TWI357603B - Reducing program disturb in non-volatile memory us - Google Patents

Reducing program disturb in non-volatile memory us Download PDF

Info

Publication number
TWI357603B
TWI357603B TW096141503A TW96141503A TWI357603B TW I357603 B TWI357603 B TW I357603B TW 096141503 A TW096141503 A TW 096141503A TW 96141503 A TW96141503 A TW 96141503A TW I357603 B TWI357603 B TW I357603B
Authority
TW
Taiwan
Prior art keywords
stylized
reverse
word line
storage element
voltage
Prior art date
Application number
TW096141503A
Other languages
Chinese (zh)
Other versions
TW200836203A (en
Inventor
Jeffrey W Lutze
Yingda Dong
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 US11/555,850 external-priority patent/US7440323B2/en
Priority claimed from US11/555,856 external-priority patent/US7468911B2/en
Application filed by Sandisk Corp filed Critical Sandisk Corp
Publication of TW200836203A publication Critical patent/TW200836203A/en
Application granted granted Critical
Publication of TWI357603B publication Critical patent/TWI357603B/en

Links

Classifications

    • 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
    • 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
    • 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
    • G11C16/08Address circuits; Decoders; Word-line control 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
    • G11C16/30Power supply 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

Landscapes

  • Read Only Memory (AREA)
  • Non-Volatile Memory (AREA)

Description

1357603 九、發明說明: 【發明所屬之技術領域】 本發明係關於非揮發記憶體。 【先前技術】 半導體3己憶體用於各種電子裝置中已變得日益風行。舉 例而言,非揮發半導體記憶體用於蜂巢式電話、數位攝影 機、個人數位助理、行動計算裝置、非行動計算裝置及其 他裝置中。電可擦可程式化唯讀記憶體(EEpR〇M)及快閃 • 記憶體在最風行之非揮發半導體記憶體之列。與傳統之全 特徵EEPROM相比,在快閃記憶體(亦為一類型之 EEPROM)之情況下,可單步地擦除整個記憶體陣列之内容 或記憶體之一部分之内容。 傳統EEPROM與快閃記憶體使用位於半導體基板中之通 道區上方且與其絕緣之浮動閘極。浮動閘極位於源極區與 汲極區之間。控制閘極提供於浮動閘極上方且與其絕緣。 籲 如此形成之電晶體之臨限電壓(VTH)受保留於浮動閘極上 之電荷量的控制。亦即,在接通電晶體之前必須施加至控 制閘極以准許電晶體之源極與汲極之間的傳導之電壓之最 小量受浮動閘極上之電荷含量的控制。 某些EEPROM及快閃記憶體裝置具有用於儲存兩個範圍 " 之電荷的浮動閘極,且因此記憶體元件可在兩個狀態(例 如,擦除狀態及程式化狀態)之間得以程式化/擦除。因為 每一記憶體元件可儲存一個位元之資料,所以此種快閃記 憶體裝置有時被稱為二進位快閃記憶體裝置。 126206.doc 1357603 多態(亦被稱為多階)快閃記憶體裝置藉由識別多個相異 容許/有效程式化之臨限電壓範圍而予以實施。每一相異 臨限電壓範圍對應於在記憶體裝置中被編碼之資料位元之 集合的預疋值。舉例而言,當每一記憶體元件可置於對應 於四個相異臨限電壓範圍之四個離散電荷帶中之一者中 時,e亥元件可儲存兩個位元之資料。1357603 IX. Description of the Invention: TECHNICAL FIELD OF THE INVENTION The present invention relates to non-volatile memory. [Prior Art] Semiconductor 3 memory has become increasingly popular in various electronic devices. For example, non-volatile semiconductor memory is used in cellular phones, digital cameras, personal digital assistants, mobile computing devices, inactive computing devices, and other devices. Erasable Programmable Read Only Memory (EEpR〇M) and Flash • Memory is among the most popular non-volatile semiconductor memories. In the case of flash memory (also a type of EEPROM), the contents of the entire memory array or the contents of a portion of the memory can be erased in a single step compared to a conventional full-featured EEPROM. Conventional EEPROMs and flash memories use floating gates that are above and insulated from the channel regions in the semiconductor substrate. The floating gate is located between the source and drain regions. The control gate is provided above and insulated from the floating gate. The threshold voltage (VTH) of the thus formed transistor is controlled by the amount of charge remaining 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 permit conduction between the source and the drain of the transistor is controlled by the charge level on the floating gate. Some EEPROM and flash memory devices have floating gates for storing two ranges of "charges, and thus the memory elements can be programmed between two states (eg, erased state and stylized state) / erase. Since each memory element can store one bit of data, such a flash memory device is sometimes referred to as a binary flash memory device. 126206.doc 1357603 A polymorphic (also known as multi-order) flash memory device is implemented by identifying a plurality of distinct allowable/effectively programmed threshold voltage ranges. Each distinct threshold voltage range corresponds to an expected value of a set of data bits encoded in the memory device. For example, when each memory element can be placed in one of four discrete charge bands corresponding to four distinct threshold voltage ranges, the e-element can store two bits of data.

通常’將在程式化操作期間施加至控制閘極的程式化電 壓vPGM係作為隨著時間逝去增加量值之一連串脈衝而施 加。在一種可能方法中,脈衝之量值隨著每一連續脈衝而 被增加預定步長,例如,G2^.4 v。I可施加至快閃 记憶體7C件之控制閛極。在程式化脈衝之間的週期中,執 行驗也操作亦即,在連續程式化脈衝之間讀取正並行地 耘式化之群元件中之每一元件的程式化位準以判定其是 等於還疋大於兀件正被程式化至之驗證位準。對於多態快 閃記憶體元件陣列而言,可對元件之每-狀態執行驗證步 驟以判定元件是否已達到其與資料相關聯之驗證位準。舉 例而%夠在四個狀態中儲存資料之多態記憶體元件可 倉匕而要對二個比較點執行驗證操作。 此外,當程式化EEPR〇M或快閃記憶體裝置(諸如「反 及」串巾之「反及」快閃記憶體裝置)時,通常將Vp⑽施 加至控制閘極且使位元線接地,進而使得來自一單元或記 憶體元件(例如,儲存元件)之通道之電子注入至浮動間極 中。當電子累積於浮動閘極中時,浮動閘極變為帶負電且 記憶體元件之臨限電壓升高,使得記憶體元件被認為係處 126206.doc 1357603 於程式化狀態。可在標題為"source Side Self B〇〇stingTypically, the stylized voltage vPGM applied to the control gate during the stylization operation is applied as a series of pulses that increase in magnitude over time. In one possible approach, the magnitude of the pulse is increased by a predetermined step size with each successive pulse, for example, G2^.4v. I can be applied to the control bucks of the flash memory 7C device. In the period between the stylized pulses, the performing operation is also performed, that is, reading the stylized level of each of the group elements that are being parallelized in parallel between successively stylized pulses to determine that it is equal to It is also greater than the verification level to which the component is being programmed. For a multi-state flash memory device array, a verification step can be performed on each state of the component to determine if the component has reached its verification level associated with the data. For example, a polymorphic memory component that is capable of storing data in four states can perform a verification operation on two comparison points. In addition, when stylizing an EEPR〇M or a flash memory device (such as a "reverse" flash memory device), Vp(10) is typically applied to the control gate and the bit line is grounded. In turn, electrons from a channel of a cell or memory element (eg, a storage element) are injected into the floating interpole. When electrons accumulate in the floating gate, the floating gate becomes negatively charged and the threshold voltage of the memory component rises, causing the memory component to be considered to be in the stylized state of 126206.doc 1357603. Available under the heading "source Side Self B〇〇sting

Technique For Non-Volatile Memory"之美國專利第 6,859,397號及2005年2月3日公布之標題為” DetecUng 〇ver Programmed Mem〇ry"之美國專利申請公開案第 2005/0024939號中找到關於此程式化之更多資訊;兩個專 利文獻以引用之方式全文併入本文中。This stylization can be found in US Patent Application Publication No. 2005/0024939, entitled "DetecUng 〇ver Programmed Mem〇ry", published in U.S. Patent No. 6,859,397, issued to, the entire entire entire entire entire content Further information; both patent documents are herein incorporated by reference in their entirety.

然而,歸因於非揮發儲存元件彼此之接近,已在程式化 期間經歷各種形式之程式干擾。此外,預期此問題隨著 「反及」技術之進一步擴展而惡化。當未選定之非揮發儲 存元件之臨限電壓歸因於其他非揮發儲存元件之程式化而 移位時,發生程式干擾。各種程式干擾機制可限㈣揮發 儲存裝置(諸如,「反及」快閃記憶體)之可用操作窗。升壓 技術試圖藉由使被抑制程式化之「反及」$之通道區域升 壓至高電位並將含有待程式化之儲存元件的「反及」串之 通道區域連接至低電位(諸如,G v)來解決此_ n 給定升壓模式不可最佳地解決多個故障機制。 π 【發明内容】 本發明藉由提供用於操作減少程式干擾之非揮發儲存系 統之方法來解決上述及其他問題。 ’' 在一實施例中,一種用於操作非揮發儲存器之方法包括 程式化在一非揮發儲存元件集合中之一健存元件,其中該 非揮發儲存元件集合與許多字線通信,且該储存元件與— 選^字線通信。該方法進-步包括在程式化_將電I 之-集合施加至未選定之字線及基於升壓模式切換標準 126206.doc 1357603 自將電壓之第一集合施加至未選定之字線切換至將電壓之 第二集合施加至未選定之字線。電壓之第一集合至少部分 地不同於電壓之第二集合。舉例而言,該程式化可包括將 一脈衝串施加至敎之字線,其中當將該脈衝串中具有指 定振幅之程式化脈衝施加至選定之字線時,或當已將該脈 衝串中指定數目之程式化脈衝施加至選定之字線時,觸發 升壓模式切換標準。However, due to the proximity of the non-volatile storage elements to each other, various forms of program disturb have been experienced during stylization. In addition, it is expected that this problem will worsen with the further expansion of the "reverse" technology. Program disturb occurs when the threshold voltage of an unselected non-volatile storage element is shifted due to the stylization of other non-volatile storage elements. Various program interference mechanisms may limit (4) available operating windows for volatile storage devices (such as "reverse" flash memory). The boost technique attempts to boost the channel region of the "reverse" channel that is suppressed by stabilizing to a high potential and connect the channel region of the "reverse" string containing the storage element to be programmed to a low potential (such as G v) to solve this _ n given boost mode does not optimally solve multiple failure mechanisms. π [ SUMMARY OF THE INVENTION The present invention addresses these and other problems by providing a method for operating a non-volatile storage system that reduces program disturb. In one embodiment, a method for operating a non-volatile reservoir includes programming one of a set of non-volatile storage elements, wherein the set of non-volatile storage elements is in communication with a plurality of word lines, and the storing The component communicates with the - word line. The method further includes applying a set of the set of voltages to the unselected word lines and applying a first set of voltages to the unselected word lines to the unselected word lines based on the boost mode switching standard 126206.doc 1357603 A second set of voltages is applied to the unselected word lines. The first set of voltages is at least partially different from the second set of voltages. For example, the stylization can include applying a burst to a word line of 敎, wherein when a stylized pulse having a specified amplitude in the burst is applied to the selected word line, or when the pulse has been The boost mode switching criterion is triggered when a specified number of stylized pulses are applied to the selected word line.

在另一實施例中,一種用於操作非揮發儲存器之方法包 括在發生-非揮發儲存元件集合中之儲存元件之程式化的 第-程式化階段期間實施第一升壓模式,及在繼續該健存 疋件之程式化的第二程式化階段期間實施第二升壓模式。 該儲存元件之臨限電壓在第一程式化階段期間自第一位準 增加至第二位準且在第二程式化階段期間自第二位準增加 至第三位準。另外’第—程式化階段可包括多遍次程式化 技術中之帛次’且第二程式化階段可包括多遍次程式 化技術中之第二遍次。 在一方法中,在第一程式化階段中,將—脈衝串中之脈 衝之第—子集施加至該儲存元件’且在第二程式化階段 中,將該脈衝串中之脈衝之第二子集施加至該儲存元件。 在另-方法中,在第一程式化階段中,將第一脈衝串施 加至該儲存元件,且在第二程式化階段中,將第二脈衝串 施加至該儲存元件。 在另一實施例中,一種用於操作非揮發儲存器之方法包 括程式化在-非揮發儲存元件集合中之一儲存元件,其中 126206.doc 1357603 以非揮發儲存元❹合與 -脈衝串施加至…儲卜線通仏。該程式化包括將 法進存70件通信之選定之字線。該方 加至該= 衝串中之程式化脈衝之第-子集施 升壓模= = ;:之非揮發儲存㈣實施第-加至之式化脈衝之第二子集施 -升對未選定之非揮發儲存元件實施第 壓=料切換至對未選定之非揮發儲存元❹施第二升In another embodiment, a method for operating a non-volatile reservoir includes performing a first boost mode during a stylized first stylization phase of a storage component in a set of non-volatile storage components, and continuing The second boost mode is implemented during the stylized second stylization phase of the health condition. The threshold voltage of the storage element increases from a first level to a second level during the first stylization phase and from a second level to a third level during the second stylization phase. In addition, the 'first stylization stage may include the number of times in the multi-pass stylization technique' and the second stylization stage may include the second pass of the multi-pass stylization technique. In one method, in a first stylization phase, a first subset of pulses in a burst is applied to the storage element 'and in a second stylized phase, a second pulse in the burst A subset is applied to the storage element. In another method, a first pulse train is applied to the storage element in a first stylization phase, and a second pulse train is applied to the storage element in a second stylization phase. In another embodiment, a method for operating a non-volatile reservoir includes staging a storage element in a set of non-volatile storage elements, wherein 126206.doc 1357603 is coupled with a non-volatile storage element and a pulse train. To... the storage line is wanted. The stylization includes the method of depositing the selected word line for 70 communications. The square is added to the first subset of the stylized pulses in the = string to apply the boost mode = = ;: the non-volatile storage (four) implements the second subset of the first-plus-formed pulse The selected non-volatile storage element is subjected to the first pressure switch to switch to the unselected non-volatile storage element to apply the second liter

該非揮發儲存Μ集合可提供於許多「反及」串中,包 括提供該儲存元件之選定之「反及」卜及未選定之「反 及」串,其中第-及第二升a模式將該未選定之「反及」 串之通道升壓。另夕卜在一方法中’實施第一升壓模式包 括將通道升^不使在「反及」串之源極側上之通道之部 分與在「反及」串之汲極側上之通道之部分隔離,且實施 第二升壓模式包括使在「反及」串之源極側上之通道之部 分與在「反及」串之汲極側上之通道之部分隔離。 【實施方式】 本發明提供減少程式干擾之非揮發儲存系統及方法。 適用於實施本發明之記憶體系統之一實例使用「反及 快閃記憶體結構,該結構包括在兩個選擇閘極之間串聯配 置多個電晶體。該等串聯電晶體及該等選擇閘極被稱為 「反及」串。圖1為展示「反及」串之俯視圖。圖2為其等 效電路。圖1及圖2中所描繪之「反及」串包括串聯的且炎 於第一選擇閘極120與第二選擇閘極122之間的四個電晶體 126206.doc • 11· 1357603 100、102、104及106。選擇閘極120閘控「反及」串至位 元線126之連接。選擇閘極122閘控「反及」串至源極線 128之連接。藉由將適當電壓施加至控制閘極12〇CG而控The non-volatile storage stack can be provided in a plurality of "reverse" strings, including a selected "reverse" of the storage element and an unselected "reverse" string, wherein the first and second rise a modes will The channel of the "reverse" string that is not selected is boosted. In addition, in one method, 'implementing the first boost mode includes raising the channel so that the portion of the channel on the source side of the "reverse" string and the channel on the drain side of the "reverse" string Partial isolation, and implementing the second boost mode includes isolating portions of the channels on the source side of the "reverse" string from portions of the channels on the drain side of the "reverse" string. [Embodiment] The present invention provides a non-volatile storage system and method for reducing program interference. One example of a memory system suitable for use in practicing the present invention uses a "reverse flash memory structure that includes a plurality of transistors arranged in series between two select gates. The series transistors and the select gates It is called the "reverse" string. Figure 1 is a top view showing the "reverse" string. Figure 2 is its equivalent circuit. The "reverse" string depicted in Figures 1 and 2 includes four transistors 126206.doc • 11· 1357603 100, 102 connected in series and igniting between the first selection gate 120 and the second selection gate 122 , 104 and 106. The connection of the gate 120 gate control "reverse" string to the bit line 126 is selected. The connection of the gate 122 gate control "reverse" string to the source line 128 is selected. Controlled by applying an appropriate voltage to the control gate 12 〇 CG

. 制選擇閘極120。藉由將適當電壓施加至控制閘極122CG • 而控制選擇閘極122。電晶體1〇〇、1〇2、104及106中之每 一者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OFC^電晶體1〇2包括控制閘極 籲 l〇2CG及浮動閘極102FG。電晶體104包括控制閘極i〇4C(} 及浮動閘極104FG。電晶體106包括控制閘極106CG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102CG連接至字線WL2,控制閘極1〇4(:(}連接至字線 WL1,且控制閘極1〇6CG連接至字線WL〇。在一實施例 令,電晶體100、102、104及106為每-儲存元件,亦被稱 為記憶體單元。在其他實施例中,儲存元件可包括多個電 晶體或可不同於圖i及圖2中所描繪之記憶體元件。選擇閘 • 極120連接至選擇線SGD。選擇閘極122連接至選擇線 SGS。 圖3為据緣三個「反及」串之電路圖。使用「反及」結 構之快閃記憶體系統之典型架構將包括若干「反及」串。 舉:而言,在具有更多「反及」_之記憶體陣列中展示二 個「反及」串320、340及360。該等r反及」串中之每一 者包括兩個選擇閘極及四個儲存元件。雖然為簡單起見而 說:四個儲存元件,但現代「反及」,可具有高達(例如) 一十一個或六十四個儲存元件。 126206.doc •12- 1357603 舉例而&,「反及」串320包括選擇閘極322及327以及儲 存兀件323至326 ’「反及」串340包括選擇閘極342及347以 及儲存元件343至346,「反及」串36〇包括選擇閘極362及 367以及儲存元件至366。每一「反及」串藉由其選擇 閘極(例如,選擇閘極327、347或367)而連接至源極線。使 用選擇線SGS來控制源極側選擇閘極。各個「反及」串 32〇、34〇及360猎由選擇閘極322、342、362等中之選擇電 晶體而連接至各別位元線321、341及361。此等選擇電晶 體由汲極選擇線SGD控制。在其他實施例中,選擇線在 反及」串中未必為共同的。亦即,可對不同「反及」串 提供不同選擇線。字線WL3連接至儲存元件323、343及 363之控制閘極。字線WL2連接至儲存元件^ 及π# 之控制閘極。字線WL1連接至儲存元件325、Μ及365之 控制閘極。字線WL〇連接至儲存元件326、346及鳩之控 制閘極。如可看出,每一位元線及各別「反及」串包含儲 存凡件之陣列或集合之行°字線(WL3、WL2、WL1及 WL0)包含該陣列或集合之列。每一字線連接行中之每一 儲存元件之控制閘極。$,控制極可由字線自身提供。 舉例而言’字線WL2對健存元件324、344及⑽提供控制 閘極。實務上’在一字線上可存在上千個儲存元件。 每儲存7L件可儲存資料。舉例而言,當儲存一個位元 之數位資料暗,技Μ + 八 ,儲存几件之可能臨限電壓(VTH)之範圍 7兩個範圍’對兩個範圍指派邏輯資料"1”及"0”。在 反及」型快閃記憶體之—實例中,Vth在擦除儲存元件 126206.doc •13- 丄乃7603 =為負,純定義為邏輯,,卜Vth在程式化操作後為正 且被定義為邏輯"〇"。當Vth為負且試圖進行讀取時,儲存 ^件將接通則以正料邏輯"卜當Vth為正且試圖進行 讀取操作時,射元件㈣接通,此指㈣存邏輯"〇"。 儲存元件亦可儲存多個階層之資訊,例如,多個位元之數 位資料。在此狀況下,將Vth值之範圍分成該數目之資料 階層。舉例而tr ’若儲存四個階層之資訊,則將存在四個 VTH範圍,對其指派資料值,,u "、”ι〇"' ,,〇1"及"⑻"。在 「反及」$記憶體之-實例中,Vth在擦除操作後為負且 被定義為,’u”。正Vth值用於狀態” 1〇"、"〇1"及"⑼"。程式 化至儲存元件中之資料與元件之臨限電壓範圍之間的特1 關係取決於對儲存元件採用之資料編碼機制。舉例而言, 均以引用之方式全文併人本文中的美國專利第6,222,762號 及美國專利申凊公開案第2〇〇4/〇255〇9〇號描述用於多態快 閃儲存元件之各種資料編碼機制。. Select the gate 120. The selection gate 122 is controlled by applying an appropriate voltage to the control gate 122CG. Each of the transistors 1〇〇, 1〇2, 104, and 106 has a control gate and a floating gate. The transistor 1 has a control gate 100CG and a floating gate i〇OFC^ transistor 1〇2 including a control gate 〇1〇2CG and a floating gate 102FG. The transistor 104 includes a control gate i〇4C(} and a floating gate 104FG. The transistor 106 includes a control gate 106CG and a floating gate 106FG. The control gate i〇〇cG is connected to (or is) a word line WL3, and is controlled. The gate 102CG is connected to the word line WL2, the control gate 1〇4 (:(} is connected to the word line WL1, and the control gate 1〇6CG is connected to the word line WL〇. In an embodiment, the transistors 100, 102 104 and 106 are each storage element, also referred to as a memory unit. In other embodiments, the storage element may comprise a plurality of transistors or may be different from the memory elements depicted in Figures i and 2. Gate 120 is connected to select line SGD. Select gate 122 is connected to select line SGS. Figure 3 is a circuit diagram of three "reverse" strings according to the edge. Typical architecture of flash memory system using "reverse" structure It will include a number of "reverse" strings. For example, in the memory array with more "reverse" _, two "reverse" strings 320, 340 and 360 are displayed. These are reversed in the string. Each of them includes two select gates and four storage elements. Although for the sake of simplicity: four storage elements Pieces, but modern "reverse", can have up to (for example) eleven or sixty-four storage elements. 126206.doc • 12- 1357603 For example and &, "reverse" string 320 includes selection gate 322 And 327 and storage elements 323 to 326 'reverse' string 340 includes selection gates 342 and 347 and storage elements 343 to 346. The "reverse" string 36 includes select gates 362 and 367 and storage elements to 366. Each "reverse" string is connected to the source line by its selection gate (for example, selection gate 327, 347 or 367). The selection line SGS is used to control the source side selection gate. The strings 32〇, 34〇 and 360 are connected to the respective bit lines 321, 341 and 361 by the selection transistors of the selection gates 322, 342, 362, etc. These selection transistors are controlled by the drain selection line SGD. In other embodiments, the select lines are not necessarily common in the "reverse" string. That is, different select lines may be provided for different "reverse" strings. Control of word lines WL3 connected to storage elements 323, 343, and 363 Gate. WL2 is connected to the control gate of the storage component ^ and π#. The word line WL1 is connected to the gate. The control gates of the components 325, Μ and 365. The word line WL is connected to the storage elements 326, 346 and the control gate of the 。. As can be seen, each bit line and the respective "reverse" string contain the memory. The array or set of lines of the word line (WL3, WL2, WL1, and WL0) contains the array or set of columns. Each word line connects the control gate of each storage element in the row. $, the control word can be word The line itself provides. For example, the word line WL2 provides control gates for the health elements 324, 344, and (10). In practice, there can be thousands of storage elements on a word line. Data can be stored for every 7L of storage. For example, when storing a bit of digital data dark, tech + eight, store a few pieces of possible threshold voltage (VTH) range of 7 two ranges 'assigning logical data to both ranges"1" and &quot ;0". In the case of the "Flash Memory" example, Vth is in the erase storage element 126206.doc • 13- 丄 is 7603 = negative, purely defined as logic, and Vth is positive after the stylization operation and is Defined as logical "〇". When Vth is negative and an attempt is made to read, the storage component will be turned on and the logic is "when Vth is positive and an attempt is made to perform a read operation. The firing element (4) is turned on. This refers to (4) storing logic"". The storage element can also store information of multiple levels, for example, digital data of multiple bits. In this case, the range of Vth values is divided into the data hierarchy of the number. For example, if you store four levels of information, there will be four VTH ranges, assigning data values to them, u ", "ι〇"', 〇1" and "(8)". In contrast to the "$memory-instance" example, Vth is negative after the erase operation and is defined as 'u'. The positive Vth value is used for the state "1" ", "〇1" and "(9)" . The special relationship between the data programmed into the storage element and the threshold voltage range of the component depends on the data encoding mechanism used for the storage component. For example, U.S. Patent No. 6,222,762 and U.S. Patent Application Serial No. 2/4/255, filed to the entire disclosure of Data encoding mechanism.

反及J型快閃記憶體及其操作之相關實例提供於美國 專利第 5,386,422號、第 5,522,580號、帛 5,570,315號、第 5’774,397 號、帛 6,〇46,935 號、帛 6,456,528 號及第 6,522,580號中,其中之每一者以引用之方式併入本文中。 當程式化快閃儲存元件時,將程式化電壓施加至該儲存 凡件之控制閘極且將與該儲存元件相關聯之位元線接地。 來自通道之電子被〉主入浮動閘極中。當電子在浮動閘極中 累積時,浮動閘極變為帶負電且儲存元件之Vth上升。為 了將程式化電壓施加至正被程式化之儲存元件的控制閘 126206.doc 1357603 極,將此程式化電壓施加於適當之字線上。如上所述,在 該等「反及」串中之每一者中之一儲存元件共用同一字 線。舉例而言,當程式化圖3之儲存元件324時,程式化電 • 壓亦將施加至儲存元件344及364之控制閘極。 , 然而’在程式化其他「反及」串期間,在經抑制之「反 及」串處可能發生程式干擾,且有時在經程式化之「反 及」串自身處發生程式干擾。舉例而言,若「反及」串 320被抑制(例如,其為不含有當前正程式化之儲存元件的 未選定之「反及」_)且「反及」串34〇正被程式化(例如, 其為含有當前正程式化之儲存元件的選定之「反及」 串),則在「反及」串320處可能發生程式干擾。舉例而 言,若導通電壓VPASS較低,則不會將經抑制之「反及」 _之通道良好地升壓,且可無意地程式化未選定之「反 及」串的選定之字線。在另一種可能情況下,所升壓之電 壓可由閘極引發汲極漏電(GIDL)或其他漏電機制降低,進 φ 而導致相同問題。其他效應(諸如,歸因於儲存元件之間 的電容性耦合的經程式化之儲存元件中所儲存之電荷的移 位)亦可為有問題的。 . 圖4描繪展示升壓模式決定過程之概念圖。如開頭所提 及,程式干擾仍然為非揮發儲存系統之顯著問題。當未選 定之非揮發儲存元件之臨限電壓歸因於其他非揮發儲存元 件之程式化而移位時,發生程式干擾。可在先前程式化之 儲存元件以及尚未程式化之經擦除之儲存元件上發生程式 干擾。各種程式干擾機制可限制非揮發儲存裝置(諸如, 126206.doc -15· 1JJ /ou^ 反及」快閃記憶體)之可用操作窗。舉例而言,升壓技 術試圖藉由將經抑制之「反及」串之通道區域升壓至高電 位並將含有待程式化之儲存元件之「反及」串的通道區域 •連接至低電位(諸如,QV)來解決此問題。心,給定升壓 • _式不可最佳地解決多種故障機制。亦即,給定升愿模式 σ有效地解决特定程式干擾故障機制但在解決其他故障機 制時可能為低效的。通常,對升壓模式進行折衷或最佳化 • 以給出最好操作窗。此處,建議在程式化期間使用不同升 壓模式以更好地最佳化升壓。舉例而言,在一方法中,在 初始程式化期間使用一升壓模式且在程式化單個頁面或字 線接近結束時使用第二升壓模式以相對於程式干擾而改良 總裕度(margin)。 *可使用各種標準來決定使用哪一升壓模式,&自一升壓 模式切換至另-升壓模式之時刻。作為實例,可由升屋模 式決定過程(區塊415)來選擇在區塊4〇〇、4〇5及41〇處指示 • 之三個不同升壓模式。升壓模式包括(例如)在下文進一步 論述之自升壓(SB)、局部自升壓(LSB)、擦除區域自升屋 (EASB)及修正擦除區域自升壓(REASB)。一旦作出決定, . (例如)藉由將對應於選定之升壓模式之電壓集合施加至未 . 選定之字線來應用選定之升壓模式(區塊420)。舉例而言, 可由升壓模式切換決定過程(區塊415)使用一或多個升壓模 式切換標準(區塊425)。此等標準可包括程式化脈衝數目 (區塊430)、程式化脈衝振幅(區塊435)、程式化遍次號(區 塊440)、選定之字線之位置(區塊445)、粗略/精細程式化 126206.doc •16- 1357603 模式狀態(區塊450)、儲存元件是否 455),及由記拎體穿置 ,程式化條件(區塊 460)。 盾衣之數目(區塊 遍=次=:(例W程式化過…第-過-人還疋第一遍次在進行中。關於 化條件之標準可(例如)藉由偵测一群儲二:(=到程:Examples of anti-J-type flash memory and its operation are provided in U.S. Patent Nos. 5,386,422, 5,522,580, 5,570,315, 5'774,397, 帛6, 〇46,935, 帛6,456,528, and 6,522,580. Each of these is incorporated herein by reference. When the flash storage component is programmed, a programmed voltage is applied to the control gate of the storage device and the bit line associated with the storage component is grounded. The electrons from the channel are > mastered into the floating gate. When electrons accumulate in the floating gate, the floating gate becomes negatively charged and the Vth of the storage element rises. To apply a stylized voltage to the control gate 126206.doc 1357603 pole of the memory element being programmed, this stylized voltage is applied to the appropriate word line. As described above, one of the storage elements in each of the "reverse" strings shares the same word line. For example, when the storage element 324 of Figure 3 is programmed, the programmed voltage will also be applied to the control gates of storage elements 344 and 364. However, during the stylization of other "reverse" strings, program disturb may occur at the suppressed "reverse" string, and sometimes program disturb occurs at the stylized "reverse" string itself. For example, if the "reverse" string 320 is suppressed (eg, it is an unselected "reverse" _) that does not contain the currently stylized storage element and the "reverse" string 34 is being stylized ( For example, if it is a selected "reverse" string containing the currently stylized storage element, program disturb may occur at the "reverse" string 320. For example, if the turn-on voltage VPASS is low, the suppressed "reverse" channel will not be boosted well, and the selected word line of the unselected "reverse" string can be unintentionally programmed. In another possible case, the boosted voltage can be reduced by gate-induced drain leakage (GIDL) or other leakage mechanisms, which can cause the same problem. Other effects, such as shifting of charge stored in a stylized storage element due to capacitive coupling between storage elements, can also be problematic. Figure 4 depicts a conceptual diagram showing the boost mode decision process. As mentioned at the outset, program disturb remains a significant problem for non-volatile storage systems. Program disturb occurs when the threshold voltage of an unselected non-volatile storage element is shifted due to the stylization of other non-volatile storage elements. Program disturb can occur on previously stylized storage elements and on erased storage elements that have not been programmed. Various program disturb mechanisms can limit the available operating windows of non-volatile storage devices (such as 126206.doc -15·1JJ / ou^). For example, boosting techniques attempt to boost the channel region of the suppressed "reverse" string to a high potential and connect the channel region containing the "reverse" string of the storage element to be programmed to a low potential ( For example, QV) to solve this problem. Heart, given boost • _ type does not optimally solve multiple failure mechanisms. That is, the given wish mode σ effectively addresses the specific program interference failure mechanism but may be inefficient in solving other failure mechanisms. Typically, the boost mode is traded off or optimized to give the best operating window. Here, it is recommended to use different boost modes during stylization to better optimize boost. For example, in one method, a boost mode is used during initial stylization and a second boost mode is used at the end of a stylized single page or word line to improve the total margin relative to program disturb. . * Various standards can be used to determine which boost mode to use, & when switching from a boost mode to another boost mode. As an example, the three different boost modes indicated at blocks 4A, 4〇5, and 41〇 can be selected by the lift mode decision process (block 415). The boost mode includes, for example, self-boost (SB), local self-boost (LSB), erased area self-elevation (EASB), and modified erase area self-boost (REASB), discussed further below. Once a decision is made, the selected boost mode is applied (e.g., block 420) by applying a set of voltages corresponding to the selected boost mode to the unselected word line. For example, the boost mode switching decision process (block 415) may use one or more boost mode switching criteria (block 425). Such criteria may include the number of programmed pulses (block 430), the programmed pulse amplitude (block 435), the programmed pass number (block 440), the location of the selected word line (block 445), rough/ Fine stylization 126206.doc • 16-1357603 mode state (block 450), storage component 455), and pinning through the body, stylized condition (block 460). The number of shields (blocks = times =: (example W stylized ... first - over - people are still in progress for the first time. The criteria for the conditions can be (for example) by detecting a group of two :(=To Cheng:

塊或陣列)中之第-儲存元件或儲存元件之部分達到驗: ==刻來實施。在達到驗證條件時,可發 數目的標準可(例如)藉由追縱程式化循環之數目及以2 基礎來調整切換點而實施。舉例而言,若在脈衝串期間出 現切換點,則在記憶體裝置已經歷相對較多之循環後,在 脈衝串中可相對較早地出現該切換點,因為儲存元件在經 文額外程式化循環時傾向於較快地程式化。纟下文中更詳 細地描述升壓模式切換標準。 = 圖5描綠用於在程式化期間切換升壓模式之過程。可根 據流程圖來進一步理解上文所呈現之概念圖。在步驟5〇〇 處,程式化開始,且在步驟51〇處,應用第一升壓模式。 在決定步驟520處,若滿足切換標準,則切換至第二升壓 模式(步驟530)且程式化繼續(步驟54〇)直至其完成(步驟 550)為止。若在決定步驟52〇處不滿足切換標準,則繼續 應用第一升壓模式且程式化繼續(步驟525)。通常,藉由組 態記憶體裝置之一或多個控制電路以將適當電壓施加至與 儲存元件集合通信之字線而實施升壓模式。 126206.doc 1357603 切換升壓模式之決定可基於許多因素。通常,需要實施 對於當前程式化機制以及儲存元件及「反及」串之當前條 件而言為最佳的升壓模式。例如,非EASB#壓模式(諸 如,SB或LSB)對於初始程式化脈衝(在VpGM較低時)可相對 較有效,而EASB升壓模式(包括REASB)對於較高程式化 脈衝(在VPGM較高時)可相對較有效。在此狀況下,可基於The first storage element or part of the storage element in the block or array is tested: == implemented. When the verification condition is reached, the number of standards that can be issued can be implemented, for example, by tracking the number of stylized cycles and adjusting the switching point on a 2 basis. For example, if a switching point occurs during a burst, the switching point can occur relatively early in the burst after the memory device has experienced a relatively large number of cycles because the storage element is additionally stylized in the scripture. Loops tend to be faster programmed. The boost mode switching standard is described in more detail below. = Figure 5 depicts the process used to switch the boost mode during stylization. The conceptual diagram presented above can be further understood based on the flow chart. At step 5, stylization begins, and at step 51, the first boost mode is applied. At decision step 520, if the switching criteria are met, then switching to the second boost mode (step 530) and the programming continues (step 54A) until it is complete (step 550). If the switching criteria are not met at decision step 52, the first boost mode continues to be applied and the stylization continues (step 525). Typically, the boost mode is implemented by one or more control circuits of the configuration memory device to apply an appropriate voltage to the word line in communication with the set of storage elements. 126206.doc 1357603 The decision to switch the boost mode can be based on many factors. In general, it is necessary to implement a boost mode that is optimal for the current stylization mechanism and the current conditions of the storage component and the "reverse" string. For example, a non-EASB# voltage mode (such as SB or LSB) can be relatively efficient for initial stylized pulses (when VpGM is low), while EASB boost mode (including REASB) is for higher stylized pulses (in VPGM) High time) can be relatively effective. In this case, based on

Vpgm之振幅而作出自非EASB模式至EASB模式的切換。另The amplitude of Vpgm is switched from non-EASB mode to EASB mode. another

外,除程式化脈衝振幅之外,故障模式可對許多程式化脈 衝作出響應。在此狀況下,可基於程式化脈衝之數目(其 通常又與VpGM相關)來作出自非easb模式至easb模式的 刀換此外’某些升廢模式可較有利地基於選定之字線在 他子線中之位置。通常,取決於給定非揮發儲存裝置之 特!生’可使用產生可接受之較低故障率的多個升壓模式來 界定操作窗。 圖6描繪經由複數個字線而實施之自升壓模式。如所提 及’已開發各種類型<升壓模式以對抗程式干《。在儲存 兀件於選定之字線上之程式化期間,藉由將電壓集合施加 至與當前未程式化之儲件通信之未選m 線來實施 升壓模式。正被程式化之儲存元件與選定之「反及」串相 關聯’而其他儲存元件與未選定之「反及」串相關聯。程 式干擾通常涉及未選定之「反及」串中之儲存元件,但亦 可因在相同「反及」串中之其他儲存元件而發生。 在方法中’自升壓模式由與配置於「反及」串中之儲 存元件集合通信之實例字線描繪600。在此實例中,存在 126206.doc 1357603 標註為WLO至WL7之八個字線(例如,控制線)、標註為 SGS之源極側選擇閘極控制線及標註為sgd之没極側選擇 間極控制線。亦描綠施加至該等控制線之電壓集合。作為 ‘說明,將WL4指定為選定之字線。自「反及」串之源極側 , 錢極側’程式化通常每次前進-字線^所施加之電壓包 括:vSGS ’其施加至源極侧選擇閘極控制線sgs ;導通電 壓VPASS,其施加至未選定之字線WL〇至WL3及WL5至 φ 中之母者,程式化電壓vPGM,其施加至選定之字線 WL4 ;及VSGD,其經由汲極側選擇閘極控制線sqd來施 加。通常,VSGS為0 V,使得源極側選擇閘極關閉。 為約2.5 V,使得歸因於相應低位元線電壓Vbl(諸如,〇至i V)之細加,對於選疋之「反及」串而言,汲極側選擇閘極 開啟。歸因於相應較高VBL(諸如1.5至3 V)之施加,對於未 選疋之「反及」串而言’沒極側選擇閘極關閉。 另外,VPASS可為約7至10 v,且VpGM可在約12至2〇 v間 φ 變化。在一程式化機制中,將程式化電壓之脈衝串施加至 選定之字線。亦見圖23及圖24。該脈衝串中之每一連續程 式化脈衝之振幅以階梯方式增加,通常每一脈衝增加約 • 〇.3至0.5 V。另外,可在程式化脈衝之間施加驗證脈衝以 驗證選定之儲存元件是否已達到目標程式化條件。注意, 每一個別程式化脈衝亦可具有固定振幅’或可具有變化振 幅。舉例而§,一些程式化機制施加振幅以斜坡或階梯方 式變化之脈衝。可使用任一類型之程式化脈衝。 在WL4為程式化字線且程式化自每一「反及」串之源極 126206.doc 19 1357603 側至汲極側前進之情況下,在正程式化WL4上之儲存元件 時,已程式化與WLO至WL3相關聯之儲存元件,且將擦除 與WL5至WL7相關聯之儲存元件。未選定之字線上之導通 電壓耦合至與未選定之「反及」串相關聯之通道,使在通 道中存在一電壓,該電壓傾向於藉由降低儲存元件之穿隨 氧化物上之電壓來減少程式干擾。 圖7描繪經由複數個字線而實施之局部自升壓(LSB)模 式。在一方法中’局部自升壓模式由與配置於「反及」串 中之儲存元件集合通信之實例字線描繪700。局部自升壓 與圖6之自升壓模式的相異之處在於:相鄰於選定之字線 之字線接收0 V之隔離電壓VIS0或接近0 V之另一電壓而非 VPASS。剩餘之未選定之字線處於VpASS。局部自升壓試圖 藉由使先前程式化之儲存元件之通道與正被抑制之儲存元 件之通道隔離來減少程式干擾。雖然LSB模式對於較低值 之vPGM而言為有效的,但LSB模式之缺點在於:當VpGM較 高時’在選定之字線下方經升壓之通道的電壓可為非常 高’因為通道之彼部分與在未選定之字線下方之其他通道 區域隔離。因此’升壓電壓主要由較高程式化電壓vpgm判 定。歸因於較高升壓’在偏壓至〇 V之字線附近,可發生 帶對帶穿隧或閘極引發汲極漏電(GIDL)。可藉由使用下文 論述之擦除區域自升壓(EASB)或修正EASB(REASB)模式 將通道升壓量限制於較低值。 圖8描繪經由複數個字線而實施之擦除區域自升壓模 式。在一方法中,EASB模式由與配置於「反及」串中之 126206.doc -20- 1357603 儲存元件集合通信之實例字線描繪8〇〇。easb類似於 LSB ’不同之處在於:僅源極側鄰近字線wL3處於隔離電 【Vis〇-0 V ’使得未選定之「反及」串之源極及汲極側 上之經升壓之通道隔離。選定之字線下方之通道區域與選 疋之儲存元件之汲極側處之通道區域被連接,使得通道升 壓主要由施加至未選定之字線之VpAss替代VpGM來判定。 亦見圖13。汲極側鄰近字線WL5處於VpAss。若να”過In addition to the programmed pulse amplitude, the fault mode responds to many stylized pulses. In this case, the knife change from the non-easb mode to the easb mode can be made based on the number of stylized pulses (which are usually associated with VpGM). In addition, some of the ups and downs modes can be advantageously based on the selected word line in him. The position in the subline. Typically, the operating window can be defined by a plurality of boosting modes that result in an acceptable lower failure rate, depending on the particular non-volatile storage device. Figure 6 depicts a self boosting mode implemented via a plurality of word lines. As mentioned, 'the various types of <boost mode have been developed to combat the program." The boost mode is implemented during the staging of the storage element on the selected word line by applying a set of voltages to the unselected m lines in communication with the currently unprogrammed storage. The storage element being coded is associated with the selected "reverse" string and the other storage elements are associated with the unselected "reverse" string. Programmatic interference usually involves storage elements in unselected "reverse" strings, but may also occur due to other storage elements in the same "reverse" string. In the method, the self-boost mode is depicted 600 by an example word line that communicates with the set of storage elements disposed in the "reverse" string. In this example, there are eight word lines labeled 126206.doc 1357603 as WLO to WL7 (eg, control lines), source side select gate control lines labeled SGS, and immersed side select poles labeled sgd Control line. A set of voltages applied to the control lines is also depicted. As the 'description, specify WL4 as the selected word line. From the source side of the "reverse" string, the voltage on the side of the money side is usually 'stamped every time the voltage is applied to the word line ^ includes: vSGS ' applied to the source side select gate control line sgs; turn-on voltage VPASS, It is applied to the mother of the unselected word lines WL〇 to WL3 and WL5 to φ, the programmed voltage vPGM applied to the selected word line WL4; and VSGD, which is selected via the drain side selection gate control line sqd Apply. Typically, VSGS is 0 V, causing the source side select gate to turn off. It is about 2.5 V, so that due to the fine addition of the corresponding low bit line voltage Vbl (such as 〇 to i V), the gate selection gate is turned on for the "reverse" string of the selected one. Due to the application of a correspondingly higher VBL (such as 1.5 to 3 V), the gate is closed for the "reverse" string of the unselected 没. In addition, VPASS can be about 7 to 10 v, and VpGM can vary between about 12 and 2 〇 v. In a stylized mechanism, a burst of stylized voltage is applied to the selected word line. See also Figure 23 and Figure 24. The amplitude of each successively programmed pulse in the burst increases in a stepwise manner, typically increasing by about 〇.3 to 0.5 V per pulse. Alternatively, a verify pulse can be applied between the stylized pulses to verify that the selected storage element has reached the target stylized condition. Note that each individual stylized pulse can also have a fixed amplitude ' or can have a varying amplitude. For example, §, some stylized mechanisms apply pulses whose amplitude changes in a ramp or step manner. Any type of stylized pulse can be used. In the case where WL4 is a stylized word line and is programmed to advance from the source 126206.doc 19 1357603 side of each "reverse" string to the drain side, it is stylized when the component is being programmed on WL4. A storage element associated with WLO to WL3, and the storage elements associated with WL5 through WL7 will be erased. The turn-on voltage on the unselected word line is coupled to the channel associated with the unselected "reverse" string such that there is a voltage in the channel that tends to reduce the voltage across the oxide of the storage element. Reduce program interference. Figure 7 depicts a local self-boosting (LSB) mode implemented via a plurality of word lines. In one method, the local auto-boost mode is depicted 700 by an example word line that communicates with a set of storage elements disposed in a "reverse" string. The local auto-boost differs from the self-boost mode of Figure 6 in that the word line adjacent to the selected word line receives an isolation voltage VIS0 of 0 V or another voltage close to 0 V instead of VPASS. The remaining unselected word lines are in VpASS. Local auto-boost attempts to reduce program disturb by isolating the channels of previously stylized storage elements from the channels of the storage elements being suppressed. Although the LSB mode is effective for lower value vPGMs, the disadvantage of the LSB mode is that the voltage of the boosted channel below the selected word line can be very high when VpGM is high' because the channel is the other one. Some are isolated from other channel areas below the unselected word line. Therefore, the boost voltage is mainly determined by the higher stylized voltage vpgm. Due to the higher boost' near the word line biased to 〇 V, band-to-band tunneling or gate-induced drain leakage (GIDL) can occur. The channel boost can be limited to a lower value by using the erase region self-boost (EASB) or modified EASB (REASB) modes discussed below. Figure 8 depicts an erase region self-boost mode implemented via a plurality of word lines. In one method, the EASB mode is depicted by an example word line that communicates with the set of 126206.doc -20- 1357603 storage elements disposed in the "reverse" string. Easb is similar to LSB' in that only the source side adjacent word line wL3 is in isolation [Vis〇-0 V ' such that the undesired "reverse" string is boosted on the source and drain sides. Channel isolation. The channel area below the selected word line is connected to the channel area at the drain side of the selected memory element such that the channel boost is primarily determined by VpAss applied to the unselected word lines instead of VpGM. See also Figure 13. The drain side adjacent word line WL5 is at VpAss. If να"

低,則在通道中升壓將不足以防止程式干擾。然而,若 VPASS過高,則可程式化選定之「反及」串中的未選定之 子線(其中位元線處於〇 V),或可發生歸因於GIDL之程式 干擾。 圖9描繪經由複數個字線而實施之第一修正擦除區域自 升壓模式。在一方法中,第一 !^八犯模式由與配置於「反 及」串中之儲存元件集合通信之實例字線描繪9〇〇。 REASB類似於EASB但將較小隔離電壓Vls〇(諸如,25 v) 施加至相鄰隔離字線(例如,WL3)。 圖10描繪經由複數個字線而實施之第二修正擦除區域自 升壓模式。在一方法中,第二REASB模式由與配置於「反 及」串中之儲存元件集合通信之實例字線描繪1〇〇〇。在此 狀況下,VIS0施加至在選定之字線WL4之源極側上的多個 vIS0可以漸進方式減小,例如,自wl3上 字線,諸如,WL2及WL3。 值。舉例而言 可使用相同Viso或不同V丨s〇 之4 V減小至WL2上之2_5 V。亦可使用各種其他方法。舉 例而言’可將vIS0施加於三個相鄰字線(例如,至 126206.doc -21· 1357603 WL3)上’在此狀況下’最末字線(WL1)接收最低之vls〇, 且WL2及WL3接收共同VIS0。 圖11 a描繪經由複數個字線而實施之第三修正擦除區域 自升壓模式。在一方法中’第三REASB模式由與配置於 「反及」串中之儲存元件集合通信之實例字線描繪11〇〇。 在此狀況下,當vPGM具有相對較低之值(由Vpgm l〇w表示) 時’將相對較低之導通電壓(由vPASS_LOW表示)施加至末端 字線(例如,WL0及WL7)中之一者或兩者,而將通常、較 高之vPASS施加至其他未選定之字線。舉例而言,若 在12至20V間變化’則VPGM-L0W可表示12至16V之範圍。 此升壓模式可解決影響末端字線之程式干擾機制。具體言 之’右將具相同值之VPASS施加至所有未選定之字線(包括 末端字線),則歸因於將電子注入至與末端字線相關聯之 儲存元件中的缓慢速率’在選擇閘極上可發生漏電或 GIDL。所描繪之升壓模式可解決此問題。 另外’荽VpGM在車父南耗圍(由VpgM-HIGH表示)中時,例 如,在16至20 V之範圍中時,如圖1113中所描繪,可使 末端字線上之導通電壓升高返回至其他未選定之字線之 位準,例如,至VPASS。或,可使末端字線之導通電壓升 高至中間位準VPASS-INT,該中間位準小於vPASS但大於 VpASS-LOW 0 圖lib描繪經由複數個字線而實施之第四修正擦除區域 自升壓模式。在一方法中’第四REASB模式由與配置於 「反及」串中之儲存元件集合通信之實例字線描繪115〇。 126206.doc •22· 1357603 此處,當選定之字線WL4上之VPGM在值之較高範圍(由 VPGM-High表示)中時’使末端字線(WLO及WL7)上之導通電 壓升高返回至其他未選定之字線之位準,例如,至 Vpass ° 另外,可基於選定之字線之位置來實施不同升壓模式。 舉例而言’當在脈衝串期間發生升壓模式切換時,可在脈 衝串中之基於選定之字線之相對位置的位置處發生切換。 在一方法中’當選定之字線之位置相對較接近於未選定之 「反及」串之沒極側時,在脈衝串中相對較遲地發生自SB 或LSB至EASB或REASB之切換。 圖11 c描繪經由複數個字線而實施之第五修正擦除區域 自升壓模式。在一方法中,第五REASB模式由與配置於 「反及」串中之儲存元件集合通信之實例字線描繪117〇。 此升壓模式類似於圖11a之升壓模式,但當VpGM在較低範 圍(由Vpgm_l〇w表示)中時’對未選定之字線中之每一者使 用較低vPASS(vPASS.L0W)。當vPGM達到較高範圍(由Vpgm high 表示)時,此模式之後可為圖lib之升壓模式。亦可使用各 種其他組合。舉例而言,除末端字線外之未選定之字線的 VpASS可高於末端字線之VPASS,而與vPGM無關。另外,可 存在觸發升壓模式之改變的兩個以上之VpGM範圍。 圖12描繪展示如何藉由設定位元線抑制電壓來達成粗略 及精細程式化之時間線。如所提及,可基於粗略/精細模 式程式化狀態而發生升壓模式之切換。粗略/精細程式化 允許儲存元件之臨限電壓(VTH)首先在粗略程式化期間較 126206.doc -23- 1357603 快地且接著在精細程式化期間較慢地增加至所要位準。為 此’針對給定程式化狀態,可分別使用較低驗證位準乂及 較高驗證位準vH。特定言之,當電壓臨限值低於Vl時發生 粗略程式化,而當電壓臨限值在&與¥11之間時發生精細程 式化。粗略/精細程式化可對經程式化之儲存元件提供緊 密之電壓分布。亦見圖21d。 曲線1200指示儲存元件之Vth隨著時間之改變,而曲線 1250指示施加至與該儲存元件相關聯之位元線之位元線電 壓(vBL)。可藉由提供位元線抑制電壓vpartial ινηιβιτ來使 儲存元件之程式化減慢,此抵制所施加之程式化電壓脈衝 之效應。當vTH超過%時,將vFULL ΙΝΗΙΒ1Τ施加至位元線以 將該儲存元件置於抑制模式下,在抑制模式下,儲存元件 被鎖定以防進一步程式化及驗證。不同1及VH值可與多態 儲存元件之不同狀態(例如,狀態A、8及C)相關聯以允許 不同狀態之粗略/精細程式化。抑制電壓使程式化減慢且 藉此允許較精確地控制程式化電壓臨限位準。在一方法 中’ VPARTIAL INHIBIT(通常為0 5至1() v)減少跨越氧化物之 電場且在程式化期間傳遞至「反及」串。此情形要求選擇 閘極電壓足夠高以傳遞此電壓,通常為25 V。此外, VPGM脈衝串中減少之步長亦可用於提供精細程式化模式。 此可在位元線上具有或不具有抑制電壓之情況下實現。 因此,在一方法中,當將程式化脈衝之單個脈衝串施加 至選定之字線時,可藉由在判定某數目之儲存元件(例 如,一或多個)已達到較低驗證位準時自粗略程式化模式 126206.doc -24- Γ=細程式化模式來使用粗略/精細程式化。另外, :、:程式化機制中可使用粗略/精細程式化,其中在 π遍人中使用粗略程式化而將儲存元件程式化至接近於 臨時㈣化條件,且在第二遍次中使用 “程式化而將儲存元件自臨時程式化條件程式化至最終 气化條件夕遍次程式化亦可使用不同VPGM範圍。舉例 而δ ’ VPGM範圍可(例如)自使用粗略程式化時在第-遍次 至20 V減少至使用精細程式化時在第二遍次中之i 4 至 20 V。 圖13描繪展t^Easb(諸如’圖8中所描繪)或reasb(諸 如圖9中所拖繪)之情況下程式化區域及擦除區域的未選 疋之「反及」串之橫截面圖。該視圖為簡化的且未按比例 繪製。「反及」串13〇〇包括形成於基板139〇上之源極側選 擇閉極1306、沒極側選擇閘極1324及八個儲存元件13〇8、 1310、1312、1314、1316、1318、1320及 1322。該等組件 可形成於基板之p井區上之η井區上。除具電位vdd之位元 線1326(位元線)外,提供具電位Vs〇urce之源極供應線 1304。在程式化期間,將VpGM提供於選定之字線(在此狀 況下,WL4)上,該字線與儲存元件1316相關聯。另外, 重申儲存元件之控制閘極可作為字線之一部分而提供。舉 例而言 ’ WL0、WL1、WL2、WL3、WL4、WL5、WL6及 WL7可分別經由儲存元件1308、1310、1312、1314、 1316、1318、1320及1322之控制閘極而延伸。MVIS0施加 至選定之字線之源極侧字線(WL3,被稱為隔離字線)。將 126206.doc -25- 1357603 vPASS施加至與「反及」串1300相關聯之剩餘字線。將VsGs 施加至選擇閘極1306,且將乂⑽施加至選擇閑極1324。 假定沿著「反及」串丨则之儲存元件之程式化自儲存元 • 件1308前進至儲存元件1322,當正程式化其他「反及」串 、 t與WL4相關聯之儲存元件時,儲存元件丨则至將已 被程式化,且儲存元件1318至1322將尚未程式化。注意, 當抑制「反及」串1300時,未程式化儲存元件1316。因 ,此,取決於程式化模式,儲存元件13〇8至1314中之所有或 些將具有程式化至且儲存於其各別浮動閘極中之電子, 且可擦除或部分地程式化儲存元件1318至1322。舉例而 言,在兩步程式化技術中之第一步中可能在先前已程式化 儲存元件1318至1322。 另外,在EASB或REASB升壓模式之情況下,將足夠低 之隔離電壓VIS0施加至選定之字線之源極侧鄰近者以使基 板中之私式化及擦除通道區域隔離。亦即,在未選定之 I 「反及」串之源極侧或程式化侧上的基板之通道之一部分 (例如,區域1350)與在未選定之「反及」串之汲極側或未 程式化側上的通道之一部分(例如,區域136〇)隔離。藉由 .將VPASS施加於WL0至WL2上而將通道區域135〇升壓,而藉 由將Vpgm施加於WL4上及將vPASS施加於WL5至WL7上而將 通道區域13 60升壓。因為VPGM佔優勢,所以擦除區域1360 將經歷比程式化區域13 5 0相對較高之升壓。 圖14說明「反及」儲存元件之陣列14〇〇的實例,諸如圖 1及圖2所示之陣列。沿著每一行,位元線丨4〇6耦接至「反 126206.doc • 26 - 1357603 及」串1450之汲極選擇閘極之汲極端子1426。沿著「反 及」串之每一列,源極線14〇4可連接「反及」串之源極選 擇閘極之所有源極端子1428。在美國專利第5,57〇,315號,· 第5,774,397號;及第6,〇46,935號中會找到「反及」架構陣 列及其作為記憶體系統之部分之操作的實例。 將儲存元件陣列分成大量儲存元件區塊。如對於快閃 EEPROM系統而言為共同的,區塊為擦除之單位。亦即, 每一區塊含有被一起擦除之最小數目的儲存元件。通常將 每一區塊分成許多頁面。頁面為程式化單位。在一實施例 中,個別頁面可被分為區段且該等區段可含有隨著基本程 式化操作而一次寫入的最少數目的儲存元件。通常將一或 多個頁面之資料儲存於一列儲存元件中。頁面可儲存一或 多個扇區。扇區包括使用者資料及附加資料。附加資料通 韦匕括自扇區之使用者資料計算出的錯誤校正碼(ecC)。 控制器(下文描述)之一部分在將資料程式化至陣列中時計 算ECC,且亦在自陣列讀取資料時檢查£(:(:。或者,將 ECC及/或其他附加資料儲存於與其所屬之使用者資料不同 的頁面或甚至不同的區塊中。 一扇區之使用者資料通常為512個位元組,此對應於磁 碟機中之扇n的大附加資料通f為額外的16至2〇個位 兀組。大量頁面形成包括自8個頁面(例如)直至32、64、 或更多頁面的區塊。在一些實施例中,一列「反及」 串包含一區塊。 在實知例中,藉由使P井升高至擦除電壓(例如,20 V) 126206.doc -27· 丄妁7603 持續足夠之時間週期及在源極線及位元線為浮動時使選定 區塊之字線接地來擦除記憶體儲存元件。歸因於電容性耦 ° 未選疋之子線、位元線、選擇線及C·源極亦升高至該 , 擦除電壓之大部分。因此將強電場施加至選定之儲存元件 ' 之穿隧氧化層,且在通常藉由Fowler-Nordheim穿隧機制 將 >予動閘極之電子發射至基板側時擦除選定之儲存元件之 資料。在電子自浮動閘極傳送至ρ井區時,選定之儲存元 • 狀臨限電壓降低。可對整個記憶體陣列、獨立區塊或儲 存元件之另一單位執行擦除。 圖15為使用單列/行解碼器及讀取/寫入電路之非揮發記 憶體系統之方城圖。該圖說明根據本發明之一實施例之且 有用於並行地讀取及程式化一頁面之儲存元件的讀取/寫 入電路的記憶體裝置1596。記憶體裝置1596可包括一或多 個記憶體晶粒1 598。記憶體晶粒1 598包括儲存元件之2•維 陣列剛、控制電路151〇及讀取/寫入電路1565。在一此 瞻㈣财,該料元件陣列可為^維的。可經由列解碼器 1530由子線及經由行解碼装και, 订鮮碼器1560由位元線來對記憶 1400定址。讀取/寫 干 -電路1565包括多個感測區塊15〇〇且 . 允許並行地讀取或程式化一百品★ 八化頁面之儲存元件。通常,控制 器1550包括於與該或該等纪 控制 *' ψ 〇 Μ體日日粒1598相同之記憶體裝 置1596(例如,抽取式儲存 ρ» λ-., )中。在主機與控制器1550之 間經由線1520來傳送命令 圮恃沪曰# km 及資枓,且在控制器與該或該等 5己隱體晶粒15 9 8之間姆λ祕! c , 間左由線1518來傳送命令及資料。 控制電路1510與讀取/宜λ $ ‘‘、、電路1 565協作以對記憶體陣 126206.doc •28· 1357603 列1400執行記憶操作。控制電路1 5 10包括狀態機1 5 12、晶 載位址解碼器15 14及功率控制模組1516。狀態機15 12提供 對s己憶操作之晶片級控制β晶載位址解碼器丨5 14在由主機 或§己憶體控制器使用之位址與由解碼器1530及1560使用之 硬體位址之間提供位址介面。功率控制模組1 5 16控制在記 憶操作期間供應至字線及位元線之功率及電壓。Low, the boost in the channel will not be enough to prevent program interference. However, if VPASS is too high, the unselected sub-lines in the selected "reverse" string can be programmed (where the bit line is at 〇 V), or program disturb due to GIDL can occur. Figure 9 depicts a first modified erase region self-boost mode implemented via a plurality of word lines. In one method, first! The eight-figure mode is depicted by an example word line that communicates with a set of storage elements disposed in the "reverse" string. The REASB is similar to the EASB but applies a smaller isolation voltage Vls 〇 (such as 25 v) to the adjacent isolated word line (eg, WL3). Figure 10 depicts a second modified erase region self-boost mode implemented via a plurality of word lines. In one method, the second REASB mode is depicted by an example word line in communication with the set of storage elements disposed in the "reverse" string. In this case, the plurality of VS0s applied by VIS0 to the source side of the selected word line WL4 may be progressively reduced, e.g., from word lines on wl3, such as WL2 and WL3. value. For example, 4 V of the same Viso or different V丨s〇 can be used to reduce to 2_5 V on WL2. Various other methods are also available. For example, 'vIS0 can be applied to three adjacent word lines (eg, to 126206.doc -21·1357603 WL3). 'In this case' the last word line (WL1) receives the lowest vls〇, and WL2 And WL3 receives the common VIS0. Figure 11a depicts a third modified erase region self-boost mode implemented via a plurality of word lines. In one method, the 'third REASB mode is depicted by an example word line that communicates with the set of storage elements disposed in the "reverse" string. In this case, when vPGM has a relatively low value (represented by Vpgm l〇w), a relatively low on-voltage (represented by vPASS_LOW) is applied to one of the end word lines (eg, WL0 and WL7). Or both, and the usual, higher vPASS is applied to other unselected word lines. For example, if it varies between 12 and 20V' then VPGM-L0W can represent a range of 12 to 16V. This boost mode resolves the program interference mechanism that affects the end word line. Specifically, 'the right VPASS with the same value is applied to all unselected word lines (including the end word lines), due to the slow rate of injecting electrons into the storage elements associated with the end word lines' Leakage or GIDL can occur on the gate. The boost mode depicted can solve this problem. In addition, '荽VpGM is in the range of the car's south (expressed by VpgM-HIGH), for example, in the range of 16 to 20 V, as depicted in Figure 1113, the turn-on voltage on the end word line can be raised back. To the level of other unselected word lines, for example, to VPASS. Alternatively, the turn-on voltage of the end word line can be raised to the intermediate level VPASS-INT, which is less than vPASS but greater than VpASS-LOW 0. Figure lib depicts the fourth modified erase region implemented via a plurality of word lines. Boost mode. In one method, the fourth REASB mode is depicted by an example word line that communicates with the set of storage elements disposed in the "reverse" string. 126206.doc •22· 1357603 Here, when the VPGM on the selected word line WL4 is in the higher range of values (represented by VPGM-High), the on-voltage of the end word lines (WLO and WL7) is raised back. To other unselected word lines, for example, to Vpass ° In addition, different boost modes can be implemented based on the position of the selected word line. For example, when a boost mode switch occurs during a burst, switching can occur at a location in the burst based on the relative position of the selected word line. In a method, the switching from SB or LSB to EASB or REASB occurs relatively late in the burst when the selected word line is relatively close to the unpolarized side of the unselected "reverse" string. Figure 11c depicts a fifth modified erase region self-boost mode implemented via a plurality of word lines. In one method, the fifth REASB mode is depicted by an example word line that communicates with the set of storage elements disposed in the "reverse" string. This boost mode is similar to the boost mode of Figure 11a, but uses lower vPASS (vPASS.L0W) for each of the unselected word lines when VpGM is in the lower range (represented by Vpgm_l〇w) . When vPGM reaches a higher range (represented by Vpgm high), this mode can be followed by the boost mode of Figure lib. Various other combinations are also possible. For example, the VpASS of unselected word lines other than the end word line can be higher than the VPASS of the end word line, regardless of vPGM. Additionally, there may be more than two VpGM ranges that trigger a change in boost mode. Figure 12 depicts a timeline showing how rough and fine stylization can be achieved by setting the bit line suppression voltage. As mentioned, the switching of the boost mode can occur based on the coarse/fine mode stylized state. Rough/Fine Stylization The threshold voltage (VTH) of the allowed storage element is first increased to a desired level faster during the coarse stylization than 126206.doc -23- 1357603 and then slowly during fine stylization. For this given a lower stylized state and a higher verify level vH, respectively, for a given stylized state. In particular, coarse stylization occurs when the voltage threshold is lower than V1, and fine-grained when the voltage threshold is between & and ¥11. Rough/fine stylization provides a tight voltage distribution to the stylized storage elements. See also Figure 21d. Curve 1200 indicates the change in Vth of the storage element over time, while curve 1250 indicates the bit line voltage (vBL) applied to the bit line associated with the storage element. The stylization of the storage element can be slowed down by providing a bit line suppression voltage vpartial ινηιβιτ, which counteracts the effects of the applied stylized voltage pulses. When vTH exceeds %, vFULL ΙΝΗΙΒ1Τ is applied to the bit line to place the storage element in the inhibit mode, in which the storage element is locked from further stylization and verification. Different 1 and VH values can be associated with different states of the polymorphic storage element (e. g., states A, 8 and C) to allow for coarse/fine stylization of different states. Suppressing the voltage slows down the stylization and thereby allows for more precise control of the stylized voltage threshold level. In one method, 'VPARTIAL INHIBIT (usually 0 5 to 1 () v) reduces the electric field across the oxide and passes it to the "reverse" string during stylization. This situation requires that the gate voltage be chosen high enough to pass this voltage, typically 25 V. In addition, the reduced step size in the VPGM burst can also be used to provide a fine stylized mode. This can be done with or without a suppression voltage on the bit line. Thus, in one method, when a single burst of stylized pulses is applied to a selected word line, it can be determined by determining that a certain number of storage elements (eg, one or more) have reached a lower verify level. Rough stylized mode 126206.doc -24- Γ=fine stylized mode to use coarse/fine stylization. In addition, the :,: stylization mechanism can use coarse/fine stylization, in which the staging is used in π-pass people to program the storage element to be close to the temporary (four) condition, and use in the second pass. Stylized to stylize storage elements from temporary stylization conditions to final gasification conditions. Different VPGM ranges can also be used for stylized programming. For example, δ ' VPGM range can be (for example) used in the first pass since the rough stylization The next time to 20 V is reduced to i 4 to 20 V in the second pass when fine-stylization is used. Figure 13 depicts a t^Easb (such as depicted in Figure 8) or reasb (such as the one depicted in Figure 9) In the case of a stylized area and an erased area, the unrestricted cross-section of the "reverse" string. This view is simplified and not drawn to scale. The "reverse" string 13 includes a source side selective closed pole 1306 formed on the substrate 139, a gateless select gate 1324, and eight storage elements 13A, 1310, 1312, 1314, 1316, 1318, 1320 and 1322. The components can be formed on the n well region on the p-well region of the substrate. In addition to the bit line 1326 (bit line) having the potential vdd, a source supply line 1304 having a potential Vs〇urce is provided. During stylization, VpGM is provided on the selected word line (in this case, WL4), which is associated with storage element 1316. In addition, it is reiterated that the control gate of the storage element can be provided as part of the word line. For example, 'WL0, WL1, WL2, WL3, WL4, WL5, WL6, and WL7 may extend through the control gates of storage elements 1308, 1310, 1312, 1314, 1316, 1318, 1320, and 1322, respectively. MVIS0 is applied to the source side word line (WL3, referred to as the isolated word line) of the selected word line. 126206.doc -25 - 1357603 vPASS is applied to the remaining word lines associated with the "reverse" string 1300. VsGs is applied to select gate 1306 and germanium (10) is applied to select idle 1324. It is assumed that the stylized storage element of the storage element along the "reverse" string is advanced from the storage element 1308 to the storage element 1322, and is stored when other "reverse" strings, t and WL4 associated storage elements are being programmed. The components will then be programmed, and the storage elements 1318 through 1322 will not be programmed. Note that when the "reverse" string 1300 is suppressed, the storage element 1316 is not programmed. Thus, depending on the stylized mode, all or some of the storage elements 13〇8 to 1314 will have electrons that are programmed into and stored in their respective floating gates, and can be erased or partially programmed to be stored. Elements 1318 to 1322. For example, storage elements 1318 through 1322 may have been previously programmed in the first step of the two-step stylization technique. Additionally, in the case of the EASB or REASB boost mode, a sufficiently low isolation voltage VIS0 is applied to the source side neighbors of the selected word line to isolate the private and erase channel regions in the substrate. That is, a portion of the channel (eg, region 1350) of the substrate on the source side or the stylized side of the unselected I "reverse" string and the drain side of the unselected "reverse" string or not One of the channels on the stylized side (eg, area 136〇) is isolated. The channel region 135 is boosted by applying VPASS to WL0 to WL2, and the channel region 13 60 is boosted by applying Vpgm to WL4 and applying vPASS to WL5 to WL7. Because VPGM is dominant, erase region 1360 will experience a relatively higher boost than stylized region 135. Figure 14 illustrates an example of an array of "reverse" storage elements, such as the arrays shown in Figures 1 and 2. Along each row, the bit line 丨4〇6 is coupled to the “reverse 126206.doc • 26 - 1357603 and the string 1450 of the drain select terminal of the drain terminal 1426. Along the respective columns of the "reverse" string, the source line 14〇4 can be connected to all source terminals 1428 of the source selection gate of the "reverse" string. Examples of "reverse" architecture arrays and their operation as part of a memory system are found in U.S. Patent Nos. 5,57,315, 5,774,397, and 6,6,935. The array of storage elements is divided into a plurality of storage element blocks. As is common to flash EEPROM systems, the block is the unit of erasure. That is, each block contains a minimum number of storage elements that are erased together. Each block is usually divided into a number of pages. The page is a stylized unit. In an embodiment, individual pages may be divided into segments and the segments may contain a minimum number of storage elements that are written once with the basic programming operation. The data of one or more pages is typically stored in a list of storage elements. The page can store one or more sectors. The sector includes user data and additional information. The additional data is an error correction code (ecC) calculated from the user data of the sector. One part of the controller (described below) calculates the ECC when the data is programmed into the array, and also checks £(:(:. or, if ECC and/or other additional data is stored in it) when reading the data from the array. The user data is in different pages or even different blocks. The user data of one sector is usually 512 bytes, which corresponds to the large additional data of the fan n in the disk drive. Up to 2 locations. A large number of pages form blocks from 8 pages (for example) up to 32, 64, or more pages. In some embodiments, a column of "reverse" strings contains a block. In the practical example, by raising the P well to the erase voltage (for example, 20 V) 126206.doc -27· 丄妁 7603 for a sufficient period of time and when the source line and the bit line are floating, the selection is made. The word line of the block is grounded to erase the memory storage element. Due to the capacitive coupling, the unselected sub-line, bit line, select line, and C· source are also raised to this, and most of the erase voltage Therefore, a strong electric field is applied to the tunneling oxide layer of the selected storage element, and The data of the selected storage element is usually erased by the Fowler-Nordheim tunneling mechanism when the electrons of the pre-grating gate are emitted to the substrate side. When the electron self-floating gate is transmitted to the ρ well area, the selected storage element is selected. The threshold voltage is reduced. The erase can be performed on the entire memory array, the independent block, or another unit of the storage element. Figure 15 is a non-volatile memory system using a single column/row decoder and a read/write circuit. The present invention illustrates a memory device 1596 having a read/write circuit for reading and programming a page of storage elements in parallel, in accordance with an embodiment of the present invention. The memory device 1596 can include a Or a plurality of memory crystal grains 1 598. The memory crystal grains 1 598 include a 2D dimensional array of storage elements, a control circuit 151A, and a read/write circuit 1565. In this case, the material array is The memory 1400 can be addressed by a bit line from the sub-line and via the row decoding device και. The read/write dry-circuit 1565 includes a plurality of sensing regions. Block 15 and. Allow reading in parallel Or stylized a storage element of a hundred products. Typically, the controller 1550 is included in the same memory device 1596 as the one or the other control *' 〇Μ 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日ρ» λ-., ). Between the host and the controller 1550, the command 15K and the resource are transmitted via the line 1520, and the controller and the 5 or more hidden crystals 15 9 8, λ 密 secret! c, left to send commands and data by line 1518. Control circuit 1510 and read / λ λ $ '', circuit 1 565 cooperate to memory array 126206.doc • 28 · 1357603 Column 1400 performs a memory operation. The control circuit 1 5 10 includes a state machine 1 52, a crystal address decoder 15 14 and a power control module 1516. The state machine 15 12 provides a wafer level control for the memory operation of the β crystal carrier address decoder 丨 5 14 at the address used by the host or § memory controller and the hardware address used by the decoders 1530 and 1560. Provide a bit interface between them. Power control module 1 5 16 controls the power and voltage supplied to the word lines and bit lines during the memory operation.

在某些實施例中,可組合圖丨5之組件中的某些。在各種 设計中’可將除儲存元件陣列丨4〇〇外的組件中之一或多者 (單獨或組合)視為管理電路。舉例而言,一或多個管理電 路可包括控制電路1510、狀態機1512、解碼器 1514/1560、功率控制1516、感測區塊15〇〇、讀取/寫入電 路1565、控制器1550等中之任一者或組合。In some embodiments, some of the components of Figure 5 can be combined. One or more of the components (individually or in combination) other than the array of storage elements can be considered as management circuits in various designs. For example, one or more management circuits may include control circuit 1510, state machine 1512, decoder 1514/1560, power control 1516, sensing block 15A, read/write circuit 1565, controller 1550, etc. Any one or combination.

圖16為使用雙列/行解碼器及讀取/寫入電路之非揮發記 憶體系統之方塊圖。此處,提供圖丨5中所示之記憶體裝置 1596之另一配置。以對稱方式在陣列之相對侧上實施由各 種周邊電路對記憶體陣列;U00之存取,以使得每一側上之 存取線及電路的密度減少—半。因此,列解碼器分成列解 碼器1530A及1530B,且行解碼器分成行解碼器15嶋及 B60B。類似地,讀取/寫入電路分成自陣列14〇〇之底部連 接至位元線的讀取/寫入電路1565八及自陣列14〇〇之頂部連 接至位元線的讀取/寫入電路1565Ββ以此方式,讀取/寫 入模組之密度基本上減少—半。如針對圖15之裝置在上: 所描述,圖16中之裝置亦可包括一控制器。 圖1 7為描緣感測區塊之一奢a 實施例之方塊圖。將個別感測 126206.doc •29· 1357603Figure 16 is a block diagram of a non-volatile memory system using a dual column/row decoder and a read/write circuit. Here, another configuration of the memory device 1596 shown in FIG. 5 is provided. Access to the memory array; U00 by various peripheral circuits is performed symmetrically on opposite sides of the array such that the density of access lines and circuitry on each side is reduced by half. Therefore, the column decoder is divided into column decoders 1530A and 1530B, and the row decoder is divided into row decoders 15A and B60B. Similarly, the read/write circuit is divided into a read/write circuit 1565 connected to the bit line from the bottom of the array 14A and a read/write connected to the bit line from the top of the array 14A. In this manner, the circuit 1565 Β β has substantially reduced the density of the read/write module—half. As described above with respect to the apparatus of Fig. 15, the apparatus of Fig. 16 may also include a controller. Figure 17 is a block diagram of an embodiment of a marginal sensing block. Will be individually sensed 126206.doc •29· 1357603

區塊1500分割成一核心部分(被稱為感測模組I〗8〇)及一共 同部分1590。在一實施例中,針對每一位元線將存在一獨 立感測模組1580,且針對多個感測模組1580之一集合將具 有一共同部分1590。在一實例中,感測區塊將包括一個共 同部分1 590及八個感測模組1580 « —群組中之感測模組中 之每一者將經由資料匯流排1572而與相關聯之共同部分通 仏。關於更多細節請參考於2〇〇6年6月29曰公布且以引用 之方式全文併入本文中的標題為” N〇n_v〇Utile Mem〇ry &The block 1500 is divided into a core portion (referred to as a sensing module I) and a common portion 1590. In one embodiment, a separate sensing module 1580 will be present for each bit line, and a common portion 1590 will be provided for one of the plurality of sensing modules 1580. In one example, the sensing block will include a common portion 1 590 and eight sensing modules 1580 « - each of the sensing modules in the group will be associated via data bus 1572 The common part is wanted. For more details, please refer to the title of "N〇n_v〇Utile Mem〇ry & published in the text of June 29, 2002 and incorporated by reference in its entirety.

Method with Shared Processing for an Aggregate of Sense Amphflers"之美國專利申請公開案第2〇〇6/〇i4〇〇〇7號。 一感測模組1580包含判定所連接之位元線中之傳導電流是 间於還疋低於預定臨限位準的感測電路丨57〇。感測模組 1則亦包括用於對所連接之位元線設定電壓條件的位元線 鎖存器1582。舉例而言,位元線鎖存器1582中所鎖存之預Method with Shared Processing for an Aggregate of Sense Amphflers " US Patent Application Publication No. 2/6/〇i4〇〇〇7. A sensing module 1580 includes a sensing circuit 判定 57 判定 that determines that the conduction current in the connected bit line is still below a predetermined threshold level. The sensing module 1 also includes a bit line latch 1582 for setting a voltage condition to the connected bit line. For example, the pre-locked bit in the bit line latch 1582

疋狀L將導致所連接之位元線被拉向指定程式化抑制之狀 態(例如,Vdd)。 共同部分1590包含—處理器1592、— f料鎖存器集 1594及—純於該資料鎖存器集合⑽與資料匯流排15 之間的!/〇介面1596。處理器1592執行計算,而言 其功能中之—者為料儲存於所感測之儲存元件中之資 且將所判定之資料儲存於 存器集合1594用於儲存在鎖存器集合中。該資料: _ 在°賣取操作期間由處理器1592判: 之資料位元。其亦用於妙+丄 次用於儲存在程式化操作期間自資料匯; 入之貝料位元。所輸入之資料位元表示意欲程』 126206.doc -30* 1357603 化至記憶體中之寫入資料。I/0介面1596在資料鎖存器 1594與資料匯流排152〇之間提供介面。 在讀取或感測期間,系統之操作受狀態機1512控制,狀 態機15 12控制不同控制閘極電壓對所定址之儲存元件的供 應。在其單步遍曆(step thr〇ugh)對應於由記憶體支援之各 種記憶體狀態之各種預定控制閘極電壓時,感測模組丨58〇 可在此等電壓中之一者處跳脫,且輸出將經由匯流排丨572 而自感測模組1580提供至處理器1592。此時,處理器1592 藉由考慮感測模組之跳脫事件及經由輸入線丨593來自狀態 機之關於所施加之控制閘極電壓的資訊來判定所得記憶體 狀I、接著其關於5己憶體狀癌而計算二進位編碼且將所得 資料位元儲存至資料鎖存器1 594中。在核心部分之另一實 施例中,位元線鎖存器1582用於雙重用途,既用作用於鎖 存感測模組15 8 0之輸出的鎖存器且亦用作如上所述之位元 線鎖存器。 預期一些實施例將包括多個處理器1592。在—實施例 中,每一處理器1592將包括一輸出線(未描繪於圖7中),使 得輸出線中之每一者共同被線或連接。在一些實施例中, 在將輸出線連接至線或線之前將輸出線反相。此組態允許 在程式化驗證過程期間快速地判定完成程式化過程之時 刻’因為接收線或之狀態機可判定正被程式化之所有位元 達到所要位準之時刻》舉例而言,當每一位元已達到其所 要位準時,該位元之邏輯零將被發送至線或線(或資料一 被反相)。當所有位元輸出資料〇(或經反相之資料一)時, 126206.doc •31 · 1357603 既而狀態機知曉終止程式化過程。因為每—處理器與八個 感測模組通信’所以狀態機需要讀取線或線人次或邏輯 被添加至處理器1592以累積相關聯之位元線的結果,使得 狀態機僅需要讀取線或線—次。類似地,藉由正確地選擇 邏輯位準,整體狀態機可偵測第一位元改變其狀態之時刻 且因此改變演算法。 在程式化或驗證期間,將來自資料匯流排152〇的待程式The shape L will cause the connected bit line to be pulled toward the specified stylized suppression state (e.g., Vdd). The common portion 1590 includes a processor 1592, a f-fed set 1594, and a !/〇 interface 1596 between the data latch set (10) and the data bus 15 . The processor 1592 performs computations in which the functions are stored in the sensed storage elements and the determined data is stored in the set of registers 1594 for storage in the set of latches. The data: _ is determined by the processor 1592 during the ° fetch operation: the data bit. It is also used to store the data from the data stream during the stylized operation; The input data bit indicates the intended process 126206.doc -30* 1357603 The data is written into the memory. The I/O interface 1596 provides an interface between the data latch 1594 and the data bus 152A. During reading or sensing, the operation of the system is controlled by state machine 1512, which controls the supply of different control gate voltages to the addressed storage elements. The sensing module 丨58〇 can jump at one of the voltages when it is stepped through various predetermined control gate voltages corresponding to various memory states supported by the memory. Off, and the output will be provided from the sensing module 1580 to the processor 1592 via the bus bar 572. At this time, the processor 1592 determines the obtained memory state I by considering the tripping event of the sensing module and the information about the applied control gate voltage from the state machine via the input line 丨593, and then about the 5 The binary code is calculated by recalling the body cancer and the resulting data bits are stored in the data latch 1 594. In another embodiment of the core portion, bit line latch 1582 is used for dual purposes, both as a latch for latching the output of sense module 580 and also as described above. Meta line latch. It is contemplated that some embodiments will include multiple processors 1592. In an embodiment, each processor 1592 will include an output line (not depicted in Figure 7) such that each of the output lines are commonly lined or connected. In some embodiments, the output line is inverted before the output line is connected to the line or line. This configuration allows for a quick determination of the time at which the stylization process is completed during the stylized verification process 'because the receive line or state machine can determine when all bits being programmed have reached the desired level," for example, each When a bit has reached its desired level, the logical zero of that bit will be sent to the line or line (or the data is inverted). When all bits output data (or reversed data one), 126206.doc •31 · 1357603 and the state machine knows to terminate the stylization process. Because each processor communicates with eight sensing modules' so the state machine needs to read lines or line people or logic is added to processor 1592 to accumulate the results of the associated bit lines, so that the state machine only needs to read Line or line - times. Similarly, by correctly selecting the logic level, the overall state machine can detect when the first bit changes its state and thus change the algorithm. During the stylization or verification, the program will be sent from the data bus 152

化之資料儲存於該資料鎖存器集^別中。在狀態機之控 制下’程式化操作包含施加至所定址之儲存元件之控制閑 極的連串程式化電祕衝。每-程式化脈衝之後為回讀 (驗證。)以判定儲存元件是否已程式化至所要記憶體狀態。 處理器1592相對於所要記憶體狀態而監視回讀之記憶體狀 態。當該兩個狀態一致時,處理器1592設定位元線鎖存器 以使位元線被拉向指定程式化抑制之狀態。此情形抑 制輕接至位元線之儲存元件免受進—步程式化,即使程式The data is stored in the data latch set. Under the control of the state machine, the stylized operation includes a series of stylized electrical secrets applied to the control idles of the addressed storage elements. Each stylized pulse is read back (verified.) to determine if the storage element has been programmed to the desired memory state. The processor 1592 monitors the memory state of the readback with respect to the desired memory state. When the two states coincide, processor 1592 sets the bit line latch to cause the bit line to be pulled toward the specified stylized suppression state. This situation suppresses storage elements that are lightly connected to the bit line from being step-by-step programmed, even if the program

化脈衝出現於其控制閘極上亦如此。在其他實施例中,處 理器最初載人位元線鎖存器1582,且感測電路在驗證過程 期間將其設定至抑制值。 資料鎖存器堆疊1594含有對應於感測模組之資料鎖存器 之堆疊。在一實施例中,每一感測模組158〇存在三個資料 鎖存器。在-些實施例中(但並非必需),將資料鎖存=實 施為移位暫存器’使得儲存於其中之並行資料被轉換至用 於資料匯流排1520之串行資料’且反之亦然。在較佳實施 例中,可將對應於m個儲存元件之讀取/寫入區塊之所有資 126206.doc •32- 1357603 料鎖存器鏈接在一起以形成區塊移位暫存器,使得可藉由 串行傳送來輸入或輸出資料之區塊。詳言之,調適該組I 個讀取/寫入模組,以使得其資料鎖存器之集合中之每一 者將資料順序地移進或移出資料匯流排,如同其係用於整 個讀取/寫入區塊之移位暫存器之部分一樣。 可在以下各者中找到關於非揮發儲存裝置之各種實施例 之結構及/或操作的額外資訊:(1)於2004年3月25日公布之 美國專利申請公開案第2〇〇4/〇〇57287號,nN〇n V〇iatile Memory And Method With Reduced Source Line Bias Errors,(2)於2004年6月10曰公布之美國專利申請公開案 第 2004/0109357 號 ’ "Non-Volatile Memory And Method with Improved Sensing" ; (3)於2004年 12月 16 日申請之美國 專利申請案第11/015,199號,標題為”Improved Memory Sensing Circuit And Method For Low Voltage Operation"; (4)於2005年4月5曰申請之美國專利申請案第i ^99^33 號’標題為”Compensating for Coupling During Read Operations of Non-Volatile Memory” ;及(5)於 2005 年 12 月 28曰申請之美國專利申請案第11/321,953號,標題為 Reference Sense Amplifier For Non-Volatile Memory"。剛 在上文列出之所有五個專利文獻以引用之方式全文併入本 文中。 圖18說明針對全位元線記憶體架構或針對奇偶記憶體架 構而將記憶體陣列組織成區塊的實例。描述儲存元件陣列 1400之例示性結構。作為一實例,描述分割成ι,〇24個區 126206.doc -33· 1357603 塊之「反及」快閃EEPR0M。可同時擦除儲存於每一區塊 中之資料。在-實施例中,區塊為同時被擦除之儲存元件 之最小單位。在每-區塊中,在此實例中,存在對應於位 •①線BL0、犯······BL85U之8,512個行。在被稱為全位元 、 線(ABL)架構(架構181G)之-實施例中,在讀取及程式化 操作期間可同時選擇一區塊之所有位元線。可同時程式化 沿著共同字線且連接至任一位元線的儲存元件。 φ 「在所提供之實例十,四個儲存元件_聯地連接以形成 「反及」串。雖然展示四個儲存元件被包括於每一「反 及」串中,但可使用多於或少於四個(例如,i6、U、Μ 或另數目)。該「反及」率之—端子經由没極選擇問極 (連接至選擇閘極汲極線SGD)而連接至相應位元線,且另 一端子經由源極選擇閘極(連接至選擇閘極源極線SGS)連 接至c-源極。 在被稱為奇偶架構(架構18〇〇)之另一實施例中,將位元 藝線分成偶數位元線(BLe)及奇數位元線(BL〇)。在奇/偶位元 線架構中,在一時間程式化沿著共同字線且連接至奇數位 元線之儲存元件’而在另一時間程式化沿著共同字線且連 • 接至偶數位元線之儲存元件。可同時將資料程式化至不同 . 品鬼中自不同區塊讀取資料。在此實例中,在每一區塊 中存在8,512個行,其被分成偶數行及奇數行。在此實例 中展不四個儲存元件被串聯連接以形成「反及」串。雖 然展不四個儲存元件被包括於每-「反及」_中,但可使 用多於或少於四個之儲存元件。 126206.doc •34· 1357603 在貝取及程式化操作之一組態期間,同時選 儲存元件。所選之儲存元件且有鬥—姑 b個 仔騎具有同一帛線及同一種類之位 7L線(例如,偶數或奇數函此 同時讀取或程式化形 成邏輯頁面的532個位元組之資料,且記憶體之—區機可 儲存至少八個邏輯頁面(四個字線,每一者具有奇數頁面 及偶數頁面)。對於多態儲存元件而言,當每-儲存元件 儲存兩個位元之資料時(其中,個位元中之The same applies to the presence of a pulse on its control gate. In other embodiments, the processor initially carries the bit line latch 1582 and the sensing circuit sets it to the suppression value during the verification process. The data latch stack 1594 contains a stack of data latches corresponding to the sense modules. In one embodiment, there are three data latches per sensing module 158. In some embodiments (but not necessarily), the data latch = implemented as a shift register 'so that parallel data stored therein is converted to serial data for data bus 1520' and vice versa . In a preferred embodiment, all of the 126206.doc • 32-1357603 material latches corresponding to the read/write blocks of the m storage elements can be linked together to form a block shift register. This allows the block of data to be input or output by serial transfer. In particular, the set of I read/write modules is adapted such that each of its set of data latches sequentially shifts data into or out of the data bus as if it were used for the entire read. The part of the shift register of the fetch/write block is the same. Additional information regarding the structure and/or operation of various embodiments of the non-volatile storage device can be found in the following: (1) U.S. Patent Application Publication No. 2/4/ filed on March 25, 2004. U.S. Patent No. 57,287, NN-Volatile Memory And Method; (3) U.S. Patent Application Serial No. 11/015,199, filed on Dec. 16, 2004, entitled "Improved Memory Sensing Circuit And Method For Low Voltage Operation"; (4) in 2005 4 U.S. Patent Application Serial No. i ^99^33, entitled "Compensating for Coupling During Read Operations of Non-Volatile Memory"; and (5) U.S. Patent Application filed on December 28, 2005 No. 11/321,953, entitled "Reference Sense Amplifier For Non-Volatile Memory". All of the five patent documents listed above are hereby incorporated by reference in their entirety. 18 illustrates an example of organizing a memory array into blocks for a full bit line memory architecture or for a parity memory architecture. An exemplary structure of the storage element array 1400 is described. As an example, the description is divided into ι, 〇 24 District 126206.doc -33· 1357603 Block "Reverse" flash EEPR0M. The data stored in each block can be erased at the same time. In an embodiment, the block is the smallest unit of storage elements that are simultaneously erased. In each block, in this example, there are 8,512 rows corresponding to the bit •1 line BL0, the sin...·BL85U. In an embodiment referred to as a full bit, line (ABL) architecture (architecture 181G), all bit lines of a block can be selected simultaneously during read and program operations. Storage elements along a common word line and connected to any bit line can be programmed simultaneously. φ "In the example provided, four storage elements_ are connected in series to form a "reverse" string. Although four storage elements are shown to be included in each "reverse" string, more or less than four (e.g., i6, U, 或 or another number) may be used. The "reverse" rate - the terminal is connected to the corresponding bit line via the immersive selection pole (connected to the selected gate drain line SGD), and the other terminal is connected to the selection gate via the source selection gate The source line SGS) is connected to the c-source. In another embodiment, referred to as an odd-even architecture (architecture 18), the bit art lines are divided into even bit lines (BLe) and odd bit lines (BL〇). In the odd/even bit line architecture, the storage elements along the common word line and connected to the odd bit lines are programmed at one time and stylized along the common word line and connected to the even bits at another time. The storage component of the meta line. The data can be stylized at the same time. The product is read from different blocks. In this example, there are 8,512 rows in each block, which are divided into even rows and odd rows. In this example, four storage elements are connected in series to form a "reverse" string. Although not four storage elements are included in each "reverse", more or less than four storage elements can be used. 126206.doc •34· 1357603 During the configuration of one of the fetching and stylizing operations, the component is also selected. The selected storage component has a bucket - a b-bit with the same line and the same type of 7L line (for example, even or odd letters read or programmatically form a logical page of 532 bytes) And the memory-area machine can store at least eight logical pages (four word lines, each with odd and even pages). For polymorphic storage elements, when each storage element stores two bits Information (in which one of the bits

儲存於不同1面中),—區塊儲存十六個邏輯頁面。亦可 使用具有其他大小之區塊及頁面。 對於ABL架構或奇偶架構而言,可藉由使p井升高 除電壓(例如,20 V)且使選定之區塊之字線錢來擦除儲 存70件。源極線及位元線係浮動的。可對整個記憶體陣 列、獨立區塊或為記憶體裝置之—部分的儲存元件之另一 單位執行擦除。電子自儲存元件之浮動閘極傳送至P井 區,使得儲存元件之VTH變為負的。 在讀取及驗證操作中,選擇閘極(SGD及SGS)連接至在 2·5 V至4·5 v之範圍中的電壓’且未選定之字線(例如,當 WL2為選定之字線時為·、和及机3)升高至讀取導: 電壓Vread(通常為在4.5 乂至6 V之範圍中之電壓)以使電晶 體作為導通閘極而操作。選定之字線乳2連接至—電壓, 針對每-讀取及驗證操作而指^該㈣之料以便判定相 關儲存元件之VTH是高於還是低於此位準。舉例而言,在 雙階儲存it件之讀取操作中,選定之字線WL2可 的’使得谓測VTH是否高於〇 V。舉例而言,在雙階儲存元 126206.doc -35- 1357603 件之驗證操作中,選定之字線WL2連接至〇 8 v,使得驗證 VTH是否已達到至少〇.8 V。源極及p井為〇 選定之位I 線(假定為偶數位元線(BLe))預先充電至(例如)〇 7乂之2 準。若VtH高於字線上之讀取或驗證位準,則與有關儲存 元件相關聯之位元線(BLe)的電位位準由於非傳導健存元 件而維持向位準。另一方;,甚v . 为方面右Vth低於讀取或驗證位 準,則相關位元線(BLe)之電位位準減小至低位準(例如, 小於〇.5 V),因為傳導料元件使位元線放電。儲存元件 之狀態可藉此由連接至位元線之電壓比較器感 測。 根據此項技術中已知之技術來執行上述擦除、讀取及驗 :插作目此,所解釋之細節中之許多可由熟習此項技術 亦可使用此項技術中已知之其他擦除、讀取及驗 證技術。 圖19描續·臨限電壓分右 电堅刀布之實例集合。針對每一儲存元件 子八兩個位元之資料之狀況而提供儲存元件陣列之實例 :布。對_除之儲存元件提供第—臨限職分布E。 ρ . 儲存兀件之三個臨限電壓分布A、B及 L °在一實施例中, ^ . b刀布中之臨限電壓為負且Λ、B及(:分 布中之臨限電壓為正。 值: ^限電壓圍對應於用於資料位元集合之預定 準之/ 1储存疋件中之資料與儲存元件之臨限電壓位 制。兴特疋關係取決於對儲存元件採用之資料編碼機 舉例而言,均以引用之方式全文併入本文中的美國專 126206.doc -36 ^ 1357603 利第6,222,762號及於2004年I2月16曰公布之美國專利申靖 公開案第2004/0255090號描述用於多態快閃儲存元件之各 種資料編碼機制。在一實施例中,使用格雷(gray)碼指派 • 來將資料值指派給臨限電壓範圍,使得在浮動閘極之臨限 、 電壓錯誤地移位至其鄰近實體狀態時,將僅影響一位元。 一貝例將"11 ”指派給臨限電壓範圍E(狀態E),將"1 〇"指派 給臨限電壓範圍A(狀態A),將”〇〇"指派給臨限電壓範圍 鲁 B(狀態B)且將"01"指派給臨限電壓範圍C(狀態c)。然而, 在其他實施例中,不使用格雷碼。雖然展示了四個狀態, 但本發明亦可用於其他多態結構,包括會包括多於或少於 四個狀態之多態結構。 亦提供三個讀取參考電壓Vra、Vrb及Vrc以用於自儲存 兀件讀取資料。藉由測試給定儲存元件之臨限電壓是高於 還是低於Vra、Vrb及Vrc,系統可判定儲存元件所處之狀 態’例如’程式化條件。 • 另外,提供三個驗證參考電壓Vva、Vvb及VVC。當將儲 存70件程式化至狀態A時,系統將測試此等儲存元件是否 具有大於或等於Vva之臨限電塵。當將儲存元件程式化至 • 狀態B時,系統將測試儲存元件是否具有大於或等於μ • = &限電M °當將儲存元件程式化至狀態C時,系統將判 定儲存元件是否具有其大於或等於Vvc之臨限電壓。 在被稱為全序列程式化之―實施射,可㈣存元件自 擦除狀態』直接冑式化至程式化狀態之任一 者舉例而。,γ首先擦除待程式化之儲存元件之群體, I26206.doc -37- 1357603 使得該群體中之所有儲存元件處於擦除狀態E。接著將使 用諸如由圖23之控制閘極電壓序列描繪之一連串程式化脈 衝來將儲存元件直接程式化至狀態A、B或C ^雖然一些儲 • 存元件係自狀態E程式化至狀態a,但其他儲存元件係自 , 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(^時,對wLnj下方之相鄰浮 動閘極之寄生耦合的量為最大的,因為與自狀態E程式化 至狀態A或自狀態E程式化至狀態B時之電壓之改變相比, 在WLn下方之浮動閘極上的電荷量之改變為最大的。當自 狀態E程式化至狀態B時,對相鄰浮動閘極之耦合之量減少 但仍為顯著的。當自狀態E程式化至狀態A時,耦合之量 更進一步地減少。因此,隨後讀取WLn—丨之每一狀態所需 的杈正之量將取決於WLn上之相鄰之儲存元件的狀態而改 變。 圖20說明程式化針對兩個不同頁面(下部頁面及上部頁 鲁 面)而儲存負料之多態儲存元件的兩遍次技術之實例β描 繪四個狀態:狀態Ε(11)、狀態Α(1〇)、狀態Β(〇〇)及狀態 c(01)。對於狀態Ε,兩個頁面皆儲存”丨"。對於狀態Α,下 • 部頁面儲存"0"且上部頁面儲存"1"。對於狀態B,兩個頁 τ 面皆儲存"0"。對於狀態c,下部頁面儲存"i”且上部頁面 儲存0 »庄意’雖然已將特定位元樣式指派給該等狀態 中之每一者’但亦可指派不同位元樣式。 在第一程式化遍次中,根據待程式化至下部邏輯頁面中 之位元來設定儲存元件之臨限電壓位準。若此位元為邏輯 126206.doc •38、 1357603 ·· 1" ’則不改變臨限雷厭 m法, 電[因為此位元由於已較早捧除而 處於適當狀態。然而,若a 不除而 右#程式化之位兀為邏輯,,〇”,則 儲存元件之臨限位準掷士杏 、 徂旱增加而處於狀態A,如由箭頭11〇〇所 示。第一程式化遍次結束。Stored in different sides)—The block stores sixteen logical pages. Blocks and pages of other sizes can also be used. For an ABL architecture or an odd-even architecture, 70 pieces of memory can be erased by raising the p-well by a voltage (e.g., 20 V) and causing the word of the selected block to be erased. The source line and the bit line are floating. Erasing can be performed on the entire memory array, on a separate block, or on another unit of the storage element of the memory device. The floating gate of the electronic self-storage component is transferred to the P well region such that the VTH of the storage component becomes negative. In read and verify operations, select gates (SGD and SGS) connected to voltages in the range of 2. 5 V to 4·5 v and unselected word lines (eg, when WL2 is the selected word line) The time is up, and the machine 3) is raised to the read: voltage Vread (typically a voltage in the range of 4.5 乂 to 6 V) to operate the transistor as a turn-on gate. The selected word line milk 2 is connected to a voltage, which is referred to for each read and verify operation to determine whether the VTH of the associated storage element is above or below this level. For example, in a read operation of a two-stage storage device, the selected word line WL2 can be 'whether or not the VTH is higher than 〇 V. For example, in the verify operation of the double-order storage element 126206.doc -35 - 1357603, the selected word line WL2 is connected to 〇 8 v so that it is verified whether the VTH has reached at least 〇.8 V. The source and p wells are 选定 The selected I line (assumed to be an even bit line (BLe)) is precharged to, for example, 〇 7乂2. If VtH is above the read or verify level on the word line, the potential level of the bit line (BLe) associated with the associated memory element is maintained at the level due to the non-conductive memory element. The other side;, very v. For the aspect right Vth is lower than the read or verify level, the potential level of the associated bit line (BLe) is reduced to a low level (for example, less than 〇.5 V) because the conductive material The component discharges the bit line. The state of the storage element can thereby be sensed by a voltage comparator connected to the bit line. The above-described erasing, reading, and inspection are performed in accordance with techniques known in the art: intervening, many of the details explained may be used in the art, as well as other erasing, readings known in the art. Take the verification technology. Figure 19 depicts the continuation of the threshold voltage. An example of a storage element array is provided for each storage element in the condition of eight bits of data: cloth. Provide a first-to-be-limited distribution E for the storage elements other than _. ρ . Three threshold voltage distributions A, B and L ° of the storage element In one embodiment, the threshold voltage in the ^ . b blade is negative and the threshold voltage in the distribution is Λ, B and (: Positive value: ^ The limit voltage range corresponds to the threshold voltage level of the data and storage elements used in the predetermined data set of the data bits. The relationship between the parameters and the storage elements depends on the data used for the storage elements. For example, the encoders are incorporated by reference in their entirety in U.S. Patent No. 126,206, filed on Jun. No. 6,222,762, filed on Jan. No. describes various data encoding mechanisms for multi-state flash storage elements. In one embodiment, gray code assignments are used to assign data values to a threshold voltage range such that the threshold of the floating gate is When the voltage is erroneously shifted to its neighboring entity state, it will only affect one bit. One case assigns "11" to the threshold voltage range E (state E), and assigns "1 〇" to the threshold Voltage range A (state A), assigning "〇〇" to the threshold voltage Lu B (state B) and assign "01" to the threshold voltage range C (state c). However, in other embodiments, the Gray code is not used. Although four states are shown, the invention may also be used For other polymorphic structures, including polymorphic structures that include more or less than four states. Three read reference voltages Vra, Vrb, and Vrc are also provided for reading data from the storage element. Whether the threshold voltage of the storage element is higher or lower than Vra, Vrb and Vrc, the system can determine the state in which the storage element is located, such as a stylized condition. • In addition, three verification reference voltages Vva, Vvb and VVC are provided. When storing 70 pieces of code into state A, the system will test whether these storage elements have a threshold dust greater than or equal to Vva. When the storage element is programmed to • state B, the system will test whether the storage element has Greater than or equal to μ • = & power limit M ° When the storage component is programmed to state C, the system will determine whether the storage component has its threshold voltage greater than or equal to Vvc. Implementing a shot (4) The self-erasing state of the memory component can be directly simplified to the stylized state. For example, γ first erases the population of the storage component to be programmed, I26206.doc -37- 1357603 makes the group All of the storage elements are in an erased state E. The storage elements are then directly programmed to state A, B or C using a series of stylized pulses such as those depicted by the control gate voltage sequence of Figure 23, although some of the storage elements are From state E to state a, but other storage elements are from, state E is programmed to state B and/or from state e to state c. When staging from state E to state on WLn, the amount of parasitic coupling to the adjacent floating gate below wLnj is the largest because it is stylized from state E to state A or from state E to The change in voltage at state B is greater than the change in the amount of charge on the floating gate below WLn. When staging from state E to state B, the amount of coupling to adjacent floating gates is reduced but still Significantly, the amount of coupling is further reduced when staging from state E to state A. Therefore, the amount of positives required to subsequently read each state of WLn-丨 will depend on the adjacent WLn. The state of the storage element changes. Figure 20 illustrates an example of a two-pass technique that stylizes a polymorphic storage element that stores negative materials for two different pages (lower page and upper page), depicting four states: state Ε (11), status Α (1〇), status Β (〇〇), and status c (01). For status Ε, both pages store “丨". For status Α, the next page stores "0&quot And the upper page stores "1". For state B, two τ faces are stored "0". For state c, the lower page stores "i" and the upper page stores 0 »Zhuang Yi' although a specific bit pattern has been assigned to each of these states' but also Assigning different bit styles. In the first stylized pass, the threshold voltage level of the storage element is set according to the bit to be programmed into the lower logical page. If the bit is logic 126206.doc • 38, 1357603 ·· 1" 'There is no change to the thunder, the m method, electricity [because this bit is in the proper state because it has been taken earlier. However, if a does not divide and the right #stylized position is logical, , 〇", then the storage component's threshold is a bit of apricot, and the drought increases and is in state A, as indicated by arrow 11〇〇. The first stylization is over.

在第一程式化遍次中’根據正程式化至上部邏輯頁面中 之位元來設定儲存元件之臨限電壓位準。若上部邏輯頁面 位元將儲存邏輯T,則不發生程式化,因為取決於下部 頁面位元之程式化,儲存元件處於狀態E或A中之一者, 兩個狀態皆载運上部頁面位元’’”。若上部頁面位元為邏 輯T ’則將臨限電壓移位。若第—遍次使健存元件維持 在擦除狀態E,則在第二階段中程式化儲存元件,使得臨 限電壓增加以在狀態0内,如由箭頭2〇2〇所描繪。若由於 第一程式化遍次,儲存元件已程式化至狀態A,則在第二 遍次中進一步程式化儲存元件,使得臨限電壓增加以在狀 態B内,如由箭頭2010所描繪。第二遍次之結果為將儲存 元件程式化至經指定以對上部頁面儲存邏輯"〇"而不改變 下部頁面之資料的狀態。在圖19及圖20中,對相鄰字線上 之浮動閘極之耦合的量取決於最終狀態。 在一貫施例中,可設置系統以執行全序列寫入(若寫入 足夠資料而填滿整個頁面)。若針對全頁未寫入足夠資 料,則程式化過程可程式化以所接收之資料來程式化之下 部頁面。當接收到後續資料時’系統將接著程式化上部頁 面。在又—實施例中,系統可以程式化下部頁面之模式開 始寫入,且若隨後接收到足以填滿整個(或大部分)字線之 126206.doc -39- 1357603 儲存元件的資料,則可轉換至全序列程式化模式。此實施 例之更多細節揭示於2〇〇6年6月15日公布的以引用之方式 王文併入本文中的標題為"pipelined Pr〇gramming Non-Volatile Memories Using EaHy Data"之美 國 專利申 請公開 案第 2006/0126390號中。In the first stylized pass, the threshold voltage level of the storage element is set according to the bit being programmed into the upper logical page. If the upper logical page bit will store the logic T, no stylization will occur, because depending on the stylization of the lower page bit, the storage element is in one of the states E or A, both of which carry the upper page bit. ''". If the upper page bit is logic T', the threshold voltage is shifted. If the first time the dummy element is maintained in the erased state E, then the storage element is programmed in the second stage, so that The limit voltage is increased in state 0, as depicted by the arrow 2〇2〇. If the storage element has been programmed to state A due to the first stylization pass, the storage element is further stylized in the second pass, The threshold voltage is increased to be in state B, as depicted by arrow 2010. The result of the second pass is to program the storage element to be designated to store the logic for the upper page without changing the lower page. The state of the data. In Figures 19 and 20, the amount of coupling to the floating gates on adjacent word lines depends on the final state. In a consistent example, the system can be set to perform full sequence writes (if enough writes) Fill in the information Page). If enough data is not written for the full page, the stylization process can be programmed to program the lower page with the received data. When the subsequent data is received, the system will then program the upper page. In an embodiment, the system can program the lower page mode to start writing, and if it subsequently receives enough data to fill the entire (or most) word line of the 126206.doc -39-1357603 storage element, it can be converted to full Sequence stylized mode. More details of this embodiment are disclosed in the article published on June 15, 2002. The article titled "pipelined Pr〇gramming Non-Volatile Memories Using EaHy U.S. Patent Application Publication No. 2006/0126390 to Data".

圖21 a至21c揭示用於程式化非揮發記憶體之另一過程, 其藉由以下方式來減少浮動閘極至浮動閘極之耦合的效 應:對於任-特定儲存元件,在針對先前頁面將資料寫入 至相鄰儲存元件之後關於特定頁面將㈣寫人至此特定儲 存元件。在-實例實施例中,非揮發儲存元件使用四個資 料狀態而對每一儲存元件儲存兩個位元之資料。舉例而 言,假定狀態E為擦除狀態而狀態A ' MC為程式化狀 態。狀態E儲存資㈣。狀態續存資料〇卜狀態B儲存資 料Π)。狀態C儲存資制。此為非格雷編碼之實例,因為 兩個位元在相鄰狀態八與5之間改變。亦可使用使資料至 實體資料狀態之其他編碼。每-儲存元件儲存兩個頁面之 資料。出於參考之目的,將此等資料頁面稱作上部頁面及 下部頁面;然巾’其可被給予其他標記。關於狀態A,上 部頁面儲存位元0且下部頁面儲存位元1。關於狀_,上 部頁面儲存位元1且下部頁而被古_ j貝面儲存位兀0。關於狀態c,兩 個頁面皆儲存位元資料〇。 該程式化過程為兩步過程 頁面。若下部頁面將保持資 狀態E。若資料將程式化至〇 。在第一步驟中,程式化下部 料1 ’則儲存元件狀態保持於 ’則儲存元件之電壓臨限值升 126206.doc •40· 1357603 高’使得儲存元件被程式化至狀態B、圖…因此展示將 儲存疋件自狀態E程式化至狀態B,。狀態B,為臨時狀態B。 因此,將驗證點描繪為Vvb,,其低於Vvb。 在一實施例中’在儲存元件自狀態』程式化至狀態Β· 在「反及」串中的鄰近之儲存元件(WLn+1)將接著 關於”下部頁面來程式化。舉例而言,返回參看圖2,在 程式化儲存元件1〇6之下部頁面後,將程式化儲存元件_ 之下部頁面。在程式化儲存元件1〇4後,若儲存元件ι〇4且 有自狀態E升高至狀態B,之臨限電壓,則浮動間極至浮動 閘極之耦合效應將使儲存元件1〇6之表觀臨限電壓升高。 此將具有使針對狀態B,之臨限電壓分布加寬為如圖⑽之 臨限電壓分布215〇所描繪之分布的效應。臨限電I分布之 此明顯加寬將在程式化上部頁面時得以糾正。 圖…騎程式化上部頁面之過程。若儲存元件處於擦 除狀態E且上部頁面將保持於i,則儲存元件將保持於狀態 E。若儲存元件處於狀態E且其上部頁面資料將程式化^ y則儲存元件之臨限電壓將升高’使得儲存元件處於狀 態A。若儲存元件處於中間臨限電壓分布2丨5〇中且上部頁 面資料保持於i,則儲存元件將程式化至最終狀態b。二儲 存元件處於中間臨限電壓分布215〇中且上部頁面資料將變 為資料〇’則儲存元件之臨限電壓將升高,使得健存元件 處於狀態C。由圖⑴至⑴描繪之過程減少浮動閘極至浮 動閘極之耦合的效應,因為僅鄰近儲存元件之上部頁面程 式化將對給定儲存元件之表觀臨限電壓具有效應。替代狀 126206.doc •41 · 1357603 態編碼之實例為在上部頁面資料為】時自分布以別移至狀 態C ’且在上部頁面資料為〇時移至狀態B。 雖然圖21 a至21c提供關於四個資料狀態及兩個資料頁面 的實例,但所教示之概念可適用於具有多於或少於四個之 狀態及不同於兩個頁面之其他實施例。 圖21d描述粗略/精細程式化過程。如先前結合圖^所提 及,最初可以粗略模式來程式化儲存元件以將其快速地移 向目標程式化條件且接著以精細模式來程式化以按照較大 之準確性較慢地將其移至目標程式化條件。精細程式化模 式可涉及(例如)在VPGM脈衝串中使用減少之步長及/或對選 定之「反及」_之位元線施加抑制電壓。另外,可在一遍 次或多遍次程式化中發生粗略_精細程式化。在一遍次粗 略/精細程式化中,如圖23中所指示,在VpGM脈衝串期間 存在自粗略程式化至精細程式化之切換。相反,在多遍次 粗略/精細程式化中,例如,可在第一遍次期間使用粗略 程式化,而在第二遍次期間使用精細程式化。如圖24中所 指示’自粗略程式化至精細程式化之切換可發生於(例如) 完整VPGM脈衝串之間。另外或其他,%心脈衝串在第二遍 次或其他額外遍次之程式化中可使用值之較低範圍。可將 多遍次粗略/精細程式化視為特定類型之多遍次程式化, 其通常涉及(例如)使用一個以上之脈衝串在一個以上之遍 次中將儲存元件程式化至目標程式化條件。 舉例而言’可將儲存元件自擦除狀態(狀態幻程式化至 目標程式化狀態A、B或C。在一方法中,使用粗略程式化 126206.doc -42 - 將儲存元件程式化至臨時狀態A,、B,或c,,該等狀態分別 ”有相關聯之驗證位準VvaL、VvbL或VvcL。下標,,L"表示 驗3登位準與低於目標狀態之較低狀態相關聯。隨後,使用 精細程式化將儲存元件自臨時狀態程式化至狀態A、b或 C ’該等狀態分別具有相關聯之驗證位準Vvail、VvbH或 Vvch。下標”Η”表示驗證位準與為最終目標狀態之較高狀 癌相關聯。經程式化之儲存元件之臨限電壓因此在第一程 式化階段期間自第一位準(例如,狀態Α)增加至第二位準 (例如,VvaL、VvbL或VvcL)且在第二程式化階段期間自第 一位準增加至第三位準(例如,vvaH、VvbH或VvcH)。 圖22為描述用於程式化非揮發記憶體之方法之一實施例 的流程圖。在一實施例中’在程式化之前擦除儲存元件 (以區塊或其他單位)。在步驟2200中,由控制器發出,,資料 載入"命令且由控制電路1510接收輸入。在步驟2205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 1514。在步驟2210中,將用於所定址之頁面的一頁面之程 式化資料輸入至資料緩衝器以供程式化。將此資料鎖存於 適當之鎖存器集合中。在步驟2215中,由控制器將,,程式 化"命令發出至狀態機1512。 由"程式化"命令觸發,將使用施加至適當之選定之字線 的圖23之脈衝串2300的步進式程式化脈衝2305、2310、 2315、2320、2325、2330、2335、2340、2345、2350......Figures 21a through 21c illustrate another process for staging non-volatile memory that reduces the effects of floating gate-to-floating gate coupling by: for any-specific storage elements, for the previous page After the data is written to the adjacent storage element, (4) is written to the specific storage element with respect to the specific page. In the example embodiment, the non-volatile storage element stores four bits of data for each storage element using four data states. For example, assume that state E is an erased state and state A ' MC is a stylized state. State E storage (4). Status renewal data 状态 status B storage data Π). State C stores the system. This is an example of non-Gray coding because the two bits change between adjacent states eight and five. Other codes that enable the data to be in the state of the entity data can also be used. Each storage element stores two pages of information. For the purposes of this reference, such data pages are referred to as upper and lower pages; however, they may be given other indicia. Regarding state A, the upper page stores bit 0 and the lower page stores bit 1. Regarding the __, the upper page stores the bit 1 and the lower page is stored by the ancient _j. Regarding state c, both pages store bit data 〇. This stylization process is a two-step process page. If the lower page will remain in the status E. If the data will be stylized to 〇. In the first step, the stylized lower material 1 'the storage element state is maintained at 'the storage element voltage threshold 126206.doc •40·1357603 high' causes the storage element to be programmed to state B, figure... Shows that the storage condition is stylized from state E to state B. State B is the temporary state B. Therefore, the verification point is depicted as Vvb, which is lower than Vvb. In one embodiment, 'in the storage element self-state' is stylized to state Β. The adjacent storage element (WLn+1) in the "reverse" string will then be programmed with respect to the "lower page. For example, return Referring to Figure 2, after the page below the stylized storage element 1〇6, the lower part of the page will be stylized. After the stylized storage element 1〇4, if the element 〇4 is stored and the self-state E rises To state B, the threshold voltage, the coupling effect of the floating-to-floating gate will increase the apparent threshold voltage of the storage element 1〇6. This will have the threshold voltage distribution for state B, The width is the effect of the distribution depicted by the threshold voltage distribution 215〇 of Figure 10. The apparent widening of the threshold I distribution will be corrected when the upper page is programmed. Figure...The process of riding the upper page. When the storage element is in the erased state E and the upper page will remain at i, the storage element will remain in state E. If the storage element is in state E and its upper page data will be programmed, then the threshold voltage of the storage element will rise. 'Make the storage component in State A. If the storage element is in the middle threshold voltage distribution 2丨5〇 and the upper page data remains at i, the storage element will be programmed to the final state b. The second storage element is in the middle threshold voltage distribution 215〇 and upper The page data will become data 〇 'The threshold voltage of the storage component will rise, causing the stagnation component to be in state C. The process depicted by Figures (1) through (1) reduces the effect of the floating gate to floating gate coupling because only Program stylization on top of adjacent storage elements will have an effect on the apparent threshold voltage of a given storage element. Alternative 126206.doc •41 · 1357603 An example of state coding is self-distribution when the upper page data is] State C' and move to state B when the upper page data is 。. Although Figures 21a through 21c provide examples of four data states and two profile pages, the concepts taught can be applied to have more or less Four states and other embodiments different from the two pages. Figure 21d depicts a rough/fine stylization process. As previously mentioned in conjunction with Figure ^, it can be initially stylized in a coarse mode. The component is stored to move it quickly to the target stylized condition and then programmed in a fine mode to move it to the target stylized condition more slowly with greater accuracy. The fine stylized mode may involve, for example, The reduced step size is used in the VPGM burst and/or the suppression voltage is applied to the selected "reverse" bit line. In addition, coarse_fine stylization can occur in one or more passes of stylization. In one pass coarse/fine stylization, as indicated in Figure 23, there is a switch from coarse stylization to fine stylization during the VpGM burst. In contrast, in multi-pass rough/fine stylization, for example, coarse stylization can be used during the first pass and fine stylization used during the second pass. Switching from coarse stylization to fine stylization as indicated in Figure 24 can occur, for example, between full VPGM bursts. Additionally or alternatively, the % heart burst may use a lower range of values in the second pass or other extra pass stylization. Multiple pass coarse/fine stylization can be considered as a multi-pass stylization of a particular type, which typically involves, for example, using more than one burst to program the storage element to the target stylized condition in more than one pass. . For example, the storage element can be self-erased (the state is simplistically programmed to the target stylized state A, B or C. In one method, the rough stylized 126206.doc -42 - stylizes the storage element to the temporary State A, B, or c, these states respectively have an associated verification level VvaL, VvbL or VvcL. Subscript, L" indicates that the 3rd rank is associated with a lower state than the target state Subsequently, the fine-grained stylization is used to program the storage elements from the temporary state to state A, b or C'. These states respectively have associated verification levels Vvail, VvbH or Vvch. The subscript "Η" indicates the verification level. Associated with a higher cancer that is the final target state. The threshold voltage of the programmed storage element is thus increased from the first level (eg, state Α) to the second level during the first stylization phase (eg , VvaL, VvbL or VvcL) and increase from the first level to the third level (eg, vvaH, VvbH or VvcH) during the second stylization phase. Figure 22 is a diagram for describing a method for staging non-volatile memory Flowchart of one embodiment. In an embodiment 'Erase the storage element (in blocks or other units) prior to stylization. In step 2200, the controller issues a data load " command and receives input by control circuitry 1510. In step 2205, the The address information of the page address is input from the controller or host to the decoder 1514. In step 2210, the stylized data for a page of the addressed page is input to the data buffer for stylization. Latched into the appropriate set of latches. In step 2215, the controller will issue the stylized " command to state machine 1512. Triggered by the "stylized" command, the use is applied to the appropriate selection The stepped stylized pulses 2305, 2310, 2315, 2320, 2325, 2330, 2335, 2340, 2345, 2350 of the pulse train 2300 of FIG. 23 of the zigzag line...

來將在步驟2210中鎖存之資料程式化至由狀態機1512控制 之選定的儲存元件中。在步驟2220中,將程式化電壓VPGM 126206.doc •43· 1357603 初始化至起始脈衝(例如,12 v或另一值)且將由狀態機 1512維持之程式計數器(pc)初始化為零。在步驟Μ。中, 應用初始升壓模式,且在步驟中,將第脈衝施 加至選定之字線以開始程式化與選定之字線相關聯之儲存 元件。若邏輯"0"儲存於特定資料鎖存器中指示應程式化 相應儲存元件,則將相應位元線接地。另一方面,若邏輯 1 _儲存於特鎖存器中指示相應儲存元件應保持於其當 前資料I態’則將相應位元線連接至Vdd以抑制程式化。田 2步驟2235中,驗證選定之儲存元件之狀態。若偵測到 選定之儲存元件之目標臨限電壓已達到適t位準,則儲存 於相應資料鎖存器中之資料變為邏輯”,、若偵測到該臨 限電壓尚未達到適當位準,則儲存於相應資料鎖存器中之 貝料不改變。以此方式’不必程式化在相應資料鎖存器中 儲存有邏輯T之位元線。當所有資料鎖存器儲存邏輯M” 時’狀態機(經由上述之線或型機制)知曉所有選定之儲存 元件已被程式化。在步驟224〇中,作出關於所有資料鎖存 器是否儲存邏輯”Γ’的檢查。若所有資料鎖存器儲存邏輯 1則程式化過程完成且成功,因為所有選定之儲存元 件被程式化並驗證。在步驟2245中報告"通過"狀態。 右在步驟224G中判定並非所有資料鎖存器儲存邏輯 ”1",則程式化過程繼續。在步驟2250中,對照程式化極 限值PCmax來檢查程式計數器PC。程式化極限值之一實例 為二十;然而’亦可使用其他數目。若程式計數器PC不小 於PCmax’ _式化過程發生故障且在步驟2255中報告 126206.doc 1357603 ’’故障"狀態。若程式計數器PC小於pCmax,則在步驟226〇 中使VPGM增加了步長且程式計數器Pc遞增。在步驟2265 處,作出關於是否滿足升壓模式切換標準(例如,見圖4)之 判定。若滿足此標準,則在步驟2270處切換升壓模式,且 過程返回至步驟2230以施加下一 VPGM脈衝。若在步驟2265 處不滿足升壓模式切換標準,則過程返回至步驟223〇以在 未改變升壓模式之情況不施加不一 VpGM脈衝。 圖23描繪在程式化期間施加至非揮發儲存元件之控制閘 極的實例脈衝串2300,及在脈衝串期間發生的升壓模式切 換。脈衝串2300包括施加至針對程式化而選擇之字線的一 連串程式化脈衝 2305、2310、2315、2320、2325、2330、 2335、2340、2345、2350……。在一實施例中,程式化脈 衝具有電壓VPGM,該電壓開始於12 v且對每一連續程式化 脈衝增加增量(例如,〇 5 v)直至達到最大值2〇 v為止。在 程式化脈衝之間存在驗證脈衝。舉例而言,驗證脈衝集合 23 06包括二個驗證脈衝。在一些實施例中,針對資料正被 程式化至之每一狀態(例如,狀態A、8及c),可存在一驗 證脈衝。在其他實施例中,可存在更多或更少之驗證脈 衝。舉例而言’每一集合中之驗證脈衝可具有振幅Vva、The data latched in step 2210 is programmed into the selected storage element controlled by state machine 1512. In step 2220, the programmed voltage VPGM 126206.doc • 43· 1357603 is initialized to a start pulse (e.g., 12 v or another value) and the program counter (pc) maintained by state machine 1512 is initialized to zero. In the steps Μ. The initial boost mode is applied, and in the step, the first pulse is applied to the selected word line to begin programming the storage elements associated with the selected word line. If the logic "0" is stored in a specific data latch indicating that the corresponding storage element should be programmed, the corresponding bit line is grounded. On the other hand, if the logic 1 _ is stored in the special latch indicating that the corresponding storage element should remain in its current data I state, then the corresponding bit line is connected to Vdd to suppress stylization. In field 2 step 2235, the status of the selected storage element is verified. If it is detected that the target threshold voltage of the selected storage component has reached the appropriate t level, the data stored in the corresponding data latch becomes logic", and if the threshold voltage is detected, the appropriate level has not been reached. , the material stored in the corresponding data latch does not change. In this way, it is not necessary to programmatically store the bit line of logic T in the corresponding data latch. When all data latches store logic M" The 'state machine (via the line or type mechanism described above) knows that all selected storage elements have been programmed. In step 224, a check is made as to whether all data latches store logic "Γ". If all data latches store logic 1, the stylization process is completed and successful because all selected storage elements are programmed and verified. The "pass" status is reported in step 2245. Right in step 224G, it is determined that not all of the data latches store logic "1", then the stylization process continues. In step 2250, the program counter PC is checked against the programmed limit PCmax. An example of one of the stylized limits is twenty; however, other numbers may be used. If the program counter PC is not less than PCmax', the process fails and the 126206.doc 1357603 ''fault" state is reported in step 2255. If the program counter PC is less than pCmax, then in step 226, VPGM is incremented by the step size and the program counter Pc is incremented. At step 2265, a determination is made as to whether the boost mode switching criteria (e.g., see Figure 4) is met. If this criterion is met, the boost mode is switched at step 2270 and the process returns to step 2230 to apply the next VPGM pulse. If the boost mode switching criterion is not met at step 2265, the process returns to step 223 to not apply a different VpGM pulse without changing the boost mode. Figure 23 depicts an example pulse train 2300 applied to the control gate of the non-volatile storage element during stylization, and boost mode switching that occurs during the burst. Burst 2300 includes a series of stylized pulses 2305, 2310, 2315, 2320, 2325, 2330, 2335, 2340, 2345, 2350, ... applied to the word line selected for stylization. In one embodiment, the stylized pulse has a voltage VPGM that begins at 12 v and increments (e.g., 〇 5 v) for each successive stylized pulse until a maximum of 2 〇 v is reached. There is a verify pulse between the stylized pulses. For example, the set of verification pulses 23 06 includes two verify pulses. In some embodiments, there may be an authentication pulse for each state to which the material is being programmed (e.g., states A, 8 and c). In other embodiments, there may be more or fewer verification pulses. For example, the verification pulse in each set may have an amplitude Vva,

Vvb 及 Vvc(圖 20)、Vvb’(圖 21a) ' 或 VvaL、VvbL及 VvcL 或 VvaH、VvbH及 VvcH(圖 21d) 〇 將升壓模式之切換描繪為在施加程式化脈衝2335之前發 生在切換之别,應用第一升壓模式,而在切換之後,應 用第一升壓模式。如所提及,當程式化發生時(例如,當 126206.doc •45- 1357603Vvb and Vvc (Fig. 20), Vvb' (Fig. 21a)' or VvaL, VvbL and VvcL or VvaH, VvbH and VvcH (Fig. 21d) 描绘 The switching of the boost mode is depicted as occurring before the application of the stylized pulse 2335 The first boost mode is applied, and after the switch, the first boost mode is applied. As mentioned, when stylization occurs (for example, when 126206.doc •45-1357603

施加程式化脈衝時)’施加被施加至字線以實施升壓模式 的電壓。實務上’在每一程式化脈衝之前可稍微地起始升 壓模式之升壓電壓且在每一程式化脈衝之後將其移除。因 此,在驗證過程期間(例#,其發生於程式化脈衝之間), 不施加升壓電壓。而是,將通常小於升壓電壓之讀取電壓 施$至未選定之字線。讀取電壓具有振幅,該振幅足夠在 當前程式化之儲存元件之臨限電壓正與驗證位準相比時將 「反及」率中之先前程式化之儲存元件維持為開啟的。 一程式化階段中,將脈衝串 一子集(例如,脈衝2305、 因此,在一方法中,在第 2300中之程式化脈衝之第 2310、2315、2320、2325及2330)施加至一或多個儲存元 件,且在第二程式化階段中,將該脈衝串中之脈衝之第二 子集(例如,脈衝2335、2340、2345、235〇)施加至該或: 等儲存7L件。每一程式化遍次可因此包括多個程式化 段。 • 圖24描繪在程式化期間施加至非揮發儲存元件之控制閘 極的實例脈衝串,及在脈衝串之間發生的升壓模式之切 換。特定言之,將升壓模式之切換描綠為在脈衝串24_ • 期之間發生。在切換之前,在第-脈衝串2_期間,應 •用第-升壓模式’而在切換之後,在第二脈衝串245〇期 間,應用第二升壓模式。舉例而言,在多遍次程式化過程 中之第-遍次期間可施加脈衝串24〇〇,而在此種程式化過 程中之第二遍次期間施加脈衝串245〇。因此,在一方法 中’在第-程式化階段中,將第一脈衝串(例如,:衝串 126206.doc -46· 1357603 2400)施加至選定之字線上之—或多個館存元件,且在第 二程式化階財,將第二脈衝串(例如,脈衝串2450)施加 至該或該㈣存元件。每-程式化遍次可因此與程式 段一致。When a stylized pulse is applied, the voltage applied to the word line to apply the boost mode is applied. Practically, the boost voltage of the boost mode can be initiated slightly before each stylized pulse and removed after each stylized pulse. Therefore, during the verification process (example #, which occurs between the stylized pulses), no boost voltage is applied. Instead, a read voltage, typically less than the boost voltage, is applied to the unselected word line. The read voltage has an amplitude sufficient to maintain the previously stylized storage element in the "reverse" rate on when the threshold voltage of the currently staging storage element is being compared to the verify level. In a stylized phase, a subset of the bursts (e.g., pulse 2305, and therefore, in a method, 2310, 2315, 2320, 2325, and 2330 of the stylized pulses in 2300) are applied to one or more A storage element, and in a second stylization phase, a second subset of pulses in the burst (eg, pulses 2335, 2340, 2345, 235〇) is applied to the OR: etc. 7L pieces are stored. Each stylized pass can thus include multiple stylized segments. • Figure 24 depicts an example pulse train applied to the control gate of the non-volatile storage element during stylization and the switching of the boost mode that occurs between the bursts. In particular, switching the boost mode to green occurs between bursts 24_ •. Prior to the switching, during the first burst 2_, the second boost mode should be applied during the second burst 245 during the second burst 245. For example, the burst 24 可 may be applied during the first pass of the multi-pass stylization process, while the burst 245 施加 is applied during the second pass of the stylization process. Thus, in a method, 'in the first stylization phase, the first burst (eg, burst 126206.doc -46·1357603 2400) is applied to the selected word line - or a plurality of library elements, And in the second stylized order, a second burst (eg, burst 2450) is applied to the or the (four) memory component. Each stylized pass can therefore be consistent with the program.

、,為達成說明及描述之目的已呈現本發明之前述詳細描 述。其並不意欲為詳盡的或將本發明限於所揭示之精確形 式。按照上述教示,許多修改及變化為可能的。選擇所描 述之實施例以便最佳地解釋本發明之原理及其實際應用田 以猎此使熟習此項技術者能夠在各種實施例中且在進行適 於所預期之特^用途之各種修改的情況下最佳地使用本發 明。意欲由所附申請專利範圍來界定本發明之範疇。 【圖式簡單說明】 t 圖1為「反及」串之俯視圖。 圖2為圖1之「反及」串之等效電路圖。 圖3為「反及」快閃儲存元件陣列之方塊圖。 圖4描繪展示升壓模式決定過程之概念圖。 圖5描繪用於在程式化期間切換升壓模式之過程。 圖6描繪經由複數個字線而實施之自升壓模式。 圖7描繪經由複數個字線而實施之局部自升壓模式。 圖8描繪經由複數個字線而實施之擦除區域自升壓模 式0 圖9描繪經由複數個字線而實施之第一修正擦除區域 升壓模式。 圖1 〇描繪經由複數個字線而實施之第二修正擦除區域 126206.doc •47- 1357603 升壓模式。 圖1 la描繪經由複數個字線而實施之第三修正擦除區域 自升壓模式。 圖lib描繪經由複數個字線而實施之第四修正擦除區域 自升壓模式。 圖11c描繪經由複數個字線而實施之第五修正擦除區域 自升壓模式。 圖12描繪展示如何藉由設定位元線抑制電壓而達成粗略 及精細程式化之時間線。 圖13描繪展示程式化及擦除區域之未選定之「反及」串 的橫截面圖。 圖14為「反及」快閃儲存元件陣列之方塊圖。 圖15為使用單列/行解碼器及讀取/寫入電路之非揮發記 憶體系統之方塊圖。 圖16為使用雙列/行解碼器及讀取/寫入電路之非揮發記 憶體系統之方塊圖。 圖17為描繪感測區塊之一實施例之方塊圖。 圖18說明針對全位元線記憶體架槿 朱穉或針對奇偶記憶體架 構而將記憶體陣列組織成區塊的實例。 圖19森續臨限電壓分布·之實例集合。 圖20描繪臨限電壓分布之實例集合。 圖21 a至圖21c展示各種臨限電壓分 电I刀帝並描述用於程式化 非揮發記憶體的過程。 圖21d描述粗略/精細程式化過程。 126206.doc _48, 1357603 圖22為描述用於程式化非揮發記憶體之過程之一實施例 的流程圖。 圖23描繪在程式化期間施加至非揮發儲存元件之控制閘 極的實例脈衝串,及在脈衝串期間發生的升壓模式切換。The foregoing detailed description of the invention has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations are possible in light of the above teachings. The embodiments described are chosen to best explain the principles of the invention and the application of the embodiments of the invention in the various embodiments of the invention. The present invention is optimally used in the case. The scope of the invention is intended to be defined by the scope of the appended claims. [Simple description of the diagram] t Figure 1 is a top view of the "reverse" string. Figure 2 is an equivalent circuit diagram of the "reverse" string of Figure 1. Figure 3 is a block diagram of an "reverse" flash memory device array. Figure 4 depicts a conceptual diagram showing the boost mode decision process. Figure 5 depicts a process for switching the boost mode during stylization. Figure 6 depicts a self boosting mode implemented via a plurality of word lines. Figure 7 depicts a local self-boost mode implemented via a plurality of word lines. Figure 8 depicts an erase region self-boost mode implemented via a plurality of word lines. Figure 9 depicts a first modified erase region boost mode implemented via a plurality of word lines. Figure 1 〇 depicts a second modified erase region implemented via a plurality of word lines 126206.doc • 47-1357603 Boost mode. Figure la depicts a third modified erase region self-boost mode implemented via a plurality of word lines. Figure lib depicts a fourth modified erase region self-boost mode implemented via a plurality of word lines. Figure 11c depicts a fifth modified erase region self-boost mode implemented via a plurality of word lines. Figure 12 depicts a timeline showing how rough and fine stylization can be achieved by setting the bit line suppression voltage. Figure 13 depicts a cross-sectional view of an unselected "reverse" string showing stylized and erased regions. Figure 14 is a block diagram of an "reverse" flash memory device array. Figure 15 is a block diagram of a non-volatile memory system using a single column/row decoder and a read/write circuit. Figure 16 is a block diagram of a non-volatile memory system using a dual column/row decoder and a read/write circuit. 17 is a block diagram depicting one embodiment of a sensing block. Figure 18 illustrates an example of organizing memory arrays into blocks for a full bit line memory frame or for an even and odd memory frame. Figure 19 shows an example collection of the threshold voltage distribution. Figure 20 depicts an example set of threshold voltage distributions. Figures 21a through 21c show various threshold voltage divisions and describe the process for staging non-volatile memory. Figure 21d depicts a rough/fine stylization process. 126206.doc _48, 1357603 FIG. 22 is a flow chart depicting one embodiment of a process for programming non-volatile memory. Figure 23 depicts an example pulse train applied to the control gate of the non-volatile storage element during stylization, and boost mode switching that occurs during the burst.

圖24描繪在程式化期間施加至非揮發儲存元件之控制閘 極的實例脈衝串,及在脈衝串之間發生的升壓模式切換。 【主要元件符號說明】 100 電晶體 100CG 控制閘極 100FG 浮動閘極 102 電晶體 102CG 控制閘極 102FG 浮動閘極 104 電晶體 104CG 控制閘極 104FG 浮動閘極 106 電晶體 106CG 控制閘極 106FG 浮動閘極 120 第一選擇閘極 120CG 控制閘極 122 第二選擇閘極 122CG 控制閘極 126 位元線 126206.doc -49- 1357603Figure 24 depicts an example pulse train applied to the control gate of the non-volatile storage element during stylization, and boost mode switching occurring between the bursts. [Main component symbol description] 100 transistor 100CG control gate 100FG floating gate 102 transistor 102CG control gate 102FG floating gate 104 transistor 104CG control gate 104FG floating gate 106 transistor 106CG control gate 106FG floating gate 120 first selection gate 120CG control gate 122 second selection gate 122CG control gate 126 bit line 126206.doc -49- 1357603

128 源極線 320 「反及」 321 位元線 322 選擇閘極 323 儲存元件 324 儲存元件 325 儲存元件 326 儲存元件 327 選擇閘極 340 「反及」 341 位元線 342 選擇閘極 343 儲存元件 344 儲存元件 345 儲存元件 346 儲存元件 347 選擇閘極 360 「反及」 361 位元線 362 選擇閘極 363 儲存元件 364 儲存元件 365 儲存元件 366 儲存元件 126206.doc -50- 1357603128 source line 320 "reverse" 321 bit line 322 select gate 323 storage element 324 storage element 325 storage element 326 storage element 327 select gate 340 "reverse" 341 bit line 342 select gate 343 storage element 344 Storage Element 345 Storage Element 346 Storage Element 347 Select Gate 360 "Reverse" 361 Bit Line 362 Select Gate 363 Storage Element 364 Storage Element 365 Storage Element 366 Storage Element 126206.doc -50- 1357603

367 選擇閘極 400 區塊 405 區塊 410 區塊 415 區塊 420 區塊 425 區塊 430 區塊 435 區塊 440 區塊 445 區塊 460 區塊 1200 曲線 1250 曲線 1300 「反及」串 1304 源極供應線 1306 選擇閘極 1308 儲存元件 1310 儲存元件 1312 儲存元件 1314 儲存元件 13 16 儲存元件 1320 儲存元件 1322 儲存元件 126206.doc •51 · 1357603 1324 汲極側選擇閘; 1326 位元線 1350 區域 1360 區域 1390 基板 1400 「反及」儲存: 1404 源極線 1406 位元線 1426 汲極端子 1428 源極端子 1450 「反及」串 1500 感測區塊 1510 控制電路 1512 狀態機 1514 晶載位址解碼5 1516 功率控制模組 1518 線 1520 資料匯流排 1530A 列解碼器 1530B 列解碼器 1560 行解碼器 1560A 行解碼器 1560B 行解碼器 1565 言買取/寫入電路 126206.doc -52- 1357603367 Select Gate 400 Block 405 Block 410 Block 415 Block 420 Block 425 Block 430 Block 435 Block 440 Block 445 Block 460 Block 1200 Curve 1250 Curve 1300 "Reverse" String 1304 Source Supply line 1306 select gate 1308 storage element 1310 storage element 1312 storage element 1314 storage element 13 16 storage element 1320 storage element 1322 storage element 126206.doc • 51 · 1357603 1324 bungee side selection gate; 1326 bit line 1350 area 1360 area 1390 Substrate 1400 "Reverse" Storage: 1404 Source Line 1406 Bit Line 1426 汲 Terminal 1428 Source Terminal 1450 "Reverse" String 1500 Sensing Block 1510 Control Circuit 1512 State Machine 1514 Crystal Address Address Decoding 5 1516 Power Control Module 1518 Line 1520 Data Bus 1530A Column Decoder 1530B Column Decoder 1560 Line Decoder 1560A Line Decoder 1560B Line Decoder 1565 Word Buy/Write Circuit 126206.doc -52- 1357603

1565A 讀取/寫入電路 1565B 讀取/寫入電路 1570 感測電路 1572 匯流排 1580 感測模組 1582 位元線鎖存器 1592 處理器 1593 輸入線 1594 資料鎖存器 1596 記憶體裝置 1598 記憶體晶粒 1800 奇偶架構 1810 全位元線架構 2010 箭頭 2020 箭頭 2150 臨限電壓分布 2300 實例脈衝串 2305 程式化脈衝 2310 程式化脈衝 2315 程式化脈衝 2320 程式化脈衝 2325 程式化脈衝 2340 程式化脈衝 2345 程式化脈衝 126206.doc -53- 1357603 2350 2400 2450 A A' B B, BLe1565A Read/Write Circuit 1565B Read/Write Circuit 1570 Sensing Circuit 1572 Busbar 1580 Sensing Module 1582 Bit Line Latch 1592 Processor 1593 Input Line 1594 Data Latch 1596 Memory Device 1598 Memory Body Grain 1800 Parity Structure 1810 Full Bit Line Architecture 2010 Arrow 2020 Arrow 2150 Threshold Voltage Distribution 2300 Example Burst 2305 Stylized Pulse 2310 Stylized Pulse 2315 Stylized Pulse 2320 Stylized Pulse 2325 Stylized Pulse 2340 Stylized Pulse 2345 Stylized pulse 126206.doc -53- 1357603 2350 2400 2450 AA' BB, BLe

C C' E SGD SGS Vbl Vdd VH V iso VL VPARTIAL inhibit V PASS v PASS-LOW V PGM VpoM-HIGH 126206.doc 程式化脈衝 脈衝串 脈衝串 程式化狀態 臨時狀態 程式化狀態 臨時狀態 偶數位元線 奇數位元線 程式化狀態 臨時狀態 擦除狀態 汲極側選擇閘極控制線 源極側選擇閘極控制線 位元線電壓 電位 較高驗證位準 隔離電壓 較低驗證位準 位元線抑制電壓 導通電壓 相對較低之導通電壓 程式化電壓 較高範圍 -54- 1357603CC' E SGD SGS Vbl Vdd VH V iso VL VPARTIAL inhibit V PASS v PASS-LOW V PGM VpoM-HIGH 126206.doc Stylized pulse burst burst stylized state temporary state stylized state temporary state even bit line odd bit Meta-threaded state Temporary state Erase state Bole side Select gate control line Source side Select gate control line Bit line Voltage potential higher Verify level isolation voltage Lower verify level bit line suppress voltage turn-on voltage Relatively low turn-on voltage stylized voltage higher range -54 - 1357603

VpGM-LOW 較低範圍 VsGD 電壓 VsGS 電壓 VsOURCE 電位 V th 臨限電壓 Vra、Vrb、Vrc 讀取參考電壓 Vva、Vvb、Vvc、Vvb' 振幅 WL1 至 WL7 字線VpGM-LOW Lower range VsGD Voltage VsGS Voltage VsOURCE Potential V th Threshold voltage Vra, Vrb, Vrc Read reference voltage Vva, Vvb, Vvc, Vvb' Amplitude WL1 to WL7 Word line

126206.doc -55-126206.doc -55-

Claims (1)

第096141503號專利申請案 令文申請專利範圍替換本(100年8月) 、申請專利範圍: —種用於操作非揮發儲存器之方法,其包含: 卜啤&月 曰修正本 程式化一非揮發儲存元件集合中之至少一儲存元件, 該非揮發儲存元件集合係提供於複數個「反及」串中並 〃複數個子線通信,該至少一儲存元件與該複數個字線 中之一選定之字線通信,該程式化包括將一第一連串程 式化脈衝施加至該選定之字線,且隨後將一第二連串程 式化脈衝施加至該選定之字線; 在°亥第一連串程式化脈衝之每一程式化脈衝期間,根 2一第一升壓模式將電壓之一第一集合施加至該複數個 字線中之未選定之字線; 在該第一連申程式化脈衝之後,基於一升壓模式切換 標準而自該第一升壓模式切換至一第二升壓模式;及 在該第二連串程式化脈衝之每一程式化脈衝期間,根 據該第二升壓帛式將電壓<一第二集合施加至該等未選 定之予線,電壓之該第一集合至少部分地不同於電壓之 該第二集合。 2·如請求項丨之方法,其中: 該程式化包含將一脈衝串施加至該選定之字線,該脈 衝串包括該第一及該第二連串程式化脈衝,該切換在施 加《亥脈衝串中之一第一脈衝之後且在施加該脈衝串中之 一最末脈衝之前發生。 3.如請求項1之方法,其中: 该程式化包含將一脈衝串施加至該選定之字線,該脈 126206-1000812.doc 衝串包括該第-及該第二連串程式化脈衝,該升塵模式 切換標準係基於將該脈衝串中之具有一指定振幅之一程 式化脈衝施加至該選定之字線的時刻。 4. 如請求項1之方法,其中: ,該程式化包含將一脈衝串施加至該選定之字線,該脈 衝串匕括η亥第一及該第二連串程式化脈衝,該升壓模式 切換標準係基於已將該脈衝串巾之—指定數目之程式化 脈衝施加至該選定之字線的時刻。 5. 如請求項1之方法,其中: 該升壓模式切換標準係基於該選定之字線在該複數個 字線中之一位置。 6. 如請求項1之方法,其中: 該至少一儲存元件之一臨限電壓在該切換之前自一第 -位準增加至一第二位準且在該切換之後自該第二位準 增加至一第三位準。 7·如請求項1之方法,其中: 該程式化涉及在該切換之前的粗略程式化及在該切換 之後的精細程式化。 如請求項1之方法,其中: 該升壓模式切換標準係基於該非揮發儲存元件集合中 之至少—其他儲存元件達到一指定程式化條件的時刻\ 9.如請求項1之方法,其中: 一該升壓模式切換標準係基於由該非揮發儲存元件集合 經歷之程式化循環的一數目。 以 126206-1000812.doc 1357603 i〇.如請求項1之方法,其中: 該至少一儲存元件係提供於該複數個「反及」串之一 選定之「反及」串中,· 在肩第一升壓模式中,在該切換之前,一特定未選定 之字線接收-電壓,該電壓不會使在該特定未選定之字 線之一側上之一通道區與在該特定未選定之字線之另— 側上之一通道區隔離;及 在該第_升虔模式中,纟該切換之後,該特定未選定 之字線接收一電壓,該電壓會使在該特定未選定之字線 之該一側上之該通道區與該特定未選定之字線之該另一 側上之該通道區隔離。 11 ·如請求項1之方法,其中: 該至少一儲存元件係提供於該複數個「反及」串之— 選定之「反及」串中; 電壓升壓通道區之該第一集合與該複數個「反及」串 之未選定之「反及」串相關聯; 田由於電壓之該第一集合而將該通道區升壓時,施加 該第一連串程式化脈衝之每—程式化脈衝; 電壓升壓通道區之該第二集合與該複數個「反及」串 之未選定之「反及」串相關聯;及 田由於電壓之該第二集合而將該通道區升壓時,施加 該第二連串程式化脈衝之每一程式化脈衝。 12_ —種非揮發儲存系統,其包含: 提供於複數個「反及」串中之一非揮發儲存元件集 126206-1000812.doc 與非揮發儲存元件集合通信之複數個字線’至 存兀•件與該複數個字線中之一選定之字線通信;及 與該非揮發儲存元件集合通信之一或多個控制電路, 該一或多個控制電路程式化該至少一儲存元件,該程式 化包括將一第一連串程式化脈衝施加至該選定之字線, 且隨後將一第二連串程式化脈衝施加至該選定之字線, 在a亥第一連串程式化脈衝之每一程式化脈衝期間,該一 或多個控制電路根據一第一升壓模式將電壓之一第:集 合施加至該複數個字線中之未選定之字線,且在該第1 連串程式化脈衝之後,該—或多個控制電路基於—升壓 模式切換標準而自該第一升壓模式切換至一第二升壓模 式,且在該第二連串程式化脈衝之每一程式化脈衝期 間,將電壓之一第二集合施加至該等未選定之字線,電 壓之該第一集合至少部分地不同於電壓之該第二集合。 13. 如請求項12之非揮發儲存系統,其十: 。亥或夕個控制電路藉由將一脈衝串施加至該選定之 字線來程式化該至少一儲存元件,該脈衝串包括該第— 及該第二連串程式化脈衝,該切換在施加該脈衝串中之 一第一脈衝之後且在施加該脈衝季中之一最末脈衝之前 發生。 14. 如請求項12之非揮發儲存系統,其中: 忒一或多個控制電路藉由將—脈衝串施加至該選定之 字線來程式化該至少一儲存元件,該脈衝串包括該第— 126206-1000812.doc -4 - 1357603 及該第—連串程式化脈衝,該升壓模式切換標準係基於 將》亥脈衝_中之具有—指定振幅之_程式化脈衝施加至 該選定之字線的時刻。 15.如請求項12之非揮發儲存系統,其中: •Λ或夕個控制電路藉由將一脈衝串施加至該選定之 字線來程式化該至少-儲存元件,該脈衝串包括該第一 及。亥第—連串程式化脈衝,該升壓模式切換標準係基於Patent Application No. 096,141,503, the disclosure of which is incorporated herein by reference in its entirety in its entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire At least one storage element of the non-volatile storage element set, the non-volatile storage element set is provided in a plurality of "reverse" strings and is configured by a plurality of sub-line communication, and the at least one storage element and one of the plurality of word lines are selected Word line communication, the stylization comprising applying a first series of stylized pulses to the selected word line, and then applying a second series of stylized pulses to the selected word line; first series of stylized During each of the stylized pulses of the pulse, the root 2 - first boost mode applies a first set of voltages to the unselected word lines of the plurality of word lines; after the first consecutive programmed pulse, Switching from the first boost mode to a second boost mode based on a boost mode switching criterion; and during each of the stylized pulses of the second series of programmed pulses, The second voltage boosting silk formula < applied to a second set of such non-selected line to the first voltage set at least partially different from the second set voltage. 2. The method of claim 1, wherein: the stylizing comprises applying a burst to the selected word line, the burst comprising the first and second series of stylized pulses, the switching being applied One of the pulse trains occurs after the first pulse and before one of the last pulses in the pulse train is applied. 3. The method of claim 1, wherein: the stylizing comprises applying a burst to the selected word line, the pulse 126206-1000812.doc comprising the first and the second series of stylized pulses, The dusting mode switching criterion is based on the time at which a programmed pulse having one of the specified amplitudes is applied to the selected word line. 4. The method of claim 1, wherein: the stylizing comprises applying a burst to the selected word line, the pulse string comprising a first and a second series of stylized pulses, the boosting The mode switching criteria is based on the time at which the specified number of stylized pulses have been applied to the selected word line. 5. The method of claim 1, wherein: the boost mode switching criterion is based on a position of the selected word line at one of the plurality of word lines. 6. The method of claim 1, wherein: the threshold voltage of the at least one storage element increases from a first level to a second level before the switching and increases from the second level after the switching To the third level. 7. The method of claim 1, wherein: the stylization involves coarse stylization prior to the switching and fine stylization after the switching. The method of claim 1, wherein: the boost mode switching criterion is based on at least one of the non-volatile storage element sets - the time at which the other storage elements reach a specified stylization condition. 9. The method of claim 1, wherein: The boost mode switching criterion is based on a number of stylized cycles experienced by the set of non-volatile storage elements. The method of claim 1, wherein: the at least one storage component is provided in a selected "reverse" string of the plurality of "reverse" strings, In a boost mode, prior to the switching, a particular unselected word line receives a voltage that does not cause a channel region on one of the particular unselected word lines to be unselected at that particular The other of the word lines is isolated from one of the channel regions; and in the first _ 虔 mode, after the switching, the particular unselected word line receives a voltage that causes the particular unselected word The channel region on one side of the line is isolated from the channel region on the other side of the particular unselected word line. 11. The method of claim 1, wherein: the at least one storage component is provided in the plurality of "reverse" strings - the selected "reverse" string; the first set of voltage boost channel regions and the A plurality of "reverse" strings are associated with an unselected "reverse" string; when the field is boosted by the first set of voltages, each stylized pulse of the first series of stylized pulses is applied; The second set of voltage boost channel regions is associated with an unselected "reverse" string of the plurality of "reverse" strings; and when the field is boosted by the second set of voltages, Each of the stylized pulses of the second series of stylized pulses. 12_ - a non-volatile storage system comprising: a set of non-volatile storage elements provided in a plurality of "reverse" strings 126206-1000812.doc a plurality of word lines 'communicating with a collection of non-volatile storage elements' Communicating with a selected one of the plurality of word lines; and communicating with the non-volatile storage element set of one or more control circuits, the one or more control circuits stylizing the at least one storage element, the stylized The method includes applying a first series of stylized pulses to the selected word line, and then applying a second series of stylized pulses to the selected word line during each stylized pulse of the first series of programmed pulses And the one or more control circuits apply a set of voltages to the unselected word lines of the plurality of word lines according to a first boost mode, and after the first series of stylized pulses, the one or more control circuits - or a plurality of control circuits switching from the first boost mode to a second boost mode based on the -boost mode switching criterion, and during each of the stylized pulses of the second series of programmed pulses, The second set of pressure applied to one of these unselected word lines, the set pressure of the first electrically least partially different from the second set voltage. 13. For the non-volatile storage system of claim 12, its ten: a control circuit for programming the at least one storage element by applying a burst to the selected word line, the pulse train including the first and the second series of stylized pulses, the switching is applying One of the first pulses of the pulse train occurs and before one of the last pulses of the pulse season is applied. 14. The non-volatile storage system of claim 12, wherein: one or more control circuits program the at least one storage element by applying a pulse train to the selected word line, the pulse train including the first 126206-1000812.doc -4 - 1357603 and the first series of stylized pulses, the boost mode switching criterion is based on applying a stylized pulse having a specified amplitude to the selected word line Moment. 15. The non-volatile storage system of claim 12, wherein: ??? or a control circuit to program the at least one storage element by applying a burst to the selected word line, the pulse train comprising the first and. Haidi - a series of stylized pulses based on the boost mode switching criteria 已將該脈衝串中之一指定數目之程式化脈衝施加至該選 定之字線的時刻。 16_如請求項12之非揮發儲存系統,其中: 該升壓模❹純料基於該敎之字縣該複數個 字線中之一位置。 17.如請求項12之非揮發儲存系統,其中: 該至少一儲存元件之一臨限電壓在該切換之前自一第 -位準增加至一第二位準且在該切換之後自該第二位準 增加至一第三位準。 18‘如請求項12之非揮發儲存系統,其中: 杨式化涉及在該切換之前的粗略程式化及在該切換 之後的精細程式化。 ' 19. 如請求項12之非揮發儲存系統,其中·· 該升壓模式切換標準係基於該非揮發儲存元件集合中 之至少-其他儲存元件達到—指定程式化條件的時刻。 20. 如請求項12之非揮發儲存系統,其中·· δ亥升塵模式切換標準係基於由該非揮發儲存元件集合 126206·丨 0008 丨 2.d〇c 經歷之程式化循環的一數目。 21.如請求項12之非揮發儲存系統,其中: 該至少-儲存元件係提供於該複數個「反及」一 選定之「反及」串中; 亥第升屋模式中,在該切換之前,-特定未選定 之字線接收H該電壓不會使在該特定未選定之字 線之一側上之一通道區與在該特定未Μ定之字線之另一 側上之一通道區隔離;及 2該第二升壓模式中,在該切換之後,該㈣未選定 之予線接收-電壓,該電壓會使在該特^未選定之字線 之該一側上之該通道區與該特定未敎之字線之該另L 側上之該通道區隔離。 22.如請求項12之非揮發儲存系統,其中: 該至少一儲存元件係提供於該複數個「反及」串、一 選·定·之「反及」串中; 電壓升壓通道區之該第一集合與該複數個「反及」串 之未選定之「反及」串相關聯; 當由於電壓之該第一集合而將該通道區升壓時,施加 忒第一連串程式化脈衝之每一程式化脈衝; 電壓升壓通道區之該第二集合與該複數個「反及」串 之未選定之「反及」串相關聯;及 當由於電壓之該第二集合而將該通道區升壓時,施加 s亥第二連_程式化脈衝之每一程式化脈衝。 126206-1000812.doc • 6 ·A time has been applied to the selected word line by a specified number of stylized pulses in the burst. 16_ The non-volatile storage system of claim 12, wherein: the boost mode is based on one of the plurality of word lines of the word count. 17. The non-volatile storage system of claim 12, wherein: the threshold voltage of the at least one storage component increases from a first level to a second level prior to the switching and from the second after the switching The level is increased to a third level. 18 'A non-volatile storage system as claimed in claim 12, wherein: the simplification involves a rough stylization prior to the switch and a fine stylization after the switch. 19. The non-volatile storage system of claim 12, wherein the boost mode switching criterion is based on at least the other of the non-volatile storage element sets - the time at which the stylized condition is specified. 20. The non-volatile storage system of claim 12, wherein the delta dust mode switching criterion is based on a number of stylized cycles experienced by the non-volatile storage element set 126206 · 0008 0008 丨 2.d〇c. 21. The non-volatile storage system of claim 12, wherein: the at least-storage component is provided in the plurality of "reverse" and selected "reverse" strings; in the Haishengsheng mode, before the switching, - a particular unselected word line receiving H that does not isolate one of the channel regions on one side of the particular unselected word line from one of the channel regions on the other side of the particular undefined word line; And 2 in the second boost mode, after the switching, the (4) unselected pre-wire receives a voltage, the voltage causing the channel region on the side of the unselected word line The channel region on the other L side of the particular unexamined word line is isolated. 22. The non-volatile storage system of claim 12, wherein: the at least one storage component is provided in the "reverse" string of the plurality of "reverse" strings, and the selection and determination of the "reverse" string; The first set is associated with an unselected "reverse" string of the plurality of "reverse" strings; when the channel region is boosted due to the first set of voltages, the first series of stylized pulses are applied Each of the stylized pulses; the second set of voltage boost channel regions being associated with an unselected "reverse" string of the plurality of "reverse" strings; and when the channel is due to the second set of voltages When the zone is boosted, each stylized pulse of the second consecutive _stylized pulse is applied. 126206-1000812.doc • 6 ·
TW096141503A 2006-11-02 2007-11-02 Reducing program disturb in non-volatile memory us TWI357603B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US11/555,850 US7440323B2 (en) 2006-11-02 2006-11-02 Reducing program disturb in non-volatile memory using multiple boosting modes
US11/555,856 US7468911B2 (en) 2006-11-02 2006-11-02 Non-volatile memory using multiple boosting modes for reduced program disturb

Publications (2)

Publication Number Publication Date
TW200836203A TW200836203A (en) 2008-09-01
TWI357603B true TWI357603B (en) 2012-02-01

Family

ID=39365235

Family Applications (1)

Application Number Title Priority Date Filing Date
TW096141503A TWI357603B (en) 2006-11-02 2007-11-02 Reducing program disturb in non-volatile memory us

Country Status (4)

Country Link
JP (1) JP4950299B2 (en)
KR (1) KR101012133B1 (en)
TW (1) TWI357603B (en)
WO (1) WO2008057927A2 (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013254537A (en) * 2012-06-06 2013-12-19 Toshiba Corp Semiconductor memory and controller
KR102670210B1 (en) * 2019-10-23 2024-05-28 양쯔 메모리 테크놀로지스 씨오., 엘티디. How to program memory devices and related memory devices

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5917757A (en) * 1996-08-01 1999-06-29 Aplus Flash Technology, Inc. Flash memory with high speed erasing structure using thin oxide semiconductor devices
JP3610691B2 (en) * 1996-09-09 2005-01-19 ソニー株式会社 Semiconductor nonvolatile memory device
KR100502412B1 (en) * 2002-10-23 2005-07-19 삼성전자주식회사 Non-volatile semiconductor memory device and program method thereof
JP3863485B2 (en) * 2002-11-29 2006-12-27 株式会社東芝 Nonvolatile semiconductor memory device
US7023739B2 (en) * 2003-12-05 2006-04-04 Matrix Semiconductor, Inc. NAND memory array incorporating multiple write pulse programming of individual memory cells and method for operation of same
US7212435B2 (en) * 2004-06-30 2007-05-01 Micron Technology, Inc. Minimizing adjacent wordline disturb in a memory device
JP5452851B2 (en) * 2006-08-10 2014-03-26 三星電子株式会社 Memory device selectively using self-boost program operation and program method thereof

Also Published As

Publication number Publication date
WO2008057927A2 (en) 2008-05-15
WO2008057927A3 (en) 2008-07-24
KR101012133B1 (en) 2011-02-07
JP4950299B2 (en) 2012-06-13
KR20090089348A (en) 2009-08-21
JP2010509701A (en) 2010-03-25
TW200836203A (en) 2008-09-01

Similar Documents

Publication Publication Date Title
TWI416527B (en) Method for operating non-volatile storage and non-volatile storage device
TWI402856B (en) Enhanced bit-line pre-charge scheme for increasing channel boosting in non-volatile storage
TWI424436B (en) Compensating for coupling during read operations in non-volatile storage
EP2458592B1 (en) Multiple pass write sequence for non-volatile storage
TWI494933B (en) Forecasting program disturb in memory by detecting natural threshold voltage distribution
EP2577671B1 (en) Programming non-volatile storage with synchronized coupling
US7468911B2 (en) Non-volatile memory using multiple boosting modes for reduced program disturb
US7440323B2 (en) Reducing program disturb in non-volatile memory using multiple boosting modes
JP4808784B2 (en) Method for programming non-volatile memory using improved pass voltage to reduce program disturb
EP2556509B1 (en) Saw-shaped multi-pulse programming for program noise reduction in memory
EP2140457B1 (en) Reducing power consumption during read operations in non-volatile storage
JP5174829B2 (en) Reading of non-volatile memory cell in consideration of storage state of adjacent memory cell
TW201324513A (en) Substrate bias during program of non-volatile storage
JP2009514138A (en) How to program multi-state non-volatile memory using smart verification
JP4726958B2 (en) Last-fast mode and method for programming NAND-type non-volatile memory with reduced program disturbance
TWI386944B (en) Non-volatile storage with boosting using channel isolation switching and method thereof
TWI357603B (en) Reducing program disturb in non-volatile memory us
TWI352404B (en) Providing local boosting control implant for non-v

Legal Events

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