[go: up one dir, main page]

TWI309875B - Non-voltaile memory cells, memory arrays including the same and methods of operating cells and arrays - Google Patents

Non-voltaile memory cells, memory arrays including the same and methods of operating cells and arrays Download PDF

Info

Publication number
TWI309875B
TWI309875B TW95100248A TW95100248A TWI309875B TW I309875 B TWI309875 B TW I309875B TW 95100248 A TW95100248 A TW 95100248A TW 95100248 A TW95100248 A TW 95100248A TW I309875 B TWI309875 B TW I309875B
Authority
TW
Taiwan
Prior art keywords
layer
memory
gate
array
erase
Prior art date
Application number
TW95100248A
Other languages
Chinese (zh)
Other versions
TW200625550A (en
Inventor
Hang Ting Lue
Sheng Chih Lai
Original Assignee
Macronix Int Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Macronix Int Co Ltd filed Critical Macronix Int Co Ltd
Publication of TW200625550A publication Critical patent/TW200625550A/en
Application granted granted Critical
Publication of TWI309875B publication Critical patent/TWI309875B/en

Links

Landscapes

  • Non-Volatile Memory (AREA)
  • Semiconductor Memories (AREA)

Description

1309875 九、發明說明: 【發明所屬之技術領域】 本申請案係根據且在35U_S.C.§119(e)條款下主張: 2005年1月3日申請之美國專利臨時申請案第60/640,229 號;2005年1月27日申請之美國專利臨時申請案第 60/647,012號;2005年6月10日申請之美國專利臨時申請 案第60/689,231號;及2005年6月10日申請之美國專利 臨時申請案第60/689,314號之優先權,上述各專利之全部 内容在此以引用方式全數併入。 【先前技術】 非揮發性記憶體(NVM)指即使當自含有NVM單元之 几件移走電力供應時亦能持續儲存資訊之半導體記憶體。 NVM包括遮罩唯讀記憶體(Mask R〇M)、可程式化唯讀記 憶體(PRGM)、可抹除可程仏唯讀記龍(EpRQM)、電可 抹除可程式化唯讀記憶體(EEPROM)、及快閃記憶體。非揮 發=記憶體係歧地使用於半導體產#且餘發展以防止 拍二弋化貝料知失之一類記憶體。通常非揮發性記憶體可 終端使用者f求加以程式化、讀取及/或抹除, 且该已程式化的資料可儲存達—段長時間。 單元的’ if發性記憶元件可具有各種設計。NVM 化物=:所? .S,-氧化物-氮化物-氧 穿隨抹除操作。雖然此;以提供電同直接 681939-27U3 1309875 態期間可能存在之低電場強度下亦會發生直接穿隧。 另一 NVM設計係NROM(氮化唯讀記憶體),其使用較 厚的隧道氧化層以在保持狀態期間防止電荷損失。然而, 較厚之隧道氧化層可能影響通道抹除速率。結果,能帶間 穿隧熱電洞(BTBTHH)抹除方法可用來注入電洞陷阱以補 償電子。然而,BTBTHH抹除方法可能產生一些可靠性問 題。例如,利用BTBTHH抹除方法之NROM元件的特徵 可能在多次P/E(程式化/抹除)循環後退化。 因此,在此項技術中存在對以改進的資料保持效能及 增加操作速率來操作多次(程式化/抹除/讀取)之非揮發性 記憶單元設計及陣列的需要。 【發明内容】 本發明關於非揮發性記憶元件,且更明確言之係關於 包括一隧道介電結構的非揮發性記憶元件,其促進自收斂 抹除操作,同時亦在保持狀態期間維持記憶元件之電荷儲 存層中的電荷保持。 本發明的一具體實施例包括記憶單元,其包含:一半 導體基體,其具有設置於該基體之一表面下且由一通道區 分離的一源極區及一汲極區;一隧道介電結構,其係設置 於該通道區上,該隧道介電結構包含具有一小電洞穿隧阻 障高度之至少一層;一電荷儲存層,其係設置於該隧道介 電結構上;一絕緣層,其係設置於該電荷儲存層上;及一 閘極電極,其係設置於該絕緣層上。 本發明另一具體實施例包含記憶單元,其包含:一半 681939-27U3 7 1309875 導體基體 穴开负罝於孩丞籐 < —表面下且由一、$、若广 離的一源極區及-汲極區;-多層隧道介電結二:f 置於a亥通道區上,該多層隧道介電結構包含I W、a 穿隨阻障高度之至少一層;一電荷儲存層,; 多層隨道介電結構上;—絕緣層,其係設置於= 層上,及一閘極電極,其係設置於該絕緣層上。 料 在一些較佳具體實施例中,設置有一小電洞穿隧阻障 问度的層可含有諸如氮化矽⑼办4)或氧化給(Hf〇2)的材 。在本發明一些較佳具體實施例中’該等記憶單元包括 一具有多層之隧道介電結構,例如氧化矽、氮化矽及氧化 矽(ΟΝΟ)的一堆疊介電質三層結構。此等隧道介電結構提 供一 S0N0N0S(石夕-氧化物-氮化物-氧化物—化物-氧化物-矽)或超晶格S0N0N0S設計。 在本發明一些較佳具體實施例中,該隧道介電結構可 包含至少二介電層,各層具有至高達約4奈米之厚度。此 外,在本發明一些較佳具體實施例中,該閘極電極包含一 功函數值大於Ν+多晶矽之材料。 在一些較佳具體實施例中,該隨道介電結構可包括一 層包含具有一小電洞穿隧阻障高度之材料’其中該材料係 以濃度梯度出現在該層中,以致該材料的濃度在該層内之 一深點處係最大值。 本發明亦包括非揮發性記憶元件,其包含依據在此所 述一或多個具體實施例之複數個記憶單元(即一陣列)。如 在此所用’「複數個」指二個或二個以上。依據本發明的記 681939-27U3 8 1309875 顯現明顯改進之操作性f,包括增加抹除速率、改 進電何保持及更大的操作窗口。 本發明亦包括操作非揮發性記憶單元及 ::::明的操作方法包括藉由應用自收心 牛之vt分布緊湊而重設該記憶元件;藉由通道刑注入 〇、化邊等讀兀件至少其_ ;及藉由施加—在該等記惊 :件至少其一之抹除狀態位準和程式化狀態位準間二 ^以讀取該等記憶元件至少其一。如在此所用準名㈡ 二」係#曰使在-陣列之許多記憶單元中的臨限電壓分布變 乍。一般而言,臨限電壓分布「緊湊」係其中若 =電壓彼此在一狹窄範圍内,以致該陣列的操作比習知 二改進。例如’在一些較佳具體實施例中’如在包含依 發明之-或多個具體實施例中的記憶單元之陣。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 US Patent Provisional Application No. 60/647,012, filed on Jan. 27, 2005, and U.S. Patent Application Serial No. 60/689,231, filed on June 10, 2005; Priority of the Patent Provisional Application No. 60/689,314, the entire disclosure of each of which is incorporated herein by reference. [Prior Art] Non-volatile memory (NVM) refers to a semiconductor memory that continuously stores information even when a power supply is removed from a few pieces of the NVM unit. NVM includes masked read-only memory (Mask R〇M), programmable read-only memory (PRGM), erasable programmable read-only memory (EpRQM), and electrically erasable programmable read-only memory Body (EEPROM), and flash memory. Non-swing = memory system is used in semiconductor production # and the rest is developed to prevent the loss of a type of memory. Typically, non-volatile memory can be programmed, read, and/or erased by the end user, and the stylized data can be stored for a long period of time. The unit's if-like memory element can have a variety of designs. NVM compound =: ??, - oxide-nitride-oxygen wear and erase operation. Although this; direct tunneling also occurs at low electric field strengths that may exist during the period of providing the same 681939-27U3 1309875 state. Another NVM design is NROM (nitrided read-only memory), which uses a thicker tunnel oxide layer to prevent charge loss during the hold state. However, a thicker tunnel oxide layer may affect the channel erase rate. As a result, an interband tunneling thermal tunnel (BTBTHH) erasing method can be used to inject a hole trap to compensate for electrons. However, the BTBTHH erasing method may cause some reliability problems. For example, features of NROM components utilizing the BTBTHH erase method may degrade after multiple P/E (stylization/erasing) cycles. Therefore, there is a need in the art for non-volatile memory cell designs and arrays that operate multiple times (stylized/erased/read) with improved data retention and increased operating rates. SUMMARY OF THE INVENTION The present invention is directed to non-volatile memory elements, and more particularly to non-volatile memory elements including a tunnel dielectric structure that facilitate self-convergence erase operations while also maintaining memory elements during the hold state The charge in the charge storage layer remains. A specific embodiment of the present invention includes a memory unit including: a semiconductor substrate having a source region and a drain region disposed under a surface of the substrate and separated by a channel region; a tunnel dielectric structure Provided on the channel region, the tunnel dielectric structure includes at least one layer having a small hole tunneling barrier height; a charge storage layer disposed on the tunnel dielectric structure; an insulating layer The device is disposed on the charge storage layer; and a gate electrode is disposed on the insulating layer. Another embodiment of the present invention comprises a memory unit comprising: a half of a 681939-27U3 7 1309875 conductor base hole open to the 丞 丞 — — — — — — — — — — — — — 且 且 且 且 且 且 且 且 且 且 且 且- bungee region; - multilayer tunnel dielectric junction 2: f placed on the ahai channel region, the multilayer tunnel dielectric structure comprising IW, a at least one layer of the barrier height; a charge storage layer; a dielectric structure; an insulating layer disposed on the = layer and a gate electrode disposed on the insulating layer. In some preferred embodiments, the layer provided with a small hole tunneling barrier may contain materials such as tantalum nitride (9) or oxidized (Hf2). In some preferred embodiments of the invention, the memory cells comprise a stacked dielectric dielectric structure having a plurality of layers of tunnel dielectric structures, such as hafnium oxide, tantalum nitride, and hafnium oxide. These tunnel dielectric structures provide a S0N0N0S (Shixi-Oxide-Nitride-Oxide-Oxide-Oxide-Y) or Superlattice S0N0N0S design. In some preferred embodiments of the invention, the tunnel dielectric structure can comprise at least two dielectric layers, each layer having a thickness of up to about 4 nanometers. Moreover, in some preferred embodiments of the invention, the gate electrode comprises a material having a work function value greater than Ν + polysilicon. In some preferred embodiments, the interposable dielectric structure can include a layer of material comprising a small hole tunneling barrier height wherein the material is present in the layer with a concentration gradient such that the concentration of the material is One of the deep points in the layer is the maximum. The invention also includes a non-volatile memory element comprising a plurality of memory cells (i.e., an array) in accordance with one or more embodiments described herein. As used herein, "plural" means two or more. The 681939-27U3 8 1309875 in accordance with the present invention exhibits significantly improved operability f, including increased erase rate, improved hold, and larger operational window. The invention also includes operating the non-volatile memory unit and:::: The method of operation includes resetting the memory element by applying a compact vt distribution of the self-contained cow; and reading the channel by means of channel injection And at least one of the memory elements is read by at least one of the erased state level and the stylized state level. As used herein, the quasi-name (2) 2" system causes the threshold voltage distribution in many memory cells of the array to be degraded. In general, the threshold voltage distribution is "compact" in which if the voltages are within a narrow range of each other, the operation of the array is improved over the conventional one. For example, 'in some preferred embodiments', as in the array comprising memory cells according to the invention or in the specific embodiments

It’緊凑」之臨限電壓分布指示各種記憶單元的臨限電 塗彼此係在G.5Vft_。在其他使用依據本發明之記憶單 疋的陣列架構中’該「緊湊」臨限電壓分布可具有從上限 到下限約1.0V的範圍。 a 依據本發明之—操作方法的具體實施例包括操作依據 發月之陣列’其係藉由施力口自收斂重設/抹除電壓至欲 ,設/抹除之各記憶單元中的基體及閘極電極;程式化該複 數個記憶單元至少其—;由施加—在料記憶元件中 f少其一之抹除狀態位準和程式化狀態位準間的電壓,以 讀取該複數個記憶單元中至少其一。 本發明亦包括形成一記憶單元之方法,其包含:提供 681939-27U3 1309875 半‘體基體’其具有形成於該基體之—表面下且由—通 =區分離的—源極區及—汲極區;形成—随道介電結構在 =通道區上,其巾形成該騎介電結構包含形成至少二介 电層’其:該至少二介電層其一層具有一比該至少二介電 匕另層還小之電洞穿隨阻障高度;在該隧道介電結構上 元成電荷儲存層;在該電荷儲存層上形成-絕緣層;及 在該絕緣層上形成一閘極電極。 如,此所用’片語「小電洞穿隨阻障高度」一般指係 ;或專於一氧化石夕之近似電洞穿隨阻障高度之值。尤其 了小電洞穿隨阻障高度最好係小於或等於約4.5eV。更佳的 係一小電洞穿隧阻障高度係小於或等於約i 9eV。 【實施方式】 現將詳細參照本發明及其較佳具體實施例,其實例圖 解於附圖之中。若可能的話,所有圖式中將以相同或類似 元件符旒來代表相同或類似的部件。應注意的係非圖形之 • 繪圖係依大幅簡化之形式並且不按照精確之比例。關於在 此所揭,單純為了方便及清楚目的,方向性名詞(諸如頂 部、底部、左、右、上、下、以上、以下、位於下方、後 及月ίι)係針對附圖使用。併同附圖之以下說明所使用之此等 -方向性名詞不應被視為以任何未在隨附申請專利範圍中明 顯提出之方式限制本發明。雖然在此所揭參考一些示範性 具體實施例,應瞭解此等具體實施例係舉例說明且非限 制。應理解在此所揭之過程步驟及結構不涵蓋用於製造整 個積體電路之完整流程。本發明可與此項技術中為人熟知 681939*27U3 10 1309875 之各種積體電路製造技術—起實現或發展。 依據本發明的§己憶單元可克服在S〇N〇s及nr〇m元 =中之士 Α可菲性問題。例如,依據本發明之記憶單元結 L:允許^ ™通道抹除方法,同時保持良好電荷保持 BTBTHH抹除方法之具體實施例亦可減輕對 依賴攸而避免在多次P/E循環後元 仵之退化。 具二為可在—其中隧道介電結構係-多層結構之 人i、㈣、’使超薄隨道介電質或―超薄氧化層結 二二2阻障高度層。此可提供更好的應力免除。在 =量退後’根據本發明之非揮發性記憶單元亦顯 ,據本=的記憶單元可使用n通道或p通道設計, 通L二::示。圖1“苗述本發明-具體實施例之η : 斷面圖。該記憶單元包括—含有至少 :型=區102和104…基體ι〇ι,其中各推雜: 如圖3 能可_龍加之電壓㈣祕或汲極。 為參考目的,摻雜區102可作為源極 雜(he 104可作為汲極。其 竹 包括-通道區106。在通 =在:方= 參雜區間進-步 孫d 遇逼區106上方(在基體101表面上、 係-隧道介電結構120 :面上) 介電結構12。可包含三層薄構:;;:ΐ道 阻障高度氮化層m係爽置在一下方薄氧二;2= 薄氣化層126間。記憶單元⑽進-步包括,道介= 681939-27U3 11 1309875 結構120上之電荷陷獲(或電荷儲存)層130(較佳係氮化 物),且一絕緣層140(較佳係包含阻隔氧化物)設置在電荷 陷獲層130上。一閘極150係設置在絕緣層140上。 圖lb描述依據本發明一具體實施例的p通道記憶單元 200之斷面圖。該記憶單元包括一含有至少二p型摻雜區 202和204的η型基體201,其中各摻雜區202和204之功 能可為源極或汲極。基體201在二ρ型摻雜區間進一步包 括一通道區206。ρ通道記憶單元200同樣地包括一包含三 層薄ΟΝΟ結構之隧道介電結構220(其中一小電洞穿隧阻 障高度氮化層224係夾置在一下方薄氧化層222及上方薄 氧化層226之間)、一電荷陷獲(或電荷儲存)層230、一絕 緣層240及一閘極250。 因此,例如在圖1 a及lb中所述,依據本發明的記憶 單元可包括:一多層薄膜隧道介電結構,其包括一第一氧 化梦層01、一第一氮化梦層N1及一第二氧化珍層02 ; — 電荷儲存層,例如一第二氮化石夕層N2 ;及一例如第三氧化 矽層03之絕緣層,其係在一如半導體基體(例如矽基體) 之基體上或上方。穿隧介電結構允許電洞在記憶元件抹除/ 重設操作期間自基體穿隧到電荷儲存層。較佳的係,在本 發明之一非揮發性記憶單元中的隧道介電結構具有可忽略 之電荷陷獲效率,且更佳的係在記憶體操作期間完全不捕 獲電荷。 諸如氮化矽層、Hf02和Al2〇3之電荷儲存材料可用作 隧道介電結構中之小電洞穿隧阻障高度層。在本發明一些 681939-27U3 12 1309875 較佳具體實施例中,諸如氮化 作記憶元件中之電荷儲存層。防止儲存材料可用 可用作絕緣層’例如第三氧切心3°了 阻隔氧化物 單元在絕緣層上亦包括1極或,據本發明的記憶 極。隧道介電結構、電荷儲存岸“極,例如多晶矽閘 基體上至少一通道區及閘極可形成在 極區界U係設置在·。 其係由—源極區及-汲 電4據in各種具體實施例之記憶單元包含-隧道介The threshold voltage distribution of It's compact indicates that the threshold coatings of the various memory cells are tied to each other at G.5Vft_. In other array architectures using memory cartridges in accordance with the present invention, the "compact" threshold voltage distribution can have a range from about an upper limit to a lower limit of about 1.0 volts. a specific embodiment of the method of operation according to the present invention includes an operation according to the array of the moons, which is to reset/erase the voltage by applying a self-convergence to the substrate, and to set/erase the substrate in each memory unit and a gate electrode; stylizing the plurality of memory cells at least - the voltage between the erased state level and the stylized state level is applied by the application-in-memory memory element to read the plurality of memories At least one of the units. The invention also includes a method of forming a memory cell, comprising: providing a 681939-27U3 1309875 semi-body substrate having a source region and a drain region formed under the surface of the substrate and separated by a - pass region Forming a channel-forming dielectric structure on the =channel region, the towel forming the saddle dielectric structure comprising forming at least two dielectric layers': the at least two dielectric layers have a layer having a ratio of at least two dielectric layers The other layer also has a small hole through the barrier height; a charge storage layer is formed on the tunnel dielectric structure; an insulating layer is formed on the charge storage layer; and a gate electrode is formed on the insulating layer. For example, the phrase "small hole penetration with barrier height" is generally used to refer to the system; or it is specific to the value of the barrier height of the oxidized stone. In particular, the small hole penetration height is preferably less than or equal to about 4.5 eV. More preferably, the small hole tunneling barrier height is less than or equal to about i 9 eV. [Embodiment] The present invention and its preferred embodiments will now be described in detail, and examples thereof are illustrated in the accompanying drawings. Wherever possible, the same or similar components will be referred to in the drawings. It should be noted that the graphics are in a greatly simplified form and are not in exact proportions. With respect to this disclosure, directional nouns (such as top, bottom, left, right, top, bottom, above, below, below, after, and month) are used for the sake of convenience and clarity. The directional terminology used in the following description with reference to the accompanying drawings is not to be construed as limiting the invention in any manner that is not explicitly described in the appended claims. Although some exemplary embodiments are described herein, it is to be understood that the specific embodiments are illustrative and not limiting. It should be understood that the process steps and structures disclosed herein do not cover the complete process for fabricating the entire integrated circuit. The present invention can be implemented or developed in conjunction with various integrated circuit fabrication techniques well known in the art as 681939*27U3 10 1309875. The § memory unit according to the present invention can overcome the problem of the 之 Α 性 性 in the S〇N〇s and nr〇m elements. For example, the memory cell junction L according to the present invention allows the TM channel erasing method while maintaining a good charge retention. The specific embodiment of the BTBTHH erasing method can also alleviate the dependency and avoid the multi-P/E cycle. Degradation. The second is a barrier layer in which the tunnel dielectric structure-multilayer structure of the person i, (4), 'Ultra-thin channel dielectric or 'ultra-thin oxide layer junction 222. This provides better stress relief. After the = amount retreat, the non-volatile memory unit according to the present invention is also displayed. According to the memory unit of the present = n-channel or p-channel design can be used, and L2:: is shown. Figure 1 "Miao's invention - a specific embodiment of η: a cross-sectional view. The memory unit includes - contains at least: type = area 102 and 104... base ι〇ι, wherein each of the pushes: as shown in Figure 3 In addition, the voltage (4) is secret or bungee. For reference purposes, the doping region 102 can serve as a source impurity (he 104 can serve as a drain. The bamboo includes a channel region 106. In the pass = in: square = doping interval - step The dielectric structure 12 can be contained above the surface of the substrate 101 (on the surface of the substrate 101, the system-channel dielectric structure 120: surface). It can comprise three layers of thin structure:;;: barrier high barrier nitride layer m system Cooling under a thin oxygen dioxide; 2 = thin gasification layer 126. Memory unit (10) further includes: dao = 681939-27U3 11 1309875 structure 120 on the charge trapping (or charge storage) layer 130 (more Preferably, an insulating layer 140 (preferably comprising a barrier oxide) is disposed on the charge trapping layer 130. A gate 150 is disposed over the insulating layer 140. Figure lb depicts an embodiment in accordance with the present invention A cross-sectional view of a p-channel memory cell 200. The memory cell includes an n-type substrate 201 having at least two p-type doped regions 202 and 204, each of which The doping regions 202 and 204 may function as a source or a drain. The substrate 201 further includes a channel region 206 in the two p-type doping region. The p-channel memory cell 200 also includes a tunneling layer including a three-layer thin germanium structure. The electrical structure 220 (where a small hole tunneling barrier nitride layer 224 is sandwiched between a lower thin oxide layer 222 and the upper thin oxide layer 226), a charge trapping (or charge storage) layer 230, The insulating layer 240 and a gate 250. Therefore, as described in, for example, FIGS. 1a and 1b, the memory unit according to the present invention may include: a multilayer film tunnel dielectric structure including a first oxide layer 01, a first nitride layer N1 and a second oxide layer 02; a charge storage layer, such as a second nitride layer N2; and an insulating layer such as a third tantalum layer 03, On or above the substrate of the semiconductor body (e.g., germanium substrate). The tunneling dielectric structure allows holes to tunnel from the substrate to the charge storage layer during memory device erase/reset operations. Preferably, one of the present inventions The tunnel dielectric structure in the non-volatile memory unit has Neglected charge trapping efficiency, and more preferably no charge is trapped during memory operation. Charge storage materials such as tantalum nitride layer, HfO 2 and Al 2 〇 3 can be used as small hole tunneling in tunnel dielectric structures A barrier level layer. In some preferred embodiments of the invention 681939-27 U3 12 1309875, such as nitriding as a charge storage layer in a memory element. Preventing a storage material from being used as an insulating layer 'eg, a third oxygen cut 3 The barrier oxide unit also includes a pole or a memory electrode according to the present invention on the insulating layer. The tunnel dielectric structure and the charge storage bank "pole", for example, at least one channel region and the gate electrode on the polysilicon gate base can be formed in the pole region U system is disposed in the system. The source region and the source region are in various The memory unit of the specific embodiment includes - tunneling

10^ΐΓΓ "J"'2〇V ^^a(Vg)TT ™抹除速率。另—方面,仍可維持 並;在一些範例中,可能比許多習知s_ 電:同抹Γ ^發明的錢單元亦可避免使用能帶間熱 ί==:般係用於概0M元件中。避免此能帶 :避可大幅地免除熱電洞引入損害,且所以 此避免係付合需求的。 參考圖21於依據本發明—㈣實關㈣道介電 、、、口構之祕電壓的實驗測量值,顯示-超薄01/Ν1/〇2錄 構可具有—可忽略的_效率,如錢續程式化脈衝下之 不變臨限電壓位準所證。在針對目2測試的範例中, 〇1/Ν1/〇2層厚度分別為3G、3〇及35埃(Α)。如圖2顯示, 在使用程式化之各種方法(即_FN程式化、爛程式化及 CHE(通道熱電子)程式化)於轩m歧㈣程中,臨 限電屢Vt維持穩定在近㈨19伏特。因此,此一超薄 01/N1/02膜可作為-調變隧道介電結構。在包括cm、 681939-27U3 1309875 +FN及-FN之各種電荷注入方法下的結果皆顯示可忽視的 電% 獲。製程或元件結構可加以設計以使介面性陷牌減 到最少,以致01/N1或N1/02介面係有作用。 圖3顯示依據本發明一具體實施例具有s〇N〇N〇s設 计的記憶單元的抹除特徵。圖3所述之具體實施例中的記 憶單元包含—厚度分別為15埃、20埃及18埃之ΟΝΟ隧 C二電、、、。構的n_MOSFET設計。此具體實施例之記憶單元 包含一厚度約70埃之氮化矽電荷儲存層、一厚度約90埃 之絕、、彖氧化石夕層、及一包含任何合適導電材料之閘極,例 如=¾•摻雜多晶石夕。參考圖3,可達到快速fn抹除(如在 10笔移内)’且亦可獲得—極佳的自收斂抹除性質。 圖4顯不根據參考圖3所述之本發明記憶單元的具體 Λ施例之S0N0N0S元件的電荷保持特徵。如圖示,該等 = & S_s元件更佳,且就電流值 可 能高好多個等級。 賴示㈣含有至少—狀隧道介電結構 的應之能帶圖,其中該至少一層具有一小電洞穿随 阻障在一記憶體資料保持期間可能存在之低電場下 的隧道介電結構(此範例中的 MKFl U 1/〇2三層)之能帶圖,係 Ί a中。可除去如由點狀箭頭表示在低 好n?的阻n處帶偏移圖5b中顯示)可減少N1 及⑺的一阻障效應’使得通過〇1之直接穿隨可能發生。旦 有至^電科咖障高度層之騎介電賴可允料 6S1939-27U3 14 1309875 效FN抹除操作。 圖5c及5d顯示在一範例中之另一組能帶圖。對於一 範例中之較佳能帶偏移條件,N1的厚度可能大於〇ι。價 能帶之能帶圖係在相同之電場E_l4Mv/em處繪出。根 據WKB近似之穿隨可能性係與陰影區域相關連。在此範 例中,對於厚度Nl=()1,能帶偏移不完全賴02的阻障。 另-方面,對於N1>01,能帶偏移可較易於遮擋⑺。因此, 對於厚度中Ν1>〇1,在⑴巾相同電場下,電洞穿隨電流 可能較大。 一具有經測量及模擬電洞穿隨電流的實驗(如圖6顯示) 進-步描述根據本發明-些具體實施例通過随道介電結構 之電时隧。例如,通過01/m/〇2介電質的電洞穿随電 流可落在一超薄氧化物及一厚氧化物間。在一範例中,在 高電場下,電洞穿隨電流可近似超薄氧化物。然而,在低 電場下,直接穿随可受抑制。如圖6顯示,即使在僅mv/cm # ㈣電場強度下’電洞雜電流亦可透過―薄氧化層偵測 到。電洞穿隧電流在例如n_13MV/cm之相對較地高電場 強度下可透過-厚氧化物忽略。然而,當高電場強度出現 時,通過- ΟΝΟ隨道介電結構的電洞穿隨電流會到達一薄 :氧化層。在圖6中,由於在低電場電洞穿隧通過一薄氧化 物造成之大電流洩漏可在圖中的區域Α看見。在圖6中, 在高電場強度處通過-q疆/〇2㈣介t結構的電洞穿 隨電流可在圖中區域B看見。在圖6中,在低電場處通過 - 01/N1/02隨道介電結構和厚氧化物而實質上不存在的 681939-27U3 15 1309875 穿隧電流可在圖中區域c看見。 可將依據本發明的記憶單元設計應用於各種記憶體類 型,包括但不限於,NOR及/或NAND型快閃記憶體。 如上述,隧道介電層可包括二層或更多層以上,包括 可提供小電洞穿隧阻障高度之一層。在一範例中,提供小 電洞穿隧阻障高度之該層可含有氮化矽。該層可夾置在二 層氧化石夕層之間,若將氮化矽用作中間層時可從而形成一 O/N/O随道介電質。在本發明一些較佳具體實施例中,隧 道介電結構中的各層至高達約4奈米厚。在一些較佳具體 實施例中’隨道介電結構中的各層厚度可約1奈米至3奈 米。在一範例性元件中,一三層結構可具有一約1Q埃至 30埃之底部層(例如氧化矽層)、一約1〇埃至3〇埃之中間 層(例如氮化矽層)、及一約1〇埃至3〇埃之頂層(例如另一 氧化矽層)。在一特定範例中,可使用一 三層結構, 其具有- 15埃的底部氧化砍層、—2()埃的中間氮化石夕層、 及一 18埃的頂部氧化矽層。 在一範例中,一薄〇/N/〇三層結構顯示可忽略的電荷 陷獲,參考目5a、5b及6所述之理論能帶圖及穿随電流 分析,可能建議一隧道介電結構(例如一 或更少之⑴卿02結構),可在保持期間= = 電洞直接㈣。㈣’在高電場仍可允許有效電洞穿随。 此可能係因能帶偏移可有效地遮擒犯及〇2穿随阻障。因 此,此建躺元件可提縣速電洞?_除, 習知謂QS元件之保制題。實驗分析顯示依據本發明各 681939-27U3 16 1309875 種具體實施例之記憶單元的極佳耐久及保持性質。 在一些較佳具體實施例中,隧道介電結構包括至少一 中間層及在中間層相對側上相鄰的二層,其中中間層及二 相鄰層各包含一第一材料和一第二材料,其中該第二材料 之價能帶位準大於第一材料之價能帶位準,且第二材料之 傳導能帶位準小於第一材料的傳導能帶位準;且其中第二 材料之濃度係高於二相鄰層間之中間層,且第一材料的濃 度在二相鄰層中係高於中間層。較佳的係,在依據本發明 之此具體實施例的一隧道介電結構中,第一材料包括氧及/ 或含氧化合物,且第二材料包括氮及/或含氮化合物.。例 如,第一材料可包括氧化物(例如氧化矽),且第二材料可 包括氮化物,例如Si3N4或SixOyNz。 依據本發明此方面之隧道介電質可由三或更多層構 成,所有此等層可含有類似元素(例如Si、N及Ο),只要 具有最小電洞隧道阻障高度之材料的濃度在中間層内係高 於二相鄰層即可。 在依據本發明先前具體實施例的隧道介電結構中,該 第二材料可依梯度濃度出現在中間層中,使得在中間層中 第二材料之濃度從一相鄰層/中間層介面增加到在中間層 内一深點處之最大濃度,且從該最大濃度深點降低到一在 該另一相鄰層/中間層介面處之較低濃度。濃度中的增及減 較佳係漸進式的。 在本發明又其他具體實施例中,隧道介電結構包括至 少一中間層及在中間層相對侧上的二相鄰層,其中二相鄰 681939-27U3 17 1309875 層包含一第一材料且中間層包含一第二材料,其 料之價能帶位準大於第一材料严处册 、弗材枓之仏此▼位準,且第二材料 =傳v能帶位準小於第—材料的傳導能帶位準;且其中該 第了材料係依梯度濃度出現在中間層中,使得在中間層中 第材料/辰度從一相鄰層/中間層介面增加到在中間層 -内’罙點處之最大濃度,且從該最大濃度深點降低到在該 另:相鄰層/中間層介面處之-較低濃度。濃度中的增及減 較佳係漸進式的。較佳的係,在依據本發明之此具體實施 例的-―随道介電結構中,第一材料包括氧及/或含氧化合 物且苐一材料包含氮及/或含氮化合物。例如,第一材料 可包括一氧化物(例如氧化矽),且第二材料可包括一氮化 物(例如 Si3N4 或 Six〇yNz)。 例如,在其中隧道介電層包含一三層〇N〇結構之本發 明的具體實施例中,該底部氧化層及頂部氧化層可包含二 氣化石夕,且中間氮化層可由例如氮氧化石夕及氣化石夕構成, •其中氮切的濃度(即,二者中具有較小電洞穿随阻障高度 的材料)在此層内非固定,而係在具有夾置氧化層之二介面 間的該層内某些深點處達到最大值。 在其中具有最小電洞穿隨ρ且障高度之材料達到其最大 —濃度的中間層内之精確點並非_,只要其依梯度出現且 在中間層内某些點處之随道介電層中達到其最大濃度。 具有取小電洞穿隨阻障高度之材料的梯度濃度可有利 於改進非揮發性記憶元件之各種性質,尤其係且有 SONONOS或SONONQS狀結構者。例如,可縮小保持狀 681939-27U3 18 1309875 ^電何域、可改進在高電場下之電洞㈣、及在其可能 程度下可避免麵道介電質中之電荷㈣。 ” 改:===卜咖有利地修 的材料浪度變化:= 遺者:二最小電洞穿随阻障高度 係由氮化㈣I係透過一能帶圖顯示。中間層(層2) 層2中氣化砂層(層1及層3)係由二氧化石夕構成。 位準分别錢巾會料,㈣價能帶位準料導能帶 及ti、值。^氮Γ 為最高之層2㈣度達到最大 士矣-g濃;^中顯不二種可能的氮化梦濃度梯度,其係 的产^述=度產生之可變價能帶位準及傳導能帶位準 “性—J le中顯示,藉由在虛線上表示層2中三 傳導能帶位準輿值=_帶位準及最高 依許此等具體實施例之多層的隨道介電結構可 二d::=^r該方法包括但不 中f1戶操著相沈積過程。—具有_之梯度濃度的 中曰s例如經由化學汽相沈積方法,或另—選摆孫 =二'!形成之過量氧化物或_的電裝氮 用或化學汽相上氧化層)可接著例如藉由氧化作 6S1939-27U3 19 1309875 一電荷儲存層接著 例中,可在_介魏H 道介―構上。在一範 儲存層。在-特㈣例中,=約5奈米至1G奈米的電荷 石夕層。在電荷儲存層上的H7奈米或更厚的氮化 例如,可使㈣9奈米錢厚的氧層、 理轉換氧化矽層的至少—却八 ^ «且』错由熱處 用以形成適合材料的形成氧切層。在此描述 比m ㈣層之任何已知或待開發的方法, 上可,積或形成随道介電層、電荷儲存層及/或絕緣 層。適&方法包括例如熱成長方法及化學汽相沈積方法。 在一範例中,熱轉換過程可提供高密度或濃度之介面 陷牌,其可提升記憶元件的陷獲效率。例如,氮化物的熱 轉換可在、約1000〇C處進行,同時閘極流量比係H2 : 02=1000 : 4000sccm。 此外,因為氮化矽大體上具有極低(大約19eV)之電洞 阻障,故在高電場下其對電洞穿隧可變得無障礙。同時, 一隧道介電質(例如ΟΝΟ結構)的總厚度,可防止電子在低 電場下直接穿隧。在一範例中,此不對稱行為可提供使一 5己‘丨思元件不僅提供快速電洞穿隧抹除,而且在保持期間減 少或免除電荷泡漏。 可藉由0.12微米NROM/NBit技術製造一範例性元 件。表1顯示在一範例中之元件結構及參數。所揭具有一 超薄0/N/0之隧道介電質可改變電洞穿隧電流。在一範例 中’ 一較厚(7奈米)N2層可作為一電荷陷獲層,並且一〇3(9 奈米)層可作為阻隔層。N2及03二者皆可使用NROM/NBit 681939-27U3 20 1309875 技術製造。 表1 層 近似厚度(埃) 底部氧化物(01) 15 中間氮化物(N1) 20 中間氧化物(02) 18 陷獲氮化物(N2) 70 阻隔氧化物(03) 90 閘極:N+多晶石夕 通道長度: 0.22微米 通道寬度: 0.16微米 在本發明一些較佳具體實施例中,一閘極可包含功函 數大於N+多晶矽之材料。在本發明一些較佳具體實施例 中,此高功函數閘極材料可包含例如鉑、銥、鎢、及其他 貴金屬之金屬。較佳的係,此等具體實施例中之閘極材料 的功函數大於或等於約4.5eV。在尤其較佳具體實施例中, 閘極材料包令—高功函數金屬,例如始或銀。此外,較佳 之高功函數材料包含但不限於P+多晶矽,及諸如氮化鈦及 氮化钽之金屬氮化物。在本發明之尤其較佳具體實施例 中,閘極材料包含在白。 依據本發明一較佳具體實施例具有高功函數閘極材料 之範例性元件,亦可由0.12微米NROM/NBit技術製成。 表2顯示在一範例中之元件結構及參數。所揭具有一超薄 0/N/0之隧道介電質可改變電洞穿隧電流。在一範例中, 681939-27U3 21 1309875 -較厚(7奈米)的犯層可作為—電荷陷獲層,並且一⑴(9 奈米)層可作為阻隔層。N2及03二者皆可使用撒仙 技術製造。 表210^ΐΓΓ "J" '2〇V ^^a(Vg)TT TM erasing rate. On the other hand, it can still be maintained; in some cases, it may be more than many conventional s_ electricity: the same money Γ ^ invented money unit can also avoid the use of band-to-band heat ί==: is used in the general 0M component . Avoid this band: avoiding the large loss of thermoelectric holes to introduce damage, and therefore avoiding the need to meet the demand. Referring to FIG. 21, the experimental measurement value of the voltage of the dielectric, and the mouth structure according to the present invention - (4) the real (4) channel, the display - ultra-thin 01 / Ν 1 / 〇 2 recording structure can have - negligible _ efficiency, such as The money continues to be verified by the constant threshold voltage level under the stylized pulse. In the example for the purpose of the test of 2, the thickness of 〇1/Ν1/〇2 is 3G, 3〇 and 35 Å, respectively. As shown in Figure 2, in the use of various methods of stylization (ie _FN stylization, rotten stylization and CHE (channel hot electron) stylized) in the Xuan m (four) process, the limit power Vt remains stable at near (nine) 19 volt. Therefore, this ultrathin 01/N1/02 film can be used as a -modulation tunnel dielectric structure. The results under various charge injection methods including cm, 681939-27U3 1309875 + FN and -FN all show negligible power gain. The process or component structure can be designed to minimize interface trapping so that the 01/N1 or N1/02 interface is functional. 3 shows an erase feature of a memory cell having a design of s〇N〇N〇s in accordance with an embodiment of the present invention. The memory cell of the embodiment illustrated in Figure 3 comprises a tunnel C C2, thickness of 15 angstroms, 20 angstroms and 18 angstroms, respectively. The n_MOSFET design. The memory cell of this embodiment comprises a tantalum nitride charge storage layer having a thickness of about 70 angstroms, a thickness of about 90 angstroms, a tantalum oxide layer, and a gate comprising any suitable conductive material, for example, =3⁄4 • Doped polycrystalline stone. Referring to Figure 3, a fast fn erase (e.g., within 10 strokes) can be achieved and can also be obtained - excellent self-convergent erase properties. Figure 4 shows the charge retention feature of the SONOS element of a particular embodiment of the memory cell of the present invention as described with reference to Figure 3. As shown, these = & S_s components are better, and the current value may be higher than multiple levels. (4) an energy band diagram containing at least a tunnel-like dielectric structure, wherein the at least one layer has a small dielectric hole with a barrier dielectric structure under a low electric field that may exist during memory data retention (this) In the example, the energy band diagram of MKFl U 1 /〇2 three layers) is in Ί a. It is possible to remove the band offset as shown by the dotted arrow at the low n? resistance n. Figure 5b shows that a barrier effect can be reduced by N1 and (7) so that direct wear through 〇1 can occur. Once there is a level of electric power barrier to the level of riding the dielectric can be allowed to feed 6S1939-27U3 14 1309875 effect FN erase operation. Figures 5c and 5d show another set of energy band diagrams in one example. For a preferred band offset condition in an example, the thickness of N1 may be greater than 〇ι. The energy band diagram of the valence band is plotted at the same electric field E_l4Mv/em. According to the WKB approximation, the possibility of wear is related to the shaded area. In this example, for the thickness Nl = () 1, the band offset is not completely dependent on the barrier of 02. On the other hand, for N1 > 01, the band offset can be easier to block (7). Therefore, for the thickness Ν1 > 〇1, the hole penetration current may be large under the same electric field of the (1) towel. An experiment with measured and simulated hole-through currents (shown in Figure 6) further describes the electrical-time tunneling of a pass-by-channel dielectric structure in accordance with certain embodiments of the present invention. For example, a hole penetration current of 01/m/〇2 dielectric can fall between an ultra-thin oxide and a thick oxide. In one example, under high electric fields, the hole-through current can approximate an ultra-thin oxide. However, under low electric fields, direct wear can be suppressed. As shown in Fig. 6, even in the electric field strength of only mv/cm # (four), the hole current can be detected through the thin oxide layer. The hole tunneling current is negligible through the thick oxide at a relatively high electric field strength of, for example, n_13 MV/cm. However, when high electric field strength occurs, the hole through the - ΟΝΟ dielectric structure will follow a current to a thin: oxide layer. In Fig. 6, a large current leakage due to tunneling through a thin oxide in a low electric field hole can be seen in the area 图 in the figure. In Fig. 6, the hole-through current through the -q/〇2(d)-t structure at high electric field strength can be seen in the region B in the figure. In Figure 6, the tunneling current of 681939-27U3 15 1309875, which is substantially absent through the - 01/N1/02 channel dielectric structure and thick oxide at low electric fields, can be seen in region c of the figure. The memory cell design in accordance with the present invention can be applied to a variety of memory types including, but not limited to, NOR and/or NAND type flash memory. As described above, the tunnel dielectric layer may include two or more layers, including one layer that provides a small hole tunneling barrier height. In one example, the layer providing a small hole tunneling barrier height may contain tantalum nitride. The layer can be sandwiched between the two layers of oxidized stone layers, and if the tantalum nitride is used as the intermediate layer, an O/N/O channel dielectric can be formed. In some preferred embodiments of the invention, the layers in the tunnel dielectric structure are up to about 4 nanometers thick. In some preferred embodiments, the thickness of each layer in the intervening dielectric structure can range from about 1 nm to about 3 nm. In an exemplary component, a three-layer structure may have a bottom layer (eg, a hafnium oxide layer) of about 1 Q angstrom to 30 angstroms, an intermediate layer (eg, a tantalum nitride layer) of about 1 〇 to 3 〇, And a top layer of about 1 〇 to 3 〇 (for example, another layer of yttrium oxide). In a specific example, a three-layer structure having a bottom oxide oxidized layer of -15 angstroms, an intermediate nitride layer of -2 angstroms, and a top yttrium oxide layer of 18 angstroms may be used. In one example, a thin tantalum/N/〇 three-layer structure exhibits negligible charge trapping. The theoretical energy band diagram and the follow-through current analysis described in references 5a, 5b, and 6 may suggest a tunnel dielectric structure. (eg one or less (1) Qing 02 structure), can be held during the hold = = hole directly (four). (d) 'In the high electric field, effective holes can still be allowed to wear. This may be due to the band offset, which can effectively conceal the shackles and the obstacles. Therefore, this built-in component can mention the county speed hole? _ In addition, the customary QS component protection issues. Experimental analysis shows the excellent durability and retention properties of the memory cells of the various embodiments of 681939-27U3 16 1309875 in accordance with the present invention. In some preferred embodiments, the tunnel dielectric structure includes at least one intermediate layer and two adjacent layers on opposite sides of the intermediate layer, wherein the intermediate layer and the two adjacent layers each comprise a first material and a second material Wherein the second material has a valence band level greater than a valence band level of the first material, and the second material has a conduction band level that is less than a conduction band level of the first material; and wherein the second material The concentration is higher than the intermediate layer between two adjacent layers, and the concentration of the first material is higher in the two adjacent layers than in the intermediate layer. Preferably, in a tunnel dielectric structure in accordance with this embodiment of the invention, the first material comprises oxygen and/or oxygenates and the second material comprises nitrogen and/or nitrogen containing compounds. For example, the first material can include an oxide (e.g., hafnium oxide) and the second material can include a nitride, such as Si3N4 or SixOyNz. The tunnel dielectric in accordance with this aspect of the invention may be comprised of three or more layers, all of which may contain similar elements (e.g., Si, N, and yttrium) as long as the concentration of the material having the minimum hole tunnel barrier height is intermediate The inner layer of the layer is higher than the two adjacent layers. In a tunnel dielectric structure according to a prior embodiment of the present invention, the second material may be present in the intermediate layer in a gradient concentration such that the concentration of the second material in the intermediate layer is increased from an adjacent layer/intermediate layer interface to The maximum concentration at a deep point in the intermediate layer, and from the deep point of the maximum concentration to a lower concentration at the interface of the other adjacent layer/intermediate layer. The increase and decrease in concentration is preferred to be progressive. In still another embodiment of the present invention, the tunnel dielectric structure includes at least one intermediate layer and two adjacent layers on opposite sides of the intermediate layer, wherein the two adjacent 681939-27U3 17 1309875 layers comprise a first material and an intermediate layer Including a second material, the price of the material can be higher than the first material, the material level of the material, and the second material = the energy level of the v material is less than the conduction energy of the first material. With a level; and wherein the first material is present in the intermediate layer according to a gradient concentration, such that the material/length in the intermediate layer increases from an adjacent layer/intermediate layer interface to an intermediate layer-inside point The maximum concentration, and from the deep point of the maximum concentration to the lower concentration at the other adjacent layer/intermediate interface. The increase and decrease in concentration is preferred to be progressive. Preferably, in the conventional dielectric structure according to this embodiment of the invention, the first material comprises oxygen and/or an oxygen-containing compound and the first material comprises nitrogen and/or nitrogen-containing compounds. For example, the first material may comprise an oxide (e.g., hafnium oxide) and the second material may comprise a nitride (e.g., Si3N4 or Six〇yNz). For example, in a specific embodiment of the invention in which the tunnel dielectric layer comprises a three-layer 〇N〇 structure, the bottom oxide layer and the top oxide layer may comprise two gas fossils, and the intermediate nitride layer may be, for example, nitrogen oxynitride. And the composition of the gas fossil eve, • wherein the concentration of nitrogen cut (ie, the material having a smaller hole with the height of the barrier) is not fixed in the layer, but is between the two interfaces with the sandwiched oxide layer. The maximum point is reached at some deep points in this layer. The precise point in the intermediate layer in which the material having the smallest hole penetration ρ and the height of the barrier reaches its maximum concentration is not _ as long as it occurs according to the gradient and reaches in the intervening dielectric layer at some point in the intermediate layer Its maximum concentration. Gradient concentrations of materials having small holes that follow the barrier height can be beneficial for improving various properties of non-volatile memory elements, especially those having SONONOS or SONONQS-like structures. For example, it can be reduced to maintain the shape of 681939-27U3 18 1309875 ^ electric field, can improve the hole in the high electric field (four), and to the extent possible to avoid the charge in the surface dielectric (four). Change: === Buha beneficially repair material wave change: = Legacy: Two minimum hole penetration with barrier height is shown by nitriding (4) I system through a band diagram. Intermediate layer (layer 2) Layer 2 The middle gasification sand layer (layer 1 and layer 3) is composed of sulphur dioxide eve. The level of the money towel will be materialized, (4) the valence band can be used to guide the material and the ti value. The nitrogen Γ is the highest layer 2 (4) Degree reaches the maximum gentry-g concentration; ^ is the two possible nitriding dream concentration gradients, the system's production = the variable price band and the conduction band level "sex - J le It is shown that by means of a dotted line, the three-conducting energy level in the layer 2, the level of the band, and the highest level of the dielectric structure of the embodiment can be two d::=^r The method includes, but not f1, the phase deposition process. - the intermediate 曰 with a gradient concentration of _, for example via a chemical vapor deposition method, or alternatively an excess oxide formed by the sun or the second oxide or an oxide layer on the chemical vapor phase) For example, by oxidizing as a charge storage layer of 6S1939-27U3 19 1309875, in the following example, it can be constructed on the medium. In a van storage layer. In the case of - (4), = a charge of about 5 nm to 1 G nm. H7 nanometer or thicker nitridation on the charge storage layer, for example, can make the (four) 9 nm thick oxygen layer, the at least one of the ytterbium oxide layer, and the thermal layer be used to form a suitable The material forms an oxygen cut layer. Any known or to be developed method of the m (four) layer is described herein to form or form an intervening dielectric layer, a charge storage layer and/or an insulating layer. Suitable & methods include, for example, thermal growth methods and chemical vapor deposition methods. In one example, the thermal conversion process can provide a high density or concentration interface trap that can increase the trapping efficiency of the memory element. For example, the thermal conversion of the nitride can be carried out at about 1000 〇C, while the gate flow ratio is H2: 02 = 1000: 4000 sccm. In addition, since tantalum nitride has a very low (about 19 eV) hole barrier, its tunneling to the hole can be made unobstructed under a high electric field. At the same time, the total thickness of a tunnel dielectric (e.g., germanium structure) prevents electrons from tunneling directly under low electric fields. In one example, this asymmetrical behavior can provide that the 丨 丨 元件 element not only provides fast hole tunneling erase, but also reduces or eliminates charge bubble leakage during retention. An exemplary component can be fabricated by 0.12 micron NROM/NBit technology. Table 1 shows the component structure and parameters in an example. The tunnel dielectric with an ultra-thin 0/N/0 can change the tunnel tunneling current. In one example, a thicker (7 nm) N2 layer acts as a charge trapping layer and a 3 (9 nm) layer acts as a barrier layer. Both N2 and 03 can be fabricated using NROM/NBit 681939-27U3 20 1309875 technology. Table 1 Layer Approximate Thickness (Angstrom) Bottom Oxide (01) 15 Intermediate Nitride (N1) 20 Intermediate Oxide (02) 18 Occupied Nitride (N2) 70 Barrier Oxide (03) 90 Gate: N+ Polycrystalline Length of the stone channel: 0.22 micron channel width: 0.16 micrometers In some preferred embodiments of the invention, a gate may comprise a material having a work function greater than that of the N+ polysilicon. In some preferred embodiments of the invention, the high work function gate material may comprise a metal such as platinum, rhodium, tungsten, and other precious metals. Preferably, the work function of the gate material in these embodiments is greater than or equal to about 4.5 eV. In a particularly preferred embodiment, the gate material is a high work function metal such as a start or a silver. In addition, preferred high work function materials include, but are not limited to, P+ polysilicon, and metal nitrides such as titanium nitride and tantalum nitride. In a particularly preferred embodiment of the invention, the gate material is included in white. Exemplary components having a high work function gate material in accordance with a preferred embodiment of the present invention may also be fabricated using 0.12 micron NROM/NBit technology. Table 2 shows the component structure and parameters in an example. The tunnel dielectric with an ultra-thin 0/N/0 can change the tunneling current. In one example, 681939-27U3 21 1309875 - a thicker (7 nm) layer can serve as a charge trapping layer and a (1) (9 nm) layer can serve as a barrier layer. Both N2 and 03 can be made using Saxian technology. Table 2

底部氧化物 中間氮化物 20 中間氧化物 18 陷獲氮化物asm 70 _阻隔氧化物_ _______閘極 通道長度 ~__ :翻 ----------- 0.22微米 通道寬度:0.16微米 依據本發明具體實施例具有高功函數閘極材料之記憶 單兀顯現比其他具體實施例改進甚多的抹除性質。高功函 ❿ 數閘極材料抑制閘極電子注入陷獲層中。在本發明一些具 體實施例中,其中記憶單元包含一 N+多晶矽閘極,在抹除 -期間電洞穿隧到電荷陷獲層且同時閘極電子注入。此自收 斂抹除效應導致在抹除狀態中更高的臨限電壓位準,其在 -NAND應用中可能不符合需求。可將依據本發明具有高功 -函數閘極材料具體實施例之記憶單元用於各種類型的記憶 體應用’包括例如NOR及NAND型記憶體。然而,依據 本發明具有高功函數閘極材料具體實施例的記憶單元,係 尤其適用於NAND應用,其中在抹除/重設狀態中提升臨限 681939-27U3 22 1309875 電壓可能不符合需求。依據本發明具有高功函數閘極材料 具體實施例的記憶單元,可經由電洞穿隧方法及較佳係細 由-FN抹除操作來抹除。 一具有一ΟΝΟ穿隧介電質及一 N+多晶矽閘極之範例 性元件,可藉由習知SONOS或NROM方法程式化,且由 通道FN電洞穿隨抹除。圖7a顯示在一範例中具有一 穿隧介電質之範例性SONONOS元件的抹除特徵。參考圖 ^ ’ 一較高的閘極電壓導致更快速之抹除速率。其亦具有 更高的飽和Vt,因為閘極注入亦係更強並且產生之動態平 衡』(其决疋Vt)更高。圖式右手側顯示當臨限電壓根據袜 除閘極包壓達到約3到約5伏特之最小值。藉由微分圖h 中的曲線可由—暫態分析方法抽取電洞穿隨電流。來自圖 7a。中測量值的抽取電洞電流係顯示在如以上討論之圖6 $二比車又’亦使用WKB近似繪出模擬之電洞 ::過=果與預:則合理地-致。在高電場下,穿随電; π 02堆豐到達超'薄ο卜同時其係在低電場下關 閉。 旦體月具有高功函數閘極材料之記憶單元的-些 極抑制閘極電子注入)中,取 能低許多,且甚狀態中該元件的臨限電壓可 元件(其中閘極係由—依據本發明—具體實施例之記憶 之〇構)的臨限^值且隨道介電層包括1娜1 顯示,_FN抹除择值係顯示在圖7b中。如圖作中 細作期間在類似閑極電壓(-18V)處,該元件 681939-27U3 23 75Bottom oxide intermediate nitride 20 intermediate oxide 18 trapping nitride asm 70 _ barrier oxide _ _______ gate channel length ~__ : turn----------- 0.22 micron channel width: 0.16 micron A memory unit having a high work function gate material exhibits much improved erase properties than other embodiments in accordance with an embodiment of the present invention. The high work function ❿ gate material suppresses the gate electron injection trapping layer. In some embodiments of the invention, the memory cell includes an N+ polysilicon gate during which the hole tunnels to the charge trapping layer and simultaneously gate electron injection. This self-convergence erase effect results in a higher threshold voltage level in the erased state, which may not meet the requirements in a -NAND application. A memory cell having a high power-function gate material embodiment in accordance with the present invention can be used in various types of memory applications, including, for example, NOR and NAND type memory. However, a memory cell having a high work function gate material embodiment in accordance with the present invention is particularly suitable for use in NAND applications where the threshold is raised in the erase/reset state. 681939-27U3 22 1309875 The voltage may not meet the requirements. The memory cell having the high work function gate material according to the present invention can be erased by a hole tunneling method and preferably by a -FN erase operation. An exemplary component having a tunneling dielectric and an N+ polysilicon gate can be programmed by conventional SONOS or NROM methods and erased by a channel FN via. Figure 7a shows an erase feature of an exemplary SONONOS component having a tunneling dielectric in an example. Refer to Figure ^' A higher gate voltage results in a faster erase rate. It also has a higher saturation Vt because the gate implant is also stronger and produces a dynamic balance (which is higher than Vt). The right hand side of the figure shows that the threshold voltage is a minimum of about 3 to about 5 volts depending on the thickness of the shoe. The hole passing current can be extracted by the transient analysis method by the curve in the differential graph h. From Figure 7a. The measured hole current system of the measured values is shown in Fig. 6 as discussed above. The two-dimensional vehicle also uses the WKB approximation to draw the simulated hole. :: Over = fruit and pre: It is reasonable. In the high electric field, wear with electricity; π 02 heap to reach the super thin, while its system is closed under low electric field. In the memory cell of the high-function-function gate material, some of which are low-intensity gate-injection, the energy is much lower, and in some cases, the threshold voltage component of the component (where the gate is based) The threshold value of the present invention - the memory structure of the specific embodiment and the channel dielectric layer including the 1 Na 1 display, the _FN erase selection system is shown in Figure 7b. As shown in the figure, during the fine-grained voltage (-18V), the component 681939-27U3 23 75

的臨限電壓可設定在_3v以下 電容相對於閘極電壓值。 。圖7c中顯示該元件之對應 此外’依據本發明I士上 ,A ^昇有鬲功函數閘極材料具體實施例 的3己憶元件之保持性暫总a M係已改進。具有鉑閘極之記憶元件 的保持性負係顯示在圖7d ^ ^ + 嘲W中,其中電容係圖示為在抹除及 ^式化後^接者在各操作後3G分鐘後及各操作後二小時 〜閘極電Μ成函數。已觀_最小偏差。The threshold voltage can be set below _3v. Capacitance is relative to the gate voltage value. . Correspondence of the element is shown in Fig. 7c. Further, in accordance with the present invention, the retention of the 3M element of the A^liter 鬲 work function gate material embodiment has been improved. The negative retention of the memory element with platinum gate is shown in Figure 7d ^ ^ + 嘲 W W, where the capacitance is shown after the erase and the ^ 3 after 3G minutes after each operation and each operation After two hours ~ the gate is turned into a function. Observed _ minimum deviation.

依據本發明各種具I*餘 ^ 離方案操作。例如,且=1&例之記憶单元可用至少二分 可用來執行-2位元/單有=讀_式υ的CHE程式化 凡/早几操作。此外’亦可將低功率+FN 裎式化(模式2)用作_ 1 Α — 7 2位兀/%元操作。二模式皆可使用 相同電洞穿随技& t、、上h , , 禾除方法。镇式1較佳係可用作NOR型快閃 ^思-之虛擬接地陣列架構。模較佳係可用於nand 型之快閃記憶體。According to the invention, various I* residual scheme operations are employed. For example, the memory unit of the =1 & example can use at least two points to perform the CHE stylization of the -2 bit/single = read _ type . In addition, low power + FN (mode 2) can also be used as _ 1 Α - 7 2 兀 / % element operation. Both modes can use the same hole to wear the technology & t,, h, , and method. The town type 1 is preferably used as a virtual ground array structure of the NOR type flash memory. The mold is preferably used for nand type flash memory.

杂^圖8之範例顯示在模式1操作下,依據本發明一具體 只細例的虛擬接地陣列架構NOR型快閃記憶體的極佳耐 久性質。具有隧道介電結構之此等記憶元件的抹除退化不 會發生’因為電洞穿隧抹除(Vg=-15V)係一均勻通道抹除方 法。圖9中亦顯示對應的IV曲線,其顯示在多次P/E循環 後該元件的少許退化。在一範例中,此可能因超薄氧化層/ 氮化層擁有良好之應力免除性質。此外,該記憶元件不會 有熱電洞引入之損害。圖10顯示依據本發明一具體實施例 之NAND型快閃記憶體在模式2下操作中的耐久性質。為 了更快速的收斂抹除時間,可使用更大的偏壓(Vg=-16V)。 681939-27U3 24 1309875 在此範例中亦可獲得極佳耐久性。 SON(^rl顯示依據本㈣―具體實施例之範例性 70件的電荷保持,其中在1GG小時後僅觀察到The example of Fig. 8 shows the excellent durability of the virtual ground array architecture NOR type flash memory in accordance with a specific example of the present invention under mode 1 operation. Erasing degradation of such memory elements having a tunnel dielectric structure does not occur 'because hole tunneling erase (Vg = -15 V) is a uniform channel erasing method. A corresponding IV curve is also shown in Figure 9, which shows a slight degradation of the element after multiple P/E cycles. In one example, this may have good stress relief properties due to the ultra-thin oxide/nitride layer. In addition, the memory element does not suffer from the introduction of thermoelectric holes. Figure 10 is a diagram showing the durability of a NAND type flash memory in mode 2 operation in accordance with an embodiment of the present invention. For faster convergence erase time, a larger bias voltage (Vg = -16V) can be used. 681939-27U3 24 1309875 Excellent durability is also obtained in this example. SON (^rl shows the charge retention of the example 70 pieces according to this (4)-specific embodiment, in which only observed after 1GG hours

電制失。此改進方案的電流值等級比習知 〇S凡件焉。VG加速保持測試亦顯示可在低電場抑制 2穿隧二圖11顯示針對10ΚΡ/Ε循環元件的VG加速保 、通試之範例。電荷損失於1麵秒應力後在應力下係 J八扣示可抑制在小電場處之電洞直接穿隧。 因此,在上述範例中所指的S0N0N0S設計 有極佳耐久性質之快速電洞㈣抹除。如以上所指,^ 二與外N=類型氮化物儲細記憶體中實施該設 夕個具有類似或不同組態之記憶元件。 匕括 在根據本發明之陣列的各種具體實施例中, 據本發明之讀、單减取代在-虛浦地陣列架構中之習 或SONOS元件。可藉由使用FN電洞穿隨而非 ‘、、、電洞注人解決或減輕可靠性問題和抹除退化。在不用以 下描述的特定結構限制本㈣之範打,町將描述 本么明之心隱體陣列的各種操作方法,用於範例性⑽ 擬接地陣列架構。 CHE或CHISEL(通道激始次要電子)程式化及反向讀 出可用於2位元/單元記憶體陣列。並且抹除方法可為一二 勻通道FN電洞穿随抹除。在一範例中,該陣列架構可為 虛擬接地陣列或加X陣列。參考圖12a_2G,可將二 681939-27U3 25 1309875 01/N1/02三層結構用作_道介電質,各層厚度約3务米哎 更少以提供電洞直接穿隧。參考圖12a_20,N2可比 厚以提供一高陷獲效率。一絕緣層(〇3)可為由溼氧化形成 之氧化矽層,例如溼轉換之頂部氧化物(氧化矽),以在〇3 及N2間之介面處提供密度大的陷拼。〇3可為約6奈米戍 • 更厚以防止電荷自此氧化矽層損失。 圖12a及12b顯示一併入以上討論之記憶單元的虛擬 鲁 接地陣列架構範例,諸如具有一三層ΟΝΟ隧道介電質的纪 憶單元。尤其係,圖12a顯示記憶體陣列之一部分的等效 電路’並且圖12b顯示該記憶體陣列之一部分的範例性布 局。 此外,圖13顯示併入該陣列中之若干記憶單元的斷面 示意圖。在一範例中,埋入式擴散(BD)區域可為用於記情 單元之源極或汲極區的N+摻雜接面。基體可為p型基體。 為了避免BD0X區域(BD上的氧化物)在-FN抹除期間的可 φ 能崩潰,在一範例中可使用一厚BDOX(>50奈米)。 圖14a及14b顯示一用於併入具有上述隨道介電質 - 计之2位元/單元記憶單元的範例性虛擬接地陣列之可能電 子重置(RESET)方案。在執行進一步P/E循環前,所有元件 - 可首先經歷一電子「RESET」。一 RESET過程可確保在相 -同陣列中之記憶單元的Vt —致性且將元件Vt提升至收斂 抹除狀態。例如,施加Vg=-15V達1秒(如圖I4a中顯示), 可有將某些電荷注入氮化矽之電荷陷獲層以達到動態平衡 條件的效果。使用RESET,儘管記憶單元因例如在其製程 681939-27U3 26 1309875 中電漿充電效應造成之不均勻充電,亦可使其vt收斂。用 於產生自收斂偏壓條件之替代方式係提供閘極及基體電壓 一者之偏壓。例如參考圖14b,可施加Vg=-8V和P井=+7V。 圖15a及15b顯示用於併入具有上述隧道介電質設計 ' 之2位元/單元記憶單元的範例性虛擬接地陣列之程式化方 - 案。通道熱電子(CHE)程式化可用於程式化該元件。對於圖 15a中顯示的Bit-Ι程式化,電子係局部地注入BLN(位元 鲁 線N)上之接面邊緣。對於圖15b中顯示的Bit-2程式化, 電子係儲存在BLN-I上。用於WL(字元線)的典型程式化電 壓係約6V至12V°BL(位元線)的典型程式化電壓係約3 至7V ’且可使p井保持接地。 圖16a及16b顯示用於併入具有上述隧道介電質設計 之2位元/單元記憶單元的範例性虛擬接地陣列之讀取方 案。在一範例中’反向讀出係用來讀取此元件以執行2位 元/單元操作。參考圖16a,對於讀取Bit-1,BLN-I係用〆 鲁 適合之讀取電壓(例如1.6V)施加。參考圖16b,對於讀取 Bit-2 ’ BLN係用一適合之讀取電壓(例如1 6V)施加。在一 範例中’讀取電壓可在約i至2V的範圍中。字元線及P 井可保持接地。然而,亦可執行其他已修改的讀取方案, - 诸如 ^升Vs反向讀出方法。例如,一提升vs反向言買出 方法可將 Vd/Vs=1.8/0.2V 用於讀取 Bit-2,且 Vd/Vs=〇.2/1.8 用於讀取Bit-1。 圖14a及14b亦顯示用於併入具有上述隧道介電質設 計之2位元/單元記憶單元的範例性虛擬接地陣列之扇區抹 681939-27U3 27 1309875 除方案。在一範例中,可同時施加扇區抹除與通道電洞穿 隨抹除以抹除記憶單元。在§己憶單元中具有S〇n〇n〇S結 構之ΟΝΟ隧道介電質可提供快速抹除,其可在約1〇至5〇 亳秒中和自收斂通道抹除速率中發生。在一範例中,扇區 抹除操作條件可類似RESET過程。例如,參考圖Ma,在 WL處同時地施加VG=約-15V及留下所有BL為浮動可達 到扇區抹除。且p井可保持接地。 或者是,參考圖14b,施加約-8V至WL且約^乂至口 井亦可達到扇區抹除。在一些範例中,完全扇區抹除操作 可在100毫秒或更少時間内實現,而不會有任何過抹除或 難以抹除之單元。上述的元件設計可有利於一提供極佳自 收斂性質的通道抹除。 λ圖17顯示在使用一 SONONOS元件之範例中的抹除特 徵。一 SONONOS 元件之範例可使 〇ι/Ν1/02/Ν2/〇3 = β 度分別為、約15/20/18/70/90埃,具有一 Ν+多晶石夕閑極並1 熱轉換頂部氧化物為〇3。已顯示用於各種閘極電壓之抹除 速率。較向之閘極電壓導致更快速的抹除速率。 ν、 然而,收斂Vt亦更高。此係因閘極注入在較高閘極 Μ了更活躍。為減少閘極注人,可替代地使用高功函數的 ρ+多晶矽閘極或其他金屬閘極作為閘極材料,以在技、 間減少閘極注入電子。 禾除期 圖18顯示將SONONOS元件用於虛擬接地陣列架構之 耐久性質。在某些範例中之耐久性質極好。用於Bitq的浐 式化條件係Vg/Vd=8.5/4.4V、0.1微秒,用於出 , 2係 681939-27U3 28 1309875Loss of electricity. The current rating of this improvement scheme is better than the conventional 〇S. The VG Acceleration Hold Test also shows that it can be suppressed at low electric fields. 2 Tunneling. Figure 11 shows an example of a VG acceleration guarantee for a 10 ΚΡ/Ε cycle component. The charge loss is under the stress of one-sided second stress, which can suppress the direct tunneling of the hole at the small electric field. Therefore, the S0N0N0S referred to in the above example is designed to have a fast cavity (4) with excellent durability. As indicated above, the memory elements having similar or different configurations are implemented in the N=type nitride memory memory. Included in various embodiments of the array in accordance with the present invention, the read or single subtractive replacement of the learned or SONOS component in a virtual array architecture. It can be solved by using FN holes instead of ‘, , , hole holes to solve or mitigate reliability problems and eradicate degradation. Without limiting the specific structure described below, the town will describe the various methods of operation of the invisible array for the exemplary (10) pseudo-grounded array architecture. CHE or CHISEL (channel abrupt secondary electrons) stylized and reverse readouts are available for 2-bit/cell memory arrays. And the erasing method can be used for the one-two uniform channel FN hole to be erased. In one example, the array architecture can be a virtual ground array or an X array. Referring to Figures 12a-2G, a two-layer structure of two 681939-27U3 25 1309875 01/N1/02 can be used as the dielectric material, and the thickness of each layer is less than about three dimensions to provide direct tunneling of the holes. Referring to Figures 12a-20, N2 can be thicker to provide a high trapping efficiency. An insulating layer (〇3) may be a ruthenium oxide layer formed by wet oxidation, such as a wet-converted top oxide (ruthenium oxide) to provide a dense trap at the interface between 〇3 and N2. 〇3 can be about 6 nm 戍 • Thicker to prevent charge loss from this ruthenium oxide layer. Figures 12a and 12b show an example of a virtual lu ground array architecture incorporating the memory cells discussed above, such as a memory cell having a three layer germanium tunnel dielectric. In particular, Figure 12a shows an equivalent circuit 'of a portion of a memory array' and Figure 12b shows an exemplary layout of a portion of the memory array. In addition, Figure 13 shows a schematic cross-sectional view of several memory cells incorporated into the array. In one example, the buried diffusion (BD) region can be an N+ doped junction for the source or drain region of the quotient cell. The matrix can be a p-type matrix. In order to avoid the collapse of the BD0X region (the oxide on the BD) during the -FN erasing, a thick BDOX (> 50 nm) can be used in one example. Figures 14a and 14b show a possible electronic reset (RESET) scheme for incorporating an exemplary virtual ground array having the above described 2-bit/cell memory cells. All components - before experiencing further P/E cycles - can first go through an electronic "RESET". A RESET process ensures that the Vt of the memory cells in the phase-same array is consistent and the component Vt is raised to a clear erase state. For example, applying Vg = -15 V for 1 second (as shown in Figure I4a) may have the effect of injecting some charge into the charge trapping layer of tantalum nitride to achieve dynamic equilibrium conditions. With RESET, although the memory cell is unevenly charged due to, for example, the plasma charging effect in its process 681939-27U3 26 1309875, its vt can be converged. An alternative to generating a self-converging bias condition is to provide a bias voltage for both the gate and the base voltage. For example, referring to Figure 14b, Vg = -8V and P well = +7V can be applied. Figures 15a and 15b show a stylized recipe for an exemplary virtual ground array for incorporating a 2-bit/cell memory cell having the tunnel dielectric design described above. Channel hot electron (CHE) stylization can be used to program the component. For the Bit-Ι stylization shown in Figure 15a, the electrons are locally injected into the junction edges of the BLN (bitline Lu N). For the Bit-2 stylization shown in Figure 15b, the electronics are stored on the BLN-I. A typical stylized voltage for WL (word line) is typically about 3 to 7V' with a typical stylized voltage of about 6V to 12V° BL (bit line) and keeps the p well grounded. Figures 16a and 16b show a read scheme for an exemplary virtual ground array for incorporating a 2-bit/cell memory cell having the tunnel dielectric design described above. In an example, the 'reverse readout' is used to read this element to perform a 2-bit/cell operation. Referring to Figure 16a, for reading Bit-1, the BLN-I is applied with a suitable read voltage (e.g., 1.6V). Referring to Figure 16b, the read bit-2' BLN is applied with a suitable read voltage (e.g., 16 V). In one example, the read voltage can be in the range of about i to 2V. The word line and P well can be kept grounded. However, other modified read schemes can also be implemented, such as the ^L Vs reverse readout method. For example, a boost vs. reverse buy method can use Vd/Vs=1.8/0.2V for reading Bit-2, and Vd/Vs=〇.2/1.8 for reading Bit-1. Figures 14a and 14b also show a sector wipe 681939-27U3 27 1309875 division scheme for incorporating an exemplary virtual ground array of a 2-bit/cell memory cell having the tunnel dielectric design described above. In one example, a sector erase and channel hole wear erase can be applied simultaneously to erase the memory cell. The tunnel dielectric having the S〇n〇n〇S structure in the § memory cell provides a fast erase which can occur in about 1 〇 to 5 亳 亳 and in the self-converging channel erase rate. In one example, the sector erase operation condition can be similar to the RESET process. For example, referring to Figure Ma, VG = about -15V is applied simultaneously at WL and all BLs are left floating up to sector erase. And p well can be kept grounded. Alternatively, referring to Figure 14b, application of about -8V to WL and about to the well can also achieve sector erasure. In some examples, a full sector erase operation can be implemented in 100 milliseconds or less without any cells that are erased or difficult to erase. The above component design can facilitate a channel erase that provides excellent self-converging properties. λ Figure 17 shows the erase feature in the example using a SONONOS component. An example of a SONONOS component can be 〇ι/Ν1/02/Ν2/〇3 = β degrees, respectively, about 15/20/18/70/90 angstroms, with one Ν + polylithic idle pole and 1 thermal conversion The top oxide is 〇3. The erase rate for various gate voltages has been shown. A more gate voltage results in a faster erase rate. ν, However, the convergence Vt is also higher. This is more active due to the gate implant at the higher gate. In order to reduce the gate injection, a high work function ρ+ polysilicon gate or other metal gate can be used as the gate material to reduce the gate injection electrons. Drainage Figure 18 shows the durability of the SONONOS component for a virtual grounded array architecture. In some cases the durability is excellent. The enthalpy conditions for Bitq are Vg/Vd=8.5/4.4V, 0.1 microseconds for use, 2 series 681939-27U3 28 1309875

vg/Vs-8.5/4.6V、0]微秒。FN 50亳秒以同眭社 抹除可使用達約 道抹除,難以除一位兀。因為FN抹除係自收斂均勻通 範例Li:除或過抹除之單元通常不會出現。在-些 抹除驗證或絕佳耐久性f ’即使不使用程式化,Vg/Vs-8.5/4.6V, 0] microseconds. FN 50 亳 以 以 以 以 抹 抹 抹 抹 抹 抹 抹 抹 抹 抹 抹 抹 抹 抹 抹 抹 抹 抹Because the FN erase is self-convergent and uniform. Example Li: Units that are or are not erased usually do not appear. In some erase verification or excellent durability f ’ even if not stylized,

特徵圈Γ二Γ: f示在一範例中於p/e循環期間之1-V 的對應1^曲=,度L圖19收線性標度(圖19b)二者中 P/E循環後I有 賴中,—S_N〇S科在多次 導(㈣許退化,使得該次限定值擺動邮.)及跨 0 Λ SON〇N〇s 熱電洞,、主入Γ更優異之财久性質。其—原因可為未使用 、#11榀。此外,上揭的一超薄氧化物可具有比一厚隧 道乳化物更佳之應力免除性質。 ,20 1 員不在一範例中之chisel程式化方案。程式化 兀牛的—替代方法係使用CHISEL·程式化方案,其使用 負基,偏壓增強撞擊離子化作絲增加熱載體效率。程式 化電流由於體欵應亦可減少。此圖中顯示典型條件,其中 基體係用負電摩(-2V)施加,並且將接面電壓減少到約 3’5V。對於習知NROM元件及技術,CHISEL·程式化不可 應用’因為其在靠近通道中心區可能注入較多電子。並且 熱電洞抹除對於移走習知NROM元件中靠近通道中心區之 電子係無效率。 圖21a及21b顯示一範例中之JTOX虛擬接地陣列的 設計。JT〇x虛擬接地陣列提供在記憶體陣列中使用 681939-27U3 29 1309875 SONONOS記憶單元之替代性實施。在一範例中,jT〇x結 構及虛擬接地陣列間其一差別係JT〇x結構中的元件係由 STI方法隔離。一典型布局範例係顯示在圖21a中。圖2lb 顯示一對應的等效電路,其係與一虛擬接地陣列相同。The characteristic circle Γ2Γ: f shows the corresponding 1^ 曲 of the 1-V during the p/e cycle in an example, the degree L is in the linear scale (Fig. 19b) and the P/E cycle is followed by I. It depends on the fact that the S_N〇S family has been guided many times ((4) to degenerate, so that the limit value is oscillated.) and across the 0 Λ SON〇N〇s thermoelectric hole, and the main entry is more excellent for the long-term nature. The reason - can be unused, #11榀. In addition, an ultrathin oxide disclosed above may have better stress relief properties than a thick tunnel emulsion. , 20 1 member is not a chisel stylized scheme in an example. The stylized yak-alternative method uses the CHISEL·stylization scheme, which uses a negative base and bias-enhanced impact ionization to increase the heat carrier efficiency. The stylized current should also be reduced due to body volume. Typical conditions are shown in this figure, where the base system is applied with a negative motor (-2V) and the junction voltage is reduced to about 3'5V. For conventional NROM components and techniques, CHISEL·stylization is not applicable because it may inject more electrons near the center of the channel. And the thermal hole erase is inefficient for removing the electronic system near the center of the channel in the conventional NROM device. Figures 21a and 21b show the design of a JTOX virtual ground array in an example. The JT〇x virtual ground array provides an alternative implementation of the 681939-27U3 29 1309875 SONONOS memory unit in a memory array. In one example, the difference between the jT〇x structure and the virtual ground array is that the components in the JT〇x structure are isolated by the STI method. A typical layout example is shown in Figure 21a. Figure 2b shows a corresponding equivalent circuit that is identical to a virtual ground array.

如上揭,依據本發明之記憶單元結構係適於N0R與 NAND型快閃記憶體二者。以下將描述記憶辦列設計及 其操作方法的額外範例。在不用以下描述的特定結構限制 本發明之鉍疇下,以下將描述依據本發明之記憶體陣列的 各種操作方法’用於範例性NAND架構。 如上述,可將具有〇N〇隧道介電質的n通道 SONONOS記憶元件用於_記憶元件。圖瓜及m顯系 NAND陣列架構之範例。圖2如及⑽自二不同方向顯禾 -範例性記憶體師m計之斷面圖。在—些範例中,記憶 體陣列的操作方法可包括+FN程式化、自㈣重設/抹除及 讀取方法。此外,在-些範例中可包括電路操作方法以避 免程式化干擾。 除了早塊閘極結構設神々卜 再又。f外,亦可使用一分裂閘極 (spm-gate)陣列,諸如位在靠近源極/沒極區之二 極間使用S〇N〇N〇S元件之咖d陣列。在一此範例中, 分裂閘極設計可調整元収指減^ 外,可設計該等元件以獲得 此 動閘極間耦合效應,或-者二“;罪性,以減少或除去洋 # 達]° 如上揭,—SONON〇s ,己隐兀件可k供極佳自收數抹除 及vt分布控制。再者,緊杳 、了協助扇&抹除扭作 緊凑的抹除狀態分布可有利於多位 681939-27U3 30 1309875 準應用(MLC)。 藉由將某些設計用作記憶體陣列結構,有效通道長度 (Leff)可被擴大,以減少或者消除短通道效應。可設計〆些 範例以不使用擴散接面,從而避免在記憶元件製程期間提 供淺接面或使用袋狀植入的挑戰。 圖1顯示具有SONONOS設計之記憶元件的範例。此 外,表1註釋上述用作不同層的材料及其厚度之範例。在 一些範例中,可用P+多晶矽閘極來提供較低飽和重設/抹除 電壓Vt,其可藉由減少閘極注入達到。 圖22a及22b顯示一記憶體陣列的範例,諸如具有依 據表1所述具體實施例之記憶單元的S〇n〇n〇S-NAND陣· 列,其具有擴散接面。在一範例中,分離的元件可藉由各 種隔離技術彼此隔離,例如藉由使用淺溝渠隔離(STI)或絕 緣物上石夕(SOI)之隔離技術。參考圖22a,一記憶體陣列可 包括多條位元線(例如BL1及BL2),及多條字元線(諸如 WL1、WL>M、及WLN)。此外,該陣列可包括源極線電 晶體(或源極線選擇電晶體或SLT)及位元線電晶體(或位元 線選擇電晶體或BLT)。如舉例,該陣列中之記憶單元可使 用SONONOS設計,並且SLT及BLT可包括n型金氧半導 體場效應電晶體(NMOSFET)。 圖22b顯示一記憶體陣列(如NAND陣列)的範例性布 局。參考圖22b,Lg係記憶單元的通道長度,並且Ls係記 憶元件之各分離線間的空間。此外,w係記憶單元的通道 I度,並且Ws係分離位元線或源極/没極區間之隔離區寬 681939-27U3 31 1309875 度’其在一範例可為STI寬度。 再次參考圖22a及22b,記憶元件可串聯連接且形成 NAND陣列。例如,一串記憶元件可包括16或幻個記憶 疋件’提供16或32的串數目。可使用BLT及SLT作為選 擇電晶體以控制對應的]SfAND串。在一範例中,用於BLT '及SLT的閘極介電質可為不包括一氮化矽陷獲層的氧化矽 層。此組態在一些範例中(雖然在所有情況中不一定需要) φ 可避免在記憶體陣列操作期間BLT和SLT的可能vt偏移。 另一選擇係BLT及SLT可將複數層ONONO層的結合用作 其閘極介電層。 在一些範例中,施加於BLT及SLT的閘極電壓可能小 於10V,其可能造成較少的閘極干擾。若BLT及SLT的閘 極;I電層可此被充電或陷獲電荷時,額外的_Vg抹除可施 加於BLT或SLT之閘極,以使其閘極介電層放電。 再參考圖,各BLT可與一位元線(BL)耦合。在一 φ 範例中,BL可為具有與STI相同或近似相同間距的金屬 線。同樣地,各SLT係連接至一源極線(SL)。源極線係與 WL平行且連接至用於讀取感測之感測放大器。源極線可 為—金屬(例如鎢),或多晶矽線,或一擴散N+摻雜線。As disclosed above, the memory cell structure in accordance with the present invention is suitable for both NOR and NAND type flash memories. Additional examples of memory programming designs and methods of operation thereof will be described below. Without departing from the scope of the present invention by the specific structure described below, various operational methods of the memory array in accordance with the present invention will be described hereinafter for an exemplary NAND architecture. As described above, an n-channel SONONOS memory element having a 〇N〇 tunnel dielectric can be used for the _memory element. An example of a NAND array architecture. Figure 2 and (10) are cross-sectional views of the exemplary memory technicians from two different directions. In some examples, the memory array can be operated by +FN stylization, self (four) reset/erase, and read methods. In addition, circuit operation methods may be included in some examples to avoid stylized interference. In addition to the early block gate structure, the gods are set again. In addition to f, an array of spm-gates may be used, such as an array of coffee cells using S〇N〇N〇S elements between the two poles adjacent to the source/no-polar regions. In this example, the split gate design can be adjusted to reduce the number of fingers, and these components can be designed to obtain the coupling effect between the dynamic gates, or - "second"; sinfulness to reduce or remove the ocean #达达] ° As revealed above, —SONON〇s, the hidden parts can be used for excellent self-receiving erase and vt distribution control. Moreover, close, assist fan & erase the twist to make a compact erase state The distribution can be beneficial to multiple 681939-27U3 30 1309875 quasi-applications (MLC). By using some designs as memory array structures, the effective channel length (Leff) can be expanded to reduce or eliminate short channel effects. These examples do not use a diffusion junction to avoid the challenge of providing shallow junctions or using pocket implants during memory cell processing. Figure 1 shows an example of a memory element with a SONONOS design. Examples of different layers of material and their thickness. In some examples, a P+ polysilicon gate can be used to provide a lower saturation reset/erase voltage Vt, which can be achieved by reducing gate implants. Figures 22a and 22b show a memory. An example of a volume array, such as having The S〇n〇n〇S-NAND array of memory cells according to the specific embodiment of Table 1 has a diffusion junction. In an example, the separated components can be isolated from each other by various isolation techniques, such as By isolation technique using shallow trench isolation (STI) or insulator on-board (SOI). Referring to Figure 22a, a memory array can include multiple bit lines (e.g., BL1 and BL2), and multiple word lines ( In addition, the array may include a source line transistor (or source line select transistor or SLT) and a bit line transistor (or bit line select transistor or BLT). For example, the memory cells in the array can be designed using SONONOS, and the SLT and BLT can include n-type MOSFETs (NMOSFET). Figure 22b shows an exemplary layout of a memory array such as a NAND array. Referring to Fig. 22b, the Lg is the channel length of the memory cell, and the Ls is the space between the separation lines of the memory element. In addition, the w is the channel of the memory cell, and the Ws is separated by the bit line or the source/no pole. The isolation area of the interval is 681939-27U3 31 1309875 degrees ' it is in one An example may be the STI width. Referring again to Figures 22a and 22b, the memory elements may be connected in series and form a NAND array. For example, a string of memory elements may include 16 or a magic memory element 'providing a number of strings of 16 or 32. BLT may be used And SLT as a selection transistor to control the corresponding SfAND string. In one example, the gate dielectric for BLT 'and SLT can be a hafnium oxide layer that does not include a tantalum nitride trapping layer. In some examples (although not necessarily required in all cases) φ may avoid possible vt shifts of BLT and SLT during memory array operation. Another option, BLT and SLT, can be used as a gate dielectric layer by combining a plurality of layers of ONONO layers. In some examples, the gate voltage applied to the BLT and SLT may be less than 10V, which may result in less gate interference. If the gates of BLT and SLT; the I electrical layer can be charged or trapped, an additional _Vg erase can be applied to the gate of the BLT or SLT to discharge its gate dielectric. Referring again to the figures, each BLT can be coupled to a bit line (BL). In a φ paradigm, BL can be a metal line having the same or approximately the same pitch as the STI. Similarly, each SLT is connected to a source line (SL). The source line is parallel to the WL and is connected to a sense amplifier for read sensing. The source line can be a metal (e.g., tungsten), or a polysilicon line, or a diffused N+ doped line.

• 圖23a顯示一範例性記憶體陣列(如SONONOS-NAND -把憶體陣列)沿通道長度方向的斷面圖。通常,Lg& Ls近 似等於F,其-般表示一元件(或節點)之關鍵尺寸。關鍵尺 寸可隨著用於製造的技術而變化。例如,F=5Q奈米代表使 用50奈米節點。圖23b顯示範例性記憶體陣列(如 681939-27U3 32 1309875 S〇N〇N〇S_NAND記憶體陣⑴沿通道寬0向的斷面 圖。參考圖23b,通道寬度方向的間距近似等於或稍大於 通道長度方向中的間距。因此,一記憶單元的 4F2/單元。 在製造記憶體陣列(諸如上揭陣列)的範例巾,該等過 -程可能有關僅使用二主要遮罩或微影钮刻過程,諸=其一 用於多晶石夕(字元線)且另-用於STI(位元線)。反之,臓d 奉型洋動閘極元件之製造可能需要至少二多晶發處理及另一 多晶石夕ΟΝΟ間處理。因此,所揭元件的結構及製程可比該 等NAND型浮動閘極記憶體更簡單。 參考圖23a,在一範例中,字元線(WL)間之空間(Ls) 可形成有淺接面(如N+摻雜區的淺接面),其可作為記憶元 件之源極或汲極區。如圖23a中顯示,彳實行額外植入及/ 或擴散過程(例如斜角的袋狀植入),以提供鄰近一或多個 淺接面區之接面的一或多個「袋狀」區或袋狀延伸。在一 鲁 些範例中,此組態可提供較佳的元件特徵。 在其中sti係用於隔離分離記憶元件之範例中,sti _區的溝渠深度可大於p井中之空泛寬度,尤其係當所用的 接面偏壓被提升得更高時。例如,接面偏壓可高達7V,用 於程式化禁止的位元線(程式化期間未選擇的位元線)。在 -一範例中,STI區之深度可在200至400奈米的範圍中。 在記憶體陣列製成後,可在記憶體陣列的其他操作前 先執行重設操作以使Vt分布緊湊。圖24a顯示此操作之範 例。在一範例中,在其他操作開始前,首先可施加VG=約 681939-27U3 33 1309875 _7V且P井=+8V以重設陣列(VG和P井之電壓降可分到閘 極電壓進入各WL和p井中)。在RESET期間,bl可浮動, 或挺升到與p井相同的電壓。如圖24b中顯示,重設操作 可提供極佳自收斂性質。在一範例中,即使—開始將 SONONOS元件充電至各種Vt’此重設操作可使其「緊湊」 至重設/抹除狀態。在一範例中,重設時間係約1〇〇毫秒。 在該範例中,記憶體陣列可使用具有 埃之 η 通道 SONONOS 元件,其具 有Lg/W=0.22/0.16微米之N+多晶矽閘極。 ' 一般而言,傳統浮動閘極元件係無法提供自收斂抹 除。反之,SONONOS元件可用收斂重設/抹除方法操作。 在-些_巾,為減Vt分转常由於蚊製程問題(諸 如過程不-致性或電漿充電效應)而在相當廣的範圍中,此 操作可能變得十分重要。範例性自收斂「重設」可協助使 記憶兀件的初始Vt分布範圍緊湊或變窄。• Figure 23a shows a cross-sectional view of an exemplary memory array (such as SONONOS-NAND - array of memory) along the length of the channel. In general, Lg & Ls is approximately equal to F, which generally represents the critical dimensions of a component (or node). The critical dimensions can vary with the technology used for manufacturing. For example, F = 5Q nm represents the use of a 50 nm node. Figure 23b shows a cross-sectional view of an exemplary memory array (e.g., 681939-27U3 32 1309875 S〇N〇N〇S_NAND memory array (1) along the channel width 0 direction. Referring to Figure 23b, the channel width direction spacing is approximately equal to or slightly larger than The spacing in the length direction of the channel. Therefore, the 4F2/cell of a memory unit. In the case of manufacturing a memory array (such as an array of images), the over-travel may be related to using only two main masks or lithography buttons. Process, one for polycrystalline stone (word line) and the other for STI (bit line). Conversely, the fabrication of 臓d-type galvanic gate components may require at least two polycrystalline processing And another polycrystalline stone processing. Therefore, the structure and process of the disclosed components can be simpler than the NAND type floating gate memory. Referring to Figure 23a, in an example, between the word lines (WL) The space (Ls) can be formed with a shallow junction (such as the shallow junction of the N+ doped region), which can serve as the source or drain region of the memory element. As shown in Figure 23a, the 彳 performs additional implantation and/or diffusion. Process (eg, beveled pocket implant) to provide proximity to one or more One or more "pocket" areas or pocket extensions of the junction of the junction areas. In some examples, this configuration provides better component characteristics. The sti is used to isolate the isolated memory elements. In the example, the ditch depth of the sti _ zone can be greater than the vacant width of the p well, especially when the junction bias used is raised higher. For example, the junction bias can be as high as 7V for stylized forbidden bits. Meta-line (bit line not selected during stylization). In an example, the depth of the STI region can be in the range of 200 to 400 nm. After the memory array is fabricated, other memory arrays can be used. The reset operation is performed before the operation to make the Vt distribution compact. Figure 24a shows an example of this operation. In an example, before other operations start, VG = about 681939-27U3 33 1309875 _7V and P well = +8V can be applied first. To reset the array (the voltage drop of the VG and P wells can be divided into the gate voltage into each of the WL and p wells). During RESET, bl can float, or rise to the same voltage as the p-well, as shown in Figure 24b. The reset operation provides excellent self-convergence properties. In one example, even— Begin charging the SONONOS components to various Vt's. This reset operation can make it "compact" to the reset/erase state. In one example, the reset time is about 1 millisecond. In this example, the memory array An η-channel SONONOS component with Å can be used with an N+ polysilicon gate of Lg/W=0.22/0.16 μm. 'Generally, conventional floating gate components are not available. Self-convergence erase. Conversely, SONONOS components can be operated with a convergence reset/erase method. In some cases, the Vt conversion is often due to mosquito process problems (such as process non-induced or plasma charging effects). This operation can become very important in a fairly wide range. An exemplary self-convergence "reset" can help to make the initial Vt distribution of the memory element compact or narrow.

在程式化操作之範例中,已逻宗沾 .^ + 選疋的WL·可用高電壓施 加(例如約+ 16V至+20V之電壓),以弓I發通道側注入。 其他PASS閘極(其他未選定WL) 虫士 2丨议G7 J加从開啟以在一 NAND 串中引發反轉層。.程式化在 法。在-範例中,平行程式化方法諸‘ T為低功羊方 承分百而沪4·,儿 次4如以4Κ位元組單元 千仃頁面輊式化,可使程式化通 干 同時總電流消耗可控制在lmA内里^至多於腦版C, 在其他肌巾之料針擾,—高些範财,為避免 施加於其他BL,以致反轉層電位^4(如約7V之電壓)可 更高以抑制在未選定 681939-27U3 1309875 BL(例如圖25中單元B)中的電壓降。 在讀取操作的範例中,已選定WL可提升至在一抹除 狀態位準(EV)及一程式化狀態位準(PV)間之電壓。其他WL 可作為「PASS閘極」,以致其閘極電壓可提升至高於pv 之電壓。在一些範例中,抹除操作可與上述重設操作類似, 其可允許自收斂至相同或類似重設Vt。 圖25顯示操作記憶體陣列之範例。程式化可包括通道 +FN電子注入進入SON〇N〇S氮化物陷獲層。一些範例可 包括施加Vg=約+18V至已選定WLN-1,且施加vg=約 + 10V至其他WL以及BLT〇SLT可關閉以避免在單元B中 之通道熱電子注入。在此範例中,因為在NAND串中的所 有電晶體被開啟’此反轉層穿過該等串。此外,因為Bl 1 係接地’ BL1中之反轉層具有零電位。另一方面,其他bl 提升至高電位(如約+7V之電壓)’以致其他BL的反轉層的 電位較高。 尤其係對於單元A(其係選定程式化的單元),電壓降係 約+ 18V,故造成+FN注入。並且Vt可提升到pv。至於單 元B,電壓降係+11V,造成少許多的+FN注入,因為fn 注入係對Vg敏感。至於單元C,僅施加+ 10V,造成沒有 或可忽略的+FN注入。在一些範例中,程式化操作不限於 已說明之技術。換句話說’可應用其他適當程式化抑制技 術。 圖24a、26及27進一步顯示陣列操作的一些範例,且 顯示一些範例的耐久及保持性質。如舉例,在一些操作循 681939-27U3 35 1309875 環後的元件退化可保持極小。圖24a顯示範例性抹除操 作,其可與重設操作類似。在一範例中,抹除係由扇區或 區塊執行。如上揭,該等記憶元件可具有良好自收斂抹除 性質。在一些範例中,抹除飽和Vt可取決於Vg。例如, 較高的Vg可造成較高的飽和Vt。如圖26中所示,收斂時 間可約10到100毫秒。 圖27顯示讀取操作的範例。在一範例中,讀取可藉由 施加在一抹除狀態Vt(EV)及一程式化狀態Vt(PV)間之閘 極電壓而執行。例如,閘極電壓可為約5V。另一方面,其 他WL及BLT和SLT係用一更高的閘極電壓(如約+9V)施 加,以開啟所有其他記憶單元。在一範例中,若單元A的 Vt比5V高,讀取電流可能極小(<0.1uA)。若單元A的Vt 比5V低,讀取電流可能較高(>0.1uA)。結果,可識別記憶 體狀態(即已儲存的資訊)。 在一些範例中,用於其他WL的通過閘極電壓應高於 高Vt狀態或程式化狀態Vt,但不要太高以免觸發閘極干 擾。在一範例中,PASS電壓係在約7至10V的範圍内。 BL處的施加電壓可為約IV。儘管較大讀取電壓可引發更 多電流,但讀取干擾在一些範例可能變得更明顯。在一些 範例中,感測放大器可放在源極線(源極感測)上或一位元 線上(汲極感測)。 NAND串的一些範例可具有每串8、16或32個記憶元 件。一較大的NAND串可節省更多額外負擔且增加陣列效 率。然而,在一些範例中,讀取電流可能較小且干擾可能 681939-27U3 36 1309875 變得更明顯。因此,應基於各種設計、製造及操作因子選 擇NAND串的適當數目。 、 圖28顯示某些範例性元件的循環耐久性。參考圖, 可貫行具有+FN程式化及-FN抹除的p/E循環,並且結果 顯示良好的耐久特徵。在此範例中,抹除條件係Vg=約_16v 達毫秒。在一些範例中,僅需要單次抹除並且並不必要 狀態的驗證。記憶體Vt窗口良好而無退化。 圖29a及29b顯示使用不同標度的範例性記憶元件的 IV特徵。尤其係圖29a中顯示元件的小擺動退化,並且圖 29b顯示元件的小跨導退化。圖3〇顯示一範例性s〇n〇n〇s 元件的保持特徵。參考圖30,藉由對於在ιοκ循環後且在 至 離開200小時後之元件具有少於1 〇〇mV之電荷損失而 提供良好保持。圖30亦顯示在高溫處之可接受電荷損失。 在一些範例中’分裂閘極設計(例如分裂閘極 SONONOS-NAND設計)可用來達成記憶體陣列的更進— 步按比例縮小。圖31顯示使用此設計之範例。參考圖31, 可縮小各字元線間、或共享相同位元線的二相鄰記憶元件 間之空間(Ls)。在一範例中,Ls可縮小到約3〇奈米或更少。 如範例中’使用分裂閘極設計之記憶元件沿相同位元線可 能僅共旱一源極區或一汲極區。換句話說,對於一些記憶 凡件而言,分裂閘極SONONOS-NAND陣列可不使用擴散 區或接® (❹摻雜區)。在—範例中,該設計亦可減少 或免除淺接面及鄰近「袋狀」的需要,其在一些範例中可 此涉及更複雜的製程。此外,在—些範例巾,該設計較少 681939-27U3 37 1309875 受短通道效應的影響,因為已增加通道長度,諸如在一範 例中增加到Lg=2F-Ls。 圖32顯示一使用分裂閘極設計之記憶體陣列的範例 性製程。該示意圖僅係示範性範例,並且該記憶體陣列可 以各種不同方法設計及製造。參考圖32,在形成用於提供 記憶元件之多層材料後,可使用一氧化矽結構作為形成於 該等層上之硬遮罩將該等層圖案化。例如,可藉由微影及 蝕刻過程以界定該等氧化矽區。在一範例中,用於界定初 始氧化矽區之圖案可具有約F的寬度且氧化矽區間之空間 約F,產生約2F之間距。在圖案化初始氧化矽區後,氧化 矽間隔件可接著形成,以圍繞已圖案化區而擴大各氧化矽 區且窄化其間距。 再次參考圖32,在形成氧化矽區後,其等被用作硬遮 罩以界定或圖案化其底層以提供一或多個記憶元件,如同 多個NAND串。此外,絕緣材料(例如氧化矽)可用來填充 相鄰記憶元件間之空間,例如圖32顯示的空間Ls。 在一範例中,沿相同位元線之相鄰記憶元件間的空間 Ls可在約15奈米到約30奈米的範圍中。如上述,在此範 例中,有效通道長度可擴大到2F-LS。在一範例中,若F 係約30奈米且Ls係約15奈米,則Leff係約45奈米。對 於該等範例性記憶元件的操作,閘極電壓可減少到15V以 下。此外,字元線間之多晶矽間電壓降可經設計成不大於 7V,以避免在Ls空間中之間隔件崩潰。在一範例中,此 可藉由在相鄰字元線間具有少於5MV/cm之電場而達到。 681939-27U3 38 1309875 用於習知NAND浮動閘極元件之擴散接面的Leff係其 閘極長度的大約一半。相反地,在一範例中,若F係約5〇 奈米並且Leff係約3〇奈米,Leff係所建議設計(分裂閘極 NAND)的大、約80奈米。更長的Leff可藉由減少或免除短 通道效應的影響而提供更佳的元件特徵。 如上述’分裂閘極的NAND設計可進一步縮小相同位 元線之相鄰圮憶單元間之空間(Ls)。反之,傳統nand型 =動閘,的元件可能不提供小間距,因為浮動閘極間輕合 =可砲失去記憶體窗口。當相鄰浮動閘極間的耦合電容 冋時’ e動閘極_合係相鄰記憶單元間之干擾(浮動閉極 2的工間〗、’以致相鄰浮動閘極間的耦合電容極高,使得 貝^干擾A生)。如上揭’該設計可消除製造—些擴散接面 之南要’並且若開啟所有字元線麻轉層可直接連接。因 此,該設計可簡化記憶元件的製程。 i述包括結構化設計、陣列設計及記憶元件 例,可提供符合需求之陣列尺寸、優良可靠 於按尺其任何的結合。所述之—些範例亦可應用 閃—己㈣^小非揮發性㈣記憶體的尺寸,例如似勵快 於資料應用之快閃記憶體。某些範例可提供 t1钱2^電洞穿隧抹除的sc)nqnqs元件。 某二犯例亦可提供記憶元件 過抹㈣_〜 的良敎且減少難以抹除或 示的問《°同樣地’可提供良In the example of stylized operation, the WL of the selected device can be applied with a high voltage (for example, a voltage of about +16V to +20V), and injected into the channel side. Other PASS Gates (Other Unselected WLs) 2 Insects G7 J Plus is turned on to induce an inversion layer in a NAND string. Stylized in law. In the example, the parallel stylized method of 'T is a low-powered sheep, and the other is 4, and the fourth is like a 4Κ byte unit, which can be programmed to make the program The current consumption can be controlled within the lmA ^ to more than the brain version C, in the other muscle towel material interference, - higher fan, in order to avoid application to other BL, so that the inversion layer potential ^ 4 (such as about 7V voltage) It can be higher to suppress the voltage drop in the unselected 681939-27U3 1309875 BL (eg, cell B in Figure 25). In the example of a read operation, the selected WL can be boosted to a voltage between an erased state level (EV) and a stylized state level (PV). The other WL can be used as a "PASS gate" so that its gate voltage can be raised to a voltage higher than pv. In some examples, the erase operation can be similar to the reset operation described above, which can allow self-convergence to the same or similar reset Vt. Figure 25 shows an example of operating an array of memory. Stylization can include channel +FN electron injection into the SON〇N〇S nitride trapping layer. Some examples may include applying Vg = about +18V to selected WLN-1, and applying vg = about +10V to other WLs and BLT〇SLT can be turned off to avoid channel hot electron injection in cell B. In this example, because all of the transistors in the NAND string are turned on, the inverted layer passes through the strings. Further, since the inverted layer in the Bl 1 grounding 'BL1 has a zero potential. On the other hand, the other bl is raised to a high potential (e.g., a voltage of about +7 V) so that the potential of the inversion layer of the other BL is higher. In particular, for cell A, which is a selected stylized cell, the voltage drop is about +18V, resulting in a +FN implant. And Vt can be promoted to pv. As for cell B, the voltage drop is +11V, resulting in much less +FN implant because the fn implant is sensitive to Vg. As for cell C, only +10V is applied, resulting in no or negligible +FN injection. In some examples, the stylization operation is not limited to the techniques described. In other words, other suitable stylization suppression techniques can be applied. Figures 24a, 26 and 27 further show some examples of array operation and show some examples of durability and retention properties. As an example, component degradation after some cycles of 681939-27U3 35 1309875 can be kept to a minimum. Figure 24a shows an exemplary erase operation that can be similar to the reset operation. In one example, the erase is performed by a sector or block. As noted above, the memory elements can have good self-convergent erase properties. In some examples, erasing saturation Vt may depend on Vg. For example, a higher Vg can result in a higher saturation Vt. As shown in Fig. 26, the convergence time can be about 10 to 100 milliseconds. Figure 27 shows an example of a read operation. In one example, the reading can be performed by applying a gate voltage between a erased state Vt (EV) and a stylized state Vt (PV). For example, the gate voltage can be about 5V. On the other hand, other WL and BLT and SLT systems are applied with a higher gate voltage (e.g., about +9V) to turn on all other memory cells. In an example, if the Vt of cell A is higher than 5V, the read current may be extremely small (<0.1uA). If the Vt of cell A is lower than 5V, the read current may be higher (>0.1uA). As a result, the state of the memory (i.e., the stored information) can be identified. In some examples, the pass gate voltage for other WLs should be higher than the high Vt state or the stylized state Vt, but not too high to trigger gate interference. In one example, the PASS voltage is in the range of about 7 to 10V. The applied voltage at BL can be about IV. Although larger read voltages can induce more current, read disturb may become more apparent in some examples. In some examples, the sense amplifier can be placed on the source line (source sense) or on a bit line (drain sense). Some examples of NAND strings can have 8, 16, or 32 memory elements per string. A larger NAND string saves additional overhead and increases array efficiency. However, in some examples, the read current may be small and the interference may become more noticeable 681939-27U3 36 1309875. Therefore, the appropriate number of NAND strings should be selected based on various design, fabrication, and operational factors. Figure 28 shows the cycle durability of certain exemplary components. Referring to the figure, a p/E cycle with +FN stylization and -FN erasing can be performed and the results show good endurance characteristics. In this example, the erase condition is Vg = about _16v for milliseconds. In some examples, only a single erase is required and verification of the state is not necessary. The memory Vt window is good without degradation. Figures 29a and 29b show IV features of exemplary memory elements using different scales. In particular, the small wobble degradation of the display elements in Figure 29a, and Figure 29b shows the small transconductance degradation of the elements. Figure 3A shows the retention characteristics of an exemplary s〇n〇n〇s element. Referring to Figure 30, good retention is provided by having a charge loss of less than 1 〇〇mV for the component after the ιοκ cycle and after 200 hours of leaving. Figure 30 also shows the acceptable charge loss at high temperatures. In some examples, a 'split gate design (e.g., a split gate SONONOS-NAND design) can be used to achieve a further step-by-step scaling of the memory array. Figure 31 shows an example of using this design. Referring to Fig. 31, the space (Ls) between two adjacent memory elements between the word lines or sharing the same bit line can be reduced. In one example, Ls can be reduced to about 3 nanometers or less. As in the example, a memory element using a split gate design may only have a source region or a drain region along the same bit line. In other words, for some memory parts, the split gate SONONOS-NAND array can be used without the diffusion region or the ❹ (❹ doped region). In the example, the design also reduces or eliminates the need for shallow joints and adjacent "pockets", which in some examples may involve more complex processes. In addition, in some sample cases, the design is less 681939-27U3 37 1309875 due to the short channel effect because the channel length has been increased, such as increasing to Lg = 2F-Ls in an example. Figure 32 shows an exemplary process for a memory array using a split gate design. This schematic is merely an exemplary example, and the memory array can be designed and fabricated in a variety of different ways. Referring to Fig. 32, after forming a multilayer material for providing a memory element, the layers may be patterned using a niobium oxide structure as a hard mask formed on the layers. For example, the yttrium oxide regions can be defined by lithography and etching processes. In one example, the pattern used to define the initial yttrium oxide region may have a width of about F and a space of about F of the yttrium oxide region, resulting in a distance of about 2F. After patterning the initial yttrium oxide region, a hafnium oxide spacer can then be formed to expand the respective hafnium oxide regions around the patterned regions and narrow the spacing thereof. Referring again to Figure 32, after forming the hafnium oxide region, it is used as a hard mask to define or pattern its underlying layer to provide one or more memory elements, such as a plurality of NAND strings. In addition, an insulating material such as hafnium oxide can be used to fill the space between adjacent memory elements, such as the space Ls shown in FIG. In one example, the space Ls between adjacent memory elements along the same bit line can range from about 15 nanometers to about 30 nanometers. As described above, in this example, the effective channel length can be expanded to 2F-LS. In one example, if F is about 30 nm and Ls is about 15 nm, then Leff is about 45 nm. For the operation of these exemplary memory elements, the gate voltage can be reduced to less than 15V. In addition, the polysilicon voltage drop between the word lines can be designed to be no greater than 7V to avoid spacer collapse in the Ls space. In an example, this can be achieved by having an electric field of less than 5 MV/cm between adjacent word lines. 681939-27U3 38 1309875 Leff for the diffusion junction of a conventional NAND floating gate element is about half of its gate length. Conversely, in one example, if F is about 5 nanometers and Leff is about 3 nanometers, Leff is the recommended design (split gate NAND) of about 80 nanometers. Longer Leffs provide better component characteristics by reducing or eliminating the effects of short channel effects. The NAND design of the split gate as described above can further reduce the space (Ls) between adjacent memory cells of the same bit line. Conversely, the traditional nand type = the gate, the components may not provide a small pitch, because the floating gates are lightly combined = the cannon loses the memory window. When the coupling capacitance between adjacent floating gates is ', the e-gate _ is the interference between adjacent memory cells (the work of floating closed-pole 2, 'so that the coupling capacitance between adjacent floating gates is extremely high So that the shell ^ interferes with A). As noted above, the design eliminates the need for manufacturing - some of the diffusion junctions and can be directly connected if all of the word lines are turned on. Therefore, this design simplifies the process of memory components. The description includes structured design, array design and memory components, which can provide an array size that meets the requirements, and is excellent and reliable in any combination of scales. The above-mentioned examples can also be applied to the size of the flash-small (four)^small non-volatile (four) memory, such as flash memory that is faster than the data application. Some examples provide sc)nqnqs components that are t1 money 2^ hole tunneling erased. A second case can also provide a memory element to over-improve (4) _~ and reduce the difficulty of erasing or showing the question "° the same place" can provide good

循環後之小退化以;3僖*卞行傲諸如在P/E 内之元件—致性而不會I 杈仏。己體陣列 /、有不%疋的位元或單元。再者, 681939-27U3 1309875Small degradation after the cycle; 3僖*卞 proud of components such as in P/E—not II. Array of own /, with no bits or units of 疋. Furthermore, 681939-27U3 1309875

〜些範例可經由分裂閘極NAN 特徵,其可名印产叹冲美供良好短通道元件 、了在圮feTC件操作期間提供更 有關本發明之較佳罝體•以的關裕度。 及說明之目的。並非毫之前揭内容,係供例示 精確形式。:習St:遺漏或欲限制本發明為所揭露之 施例進行’m不上:者應即瞭射對上述各項具體實 應瞭解本發㈣錢於^其廣狀發雜概念。因此, 蓋歸屬如後載各請求項所定體實施例,而係為涵 傅。 、7疋義之本發明精神及範圍内的修 【圖式簡單說明】 前揭覽:遠:可f佳瞭, 圖式裏圖、%有現屬較佳'、、、達本發明之5兒明目的,各 並不限於所緣之精破排具體實施例。然應瞭解本發明 在各圖式中:*置方式及設備裝置。 記憶^ ^及^分別係依據本發明一具體實施例的N通道 之斷而疋&依據本發明—具體實施例的p通道記憶單元 研m不意圖; 各種:式系依據本發明之一具體實施例的隨道介電結構在 圖^化方法下之臨限電壓(電荷陷獲容量)的圖示; 輩开夕的系依據本發明之—具體實施例的SONONOS記憶 限電壓在抹除期間隨時間改變的圖示; 元之的ΡΡ Ϊ依據本發明1體實施例的S〇NC>NC)S記憶單 —在保軸間_間改變的圖示; 681939-27U3 40 1309875 圖5a-5e係依據本發明各種具體實施例的ΟΝΟ隧道介 電結構之能帶圖; 圖6係用於三種不同隧道介電結構之電洞穿隧電流相 對於電場強度的圖示; 圖7a係依據本發明一具體實施例的記憶單元在各種 - 類型之程式化後的抹除期間隨時間改變之臨限電壓的圖 示; φ 圖7b係依據本發明一具體實施例具有一鉑閘極的記 憶單元在抹除期間隨時間改變之臨限電壓的圖示; 圖7c及7d係有關圖7b中之記憶單元的電容相對於電 壓的圖示; 圖8係依據本發明一具體實施例的記憶單元在各種操 作條件下於許多程式化/抹除循環過程中的臨限電壓之圖 示; 圖9係依據本發明一具體實施例的記憶單元在1循環 Φ 和ίο3循環後之電流-電壓(iv)關係圖示; 圖10係依據本發明一具體實施例的記憶單元在一組 程式化及抹除條件下於許多程式化/抹除循環過程中的臨 限電壓之圖示; - 圖11係依據本發明一具體實施例的記憶單元在VG加 - 速保持測試下之臨限電壓隨時間改變的圖示; 圖12a及12b分別係依據本發明一具體實施例的記憶 單元之虛擬接地陣列的等效電路圖及布局圖; 圖13係圖12b中所示依據本發明一具體實施例的記憶 681939-27U3 41 1309875 單元之虛擬接地陣列沿線12B-12B取得的斷面示意圖; 〇圖Ma及Mb係包含依據本發明—具體實施例的記憶 單元之記憶體陣列的等效電路圖,且描述依據本發明之操 作的二具體實施例之適合的重設/抹除電壓; 圖15a及15b係包含依據本發明一具體實施例的記憶 早元之記憶體陣列的等效電路圖,其描述依據本發明程 化之一方法; „ 一圖⑽及16b係包含依據本發明-具體實施例的記憶 早凡之記憶體陣列的等效電路圖,其描述依據 —位元之方法; 貝% 抹J二Ϊ依據本發明—具體實施例的記憶單元在各種 矛'“下隨時間變化的臨限電壓圖示; 程式I / 本^明—具體纽例的記料元在許多 圖19a=%過程中的臨限電壓之圖示; 元,在各種m 1%係依據本發明一具體實施例的記憶單 線性標度的^電壓下肢極處之1流分縣對數標度及 陣列Ξ等包括依據本發明一具體實施例的記憶單元之 法.、 路圖,其描述依據本發明程式化一位元的方 陣列的布Z 21b係依據本發明—具體實施例之虛擬接地 、局圖和等效電路圖; 置开I ΝΛ= 22b分別係依據本發明一具體實施例的記憶 W AND_料效電關及布局圖; 681939-27U3 42 1309875 單—圖23a及23b分別係依據本發明一具體實施例的記憶 凡之nand陣列沿圖22b中所示線22A_22A及22B_22b 取得的斷面圖; · 24a係依據本發明一兴篮施例的ρ 效雷败固 圖’其描述依據本發明之操作方法; 具體實施例的NAND陣列之 對一圖24b係依據本發明一具體實施例在重設操作期間針Some examples can be based on the split gate NAN feature, which can be used to name a good short channel component, and provide a better margin for the preferred body of the present invention during operation of the TCfeTC device. And the purpose of the description. It is not intended to reveal the precise form. :SSt: omission or desire to limit the invention to the disclosed embodiment. The person should be able to understand the specifics of the above (4) money in the wide-ranging concept. Therefore, the cover belongs to the embodiment of the request, and is embossed. 7 疋 之 本 本 本 本 本 本 本 本 本 本 本 本 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前 前For the purpose of the disclosure, each embodiment is not limited to the specific embodiment. However, it should be understood that the present invention resides in various drawings: * means and apparatus. The memory ^ ^ and ^ are respectively according to an embodiment of the present invention, the N-channel is broken and the P-channel memory unit according to the present invention - the specific embodiment is not intended; A schematic diagram of the threshold voltage (charge trapping capacity) of the associated dielectric structure of the embodiment under the method of the present invention; the SONONOS memory limit voltage of the embodiment according to the present invention is during the erasing period Illustrated with time; 〇 Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ Ϊ 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 681 An energy band diagram of a germanium tunnel dielectric structure in accordance with various embodiments of the present invention; FIG. 6 is a graphical representation of hole tunneling current versus electric field strength for three different tunnel dielectric structures; FIG. 7a is a diagram in accordance with the present invention A graphical representation of the threshold voltage of a memory cell of a particular embodiment over time during the erasing of various types of stylization; φ Figure 7b is a memory cell having a platinum gate in accordance with an embodiment of the present invention. In addition to changing over time Figure 7c and 7d are diagrams showing the capacitance versus voltage of the memory cell of Figure 7b; Figure 8 is a diagram of a plurality of stylized memory cells in various operating conditions in accordance with an embodiment of the present invention. FIG. 9 is a diagram showing a current-voltage (iv) relationship of a memory cell after a cycle of φ and ίο3 according to an embodiment of the present invention; FIG. 10 is based on the present invention. A diagram of a threshold voltage of a memory unit in a set of stylization and erasing conditions during a number of stylization/erasing cycles; and FIG. 11 is a memory unit in accordance with an embodiment of the present invention. FIG. 12a and 12b are respectively an equivalent circuit diagram and a layout diagram of a virtual ground array of a memory unit according to an embodiment of the present invention; FIG. 13 is a diagram showing a change of a threshold voltage with time under a VG plus-speed hold test; FIG. 12b is a cross-sectional view of the virtual ground array of the memory 681939-27U3 41 1309875 unit taken along line 12B-12B in accordance with an embodiment of the present invention; FIGS. Ma and Mb are included in accordance with the present invention - An equivalent circuit diagram of a memory array of memory cells of the embodiment, and describing suitable reset/erase voltages for two embodiments in accordance with the operation of the present invention; Figures 15a and 15b are included in accordance with an embodiment of the present invention. An equivalent circuit diagram of a memory array of memory early, which describes one of the methods according to the present invention; „ a diagram (10) and 16b includes an equivalent circuit diagram of an early memory memory array according to the present invention. The description is based on the method of the bit; the method of the present invention is based on the threshold voltage of the memory unit according to the present invention in the various spears of the present invention; the program I / this ^ - specific The icon of the new example is shown in a number of threshold voltages in the process of Figure 19a = %; in various m 1%, according to a memory of a specific embodiment of the present invention, the voltage of the lower extremity 1 stream county logarithmic scale and array Ξ, etc. including a memory unit according to an embodiment of the present invention. A road diagram describing a fabric Z 21b of a square array of one-dimensional elements according to the present invention is in accordance with the present invention. - specific implementation Virtual ground, local map and equivalent circuit diagram; set I ΝΛ = 22b are respectively memory W AND_ material efficiency and layout according to an embodiment of the present invention; 681939-27U3 42 1309875 single - Figure 23a and 23b A cross-sectional view taken along lines 22A_22A and 22B_22b shown in FIG. 22b, respectively, according to an embodiment of the present invention; 24a is a ρ-effect thunder-fixation diagram according to a preferred embodiment of the present invention. It describes a method of operation in accordance with the present invention; a pair of NAND arrays of a particular embodiment is shown in Figure 24b during a reset operation in accordance with an embodiment of the present invention.

带—具有不同初始臨限電壓的記憶單元隨時間改變之臨限 電壓的圖示; 圖25係依據本發明一具體實施例的操作方法之 電路圖; 圖26係依據本發明一具體實施例的記憶單元之臨限 壓在各種抹除條件下隨時間改變的圖示; 圖27係描述依據本發明一具體實施例的操作方法之 等效電路圖; 圖28係依據本發明一具體實施例的記憶單元在一組 • 程式化及抹除條件下於許多程式化/抹除循環過程中的臨 限電壓之圖示; 圖29a及29b係在依據本發明—具體實施例的記憶單 元,在各種閘極電壓下於汲極處之電流在三不同循環數目 - 處分别依照對數標度及線性標度的圖示; 圖係依據本發明一具體實施例的記憶單元之臨限 電壓在三不同溫度和循環條件下於保持期間隨時間變化的 圖示; 圖31係依據本發明一具體實施例的NAND陣列字元 681939-27U3 43 1309875 線之斷面示意圖;及 圖32係依據本發明一具體實施例的NAND陣列字元 線形成技術之斷面示意圖。 【主要元件符號說明】 100 η通道記憶單元 101 Ρ型基體 102 Ν型摻雜區 104 η型掺雜區 106 通道區 120 隧道介電結構 122 下方薄氧化層 124 小電洞穿隧阻障高度氮化層 126 上方薄氧化層 130 電荷陷獲/電荷儲存層 140 絕緣層 150 閘極 200 ρ通道記憶單元 201 η型基體 202 Ρ型摻雜區 204 Ρ型換雜區 206 通道區 220 隨道介電結構 222 下方薄氧化層 224 小電洞穿隧阻障高度氮化層 681939-27U3 44 1309875FIG. 25 is a circuit diagram of an operational method in accordance with an embodiment of the present invention; FIG. 26 is a memory diagram in accordance with an embodiment of the present invention; FIG. 26 is a circuit diagram of a method for operating a memory unit having different initial threshold voltages; FIG. 27 is an equivalent circuit diagram depicting an operation method in accordance with an embodiment of the present invention; FIG. 28 is a memory unit in accordance with an embodiment of the present invention; FIG. Illustration of threshold voltages during a number of stylization and erasing cycles in a set of stylization and erasing cycles; Figures 29a and 29b are memory cells in accordance with the present invention - in various gates The current at the voltage at the drain is at a number of three different cycles - in accordance with the logarithmic scale and the linear scale, respectively; the threshold voltage of the memory cell according to an embodiment of the invention is at three different temperatures and cycles Figure 31 is a cross-sectional view of a NAND array character 681939-27U3 43 1309875 line in accordance with an embodiment of the present invention; 32 a schematic sectional view of line art NAND array of word lines is formed based on a specific embodiment of the present invention. [Main component symbol description] 100 η channel memory unit 101 Ρ type substrate 102 Ν type doped region 104 η type doped region 106 channel region 120 tunnel dielectric structure 122 underlying thin oxide layer 124 small hole tunneling barrier height nitridation Thin oxide layer 130 above layer 126 charge trapping/charge storage layer 140 insulating layer 150 gate 200 ρ channel memory unit 201 n-type substrate 202 Ρ-type doping region 204 换-type replacement region 206 channel region 220 channel dielectric structure 222 Lower thin oxide layer 224 Small hole tunneling barrier High nitride layer 681939-27U3 44 1309875

226 上方薄氧化層 230 電荷陷獲/電荷儲存層 240 絕緣層 250 閘極 681939-27U3 45226 Upper thin oxide layer 230 Charge trapping/charge storage layer 240 Insulation layer 250 Gate 681939-27U3 45

Claims (1)

鬌 lJ〇987s7 '申請專利範園: .、種記憶單元,其包含: —半導體基體,其 〜汲桎區,· 、由—通道區分離的一源極區 —隧道介電結構,发/ 電結構包括複數層,其係设置於該通道區上,該隧道 電洞穿隨阻障高度之了5亥隨道介電結構包含具有- 7荷儲存層,其係:置:; —絕緣層,其係毁置於於该隧道介電結構上; —閘極電極,其係执㈠亥電荷儲存層上;及 電極包含a 置於該絕緣岸j·., =—功函數值大於Ν + ^層上’其中該閘極 …其中,該具複數芦、、矽的材料 二=7存結構、==構:口電荷穿 結構層之隨道介電結構之後並陷獲二電;:: 如請求項 如請求項丨之:二,其"閉極電極包含銘。 少二介電層道介電結構Mi 如請求項丨之』夕達約奈米之厚度。 弟—虱化矽層、一在嗲 、"電結構包含一 層、及IP 夕層上之第石々 , 在该第一氮化矽層上之第-& 虱化矽=求項1之記憶單元,i中該以:層。 亂主匕矽、八喊及Hf〇2,组成之族尹選了储存層包括從由 如請求項i 出的至少— 之讀早元,其”絕緣層 p ' ' ' II— I l_ I _ _ ~ i p年//月相修(更)正替換Mj ^— 一 介 2. 4. 5. 681939-27U3 46 *1309875鬌lJ〇987s7 'Application for patent garden: ., a memory unit, which comprises: - a semiconductor substrate, its ~ 汲桎 region, · a source region separated by a channel region - tunnel dielectric structure, hair / electricity The structure includes a plurality of layers disposed on the channel region, the tunnel hole passing through the barrier height, and the dielectric structure comprises a storage layer having a -7 load, which is: an insulating layer, The ruin is placed on the dielectric structure of the tunnel; - the gate electrode is held on (1) the charge storage layer; and the electrode contains a placed on the insulating bank j., = - the work function value is greater than the Ν + ^ layer Where 'the gate is ... wherein the material with a plurality of reeds, bismuth = 7 storage structure, = = structure: the charge passes through the intervening dielectric structure of the structural layer and traps the second electricity;:: as requested The item is as requested: Second, its "closed electrode contains the name. The second dielectric layer dielectric structure Mi is as thick as the thickness of the request. Brother-虱化矽层,一在嗲,"Electrical structure consists of a layer, and the first stone layer on the IP layer, the first-&虱 矽==1 Memory unit, i should be: layer. The chaotic master, the eight shouts, and the Hf〇2, the composition of the family chose the storage layer to include at least the read element from the request item i, the "insulation layer p ' ' ' II - I l_ I _ _ ~ ip year / / month phase repair (more) is replacing Mj ^ - one 2. 4. 5. 681939-27U3 46 *1309875 8. 如。月求項1之記憶單元’其中該隨道介带 可忽略之陷獲致率。 电 種5己憶體陣列,其包含複數個如請求項 元。 ' 結構具有一 1之記憶單 9. 10. 11 12. 13, =請求項8之記憶體_,其巾職數個 至少二記憶單元係藉由一淺溝準早兀中 隔離中至少-者魏絲 _及—絕緣物切 如請求項8之記憶體陣列’其中該記憶體陣列包含至 少::字元線、至少二條位元線及至少_條源極線。 如Μ求項10之記憶體陣列,其中該記憶體陣列包含至 少ϋ線選擇電晶體’其_合至—對應位元線。 如印求項10之記憶體陣列,其中該記憶體陣列包含至 )-源極線選擇電晶體,其軸合至—對應源極線。 如請求項10之記憶體陣列,其中該基體包含至少一對 用於該記憶元件之淺接面。8. For example. The memory unit of the month 1 is the negligible trapping rate. A battery 5 array of memories, comprising a plurality of such as request elements. ' Structure has a memory list of 1. 9. 10. 12. 12. 13, = memory of request 8 _, the number of its jobs is at least two memory cells are separated by at least one of the shallow ditch Weiss _ and - the insulator is cut as in the memory array of claim 8 wherein the memory array comprises at least: a word line, at least two bit lines, and at least _ source lines. For example, the memory array of item 10, wherein the memory array includes at least a 选择 line selection transistor ' _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The memory array of claim 10, wherein the memory array comprises a source-selective transistor that is coupled to the corresponding source line. The memory array of claim 10, wherein the substrate comprises at least one pair of shallow junctions for the memory element. 681939-27U3 47681939-27U3 47
TW95100248A 2005-01-03 2006-01-03 Non-voltaile memory cells, memory arrays including the same and methods of operating cells and arrays TWI309875B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US64022905P 2005-01-03 2005-01-03
US68923105P 2005-06-10 2005-06-10

Publications (2)

Publication Number Publication Date
TW200625550A TW200625550A (en) 2006-07-16
TWI309875B true TWI309875B (en) 2009-05-11

Family

ID=45072125

Family Applications (1)

Application Number Title Priority Date Filing Date
TW95100248A TWI309875B (en) 2005-01-03 2006-01-03 Non-voltaile memory cells, memory arrays including the same and methods of operating cells and arrays

Country Status (1)

Country Link
TW (1) TWI309875B (en)

Also Published As

Publication number Publication date
TW200625550A (en) 2006-07-16

Similar Documents

Publication Publication Date Title
TWI306669B (en) Non-volatile memory cells, memory arrays including the same and methods of operating cells and arrays
JP5712420B2 (en) Nonvolatile memory cell, memory array having the same, and cell and array operating method
EP1603137B1 (en) Program/erase method for p-channel charge trapping memory device
US7688626B2 (en) Depletion mode bandgap engineered memory
CN101964209A (en) Method of operating P-channel non-volatile memory device
TWI297154B (en) Non-volatile memory cells, memory arrays including the same and methods of operating cells and arrays
US20060202261A1 (en) Non-volatile memory cells, memory arrays including the same and methods of operating cells and arrays
US20060202252A1 (en) Non-volatile memory cells, memory arrays including the same and methods of operating cells and arrays
US20060198189A1 (en) Non-volatile memory cells, memory arrays including the same and methods of operating cells and arrays
EP1408511A1 (en) Single bit nonvolatile memory cell and methods for programming and erasing thereof
WO2008122919A1 (en) A memory cell, a memory array and a method of programming a memory cell
JP4907173B2 (en) Nonvolatile memory cell, memory array having the same, and cell and array operating method
WO2003044869A1 (en) Transistor and semiconductor memory comprising it
TWI309875B (en) Non-voltaile memory cells, memory arrays including the same and methods of operating cells and arrays
TWI506769B (en) Silicon on insulator and thin film transistor bandgap engineered split gate memory
CN101587898B (en) Semiconductor structures with integrated circuit elements and methods of forming and operating the same
TWI298494B (en) Non-volatile memory cells, memory arrays including the same and methods of operating cells and arrays
JP2011139081A (en) Nonvolatile memory cell, memory array having the same, and method of operating the cell and the array
JP2006236424A (en) Nonvolatile memory device and charge injection method thereof