Engineered dielectric profiles for high aspect ratio 3D NAND structures
By adopting an engineered bottom thick dielectric profile method in 3D NAND memory devices, the problem of dielectric thinning in high aspect ratio memory holes is solved, and the reliability of memory cells is improved.
Patent Information
- Application Number
- CN202311828242.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-05-12
- Filing Date
- 2023-12-28
- Publication Date
- 2024-11-12
AI Technical Summary
In manufacturing high aspect ratio 3D NAND memory devices, it becomes difficult to deposit and grow dielectric materials to achieve sufficient thickness at the bottom of the memory hole, resulting in thinning of the dielectric, thereby reducing the reliability of the memory cell.
Using an engineered bottom thick dielectric profile method, the thickness of the dielectric material remains basically consistent from the top to the bottom by forming a base dielectric film on the side walls of the memory hole and selectively deposition and oxidation on it. Thinning problems caused by increase.
The tunnel dielectric film thickness at the bottom part of the memory hole is effectively increased, the reliability of the memory cell is improved, and the step coverage trade-offs are overcome depending on the aspect ratio.
Smart Images

Figure CN118946150A_ABST
Abstract
Description
Background Art
[0001] Three-dimensional (3D) NAND (non-AND) technology is commonly used to create non-volatile (NV) storage devices, such as solid-state drives (SSDs). References to 3D NAND may more specifically refer to NAND flash memory. Unlike conventional 2D memory devices, 3D NAND memory devices have one or more decks of tiers of circuit elements stacked on top of each other. The circuit elements are connected via channels in vertical structures (e.g., memory holes or pillars) with a high aspect ratio (AR).
[0002] Manufacturing 3D memory devices with high AR vertical structures presents significant challenges. One issue is that as AR increases, it becomes more difficult to deposit and / or grow dielectric materials (such as tunnel dielectrics and / or blocking dielectrics) to have sufficient thickness for the NAND cell in the bottom portion of the memory hole compared to the rest of the cell. This typically results in a degradation in reliability for cells with thinner dielectrics. BRIEF DESCRIPTION OF THE DRAWINGS
[0003] The foregoing aspects of the present invention and many of the attendant advantages will become more readily appreciated as they become better understood by reference to the following detailed description when taken in conjunction with the accompanying drawings, wherein like reference numerals refer to like parts throughout the various views unless otherwise specified:
[0004] Figure 1 is a block diagram of an example of a system that stores data in non-volatile (NV) media;
[0005] Figure 2 is a block diagram of an example of a system including a three-dimensional (3D) memory device structure;
[0006] Figure 3 is a diagram showing an abstract representation of a portion of a solid-state memory component according to one example;
[0007] Figure 4 It is shown in Figure 3 A simplified view of a memory cell implemented in a solid-state memory component of FIG.
[0008] Figure 5 is a diagram showing a cross-sectional view of a portion of a vertical channel in a 3D memory structure according to one embodiment;
[0009] Figure 6a shows an isometric cross-sectional view of an abstract representation of a semiconductor structure in which a memory hole is formed through a stack 604 of memory levels and in which the thickness of a dielectric material formed over a sidewall of the memory hole decreases for a lower wordline;
[0010] Figure 6b An isometric cross-sectional view of an abstract representation of a semiconductor structure having an engineered dielectric profile in which the thickness of the dielectric remains substantially the same from top to bottom of a memory hole for a word line;
[0011] Figure 7a-7h is a schematic diagram of a cross section of a processing stage of a 3D memory structure according to one embodiment;
[0012] Figure 8 is a flow chart showing top-level operations for fabricating an oxide having a BTO profile according to one embodiment;
[0013] Fig. 9 is a flow chart illustrating detailed operations for performing selective oxide deposition according to one embodiment;
[0014] Fig.10 An isometric cross-sectional view of a 3D memory structure and a pair of 2D cross-sectional details illustrating the tunnel dielectric profile are shown;
[0015] Fig.11 A cross section showing the structure of a 3D memory device using a charge trap flash (CTF) cell;
[0016] Fig.12 shows an isometric cross-sectional view of a semi-cylindrical cell employing an engineered dielectric profile according to one embodiment;
[0017] Fig.13 shows an isometric cross-sectional view of a planar trench cell employing an engineered dielectric profile according to one embodiment;
[0018] Fig.14a is a block diagram of an example of a system having a hardware view of a solid state drive (SSD) with a non-volatile array having a vertical channel with a conductive structure; and
[0019] Fig.14b is a block diagram of an example of a logical view of a system having a solid state drive (SSD) with a nonvolatile array having vertical channels with conductive structures. DETAILED DESCRIPTION
[0020] Embodiments of methods and apparatus for bottom thick oxides of high aspect ratio 3D NAND structures are described herein. In the following description, numerous specific details are set forth to provide a thorough understanding of embodiments of the invention. However, one skilled in the relevant art will recognize that the invention may be practiced without one or more of the specific details, or with other methods, components, materials, etc. In other cases, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of the invention.
[0021] References throughout this specification to "one embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in at least one embodiment of the present invention. Thus, the phrases "in one embodiment" or "in an embodiment" appearing in various places throughout this specification do not necessarily all refer to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
[0022] For the sake of clarity, various components in the figures herein may also be referred to by their markings in the figures rather than by specific reference numerals. In addition, reference numerals relating to specific types of components (as opposed to specific components) may be illustrated with reference numerals followed by "(typ)" indicating "typical". It should be understood that the configuration of these components will be typical similar components that may exist but are not shown in the figures for simplicity and clarity, or similar components that are not marked with separate reference numerals. On the contrary, "(typ)" should not be interpreted as meaning that components, elements, etc. are typically used for their disclosed functions, embodiments, purposes, etc.
[0023] Figure 1 1 is a block diagram of an example of a system that stores data in a non-volatile (NV) medium. System 100 includes a host 110 coupled to an NV device 120. Host 110 represents a computing device. Host 110 includes I / O (input / output) 112, which represents hardware interconnected with NV device 120. NV device 120 includes I / O 122 corresponding to I / O 112. I / O 122 represents hardware interconnected with host 110.
[0024] The host 110 provides a hardware platform to operate the NV device 120. The host 110 includes one or more processors 114 to perform the operations of the host 110. The processor 114 executes a host operating system (OS), which provides a software platform for the operation of the NV device 120. The hardware platform provides hardware resources for interfacing with the NV device 120, including transceiver hardware for performing access to the device. The software platform includes control software to execute other software elements, such as applications or other agents that execute under the OS and create requests to access the NV device 120.
[0025] I / O 112 and I / O 122 are interconnected via one or more signal lines 150. Signal lines 150 typically include multiple individual lines and may be considered to be one or more buses that connect host 110 to NV device 120. Host 110 may send a host read command to NV device 120 via signal line 150. According to any of the examples provided, in response to the read command, NV device 120 services the request from a transient Vt state.
[0026] In one example, the host 110 includes a controller 116. The controller 116 represents a memory controller or storage controller. In one example, the controller 116 is integrated with the processor 114. In one example, the controller 116 is separate from the processor 114. The controller 116 enables the host 110 to manage access to the NV device 120. In response to a host operation of the processor 114 requesting access to data on the NV device 120, the controller 116 provides access to the NV device 120. The controller 116 may represent hardware and firmware elements of the host 110 to enable interaction with the NV device 120.
[0027] The NV device 120 includes a controller 124, which represents a storage controller on the storage device side, which is separate from the controller 116 of the host 110. The controller 116 of the host 110 represents a component of the host system. The controller 124 represents a component of the storage device or memory device into which the NV medium is incorporated. The controller 124 receives a command sent from the host 110 and determines how to service the command or request from the host. The controller 124 performs operations to access (e.g., read or write) the NV medium 130 in response to the host command.
[0028] NV medium 130 represents a non-volatile storage medium of NV device 120. In one example, NV medium 130 includes three-dimensional (3D) NAND memory cells. In one example, NV medium 130 includes 3D NOR memory cells. In one example, NV medium 130 includes 3D XPOINT™ memory cells.
[0029] NV medium 130 includes bit cells or memory cells organized as blocks 132. A memory block refers to a portion of NV medium 130 that is jointly charged or activated for access operations. In one example, block 132 is subdivided into sub-blocks. In one example, a block refers to a bit cell that shares a select gate line. In one example, multiple sub-blocks share a select gate (e.g., a common select gate source (SGS) or a common select gate drain (SGD)) connector.
[0030] In one example, a block refers to an erase unit or a unit size of NV media 130 that is erased together and the number of writes is monitored by controller 124. In one example, NV media 130 includes single-level cell (SLC) and multi-level cell (MLC) media. For example, NV media 130 may include SLC and QLC (quad-level cell) or SLC and TLC (tri-level cell) bit cells. Block size may vary depending on the media type.
[0031] In one example, the controller 124 is an ASIC (Application Specific Integrated Circuit) that controls the operation of the NV device 120. In one example, the controller 124 is a CPU (Central Processing Unit) core or processor device on the NV device 120. In one example, the NV device 120 represents an SSD, and the controller 124 controls multiple NV media dies or NV media chips integrated into the SSD. In one example, the NV device 120 represents a module or PCB (Printed Circuit Board) including multiple NV media dies or NV media chips integrated thereon, and the controller 124 controls the NV media dies of the module. In one example, the controller 124 executes firmware to manage the NV device 120. In one example, the controller 124 executes firmware to manage the NV device 120, including firmware for controlling the service of a read command based on whether the NV medium is in thermal balance.
[0032] In one example, the controller 124 manages Vt state detection and read command servicing based on idle time or delay between consecutive read commands. In one example, the controller 124 monitors one or more media states 126. The media state 126 represents the state of a portion of a memory, such as a block, and how to access the media can be determined based on the media state 126. For example, if the media state 126 indicates that the target block is in a stable state, the controller 124 can first issue a pseudo read before accessing the target block. In one example, the NV media device 120 can include one or more timers 142 and counters 144.
[0033] Figure 22 is a block diagram of an example system showing further details of the 3D memory device structure. System 200 represents a computing device including a 3D memory. Host 210 represents a hardware platform that performs operations to control the functions of system 200. Host 210 includes processor 212, which is a host processor that performs the operations of the host. In one example, processor 212 is a single-core processor. In one example, processor 212 is a multi-core processor device. Processor 212 can be a general-purpose processor that executes a host operating system or software platform for system 200. In one example, processor 212 is a dedicated processor, graphics processor, peripheral processor, or other controller or processing unit on host 210. Processor 212 executes multiple agents or software programs (not specifically shown). An agent can be an independent program and / or thread, process, software module, or other code and data to be operated by processor 212.
[0034] During operations performed by the processor 212, an agent executed by the processor may request data and / or code that is not stored at the host 210 (e.g., in a cache or main memory) and should therefore be obtained from the memory 220. The memory controller 214 generates and processes memory access commands to the memory 220 to perform memory accesses. The memory controller 214 represents a circuit or logic or processor that manages access to the memory 220. In one example, the memory controller 214 is part of the host 210. In one example, the memory controller 214 is part of the processor 212. In one example, the memory controller 214 is integrated with the processor 212 on a common substrate. In one example, the memory controller 214 separates the chip from the processor 212 and can be integrated with the processor 212 in a multi-chip package (MCP).
[0035] The memory 220 includes a controller 240, which represents a controller at a memory or storage device for processing and servicing commands from the memory controller 214. In one example, the controller 240 represents a controller for a memory device. In one example, the controller 240 represents a controller for a memory module. The memory 220 includes a 3D array 222. In one example, the 3D array 222 includes NAND memory blocks. In one example, the 3D array 222 includes QLC NAND memory blocks.
[0036] As shown, the bit lines (BL) intersect the planes of the word line (WL) hierarchy. As an example, each word line WL[0:(N-1)] is a hierarchy. There may be P bit lines (BL[0:(P-1)]). In one example, the 3D array 222 is also divided into sub-blocks by SGD[0:(M-1)], which divides each word line into separate segments within the hierarchy or plane of the word line. Alternatively, the SGS may be subdivided to provide sub-blocks. In such a configuration, although the SGS is shown as being applied to multiple SGD lines, there may be multiple SGS lines to a single SGD line. SRC represents a common source.
[0037] Channel 250 represents a vertical channel of a 3D array. A channel refers to a vertical stack of bit cells that can be charged through a channel connector. In one example, the channel is coupled to a bit line. It should be understood that there can be spatial dependencies in the stable Vt state of the channel. For example, the flow of charge carriers in the channel may be different at different ends of the channel. Therefore, a block with a particular word line may show worse degradation than other blocks. The operation of the controller 240 to mitigate read disturbances due to stable Vt in the channel can be set by mitigating thresholds and operations for the most sensitive word lines.
[0038] Each label WL[0], WL[1], SGD[0], etc. indicates a selection signal provided by the control logic of the decode logic 224 or a selection signal provided by the control logic of the sense / output logic 226. In one example, the decode logic 224 includes selection logic for selecting each of the illustrated signal lines. In one example, the sense / output logic 226 enables sensing of the contents of the bit cells of the 3D array 222 for a read operation or writing a value back to the array. The outputs may be used for a read operation to send data back to the host 210. A write operation would include a write buffer to apply a value to the array.
[0039] It should be understood that the signal lines in the 3D array 222 are wires or traces or other conductors that provide charge from the driver to various elements or components. The driver circuit decoding logic 224 provides charge to charge each signal line to a desired voltage for a desired operation. Each signal line may have an associated voltage level associated with certain operations. For example, each word line may have a select voltage and an unselect voltage to indicate a word line that is selected for operation and a word line that is not selected for operation, respectively.
[0040] In the 3D array 222, it will be appreciated that the length of the word lines can be quite large. In one example, the number of levels of word lines is on the order of tens or dozens of word lines (e.g., N=28, 32, 36, 70, or more). In one example, the number of sub-blocks is on the order of a few or tens (e.g., M=8, 76, or more). Typically, the number of bit lines in the 3D array 222 will be on the order of hundreds to thousands (e.g., P=2K). Therefore, in one example, each bit line is relatively short compared to the length of the word line.
[0041] Figure 3 An abstract representation of a portion of a solid-state memory component 300 according to an example is shown. Typically, a portion of a solid-state memory component includes a memory column 310 and a memory cell 320a-n (i.e., a string 325 of memory cells, such as a NAND string) positioned adjacent to the memory column 310. In some embodiments, the memory column 310 may also be referred to as a "memory hole". Any suitable number of memory cells may be included. The memory column 310 may act as a channel region of the memory cells 320a-n that may be coupled in series. For example, during operation of one or more of the memory cells 320a-n of the string, a channel may be formed in the memory column 310. The memory column 310 and the string of memory cells 320a-n may be oriented vertically, such as in a three-dimensional memory array. For example, the memory cell 320a is located at a vertical level (e.g., near the top of the memory array) above the vertical level (e.g., near the bottom of the memory array) where the memory cell 320n is located. In general, the memory cells 320a-n may have any suitable structure. A simplified memory cell structure is provided for context and as an example. Therefore, it will be appreciated that suitable memory cell structures may vary from Figure 3 The memory cell structure shown in .
[0042] Each memory cell 320a-n in this example may have a charge storage structure (eg, it may be a conductive floating gate, a dielectric charge trap, etc.). For example, as shown in the cross-sectional side view Figure 4As shown in , a memory pillar 310 and a representative memory cell 320, the memory cell 320 may have a charge storage structure 321. Each memory cell 320a-n may also have a tunnel dielectric between its charge storage structure and the memory pillar 310. For example, the memory cell 320 may have a tunnel dielectric 313 between the charge storage structure 321 and the memory pillar 310. In addition, each memory cell 320a-n may have a control gate (e.g., as part of an access line (e.g., a word line) or coupled to an access line). For example, the memory cell 320 may include a control gate 330. Each memory cell may have one or more dielectric materials or dielectric layers between its charge storage structure and the control gate. For example, the memory cell 320 may include a dielectric layer 323 between the charge storage structure 321 and the control gate 330, which is referred to as an inter-polysilicon dielectric (IPD) in other parts of this document.
[0043] Each memory cell 320 may be a non-volatile memory cell and may have a charge storage structure 321, for example, a floating gate which may be a semiconductor (e.g., polysilicon), a charge trapping layer which may be a dielectric film, etc. Non-limiting examples of dielectrics suitable for charge trapping include nitrides, high dielectric constant (high-K) dielectrics (e.g., aluminum oxide (Al2O3) with a K of about 10, which has embedded conductive particles (e.g., nanodots), such as embedded metal particles or embedded nanocrystals (e.g., silicon, germanium, or metal crystals)), silicon-rich dielectrics, or SiON / Si3N4. Embodiments of floating gates and charge trapping cells are described and illustrated below.
[0044] Further references Figure 3 , dielectric 340 (also referred to as an isolation layer) may be interposed between consecutively adjacent memory cells 320a-n in string 325. For example, dielectric 340 may be interposed between charge storage structure 321, dielectric 323, and control gate 330 of at least consecutively adjacent memory cells 320a-n. Dielectric 341 may be interposed between an end of string 325 and select gate 311 (e.g., between memory cell 320a and select gate 311), and dielectric 342 may be interposed between an opposite end of string 325 and select gate 312 (e.g., between memory cell 320n and select gate 312), as shown in FIG. Figure 4 as shown in .
[0045] In some embodiments, where the charge storage structure 321 is a charge trap, the tunnel dielectric 322, the charge storage structure 321, and the dielectric 323 may form a continuous structure that may be shared by two or more of the memory cells 320a-n (e.g., may be shared by two or more of the memory cells 320a-n). For example, such a structure may be shared or shared by all of the memory cells 320a-n.
[0046] Each of the memory cells 320a-n can have a thickness (e.g., channel length) 326. For example, the memory cells 320a-n can have the same channel length regardless of where the memory cells are located in the string 325. In some embodiments, at least one channel length of a memory cell can be different from another channel length of another memory cell.
[0047] Each memory cell 320a-n of the string 325 can be coupled in series with a select gate (e.g., a drain select gate) 311 adjacent to (e.g., in contact with) the column 310 and a select gate (e.g., a source select gate) 312 adjacent to (e.g., in contact with) the column 310 and can be between the select gate 311 and the select gate 312. For a functional memory column, the column 310 is electrically coupled to a data line (e.g., a bit line 316), as indicated at 317a and 317b. Thus, the select gate 311 can selectively couple the string 325 to the data line (e.g., the bit line 316). Additionally, for a functional memory column, the column 310 is electrically coupled to a source line 318, as indicated at 319a and 319b. Thus, the select gate 312 can selectively couple the string 325 to the source line 318. For example, select gate 311 can be coupled in series with memory cell 320a, and select gate 312 can be coupled in series with memory cell 320n. Select gates 311 and 312 can each include a gate dielectric 313 adjacent to (e.g., contacting) pillar 310 and a control gate 314 adjacent to (e.g., contacting) the corresponding gate dielectric 313.
[0048] Figure 5 A circuit 500 is shown that represents a portion of a vertical channel in a 3D memory structure, such as a 3D NAND array. The circuit 500 is not necessarily drawn to scale, and shows non-limiting examples of features rather than providing an accurate representation of the features. In addition, the shapes of some cell structures are simplified for illustration purposes.
[0049] Circuit 500 shows two memory cells (cell 510 and cell 520) and three isolation layers 502, 504 and 506 (which may also be referred to as separation layers). Although circuit 500 is not necessarily drawn to scale, the isolation layers between cells are typically thinner than the cells themselves. The cells show an example of a memory cell structure, where semiconductors are indicated as storage node 512 and storage node 522, respectively. Storage node 512 is separated from control gate polysilicon by one or more IPD (inter-polysilicon dielectric) layers 514. Conductor layer polysilicon is a conductor layer for controlling access to the storage node. The conductor layer polysilicon for storage node 512 is represented as control gate 516. Similarly, storage node 522 is separated from the conductor layer polysilicon by one or more IPD layers 524 represented as control gate 526. As long as the storage node is electrically isolated from the conductor layer, the number of IPD layers and the structure of those layers are not important to circuit 500.
[0050] In one example, circuit 500 includes a channel conductor 530 with a dielectric fill 532. 3D NAND typically uses polycrystalline (poly) materials for channel 530, such as but not limited to polycrystalline silicon (also known as polysilicon). In one example, channel 530 can be p-type or n-type doped polysilicon.
[0051] Figure 3 and Figure 4 An example of an ideal structure is shown, where the memory hole / pillar has perfectly straight sidewalls (ie, constant diameter), and the thickness of the dielectric layer remains the same from the top to the bottom of the memory hole / pillar. For illustration purposes and clarity, Figure 3 The aspect ratio (height to width ratio or hole depth to hole diameter ratio) is reduced in the PCB - the aspect ratio in actual devices is larger, such as shown in some of the following figures.
[0052] Typically, it is not possible or practical to form a memory hole with a high aspect ratio having completely straight sidewalls. Instead, the memory hole has a small amount of taper, where the diameter at the top of the memory hole is slightly larger than the diameter at the bottom. Likewise, the thickness of the dielectric material formed on the sidewalls of the memory hole also varies slightly from top to bottom.
[0053] The problem is Figure 6a 600a, which shows an abstract representation of a semiconductor structure, wherein a memory hole 602 is formed through a stack 604 of memory levels. A dielectric material 606 is formed on the sidewalls of the memory hole 602. As shown, both the memory hole 602 and the dielectric material 606 have a tapered profile (enlarged in the diagram 600a for clarity). More notably, the cross-sectional width of the dielectric material 606 narrows from top to bottom.
[0054] The grey layers in diagram 600a represent memory levels in which circuit elements including word lines (WLs) 608 are formed, while the white layers represent layers (e.g., isolation or separation layers) used to separate memory levels. The thickness of dielectric material 606 is thicker for word lines toward the top of memory hole 602 and thinner for lower word lines. The thinner dielectric material (e.g., oxide) on the lower WL can be attributed to the fundamental process capability limitations, i.e., the reduction of oxidizing species transport / reaction rate limitations at the bottom of the memory hole to produce such dielectric films in high aspect ratio structures. Therefore, as the aspect ratio of the memory hole increases, the thickness of the dielectric film deposited at the bottom of the structure becomes a critically important parameter, which is generally proportional to the reliability of the memory cells on the lower word line after a threshold amount of program-erase cycles of the memory product.
[0055] To address this thinned dielectric problem for high aspect ratio memory holes, the proposed method produces an engineered bottom thick dielectric profile where the thickness and quality of the dielectric remains essentially the same from the top to the bottom of the memory hole for the word line, thus overcoming the step coverage trade-off limitation that depends on aspect ratio.
[0056] exist Figure 6b An example of the application of this method is shown in diagram 600b in FIG. As shown, the width of dielectric material 606 is (substantially) matched for both the top and lower word lines, while the width of dielectric material 606 in the middle word lines is slightly reduced relative to the top and lower word lines.
[0057] Figure 7a-7h is a cross-sectional illustration of a processing stage of a 3D memory structure using an engineered dielectric profile according to one embodiment. For purposes of illustration, Figure 7a-7h A process for creating a vertical channel with low resistance regions in recesses along the channel is shown. Figure 7a-7h The circuit states shown in can be applied to any example 3D memory structure, such as but not limited to circuit 500. Processing is shown only for a few layers of the 3D array. The processing will be understood to be applied to other layers of the 3D array. The drawings are for discussion purposes and are not necessarily drawn to scale.
[0058] Figure 7a A circuit state 700 is shown in which the layers of a 3D array are formed in a vertical stack. Circuit state 700 shows the entire vertical stack or a portion of the vertical stack of layers of a memory array, where the substrate or wafer on which the circuit is formed is not shown. Other layers related to the selector are also not shown. The stack may have dozens of conductor layers and isolation layers.
[0059] Polysilicon 730 represents a conductor layer for a memory cell. Isolation layer 720 represents an electrical isolation layer between the conductor of polysilicon 730 and the layer below it (not shown). Circuit state 700 also shows polysilicon 750, polysilicon 770, and polysilicon 785 as conductor layers, and isolation 740, isolation 760, isolation 780, and isolation 790 alternating between the conductor layers. In one example, the conductor layer is a highly doped polysilicon layer. In one example, the isolation layer is a dielectric layer (e.g., silicon oxide). In one example, the isolation layer can be or include a nitride material.
[0060] Figure 7b Circuit state 702 is shown, where the process etches or removes a portion of the layer. The resulting etched isolation layer is represented as cavity 791. Cavity 791 represents a vertical cavity or vertical opening through the conductor layer and isolation layer. For example, cavity 791 can be formed by an etching operation that forms a hole in the 3D array structure. Figure 6a and Figure 6b As shown, the etch can be performed in stages to achieve vertical channels cut into the 3D array, recognizing that the sidewalls of the channels will not be perfectly vertical due to process limitations.
[0061] Figure 7c Circuit state 704 is shown, where the process removes more material from the polysilicon layer. The resulting grooves are shown as grooves 732 for layer polysilicon 730, grooves 752 for layer polysilicon 750, grooves 772 for layer polysilicon 770, and grooves 792 for polysilicon 785. The grooves are formed from cavity 791 into the polysilicon layer and between the isolation layers. The depth of the grooves can be controlled by selective etching of the polysilicon conductor rather than the isolation material. The grooves extend away from the center of cavity 791 and will therefore extend away from the vertical channel to be formed. The grooves can extend away from the center of cavity 791 toward the corresponding control gate of the memory cell or the control gate that controls access to the storage node. The depth of the grooves depends on the structure of the storage node to be formed for the memory cell. Circuit state 704 shows grooves on both sides of cavity 791. It should be understood that cavity 791 is a three-dimensional structure, such as a vertical cylinder or annular column, and the view only shows a two-dimensional cross-sectional view.
[0062] Figure 7dCircuit state 706 is shown, where processing forms material for memory cell storage in the exposed area. Storage node 736 represents a storage area for layer polysilicon 730, which is identified as control gate 734 in circuit state 706. The polysilicon layer is the control gate of storage node 736 to control access to the storage element. Isolation layer 735 represents one or more layers of isolation material used to isolate the conductor of control gate 734 from the conductor of storage node 736. Storage node 756 represents a storage area for layer polysilicon 750, which is identified as control gate 754. The polysilicon layer is the control gate of storage node 756 to control access to the storage element. Isolation layer 755 represents one or more layers of isolation material used to isolate the conductor of control gate 754 from the conductor of storage node 756. Storage node 776 represents a storage area for layer polysilicon 770, which is identified as control gate 774. The polysilicon layer is the control gate of the storage node 776 to control access to the storage element. The isolation layer 775 represents one or more layers of isolation material used to isolate the conductor of the control gate 774 from the conductor of the storage node 776. The storage node 796 represents a storage area for the layer of polysilicon 785, which is identified as the control gate 794. The polysilicon layer is the control gate of the storage node 796 to control access to the storage element. The isolation layer 795 represents one or more layers of isolation material used to isolate the conductor of the control gate 794 from the conductor of the storage node 796. It should be understood that the various storage nodes will be formed in parallel. The formation of the storage node may include multiple processing operations performed in series. The structure of the storage node is not necessarily exactly as shown. In one example, the storage node represents a floating gate. In another example, the storage node represents a charge trapping layer. The following presents a model of a floating gate cell and a charge trapping cell that shows further details of the internal structure of the memory cell. After the storage node is formed, a dielectric film can be deposited or grown from the storage node to form the tunnel dielectric 725, such as Figure 7e shown.
[0063] Figure 7e Circuit state 708 is shown where processing may remove material from the isolation layer. The polysilicon layer including the gate and storage nodes is represented in circuit state 708 as memory cells 738, 758, 778, and 798.
[0064] As discussed above, an engineered dielectric profile method is used to increase the thickness of the tunnel dielectric film at the bottom portion of the memory hole. Figure 7e-7g The circuit state shown in Figure 8 and Fig. 9 800 and 900 , according to one embodiment, the proposed method is implemented as follows.
[0065] The top-level flow is shown in flow chart 800 and begins by performing an in-situ steam generation (ISSG) oxidation process to form a base dielectric film on the sidewalls of the memory hole, as shown in blocks 802 and Figure 7e As shown. Before the selective oxide is deposited (in block 804) at the bottom of the memory hole, the base dielectric film acts as a structural barrier, thereby protecting the underlying floating gate cell film from potential precursor contamination during subsequent processing. Circuit state 708 and Figure 7e The circuit state is shown after a base dielectric film 725 has been formed. Generally, the thickness of the base dielectric film 725 is exaggerated for illustration purposes; in practice, the base dielectric film 725 may be on the order of 4 nm-6 nm.
[0066] Next, in block 804, selective dielectric film deposition is performed. Figure 7f As shown in the circuit state 710 in FIG. 1 , the selectively deposited dielectric film 737 is selectively deposited at the bottom portion with a desired thickness using an inhibitor-based deposition process without interfering with the remaining portion of the memory hole. Further details of the selective deposition process are described in Fig. 9 As will be explained below, in practice, the thickness of the selective dielectric film will begin at a starting point that may be selected as the middle portion of the memory hole and gradually increase toward the bottom of the hole. For illustration purposes, Figure 7f-7h This gradual increase is exaggerated in FIG. 7 , as is the thickness of the selectively deposited dielectric film. Furthermore, in one embodiment, the base dielectric film 725 and the selectively deposited dielectric film 737 comprise the same material - for purposes of illustration, the selectively deposited dielectric film 737 is Figure 7f and Figure 7g are shown in different cross-hatching patterns.
[0067] The third step in the top-level process flow is to perform ISSG oxidation to form a capping dielectric over the base dielectric film and the selectively deposited dielectric film, as shown in block 806. This is done by Figure 7g 712 in the circuit state 712 is graphically illustrated by adding a cover dielectric film 745. Since ISSG is used for both the base dielectric film and the cover dielectric film, the cross section of the cover film 745 does not show the layer separation line, but the base dielectric film 725 and the cover dielectric film 745 are shown as a single dielectric material. In addition, as shown in FIG. Figure 7hAs shown, the base dielectric film 725, the selectively deposited dielectric film 737, and the capping dielectric film 745 are shown mixed together (because they are the same material) to form a tunnel dielectric film 753 having the profile shown. Since the ISSG oxidation acts as a repair process to densify the film and remove any potential contaminants trapped within the film, this capping process is necessary to optimize the electrical quality of the selectively deposited dielectric film. Typically, the ISSG oxidation process can be performed between 800°C-100°C to obtain the best dielectric quality. Similar to the base dielectric film, the thickness of the capping dielectric film can be ~2nm-5nm. Therefore, the total thickness of all three steps combined will form the tunnel dielectric film of the 3D NAND memory cell.
[0068] Moving to flow chart 900, a selective deposition process comprising two loops is explained. The outer loop is called a group cycle (GC), which can be performed one or more times based on the desired film thickness to be deposited. The inner loop is called a PEALD (plasma enhanced atomic layer deposition) cycle, where the actual precursor gases react under plasma to produce a single layer of dielectric film in each cycle, which is repeated multiple times in each group cycle.
[0069] The group cycle begins with a plasma etch in box 902, followed by a plasma purge in box 904. In one embodiment, the plasma etch employs nitrogen trifluoride (NF3). The selective deposition process capability is a customized atomic layer deposition (ALD) process to selectively deposit ultra-thin dielectric films (thickness ~0.4-3nm) at the bottom of the memory hole while suppressing film deposition on the top area of the memory hole. NF3 suppresses ALD growth by blocking precursor adsorption on the upper portion of the memory hole, thereby enabling preferential deposition of the dielectric film at the bottom portion of the memory hole. The effectiveness of plasma etching varies with depth, with a greater impact on the upper portion of the memory hole. The suppression and deposition in the transition region (mid-depth portion of the memory hole) results in a gradual thickness profile increase of the film, which mimics a conventional dielectric thickness profile taper effect, such as Figure 7b shown.
[0070] The PEALD cycle includes a dose operation, a dose purge, a radio frequency (RF) conversion, and an RF purge, as shown in boxes 906, 908, 910, and 912, respectively. The dose operation refers to flowing / injecting a precursor onto the surface of the reservoir hole wall to cause precursor absorption. In one embodiment, the dose operation uses tert-butylaminosilane (also known as BTBAS), noting that other compounds with similar properties can be used, but are not limited to other aminosilanes. During the dose purge, an inert gas (e.g., nitrogen) is introduced to purge away excess precursor. The RF conversion and RF purge operations employ RF plasma cycles and purges. For example, in one embodiment, the RF plasma cycle uses oxygen as a precursor to react with the precursor to produce SiO2 (or SiO X ) reactants. Each iteration of the PEALD cycle increases the thickness of the deposited dielectric thickness by . In one embodiment, the preferred thickness is It may be used as part of a tunnel dielectric, although different thicknesses of this selectively deposited film may be used based on the application.
[0071] After the last PEALD cycle, a plasma passivation operation is performed in block 914. Plasma passivation is used to reduce fluoride contamination that may result from the use of NF3.
[0072] As discussed above, in block 806, the selective deposition process is followed by forming a capping dielectric film (eg, a capping oxide). Figure 5 When used in the floating gate structure of the circuit 500, the tunnel dielectric film 753 corresponds to the oxide 550. Subsequently, a channel conductor is formed on the tunnel dielectric film 753, such as Figure 7h The circuit state 714 in FIG. 1 is shown. The channel conductor is shown as channel 757. In one example, channel 757 is lightly doped (1e17 / cm 3 -5e17 / cm 3 ) polysilicon material, such as an n-type (majority of electron carriers) or p-type (majority of hole carriers) channel.
[0073] Fig.10An isometric cross-sectional view of a 3D memory structure 1000 is shown showing a memory hole and a pair of 2D cross-sectional details 1002 and 1004. For illustrative purposes, the height of the structure is not to scale, i.e., the number of memory levels may actually be higher. As shown, the depth of the memory hole is divided into three regions for the purpose of illustrating the engineered tunnel dielectric profile described in the previous section. In the upper region toward the top of the memory hole, the tunnel dielectric film thickness is unchanged and is the result of the ISSG oxidation process. In the transition region, toward the middle portion, the tunnel dielectric film thickness consists of a tapered profile that is the result of a selective deposition process from thinner to thicker as the depth of the memory hole increases. The lower region toward the bottom of the memory hole has the full thickness of the selective dielectric deposition sandwiched between the base layer and the capping layer formed by ISSG oxidation, as described in the previous section.
[0074] In 2D cross-sectional detail 1002 , which shows memory cells at three word lines in an intermediate depth and transition region, the thickness of the tunnel dielectric is less than the tunnel dielectric shown in 2D cross-sectional detail 1004 , which shows memory cells at three word lines toward the bottom of 3D memory structure 1000 .
[0075] The engineered dielectric profile approach can be applied to other memory cell structures, such as charge trap flash (CTF) cells, half-cylinder cells, and trench cells. Fig.11 Cross sections 1100 and 1102 of the structure of a 3D memory device using CTF cells are shown. As shown in cross section 1100, the 3D memory structure includes cells 1110, 1130, and 1150 separated by isolation (layers) 1120 and 1140. The added materials (e.g., via deposition and / or growth) include a blocking dielectric film 1160, a charge trapping dielectric film 1165, a tunnel dielectric 1170, and a polysilicon channel 1180. The memory hole is then filled with a channel oxide. An engineered dielectric profile method can be applied to form one or more CTF cell films (blocking oxide 1160 and tunnel oxide 1170) to adjust the step coverage profile on the limiting word line in a manner similar to that described and shown above for a floating gate (FG) charge storage cell. Therefore, this method can help alleviate the step coverage limitation that depends on the aspect ratio and achieve vertical 3D NAND cell scaling.
[0076] The engineered dielectric profile approach can be extended to alternative cell topologies with high aspect ratio features. For example, Fig.12 and Fig.13A semi-cylindrical cell 1200 and a trench cell 1300 are shown, respectively, for which the proposed method has been applied to adjust the profile of the dielectric along the height of the structure. The semi-cylindrical cell 1200 shows a hierarchy of alternating circuit layers 1202 and isolation layers 1204 in which word lines and cells are formed. A cavity that will become a memory hole 1206 is formed in the hierarchy, followed by the creation of external memory cell features (not shown for clarity). Fig.12 Then, the proposed method (e.g., such as Figure 7e-7g 800 and 900) to form a tunnel dielectric 1208 with an engineered profile to have a gradual change in thickness. Subsequently, a channel material 1210 may be formed over the tunnel dielectric to complete the device formation. This method may be used for both floating gate (FG) charge storage cells and charge trapping (CTF) cells.
[0077] A similar approach is used for trench cell 1300. As before, the hierarchical structure includes alternating circuit layers 1302 and isolation layers 1304. A trench 1306 is then formed in the hierarchical structure. For purposes of illustration, the angle of the sidewalls of trench 1306 is exaggerated - in practice, the trench walls will be nearly vertical with a small taper from top to bottom. Fig.13 After processing (shown in FIG. 1 ), a tunnel oxide 1308 is formed over the trench walls. A channel material 1310 is formed over a portion of the tunnel oxide 1308. The profile of the tunnel oxide can have a step change (as shown), or an engineered dielectric profile can be used so that the thickness of the tunnel dielectric gradually increases with depth in a manner similar to that achieved for the memory holes in the 3D memory structure described and shown in the above embodiments. This method can be used for both floating gate (FG) cells and charge trapping (CTF) cells.
[0078] Fig.14a is a block diagram of an example of a system having a hardware view of a solid state drive (SSD) with a non-volatile array having a vertical channel with a conductive structure. System 1402 represents components of a non-volatile storage system that can implement a non-volatile medium with a high AR memory hole structure according to any of the examples described or according to the above-described processing sequence.
[0079] System 1402 includes SSD 1420 coupled to host 1410. Host 1410 represents a host hardware platform connected to SSD 1420. Host 1410 includes CPU (central processing unit) 1412 or other processor as a host processor or host processor device. CPU 1412 represents any host processor that generates a request to access data stored on SSD 1420, read data, or write data to a storage body. Such a processor may include a single-core or multi-core processor, a main processor for a computing device, a graphics processor, a peripheral processor, or a supplemental or auxiliary processor, or a combination. CPU 1412 can execute a host OS and other applications to cause the operation of system 1402.
[0080] The host 1410 includes a chipset 1414, which represents hardware components that may be included in the connection between the CPU 1412 and the SSD 1420. For example, the chipset 1414 may include interconnect circuits and logic to enable access to the SSD 1420. Thus, the host 1410 may include a hardware platform driver interconnect for coupling the SSD 1420 to the host 1410. The host 1410 includes hardware for interconnecting to the SSD. Likewise, the SSD 1420 includes corresponding hardware for interconnecting to the host 1410.
[0081] The host 1410 includes a controller 1416, which represents a storage controller or memory controller on the host side for controlling access to the SSD 1420. In one example, the controller 1416 is included in the chipset 1414. In one example, the controller 1416 is included in the CPU 1412. The controller 1416 may be referred to as an NV memory controller to enable the host 1410 to schedule and organize commands to the SSD 1420 to read and write data.
[0082] SSD 1420 represents a solid state drive or other storage system or module including a non-volatile (NV) medium 1430 for storing data. SSD 1420 includes HW (hardware) interface 1422, which represents a hardware component that interfaces with host 1410. For example, HW interface 1422 can be connected to one or more bus interfaces to implement a high-speed interface standard such as NVMe (Non-Volatile Memory Express) or PCIe (Peripheral Component Interconnect Express).
[0083] In one example, SSD 1420 includes NV (non-volatile) media 1430 as the main storage of SSD 1420. In one example, NV media 1430 is or includes block addressable memory technology, such as NAND (not AND) or NOR (not OR). In one example, NV media 1430 may include non-volatile media that may be block addressable or byte addressable, which stores data based on the resistance state of the memory cell or the phase of the memory cell. For example, NV media 1430 may be or include 3D XPOINT TM (3DXP) memory or a memory array based on chalcogenide phase change material (e.g., chalcogenide glass). In one example, the NV medium can be or include multi-threshold level NAND flash memory, NOR flash memory, single-level or multi-level phase change memory (PCM) or phase change memory with switch (PCMS), resistive memory, nanowire memory, ferroelectric transistor random access memory (FeTRAM), magnetoresistive random access memory (MRAM) combined with memristor technology, or spin transfer torque (STT)-MRAM, or any combination of the above, or other memory.
[0084] In one example, NV media 1430 is implemented as a plurality of dies, shown as N dies, die [0:{N-1}]. N can be any number of devices, and is typically a binary number. SSD 1420 includes a controller 1440 for controlling access to NV media 1430. Controller 1440 represents hardware and control logic within SSD 1420 for performing control of the media. Controller 1440 is internal to a non-volatile storage device or module and is separate from controller 1416 of host 1410.
[0085] The NV die of NV medium 1430 includes NV array 1432, which represents a 3D array of NV medium-based memory cells. NV array 1432 includes an array of memory holes having an engineered dielectric profile and labeled as memory holes 1434. Memory holes 1434 include one or more layers of dielectric material, such as but not limited to tunnel oxide, having an engineered dielectric profile according to one or more embodiments described and illustrated herein.
[0086] Fig.14b is a block diagram of an example of a logical view of a system having a solid state drive (SSD) with a non-volatile array having a vertical channel with a conductive structure. The system 1404 provides a Fig.14b1404 illustrates a logical layer of a host and an SSD according to a hardware platform of system 1402. System 1404 can represent software and firmware components, as well as physical components, of an example of system 1402. In one example, host 1450 provides an example of host 1410. In one example, SSD 1460 provides an example of SSD 1420.
[0087] In one example, host 1450 includes host OS 1452, which represents the host operating system or software platform of the host. Host OS 1452 can include a platform on which applications, services, agents and / or other software are executed and executed by a processor. File system 1454 represents control logic for controlling access to NV media. File system 1454 can manage what address or memory location is used to store what data. There are many known file systems, and file system 1454 can implement known file systems or other proprietary systems. In one example, file system 1454 is part of host OS 1452.
[0088] Storage driver 1456 represents one or more system-level modules that control the hardware of host 1450. In one example, driver 1456 includes a software application to control the interface to SSD 1460, and thus control the hardware of SSD 1460. Storage driver 1456 may provide a communication interface between the host and the SSD.
[0089] Controller 1470 of SSD 1460 includes firmware 1474, which represents control software / firmware for the controller. In one example, controller 1470 includes host interface 1472, which represents an interface to host 1450. In one example, controller 1470 includes media interface 1476, which represents an interface to NAND die 1462. NAND die 1462 represents a specific example of NV media and includes associated NAND array 1464, which represents a 3D NAND array.
[0090] NAND array 1464 includes an array of memory holes having an engineered dielectric profile and labeled as memory holes 1466. Memory holes 1466 include one or more layers of dielectric material, such as but not limited to tunnel oxide, having an engineered dielectric profile according to one or more embodiments described and illustrated herein.
[0091] The media interface 1476 represents the control performed on the hardware of the controller 1470. It should be understood that the controller 1470 includes hardware that interfaces with the host 1450, which can be considered to be controlled by the host interface software / firmware 1474. Likewise, it should be understood that the controller 1470 includes hardware that interfaces with the NAND die 1462. In one example, the code for the host interface 1472 can be part of the firmware 1474. In one example, the code for the media interface 1476 can be part of the firmware 1474.
[0092] In one example, the controller 1470 includes error control 1480 to handle data errors in the accessed data, as well as corner cases in compliance with signaling and communication interfaces. The error control 1480 may include implementation in hardware or firmware, or a combination of hardware and software.
[0093] Although some embodiments have been described with reference to specific embodiments, other embodiments are possible according to some embodiments. In addition, the arrangement and / or order of the elements or other features shown in the drawings and / or described herein need not be arranged in the particular manner shown and described. According to some embodiments, many other arrangements are possible.
[0094] In each system shown in the figure, in some cases, elements can each have the same reference numeral or different reference numerals to indicate that the elements represented can be different and / or similar. However, elements can be flexible enough to have different embodiments and work together with some or all of the systems shown or described herein. The various elements shown in the figure can be the same or different. Which one is called the first element and which one is called the second element is arbitrary.
[0095] In the specification and claims, the terms "coupled" and "connected" and their derivatives may be used. It should be understood that these terms are not intended to be synonyms for each other. On the contrary, in a particular embodiment, "connected" may be used to indicate that two or more elements are in direct physical or electrical contact with each other. "Coupled" may represent that two or more elements are in direct physical or electrical contact. However, "coupled" may also represent that two or more elements are not in direct contact with each other, but still cooperate or interact with each other. In addition, "communicatively coupled" means that two or more elements that may or may not be in direct contact with each other can communicate with each other. For example, if component A is connected to component B, and component B is connected to component C, then component A can be communicatively coupled to component C using component B as an intermediate component.
[0096] An embodiment is an implementation or example of the present invention. References in the specification to "an embodiment," "one embodiment," "some embodiments," or "other embodiments" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least some embodiments of the present invention, but not necessarily in all embodiments. Various appearances of "an embodiment," "one embodiment," or "some embodiments" do not necessarily refer to the same embodiment.
[0097] Not all components, features, structures, characteristics, etc. described and illustrated herein need to be included in one or more specific embodiments. For example, if the specification states that a component, feature, structure, or characteristic "may", "might", "can", or "can" be included, then that particular component, feature, structure, or characteristic is not necessarily included. If the specification or claims refer to "a" or "an" element, this does not mean that there is only one element. If the specification or claims refer to "additional" elements, it does not exclude the presence of more than one additional element.
[0098] As used herein, a list of items connected by the term "at least one of" may mean any combination of the listed terms. For example, the phrase "at least one of A, B, or C" may mean A; B; C; A and B; A and C; B and C; or A, B, and C.
[0099] The above description of the illustrated embodiments of the present invention (including what is described in the Abstract) is not intended to be exhaustive or to limit the present invention to the precise forms disclosed. Although specific embodiments and examples of the present invention are described herein for illustrative purposes, various equivalent modifications within the scope of the present invention are possible, as will be appreciated by those skilled in the relevant art.
[0100] These modifications may be made to the present invention in light of the above specific embodiments. The terms used in the following claims should not be interpreted as limiting the present invention to the specific embodiments disclosed in the specification and the drawings. On the contrary, the scope of the present invention will be determined entirely by the appended claims, which will be interpreted in accordance with the established principles of claim interpretation.
Claims
1. A method of manufacturing a three-dimensional (3D) NAND memory structure, comprising: forming a memory hole in a semiconductor structure including a plurality of stacked layers, the memory hole having a depth to diameter aspect ratio of at least 25:1; as well as A dielectric having a cross-sectional profile is formed on a sidewall of the memory hole, wherein a first thickness of the dielectric near a bottom of the memory hole is greater than or equal to a second thickness of the dielectric in at least a portion of the memory hole above the bottom of the memory hole.
2. The method according to claim 1, wherein: The dielectric includes a tunnel dielectric or a blocking dielectric film.
3. The method according to claim 1 or 2, wherein: The first thickness substantially matches a second thickness of the dielectric proximate a top of the memory hole.
4. A method according to any one of the preceding claims, wherein: The 3D NAND structure includes a stacked memory level, within which a word line layer is interposed between a pair of isolation layers, wherein fabrication of the word line layer includes: forming a cavity / aperture in the semiconductor structure, the cavity passing through a word line layer and the isolation layer in the stacked memory level; forming a groove in the cavity in the word line layer; fabricating a floating gate memory cell structure in the recess; and A tunnel dielectric is formed over sidewalls of the cavity in the isolation layer and adjacent to the memory cell structure in the word line layer, wherein the tunnel dielectric has a cross-sectional profile having a first thickness for a bottom word line layer that is thicker than a second thickness of a tunnel oxide for at least a portion of the word line layer above the bottom word line layer.
5. A method according to any one of the preceding claims, wherein: The memory cell is a floating gate memory cell or a charge trap flash (CTF) memory cell.
6. A method according to any one of the preceding claims, wherein: The dielectric is formed by: performing a first oxidation process to form a base dielectric film on the sidewall of the memory hole; performing a deposition process including performing a selective dielectric deposition over a portion of the base dielectric film in a bottom section of the memory hole; as well as A second oxidation process is performed to form a capping oxide over the base oxide and the oxide selectively deposited in the bottom section of the memory hole.
7. The method according to claim 6, wherein: Performing the selective dielectric deposition includes: performing a plasma etch to inhibit atomic layer deposition (ALD) growth of a dielectric by blocking precursor adsorption on an upper section of the memory hole; and ALD growth is performed over the base dielectric to obtain a desired dielectric thickness at a desired bottom section, wherein the ALD growth is substantially suppressed over the base dielectric film in the upper section of the memory hole.
8. The method according to claim 6 or 7, wherein: The thickness of the selective oxide deposition is 4-12 angstroms.
9. The method according to any one of claims 6 to 8, wherein: The first oxidation process and the second oxidation process include in-situ steam generation (ISSG) oxidation processes.
10. A three-dimensional (3D) memory device comprising: a plurality of memory levels, the plurality of memory levels comprising word line layers between isolation layers, the memory levels comprising a two-dimensional (2D) array of memory cells; a plurality of vertical structures passing through memory cells in the word line layer and through the isolation layer; A tunnel dielectric is formed over the sidewalls of the vertical structure, the tunnel dielectric having a cross-sectional profile wherein a first thickness for a bottom word line layer is thicker than a second thickness of the tunnel dielectric for at least a portion of the word line layer above the bottom word line layer.
11. The 3D memory device according to claim 10, wherein: A quality of the tunnel dielectric at the bottom word line layer matches a quality of the tunnel dielectric for at least one or more of an intermediate word line layer and a top word line layer.
12. The 3D memory device according to claim 10 or 11, wherein: The vertical structure includes a memory hole.
13. The 3D memory device according to any one of claims 10 to 12, wherein: The vertical structure includes a trench.
14. The 3D memory device according to any one of claims 10 to 13, wherein: The cross-sectional profile of the tunnel oxide has a third thickness for the word lines near the top of the vertical structures, the third thickness substantially matching the first thickness.
15. The 3D memory structure according to any one of claims 10 to 14, wherein: The plurality of memory cells include floating gate memory cells or charge trap flash (CTF) memory cells.
16. A system comprising: A host, the host comprising a processor; A three-dimensional (3D) NAND memory device, the 3D NAND memory device coupled to the host, the 3D NAND memory device having: a plurality of memory levels, the plurality of memory levels comprising word line layers between isolation layers, the memory levels comprising a two-dimensional (2D) array of memory cells; a plurality of memory holes passing through the memory cells in the word line layer and through the isolation layer; A dielectric is formed over the sidewalls of the memory hole, the dielectric having a cross-sectional profile having a first thickness for a bottom word line layer, the first thickness being thicker than a second thickness of the dielectric for at least a portion of the word line layer above the bottom word line layer.
17. The system of claim 16, wherein: The memory hole has a depth to diameter aspect ratio of at least 25:
1.
18. The system according to claim 16 or 17, wherein: The cross-sectional profile of the dielectric has a third thickness for the wordline layer proximate a top of the memory hole, the third thickness substantially matching the first thickness.
19. The system according to any one of claims 16 to 18, wherein: The plurality of memory cells include floating gate memory cells or charge trap flash (CTF) memory cells.
20. The system according to any one of claims 16 to 19, wherein: The plurality of memory cells include charge trap flash (CTF) memory cells, and wherein the dielectric includes a blocking dielectric and / or a tunnel dielectric.
