TW587282B - Method for improving the performance of flash memory by using microcrystalline polysilicon film as a floating gate - Google Patents

Method for improving the performance of flash memory by using microcrystalline polysilicon film as a floating gate Download PDF

Info

Publication number
TW587282B
TW587282B TW91114467A TW91114467A TW587282B TW 587282 B TW587282 B TW 587282B TW 91114467 A TW91114467 A TW 91114467A TW 91114467 A TW91114467 A TW 91114467A TW 587282 B TW587282 B TW 587282B
Authority
TW
Taiwan
Prior art keywords
microcrystalline silicon
patent application
floating gate
silicon layer
scope
Prior art date
Application number
TW91114467A
Other languages
Chinese (zh)
Inventor
Tzung-Ting Han
Chin-Ta Su
Yun-Chi Yang
Original Assignee
Macronix Int Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Macronix Int Co Ltd filed Critical Macronix Int Co Ltd
Priority to TW91114467A priority Critical patent/TW587282B/en
Application granted granted Critical
Publication of TW587282B publication Critical patent/TW587282B/en

Links

Abstract

This invention provides a method for forming polysilicon by using silane with introducing hydrogen, such that polysilicon is microcrystalline. This microcrystal polysilicon can be applied to floating gate of flash memory to improve the character of flash memory.

Description

587282 五、發明說明(1) 5 _ 1發明領域: 一法 於方 關之 有能 是性 別體 特憶 ,記 體式 憶閃 記快 式之 閃進 快促 USul BhJ- 種閘 一置 於浮 關之 有膜 係碎 明晶 發微 本用 利 .1PC- 種 景 背 明 發 2 1 5 排電中 列應塊 和效區 行場每 以 閑 , 其置塊 ,浮區 列一 一 陣含為 憶包組 記胞群 一 晶呈 括個常 包一通 體每胞 憶,晶 己包, =0 式晶荷 閃憶電 快記存 的個儲 型數以 典複用 成體 列晶 用 隨 為 作 中存 閘儲 置的 浮據 從數 作被中 電可胞 充荷晶 的電定 閘式決 置程除 浮除抹 用抹或 利塊在 可區存 胞由的 晶藉荷 個,電 一式中 每程閘 之作置 運浮 性, 電除 機移 第一圖所示為猶如使用於快閃式記憶體中,典型的記 憶晶胞5的截面圖。記憶晶胞5包含有一源極6 0區域和一汲 極7 0區域,源極6 0和沒極7 0以一通道區域8 0加以隔離開。 記憶晶胞5更包括一浮置閘3 0其以第一多晶矽層形成、一 控制閘5 0其以第二多晶矽層形成,浮置閘3 0和控制閘5 0則 以一内多晶介電層4 0加以隔離,浮置閘3 0和通道8 0區域則 利用一約厚1 0 0埃之薄氧化層(穿遂氧化層)2 0加以分開。 第二圖所示為利用多晶矽層8 4形成之浮置閘位於一底587282 V. Description of the invention (1) 5 _ 1 Field of invention: One method that can be used in the Fangguan is the special recollection of the gender, the flash of the remembering style, the flashing of the quick style, and the rapid advancement of the USul BhJ- the gate is placed in the floating barrier There are film-type shattered crystal hair micro-costs. 1PC- Species back bright hair 2 1 5 The row of the power block and the effect area are always idle, and the block and the floating area are included in the memory for a while. The package group records a group of cells, including a regular package, a single cell, a cell, and a crystal, which is equal to 0. The number of storage types recorded by the crystal charge flash memory electric memory is multiplexed into a series of crystals. The floating data stored in the gate are removed from the electric fixed-gate decisive process that is charged by CLP's cell, and the wiper wipes or wipes the crystals that can be stored in the area. The float operation of each gate is floating. The first figure shows a cross-sectional view of a typical memory cell 5 as if it is used in flash memory. The memory cell 5 includes a source 60 area and a drain 70 area, and the source 60 and the end 70 are separated by a channel area 80. The memory cell 5 further includes a floating gate 30 formed by a first polycrystalline silicon layer, a control gate 50 formed by a second polycrystalline silicon layer, and the floating gate 30 and the control gate 50 are formed by a The inner polycrystalline dielectric layer 40 is isolated, and the area between the floating gate 30 and the channel 80 is separated by a thin oxide layer (passage oxide layer) 20 with a thickness of about 100 angstroms. The second figure shows a floating gate formed by using a polycrystalline silicon layer 84.

第4頁Page 4

587282 五、發明說明(2) 材8 2如穿遂氧化層上之截面圖。多晶矽層8 4的形成,利用 低壓化學氣相沉積法、溫度約6 0 0°C,以矽烷(S i Η 4)為材 質沉積多晶矽在底材8 2表面。因大的柱狀結晶造成多晶矽 層8 4的表面不規則性,同時此種表面的不規則性將很難得 到良好的圖案轉移圖形,致使光阻層内部因多晶矽層8 4表 面的粒狀產生顯著的形變,及造成圖案轉移過程中非均勻 性的反射,光阻層的非均勻性反射導致產生較差的蝕刻圖 案,以及容易產生多晶矽層2 5的殘留。 第三圖所示為另一種多晶矽層8 8作為浮置閘形成於底 材8 6如穿遂氧化層上之截面圖。多晶矽層8 8的形成,利用 i氐壓化學氣相沉積法、溫度約5 5 0°C,沉積碎在底材8 6表 面上。在此低溫下沉積矽,乃產生非晶形矽,因為晶粒無 法在此低溫下形成。此非晶形矽接著暴露在溫度約6 0 0°C 再結晶。最後的再結晶結構如第三圖所示,其具有大的晶 粒形成。然此第三圖之多晶矽層8 8同樣具有如第二圖中所 示之表面不規則性的缺點,多晶矽層8 8中大的晶粒尺寸會 降低膜中的晶界密度。除此之外,也因為多晶矽層8 8於低 溫下沉積,沉積速率相對降低,產生慢速的產出週期。 快閃記憶體的 N 〇 r d h e i πι 模式,如 層’穿遂氧化層介 層通常為1 0 0埃厚587282 V. Description of the invention (2) Material 8 2 Cross-section view on the oxide layer. The polycrystalline silicon layer 8 4 is formed by depositing polycrystalline silicon on the surface of the substrate 8 2 by using a low pressure chemical vapor deposition method at a temperature of about 600 ° C and using silane (S i Η 4) as a material. The surface irregularity of the polycrystalline silicon layer 84 is caused by large columnar crystals. At the same time, it is difficult to obtain a good pattern transfer pattern on this surface irregularity, which causes the interior of the photoresist layer to be caused by the graininess of the polycrystalline silicon layer 84 Significant deformation and non-uniform reflection during the pattern transfer process, non-uniform reflection of the photoresist layer result in poor etching patterns, and the residue of the polycrystalline silicon layer 25 is easily generated. The third figure shows a cross-sectional view of another polycrystalline silicon layer 88 formed as a floating gate on a substrate 86 such as a tunnel oxide layer. The polycrystalline silicon layer 88 is formed on the surface of the substrate 86 by using an i-pressure chemical vapor deposition method at a temperature of about 550 ° C. Depositing silicon at this low temperature produces amorphous silicon because grains cannot form at this low temperature. This amorphous silicon is then recrystallized by exposure to a temperature of about 600 ° C. The final recrystallized structure is shown in the third figure, which has large crystal grain formation. However, the polycrystalline silicon layer 88 of the third figure also has the disadvantage of surface irregularities as shown in the second figure. The large grain size in the polycrystalline silicon layer 88 may reduce the grain boundary density in the film. In addition, because the polycrystalline silicon layer 88 is deposited at a low temperature, the deposition rate is relatively reduced, resulting in a slow output cycle. The flash memory ’s N 0 r d h e i π mode, such as the layer ’s tunneling oxide interlayer, is usually 100 Angstroms thick.

F 化 氧 遂 穿 的 薄 過 由經 藉穿 常遂 式子 方電 除, 抹中 或圖 存一 儲第 置 浮 於 快 於 對 和 之 道 此 化 氧 遂 穿 的 可 亦 式 模 電 充 的 體 憶 記 式F The thinning of the oxygen oxidative wear is removed by the regular wear and tear formula, and the storage or storage of a storage capacitor is faster than the method of reconciliation. Body memory

第5頁Page 5

五、發明說明(3) 使用熱載子經由通道遂穿到% 。快閃記憶體利用熱載子寫’予置鬧,並且儲存在浮置閘中 正電荷,源極則連接地線Y入動作中,控制閘和汲極連接 N 〇 r d h e i m方式在控制閘和通、、,抹除動作常利用F 〇 w 1 e r -除程序。 ^:間給予高電壓,使之完成抹 圮憶晶胞的過度抹除動 · 意味著較高的電流通過穿下為有較短週期的抹除速度, 晶矽浮置閘時,會引起^々氧化層。。當使用大晶粒的多 過度抹除區域存^時,/ ^範圍的起始電壓(Vt)分佈。若 有尾巴狀分佈和較大的 f起始電壓分佈範圍外,會帶 起始電1分佈與大晶:多1;:;,另-方面1寬的抹除 夕日日矽有關。 傳統的快閃式記恃#制# 士 t 匕丄 U體製程中,利用大晶粒的多晶功π 作為浮置閘乃產生諸多缺4 a扣、网也4士 ^ 4 日日石夕膜 夕缺點,如引起過快抹除動作(pV. Description of the invention (3) The hot carrier is penetrated to% through the channel. The flash memory uses hot carriers to write 'pre-alarm', and stores the positive charge in the floating gate, the source is connected to the ground Y to enter the action, and the control gate and the drain are connected to the gate. The erasing action often uses F 0w 1 er-erasing program. ^: High voltage is occasionally applied to complete the erasing of the memory cell. It means that a higher current passes through the cell for a shorter period of erasing speed. When the silicon floats on the gate, it will cause ^ Gadolinium oxide layer. . When large over-erased areas with large grains are used, the starting voltage (Vt) of the / ^ range is distributed. If there is a tail-like distribution and a large f-start voltage distribution range, the distribution of the initial charge 1 will be related to the large crystal: more than 1;:;, and the other-the 1-wide erasure of the silicon. In the traditional flash-style recording of # 恃 #, the use of large-crystalline polycrystalline work π as a floating gate in the process of the U.S. system has a lot of shortcomings. Disadvantages of the membrane, such as causing too fast erasing action (p

Erase)或過度寫入(Fast pr〇gram)、寬的起始電壓 st 附帶有尾巴狀分佈產生、穿遂氧化層性能下降、多€及 值較咼和較差的蝕刻外觀而導致多晶矽殘存等問題。阻 3發明目的及概述 鑒於上述之發明背景中,本發明提供以微晶矽 憶晶胞之浮置閘,改善傳統的快閃式記憶體製程為記 諸多缺點。 生的Erase) or overwrite (Fast pr0gram), a wide starting voltage st with tail-like distribution, reduced performance of the tunneling oxide layer, more polysilicon, and poor etching appearance, resulting in polysilicon residues, etc. . In view of the above-mentioned background of the invention, the present invention provides a number of shortcomings for improving the traditional flash memory system by using a floating gate of a microcrystalline silicon memory cell. raw

第6頁 587282 五、發明說明(4) 本發明的另一目的在使用微晶矽浮置閘以改善記憶晶 胞之較寬的抹除起始電壓分佈及帶有尾巴狀分佈問題。 本發明的再一目的,係用以微晶矽浮置閘改善記憶晶 胞過度抹除問題。 · 本發明的另一目的,係用以微晶矽浮置閘促進記憶晶 胞之穿遂氧化層之性能。 本發明的另一目的,係用以微晶矽浮置閘促進快閃式 記憶體的使用性。 本發明的另一目的,係用以微晶矽改善浮置閘的蝕刻 外觀。 本發明的另一目的,係用以微晶矽降低浮置閘的阻值 本發明揭露一記憶晶胞使用微晶矽作為浮置閘材質, 低電子親和性是使用在快閃式記憶體的特性,微晶粒更突 顯低電子親和性。使用低電子親和性材質作為浮置閘,以 提供降低與穿遂氧化層介面的阻障,降低穿遂距離通常可 增加電子穿遂的可行性。使用低電子親和性的浮置閘以解Page 6 587282 V. Description of the invention (4) Another object of the present invention is to use a microcrystalline silicon floating gate to improve the wider erased initial voltage distribution of the memory cell and the problem of having tail-like distribution. Another object of the present invention is to improve the erasure of memory cells by using a microcrystalline silicon floating gate. · Another object of the present invention is to use a microcrystalline silicon floating gate to promote the performance of a tunneling oxide layer of a memory cell. Another object of the present invention is to use a microcrystalline silicon floating gate to promote the usability of a flash memory. Another object of the present invention is to improve the etching appearance of a floating gate by using microcrystalline silicon. Another object of the present invention is to reduce the resistance of a floating gate by using microcrystalline silicon. The present invention discloses that a memory cell uses microcrystalline silicon as the material of the floating gate. The low electronic affinity is used in flash memory. Characteristics, the microcrystals are more prominent with low electron affinity. The use of a low electron affinity material as a floating gate provides a lower barrier to the tunneling oxide interface, and lowering the tunneling distance generally increases the feasibility of electron tunneling. Use a low electron affinity floating gate to solve

第7頁 587282 五、發明說明(5) 決這些問題,因此,低阻障促使電子逃脫或藉由穿移出。 低阻障所需較低電壓,類似於穿遂距離,作為電子抹除動 作,此結果導致較快的抹除時間和減少損害。 於本發明之利用微晶矽浮置閘以改善快閃式記憶體之 性能,其方法為提供一底材,其中該底材包含有源極區域 和汲極區域,並以通道區域相隔離。一穿遂氧化層形成在 該底材上。以微晶矽層作為浮置閘形成於穿遂氧化層上, 微晶矽膜以低壓化學氣相沉積法、選擇矽烷或乙矽烷為材 質,並同時導入氫氣形成。一内多晶介電層形成於微晶矽 層上。一多晶矽層作為控制閘形成於内多晶介電層上。最 後,對多晶矽層、内多晶介電層、微晶矽層及穿遂氧化層 進行蝕刻而形成一閘極構成一記憶晶胞。 5 - 4發明詳細說明: 本發明將配合圖示說明詳細揭露如下所述,本實施例中提 供一種利用微晶石夕膜(m i c r 〇 c r y s t a 1 1 i n e ρ ο 1 y s i 1 i c ο η f i 1 m)作為浮置閘,以促進快閃式記憶體性能之方法。 第四圖所示之記憶晶胞1 0 0截面圖,其中以微晶矽膜 形成之浮置閘1 2 0取代先前技術之多晶矽膜浮置閘。記憶 晶胞1 0 0包含一源極1 5 0區域和一汲極1 6 0區域,源極1 5 0和 汲極1 6 0以一通道1 7 0區域加以隔離開。記憶晶胞1 0 0更包Page 7 587282 V. Description of the invention (5) These problems are solved. Therefore, low barriers promote the escape of electrons or removal by penetration. The lower voltage required for low barriers, similar to the tunneling distance, acts as an electronic wipe, which results in faster wipe time and reduced damage. In the present invention, a microcrystalline silicon floating gate is used to improve the performance of a flash memory. The method is to provide a substrate, wherein the substrate includes a source region and a drain region, and is separated by a channel region. A tunneling oxide layer is formed on the substrate. A microcrystalline silicon layer is formed on the tunneling oxide layer as a floating gate. The microcrystalline silicon film is formed by using a low pressure chemical vapor deposition method, selecting silane or disilane, and introducing hydrogen gas at the same time. An inner polycrystalline dielectric layer is formed on the microcrystalline silicon layer. A polycrystalline silicon layer is formed on the inner polycrystalline dielectric layer as a control gate. Finally, the polycrystalline silicon layer, the inner polycrystalline dielectric layer, the microcrystalline silicon layer, and the tunneling oxide layer are etched to form a gate to form a memory cell. 5-4 Detailed description of the invention: The present invention will be disclosed in detail with the illustrations as follows. In this embodiment, a microcrystalline stone evening film (micr 〇crysta 1 1 ine ρ ο 1 ysi 1 ic ο η fi 1 m ) As a floating gate to promote the performance of flash memory. A cross-sectional view of the memory cell 100 shown in the fourth figure, wherein the floating gate 120 formed by a microcrystalline silicon film replaces the floating gate of the polycrystalline silicon film of the prior art. The memory cell 100 includes a source region 150 and a drain region 160, and the source 150 and the drain region 160 are separated by a channel 170 region. Memory cell 1 0 0 more pack

587282 五、發明說明(6) 括一浮置閘1 2 0其以微晶矽膜形成、一控制閘1 4 0其以一多 晶矽層形成,浮置閘1 2 0和控制閘1 4 0則以一内多晶介電層 13 0如氧-氮-氧層(ONO layer)加以隔離,浮置閘120和 通道1 7 0區域則利用一閘介電層1 1 0如氧化矽加以分開,内 多晶介電層1 3 0和閘介電層1 1 0為絕緣層,閘介電層功能作 為電子穿遂氧化層。 · 第五圖所示為本實施例中利用微晶矽膜1 7 4作為浮置 閘截面放大圖,微晶矽膜1 7 4形成於一底材1 7 2上,微晶矽 膜1 7 4利用氫氣導入矽沉積反應室中控制膜的形成。本實 施例中,利用矽烷為材質並同時導入氫氣沉積微晶矽膜, 另一實施例為利用乙矽烷為材質並同時導入氫氣沉積微晶 矽膜。微晶矽膜1 7 4晶粒的尺寸平均約為5 0 0埃到1 0 0 0埃之 間,微晶碎膜1 7 4的晶粒大小乃以晶粒之平均直徑定義之 雖然晶粒無法具有均一性的外形,但具有平均直徑約 5 0 0埃到1 0 0 0埃。低壓化學氣相沉積方法或其他之沉積方 法皆可應用於此技術沉積薄膜。本實施例中,利用低壓化 學氣相沉積法之參數為:壓力控制約2 0 0〜4 0 0托、溫度控 制約7 0 0〜7 5 0°C、氫氣和氮氣的流量比約為5〜6 0 % ( 5〜6 0 % 氫氣/1 0 0% (氫氣+氮氣))。浮置閘可以經由相同於傳 統元件之多晶矽浮置閘形成方法中的圖案轉移和蝕刻方式 等製程來形成。587282 V. Description of the invention (6) Including a floating gate 1 2 0 formed by a microcrystalline silicon film, a control gate 1 4 0 formed by a polycrystalline silicon layer, a floating gate 1 2 0 and a control gate 1 4 0 An internal polycrystalline dielectric layer 130 such as an oxygen-nitrogen-oxygen layer (ONO layer) is used for isolation, and the area of the floating gate 120 and the channel 170 is separated by a gate dielectric layer 110 such as silicon oxide. The inner polycrystalline dielectric layer 130 and the gate dielectric layer 110 are insulating layers, and the gate dielectric layer functions as an electron tunneling oxide layer. · The fifth figure shows an enlarged view of the cross section of the floating gate using the microcrystalline silicon film 1 7 4 in this embodiment. The microcrystalline silicon film 1 7 4 is formed on a substrate 1 7 2 and the microcrystalline silicon film 1 7 4 Control the film formation by introducing hydrogen into the silicon deposition reaction chamber. In this embodiment, a crystalline silicon film is deposited by using silane as a material and hydrogen is introduced simultaneously, and another embodiment is used to deposit a microcrystalline silicon film by using silane as a material and simultaneously introduced hydrogen. The average grain size of microcrystalline silicon film 174 is about 500 angstroms to 100 angstroms. The grain size of microcrystalline broken film 174 is defined by the average diameter of the crystal grains. It cannot have a uniform shape, but has an average diameter of about 500 Angstroms to 100 Angstroms. Low pressure chemical vapor deposition methods or other deposition methods can be applied to this technology to deposit thin films. In this embodiment, the parameters using the low-pressure chemical vapor deposition method are: pressure control of about 200 ~ 4 0 0 Torr, temperature control of about 7 0 ~ 7 5 0 ° C, and the flow ratio of hydrogen and nitrogen is about 5 ~ 60% (5 ~ 60% hydrogen / 1 0% (hydrogen + nitrogen)). The floating gate can be formed by a process such as pattern transfer and etching in the method for forming a polycrystalline silicon floating gate similar to a conventional device.

587282 五、發明說明(7) 主要的反應方程式為SiH4_5 多晶矽膜中,矽原子高的表面擴 1 2 + 2H2。在傳 導致大的晶粒和成長速率。對二、率勝過於 '玄的 響成核速率。因此,當成核速=成長速率,』:二具 尺寸的晶粒形成。 々达擴散逮率將產生較了 注意本實施例與第二圖之 石夕膜"4中顯著的降低晶粒尺寸。摻質: = := ,列Ϊ第如Μ的晶界密度明顯大於多晶㈣ 所仏被 多晶矽膜的晶界密度,微晶矽膜1 74中摻 二的,散可得到改善,摻質滲入接近微晶矽膜1 74的表面 可Γ夺易沿著高密度的晶界穿透微晶矽膜17 4。結果,穿 =微晶矽膜1 74的摻質濃度,促進微晶矽膜1 74的導電度, 其比第二圖所示之再結晶非晶矽膜更加具電場均勻性。 曰 另一方面,與第二圖所示之傳統的多晶矽膜相比,微 晶矽膜1 74降低平均晶粒尺寸,更促進微晶膜1 74表面的平 坦化。微晶矽膜1 7 4的晶粒尺寸小型化於本發明中顯而易 見的是對膜表面的影響。除此之外,當微晶矽膜1 74製程 茶數加以調整增加或降低晶粒平均尺寸,膜的表面不規則 性也分別增加或降低。藉由微晶矽膜i 74表面的平坦化, 第10頁 587282 五、發明說明(8) 在微影製程中光阻的内部厚度的偏差也相對降低,與第二 圖所示之傳統的多晶矽膜相比,其反射達到更均勻性,改 善閘極邊緣的解析度。 最後可藉由”氧化縫隙’’(ο X i d e v a 1 1 e y)在穿遂氧化 層180、180A中說明,如第六A圖、第六B圖所示。在傳統 製程中,高濃度磷摻雜在氧化矽區域,氧化縫隙1 9 4形成 於多晶矽晶粒1 9 2介面。所有的多晶矽晶粒1 9 2相對為氧化 縫隙1 9 4的總長度。當使用大尺寸晶粒的多晶矽時,只有 少數的晶粒存在於一個抹除單位,舉例說明,當只有五個 多晶矽晶粒存在於一個抹除單位,其每一個多晶矽晶粒 1 9 2分別佔2 0%的抹除動作,因此一個多晶矽晶粒1 9 2在抹 除速度上有較大誤差值,並且有較寬的抹除起始電壓分佈 其次,本實施例中如第六B圖所示,眾多數的氧化縫 隙1 9 4 A可確實達到在每一記憶陣列之晶胞具有相同的抹除 速度。另一方面,使用小晶粒1 9 2 A的微晶矽,有眾多的晶 粒1 9 2 A存在於一個抹除單位内。舉例說明,當五十個微晶 矽晶粒1 9 2 A存在一個抹除單位内,每一個微晶矽晶粒1 9 2 A 的抹除動作僅有佔2% 。所以,每一個晶粒1 9 2 A具有相同 的抹除速度,可達到較窄的抹除起始電壓分佈範圍大的抹 除區域含有多數量的微晶矽晶粒1 9 2 A,可獲得均勻性的抹 除速度和窄範圍抹除起始電壓分佈。587282 V. Description of the invention (7) The main reaction equation is SiH4_5 polycrystalline silicon film, the surface area of which is high with silicon atoms is 1 2 + 2H2. This leads to large grains and growth rates. For two, the rate is better than the mysterious nucleation rate. Therefore, when the nucleation rate = growth rate, 』: two sized grains are formed. The Trent diffusion rate will result in a noticeable reduction in grain size in the Shi Xi film " 4 of this embodiment and the second figure. Doping: =: =, the grain boundary density of column M is significantly higher than that of polycrystalline silicon. The grain boundary density of polycrystalline silicon film is doped with two, and the dispersion can be improved. The surface close to the microcrystalline silicon film 1 74 can easily penetrate the microcrystalline silicon film 17 4 along the high-density grain boundaries. As a result, the dopant concentration of the microcrystalline silicon film 1 74 promotes the conductivity of the microcrystalline silicon film 1 74 and has more electric field uniformity than the recrystallized amorphous silicon film shown in the second figure. On the other hand, compared with the conventional polycrystalline silicon film shown in the second figure, the microcrystalline silicon film 1 74 reduces the average grain size and promotes the flattening of the surface of the microcrystalline film 1 74. The miniaturization of the grain size of the microcrystalline silicon film 174 is apparent in the present invention, and the effect on the film surface is obvious. In addition, when the microcrystalline silicon film 174 process is adjusted to increase or decrease the average grain size, the surface irregularities of the film also increase or decrease, respectively. By flattening the surface of the microcrystalline silicon film i 74, p. 10 587282 V. Description of the invention (8) The deviation of the internal thickness of the photoresist in the lithography process is also relatively reduced, which is similar to the traditional polycrystalline silicon shown in the second figure Compared with the film, its reflection achieves more uniformity, improving the resolution of the gate edge. Finally, it can be described in the tunneling oxide layers 180 and 180A by "oxidation gap" (ο X ideva 1 1 ey), as shown in Figures 6A and 6B. In the traditional process, high-concentration phosphorus is doped Miscellaneous in the silicon oxide region, the oxidation gap 194 is formed at the interface of the polycrystalline silicon grains 192. All polycrystalline silicon grains 192 are relative to the total length of the oxidized gaps 192. When using polycrystalline silicon with large grain sizes, Only a few grains exist in one erasing unit. For example, when there are only five polycrystalline silicon grains in one erasing unit, each of the polycrystalline silicon grains 192 accounts for 20% of the erasing action, so one The polycrystalline silicon grain 1 9 2 has a large error value in the erasing speed, and has a wide erase starting voltage distribution. Second, in this embodiment, as shown in FIG. 6B, a large number of oxidation gaps 1 9 4 A can indeed achieve the same erasure speed in the unit cell of each memory array. On the other hand, using microcrystalline silicon with small grains of 192 A, there are many grains of 192 A in one erasure. In the unit, for example, when fifty microcrystalline silicon grains 1 9 2 A exist one In the erasing unit, the erasing action of each microcrystalline silicon grain 192 A is only 2%. Therefore, each grain of 192 A has the same erasing speed and can achieve a narrow erasing. The erasing area with a large starting voltage distribution range contains a large number of microcrystalline silicon grains 192 A, and a uniform erasing speed and a narrow-range erasing starting voltage distribution can be obtained.

587282 五、發明說明(9) 閘極電壓直接決定閘極電流的強度,和此閘極電流是 最主要影響程序中中穿遂氧化層介電值衰退致使崩潰。本 實施例中,分別使用微晶矽浮置閘和先前技術的多晶矽浮 置閘,測試穿遂氧化層崩潰電壓(charge-to breakdown, Qbd)和電子捕捉率(trapping rate)。如表一所示,其 使用微晶矽和多晶矽的穿遂氧化層的厚度利用CV方式和FV 方式測得厚度均相同,且表中每一個數據均以每一片晶圓 測試三次所得平均值。 如第七圖所示,A柱為使用微晶矽浮置閘之穿遂氧化 層,B柱為使用多晶矽浮置閘之穿遂氧化層,圖中可看出 本發明使用微晶矽浮置閘其穿遂氧化層具有較高的電子注 入效率和較大的崩潰電壓值。另外,如第八圖所示,C柱 為使用微晶矽浮置閘之穿遂氧化層,D柱為使用多晶矽浮 置閘之穿遂氧化層,圖中可看出本發明使用微晶矽浮置閘 其穿遂氧化層具有較低電子捕捉率,可獲得較佳的快閃式 記憶體之元件可靠度。使用較小尺寸的微晶矽,具有多數 的晶粒存在於浮置閘,利用小尺寸晶粒的浮置閘可降低穿 遂氧化層的電子捕捉率,因此,可促進快閃記憶體的功能 性。 以上所述僅為本發明之較佳實施例而已,並非用以限 定本發明之申請專利範圍;凡其它未脫離本發明所揭示之587282 V. Description of the invention (9) The gate voltage directly determines the strength of the gate current, and this gate current is the most important factor that affects the breakdown of the dielectric value of the tunneling oxide layer in the program and causes collapse. In this embodiment, a microcrystalline silicon floating gate and a prior art polycrystalline silicon floating gate are respectively used to test charge-to breakdown (Qbd) and electron trapping rate. As shown in Table 1, the thickness of the tunneling oxide layer using microcrystalline silicon and polycrystalline silicon is the same as measured by the CV method and the FV method, and each data in the table is an average value obtained by testing three times per wafer. As shown in the seventh figure, the pillar A is a tunneling oxide layer using a microcrystalline silicon floating gate, and the pillar B is a tunneling oxide layer using a polycrystalline silicon floating gate. As can be seen in the figure, the present invention uses a microcrystalline silicon floating gate. The gate tunneling oxide layer has a higher electron injection efficiency and a larger breakdown voltage value. In addition, as shown in the eighth figure, the C pillar is a tunneling oxide layer using a microcrystalline silicon floating gate, and the D pillar is a tunneling oxide layer using a polycrystalline silicon floating gate. It can be seen in the figure that the present invention uses a microcrystalline silicon The floating gate has a lower electron capture rate for its tunneling oxide layer, which can obtain better reliability of the flash memory device. The use of smaller-sized microcrystalline silicon has most of the grains present in the floating gate. Using the floating gate of small-sized grains can reduce the electron capture rate of the tunneling oxide layer, so it can promote the function of flash memory Sex. The above are only the preferred embodiments of the present invention, and are not intended to limit the scope of patent application of the present invention; all others do not depart from the disclosure of the present invention.

第12頁 587282 五、發明說明(10) 精神下所完成之等效改變或修飾’均應包含在下述之申請 專利範圍内。 - 表^一所示為使用微晶石夕和多晶石夕浮置閘’其各牙遂氧 化層之厚度、崩潰電壓值、電子捕捉率之測試數據。 曰m 膜 厚度(CV) 厚度(FV) 崩潰電壓 電子捕捉率 曰曰圓 (埃) (埃) 庫侖/平方公分 電子/十億 1 105.0 103.6 23.50 0.030 2 便用 104.2 103.6 23.02 0,033 3 微曰曰 Th 104.6 103,6 22.04 0.033 4 104,6 103.6 21.66 0,027 11 104.8 103.6 13.82 0.028 12 104,6 103.6 14,14 0.031 13 使 104,6 103.6 14.74 0.031 14 用 104,8 103.6 14.94 0.032 15 多 104,6 103.6 14.47 0.031 16 晶 104.4 103.6 15.12 0,035 17 石夕 104,8 103.6 16.34 0.037 18 膜 104,4 103.6 13.70 0.036 19 104.8 103,6 16.10 0.036 20 104.4 103,6 16,02 0,035 21 104.4 103.6 15.79 0.035Page 12 587282 V. Description of the invention (10) Equivalent changes or modifications under the spirit of the invention shall be included in the scope of the patent application as described below. -Table ^ 1 shows the test data of the thickness, breakdown voltage, and electron capture rate of the oxidized oxide layer of the microcrystalline and polycrystalline floating gates. Membrane thickness (CV) Thickness (FV) Electron capture rate of breakdown voltage 104.6 103,6 22.04 0.033 4 104,6 103.6 21.66 0,027 11 104.8 103.6 13.82 0.028 12 104,6 103.6 14,14 0.031 13 use 104,6 103.6 14.74 0.031 14 use 104,8 103.6 14.94 0.032 15 more 104,6 103.6 14.47 0.031 16 Crystal 104.4 103.6 15.12 0,035 17 Shi Xi 104,8 103.6 16.34 0.037 18 Film 104,4 103.6 13.70 0.036 19 104.8 103,6 16.10 0.036 20 104.4 103,6 16,02 0,035 21 104.4 103.6 15.79 0.035

第13頁 587282 圖式簡單說明 第一圖所示為先前技術之記憶晶胞之截面圖; 第二圖所示為使用多晶矽膜浮置閘於一底材上之截面 放大圖; 第三圖所示為使用再結晶矽膜浮置閘於底材上之截 面放大圖; 第四圖所示為本發明之記憶晶胞之截面圖,其中使用 微晶矽膜作為浮置閘。Page 587 282 Brief description of the diagram The first picture shows a cross-sectional view of a memory cell of the prior art; the second picture shows an enlarged cross-sectional view of a polysilicon film floating gate on a substrate; the third picture shows Shown is an enlarged cross-sectional view of a floating gate using a recrystallized silicon film on a substrate; the fourth figure is a cross-sectional view of a memory cell of the present invention, in which a microcrystalline silicon film is used as the floating gate.

第五圖所示為本發明使用微晶矽膜作為浮置閘於一底 材上之截面放大圖; 第六A圖所示為氧化縫隙之示意圖,圖中所示有大晶 粒而少氧化縫隙形成存在一個記憶晶胞中; 第六B圖所示為氧化縫隙之示意圖,圖中所示有小晶 粒而多氧化縫隙形成存在一個記憶晶胞中;The fifth figure is an enlarged cross-sectional view of the present invention using a microcrystalline silicon film as a floating gate on a substrate; the sixth diagram A shows a schematic diagram of an oxidation gap, showing large grains and less oxidation. Gap formation exists in a memory cell; Figure 6B shows a schematic diagram of an oxidation gap, with small grains shown in the figure and multiple oxidation gaps formed in a memory cell;

第七圖所示為使用微晶矽和多晶矽浮置閘,其穿遂氧 化層之崩潰電壓值比較。 第八圖所示為使用微晶矽和多晶矽浮置閘,其穿遂氧Figure 7 shows a comparison of the breakdown voltage of the tunneling oxide layer using microcrystalline and polycrystalline silicon floating gates. The eighth figure shows the use of microcrystalline and polycrystalline silicon floating gates.

第14頁 587282 圖式簡單說明 化層之電子捕捉率比較圖。 主要部分之代表符號: 5記憶晶胞 10底材 2 0穿遂氧化層 3 0多晶矽浮置閘 4 0内多晶介電層 5 0控制閘 6 0源極 7 0汲極 8 0通道 8 2底材 8 4柱狀多晶矽 8 6底材 8 8再結晶矽 1 0 0記憶晶胞 I 0 5底材 II 0穿遂氧化層 1 2 0微晶矽浮置閘 1 3 0内多晶介電層 1 4 0控制閘 1 5 0源極 1 6 0汲極Page 14 587282 The figure briefly illustrates the comparison diagram of the electron capture rate of the chemical layer. Representative symbols of main parts: 5 memory cell 10 substrate 2 0 tunneling oxide layer 3 0 polycrystalline silicon floating gate 4 0 polycrystalline dielectric layer 5 0 control gate 6 0 source 7 0 drain 8 0 channel 8 2 Substrate 8 4 Cylindrical polycrystalline silicon 8 6 Substrate 8 8 Recrystallized silicon 1 0 0 Memory cell I 0 5 Substrate II 0 Passive oxide layer 1 2 0 Microcrystalline silicon floating gate 1 3 0 Polycrystalline dielectric inside Layer 1 4 0 Control gate 1 5 0 Source 1 6 0 Drain

第15頁 587282 圖式簡單說明 17 0通道 1 7 2底材 1 7 4微晶矽 1 7 6底材 1 7 8源極/汲極 18 0穿遂氧化層 « 190多晶矽Page 15 587282 Simple illustration of the drawing 17 0 channel 1 7 2 substrate 1 7 4 microcrystalline silicon 1 7 6 substrate 1 7 8 source / drain 18 0 tunneling oxide layer «190 polycrystalline silicon

192多晶碎晶粒 1 9 4氧化縫隙 176A底材 1 7 8 A源極/汲極 1 8 0A穿遂氧化層 1 9 0 A微晶矽 1 9 2 A微晶碎晶粒 194A氧化縫隙 A柱使用微晶矽浮置閘之穿遂氧化層崩潰電壓 B柱使用多晶矽浮置閘之穿遂氧化層崩潰電壓 C柱使用微晶矽浮置閘之穿遂氧化層電子捕捉率 D柱使用多晶矽浮置閘之穿遂氧化層電子捕捉率192 polycrystalline grains 1 9 4 oxidation gaps 176A substrate 1 7 8 A source / drain 1 8 0A tunneling oxide layer 1 9 0 A microcrystalline silicon 1 9 2 A microcrystalline grains 194A oxidation gaps A The column uses microcrystalline silicon floating gate breakdown oxide breakdown voltage B column uses polycrystalline silicon floating gate breakdown oxide breakdown voltage C pillar uses microcrystalline silicon floating gate penetration oxide electron capture rate D pillar uses polycrystalline silicon Electron capture rate of the through oxide layer of the floating gate

第16頁Page 16

Claims (1)

587282 案號 91114467 _案號91114467_年 > 月q日 修正_ 六、申請專利範圍 1 . 一種形成一微晶矽層作為一浮置閘之方法,包含: 提供一底材,具有一閘介電層於該底材上; ,藉由分解矽烷並導入氩氣於該閘介電層上形成該微 晶矽層,以作為該浮置閘。 2.如申請專利範圍第1項之形成一微晶矽層作為一浮置閘 之方法,其中該方法包含低壓化學氣相沉積法。 3. 如申請專利範圍第2項之形成一微晶矽層作為一浮置閘 之方法,其中該方法包含溫度控制約7 0 0〜7 5 0°C。587282 Case No. 91114467 _ Case No. 91114467_year > Amendment on the qth of the month _ 6. Application for patent scope 1. A method for forming a microcrystalline silicon layer as a floating gate, comprising: providing a substrate with a gate An electric layer is formed on the substrate; the microcrystalline silicon layer is formed on the gate dielectric layer by decomposing silane and introducing argon gas, and the microcrystalline silicon layer is used as the floating gate. 2. A method of forming a microcrystalline silicon layer as a floating gate as described in the first patent application scope, wherein the method includes a low pressure chemical vapor deposition method. 3. The method of forming a microcrystalline silicon layer as a floating gate as described in item 2 of the scope of patent application, wherein the method includes temperature control of about 700 ~ 750 ° C. 4. 如申請專利範圍第2項之形成一微晶矽層作為一浮置閘 之方法,其中該方法包含壓力控制約2 0 0托到4 0 0托。 5.如申請專利範圍第1項之形成一微晶矽層作為一浮置閘 之方法,其中該方法包含氣體流量為氫氣/(氫氣+氮氣) 比值為5〜6 0 %。 6.如申請專利範圍第1項之形成一微晶矽層作為一浮置閘 之方法,其中該微晶矽物質可選擇包含矽烷和乙矽烷。 7. —種促進快閃記憶體性能的方法,包含: 形成一底材,其中該底材包含一源極區域和一汲極區 域並以一通道區域加以隔離;4. The method of forming a microcrystalline silicon layer as a floating gate as described in item 2 of the patent application scope, wherein the method includes pressure control of about 200 Torr to 400 Torr. 5. The method of forming a microcrystalline silicon layer as a floating gate according to item 1 of the patent application scope, wherein the method includes a gas flow rate of hydrogen / (hydrogen + nitrogen) ratio of 5 to 60%. 6. The method of forming a microcrystalline silicon layer as a floating gate according to item 1 of the scope of the patent application, wherein the microcrystalline silicon substance may optionally include silane and ethane. 7. A method for promoting flash memory performance, comprising: forming a substrate, wherein the substrate includes a source region and a drain region and is isolated by a channel region; 第17頁 587282 _案號91114467_年月曰 修正_ 六、申請專利範圍 形成一穿遂氧化層於該底材上; 分解矽烷並導入氫氣於該穿遂氧化層上,形成一微晶 矽層; 形成一内多晶介電層於該微晶石夕層上; 形成一多晶矽層於該内多晶介電層上;以及 蝕刻該多晶矽層、該内多晶介電層、該微晶矽層和該 穿遂氧化層以形成一閘極結構。 8. 如申請專利範圍第7項之促進快閃記憶體性能的方法, 其中上述該微晶矽的晶粒尺寸約5 0 0埃到1 0 0 0埃。Page 17 587282 _Case No. 91114467_Amended in January _ Sixth, the scope of the patent application forms a tunneling oxide layer on the substrate; decomposes silane and introduces hydrogen to the tunneling oxide layer to form a microcrystalline silicon layer Forming an inner polycrystalline dielectric layer on the microcrystalline silicon layer; forming a polycrystalline silicon layer on the inner polycrystalline dielectric layer; and etching the polycrystalline silicon layer, the inner polycrystalline dielectric layer, and the microcrystalline silicon Layer and the tunneling oxide layer to form a gate structure. 8. The method for promoting flash memory performance according to item 7 of the scope of patent application, wherein the crystallite size of the microcrystalline silicon is about 500 angstroms to 100 angstroms. 9. 如申請專利範圍第7項之促進快閃記憶體性能的方法, 其中上述該微晶矽層乃藉由低壓化學氣相沉積法形成。 1 0.如申請專利範圍第9項之促進快閃記憶體性能的方 法,其中上述該方法溫度控制約7 0 0〜7 5 0°C。 1 1.如申請專利範圍第9項之促進快閃記憶體性能的方 法,其中上述該方法壓力控制約2 0 0托到4 0 0托。9. The method for promoting flash memory performance according to item 7 of the patent application, wherein the microcrystalline silicon layer is formed by a low pressure chemical vapor deposition method. 10. The method for promoting the performance of flash memory according to item 9 of the scope of patent application, wherein the temperature of the method is about 700 to 75 ° C. 1 1. The method for promoting the performance of flash memory according to item 9 of the scope of patent application, wherein the method described above is pressure-controlled from about 200 Torr to 400 Torr. 1 2.如申請專利範圍第9項之促進快閃記憶體性能的方 法,其中該方法包含氣體流量為氫氣/(氫氣+氮氣)比值 為5〜60%〇1 2. The method for promoting flash memory performance according to item 9 of the scope of patent application, wherein the method includes a gas flow rate of hydrogen / (hydrogen + nitrogen) ratio of 5 to 60%. 第18頁 587282 _案號91114467_年月曰 修正_ 六、申請專利範圍 1 3 .如申請專利範圍第7項之促進快閃記憶體性能的方 法,其中該微晶矽物質可選擇包含矽烷和乙矽烷。 1 4.如申請專利範圍第7項之促進快閃記憶體性能的方 法,其中上述該微晶矽層為浮置閘。 ~ 1 5.如申請專利範圍第7項之促進快閃記憶體性能的方 法,其中上述該多晶矽層為控制閘。Page 18 587282 _ Case No. 91114467 _ month and month amend _ six, the scope of the patent application 1 3, such as the scope of the patent application for the seventh method to promote flash memory performance, wherein the microcrystalline silicon material can optionally contain silane and Ethylsilane. 1 4. The method for promoting flash memory performance according to item 7 of the scope of patent application, wherein the microcrystalline silicon layer is a floating gate. ~ 1 5. The method for promoting flash memory performance according to item 7 of the scope of the patent application, wherein the polycrystalline silicon layer is a control gate. 第19頁Page 19
TW91114467A 2002-06-28 2002-06-28 Method for improving the performance of flash memory by using microcrystalline polysilicon film as a floating gate TW587282B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW91114467A TW587282B (en) 2002-06-28 2002-06-28 Method for improving the performance of flash memory by using microcrystalline polysilicon film as a floating gate

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW91114467A TW587282B (en) 2002-06-28 2002-06-28 Method for improving the performance of flash memory by using microcrystalline polysilicon film as a floating gate

Publications (1)

Publication Number Publication Date
TW587282B true TW587282B (en) 2004-05-11

Family

ID=34057760

Family Applications (1)

Application Number Title Priority Date Filing Date
TW91114467A TW587282B (en) 2002-06-28 2002-06-28 Method for improving the performance of flash memory by using microcrystalline polysilicon film as a floating gate

Country Status (1)

Country Link
TW (1) TW587282B (en)

Similar Documents

Publication Publication Date Title
US6455372B1 (en) Nucleation for improved flash erase characteristics
JP4907815B2 (en) Manufacturing method of ONO floating gate electrode in 2-bit EEPROM device
CN102208442B (en) Gate stack structure suitable for semiconductor flash memory device and manufacturing method of gate stack structure
JPH10261725A (en) Semiconductor device and its manufacture
JPH1140682A (en) Non-volatile semiconductor memory and its manufacture
TWI376773B (en) Method for manufacturing non-volatile memory and structure threrof
US20040113196A1 (en) Method and structure for an oxide layer overlying an oxidation-resistant layer
KR100636022B1 (en) Method for forming a thin film in semiconductor device and manufacturing a non-volatile memory device using the same
JP4792620B2 (en) Nonvolatile semiconductor memory device and manufacturing method thereof
US6461984B1 (en) Semiconductor device using N2O plasma oxide and a method of fabricating the same
CN110098323A (en) A kind of low-power consumption silicon nitride tunnel junctions resistance-variable storing device and preparation method thereof
TW200404339A (en) Method for forming a protective buffer layer for high temperature oxide processing
CN101312212A (en) Non-volatile memory utilizing high K medium and nanocrystalline floating gate and its manufacture method
TW587282B (en) Method for improving the performance of flash memory by using microcrystalline polysilicon film as a floating gate
US20060205205A1 (en) Method for improving the performance of flash memory by using microcrystalline silicon film as a floating gate
US20070249121A1 (en) Method of fabricating non-volatile memory
CN101399289A (en) Nanocrystalline floating gate non-vaporability memory with double layer tunneling medium structure and manufacturing method
TWI282149B (en) Method of forming a floating gate in a flash memory device
WO2006095890A1 (en) Semiconductor device and method for manufacturing same
JPH0766305A (en) Nonvolatile semiconductor storage device
JPH09223752A (en) Manufacture of nonvolatile semiconductor storage device
CN100349283C (en) Method for promoting performance of flash memory by using microcrystalline silicon film as floating gate
CN110165047A (en) A kind of high density nano-silicon resistance-variable storing device and preparation method thereof
JP4224000B2 (en) Manufacturing method of semiconductor device
CN100511714C (en) Method for preparing non-volatile floating bar memory based on germanium and silicon heterogeneous nano structure

Legal Events

Date Code Title Description
MK4A Expiration of patent term of an invention patent