性儲存系統,其中: 在:一第一電壓包括一讀取電壓,該讀取電壓用於 第一非揮發性儲存元件已程式化後讀取該至 …I揮發性儲存元件之該程式化狀態。 8·如吻求項1之非揮發性儲存系統,其中: 亥至少一第一電壓包括—驗證電壓,該 確定該至少一第一非 电铿用於 望之程式化狀態。 所期 9. 如請求们之非揮發性儲存系統,其中: 該至少一第一電壓經溫度補償。 10. 如請求項1之非揮發性儲存系統,其中·· :二=置:_選擇控制線在該複數個控 不目對位置來加以溫度補償。 Π·如請求項1之非揮發性儲存系統,其中: 該組非揮發性儲存元件包括多位準儲存元件。 12. -種用於操作非揮發性儲存器之方法,其包括: 121185.doc 200805382 將至少_第—電壓施加至-選擇字線以確定至少-盘 =擇字線相關聯之卜非揮發性儲存元件之—程心匕 欢。4至少一第一非揮發性儲存元件設置於一組非揮 發性儲存元件中;及 13. 14. \ / 15 16. 爱在—其中施加該至少一第一電壓之時間之至少一部分 ^ 將一經溫度補償之電壓施加至至少一與該組非揮 發性健存元件相關聯H選擇字線。 如請求項12之方法,其中: 將系列忒等第一電壓施加至該選擇字線以確定該至 '第一非揮發性儲存元件之該程式化狀態,該經溫度 補償之電壓係在將該系列第一電壓中之每一第一電壓施 加至該選擇字線時施加。 如請求項12之方法,其中: 據為至y 弟一非選擇字線在複數個與該組非揮發 性儲存元件相關聯之字線中之一相對位置來設定該經溫 度補償電壓之一位準。 如請求項12之方法,其進一步包括: 在一其十施加該至少一第一電壓之時間之至少一部分 期間,將一電壓施加至一直接鄰近該選擇字線之第二非 選擇字線,該電壓未經溫度補償或經量溫度補償一相對 於施加至該第一非選擇字線之該經溫度補償電壓之減小 之量。 如請求項12之方法,其進一步包括: 在一其令施加該至少一第一電壓之時間之至少一部分 121185.doc 200805382 期間’將一經溫度補償之電壓施加至包括該至少一第一 非揮务性儲存元件的一 NAND串之至少一個選擇閘極之 一控制閘極。 17. 如請求項12之方法,其中: /、、、工/皿度補彳貞之電壓足以將與該至少一第一非選擇字 線相關聯之非揮發性儲存元件維持在一導電狀態中。 18. 如請求項12之方法,其中:2. [Knowing to the selection control line to determine at least one of the first non-volatile storage (four) associated with the selection control line - a stylized state, and (8) at - the time during which the at least - first voltage is applied At least a portion of the period 'will: a temperature compensated voltage is applied to at least - a first non-selected control line associated with the set of non-volatile storage elements. The non-volatile storage system of claim 1, wherein the one or more circuits apply a series of the first voltages to the selection control line to determine the at least one first non-volatile storage element Stylized, the temperature compensated voltage is applied when each of the series of first voltages is applied to the select control line. 3. The non-volatile storage system of claim 1, wherein: the temperature compensated one of the levels is set according to the relative position of the at least one first non-selected control line in one of the plurality of control lines . 4. The non-volatile storage system of claim 1, wherein the one or more circuits apply a voltage to a direct proximity to the selection during at least a portion of the time during which the at least one first voltage is applied a second non-selected control line of the control line, the (four) temperature compensated or temperature compensated - relative to the reduced amount of the temperature compensated voltage applied to the first non-selected control line. The non-volatile storage system of claim 1, wherein: wherein: at least a portion of the time during which the at least one first voltage is applied, or a plurality of circuits, the temperature-compensated voltage is applied. One of the gates of the NAND string including the at least one _ _ non-volatile storage element controls the gate. 6_ The non-volatile storage system of claim 1, wherein the non-volatile storage (four) of the at least - the first non-selective control is maintained in a conductive state. The non-volatile storage system of item 1, wherein: the first voltage comprises a read voltage, and the read voltage is used to read the first non-volatile storage element after the program is programmed to... The stylized state of the sexual storage element. 8. The non-volatile storage system of claim 1, wherein: at least a first voltage comprises a verification voltage, and wherein the at least one first non-electricity is determined to be used in a stylized state. Period 9. The non-volatile storage system of the request, wherein: the at least one first voltage is temperature compensated. 10. In the non-volatile storage system of claim 1, wherein: ·===: The selection control line is temperature compensated at the plurality of control positions. The non-volatile storage system of claim 1, wherein: the set of non-volatile storage elements comprises a plurality of level storage elements. 12. A method for operating a non-volatile reservoir, comprising: 121185.doc 200805382 applying at least a _th voltage to a select word line to determine at least a disc = a word line associated with a non-volatile The storage component - Cheng Xinhuan. 4 at least one first non-volatile storage element is disposed in a set of non-volatile storage elements; and 13. 14. / / 15. 16. Love at - at least a portion of the time during which the at least one first voltage is applied A temperature compensated voltage is applied to at least one H select word line associated with the set of non-volatile storage elements. The method of claim 12, wherein: applying a first voltage such as series 忒 to the selected word line to determine the stylized state to the 'first non-volatile storage element, the temperature compensated voltage is Applied when each of the series first voltages is applied to the selected word line. The method of claim 12, wherein: the one of the temperature compensated voltages is set according to a relative position of a non-selected word line to a plurality of word lines associated with the set of non-volatile storage elements. quasi. The method of claim 12, further comprising: applying a voltage to a second unselected word line directly adjacent to the selected word line during at least a portion of a time during which the at least one first voltage is applied The voltage is not temperature compensated or temperature compensated by an amount relative to the decrease in the temperature compensated voltage applied to the first unselected word line. The method of claim 12, further comprising: applying a temperature compensated voltage to the at least one first non-spreading during at least a portion of time 121185.doc 200805382 at which the at least one first voltage is applied One of the at least one selection gate of a NAND string of the storage element controls the gate. 17. The method of claim 12, wherein: /,,, and the voltage of the device is sufficient to maintain the non-volatile storage element associated with the at least one first unselected word line in a conductive state. 18. The method of claim 12, wherein: 名至夕一第一電壓包括一讀取電壓,該讀取電壓用於 在^至少ϋ揮發性_存元件已程式化後讀取該至 少第非揮發性儲存元件之該程式化狀態。 19·如請求項12之方法,其中: 二至夕帛電壓包括一驗證電壓,該驗證電壓用於 確疋口亥至)-第-非揮發性儲存元件是否已達到一所期 望之程式化狀態。 ^ 20·如請求項12之方法,其中·· 對該至少一第—電壓進行溫度補償。 21.如請求項12之方法,其中: 根據該選擇字線在葙* 關聯之字線中之一相:與該組非揮發性儲存元件柄 溫度補償。相對位置來對該至少一第-電壓進行 121185.docThe first voltage includes a read voltage for reading the stylized state of the at least non-volatile storage element after at least the volatile component has been programmed. The method of claim 12, wherein: the second to fourth voltage comprises a verification voltage for determining whether the first non-volatile storage element has reached a desired stylized state . The method of claim 12, wherein the at least one first voltage is temperature compensated. 21. The method of claim 12, wherein: one of the word lines associated with the selected word line is associated with the set of non-volatile storage element handles. Relative position to perform the at least one first-voltage 121185.doc
TW96119150A 2006-06-16 2007-05-29 System and method for operating non-volatile memory using temperature compensation of voltages of unselected word lines and select gates TWI336081B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US11/424,812 US7342831B2 (en) 2006-06-16 2006-06-16 System for operating non-volatile memory using temperature compensation of voltages of unselected word lines and select gates
US11/424,800 US7391650B2 (en) 2006-06-16 2006-06-16 Method for operating non-volatile memory using temperature compensation of voltages of unselected word lines and select gates

Publications (2)

Publication Number Publication Date
TW200805382A true TW200805382A (en) 2008-01-16
TWI336081B TWI336081B (en) 2011-01-11

Family

ID=38834211

Family Applications (1)

Application Number Title Priority Date Filing Date
TW96119150A TWI336081B (en) 2006-06-16 2007-05-29 System and method for operating non-volatile memory using temperature compensation of voltages of unselected word lines and select gates

Country Status (2)

Country Link
TW (1) TWI336081B (en)
WO (1) WO2007149676A2 (en)

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3954245B2 (en) * 1999-07-22 2007-08-08 株式会社東芝 Voltage generation circuit
JP3829088B2 (en) * 2001-03-29 2006-10-04 株式会社東芝 Semiconductor memory device

Also Published As

Publication number Publication date
WO2007149676A3 (en) 2008-03-13
WO2007149676A2 (en) 2007-12-27
TWI336081B (en) 2011-01-11

Similar Documents

Publication Publication Date Title
US7342831B2 (en) System for operating non-volatile memory using temperature compensation of voltages of unselected word lines and select gates
US7391650B2 (en) Method for operating non-volatile memory using temperature compensation of voltages of unselected word lines and select gates
US7590002B2 (en) Resistance sensing and compensation for non-volatile storage
US7616498B2 (en) Non-volatile storage system with resistance sensing and compensation
JP5575244B2 (en) Selective memory cell program and erase
US8942043B2 (en) Non-volatile storage with process that reduces read disturb on end wordlines
TWI330848B (en) System and method for read opeartion for non-volatile storage with compensation for coupling
KR20110061625A (en) Improving data retention of last word line of non-volatile memory arrays
WO2013126109A1 (en) Temperature based compensation during verify operations for non-volatile storage
EP2078303B1 (en) Reading of a nonvolatile memory cell by taking account of the stored state of a neighboring memory cell
EP2577671A1 (en) Programming non-volatile storage with synchronized coupling
KR20090073083A (en) Compensating for coupling during programming
JP4995273B2 (en) Verification operation for non-volatile storage devices using different voltages
KR20090073082A (en) Compensating for coupling based on sensing a neighbor using coupling
TWI378457B (en) Non-volatile storage system with intelligent control of program pulse duration and method thereof
WO2013138199A1 (en) Non-volatile storage with read process that reduces disturb
KR101080394B1 (en) Resistance sensing and compensation for non-volatile storage
TWI356417B (en) Method and system for reverse reading in non-volat
TW200805382A (en) System and method for operating non-volatile memory using temperature compensation of voltages of unselected word lines and select gates
TW200822344A (en) Method and system for reducing program disturb in non-volatile storage
TWI355663B (en) Method and system for configuring compensation

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees