This application is based upon and claims the benefit of priority from Japanese Patent Application No. 2016-183295, filed Sep. 20, 2016, the entire contents of which are incorporated herein by reference.
Embodiments described herein relate generally to a non-volatile semiconductor memory device.
A NAND type flash memory employs a page buffer as a buffer memory to reduce programming time. The page buffer includes, for example, an inverter latch circuit.
In general, according to one embodiment, a non-volatile semiconductor memory device is disclosed. The device includes a semiconductor substrate, and a memory cell array provided on the semiconductor substrate. The memory cell array includes a plurality of memory transistors which are electrically rewritable and arranged in a three-dimensional manner. The device further includes a latch provided above the semiconductor substrate and configured to hold data that is to be written in the memory cell array. The latch includes a capacitor and a first field-effect transistor which is connected to the capacitor and includes a first oxide semiconductor layer.
Embodiments will be described hereinafter with reference to the accompanying drawings. The drawings are schematic and conceptual, and the dimensions, the proportions, etc., of each of the drawings are not necessarily the same as those in reality. Further, in the drawings, the same reference symbols denote the same or corresponding portions, and overlapping explanations thereof will be made as necessary. In addition, as used in the description and the appended claims, the singular forms “a”, “an”, and “the” are intended to include the plural forms as well, unless clearly described.
As shown in
The memory cell array MA comprises a plurality of memory transistor (not shown) which are electrically rewritable and three-dimensionally arranged as described later. In addition, the memory cell array MA comprises a plurality of word lines WL extending along the X axis of
The memory cell array MA comprises a plurality of conductive layers and a plurality of insulating layers, and the conductive layers and the insulating layers are alternately stacked. A stair-like step portion is provided on the periphery of the memory cell array MA. The stair-like step portion includes a plurality of steps.
As shown in
As shown in
As shown in
As shown in
The stair-like step portion SR (SR1, SR1′, SR2, SR2′) is formed by sliming a resist while etching the stacked conductive layer CL and the insulating layer IL in accordance with a well-known method. Consequently, the stair-like step portion SR (SR1, SR1′, SR2, SR2′) is generally formed to surround four sides of the memory cell array MA (
Referring to
The word line connection circuit SW is a switching circuit to connect the word line WL and the row recorder RD.
The bit connection circuit BLHU and the page buffer PG are arranged in a direction parallel to the Y axis of the stair-like step portion SR.
The page buffer PG includes a latch 11 and a sense amplifier 12. For example, the latch 11 and the sense amplifier 12 are provided for each of the bit lines BL. It is noted that the latch 11 and the sense amplifier 12 may not provided for each of the bit lines BL.
In the data write operation, the page buffer PG temporarily holds data supplied from the outside and transfers the data to the bit line BL, and writes the data collectively in a unit of page. In the data read operation, the page buffer PG senses and amplifies the data read on the bit line BL in the unit of page, and temporarily hold the data in the latch 11 and outputs the data to the outside. The bit connection circuit BLHU is configured to connect the bit line BL to the latch 11, or disconnect the bit line BL from the latch 11. The bit line BL is connected to the latch 11 in the data write operation and the data read operation.
The latch 11 is provided on the stair-like step portion SR, and the sense amplifier 12 is provided on a surface of a semiconductor substrate such as a silicon substrate.
Here, when the latch is formed on the surface of the semiconductor substrate, the latch is required to be formed in an area which is different from an area in which the sense amplifier is to be formed. Consequently, the chip area increases by the area for the latch. In the present embodiment, the latch is provided on the stair-like step portion SR, so that the increase of the chip area (area penalty) caused by latch 11 is suppressed. Therefore, the increase of chip area of the page buffer PG is suppressed even the capacity of the page buffer PG increases.
The capacity of the page buffer PG has increased in order to cope with the advances of the NAND type memory brought by the three-dimensional stacking structure, the high capacity by multiple-valued, or the high bandwidth. Consequently, the NAND type flash memory 10 of the present embodiment, which is capable of suppressing the increase of the chip area, has an advantage in reducing bit cost.
The transistor Tr1 comprises a semiconductor layer which contains oxide semiconductor as a material (hereafter referred to as an oxide semiconductor layer) 1, a gate insulating film 2, and a gate electrode 3. The transistor Tr2 comprises a oxide semiconductor layer 1, a gate insulating film 2 and, a gate electrode 3 similar to the transistor Tr1. The transistor Tr1 and the transistor Tr2 have the same conductivity type for instance.
The oxide semiconductor layer contains, for example, oxide semiconductor such as InGaZnO, InSnZnO, InGaSnZnO, InGaSnO, AlInGaZnO, AlInSnZnO, In2O3, Ga2O3, TiO2, ZnO or SnO2 or the like. The oxide semiconductor layer exhibits n-type conductivity. InGaZnO, ZnO, TiO2, Ga2O3, and SnO2 are wide band-gap semiconductors. The gate insulating film 2 includes, for example, a silicon oxide film. The gate electrode 3 is a metal film (conductive film) which includes tantalum nitride, titanium nitride, tungsten nitride, tantalum, titanium, tungsten, aluminum or molybdenum, or alloy of tantalum, titanium, tungsten, aluminum or molybdenum.
The transistors Tr1 and Tr2 employ the oxide semiconductor layers, so that the transistors Tr1 and Tr2 can be so called junctionless field-effect transistors which do not require p-n junctions. The junctionless field-effect transistor comprises source/drain regions and a channel region, which have the same conductivity type. Consequently, the junctionless field-effect transistor in on-state is turned off by a depletion layer formed in the channel region under the gate electrode.
The transistors Tr1 and Tr2 do not need p-n junctions, so that the manufacturing process of the transistors Tr1 and Tr2 is simplified. Moreover, the transistors Tr1 and Tr2 employ the oxide semiconductor layers, so that the transistors Tr1 and Tr2 maintain high conductivity even the miniaturization of the transistors Tr1 and Tr2 is developed.
The capacitor C includes a first capacitor electrode 4, a capacitor insulating film 5 covering the side surface and the bottom surface of the first capacitor electrode 4, and a second capacitor electrode 6 covering the side surface and the bottom surface of the capacitor insulating film 5. The capacitor C is a trench capacitor formed in the insulating film 7.
One of the source/drain regions in the oxide semiconductor layer 1 of the transistor Tr1 is connected to the upper end of the second capacitor electrode 6 of the capacitor C. The other of the source/drain regions in oxide semiconductor layer 1 of the transistor Tr1 is connected to the sense amplifier (not shown).
One of the source/drain regions in the oxide semiconductor layer 1 of the transistor Tr2 is connected to the upper end of the second capacitor electrode 6 of the capacitor C. The other of the source/drain regions in oxide semiconductor layer 1 of the transistor Tr2 is connected to the bit line (not shown).
The transistors Tr1 and Tr2, which is configured to form the channel by the oxide semiconductor, is expected to have a very low off leak current characteristic of 1×10−22 (A/μm). Consequently, the latch 11 including the capacitor C has a long retention time that is greater than or equal to ten days. That is, it becomes clear that the latch 11 of the embodiment can be used as almost a non-volatile memory. Thus, the lost of data held by the latch 11 is suppressed even the power supply for the page buffer PG is shut down.
On the other hand, a conventional page buffer comprises, for example, an inverter latch (volatile memory) which is constituted by inverters. The data held by the inverter latch may be lost when the power supply is shutdown. Consequently, as the page buffer has a larger memory capacity, the inverter latch may lose larger size of data. However, the latch 11 of the embodiment does not have such a problem (data lost).
Another page buffer further comprises a Fe-RAM (non-volatile memory) in addition to the inverter latch. The page buffer relocates the data in the inverter latch to the Fe-RAM when the power supply is shutdown, thereby suppressing the loss of the data. However, this page buffer requires a larger chip area. Because the Fe-RAM is formed in an area different from the area where the inverter latch is formed. In addition, the page buffer is required to relocate the data at the time of power shutdown and write back the data at the time of power recovery, thereby complicating operations of the page buffer. However, the page buffer PG of the embodiment does not have such problems (larger chip area, complicated operations).
Next, the memory cell array will be explained.
The memory cell array MA includes a plurality of memory blocks MB. The memory blocks MB is formed on a semiconductor substrate (not shown), and arranged in a direction parallel to the Y axis.
The memory blocks MB comprises a plurality of memory strings MS, a plurality of source side select transistors SSTr, and a plurality of drain side select transistors SDTr. The memory string MS comprises a plurality of memory transistors (memory cells) MTr1 to MTr4 connected in series. In
One of the source/drains of the drain side select transistor SDTr is connected to one end of the memory string MS (memory transistor MTr4). One of the source/drains of the source side select transistor SSTr is connected to the other end of the memory string MS (memory transistor MTr1). The memory strings MS are disposed in a matrix in the XY plane over the plurality of rows and the plurality of columns for each of the memory blocks MB.
With regard to the memory block MB, the control gates of the memory transistors MTr1 disposed in a matrix are commonly connected to the word line WL1. Similarly, the control gates of the memory transistors MTr2 are commonly connected to the word line WL2, the control gates of the memory transistors MTr3 are commonly connected to the word line WL3, and the control gates of the memory transistors MTr4 are commonly connected to the word line WL4.
With regard to the memory block MB, the control gates of the drain side select transistors SDTr arranged in a line along the X axis are commonly connected to a drain side select gate line SGD. A plurality of the drain side select gate lines SGD in one memory block MB are arranged in a line along the Y axis by a predetermined pitch. In addition, the others of source/drains of the drain side select transistors SDTr arrange in a line along the Y are commonly connected to the bit line BL. The bit line BL is formed in a manner that the bit line BL extends along the Y axis across the memory blocks MB. A plurality of bit lines BL are provided along the X axis by a predetermined pitch.
With regard to one memory block MB, the control gates of all of the source side select transistors SSTr are commonly connected to a source side select transistor SGS. In addition, the others of the source/drains of the source side select transistors SSTr arranged along the Y axis are commonly connected to the source line SL.
As shown in
The source side selection transistor layer 20 is a layer that functions as the source side select transistor SSTr. The memory transistor layer 30 is a layer that functions as the memory string MS (memory transistor MTr1 to MTr4). The drain side selection transistor 40 is a layer that functions as the drain side select transistor SDTr. The interconnect layers 50 are layers function as plural kinds of interconnects.
As shown in
The source side first insulating layer 21 and the source side second insulating layer 23 include, for example, silicon oxide films such as SiO2 films. In addition, as shown in
As shown in
The source side pillar-shaped semiconductor layer 26 has a pillar shape which extends along the Z axis (stacking direction). The upper surface of the source side pillar-shaped semiconductor layer 26 is in contact with the lower surface of a pillar-shaped semiconductor layer which is to be described later. The lower surface of the source side pillar-shaped semiconductor layer 26 is in contact with a diffusion layer Ba1 on the surface of the semiconductor substrate Ba. The diffusion layer Ba1 functions as a source line SL. The source side gate insulating layer 25 includes, for example, a silicon oxide film such as a SiO2 film. The source side pillar-shaped semiconductor layer 26 includes, for example, a polycrystalline film.
The source side conductive layer 22 in the source side select transistor layer 20 functions as the control gate of the source side selection transistor SSTr and the source side select gate line SGS.
As shown in
The first word line conductive layer 31a to the fourth word line conductive layer 31d, and the first inter-word line insulating layer 32a to the fourth inter-word line insulating layer 32d have shapes extending two dimensionally in a plane defined by the X axis and the Y axis (plate-like shapes). The first word line conductive layer 31a to the fourth word line conductive layer 31d, and the first inter-word line insulating layer 32a to the fourth inter-word line insulating layer 32d are segmented into memory blocks.
As shown in
Moreover, as shown in
The block insulating layer 34a is provided with a predetermined thickness on the side wall defining the memory hole 33. The charge storage layer 34b is provided with a predetermined thickness on a side wall of the block insulating layer 34a. The tunnel insulating layer 34c is provided with a predetermined thickness on a side wall of the charge storage layer 34b. The pillar-shaped semiconductor layer 35 is provided to fill the memory hole 33 via the block insulating layer 34a, the charge storage layer 34b, and the tunnel insulating layer 34c.
The pillar-shaped semiconductor layer 35 has a pillar shape which extends along the Z axis (stacking direction). The lower surface of the pillar-shaped semiconductor layer 35 is in contact with the upper surface of the source side pillar-shaped semiconductor layer 26. In the meantime, the upper surface of the pillar-shaped semiconductor layer 35 is in contact with the lower surface of a drain side pillar-shaped semiconductor layer 44 which is to be described later.
The block insulating layer 34a and the tunnel insulating layer 34c include, for example, silicon oxide films such as a SiO2 films. The charge storage layer 34b includes, for example, a silicon nitride film such as a Si3N4 film. The pillar-shaped semiconductor layer 35 includes, for example, a polycrystalline film.
The first word line conductive layer 31a to the fourth word line conductive layer 31d in the memory transistor layer 30 function as control gates of the memory transistors MTr 1 to MTr 4 and the word lines WL1 to WL4.
As shown in
In addition, as shown
Moreover, as shown
The drain side gate insulating layer 43 includes, for example, a silicon oxide film such as a SiO2 film. The drain side pillar-shaped semiconductor layer 44 includes, for example, a polycrystalline film. The drain side conductive layer 41 functions as the control gate of the drain side select transistor SDTr and the drain side select gate line SGD.
As shown in
As shown in
The interconnect layer 31a′ is provide on the same layer as the first word line conductive layer 31a, and is electrically and physically connected to the first word line conductive layer 31a. Similarly, the interconnect layer 31b′ to interconnect layer 31d′ are provided on the same layers as the second word line conductive layer 31b to the fourth word line conductive layer 31d, respectively, and are electrically and physically connected to the second word line conductive layer 31b to the fourth word line conductive layer 31d, respectively. In other words, the interconnect layers 31a′ to 31d′ are integrally formed with the word line conductive layer 31a to 31d, respectively.
The stair-like step portion SR further includes interlayer insulating layers 32a′ to 32d′ which are formed by extending the first to fourth inter-word line insulating layers 32a to 32d. Edge portions of the interlayer insulating layers 32a′ to 32d′ along the X axis are aligned with edge portions of the interconnect layers 31a′ to 31d′, respectively.
The interlayer insulating layers 32a′ is provided on the same layers as the first inter-word line insulating layers 32a, and is physically connected to the first inter-word line insulating layers 32a. Similarly, the interlayer insulating layers 32b′ to 32d′ are provided on the same layers as the second to fourth inter-word line insulating layers 32b to 32d, respectively, and are physically connected to the second to fourth inter-word line insulating layers 32b to 32d, respectively. In other words, the interlayer insulating layers 32a′ to 32d′ are integrally formed with the first to fourth inter-word line insulating layers 32a to 32d, respectively.
The interconnect layer 31a′, the interlayer insulating layers 32a′, the interconnect layer 31b′, the interlayer insulating layers 32b′, the interconnect layer 31c′, the interlayer insulating layers 32c′, the interconnect layer 31d′ and the interlayer insulating layers 32d′ are stacked in this order, and these layers 31a′, 32a′, 31b′, 32b′, 31c′, 32c′, 31d′ and 32d′ constitute a step portion ST. Specifically, the step portion ST shown in
Four conductive layers constituted by layers 31a to 31d and 31a′ to 31d′ correspond to the four conductive layers CL shown in
As shown in
[
The diffusion layer Ba1, the MOS transistor Tr3 constituting the sense amplifier, the source side select transistor layer 20, the memory transistor layer 30, the drain side select transistor layer 40 and the insulating film 71 are formed on the semiconductor substrate Ba by using well-know method.
[
In the step of forming the hole that is to be filled with the contact plug C1 of
In the present embodiment, the trench 81 is formed to reach the upper surface of the conductive layer 31a. However, the trench 81 may be formed to reach the upper surface of the conductive layer 31b, the upper surface of the conductive layer 31c or the upper surface of the conductive layer 31d. That is, the trench 81 is formed to reach the upper surface of the conductive layer corresponding to the depth where necessary capacitance for the capacitor is ensured. The depth of the trench 81 is, for example, 3 μm.
[
The insulating film 70 is formed on an area including the insulating film 71 to cover the side surface and the bottom surface of the trench 81, and then a conductive film 6 to be processed into the second capacitor electrode, an insulting film 5 to be processed into the capacitor insulating film and a conductive film 4 to be processed into the first capacitor electrode are sequentially formed on the insulating film 70. The conductive film 4 is formed to fill the trench 81.
[
The conductive film 4, the insulating film 5, the conductive film 6 and the insulating film 70, which are outside the trench 81, is removed by using chemical mechanical polishing (CMP), and the surfaces exposed by removing these films 4, 5, 6, and 70 are planarized by the CMP. In this process, the formation of the capacitor C constituting the latch is completed. In the following process, the conductive film 4, the insulating film 5 and the conductive film 6 are referred to as the first capacitor electrode 4, the capacitor insulating film 5 and the second capacitor electrode 6, respectively.
[
The oxide semiconductor layer 1 is formed on an area including the insulating film 71, and then an insulating film 2 to be processed into the gate insulating film, a conductive film 3 to be processed into the conductive film are sequentially formed on the oxide semiconductor layer 1. The oxide semiconductor layer 1 is formed to be in contact with the upper surfaces of the oxide semiconductor layer 1, the first capacitor electrode 4, the capacitor insulating film 5 and the second capacitor electrode 6.
[
Two gate electrodes 3 and two gate insulating films 3 are formed by sequentially patterning the conductive film 3 and the insulating film 2. More particularly, a resist pattern (not shown) is formed on the conductive film 3 shown in
Thereafter two transistors Tr1 and Tr2 are obtained by patterning the oxide semiconductor layer 1. More particularly, a resist pattern (not shown) is formed on the oxide semiconductor 1 and the gate electrode 3, and then the oxide semiconductor layer 1 is etched by using the resist pattern as a mask, thereby forming the two transistors Tr1 and Tr2.
Thereafter the interlayer insulating film 72, the contact plugs 8a, 8b, 51 and 74, the interlayer insulating film 73, the first interconnect layer (bit line) 52, the second interconnect layer (not shown), and the third interconnect layer 54 are formed by using well-known method, thereby obtaining the structure shown in
As shown in
The source side select transistor layer 20, the memory transistor layer 30 and the drain side select transistor layer 40 in
In the present embodiment, the latch 11 is provided between the bit line and the stair-like step portion, however the latch 11 may be provided in a region above the first interconnect layer (bit line) 52 as shown in
Noted that, in
Moreover, the transistors Tr1 and Tr2, and the capacitor C may be formed in a region 92 shown in
Furthermore, in the present embodiment, the NAND type flash memory 10 employing a linear type memory string MS is explained, however a NAND type flash memory employing a pipe type memory string may be used.
While certain embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions. Indeed, the novel embodiments described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the embodiments described herein may be made without departing from the spirit of the inventions. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the inventions.
Number | Date | Country | Kind |
---|---|---|---|
2016-183295 | Sep 2016 | JP | national |
Number | Name | Date | Kind |
---|---|---|---|
9152350 | Marukame et al. | Oct 2015 | B2 |
9682272 | Januszek | Jun 2017 | B2 |
9698272 | Ikeda et al. | Jul 2017 | B1 |
20040032764 | Koga | Feb 2004 | A1 |
20050162895 | Kuhr | Jul 2005 | A1 |
20050207229 | Takeuchi | Sep 2005 | A1 |
20100027353 | Kwon, II | Feb 2010 | A1 |
20120155190 | Kim | Jun 2012 | A1 |
20120182804 | Hung | Jul 2012 | A1 |
20130135934 | Kim et al. | May 2013 | A1 |
20160071599 | Hsu | Mar 2016 | A1 |
20160267988 | Ogawa | Sep 2016 | A1 |
Number | Date | Country |
---|---|---|
2014-116516 | Jun 2014 | JP |
5631938 | Nov 2014 | JP |
2017-168623 | Sep 2017 | JP |
Entry |
---|
R. Brain et al., “A 22nm High Performance Embedded DRAM SoC Technology Featuring Tri-gate Transistors and MIMCAP COB” 2013 Symposium on VLSI Technology (VLSIT) Digest of Technical Papers, Jun. 11-13, 2013, pp. T16-T17. |
Seon Yong Cha, “DRAM Technology—History & Challenges,” 2011 IEDM Short Course : Advanced Memory Technology, Dec. 4, 2011, pp. 1-73. |
Teruyoshi Hatanaka et al., “Ferroelectric (Fe)-NAND Flash Memory With Batch Write Algorithm and Smart Data Store to the Nonvolatile Page Buffer for Data Center Application High-Speed and Highly Reliable Enterprise Solid-State Drives” IEEE Journal of Solid-State Circuits, vol. 45, No. 10, Oct. 2010, pp. 2156-2164. |
Sungjoo Hong, “Memory Technology Trend and Future Challenges” 2010 IEEE International Electron Devices Meeting (IEDM), Dec. 6-8, 2010, pp. 12.4.1-12.4.4. |
Deok-Sin Kil, et al., “Development of New TiN/ZrO2/Al2O3/ZrO2/TiN Capacitors Extendable to 45nm Generation DRAMs Replacing HfO2 based Dielectrics” 2006 Symposium on VLSI Technology Digest of Technical Papers, Jun. 2006, 2 pages. |
Noboru Shibata, et al., “A 70nm 16Gb 16-level-cell NAND Flash Memory” 2007 Symposium on VLSI Circuits Digest of Technical Papers, Jun. 2007, pp. 190-191. |
Daisaburo Takashima, et al., “An Embedded DRAM Technology for High-Performance NAND Flash Memories” 2011 IEEE International Solid-State Circuits Conference Digest of Technical Papers (ISSCC), Session 28, Feb. 2011, pp. 504-505. |
Ken Takeuchi, et al., “A Dual-p. Programming Scheme for High-Speed Multigigabit-Scale NAND Flash Memories” IEEE Journal of Solid-State Circuits, vol. 36, No. 5, May 2001, pp. 744-751. |
Ken Takeuchi, et al., “A 56-nm CMOS 99-mm2 8-Gb Multi-Level NAND Flash Memory With 10-MB/s Program Throughput” IEEE Journal of Solid-State Circuits, vol. 42, No. 1, Jan. 2007, pp. 219-232. |
Number | Date | Country | |
---|---|---|---|
20180082750 A1 | Mar 2018 | US |