TWI338375B - Memory unit structure and operation method thereof - Google Patents

Memory unit structure and operation method thereof Download PDF

Info

Publication number
TWI338375B
TWI338375B TW96107469A TW96107469A TWI338375B TW I338375 B TWI338375 B TW I338375B TW 96107469 A TW96107469 A TW 96107469A TW 96107469 A TW96107469 A TW 96107469A TW I338375 B TWI338375 B TW I338375B
Authority
TW
Taiwan
Prior art keywords
doped region
layer
memory cell
substrate
germanium
Prior art date
Application number
TW96107469A
Other languages
Chinese (zh)
Other versions
TW200837963A (en
Inventor
Chao I Wu
Original Assignee
Macronix Int Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Macronix Int Co Ltd filed Critical Macronix Int Co Ltd
Priority to TW96107469A priority Critical patent/TWI338375B/en
Publication of TW200837963A publication Critical patent/TW200837963A/en
Application granted granted Critical
Publication of TWI338375B publication Critical patent/TWI338375B/en

Links

Landscapes

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

Description

P950202 22692twfdoc/n 九、發明說明: 【發明所屬之技術領域】 本發明是有關於一種記憶單元結構(memory unit structure),且特別是有關於一種在操作後降低擺動劣化 (swing degradation)衝擊(impact)的記憶單元結構及其操作 方法。 【先前技術】 目前的非揮發性記憶體產品中,具有可進行多次資料 之存入、讀取、抹除等動作,且可在一個記憶單元中進行 二位元(2 bit)操作的SONOS記憶單元,已成為個人電腦和 電子設備所廣泛採用的一種記憶體元件。 一般而言,SONOS記憶單元是採用一層電荷捕捉層 (charge trapping layer)取代習知快閃記憶體的多晶矽浮置 閘極(floating gate) ’並在這種電荷捕捉層上下通常各有一 層氡化矽層,以形成由氧化矽/氮化矽/氧化矽(ON〇)層所構 成之堆疊式結構(stacked structure)。再者,於0N0層兩側 基底中會有源極與汲極,而在0Ν0層上則設有閘極。 由於傳統的SONOS記憶單元之0Ν0層中的底氧化層 (bottom oxide)是直接利用熱氧化法形成在基底上,所以其 與石夕晶片之基底間是低的界面捕捉密度(丨nterface trap density,Dit) ’約】〇wcm-2eV·〗,但是這個Dit值被發現會 隨§己憶單元的循環次數增加而逐漸增加,如圖1所示,其 中不管是最初抹除狀態(initial erase state)還是最初程式化 狀態(initial program state)都與循環一萬次的抹除狀態及程 P950202 22692twfdoc/n 式化狀態騎率差異極大。因料致 perf〇rma㈣)降低’進而影響記憶單元的摔作^^ (cycle end腦ce)以及資料保持性㈣邮 衣持久性 【發明内容】 本發明提供一種記憶單元結構, 擺動效能。 # 了 ^物操作後的 本發=提供—種記憶單元的操作方法,可對記憶單 凡作一位元而持久性的操作。 心… 電壓種記憶單元的操作方法,崎決啟始 以降低在操作後擺 本發明另提供一種記憶單元結構, 動劣化之衝擊。 本發明又提供—種記憶單元的操作方法, 單元的資料保持性。 本發明再提供—種記憶單元的操作方法, 位元操作且可改善記憶單元之概持久性。 可維持記憶 以進行單一 本發明提出—種記憶單元,包括基底、位於石夕 層捕捉層、分別位於捕捉層兩側㈣基底内的 一與第二摻雜區、位於捕捉層上的一個閘極、位於閘極 2?層?間的—層第—氧化層、位於石夕基底與捕捉層之 層间界面捕捉密度((interface trap density,Dit)材料 :以及位於高界面捕捉密度材料層與捕捉層之間的一層第 氧化層,其中高界面捕捉密度材料層與矽基底之間的界 面捕捉密度在1〇丨】〜loWv-1之間。 P950202 22692twfdoc/n 在本發明之一實施例中,上述高界面捕捉密度材料層 與矽基底之間的界面捕捉密度為。 在本發明之一實施例中,上述高界面捕捉密度材料層 的厚度在10〜70埃之間。 在本發明之一實施例中,上述高界面捕捉密度材料層 的材料包括氮化石夕。 在本發明之一實施例中,上述高界面捕捉密度材料層 的材料包括二氧化姶(Hf〇2)、二氧化錯(Zr〇2)、氮氧化锆 (ZrOxNy)、氮氧化銓(Hf〇xNy)、矽酸铪(H⑸Α)、矽酸锆 (△sixoy)、氮氧矽铪(Hfsix〇yNz)、三氡化二鋁(A丨2〇3)、二 氧化鈦(Ti〇2)、五氧化鈕(Ta2〇5)、三氧化二鑭(La2〇3)、二 氡化鈽(Ce〇2)、矽酸鉍(BLtSbO]2)、氧化鎢(w〇3)、氧化釔 (Υζ〇3)、铭酸鑭(LaAl〇3)、鈦酸鋇鏍(% χδΓχΤί〇3)、鈦酸鋇 (BaTi03)、鍅酸鉛(pbZr03)、鈕酸銳鉛(pbSCzTai z〇3,簡稱 PST)、铌酸鋅鉛(pbZi^Nb^O3 ’簡稱PZN)、锆鈦酸鉛 (PbZr〇3_PbTi〇3 ’ 簡稱 PZT)或鈮酸鎂鉛(PbMgzNbi z〇3,簡 稱 PMN) 〇 在本發明之一實施例中,上述矽基底是p型矽基底, 且第一摻雜區與第二摻雜區型摻雜區。 本發明又提出一種記憶單元的操作方法,適用於上述 在问界面捕捉密度材料層與矽基底之間的界面捕捉密度 (Dn)為1〇11〜1〇13心^之間的記憶單元。其操作方法包 括.程式化記憶單元時,於閘極上施加第一正電壓,於第 二摻雜區施加第二正電壓,並使第一摻雜區為〇伏特,以 P950202 22692twfd〇c/n 利用通道熱電子(channel hot electron,CHE)方式程式化記 憶單元之一側的位元;抹除記憶單元時,於閘極上施加一 第一負電壓,於第二摻雜區施加一第三正電壓,並使第一 摻雜區為0伏特,以利用價帶對價帶熱電洞(band-to-band tunneling hot hole,BTBTHH)方式抹除記憶單元之該側的 位元。 本發明再提出一種記憶單元的操作方法,適用於上述 在高界面捕捉密度材料層與矽基底之間的界面捕捉密度 (Dit)為1011〜i〇Bcm-2evi之間的記憶單元。其操作方法包 括.程式化記憶單元時’於閘極上施加一第四正電壓,並 使第一摻雜區與第二摻雜區為〇伏特’以利用富勒_諾德漢 (Fowler-Nordheim ’ FN)方式程式化所述記憶單元;而在抹 除記憶單元時,於閘極上施加一第二負電壓,並使第一摻 雜區與第二摻雜區為〇伏特,以利用富勒-諾德漢(FN)方式 抹除所述記憶單元。 在本發明之一實施例中,上述第一摻雜區是源極、第 一推雜區是汲極。 在本發明之一實施例中,上述第一摻雜區是汲極、第 二摻雜區是源極。 在本發明之一實施例中,上述第一正電壓大於第二正 電壓。 本發明另提出一種記憶單元,包括一個矽基底、位於 矽^底上的一層捕捉層、分別位於捕捉層兩側的矽基底内 的第與第二摻雜區、位於捕捉層上的一個閘極、位於閘 P950202 22692twfdoc/n 極該捕捉層之_-層[介電層以及位於 層之間的-層第二介,其巾第二介電層細基底^ 的界面捕捉密度(Dit)在ι〇η〜i〇13cm-2eV-i之間。 在本發明之另-實施例中,上述第二介電廣與石夕 之間的界面捕捉密度為l〇12cm-2eV-i。 _ 在本發明之另-實施例中,上述第二介電層包括氧化 盾0 在本發明之另-實施财’上述雜底是p型石夕基 底,且第-摻雜區與第二摻雜區是n型摻雜區。 土 在本發明之另-實施例中,上述第一介電層包括氧化 層。 本發明又提出-航憶單元哺作方法,翻於上述 t第二ί電層與碎基底之間的界面捕捉錢⑽)為 〜j〇 cp2eV-i之間的記憶單元。其操作方法包括:程 ,化°己隐單元時,於閘極上施加第一正電壓,於第二摻雜 區施加第二正電壓’並使第-摻雜區為0伏特,以利用通 子(CHE)方式程式化記憶單元之一側的位元;抹除 ^己憶j時,於開極上施加—第-負電壓,於第二摻雜區 ί加"ΐί二正電壓,並使第一推雜區$ 0伏特,以利用價 帶對價帶熱t洞(BTBTHH)料雜誠單狀該側的位 元。 γ么月再提出一種記憶單元的操作方法,適用於上述 二二Ϊ電f與矽基底之間的界面捕捉密度_為 10〜1013cnr2eV-i之間的記憶單元。其操作方法包括:程 P950202 22692twfdoc/n 2 = ”時,於閘極上施加—第四正電壓,並使第一 品”弟一摻雜區為〇伏特,以利用 方式程式化記鮮元^在齡〜/莫(FN) ^ )你外陆。己隐早兀時,於閘極上施 弟一負電壓’並使第一摻雜區與第二換雜區為0伏 特,以利用富勒-諾德漢(FN)方式抹除記憶單元。 f本發明之P實施射,上述第—摻雜是源極、 第一推雜區是〉及極。 从_在本發明之另—實施例中,上述第—摻雜區是汲極、 弟一推雜區是源極。 在本發明之另一實施例中,上述第一正電壓大於第二 正電壓。 、一 本發明因為將矽基底與其上層之間的界面設為在 〜lOin^eV1之間的高界面捕捉密度(Dit),所以當操 作循環次數逐漸增加後,其擺動效能能可維持在一定的程 度,因此記憶單元的操作、循環持久性以及資料保持性都 會儘可能的維持在可容許的範圍内,而不至於使記憶單元 失效。 為讓本發明之上述特徵和優點能更明顯易僅,下文特 舉較佳實施例’並配合所附圖式,作詳細說明如下。 【實施方式】 以下將隨所附圖式來更充分地描述本發明之實施 例。不過,本發明尚可以多種不同形式來實踐,且不應將 其解釋為限於說明書所陳述之實施例。而且,在圖式中, 為明確起見可能誇示各層以及區域的尺寸,而未按照實際 P950202 22692twfdoc/n 比例繪示。 用:ίΪ:書中所使用之用語僅是為描述以下的應 的「第-」、「第二」等用語,只是用來將某—d 部分與另-區域、層或部分作區別,並不 = 或是操作順序。 /、〜欣m斤 第一實施例 谨、隹圖-是本發明之第一實施例之一種記憶單元結 構進订-位兀%式化操作時的剖面示意圖。 请參照圖2A,第—實施例的記憶單元包括—個石夕基 底200、-層捕捉層2〇2、第一摻雜區2〇4a與第二捧雜區 204b、-個閘極寫、-層第一氧化層2〇8、一層高界面捕 捉始、度(high-Dit)材料層21〇以及一層第二氧化層212。在 本實施例中,矽基底200是p型矽基底,第一摻雜區2〇如 與第二摻雜區204b是n型摻雜區。而捕捉層2〇2是位於矽 基底200上、第一與第二摻雜區2〇知和2〇牝則分別是位 於捕捉層202兩側的矽基底2〇〇、閘極206位於捕捉層2〇2 上、第一軋化層208則位於閘極206與捕捉層202之間。 至於高界面捕捉密度材料層210是位於矽基底200與捕捉 層202之間,第二氧化層212則位於高界面捕捉密度材料 層210與捕捉層202之間,其中高界面捕捉密度材料層21〇 與石夕基底200之間的界面捕捉密度(interface trap density, Dit)在ΙΟ11〜l〇13cm-2evi之間;較佳為lOimiv-丨。此外, 高界面捕捉密度材料層210的厚度譬如是在1〇〜7〇埃之 1338375 P9 50202 22692twfdoc/nP950202 22692twfdoc/n IX. Description of the Invention: [Technical Field] The present invention relates to a memory unit structure, and in particular to a method of reducing a swing degradation impact after operation (impact Memory cell structure and its operation method. [Prior Art] Currently, non-volatile memory products have SONOS that can perform multiple data storage, reading, erasing, etc., and can perform two-bit (2 bit) operation in one memory unit. The memory unit has become a memory component widely used in personal computers and electronic devices. In general, a SONOS memory cell is a polysilicon floating gate that replaces a conventional flash memory with a charge trapping layer and has a layer of deuteration above and below the charge trapping layer. The tantalum layer is formed to form a stacked structure composed of a tantalum oxide/tantalum nitride/yttria (ON〇) layer. Furthermore, the source and the drain are in the substrate on both sides of the 0N0 layer, and the gate is provided on the 0Ν0 layer. Since the bottom oxide layer in the 0Ν0 layer of the conventional SONOS memory cell is directly formed on the substrate by thermal oxidation, it has a low interface trap density (丨nterface trap density) between the substrate and the base of the Shixi wafer. Dit) 'about】 〇 wcm-2eV ·〗, but this Dit value is found to increase with the number of cycles of the hex element, as shown in Figure 1, which is the initial erase state (initial erase state) Or the initial program state is very different from the 10,000-time erase state and the P950202 22692twfdoc/n state ride rate. According to the material, perf〇rma(4)) is reduced, which in turn affects the memory unit's fall end (ceend end ce) and data retention (4) mailing durability. SUMMARY OF THE INVENTION The present invention provides a memory cell structure, swing performance. #了 ^ After the operation of the object = provide - the operation method of the memory unit, can be a one-dimensional and persistent operation on the memory. The heart... The operation method of the voltage type memory unit, the start of the process to reduce the pendulum after the operation. The present invention further provides a memory cell structure, the impact of dynamic degradation. The invention further provides an operation method of the memory unit, and data retention of the unit. The present invention further provides a method of operating a memory cell that operates in a bit and improves the persistence of the memory cell. The memory can be maintained for a single memory unit, including a substrate, a trap layer in the layer, a first and second doped regions respectively located on the sides (4) of the capture layer, and a gate on the capture layer. Located on the 2nd floor of the gate? The inter-layer-oxide layer, the interface trap density (Dit) material at the base of the Shixia substrate and the trap layer: and a layer of oxide layer between the high-interface capture density material layer and the capture layer Wherein the interface capture density between the high interface capture density material layer and the germanium substrate is between 1 and LoWv-1. P950202 22692twfdoc/n In one embodiment of the invention, the high interface captures the density material layer and The interface capture density between the ruthenium substrates is. In one embodiment of the invention, the high interface capture density material layer has a thickness between 10 and 70 angstroms. In one embodiment of the invention, the high interface capture density is The material of the material layer includes nitrite. In one embodiment of the present invention, the material of the high interface-capture material layer includes cerium oxide (Hf〇2), dioxin (Zr〇2), zirconium oxynitride ( ZrOxNy), bismuth oxynitride (Hf〇xNy), bismuth ruthenate (H(5) Α), zirconium ruthenate (Δsixoy), bismuth oxynitride (Hfsix〇yNz), bismuth hydride (A丨2〇3), Titanium dioxide (Ti〇2), pentoxide button (Ta2〇5), trioxide镧(La2〇3), bismuth telluride (Ce〇2), bismuth ruthenate (BLtSbO)2), tungsten oxide (w〇3), cerium oxide (Υζ〇3), lanthanum strontium (LaAl〇3) , barium titanate (% χδΓχΤί〇3), barium titanate (BaTi03), lead citrate (pbZr03), lead acid sharp lead (pbSCzTai z〇3, referred to as PST), zinc citrate lead (pbZi^Nb^O3 'PZN for short), lead zirconate titanate (PbZr〇3_PbTi〇3 'abbreviated as PZT) or lead magnesium niobate (PbMgzNbi z〇3, abbreviated as PMN) 〇 In one embodiment of the invention, the ruthenium substrate is p-type 矽a substrate, and a first doped region and a second doped region doped region. The present invention further provides a method for operating a memory cell, which is suitable for the interface capture density between the density interface material layer and the germanium substrate at the interface ( Dn) is a memory unit between 1 〇 11 〜 1 〇 13 ^ ^. The operation method includes: when the memory unit is programmed, a first positive voltage is applied to the gate, and a second positive voltage is applied to the second doped region. And the first doped region is 〇Vot, and the channel hot electron (CHE) is used to program the bit on one side of the memory cell by using P950202 22692twfd〇c/n; erasing In the cell, a first negative voltage is applied to the gate, a third positive voltage is applied to the second doped region, and the first doped region is 0 volts to utilize the valence band valence band thermal hole (band-to- The band tunneling hot hole (BTBTHH) method erases the bit on the side of the memory unit. The invention further proposes a method for operating a memory cell, which is suitable for the memory cell between the high interface capture density material layer and the germanium substrate having an interface capture density (Dit) of 1011~i〇Bcm-2evi. The method of operation includes: when programming the memory cell, 'applying a fourth positive voltage to the gate and making the first doped region and the second doped region 〇 volts' to utilize Fowler-Nordheim (Fowler-Nordheim) The 'FN) mode stylizes the memory cell; and when the memory cell is erased, a second negative voltage is applied to the gate, and the first doped region and the second doped region are volts to utilize the Fuller - The Nordhan (FN) mode erases the memory unit. In an embodiment of the invention, the first doped region is a source and the first doped region is a drain. In an embodiment of the invention, the first doped region is a drain and the second doped region is a source. In an embodiment of the invention, the first positive voltage is greater than the second positive voltage. The invention further provides a memory unit comprising a germanium substrate, a capture layer on the bottom of the substrate, first and second doped regions in the germanium substrate respectively located on both sides of the capture layer, and a gate on the capture layer. In the gate P950202 22692twfdoc/n pole of the capture layer _-layer [dielectric layer and layer between the layers of the second layer, the second dielectric layer of the substrate, the interface density (Dit) in the ι 〇η~i〇13cm-2eV-i between. In another embodiment of the present invention, the interface capture density between the second dielectric broad and the stone eve is l〇12cm-2eV-i. In another embodiment of the present invention, the second dielectric layer comprises an oxidized shield 0. In the present invention, the above-mentioned hetero-substrate is a p-type stellite substrate, and the first doped region and the second doped region The impurity region is an n-type doped region. Soil In another embodiment of the invention, the first dielectric layer comprises an oxide layer. The invention further proposes a method for feeding the aeronautical memory unit, wherein the interface between the second electric layer and the broken substrate captures money (10) is a memory unit between ~j〇 cp2eV-i. The method includes the following steps: applying a first positive voltage to the gate, applying a second positive voltage to the second doped region and making the first doped region 0 volts to utilize the passer. (CHE) mode stylizes one of the bits on the side of the memory cell; when erasing ^jj, applying a -first-negative voltage on the open-pole, and adding a positive voltage to the second doped region The first doping area is $0 volt, in order to utilize the valence band to measure the hot t-hole (BTBTHH) material. γ Mouyue proposes a method of operating a memory cell, which is suitable for the memory cell between the above-mentioned interface and the germanium substrate f-density_between 10~1013cnr2eV-i. The operation method includes: when the process P950202 22692twfdoc/n 2 = ”, the fourth positive voltage is applied to the gate, and the first product is doped to be a volt, so as to use the way to program the fresh element ^ Age ~ / Mo (FN) ^) You are outbound. When it is already early, a negative voltage is applied to the gate and the first doped region and the second doped region are 0 volts to erase the memory cell using the Fuller-Nordheim (FN) method. f is the P of the present invention, wherein the first doping is the source, the first doping region is > and the pole. In another embodiment of the invention, the first doped region is a drain, and the dipole region is a source. In another embodiment of the invention, the first positive voltage is greater than the second positive voltage. According to the invention, since the interface between the ruthenium substrate and the upper layer is set to a high interface capture density (Dit) between 〜10in^eV1, the swing performance can be maintained at a certain level when the number of operation cycles is gradually increased. To the extent that the operation of the memory unit, cycle persistence, and data retention are maintained as far as practicable, without causing the memory unit to fail. The above features and advantages of the present invention will become more apparent from the following detailed description. [Embodiment] Hereinafter, embodiments of the present invention will be described more fully with the accompanying drawings. However, the invention may be practiced in many different forms and should not be construed as limited to the embodiments set forth herein. Moreover, in the drawings, the dimensions of the various layers and regions may be exaggerated for clarity, and are not drawn to the actual P950202 22692twfdoc/n ratio. Use: Ϊ: The terms used in the book are only used to describe the following terms such as "--" and "second", but only to distinguish a-d part from another-area, layer or part, and Not = or the order of operations. The first embodiment is a schematic cross-sectional view of a memory cell structure in the first embodiment of the present invention. Referring to FIG. 2A, the memory unit of the first embodiment includes a stone substrate 200, a layer capture layer 2, a first doped region 2〇4a and a second doping region 204b, and a gate write. a layer of first oxide layer 2〇8, a layer of high interface capture high-Dit material layer 21〇 and a second oxide layer 212. In the present embodiment, the germanium substrate 200 is a p-type germanium substrate, and the first doping region 2, for example, and the second doping region 204b are n-type doping regions. The capture layer 2〇2 is located on the germanium substrate 200, the first and second doped regions 2 are known, and the second and second doped regions 2 are respectively located on both sides of the capture layer 202, and the gate 206 is located on the capture layer. The upper and first rolling layers 208 are located between the gate 206 and the capture layer 202. As for the high interface capture density material layer 210 is located between the germanium substrate 200 and the capture layer 202, the second oxide layer 212 is between the high interface capture density material layer 210 and the capture layer 202, wherein the high interface captures the density material layer 21〇 The interface trap density (Dit) between the stone and the base substrate 200 is between ΙΟ11~l〇13cm-2evi; preferably lOimiv-丨. In addition, the thickness of the high interface capture density material layer 210 is, for example, 1 375 〇 7 〇 之 1338375 P9 50202 22692 twfdoc/n

間;較佳為30埃。至於高界面捕捉密度材料層210的材料 可為氮化矽;抑或,二氧化铪(Hf02)、二氧化锆(Zr〇2)、氮 氧化鍅(ZrOxNy)、氮氧化铪(HfOxNy) '矽酸铪(HfSix〇y)、 矽酸锆(ZrSixOy)、氮氧矽铪(HfSix〇yNz)、三氧化二紐 (AI2O3) 一氧化欽(Ti〇2)、五氧化组(丁&2〇5)、三氧化二鑭 (La203)、二氧化鈽(ce〇2)、矽酸鉍(Bi4si2〇12)、氧化鎢 (W03)、氧化釔(γ2〇3)、鋁酸鑭(LaA1〇3)、鈦酸鋇鳃 (Ba,—xSrxTi〇3)、鈦酸鋇(BaTi〇3)、鍅酸叙(pbZr〇3)、鈕酸筑 鉛(PbSCzTakO〗’簡稱 pst)、鈮酸鋅鉛(PbZnzNbl z〇3,簡 柄PZN)、鍅鈦酸鉛(PbZr〇3_pbTi〇3,簡稱ρζτ)與鈮酸鎂 紹(PbMgzNbi_z〇3 ’ 簡稱 ρμν)其中之一。 Μ迤頊麥照圖2A,當對本實施例之記憶單元進行如 圖2Α之二位元程式化的操作時,可於閘極2〇6上施加 二正電壓(如Vg,伏特),於第二推雜區2〇扑(此時 為汲極)施加第二正電壓(如Vd=5伏特),並使第—Between; preferably 30 angstroms. The material of the high interface capture density material layer 210 may be tantalum nitride; or, hafnium oxide (Hf02), zirconium dioxide (Zr〇2), niobium oxynitride (ZrOxNy), niobium oxynitride (HfOxNy) 'tannic acid铪(HfSix〇y), zirconium citrate (ZrSixOy), oxynitride (HfSix〇yNz), bismuth oxide (AI2O3), oxidized bismuth (Ti〇2), pentoxide group (Ding & 2〇5 ), antimony trioxide (La203), cerium oxide (ce〇2), bismuth ruthenate (Bi4si2〇12), tungsten oxide (W03), cerium oxide (γ2〇3), lanthanum aluminate (LaA1〇3) , barium titanate (Ba, -xSrxTi〇3), barium titanate (BaTi〇3), bismuth acid (pbZr〇3), lead acid (PbSCzTakO) 'pst), zinc lead citrate (PbZnzNbl) Z〇3, simple handle PZN), lead barium titanate (PbZr〇3_pbTi〇3, abbreviated as ρζτ) and magnesium citrate (PbMgzNbi_z〇3 'abbreviated as ρμν). Referring to FIG. 2A, when the memory unit of the embodiment is operated by the two-digit stylization of FIG. 2, a positive voltage (eg, Vg, volt) can be applied to the gate 2〇6. The second push zone 2 (when the bungee is applied) applies a second positive voltage (such as Vd = 5 volts) and makes the first

204a(此時可作為源極)為v㈣伏特,以利用通道妖電; k annd hot electron ’ CHE)方式程式化記憶單元之一 通常!基底200是接地,所以其 極、第二_ £ 2Q4lf ’ t第—摻雜區204a也可作為汲 ❹第—正電壓、在_σ第二正上 大於第二正電壓。 且上述弟—正電壓 圖2β則是圖2八之記憶單元 結構進行二位元抹除操 作204a (which can be used as the source at this time) is v (four) volts to utilize the channel demon power; k annd hot electron ' CHE) way to program one of the memory cells usually! The base 200 is grounded, so its pole, the second _ £ 2Q4lf ' The t-doped region 204a may also be a 汲❹-positive voltage and a second positive voltage at _σ second positive. And the above-mentioned brother - positive voltage Figure 2β is the memory cell structure of Figure 2-8 for two-bit erase operation

12 P950202 22692twfd〇c/n 時的剖面示意圖。 請參照圖2B,當對本實施例之記憶單元進行二位元 抹除操作時,需要在閘⑯2%上施加—第一負電壓(如 vg 10伏特)’於第—摻雜區之隱施加—第三正電壓(如12 P950202 22692twfd 〇c / n schematic diagram. Referring to FIG. 2B, when the memory cell of the embodiment is subjected to the two-bit erase operation, it is required to apply a first negative voltage (eg, vg 10 volts) on the gate 162% to the hidden application of the first doped region. Third positive voltage (eg

Vd 5伏特)’並使第一摻雜區2〇4a為〇伏特(如伏 特)’以利用價帶對價帶熱電洞(band*band _eiing h〇t hole ’ BTBTHH)方式抹除記憶單元之—側位元(即没極側位 元)。 除I一位兀钿作之外,第一實施例的記憶單元結構還 可利用單一位元操作的方式進行,如圖2匚與圖2D所示。 士圖2C疋圖2A之记憶單元結構進行單一位元程式化操 作日的剖面示意圖。請參照圖兀,當對本實施例之記憶單 :進行單一位元程式化操作時,需要在於閘極2〇6上施加 —第四正電壓(如Vg=20伏特),並使第一摻雜區2〇如與第 二摻雜區204b為0伏特(Vs=0伏特、Vd=〇伏特),以利用 富勒-諾德漢(Fowler-Nordheim,FN)方式程式化所述記憶單 元。 圖2D是圖2A之記憶單元結構進行單一位元抹除操 作時的剖面示意圖。請參照圖2D,當對本實施例之記憶單 凡進行單一位元抹除操作時,需於閘極2〇4上施加一第二 負電壓(如Vg=-20伏特)’並使第—摻雜區2〇乜與第二摻 雜區2〇4b為0伏特(Vs=0伏特、VdW伏特),以利用富勒 -諾德漢(FN)方式抹除所述記憶單元。 為證實第一實施例的記憶單元確實能在多次循環後 〆· 13 P950202 22692twfdoc/n 仍維持敎的擺動效能,請參考圖3A與圖3β。 圖3A是圖2A之^»,|•立…-仏 作最初與經-萬五千(1^Γ^進行二位元程式化操Vd 5 volts) 'and the first doped region 2 〇 4a is 〇 volt (such as volts)' to erase the memory cell by means of a band*band _eiing h〇t hole 'BTBTHH' Side bit (ie, no side bit). In addition to the one-bit operation, the memory cell structure of the first embodiment can also be performed by a single bit operation, as shown in Figs. 2A and 2D. Figure 2C is a cross-sectional view of the memory cell structure of Figure 2A for a single bit stylized operation day. Referring to the figure, when performing the single bit stylization operation on the memory sheet of the embodiment, it is necessary to apply a fourth positive voltage (such as Vg=20 volts) on the gate 2〇6 and make the first doping. The region 2, for example, and the second doped region 204b are 0 volts (Vs = 0 volts, Vd = 〇 volts) to program the memory cells using the Fowler-Nordheim (FN) approach. Figure 2D is a cross-sectional view showing the memory cell structure of Figure 2A in a single bit erase operation. Referring to FIG. 2D, when a single bit erase operation is performed on the memory of the embodiment, a second negative voltage (such as Vg=-20 volts) is applied to the gate 2〇4 and the first doping is performed. The impurity region 2 〇乜 and the second doping region 2 〇 4b are 0 volts (Vs = 0 volts, VdW volts) to erase the memory cell by using a Fuller-Nordheim (FN) method. To confirm that the memory unit of the first embodiment can maintain the oscillating performance of 敎·13 P950202 22692 twfdoc/n after multiple cycles, please refer to FIG. 3A and FIG. 3β. Fig. 3A is the ^»,|•立...-仏 of Fig. 2A, and the first two-dimensional stylized operation with the -500 thousand (1^Γ^)

*㈣θ扣曰 )三萬(3叹)次循環後的1-V 曲線圖。圖3Β是圖2Β之記憶單元* (4) θ buckle 曰 ) 1-V curve after 30,000 (3 s) cycles. Figure 3Β is the memory unit of Figure 2

作最初與經15Κ次、3GK 士鮮w T f Κ凡抹_ ^闰叩叮, μ κ_人循5衣後的i-v曲線圖。從圖3A ":U官疋圖3Α中的最初程式化狀態(initial program s ate還疋圖3B *的最初抹除狀態㈣^卜職 ^的=i^15K ^和胤次的程式化狀態以及抹除 在夕.夕換句話說,第一實施例的記憶單元能 在夕次㈣循職⑽持敎的擺動效能。 第一實施例 n圖_4A是細本發明之第二實_之—種記憶單元結 行一位元程式化操作時的剖面示意圖。 π參照圖4A’第二實施例的記憶單元包括—個石夕基 二400、-層捕捉層術、第一捧雜區她與第二捧雜區 人 個閘極400、—層第—介電層4〇8以及-層第二 二二層410 〇在本實施例+,石夕基底柳是p型石夕基底, 、,雜區4G4a與第二摻雜區4Q4b是n型摻雜區其中 介電層408例如是氧化層。至於,第二介電層41〇 =仓土底4〇0之間則具有高界面捕捉(high interface trap, 101; 特f的界面412,其界面捕捉密度(Dit)在 二〜1〇⑽eV之間;較佳為H)】W2eV-】。此外,第二 1 ’層410可以疋氧化層;舉例來說,可選擇用最差的熱 U3S375 P950202 22692twfdoc/n ,化製程或者在熱氧化製程後祕人方式(impla福ion)使 1面412的Dlt介於10】]〜lOUcn^eV·1之間。而在本實施 ,中’ ^述石夕基底40(M列如是p型石夕基底,且第一換雜區 〇4a與第二摻雜區4〇4b例如是打型換雜區。The initial i-v curve after 15 times, 3GK, fresh w T f Κ 抹 _ 闰叩叮 μ μ μ μ 。 。 。 。 。 。 。. From the initial stylized state in Figure 3A ":U official map 3Α (initial program s ate also in the initial erase state of Figure 3B * (four) ^ ^ ^ ^ ^ ^ 15K ^ and the stylized state of the order And in the evening, in other words, the memory unit of the first embodiment can perform the (10) continuous swing performance in the evening (four). The first embodiment n FIG. 4A is the second embodiment of the present invention. A cross-sectional schematic diagram of a memory unit in a one-dimensional stylized operation. π Referring to FIG. 4A' The second embodiment of the memory unit includes a Shi Xiji II 400, a layer capture layer, and a first holding area. In the second embodiment, the second layer of the gate is 400, the layer-dielectric layer 4〇8, and the second layer of the second layer 220. In this embodiment, the stone basal willow is a p-type stone slab base, The impurity region 4G4a and the second doping region 4Q4b are n-type doped regions, wherein the dielectric layer 408 is, for example, an oxide layer. As a result, the second dielectric layer 41〇=the bottom of the silt soil has a high interface capture. (high interface trap, 101; interface f of the special f, the interface capture density (Dit) is between two ~ 1 〇 (10) eV; preferably H)] W2eV-]. In addition, the second 1 'layer 410 For example, the worst thermal U3S375 P950202 22692twfdoc/n can be selected, or the process of the thermal oxidation process or the secret mode of the thermal oxidation process (impla blessing) can make the Dlt of one side 412 between 10]]~ l OUcn^eV·1. In the present embodiment, the description is given to the radiant substrate 40 (the M column is a p-type slab substrate, and the first variator region a4a and the second doping region 〇4b are, for example, Type change zone.

_請繼續參照圖4A ’當對本實施例之記憶單元進行二 位π程式化的操作時,可於閉極概上施加第一正電壓(如 一g 10=特)’於第二摻雜區4〇4b(此時可作為汲極)施加第 -正電壓(如Vd=5伏特)’並使第一摻雜區4〇知(此時可作 為源極)D為一Vs=0伏特,以利用通道熱電子(CHE)方式程式 匕。己隐單元之一侧的位元(即沒極側位元),且通常 伏特。再者,上述第一播雜區刚a和第二摻雜區10仆也 =奐作汲極與源極,而不限於第二實施例所描述;換句話 况’當在閘極上施加第—正電壓、在源極加第二正電壓、 為〇伏特時,可程式化記憶單元之源極條元。而上 义第一正電壓一般而言是大於第二正電壓。_Continue to refer to FIG. 4A 'When the two-bit π-stylization operation is performed on the memory unit of the embodiment, a first positive voltage (eg, a g 10=special) may be applied to the second doping region 4 〇4b (which can be used as a drain) can apply a positive-positive voltage (eg, Vd=5 volts) and make the first doped region 4 known (which can be used as a source) D is a Vs=0 volt to Use channel hot electron (CHE) mode to program. The bit on one side of the cell (ie, the non-polar side bit), and usually volts. Furthermore, the first doping region a and the second doping region 10 are also used as the drain and the source, and are not limited to the second embodiment; in other words, when the gate is applied - The source voltage of the memory cell can be programmed when a positive voltage, a second positive voltage is applied to the source, and is volts. The first positive voltage is generally greater than the second positive voltage.

圖4B則是圖 時的剖面示意圖。 4A之記憶單元結構進行二位元抹除操作 口月多,备蚵个頁地1夕死憶單元進行二 、乐操作時,需在閘極406上施加第—負電壓(如Vg 伏特)’於第—摻雜區404b施加第三正電壓(如 特)’並使第一摻雜區4〇4a為0伏特(如Vs==〇伏特 用價帶對價帶熱電洞(BTBTHH)方式抹除記愔斤 位兀(即汲極側位元)。 關於第二實施例的記憶單元結構,尚可利用單—Fig. 4B is a schematic cross-sectional view of the figure. The memory cell structure of 4A performs a two-bit erase operation operation for more than one month, and prepares a page for the first time to perform a second memory operation. When the music operation is performed, a first-negative voltage (such as Vg volt) is applied to the gate 406. Applying a third positive voltage (eg, special) to the first doped region 404b and making the first doped region 4〇4a 0 volts (eg, Vs==〇Vot is valence band valence band thermal hole (BTBTHH))愔 兀 兀 汲 汲 汲 汲 汲 汲 。 。 关于 关于 关于 关于 关于 关于 关于 关于 关于 关于 关于 关于 关于 关于 关于 关于 关于 关于

15 1338375 P950202 22692tNvfd〇c/n ==進行程;化與抹除,如圖扣與圖奶所示,其 .操作二=單元結構進行單-位元裎式化及抹除 =參照® 4C,在閘極4〇6上施加一第四正 Γ伏:Γη) ’並使第—摻雜區4〇4a與第二摻雜區為 德綱 # $參照® 4D ’於間極404上施加一第二負 為^特^),並使第—摻雜區_與第二摻雜 ^伙特(Vs=0伏特' v㈣伏特),即可利用富勒_諾德漢 (FN)方式抹除所述記憶單元。 心、 設二1述/本2發明由W基底及其上層之間的界面 憶單元操作次數逐漸增加後,Dlt_‘二 ΐ憶能不會義降低,和帥單元的操 參 得改善。。&貧料保持性在循環次數逐漸增加後可獲 限定Γΐΐ發Γ已以較佳實施例揭露如上,然其並非用以 ’任何糊技術領射 為準。x保當視後附之中請專利範圍所界定者 【圖式簡單說明】 圖1是習知SONOS記憶單元在操作最初與經一萬次 ^ :·.. 16 1338375 P950202 22692twfdoc/n 循環後的i-v曲線圖。 圖2A是依照本發明之第一實施例之一種記憶單元結 構進行二位元程式化操作時的剖面示意圖。 圖2B是圖2A之記憶單元結構進行二位元抹除操作時 的剖面示意圖。 圖2C是圖2A之記憶單元結構進行單一位元程式化操 作時的剖面示意圖。 圖2D是圖2A之記憶單元結構進行單一位元抹除操 作時的剖面示意圖。 圖3A是圖2A之記憶單元結構進行二位元程式化操 作最初與經15K次、30K次循環後的I-V曲線圖。 圖3B是圖2A之記憶單元結構進行二位元抹除操作最 初與經15K次、30K次循環後的I-V曲線圖。 圖4A是依照本發明之第二實施例之一種記憶單元結 構進行二位元程式化操作時的剖面示意圖。 圖4B是圖4A之記憶單元結構進行二位元抹除操作時 的剖面示意圖。 圖4C是圖4A之記憶單元結構進行單一位元程式化操 作時的剖面示意圖。 圖4D是圖4A之記憶單元結構進行單一位元抹除操 作時的剖面示意圖。 【主要元件符號說明】 200、400 :矽基底 17 1338375 P950202 22692twfdoc/n 202、402 :捕捉層 204a、404a :第一摻雜區 2〇4b、404b :第二摻雜區 206、406 :閘極 208 :第一氧化層 210 :高界面捕捉密度材料層 212 :第二氧化層 408 :第一介電層 410 :第二介電層 412 :界面15 1338375 P950202 22692tNvfd〇c/n == proceeding; characterization and erasing, as shown in the figure and figure milk, operation 2 = unit structure for single-bit 裎 及 and erase = reference о 4C, Applying a fourth positive undulation on the gate 4〇6: Γη)' and applying a first doping region 4〇4a and a second doping region to the dynamite #$reference® 4D' The second negative is ^^^), and the first doping region _ and the second doping region (Vs=0 volts 'v (four) volts) can be erased by the Fuller_Nordheim (FN) method. The memory unit. After the number of operations of the interface between the W substrate and its upper layer is gradually increased, the Dlt_'2 memory can not be reduced, and the operation of the handsome unit is improved. . & Poor retention can be limited by the gradual increase in the number of cycles. The above example has been disclosed in the preferred embodiment, but it is not based on any paste technique. x Guardian's stipulations, please define the patent scope [Simplified illustration] Figure 1 is a conventional SONOS memory unit after the initial operation and after 10,000 times ^ :·.. 16 1338375 P950202 22692twfdoc / n cycle Iv chart. Fig. 2A is a schematic cross-sectional view showing a memory cell structure in a two-bit stylized operation in accordance with a first embodiment of the present invention. Fig. 2B is a schematic cross-sectional view showing the structure of the memory cell of Fig. 2A in a two-bit erase operation. Figure 2C is a cross-sectional view showing the memory cell structure of Figure 2A in a single bit stylized operation. Figure 2D is a cross-sectional view showing the memory cell structure of Figure 2A in a single bit erase operation. Fig. 3A is an I-V graph of the memory cell structure of Fig. 2A after the two-element stylized operation is initially performed with 15K cycles and 30K cycles. Fig. 3B is an I-V graph of the memory cell structure of Fig. 2A after the two-bit erase operation is performed at the beginning and after 15K times and 30K cycles. Fig. 4A is a cross-sectional view showing a memory cell structure in a two-element stylization operation in accordance with a second embodiment of the present invention. Fig. 4B is a schematic cross-sectional view showing the structure of the memory cell of Fig. 4A in a two-bit erase operation. Figure 4C is a cross-sectional view showing the memory cell structure of Figure 4A in a single bit stylized operation. Figure 4D is a cross-sectional view showing the memory cell structure of Figure 4A in a single bit erase operation. [Main component symbol description] 200, 400: germanium substrate 17 1338375 P950202 22692twfdoc/n 202, 402: capture layer 204a, 404a: first doped region 2〇4b, 404b: second doped region 206, 406: gate 208: first oxide layer 210: high interface capture density material layer 212: second oxide layer 408: first dielectric layer 410: second dielectric layer 412: interface

1818

Claims (1)

99-11-1199-11-11 1338375 十、申請專利範園: 1. 一種記憶單元結構,包括: 一矽基底; 一 ΟΝΟ電荷捕捉層,位於該矽基底上; 一第一摻雜區與一第二摻雜區,分別位於該〇Ν〇電 荷捕捉層兩側的該>5夕基底内; 一閘極,位於該0Ν0電荷捕捉層上;以及 ^ 一高界面捕捉密度(high-Dit)材料層,位於該矽基底與 該ΟΝΟ電荷捕捉層之間,其中該高界面捕捉密度材料層 與该石夕基底之間的界面捕捉密度(interface trap density,Dit) 在 1011 〜lOAm^eV·1 之間。 2. 如申請專利範圍第1項所述之記憶單元結構,其中 該高界面捕捉密度材料層與該矽基底之間的界面捕捉密度 為 lOUcn^eV·1。 3. 如申請專利範圍第1項所述之記憶單元結構,其中 或南界面捕捉街度材料層的厚度在1 〇〜7〇埃之間。 > 4.如申請專利範圍第1項所述之記憶單元結構,其中 該ifj界面捕捉後度材料層的材料包括氮化碎。 5.如申請專利範圍第1項所述之記憶單元結構,其中 該尚界面捕捉密度材料層的材料包括二氧化給(Hf〇2)、二 氧化锆(Zr〇2)、氮氧化鍅(Zr〇xNy)、氮氧化銓(Hf〇xNy)、矽 酸铪(HfSixOy)、矽酸鍅(ZrSix〇y)、氮氧矽铪(Hfsix〇yNz)、 三氧化二鋁(AhO3)、二氧化鈦(Ti〇2)、五氧化钽(Ta2〇5)、 三氧化二鑭(La203)、二氧化鈽(Ce〇2)、矽酸鉍(Bi4Si2〇丨2)、 19 1338375 99-11-11 氧化鎢(W03)、氧化釔(Y2〇3)、鋁酸鑭(LaA103)、鈦酸鋇鰓 (Ba〗-xSrxTi03)、鈦酸鋇(BaTi03)、锆酸鉛(PbZr03)、鈕酸銳 鉛(PbSczTai.z〇3,簡稱 PST)、鈮酸鋅鉛(PbZnzNbkC^,簡 稱PZN)、鍅鈦酸鉛(PbZrOrPbTi〇3,簡稱PZT)或鈮酸鎂 鉛(PbMgzNb】-z〇3,簡稱 PMN)。 6. 如申請專利範圍第1項所述之記憶單元結構,其中 該矽基底是p型矽基底,且該第一摻雜區與該第二摻雜區 是η型摻雜區。 7. —種記憶單元的操作方法,適用於一記憶單元,該 記憶單元具有一矽基底、位於該矽基底上的一 ΟΝΟ電荷 捕捉層、分別位於該0Ν0電荷捕捉層兩側的該矽基底内 的一第一摻雜區與一第二摻雜區、位於該0Ν0電荷捕捉 層上的一閘極、以及位於該矽基底與該ΟΝ〇電荷捕捉層 之間的一高界面捕捉密度材料層,其中該高界面捕捉密度 材料層與該矽基底之間的界面捕捉密度(Dit)在 1011〜H^cm'V·1之間,該操作方法包括: 私式化5亥5己憶早元時’於該間極上施加--'第一正電 廢’於該第二摻雜區施加一第二正電壓,並使該第一摻雜 區為0伏特’以利用通道熱電子(channei hot electron,CHE) 方式程式化該記憶單元之一側的位元; 抹除該記憶單元時,於該閘極上施加一第一負電壓, 於該第二摻雜區施加一第三正電壓,並使該第一推雜區為 0伏特’以利用價帶對價帶熱電洞(ban(j-t〇-band tunneling hot hole ’ BTBTHH)方式抹除該記憶單元之該側的位元。 20 1338375 99-11-11 請專利範圍第7項所述之記憶單元的操作方 /、汶第一摻雜區是源極,且該第二摻雜區是汲極。 9·如中請專利範圍第7項所述之記憶單元的操作方 '’其中該第-摻雜區是祕,且該第二摻雜區是源極。 10.如申請專利範圍第7項所述之記憶單元的操作方 法,其中該第一正電壓大於該第二正電壓。 11· 一種記憶單元的操作方法,適用於一記憶單元, • ,°己憶單元具有一矽基底、位於該矽基底上的一 ΟΝΟ電 何捕捉層、分別位於該〇Ν〇電荷捕捉層兩側的該矽基底 内的—第一摻雜區與一第二摻雜區、位於該ΟΝΟ電荷捕 捉層上的一閘極、以及位於該矽基底與該〇Ν〇電荷捕捉 層之間的一高界面捕捉密度材料層,其中該高界面捕捉密 度材料層與該矽基底之間的界面捕捉密度(DU)在 1011〜ΙΟΑπΓ2^1之間,該操作方法包括: 程式化該記憶單元時,於該閘極上施加一第四正電 壓,並使該第一摻雜區與該第二掺雜區為0伏特,以利用 鲁 虽勒-諾德漢(Fowler-Nordheim ’ FN)方式程式化該記恨單 元; … 抹除該&己憶早元時,於該閘極上施加一第二負電壓, 並使δ玄第一換雜區與該第二摻雜區為〇伏特,以利用富勒_ 諾德漢(FN)方式抹除該記憶單元。 12·如申請專利範圍第η項所述之記憶單元的操作方 法,其中該第一摻雜區是源極,且該第二摻雜區是汲極。 13.如申請專利範圍第U項所述之記憶單元的操作方 法,其中該第一摻雜區是汲極,且該第二摻雜區是源極。 211338375 X. Patent application garden: 1. A memory cell structure comprising: a germanium substrate; a germanium charge trapping layer on the germanium substrate; a first doped region and a second doped region respectively located a 闸 〇Ν〇 charge trap layer on both sides of the substrate; a gate on the 0 Ν 0 charge trap layer; and a high interface capture density (high-Dit) material layer, located on the 矽 substrate and Between the charge trapping layers, wherein the interface trap density (Dit) between the high interface capture density material layer and the stone substrate is between 1011 and lOAm^eV·1. 2. The memory cell structure of claim 1, wherein the interface capture density between the high interface capture density material layer and the germanium substrate is lOUcn^eV·1. 3. The memory cell structure according to item 1 of the patent application, wherein the south interface captures a thickness of the street material layer between 1 〇 and 7 〇. 4. The memory cell structure of claim 1, wherein the ifj interface captures a material of the late material layer comprising nitriding. 5. The memory cell structure according to claim 1, wherein the material of the interface to capture the density material layer comprises: (2), (2, 2, 2, 2, 2, 2, 2, 2, 2 〇xNy), Hf〇xNy, HfSixOy, ZrSix〇y, Hfsix〇yNz, Al2O3, Titanium Dioxide 〇2), bismuth pentoxide (Ta2〇5), antimony trioxide (La203), cerium oxide (Ce〇2), bismuth ruthenate (Bi4Si2〇丨2), 19 1338375 99-11-11 tungsten oxide ( W03), yttrium oxide (Y2〇3), lanthanum aluminate (LaA103), barium titanate (Ba-xSrxTi03), barium titanate (BaTi03), lead zirconate (PbZr03), lead acid sharp lead (PbSczTai. Z〇3, abbreviated as PST), lead bismuth citrate (PbZnzNbkC^, PZN for short), lead bismuth titanate (PbZrOrPbTi〇3, PZT for short) or lead bismuth citrate (PbMgzNb)-z〇3, referred to as PMN). 6. The memory cell structure of claim 1, wherein the germanium substrate is a p-type germanium substrate, and the first doped region and the second doped region are n-type doped regions. 7. A method of operating a memory cell, the memory cell having a germanium substrate, a germanium charge trapping layer on the germanium substrate, respectively located in the germanium substrate on both sides of the germanium charge trapping layer a first doped region and a second doped region, a gate on the 0Ν0 charge trap layer, and a high interface capture density material layer between the germanium substrate and the germanium charge trap layer, Wherein the interface capture density (Dit) between the high interface capture density material layer and the germanium substrate is between 1011 and H^cm'V·1, and the operation method comprises: privately arranging 5 Applying a 'positive positive waste' to the second doped region, applying a second positive voltage to the second doped region, and making the first doped region 0 volts' to utilize channel hot electrons (channei hot electron , CHE) mode stylizes a bit on one side of the memory cell; when erasing the memory cell, applying a first negative voltage to the gate, applying a third positive voltage to the second doped region, and The first doping region is 0 volts' to take advantage of the valence band The hole (jt〇-band tunneling hot hole 'BTBTHH) method erases the bit on the side of the memory unit. 20 1338375 99-11-11 The operation unit of the memory unit described in item 7 of the patent scope is The first doped region is a source, and the second doped region is a drain. 9· The operating unit of the memory unit described in claim 7 of the patent scope, wherein the first doped region is secret The method of operating the memory unit of claim 7, wherein the first positive voltage is greater than the second positive voltage. 11· Operation of a memory unit The method is applicable to a memory unit, wherein the memory unit has a substrate, and a capture layer on the substrate, respectively located in the substrate on both sides of the charge trap layer. a first doped region and a second doped region, a gate on the germanium charge trap layer, and a high interface capture density material layer between the germanium substrate and the germanium charge trap layer, wherein The high interface captures the boundary between the layer of density material and the substrate The capture density (DU) is between 1011 and ΙΟΑπΓ2^1. The operation method includes: when the memory unit is programmed, applying a fourth positive voltage to the gate, and the first doped region and the second doping The miscellaneous area is 0 volts to program the hate unit using the Fowler-Nordheim 'FN method; ... when erasing the & early memory, a second negative is applied to the gate The voltage, and the δ 第一 first mismatch region and the second doped region are 〇 volts, to erase the memory cell by using a Fuller_Nordheim (FN) method. The method of operating a memory cell, wherein the first doped region is a source and the second doped region is a drain. 13. The method of operating a memory cell of claim U, wherein the first doped region is a drain and the second doped region is a source. twenty one
TW96107469A 2007-03-05 2007-03-05 Memory unit structure and operation method thereof TWI338375B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW96107469A TWI338375B (en) 2007-03-05 2007-03-05 Memory unit structure and operation method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW96107469A TWI338375B (en) 2007-03-05 2007-03-05 Memory unit structure and operation method thereof

Publications (2)

Publication Number Publication Date
TW200837963A TW200837963A (en) 2008-09-16
TWI338375B true TWI338375B (en) 2011-03-01

Family

ID=44820381

Family Applications (1)

Application Number Title Priority Date Filing Date
TW96107469A TWI338375B (en) 2007-03-05 2007-03-05 Memory unit structure and operation method thereof

Country Status (1)

Country Link
TW (1) TWI338375B (en)

Also Published As

Publication number Publication date
TW200837963A (en) 2008-09-16

Similar Documents

Publication Publication Date Title
TWI266321B (en) Operation scheme with charge balancing erase for charge trapping non-volatile memory
Lee et al. Metal nanocrystal memory with high-/spl kappa/tunneling barrier for improved data retention
CN100502009C (en) SONOS type memory device
Zhao et al. Review on non-volatile memory with high-k dielectrics: Flash for generation beyond 32 nm
TWI358834B (en)
US20180068723A1 (en) Oxide based memory
TWI436455B (en) Sonos ono stack scaling
US20090189215A1 (en) Nonvolatile flash memory device and method for producing the same
EP2110854A2 (en) Floating Gate Memory Device with Interpoly Charge Trapping Structure
JP2005311379A (en) Memory device with multiple dielectric layer, and its manufacturing method
TWI440189B (en) Sonos memory cell having high-k dielectric
JP2006114905A (en) Non-volatile semiconductor memory element
TW200935606A (en) Memory cells, electronic systems, methods of forming memory cells, and methods of programming memory cells
TW200919708A (en) Memory devices with split gate and blocking layer
TW200910532A (en) Non-volatile memory device and method for manufacturing the same
TW201114022A (en) Charge trapping memory cell having bandgap engineered tunneling structure with oxynitride isolation layer
US20090127611A1 (en) Non-volatile memory device and memory card and system including the same
JP2011155266A (en) Nonvolatile memory cell, memory array having the same, and method of operating cell and array
JP2005183940A (en) Manufacturing method of nonvolatile memory using plasma oxidized high-k charge trapping layer
CN101283448B (en) Semiconductor storage device and method for manufacturing same
JP2007184380A (en) Nonvolatile memory cell, memory array having the same, and method of operating the cell and the array
US20080185633A1 (en) Charge trap memory device with blocking insulating layer having higher-dielectric constant and larger energy band-gap and method of manufacturing the same
TWI338375B (en) Memory unit structure and operation method thereof
US20070114572A1 (en) Gate structure including multi-tunneling layer and method of fabricating the same, non-volatile memory device and method of fabricating the same
JP2008227492A (en) Charge trap memory device