• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:

    公开号:FR3017746A1
    申请号:FR1451297
    申请日:2014-02-18
    公开日:2015-08-21
    发明作者:Marc Mantelli；Stephan Niel；Arnaud Regnier；Rosa Francesco La；Julien Delalleau
    申请人:STMicroelectronics Rousset SAS；
    IPC主号:
    专利说明:

    [0001] The present invention relates to split-grid memory cells each comprising a selection transistor section and a floating gate transistor section. The selection transistor section has a selection gate and the floating gate transistor section has a floating gate and a control gate. Memory cells called "split gate" are conventionally programmed by hot electron injection (or "hot carrier injection"). Hot electron programming has the advantage over tunneling programming of being short-lived, typically 100 times shorter than tunneling programming. The programming time of a memory cell by hot electron injection is typically of the order of a few microseconds against a few milliseconds for tunneling programming.
    [0002] During the hot electron programming, the two transistor sections of the memory cell cooperate to inject electric charges into the floating gate. The selection transistor section has a conductive channel in which a current is formed comprising electrons with high kinetic energy, known as "hot electrons". When this current reaches the conductive channel of the floating gate transistor section, an injection zone is formed where the high energy electrons are injected into the floating gate under the effect of a transverse electric field created by the applied voltage. to the control grid.
    [0003] Figure 1 shows the arrangement of a conventional split gate memory cell C1, in a word line WLi of a memory plane. The selection gate SG of the selection transistor section ST of the memory cell is connected to a selection line SLi and the control gate CG of the floating gate transistor section FGT is connected to a gate control line CGLi . The drain D of the selection transistor section is connected to a bit line BLi and the source S of the floating gate transistor section FGT is connected to a source line SCLi. The selection lines SLi, gate control CGLi and source SCLi are parallel and connected to all the memory cells of the word line. Bit line BLi is transverse to lines SLi, CGLi, SCLi and is also connected to memory cells belonging to other word lines (not shown).
    [0004] The selection line SLi receives a selection voltage VSi, the gate control line CGLi receives a gate voltage VGi and the source line SCLi receives a source voltage VSC. The voltage VG is generally high, for example 10 V, to show, in the channel of the floating gate transistor section FGT, a transverse electric field favoring the injection of electrons into the floating gate. The voltage VSC is sufficiently high, for example 4 V, to ensure the conduction of the memory cell. The voltage VS is generally set at a value greater than the threshold voltage of the selection transistor section, for example between 1V and 3V. A programming current flows through the memory cell and the bit line BLi. An electron flow circulating in the opposite direction of the current flows through the channel of the selection transistor section until it reaches the injection point in the channel of the floating gate transistor section. In return for their good injection efficiency, the divided-grid memory cells have the disadvantage of occupying a larger semiconductor area than conventional flash memory cells, also programmed by hot electron injection but having only one control grid.
    [0005] US Patent 5,495,441 discloses a so-called "split gate" memory cell whose selection transistor section is arranged vertically to reduce the size of the memory cell. Figure 2 corresponds to Figure 7 of this document and shows a sectional view of the structure of such a memory cell. The reference numerals in FIG. 2 are those of the original FIG. 7 of the aforementioned document. The memory cell C2 shown in FIG. 2 comprises a trench etched in a substrate (27) after the formation of a floating gate FG (28) made of polysilicon (polycrystalline silicon) above the substrate. The trench was then covered with an oxide layer (200a, 200b). A conductive polysilicon layer (26) was then deposited on the entire memory cell. The conductive layer (26) has a portion extending into the trench and forming a vertical selection gate SG, a portion extending over the floating gate FG (28) forming a horizontal control gate CG, the remainder of the layer conductor forming a selection line SL of the memory cell. A doped region (21) implanted in the substrate forms a bit line BL and doped regions (20) located at the bottom of the trench form "source bit lines" SBL ("source bit liges") which are parallel to the bit line BL (21). The memory cell C2 thus has a selection transistor section ST having a vertical channel of length L1, and a floating gate transistor section FGT having a horizontal channel of length L2, which cooperate to form a transistor having a channel of length L1. + L2. The control gates GC and SG selection of the two transistor sections FGT, ST are formed by the same conductive layer (26) and are merged. The memory cell C2 is formed together with a memory cell C2 'connected to the same selection line SL (26) and to the same bit line BL (21), but to a "source bit line" SBL' (20). ) different. As shown in FIG. 3, this memory cell structure C2, C2 'imposes a memory plane architecture very different from the conventional architecture shown in FIG. 1. The sources S of the selection transistor sections ST of the two twin memory cells are connected to the "source bit lines" SBL (20), SBL '(20) which are parallel to the bit line BL (21). The selection line SL (26), the grids SG (26) and CG (26) of the memory cells are at the same electrical potential, the grids SG and CG thus forming only a single selection / control grid.
    [0006] This memory cell structure provides a small footprint due to the vertical arrangement of the select transistor section. In return, it involves a multiplication of the number of source lines in the form of "source bit lines" SBL, which results in a multiplication of the voltage switching means in the memory plane. For example, a word line comprising 1024 memory cells will require 512 bit lines and 1024 "source bit lines" parallel to the bit lines, against 1024 bit lines and a single source line in a conventional architecture of the type shown on FIG. 1. On the other hand, since the control gates CG and the selection gates 5 are at the same electrical potential because they are formed by the same layer of polysilicon (26), it is not possible to apply to them different voltages allowing to optimize the injection efficiency with the efficiency afforded by a conventional split-grid memory cell of the type shown in FIG. 1. Finally, the gate oxide 200a which covers the trench is formed at the same time as a lateral oxide 200b which isolates the selection gate SG from the floating gate 15 FG. It is therefore not possible to separately control the thickness of the gate oxide 200a and that of the lateral oxide 200b. This manufacturing method therefore offers little flexibility for controlling the electrical characteristics of the memory cell, in particular its injection efficiency, its threshold voltage in the vertical channel region L1, and its breakdown voltage. It may therefore be desired to provide an improved split gate memory cell structure, and a method of manufacturing such a memory cell. Embodiments of the invention relate to a memory cell formed on a semiconductor substrate, comprising a vertical selection gate extending in a trench made in the substrate, a floating gate extending above the substrate, and a horizontal control grid extending above the floating gate, in which the floating gate also extends over a portion of the vertical selection grid, over a non-zero overlap distance, the cell memory comprising an electrically floating doped region, located at the intersection of a vertical channel region extending opposite the selection gate and a horizontal channel region extending opposite the floating gate. According to one embodiment, the trench is covered with a dielectric layer comprising a region of greater thickness near the surface of the substrate, and the floating gate has a protuberance which extends below the surface of the substrate in the region of greater thickness of the dielectric layer and has a face opposite a portion of the vertical selection gate.
    [0007] According to one embodiment, the memory cell comprises a buried layer forming a collective source plane for collecting programming currents from the memory cell and memory cells formed on the same substrate. Embodiments of the invention also relate to a group of memory cells comprising first and second memory cells according to the invention, sharing the same vertical selection grid. Embodiments of the invention also relate to a memory circuit, comprising a memory plane comprising a plurality of memory cells according to the invention. Embodiments of the invention also relate to a memory circuit comprising at least one memory cell according to the invention, and memory cell programming means configured to apply to the substrate, to the vertical selection gate, to the memory cell. horizontal control grid and to the drain and source regions of the memory cell, electrical potentials such as hot electrons are injected into the floating gate. Embodiments of the invention also relate to a memory circuit comprising memory cells according to the above-described embodiment having a protrusion which extends below the surface of the substrate in the region of greater thickness of the dielectric layer and which has a face opposite a part of the vertical selection gate, and means for erasing the tunneling memory cell configured to apply to the vertical selection gate and the horizontal control gate of the memory cell. Electrical potentials such as electrical charges are extracted from the floating gate and collected by the vertical selection gate via the protuberance of the floating gate and the dielectric material extending between the protuberance and the vertical selection gate.
    [0008] Embodiments of the invention also relate to a method of manufacturing on a semiconductor substrate an electrically programmable memory cell, comprising the steps of: etching a trench in the substrate via a hard mask depositing on the walls of the trench a first dielectric layer, depositing on the substrate a first conductive layer and etching the first conductive layer to form a vertical selection gate extending in the trench, depositing on the substrate a second dielectric layer, depositing on the second dielectric layer a second conductive layer, and etching the second conductive layer to form a floating gate, wherein the second conductive layer is etched so that the floating gate partially overlaps the vertical selection gate over a distance non-zero recovery, and including, before the deposit of the first dielectric layer on the walls of the trench, a step of inclined implantation of dopants in a region of the substrate located at the upper edge of the trench and under the hard mask, the implantation being made via a vertical wall of the trench, to form in the memory cell an electrically floating doped region located at the intersection of a vertical channel region extending in front of the selection gate and a horizontal channel region extending in front of the floating gate. According to one embodiment, the second conductive layer is etched from a photolithography plane defining between the proximal edge of the floating gate and the corresponding proximal edge of the vertical selection gate a theoretical overlap distance at least equal to one photolithography tolerance of the manufacturing process. According to one embodiment, the method comprises a preliminary step of implanting in the substrate a conductive plane forming a source line 30 for the memory cell. According to one embodiment, the method comprises a step of making in the dielectric layer covering the trench, a region of greater thickness located near the surface of the substrate. According to one embodiment, the method comprises a step of making a hollow in the region of greater thickness of the dielectric layer. According to one embodiment, the recess is shaped to extend below the surface of the substrate and so that the floating gate has a protuberance extending into the recess and having a face facing a portion vertical selection grid. According to one embodiment, the method comprises steps of depositing a third dielectric layer on the second conductive layer and depositing a third conductive layer on the third dielectric layer, and a step of simultaneous etching of the third conductive layer and the second conductive layer, to form a horizontal control grid on the floating gate. Embodiments of the invention also relate to a method of manufacturing an integrated circuit on a semiconductor wafer, including the method of manufacturing a memory cell according to the invention. These objects and features of the present invention will be better understood on reading the following description of embodiments of a method for manufacturing a memory cell according to the invention, and examples of memory cells produced according to this method. , in a nonlimiting manner in relation to the attached figures among which: FIG. 1 previously described shows a conventional memory plane architecture comprising divided grid memory cells; FIG. 2 previously described is a sectional view of FIG. A conventional split-grid memory cell having a vertical selection gate, FIG. 3 previously described shows a memory array architecture receiving the memory cell of FIG. 2; FIGS. 4 to 18 are sectional views showing FIGS. steps of a method of manufacturing a memory cell according to the invention, - Figures 19A to 19C show memory cells r 20 and 21 show complementary steps of manufacturing an integrated circuit comprising a memory cell according to the invention, FIG. 22 illustrates a method of programming a memory cell according to the invention. FIG. 23 illustrates a method of erasing a memory cell according to the invention; FIGS. 24 and 25 illustrate another method of erasing a memory cell according to the invention, FIG. 26 shows a memory array architecture 25 comprising a memory cell according to the invention, and - 27 shows an example of a memory circuit comprising memory cells according to the invention. Figures 4 to 18 show, in sectional views, the steps of a method of manufacturing a memory cell according to the invention. Figures 13A and 13B show two variants of a step of this method. FIGS. 14A-14C and 15A-15C show three variants of two other steps of this method. FIGS. 19A to 19C show the three variants C31, C32, C33 of a memory cell C3 according to the invention realized by means of this method and its variants. The memory cell C3 (C31, C32, C33) shown in Figs. 19A, 19B, 19C comprises a horizontal floating gate FG formed on a P-type PW substrate, a horizontal control gate CG extending over the floating gate FG , the assembly forming a grid stack FG / CG, and an SG vertical selection grid 10 formed in a trench 10 made in the substrate, the trench being covered with a dielectric layer D1. The memory cell C3 is produced here at the same time as a twin memory cell C3 '(C31', C32 ', C33') using the same selection grid SG. The floating gate FG extends over a portion of the selection grid SG. The distance Dov between the proximal edge of the floating gate FG and the corresponding proximal edge of the selection gate SG is here negative, and is referred to as the "overlap distance" in the following. The memory cell C3 also comprises a n-type buried nO source region, connected to an SCL source line, a n-type floating drain-source region n1 which is not connected to any conductive line or source of potential, and an N-type drain region n2 to be connected to a bit line. The memory cell C3 can be seen as a combination of a horizontal floating gate transistor and a vertical selection transistor, the horizontal floating gate transistor comprising the gate stack FG / CG, the region n2 as the region of drain and region n1 as the source region, the vertical selection transistor comprising the vertical selection gate SG, the region n1 as the drain region, and the region n0 as the source region. However, the drain-source region n1 is here floating and the assembly actually forms the equivalent of a single transistor of the split gate type, programmable by hot electron injection and extending between the source region nO and the drain region n2. Thus, the region n1 is called the "drain / source" region by analogy with a memory cell structure with two transistors of the aforementioned type, but does not provide electron collection here. The region nl is intended to improve the reliability of the memory cell over time, protecting it against certain types of degradation occurring with its aging, as will be detailed below.
    [0009] The variants C31, C32 of the memory cell C3, shown in Figures 19A, 19B, comprise, in the dielectric layer D1, a region D1 'of greater thickness, located near the surface of the substrate PW. In the embodiment shown, the thickness of the region D1 'increases as one approaches the surface of the substrate and has, in sectional view, a profile substantially in the shape of a "V", the D1 dielectric layer having a corresponding profile substantially shaped "Y". As a numerical example, the dielectric layer D1 has a thickness of the order of 5 to 10 nm (nanometers), and the region D1 'a thickness of the order of 15 to 20 nm without its close to the surface of the substrate.
    [0010] Moreover, the lower face of the floating gate FG of the variant C31 of the memory cell C3 comprises a protuberance p15 which extends below the surface of the substrate PW, in the region D1 'of the layer D1, and presents a face opposite a part of the selection grid SG. Figure 4 shows a preliminary stage of realization of the memory cell C3. A deep doped layer NL has been implanted in a semiconductor substrate wafer or WF wafer. This layer is for example the N-type insulation layer of a P type box forming the PW substrate in which the memory cell is made. This layer will serve as SCL source line to all the memory cells implanted in the PW substrate, more specifically a collective source plane, able to collect the programming currents of several memory cells. Shallow trench insulation (STI) type trenches, not visible in FIG. 4 because located in a section plane parallel to that of the figure, may have been made on the surface of the substrate, in the part of a collective fabrication of several rows of memory cells. A SOX sacrificial oxide layer was then deposited on the surface of the PW substrate. During a step illustrated in FIG. 5, a hard mask HM1 ("hard mask") is formed on the SOX oxide layer, by deposition or growth of one or more solid layers, for example silicon oxide or silicon nitride or a combination of these materials. A photoresist mask PH is then deposited on the mask HM1, and is then developed to form an opening 1 in the mask PH. During a step whose result is illustrated in FIG. 6, the mask HM1 was etched through the resin mask PH so as to form a corresponding opening 1 in the mask HM1, and the mask PH was then removed. . In a step illustrated in FIG. 7, the trench 10 is formed in the PW substrate by etching it through the aperture 1 of the HM1 mask. The etching process used is preferably a nonselective and anisotropic dry etching process, such as a plasma etching process. The depth of the trench 10 is here less than the implantation depth of the doped layer NL. As a numerical example, the trench has a depth of 450 nm and the NL layer is implanted at 750 nm from the surface of the substrate. In a step illustrated in FIG. 8A, a deep doped pocket forming the nO region is implanted into the substrate via the trench 10, near the bottom of the trench 10. The nO region is formed by vertical ion implantation, and remains located in the region of the substrate located in the vicinity of the bottom of the trench 10. The nO region extends to the doped layer NL and thus serve as a source region to the memory cell in the process of formation, while the NL doped layer will serve as an SCL source line in the continuity of the nO source region. In an alternative embodiment, the region n0 is not implanted and the trench 10 is etched to a greater depth so as to reach the NL layer, which will serve as the source and source line region. In a step illustrated in FIG. 8B, a shallow doped pocket forming the floating drain-source region n1 is implanted in a region of the substrate PW extending under the hard mask HM1 and the oxide layer SOX. . Implantation is done by ion implantation with a tilt angle (implantation angle), through the vertical wall of the trench and using the hard mask as a doping mask, along an inclined implantation axis. shown by an arrow A1. The angle of inclination of the implantation axis A1 relative to a vertical axis VA perpendicular to the substrate is chosen so that the region n1 extends below the surface of the substrate PW and that does not extend too much in the depth of the substrate, to subsequently allow the appearance, between the regions nO and nl and in front of the trench 10, of a vertical conductive channel. Current technologies offer an angle of implantation of up to 45 ° relative to the vertical axis VA. This implantation step is followed here by the implantation of the corresponding floating region n1 of the twin cell C3 ', via the other vertical wall of the trench 10, along an implantation axis Al' having a angle of inclination opposite that of the implantation axis A1, relative to the vertical axis VA. During a step illustrated in FIG. 9, the dielectric layer DI is then formed on the walls of the trench 10, for example by oxide growth. In a step illustrated in FIG. 10, a conductive layer P1, for example made of polysilicon, is deposited on the entire substrate, as well as inside the trench 10. During a step illustrated in FIG. 11, the layer P3 is etched so as to no longer remain on the surface of the substrate, except inside the trench 10 where it forms the selection gate SG. This step includes simultaneous etching of the SOX sacrificial oxide layer, or is followed by a wet etch step of the SOX layer.
    [0011] The following steps, illustrated in FIGS. 12, 13A, 13B, aim at producing the thicker region D1 'in the dielectric layer D1 of cells C31 (FIG 19A) and C32 (FIG 19B).
    [0012] In the step illustrated in FIG. 12, a layer of high voltage dielectric DHV is deposited on the whole of the substrate, here by growth of a thermal oxide such as silicon dioxide SiO 2, for example over a thickness of order of 10 to 15 nm. This deposition can be carried out in one or more steps and the oxide formed comes in part from the oxidation of the material forming the substrate PW, here silicon. This oxidation causes the region D1 'to appear in the vicinity of the surface of the substrate, on the one hand by oxidation of the material of the substrate facing the trench 10, in this case silicon, and on the other hand by oxidation of the material forming the vertical grid SG, here polysilicon. Equal oxidation of silicon and polysilicon on each side of the dielectric layer D1 results in the substantially V-shaped region of the D1 'region. The duration of this oxidation step, the thickness of the DHV layer, and the conditions of its implementation make it possible to control the depth and the width of the region Dl '.
    [0013] It will be noted that this step is optional with respect to the manufacturing method of the memory cell, but may be necessary in the context of the simultaneous realization of high voltage transistors present in other parts of the circuit in which the memory cell is integrated.
    [0014] In the step shown in any one of FIGS. 13A, 13B, the DHV dielectric layer is removed by etching, for example by means of a technique called "BOE" ("Buffered Oxide Etch") based on hydrofluoric acid. (HF). It follows a partial removal of the dielectric of the region Dl ', leading to the appearance of a hollow 15 (Figure 13A) or a hollow 16 (Figure 13B) whose depth depends on the conditions in which this step of etching is carried out, and can be controlled by those skilled in the art. Thus, in the exemplary embodiments shown, the region D1 'of the memory cell C31 in formation (FIG 13A) has a recess 15 which extends opposite the buried gate SG, below the surface of the substrate. The hollow is here in the shape of a point, because of the "V" shape of the region D1 '. On the other hand, the region D1 'of the memory cell C32 in formation (FIG 13B) only has a superficial hollow 16.
    [0015] In the step shown in any one of FIGS. 14A, 14B, 14C, a tunnel dielectric layer D2, for example silicon oxide, is deposited on the whole of the substrate, for example over a thickness of order of 7 to 10 nm. FIG. 14A shows the profile of the region D1 'after deposition of the dielectric layer D2 on the memory cell in formation C31. The hollow 15 is only partially filled by the dielectric D2 and always extends opposite the buried vertical grid SG, below the surface of the substrate. FIG. 14B shows the profile of the region D1 'after deposition of the dielectric layer D2 on the memory cell in formation C32. The surface hollow 16 is almost completely filled by the dielectric D2. FIG. 14C shows the profile of the layer D2 after it has been deposited on the memory cell in formation C33. The substrate of the memory cell C33, as shown in FIG. 11, has not been subjected to the high-voltage dielectric deposition step and therefore does not include the region D1 '. In the step shown in any one of FIGS. 15A, 15B, 15C, a conductive layer P2, for example made of polysilicon, is deposited on the entire substrate. FIG. 15A shows the profile of the layer P2 after depositing it on the memory cell in formation C31. The lower face of the layer P2 has the aforementioned protuberance p15 which extends into the recess 15 below the surface of the substrate, and has a face opposite a part of the selection gate SG. FIG. 15B shows the profile of the layer P2 after depositing it on the forming memory cell C32. The lower face of the layer P2 has a protuberance p16 of small extent which extends in the surface recess 16. FIG. 15C shows the profile of the layer P2 after depositing it on the memory cell in formation C33. In this embodiment, the underside of the layer P2 has no irregularities. In what follows, the protuberance p16 will be considered negligible and the memory cell C32 as equivalent to the memory cell C33. Figures 16, 17 and 18 illustrate the following steps of manufacturing cell C3. The memory cell represented is the cell C31, but these steps are also applied to the memory cells C32, C33, which are not represented for the sake of simplicity. In the step illustrated in FIG. 16, the layer D2 is covered with a dielectric layer D3, for example an oxide called "interpoly" of the ONO type (oxide-nitride-oxide). The layer D3 is then covered with a conducting layer P3, here in polysilicon, and the layer P3 is then covered by a hard mask HM2. In the step illustrated in FIG. 17, the hard mask HM2 is etched by photolithography so as to keep only two mask portions HM2-1, HM2-2 corresponding to the gate stack FG / CG to be produced. In the step illustrated in FIG. 18, the layers D2, P2, D3 and P3 are etched by anisotropic dry etching. The regions protected by the mask portions HM2-1, HM2-2 are not etched and form the stack of grids FG / CG comprising the tunnel dielectric layer D2, the floating gate FG, the dielectric layer D3, and the gate CG control. The positioning of the mask portions HM2-1, HM2-2 determines the position of the gate stack FG / CG relative to the vertical gate SG. This positioning must be determined during the design of the photolithography plane ("layout") of the memory cell. For this purpose, the designer must define a theoretical overlap distance Dovt corresponding to the targeted recovery distance Dov taking into account a tolerance "T" of the manufacturing process. The obtained overlap distance Dov is equal to the theoretical overlap distance Dovt plus or minus this tolerance, and thus lies within the open interval] Dovt-T; Dovt + T [(the tolerance T considered here being a limit error which is not supposed to be achieved by the manufacturing process).
    [0016] In one embodiment, the theoretical overlap distance is Dovt = T, to obtain a recovery distance Dov in the range] 0; 2T [. In other words, the memory cells produced have, between the proximal edge of the gate stack FG / CG and the corresponding proximal edge of the vertical grid SG, a recovery distance Dov ranging from a value close to zero. at a value close to 10 2T, the value close to zero corresponding to an almost perfect alignment of the stack of grids relative to the vertical grid SG. By way of example, with a manufacturing method 15 making it possible to produce a floating gate FG with a length of the order of 120 to 150 nm, a typical tolerance value T is of the order of 20 nm, and the The width of the vertical grid SG is of the order of 150 to 300 nm. The overlap distance Dov is then in the range] 0; 40 nm [. FIGS. 19A, 19B, 19C previously described represent the three variants C31, C32, C33 of the memory cell C3 obtained after the finalization steps of the manufacturing method, in particular: the removal of the mask portions HM2-1, HM2 -2, - the deposition of a lateral dielectric layer D4 on the vertical walls of the gate stack FG / DG, and - the formation in the PW substrate of a doped pocket 30 forming the drain region n2, by implantation vertically self-aligned on an outer edge of the gate stack FG / CG, this step for forming simultaneously the drain region n2 of the twin memory cell C3 '(C31', C32 ', C33'). These steps may be followed by complementary steps to achieve a complete integrated circuit. For example, as shown in FIG. 20, the drain regions n2 of the twin cells C3, C3 'are then connected to a bit line BL made in a first level of metal or "metal 1", via a conducting vias V1 passing through a dielectric layer D5 covering the memory cells. Similarly, the selection grid SG can be connected to a selection line SL formed in a second level of metal or "metal 2" via a set of conductive vias V2 passing through the dielectric layer D5 and to through a dielectric layer D6 covering the level of metal "metal 1".
    [0017] As shown in FIG. 21, the doped layer NL as source line SCL, here a source plane, can be connected by conductive vias V3 to a set of surface contacts making it possible to apply to the layer NL a potential source line. FIG. 22 illustrates a method of programming the memory cell C3 by hot electron injection. The memory cell represented is the cell C33 but the method is applicable to the other variants C31, C32 of the memory cell. The drain region n2 receives a positive drain voltage VD1, for example 4V. The control gate CG receives a positive programming voltage VG11, for example 10V. The selection gate SG receives a positive selection voltage VS1, for example between 1 and 3V. The doped layer NL receives a source voltage VSC1 of zero value (circuit ground). The twin memory cell C3 ', which is connected to the same bit line and thus also receives the voltage VD1, receives on its control gate CG a negative or zero programming inhibition voltage VG12, for example between -2V and OV. The biasing of the gates CG, SG causes the C3 memory cell to show a horizontal channel region CH1 extending below the floating gate FG, a vertical channel region CH2 extending opposite the selection gate SG. A current flows from the drain region n2 to the source region n0 of the memory cell. An electron flow EF flows in the opposite direction of this current, shown schematically by a thick line in the figure. The electron flow EF passes through the vertical channel region CH2 and then the channel region CH1 to join the drain region n2. The floating drain-source region n1 is located at the vertical channel region intersection CH2 and the horizontal channel region CH1. Studies show that the EF electron flow bypasses the nl region, as shown very schematically in Fig. 22, because the nl region, due to its N-type doping, weakens the kinetic energy of the electrons and forms a region so-called "cold" or few electrons circulate. The channel region CH1 also has an injection zone where hot electrons present in the electron flow EF are injected into the floating gate FG under the effect of a transverse electric field created by the voltage VG11. This region is substantially located at the intersection of the vertical channel CH2 and the horizontal channel CH1, near the floating gate FG, and is here at the edge of the region n1 since the hot electrons do not pass through it.
    [0018] The nl-doped region makes it possible to reduce the appearance of defects in the memory cell with the aging thereof, or "cycling" (that is to say, as the latter undergoes cycles of erasure and programming), and also reduces the effect of such defects on the electrical properties of the memory cell. Such "defects" may include fixed charges, trapped charges, trapped charges of interface states, and other types of electrical charges: - fixed charges are electrical charges occurring at the interface between the PW substrate and the dielectric D1 in response to electrical stresses experienced by the memory cell or induced by oxidation phenomenon, and which can be caused by breaks in atomic bonds such as Si-H silicon-hydrogen bonds, the trapped charges are charges. 20 stored in the volume of the dielectric layer Dl. The appearance of these trapped electrical charges can be caused by the breaking of atomic bonds, for example Si-O bonds in the case of a SiO silicon dioxide dielectric, or oxygen vacancies in the SiO 2 bond. oxide, the trapped charges of the interface states are electric charges trapped by the defects which result from the discontinuity of the crystal lattice, for example at the Si-SiO 2 interface, electrical charges can also be generated by impurities metal or by defects caused by radiation, or by mechanisms of breaking atomic bonds. Unlike fixed charges or trapped charges, the latter two types of charges can change with the electric potential of the substrate and can strongly influence the electrical characteristics of the memory cell. In the absence of region n1, these various "defects" progressively modify the electrical properties of the memory cell, for example its programming or erasing efficiency, its threshold voltage, its channel length, etc. Thus, for example, charges entrapped in the volume of the oxide can affect the conduction modes in the oxide and decrease the performance of the cell in erasure. They are also liable to cause an increase in the threshold voltages of the floating gate transistor and selection transistor regions, which then leads to a decrease in the saturation current and therefore a reduction in the injection efficiency (ie slower programming or requiring higher voltages). The trapped charges of the interface states can also affect the conduction properties of the two transistor regions, in particular their slope under the threshold voltage, which also results in a reduction of the injection efficiency and a degradation of the characteristics of the transistor. reading of the memory cell.
    [0019] Thus, region n1 makes it possible to "mask" or electrically neutralize an area of the memory cell that is conducive to the generation of interface states and trapped charges. Indeed, this zone is on the one hand covered with the lateral dielectric layer D4 (FIG 19A to 19C) which extends on the edges of the stack of grids FG / CG and is capable of receiving a layer of nitride which can aggravate the "trapping" of charges. Also, in the case of the embodiment shown in FIG. 19B, the "V" -shaped region D1 'extending opposite the trench 10 undergoes, during the production of the memory cell, a succession of oxidations. and wet etchings that can alter its reliability.
    [0020] This zone being rendered electrically inactive by the region n1, it no longer intervenes in the formation of the channel CH1 or CH2 when the memory cell is put in the conductive state (programming or reading of the memory cell) As shown schematically in the figure 22, the flow of electrons EF through the memory cell is deflected and away from this area. Its possible degradation therefore no longer affects the electrical properties of the memory cell.
    [0021] The region n1 also makes it possible to locally reduce the intensity of the electric fields during the programming and erasure phases, which can accelerate the formation of the aforementioned defects.
    [0022] In return, the prediction of the region n1 may have the effect of reducing the injection efficiency of the memory cell relative to a memory cell which would be devoid of the region n1. However, this is largely offset by the advantages offered in terms of the stability of its electrical properties over time. FIG. 23 illustrates a method of erasing via the channel of the memory cell C3. The memory cell shown is the C33 cell but the method is applicable to the variants C31, C32 of the memory cell. Region n2 receives a drain voltage VD2 of zero value. The control gate CG receives a negative erase voltage VG21, for example -10V. The selection gate SG receives a positive selection voltage VS2, for example 5 V. The doped layer NL receives a positive source voltage VSC2, for example 5V. The PW substrate is then brought to a voltage VB equal to the voltages VS2 and VSC2, for example 5 V. Electrons are torn from the floating gate FG via the substrate and are collected by the source line NL / SCL. The twin memory cell C3 'receives on its control gate CG a positive clearing inhibition voltage VG22, for example 2.5 V.
    [0023] FIG. 24 illustrates an erasing method according to the invention of the memory cell C31, carried out via the vertical grid SG. This method is applicable only to the cell C31 and involves the protrusion p15 of the floating gate FG. The region n2 receives a drain voltage VD3 of zero value. The control gate CG receives a negative erase voltage VG31, for example -10V. The selection gate SG receives a positive selection voltage VS3, for example 5 V. The doped layer NL here receives a source voltage VSC3 of zero value. The voltage VB of the substrate PW is therefore zero. The twin memory cell C33 'receives on its control gate CG an erase inhibition voltage VG32 which is not necessarily positive and can be zero because the voltage VB is itself zero. Under the effect of the voltage VS3, electrons are torn from the floating gate FG by the vertical grid SG, and are collected by the word line to which it is connected. This process is illustrated in more detail in FIG. 25. Between the protuberance p15 and the vertical grid SG extends a dielectric material forming part of the region D1 ', which comprises in part dielectric material of the dielectric layer D2, and a composite dielectric material D12 which comprises the original dielectric material of the layer D1 combined with residues of the high voltage dielectric material DHV. The distance between the protuberance p15 and the gate SG, of the order of a few tens of nanometers, allows the appearance of the tunnel effect between these two elements. This method of erasure by the gate SG, compared to the erasure method by the channel, has various advantages. In particular, the erasure does not take place through the same dielectric region as the programming, which reduces the electrical stress of the dielectric material and its aging. Furthermore, the twin memory cell does not undergo erasure stress (slow parasitic erasure) because the substrate voltage remains zero, unlike the channel erase method.
    [0024] It will be noted that the sectional plane of FIG. 20, and generally the sectional plane of FIGS. 4 to 19, 22 to 25, is perpendicular to the sectional plane of FIG. 2. In FIG. bit BL is parallel to the section plane while the bit line (21) of FIG. 2 is perpendicular to the section plane. Similarly, the selection line SL is perpendicular to the sectional plane in FIG. 20 and the selection line formed by the gate material (26) is parallel to the sectional plane in FIG. "Source bit lines" (20) imposed by the memory cell structure of FIG. 2 are replaced, with a memory cell structure according to the invention, by the SCL source line doped NL layer and more precisely a plane source for all the memory cells implanted in the same box PW, able to collect the programming currents of several memory cells. The memory cell structure according to the invention therefore leads to a simpler programmable memory array architecture by hot electron injection than that imposed by the memory cell structure of FIG. 2. FIG. 26 shows such a plan architecture. memory. Only two pairs of twin memory cells C3i ,, C3i + 1, i, respectively C3i, j + 1, C3i + 1, j + 1 are represented. The memory cells C3i ,, C3i, + 1 belong to a word line WLi and the memory cells C3i, 1, j, C3i, 1, j, 1 belong to a twin word line WLili. The selection gates SG of the selection transistor sections ST of the memory cells are connected to the same selection line SLi, i + 1 and the sources S of the four memory cells are connected to the same source SCL (formed by the buried layer NL). The control gates CG of the floating gate transistor sections FGT of the memory cells C3i, j and C3i + 1, i are connected to a gate control line CGLi, and the control gates CG of the memory cells C3i + 1, i and C3i + 1, i + 1 are connected to a CGLiil gate control line. The drains of the twin memory cells C3i ,, C3i + 1, i are connected to a bit line BLi and the drains of the twin memory cells C3i, + 1, C3i + 1, j + 1 are connected to a bit line BLi laughed. The memory plane therefore comprises only one bit line per vertical row of memory cells. Each word line WLi, WLi + i comprises only one CGLi, CGLiil and a selection line CLi, i + 1 common to the twin word line. The bit line BLi receives a drain voltage VDi and the bit line BLj + i receives a drain voltage VDiil. The gate control line CGLi receives a gate voltage VGi and the gate control line CGLiil receives a gate voltage VGili. The selection line SLi, i1 receives a selection voltage VSj_iil. The common source line SCL, here a source plane, receives the source voltage VSC. The memory array therefore comprises a reduced number of interconnection lines and its structure is similar to that of a memory array of the type shown in FIG. 1, while benefiting on the one hand from the advantage offered by a cell. memory having a vertical selection transistor section, in terms of bulk, and secondly the advantage of having separate control and selection gates for optimization of the programming process. It will be clear to one skilled in the art that the method according to the invention is susceptible of various other embodiments and applications.
    [0025] In particular, although the formation of two twin memory cells has been described in the foregoing, one embodiment of the method according to the invention can aim at the realization of "unitary" memory cells, that is to say without twin memory cell sharing the same vertical selection grid SG. Conversely, embodiments can aim at the collective and simultaneous realization of one or more rows of twin memory cells, for example in the context of the realization of an electrically erasable and programmable MEM1 memory circuit of the type shown in FIG. 27. The MEM1 circuit is formed on a semiconductor wafer and forms an integrated circuit IC. It comprises twin word lines WLi, WLi + i of the type shown in FIG. 26, made on the PW substrate, two twin word lines WLi, WLi + i comprising twin memory cells sharing the same selection line SLi, i .
    [0026] The selection lines SL and the control lines CGL are connected to a word line decoder WLDEC which applies erase voltages, programming and read memory cells. The bit lines BL connected to the drain regions n2 of the memory cells are connected to a set of programming locks BLT and to a set of read amplifiers SA via a column decoder CDEC. These elements are connected to a control circuit CCT which provides the sequencing of programming and erasing operations in accordance with one of the methods described above. It will be clear to those skilled in the art that a memory cell according to the invention is capable of being produced in other technological sectors, the materials cited in the foregoing description, in particular silicon, of silicon dioxide. , polysilicon, being only examples.
    [0027] Similarly, the method described above, forming the Dl 'region of the dielectric layer Dl and forming the protrusion p15, is only one embodiment. Other techniques may make it possible to produce a floating gate FG comprising a protuberance making it possible to erase the memory cell via the selection gate. The method described merely has the advantage of not requiring an additional manufacturing step to make the protrusion, when the deposition of a high voltage dielectric material on the substrate is rendered.
    权利要求:
    Claims (15)
    [0001]
    REVENDICATIONS1. Memory cell (C3, C31, C32, C33) formed on a semiconductor substrate (WF, PW), comprising a vertical selection grid (SG) extending in a trench (10) formed in the substrate, a floating gate (FG) extending above the substrate, and a horizontal control grid (CG) extending above the floating gate (FG), characterized in that the floating gate (FG) also extends to the above a portion of the vertical selection grid (SG), over a non-zero overlap distance (Dov), and in that it comprises an electrically floating doped region (nl) implanted at the intersection of a vertical channel region (CH2) extending opposite the selection gate and a horizontal channel region (CH1) extending opposite the floating gate.
    [0002]
    2. memory cell (C31, C32) according to claim 1, wherein the trench (10) is covered with a dielectric layer (D1) comprising a region (Dl ') of greater thickness near the surface of the substrate, and the floating gate has a protuberance (p15) which extends below the surface of the substrate in the region (Dl ') of greater thickness of the dielectric layer (D1) and has a face facing a portion of the vertical selection grid (SG).
    [0003]
    3. Memory cell according to one of claims 1 and 2, comprising a buried layer (NL) forming a collective source plane (SCL) for collecting programming currents of the memory cell and memory cells formed on the same substrate. .
    [0004]
    4. A group of memory cells comprising a first (C3) and a second (C3 ') memory cells according to one of claims 1 to 3, sharing the same vertical selection grid (SG).
    [0005]
    5. Memory circuit (IC, MEM1), characterized in that it comprises a memory array comprising a plurality of memory cells (C31, C32, C33) according to one of claims 1 to 4.
    [0006]
    6. Memory circuit (IC, MEM1) comprising at least one memory cell according to one of claims 1 to 3, and means (CCT) for programming the memory cell configured to apply to the substrate (PW), to the vertical selection grid (SG), to the horizontal control grid (CG) and to drain regions (n2) and source (nO) of the memory cell, electrical potentials such as hot electrons are injected into the floating gate (FG).
    [0007]
    A memory circuit (IC, MEM1) comprising memory cells according to claim 2 and means (CCT) for deleting the tunneling memory cell configured to apply to the vertical selection grid (SG) and the gate Horizontal control (CG) of the memory cell of electric potentials such as electrical charges are extracted from the floating gate and collected by the vertical selection grid (SG) via the protrusion (p15) of the floating gate and dielectric material (D1 ', D2) extending between the protuberance (p15) and the vertical selection gate (SG).
    [0008]
    A method of manufacturing on a semiconductor substrate (WF, PW) an electrically programmable memory cell (C3, C31, C32, C33), comprising the steps of: etching a trench (10) in the substrate, by intermediate a hard mask (HM1), - deposit on the walls of the trench a first dielectric layer (D1), - deposit on the substrate a first conductive layer (P1) and etch the first conductive layer to form a grid of vertical selection (SG) extending in the trench (10), - depositing on the substrate a second dielectric layer (D2), - depositing on the second dielectric layer (D2) a second conductive layer (P2), and - etching the second conductive layer so as to form a floating gate (FG), characterized in that the second conductive layer (P2) is etched so that the floating gate (FG) partially overlaps the vertical selection gate (SG) on a distance from overlap (Dov) nonzero, and in that it comprises, before the deposition of the first dielectric layer (Dl) on the walls of the trench, a step of inclined implantation of dopants in a region (nl) of the substrate (PW) located at the upper edge of the trench (10) and under the hard mask (HM1), the implantation being made via a vertical wall of the trench (10), to form in the memory cell an electrically floating doped region (nl) located at the intersection of a vertical channel region (CH2) extending in front of the selection gate (SG) and a horizontal channel region (CH1) extending in front of the floating gate (FG). 35
    [0009]
    The method of claim 8, wherein the second conductive layer (P2) is etched from a photolithography plane defining between the proximal edge of the floating gate (FG) and the corresponding proximal edge of the vertical selection grid. (SG) a theoretical overlap distance (Dovt) at least equal to a photolithography tolerance (T) of the manufacturing process.
    [0010]
    10. Method according to one of claims 8 and 9, comprising a preliminary step of implanting in the substrate a conductive plane (NO, NL) forming a source line for the memory cell.
    [0011]
    11. Method according to one of claims 8 to 10, comprising a step of making in the dielectric layer (Dl) covering the trench, a region (Dl ') of greater thickness located near the surface of the substrate.
    [0012]
    12. The method of claim 11 including a step of providing a recess (15, 16) in the thicker region (D1 ') of the dielectric layer (D1).
    [0013]
    The method of claim 12, wherein the recess (15) is shaped to extend below the surface of the substrate and so that the floating gate (FG) has a protrusion (p15) extending in the hollow (15) and having a side facing a portion of the vertical selection grid (SG).
    [0014]
    14. Method according to one of claims 8 to 13, comprising steps of depositing a third dielectric layer (D3) on the second conductive layer (P2) and depositing a third conductive layer (P3) on the third dielectric layer (D3), and a step of simultaneously etching the third conductive layer (P3) and the second conductive layer (P2), to form a horizontal control grid (CG) on the floating gate (FG). 10
    [0015]
    15. A method of manufacturing an integrated circuit (IC) on a semiconductor wafer (WF), including the method of manufacturing a memory cell according to one of claims 8 to 14.
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3017746A1|2015-08-21|VERTICAL MEMORY CELL HAVING NON-SELF-ALIGNED FLOATING DRAIN-SOURCE IMPLANT
    FR3012672A1|2015-05-01|MEMORY CELL COMPRISING NON-SELF-ALIGNED HORIZONTAL AND VERTICAL CONTROL GRIDS
    EP1495496B1|2009-11-18|Non-volatile, programmable, electrically erasable memory semiconductor device having a single grid material layer and corresponding magnetic core plane
    FR3012673A1|2015-05-01|HOT CARRIER INJECTION PROGRAMMABLE MEMORY AND METHOD OF PROGRAMMING SUCH A MEMORY
    FR3030883A1|2016-06-24|VERTICAL SELECTION GRID MEMORY CELL FORMED IN A FDSOI TYPE SUBSTRATE
    FR2996680A1|2014-04-11|NON-VOLATILE MEMORY HAVING VERTICAL SELECTION TRANSISTORS
    FR2692720A1|1993-12-24|EPROM device with single layer of polycrystalline silicon with fast erasure.
    EP0648375B1|1999-12-15|Triple-gate flash eeprom memory and method for making same
    FR2987696A1|2013-09-06|METHOD FOR READING WRITTEN NON-VOLATILE MEMORY CELLS
    FR3021804A1|2015-12-04|DUAL NON-VOLATILE MEMORY CELL COMPRISING AN ERASING TRANSISTOR
    FR3021803A1|2015-12-04|MEMORY CELLS TWINLY ACCESSIBLE INDIVIDUALLY READING
    FR2953643A1|2011-06-10|MEMORY CELL FLASH ON SEOI HAVING A SECOND CHECK GRID ENTERREE UNDER THE INSULATING LAYER
    FR2987700A1|2013-09-06|NON-VOLATILE MEMORY COMPRISING MINI FLOATING POTENTIAL HOUSINGS
    EP0675547B1|1999-05-12|Electrically programmable memory cell
    WO2009147347A1|2009-12-10|Read-only memory with eeprom structure
    EP1994567A2|2008-11-26|Mos transistor with adjustable threshold
    EP1863035B1|2011-07-27|EEPROM memory with an improved resistor against transistors breakdown
    EP0543703A1|1993-05-26|Single-transistor electrically programmable integrated memory
    EP0282520A1|1988-09-21|Non-volatile memory with floating grid and without thick oxide.
    EP2613342A2|2013-07-10|Method for manufacturing a double-gate non-volatile memory cell
    FR3054723A1|2018-02-02|COMPACT EEPROM CELL-MEMORY WITH REDUCED TUNNEL INJECTION AREA
    FR3071355B1|2019-08-30|MEMORY CELL-COMPACT EEPROM
    EP0886320A1|1998-12-23|One polysilicon level EEPROM cell with self-aligned tunnel zone
    FR3080949A1|2019-11-08|NON-VOLATILE MEMORY DEVICE OF LOAD TRAPPING TYPE AND METHOD OF MANUFACTURE
    FR2978294A1|2013-01-25|PROCESS FOR MANUFACTURING A NANOCRYSTAL TRANSISTOR
    同族专利:
    公开号 | 公开日
    US10192999B2|2019-01-29|
    US9543311B2|2017-01-10|
    US20150236031A1|2015-08-20|
    FR3017746B1|2016-05-27|
    US20180145183A1|2018-05-24|
    US9876122B2|2018-01-23|
    US20170084749A1|2017-03-23|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20020153546A1|2001-04-20|2002-10-24|Koninklijke Philips Electronics N.V.|Two-transistor flash cell|
    US20060280000A1|2005-01-18|2006-12-14|Mammen Thomas|CAcT-Tg low voltage NVM cells|
    FR2987697A1|2012-03-05|2013-09-06|St Microelectronics Rousset|Method for manufacturing wafer of integrated circuit, involves extending doped area along lower edges from vertical gate to form source area in transistors of cell memories so that transistors comprise channel areas|
    US5278438A|1991-12-19|1994-01-11|North American Philips Corporation|Electrically erasable and programmable read-only memory with source and drain regions along sidewalls of a trench structure|
    US5495441A|1994-05-18|1996-02-27|United Microelectronics Corporation|Split-gate flash memory cell|
    US5675161A|1995-03-28|1997-10-07|Thomas; Mammen|Channel accelerated tunneling electron cell, with a select region incorporated, for high density low power applications|
    US5745410A|1995-11-17|1998-04-28|Macronix International Co., Ltd.|Method and system for soft programming algorithm|
    JPH11162181A|1997-11-26|1999-06-18|Sanyo Electric Co Ltd|Non-volatile semiconductor storage|
    US6496417B1|1999-06-08|2002-12-17|Macronix International Co., Ltd.|Method and integrated circuit for bit line soft programming |
    FR2805653A1|2000-02-28|2001-08-31|St Microelectronics Sa|Serial memory of electrically erasable and programmable read-only type with advanced data reading, for use in computing|
    TW484213B|2001-04-24|2002-04-21|Ememory Technology Inc|Forming method and operation method of trench type separation gate nonvolatile flash memory cell structure|
    EP1451969A2|2001-11-27|2004-09-01|Koninklijke Philips Electronics N.V.|Semiconductor device having a byte-erasable eeprom memory|
    US7339822B2|2002-12-06|2008-03-04|Sandisk Corporation|Current-limited latch|
    US6894339B2|2003-01-02|2005-05-17|Actrans System Inc.|Flash memory with trench select gate and fabrication process|
    CN100514495C|2005-01-20|2009-07-15|义隆电子股份有限公司|Write operation circuit and method applied in nonvolatile separate gate memory|
    US7342272B2|2005-08-31|2008-03-11|Micron Technology, Inc.|Flash memory with recessed floating gate|
    CN100477232C|2006-01-19|2009-04-08|力晶半导体股份有限公司|Non-volatile storage, producing method and operation method thereof|
    US7253057B1|2006-04-06|2007-08-07|Atmel Corporation|Memory cell with reduced size and standby current|
    US7696044B2|2006-09-19|2010-04-13|Sandisk Corporation|Method of making an array of non-volatile memory cells with floating gates formed of spacers in substrate trenches|
    US8093649B2|2008-03-28|2012-01-10|National Tsing Hua University|Flash memory cell|
    US7851846B2|2008-12-03|2010-12-14|Silicon Storage Technology, Inc.|Non-volatile memory cell with buried select gate, and method of making same|
    US8940604B2|2012-03-05|2015-01-27|Stmicroelectronics Sas|Nonvolatile memory comprising mini wells at a floating potential|
    US8901634B2|2012-03-05|2014-12-02|Stmicroelectronics Sas|Nonvolatile memory cells with a vertical selection gate of variable depth|
    FR3012672B1|2013-10-31|2017-04-14|Stmicroelectronics Sas|MEMORY CELL COMPRISING NON-SELF-ALIGNED HORIZONTAL AND VERTICAL CONTROL GRIDS|
    FR3012673B1|2013-10-31|2017-04-14|St Microelectronics Rousset|HOT CARRIER INJECTION PROGRAMMABLE MEMORY AND METHOD OF PROGRAMMING SUCH A MEMORY|
    FR3017746B1|2014-02-18|2016-05-27|Stmicroelectronics Sas|VERTICAL MEMORY CELL HAVING NON-SELF-ALIGNED FLOATING DRAIN-SOURCE IMPLANT|FR3012672B1|2013-10-31|2017-04-14|StmicroelectronicsSas|MEMORY CELL COMPRISING NON-SELF-ALIGNED HORIZONTAL AND VERTICAL CONTROL GRIDS|
    FR3017746B1|2014-02-18|2016-05-27|StmicroelectronicsSas|VERTICAL MEMORY CELL HAVING NON-SELF-ALIGNED FLOATING DRAIN-SOURCE IMPLANT|
    US9917165B2|2015-05-15|2018-03-13|Taiwan Semiconductor Manufacturing Co., Ltd.|Memory cell structure for improving erase speed|
    FR3057393A1|2016-10-11|2018-04-13|StmicroelectronicsSas|INTEGRATED CIRCUIT WITH DECOUPLING CAPACITOR IN A TYPE TRIPLE CAISSON STRUCTURE|
    CN108806751B|2017-04-26|2021-04-09|中芯国际集成电路制造有限公司|Multi-time programmable flash memory cell array, operation method thereof and memory device|
    FR3070534A1|2017-08-28|2019-03-01|StmicroelectronicsSas|PROCESS FOR PRODUCING CAPACITIVE ELEMENTS IN TRENCHES|
    FR3070535A1|2017-08-28|2019-03-01|StmicroelectronicsSas|INTEGRATED CIRCUIT WITH VERTICAL STRUCTURE CAPACITIVE ELEMENT AND METHOD OF MANUFACTURING THE SAME|
    FR3076660B1|2018-01-09|2020-02-07|StmicroelectronicsSas|INTEGRATED CAPACITIVE FILLING CELL DEVICE AND MANUFACTURING METHOD THEREOF|
    US10482975B2|2018-03-16|2019-11-19|Microchip Technology Incorporated|Flash memory cell with dual erase modes for increased cell endurance|
    FR3080949B1|2018-05-04|2021-05-28|St Microelectronics Rousset|NON-VOLATILE LOAD TRAP TYPE MEMORY DEVICE AND METHOD OF MANUFACTURING|
    FR3087027A1|2018-10-08|2020-04-10|StmicroelectronicsSas|CAPACITIVE ELEMENT OF ELECTRONIC CHIP|
    US10762966B2|2018-10-30|2020-09-01|Globalfoundries Singapore Pte. Ltd.|Memory arrays and methods of forming the same|
    US11004785B2|2019-08-21|2021-05-11|StmicroelectronicsSas|Co-integrated vertically structured capacitive element and fabrication process|
    法律状态:
    2015-02-20| PLFP| Fee payment|Year of fee payment: 2 |
    2016-01-21| PLFP| Fee payment|Year of fee payment: 3 |
    2017-01-24| PLFP| Fee payment|Year of fee payment: 4 |
    2018-01-23| PLFP| Fee payment|Year of fee payment: 5 |
    2020-01-22| PLFP| Fee payment|Year of fee payment: 7 |
    2021-01-20| PLFP| Fee payment|Year of fee payment: 8 |
    2022-01-19| PLFP| Fee payment|Year of fee payment: 9 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1451297A|FR3017746B1|2014-02-18|2014-02-18|VERTICAL MEMORY CELL HAVING NON-SELF-ALIGNED FLOATING DRAIN-SOURCE IMPLANT|FR1451297A| FR3017746B1|2014-02-18|2014-02-18|VERTICAL MEMORY CELL HAVING NON-SELF-ALIGNED FLOATING DRAIN-SOURCE IMPLANT|
    US14/625,356| US9543311B2|2014-02-18|2015-02-18|Vertical memory cell with non-self-aligned floating drain-source implant|
    US15/365,768| US9876122B2|2014-02-18|2016-11-30|Vertical memory cell with non-self-aligned floating drain-source implant|
    US15/852,826| US10192999B2|2014-02-18|2017-12-22|Vertical memory cell with non-self-aligned floating drain-source implant|
    [返回顶部]