3280twf.doc / 0063280twf.doc / 006
經濟部中央標準局員工消费合作社印製 五、發明説明(/ ) 本發明是有關於一種記憶體結構、製造方法及其操作 方法,且特別是有關於一種多重狀態快閃記憶胞結構、製 造方法及其操作方法。 請參照第1圖,其繪示習知的一種AND型快閃記憶 體。以此圖爲例,吾人可在P型基底(P-substrate) 100中 形成N型深井(deep N-well) 110,再於N型深井110中 形成P型井區(P-well) 120,以作爲架構快閃記憶體之主 動區。接著,吾人可以P型井區120爲基礎,架構出快閃 記憶體結構,包括汲極130、淺摻雜區140、源極150、及 堆疊閘180,其中堆疊閘180可由控制閘160及浮置閘170 所組成,且汲極130、淺摻雜區140與源極150均爲N型 導電型,淺摻雜區140的離子摻雜濃度較汲極130與源極 150爲低。在後續的金屬連線製程中,可將控制閘160接 至字兀線(word line),汲極130接至位元線(bit line), 源極150則稱接至源極線(source line);此等AND型快 閃記憶體之操作方法將於下文中加以敘述。 AND型快閃記憶體之操作方法如下:當進行抹除 (erase)操作時,係利用“電子通道穿遂注入”(channel FN injection of electrons)加以實現,此時,各端點電壓分別 爲控制閘160施加10伏特電壓,源極150與P型井區120 則施加·8伏特電壓,汲極130爲浮接(floating)狀態。程 式(programming)操作時,係利用“電子邊緣穿遂射出” (edage FN ejection of electrons)加以實現,此時,各端點 電壓分別爲控制閘160施加-10伏特電壓,汲極130施加6 -------Ί,--装-- ·to (請先閱讀背面之注意事項再填朽本頁)Printed by the Employees' Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs 5. Description of the Invention (/) The present invention relates to a memory structure, a manufacturing method and an operating method thereof, and in particular to a multi-state flash memory cell structure and a manufacturing method And how to do it. Please refer to FIG. 1, which shows a conventional AND flash memory. Taking this figure as an example, we can form a deep N-well 110 in a P-substrate 100, and then form a P-well 120 in the deep N-well 110. As the active area of the architecture flash memory. Then, we can use the P-type well region 120 as the basis to construct a flash memory structure, including a drain 130, a lightly doped region 140, a source 150, and a stack gate 180. The stack gate 180 can be controlled by the control gate 160 and the floating gate. The gate 170 is formed, and the drain 130, the shallow doped region 140, and the source 150 are all N-type conductive types. The ion doping concentration of the shallow doped region 140 is lower than that of the drain 130 and the source 150. In the subsequent metal connection process, the control gate 160 can be connected to the word line, the drain 130 can be connected to the bit line, and the source 150 can be connected to the source line. ); The operation method of these AND-type flash memories will be described below. The operation method of AND flash memory is as follows: When erasing (erase) operation, it is realized by using "channel FN injection of electrons". At this time, the voltages at each end point are controlled separately. The gate 160 applies a voltage of 10 volts, the source 150 and the P-well region 120 apply a voltage of 8 volts, and the drain 130 is in a floating state. In programming operation, it is realized by "edage FN ejection of electrons". At this time, the terminal voltages are respectively -10 volts applied to the control gate 160 and 6- ------ Ί,-装-· to (Please read the precautions on the back before filling out this page)
-、1T '丨線 本紙張尺度適用中國國家標準(CNS ) Μ規格(210Χ297公兑) 3280twf.doc/006 3280twf.doc/006 經濟部中央標準局員工消资合作社印製 B? 五、發明说明(之) 伏特電壓,P型井區120接地,源極150爲浮接狀態。在 過去的文獻中’證明了當快閃記憶體在執行“電子邊緣穿 遂射出”操作時,經常會伴隨“能帶一能帶穿遂效應” (band-to-band tunneling),此“能帶一能帶穿遂效應”所 造成的漏電流受橫向電場加速的影響,容易衍生出“熱電 洞注入”的不利影響’造成元件特性的退化。 綜觀以上所述,習知上所採用之快閃記憶體在執行 “電子邊緣穿遂射出”操作時,經常會伴隨“能帶一能帶 穿遂效應”,此“能帶一能帶穿遂效應”所造成的漏電流 受橫向電場加速的影響,容易衍生出“熱電洞注入”的不 利影響,造成穿遂氧化層的退化。 因此本發明的目的就是在提供一種多重狀態互補式快 閃記億胞結構’以避免元件特性退化,並藉由寄生之放大 電路提昇記憶胞之存取效能。 因此本發明的另一目的就是在提供一種多重狀態互補 式快閃記憶胞之製造方法,其製造方法係相對於本發明所 提之結構。 因此本發明的另一目的就是在提供一種多重狀態互補 式快閃記憶胞之操作方法,其操作方法係相對於本發明所 提之結構。 爲達成上述及其他目的,本發明提供一種多重狀態互 補式快閃記憶胞’吾人可在P型基底中形成N深井,再於 N型深井中形成P型井區,以作爲架構互補式快閃記憶胞 之主動區。接著’吾人可以p型井區爲基礎,架構出互補 4 適用中國ΐ家標準(CNS ) --- ----------^------1T----- »*· (請先閱讀背面之注意事項再填寫本页) 3280twf.d〇c/〇〇6 A? H7 —… .............------ 五、發明説明(>) 式快閃記憶胞,包括汲極、浮置摻雜區'源極、及推疊閘。 堆疊閘可由控制閘及浮置閘所組成,浮置摻雜區及源極之 電性與P型井區不同,爲N型導電型,且浮置摻雜區的離 子摻雜濃度較源極爲低;汲極之電性與P型井區相同,爲 P型導電型。此等互補式快閃記憶胞結構可寄生有雙戰子 電晶體,以作爲電流放大之用,並且相對於此等結構,接 露出其相關的製造方法及操作方法。 爲讓本發明之上述和其他目的、特徵、和優點能更明 顯易懂,下文特舉一較佳實施例,並配合所附圖式,作詳 細說明如下: 圖式之簡單說明: 第1圖繪示習知的一種AND型快閃記憶體; 第2圖繪示依照本發明之一較佳實施例,所提供的一 種多重狀態互補式快閃記憶胞結構圖; 第3A至3E圖繪示對應第2圖所示之互補式快閃記憶 胞結構,所提供之一種多重狀態互補式快閃記憶胞的製作 方法; 第4圖^第2圖所提供之多重狀態互補式快閃 記億的寄生電路圖;以及 S 乙4) ||^^^用以說明利用本發明之多重狀態互補式快閃 記憶胞之操作方法。 圖式之標記說明: 30 堆疊閘; 100 P型基底; 110 N型深井; 120 P型井; 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公沒) ---------拍衣-------IT-----—漆. (請先閱讀背面之注意事Jii再填寫本頁} 經濟部中央標準局貝工消费合作社印製 3280twf.doc/006 Λ7 五、發明説明(f) 130 汲極; 140 淺摻雜區; 150 源極; 160 控制閘; 170 浮置閘; 180 堆疊閘; 200 P型基底; 210 N型深井; 220 P型井區; 230 汲極; 240 浮置摻雜區; 250 源極; 260 控制閘; 270 浮置閘; 280 堆疊閘; 300 P型基底; 310 N型深井; 320 P型井區; 330, 330a 穿遂氧化層; 340 多晶矽; 340a 浮置閘; 350, 350a 介電層; 360 多晶砂; 360a 控制閘; 370 N型摻雜區; 370a 浮置摻雜區 380 汲極;以及 390 源極。 較佳實施例-、 1T '丨 The size of the paper is applicable to the Chinese National Standard (CNS) M specification (210 × 297). 3280twf.doc / 006 3280twf.doc / 006 Printed by the Consumers ’Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs. 5. Description of the invention (A) Volt voltage, the P-type well area 120 is grounded, and the source electrode 150 is in a floating state. In the past literature, it has been proven that when flash memory performs the "electronic edge tunneling ejection" operation, it is often accompanied by "band-to-band tunneling". The leakage current caused by the "band-band energy tunneling effect" is affected by the acceleration of the lateral electric field, and it is easy to derive the adverse effect of "hot hole injection", which leads to the degradation of element characteristics. In summary, the flash memory used in practice is often accompanied by the "band-band-band-pass effect" when performing the "electronic edge pass-through shot" operation. This "band-band pass-through effect" The "leakage current" caused by the "effect" is affected by the acceleration of the lateral electric field, which can easily lead to the adverse effect of "hot hole injection" and cause the degradation of the tunneling oxide layer. Therefore, the object of the present invention is to provide a multi-state complementary flash memory cell structure 'to avoid the degradation of element characteristics, and to improve the memory cell access performance by a parasitic amplification circuit. Therefore, another object of the present invention is to provide a method for manufacturing a multi-state complementary flash memory cell. The manufacturing method is relative to the structure provided by the present invention. Therefore, another object of the present invention is to provide a method for operating a multi-state complementary flash memory cell. The operation method is relative to the structure provided by the present invention. In order to achieve the above and other objectives, the present invention provides a multi-state complementary flash memory cell. We can form an N-deep well in a P-type substrate, and then form a P-well area in the N-type deep well, as a complementary flash architecture. Active area of memory cells. Then 'I can use p-type wells as a basis, and complement each other. 4 Applicable Chinese Standards (CNS) --- ---------- ^ ------ 1T ----- » * · (Please read the precautions on the back before filling this page) 3280twf.d〇c / 〇〇6 A? H7 —... .............-------- V. DESCRIPTION OF THE INVENTION A > type flash memory cell includes a drain, a floating doped region 'source, and a push-pull gate. The stacking gate can be composed of a control gate and a floating gate. The electrical properties of the floating doped region and the source are different from those of the P-type well region. Low; the electrical property of the drain electrode is the same as that of the P-type well area, which is a P-type conductive type. These complementary flash memory cell structures can be parasitic with a bipolar transistor for current amplification, and relative to these structures, the related manufacturing methods and operating methods are exposed. In order to make the above and other objects, features, and advantages of the present invention more comprehensible, a preferred embodiment is given below in conjunction with the accompanying drawings for detailed description as follows: Brief description of the drawings: FIG. 1 FIG. 2 shows a conventional AND flash memory. FIG. 2 shows a structure diagram of a multi-state complementary flash memory cell provided according to a preferred embodiment of the present invention. FIGS. 3A to 3E show Corresponding to the structure of the complementary flash memory cell shown in FIG. 2, a method for manufacturing a multi-state complementary flash memory cell is provided; FIG. 4 ^ Paragraph of the multi-state complementary flash memory provided in FIG. 2 The circuit diagram; and S B 4) || ^^^ are used to explain the operation method using the multi-state complementary flash memory cell of the present invention. Explanation of the marks in the drawings: 30 stack gates; 100 P-type bases; 110 N-type deep wells; 120 P-type wells; This paper size applies to China National Standard (CNS) A4 specification (210X297 public) -------- -拍 衣 ------- IT -----— lacquer. (Please read the notice on the back Jii before filling out this page} Printed by the Shellfish Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs 3280twf.doc / 006 Λ7 V. Description of the invention (f) 130 drain; 140 shallow doped region; 150 source; 160 control gate; 170 floating gate; 180 stacked gate; 200 P-type base; 210 N-type deep well; 220 P-type well area; 230 drain; 240 floating doped region; 250 source; 260 control gate; 270 floating gate; 280 stacked gate; 300 P-type base; 310 N-type deep well; 320 P-type well area; 330, 330a through oxidation Layer; 340 polycrystalline silicon; 340a floating gate; 350, 350a dielectric layer; 360 polycrystalline sand; 360a control gate; 370 N-type doped region; 370a floating doped region; 380 drain; and 390 source. Better Examples
請參照第2圖,其繪示依照本發明之一較佳實施例, 所提供的一種多重狀態互補式快閃記憶胞結構圖。以此圖 爲例,吾人可在P型基底200中形成N型深井210,再於 N型深井210中形成P型井區220,以作爲架構互補式快 閃記憶胞之主動區。接著,吾人可以P型井區220爲基礎, 架構出互補式快閃記憶胞,包括汲極230、浮置摻雜區240、 源極250、及堆疊閘280。堆疊閘280可由控制閘260及浮 置閘270所組成,浮置摻雜區240及源極250之電性與P 本紙張尺度適用中國國家標準(CNS ) Λ4規格(2丨0X297公处) 3280 twf. doc /006 3280 twf. doc /006 經濟部中央標準局貝工消费合作社印製 五、發明説明(乡) 型井區220不同,爲N型導電型’且浮置摻雜區240的離 子摻雜濃度較源極250爲低;汲極230之電性與P型井區 220相同,爲P型導電型。 整體而言,此等互補式快閃記憶胞係架構於P型井區 220處,源極250位於P型井區220之表面,汲極230亦 位於P型井區220表面並以一距離與源極250相鄰;堆疊 閘280則覆蓋源極250與汲極230間之P型井區220表面。 此外,浮置摻雜區240係位於汲極230周圍並延伸至源極 250與汲極230間之P型井區220表面處,並與堆疊閘280 耦接;浮置摻雜區240之電性與源極250相同,並介於P 型井區220與汲極230之間。 在後續的金屬連線製程中,可將控制閘260接至字元 線(word line),汲極230接至位元線(bit line),源極250 則耦接至源極線(source line);此等互補式快閃記憶胞 之製作方法將於下文中加以敘述。 請參照第3A至3E圖,其繪示對應第2圖所示之互補 式快閃記憶胞結構,所提供之一種互補式快閃記憶胞的製 作方法。請參照第3A圖,首先,吾人可利用N型離子植 入P型基底300中,以形成N型深井310,並將P型離子 植入N型深井310中,以形成P型井區320。以P型井區 320爲基礎,即可架構出此等互補式快閃記憶胞結構。 請參照第3B圖,吾人可於P型井區320表面上形成隧 穿氧化層330,並形成一多晶矽層340,覆蓋於隧穿氧化 層330之表面。接著,定義多晶矽層340及隧穿氧化層33〇, ---------裝------訂----->.丨線 -/ (請先閱讀背而之注意事項再填寫本s ) 本紙張尺度適用中國國家縣(CNS) A4規格(21〇χ297公楚〉 3280twf.doc/ 006 ΑΊ ____ H? 五、發明説明(έ ) 即可成爲浮置閘340a,且隧穿氧化層330a介於浮置閘340a 與P型井區320之間。 請參照第3C圖,吾人可形成一介電層350覆蓋於浮 置閘340a表面,並形成一多晶矽層360覆蓋於介電層350 表面;如第3D圖所示,定義多晶矽層360後,即可成爲 控制閘360 a’且介電層350a介於控制閘360 a與浮置閘340a 之間。此外,爲簡化說明起見,吾人可將控制閘360 a、 介電層350a、浮置閘340a與隧穿氧化層330a總稱爲堆疊 閘30,以利後續之說明。 堆疊閘30形成後,可利用離子植入法,於P型井區320 表面處形成N型摻雜區370,N型摻雜區370與堆疊閘30 之一側相鄰並延伸至堆疊閘30之底部,與隧穿氧化層330a 相接觸。 請參照第3E圖,吾人可利用離子植入法,於N型摻 雜區370內形成P型摻雜區,·以作爲互補式快閃記憶胞之 汲極380,其中,汲極380之離子摻雜濃度較P型井區320 爲高;需要注意的是,當汲極380形成的同時,亦將原有 的N型摻雜區370定義爲浮置摻雜區370a,亦即浮置摻雜 區370a係位於汲極380周圍,並延伸至堆疊閘30底部。 此外,吾人並可於於P型井區320表面處形成另一N型摻 雜區,與堆疊閘30之另一側相鄰,以作爲互補式快閃記 憶胞之源極390 ;其中,源極390之離子摻雜濃度較浮置 摻雜區370a爲高。 最後,利用金屬連線製程,可將控制閘360a接至字 本紙張尺度適用中國國家標準(CNS ) A4規格(21〇Χ297公犮) 3280twf. doc / 006 hi H7 _ _____ 五、發明説明(7) 元線,汲極380接至位元線,源極390則耦接至源極線, 以執行資料編碼、讀取及抹除等操作。 請參照第4圖,其繪示依照第2圖所提供之互補式快 閃記憶胞結構,所產生的寄生電路圖。首先吾人必須瞭解, 本發明與第1圖所示之習知技藝相異之處,在於習知的記 憶胞汲極爲N型摻雜,而本發明所提供之記憶胞結構,係 以P型摻雜形成汲極,故第1圖所示之習知技藝僅爲單一 快閃記憶胞結構,而依照第2圖所提供之互補式快閃記憶 胞結構,則可衍生出如第4圖所示之等效電路;下文中將 以第4圖所示之等效電路爲基礎,針對其所獲致之利益加 以說明。 由圖示中可看出,寄生電晶體Ml與寄生電晶體M2 係共用同一控制閘,此外,寄生電晶體Ml之源極與寄生 電晶體M2之井區耦接,寄生電晶體M2之汲極與寄生電 晶體Ml之井區耦接。需要注意的是,此等互補式快閃記 憶胞結構寄生有一電晶體Q1,電晶體Q1之基極與寄生電 晶體Ml之井區耦接,電晶體Q1之射極與寄生電晶體Ml 之汲極耦接,電晶體Q1之集極與寄生電晶體Ml之源極 耦接。由於此等互補式快閃記憶胞之等效電路,除具備與 第1圖中所示結構等效之寄生電晶體M2外,更包括了寄 生電晶體Ml與電晶體Q1,故整體的電路運作將更趨複雜; 大體而言,由於寄生電晶體M2的汲極係作爲電晶體Q1 之基極,且電晶體Q1之射極係電晶體Ml之汲極,因此, 藉由電晶體Q1之放大作用,可將電晶體Ml之汲極電流 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) I I裝 訂 "線 I (請先閱讀背面之注意事K填寫本頁) 經濟部中央標準局貝工消費合作社印製 3280twf.doc/006 Η7 ______ 五、發明説明(?) 放大爲電晶體M2汲極電流之l+β倍’由於電晶體M1之汲 極係耦接至位元線,故與第1圖所示之記憶胞結構相比’ 此等互補式快閃記億胞之汲極電流將增加l+β倍’故可大 幅提昇資料存取之效能。 下文中將說明與本發明之互補式快閃記憶胞相對應之 操作方法。 請參照第5圖,第5圖係用以說明利用本發明之互補 式快閃記憶胞之操作方法,下文中將針對資料的抹除 (erase)、讀取(read)及程式編碼(programing)加以說 明。爲簡化圖示及便於說明起見,下文及圖示中將以Vwl 表示字元線電壓,其中字元線係與堆疊閘中的控制閘耦 接;V&表示位元線電壓,其中位元線係與記憶胞的汲極 耦接;Vslj表示源極線電壓,其中源極線係與記憶胞的源 極耦接;VPW表示井區電壓,用以表示P型井區的電位。 詳細的操作方法將於下文中加以敘述。 請參照第5圖(1),其繪示互補式快閃記憶胞之抹 除操作示意圖。當執行抹除操作時,係利用“電子通道穿 遂射出’’ (channel FN ejunction of electrons)來完成,其 端點偏壓分別爲VWL=-l〇V ’ VSL=8V,VPW=8V,汲極則爲 浮接狀態(Hoating);藉此,可將電子由浮置閘中射出互 補式快閃記憶胞,完成抹除之操作。 請參照第5圖(2),其繪示互補式快閃記憶胞之讀 取操作示意圖。當執行讀取操作時,各端點偏壓分別爲 VWL=3〜5V ’ VBL=1V,VSL=〇V ’ VPW=〇V ;藉此,即可完成 10 本紙張尺度適用中國國家標準(CNS ) Λ4規格(210X297公徒) 3280twf. doc / 006 3280twf. doc / 006 經濟部中央標準局貝工消f合作社印製 五、發明説明(?) 互補式快閃記憶胞之讀取操作。需要注意的是’由於寄生 雙載子電晶體(第4圖),故而當互補式快閃記憶胞進行 讀取操作時,導通電流會經由此寄生之雙載子電晶體放 大-,因此在位元線所得之讀取電流較傳統之快閃記憶胞 大,約爲過去之l + β倍,如此,即可以較簡單之周邊電路 (peripheral circuit)進行較快的讀取操作,減低周邊電路 設計成本。 請參照第5圖(3),其繪示互補式快閃記憶胞之程 式編碼操作示意圖。當執行程式編碼操作時’係利用“能 帶一能帶穿遂效應感生熱電子注入” (band-to-band tunneling induced hot electron injuntion)來完成,其端點偏 壓分別爲 VWL=8〜10V,VBL=-6〜-3V,VSL=0〜3V,VPW=0V ; 藉此,可將電子由浮置摻雜區中射入浮置閘,完成程式編 碼之操作。 請參照第5圖(4),其繪示互補式快閃記憶胞之多 重狀態程式編碼操作時序圖。此種多重狀態程式編碼可分 爲兩階段完成,第一階段係利用“能帶一能帶穿遂效應感 生熱電子注入"(band-to-band tunneling induced hot electron injunUoii)來進行電子注入,其端點偏壓分別是在控制閘 加上一程式電壓,例如偏壓VPC]M,在汲極加上偏壓VBL, 在源極加上偏壓Vsl_,P型井區接地,其中偏壓大小如前 所述。第二階段則是藉由寄生之P型通道電晶體(寄生電 晶體M2 ’第4圖)的導通,將位元線的電壓放電至P型 井區的電位,如此電子無法再注入,進而達到第一自我收 本紙張尺度適用中國國豕標準(CNS ) A4規格(2ΐ〇χ 297公势) ---------^------II------^ {褚先閱讀背而之_注意事^:填寫本頁) 3280twf.doc/006 五、發明説明(/C ) 斂狀態;其端點偏壓分別是在控制閘加上一放電電壓,例 如偏壓vDISI ’在汲極加上偏壓VBL,在源極加上偏壓Vn, P型井區接地。以不同的控制閘偏壓vDIS2重複上述多重狀 態程式編碼操作,可達到第二自我收斂狀態;同理,以不 同之控制閘偏壓VDISn,可達到第η自我收斂狀態,達到多 重狀態的程式編碼操作。 綜觀以上所述,與習知上所採用之快閃記'億胞相比 較’本發明所提供之互補式快閃記憶胞至少具有下列優 點: 一、 當互補式快閃記憶胞進行讀取操作時,導通電流 會經由此寄生之雙載子電晶體放大,因此在位元線所得之 讀取電流較傳統之快閃記憶胞約增加l+β倍,可以較簡單 之周邊電路進行較快的讀取操作,減低周邊電路設計成 本。 二、 利用此等互補式快閃記憶胞結構,當進行“抹除” 與“程式編碼”操作時並無熱電洞產生,可大幅提昇穿遂 氧化層的可靠度,使互補式快閃記憶胞結構不易退化》 經濟部中央標隼局貝工消費合作社印製 三、 利用此等互補式快閃記憶胞結構,當進行“程式 編碼”操作時係利用“能帶一能帶穿遂效應感生熱電子注 入”來完成,可提昇記憶胞之工作效能。 四、 利用不同之控制閘偏壓VDISn,可達到第η自我收 斂狀態,達到多重狀態的程式編碼操作。 雖上文中係針對以Ρ型井區所架構出的互補式快閃記 憶胞結構、製作方法及操作方法加以說明,然並非用以限 本紙張尺度適用中國國家標準(CNS ) Α4规格(210Χ297公鏟) 3280twf.doc/006 1Γ 五、發明説明(// ) 定本發明,熟悉此技藝之人士應用N型井區架構出與本發 明相對之互補式快閃記憶胞結構、製作方法及操作方法, 應不脫離本發明之精神。 •以上所述僅爲本發明之較佳實施例,然其並非用以限 定本發明,任何熟習此技藝者,在不脫離本發明之精神和 範圍內,當可作各種之更動與潤飾,因此本發明之保護範 圍當視後附之申請專利範圍所界定者爲準,凡依本發明申 請專利範圍所做之均等變化與修飾,皆應屬本發明專利之 涵蓋範圍。 . 訂1·^ 裊 . (請先閱讀背面之注意事項再填寫本頁) 經濟部中央標準局員工消资合作社印製 本紙張尺度適用中國國家標率(〇阳),\4規格(2丨0乂297公赴)Please refer to FIG. 2, which illustrates a structure diagram of a multi-state complementary flash memory cell according to a preferred embodiment of the present invention. Taking this figure as an example, we can form an N-type deep well 210 in the P-type substrate 200, and then form a P-type well region 220 in the N-type deep well 210 as the active area of the complementary flash memory structure. Then, we can use the P-type well region 220 as the basis to construct a complementary flash memory cell, including a drain 230, a floating doped region 240, a source 250, and a stack gate 280. The stacking gate 280 can be composed of a control gate 260 and a floating gate 270. The electrical properties and P of the floating doped region 240 and the source 250 are in accordance with the Chinese National Standard (CNS) Λ4 specification (2 丨 0X297) 3280 twf. doc / 006 3280 twf. doc / 006 Printed by the Central Laboratories of the Ministry of Economic Affairs, Shelley Consumer Cooperative, V. Description of Invention (Village) Well type 220 is different, it is N-type conductivity type, and floating doped area 240 ions The doping concentration is lower than the source 250; the drain 230 has the same electrical properties as the P-type well region 220 and is a P-type conductive type. Generally speaking, these complementary flash memory cell systems are structured at the P-type well region 220, the source 250 is located on the surface of the P-type well region 220, and the drain 230 is also located on the surface of the P-type well region 220 with a distance from The source electrodes 250 are adjacent; the stack gate 280 covers the surface of the P-type well region 220 between the source electrodes 250 and the drain electrodes 230. In addition, the floating doped region 240 is located around the drain 230 and extends to the surface of the P-type well region 220 between the source 250 and the drain 230, and is coupled to the stack gate 280; The property is the same as that of the source 250 and is between the P-well region 220 and the drain 230. In the subsequent metal connection process, the control gate 260 can be connected to a word line, the drain 230 can be connected to a bit line, and the source 250 can be coupled to a source line. ); The manufacturing method of these complementary flash memory cells will be described below. Please refer to FIGS. 3A to 3E, which illustrate a method for manufacturing a complementary flash memory cell corresponding to the complementary flash memory cell structure shown in FIG. 2. Referring to FIG. 3A, first, we can implant N-type ions into P-type substrate 300 to form N-type deep well 310, and implant P-type ions into N-type deep well 310 to form P-type well region 320. Based on the P-type well area 320, these complementary flash memory cell structures can be constructed. Referring to FIG. 3B, we can form a tunneling oxide layer 330 on the surface of the P-type well region 320, and form a polycrystalline silicon layer 340 to cover the surface of the tunneling oxide layer 330. Next, define the polycrystalline silicon layer 340 and the tunneling oxide layer 33. --------- Installation ------ Order ----- >. Line- / (Please read the back first Note: Please fill in this s) This paper size is applicable to China National County (CNS) A4 specifications (21〇297297) 3280twf.doc / 006 ΑΊ ____ H? V. Description of the invention (έ) can become a floating gate 340a, And the tunneling oxide layer 330a is between the floating gate 340a and the P-type well region 320. Please refer to FIG. 3C, we can form a dielectric layer 350 to cover the surface of the floating gate 340a, and form a polycrystalline silicon layer 360 to cover On the surface of the dielectric layer 350; as shown in FIG. 3D, after the polycrystalline silicon layer 360 is defined, it can become the control gate 360a 'and the dielectric layer 350a is between the control gate 360a and the floating gate 340a. In addition, For the sake of simplicity, we can collectively refer to the control gate 360a, the dielectric layer 350a, the floating gate 340a, and the tunneling oxide layer 330a as the stack gate 30 for the convenience of subsequent explanations. After the stack gate 30 is formed, the ion implantation can be used. In the method, an N-type doped region 370 is formed at the surface of the P-type well region 320. The N-type doped region 370 is adjacent to one side of the stack gate 30 and extends to the bottom of the stack gate 30. The tunneling oxide layer 330a is in contact. Please refer to FIG. 3E. We can use the ion implantation method to form a P-type doped region in the N-type doped region 370 as the drain 380 of the complementary flash memory cell. Among them, the ion doping concentration of the drain 380 is higher than that of the P-type well region 320. It should be noted that when the drain 380 is formed, the original N-type doped region 370 is also defined as a floating dopant. The region 370a, that is, the floating doped region 370a is located around the drain electrode 380 and extends to the bottom of the stack gate 30. In addition, we can form another N-type doped region on the surface of the P-type well region 320, and The other side of the stack gate 30 is adjacent to serve as the source 390 of the complementary flash memory cell; wherein the ion doping concentration of the source 390 is higher than that of the floating doped region 370a. Finally, a metal connection process is used. , Can connect the control brake 360a to the copybook paper size applicable Chinese National Standard (CNS) A4 specification (21〇297297) 3280twf. Doc / 006 hi H7 _ _____ 5. Description of the invention (7) Yuan line, drain electrode 380 Connected to the bit line, and source 390 is coupled to the source line to perform data encoding, reading, and erasing operations Please refer to Fig. 4, which shows the parasitic circuit diagram generated according to the complementary flash memory cell structure provided in Fig. 2. First of all, we must understand that the present invention is different from the conventional technique shown in Fig. 1 The reason is that the conventional memory cell drain is an N-type dopant, and the memory cell structure provided by the present invention is formed by a P-type dopant to form a drain. Therefore, the conventional technique shown in FIG. Flash memory cell structure, and according to the complementary flash memory cell structure provided in Figure 2, the equivalent circuit shown in Figure 4 can be derived; the equivalent circuit shown in Figure 4 will be used below Basis, explain the benefits it has obtained. It can be seen from the figure that the parasitic transistor M1 and the parasitic transistor M2 share the same control gate. In addition, the source of the parasitic transistor M1 is coupled to the well region of the parasitic transistor M2 and the drain of the parasitic transistor M2 It is coupled to the well region of the parasitic transistor M1. It should be noted that these complementary flash memory cell structures parasitic a transistor Q1, the base of transistor Q1 is coupled to the well region of parasitic transistor M1, the emitter of transistor Q1 and the parasitic transistor M1 are drawn. The collector of transistor Q1 is coupled to the source of parasitic transistor M1. The equivalent circuit of these complementary flash memory cells, in addition to having a parasitic transistor M2 equivalent to the structure shown in Figure 1, includes a parasitic transistor M1 and a transistor Q1, so the overall circuit operation Will be more complicated; generally speaking, since the drain of the parasitic transistor M2 is used as the base of transistor Q1, and the emitter of transistor Q1 is the drain of transistor M1, therefore, the amplification of transistor Q1 Function, the drain current of the transistor M1 can be applied to the paper size of the Chinese National Standard (CNS) A4 (210X297 mm) II binding " line I (please read the note on the back first and fill in this page) Printed by the Bureau of Standardization and Consumer Electronics Co., Ltd. 3280twf.doc / 006 Η7 ______ V. Description of the invention (?) Enlarged to 1 + β times of the drain current of transistor M2 because the drain of transistor M1 is coupled to the bit line Therefore, compared with the memory cell structure shown in Figure 1, the drain current of these complementary flash memory cells will increase by l + β times, so the performance of data access can be greatly improved. The operation method corresponding to the complementary flash memory cell of the present invention will be described below. Please refer to FIG. 5. FIG. 5 is a diagram for explaining an operation method using the complementary flash memory cell of the present invention. In the following, data erasure, read, and programming are described. Explain. In order to simplify the illustration and facilitate the description, the word line voltage is represented by Vwl in the following and in the figure, where the word line is coupled to the control gate in the stacking gate; V & represents the bit line voltage, where the bit The line is coupled to the drain of the memory cell; Vslj represents the source line voltage, where the source line is coupled to the source of the memory cell; VPW represents the well area voltage and is used to represent the potential of the P-type well area. The detailed operation method will be described below. Please refer to FIG. 5 (1), which shows the erase operation of the complementary flash memory cell. When the erasing operation is performed, it is completed by "channel FN ejunction of electrons", and the end point bias voltages are VWL = -l0V 'VSL = 8V, VPW = 8V, drain The pole is in a floating state (Hoating); by this, electrons can be ejected from the floating gate into a complementary flash memory cell to complete the erasing operation. Please refer to FIG. 5 (2), which shows the complementary fast Schematic diagram of the read operation of the flash memory cell. When the read operation is performed, the bias voltage of each terminal is VWL = 3 ~ 5V 'VBL = 1V, VSL = 〇V' VPW = 〇V; by this, 10 can be completed. This paper size applies the Chinese National Standard (CNS) Λ4 specification (210X297 male) 3280twf. Doc / 006 3280twf. Doc / 006 Printed by the Central Standards Bureau of the Ministry of Economic Affairs, Bei Gong Xiao F Cooperatives 5. Description of the invention (?) Complementary flash Read operation of the memory cell. It should be noted that 'due to the parasitic bipolar transistor (Figure 4), when a complementary flash memory cell performs a read operation, the on-state current will pass through this parasitic bipolar transistor. The crystal is amplified-so the read current obtained on the bit line is faster than the traditional flash The memory cell is large, about l + β times of the past. In this way, a simpler peripheral circuit can perform faster reading operations and reduce the cost of peripheral circuit design. Please refer to Figure 5 (3), which Schematic diagram of program coding operation of complementary flash memory cells. When performing the program coding operation, 'band-to-band tunneling induced hot electron injuntion' is used. To complete, the terminal bias voltages are VWL = 8 ~ 10V, VBL = -6 ~ -3V, VSL = 0 ~ 3V, VPW = 0V; by this, electrons can be injected into the floating doped region from the floating doped region. Set the brake to complete the program coding operation. Please refer to Figure 5 (4), which shows the timing diagram of the multi-state program coding operation of the complementary flash memory cell. This multi-state program coding can be completed in two stages. In the first stage, the electron injection is performed using "band-to-band tunneling induced hot electron injunUoii". The end point biases are added to the control gate plus one. Program voltage, such as bias VPC] M, in The drain is biased with VBL, the source is biased with Vsl_, and the P-well area is grounded, where the bias is as described above. In the second stage, the voltage of the bit line is discharged to the potential of the P-type well region by the conduction of the parasitic P-type channel transistor (parasitic transistor M2 'Figure 4), so that the electrons can no longer be injected, and then reach The first self-receiving paper size applies the Chinese National Standard (CNS) A4 specification (2ΐ〇χ 297 public power) --------- ^ ------ II ------ ^ { Chu Xian read the back _Attentions ^: fill in this page) 3280twf.doc / 006 V. Description of the invention (/ C) Convergence state; the end point bias voltage is a control voltage plus a discharge voltage, such as bias voltage vDISI 'The bias voltage VBL is applied to the drain and the bias voltage Vn is applied to the source, and the P-type well region is grounded. Repeat the above multi-state program coding operation with different control gate bias voltages vDIS2 to achieve the second self-convergent state; similarly, with different control gate bias voltages VDISn, the η self-convergence state can be achieved, and the multi-state program coding can be achieved. operating. In summary, compared with the conventional flash memory "billion cells", the complementary flash memory cell provided by the present invention has at least the following advantages: 1. When the complementary flash memory cell performs a read operation The on-current will be amplified by this parasitic bipolar transistor, so the read current obtained on the bit line is increased by about l + β times compared with the traditional flash memory cell, which can be read faster than simple peripheral circuits. Fetch operation to reduce the cost of peripheral circuit design. 2. Using these complementary flash memory cell structures, no thermal holes are generated during the "erase" and "programming" operations, which can greatly improve the reliability of the tunneling oxide layer and make the complementary flash memory cell The structure is not easy to degenerate. ”Printed by the Shell Standard Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economic Affairs. 3. Using these complementary flash memory cell structures, when performing“ programming ”operations, it is induced by the“ band-band-band tunneling effect ”. "Hot electron injection" to improve the working efficiency of memory cells. 4. Using different control gate bias voltages VDISn, it can achieve the η self-convergence state, and achieve multi-state programming operation. Although the above is a description of the structure, manufacturing method and operation method of the complementary flash memory cell constructed by the P-type well area, it is not intended to limit the paper size to the Chinese National Standard (CNS) A4 specification (210 × 297). Shovel) 3280twf.doc / 006 1Γ V. Description of the invention (//) The person who is familiar with this technology uses the N-type well area to construct a complementary flash memory cell structure, manufacturing method and operation method that are opposite to the present invention. Without departing from the spirit of the invention. • The above description is only a preferred embodiment of the present invention, but it is not intended to limit the present invention. Any person skilled in the art can make various modifications and retouches without departing from the spirit and scope of the present invention. The protection scope of the present invention shall be determined by the scope of the appended patent application. Any equivalent changes and modifications made according to the patent application scope of the present invention shall fall within the scope of the invention patent. Order 1 · ^ 袅. (Please read the notes on the back before filling out this page) The paper standard printed by the Consumers' Cooperatives of the Central Standards Bureau of the Ministry of Economic Affairs applies the Chinese national standard rate (〇 阳), \ 4 specifications (2 丨0 乂 297 to go)