TWI489593B - Hot carrier programming of nand flash memory - Google Patents

Hot carrier programming of nand flash memory Download PDF

Info

Publication number
TWI489593B
TWI489593B TW100114082A TW100114082A TWI489593B TW I489593 B TWI489593 B TW I489593B TW 100114082 A TW100114082 A TW 100114082A TW 100114082 A TW100114082 A TW 100114082A TW I489593 B TWI489593 B TW I489593B
Authority
TW
Taiwan
Prior art keywords
voltage
memory
memory cells
series
memory cell
Prior art date
Application number
TW100114082A
Other languages
Chinese (zh)
Other versions
TW201244017A (en
Inventor
Shaw Hung Ku
I Chen 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 TW100114082A priority Critical patent/TWI489593B/en
Publication of TW201244017A publication Critical patent/TW201244017A/en
Application granted granted Critical
Publication of TWI489593B publication Critical patent/TWI489593B/en

Links

Landscapes

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

Description

反及閘快閃記憶體之熱載子程式化Hot-loading stylization of the gate flash memory

本發明係關於快閃記憶體技術,特別是關於在反及閘組態中合適作為低電壓程式化及抹除操作的快閃記憶體。This invention relates to flash memory technology, and more particularly to flash memory suitable for low voltage stylization and erase operations in an anti-gate configuration.

快閃記憶體是非揮發積體電路記憶體技術的一類。傳統的快閃記憶體使用浮動閘極記憶胞。隨著記憶裝置之密度提升,浮動閘極記憶胞之間逾加靠近,儲存在相鄰浮動閘極中的電荷交互影響即造成問題,因此形成限制,使得採用浮動閘極之快閃記憶體密度無法提升。另一種快閃記憶體所使用之記憶胞稱為電荷捕捉記憶胞,其採用電荷捕捉層取代浮動閘極。電荷捕捉記憶胞係利用電荷捕捉材料,不會如浮動閘極造成個別記憶胞之間的相互影響,並且可以應用於高密度的快閃記憶體。Flash memory is a class of non-volatile integrated circuit memory technology. Traditional flash memory uses floating gate memory cells. As the density of the memory device increases, the floating gate memory cells are closer together, and the charge interaction stored in the adjacent floating gates causes a problem, thus forming a limitation, so that the floating gate density of the floating gate is used. Unable to upgrade. Another type of memory cell used in flash memory is called a charge trapping memory cell, which uses a charge trapping layer instead of a floating gate. The charge trapping memory cell utilizes a charge trapping material that does not cause interaction between individual memory cells as a floating gate, and can be applied to high density flash memory.

典型的電荷儲存記憶胞包含一場效電晶體(FET)結構,其中包含由通道所分隔之源極與汲極,以及藉由一電荷儲存結構而與通道分離的閘極,其中該電荷儲存結構包含穿隧介電層、電荷儲存層(浮動閘極或介電層)、與阻障介電層。較早的傳統設計如SONOS裝置,其中源極、汲極與通道形成於矽基材(S)上,穿隧介電層則由氧化矽(O)之上,電荷儲存層由氮化矽形成(N),阻障介電層由氧化矽(O)形成,而閘極則為多晶矽(S)。A typical charge storage memory cell includes a field effect transistor (FET) structure including a source and a drain separated by a channel, and a gate separated from the channel by a charge storage structure, wherein the charge storage structure comprises A tunneling dielectric layer, a charge storage layer (floating gate or dielectric layer), and a barrier dielectric layer. Earlier conventional designs, such as SONOS devices, in which the source, drain and channel are formed on the germanium substrate (S), the tunneling dielectric layer is over the germanium oxide (O), and the charge storage layer is formed of tantalum nitride. (N), the barrier dielectric layer is formed of ruthenium oxide (O), and the gate is polycrystalline germanium (S).

快閃記憶體裝置通常可以使用反及閘(NAND)或是反或閘(NOR)架構來施作,但也可以是其他的架構,包括及閘(AND)架構。此反及閘(NAND)架構特別因為其在資料儲存應用方面的高密度及高速的優點而受到青睞。而反或閘(NOR)架構則是適合於例如是程式法儲存等其他應用上,因為隨機存取是重要的功能需求。在一反及閘(NAND)架構中,程式化過程通常是依賴富勒-諾得漢(FN)穿隧,且需要高電壓,通常是在20伏特數量級,且需要高電壓電晶體來處理。此額外的高電壓電晶體及搭配使用於邏輯和其他資料流的電晶體於同一積體電路中,會造成製程的複雜性增加。如此則會增加此裝置的製造成本。Flash memory devices can typically be implemented using either a NAND or a NOR architecture, but can be other architectures, including an AND architecture. This NAND architecture is favored especially for its high density and high speed advantages in data storage applications. The inverse OR gate (NOR) architecture is suitable for other applications such as program storage, because random access is an important functional requirement. In a NAND architecture, the stylization process typically relies on Fuller-Nordheim (FN) tunneling and requires high voltages, typically on the order of 20 volts, and requires high voltage transistors for processing. This extra high-voltage transistor and the transistors used in logic and other data streams are in the same integrated circuit, which increases the complexity of the process. This will increase the manufacturing cost of the device.

因此,需要提供一種新的記憶體技術,其可以在反及閘(NAND)架構中利用低電壓即可程式化操作。Therefore, there is a need to provide a new memory technology that can be programmed with low voltage in a NAND architecture.

本發明係關於記憶裝置及其操作方法。根據本發明之一目的,揭露一種記憶體,包含複數個記憶胞串聯安排於一半導體主體中;複數條字元線,該複數條字元線中的每一條字元線與該複數個記憶胞中對應的記憶胞耦接;以及與該複數條字元線耦接的控制電路。該控制電路適用於程式化該複數個記憶胞中與一選取字元線對應的一選取記憶胞。該程式化藉由偏壓該複數個記憶胞的一第一及一第二端之一者至一設定電壓;降低施加至該複數個記憶胞的該第一及第二端之該一者的電壓階級自該設定電壓至一位元線程式化電壓;施加一導通電壓至與未選取記憶胞所對應的字元線;以及施加一程式化電壓至與該選取記憶胞所對應的該選取字元線。The present invention relates to a memory device and a method of operating the same. According to one aspect of the present invention, a memory includes a plurality of memory cells arranged in series in a semiconductor body; a plurality of word lines, each of the plurality of word lines and the plurality of memory cells a corresponding memory cell coupling; and a control circuit coupled to the plurality of word lines. The control circuit is adapted to program a selected memory cell corresponding to a selected word line in the plurality of memory cells. The programming is performed by biasing one of the first and second ends of the plurality of memory cells to a set voltage; reducing the one of the first and second ends applied to the plurality of memory cells The voltage level is from the set voltage to the one-bit threaded voltage; applying a turn-on voltage to the word line corresponding to the unselected memory cell; and applying a stylized voltage to the selected word corresponding to the selected memory cell Yuan line.

該半導體主體包含一淡摻雜基板區域。該淡摻雜基板區域的摻雜濃度小於或等於5x1012 cm-2 。該淡摻雜基板區域包括一N-型態摻雜區域。The semiconductor body includes a lightly doped substrate region. The doping concentration of the lightly doped substrate region is less than or equal to 5 x 10 12 cm -2 . The lightly doped substrate region includes an N-type doped region.

每一個記憶胞包括一各自的電荷捕捉結構。該電荷捕捉結構可以形成於一淡摻雜基板區域之上。該電荷捕捉結構包括各自的通道氧化層,每一個通道氧化層的厚度小於90埃。Each memory cell includes a respective charge trapping structure. The charge trapping structure can be formed over a lightly doped substrate region. The charge trapping structure includes respective channel oxide layers, each channel having a thickness of less than 90 angstroms.

施加至該選取字元線的該程式化電壓小於或等於17伏特。該導通電壓係在3~8伏特範圍間。The stylized voltage applied to the selected word line is less than or equal to 17 volts. The turn-on voltage is in the range of 3 to 8 volts.

施加該設定電壓導致該半導體主體中的反轉。Applying the set voltage causes an inversion in the semiconductor body.

偏壓該複數個記憶胞的第一及一第二端之一者的步驟是在一第一時間區間內進行,且其中降低該電壓階級、施加該導通電壓以及施加該程式化電壓的步驟是在該第一時間區間後的一第二時間區間內進行。The step of biasing one of the first and second ends of the plurality of memory cells is performed in a first time interval, and wherein the step of lowering the voltage level, applying the turn-on voltage, and applying the stylized voltage is It is performed in a second time interval after the first time interval.

進行偏壓該複數個記憶胞的第一及一第二端之一者的同時施加一接地電壓階級至該複數個記憶胞的第一及一第二端之另一者以及至該複數條字元線中的每一條。Performing biasing one of the first and second ends of the plurality of memory cells while applying a ground voltage level to the other of the first and second ends of the plurality of memory cells and to the plurality of words Each of the lines.

根據本發明之另一目的,一種記憶體,包含一第一串列串聯安排於一半導體主體中;一第二串列串聯安排於該半導體主體中;複數條字元線,該複數條字元線中的每一條字元線與該複數個記憶胞中各自的該第一串列記憶胞之一及該第二串列記憶胞之一耦接;以及與該複數條字元線耦接的控制電路。該控制電路適用於程式化該第一串列中與一選取字元線對應的一選取記憶胞。此程式化可以藉由以下步驟達成:施加一位元線程式化電壓至該第一串列記憶胞中的一第一及一第二端之一者;維持該第二串列記憶胞中的該第一及第二端兩者在該接地階級電壓;施加一導通電壓至與未選取記憶胞所對應的字元線;以及施加一程式化電壓至與該選取記憶胞所對應的該選取字元線。According to another aspect of the present invention, a memory includes a first series arranged in series in a semiconductor body; a second series arranged in series in the semiconductor body; a plurality of word lines, the plurality of characters Each of the word lines in the line is coupled to one of the first series of memory cells and one of the second series of memory cells of the plurality of memory cells; and coupled to the plurality of word lines Control circuit. The control circuit is adapted to program a selected memory cell corresponding to a selected word line in the first string. The stylization can be achieved by applying a one-bit threaded voltage to one of the first and second ends of the first series of memory cells; maintaining the second string of memory cells The first and second ends are at the ground class voltage; applying a turn-on voltage to a word line corresponding to the unselected memory cell; and applying a stylized voltage to the selected word corresponding to the selected memory cell Yuan line.

該半導體主體包含一淡摻雜基板區域。該淡摻雜基板區域的摻雜濃度小於或等於5x1012 cm-2 。該淡摻雜基板區域包括一N-型態摻雜區域。The semiconductor body includes a lightly doped substrate region. The doping concentration of the lightly doped substrate region is less than or equal to 5 x 10 12 cm -2 . The lightly doped substrate region includes an N-type doped region.

每一個記憶胞包括一各自的電荷捕捉結構。該電荷捕捉結構可以形成於一淡摻雜基板區域之上。該電荷捕捉結構包括各自的通道氧化層,每一個通道氧化層的厚度小於90埃。Each memory cell includes a respective charge trapping structure. The charge trapping structure can be formed over a lightly doped substrate region. The charge trapping structure includes respective channel oxide layers, each channel having a thickness of less than 90 angstroms.

施加至該選取字元線的該程式化電壓小於或等於17伏特。該導通電壓係在3~8伏特範圍間。The stylized voltage applied to the selected word line is less than or equal to 17 volts. The turn-on voltage is in the range of 3 to 8 volts.

施加該設定電壓導致該半導體主體中的反轉。Applying the set voltage causes an inversion in the semiconductor body.

該控制電路更進一步組態為,於一第一時間區間內偏壓該第一串列記憶胞中的第一及一第二端之一者至一設定電壓,並且同時施加一接地電壓階級至該第一串列記憶胞中的第一及一第二端之另一者、該複數條字元線中的每一條、以及該第二串列記憶胞中的第一及第二端兩者。施加該位元線程式化電壓、維持該第二串列記憶胞中的該第一及第二端兩者在該接地階級電壓、施加該導通電壓以及施加該程式化電壓的步驟皆是在該第一時間區間後的一第二時間區間內進行。The control circuit is further configured to bias one of the first and second terminals of the first series of memory cells to a set voltage in a first time interval, and simultaneously apply a ground voltage level to The other of the first and second ends of the first series of memory cells, each of the plurality of word line lines, and the first and second ends of the second series of memory cells . Applying the bit threading voltage, maintaining both the first and second ends of the second serial memory cell at the ground level voltage, applying the turn-on voltage, and applying the stylized voltage are all The second time interval after the first time interval is performed.

施加該位元線程式化電壓包括降低施加至該第一串列記憶胞中的該第一及第二端之該一者的電壓階級自該設定電壓至該位元線程式化電壓。Applying the bit to thread the voltage includes decreasing a voltage level of the one of the first and second ends applied to the first series of memory cells from the set voltage to the bit threaded voltage.

本發明以下的實施例描述係搭配圖式1到8進行說明。The following description of the embodiments of the present invention is described in conjunction with Figures 1 through 8.

第1A和1B圖分別顯示複數個電荷捕捉快閃記憶胞串聯在一起成為反及閘串列的剖面示意圖,及進行FN穿隧程式化的偏壓示意圖,其是在反及閘快閃記憶體架構中的典型操作。第1C圖顯示第1A和1B圖中所示反及閘串列的簡要示意圖。Figures 1A and 1B respectively show a cross-sectional view of a plurality of charge-trapping flash memory cells connected in series to form an anti-gate sequence, and a bias diagram for performing FN tunneling stylization, which is in the inverse gate flash memory. Typical operations in the architecture. Figure 1C shows a schematic diagram of the inverted gate train shown in Figures 1A and 1B.

第1A圖顯示對一包括目標記憶胞(第1C圖中的記憶胞A)於一選取位元線上的反及閘串列之偏壓示意圖,而第1B圖顯示對一位於未選取位元線上的反及閘串列之偏壓示意圖。使用能隙工程SONOS電荷捕捉技術以實施反及閘快閃記憶體的一技術可參閱Lue之美國專利第7315474號,其在此引為參考資料。反及閘串列可以使用許多不同的組態實施,包括鰭形場效電晶體技術、淺溝渠隔離技術、垂直反及閘技術等等。某些垂直反及閘結構的範例,請參閱Kim等人標題為"Non-volatile memory device,method of operating same and method of fabricating the same"的歐洲專利第EP 2048709號。FIG. 1A shows a bias diagram of a reverse-to-gate sequence including a target memory cell (memory cell A in FIG. 1C) on a selected bit line, and FIG. 1B shows that the pair is located on an unselected bit line. The reverse bias diagram of the gate series. A technique for implementing a backlash flash memory using an energy gap engineering SONOS charge trapping technique can be found in U.S. Patent No. 7,315,474, issued toU.S. The reverse gate series can be implemented using a number of different configurations, including fin field effect transistor technology, shallow trench isolation technology, vertical reverse gate technology, and the like. For an example of certain vertical reversal gate structures, see European Patent No. EP 2048709 to Kim et al., entitled "Non-volatile memory device, method of operating same and method of fabricating the same."

請參閱第1A圖,此記憶胞形成於一半導體主體10中。對n通道記憶胞而言,半導體主體10可以是一個於半導體晶片的更深的n井內之隔離的p井。替代地,半導體主體10可以由絕緣層或是其他類似地方式隔離。某些實施例可以使用p通道記憶胞,其中半導體主體10中的摻雜是n型。Referring to FIG. 1A, the memory cell is formed in a semiconductor body 10. For n-channel memory cells, semiconductor body 10 can be an isolated p-well within a deeper n-well of a semiconductor wafer. Alternatively, the semiconductor body 10 can be isolated by an insulating layer or other similar means. Some embodiments may use p-channel memory cells in which the doping in the semiconductor body 10 is n-type.

複數個快閃記憶胞可以安排成沿著一個與字元線方向正交的位元線方向排列之串列。字元線22-27沿伸通過一些平行的反及閘串列。節點12-18是由半導體主體中的n型區域(對n通道裝置而言),且作為記憶胞的源/汲極區域。一個由金氧半電晶體形成的第一切換開關具有一閘極於接地選擇線GSL 21中,其連接於具有第一字元線22(第1C圖中的WL0)的對應記憶胞與由半導體主體10中的n型區域形成之一接點11之間。此接點11與共同源極線CS 30連接。一個由金氧半電晶體形成的第二切換開關具有一閘極於串列選擇線SSL 28中,其連接於具有最後字元線27的對應記憶胞與由半導體主體10中的n型區域形成之一接點19之間。此接點19與位元線BL 31連接。在此例示實施例中的第一及第二切換開關是金氧半電晶體,此範例中具有二氧化矽的閘介電層7和8。The plurality of flash memory cells can be arranged in a series of bit line directions that are orthogonal to the direction of the word line. The word lines 22-27 extend through a number of parallel anti-gate trains. Nodes 12-18 are n-type regions (for n-channel devices) in the semiconductor body and serve as source/drain regions for the memory cells. A first switch formed by a MOS transistor has a gate in the ground select line GSL 21 coupled to a corresponding memory cell having a first word line 22 (WL0 in FIG. 1C) and a semiconductor The n-type region in the body 10 is formed between one of the contacts 11. This contact 11 is connected to the common source line CS 30. A second switch formed by the MOS transistor has a gate in the string select line SSL 28 coupled to the corresponding memory cell having the last word line 27 and formed by the n-type region in the semiconductor body 10. One of the contacts 19 is between. This contact 19 is connected to the bit line BL 31. The first and second switching switches in this exemplary embodiment are MOS transistors, in this example thyristor dielectric layers 7 and 8.

在此例示中,為了簡化起見此串列中具有六個記憶胞。在典型的組態中,一個反及閘串列可以包含16、32或更多個記憶胞串聯安排。這些記憶胞所對應的字元線22-27具有電荷捕捉結構9於字元線與半導體主體10中通道區域之間。此記憶胞中的電荷捕捉結構9可以是介電電荷捕捉結構、浮動閘極電荷捕捉結構、或是其他合適作為使用此處所描述技術來程式化的快閃記憶體結構。此外,反及閘快閃結構的實施例中已經開發出沒有接面的樣態,其中節點13-17,且選擇性地包括節點12和18可以自此結構中省略。In this illustration, there are six memory cells in this series for the sake of simplicity. In a typical configuration, an inverse gate sequence can contain 16, 32 or more memory bank arrangements. The word lines 22-27 corresponding to these memory cells have a charge trapping structure 9 between the word lines and the channel regions in the semiconductor body 10. The charge trapping structure 9 in this memory cell can be a dielectric charge trapping structure, a floating gate charge trapping structure, or other suitable flash memory structure suitable for programming using the techniques described herein. In addition, no junctions have been developed in embodiments that are inverse gate flash structures in which nodes 13-17, and optionally nodes 12 and 18, may be omitted from this configuration.

第1A圖顯示一習知技術反及閘(NAND)架構快閃記憶體的剖面圖,其中誘發FN穿隧以對與字元線24對應之記憶胞進行程式化的偏壓示意圖。根據此處所顯示的偏壓,接地選擇線GSL偏壓至大約為0V而共同源極線接地,使得與接地選擇線GSL 21對應之第一切換開關是關閉的,且串列選擇線SSL偏壓至約VCC 而所選取位元線也是接地,使得與串列選擇線SSL 28對應之第二切換開關是開啟的。在這些條件下,與反及閘串列相關的區域33中的半導體主體是預充電至約0V。此選取字元線24被偏壓至一高電壓程式化階級V-PGM,在某些實施例中可以高達20~22伏特的數量級。選擇如此高的電壓足以導致主體10中的熱電子穿隧進入所選取記憶胞的電荷捕捉結構9中。同時,未選取字元線22、23、25~27被偏壓至一導通電壓V-PASS,其係比V-PGM還小於一個可以抑制此串列中未選取細胞的程式化之電壓。舉例而言,記憶胞C的閘極自字元線25接收導通電壓V-PASS,且雖然記憶胞C具有被設定為程式化的一主體區域,此低的導通電壓V-PASS仍足以干擾記憶胞C的程式化過程。FIG. 1A shows a cross-sectional view of a conventional technique NAND architecture flash memory in which FN tunneling is induced to program a biased pattern of memory cells corresponding to word line 24. According to the bias voltage shown here, the ground select line GSL is biased to approximately 0V and the common source line is grounded such that the first switch corresponding to the ground select line GSL 21 is off and the tandem select line SSL bias To about V CC and the selected bit line is also grounded, so that the second switch corresponding to the serial select line SSL 28 is turned on. Under these conditions, the semiconductor body in region 33 associated with the gate series is precharged to about 0V. The selected word line 24 is biased to a high voltage stylized class V-PGM, which in some embodiments can be on the order of 20-22 volts. Selecting such a high voltage is sufficient to cause hot electrons in the body 10 to tunnel into the charge trapping structure 9 of the selected memory cell. At the same time, the unselected word lines 22, 23, 25-27 are biased to a turn-on voltage V-PASS which is less than a V-PGM that is less than a stylized voltage that can suppress unselected cells in the series. For example, the gate of the memory cell C receives the turn-on voltage V-PASS from the word line 25, and although the memory cell C has a body region that is set to be programmed, the low turn-on voltage V-PASS is still sufficient to interfere with the memory. The stylization process of cell C.

第1B圖顯示一習知技術反及閘(NAND)架構快閃記憶體的剖面圖,其係對分享第1A圖中字元線22~27之反及閘串列未選取位元線的偏壓示意圖。由圖中可以發現,所有字元線、接地選擇線GSL與串列選擇線SSL皆與第1A圖所示的偏壓相同。類似地,共同源極線30也是接地。然而,未選取的位元線偏壓至約為VCC 的階級。如此會將第二切換開關關閉,其與串列選擇線SSL對應,且將區域35中的半導體主體與未選取的位元線BL 32解除耦接。其結果是,區域35中的半導體主體會由施加至字元線22~27電壓所產生的電容耦合自我壓升,其可以防止足以干擾未選取反及閘串列之記憶胞中電荷捕捉結構的電場形成。根據電容性自我壓升之所謂的遞增步進脈衝程式化(ISSP)操作是業界所熟知的。FIG. 1B is a cross-sectional view showing a conventional flash memory (NAND) architecture flash memory, which shares the bias of the uncharacterized bit lines of the gate lines 22 to 27 and the gate series of the gate array. schematic diagram. As can be seen from the figure, all word lines, ground selection lines GSL and serial selection lines SSL are the same as those shown in FIG. 1A. Similarly, the common source line 30 is also grounded. However, the unselected bit lines are biased to a level of approximately V CC . This will turn off the second switch, which corresponds to the string select line SSL, and decouples the semiconductor body in region 35 from the unselected bit line BL32. As a result, the semiconductor body in region 35 will self-pressurize by the capacitive coupling generated by the voltage applied to word lines 22-27, which can prevent the charge trapping structure in the memory cells of the unselected inverse gate series from being sufficiently disturbed. The electric field is formed. The so-called incremental step pulse programming (ISSP) operation based on capacitive self-pressure is well known in the art.

第1A~1C圖中所描述的程式化操作雖然可以有效率,但是其仍具有某些缺點。一個問題是程式化電壓階級V-PGM需要如此的高電壓階級(例如20~22伏特)。如此的高電壓需要會對某些記憶裝置的半導體結構造成設計上的限制,而使得半導體結構的微縮變得很困難。Although the stylized operations described in Figures 1A-1C can be efficient, they still have certain disadvantages. One problem is that the stylized voltage class V-PGM requires such a high voltage class (eg 20-22 volts). Such high voltages necessitate design limitations on the semiconductor structure of certain memory devices, making the shrinking of semiconductor structures difficult.

例如第1A~1C圖中所描述的傳統程式化操作的另一個問題是其僅允許導通電壓V-PASS很小範圍的變動以防止干擾。另一方面,假如導通電壓V-PASS太低的話,則不會有足夠的電容耦合效應來對例如第1B圖中所示的未選取反及閘串列升壓,而會對與目標記憶胞(第1C圖中的記憶胞A)分享字元線的記憶胞(第1C圖中的記憶胞B)產生干擾。另一方面,假如導通電壓V-PASS太高的話,則可能會在選取反及閘串列中的未選取記憶胞(第1C圖中的記憶胞C)產生熱載子注射。其結果是,導通電壓V-PASS必須小心地控制在介於其上下邊界之間。舉例而言,典型的導通電壓V-PASS範圍是介於9~11伏特之間。如此嚴謹的控制會因為製程或環境變動等因素而十分困難。Another problem with conventional stylized operations such as those described in Figures 1A-1C is that it only allows a small range of variations in the turn-on voltage V-PASS to prevent interference. On the other hand, if the turn-on voltage V-PASS is too low, there will not be enough capacitive coupling effect to boost the unselected inverse gate series as shown in FIG. 1B, but to the target memory cell. (Memory cell A in Fig. 1C) Memory cells sharing the word line (memory cell B in Fig. 1C) cause interference. On the other hand, if the turn-on voltage V-PASS is too high, hot carrier injection may be generated in the unselected memory cells (memory cell C in Fig. 1C) in the selected gate sequence. As a result, the turn-on voltage V-PASS must be carefully controlled between its upper and lower boundaries. For example, a typical turn-on voltage V-PASS range is between 9 and 11 volts. Such rigorous control can be difficult due to factors such as process or environmental changes.

如此傳統程式化操作的另一個問題是有時候會發生所謂的閘極誘發汲極漏電(GIDL)問題,舉例而言在接地選擇線GSL與字元線WL0的記憶胞之間的接面發生。此閘極誘發汲極漏電(GIDL)問題是難以避免的,且會在裝置微縮後變得更嚴重。Another problem with such conventional stylized operations is that sometimes a so-called gate induced drain leakage (GIDL) problem occurs, for example, at the junction between the ground select line GSL and the memory cell of the word line WL0. This gate induced dipole leakage (GIDL) problem is difficult to avoid and can become more severe after the device is miniature.

傳統反及閘記憶裝置及程式化操作的這些及其他的缺點可以藉由使用本發明所描述的裝置及方法加以克服。一種改良的反及閘記憶裝置可以具有與第1A和1B圖中類似的反及閘記憶胞達成,其中每一個記憶胞包括一個電荷捕捉結構於摻雜源極/汲極區域之間。然而,此處所揭露的反及閘記憶胞最好是形成於淡摻雜基板區域之上,舉例而言,具有摻雜濃度小於5x1012 cm-2 ,最好是大於零使得存在有少量的雜質。此淡摻雜基板允許在較低電壓階級的反轉。通常而言,此反及閘記憶胞是N型裝置,雖然P型裝置或許也可行且可以由熟習本技藝人士根據本發明的精神來實施。總而言之,本揭露主要係專注於N型裝置。在N型裝置中,源極/汲極區域包含N+ 摻雜區域,例如形成作為埋藏擴散區域。在如此的情況中,淡摻雜區域是N- 型態摻雜,如此可幫助電子反轉。These and other shortcomings of conventional anti-gate memory devices and stylized operations can be overcome by using the devices and methods described herein. An improved inverse thyristor device can have an inverse thyristor cell similar to that of Figures 1A and 1B, wherein each memory cell includes a charge trapping structure between the doped source/drain regions. However, the anti-gate memory cells disclosed herein are preferably formed over the lightly doped substrate region, for example, having a doping concentration of less than 5 x 10 12 cm -2 , preferably greater than zero such that a small amount of impurities are present. . This lightly doped substrate allows for inversion at lower voltage levels. In general, the anti-gate memory cell is an N-type device, although a P-type device may also be feasible and can be implemented by those skilled in the art in accordance with the spirit of the present invention. In summary, this disclosure focuses primarily on N-type devices. In an N-type device, the source/drain regions comprise N + doped regions, for example formed as buried diffusion regions. In such cases, the lightly doped regions are N - type doped, which can help with electron inversion.

此外,此處所揭露的記憶裝置及程式化操作可以允許降低程式化電壓階級V-PGM,例如程式化電壓階級V-PGM可以小於等於17伏特。舉例而言,程式化電壓階級V-PGM可以在13V≦V-PGM≦17V之間達成。在此處所描述的裝置及程式化操作中,通道電位(Vch)可以因為非常低的基板摻雜而被升壓至程式化電壓階級V-PGM的0.6倍或0.7倍。舉例而言,13V的程式化電壓可以將通道電位升壓至大約是7或8V,其可以誘發熱載子注入記憶胞的儲存節點。其結果是,可以使用一個較小的導通電壓V-PASS,舉例而言,在3V≦V-PASS≦8V之間,其可以幫助抑制閘極誘發汲極漏電(GIDL)。此外,程式化和讀取操作可以使用相同的導通電壓V-PASS。Moreover, the memory devices and stylized operations disclosed herein may allow for the reduction of the stylized voltage class V-PGM, for example, the stylized voltage class V-PGM may be less than or equal to 17 volts. For example, a stylized voltage class V-PGM can be achieved between 13V ≦V-PGM ≦ 17V. In the apparatus and stylized operations described herein, the channel potential (Vch) can be boosted to 0.6 times or 0.7 times the stylized voltage class V-PGM due to very low substrate doping. For example, a 13V stylized voltage can boost the channel potential to approximately 7 or 8V, which can induce a hot carrier to be injected into the memory cell's storage node. As a result, a smaller turn-on voltage V-PASS can be used, for example, between 3V ≦V-PASS ≦ 8V, which can help suppress gate induced gate leakage (GIDL). In addition, the same turn-on voltage V-PASS can be used for stylization and read operations.

第2A和2B圖顯示如此反及閘快閃記憶裝置的一個實施例。第2A圖顯示將複數個介電電荷捕捉快閃記憶胞串聯安排形成反及閘串列一部分的剖面圖,而第2B圖顯示包括第2A圖所示記憶胞的反及閘串列101和103之示意圖。Figures 2A and 2B show an embodiment of such a reverse flash memory device. Figure 2A shows a cross-sectional view of a plurality of dielectric charge trapping flash memory cells arranged in series to form a portion of the gate sequence, and Figure 2B shows the inverse gate series 101 and 103 including the memory cells shown in Figure 2A. Schematic diagram.

此反及閘串列101和103包括第一及第二切換開關,分別與接地選擇線GSL和串列選擇線SSL對應,其類似於第1A和1B圖所示,每一個係由金氧半電晶體形成,此電晶體具有閘極連接於一記憶胞與接點之間,其中此接點由一n型區域形成於半導體主體10內所構成。此狀況中的接地選擇線GSL,此接點可以與共同源極線CS連接;此狀況中的串列選擇線SSL,此接點可以與位元線BL連接。請參閱第2A圖,每一個記憶胞苦以與各自的字元線WL連接,例如第2A圖中所示的字元線23-25。這些記憶胞每一個也包括各自的電荷捕捉結構9介於字元線WL與半導體主體10內的通道區域之間。對n通道記憶胞而言,半導體主體10可以是一個於半導體晶片的更深的n井內之隔離的p井。替代地,半導體主體10可以由絕緣層或是其他類似地方式隔離。某些實施例可以使用p通道記憶胞,其中半導體主體10中的摻雜是n型。The reverse gate trains 101 and 103 include first and second switch switches respectively corresponding to the ground selection line GSL and the tandem selection line SSL, which are similar to those shown in FIGS. 1A and 1B, each of which is composed of a gold oxide half. A transistor is formed, the transistor having a gate connected between a memory cell and a contact, wherein the contact is formed by an n-type region formed in the semiconductor body 10. The ground selection line GSL in this case can be connected to the common source line CS; in this case, the series selects the line SSL, which can be connected to the bit line BL. Referring to Figure 2A, each memory cell is connected to a respective word line WL, such as word line 23-25 shown in Figure 2A. Each of these memory cells also includes a respective charge trapping structure 9 between the word line WL and the channel region within the semiconductor body 10. For n-channel memory cells, semiconductor body 10 can be an isolated p-well within a deeper n-well of a semiconductor wafer. Alternatively, the semiconductor body 10 can be isolated by an insulating layer or other similar means. Some embodiments may use p-channel memory cells in which the doping in the semiconductor body 10 is n-type.

複數個快閃記憶胞安排成沿著一個與字元線方向正交的位元線方向排列之串列。字元線WL沿伸通過一些平行的反及閘串列。例如是第2A圖中所示的節點14-15,是由半導體主體10中的n型區域(對n通道裝置而言)形成,且作為記憶胞的源/汲極區域。The plurality of flash memory cells are arranged in a series arranged along a direction of the bit line orthogonal to the direction of the word line. The word line WL extends through a number of parallel anti-gate trains. For example, the node 14-15 shown in FIG. 2A is formed by an n-type region (for an n-channel device) in the semiconductor body 10 and serves as a source/drain region of the memory cell.

在此例示中,為了簡化起見此串列中具有六個記憶胞。在典型的組態中,一個反及閘串列可以包含16、32或更多個記憶胞串聯安排。這些記憶胞所對應的字元線WL0-WL5具有電荷捕捉結構9於字元線與半導體主體10中通道區域之間。In this illustration, there are six memory cells in this series for the sake of simplicity. In a typical configuration, an inverse gate sequence can contain 16, 32 or more memory bank arrangements. The word lines WL0-WL5 corresponding to these memory cells have a charge trapping structure 9 between the word lines and the channel regions in the semiconductor body 10.

必須注意的是,第2A圖中所示的記憶裝置可以包括淡摻雜基板區域40以抹除例如是n型通道裝置中的電子之少數載子的反轉。換句話說,反轉過程可以在與傳統沒有淡摻雜基板區域40之裝置相較相對更低的電壓階級發生。此淡摻雜基板區域40可以摻雜與源/汲極區域14和15相同的導電型態。舉例而言,對一n通道裝置,此淡摻雜基板區域40可以是n- 摻雜區域。對包括淡摻雜基板區域40的實施例,此淡摻雜區域可以具有摻雜濃度小於等於5x1012 cm-2 。此淡摻雜區域40可以使用例如是已知的擴散製程方式形成。It must be noted that the memory device shown in FIG. 2A may include a lightly doped substrate region 40 to erase the inversion of a minority carrier such as an electron in an n-channel device. In other words, the inversion process can occur at a relatively lower voltage level than devices that are conventionally do not have a lightly doped substrate region 40. This lightly doped substrate region 40 can be doped with the same conductivity type as the source/drain regions 14 and 15. For example, for an n-channel device, the lightly doped substrate region 40 can be an n - doped region. For embodiments comprising the lightly doped substrate region 40, the lightly doped region may have a doping concentration of less than or equal to 5 x 10 12 cm -2 . This lightly doped region 40 can be formed using, for example, a known diffusion process.

此記憶胞中的電荷捕捉結構9可以是介電電荷捕捉結構、浮動閘極電荷捕捉結構、或是其他合適作為使用此處所描述技術來程式化的快閃記憶體結構。此外,反及閘快閃結構的實施例中已經開發出沒有接面的樣態,其中節點13-17,且選擇性地包括節點12和18可以自此結構中省略。The charge trapping structure 9 in this memory cell can be a dielectric charge trapping structure, a floating gate charge trapping structure, or other suitable flash memory structure suitable for programming using the techniques described herein. In addition, no junctions have been developed in embodiments that are inverse gate flash structures in which nodes 13-17, and optionally nodes 12 and 18, may be omitted from this configuration.

第2B圖顯示本發明所揭露之將複數個介電電荷捕捉快閃記憶胞串聯安排形成反及閘串列101和103,及其程式化操作時偏壓之示意圖。在第2B圖中,反及閘串列101是一選取記憶串列,其包括一目標記憶胞(記憶胞A)於字元線WL2上以進行程式化操作。反及閘串列103是一未選取記憶串列與一未選取位元線BL1連接。與第1C圖相較,可以發現在第2B圖中的程式化位元線與程式化干擾位元線電壓係與第1C圖中的相反。FIG. 2B is a schematic diagram showing the biasing of a plurality of dielectric charge trapping flash memory cells in series to form the anti-gate trains 101 and 103 and their stylized operation. In FIG. 2B, the inverse gate train 101 is a selected memory string including a target memory cell (memory cell A) on the word line WL2 for program operation. The inverse gate train 103 is an unselected memory string connected to an unselected bit line BL1. As compared with Figure 1C, it can be seen that the stylized bit line and the stylized interfering bit line voltage system in Figure 2B are the opposite of those in Figure 1C.

第3圖顯示第2圖之選取反及閘串列101的程式化操作時其操作信號的一範例時序示意圖。更具體而言,第3圖中顯示未選取字元線信號105、選取字元線信號106、串列選擇線SSL信號107及接地選擇線GSL信號108。為了程式化目標記憶胞A,未選取字元線信號105施加至未選取字元線WL0、WL1和WL3~WL5,選取字元線信號106施加至選取字元線WL2,串列選擇線SSL信號107自位元線BL0通過串列選擇切換開關施加至基板,而接地選擇線GSL信號108自共同源極線CS通過接地選擇切換開關施加至基板。FIG. 3 is a timing diagram showing an example of the operation signal of the second figure and the operation of the gate sequence 101. More specifically, the unselected word line signal 105, the selected word line signal 106, the tandem selection line SSL signal 107, and the ground selection line GSL signal 108 are shown in FIG. To program the target memory cell A, the unselected word line signal 105 is applied to the unselected word lines WL0, WL1 and WL3~WL5, and the selected word line signal 106 is applied to the selected word line WL2, the serial selection line SSL signal. The self-bit line BL0 is applied to the substrate through the serial selection switch, and the ground selection line GSL signal 108 is applied to the substrate from the common source line CS through the ground selection switch.

在時間t0時,反及閘串列101是在準備狀態,其信號105到108皆設定為0V。在時間t1時或之前,程式化操作被,舉例而言根據已知記憶體控制系統藉由內部命令,而初始啟動。在響應時,如第2B圖所示,施加一個約為Vcc的電壓至串列選擇線SSL,開啟此串列選擇切換開關,且施加≦0V的電壓至接地選擇線GSL,關閉此接地選擇切換開關。在時間t1時,串列選擇線SSL信號107包括一設定脈衝111通過位元線BL0施加至選取反及閘串列101的基板。此設定脈衝111超出Vcc一個預定的值,舉例而言,某些電壓範圍在串列選擇線SSL的臨界電壓<Vcc<設定脈衝111。此設定脈衝111創造一種情況是其中串列選擇切換開關的汲極電壓較閘極處更高,具有將電子吸入通道區域的效果,換句話說,其在此選取串列中初始此反轉過程。注意的是此效應並不會於未選取串列103發生,其中位元線BL1係施加0V。At time t0, the reverse gate train 101 is in a ready state, and its signals 105 to 108 are all set to 0V. At or before time t1, the stylized operation is initiated, for example, by a known memory control system by internal commands. In response, as shown in FIG. 2B, a voltage of about Vcc is applied to the string selection line SSL, the series selection switch is turned on, and a voltage of ≦0V is applied to the ground selection line GSL to turn off the ground selection switching. switch. At time t1, the serial select line SSL signal 107 includes a set pulse 111 applied to the substrate of the selected reverse gate train 101 through the bit line BL0. This set pulse 111 exceeds Vcc by a predetermined value. For example, some voltage ranges are at the threshold voltage <Vcc < set pulse 111 of the serial select line SSL. The set pulse 111 creates a situation in which the threshold voltage of the tandem selection switch is higher than that of the gate, and has the effect of drawing electrons into the channel region, in other words, it initially initiates this inversion process in the series. . Note that this effect does not occur in the unselected series 103 where the bit line BL1 is applied with 0V.

在時間t2時,當施加程式化電壓至記憶胞A的選取字元線WL2時,記憶胞A靠近或通道中的熱電子被拉至記憶胞A的電荷捕捉結構9中。必須注意的是第3圖中的電壓階級並未依比例繪製,且必須理解V-PGM>V-PASS。在時間t2時,V-PASS施加至未選取的記憶胞,但是V-PASS並沒有強到允許熱電子可以克服被捕捉至例如是記憶胞B和C的未選取記憶胞中的電荷捕捉結構9所需之能障高度。最後,在時間t3時,所有的電壓皆回到0V且此程式化操作完成。熟習本記憶人士應當可以理解在某些介於t2和t3之間的時段中可以選取允許適當的時間使熱電子穿隧進入電荷捕捉結構9中,且可以根據不同的因素例如是裝置尺寸及材料而變動。At time t2, when the stylized voltage is applied to the selected word line WL2 of the memory cell A, the thermal electrons in the memory cell A close to or in the channel are pulled into the charge trapping structure 9 of the memory cell A. It must be noted that the voltage classes in Figure 3 are not drawn to scale and must be understood to be V-PGM > V-PASS. At time t2, V-PASS is applied to the unselected memory cells, but V-PASS is not strong enough to allow hot electrons to overcome the charge trapping structure that is captured into unselected memory cells such as memory cells B and C. The required height of the barrier. Finally, at time t3, all voltages return to 0V and this stylization is complete. Those skilled in the art should understand that certain periods of time between t2 and t3 may be selected to allow proper time for tunneling of hot electrons into the charge trapping structure 9, and may vary depending on factors such as device size and material. And change.

接著,請參閱第4A及4B圖,這些圖表顯示程式化電壓與通道氧化層厚度的範例,其可以用來作為第2A及2B圖中所示裝置的特定應用。舉例而言,如第4A圖所示,在某些實施例中,大約17V的程式化電壓可以作為V-PGM,且導通電壓V-PASS大約是7~13V。如第4A圖所示,顯著數目的電子被注入選取記憶胞A的電荷捕捉結構9中,而未選取記憶胞B僅有少數。Next, please refer to Figures 4A and 4B, which show examples of stylized voltages and channel oxide thicknesses that can be used as specific applications for the devices shown in Figures 2A and 2B. For example, as shown in FIG. 4A, in some embodiments, a staging voltage of approximately 17V can be used as the V-PGM, and the turn-on voltage V-PASS is approximately 7-13V. As shown in Fig. 4A, a significant number of electrons are injected into the charge trapping structure 9 of the selected memory cell A, while only a few memory cells B are not selected.

第4C圖顯示一實施例的記憶胞之簡要剖面示意圖,其顯示出與字元線24相關記憶胞的範例電荷捕捉結構9之放大圖。其他的記憶胞與第4C圖中所示的相同,所以為了簡化起見僅顯示一個記憶胞。此電荷捕捉結構9包括一通道氧化層9c直接於基板10之上,或是更具體而言,於基板的淡摻雜區域40之上。之後,一浮動閘極(電荷儲存)層9b直接提供於通道氧化層9c之上。一阻擋介電層9a直接提供於浮動閘極層9b之上。控制閘極24則直接於阻擋介電層9a之上。如此,舉例而言,此電荷捕捉結構9可以使用矽-氧化矽-氮化矽-氧化矽-矽(SONOS)的結構形成。然而,也可以使用其他的電荷捕捉結構。Figure 4C shows a schematic cross-sectional view of a memory cell of an embodiment showing an enlarged view of an exemplary charge trapping structure 9 associated with word line 24 memory cells. The other memory cells are the same as those shown in Fig. 4C, so only one memory cell is shown for the sake of simplicity. The charge trapping structure 9 includes a channel oxide layer 9c directly over the substrate 10 or, more specifically, over the lightly doped region 40 of the substrate. Thereafter, a floating gate (charge storage) layer 9b is directly provided over the channel oxide layer 9c. A barrier dielectric layer 9a is provided directly over the floating gate layer 9b. The control gate 24 is directly over the blocking dielectric layer 9a. Thus, for example, the charge trapping structure 9 can be formed using a structure of yttrium-yttria-yttria-yttria-yttria (SONOS). However, other charge trapping structures can also be used.

第4B圖顯示本發明程式化操作可以有益地允許相對厚的通道氧化層於第2A及2B圖中所示記憶胞的電荷捕捉結構9。舉例而言,通道氧化層9c之厚度T9c可以是在79到91埃的範圍間。較厚的通道氧化層可以需要略長的程式化時間(例如圖3中較長的時間於t2和t3之間),所以最好是T9c厚度小於90埃。然而,具有較厚通道氧化層的記憶胞可以有較長保存時間的優點,所以也可以使用其他的厚度。Figure 4B shows that the stylization operation of the present invention can advantageously allow a relatively thick channel oxide layer to be present in the charge trapping structure 9 of the memory cells shown in Figures 2A and 2B. For example, the thickness T9c of the channel oxide layer 9c may be in the range of 79 to 91 angstroms. A thicker channel oxide layer may require a slightly longer stylized time (e.g., a longer time in Figure 3 between t2 and t3), so it is preferred that the T9c thickness be less than 90 angstroms. However, memory cells with thicker channel oxide layers can have the advantage of longer shelf life, so other thicknesses can be used.

第5圖顯示傳統反及閘記憶串列與本發明之字元線WL0分佈的比較。如第5圖所示,因為本發明的反及閘快閃記憶裝置具有較傳統反及閘記憶裝置更低的程式化電壓V-PGM及導通電壓V-PASS電壓階級,本發明的反及閘快閃記憶裝置可以因為消除閘極誘發汲極漏電(GIDL)而具有顯著減少字元線WL0分佈的優點仍能足以產生熱載子於此區域中。Figure 5 shows a comparison of the conventional inverse gate memory string with the word line WL0 distribution of the present invention. As shown in FIG. 5, since the anti-gate flash memory device of the present invention has a lower stylized voltage V-PGM and a turn-on voltage V-PASS voltage class than the conventional anti-gate memory device, the inverse gate of the present invention Flash memory devices can have the advantage of significantly reducing the distribution of word line WL0 due to the elimination of gate induced drain leakage (GIDL), which is still sufficient to generate hot carriers in this region.

第6圖顯示本發明選取記憶胞A於程式化操作時如何發生熱載子注射的能帶示意圖。必須注意第6圖係顯示電子注射,因為其是在N型通道裝置中發生。熟知此技藝人士可以理解在P型通道裝置中是電洞注射。於程式化操作時,較高的串列選擇脈衝111提供能量給基板中的電子,包括於淡摻雜基板區域40中的電子。能帶間穿隧使電子加速,且這些電子變成熱電子。施加程式化電壓V-PGM吸引這些熱電子,提供足夠的能量以克服通道氧化層中的能障,使得熱電子注射進入浮動閘極(FG)層。Fig. 6 is a view showing the energy band of the present invention for selecting how to generate a hot carrier injection when the memory cell A is programmed. It must be noted that Figure 6 shows the electron injection as it occurs in the N-channel device. Those skilled in the art will appreciate that hole injection is a P-channel device. During the stylization operation, the higher tandem select pulse 111 provides energy to the electrons in the substrate, including electrons in the lightly doped substrate region 40. The inter-band tunneling accelerates the electrons and these electrons become hot electrons. The application of the stylized voltage V-PGM attracts these hot electrons, providing sufficient energy to overcome the energy barrier in the channel oxide layer, allowing hot electrons to be injected into the floating gate (FG) layer.

第7圖顯示實驗數據的結果,顯示如何達成足夠的臨界電壓Vt差異使得允許決定一記憶胞是否被程式化或抹除。舉例而言,在第7圖中,因為其間的臨界電壓差值大約是3.5V而可以將抹除記憶胞B自記憶胞A中分辨出來。Figure 7 shows the results of the experimental data showing how to achieve a sufficient threshold voltage Vt difference to allow determination of whether a memory cell is programmed or erased. For example, in Fig. 7, the erased memory cell B can be distinguished from the memory cell A because the threshold voltage difference therebetween is about 3.5V.

第8圖顯示積體電路的簡化示意圖,其使用此處所描述之熱載子注入程式化的反及閘快閃記憶體。此積體電路210包括使用電荷捕捉或是浮動閘極記憶胞的一記憶體陣列212,其形成於舉例而言,一半導體基板之上。字元線(或列)接地選擇及串列選擇解碼器(包括合適的驅動器)214與複數條字元線216、串列選擇線、和接地選擇線耦接且電性溝通,且沿著記憶陣列212的列方向排列。位元線(行)解碼器及驅動器218與複數條位元線220電性溝通且沿著記憶陣列212的行方向排列,以自陣列212的記憶胞讀取資料或寫入資料至其中。位址係由匯流排222提供給字元線及串列選擇解碼器214與位元線解碼器218。方塊224中的感測放大器與資料輸入結構,包括讀取、程式化及抹除模式的電流源,經由資料匯流排226與位元線解碼器218耦接。資料由積體電路210上的輸入/輸出埠提供給資料輸入線228,或者由積體電路210其他內部/外部的資料源,輸入至方塊224中的資料輸入結構。其他電路230係包含於積體電路210之內,例如泛用目的處理器或特殊目的應用電路,或是模組組合以提供由陣列所支援的系統單晶片功能。資料由方塊224中的感測放大器,經由資料輸出線232,提供至積體電路210,或提供至積體電路210內部/外部的其他資料終端。Figure 8 shows a simplified schematic of an integrated circuit that uses the hot carrier described herein to inject a programmed inverse gate flash memory. The integrated circuit 210 includes a memory array 212 that uses charge trapping or floating gate memory cells, which are formed, for example, on a semiconductor substrate. A word line (or column) ground selection and serial selection decoder (including a suitable driver) 214 is coupled to the plurality of word lines 216, the string selection line, and the ground selection line and electrically communicated, and along the memory The arrays 212 are arranged in the column direction. The bit line (row) decoder and driver 218 is electrically coupled to the plurality of bit lines 220 and arranged along the row direction of the memory array 212 to read data or write data from the memory cells of the array 212. The address is provided by bus bar 222 to word line and string select decoder 214 and bit line decoder 218. The sense amplifier and data input structures in block 224, including current sources for read, program, and erase modes, are coupled to bit line decoder 218 via data bus 226. The data is supplied to the data input line 228 by the input/output ports on the integrated circuit 210, or is input to the data input structure in block 224 by other internal/external data sources of the integrated circuit 210. Other circuits 230 are included within integrated circuit 210, such as a general purpose processor or special purpose application circuit, or a combination of modules to provide system single chip functionality supported by the array. The data is provided by the sense amplifier in block 224, via the data output line 232, to the integrated circuit 210, or to other data terminals internal/external to the integrated circuit 210.

在本實施例中所使用的控制器234,使用了偏壓調整狀態機構,控制了偏壓調整供應電壓及電流源236的應用,例如讀取、程式化、抹除、抹除確認以及程式化確認電壓或電流施加於字元線或位元線上,並使用存取控制流程控制了字元線/源極線的操作。該控制器也應用切換序列來誘發此處所描述之熱載子程式化。控制器234可以使用業界所熟知的特殊功能邏輯電路來實施。在替代實施例中,該控制器234包括了通用目的處理器,其可使於同一積體電路,以執行一電腦程式而控制裝置的操作。在又一實施例中,該控制器234係由特殊目的邏輯電路與通用目的處理器組合而成。The controller 234 used in this embodiment uses a bias adjustment state mechanism to control the application of the bias voltage adjustment supply voltage and current source 236, such as reading, programming, erasing, erasing confirmation, and stylization. Verify that the voltage or current is applied to the word line or bit line and control the operation of the word line/source line using the access control flow. The controller also applies a switching sequence to induce the hot carrier stylization described herein. Controller 234 can be implemented using special function logic circuitry well known in the art. In an alternate embodiment, the controller 234 includes a general purpose processor that can be used in the same integrated circuit to execute a computer program to control the operation of the device. In yet another embodiment, the controller 234 is a combination of special purpose logic circuitry and a general purpose processor.

雖然本發明係已參照實施例來加以描述,然本發明創作並未受限於其詳細描述內容。替換方式及修改樣式係已於先前描述中所建議,且其他替換方式及修改樣式將為熟習此項技藝之人士所思及。特別是,所有具有實質上相同於本發明之構件結合而達成與本發明實質上相同結果者,皆不脫離本發明之精神範疇。因此,所有此等替換方式及修改樣式係意欲落在本發明於隨附申請專利範圍及其均等物所界定的範疇之中。Although the present invention has been described with reference to the embodiments, the present invention is not limited by the detailed description thereof. Alternatives and modifications are suggested in the foregoing description, and other alternatives and modifications will be apparent to those skilled in the art. In particular, all combinations of components that are substantially identical to the invention can achieve substantially the same results as the present invention without departing from the spirit of the invention. Therefore, all such alternatives and modifications are intended to be within the scope of the invention as defined by the appended claims and their equivalents.

本段僅與US 37 CFR 1.77相關,故未翻譯This paragraph is only relevant to US 37 CFR 1.77 and is therefore not translated.

7、8...閘介電層7, 8. . . Gate dielectric layer

9...電荷捕捉結構9. . . Charge trapping structure

10...半導體主體10. . . Semiconductor body

11、19...接點11, 19. . . contact

12~18...節點12~18. . . node

21...接地選擇線GSLtwenty one. . . Ground selection line GSL

22~27...字元線22~27. . . Word line

28...串列選擇線SSL28. . . Serial selection line SSL

30...共同源極線CS30. . . Common source line CS

31...位元線31. . . Bit line

32...未選取位元線32. . . Unselected bit line

40...淡摻雜基板區域40. . . Lightly doped substrate area

101、103...反及閘串列101, 103. . . Reverse gate train

210...積體電路210. . . Integrated circuit

212...反及閘快閃記憶體212. . . Anti-gate flash memory

214...字元線/串列選擇解碼器及驅動器214. . . Word line/serial selection decoder and driver

216...字元線216. . . Word line

218...位元線解碼器218. . . Bit line decoder

220...位元線220. . . Bit line

222、226‧‧‧匯流排222, 226‧‧ ‧ busbar

224‧‧‧感測放大器/資料輸入結構224‧‧‧Sense Amplifier/Data Entry Structure

234‧‧‧控制器(熱載子注射程式化、FN抹除)234‧‧‧ Controller (hot carrier injection stylized, FN erase)

236‧‧‧偏壓調整供應電壓236‧‧‧ bias adjustment supply voltage

228‧‧‧資料輸入線228‧‧‧ data input line

230‧‧‧其他電路230‧‧‧Other circuits

232‧‧‧資料輸出線232‧‧‧ data output line

本發明係由申請專利範圍所界定。這些和其它目的,特徵,和實施例,會在下列實施方式的章節中搭配圖式被描述,其中:The invention is defined by the scope of the patent application. These and other objects, features, and embodiments are described in the following sections of the accompanying drawings, in which:

第1A和1B圖分別顯示一選取反及閘串列及一未選取反及閘串列的剖面示意圖,其是使用傳統的程式化操作偏壓。Figures 1A and 1B show a cross-sectional view of a selected reverse gate train and an unselected reverse gate train, respectively, using a conventional stylized operational bias.

第1C圖顯示第1A和1B圖中所示反及閘串列的簡要示意圖。Figure 1C shows a schematic diagram of the inverted gate train shown in Figures 1A and 1B.

第2A圖顯示根據本發明實施例之反及閘快閃記憶串列一部分的剖面圖。Figure 2A shows a cross-sectional view of a portion of a reverse flash memory string in accordance with an embodiment of the present invention.

第2B圖顯示根據本發明實施例之反及閘快閃記憶串列的示意圖。Figure 2B shows a schematic diagram of a reverse gate flash memory string in accordance with an embodiment of the present invention.

第3圖顯示第2A及2B圖之反及閘串列的程式化操作時其操作信號的一範例時序示意圖。Figure 3 is a diagram showing an example timing diagram of the operation signals of the reversed gate sequence of the 2A and 2B diagrams.

第4A及4B圖的圖表顯示程式化電壓與通道氧化層厚度的範例,其可以用來作為第2A及2B圖中所示裝置的特定應用。The graphs of Figures 4A and 4B show examples of stylized voltages and channel oxide thicknesses that can be used as specific applications for the devices shown in Figures 2A and 2B.

第4C圖顯示一實施例的記憶胞之簡要剖面示意圖,其顯示出一範例電荷捕捉結構之放大圖。Figure 4C is a schematic cross-sectional view of a memory cell of an embodiment showing an enlarged view of an exemplary charge trapping structure.

第5圖顯示傳統反及閘記憶串列與本發明之字元線WL0分佈的比較。Figure 5 shows a comparison of the conventional inverse gate memory string with the word line WL0 distribution of the present invention.

第6圖顯示本發明於程式化操作時如何發生熱載子注射的能帶示意圖。Fig. 6 is a view showing the energy band of the present invention in which a hot carrier injection occurs during a stylized operation.

第7圖顯示實驗數據的結果,顯示如何達成足夠的臨界電壓Vt差異使得允許決定一記憶胞是否被程式化或抹除。Figure 7 shows the results of the experimental data showing how to achieve a sufficient threshold voltage Vt difference to allow determination of whether a memory cell is programmed or erased.

第8圖係可應用本發明所描述熱載子注射程式化反及閘快閃記憶體之積體電路的方塊示意圖。Figure 8 is a block diagram showing the integrated circuit of a hot carrier injection stylized inverse gate flash memory as described in the present invention.

101、103...反及閘串列101, 103. . . Reverse gate train

Claims (24)

一種記憶體,包含:複數個記憶胞串聯安排於一半導體主體中;複數條字元線,該複數條字元線中的每一條字元線與該複數個記憶胞中對應的記憶胞耦接;以及控制電路,與該複數條字元線耦接,該控制電路適用於藉由下列步驟程式化該複數個記憶胞中與一選取字元線對應的一選取記憶胞:偏壓該複數個記憶胞的一串列選擇線至一設定電壓;降低施加至該複數個記憶胞的該第一及第二端之該一者的電壓階級自該設定電壓至一位元線程式化電壓;施加一導通電壓至與未選取記憶胞所對應的字元線;以及施加一程式化電壓至與該選取記憶胞所對應的該選取字元線;其中,該位元線程式化電壓大於該串列選擇線的一臨界電壓,該程式化電壓大於該導通電壓。 A memory comprising: a plurality of memory cells arranged in series in a semiconductor body; a plurality of word lines, each of the plurality of word lines being coupled to a corresponding memory cell of the plurality of memory cells And a control circuit coupled to the plurality of word lines, the control circuit being adapted to program a selected memory cell corresponding to a selected word line of the plurality of memory cells by biasing the plurality of memory cells a series of selection lines of the memory cells to a set voltage; reducing a voltage level of the one of the first and second ends applied to the plurality of memory cells from the set voltage to a bit-threaded voltage; applying a turn-on voltage to a word line corresponding to the unselected memory cell; and applying a stylized voltage to the selected word line corresponding to the selected memory cell; wherein the bit threaded voltage is greater than the string A threshold voltage of the line is selected, and the stylized voltage is greater than the turn-on voltage. 如申請專利範圍第1項所述之記憶體,其中該半導體主體包含一淡摻雜基板區域。 The memory of claim 1, wherein the semiconductor body comprises a lightly doped substrate region. 如申請專利範圍第2項所述之記憶體,其中該淡摻雜基板區域的摻雜濃度小於或等於5x1012 cm-2The memory of claim 2, wherein the doped substrate region has a doping concentration of less than or equal to 5 x 10 12 cm -2 . 如申請專利範圍第2項所述之記憶體,其中該淡摻雜基板區域包括一N- 型態摻雜區域。The memory of claim 2, wherein the lightly doped substrate region comprises an N - type doped region. 如申請專利範圍第1項所述之記憶體,其中該複數個記憶胞中 的每一個記憶胞包括一各自的電荷捕捉結構。 The memory of claim 1, wherein the plurality of memory cells are Each memory cell includes a respective charge trapping structure. 如申請專利範圍第5項所述之記憶體,其中該電荷捕捉結構係形成於一淡摻雜基板區域之上。 The memory of claim 5, wherein the charge trapping structure is formed over a lightly doped substrate region. 如申請專利範圍第5項所述之記憶體,其中該電荷捕捉結構包括各自的通道氧化層,每一個通道氧化層的厚度小於90埃。 The memory of claim 5, wherein the charge trapping structure comprises a respective channel oxide layer, each channel oxide layer having a thickness of less than 90 angstroms. 如申請專利範圍第1項所述之記憶體,其中施加至該選取字元線的該程式化電壓小於或等於17伏特。 The memory of claim 1, wherein the stylized voltage applied to the selected word line is less than or equal to 17 volts. 如申請專利範圍第8項所述之記憶體,其中該導通電壓係在3~8伏特範圍間。 The memory of claim 8, wherein the turn-on voltage is in the range of 3 to 8 volts. 如申請專利範圍第1項所述之記憶體,其中施加該設定電壓導致該半導體主體中的反轉。 The memory of claim 1, wherein applying the set voltage results in inversion in the semiconductor body. 如申請專利範圍第1項所述之記憶體,其中偏壓該複數個記憶胞的第一及一第二端之一者的步驟是在一第一時間區間內進行,且其中降低該電壓階級、施加該導通電壓以及施加該程式化電壓的步驟是在該第一時間區間後的一第二時間區間內進行。 The memory of claim 1, wherein the step of biasing one of the first and second ends of the plurality of memory cells is performed in a first time interval, and wherein the voltage class is lowered The step of applying the turn-on voltage and applying the stylized voltage is performed in a second time interval after the first time interval. 如申請專利範圍第1項所述之記憶體,其中進行偏壓該複數個記憶胞的第一及一第二端之一者的同時施加一接地電壓階級至該複數個記憶胞的第一及一第二端之另一者以及至該複數條字元線中的每一條。 The memory of claim 1, wherein the biasing of one of the first and second ends of the plurality of memory cells simultaneously applies a ground voltage level to the first of the plurality of memory cells. The other of the second ends and to each of the plurality of word lines. 一種記憶體,包含:複數個記憶胞的一第一串列串聯安排於一半導體主體中; 複數個記憶胞的一第二串列串聯安排於該半導體主體中;複數條字元線,該複數條字元線中的每一條字元線與該複數個記憶胞中各自的該第一串列記憶胞之一及該第二串列記憶胞之一耦接;以及控制電路,與該複數條字元線耦接,該控制電路適用於藉由下列步驟程式化該複數個記憶胞的該第一串列中與一選取字元線對應的一選取記憶胞:偏壓一設定電壓至該第一串列記憶胞中的一串列選擇線;降低施加於該串列選擇線的電壓階級自該設定電壓至一位元線程式化電壓,該位元線程式化電壓大於該串列選擇線的臨界電壓;維持該第二串列記憶胞中的該第一及第二端兩者在該接地階級電壓;施加一導通電壓至與未選取記憶胞所對應的字元線;以及施加一程式化電壓至與該選取記憶胞所對應的該選取字元線,該程式化電壓大於該導通電壓。 A memory comprising: a first series of a plurality of memory cells arranged in series in a semiconductor body; a second series of a plurality of memory cells are arranged in series in the semiconductor body; a plurality of word lines, each of the plurality of word lines and the first string of the plurality of memory cells One of the column memory cells and one of the second serial memory cells are coupled; and a control circuit coupled to the plurality of word line lines, the control circuit being adapted to program the plurality of memory cells by the following steps a selected memory cell corresponding to a selected word line in the first series: biasing a set voltage to a series of select lines in the first serial memory cell; reducing a voltage level applied to the serial select line From the set voltage to the one-bit threaded voltage, the bit threaded voltage is greater than a threshold voltage of the string select line; maintaining both the first and second ends of the second string of memory cells The ground-level voltage; applying a turn-on voltage to a word line corresponding to the unselected memory cell; and applying a stylized voltage to the selected word line corresponding to the selected memory cell, the stylized voltage being greater than the turn-on Voltage. 如申請專利範圍第13項所述之記憶體,其中該半導體主體包含一淡摻雜基板區域。 The memory of claim 13, wherein the semiconductor body comprises a lightly doped substrate region. 如申請專利範圍第14項所述之記憶體,其中該淡摻雜基板區域的摻雜濃度小於或等於5x1012 cm-2The memory of claim 14, wherein the doped substrate region has a doping concentration of less than or equal to 5 x 10 12 cm -2 . 如申請專利範圍第14項所述之記憶體,其中該淡摻雜基板區域包括一N- 型態摻雜區域。The memory of claim 14, wherein the lightly doped substrate region comprises an N - type doped region. 如申請專利範圍第13項所述之記憶體,其中該複數個記憶胞中的每一個記憶胞包括一各自的電荷捕捉結構。 The memory of claim 13, wherein each of the plurality of memory cells comprises a respective charge trapping structure. 如申請專利範圍第17項所述之記憶體,其中該電荷捕捉結構係形成於一淡摻雜基板區域之上。 The memory of claim 17, wherein the charge trapping structure is formed over a lightly doped substrate region. 如申請專利範圍第17項所述之記憶體,其中該電荷捕捉結構包括各自的通道氧化層,每一個通道氧化層的厚度小於90埃。 The memory of claim 17, wherein the charge trapping structure comprises a respective channel oxide layer, each channel oxide layer having a thickness of less than 90 angstroms. 如申請專利範圍第13項所述之記憶體,其中施加至該選取字元線的該程式化電壓小於或等於17伏特。 The memory of claim 13, wherein the stylized voltage applied to the selected word line is less than or equal to 17 volts. 如申請專利範圍第20項所述之記憶體,其中該導通電壓係在3~8伏特範圍間。 The memory of claim 20, wherein the turn-on voltage is in the range of 3 to 8 volts. 如申請專利範圍第13項所述之記憶體,其中施加該設定電壓導致該半導體主體中的反轉。 The memory of claim 13, wherein applying the set voltage results in inversion in the semiconductor body. 如申請專利範圍第13項所述之記憶體,其中該控制電路更進一步組態為,於一第一時間區間內偏壓該第一串列記憶胞中的第一及一第二端之一者至一設定電壓,並且同時施加一接地電壓階級至該第一串列記憶胞中的第一及一第二端之另一者、該複數條字元線中的每一條、以及該第二串列記憶胞中的第一及第二端兩者。 The memory of claim 13, wherein the control circuit is further configured to bias one of the first and second ends of the first series of memory cells in a first time interval Setting a voltage to a voltage, and simultaneously applying a ground voltage level to the other of the first and second ends of the first series of memory cells, each of the plurality of word lines, and the second Aligning both the first and second ends of the memory cell. 如申請專利範圍第23項所述之記憶體,其中施加該位元線程式化電壓、維持該第二串列記憶胞中的該第一及第二端兩者在該接地階級電壓、施加該導通電壓以及施加該程式化電壓的步驟皆是在該第一時間區間後的一第二時間區間內進行。 The memory of claim 23, wherein the bit threading voltage is applied, maintaining the first and second ends of the second serial memory cell at the ground class voltage, applying the The turn-on voltage and the step of applying the stylized voltage are all performed in a second time interval after the first time interval.
TW100114082A 2011-04-22 2011-04-22 Hot carrier programming of nand flash memory TWI489593B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW100114082A TWI489593B (en) 2011-04-22 2011-04-22 Hot carrier programming of nand flash memory

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW100114082A TWI489593B (en) 2011-04-22 2011-04-22 Hot carrier programming of nand flash memory

Publications (2)

Publication Number Publication Date
TW201244017A TW201244017A (en) 2012-11-01
TWI489593B true TWI489593B (en) 2015-06-21

Family

ID=48093968

Family Applications (1)

Application Number Title Priority Date Filing Date
TW100114082A TWI489593B (en) 2011-04-22 2011-04-22 Hot carrier programming of nand flash memory

Country Status (1)

Country Link
TW (1) TWI489593B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090086542A1 (en) * 2007-09-28 2009-04-02 Dana Lee High Voltage Generation and Control in Source-Side Injection Programming of Non-Volatile Memory
CN101449335A (en) * 2006-05-19 2009-06-03 Nxp股份有限公司 Sonos memory device and method of operating a sonos memory device
TW201007724A (en) * 2008-08-04 2010-02-16 Mediatek Inc Apparatus and method for calibrating optical storage device

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101449335A (en) * 2006-05-19 2009-06-03 Nxp股份有限公司 Sonos memory device and method of operating a sonos memory device
US20090086542A1 (en) * 2007-09-28 2009-04-02 Dana Lee High Voltage Generation and Control in Source-Side Injection Programming of Non-Volatile Memory
TW201007724A (en) * 2008-08-04 2010-02-16 Mediatek Inc Apparatus and method for calibrating optical storage device

Also Published As

Publication number Publication date
TW201244017A (en) 2012-11-01

Similar Documents

Publication Publication Date Title
JP3878681B2 (en) Nonvolatile semiconductor memory device
JP6633295B2 (en) Sub-block erase
KR102107090B1 (en) P-channel 3d memory array
US6657894B2 (en) Apparatus and method for programming virtual ground nonvolatile memory cell array without disturbing adjacent cells
US8842479B2 (en) Low voltage programming in NAND flash with two stage source side bias
US8472251B2 (en) Single-polycrystalline silicon electrically erasable and programmable nonvolatile memory device
US8947939B2 (en) Low voltage programming in NAND flash
US8971125B2 (en) Erase operations with erase-verify voltages based on where in the erase operations an erase cycle occurs
US7450418B2 (en) Non-volatile memory and operating method thereof
JP5059437B2 (en) Nonvolatile semiconductor memory device
JP2009540545A (en) Nonvolatile memory embedded in a conventional logic process and method of operating such a nonvolatile memory
US7672169B2 (en) Nonvolatile semiconductor memory and driving method thereof
CN111508541A (en) Asymmetric pass field effect transistor for non-volatile memory
TW201832241A (en) Method to reduce program disturbs in non-volatile memory cells
JP2006173565A (en) Memory cell and related operating method
US11238946B2 (en) Apparatus and methods for seeding operations concurrently with data line set operations
JP5853853B2 (en) Semiconductor memory device and driving method thereof
US7894265B2 (en) Non-volatile memory device and operation method of the same
US8913427B2 (en) Semiconductor memory device and method of operating the same
US20130080718A1 (en) Semiconductor memory device and method of operating the same
JP5483826B2 (en) Nonvolatile semiconductor memory device and writing method thereof
TWI489465B (en) Low voltage programming in nand flash with two stage source side bias
TWI292218B (en) An electrically erasable and programmable nonvolatile memory device and array and method for operating thereof
US9349461B1 (en) Applying substantially the same voltage differences across memory cells at different locations along an access line while programming
US20120236649A1 (en) Hot carrier programming of nand flash memory