TW434842B - Flash memory hot electron programming method that is suit for low drain voltage operation - Google Patents

Flash memory hot electron programming method that is suit for low drain voltage operation Download PDF

Info

Publication number
TW434842B
TW434842B TW88121352A TW88121352A TW434842B TW 434842 B TW434842 B TW 434842B TW 88121352 A TW88121352 A TW 88121352A TW 88121352 A TW88121352 A TW 88121352A TW 434842 B TW434842 B TW 434842B
Authority
TW
Taiwan
Prior art keywords
source
volts
substrate
gate
floating gate
Prior art date
Application number
TW88121352A
Other languages
Chinese (zh)
Inventor
Da-Huei Wang
Ru-Ping Jiang
Ching-Wei Tsai
Nian-Kai Tzou
Original Assignee
Wang Da Huei
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 Wang Da Huei filed Critical Wang Da Huei
Priority to TW88121352A priority Critical patent/TW434842B/en
Application granted granted Critical
Publication of TW434842B publication Critical patent/TW434842B/en

Links

Landscapes

  • Non-Volatile Memory (AREA)

Abstract

A new method for charging floating gate of flash memory is disclosed. The flash memory consists of source, drain, control gate, floating gate and substrate. Let the bias between source and drain smaller than 5 volts, the bias between floating gate and source larger than 2 volts, and the bias between substrate and source larger than 0.5 volt. Use the positive bias from the substrate to the source to inject large amount of holes into n-type channel and generate Auger re-combination. The Auger re-combination can raise the electron energy of n-channel and let electrons get enough energy to surmount the energy barrier of tunnel oxide and charge the floating gate.

Description

五、發明說明(1) 本發明係有關於一種對非揮發性記憶體的浮動閘極充 電的方法,尤指一種能在低沒極電壓操作下,對非揮發性 記億體的浮動閘極充電的方法。 非揮發性記憶體是一種記憶資料不隨著電源關閉而流 失的半導體記憶體’大致上可以分類成唯讀記憶體(rea(i only memory,ROM)、可抹除及程式化唯讀記憶體 (erasable programmable read only memory , EPROM)以 及電子式可抹除及程式化唯讀記憶體(electrically erasable programmable read only memory ,EEPR0M)。 EPROM與EEPROM都是以電子方式將信號寫入,eprom是以照 射紫外光將資料抹除’ EEPR0M是以電子的方式將資料抹 除’而習知的快閃(Π ash) E EPROM是以電子方式一次抹除 所有的或一區域中的EE PROM之資料。為了解說上的方便, 統稱為EPROM。 請參閱第1圖,第1圖為習知的EPROM的結構示意圖。 一般的EPROM包含有一基底(substrate)lO、兩個和基 底10相反電性的源極(source)區12以及没極(drain)區 14、一浮動閘(floating gate)極18以及一控制閘 (control gate)極22 形通道EPROM的基底10是一 p形基 底,源極區1 2以及汲極區1 4是兩個η形摻雜區。基底1 〇、 源極區1 2、汲極區1 4、浮動閘極1 8以及控制閘極22上的電 壓分別表示為Vb、Vs、Vd、Vi以及Vc。兩個電極上的電壓 差則定義為兩個電極間的偏壓,譬如說,Vds意指汲極對 Vs的偏壓,也就是Vd-Vs。V. Explanation of the invention (1) The present invention relates to a method for charging a floating gate of a non-volatile memory, especially a floating gate capable of charging non-volatile memory to a non-volatile memory under low-voltage operation. How to charge. Non-volatile memory is a type of semiconductor memory that does not lose data when the power is turned off. It can be roughly classified into read-only memory (rea (i only memory, ROM), erasable and programmable read-only memory). (erasable programmable read only memory (EPROM)) and electronic erasable programmable read only memory (EEPR0M). EPROM and EEPROM both write signals electronically, eprom is irradiation Ultraviolet light erases data 'EEPR0M is electronically erasing data' and the conventional flash (Π ash) E EPROM is an electronic erasure of all or EE PROM data in one area at a time. For convenience of explanation, they are collectively referred to as EPROMs. Please refer to FIG. 1. FIG. 1 is a schematic structural diagram of a conventional EPROM. A general EPROM includes a substrate 10, and two sources with opposite electrical properties to the substrate 10 ( source region 12 and drain region 14, a floating gate electrode 18 and a control gate electrode 22. The substrate 10 of the EPROM channel is a p-shaped substrate. Region 12 and drain region 14 are two n-shaped doped regions. The voltages on substrate 10, source region 12, drain region 14, floating gate 18, and control gate 22 are expressed as Vb, Vs, Vd, Vi, and Vc. The voltage difference between the two electrodes is defined as the bias between the two electrodes. For example, Vds means the bias of the drain to Vs, which is Vd-Vs.

434842 五、發明說明(2) ~ 在電性上,每個EPROM有一個臨界電壓(threshQld voltage,Vt),當控制閘極22對源極區12的偏壓值Vcs大 於Vt時,源極區1 2以及汲極區1 4便可以相互導通。EpR〇M 中存入的資料是邏輯〇或是邏輯1便是以源極區丨2和汲極區 1 4之間有無相互導通來判定。而v t會受到浮動閘極丨8中的 電荷多寡所影響’所以利用浮動閘極18中儲存電荷的量來, 定義邏輯0或是邏輯1。在EPROM中,將帶電載子(carrier) 送入浮動閘極1 8中並使帶電載子陷於浮動閘極丨8中的充電 過程稱為程式化(programming),將帶電載子移出浮動閘 極18的過程則稱為抹除(erasing)。 習知的技術中’ η形通道EPROM之熱電子程式化法所運 用的物理方法大致上可以分為兩種’ 一種稱為通道熱電子 注射(channel hot electron injection,CHEI),另一種 則稱為通道觸發之二次電子注射(channel initiate(1 secondary electron injection ,CISEI)。 請參閱第2A圖’第2A圖為習知的CHE I之示意圖。通道 熱電子注射的目的在加熱於浮動閘極18下之通道中的電 子’以使部分加熱後的電子擁有足夠的能量可以跨越第一 介電層16與基底10的導電帶(conduct丨〇n band)之差所形 成的能障(energy barrier)。因為加熱後的電子比熱平衡 下之電子具有更高能量’所以稱之為熱電子。如果基底1〇 和第一介電層1 6分別是以矽和二氧化矽所構成,那麼矽/ 二氧化矽的介面(interface)所形成的能障大約是3.2電子 伏特(eV ),也就是說,n形通道中的熱電子之能量必須接434842 V. Description of the Invention (2) ~ In terms of electrical properties, each EPROM has a threshold voltage (threshQld voltage, Vt). When the bias value Vcs of the control gate 22 to the source region 12 is greater than Vt, the source region 12 and the drain region 14 can be connected to each other. Whether the data stored in EpROM is logic 0 or logic 1 is determined by the mutual conduction between the source region 2 and the drain region 14. And v t will be affected by the amount of charge in the floating gate 丨 8 ’, so the amount of charge stored in the floating gate 18 is used to define logic 0 or logic 1. In EPROM, the charging process of sending a charged carrier into the floating gate 18 and trapping the charged carrier in the floating gate is called programming, and the charged carrier is moved out of the floating gate. The process of 18 is called erasing. In the conventional technique, the physical method used in the thermo-electron stylization method of the η-shaped channel EPROM can be roughly divided into two types. One is called channel hot electron injection (CHEI), and the other is called CHEI. Channel-initiated secondary electron injection (CISEI). Please refer to Figure 2A 'Figure 2A is a schematic diagram of the conventional CHE I. The purpose of channel hot electron injection is to heat the floating gate 18 The electrons in the lower channel 'so that some of the heated electrons have enough energy to cross the energy barrier formed by the difference between the conductive band of the first dielectric layer 16 and the substrate 10 Since the heated electrons have higher energy than the electrons under thermal equilibrium, they are called hot electrons. If the substrate 10 and the first dielectric layer 16 are made of silicon and silicon dioxide, respectively, then silicon / two The energy barrier formed by the interface of silicon oxide is about 3.2 electron volts (eV), that is, the energy of the hot electron in the n-channel must be connected

434842 五、發明說明(3) 近或大於3. 2eV才能跳過第一介電層而進入浮動閑極μ , 對浮動閘極1 8進行充電。假使忽略掉η形通道中電子散射 (s c a 11 e r i ng )的影響’那從源極流動到汲極之電子所能獲 得的最大能里約等於q*(Vds) ’其中,q為一個電子的電 量。所以’在理想上,V d s必須大於3伏特後,通道中的電 子才會有足夠機會跨越第一介電層至浮動閘極。然而,在 實際的操作上’Vds至少要大於5伏特才能提供通道中電子 足夠能量以跨越3. 2電子伏特的能障。 但疋,11¾著半導體製程的演進,晶片上的尺寸不斷的 縮小,外在的電源也由5伏特降至3 . 3伏特,甚至更低。為 了增加EPROM陣列的積集度以及減少汲極區電隔絕上的困 難’因此’程式化時汲極的電壓也希望降至3 · 3伏特左 右。 習知的CISEI ’如美國專利案號5, 659, 504號所述,就 是一種低電壓(Vd低於5伏特)的η形通道EPROM之程式化方 法。請參閱第2B圖,第2B圖為習知的CISEI之示意圖。 CISEI和CHEI的最大差別在於CiSEI之Vds降至3伏特左 右’ Vbs小於零;而CHEI之Vds大約為5伏特,Vbs等於 零。在CISEI的程式化方法中,從源極12出發的電子el在 通道中無法得到足夠的能量跨越S i /S i 02的能障,但是, 沒極到源極的偏壓Vds卻足以使電子ei有足夠的能量在汲 極區14附近的n形通道中產生撞擊離子化效應(丨mpact ionization)。因為石夕的能帶間隙(energy gap)大約為 1,12電子伏特,所以電子el只要獲得超過i 12eV的能量便434842 V. Description of the invention (3) The first dielectric layer can be skipped to enter the floating idler μ near or greater than 3.2 eV, and the floating gate 18 is charged. If the influence of the electron scattering (sca 11 eri ng) in the η-shaped channel is ignored, the maximum energy that can be obtained by the electrons flowing from the source to the drain is approximately equal to q * (Vds), where q is an electron Battery. So 'ideally, the electrons in the channel will have enough opportunity to cross the first dielectric layer to the floating gate after V d s must be greater than 3 volts. However, in actual operation, 'Vds must be at least greater than 5 volts to provide sufficient energy for the electrons in the channel to cross the energy barrier of 3.2 electron volts. However, as the semiconductor process evolved, the size on the wafer continued to shrink, and the external power supply also decreased from 5 volts to 3.3 volts or even lower. In order to increase the accumulation of the EPROM array and reduce the difficulty of galvanic isolation of the drain region ', therefore, the voltage of the drain electrode is also expected to drop to about 3.3V during programming. The conventional CISEI ', as described in U.S. Patent No. 5,659,504, is a method of stylizing a low-voltage (Vd below 5 volts) n-channel EPROM. Please refer to FIG. 2B, which is a schematic diagram of the conventional CISEI. The biggest difference between CISEI and CHEI is that the Vds of CiSEI drops to about 3 volts. Vbs is less than zero; while the Vds of CHEI is about 5 volts and Vbs is equal to zero. In the CISEI's stylized method, the electron el from the source 12 cannot get enough energy in the channel to cross the energy barrier of S i / S i 02, but the bias voltage Vds without the source to the source is enough to make the electron ei has enough energy to generate impact ionization in the n-shaped channel near the drain region 14. Because the energy gap of Shi Xi is approximately 1,12 electron volts, as long as the electron el obtains energy exceeding i 12eV,

第6頁 4 3 4 8 4 2 五、發明說明(4) 能產生電子電洞對(electron-hole pair) ,e2以及h2。 電洞h2受Vbs所形成的電場牽引朝向基底10加速。當電洞 h2獲得了足夠的能量,便在汲極區14與基底1〇的接靣 (j unction)附近產生電子電洞對,e3以及h3。在同一個電 場中,電子和電洞的受力的方向相反,因此電子e3朝向 Si/Si02的介面加速。一旦電子e3的能量接近或大於3.2電! 子俠特的能障’就有機會躍過第一介電層1 6而掉入浮動閘 極1 8 ’對浮動閘極進行充電。因為撞擊離子化效應所產生 的二次電子是由η形通道中的電子所引發的,所以稱為通 道觸發之二次電子注射CISEI。在美國專利案號5,659,504 號中的實施例中建議,Vs = 0、l.lvSVdS3_3v、以及-3ν SVbS-0.5v。 本發明的主要目的,在於提供另一種低没極電壓的η 形通道記憶體之浮動閘極充電的方法。能夠在Vds低於5伏 特的條件下,對EPROM的浮動閘極進行充電。 根據上述之目的’本發明提出一種對一η形通道記憶 體之浮動閘極充電的方法^該記憶體包含有—源極、一汲 極、一控制閘極 '一浮動閘極以及一基底。本發明之方法 稱為1彳木加強式通道熱電子注射(AUge.r_enhancecj channel hot electron injection ,AECHEI) dAECHEI從 *亥基底對靠近該源極附近之n形通道注入電洞,使電洞和n 开乂通道中電子產生歐傑再結合rec〇mbinati〇n)以 提^部分導電帶中電子之能量。之後,與chei相似,歐傑 再、结合所產生之高能量電子再被设極對源極之偏壓Vds所 電子就能進入 加熱。當電子的能量大過—固定之能障後 該浮動閘極’對浮動閘極進行充電。·"又 在本發明中,進入該浮動閘極之電子 個來源,一是該汲極對該源極的偏壓所 丨的能量有兩 則是該基底對該源極Vbs之偏壓所導致的、加逮,另一 為要跨越能障進入該浮動閘的最小能量是固木:合。因 要增加歐傑再結合後所提供的能量,便 、,所以只 的加速所需提供的能量’也就是降低了該汲極二形通道 Vds的偏壓,達到低電壓操作的目的。 ’ μ原極 ^發明❺第一實施例提供了η形通道記憶體 各個偏壓,為了產生歐傑再結合,在動冬 ,該基底對該源極vbS的偏壓為正值。而通增中 的加返由該汲極對該源極的正偏壓 =子 上的電㈣,由控制閉輕合以及其中存在電動: =大於4的程度,„形通道才會產生,也才會產生熱也 ,,本發明之第二實施例提供了 AECHEi執行時所使用的η 开^通道5己憶體之結構。吾人利用在η形通道中的下方毁_ 個重ρ形摻雜區’用以增加有效電洞注入的量以強化歐傑 再結合效應及增加程式化之速度。 * 本發明之第三實施例提供了利用AECHE ί程式化之浮動 開記憶元陣列。部分的記憶元之基底相連結至第一接點, 部分的記憶元之基底相連結至第二接點,第一接點和第二 接點在電性上相互隔絕。如此的設計可以減少進行程式化Page 6 4 3 4 8 4 2 V. Description of the invention (4) Electron-hole pair (e2 and h2) can be generated. The hole h2 is accelerated toward the substrate 10 by the electric field formed by Vbs. When sufficient energy is obtained in the hole h2, an electron hole pair, e3, and h3 are generated near the junction of the drain region 14 and the substrate 10. In the same electric field, the forces of the electrons and the holes are in opposite directions, so the electrons e3 accelerate toward the Si / Si02 interface. Once the energy of the electron e3 is close to or greater than 3.2 electricity! Ziaite's energy barrier ’has the opportunity to jump over the first dielectric layer 16 and fall into the floating gate 1 8 ′ to charge the floating gate. Because the secondary electrons produced by the impact ionization effect are caused by the electrons in the η-shaped channel, they are called channel-triggered secondary electron injection CISEI. It is suggested in the examples in US Patent No. 5,659,504 that Vs = 0, l.lvSVdS3_3v, and -3v SVbS-0.5v. The main object of the present invention is to provide another method for charging a floating gate of an n-channel channel memory with a low inverting voltage. Capable of charging EPROM floating gates with Vds below 5 volts. According to the above object, the present invention proposes a method for charging a floating gate of an n-channel channel memory. The memory includes a source, a drain, a control gate, a floating gate, and a substrate. The method of the present invention is called AUge.r_enhancecj channel hot electron injection (AECHEI). DAECHEI injects holes into the n-shaped channel near the source from the * Hai substrate, so that the holes and n The electrons generated in the open channel are combined with Ojie (recommbination) to increase the energy of the electrons in some conductive bands. After that, similar to chei, the high-energy electrons generated by the combination are then heated by the biased Vds of the set-to-source bias. When the energy of the electron is greater than a fixed energy barrier, the floating gate 'charges the floating gate. · "In the present invention, the source of the electrons entering the floating gate is that the energy biased by the drain to the source is two and the energy biased by the substrate to the source Vbs The minimum energy to cause the trap to enter the floating gate across the energy barrier is solid wood: He. Because the energy provided after recombination is increased, the energy required to accelerate is only used to reduce the bias voltage of the drain-shaped channel Vds to achieve the purpose of low voltage operation. ‘Μ original electrode ^ Invention ❺ The first embodiment provides various bias voltages for the η-shaped channel memory. In order to generate the Auger recombination, the substrate has a positive bias voltage to the source vbS in winter. The increase and increase in the pass increase is caused by the positive bias of the drain to the source = the electric voltage on the sub, by the closed circuit and the presence of electric motor: = to a degree greater than 4, the shape of the channel will be generated. Only heat is generated. The second embodiment of the present invention provides the structure of the η open channel 5 memory used in the execution of AECHEi. We use the bottom of the η channel to destroy _ heavy p-type doping The area is used to increase the amount of effective hole injection to enhance the recombination effect and increase the speed of stylization. * The third embodiment of the present invention provides a stylized floating open memory cell array. Partial memory The base of the element is connected to the first contact, and the base of some memory cells is connected to the second contact. The first contact and the second contact are electrically isolated from each other. Such a design can reduce programming.

第8頁 434842 五、發明說明(6) ^基底電流的消耗’用以增加程式化的致志 為使本發明之上述目 下文特舉一較佳實施例, 下: 的、特徵和優黑占 並配合所附圖式 能更明顯易懂, ’作詳細說明如 圖式之簡單說明: 第1圖為習知的EPROM的結構示意圖; 第2A圖為習知的CHEI之示意圖; 第2B圖為習知的CISEI之示意圖;Page 8 434842 V. Description of the invention (6) ^ Consumption of base current 'is used to increase the ambition of stylization In order to make the above item of the present invention a preferred embodiment is given below, the following: With the accompanying drawings, it can be more clearly understood. 'For a detailed description, the diagram is briefly explained: FIG. 1 is a schematic diagram of a conventional EPROM; FIG. 2A is a schematic diagram of a conventional CHEI; and FIG. 2B is Schematic diagram of the known CISEI;

第3圖表示本發明之AECHE I的原理示意圖. 第4圖為歐傑再合併的示意圖; 第5圖表示一個虛擬記憶元於導通時所發射的光之能 量分佈的實驗結果圖; 第6圖為一個虛擬記憶體的閘極電流對閘極電壓圖; 第7圖為本發明之AECHEI以及習知CHEI對一EPROM的程 式化特性圖; 第8圖為一種本發明之AECHEI所使用之EPROM的結構 圖;Fig. 3 shows the principle diagram of AECHE I of the present invention. Fig. 4 is a schematic diagram of recombination of Oujie; Fig. 5 shows an experimental result diagram of the energy distribution of light emitted by a virtual memory cell when it is turned on; Fig. 6 FIG. 7 is a graph of gate current versus gate voltage of a virtual memory. FIG. 7 is a stylized characteristic diagram of AECHEI of the present invention and conventional CHEI on an EPROM. FIG. 8 is a graph of EPROM used by the AECHEI of the present invention Structure diagram

第9A圖以及第9B圖為兩種本發明之AECHEI所使用之 ΕΜ〇Μ的結構圖; 第1 0 A圖表示本發明之浮動閘記憶元陣列示意圖; 第1 0B圖為第i 〇 A圖中記憶元陣列的晶片實施例;以及 第11圖為一種分閘式η形通道元件示意圖。 符號說明‘:FIG. 9A and FIG. 9B are structural diagrams of the EMOM used by the two AECHEIs of the present invention; FIG. 10A shows a schematic diagram of the floating gate memory cell array of the present invention; and FIG. 10B is the IOA diagram An embodiment of a chip of a memory cell array; and FIG. 11 is a schematic diagram of a gated n-channel element. Symbol Description':

第9頁 3 4 β -:: 2 五、發明說明(7) 30〜基底; 3 2 ~源極; 3 4〜汲極; 36〜第一介電層; 3 8〜浮動閘極; 40~第二介電層; 4 2〜控制閘極; 44〜重ρ形摻雜區; 4 5〜位元線; 4 6〜字元線; 4 8〜源極線; 5 0 a〜第一接點; 50b〜第二接點; 6 1〜閘氧化層; 6 2〜選擇閘極。 實施例: 請參閱第3圖,第3圖表示本發明之AECHEI的原理示意 圖。本發明提供了 一種對一η形通道記憶體(又稱EPROM)之 浮動閘極充電的方法。如第3圖所示,一 E P R 0 Μ包含有一源 極3 2、一汲_極3 4、一控制閘極4 2、一浮動閘極3 8以及一基 底30。浮動閘極38與基底30之間有一第一介電層36,一般 是以氧化物所構成,所以又稱穿透氧化層(1: U η n e 1 i n g o x i d e 1 a y e r )。控制閘極4 2與浮動閘極3 8之間有一第二介Page 9 3 4 β-:: 2 V. Description of the invention (7) 30 ~ substrate; 3 2 ~ source; 3 4 ~ drain; 36 ~ first dielectric layer; 3 8 ~ floating gate; 40 ~ Second dielectric layer; 4 2 ~ control gate; 44 ~ heavy doped region; 4 5 ~ bit line; 4 6 ~ word line; 4 8 ~ source line; 50 a ~ first connection Point; 50b ~ second contact; 6 1 ~ gate oxide layer; 6 2 ~ select gate. Example: Please refer to Fig. 3, which shows a schematic diagram of the principle of AECHEI of the present invention. The invention provides a method for charging a floating gate of an n-shaped channel memory (also called EPROM). As shown in FIG. 3, an EPROM 0M includes a source electrode 3 2, a drain_pole 3 4, a control gate electrode 4, a floating gate electrode 38, and a base 30. There is a first dielectric layer 36 between the floating gate 38 and the substrate 30. The first dielectric layer 36 is generally made of an oxide, so it is also called a penetrating oxide layer (1: U η n e 1 i n g o x i d e 1 a y e r). There is a second interface between the control gate 4 2 and the floating gate 3 8

第10頁 43 卯 3 ' ’ 五、發明說明(8) 電層40 ’以作為控制閘極42與浮動閘極38之間的電隔絕。. 本發明之AECHE I包含有下列步驟:以一第一連接線提 供一第一電壓(以下稱做Vd)至汲極34以使汲極34與源極32 之間的偏壓Vds小於5伏特;以一第二連接線提供一第二電 壓(以下稱做Vc)與控制閘極42以使浮動閘極38與源極32之 間的偏壓Vf s大於2伏特;以及以一第三連接線提供一第三, 電壓(以下稱為Vb)與基極3〇以使基底30與源極32之間的偏 壓Vbs大於〇. 5伏特。Vf s能夠使一個η形通道形成於第一介 電層下的基底30表面。因為作8大於零,所以會有電洞h由 基底30流向源極32以及源極32附近的η形通道,因而發生 \ 歐傑再合併。歐傑再合併提供了從源極32流向汲極34的電 Ρ 子e —些能量。電子e再經過Vds之電場加速後,便有機會 得到足夠的能量跨越第一介電層3 6與基底3 〇所形成的能 障’進入浮動閘極38,對浮動閘極38進行充電。 如第4圖所示,第4圖為歐傑再合併的示意圖。簡單的 說’歐傑再合併就是撞擊離子化效應(i mpac t ionization)的相反過程。因為vbs大於零,所以基底30中 的電洞h便向n形通道中擴散。電洞h到達n形通道後,便有 很大的機率會與η形通道中的電子e產生歐傑再合併而消 失’所產生的能量(至少一個能帶間隙)便提供給其餘的一 ) 顆電子。也就是說,歐傑再合併會導致—些電子,於一開 始在η形通道中運動時便有較高的能量,所以yds只要提供 —個較低的能量給通道中的電子e,便有機會使電子e獲得 足夠的能量跨越第一介電層36與基底3〇所形成的能障。Page 10 43 卯 3 ′ Ⅴ. Description of the invention (8) The electric layer 40 ′ serves as the electrical isolation between the control gate 42 and the floating gate 38. The AECHE I of the present invention includes the following steps: a first connection line is used to provide a first voltage (hereinafter referred to as Vd) to the drain 34 so that the bias voltage Vds between the drain 34 and the source 32 is less than 5 volts. A second connection line is used to provide a second voltage (hereinafter referred to as Vc) and the control gate 42 so that the bias voltage Vf s between the floating gate 38 and the source 32 is greater than 2 volts; and a third connection The line provides a third voltage (hereinafter referred to as Vb) and the base 30 so that the bias voltage Vbs between the substrate 30 and the source 32 is greater than 0.5 volts. Vf s enables an n-shaped channel to be formed on the surface of the substrate 30 under the first dielectric layer. Because 8 is greater than zero, there will be a hole h flowing from the substrate 30 to the source 32 and the n-shaped channel near the source 32, so the recombination occurs. The Auger recombination provides some energy, e, from the source 32 to the drain 34. After the electron e is accelerated by the electric field of Vds, it has a chance to get enough energy to cross the barrier formed by the first dielectric layer 36 and the substrate 30 and enter the floating gate 38 to charge the floating gate 38. As shown in Figure 4, Figure 4 is a schematic diagram of Ou Jie's recombination. To put it simply, ‘Auger recombination is the opposite process of impact ionization. Since vbs is greater than zero, the hole h in the substrate 30 diffuses into the n-shaped channel. After the hole h reaches the n-shaped channel, there is a great chance that it will merge with the electron e in the η-shaped channel and then merge and disappear. 'The energy (at least one band gap) will be provided to the remaining one) Electrons. In other words, the recombination of Ou Jie will cause some electrons to have higher energy when they initially move in the η-shaped channel, so as long as yds provides a lower energy to the electron e in the channel, there will be The opportunity allows the electron e to obtain enough energy to cross the energy barrier formed by the first dielectric layer 36 and the substrate 30.

第11頁 43i4gi4i 五、發明說明(9) ' ' ' 请參閱第5圖,第5圖表示一個虛擬記憶元於導通時所 發射光的能量分佈實驗結果圖。所謂虛擬記憶元(dufflmy =U)就是一個控制閘極與浮動閘極相短路的EpR〇M,經常 疋用來里測浮動閘上的電性。第5圖的縱座標是光強度除 以及,電流,也就是正規K(n〇rma丨後的光強度。橫 座標^光能量。第5圖之縱座標的另—層意義是電子的數 目一 k座軚的另一層意義是電子所帶的能量。由第5圖中 =了條線可知,當基底對源極的偏壓VbS漸漸增大時,高 的:子數目也漸漸地增乡’也就是說能夠有足夠能量 極的電子也越多。其中,更值得注意的是,於 1 6雷早你及几=1.5伏特的兩條線中,在電子能量等於 二,二特的附近均有一個凸起(hump)。而U電子伏特 以妙的能帶間隙㈠.12電子伏特)的1<5倍。由元件物 知,歐傑再結合時導電帶的電子最可能獲得 g 1此帶間隙之能量,這個值在以矽所構成 疋電子伏特。也就是說,第5圖士 土&上k 的雪ί二所造成的,迠也提供歐傑再合併對n形通道中 的電子供應能量的一個證明。 小通遏中· 〇月參閱第6圖,第β圖為一個虛 閉極電壓,。在此,定義虛擬記㈡=間極電流對 e ^ ^ , 〇 tvds = 3:5 Λ ^ 1 fi]" ^ 者閘極電壓由零漸漸增大,流入閘極 Vbs=0時,隨 ;的電流之最大值會出現在閑極會增大。閑Page 11 43i4gi4i V. Description of the invention (9) '' 'Please refer to Figure 5, which shows the experimental results of the energy distribution of light emitted by a virtual memory cell when it is turned on. The so-called virtual memory cell (dufflmy = U) is an EpROM that controls the short circuit between the gate and the floating gate. It is often used to measure the electrical properties on the floating gate. The ordinate of Figure 5 is the light intensity divided by the current, which is the light intensity after the normal K (n0rma). The abscissa is the light energy. The other meaning of the ordinate of Figure 5 is that the number of electrons is one The other meaning of k-block 是 is the energy carried by the electron. From Figure 5 = the line shows that when the substrate-source bias VbS gradually increases, the higher: the number of sub-substances gradually increases. That is to say, there are more electrons that can have enough energy poles. Among them, it is more noticeable that in the two lines of 16 and 1.5 volts earlier, the electron energy is equal to two, and the vicinity of the two special is equal. There is a hump, and the U electron volt is 1 < 5 times with a wonderful band gap (.12 electron volts). It is known from the components that the electrons of the conductive bands are most likely to obtain the energy of the gap of g 1 when recombination, which is a value of 疋 electron volts composed of silicon. In other words, the 5th Tu Shi Tu & k caused by Xue Erji, also provides a proof that Ou Jie recombined to supply energy to the electrons in the n-shaped channel. Xiaotong is under control. 〇 Refer to Figure 6, Figure β is a virtual closed-pole voltage. Here, define a virtual record ㈡ = inter-pole current pair e ^ ^, 〇tvds = 3: 5 Λ ^ 1 fi] " ^ the gate voltage gradually increases from zero, flowing into the gate Vbs = 0, then; The maximum value of the current will appear at the idle pole and will increase. idle

Ms時,之後便隨著閘極電壓的增加逐漸減少==Ms, then gradually decrease with the increase of the gate voltage ==

434i4g 五、發明說明(10) 一般所知的金氧主 提高狀5伏特時電晶體之熱電子的發生完全-樣。當Vb 向電場下降,進而^為本體效應(b〇dy effect)而造成橫 大於0·5伏特後,流減少°然而,當基底電壓作 便開始大幅的上Θ間f流在閘電壓大於3. 5伏特以上的部分 於削伏特時^位尤其是化2伏特時,間極電流更是高 Vb大於0.5伏特時^以上,這是相當可觀的。換言之,當 敍!的地旦、# , ‘傑再結合可以使η形通道之電子有足 狗此里進入閘極產生電流。只是,Vbs也不可以太大。 士 ,,大於2伏特’基底與源極的Pn接面將因為-個 Κ β產生大量電流,反而將降低程式化之效率。 所以Vbs的建議值介於〇 5至2伏鸦之門 u /丨於ϋ. 5至^伙特之間。—次歐傑再結合 大約了以挺什一個電子16電子伏特的能量, Η。 所需的VdS可以比CHEI所需的Vds少l 6電子伏特以上,vds 最好是大於3伏特而小於5伏特。浮動間對源極的偏壓 VfS,如第6圖中所示,最少要大於2伏特以上才會有閘極 電流產生。控制閘的電壓則端看控制閘對浮動閘的耦合係 數(coupling ratio),一般的耦合係數大約為5〇%左右口, 也就是說控制閘的電壓必須大於5伏特。各個偏壓的值整 理如第1表。 第1表434i4g V. Description of the invention (10) The commonly known gold-oxygen main raises the thermoelectric generation of the transistor when the voltage is 5 volts. When Vb decreases to the electric field, and then ^ is the body effect (b〇dy effect), the current decreases after the horizontal is greater than 0. 5 volts. However, when the base voltage is applied, the current flow at the gate voltage greater than Θ is greater than 3 When the part above 5 volts is at ^, especially at 2 volts, the interpolar current is even higher when Vb is greater than 0.5 volts, which is quite considerable. In other words, when it comes to Syria! Didan, #, ‘Jie recombination can make the electrons of the η-shaped channel sufficient. The dog enters the gate here to generate current. But Vbs can't be too big. In fact, the Pn junction between the substrate greater than 2 volts and the source will generate a large amount of current because of a κ β, which will reduce the efficiency of stylization. Therefore, the recommended value of Vbs is between 05 and 2 and the gate of Volcano u / 丨 between ϋ5 and ^ Hot. —The sub-European recombination has about the energy of 16 electron volts for one electron, eh. The required VdS may be more than 16 electron volts less than the Vds required by CHEI, and vds is preferably greater than 3 volts and less than 5 volts. As shown in Figure 6, the bias voltage VfS of the source between the floats must be greater than 2 volts for the gate current to occur. The voltage of the control gate depends on the coupling ratio of the control gate to the floating gate. The general coupling coefficient is about 50%, which means that the voltage of the control gate must be greater than 5 volts. The values of each bias are adjusted as shown in Table 1. Table 1

Vds Vbs Vfs Vcs 綠值(伏特) 3-5 0.5-2 一- — 5-Vds Vbs Vfs Vcs Green value (Volts) 3-5 0.5-2 One-— 5-

43414 i 五、發明說明(11) 請參閱第7圖,第7圖為本發明之AECHEI以及習知CHEI 對一 EPROM的程式化特性圖。由第7圖中可知,當以習知的 CHEI在Vds = 3.5伏特電壓下對一 EPROM進行程式化時,不論 是控制閘極的電壓是1 2伏特或是1 5伏特,臨界電壓都在數 十亳秒(ms)後才慢慢的上昇。而一般商業用之EPROM的程 式化時間是毫秒或是以下,因此,由第7圖中可知,以 C HE ί來程式化一個EPROM並且Vds〜3 .5時,商業上而言根本 是不可行的。當以本發明之AECHEI對一EPROM進行程式化 時,當Vb = 2伏特、Vc = 1 2或1 5伏特時,經歷了一毫秒後, EPROM的臨界電壓便快速的拉升到一個飽和電位,也就是 說’本發明之AECHEI可以在Vds = 3. 5伏特之低電壓的條件 下’在合理的時間内’對一 E P R 0 Μ進行程式化。此外,本 發明的AECHE I可以使EPROM之臨界電壓拉昇達8伏特以上, 由第7圖所示’可使邏輯1與邏輯〇擁有較大的雜訊邊際 (noise margin) 〇 為了降低源極與基底導通時電流之電子電流部份,本 發明提供了 AECHE I執行時所使用的n形通道記憶體之結 構。請參閱第8圖’第8圖為一種本發明之AECHE I所使用之 EPROM的結構圖。本發明之用於AECHEI的EPROM為一閘堆疊 η形通道元件,包含有一基底30、一源極區32、一汲極區 34、一第一介電層36、一浮動閘極38、一第二介電層38以 及一控制閘極42。如第8圖所示’基底30可以是一ρ形基底 (P substrate),源極區32和汲極區34是兩個η形參雜區, 第一介電層36通常是厚度約100埃的氧化矽層,浮動閘極43414 i V. Description of the Invention (11) Please refer to FIG. 7, which is a programmatic characteristic diagram of the AECEI of the present invention and the conventional CHEI on an EPROM. As can be seen in Figure 7, when the EPROM is programmed with the conventional CHEI at Vds = 3.5 volts, the threshold voltage is no matter whether the gate voltage is 12 volts or 15 volts. It only rises slowly after ten leap seconds (ms). The programming time of EPROM for general business is milliseconds or less. Therefore, as shown in Figure 7, when EPROM is programmed with C HE ί and Vds ~ 3.5, it is not commercially feasible at all. of. When an EPROM is programmed with the AECHEI of the present invention, when Vb = 2 volts, Vc = 12 or 15 volts, after one millisecond, the threshold voltage of the EPROM quickly rises to a saturation potential. That is to say, the AECHEI of the present invention can program an EPR 0 M under a condition of low voltage of Vds = 3.5 volts within a reasonable time. In addition, the AECHE I of the present invention can raise the threshold voltage of EPROM to more than 8 volts, as shown in FIG. 7 'to make logic 1 and logic 0 have a larger noise margin. In order to reduce the source The electronic current part of the current when conducting with the substrate, the present invention provides the structure of the n-channel memory used in the execution of AECHE I. Please refer to FIG. 8 'FIG. 8 is a structural diagram of an EPROM used in the AECHE I of the present invention. The EPROM for AECHEI of the present invention is a gate stacked n-shaped channel element, which includes a substrate 30, a source region 32, a drain region 34, a first dielectric layer 36, a floating gate 38, a first Two dielectric layers 38 and a control gate 42. As shown in FIG. 8 'the substrate 30 may be a p-shaped substrate (P substrate), the source region 32 and the drain region 34 are two n-type impurity regions, and the first dielectric layer 36 is usually about 100 angstroms thick. Silicon oxide layer, floating gate

第14頁 434i4g 五、發明說明(12) 38設於第一介電層36上,可以甩多晶矽所構成,第二介電 層40設在浮動閘極38上,控制閘42設於第二介電層4〇上。 源極區32可以是一個雙擴散汲極(double di f fusi〇n d r a i η,D D D )結構。源極區3 2的形成包含有兩個步驟,先 以磷(P)以及砷(As)對源極區32的表面進行離子植入,然 後進行熱處理。因為磷的擴散速度較砷為快,所以罐的播 雜物會形成一較淡的η形參雜區32b,碎的摻雜物會形成一 較深的η形參雜區3 2 a,如第8圖所示。DDD結構的源極區3 2 可以降低pn接面導通時的電子電流比例,減少不必要的雪 流之浪費。 請參閱第9A圖以及第9B圖,第9A圖以及第9B圖為兩種 本發明之AECHEI所使用之EPROM的結構圖。為了增加產生 歐傑再合併的機率,可以在源極32至汲極34之電子流的下 方设一重摻雜區44,周以i曾加由基;^30至源極32之電 洞流比例,如第圖所示。而且,注入的地方就是通道中 電子剛剛要開始由源極32向汲極34運動的地方,可以增大 通道t產生歐傑再合併的機率,以產生較多的熱電子並增 加程式化之速度。相同的道理,也可以合併第g圖與第9/ 圖的結構,因此產生了第9B圖。第9B圖令的源極區32是— 们DDD.、.、》構,而且,基底中有一個重p形摻雜區Μ設於η 形通道的下方。—方面,DDD結構可以減少在Vbs呈正偏壓 時的電、子電流比例,另一方面,重p形摻雜區4 4可以增加 對η形通道的電洞注入,增加通道中歐傑再結合之效應。 為了避免Vbs呈正電壓時產生大量電流,本發明另提Page 14 434i4g 5. Description of the invention (12) 38 is provided on the first dielectric layer 36, which can be formed by throwing polycrystalline silicon, the second dielectric layer 40 is provided on the floating gate 38, and the control gate 42 is provided on the second dielectric Electrical layer 40. The source region 32 may be a double diffused drain (double di f fusion d r a i η, D D D) structure. The formation of the source region 32 includes two steps. First, the surface of the source region 32 is ion-implanted with phosphorus (P) and arsenic (As), and then heat-treated. Because the diffusion rate of phosphorus is faster than that of arsenic, the impurities in the tank will form a lighter n-type impurity region 32b, and the broken dopants will form a deeper n-type impurity region 3 2 a. Figure 8 shows. The source region 3 2 of the DDD structure can reduce the proportion of the electronic current when the pn junction is turned on, and reduce the waste of unnecessary snow current. Please refer to FIG. 9A and FIG. 9B. FIG. 9A and FIG. 9B are structural diagrams of EPROMs used by the two AECHEIs of the present invention. In order to increase the probability of Oujie recombination, a heavily doped region 44 can be set under the electron flow from source 32 to drain 34, and the base current ratio is from i to 30; ^ 30 to the hole current ratio of source 32 As shown in the figure. Moreover, the injection place is where the electrons in the channel are just beginning to move from source 32 to drain 34. The probability of channel O generating recombination can be increased to generate more hot electrons and increase the speed of programming. . For the same reason, the structure of the g-th graph and the 9 / th graph can also be merged, so the 9th graph is generated. The source region 32 of FIG. 9B is a DDD,..., And D structure. Moreover, a heavy p-type doped region M in the substrate is provided below the n-shaped channel. -On the one hand, the DDD structure can reduce the proportion of electricity and sub-current when Vbs is positively biased. On the other hand, the heavy p-doped region 44 can increase the hole injection into the n-shaped channel, and increase the recombination of the channel in the channel. effect. In order to avoid generating a large amount of current when Vbs is a positive voltage, the present invention provides another

第15頁 434841 五、發明說明(13) 供一種浮動閘記憶元陣列’以進行AECHE I,如第1 ο A圖所 示。記憶體陣列包含有至少兩個記憶元,也就是先前所述 之EPROM。每個記憶元包含有一基底30、一源極區32、一 没極區34、一第一介電層36、一浮動閘極38、一第二介電 層4 0以及一控制閘極4 2。複數個E P R 〇 Μ之控制閘極4 2連接 在一起形成一條字元線(word line),即第一連接線,如 第1 0A圖中的46a以及46b所示。複數個EPROM之汲極34連接 在一起形成一條位元線(bit line)45a〜45d,即第二連接 線。複數個EPROM之源極32連接在一起形成一源極線 (source 1 ine)48a,以及48b。至少一EPROM 的基底30 連接 到一第一接點50a ’至少一EPROM的基底30連接到一第二接 點50b ’而且,第一接點50a與第二接點50b在電性上相互 隔絕。第1 0 A圖中的記憶元陣列在進行AECHE I時能減少不 必要電流的消耗。譬如說’如果要以AECHEI對EPROM Ml之 浮動閘進行充電,則字元線46a施以1 2伏特、源極線48a接 地、位元線4 5 a施以3. 5伏特、以及第一接點5 〇 a施以2伏 特。由第10A圖中可以發現’基底對源極有正偏壓的epr〇m 只有Μ1和Μ 2 ’如果源極線4 8 a和源極線48 b相互電隔絕,Μ 3 和Μ4也不會有基底流到源極之電流的問題。而右半邊的 EPR0M(M5〜Μ8),因為第一接點5〇b和第二接點50b相互電隔 絕,所以在第一接點5 0 a上的電壓根本不會傳與第二接點 50b ’自然M5至M8之EPROM也不會有Vbs成正偏壓的問題。 請參閱第10B圖’第10B圖為第10A圖中記憶元陣列的 晶片實施例。本發明之記憶元障列的特徵在於某些的Page 15 434841 V. Description of the invention (13) A floating gate memory cell array is provided for AECHE I, as shown in Fig. 1A. The memory array contains at least two memory cells, which is the EPROM described previously. Each memory cell includes a substrate 30, a source region 32, an electrodeless region 34, a first dielectric layer 36, a floating gate 38, a second dielectric layer 40, and a control gate 42. . A plurality of control gates 42 of EPROM are connected together to form a word line, that is, a first connection line, as shown by 46a and 46b in FIG. 10A. The drain electrodes 34 of the plurality of EPROMs are connected together to form a bit line 45a to 45d, that is, a second connection line. The source electrodes 32 of the plurality of EPROMs are connected together to form a source line 48a and 48b. The substrate 30 of at least one EPROM is connected to a first contact 50a 'and the substrate 30 of at least one EPROM is connected to a second contact 50b', and the first contact 50a and the second contact 50b are electrically isolated from each other. The memory cell array in Fig. 10A can reduce unnecessary current consumption when performing AECHE I. For example, 'If you want to charge the floating gate of EPROM M1 with AECHEI, the word line 46a is 12 volts, the source line 48a is grounded, the bit line 4 5 a is 3.5 volts, and the first connection The point 50a is given 2 volts. From Figure 10A, it can be found that 'the substrate has positively biased epr0m from the source, only M1 and M2'. If source line 48a and source line 48b are electrically isolated from each other, M3 and M4 will not There is a problem with the current flowing from the substrate to the source. For the EPR0M (M5 ~ M8) in the right half, because the first contact 50b and the second contact 50b are electrically isolated from each other, the voltage on the first contact 50a will not be transmitted to the second contact at all. 50b 'Natural EPROMs of M5 to M8 will not have the problem of Vbs being positively biased. Please refer to FIG. 10B 'FIG. 10B is an embodiment of a chip of the memory cell array in FIG. 10A. The memory cell barrier of the present invention is characterized by some

第16頁 434842 五、發明說明(14) EPROM之基底與某些的EPROM之基底分開。為了達成以上之 特徵,本發明提出一種三重丼結構(triple well structure),如第10B圖所示。每'—個EPROM均設於一p形 井中,也就是說每個EPROM之基底就是一p形井=譬如說, 第10A圖中的Ml至M4設於一第一p形井中,因此第一p形井 就是第一接點50a,而第10A圖中的M5至M8設於一第二p形: 井中’所以第二p形井就是第二接點5〇b。而第一接點5〇a 和第一接點5 0 b之間就以一個;朱η形井(d e e ρ η - w e 1 i > D n W ) 作為電隔絕,如第1 OB圖所示。 本發明之AECHEI也可以運用於分閘式η形通道元件, 如第11圖所示。第11圖為一種分閘SnB通道元件示意 圖。除了一般堆疊式η形通道元件所有的基底、源極^、 汲極區、浮動閘極以及控制閘外 〇 分閘式η形通道元件另 包含有一閘氧化層61以及一選擇閘極“。閘氧化層61設方 浮動閘38與源極區32之間的基底3〇表面,而選擇間極⑴ f在閘氧化層62上,如第11圖所示。選擇閘極⑸可以控脅 =源極32流到沒極34之電流的大小,亦即控制在n形通道 中的電子數量。而Vbs所產生的電洞注入決定歐俾再結合 的數量。K要減少〇形通道中的電子Page 16 434842 V. Description of the invention (14) The EPROM substrate is separate from some EPROM substrates. In order to achieve the above characteristics, the present invention proposes a triple well structure, as shown in FIG. 10B. Each EPROM is set in a p-shaped well, that is to say, the base of each EPROM is a p-shaped well = for example, M1 to M4 in Figure 10A are set in a first p-shaped well, so the first The p-shaped well is the first contact 50a, and M5 to M8 in Fig. 10A are provided in a second p-shaped well: so the second p-shaped well is the second contact 50b. There is one between the first contact 50a and the first contact 50b; Zhu η-shaped well (dee ρ η-we 1 i > D n W) is used for electrical isolation, as shown in Figure 1 OB Show. The AECHEI of the present invention can also be applied to a gate-type n-channel element, as shown in FIG. 11. Figure 11 is a schematic diagram of an open SnB channel element. In addition to all the base, source, drain, floating gate, and control gates of a general stacked n-channel element, the opening n-channel element includes a gate oxide layer 61 and a select gate. The oxide layer 61 is provided on the surface of the substrate 30 between the square floating gate 38 and the source region 32, and the selection electrode ⑴f is on the gate oxide layer 62, as shown in Fig. 11. Selecting the gate electrode ⑸ can control the threat = source The amount of current flowing from the pole 32 to the pole 34 is the number of electrons controlled in the n-shaped channel. The hole injection generated by Vbs determines the number of recombination of the europium.

結合的數量,則對浮動閘極38注^ w β _加^彳^丹 加。 ,王入電子的效率就可以增 相較於習知的對一η形通道 方法,本發明之充電的方法施、力:己\體二,極充電的 壓Vbs於基底,基底之電洞备^;、;^·5伏特之順向偏 曰A入源極中。部分的電洞將The number of combinations is 38 w β _ plus ^ 彳 ^ denga to the floating gate. Compared with the conventional method of n-shaped channel, the efficiency of Wang Jin Electronics can be increased. The charging method of the present invention has the force and force: two, the voltage Vbs of the extreme charge on the substrate, and the electrical holes of the substrate. ^ ;, ^ · 5 volts forward bias A into the source. Some holes will

五、發明說明(15) ^________ 會>主入n形通道中並虚罝& 形通道中電子的能晉、\中的電子產生歐傑再結合以增加n 極在介於3〜5伏特圍因此可以降低沒極上的電麗’使汲 行充電。 ’依然可以對η形通道之浮動閘進 本發明雖以 較佳也 jt_ 定本發明,任何熟習此^施例揭露如上’然其並非用以限 和範圍内,當可做此畔的^藝者,在不脫離本發明之精神 範圍當視後附之申;;= ;者本發明之保護V. Description of the invention (15) ^ ________ Meeting> Mainly enter the n-shaped channel and virtual amp & The energy of the electrons in the & -shaped channel, the electrons in \ are generated and recombined to increase the n pole between 3 ~ 5 Volt Circumference can therefore reduce the electric power on the poles to charge the battery. 'It is still possible to enter the floating gate of the η-shaped channel. Although the present invention is better, the present invention can be determined by anyone. Anyone familiar with this ^ embodiment is disclosed above.' However, it is not intended to be within the scope and scope. , Attached without departing from the spirit of the present invention; =; the protection of the present invention

Claims (1)

434^42 六、申請專利範圍 ' " 一~' — ^「種EPROM元件中適合低汲極電壓操作之熱 ; 法,該元件包含有一源極、一汲極、一护1 程 一浮動開極以及-基底,該方法包含有: =相極、 的偏第一電壓予於該沒極以使該没極與該源極之門 的偏壓小於5伏特; 〜 < 間 βΪ供一第二電壓予該控制閘極以使該浮動閘極鱼嗲 極之間的偏壓大於2伏特;以及 …雨、 提供一第三電壓予該基極以使該基底與該源極之間 偏壓大於0. 5伏特,其中,對該浮動閘極充電之 該源極至該汲極之電子流而產生。 ,L係由 ^ 2.如申請專利範圍第1項之方法,其中,該第三電壓 :, 係使該基底與該源極之間的偏壓介於〇. 5伏特至2伏之 間。 J , 3. 如申請專利範圍第1項之方法,其中,該第—電壓 係使該汲極與該源極之間的偏壓介於3伏特至5伏特之間。 4. 如申請專利範圍第1項之方法,其中,該第二電壓 係使該浮動閘極與該源極之間的偏壓介於2伏特至7伏特之 間。 5. 如申請專利範圍第1項之方法,其中,該第二電壓 係使該控制閘極與該源極之間的偏壓大於5伏特。 6. —種η形通道元件,形成於一基底上,包含有: 一源極區以及一汲極區,形成於該基底表面; 一第一介電層,形成於該基底上; 一浮動閘極,形成於該第一介電層上,且部分之該浮434 ^ 42 6. Scope of patent application '" a ~' — ^ "A kind of heat suitable for low-drain voltage operation in EPROM components; this component includes a source, a drain, a protection, a process, a floating open And the substrate, the method includes: a phase bias, a first voltage biased to the pole so that the bias voltage between the pole and the source gate is less than 5 volts; ~ < Two voltages to the control gate so that the bias between the floating gate and the fish poles is greater than 2 volts; and ... rain, providing a third voltage to the base to bias the substrate and the source Greater than 0.5 volts, in which the floating gate charges the source to the drain of the electron current generated., L is from ^ 2. The method according to item 1 of the scope of patent application, wherein the third Voltage :, the bias voltage between the substrate and the source is between 0.5 volts and 2 volts. J, 3. The method according to item 1 of the patent application range, wherein the first voltage is The bias voltage between the drain and the source is between 3 volts and 5 volts. 4. If the method of the first scope of the patent application, the The second voltage is such that the bias voltage between the floating gate and the source is between 2 volts and 7 volts. 5. For the method of claim 1 in the patent application range, wherein the second voltage is The bias voltage between the control gate and the source is greater than 5 volts. 6.-An n-shaped channel element formed on a substrate including: a source region and a drain region formed on the surface of the substrate. A first dielectric layer formed on the substrate; a floating gate formed on the first dielectric layer, and a portion of the floating gate 434S42 六、申請專利範圍 動閘極與該汲極區重疊; . 一第二介電層’形成該浮動閘極上; 一控制閘極’形成於該第二介電層上; 一第一連接線’連接於該汲極區,用以使該汲極與該 源極之間的偏壓小於5伏特; 一第二連接線’連接於該控制閘,兩以使該浮動閘與 該源極之間的偏壓大於2伏特;以及 一第三連接線’連接於該基底,用以使該基底與該源 極之間的偏壓大於〇 · 5伏特; 其中’對該浮動閘極充電之電流係由該源極至該ί及 0 極之電子流而產生。 ' 7. 如申請專利範圍第6項之中,該η形通道元 件係為一閘堆疊式η形通道元件(stac【ed gate n-channei device) 〇 8. 如申請專利範圍第6項之其中,該n形通道元 件另包含有:. —閘氧化層,設於該浮動閘與該源極區之間的該基底 表面;以及 —選擇閘極,設於該閘氧化層上,用以控制該源極至 該汲極之電流。 & .) 9. 如申請專利範圍第6項之其中,該源極區係 為一雙擴散汲極(double di f ,DDD)結構。 10. 如申請專利範固第6項之中,該n形通道 元件另包含有一重Ρ形摻雜區,設於該源極至該汲極之電434S42 VI. Patent application range The moving gate overlaps the drain region; a second dielectric layer is formed on the floating gate; a control gate is formed on the second dielectric layer; a first connection line 'Connected to the drain region, so that the bias voltage between the drain and the source is less than 5 volts; a second connection line' is connected to the control gate, so that the floating gate and the source The bias voltage between the substrate and the source is greater than 0.5 volts; and a third connecting line is connected to the substrate to make the bias voltage between the substrate and the source greater than 0.5 volts; It is generated by the electron current from the source to the ί and 0 poles. '7. As for item 6 in the scope of patent application, the n-shaped channel element is a gate stacked n-channel device (stac [ed gate n-channei device) 〇 8. As in item 6 of the scope of patent application The n-shaped channel element further includes:-a gate oxide layer provided on the surface of the substrate between the floating gate and the source region; and-a selected gate electrode provided on the gate oxide layer for controlling The current from the source to the drain. &.) 9. According to item 6 of the scope of patent application, the source region is a double diffusion (DDD) structure. 10. As described in item 6 of the patent application, the n-channel element further includes a heavy P-shaped doped region, which is provided between the source and the drain. 第20頁 434S4 2 六、申請專利範圍 子流的下方,用以增加該基底至該源極導通時之電洞電流 比例。 11 · 一種浮動閘記憶元陣列,包含有: 至少二記憶元,每一記憶元包含有設於一基底表面之 一源極區以及一汲極區,一第一介電層,形成於該基底 上’一浮動閘極,形成於該第一介電層上,一第二介電 * 層’形成該浮動閘極上,一控制閘極,形成於該第二介電 層上’ 一第一連接線,連接於該汲極區,用以使該汲極與 該源極之間的偏壓小於5伏特,一第二連接線,連接於該Page 20 434S4 2 6. Scope of patent application Below the sub-flow, it is used to increase the ratio of the hole current when the substrate is turned on to the source. 11 · A floating gate memory cell array comprising: at least two memory cells, each memory cell including a source region and a drain region provided on a substrate surface, a first dielectric layer formed on the substrate A 'floating gate' is formed on the first dielectric layer, a second dielectric * layer is formed on the floating gate, a control gate is formed on the second dielectric layer ', a first connection A line connected to the drain region to make the bias voltage between the drain and the source less than 5 volts, and a second connection line connected to the 控制閘,兩以使該浮動閘與該源極之間的偏壓大於2伏 特’以及一第三連接線,連接於該基底,用以使該基底與 該源極之間的偏壓大於〇. 5伏特; 其中’對該浮動閘極充電之電流係由該源極至該汲極 之電子流而產生;Control the gate, so that the bias between the floating gate and the source is greater than 2 volts, and a third connection line connected to the substrate to make the bias between the substrate and the source greater than 0 5 volts; where 'the current that charges the floating gate is generated by the flow of electrons from the source to the drain; 其中至少一記憶元之基底連接至一第一接點,至少. 記憶元之基底連接至一第二接點,該第—接點與該第二」 點在電性上相互隔絕。 丨 一 1 2.如申請專利範圍第i丨項中:每一記憶 疋之基底係為一 P形井,且連接於該—第一接點之p形井與:The base of at least one memory cell is connected to a first contact, at least. The base of the memory cell is connected to a second contact, and the first contact and the second "point are electrically isolated from each other.丨 One 1 2. As in item i 丨 of the scope of patent application: the base of each memory 系 is a P-shaped well, and the p-shaped well connected to the first contact is: 接於戎第二接點之P形井係以一深η形井作為彼此的電隔 絕。The P-shaped well connected to the second junction of Rong uses a deep η-shaped well as the electrical insulation of each other. 第21頁Page 21
TW88121352A 1999-12-07 1999-12-07 Flash memory hot electron programming method that is suit for low drain voltage operation TW434842B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW88121352A TW434842B (en) 1999-12-07 1999-12-07 Flash memory hot electron programming method that is suit for low drain voltage operation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW88121352A TW434842B (en) 1999-12-07 1999-12-07 Flash memory hot electron programming method that is suit for low drain voltage operation

Publications (1)

Publication Number Publication Date
TW434842B true TW434842B (en) 2001-05-16

Family

ID=21643281

Family Applications (1)

Application Number Title Priority Date Filing Date
TW88121352A TW434842B (en) 1999-12-07 1999-12-07 Flash memory hot electron programming method that is suit for low drain voltage operation

Country Status (1)

Country Link
TW (1) TW434842B (en)

Similar Documents

Publication Publication Date Title
EP0360504B1 (en) One transistor flash eprom cell
US8076709B2 (en) Nonvolatile semiconductor memory device
US5242848A (en) Self-aligned method of making a split gate single transistor non-volatile electrically alterable semiconductor memory device
CN100456452C (en) Method of manufacturing a nonvolatile semiconductor memory device, and a nonvolatile semiconductor memory device
US7057931B2 (en) Flash memory programming using gate induced junction leakage current
CN101165903B (en) Non-volatile memory devices including double diffused junction regions and methods of fabricating the same
KR100397048B1 (en) A self-aligned buied channel/junction stacked gate flash memory cell
US6009017A (en) Floating gate memory with substrate band-to-band tunneling induced hot electron injection
EP0649178A2 (en) Flash EEPROM cell having gap between floating gate and drain for high hot electron injection efficiency for programming
US6111286A (en) Low voltage low power n-channel flash memory cell using gate induced drain leakage current
TWI238413B (en) Methods for enhancing erase of a memory device, programmable read-only memory device and method for preventing over-erase of an NROM device
KR100366599B1 (en) High energy buried layer implant with flash providing low resistance pit-well to pyramid array
US5726933A (en) Clipped sine shaped waveform to reduce the cycling-induced electron trapping in the tunneling oxide of flash EEPROM
US5468981A (en) Self-aligned buried channel/junction stacked gate flash memory cell
US5677876A (en) Flash EEPROM with impurity diffused layer in channel area and process of production of same
JPS63300567A (en) Floating gate type insulated gate field effect transistor
JP5095081B2 (en) Nonvolatile memory device and manufacturing method thereof
TW434842B (en) Flash memory hot electron programming method that is suit for low drain voltage operation
JP2963882B2 (en) Flash memory cell programming method
JPH05167078A (en) Semiconductor device and fabrication thereof
JP2757814B2 (en) Nonvolatile semiconductor memory device and method of manufacturing the same
US7638393B2 (en) Non-volatile memory device including nitrogen pocket implants and methods for making the same
JPH0462473B2 (en)
TW519736B (en) Method of programming and erasing non-volatile memory cells
KR100200074B1 (en) Fabrication method of non-volatile semiconductor memory device

Legal Events

Date Code Title Description
GD4A Issue of patent certificate for granted invention patent
MM4A Annulment or lapse of patent due to non-payment of fees