TWI307965B - Multi-level non-volatile memory and manufacturing method thereof - Google Patents
Multi-level non-volatile memory and manufacturing method thereof Download PDFInfo
- Publication number
- TWI307965B TWI307965B TW95124452A TW95124452A TWI307965B TW I307965 B TWI307965 B TW I307965B TW 95124452 A TW95124452 A TW 95124452A TW 95124452 A TW95124452 A TW 95124452A TW I307965 B TWI307965 B TW I307965B
- Authority
- TW
- Taiwan
- Prior art keywords
- voltage
- gate
- layer
- substrate
- memory
- Prior art date
Links
- 230000015654 memory Effects 0.000 title claims description 279
- 238000004519 manufacturing process Methods 0.000 title claims description 18
- 239000000758 substrate Substances 0.000 claims description 76
- 238000000034 method Methods 0.000 claims description 63
- 238000003860 storage Methods 0.000 claims description 60
- 239000000463 material Substances 0.000 claims description 45
- 239000004020 conductor Substances 0.000 claims description 35
- 230000000694 effects Effects 0.000 claims description 17
- 125000006850 spacer group Chemical group 0.000 claims description 11
- 230000005641 tunneling Effects 0.000 claims description 9
- 230000003647 oxidation Effects 0.000 claims description 6
- 238000007254 oxidation reaction Methods 0.000 claims description 6
- 239000004575 stone Substances 0.000 claims description 6
- 239000013078 crystal Substances 0.000 claims description 4
- SIWVEOZUMHYXCS-UHFFFAOYSA-N oxo(oxoyttriooxy)yttrium Chemical compound O=[Y]O[Y]=O SIWVEOZUMHYXCS-UHFFFAOYSA-N 0.000 claims description 4
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 230000008878 coupling Effects 0.000 claims description 2
- 238000010168 coupling process Methods 0.000 claims description 2
- 238000005859 coupling reaction Methods 0.000 claims description 2
- 150000004767 nitrides Chemical class 0.000 claims description 2
- 208000001613 Gambling Diseases 0.000 claims 1
- 241000282320 Panthera leo Species 0.000 claims 1
- 238000012840 feeding operation Methods 0.000 claims 1
- 210000003127 knee Anatomy 0.000 claims 1
- 239000011159 matrix material Substances 0.000 claims 1
- 235000012054 meals Nutrition 0.000 claims 1
- 239000007800 oxidant agent Substances 0.000 claims 1
- 230000001590 oxidative effect Effects 0.000 claims 1
- 239000010410 layer Substances 0.000 description 225
- 238000002347 injection Methods 0.000 description 16
- 239000007924 injection Substances 0.000 description 16
- 238000010586 diagram Methods 0.000 description 10
- 238000005229 chemical vapour deposition Methods 0.000 description 9
- 230000005684 electric field Effects 0.000 description 7
- 239000011229 interlayer Substances 0.000 description 6
- 229910021420 polycrystalline silicon Inorganic materials 0.000 description 5
- 229920005591 polysilicon Polymers 0.000 description 5
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 4
- 239000003795 chemical substances by application Substances 0.000 description 4
- 238000005468 ion implantation Methods 0.000 description 4
- 239000001301 oxygen Substances 0.000 description 4
- 229910052760 oxygen Inorganic materials 0.000 description 4
- 229910000420 cerium oxide Inorganic materials 0.000 description 3
- 239000003989 dielectric material Substances 0.000 description 3
- 239000000428 dust Substances 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- BMMGVYCKOGBVEV-UHFFFAOYSA-N oxo(oxoceriooxy)cerium Chemical compound [Ce]=O.O=[Ce]=O BMMGVYCKOGBVEV-UHFFFAOYSA-N 0.000 description 3
- BPUBBGLMJRNUCC-UHFFFAOYSA-N oxygen(2-);tantalum(5+) Chemical compound [O-2].[O-2].[O-2].[O-2].[O-2].[Ta+5].[Ta+5] BPUBBGLMJRNUCC-UHFFFAOYSA-N 0.000 description 3
- MZLGASXMSKOWSE-UHFFFAOYSA-N tantalum nitride Chemical compound [Ta]#N MZLGASXMSKOWSE-UHFFFAOYSA-N 0.000 description 3
- 229910001936 tantalum oxide Inorganic materials 0.000 description 3
- RUDFQVOCFDJEEF-UHFFFAOYSA-N yttrium(III) oxide Inorganic materials [O-2].[O-2].[O-2].[Y+3].[Y+3] RUDFQVOCFDJEEF-UHFFFAOYSA-N 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- GQPLMRYTRLFLPF-UHFFFAOYSA-N Nitrous Oxide Chemical compound [O-][N+]#N GQPLMRYTRLFLPF-UHFFFAOYSA-N 0.000 description 2
- RTAQQCXQSZGOHL-UHFFFAOYSA-N Titanium Chemical compound [Ti] RTAQQCXQSZGOHL-UHFFFAOYSA-N 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 239000002019 doping agent Substances 0.000 description 2
- 239000011521 glass Substances 0.000 description 2
- 229910000449 hafnium oxide Inorganic materials 0.000 description 2
- WIHZLLGSGQNAGK-UHFFFAOYSA-N hafnium(4+);oxygen(2-) Chemical compound [O-2].[O-2].[Hf+4] WIHZLLGSGQNAGK-UHFFFAOYSA-N 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000001459 lithography Methods 0.000 description 2
- 238000000059 patterning Methods 0.000 description 2
- 229910052715 tantalum Inorganic materials 0.000 description 2
- GUVRBAGPIYLISA-UHFFFAOYSA-N tantalum atom Chemical compound [Ta] GUVRBAGPIYLISA-UHFFFAOYSA-N 0.000 description 2
- 241000288673 Chiroptera Species 0.000 description 1
- 235000014676 Phragmites communis Nutrition 0.000 description 1
- BQCADISMDOOEFD-UHFFFAOYSA-N Silver Chemical compound [Ag] BQCADISMDOOEFD-UHFFFAOYSA-N 0.000 description 1
- 229910052769 Ytterbium Inorganic materials 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 235000021384 green leafy vegetables Nutrition 0.000 description 1
- 235000012907 honey Nutrition 0.000 description 1
- 238000007654 immersion Methods 0.000 description 1
- 239000007943 implant Substances 0.000 description 1
- 238000002513 implantation Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000003475 lamination Methods 0.000 description 1
- 238000005121 nitriding Methods 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 239000001272 nitrous oxide Substances 0.000 description 1
- 229920002120 photoresistant polymer Polymers 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 229910001925 ruthenium oxide Inorganic materials 0.000 description 1
- WOCIAKWEIIZHES-UHFFFAOYSA-N ruthenium(iv) oxide Chemical compound O=[Ru]=O WOCIAKWEIIZHES-UHFFFAOYSA-N 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 229910052709 silver Inorganic materials 0.000 description 1
- 239000004332 silver Substances 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000009941 weaving Methods 0.000 description 1
- NAWDYIZEMPQZHO-UHFFFAOYSA-N ytterbium Chemical compound [Yb] NAWDYIZEMPQZHO-UHFFFAOYSA-N 0.000 description 1
Landscapes
- Non-Volatile Memory (AREA)
- Semiconductor Memories (AREA)
Description
I3079^〇2 19830twf.doc/e 九、發明說明: 【發明所屬之技術領域】 種半導體元件,且特別是有關於― 及其製造方法與操作方法。 本發明是有關於— 種多階非揮發性記憶體 【先前技術】 非揮發性記㈣元件由 入、讀取、技降犛叙从 丄 人選仃貝#之存 失之優點,所以已成為個=人之資料在斷電後也不會消 -種記憶體元件。1腦和電子設備所廣泛採用的 典型的非揮發性記憶體元件是以推 ”閘極(F1〇ating Gate)與控制 曰曰押= 極是直接設置在較_上,浮雜 )= 介電層相隔,而浮置_與基底之間以穿 0xlde)刪亦即所謂堆疊閘極快閃記憶體 ^己fe體在抹_ ’ _度鎌絲太過嚴重,而導致資 ==的問題。所以,會在控制閘極與浮置閘極側壁、、 =二另二選擇_(select Gate) ’而形成分離閘極 (Split Gate)結構。 般而σ ’目刖業界較常使用的非揮發性記憶體 例如是反或閘(Ν〇·_結構。在f知的反或閘(ν 型陣列中,每兩個記憶胞共用—條位元_接_以及共 用一條源極線。因此,每一個記憶胞會佔用到半個接觸^ 與半個源極線寬度之大小。由於每個記憶胞皆由位元線直 接連結,因此在反或閘(N〇R)型陣列中,記憶胞可以任意 I30796$7〇2 19830twf.d〇c/e 貝取與寫入的操作,且由於串聯之電阻較小,記恃 取操作時的電流較大,讀取的速度也就較快/心 階±NC)R _揮發性記憶體中的記憶體作為多 分布範圍 二於用於判別各個資料狀態的啟始電壓 的程式化在習知的歷型非揮發性記憶體 認步^要進行多絲疏步驟及程式化確 ’錢程式化記憶鮮確處於設定的啟始電壓分布 範圍中,如此將花費較長的時間。 【發明内容】 有鑑於此’本發明的目的就是在提供一 ^己憶體及其製造方法與操作綠,可以提高元件的3 本發明的再一目的是提供一種多階非揮發性記憶體及 ^衣仏方法與操作方法,其製程簡單,而可以增加製程裕 度。 本發明的又一目的是提供一種多階非揮發性記憶體及 ,、製造方法與操作方法,可以縮短記憶胞的程式化時間、 加快抹除速度。 本發明提供-鮮轉發性記憶體,包括設置於基 底上的第-記憶胞。此第—記憶胞包括控制閘極、電荷儲 存層、摻雜區、選擇閘極與輔助閘極。控制閘極設置於基 底上。電荷儲存層設置於控制閘極與基底之間。摻雜區設 置於控制閘極之第-側的基底中。選擇閘極設置於控制閘 極之第-侧_壁’且位於控制閘極與摻祕之間的基底 1307965702 19830twf.doc/e =。辅助雜設置雜制_之帛二觸 閘極施加電壓時’輔助閘:於辅助 =明的較佳實施=之底;二 H電層設置於電荷儲存層與基底之間。Ϊ -”電層設置於電荷儲存層鱼 ^ _ 弟一 置於辅助閘極與控制閘極;:。弟二介電層設 極與基底之間。第四介⑽於 3 /及輔助閘 電荷儲存層之間,上=:與控制閘極、 體二本-發二的二實:Γ述之多階非揮發性記憶 間隙壁。電層與控制閘極、電荷儲存層之間形成有 ^照本發明的較佳實施例所述之多階非揮發性 ’ t逑第四介電層與控制閘極之間形成有_壁。… 體,上實施觸述之多_揮發性記憶 边第"電層的材質包括氧化矽。 體 依照本發明的触實施觸述之 =第二介電層的材質包括氧化石夕/氮化:二己。隐 體 =、7的k佳實施例所述之多階_發性記憶 上述第二;丨電層與第四介電層為高溫熱氧化層。 體,:=發:的較佳實施例所述之多階非揮二生記憶 埃Sr丨電層以及第四介電層的厚度介於購埃至 依照本發_健實闕所述之乡階麵發性記憶 7 1307965 702 19830twf.doc/e 體,更包括第二記憶胞。第二記憶胞與第一記憶胞具有相 同的結構,且第二記憶胞與第一記憶胞以鏡像對稱的方 鄰接設置。 % 並且與 依照本發明的較佳實施例所述之多階非揮發性記愴 體,上述第二記憶胞設置於第一記憶胞之第一側 第一記憶胞共用摻雜區。 依照本發明的較佳實施例所述之多階非揮發性記情 體,上述第二記憶胞設置於第一記憶胞之第二側,並且^ 第一記憶胞共用輔助電極。 /、 依照本發明的較佳實施例所述之多階非揮發性記 體,更包括頂蓋層。此頂蓋層設置於控制閘極上。μ 依照本發明的較佳實施例所述之多階非揮發性記 體,上述電荷儲存層部分係水平地突出於該控制間極,= 且鄰接選擇閘極的位置具有轉角。 、’ 本發明提供-種多階非揮發性記憶體,包括多 ' ί數條位元線、多數條控制閘極線、多數條選擇Η ^夕數條獅时極線與多數個f晶體。多數個記憶二 铋二:上5又置成行/列陣列。各記憶胞包括控制閘極、電‘ 其::巧區、選擇閘極、辅助閘極。控制閘極設置於 % 荷儲存層&置於控侧極與基底之間。摻雜巴 制間極之第-側的基底中。選擇間極設置於: ::之疒側的側壁,且位於控制閑極與摻雜區之間的I =二,:閘極設置於控制閘極之第二側的側壁,其中在 仃上的讀胞以鏡像對稱的方式鄰接設置。多數條位 8 Ό2 19830twf.doc/e 元線在行方向上平行排列,連接同一行之記憶胞的換雜 區。多數條控制閘極線在列方向上平行排列,連接同二列 之記憶胞的控制閘極。多數條選擇閘極線在列方向上平行 排列,連接同-列之記憶胞的選擇閘極。多數條輔助閉= 線在列方向上平行制,連接同—狀記憶胞的輔 極。電晶體岐極分職接輔助閘極線下方的基底。 依照本發_較佳實_所述之乡 體,更包括第一介電層、第二介電層、第三介電層^ =電層。第—介電層設置於電荷儲存層與基底之間。第: "電層設置於電荷儲存層與控侧極 二 =::rrr、電荷細之間;= 電价:ίΓ 層設置於選擇間極與控制間極、 電何儲存層之間,以及麵_與基底之間。 憶 間隙壁。電I、控制極、電何儲存層之間形成有 體’二:四介: 憶 記憶 蜜,其中該第三介電層與:: '02 19830twf.doc/e 依照本發明的較佳實施例所述之多階非揮發性記憶 體,上述電荷儲存層部分係水平地突出於控制聞極,並且 鄰接選擇閘極的位置具有轉角。 依照本發明的較佳實施例所述之多階非揮發性記憶 體’上述彼此鏡像對稱之相鄰二記憶胞係共用汲極區。 依照本發明的較佳實施例所述之多階非揮發性記憶 體’上述彼此鏡像對稱之相鄰二記憶胞係共用輔助閘極。 本發明之多階非揮發性記憶體中,相鄰的兩記憶胞例 如是以鏡像對稱的方式配置,亦即相鄰的兩個記憶胞共用 輔助閘極或摻雜區。因此本發明之多階非揮發性記憶體的 、、、σ構不但可以間化製造流程、降低製造成本,同時也能夠 提高元件的積集度。 而且,當辅助閘極上施加有電壓而打開辅助閘極下方 的通道、並形成反轉層時,於反轉層施加電壓,使得輔助 閘極下方的反轉層可以處於預充電狀態。在程式化本發明 之非揮發性記憶體時,藉由自我加速電荷注入(ch^nei ^lfboosting)之方式-,利用源極侧注入效應來進行記憶胞 可以提升程式化速度。而且’當此非揮發 體作為多階記憶胞時’在程式化操作時可以準確的控制記 憶胞經程式化後的啟始電壓位於所設定的範圍内。° 本發明提供一種多階非揮發性記憶體的製造方法,勹 步驟。首先,於基底上依序形成穿时電層,、電^ 储存層,並於電荷儲存層上形成至少二堆疊層,1 堆叠層依序包括關介電層、控卿極與職層。接著, 130796^702 19830twf.doc/e 130796^702 19830twf.doc/e ft堆疊層之㈣電荷儲存層,以形成第-溝泪。π 堆®層側壁與第—、、番泪办,时 取弟溝木。於二 禮外側之電荷儲存層。銬德,你且* L… 成介電層。於二堆疊岸之門取;;姑二交,於基底上形 外側之侧壁分別形成^ ~ 疊層 疊層與摻 ::;:别形成接雜區,並且選擇閘極係二广之 依照本發明的較佳實 的製造方法,上齡電料減溫揮紐輯雜 造方:照所述之非揮發性記憶體的製 ^ ί電層的厚度介於100埃至細埃之間。 造方法 實施例所述之非揮發性記憶體的製 ,外側之侧壁分卿成選擇雜之綠如下。首先於 >形成導體材料層’此導體材料層填滿第一溝渠、、土- 仃回,製?:移除部分導體材料層,直到暴露出頂蓋層 ^ ; 一堆豐層之間形成輔助閘極,並於二堆疊 渠’然後進 的發=,實_所述之多階非揮發性記憶曰體 ^方法’上述穿隨介電層的材質包括氧化石夕。 圭實施例所述之多階非揮發性記憶體 心方法’上迷轉介電層的形成方法包括熱氧化法。 氧化石夕 依照本發明的較佳實施例所述之多階非揮發性記 的製::ΐ發明广較佳實施例所述之多階非揮發性記憶體 ,製化方法,上述_介電層的材質包括氧切/氮化石夕/ 憶體 11 130796^702 19830twf.doc/e 在本發明之非揮多:石夕。 而日,士成本、增加製程裕度。 由於所形成的電荷儲存層在靠 分具有-個轉角,因此在對記 f選擇間極之部 角使電場集中’而可提高電子::出:二此轉 極的抹除操作速度。 $仃储存層拉出至選擇閘 用;供I種多階非揮發性記憶體的操作方法,適 閘極之弟-側的基底中。選擇閘極設置於控制閘極之^ 貝1的側壁’且位於控制雜與摻雜區之間的基底上 ,極設置於控制閘極之第二侧的側壁,此方法包括在進行 表式化操作時’預先於辅助閘極施加第一電壓,以於輔助 閘極下方的基底中形成反轉層,並於反轉層施加第二電 慶。,後’於控侧極施加第三電壓,於選擇閘極施加第 四電壓,於汲極區施加第五電壓,其中第一電壓大於輔助 閘,的啟始電壓,第三電壓大於第一電壓,第四電壓大於 或等於選擇閘極的啟始電壓,第三電壓大於第五電壓,以 利用源極側注入效應程式化記憶胞。 依照本發明的較佳實施例所述之多階非揮發性記憶體 的操作方法,第一電壓為8伏特左右;第二電壓為5伏特 12 13079657 02 19830twf.doc/e ΐΐ電伏特左右如電壓為μ伏特左右; 弟五電壓為0伏特左右。 的操發,=佳實__之多階非揮發性記憶體 第二電厂:以於進仃抹除操作時’於選擇閘極施加I3079^〇2 19830twf.doc/e IX. Description of the Invention: [Technical Field of the Invention] A semiconductor element, and particularly relates to "and its manufacturing method and operation method. The present invention relates to a multi-level non-volatile memory. [Prior Art] The non-volatile recording (four) component has the advantages of being in, reading, and technically reducing the loss of the 仃人仃仃贝#, so it has become a = The data of the person will not disappear after the power is turned off - the memory component. 1 Typical non-volatile memory components widely used in brain and electronic equipment are push gates (F1〇ating Gate) and control clamps = poles are directly set on the _, 浮 ) = = = = = = The layers are separated, and the floating _ is separated from the substrate by 0xlde. That is to say, the so-called stacked gate flash memory ^ fe 体 body in the wipe _ ' _ degree 镰 silk is too serious, resulting in the problem of ==. Therefore, the gate gate and the floating gate sidewall are controlled, and the s(select gate) is selected to form a split gate structure. Generally, σ 'sees the non-volatile which is commonly used in the industry. The memory is, for example, an inverse or gate (Ν〇·_ structure. In the inverse or gate of the known (in the ν-type array, every two memory cells share - a bit_connect_) and share a source line. Therefore, Each memory cell occupies half the width of the contact ^ and half of the source line. Since each memory cell is directly connected by a bit line, in the inverse or gate (N〇R) type array, the memory cell Any I30796$7〇2 19830twf.d〇c/e can be taken and written, and since the resistance of the series is small, it is recorded during the operation. The flow is large, the reading speed is faster / the heart level ± NC) R _ the memory in the volatile memory as the multi-distribution range 2 is used to determine the starting voltage of each data state. The calendar type non-volatile memory recognition step ^ requires a multi-filing step and stylization to ensure that the 'money stylized memory is freshly in the set starting voltage distribution range, which will take a long time. [Summary of the Invention] In view of the above, the object of the present invention is to provide a memory and a method for manufacturing the same, and to improve the device. A further object of the present invention is to provide a multi-stage non-volatile memory and method. And the operation method, the process is simple, and the process margin can be increased. Another object of the present invention is to provide a multi-stage non-volatile memory and, manufacturing method and operation method, which can shorten the stylized time of the memory cell and speed up The invention provides a fresh forward memory, comprising a first memory cell disposed on the substrate. The first memory cell includes a control gate, a charge storage layer, a doped region, a select gate and a gate is provided on the substrate, and a charge storage layer is disposed between the control gate and the substrate. The doped region is disposed in the substrate on the first side of the control gate. The selection gate is disposed on the control gate The first side _ wall 'and the base between the control gate and the doping is 1307965702 19830twf.doc / e =. Auxiliary miscellaneous _ 帛 帛 帛 触 触 施加 施加 ' ' ' 辅助 辅助 辅助 辅助 辅助Preferably, the bottom layer is disposed between the charge storage layer and the substrate. Ϊ - "Electrical layer is disposed in the charge storage layer, and the first layer is placed on the auxiliary gate and the control gate; The dielectric layer is disposed between the pole and the substrate. The fourth layer (10) is between the 3/ and the auxiliary gate charge storage layer, and the upper gate is: the control gate, the body two, and the second two: the multi-order non-description Volatile memory spacers. Between the electrical layer and the control gate, the charge storage layer is formed with a _ wall formed between the multi-level non-volatile 逑t逑 fourth dielectric layer and the control gate according to the preferred embodiment of the present invention. ... body, the implementation of the description of the _ volatile memory side of the "electric layer material includes yttrium oxide. The material according to the touch implementation of the present invention = the material of the second dielectric layer comprises oxidized oxide/nitriding: dihex. The second order of the hidden body =, the k-good embodiment of the seventh embodiment is the second; the tantalum layer and the fourth dielectric layer are high temperature thermal oxide layers. The thickness of the multi-stage non-swollen memory Sr丨 electric layer and the fourth dielectric layer described in the preferred embodiment of the present invention are between the purchase and the hometown according to the present invention. Stepped facial memory 7 1307965 702 19830twf.doc / e body, including the second memory cell. The second memory cell has the same structure as the first memory cell, and the second memory cell is disposed adjacent to the first memory cell in a mirror symmetrical square. And in combination with the multi-level non-volatile memory according to the preferred embodiment of the present invention, the second memory cell is disposed on the first side of the first memory cell, and the first memory cell shares the doped region. According to a multi-stage non-volatile grammar of the preferred embodiment of the present invention, the second memory cell is disposed on a second side of the first memory cell, and the first memory cell shares an auxiliary electrode. /, a multi-order non-volatile record according to a preferred embodiment of the invention, further comprising a cap layer. This cap layer is placed on the control gate. μ In accordance with a multi-stage non-volatile record according to a preferred embodiment of the present invention, the charge storage layer portion protrudes horizontally from the control electrode, and the position adjacent to the selection gate has a rotation angle. The present invention provides a multi-level non-volatile memory including a plurality of bit lines, a plurality of control gate lines, a plurality of strips, and a plurality of f-crystals. Most of the memories are two: The upper 5 is again placed in a row/column array. Each memory cell includes a control gate, an electric ‘its:: a smart area, a selection gate, and an auxiliary gate. The control gate is placed between the % load reservoir & placed between the control side and the substrate. Doped in the substrate on the first side of the interpole. The selected interpole is disposed on the side wall of: :: on the side of the side, and is located between the control idler and the doped region, I = two, the gate is disposed on the sidewall of the second side of the control gate, wherein the gate is The read cells are contiguously arranged in a mirror symmetrical manner. Most of the strips 8 Ό 2 19830twf.doc/e are lined up in parallel in the row direction, connecting the swap cells of the same row of memory cells. A plurality of control gate lines are arranged in parallel in the column direction to connect the control gates of the memory cells of the same two columns. A plurality of strip select gate lines are arranged in parallel in the column direction to connect the select gates of the memory cells of the same column. Most of the auxiliary closed = lines are parallel in the column direction, connecting the auxiliary poles of the same-like memory cells. The transistor is divided into a sub-base under the auxiliary gate line. According to the invention, the first dielectric layer, the second dielectric layer and the third dielectric layer are electrically layered. The first dielectric layer is disposed between the charge storage layer and the substrate. The first: " electrical layer is set between the charge storage layer and the control side pole two =:: rrr, charge fine; = electricity price: ί Γ layer is set between the selection of the interpole and the control between the pole, the electrical storage layer, and Between the surface _ and the substrate. Recall the gap wall. The body I, the control electrode, and the electrical storage layer are formed with a body 'two: four: memory honey, wherein the third dielectric layer and:: '02 19830twf.doc / e in accordance with a preferred embodiment of the present invention In the multi-level non-volatile memory, the charge storage layer portion protrudes horizontally from the control horn, and the position adjacent to the selection gate has a rotation angle. The multi-level non-volatile memory according to the preferred embodiment of the present invention' The multi-level non-volatile memory according to the preferred embodiment of the present invention, the adjacent two memory cells mirror-symmetrical to each other share an auxiliary gate. In the multi-level non-volatile memory of the present invention, adjacent two memory cells are arranged, for example, in a mirror symmetrical manner, that is, adjacent two memory cells share an auxiliary gate or doped region. Therefore, the multi-stage non-volatile memory of the present invention can not only simplify the manufacturing process, reduce the manufacturing cost, but also improve the integration of components. Further, when a voltage is applied to the auxiliary gate to open the channel under the auxiliary gate and an inversion layer is formed, a voltage is applied to the inversion layer so that the inversion layer under the auxiliary gate can be in a precharged state. In the non-volatile memory of the present invention, by means of self-accelerating charge injection (ch^nei ^lfboosting), the use of the source side injection effect to perform memory cells can increase the programming speed. Moreover, when the non-volatile body is used as a multi-order memory cell, the starting voltage of the memory cell can be accurately controlled during the stylization operation within the set range. The present invention provides a method of manufacturing a multi-stage non-volatile memory, 勹 step. First, a time-penetrating layer is formed on the substrate, and the storage layer is formed, and at least two stacked layers are formed on the charge storage layer. The stacked layers sequentially include a dielectric layer, a control layer, and a layer. Next, 130796^702 19830twf.doc/e 130796^702 19830twf.doc/e ft (4) the charge storage layer of the stacked layer to form the first groove. The side wall of the π-stack® layer is the first and the third, and the tears are made. The charge storage layer on the outside of the second ceremony. Jude, you and * L... into a dielectric layer. In the second stacking of the bank gate;; the second two sides, the sidewalls on the outer side of the base are formed separately ~ ~ lamination stack and doping::;: do not form a miscellaneous area, and select the gate system of the second wide In a preferred embodiment of the invention, the age-old electrical material desuperheating wave is made up of a thickness of between 100 angstroms and fine angstroms. Method of Making Non-volatile memory according to the embodiment, the outer side wall is divided into the following greens. First, the conductor layer is formed in the > layer. The conductor material layer fills the first trench, the soil-return, and is made. : removing part of the conductor material layer until the top cover layer is exposed; a stack of auxiliary gates is formed between the stacks of layers, and the second stacking channel is then 'into the stack', then the multi-level non-volatile memory曰体^方法' The material of the above-mentioned dielectric layer includes oxidized stone. The method of forming the fused dielectric layer on the multi-stage non-volatile memory core method described in the embodiment includes a thermal oxidation method. The invention relates to a multi-stage non-volatile memory according to a preferred embodiment of the present invention: a multi-stage non-volatile memory according to a preferred embodiment of the invention, a method for preparing the same, and the above-mentioned The material of the layer includes oxygen cut/nitridite eve/memory 11 130796^702 19830 twf.doc/e in the present invention is not a wave: Shi Xi. And the cost of the day, increase the process margin. Since the formed charge storage layer has a corner at the corner, the electric field is concentrated by selecting the angle between the poles of the pair f to increase the electron::out: the speed of the erase operation of the second electrode. The $仃 storage layer is pulled out to the selection gate; the method for operating a multi-level non-volatile memory is suitable for the gate-side substrate of the gate. The selection gate is disposed on the sidewall of the control gate 1 and is located on the substrate between the control impurity and the doping region, and the pole is disposed on the sidewall of the second side of the control gate, and the method includes performing the patterning In operation, a first voltage is applied to the auxiliary gate in advance to form an inversion layer in the substrate under the auxiliary gate, and a second electrification is applied to the inversion layer. Applying a third voltage to the control side, applying a fourth voltage to the select gate, and applying a fifth voltage to the drain region, wherein the first voltage is greater than the start voltage of the auxiliary gate, and the third voltage is greater than the first voltage The fourth voltage is greater than or equal to the starting voltage of the selected gate, and the third voltage is greater than the fifth voltage to program the memory cell with the source side injection effect. According to a preferred embodiment of the preferred embodiment of the present invention, the first voltage is about 8 volts; the second voltage is 5 volts 12 13079657 02 19830 twf.doc/e ΐΐ volts or so as voltage It is about volts; the voltage of the fifth is about 0 volts. Operation, = good __ multi-level non-volatile memory, second power plant: for the purpose of the 仃 erase operation
被拉出並移除,其中第丄雷壓盘装泛^子左由選擇閘極 穿隨效應。八中第,、電壓與基底的電壓差會引發FN 的二=發ΪΪ較佳實施例所述之多階非揮發性記憶體 、知作方法,第六電壓為u至15伏特左右。 的操發佳實施例所述之多階非揮發性記憶體 第七電愿於於進订%取操作時,於輔助閘極施加 九雷麗^ 閘極施加第人電壓,於選擇閘極施加第 壓錄區施加第十電壓,以讀取記憶胞,第九電 1大於選擇閘極的啟始電壓。 作方ί照ί發明的較佳實施例所述之非揮發性記憶體的操 _ Vee(f源電壓)’第八電壓為i 4伏特 左右’第九電塵為Vcc(電源電壓),第十龍為i 5伏 右0 本發明提供-種多階非揮發性記憶體的操作方法,適 Γ,Τ多數個記憶祕狀記憶胞陣列。各記憶胞包括控 '甲1極、電荷儲存層、摻雜區、選擇問極、辅助間極。控 制間極設置於基底上。電荷儲存層設置於控制閘極與基底 之,。摻雜區設置於控制間極之第一側的基底中。間 極叹置於控制閘極之第一側的側壁,且位於控制開極與摻 13 '702 19830twf.doc/e 雜區之間的基底上。輔师極設置於控糊極之第二侧的 侧壁,其巾在同—行上的記憶胞以鏡像對翻方式鄰接設 置。多數條位元線在行方向上平行排列,連接同一行之記 憶胞的摻㈣。多數齡糊極線在财向±平行排列, 連接同i之記憶胞的控制閘極4數條卿閘極線在列 方向上平行排列,連接同—列之記憶胞的選擇閘極。多數 條辅助酿線在财向上平行排列,連接同—狀記憶胞 的辅助閘;1¾。電晶體的汲極連接輔㈤閘極線下方的基底。 t法包括在進行程式化操作時,預先於電晶體的閑極施 u電壓,於電晶體的源極施加第二電壓,於辅助閘極 屏力第―電壓’以於輔助閑極線下方的基底巾形成反轉 曰’使反轉層導通有第二電壓,之後改變該些電晶體的閘 =電^為〇伏特以關閉該些電晶體的通道,使得該反轉層 处於一預充電狀態,然後將輔助閘極電壓拉至s伏特左 ’以轉合(coupling)反轉層的電壓至5伏特左右。於選定 $制間極線施加第四電壓,於選^之選擇閘極線施加第 於…壓,於選定之位元線施加第六電壓,其中第三電壓大 芦輔助閘極的啟始電壓,第四電壓大於第三電壓,第五電 =大於或等於選擇閘極的啟始電壓,第二電壓大於第六電 以利用源極侧注入效應程式化選定記憶胞。 依照本發明的較佳實施例所述之多階非揮發性記憶體 ▽喿作方法,第一電壓為Vcc(電源電壓);第二電壓為 ^cc(電源電壓)-Vth(電晶體之起始電壓);第三電壓為 (電源電壓),第四電壓為10伏特左右;第五電壓為 14 1307嫩 〇2 19830twf.doc/e 伏特左右;第六電壓為0伏特左右。 的極發明的較佳實施例所述之多階非揮發性記憶體 包括於進行抹除操作時,於選擇閘極施加 =拉出並移除,其中W與基底的電壓 穿隧效應。 左以丨钐州 =發明的較佳實施例所述之多階非揮發性記憶體 的刼作方法,第七電壓為11至15伏特左右。 的較佳實施例所述之多階非揮發性記憶體 極i包括於進行讀取操作時’於選定之辅助閘 極線施加“電壓,於選定之控侧極線施加第九電壓, 33?_線施加第十電壓,於選定之位元線施加 始電壓 續取記憶胞,第十電壓大於選擇閘極的啟 从士?照,發㈣較佳實施綱述之非揮發性記憶體的操 乍方法,第人電壓為Vee(電源電壓)’第九電壓為t 4 左右,第十電壓為Vee(電源電壓),第十—電壓為15 左右。 可 在本發明之多階非揮發性記憶體的操作方法中 採用自我加速電荷注入(咖圖1 Self-bcK)Sting)之方式,利用 源極側注人效應來進行織_植,聽使輔助開極的 通道充電至設定電壓,而可以快速的程式化記憶胞。而且, 利用選擇閘極來抹除記憶胞,使得電子經由抹除閘極而移 除,可以減少電子穿越請介騎之錄,而提高元件 15 13079餘02 1983〇twf.doc/e 靠度。此外,電荷儲存層鄰接選擇閘極之處具 時’藉由此轉角使電場集中,而可提高電子從電 何储存層拉^至選擇閘極的抹除操作速度。 為讓本發明之上述和其他目的、特徵和優點能更明顯 易馑’下文特舉實施例’並配合所_式,作詳細說明如 下0 【實施方式】It is pulled out and removed, and the first thunder pressure plate is mounted on the left side by the selection gate. In the eighth, the voltage difference between the voltage and the substrate causes the FN to be the second-order non-volatile memory of the preferred embodiment, and the sixth voltage is about u to 15 volts. The multi-stage non-volatile memory according to the preferred embodiment of the present invention is intended to apply a nine-threshold voltage to the auxiliary gate during the advance % pick operation, and apply the first voltage to the gate. The tenth voltage is applied to the embossed area to read the memory cell, and the ninth power 1 is greater than the starting voltage of the selected gate. The eighth voltage of the non-volatile memory of the non-volatile memory described in the preferred embodiment of the invention is about 4 volts, and the ninth electric dust is Vcc (power supply voltage), The ten dragon is i 5 volts right 0 The present invention provides a method for operating multi-level non-volatile memory, which is suitable for most memory memory cell arrays. Each memory cell includes a '1 pole, a charge storage layer, a doped region, a selective pole, and an auxiliary pole. The control electrode is placed on the substrate. The charge storage layer is disposed on the control gate and the substrate. The doped region is disposed in the substrate on the first side of the control interpole. The sigh is placed on the side wall of the first side of the control gate and is located on the substrate between the control opening and the doped 13 '702 19830 twf.doc/e. The auxiliary pole is disposed on the side wall of the second side of the control paste, and the memory cells of the towel on the same line are adjacently arranged in a mirror-to-turn manner. Most of the bit lines are arranged in parallel in the row direction, connecting the cells of the same row (4). Most age paste lines are arranged in parallel in the fiscal direction. The control gates connected to the memory cells of i are arranged in parallel in the column direction, and the selection gates of the memory cells of the same column are connected. Most of the auxiliary brewing lines are arranged in parallel in the financial direction, connecting the auxiliary gates of the same-like memory cells; 13⁄4. The drain of the transistor is connected to the substrate under the auxiliary (5) gate line. The t method includes, when performing a stylization operation, applying a second voltage to the source of the transistor in advance to the voltage of the idle electrode of the transistor, and applying a second voltage to the auxiliary gate of the gate to the lower side of the auxiliary idle line. The base towel forms an inversion 曰' such that the inversion layer is turned on by a second voltage, and then the gates of the transistors are changed to be volts to close the channels of the transistors, so that the inversion layer is in a precharge State, then pull the auxiliary gate voltage to s volts left 'to coupling the voltage of the inversion layer to about 5 volts. Applying a fourth voltage to the selected inter-electrode line, applying a voltage to the selected gate line, applying a sixth voltage to the selected bit line, wherein the third voltage is a starting voltage of the auxiliary gate of the reed, The fourth voltage is greater than the third voltage, the fifth voltage is greater than or equal to the starting voltage of the selected gate, and the second voltage is greater than the sixth voltage to program the selected memory cell using the source side injection effect. According to a multi-stage non-volatile memory processing method according to a preferred embodiment of the present invention, the first voltage is Vcc (supply voltage); the second voltage is ^cc (supply voltage)-Vth (from the crystal) The starting voltage); the third voltage is (supply voltage), the fourth voltage is about 10 volts; the fifth voltage is 14 1307 〇 2 19830 twf.doc / e volts; the sixth voltage is about 0 volts. The multi-stage non-volatile memory of the preferred embodiment of the invention is included in the selection gate application = pull-out and removal, wherein the voltage tunneling effect of W and the substrate. The method of the multi-stage non-volatile memory described in the preferred embodiment of the invention is the left voltage of about 11 to 15 volts. The multi-stage non-volatile memory electrode i described in the preferred embodiment includes 'applying a voltage to the selected auxiliary gate line when performing a read operation, and applying a ninth voltage to the selected control side line, 33? Applying a tenth voltage to the _ line, applying a starting voltage to the selected bit line to renew the memory cell, the tenth voltage is greater than the selection of the gate, and the fourth embodiment of the preferred embodiment of the non-volatile memory operation In the 乍 method, the first person voltage is Vee (supply voltage), the ninth voltage is about t 4 , the tenth voltage is Vee (power supply voltage), and the tenth voltage is about 15. The multi-level non-volatile memory that can be used in the present invention The method of self-acceleration charge injection (self-bcK) Sting) uses the source-side injection effect to perform the weaving, and the auxiliary open-circuit channel is charged to the set voltage. Fast stylized memory cells. Moreover, using the selection gate to erase the memory cells, so that the electrons are removed by erasing the gates, can reduce the number of electrons crossing the phone, and improve the components 15 13079 02 1983 twf .doc/e reliability. In addition, charge storage Adjacent to the selection of the gate, the time is taken to "concentrate the electric field by the corners, and the speed of the erase operation of the electrons from the storage layer to the selection gate can be increased. To achieve the above and other objects, features and features of the present invention. The advantages can be more obvious and easy to use 'the following specific embodiment' and with the formula, the detailed description is as follows 0 [Embodiment]
圖1疋依照本發明的實施例所奢示之非揮發性記憶 的結構剖面圖。 x 〜 β請參照圖卜本發明提出之多階非揮發性記憶體例如 是由設置於基底100上的多個記憶胞Q卜q2、Q3、所 構成。各記憶胞Q卜Q2、Q3、Q4包括穿隧介電層1〇2、 電荷儲存層104、閘間介電層1〇6、控制閘極1〇8、頂蓋層 Π0、間隙壁112a、112b、摻雜區114、選擇閘極116、辅 助閘極118以及介電層120、122。閘間介電層ι〇6、控制 閘極108、頂蓋層11〇構成堆疊層1U。BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a cross-sectional view showing the structure of a non-volatile memory in accordance with an embodiment of the present invention. x to β Please refer to the multi-dimensional non-volatile memory proposed by the present invention, for example, by a plurality of memory cells Q2 and Q3 provided on the substrate 100. Each of the memory cells Qb, Q3, Q4 includes a tunneling dielectric layer 1, 2, a charge storage layer 104, a gate dielectric layer 1〇6, a control gate 1〇8, a cap layer Π0, a spacer 112a, 112b, doped region 114, select gate 116, auxiliary gate 118, and dielectric layers 120, 122. The inter-gate dielectric layer 〇6, the control gate 108, and the cap layer 11 〇 form a stacked layer 1U.
控制閘極108例如是設置於基底1〇〇上。控制閘極ι〇8 的材質例如是推雜多晶石夕。 電荷儲存層104例如是設置於控制閘極1〇8與基底 100之間。電荷儲存層104之材質例如是導體材料(如推雜 多晶矽)或能夠使電荷陷入於其中之材質,例如氮化矽、I 氧化石夕、氧化钽、鈦酸鎖或氧化給等。 閘間介電層106例如是設置於控制閘極108與電荷儲 存層104之間,閘間介電層106的材質例如是氧化矽/氮化 16 19830twf.doc/e 矽/氧化矽。 穿随介電層102例如是設置於電荷儲存層刚下方之 基底100上,穿隧介電層102的材質例如是氧化石夕。 摻雜區U4設置於控制閘極1〇8之一側的基底1〇〇 中。摻雜區114例如是作為汲極區。 選擇閘極116例如設置於控制間才亟1〇8 一侧的側壁, 且位於控咖極刚與該摻雜區114之間的基底⑽上。 選擇閘極116的材質例如是摻雜多晶 一 2層12〇例如設置於選擇_ 116與控制間極 〇、電^儲存層刚之間,以及選擇閘極116與基底卿 層12G之材質例如是以高溫熱氧化法形成之氧 選擇fU 1電層120的厚度介於120埃至130埃之間。在 擇問與基底觸之間的部分介電層料作為選 ^ 1ΠΛ . 电壓時,辅助閘極118下方的基 雜多晶發形成反轉層124。辅助閘極118的材質例如是摻 1〇8、電没置於輔助間才亟118與控制問極 之間。介電//22 7間’以及辅助閑極118與基底100 化石夕層。在輔例如是以高溫熱氧化法形成之氧 是作為辅助閘極"介電/:^底^之間的部分介電層120 130埃之間。 電層。"電層122的厚度介於120埃至 130796^702 19830twf.doc/e 在辅助閘極m與控制閘極l〇8、電荷儲存層⑺ 。在選擇閘極116與控制閘極 的材質例如是氮恤、⑽ 由於電何儲存層1〇4部分係水平地 二厚二,108並且位於間隙壁112a下方,使得電荷 ==二,閘極116之部分具有-個轉角⑼, =此在對讀、胞進行抹除時,藉由此轉角126使The control gate 108 is, for example, disposed on the substrate 1A. The material of the control gate 〇8 is, for example, a doped polycrystalline stone. The charge storage layer 104 is disposed, for example, between the control gate 1A8 and the substrate 100. The material of the charge storage layer 104 is, for example, a conductor material (e.g., a doped polysilicon) or a material capable of trapping charges therein, such as tantalum nitride, I oxidized cerium oxide, cerium oxide, titanate or oxidized. The inter-gate dielectric layer 106 is disposed, for example, between the control gate 108 and the charge storage layer 104. The material of the inter-gate dielectric layer 106 is, for example, hafnium oxide/nitride 16 19830 twf.doc/e 矽/yttria. The pass-through dielectric layer 102 is, for example, disposed on the substrate 100 just below the charge storage layer, and the material of the tunnel dielectric layer 102 is, for example, oxidized oxide. The doping region U4 is disposed in the substrate 1〇〇 on one side of the control gate 1〇8. The doping region 114 is, for example, a drain region. The selection gate 116 is disposed, for example, on a side wall of the control room 1 side, and is located on the substrate (10) between the control electrode and the doping region 114. The material of the gate 116 is selected, for example, a doped poly-two layer 12 〇, for example, disposed between the selection _116 and the control inter-electrode, the storage layer, and the material of the gate 116 and the underlying layer 12G. The thickness of the fU 1 electrical layer 120 formed by the high temperature thermal oxidation method is between 120 angstroms and 130 angstroms. The partial polysilicon underneath the auxiliary gate 118 forms an inversion layer 124 when a portion of the dielectric layer between the selective and substrate contacts is selected as the voltage. The material of the auxiliary gate 118 is, for example, 掺8, and the electricity is not placed between the auxiliary chamber 118 and the control electrode. Dielectric / / 22 7 ' and auxiliary idle 118 and substrate 100 fossil layer. In addition, for example, oxygen formed by high-temperature thermal oxidation is used as an auxiliary gate between the dielectric layer and the dielectric layer 120 between 130 angstroms. Electrical layer. " Electrical layer 122 thickness between 120 angstroms to 130796^702 19830 twf.doc / e in the auxiliary gate m and control gate l 〇 8, the charge storage layer (7). The material for selecting the gate 116 and the control gate is, for example, a nitrogen shirt, (10) because the portion of the storage layer 1 〇 4 is horizontally two thick, 108 and is located below the spacer 112a, so that the charge == two, the gate 116 The part has a corner (9), which is caused by the corner 126 when the read and the cell are erased.
=:=子從電荷儲存層104拉出至選擇間極丨;6 的抹除插作速度。 —此外’摻雜區114例如是藉由插塞130電性連接至位 兀線128。插塞13G與位元線128讀質例如是導體材料。=: = the child is pulled from the charge storage layer 104 to the selected interpole 丨; Further, the doped region 114 is electrically connected to the bit line 128, for example, by the plug 130. The plug 13G and the bit line 128 are read, for example, of a conductor material.
本毛月之夕階非揮發性記憶體中,才_的兩記憶胞例 如疋乂鏡像對稱的方式配置,亦即相鄰的兩個記憶胞共用 輔助閘極或摻雜區。舉例來說,記憶胞Q1與記憶胞Q2 共用輔助閘極;記憶胞Q2與記憶胞Q3共轉雜區;記憶 胞Q3與記憶胞Q4共用辅助閘極。因此本發明之多階非揮 發性記憶體的結構;^但可以簡化製造流程、降低製造成 本,同時也能夠提高元件的積集度。 而且,當輔助閘極118上施加有電壓而打開辅助閘極 118下方的通道、並形成反轉層124時,於反轉層I%施 加電壓,使得輔助閘極118下方的反轉層124可以處於預 充電狀態。在程式化本發明之非揮發性記憶體時,藉由自 我加速電荷注入(channel self_b〇osting)之方式,利用源極侧 注入效應來進行記憶胞程式化,可以提升程式化速度。而 18 13 07%& 19830twf.doc/e 且’當此非揮發性記憶體作為多階記憶胞時,在程式化操 作時可以準確的控制記憶胞經程式化後的啟始電壓位於所 設定的範圍内。 舉例來說,此多階記憶胞是藉由基準讀取電壓Vrefl、In the non-volatile memory of the eclipse of the month, the two memory cells of the _ _ are configured in a mirror-symmetrical manner, that is, the adjacent two memory cells share the auxiliary gate or doped region. For example, the memory cell Q1 shares the auxiliary gate with the memory cell Q2; the memory cell Q2 and the memory cell Q3 co-transform region; the memory cell Q3 shares the auxiliary gate with the memory cell Q4. Therefore, the structure of the multi-stage non-volatile memory of the present invention can simplify the manufacturing process, reduce the manufacturing cost, and improve the integration of components. Moreover, when a voltage is applied to the auxiliary gate 118 to open the channel under the auxiliary gate 118 and the inversion layer 124 is formed, a voltage is applied to the inversion layer 1% so that the inversion layer 124 under the auxiliary gate 118 can In pre-charged state. In the stylization of the non-volatile memory of the present invention, the memory cell can be programmed by the source side injection effect by means of channel self_b〇osting, which can improve the programming speed. And 18 13 07% & 19830twf.doc/e and 'when this non-volatile memory is used as a multi-level memory cell, the programmed voltage can be accurately controlled during the stylization operation. The starting voltage is programmed. In the range. For example, the multi-level memory cell is read by the reference voltage Vrefl,
VreG、Vref3 ’來判別四種不同啟始電壓(Threshold Voltage) Vthl 、Vth2、Vth3 、Vth4 ,VreG, Vref3 ' to discriminate four different threshold voltages (Threshold Voltage) Vthl, Vth2, Vth3, Vth4,
Vthl<Vrefl<Vth2<Vref2<Vth3<Vref3<Vth4。Vthl 例如為小 於〇伏特’表示為「〇〇」狀態;Vrefl例如為〇伏特左右; _ Vth2例如為0.2〜2伏特,表示為「〇1」狀態;Vref2例如 為2.2伏特;Vth3例如為2·4〜4.2伏特,表示為「1〇」狀 態;Vref3例如為4.4伏特;Vth4例如為大於4 4伏特,表 不為「11」狀態。在程式化時,要使記憶胞處於「〇1」或 「10」狀態時,需要使記憶胞的啟始電壓準確的落在ο.? 〜2伏特或2.4〜4.2伏特的範圍内。由於記憶胞處於「〇1」 或「10」狀態時的啟始電壓的範圍,因此需要進行多次程 式化步驟及程式化確認步驟,以使程式化記憶胞準確處於 • 設定的啟始電壓中,如此將花費較長的時間。然而,若採 用自我加速電荷注入(channel self_b〇〇sting)之方式,利用源 極侧注入效應來進行記憶胞程式化,預先使輔助間極的通 ,道充電至設定電壓’而可以快速的程式化記憶胞至「 或「1〇」狀態。至於「η」狀態,則可以採用自我加速電 何注入(channel self_boosting)之方式或者不採用自我加速 電荷注人(ehannd self_bGGSting)之κ進行記馳的程式 化,0 19 1307%& 02 19830twf.doc/eVthl < Vrefl < Vth2 < Vref2 < Vth3 < Vref3 < Vth4. Vthl is, for example, less than 〇 volts, denoted as "〇〇" state; Vref1 is, for example, 〇 volts; _ Vth2 is, for example, 0.2 to 2 volts, expressed as "〇1" state; Vref2 is, for example, 2.2 volts; Vth3 is, for example, 2· 4 to 4.2 volts, expressed as "1 〇" state; Vref3 is, for example, 4.4 volts; Vth4 is, for example, greater than 44 volts, and the table is not "11" state. When stylizing, in order to make the memory cell in the "〇1" or "10" state, it is necessary to make the starting voltage of the memory cell accurately fall within the range of ο.?~2 volt or 2.4~4.2 volt. Since the memory cell is in the range of the starting voltage in the "〇1" or "10" state, it is necessary to perform multiple stylization steps and stylization confirmation steps so that the stylized memory cell is accurately at the set starting voltage. This will take a long time. However, if self-accelerated charge injection (channel self_b〇〇sting) is used, the source side injection effect is used to program the memory cell, and the auxiliary inter-electrode channel is charged to the set voltage in advance, and the program can be quickly executed. The memory cell is in the state of "or "1". As for the "η" state, you can use the self-accelerating method or the self-accelerating charge (ehannd self_bGGSting) κ to perform the stylization, 0 19 1307% & 02 19830twf. Doc/e
所繪示為對記憶胞進行程式化操作之-實圖圖2A 犯所繪示為對記憶胞進行抹除操作之 2C所繪不為對記憶胞進行讀取操作之一實例的干ς圖 請參照圖2Α,錢行程式化操 結。 AG施加電壓Vag,以辅助 預先於輔助閉極It is depicted as a programmatic operation of the memory cell. Figure 2A is a diagram showing the example of the memory cell being erased by 2C. Referring to Figure 2, the money is stroked. AG applies voltage Vag to assist pre-assisted closure
^ L L 閘極AG的啟始電壓,電壓 二a】於, VI例如為5伏特。 g例如為8伙特左右。電壓 然後’於控制閘極(^施加電壓Vpi :於 ==壓=^_壓,電壓¥大於 或等於選擇閘極‘:二:伙^;,Vp2大於 特左右;電壓V1㈣例如為L5伏 M Vi大於紐VP3 ’電壓Vp3例如為〇伏特 右’以利用源極側注入效應程式化記憶胞。 加雷圖2B,於進行抹除操作時,於選擇閘極犯施 袖’並將基料置’錢财在電荷儲存層之電 ’、’! k擇開極SG被拉出並移除,其中電壓Vel =壓差會弓mFN穿隨效應。龍Vel例如為^至15 伏特左右。 請參照圖2C ’於進行讀取操作時,於輔助閘極八〇施 口 =壓Vrl ’於控制閘極CG施加電壓Vr2,於選擇閘極施 加電壓Vr2,於祕區D施加電壓—,以讀取記憶胞。 20 130796§〇2 19830twf.doc/e 其中,電壓Vrl例如是由輔助閘極下方的基底的電阻決 定,電壓Vrl例如為Vcc(電源電壓)。電壓Vr2例如為14 伏特左右。電壓Vr3大於選擇閘極的啟始電壓,電壓Vr3 例如為Vcc(電源電壓)。電壓Vr4例如為丨5伏特左右。 在上述偏壓情況下,可藉由偵測記憶胞之通道電流大 小來判斷儲存於此記憶胞中的數位資訊。 在本發明之操作方法中,由於採用自我加速電荷注入 (channel self_b〇osting)之方式’利用源極側注入效應來進行 s己憶胞程式化,預先使輔助閘極的通道充電至設定電壓, 而可以快速的程式化記憶胞。而且,利用選擇閘極來抹除 纪憶胞,使得電子經由選擇閘極而移除,可以減少電子穿 越穿隧介電層之次數,*提高元件可靠度。此外,電荷儲 存層鄰接騎閘極之處具有㈣。在進行抹除時,藉由此 轉角使電場集中,而可提高電子從電荷儲存層拉出至選擇 閘極的抹除操作速度。 ^圖3為繪示由本發明之多階非揮發性記憶胞所構成之 S己憶胞陣列之一實施例的電路簡圖。 夕如圖3所示,記憶胞陣列例如是由記憶胞QU〜Q44、 夕條辅助閘極、線AL1〜AL2、多條位元線BL1〜BL4盘多 ,控制閘極線CL1〜CL4。記憶胞Qn〜Q44的結科前 述圖1中的Q1〜Q4所示。 記憶胞Q11〜Q44排列成行/列陣列。在X方向(行方 二向)土’記憶胞Q11〜Q14例如是成鏡向配置。相鄰的兩個 ’《I Q11〜Q14會共用一個辅助閘極或沒極區。舉例來 21 Ό2 19830twf.doc/e 說,記憶胞Qll與Q12共用輔助閘極;記憶胞Q12與q13 共用汲極區;記憶胞Q13與Q14共用輔助閘極。 多條辅助閘極線AL1〜AL2在Y方向(列方向)上平行 排列’並連接同一列之記憶胞的輔助閘極。舉例來說,辅 助閘極線AL1連接記憶胞Q11〜記憶胞Q41以及記憶胞 Q12〜記憶胞Q42的輔助閘極;辅助閘極線AL2連接記憶 胞Q13〜記憶胞Q43以及記憶胞Q14〜記憶胞Q44的辅助 閘極。 多條位元線BL1〜BL4在X方向(行方向)上平行排 列,連接同一行之記憶胞的汲極區。舉例來說,位元線Bu 連接記憶胞Q11〜記憶胞Q14的汲極區;位元線BL2連接 s己憶胞Q21〜記憶胞Q24的汲極區;…;依此類推,位元 線BL4連接記憶胞q41〜記憶胞q44的汲極區。 多條控制閘極線CL1〜CL6在列方向上平行排列,並 連接同-狀記憶胞的控則極。舉例來說,控制閑極線 CL1連接記憶胞Q11〜記憶胞Q41的控制問極;控制間極 線CL2連接記憶胞φ2〜記憶胞Q42的控制閘極;…丨依 此類推,控制閘極線CL4連接記憶胞Q14〜記憶胞〇44 的控制閘極。 多條選擇閘極線SL1〜SL4在列方向上平行排列,並 連接同-狀記憶胞的選擇·。舉例來說,選擇間極線 SL1連接記憶胞Q11〜記憶胞Q41的選擇閉極;選擇間極 線SL2連接記憶胞Q12〜記憶胞Q42的選擇閘極;…;化 此類推’選擇閘極、線SL4連接記憶胞Q14〜記憶胞_ ^ 22 13079<^t)2 19830twf.doc/e 選擇閘極。 電晶體ΤΙ、T2的汲極D1、D2連接輔助閘極線AL1 〜AL2下方的基底。當輔助閘極線ALl、AL2上施加有電 壓而打開輔助閘極線AL1、AL2下方的通道、並形成反轉 層’且電晶體ΤΙ、T2的閘極Gl、G2施加有電壓而打開 電晶體Ή、T2的通道時,施加於電晶體ή、T2的源極 SI、S2的電流即會經由汲極D1、D2而流通至輔助閘極線 AU、AL2下方的反轉層。然後,關閉電晶體τ卜T2的 通道並且拉咼輔助閘極線ALl、AL2之電壓,使得輔助閘 極線ALl、AL2下方的反轉層可以處於預充電狀態,即反 轉層之電壓係大於源極SI、S2電壓減掉電晶體Ή、Τ2起 始電壓。在程^化本發明之非揮發性記憶體時,藉由自我 加,電荷注人(ehaimel self_b(K)Sting)來妨程式化可以提 。而且’當此非揮發性記憶體作為多階記憶 作時可辞確触制賴祕程式化後 的啟始電壓位於所設定的範圍内。 接著,說明本發明之非揮發性記憶 式’其係包括程式化、抹除與資料_ ί j之操作模 發明之非揮發性記紐之操作^㈣模式。就本 佳實施例作為說明。但本發明之非以下僅提供一較 作方法,並不限錄這料法。=料記憶斷列的操 記憶體陣顺行程式化操作之_ 4a崎示為對 繪示為對所有記憶胞進行抹除操作意圖。圖4B所 4C所繪示為記憶體陣列進行嘈一貫例的示意圖。圖 °锦作之一實例的示意 23 (2 19830twf.doc/e 圖。而且,在下述說明中係以圖3所示之記憶單元Q13為 實例做說明。 請參照圖3及圖4A ’在對記憶胞Q13進行程式化操 作時’預先於電晶體ΤΙ、T2的閘極施加電壓vtg ;於電晶 體τι、T2的源極SI、S2施加電壓vts;於輔助閘極線AL2 施加電壓Val,以於輔助閘極線AL2下方的基底中形成反 轉層L2,並使反轉層L2導通有電壓Vts_Vth,其中vth 為電晶體T卜T2之起始電壓,而輔助閘極線AL1之電壓 • 則維持0伏特,因此其下方的基底中不會形成反轉層。然 後’關閉電晶體ΤΙ、T2的通道’並且將辅助閘極線AL2 之電壓拉高至8伏特左右,以耦合反轉層L2之電壓至5 伏特左右,因此使得反轉層L2可以處於預充電狀態。其 中’電壓vtg例如是大於或者等於電晶體T1、T2的啟始_ 電壓Vth,以打開電晶體丁卜丁2的通道。電壓Vtg例如為^ L L The starting voltage of the gate AG, the voltage 2a is, for example, VI is 5 volts. g is, for example, about 8 bats. The voltage is then 'in the control gate (^ applied voltage Vpi: at == voltage = ^_ voltage, voltage ¥ is greater than or equal to the selected gate ': two: mate ^;, Vp2 is greater than special; voltage V1 (four) is for example L5 volt M Vi is greater than New VP3 'Voltage Vp3 is for example 〇 volt right' to program the memory cell using the source side injection effect. Adding Ray 2B, when performing the erase operation, the selection of the gate is made to apply the sleeve and set the base material 'The money in the charge storage layer', '! k select the open pole SG is pulled out and removed, where the voltage Vel = differential pressure will bow the mFN wear effect. The dragon Vel is, for example, ^ to 15 volts or so. 2C', when performing a read operation, applying a voltage Vr2 to the control gate CG at the auxiliary gate 〇 = = voltage Vrl ', applying a voltage Vr2 to the gate, and applying a voltage to the secret region D to read 20 130796§〇2 19830twf.doc/e where the voltage Vrl is determined, for example, by the resistance of the substrate under the auxiliary gate, and the voltage Vrl is, for example, Vcc (supply voltage). The voltage Vr2 is, for example, about 14 volts. The voltage Vr3 It is greater than the starting voltage of the selection gate, and the voltage Vr3 is, for example, Vcc (supply voltage). The voltage Vr4 is, for example, about 5 volts. In the case of the above bias voltage, the digital information stored in the memory cell can be judged by detecting the channel current of the memory cell. In the operation method of the present invention, The method of accelerating charge injection (channel self_b〇osting) uses the source side injection effect to program the memory of the memory, and the channel of the auxiliary gate is charged to the set voltage in advance, so that the memory cell can be quickly programmed. The selection of the gate to erase the memory cells, so that the electrons are removed through the selection gate, can reduce the number of electrons crossing the tunneling dielectric layer, * improve component reliability. In addition, the charge storage layer is adjacent to the gate Having (4). When erasing, the electric field is concentrated by the corner, and the erasing operation speed of electrons pulling out from the charge storage layer to the selection gate can be improved. FIG. 3 is a multi-stage non-volatile process of the present invention. A circuit diagram of an embodiment of a S memory cell composed of a memory cell. As shown in FIG. 3, the memory cell array is, for example, a memory cell QU~Q44, an auxiliary gate, and a line AL1. AL2, a plurality of bit lines BL1 to BL4, and control gate lines CL1 to CL4. The memory cells Qn to Q44 are shown in the above-mentioned Q1 to Q4 in Fig. 1. The memory cells Q11 to Q44 are arranged in a row/column array. In the X direction (bidirectional two-way) soil 'memory cells Q11~Q14 are, for example, mirrored. The two adjacent ''I Q11~Q14 will share an auxiliary gate or immersion area. For example, 21 Ό 2 19830twf .doc/e says that memory cells Qll and Q12 share the auxiliary gate; memory cells Q12 and q13 share the drain region; memory cells Q13 and Q14 share the auxiliary gate. The plurality of auxiliary gate lines AL1 to AL2 are arranged in parallel in the Y direction (column direction) and are connected to the auxiliary gates of the memory cells of the same column. For example, the auxiliary gate line AL1 is connected to the auxiliary gate of the memory cell Q11~memory cell Q41 and the memory cell Q12~memory cell Q42; the auxiliary gate line AL2 is connected to the memory cell Q13~memory cell Q43 and the memory cell Q14~memory cell The auxiliary gate of Q44. The plurality of bit lines BL1 to BL4 are arranged in parallel in the X direction (row direction) to connect the drain regions of the memory cells of the same row. For example, the bit line Bu is connected to the drain region of the memory cell Q11 to the memory cell Q14; the bit line BL2 is connected to the drain region of the memory cell Q21 to the memory cell Q24; and so on, the bit line BL4 Connect the bungee region of memory cell q41 ~ memory cell q44. A plurality of control gate lines CL1 to CL6 are arranged in parallel in the column direction, and are connected to the control electrode of the same-shaped memory cell. For example, the control idle line CL1 is connected to the control cell of the memory cell Q11~memory cell Q41; the control inter-pole line CL2 is connected to the control gate of the memory cell φ2~memory cell Q42; ...and thus, the gate line is controlled CL4 is connected to the control gate of memory cell Q14~memory cell 44. The plurality of selection gate lines SL1 to SL4 are arranged in parallel in the column direction, and are connected to the selection of the same-shaped memory cell. For example, the selection of the interpolar line SL1 is connected to the selective closed pole of the memory cell Q11 to the memory cell Q41; the selection of the interpolar line SL2 is connected to the selected gate of the memory cell Q12 to the memory cell Q42; Line SL4 is connected to the memory cell Q14~memory cell_^22 13079<^t)2 19830twf.doc/e Select the gate. The transistor ΤΙ, the drains D1 and D2 of T2 are connected to the substrate under the auxiliary gate lines AL1 to AL2. When a voltage is applied to the auxiliary gate lines AL1 and AL2 to open a channel under the auxiliary gate lines AL1 and AL2, and an inversion layer is formed, and the gates G1 and G2 of the transistors ΤΙ and T2 are applied with a voltage to turn on the transistor. In the case of the channel of T2 and T2, the currents applied to the source electrodes S1 and S2 of the transistor ή and T2 flow through the drain electrodes D1 and D2 to the inversion layer below the auxiliary gate lines AU and AL2. Then, the channel of the transistor τB2 is turned off and the voltages of the auxiliary gate lines AL1 and AL2 are pulled, so that the inversion layer under the auxiliary gate lines AL1 and AL2 can be in a precharge state, that is, the voltage of the inversion layer is greater than The source voltages of SI and S2 are reduced by the initial voltages of the transistors Ή and Τ2. In the case of the non-volatile memory of the present invention, by self-adding, the charge injection (ehaimel self_b(K)Sting) can be simplified. Moreover, when this non-volatile memory is used as a multi-level memory, the starting voltage after stylization is within the set range. Next, the non-volatile memory of the present invention will be described as including the operation (4) mode of the non-volatile memory of the invention, which is programmed, erased, and erased. The preferred embodiment is described as an example. However, the present invention is not limited to the following, and is not limited to the method. = The operation of the memory is interrupted. The memory array is arbitrarily operated. 4a is shown as the opposite of the memory operation. 4B is a schematic diagram showing a consistent example of a memory array. Figure 23 is an example of an example 23 (2 19830 twf.doc/e diagram. Moreover, in the following description, the memory unit Q13 shown in Fig. 3 is taken as an example. Please refer to Fig. 3 and Fig. 4A' When the memory cell Q13 is programmed, 'the voltage vtg is applied to the gate of the transistor ΤΙ, T2 in advance; the voltage vts is applied to the sources SI and S2 of the transistor τι, T2; and the voltage Val is applied to the auxiliary gate line AL2 to An inversion layer L2 is formed in the substrate under the auxiliary gate line AL2, and the inversion layer L2 is turned on by a voltage Vts_Vth, where vth is the initial voltage of the transistor Tb and T2, and the voltage of the auxiliary gate line AL1 is Maintain 0 volts, so no inversion layer is formed in the underlying substrate. Then 'turn off the transistor ΤΙ, T2 channel' and pull the voltage of the auxiliary gate line AL2 up to about 8 volts to couple the inversion layer L2 The voltage is about 5 volts, so that the inversion layer L2 can be in a precharge state, wherein the 'voltage vtg is, for example, greater than or equal to the start_voltage Vth of the transistors T1, T2 to open the channel of the transistor Ding Ding 2 The voltage Vtg is, for example,
Vcc(電源電壓)。電壓Vts例如為Vcc(電源電壓電壓W 例如是大於或者等於輔助閘極的啟始電壓,例如是開始為 參 Vcc(電源電壓)’等到-電晶體们、T2 _之後便拉高^ 伏特左右。 —^後,於選定之控制閘極線CL3施加電壓vpl,於選 定之選擇閉極線SL3施加電壓Vp2,於選定之位元線Bu 1加電m Vp3 ’電塵Vpl例如為w伏特左右;電塵^ 你ί或等於選擇閘極SG的啟始電愿,電壓Vp2例如為1.5 性士 Ϊ右’電麼VtS大於電愿Vp3 ’電屢Vp3例如為0伏 寺右,以利用源極侧注入效應程式化選定記憶胞叫。 24 1307%^〇2 19830twf.doc/e 在進行上述程式化操作時,對於與選定記憶胞QU共 用控制閘極線CL3、選擇閘極線SL3及輔助閘極線AL2 的其他非選定記憶胞Q23、Q33及Q43而言,則可以於這 些非選定記憶胞Q23、Q33及Q43所耦接的非選定位元線 BL2、BL3及BL4施加電壓,以抑制非選定記憶胞Q23、 Q33及Q43被程式化。此外’因為非選定控制閘極線cli、 CL2、CL4未施加電壓,所以其他非選定記憶胞qU〜 Q41、Q12〜Q42及Q14〜Q44不會被程式化。 §奇參照圖3及圖4B,在進行抹除操作時,於選擇閘極 線SL1〜SL4施加電壓Vel,並將基底浮置,以使儲存在 所有記憶胞Q11〜Q41、Q12〜Q42、Q13〜Q43及Q14〜Q44 之電荷儲存層之電子經由選擇閘極被拉出並移除,其中電 壓Vel與基底的電壓差會引發FN穿隧效應。電壓Vel例 如為11至15伏特左右。 請參照圖3及圖4C,於進行讀取操作時,於選定之辅 助閘極線AL2施加電壓Vd’於選定之控制閘極線CL3施 加電壓Vr2,於選定之選擇閘極線SL3施加電壓Vr3,於 選定之位元線BL1施加電壓Vr4,以讀取記憶胞Q13。其 中,電壓Vrl例如是由輔助閘極線下方的基底的電阻決 定,電壓Vrl例如為Vcc(電源電壓)。電壓Vr2例如為L4 伏特左右。電壓Vr3大於選擇閘極的啟始電壓,例如為 Vcc(電源電壓)。電壓Vr4例如為1.5伏特左右。至於非選 定之輔助閘極線AL1、非選定之控制閘極線CLI、CL2、 CL4、非選定之選擇閘極線SL1、SL2、SL4、非選定之位 25 Ό2 19830tw£doc/c Ό2 19830tw£doc/cVcc (supply voltage). The voltage Vts is, for example, Vcc (the power supply voltage W is, for example, greater than or equal to the starting voltage of the auxiliary gate, for example, starting with the reference Vcc (supply voltage)') until the transistors, T2_, and then pulling up by about volts. After ^^, a voltage vpl is applied to the selected control gate line CL3, a voltage Vp2 is applied to the selected closed line SL3, and a voltage Vp3 is applied to the selected bit line Bu1. The electric dust Vpl is, for example, about w volts; Electric dust ^You ί or equal to the starting power of the gate SG, the voltage Vp2 is for example 1.5 sex Ϊ right 'electricity VtS is greater than the electric wish Vp3 'electrical repeated Vp3 for example 0 volt temple right to use the source side The injection effect is programmed to select the memory cell. 24 1307%^〇2 19830twf.doc/e In the above stylized operation, the control gate line CL3, the selection gate line SL3 and the auxiliary gate are shared with the selected memory cell QU. For other unselected memory cells Q23, Q33, and Q43 of line AL2, voltages may be applied to the unselected positioning element lines BL2, BL3, and BL4 to which the unselected memory cells Q23, Q33, and Q43 are coupled to suppress non-selection. Memory cells Q23, Q33 and Q43 are stylized. Also 'because they are not selected The control gate lines cli, CL2, and CL4 are not applied with voltage, so other unselected memory cells qU to Q41, Q12 to Q42, and Q14 to Q44 are not programmed. § Scratchly with reference to FIG. 3 and FIG. 4B, the erase operation is performed. When a voltage Vel is applied to the selection gate lines SL1 to SL4, the substrate is floated so that electrons stored in the charge storage layers of all the memory cells Q11 to Q41, Q12 to Q42, Q13 to Q43, and Q14 to Q44 are selected. The gate is pulled out and removed, wherein the voltage difference between the voltage Vel and the substrate causes an FN tunneling effect. The voltage Vel is, for example, about 11 to 15 volts. Referring to FIG. 3 and FIG. 4C, when performing a read operation, The selected auxiliary gate line AL2 applies a voltage Vd' to apply a voltage Vr2 to the selected control gate line CL3, applies a voltage Vr3 to the selected gate line SL3, and applies a voltage Vr4 to the selected bit line BL1 to read the memory. The voltage Vrl is determined, for example, by the resistance of the substrate under the auxiliary gate line, and the voltage Vrl is, for example, Vcc (supply voltage). The voltage Vr2 is, for example, about L4 volts. The voltage Vr3 is greater than the starting voltage of the selected gate. For example, Vcc (supply voltage). The voltage Vr4 is, for example, about 1.5 volts. As for the unselected auxiliary gate line AL1, the unselected control gate lines CLI1, CL2, CL4, the unselected selection gate lines SL1, SL2, SL4, the unselected bits 25 Ό 2 19830tw£doc/c Ό2 19830tw£doc/c
« 元線BL2、BL3、BL4則不施加電壓或浮置。 在上述偏壓情況下,可藉由偵測記憶胞q13之通道電 ΛΙΙ·大小來判斷儲存於此記憶胞QU申的數位資訊。 在本發明之操作方法中,由於採用自我加速電荷注入 (self-boosted-charge injection)之方式,利用源極側注入效應 來,行s己憶胞程式化,預先使輔助閘極的通道充電至設定 電【而可以快速的程式化記憶胞。而且,利用選擇閘極 來抹除記憶胞,使得電子經由選擇閘極而移除,可以減少 電=穿越穿隧介電層之次數,而提高元件可靠度。此外, ,荷健存層雜選擇閘極之處具有㈣。在進行抹除時, 藉由此轉肖使電場集巾,而可提高電子從電荷错存層拉出 至選擇閘極的抹除操作速度。 接著,說明本發明之非揮發性記憶體的製造方法。圍 至圖5G為依照本發明之一實施例所繪示之 憶體的製造方法的流程剖面示賴。 輝發° a首先,請參照圖5A,提供基底200。此基底200例女 疋石夕基底。於基底200上形成一層介電層2〇2。介電声 的材質例如是氧切,介電層搬的形成方法例如是㈣ 化法。然後,於介電| 2〇2上形成一層電荷儲存層辦。 =儲存層204之材質例如是導體材料(如接雜多晶石夕)惠 =夠使電荷陷人於其巾之材質,例域切、氮氧化石夕、 鈦酸錄或氧化铪等。電荷儲存層綱的材質剔. 二ί矽時,其形成方法例如是利用化學氣相沈積法形居 一層未摻雜多晶㈣後,進行離子植人步驟⑽成之,^ 26 1307 19830twf.doc/e 者也可以採用臨場植入摻質的方式以化學氣相沈積法形成 之。而且,當電荷儲存層204為導體材料時,此電荷儲存 層204例如是成條狀(未纟會示)。 然後’於基底200上形成閘間介電層206。閘間介電 層206的材質例如是氧化矽/氮化矽/氧化矽,其形成方法 例如是先以熱氧化法形成一層氧化矽層後,再利用化學氣 相沈積法依序形成一層氮化矽層與另一層氧化矽層。當 然,閘間介電層206的材質也可以是氧化矽、或氮化矽/ 氧化矽等。 接著,請參照圖5B,於閘間介電層206上形成一層導 體層(未繪示)與一層頂蓋層(未繪示)。然後,圖案化頂蓋 層、導體層及閘間介電層206,以形成由閘間介電層2〇如、 導體層208及頂蓋層210構成的多個堆疊層212。導體層 208的材質例如是摻雜多晶矽,其形成方法例如是利用化 學氣相沈積法形成一層未摻雜多晶矽層後,進行離子植入 步驟以形成之,或者也可以採用臨場植入摻質的方式以化 學氣相沈積法形成之·。導體層2〇8例如是做為控制閘極。 頂蓋層210之材質例如是氮化矽或氧化矽,其形成方法例 如是化學氣相沈積法。 接著,請參照圖5C,於基底200上形成一層圖案化罩 幕層214。此圖案化罩幕層214具有開口 216,此開口 至少暴露每二堆疊層212之間的導體層204。然後,圖案 化罩幕層214為罩幕,移除每二堆疊層212之間的導體層 204 ’以形成-溝渠218。上述溝渠218的形成方法例如^ 27 13079餘 02 1983〇twf.d〇c/e 至曝露出介電堆糾212之_導體層綱 是光阻表面。圖案化罩幕層214之材質例如 程。,_化罩幕層214㈣成方法例如是微影製 案化;照圖5D,移除圖案化罩幕層214。移除圖 分的罩幕A,錢灰化餘移除大部 侧壁與溝渠2Μ側壁之後,於堆疊層212 龜日e 爾形成間隙壁220。間隙壁220的材質 如^夕或其他介電材料。間隙壁顶的形成方法例 2〇Γ。^ 相隙壁㈣層’叫形㈣住整個基底 層,以^成^行非等向性银刻餘,移除部分間隙壁材料 然後,請參照圖5Ε,以間隙壁22〇、堆疊層 =移除所裸露出之導體層綱,以於每—個堆疊層為 ^ =成電何儲存層2〇4a。其中位於電荷儲存層鳩與 二技G 0之間的介電層2 〇2例如是作為穿隧介電層。然後, 上形成共形的介電層222。介電層222的材質例如 疋1^熱氧切層’其形成方法例如是高溫熱氧化法。 隨後,請參照圖5F,於二堆疊層212之間形成導 =’並於二堆疊層212外側之側壁分別形成導體層22;。 =層224填滿溝渠218。導體層224及導體層226的材 、例如疋摻㈣晶⑦,其形成方法例如是利用化學氣相沈 積法形成一層未摻雜多晶矽層後,進行離子植入步驟以形 成之,或者也可以採用臨場植入摻質的方式以化學氣相沈 28 1307鄕2 19830twf.doc/e ft ίΐ。利用上述方法先於基底2GG上形成—層摻雜« The main lines BL2, BL3, BL4 are not applied with voltage or floating. In the above bias condition, the digital information stored in the memory cell can be determined by detecting the channel size of the memory cell q13. In the operation method of the present invention, since the source-side injection effect is used, the source side injection effect is used, and the channel of the auxiliary gate is charged in advance to the channel of the auxiliary gate. Set the battery [can quickly program the memory cell. Moreover, by using the selection gate to erase the memory cell, the electrons are removed via the selection gate, which can reduce the number of times of crossing the tunneling dielectric layer and improve component reliability. In addition, there is (4) where the gate is selected by the drain layer. When the erasing is performed, the electric field is spread by the electric field, and the erasing operation speed of the electrons from the charge dislocation layer to the selection gate can be increased. Next, a method of producing the non-volatile memory of the present invention will be described. FIG. 5G is a flow cross-sectional view showing a method of manufacturing a memory in accordance with an embodiment of the present invention. Hui Fa ° a First, please refer to FIG. 5A, a substrate 200 is provided. This base of 200 cases of female 疋石夕 base. A dielectric layer 2〇2 is formed on the substrate 200. The material of the dielectric sound is, for example, oxygen cutting, and the method of forming the dielectric layer is, for example, a (four) method. Then, a layer of charge storage layer is formed on the dielectric | 2〇2. = The material of the storage layer 204 is, for example, a conductor material (such as a polycrystalline spine). It is sufficient to cause the charge to be trapped in the material of the towel, such as smear, nitrous oxide, titanate or yttrium oxide. The material of the charge storage layer is determined by the method of chemical vapor deposition, for example, after forming an undoped polycrystal (IV) by chemical vapor deposition, and then performing the ion implantation step (10), ^ 26 1307 19830 twf.doc /e can also be formed by chemical vapor deposition by means of on-site implant doping. Moreover, when the charge storage layer 204 is a conductor material, the charge storage layer 204 is, for example, strip-shaped (not shown). An inter-gate dielectric layer 206 is then formed on the substrate 200. The material of the inter-gate dielectric layer 206 is, for example, yttrium oxide/yttria/yttria, which is formed by, for example, forming a layer of ruthenium oxide by thermal oxidation, and then sequentially forming a layer of nitridation by chemical vapor deposition. The tantalum layer and another layer of tantalum oxide. Of course, the material of the inter-gate dielectric layer 206 may also be tantalum oxide, tantalum nitride, tantalum oxide or the like. Next, referring to FIG. 5B, a conductor layer (not shown) and a top cap layer (not shown) are formed on the inter-gate dielectric layer 206. Then, the cap layer, the conductor layer and the inter-gate dielectric layer 206 are patterned to form a plurality of stacked layers 212 composed of the inter-gate dielectric layer 2, such as the conductor layer 208 and the cap layer 210. The material of the conductor layer 208 is, for example, a doped polysilicon, which is formed by, for example, forming an undoped polysilicon layer by chemical vapor deposition, performing an ion implantation step, or forming a dopant by field implantation. The method is formed by chemical vapor deposition. The conductor layer 2〇8 is used, for example, as a control gate. The material of the cap layer 210 is, for example, tantalum nitride or hafnium oxide, and its formation method is, for example, chemical vapor deposition. Next, referring to FIG. 5C, a patterned mask layer 214 is formed on the substrate 200. The patterned mask layer 214 has an opening 216 that exposes at least the conductor layer 204 between each two stacked layers 212. The patterned mask layer 214 is then a mask that removes the conductor layer 204' between each two stacked layers 212 to form a trench 218. The method for forming the trench 218 is, for example, ^ 27 13079 02 1983 twf.d 〇 c / e to expose the dielectric stack 212 - the conductor layer is the photoresist surface. The material of the patterned mask layer 214 is, for example, a process. The method of forming the mask layer 214 (four) is, for example, lithography; as shown in Fig. 5D, the patterned mask layer 214 is removed. After removing the mask A of the figure, after removing the sidewalls of the large sidewalls and the trenches, the spacers 220 are formed on the stacked layers 212. The spacer 220 is made of a material such as a dielectric material or other dielectric material. Example of forming a spacer top 2 〇Γ. ^ Phase gap wall (four) layer 'four shape' (four) live the entire base layer, to form a non-isotropic silver engraving, remove part of the spacer material and then, referring to Figure 5, with spacers 22, stacked layers = The exposed conductor layer is removed so that each of the stacked layers is ^= into the storage layer 2〇4a. The dielectric layer 2 〇 2 between the charge storage layer 鸠 and the second technique G 0 is, for example, a tunneling dielectric layer. A conformal dielectric layer 222 is then formed thereon. The material of the dielectric layer 222 is, for example, a hot-oxidized layer. The method of forming the dielectric layer 222 is, for example, a high-temperature thermal oxidation method. Subsequently, referring to FIG. 5F, a conductive layer 22 is formed between the two stacked layers 212 and formed on the sidewalls outside the two stacked layers 212; The layer 224 fills the trench 218. The conductor layer 224 and the material of the conductor layer 226, for example, ytterbium doped (tetra) crystal 7, are formed by, for example, forming an undoped polysilicon layer by chemical vapor deposition, performing an ion implantation step to form, or may be employed. The method of implanting the dopant in the field is chemical vapor deposition 28 1307鄕2 19830twf.doc/e ft ΐ. Forming layer doping prior to substrate 2GG using the above method
分:雜曰石;ίΐ亍回蝕刻製程(非等向性蝕刻製程)移除部 刀摻雜夕一層直到暴露出頂 2iG =及導體層226。導體層224例如是作為輔助閉= ,層224與基底200之間的介電層222及介電層2〇2例如 疋作為辅助閘極介電層。導體層226 極。在導體層226與基底200之間的介電層 202例如是作為選擇閘極介電層。其中介電層及介電Divided: Heterolithic; ΐ亍 etchback process (non-isotropic etching process) removal section The blade is doped to the top layer until the top 2iG = and the conductor layer 226 are exposed. The conductor layer 224 is, for example, an auxiliary closed =, a dielectric layer 222 between the layer 224 and the substrate 200, and a dielectric layer 2, for example, 疋 as an auxiliary gate dielectric layer. The conductor layer is 226 poles. The dielectric layer 202 between the conductor layer 226 and the substrate 200 is, for example, a selective gate dielectric layer. Dielectric layer and dielectric
層202的厚度總和例如是介於120埃至13〇埃之間。 然後,於二堆疊層212外侧之基底20〇中分別形成摻 雜區228。此摻雜區228例如是作為汲極區。摻雜區228The sum of the thicknesses of layer 202 is, for example, between 120 angstroms and 13 angstroms. Then, doped regions 228 are formed in the substrates 20A outside the two stacked layers 212, respectively. This doped region 228 is, for example, a drain region. Doped region 228
的形成方法例如是進行一離子植入製程。其中,導體層2 2 6 例如是形成在摻雜區228與導體層208之間的基底200上。 隨後,請參照圖5G,於基底200上形成層間絕緣層 230。層間絕緣層230的材質例如是氧化矽、磷矽玻璃、硼 磷矽玻璃或其他適合之介電材料,其形成方法例如是化學 氧相沈積法。然後,於層間絕緣層230中形成與摻雜區228 電性連接的插塞232。插塞232之形成方法例如是先圖案 化層間絕緣層230以形成暴露摻雜區228之開口,然後於 開口中填入導體材料而形成之。 之後,再於層間絕緣層230上形成與插塞232電性連 接的導體層234。導體層234例如是作為位元線。後續完 成非揮發性記憶體之製程為習知技術者所周知,在此不再 贅述。 1307暢丨 19830twf.doc/e —在本發明之非揮發性記憶體之製造方法中,由於採用 自厅對準的方式形成辅助閘極(導體層224)及選擇閘極(導 體層226) ’不需要使用到微影製程,因此可以節省製 本、增加製程裕度。 而且由於所形成的電荷储存層204a在靠近選擇閘極 (導體層226)之部分具有一個轉角,因此在對記憶胞進 藉由此轉角使電場集中,而可提高電子從電荷儲存 θ 〇,拉出至選擇間極(導體層226)的抹除操作速度。 ,上所述’本發明之多階非揮發性記憶體,相鄰的兩 例如是以鏡像對稱的方式配置,亦即相鄰的 憶胞共用辅助開極或摻雜區,因此能夠提高元件的積集度? :且:當辅助閘極上施加有電壓而打開辅助閘極下; 、、、、並形成反轉層時,於反轉層施加電壓, 反制可以處㈣充錄態。在程式化本發明 ”揮發性記憶體時’藉由自我加速電荷= _bQ〇sted_ehaFge injeetiGn)之方式 ’ _ 源極側注 來進行記憶胞程式化-,可以提升程式化迷度。而且^應 =發性記憶體作為多階記憶胞時,在_化_^此 圍内。 莖位於所扠疋的範 此外’由於所形成的電荷儲存層在靠 ^具有一個轉角,因此在對記憶胞進行抹_,二,部 角使電場集中,而可提高電子從電荷儲存 :此轉 極的抹除操作速度。 e杈出至選擇閘 30 1307965)2 19830twf.doc/e 1307965)2 19830twf.doc/eThe formation method is, for example, an ion implantation process. The conductor layer 2 26 is formed, for example, on the substrate 200 between the doped region 228 and the conductor layer 208. Subsequently, referring to FIG. 5G, an interlayer insulating layer 230 is formed on the substrate 200. The material of the interlayer insulating layer 230 is, for example, cerium oxide, phosphoric glass, borophosphon glass or other suitable dielectric material, and is formed by a chemical oxygen phase deposition method, for example. Then, a plug 232 electrically connected to the doping region 228 is formed in the interlayer insulating layer 230. The plug 232 is formed by, for example, patterning the interlayer insulating layer 230 to form an opening exposing the doped region 228, and then filling the opening with a conductor material. Thereafter, a conductor layer 234 electrically connected to the plug 232 is formed on the interlayer insulating layer 230. The conductor layer 234 is, for example, a bit line. The subsequent process of completing the non-volatile memory is well known to those skilled in the art and will not be described here. 1307丨19830twf.doc/e—In the manufacturing method of the non-volatile memory of the present invention, the auxiliary gate (conductor layer 224) and the selective gate (conductor layer 226) are formed by self-alignment. No need to use the lithography process, so you can save on the cost and increase the process margin. Moreover, since the formed charge storage layer 204a has a corner at a portion close to the selection gate (the conductor layer 226), the electric field is concentrated by the corner of the memory cell, and the electrons can be increased from the charge storage θ 〇 The erase operation speed to the selected interpole (conductor layer 226). In the above-mentioned multi-stage non-volatile memory of the present invention, adjacent two are arranged, for example, in a mirror symmetrical manner, that is, adjacent memory cells share an auxiliary open or doped region, thereby improving component Accumulation? : and: When a voltage is applied to the auxiliary gate to open the auxiliary gate; when, and, and an inversion layer is formed, a voltage is applied to the inversion layer, and the counteraction can be in (4) the charged state. When stylizing the "volatile memory" of the present invention, by self-accelerating charge = _bQ〇sted_ehaFge injeetiGn), the _ source side note is used to program the memory cell, which can improve the stylization fantasity. When the memory is used as a multi-order memory cell, it is within the range of _ _ _ ^. The stalk is located in the yoke of the fork. In addition, since the formed charge storage layer has a corner in the ^, the memory cell is wiped. _, two, the corners make the electric field concentrated, and can increase the electrons from the charge storage: the speed of the erase operation of the pole. e杈 to select the gate 30 1307965) 2 19830twf.doc / e 1307965) 2 19830twf.doc / e
» 另外,本發明之多階非揮發性記憶體製造方法簡單, 可以降低製造成本,並增加製程裕度。 雖然本發明已以實施例揭露如上’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 ,内,當可作些許之更触難,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為準。 【圖式簡單說明】 圖1是依照本發明的實施例所繪示之非揮發性記 的結構剖面圖。 X °〜 圖2A所繪示為對記憶胞進行程式化操作之一 示意圖。 、 圖2B所繪示為對記憶胞進行抹除操作之一實例的示 意圖。 圖2C所繪示為對記憶胞進行讀取操作之一實例的示 意圖。 、、 ® 3為緣示由本發明之多階非揮發性記憶胞所構成之 記憶胞陣列之一實施例的電路簡圖。 圖4A所繪示為對記憶體陣列進行裎式化操作之一實 例的示意圖。 ' 圖4B所繪示為對記憶體陣列進行抹除操作之―實例 的示意圖。 圖4C所繪示為對記憶體陣列進行讀取操作之一實例 的示意圖。 圖5A至圖5G為依照本發明之一實施例所綠示之非揮 發性記憶體的製造方法的流程剖面示意圖。曰 31 1307 9^0^02 19830twf.doc/e 【主要元件符號說明】 100、200 :基底 102 :穿隧介電層 104、204、204a :電荷儲存層 106、206、206a :閘間介電層 108、CG :控制閘極 110、 210 :頂蓋層 111、 212 :堆疊層 • 112a、112b、220 :間隙壁 114、228 :摻雜區 116、SG :選擇閘極 118、AG :輔助閘極 120、122、202、222 :介電層。 124、L、LI、L2 :反轉層 126 :轉角 128、BL1〜BL4 :位元線 修 130、232:插塞- 208、224、226、234 :導體層 214:圖案化罩幕層 216 :開口 218 :溝渠 222 :介電層 230 :層間絕緣層 AG :輔助閘極 32 1307945)2 19830twf.doc/e ALl〜AL2 :輔助閘極線 CL1〜CL4 :控制閘極線。 D .波極區 Dl、D2 :汲極 G1〜G2 :閘才亟 Q1〜Q4、Q11〜Q44 :記憶胞 SI、S2 :源極 ΤΙ、T2 :電晶體In addition, the multi-stage non-volatile memory of the present invention is simple in manufacturing method, can reduce manufacturing cost, and increase process margin. The present invention has been disclosed in the above embodiments, and it is not intended to limit the present invention. Any one skilled in the art can protect the present invention even if it can be made more difficult without departing from the spirit and scope of the present invention. The scope is subject to the definition of the scope of the patent application attached. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a cross-sectional view showing the structure of a non-volatile memory according to an embodiment of the present invention. X ° ~ Figure 2A is a schematic diagram of one of the stylized operations on the memory cell. FIG. 2B is a schematic diagram showing an example of erasing a memory cell. Fig. 2C is a diagram showing an example of a reading operation on a memory cell. And, ® 3 are circuit diagrams showing an embodiment of a memory cell array composed of the multi-stage non-volatile memory cells of the present invention. Figure 4A is a schematic diagram showing an example of a singulation operation of a memory array. Figure 4B is a schematic illustration of an example of an erase operation of a memory array. Figure 4C is a schematic diagram showing an example of a read operation on a memory array. 5A through 5G are schematic cross-sectional views showing a process of manufacturing a non-volatile memory according to an embodiment of the present invention.曰31 1307 9^0^02 19830twf.doc/e [Description of main component symbols] 100, 200: substrate 102: tunneling dielectric layers 104, 204, 204a: charge storage layers 106, 206, 206a: dielectric between gates Layer 108, CG: Control gate 110, 210: Top cover layer 111, 212: Stacked layer • 112a, 112b, 220: Clearance wall 114, 228: Doped region 116, SG: Select gate 118, AG: Auxiliary gate Pole 120, 122, 202, 222: dielectric layer. 124, L, LI, L2: inversion layer 126: corner 128, BL1~BL4: bit line repair 130, 232: plug - 208, 224, 226, 234: conductor layer 214: patterned mask layer 216: Opening 218: trench 222: dielectric layer 230: interlayer insulating layer AG: auxiliary gate 32 1307945) 2 19830twf.doc/e AL1 to AL2: auxiliary gate lines CL1 to CL4: control gate line. D. Wave zone Dl, D2: Bungee G1~G2: Gate 亟 Q1~Q4, Q11~Q44: Memory cell SI, S2: Source ΤΙ, T2: Transistor
3333
Claims (1)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW95124452A TWI307965B (en) | 2006-07-05 | 2006-07-05 | Multi-level non-volatile memory and manufacturing method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW95124452A TWI307965B (en) | 2006-07-05 | 2006-07-05 | Multi-level non-volatile memory and manufacturing method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TW200805675A TW200805675A (en) | 2008-01-16 |
| TWI307965B true TWI307965B (en) | 2009-03-21 |
Family
ID=44766141
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW95124452A TWI307965B (en) | 2006-07-05 | 2006-07-05 | Multi-level non-volatile memory and manufacturing method thereof |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI307965B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI467745B (en) * | 2012-05-07 | 2015-01-01 | Ememory Technology Inc | Non-volatile memory and manufacturing method thereof |
-
2006
- 2006-07-05 TW TW95124452A patent/TWI307965B/en active
Also Published As
| Publication number | Publication date |
|---|---|
| TW200805675A (en) | 2008-01-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI308763B (en) | Method for operating single-poly non-volatile memory device | |
| US10734408B2 (en) | Ferroelectric non-volatile memory | |
| TW512495B (en) | Nonvolatile semiconductor memory device, process of manufacturing the same and method of operating the same | |
| US10461095B2 (en) | Ferroelectric non-volatile memory | |
| TWI415249B (en) | Depletion mode charge trapping flash device | |
| CN102386188B (en) | Three-dimensional array memory architecture with diodes in memory strings | |
| CN1757114B (en) | Charge trapping memory array resistant to damage caused by contact hole formation | |
| CN102842339B (en) | A three-dimensional array memory device and its operating method | |
| TWI296440B (en) | Nov-volatile memory and method of forming thereof | |
| TWI282167B (en) | Charge trapping non-volatile memory with two trapping locations per gate, and method for operating same | |
| US10453862B1 (en) | Ferroelectric non-volatile memory | |
| TW200926184A (en) | Operation methods for memory cell and array thereof immune to punchthrough leakage | |
| TW201030945A (en) | Nonvolatile memory cell and method for fabricating the same | |
| TW200837887A (en) | Trench memory structures and operation | |
| TWI247419B (en) | Fabrication of gate dielectric in nonvolatile memories in which a memory cell has multiple floating gates | |
| TW201622111A (en) | Method of manufacturing semiconductor device | |
| TW200931648A (en) | Non-volatile semiconductor memory device and method of manufacturing the same | |
| KR100843141B1 (en) | Non-volatile memory integrated circuit device and manufacturing method thereof | |
| TWI281753B (en) | Non-volatile memory and manufacturing method and operating method thereof | |
| TW586221B (en) | Flash memory with selective gate within a substrate and method of fabricating the same | |
| TWI307965B (en) | Multi-level non-volatile memory and manufacturing method thereof | |
| TW201218320A (en) | Nonvolatile memory having raised source and drain regions | |
| CN101236787A (en) | Method for reading memory array | |
| US20110058410A1 (en) | Semiconductor memory device | |
| JP5801049B2 (en) | Method for writing data to semiconductor memory device and semiconductor memory device |