CELL-TYPE FLOATING GATE TEST STRUCTURE

The cell-like top layout of the floating gate test device in IC manufacturing addresses the height mismatch issue, preventing damage and shorts during logic device formation, ensuring compatibility with advanced process nodes at a low cost.

DE102018110841B4Active Publication Date: 2024-11-07TAIWAN SEMICONDUCTOR MANUFACTURING CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
DE102018110841
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2018-04-25
Filing Date
2018-05-06
Publication Date
2024-11-07
Estimated Expiration
2038-05-06

AI Technical Summary

Technical Problem

The challenge in integrated circuit (IC) manufacturing is that the height of the floating gate test device structure exceeds that of the memory cell structure, leading to potential damage and electrical shorts during the formation of logic device structures, particularly in advanced process nodes like 28 nanometer and below, due to uneven etch-back processes.

Method used

A method is introduced where the floating gate test device is designed with a cell-like top layout, comprising islands and bridges, ensuring uniform etch-back material deposition and preventing damage during logic device formation, without requiring additional reticles or photomasks, and maintaining reliability and performance.

Benefits of technology

This approach effectively prevents damage to the floating gate test device and electrical shorts, ensuring compatibility with future process nodes while maintaining reliability and performance at a low cost.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Method (2800) for forming an integrated circuit (IC), wherein the method comprises: Forming a floating-gate test fixture structure (102) on a semiconductor substrate (402), wherein the floating-gate test fixture structure comprises a first floating-gate electrode (216) and a first control-gate electrode (202) positioned above the first floating-gate electrode, wherein the first floating-gate electrode and the first control-gate electrode partially define an array of islands (106i) and further partially define a plurality of bridges (106b) connecting the islands; Forming a memory cell structure (204) on the semiconductor substrate (402), wherein the memory cell structure comprises a second floating-gate electrode (218) and a second control-gate electrode (208) located above the second floating-gate electrode; and Deposition of a back-etch layer (1602) covering the memory cell structure (204) and the floating-gate test device structure (102) with the array of islands and the plurality of bridges, wherein the back-etch layer has a first thickness (T1) directly over the first control gate electrode (202) and a second thickness (T2) directly over the second control gate electrode (208), wherein the first and second thicknesses are equal or substantially equal.
Need to check novelty before this filing date? Find Prior Art

Description

BACKGROUND

[0001] The integrated circuit (IC) industry has experienced exponential growth in recent decades. In IC development, the functional density (i.e., the number of interconnected devices per unit area of ​​the chip) has generally increased, while the geometric size (i.e., the smallest component that can be produced) has decreased. Some advances in IC development include embedded memory technology. Embedded memory technology is the integration of memory devices with logic devices on the same semiconductor chip, so that the memory devices support the operation of the logic devices. Embedded memory finds applications in smart cards and automotive devices, among others.

[0002] US Patent 2007 / 0026612 A1 describes a method for manufacturing a flash memory device. The flash memory device has a memory area and a test area. Several floating-gate transistors are formed in the memory area. Furthermore, several test transistors are formed in the test area, which are configured to analyze the characteristics of the memory cells within the memory area. US 5,864,501 A describes a test structure for a flash memory device. The test structure comprises two separate floating gates, each extending substantially along a first direction. At least three drain regions are arranged between the two floating gates, and source regions are arranged on opposite sides of the floating gates. A control electrode is located above the two floating gates, with the sections of the control electrode located above the two floating gates connected to each other at their ends. BRIEF DESCRIPTION OF THE DRAWINGS

[0003] Aspects of this disclosure are best understood from the following detailed description when read in conjunction with the accompanying drawings. Note that, in accordance with standard industry practice, several features are not drawn to scale. In fact, the dimensions of the various features may be enlarged or reduced as desired for clarity of description. The Fig. 1A and Fig. Figure 1B shows top views of different embodiments of an integrated circuit (IC) with a floating-gate test device having a cell-like top layout. The Fig. Figures 2A-2F show different views of the top layout of some more detailed embodiments of the IC. Fig. 1A, in which the IC comprises a memory cell array. The Fig. Figures 3A-3H show different views of the top layout of some more detailed embodiments of the IC. Fig. 1B, in which the IC comprises a memory cell array. The Fig. Figures 4A-4D show different cross-sectional views of some embodiments of the IC. Fig. 2A and Fig. 2B and / or the IC of the Fig. 3A and Fig. 3B. The Fig. Figures 5A and 5B to 27A and 27B show a series of cross-sectional views of some embodiments of a method for forming an IC comprising a floating-gate test device with a cell-like top layout. Fig. 28 shows a flowchart of some embodiments of the method of Fig. 5A and 5B to 27A and 27B. The Fig. Figures 29A and 29B to 41A and 41B show a series of cross-sectional views of some alternative embodiments of the method of Fig. 5A and 5B to 27A and 27B, in which the procedure features a gate replacement procedure. DETAILED DESCRIPTION

[0004] The present disclosure provides many different embodiments or examples for implementing various features of this disclosure. Specific examples of components and arrangements are described below to simplify the present disclosure. These are, of course, only examples. For instance, forming a first feature over or on top of a second feature in the following description may include embodiments in which the first and second features are in direct contact, and may also include embodiments in which additional features may be formed between the first and second features, so that the first and second features need not be in direct contact. Additionally, the present disclosure may repeat reference numerals and / or letters in the various examples.This repetition serves the purpose of simplicity and clarity and does not in itself imply any relationship between the various described embodiments and / or configurations.

[0005] Furthermore, spatially relative terms such as "below," "under," "lower," "above," "upper," and similar terms may be used here for the sake of simplicity to describe the relationship of a feature or device with one or more other features or devices, as shown in the figures. These spatially relative terms are intended to encompass various orientations of the device or apparatus in use or operation in addition to the orientation shown in the figures. The device or apparatus may be oriented differently (rotated by 90 degrees or in other orientations), and the spatially relative terms used herein may be interpreted accordingly. Furthermore, the terms "first," "second," "third," "fourth," and the like are merely generic designations and may be interchanged as such in different embodiments.For example, while in some embodiments an element (e.g. an opening) may be referred to as a "first" element, in other embodiments the element may be referred to as a "second" element.

[0006] According to a method for fabricating an integrated circuit (IC) with embedded memory technology, a memory cell structure and a floating-gate test fixture structure are formed on a semiconductor memory area of ​​a semiconductor substrate. The floating-gate test fixture structure comprises a floating-gate dielectric layer and a floating-gate electrode located above the floating-gate dielectric layer. The floating-gate test fixture can be used, for example, to test the quality of the floating-gate dielectric layer, as its quality can be representative of the lifetime of the memory cell structure. A hard mask layer is formed covering the semiconductor substrate, and planarization is performed on one of the top surfaces of the hard mask layer to flatten the surface.The hard mask layer is patterned to remove the hard mask layer from a semiconductor logic region of the semiconductor substrate and to define a memory hard mask that covers the memory cell structure and the floating-gate test fixture structure, but not the semiconductor logic region. Residual material remaining from the formation of the memory cell structure and the floating-gate test fixture structure is removed from the semiconductor logic region, leaving the memory hard mask in place. Then, logic fixture layers are deposited and patterned on the semiconductor logic region to form a logic fixture structure on the semiconductor logic region.

[0007] One challenge with the method is that the height of the memory cell structure and the floating-gate test fixture structure exceeds the height of the logic fixture structure, thus preventing its application in future process nodes. A solution is to reduce the height of the floating-gate test fixture structure and the memory cell structure before forming the memory hard mask. A back-etch layer is deposited covering the memory cell structure and the floating-gate test fixture structure, and then non-selective back-etching is performed to reduce the height of the memory cell structure and the floating-gate test fixture structure, respectively. The non-selective back-etching process etches the back-etch layer, the memory cell structure, and the floating-gate test fixture structure until the height of the memory cell structure and the floating-gate test fixture structure is sufficiently reduced.

[0008] While non-selective etching effectively reduces the height of the memory cell structure, it is ineffective in reducing the height of the floating-gate test fixture structure. The floating-gate test fixture structure has a top floating-gate internal layout that covers a large area of ​​the semiconductor substrate compared to the top floating-gate internal layout of the memory cell structure. For example, a top internal layout might be a single surface instead of several smaller surfaces, with a single border extending in a closed path. The top floating-gate internal layouts extend to a top surface of the floating-gate test fixture structure and a top surface of the memory cell structure, respectively, so the floating-gate test fixture structure has a large top internal layout, and the top surface of the memory cell structure has a small top internal layout.The difference in surface area between the large upper internal layout and the small upper internal layout causes the etching material of the etch layer to deposit more readily on the floating-gate test fixture structure than on the memory cell structure during deposition. This results in a thicker etch layer on the floating-gate test fixture structure than on the memory cell structure. Because more etching material is available on the floating-gate test fixture structure to etch through, the floating-gate test fixture structure is hardly etched during non-selective etching compared to the memory cell structure. Furthermore, the height of the floating-gate test fixture structure exceeds that of the memory cell structure at the end of the non-selective etching process.

[0009] Because the height of the floating-gate test fixture structure exceeds that of the memory cell structure, and the top surface of the memory hardmask is flat or substantially flat, the first thickness of the memory hardmask on the floating-gate test fixture structure is less than the second thickness of the memory hardmask on the memory cell structure. As a result, the floating-gate test fixture structure has less protection and can be destroyed or severely damaged during logic structure formation. For example, removing residual material on the semiconductor logic area can partially remove the memory hardmask. Due to the reduced thickness of the memory hardmask on the floating-gate test fixture structure, this removal can also expose the floating-gate test fixture structure but not the memory cell structure, creating an opening that extends through the floating-gate test fixture structure.The opening can extend to the floating-gate electrode and the semiconductor memory area, potentially facilitating an electrical short circuit between the semiconductor memory area and the floating-gate electrode during subsequent back-end-of-line (BEOL) processing.

[0010] In light of the foregoing, various embodiments of the present application provide a method for forming an IC comprising a floating-gate test fixture structure with a cell-like top layout. In some embodiments, a floating-gate test fixture structure is formed on a semiconductor substrate. The floating-gate test fixture structure comprises a first floating-gate electrode and a first control-gate electrode located above the first floating-gate electrode. The first floating-gate electrode and the first control-gate electrode partially define an array of islands and further partially define a plurality of bridges. The bridges connect the islands to one another. A memory cell structure is formed on the semiconductor substrate and comprises a second floating-gate electrode and a second control-gate electrode located above the second floating-gate electrode.A back-etch layer is formed, covering the floating-gate test device structure and the memory cell structure. The back-etch layer has a first thickness directly above the first control gate electrode and a second thickness directly above the second control gate electrode. The first and second thicknesses are equal or substantially equal.

[0011] By forming the first floating-gate electrode, which includes the islands and bridges, the floating-gate electrode exhibits a cell-like top layout. This cell-like top layout is characterized by multiple interconnected small areas instead of one large area and resembles the top layout of a memory cell array. The cell-like top layout can prevent etch-back material from more readily adhering to the floating-gate test fixture structure compared to the memory cell structure. This, in turn, can prevent the floating-gate test fixture structure from being destroyed or damaged during logic device formation and / or can prevent an electrical short circuit between the semiconductor substrate and the first floating-gate electrode during BEOL processing.The method can be performed at low cost, without additional reticles or photomasks, and without compromising the reliability and performance of the floating-gate test fixture structure. Furthermore, the method can be compatible with future process nodes, such as 28-nanometer, 20-nanometer, 16-nanometer, and smaller process nodes.

[0012] With reference to Fig. Figure 1A shows a top layout view 100A of some embodiments of an IC with a floating-gate test device 102 having a cell-like top layout. The floating-gate test device 102 is located on a semiconductor memory area 104m and comprises a gate stack 106. As can be seen below, the gate stack 106 is defined by several gate electrodes (e.g., a floating-gate electrode and a control-gate electrode) stacked along an axis extending into and out of the plane of the image. The gate stack 106 comprises a plurality of islands 106i and a plurality of bridges 106b. For illustration, the islands 106i and the bridges 106b are shown separately. However, it is understood that the islands 106i may be integrated and / or continuous with the bridges 106b.

[0013] The islands 106i are arranged in one or more rows and one or more columns such that they define an island array. In some embodiments, the islands 106i are arranged in a single row and multiple columns or multiple rows and a single column. For example, the islands 106i may be arranged in three columns and one row. Furthermore, in some embodiments, the islands 106i share a common top layout. For example, the islands 106i may have a rectangular top layout, a square top layout, a triangular top layout, or one or more other suitable top layouts. As used herein, an expression (e.g., top layout) may mean, for example, the singular or plural.The bridges 106b are arranged between the islands 106i to connect the islands 106i to one another, and each bridge 106b extends from direct contact with an associated first island of the islands 106i to an associated second island of the islands 106i that is adjacent to the associated first island. In some embodiments, the bridges 106b comprise one bridge for each pair of adjacent islands in the island array, and the bridge extends from direct contact with a first island of the pair to a second island of the pair. Furthermore, in some embodiments, the bridges 106b share a common top layout and / or the common top layout of the bridges 106b is smaller than the common top layout of the islands 106i. For example, the bridges 106b may have a rectangular top layout, a square top layout, a triangular top layout, or one or more other suitable top layouts.The common upper layout of the bridges 106b may, for example, be smaller than the common upper layout of the islands 106i in terms of area, width, length or one or more other suitable parameters.

[0014] In some embodiments, a plurality of test fixture selection gate electrodes 108 are adjacent to the gate stack 106. The test fixture selection gate electrodes 108 may, for example, be made of or comprise metal, doped polysilicon, or one or more other suitable conductors. In some embodiments, the test fixture selection gate electrodes 108 comprise a plurality of recessed selection gate electrodes 108r that are arranged wholly or substantially wholly within recesses defined by the gate stack 106. For the sake of simplicity, only some of the recessed selection gate electrodes 108r are designated 108r. The recessed selection gate electrodes 108r may, for example, have a U-shaped top layout or one or more other suitable top layouts.Furthermore, in some embodiments, the test device selection gate electrodes 108 comprise a plurality of peripheral selection gate electrodes 108p arranged on the sides of the gate stack 106 and located completely or substantially completely outside of recesses defined by the gate stack 106. For the sake of simplicity, only one of the peripheral selection gate electrodes 108p is designated 108p. The peripheral selection gate electrodes 108p may, for example, have a row-shaped top layout or one or more other suitable top layouts.

[0015] By forming the floating-gate test fixture 102 with the islands 106i and the bridges 106b, the floating-gate test fixture 102 has a cell-like top layout consisting of several small interconnected areas instead of one large area. As will be shown below, this can prevent the floating-gate test fixture 102 from being destroyed or damaged while a logic device of the IC is being formed, and / or it can prevent an electrical short circuit between the semiconductor memory area 104m and a floating-gate electrode (not shown) of the gate stack 106. Furthermore, as will be shown below, the floating-gate test fixture 102 with the cell-like top layout can be formed at low cost without additional reticles or photomasks and without compromising the reliability and performance of the floating-gate test fixture 102.As can be seen below, the Floating Gate Test Device 102 is also compatible with future process nodes, such as 28-nanometer process nodes, 20-nanometer process nodes, 16-nanometer process nodes and smaller process nodes.

[0016] With reference to Fig. Figure 1B shows a top layout view of Figure 100B of some further embodiments of an IC comprising a floating-gate test device 102 with a cell-like top layout. Fig. 1B is a variant of Fig. 1A, in which the islands 106i are arranged in a plurality of rows and a plurality of columns. For example, the islands 106i may be arranged in three rows and three columns. Furthermore, in some embodiments, the test device selection gate electrodes 108 comprise one or more enclosed selection gate electrodes 108e that are completely enclosed by the gate stack 106. For the sake of simplicity, only some of the enclosed selection gate electrodes 108e are designated 108e. The enclosed selection gate electrodes 108e may, for example, have an annular top layout or one or more other suitable top layouts. As used herein, an annular top layout is a ring-shaped top layout that is not restricted to circular boundaries or side walls.

[0017] With reference to Fig. 2A is a top layout view. 200A shows some more detailed embodiments of the IC. Fig. 1A provided. The floating-gate test device 102 comprises a test device control gate electrode 202, which partially defines the gate stack 106. The test device control gate electrode 202 may, for example, consist of or comprise metal, doped polysilicon, some other suitable conductors, or a combination thereof. Furthermore, the test device control gate electrode 202 comprises a plurality of control gate islands 202i and a plurality of control gate bridges 202b. For illustration, the separation of the control gate islands 202i and the control gate bridges 202b is shown separately. However, it is understood that the control gate islands 202i may be integrated and / or continuous with the control gate bridges 202b.

[0018] The control gate islands 202i each partially define the islands 106i of the gate stack 106 and are arranged in one or more rows and one or more columns to define a control gate island array. In some embodiments, the control gate island array has a single row and multiple columns or multiple rows and a single column. Furthermore, in some embodiments, the control gate islands 202i share a common top layout. For example, the control gate islands 202i may share a rectangular top layout, a square top layout, or one or more other suitable top layouts. The control gate bridges 202b each partially define the bridges 106b of the gate stack 106 and are arranged between the control gate islands 202i to connect the control gate islands 202i to one another.Furthermore, the control gate bridges 202b comprise one control gate bridge for each pair of adjacent control gate islands in the control gate island array, and the control gate bridge extends from direct contact with a first island of the pair to a second island of the pair. In some embodiments, the control gate bridges 202b share a common top layout. For example, the control gate bridges 202b may share a rectangular top layout or one or more other suitable top layouts.

[0019] In some embodiments, the floating-gate test device 102 is adjacent to a memory cell array 204. Furthermore, in some embodiments, the memory cell array 204 lies on the semiconductor memory area 104m and / or the semiconductor memory area 104m is continuous from the floating-gate test device 102 to the memory cell array 204. The memory cell array 204 comprises a plurality of memory cells 204c, which are arranged in a plurality of rows and a plurality of columns. For the sake of simplicity, only some of the memory cells 204c are labelled 204c. The memory cells 204c comprise a plurality of memory control gate electrodes 208, a plurality of memory select gate electrodes 210, and a plurality of memory erase gate electrodes 212.For the sake of simplicity, only some of the storage control gate electrodes 208 are designated 208, only some of the storage selection gate electrodes 210 are designated 210, and only some of the storage erase gate electrodes 212 are designated 212. The storage control gate electrodes 208, the storage selection gate electrodes 210, and the storage erase gate electrodes 212 can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductors.

[0020] Each of the memory cells 204c comprises an associated control gate electrode of the memory control gate electrodes 208, an associated select gate electrode of the memory select gate electrodes 210, and an associated erase gate electrode of the memory erase gate electrodes 212. The associated select gate electrode adjoins a first side wall of the associated control gate electrode. The associated erase gate electrode adjoins a second side wall of the associated control gate electrode, which is opposite the first side wall. Furthermore, in some embodiments, the memory cells 204c are grouped in pairs, and the memory cells of each pair share the associated erase gate electrode.

[0021] In some embodiments, several intermediate device lines 214 are electrically coupled to the memory cell array 204. For the sake of simplicity, only some of the intermediate device lines 214 are designated 214. The intermediate device lines 214 are elongated in a common direction and each extends in the common direction along a corresponding row or column of the memory cell array 204. In some embodiments, the intermediate device lines 214 have row-shaped top layouts or one or more other suitable elongated top layouts. The intermediate device lines 214 are conductive and can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductors.The intermediate device lines 214 comprise a plurality of selection gate intermediate device lines 214s, a plurality of deletion gate intermediate device lines 214e, and a plurality of control gate intermediate device lines 214c. For the sake of simplicity, only one of the selection gate intermediate device lines 214s is designated 214s, only one of the deletion gate intermediate device lines 214e is designated 214e, and only one of the control gate intermediate device lines 214c is designated 214c.

[0022] The selection gate intermediate device lines 214s are each integrated and continuous with selection gate electrodes of the memory selection gate electrodes 210 in the corresponding row or column of the memory cell array 204. The erase gate intermediate device lines 214e are each integrated and continuous with erase gate electrodes of the memory erase gate electrodes 212 in the corresponding row or column of the memory cell array 204. The control gate intermediate device lines 214c are each integrated and continuous with control gate electrodes of the memory control gate electrodes 208 in the corresponding row or column of the memory cell array 204. Furthermore, in some embodiments, at least one of the control gate intermediate device lines 214c extends beyond the memory cell array 204 to the floating gate test device 102 and is integrated and continuous with the test device control gate electrode 202.

[0023] It should be noted that the division along the selection gate intermediate device lines 214s is only separate to emphasize the storage selection gate electrodes 210. Furthermore, the division along the erase gate intermediate device lines 214e is only separate to emphasize the storage erase gate electrodes 212. Furthermore, the division along the control gate intermediate device lines 214c is only separate to emphasize the storage control gate electrodes 208. Therefore, the separate divisions should not be interpreted as restricting the composition of the selection gate intermediate device lines 214s, the storage selection gate electrodes 210, the erase gate intermediate device lines 214e, the storage erase gate electrodes 212, the control gate intermediate device lines 214c, and the storage control gate electrodes 208.

[0024] With reference to Fig. 2B is another top layout view. 200B shows some more detailed embodiments of the IC. Fig. 1A provided. While the upper layout view 200A of Fig. 2A shows a control gate layout of the IC, the upper layout view 200B shows. Fig. 2B shows a floating-gate layout of the IC. As will become clearer below, the control gate layout overlays the floating-gate layout along an axis that extends into and out of the image plane.

[0025] The floating-gate test device 102 comprises a test device floating-gate electrode 216, which partially defines the gate stack 106. The test device floating-gate electrode 216 can, for example, consist of or comprise metal, doped polysilicon, one or more other suitable conductors, or any combination thereof. Furthermore, the test device floating-gate electrode 216 comprises a plurality of floating-gate islands 216i and a plurality of floating-gate bridges 216b. For illustrative purposes, the division between the floating-gate islands 216i and the floating-gate bridges 216b is shown separately. However, it is understood that the floating-gate islands 216i can be continuous and / or integrated with the floating-gate bridges 216b.

[0026] The floating-gate islands 216i each partially define the islands 106i of the gate stack 106 and are arranged in one or more rows and one or more columns to define a floating-gate island array. In some embodiments, the floating-gate island array has a single row and multiple columns or multiple rows and a single column. Furthermore, in some embodiments, the floating-gate islands 216i share a common top layout. For example, the floating-gate islands 216i may share a rectangular top layout, a square top layout, or one or more other suitable top layouts. The floating-gate bridges 216b each partially define the bridges 106b of the gate stack 106 and are arranged between the floating-gate islands 216i to connect them.Furthermore, the floating-gate bridges 216b comprise one floating-gate bridge for each pair of adjacent floating-gate islands in the floating-gate island array, and the floating-gate bridge extends from direct contact with a first island of the pair to a second island of the pair. In some embodiments, the floating-gate bridges 216b share a common top layout. For example, the floating-gate bridges 216b may share a rectangular top layout or one or more other suitable top layouts. The floating-gate bridges 216b have a floating-gate bridge width W. fb In some embodiments, each of the floating-gate bridges 216b individually has the floating-gate bridge width W. fb on.

[0027] The memory cells 204c comprise the majority of memory selection gate electrodes 210 and the majority of memory erase gate electrodes 212, and further comprise a majority of memory floating gate electrodes 218. For the sake of simplicity, only some of the memory cells 204c are labelled 204c, only some of the memory floating gate electrodes 218 are labelled 218, only some of the memory selection gate electrodes 210 are labelled 210, and only some of the memory erase gate electrodes 212 are labelled 212. Each of the memory cells 204c comprises an associated floating-gate electrode of the memory floating-gate electrodes 218, an associated selection gate electrode of the memory selection gate electrodes 210, and an associated erase gate electrode of the memory erase gate electrodes 210. The associated selection gate electrode borders a first side wall of the associated floating-gate electrode.The associated quenching gate electrode borders a second side wall of the associated floating gate electrode, which is opposite the first side wall. The storage quenching gate electrodes 212 have a quenching gate width W. eg In some embodiments, each of the quenching gate electrodes 212 has individually the quenching gate width W. eg In some embodiments, the floating-gate bridge width W fb greater than or equal to the deletion gate width W eg .

[0028] With regard to the Fig. 2C and Fig. 2D are top layout views 200C, 200D of some alternative embodiments of the top layout views 200A, 200B of the Fig. Figures 2A and 2B show the control gate islands 202i and the floating gate islands 216i as square (e.g., all four sides are of equal or approximately equal length). However, other shapes and / or sizes are possible in further embodiments.

[0029] With regard to the Fig. 2E and Fig. 2F are top layout views 200E, 200F of some alternative embodiments of top layout views 200A, 200B of the Fig. 2A and 2B are provided, in which an additional island 106i and an additional bridge 106b have been added to a single side of the gate stack 106. In further embodiments, more islands and more bridges can be added to the single side of the gate stack 106 and / or to other sides of the gate stack 106.

[0030] With regard to the Fig. 3A and Fig. 3B are various top layout views 3004, 300B some more detailed embodiments of the IC of Fig. 1B provided. The upper layout views 300A, 300B of the Fig. 3A and Fig. 3B are variants of Fig. 2A and 2B, in which the floating-gate test device 102 forms the cell-like upper layout of the Fig. 1B used. Fig. Figure 3A shows a control gate layout of the IC and Fig. Figure 3B shows a floating-gate layout of the IC. The floating-gate bridges 216b have a floating-gate bridge width W. fb (see Fig. 3B) and the storage quenching gate electrodes 212 have a quenching gate width W eg In some embodiments, each of the floating-gate bridges 216b individually has the floating-gate bridge width W. fb on and / or each of the quenching gate electrodes 212 individually has the quenching gate width W eg on. In some embodiments, the floating-gate bridge width W fb greater than or equal to the deletion gate width W eg .

[0031] With regard to the Fig. 3C and Fig. 3D are top layout views 300C, 300D of some alternative embodiments of the top layout views 300A, 300B of the Fig. 3A and 3B are provided, in which the control gate islands 202i and the floating gate islands 216i are square (e.g., all four sides are of equal or approximately equal length). However, other shapes and / or sizes are possible in further embodiments.

[0032] With regard to the Fig. 3E and Fig. 3F are top layout views 300E, E00F of some alternative embodiments of top layout views 300A, 300B of the Fig. 3A and 3B are provided, in which one fewer row of islands 106i is present. In further embodiments, more or fewer rows of islands 106i and / or more or fewer columns of islands 106i may be present.

[0033] With regard to the Fig. 3G and Fig. 3H are top layout views 300G, 300H of some alternative embodiments of the top layout views 300A, 300B of the Fig. 3A and 3B are provided, in which an additional island 106i and an additional bridge 106b have been added to a single side of the gate stack 106. In further embodiments, more islands and more bridges can be added to the single side of the gate stack 106 and / or to other sides of the gate stack 106.

[0034] With reference to Fig. 4A is a cross-sectional view 400A of some embodiments of the IC of the Fig. 2A and Fig. 2B and / or the IC of the Fig. 3A and Fig. 3B provided. The cross-sectional view 400A can, for example, be viewed along line AA' in the Fig. 2A and Fig. 2B and / or along line AA' in the Fig. 3A and Fig. 3B. Additionally, the cross-sectional view 400A can be taken, for example, along line AA' in Fig. 1A and / or Fig. 1B has been taken.

[0035] As shown, a semiconductor substrate 402 comprises the semiconductor memory region 104m, and an insulating structure 404 extends into an upper surface of the semiconductor substrate 402 to delimit a boundary of the semiconductor memory region 104m. The semiconductor substrate 402 may, for example, consist of or comprise a bulk silicon substrate, a silicon-on-insulator (SOI) substrate, or one or more other suitable semiconductor substrates. The insulating structure 404 electrically isolates the semiconductor memory region 104m from adjacent regions of the semiconductor substrate and may, for example, comprise a trench filled with a dielectric. In some embodiments, the insulating structure 404 comprises a shallow trench isolation (STI) region, a deep trench isolation (DTI) region, or one or more other suitable insulating structures.

[0036] The floating-gate test fixture 102 is located above the semiconductor memory area 104m and comprises the gate stack 106. The gate stack 106 has a cell-like top layout and comprises the majority of islands 106i. Examples of the cell-like top layout are found, for example, in the Fig. 1A and Fig. 1B shown. While the islands 106i are shown as unconnected, the islands 106i can, for example, be located outside the cross-sectional view 400A of Fig. 4A connected. The gate stack 106 is defined by a test device floating gate dielectric layer 406, the test device floating gate electrode 216, a test device control gate dielectric layer 408, the test device control gate electrode 202 and a test device control gate hard mask 410, which are stacked on top of each other.

[0037] The test device floating gate electrode 216 is located above the test device floating gate dielectric layer 406, the test device control gate dielectric layer 408 is located above the test device floating gate electrode 216, the test device control gate electrode 202 is located above the test device control gate dielectric layer 408, and the test device control gate hard mask 410 is located above the test device control gate electrode 202. The test device floating gate electrode 216 and the test device control gate electrode 202 can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductive materials.The test device floating gate dielectric layer 406, the test device control gate dielectric layer 408, and the test device control gate hardmask 410 can, for example, consist of or comprise silicon oxide, silicon nitride, one or more other suitable dielectrics, or any combination thereof. Furthermore, the test device floating gate dielectric layer 406, the test device floating gate electrode 216, the test device control gate dielectric layer 408, the test device control gate electrode 202, and the test device control gate hardmask 410 have cell-like top layouts. In some embodiments, the test device floating gate dielectric layer 406 and the test device floating gate electrode 216 have the same or substantially the same cell-like top layout. Examples of this cell-like top layout can be found, for example, in the . Fig. 2B and Fig. Figure 3B shows the test device floating-gate electrode 216. In some embodiments, the test device control-gate dielectric layer 408, the test device control-gate electrode 202, and the test device control-gate hard mask 410 have the same or substantially the same cell-like upper layout. Examples of this cell-like upper layout are shown, for example, in the Fig. 2A and Fig. 3A shown in relation to the test device control gate electrode 202.

[0038] While the test fixture floating gate dielectric layer 406, the test fixture floating gate electrode 216, the test fixture control gate dielectric layer 408, the test fixture control gate electrode 202, and the test fixture control gate hard mask 410 are each shown with several segments belonging to the islands 106i, the segments can be, for example, shown outside the cross-sectional view 400A of Fig. 4A connected. To simplify the representation, only one segment was labelled for each of the test device floating gate dielectric layer 406, the test device floating gate electrode 216, the test device control gate dielectric layer 408, the test device control gate electrode 202, and the test device control gate hard mask 410.

[0039] A test device control gate spacer 412 lines the side walls of the test device control gate electrode 202, the side walls of the test device control gate dielectric layer 408, and the side walls of the test device control gate hard mask 410, and comprises a plurality of control gate spacer segments adapted to the side walls. For the sake of simplicity, only some of the control gate spacer segments are designated 412. Furthermore, the test device control gate spacer 412 lies above the test device floating gate electrode 216 and, in some embodiments, the insulation structure 404. The test device control gate spacer 412 can, for example, consist of or comprise silicon oxide, silicon nitride, one or more other suitable dielectrics, or any combination thereof.

[0040] A test fixture floating-gate spacer 414 lines the side walls of the test fixture floating-gate electrode 216, the side walls of the test fixture floating-gate dielectric layer 406, and the side walls of the test fixture control-gate spacer 412, and comprises a plurality of floating-gate spacer segments adapted to the side walls. For the sake of simplicity, only some of the floating-gate spacers are designated 414. Furthermore, the test fixture floating-gate spacer 414 lies above the semiconductor memory area 104m and, in some embodiments, above the insulation structure 404. The test fixture floating-gate spacer 414 can, for example, consist of or comprise silicon oxide, silicon nitride, one or more other suitable dielectrics, or any combination thereof.

[0041] A plurality of test device selection gate electrodes 108 abut side walls of the gate stack 106, and a plurality of test device selection gate dielectric layers 416 space the test device selection gate electrodes 108 from the semiconductor memory area 104m. For the sake of simplicity, only some of the test device selection gate electrodes 108 are designated 108, and only one of the test device selection gate dielectric layers 416 is designated 416. The test device selection gate electrodes 108 can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductive materials. The test device selection gate dielectric layers 416 can, for example, consist of or comprise silicon oxide, one or more other suitable dielectrics, or any combination thereof.

[0042] A BEOL compound structure 418 covers the floating-gate test device 102 and comprises a plurality of compound dielectric layers covering the floating-gate test device 102. The compound dielectric layers comprise an underwire dielectric layer 420a, a first ILD layer 420b located above the underwire dielectric layer 420a, and a second ILD layer 420c located above the first ILD layer 420b. The underwire dielectric layer 420a and the first and second ILD layers 420b, 420c can, for example, consist of or comprise silicon oxide, silicon nitride, silicon oxynitride, a low-k dielectric, one or more other suitable dielectrics, or any combination thereof. As used herein, a low-k dielectric is a dielectric with a dielectric constant k of less than about 3.9, 3, 2 or 1.In some embodiments, a contact etch stop layer 422 covers the floating-gate test device 102 between the floating-gate test device 102 and the under-wire dielectric layer 420a and is adapted to it. The contact etch stop layer 422 can, for example, consist of or comprise silicon nitride, silicon carbide, one or more other suitable dielectrics, or any combination thereof.

[0043] The BEOL interconnect structure 418 further comprises a wire 424 and a plurality of vias 426. For the sake of simplicity, only some of the vias 426 are labelled 426. The wire 424 covers and contacts the vias 426 and also lies above the floating-gate test device 102. Furthermore, the wire 424 is embedded in a top surface of the first ILD layer 420b and is covered by the second ILD layer 420c. The vias 426 are each located above the islands 106i of the gate stack 106. Furthermore, the vias 426 extend from an upper surface of the under-wire dielectric layer 420a through the under-wire dielectric layer 420a and the test device control gate electrode 202 such that they touch an upper surface of the test device floating gate electrode 216.It should be noted that the openings in the islands 106i, through which the vias 426 extend to the test device floating gate electrode 216, are in the . Fig. Figures 2A-2F and 3A-3H are not shown for clarity. In some embodiments, the vias 426 are electrically insulated from the test device control gate electrode 202 by dielectric linings (not shown) that separate the vias 426 from side walls of the test device control gate electrode 202. The wire 424 and the vias 426 may, for example, consist of or comprise copper, aluminum, aluminum-copper, tungsten, one or more other suitable conductors, or any combination thereof.

[0044] The vias 426 have a via width W cv and the control gate islands 202i has a control gate island width W ciIn some embodiments, each of the vias 426 has an individual via width W. cv and / or each of the control gate islands 202i has individually the control gate island width W ci In some embodiments, the control gate island width W ci at least three times as large as the via width W cv . Such embodiments can occur when the upper layouts of the control gate islands 202i are square, with examples of this shown in the Fig. 2C and Fig. 3C are shown. If the control gate island width W ci less than three times the diameter of the via W cv The reliability of the Floating-Gate Test Device 102 is low and the mass production yield is low.

[0045] In some embodiments, a plurality of test fixture source / drain regions 427 are located in the semiconductor substrate 402, and each test fixture source / drain region 427 is formed between two opposing gate electrodes of the test fixture selection gate electrodes 108. In other embodiments, the test fixture source / drain regions 427 are omitted. In some embodiments, the BEOL interconnect structure 418 defines a conductive path (not shown) that electrically couples the test fixture source / drain regions 427 to the test fixture floating gate electrode 216. The test fixture source / drain regions 427 can be, for example, of the n-type or the p-type.

[0046] In some embodiments, several silicide pads 428 are located on an upper surface of the test device floating-gate electrode 216, an upper surface of the test device source / drain regions 427, upper surfaces of the test device selection-gate electrodes 108, an upper surface of another structure, or any combination thereof. For the sake of simplicity, only some of the silicide pads 428 are designated 428. The silicide pads 428 may, for example, consist of or comprise nickel silicide or one or more other suitable silicides.

[0047] With reference to Fig. 4B is another cross-sectional view 400B of some embodiments of the IC of the Fig. 2A and Fig. 2B and / or the IC of the Fig. 3A and Fig. 3B provided. The cross-sectional view 400B of Fig. 4B can, for example, run along line BB' into the Fig. 2A and Fig. 2B and / or along line BB' in the Fig. 3A and Fig. 3B. Additionally, the cross-sectional view 400B can be taken, for example, along line BB' in Fig. 1A and / or Fig. 1B has been taken.

[0048] As shown, the gate stack 106 further comprises the majority of bridges 106b that connect the islands 106i. The bridges 106b comprise one bridge for each pair of adjacent islands, and the bridge extends from direct contact with a first island of the pair to a second island of the pair. While the islands 106i and the bridges 106b have similar layouts in cross-sectional view 400B, it should be noted that the islands 106i and the bridges 106b have different top layouts. See, for example, the Fig. 1A and Fig. 1B.

[0049] With reference to Fig. 4C is another cross-sectional view 400C of some embodiments of the IC of the Fig. 2A and Fig. 2C and / or the IC of the Fig. 3A and Fig. 3B provided. The cross-sectional view 400C of Fig. 4C can, for example, be found along line CC' in the Fig. 2A and Fig. 2B and / or along line CC' in the Fig. 3A and Fig. 3B has been taken.

[0050] As shown, a pair of memory cells 204c is located on the semiconductor memory area 104m. The memory cells 204c comprise a pair of individual memory source / drain regions 430i and a common memory source / drain region 430c. For the sake of simplicity, only one of the individual memory source / drain regions 430i is labeled 430i. The individual memory source / drain regions 430i and the common memory source / drain region 430c are located in the semiconductor substrate 402 along an upper surface of the semiconductor substrate 402. Furthermore, the individual memory source / drain regions 430i are spaced apart from the common memory source / drain region 430c and are located on opposite sides of the common memory source / drain region 430c.

[0051] A pair of selectively conductive memory channels 432 is located in the semiconductor substrate 402. The selectively conductive memory channels 432 extend along the top surface of the semiconductor substrate 402 from the common memory source / drain region 430c to the individual memory source / drain regions 430i. Furthermore, the selectively conductive memory channels 432 have an opposite doping type to the common memory source / drain region 430c and the individual memory source / drain regions 430i. For example, the selectively conductive memory channels 432 can be of the p-type, whereas the common memory source / drain region 430c and the individual memory source / drain regions 430i can be of the n-type, or vice versa.

[0052] A pair of storage floating-gate dielectric layers 434, a pair of storage floating-gate electrodes 218, a pair of storage control-gate dielectric layers 436, a pair of storage control-gate electrodes 208, and a pair of storage control-gate hard masks 438 are stacked on the selectively conductive storage channels 432. The storage floating-gate dielectric layers 434 each lie above the selectively conductive storage channels 432 and can, for example, consist of or comprise silicon oxide, one or more other suitable dielectrics, or any combination thereof. The storage floating-gate electrodes 218 each lie above the storage floating-gate dielectric layers 434 and can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductors.The storage control gate dielectric layers 436 are each located above the storage floating gate electrodes 218 and can consist of, for example, silicon dioxide, silicon nitride, one or more other suitable dielectrics, or any combination thereof. The storage control gate electrodes 208 are each located above the storage control gate dielectric layers 436 and can consist of, for example, metal, doped polysilicon, or one or more other suitable conductors. The storage control gate hard masks 438 are each located above the storage control gate electrodes 208 and can consist of, for example, silicon nitride, silicon oxide, one or more other suitable dielectrics, or any combination thereof.

[0053] A storage control gate spacer 440 lies over the storage floating gate electrodes 218 and comprises several control gate spacer segments, each lining side walls of the storage control gate electrodes 208, side walls of the storage control gate dielectric layers 436, and side walls of the storage control gate hard masks 438. For the sake of simplicity, only some of the control gate spacer segments are designated 440. The storage control gate spacer 440 can, for example, consist of or comprise silicon oxide, silicon nitride, one or more other suitable dielectrics, or any combination thereof.

[0054] A storage quenching gate electrode 212 is located above the common storage source / drain region 430c between the storage control gate electrodes 208. Furthermore, the storage quenching gate electrode 212 is electrically insulated from the storage floating gate electrodes 218 and the common storage source / drain region 430c by a storage quenching gate dielectric layer 442. The storage quenching gate dielectric layer 442 forms a bottom surface of the storage quenching gate electrode 212, such that the storage quenching gate dielectric layer 442 lines a bottom surface and side walls of the storage quenching gate electrode 212. The storage quenching gate electrode 212 can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductive materials.The storage erase gate dielectric layer 442 can, for example, consist of or comprise silicon oxide, one or more other suitable dielectrics, or any combination of the foregoing.

[0055] A pair of storage selection gate electrodes 210 are located on the selectively conductive storage channels 432. For the sake of simplicity, only one of the storage selection gate electrodes 210 is labelled 210. The storage selection gate electrodes 210 are each located above the selectively conductive storage channels 432 and each borders the individual storage source / drain regions 430i. Furthermore, the storage selection gate electrodes 210 are laterally separated from the common storage source / drain region 430c by the storage floating gate electrodes 218 and laterally separated from the storage floating gate electrodes 218 by a storage floating gate spacer 444.The storage floating-gate spacer 444 comprises several floating-gate spacer segments, each lining side walls of the storage control-gate spacer 440, side walls of the storage floating-gate electrodes 218, and side walls of the storage floating-gate dielectric layers 434. Furthermore, the storage selection gate electrodes 210 are vertically spaced from the selectively conductive storage channels 432 by a pair of storage selection gate dielectric layers 446. The storage selection gate electrodes 210 can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductors. The storage floating-gate spacer 444 can, for example, consist of or comprise silicon oxide, silicon nitride, one or more other suitable dielectrics, or any combination thereof.The memory selection gate dielectric layers 446 can, for example, consist of or comprise silicon oxide, one or more other suitable dielectrics, or any combination thereof.

[0056] A logic device 448 is located on a semiconductor logic area 104l of the semiconductor substrate 402 and can consist, for example, of a static random-access memory (SRAM) device, a core logic device, an input / output (I / O) device, or one or more other suitable logic devices. While the semiconductor logic area 1041 is shown as adjacent to the semiconductor memory area 104m, this is for illustrative purposes only. The semiconductor logic area 104l may, but need not, be adjacent to the semiconductor memory area 104m. Furthermore, it should be noted that, for the sake of simplicity and because the upper layout of the logic device 448 is outside the scope of this document or its application, the logic device 448 is not shown in the upper layout views 200A and 200B of the [reference missing]. Fig. 2A and Fig. 2B and the upper layout views 300A, 300B of the Fig. 3A and Fig. 3B is shown.

[0057] In some embodiments, the logic device 448 comprises a pair of logic source / drain regions 450 and a selectively conductive logic channel 452. For the sake of simplicity, only one of the logic source / drain regions 450 is designated 450. The logic source / drain regions 450 and the selectively conductive logic channel 452 are located in the semiconductor substrate 402 along an upper surface of the semiconductor substrate 402. The logic source / drain regions 450 are laterally spaced, and the selectively conductive logic channel 452 extends from one logic source / drain region 450 to another. Furthermore, the selectively conductive logic channel 452 has the opposite doping type to the logic source / drain regions 450. For example, the selectively conductive logic channel 452 can be of the p-type, whereas the logic source / drain regions 450 can be of the n-type, or vice versa.

[0058] A logic gate dielectric layer 456 and a logic gate electrode 458 are stacked on the selectively conductive logic channel 452. The logic gate dielectric layer 456 lies above the selectively conductive logic channel 452. The logic gate dielectric layer 456 can, for example, consist of or comprise silicon oxide, a high-k dielectric, one or more other suitable dielectrics, or any combination thereof. As used here, a high-k dielectric is a dielectric with a dielectric constant k that is higher than about 3.9, 5, 10, 15, or 20. The logic gate electrode 458 lies above the logic gate dielectric layer 456 and can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductors.

[0059] The BEOL interconnect structure 418 covers the memory cells 204c and the logic device 448 and comprises the underwire dielectric layer 420a and the first and second ILD layers 420b, 420c. In some embodiments, the contact etch stop layer 422 covers the memory cells 204c and the logic device 448 and is adapted to separate the underwire dielectric layer 420a from the memory cells 204c and the logic device. Furthermore, the BEOL interconnect structure 418 comprises a plurality of additional wires 462 and a plurality of additional vias 464. For the sake of simplicity, only some of the additional wires 462 are designated 462 and only some of the additional vias 464 are designated 464.

[0060] The additional vias 464 extend from an upper surface of the underwire dielectric layer 420a through the underwire dielectric layer 420a to the individual memory source / drain regions 430i, the memory select gate electrodes 210, the memory erase gate electrode 212, the logic source / drain regions 450, or any combination thereof. The additional wires 462 lie above the underwire dielectric layer 420a and are covered by the second ILD layer 420c. Furthermore, the additional wires 462 are embedded in a top surface of the first ILD layer 420b and lie above and contact the additional vias 464. In some embodiments, the additional wires 462 directly contact the additional vias 464.The additional wires 462 and the additional vias 464 may, for example, consist of or comprise copper, aluminum-copper, aluminum, tungsten, one or more other suitable conductors, or any combination of the foregoing.

[0061] In some embodiments, the silicide pads 428 are also located on upper surfaces of the memory select gate electrodes 210, an upper surface of the memory erase gate electrode 212, an upper surface of the individual memory source / drain regions 430i, upper surfaces of the logic source / drain regions 450, an upper surface of the logic gate electrode 458, or any combination thereof. For the sake of simplicity, only some of the silicide pads 428 are designated 428. In some embodiments, a logic gate hard mask (not shown) is located over the logic gate electrode 458, leaving an upper surface of the logic gate electrode 458 free of silicide.

[0062] With reference to Fig. 4D is another cross-sectional view 400D of some embodiments of the IC of the Fig. 2A and Fig. 2B and / or the IC of the Fig. 3A and Fig. 3B provided. The 400D cross-sectional view of Fig. 4D can, for example, be used along the line DD' in the Fig. 2A and Fig. 2B and / or along line DD' in the Fig. 3A and Fig. 3B has been taken.

[0063] As shown, a control gate intermediate device line 214c is integrated and continuous with a storage control gate electrode 208. Furthermore, the control gate intermediate device line 214c is adapted to a storage floating gate electrode 218, which lies beneath the storage control gate electrode 208. In some embodiments, the control gate intermediate device line 214c and the storage control gate electrode 208 together have an inverted U-shaped profile that covers and spans the storage floating gate electrode 218. It should be noted that the division along the control gate intermediate device line 214c is separate to highlight the storage control gate electrode 208. The separate division should not be designed in such a way as to restrict the composition of the control gate intermediate device line 214c and the storage control gate electrodes 208.

[0064] The control gate intermediate device line 214c and the memory control gate electrode 208 are electrically isolated from the semiconductor substrate 402 and the memory floating gate electrode 218 by a memory control gate dielectric layer 436. Furthermore, the control gate intermediate device line 214c and the memory control gate electrode 208 are covered and protected by a memory control gate hard mask 438. The memory control gate dielectric layer 436 and the memory control gate hard mask 438 are laterally elongated in a common direction and can, for example, have a row-shaped top layout or one or more other suitable top layouts.

[0065] While Fig. 4C shows a single pair of adjacent memory cells 204c, which are located, for example, along the line CC' in the Fig. 2A and Fig. 2B and / or along line CC' in the Fig. 3A and Fig. 3B has been taken, it should be noted that Fig. 4C, for example, represents each other pair of adjacent memory cells in the Fig. 2A and Fig. It can be 2B and / or 3A and 3B. While Fig. 4D shows a single memory cell 204c, which, for example, runs along the DD' line in the Fig. 2A and Fig. 2B and / or along line DD' in the Fig. 3A and Fig. 3B is taken, it should be noted that Fig. 4D, for example, represents every other memory cell in the Fig. 2A and Fig. It can be 2B and / or 3A and 3B.

[0066] During operation of the IC, each of the memory cells 204c stores a variable amount of charge in an associated floating-gate electrode of the memory floating-gate electrodes 218. If the associated floating-gate electrode stores a small amount of charge, it stores a first data state (e.g., a binary "1"). If the associated floating-gate electrode stores a large amount of charge, it stores a second data state (e.g., a binary "0").

[0067] The variable charge shields an electric field generated by an associated control gate electrode of the storage control gate electrodes 208 via an associated selectively conductive channel of the selectively conductive storage channels 432. This changes a threshold voltage of the associated control gate electrode between a low and a high value. Therefore, the variable charge can be read by biasing an associated selection gate electrode of the storage selection gate electrodes 210 with a voltage exceeding a threshold voltage of the associated selection gate electrode, by further biasing the associated control gate electrode with a voltage between the low and high values, and by measuring a resistance of the associated selectively conductive channel.Depending on whether the associated selectively conductive channel is in a high-resistance or low-resistance state, the memory cell is in the first data state or the second data state.

[0068] To change the variable charge quantity, programming and erasure operations are performed. Carriers (e.g., electrons) are added to the associated floating-gate electrode during programming operations and removed from the associated floating-gate electrode during erasure operations. The programming and erasure operations can be performed, for example, by Fowler-Nordheim tunneling (FNT), hot carrier injection, or other suitable methods for moving charge into and / or out of the associated floating-gate electrode. In some embodiments, the programming operations are performed by hot carrier injection, and the erasure operations are performed by FNT.

[0069] During programming operations, carriers tunnel through the associated floating-gate dielectric layer of the memory cell to the associated floating-gate electrode. This wears down the associated floating-gate dielectric layer over time, eventually causing the memory cell to fail. Therefore, the quality of the associated floating-gate dielectric layer is important for the lifetime of the memory cell, and testing the associated floating-gate dielectric layer is important for quality assurance.

[0070] The floating-gate test device 102 (see, for example, the Fig. 4A and Fig. 4B) can, for example, be used to measure the quality of the memory floating-gate dielectric layers 434 and thus estimate the lifetime of the memory cells 204c. In particular, the floating-gate test device 102 is configured to measure the quality of the test device floating-gate dielectric layer 406 (see, for example, the Fig. 4A and Fig. 4B). Furthermore, as will be shown below, the test fixture floating-gate dielectric layer 406 and the storage floating-gate dielectric layers 434 are formed by a common deposition process, whereby the test fixture floating-gate dielectric layer 406 has the same or substantially the same quality as the storage floating-gate dielectric layers 434. Accordingly, by measuring the quality of the test fixture floating-gate dielectric layer 406, the floating-gate test fixture 102 also measures the quality of the storage floating-gate dielectric layer 434. This can, in turn, be used to estimate the lifetime of the memory cells 204c and to ensure that the memory cells 204c meet design specifications.

[0071] In some embodiments, the floating-gate test device 102 can be used to determine the breakdown voltage of the test device's floating-gate dielectric layer 406. For example, a voltage can be progressively increased from the test device's floating-gate electrode 216 to the semiconductor memory area 104m until breakdown occurs. Since the breakdown voltage is representative of the quality of the test device's floating-gate dielectric layer 406, the breakdown voltage can then be used to estimate the lifetime of the memory cells 204c.

[0072] While the Fig. 4A - 4D with reference to the Fig. 2A, Fig. 2B, Fig. 3A and Fig. As described in section 3B, it is understood that the Fig. 4A - 4D also on the Fig. 2C - 2F and 3C - 3H are applicable. Therefore, the cross-sectional views 400A - 400D of the Fig. 4A - 4D Cross-sections for the ICs in the Fig. 2C - 2F and 3C - 3H to be and / or to represent them.

[0073] With regard to the Fig. Figures 5A and 5B to 27A and 27B are a series of cross-sectional views. Figures 500A, 500B to 2700A, 2700B provide cross-sectional views of some embodiments of a method for forming an IC with a floating-gate test device having a cell-like top layout. Figures with a suffix "A" are arranged along line AA' in the Fig. 1A, Fig. 1B, Fig. 2A, Fig. 2B, Fig. 3A, Fig. 3B or any combination of the preceding ones and / or can, for example, Fig. 4A corresponds. Figures with a suffix “B” are located along line CC’ in the Fig. 2A, Fig. 2B, Fig. 3A, Fig. 3B or any combination of the preceding ones and / or can, for example, Fig. 4C corresponds.

[0074] As shown in the cross-sectional views 500A, 500B of the Fig. 5A and Fig. As shown in Figure 5B, an isolation structure 404 is formed in an upper surface of a semiconductor substrate 402. The isolation structure 404 separates a semiconductor memory region 104m of the semiconductor substrate 402 and a semiconductor logic region 104l of the semiconductor substrate 402 and further defines an edge of the semiconductor memory region 104m and an edge of the semiconductor logic region 104l. The edge of the semiconductor memory region 104m can, for example, be as shown in Fig. 2A or Fig. 3A is shown. Note that the edge of the semiconductor logic area 104l is outside the cross-sectional views 500A, 500B of the Fig. 5A and Fig. 5B is delimited. The semiconductor substrate 402 can be, for example, a bulk silicon substrate, an SOI substrate, or another suitable semiconductor substrate. The insulation structure 404 can consist of, for example, an STI region, a DTI, or one or more other suitable insulation structures.

[0075] In some embodiments, a method for forming the insulating structure 404 comprises depositing a lower filler layer 502 covering the semiconductor substrate and further depositing an upper filler layer 504 covering the lower filler layer 502. The deposition of the lower and upper filler layers 502, 504 can be carried out, for example, by chemical vapor deposition (CVD), physical vapor deposition (PVD), thermal oxidation, one or more other suitable deposition methods, or any combination thereof. As used herein, a singular expression (e.g., a method) can mean singular or plural. The lower filler layer 502 is composed of a different material than the upper filler layer 504 and can, for example, consist of or comprise silicon oxide or one or more other suitable dielectrics.The upper filler layer 504 can, for example, consist of or comprise silicon nitride or one or more other suitable dielectrics. The lower and upper filler layers 502, 504, and the semiconductor substrate 402 are patterned to define a trench with an upper layout of the insulating structure 404, and the trench is subsequently filled with a dielectric layer. The patterning can be performed, for example, by a photolithography / etching process or one or more other suitable patterning methods. The trench can be filled, for example, by depositing the dielectric layer that fills the trench and covers the upper filler layer 504, and then planarizing the dielectric layer until the upper filler layer 504 is reached.The deposition of the dielectric layer can be carried out, for example, by CVD, PVD, or one or more other suitable deposition methods. Planarization can be achieved, for example, by chemical-mechanical polishing (CMP) or one or more other suitable planarization methods.

[0076] As used herein, a photolithography / etching process may, for example, involve depositing a photoresist layer onto a substrate and subsequently structuring the photoresist layer with a pattern. Deposition may be performed, for example, by rotoplating or one or more other suitable deposition methods. Structuring may be performed, for example, by photolithography or other suitable structuring methods. Furthermore, the photolithography / etching process may, for example, involve etching into the substrate with the structured photoresist in place to transfer the pattern onto the substrate and subsequently removing the structured photoresist layer. Removal may be performed, for example, by plasma ashing or one or more other suitable removal methods.

[0077] As shown in the cross-sectional views 600A, 600B of the Fig. 6A and Fig. As shown in Figure 6B, the upper filler layer 504 is patterned to remove the upper filler layer 504 from the semiconductor memory region 104m, but not from the semiconductor logic region 104l. The patterning forms a filler opening 602 in place of the removed portion of the upper filler layer 504 and can be performed, for example, by a photolithography / etching process or one or more other suitable patterning methods.

[0078] As also shown by the cross-sectional views 600A, 600B of the Fig. 6A and Fig. As shown in Figure 6B, the lower filler layer 502 is patterned to remove the lower filler layer 502 from the semiconductor memory region 104m, but not from the semiconductor logic region 104l. The patterning enlarges the filler opening 602 to fill the space previously occupied by the removed portion of the lower filler layer 502 and can be performed, for example, by an etching process or one or more other suitable patterning processes. The etching process can, for example, include a wet etching process or one or more other suitable etching processes and / or may, for example, use a wet etching agent comprising hydrofluoric acid (HF) or one or more other suitable chemicals. Furthermore, the etching process can, for example, be performed as part of a cleaning process or one or more other suitable processes.

[0079] As shown in the cross-sectional views 700A, 700B of the Fig. 7A and Fig. As shown in Figure 7B, a floating-gate dielectric layer 701 is formed in the semiconductor memory area 104m in the filler opening 602. The floating-gate dielectric layer 701 can, for example, consist of or comprise silicon oxide, one or more other suitable dielectrics, or any combination thereof. The floating-gate dielectric layer 701 can be formed, for example, by thermal oxidation, CVD, PVD, sputtering, one or more other suitable deposition methods, or any combination thereof.In some embodiments in which the floating-gate dielectric layer 701 is formed by thermal oxidation, the floating-gate dielectric layer 701 does not form or forms only minimally on the semiconductor logic area 104l and the insulation structure 404, since the oxide of thermal oxidation forms more readily on exposed semiconductor material in the semiconductor memory area 104m than on exposed material (e.g. silicon nitride) in the logic semiconductor area 102l and exposed material (e.g. oxide) in the insulation structure 404.

[0080] As also shown by the cross-sectional views 700A, 700B of the Fig. 7A and Fig. As shown in Figure 7B, a floating-gate electrode layer 702 is formed, which covers the insulation structure 404 and lines the filler layer opening 602 (see the Fig. 6A and Fig. 6B). The floating-gate electrode layer 702 can, for example, consist of metal, doped polysilicon, one or more other suitable conductors, or any combination thereof. Furthermore, the floating-gate electrode layer 702 can be formed, for example, by CVD, PVD, electroless plating, electroplating, one or more other suitable deposition processes, or one or more other suitable plating processes.

[0081] As shown in the cross-sectional views 800A, 800B of the Fig. 8A and Fig. As shown in Figure 8B, planarization is performed on the floating-gate electrode layer 702 until the insulation structure 404 is reached. The planarization separates the floating-gate electrode layer 702 from the semiconductor logic region 104l and structures the floating-gate electrode layer 702 with the same top layout as that of the semiconductor memory region 104m. Examples of this top layout are shown, for example, in the Fig. 2A, Fig. 2B, Fig. 3A and Fig. Figure 3B shows the semiconductor memory area of ​​104m. Planarization can be performed, for example, by a CMP or one or more other suitable planarization methods.

[0082] Although this is shown in the cross-sectional views 800A, 800B of the Fig. 8A and Fig. Since 8B is not visible, the floating-gate electrode layer 702 can, for example, be structured such that it is divided into a plurality of separate segments. See, for example, the Fig. 2B and Fig. 3B, and it is noted that the test fixture and the memory floating-gate electrodes 216, 218 are separated in a direction parallel to line DD'. It is this separation that is the subject of the structuring. The separated segments may, for example, comprise a test fixture segment belonging to a floating-gate test fixture currently being manufactured. Furthermore, the separated segments may, for example, comprise a plurality of row-shaped segments that have a common orientation and belong to rows or columns of a memory cell array currently being manufactured. The structuring may be carried out, for example, by a photolithography / etching process or one or more other suitable structuring methods.

[0083] As shown in the cross-sectional views 900A, 900B of the Fig. 9A and Fig. As shown in Figure 9B, a control gate dielectric layer 902, a control gate electrode layer 904, and a control gate hard mask layer 906 are stacked on top of the floating gate electrode layer 702, the upper filler layer 504, and the insulating structure 404. The control gate electrode layer 904 lies above the control gate dielectric layer 902, and the control gate hard mask layer 906 lies above the control gate electrode layer 904. The control gate electrode layer 904 can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductors, and / or can be formed, for example, by CVD, PVD, electroless plating, electroplating, or one or more other suitable deposition or plating processes.The control gate dielectric layer 902 and the control gate hard mask layer 906 can, for example, consist of or comprise silicon oxide, silicon nitride, one or more other suitable dielectrics, or any combination thereof. Furthermore, the control gate dielectric layer 902 and the control gate hard mask layer 906 can be formed, for example, by CVD, PVD, or one or more other suitable deposition methods.

[0084] As shown in the cross-sectional views 1000A, 1000B of the Fig. 10A and Fig. As shown in Figure 10B, the control gate dielectric layer 902 (see the Fig. 9A and Fig. 9B), the control gate electrode layer 904 (see the Fig. 9A and Fig. 9B) and the control gate hard mask layer 906 (see the Fig. 9A and Fig. 9B) on the semiconductor memory region 104m, but not on the semiconductor logic region 104l. The structuring defines a test fixture control gate dielectric layer 408, a test fixture control gate electrode 202, and a test fixture control gate hard mask 410, which are stacked on top of each other. See Fig. 10A. Furthermore, the structuring defines a pair of memory control gate dielectric layers 436, a pair of memory control gate electrodes 208, and a pair of memory control gate hard masks 438 stacked on top of each other. See Fig. 10B. The structuring can be carried out, for example, by a photolithography / etching process or one or more other suitable structuring methods.

[0085] The test fixture control gate dielectric layer 408 lies above the floating gate electrode layer 702, the test fixture control gate electrode 202 lies above the test fixture control gate dielectric layer 408, and the test fixture control gate hardmask 410 lies above the test fixture control gate electrode 202. Furthermore, the test fixture control gate dielectric layer 408, the test fixture control gate electrode 202, and the test fixture control gate hardmask 410 have cell-like top layouts. As described above, a cell-like top layout can, for example, consist of or comprise an array of islands and a plurality of bridges connecting the islands. In some embodiments, the test device control gate dielectric layer 408, the test device control gate electrode 202 and the test device control gate hard mask 410 have the same or substantially the same cell-like top layout.Examples of this cell-like top layout can be found, for example, in the . Fig. 2A and Fig. 3A is shown in relation to the test fixture control gate electrode 202. While the test fixture control gate dielectric layer 408, the test fixture control gate electrode 202, and the test fixture control gate hard mask 410 are each shown to have multiple separate segments, the separate segments can be, for example, outside of cross-sectional view 1000A of Fig. It must be connected to 10A.

[0086] The memory control gate dielectric layers 436 lie above the floating gate electrode layer 702 and are laterally spaced. The memory control gate electrodes 208 each lie above the memory control gate dielectric layers 436, and the memory control gate hard masks 438 each lie above the memory control gate electrodes 208. In some embodiments, the memory control gate dielectric layers 436, the memory control gate electrodes 208, and the memory control gate hard masks 438 have the same or substantially the same upper memory cell layout. Examples of this upper memory cell layout are, for example, in the Fig. 2A and Fig. 3A shown in relation to the storage control gate electrodes 208.

[0087] As shown in the cross-sectional views 1100A, 1100B of the Fig. 11A and Fig. As shown in Figure 11B, a test fixture control gate spacer 412 (see Fig. 11A) and a storage control gate spacer 440 (see Fig. 11B). The test device control gate spacer 412 comprises several segments on side walls of the test device control gate dielectric layer 408, side walls of the test device control gate electrode 202, and side walls of the test device control gate hard mask 410. Similarly, the storage control gate spacer 440 comprises several segments on side walls of the storage control gate dielectric layers 436, side walls of the storage control gate electrodes 208, and side walls of the storage control gate hard masks 438. The test device control gate spacer 412 and the storage control gate spacer 440 can, for example, consist of or comprise silicon oxide, silicon nitride, silicon oxynitride, one or more other suitable dielectrics, or any combination thereof.

[0088] In some embodiments, a method for forming the test device control gate spacer 412 and the memory control gate spacer 440 comprises depositing a spacer layer that has the structure of the Fig. 10A and Fig. 10B covers and lines the spacer layer, and then back-etching is performed into the spacer layer. The deposition can be performed conformally, for example, and / or by CVD, PVD, or one or more other suitable deposition methods. The back-etching removes horizontal segments of the spacer layer without removing vertical segments, and the remaining vertical segments correspond to the test fixture control gate spacer 412 and the storage control gate spacer 440.

[0089] As also shown by the cross-sectional views 11004, 1100B of the Fig. 11A and Fig. As shown in 11B, the floating-gate electrode layer 702 (see the Fig. 10A and Fig. 10B) and the floating-gate dielectric layer 701 (see the Fig. 10A and Fig. 10B) structured. The structuring defines a test fixture floating-gate electrode 216 and a test fixture floating-gate dielectric layer 406 stacked on top of each other. See Fig. 11A. Furthermore, the structuring defines a pair of storage floating-gate electrodes 218 and a pair of storage floating-gate dielectric layers 434 stacked on top of each other. See Fig. 11B. Structuring can be carried out, for example, by a photolithography / etching process or one or more other suitable structuring methods.

[0090] The test device floating-gate electrode 216 is located beneath the test device control-gate dielectric layer 408, and the test device floating-gate dielectric layer 406 is located beneath the test device floating-gate electrode 216. Furthermore, the test device floating-gate electrode 216 and the test device floating-gate dielectric layer 406 have cell-like upper layouts. In some embodiments, the test device floating-gate electrode 216 and the test device floating-gate dielectric layer 406 have the same or substantially the same cell-like upper layout. Examples of this cell-like upper layout are, for example, in the Fig. 2B and Fig. Figure 3B shows the test fixture floating-gate electrode 216. While the test fixture floating-gate electrode 216 and the test fixture floating-gate dielectric layer 406 are each shown with several separate segments, the separate segments can be seen, for example, outside of cross-sectional view 1100A. Fig. Be connected to 11A.

[0091] The storage floating-gate electrodes 218 are located beneath the storage control-gate dielectric layers 436, and the storage floating-gate dielectric layers 434 are located beneath the storage floating-gate electrodes 218. In some embodiments, the storage floating-gate electrodes 218 and the storage floating-gate dielectric layers 434 have the same or substantially the same upper memory cell layout. Examples of this upper memory cell layout are, for example, in the Fig. 2B and Fig. 3B shown in relation to the storage floating gate electrodes 218.

[0092] As can be seen from the cross-sectional views 1200A, 1200B of the Fig. 12A and Fig. As shown in Figure 12B, a test fixture floating gate spacer 414 (see Figure 12B) is used. Fig. 12A) and a storage floating gate spacer 444 (see Fig. 11B) trained. The test fixture floating-gate spacer 414 comprises several segments on side walls of the test fixture control-gate spacer 412, side walls of the test fixture floating-gate electrode 216, and side walls of the test fixture floating-gate dielectric layer 406. Similarly, the storage control-gate spacer 440 comprises several segments on side walls of the storage control-gate spacer 440, side walls of the storage floating-gate electrodes 218, and side walls of the storage floating-gate dielectric layers 434. The test fixture floating-gate spacer 414 and the storage floating-gate spacer 444 can be made of silicon oxide, silicon nitride, silicon oxynitride, one or more other suitable dielectrics, or a any combination of the foregoing or any combination thereof.

[0093] In some embodiments, a method for forming the test fixture floating-gate spacer 414 and the storage floating-gate spacer 444 comprises depositing a spacer layer that has the structure of the Fig. 11A and Fig. 11B covers and lines the spacer layer, and the subsequent back-etching into the spacer layer is performed. The deposition can be performed, for example, conformally and / or by CVD, PVD, or one or more other suitable deposition methods. The back-etching removes horizontal segments of the spacer layer without removing vertical segments, and the remaining vertical segments correspond to the test fixture floating-gate spacer 414 and the storage floating-gate spacer 444.

[0094] As also shown by the cross-sectional views 1200A, 1200B of the Fig. 12A and Fig. As shown in Figure 12B, a common memory source / drain region 430c is formed in an upper surface of the semiconductor substrate 402 between the memory floating-gate electrodes 218. See Fig. 12B. The common memory source / drain region 430c is a doped region of the semiconductor substrate 402 and may, for example, have an opposite doping type to an adjacent region of the semiconductor substrate 402. In some embodiments, a method for forming the common memory source / drain region 430c includes forming a photoresist layer that has the structure of Fig. 11A and Fig. 11B is covered. The photoresist layer is patterned using photolithography to define an opening that exposes the semiconductor substrate between the memory floating-gate electrodes 218. A doping procedure is then performed with the patterned photoresist layer in place to form the common memory source / drain region 430c through the opening. The doping procedure may, for example, consist of or include ion implantation or one or more other suitable doping methods. The photoresist layer is then removed.

[0095] As shown in the cross-sectional views 1300A, 1300B of the Fig. 13A and Fig. As shown in Figure 13B, segments of the memory floating-gate spacer 444 adjacent to the common memory source / drain area 430c are removed. This removal can be carried out, for example, by a photolithography / etching process or one or more other suitable etching processes.

[0096] As also shown by the cross-sectional views 1300A, 1300B of the Fig. 13A and Fig. As shown in Figure 13B, a memory erase gate dielectric layer 442 is formed, covering the common memory source / drain region 430c. See Figure 13B. Fig. 13B. Furthermore, the storage erase gate dielectric layer 442 lines the side walls of the storage floating gate electrodes 218 facing the common storage source / drain region 430c, the side walls of the storage floating gate dielectric layers 434 facing the common storage source / drain region 430c, and the side walls of the storage control gate spacer 440 facing the common storage source / drain region 430c. The storage erase gate dielectric layer 442 can, for example, consist of or comprise silicon oxide or one or more other suitable dielectrics.

[0097] In some embodiments, a method for forming the storage-erase-gate dielectric layer 442 comprises the deposition of a dielectric layer that has the structure of the Fig. 12A and Fig. 12B covers and lines the dielectric layer. The dielectric layer can be deposited, for example, by thermal oxidation, CVD, PVD, or one or more other suitable deposition methods. The dielectric layer is then structured into the storage-erase gate dielectric layer 442. The structuring can be carried out, for example, by a photolithography / etching process or one or more other suitable structuring methods.

[0098] As can be seen from the cross-sectional views 1400A, 1400B of the Fig. 14A and Fig. As shown in Figure 14B, a selection gate dielectric layer 1402 is formed on the semiconductor substrate 402 laterally to the test fixture floating-gate electrode 216 and laterally to the storage floating-gate electrodes 218. The selection gate dielectric layer 1402 can, for example, consist of or comprise silicon oxide or one or more other suitable dielectrics. Furthermore, the selection gate dielectric layer 1402 can be formed, for example, by thermal oxidation or one or more other suitable deposition methods. In some embodiments, the selection gate dielectric layer 1402 is formed by a deposition method that preferably forms the selection gate dielectric layer 1402 directly on the semiconductor material of the semiconductor substrate 402 relative to a surrounding dielectric.

[0099] As also shown by the cross-sectional views 1400A, 1400B of the Fig. 14A and Fig. Figure 14B shows a gate electrode layer 1404, which has the structure of Fig. 13A and Fig. 13B covers and lines the selection gate dielectric layer 1402. The gate electrode layer 1404 can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductors. The gate electrode layer 1404 can be formed, for example, by CVD, PVD, electroless plating, electroplating, or one or more other suitable deposition or plating processes.

[0100] As can be seen from the cross-sectional views 1500A, 1500B of the Fig. 15A and Fig. As shown in Figure 15B, the selection gate dielectric layer 1402 (see the Fig. 14A and Fig. 14B) and the gate electrode layer 1404 (see the Fig. 14A and Fig. 14B) structured to define test fixture selection gate dielectric layers 416 and test fixture selection gate electrodes 108. See Fig. 15A. Furthermore, the structuring defines memory selection gate dielectric layers 446, memory selection gate electrodes 210 and a memory erase gate electrode 212. See Fig. 15B. The test device selection gate electrodes 108 are each located above the test device selection gate dielectric layers 416 and abut the side walls of the floating gate spacer 414 of the test device. The memory selection gate electrodes 210 are each located above the memory selection gate dielectric layers 446 and abut the side walls of the memory floating gate spacer 444. The memory erase gate electrode 212 is located above the common memory source / drain area 430c.

[0101] In some embodiments, a method for structuring the selection gate dielectric layer 1402 and the gate electrode layer 1404 comprises performing a first etching into the gate electrode layer 1404 to back-etch the gate electrode layer 1404 and to remove horizontal segments of the gate electrode layer 1404 without removing vertical segments of the gate electrode layer 1404. The remaining vertical segments correspond to the test device selection gate electrodes 108, the storage selection gate electrodes 210, and the storage erase gate electrode 212. Furthermore, a second etching into the selection gate dielectric layer 1402 with the test device selection gate electrodes 108 and the storage selection gate electrodes 210 in place is carried out to form the test device selection gate dielectric layers 416 and the storage selection gate dielectric layers 446.The second etching can, for example, stop on the semiconductor substrate 402 and / or the test fixture selection gate electrodes 108, and the memory selection gate electrodes 210 can serve as a mask for the first etching.

[0102] As can be seen from the cross-sectional views 1600A, 1600B of the Fig. 16A and Fig. As shown in 16B, a back-etch layer 1602 is formed, which forms the structure of the Fig. 15A and Fig. 15B covers. The etch-back layer 1602 can, for example, consist of a flowable material, an organic material, an antireflection coating (ARC), one or more other suitable materials, or any combination thereof. Furthermore, the etch-back layer 1602 can be formed, for example, by rotational coating or one or more other suitable deposition processes.

[0103] In some embodiments, the back-etch layer 1602 deposits more readily on a large flat area than on a small flat area. Because the test device control gate hardmask 410 has a cell-like top layout, a top surface of the test device control gate hardmask 410 defines several small flat areas (e.g., corresponding to the islands and / or corresponding to the bridges) instead of a single large flat area. Examples of the cell-like top layout are shown with respect to the test device control gate electrode 202 in the Fig. 2A and Fig. Figure 3A shows that the multiple small flat areas can be similar in size to small flat areas defined by the top surfaces of the memory control gate hardmasks 438. Therefore, the back-etch layer 1602 is deposited uniformly or substantially uniformly on the test fixture control gate hardmask 410 and the memory control gate hardmasks 438. Furthermore, a first thickness T1 of the back-etch layer 1602 directly above the test fixture control gate hardmask 410 is equal to or substantially equal to a second thickness T2 of the back-etch layer 1602 directly above the memory control gate hardmasks 438. On the other hand, a top surface of the control gate hardmask layer 906 defines a large flat area on the semiconductor logic area 104l, which is large compared to the several small flat areas of the test fixture control gate hardmask 410 and the small flat areas of the memory control gate hardmasks 438.Accordingly, a third thickness T3 of the etching layer 1602 directly above the control gate hard mask layer 906 is significantly larger than the first and second thicknesses T1, T2.

[0104] As shown in the cross-sectional views 1700A, 1700B of the Fig. 17A and Fig. As shown in Figure 17B, etching is performed into the etch layer 1602, the test fixture control gate hardmask 410, and the memory control gate hardmasks 438. Etching is also performed into the test fixture control gate spacer 412, the test fixture floating gate spacer 414, the memory control gate spacer 440, the memory floating gate spacer 444, and the memory erase gate dielectric layer 442. The etching can be, for example, non-selective or other suitable. Non-selective etching, for instance, can be etching in which no material is preferentially or rapidly removed. The back etching process etches the back etch layer 1602 until the test fixture control gate hardmask 410 and the memory control gate hardmasks 438 are reached.Subsequently, the back-etching process etches the back-etch layer 1602, the test device control gate hardmask 410, and the memory control gate hardmasks 438 in parallel to each other to decrease a first height H1 of the test device control gate hardmask 410 and a second height H2 of the memory control gate hardmasks 438. In some embodiments, the first and second heights H1, H2 are equal or substantially equal after completion of the back-etching. Because the third thickness T3 of the back-etch layer 1602 is equal to the first and second thicknesses T1, T2 of the back-etch layer 1602 (see the figure). Fig. 16A and Fig. 16B) exceeds, the etching process need not have completely removed the etching layer 1602 from the semiconductor logic area 1041 before the etching process is completed, and need not have reduced a third height H3 of the control gate hard mask layer 906 directly on the semiconductor logic area 104l.

[0105] As shown in the cross-sectional views 1800A, 1800B of the Fig. 18A and Fig. As shown in 18B, the back-etch layer 1602 (see the Fig. 17A and Fig. 17B). Removal can be carried out, for example, by an etching process or one or more other suitable removal methods.

[0106] As also shown by the cross-sectional views 1800A, 1800B of the Fig. 18A and Fig. As shown in 18B, a hard mask layer 1802 is formed, which forms the structure of the Fig. 17A and Fig. 17B without the back-etch layer 1602 covered and having an upper surface 1802t that is planar or substantially planar. In some embodiments, a first thickness T1 of the hard mask layer 1802 directly above the test device control gate hard mask 410 is equal to or substantially equal to a second thickness T2 of the hard mask layer 1802 directly above the memory control gate hard masks 438. Furthermore, in some embodiments, a third thickness T3 of the hard mask layer 1802 directly above the semiconductor logic area 104l is substantially smaller than the first and second thicknesses T1 and T2. As can be seen, this is achieved by the increased thickness of the back-etch layer 1602 on the semiconductor logic area 104l during the back-etching of the Fig. 17A and Fig. 17B caused. The hard mask layer 1802 may, for example, consist of or comprise silicon oxide, silicon nitride, polysilicon or one or more other suitable hard mask materials.

[0107] In some embodiments, a method for forming the hard mask layer 1802 comprises depositing the hard mask layer 1802 and subsequently performing planarization on the upper surface 1802t of the hard mask layer 1802. The deposition can be carried out, for example, by CVD, PVD, or one or more other suitable deposition methods. The planarization can be carried out, for example, by CMP or one or more other suitable planarization methods.

[0108] As shown by the cross-sectional views 1900A, 1900B of the Fig. 19A and Fig. As shown in 19B, the hard mask layer 1802 (see Fig. 18A and Fig. 18B) structured to remove the hard mask layer 1802 from the semiconductor logic area 104l and to define a memory hard mask 1902 covering the semiconductor memory area 104m. The structuring can be carried out, for example, by a photolithography / etching process or one or more other suitable structuring methods.

[0109] As also shown by the cross-sectional views 1900A, 1900B of the Fig. 19A and Fig. As shown in 19B, the lower filler layer 502 (see the Fig. 18A and Fig. 18B), the upper filler layer 504 (see the Fig. 18A and Fig. 18B), the control gate dielectric layer 902 (see the Fig. 18A and Fig. 18B) the control gate electrode layer 904 (see the Fig. 18A and Fig. 18B), and the control gate hard mask layer 906 (see the Fig. 18A and Fig. 18B) from the semiconductor logic area 104l. In some embodiments, a method for performing the removal comprises etching the lower filler layer 502, the upper filler layer 504, the control gate dielectric layer 902, the control gate electrode layer 904, and the control gate hardmask layer 906 with the memory hardmask 1902 in place until the semiconductor logic area 1041 is reached. During etching, the memory hardmask 1902 acts as a mask and protects the structure on the semiconductor memory area 104m from damage. Furthermore, the memory hardmask 1902 is partially etched such that a first thickness T1 of the memory hardmask 1902 directly above the test device control gate hardmask 410 and a second thickness T2 of the memory hardmask 1902 directly above the memory control gate hardmasks 438 are reduced.

[0110] Since the test fixture control gate hardmask 410 has a cell-like top layout, the first thickness T1 of the memory hardmask 1902 is large and equal to, or substantially equal to, the second thickness T2 of the memory hardmask 1902. See the description above regarding the Fig. References 16A and 16B to 18A and 18B explain this. Thus, the memory hardmask 1902 is not completely removed from the test fixture control gate hardmask 410, and the initial thickness T1 is sufficient to protect the test fixture control gate hardmask 410 during the etching of the top filler layer 504, the control gate dielectric layer 902, the control gate electrode layer 904, and the control gate hardmask layer 906. This, in turn, can prevent an electrical short circuit between the semiconductor memory area 104m and the test fixture floating gate electrode 216 during the subsequent BEOL processing.

[0111] As shown in the cross-sectional views 2000A, 2000B of the Fig. 20A and Fig. As shown in Figure 20B, a logic gate dielectric layer 2002, a logic gate electrode layer 2004 and a logic gate hard mask layer 2006 are stacked on the semiconductor memory and logic areas 104m, 1041. The logic gate dielectric layer 2002 lies above the storage hard mask 1902 and the semiconductor logic area 1041, the logic gate electrode layer 2004 lies above the logic gate dielectric layer 2002, and the logic gate hard mask layer 2006 lies above the logic gate electrode layer 2004. The logic gate electrode layer 2004 can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductors or other suitable material, and / or can be formed, for example, by CVD, PVD, electroless plating, electroplating, or one or more other suitable deposition or plating processes.The logic gate dielectric layer 2002 can consist of or comprise silicon oxide, silicon nitride, a high-k dielectric, one or more other suitable dielectrics, or any combination thereof. The logic gate hard mask layer 2006, for example, can consist of or comprise silicon oxide, silicon nitride, one or more other suitable dielectrics, or any combination thereof. Furthermore, the logic gate dielectric layer 2002 and the logic gate hard mask layer 2006 can be formed, for example, by CVD, PVD, or one or more other suitable deposition methods.

[0112] As shown in the cross-sectional views 2100A, 2100B of the Fig. 21A and Fig. As shown in Figure 21B, the logic gate dielectric layer is 2002 (see the Fig. 20A and Fig. 20B), the logic gate electrode layer 2004 (see the Fig. 20A and Fig. 20B) and the logic gate hard mask layer 2006 (see the Fig. 20A and Fig. 20B). The structuring removes the logic gate dielectric layer 2002, the logic gate electrode layer 2004, and the logic gate hardmask layer 2006 from the semiconductor memory region 104m and defines a logic gate dielectric layer 456, a logic gate electrode 458, and a logic gate hardmask 2102 stacked on the semiconductor logic region 104l. The logic gate dielectric layer 456 lies above the semiconductor logic region 104l, the logic gate electrode 458 lies above the logic gate dielectric layer 456, and the logic gate hardmask 2102 lies above the logic gate electrode 458. The structuring can be performed, for example, by a photolithography / etching process or one or more other suitable structuring methods.

[0113] The memory hardmask 1902 protects the structure on the semiconductor memory area 104m from damage during the structuring of the logic gate dielectric layer 2002, the logic gate electrode layer 2004, and the logic gate hardmask layer 2006. Furthermore, in some embodiments, the memory hardmask 1902 is partially removed, reducing the first thickness T1 of the memory hardmask 1902 directly above the test fixture control gate hardmask 410 and the second thickness T2 of the memory hardmask 1902 directly above the memory control gate hardmasks 438. Since the test fixture control gate hardmask 410 has a cell-like top layout, the first thickness T1 of the memory hardmask 1902 is large, equal to, or substantially equal to the second thickness T2 of the memory hardmask 1902. See the description above regarding the Fig. Sections 16A and 16B through 18A and 18B provide an explanation. Thus, the memory hardmask 1902 is not completely removed from the test fixture control gate hardmask 410, and the initial thickness T1 is sufficient to protect the test fixture control gate hardmask 410 during structuring.

[0114] As shown in the cross-sectional views 2200A, 2200B of the Fig. 22A and Fig. As shown in 22B, the memory hard mask 1902 (see the Fig. 21A and Fig. 21B). Removal can be carried out by an etching process or one or more other suitable removal methods.

[0115] As also shown by the cross-sectional views 2200A, 2200B of the Fig. 22A and Fig. As shown in Figure 22B, the test device control gate hard mask 410, the test device control gate electrode 202, and the test device control gate dielectric layer 408 are structured to define a plurality of contact openings 2202 that expose the test device floating gate electrode 216. The structuring can be carried out, for example, by a photolithography / etching process or one or more other suitable structuring methods.

[0116] As shown in the cross-sectional views 2300A, 2300B of the Fig. 23A and Fig. As shown in Figure 23B, a pair of individual memory source / drain regions 430i and a pair of logic source / drain regions 450 are formed in an upper surface of the semiconductor substrate 402. See Figure 23B. Fig. 23B. The individual memory source / drain regions 430i each border the memory selection gate electrodes 210 and are separated from the common memory source / drain region 430c by the memory floating gate electrodes 218. Furthermore, the individual memory source / drain regions 430i have the same doping type as the common memory source / drain region 430c. The logic source / drain regions 450 each border opposite side walls of the logic gate electrode 458 and have a common doping type. In some embodiments, a plurality of test fixture source / drain regions 427 are also formed in an upper surface of the semiconductor substrate 402. The test device source / drain areas 427 are each located between two opposing gate electrodes of the test device selection gate electrodes 108 and have a common doping type.In further embodiments, the test device source / drain regions 427 are omitted. The individual memory source / drain regions 430i, the logic source / drain regions 450, and the test device source / drain regions 427 can be formed, for example, by ion implantation or one or more other suitable doping methods.

[0117] As shown in the cross-sectional views 2400A, 2400B of the Fig. 24A and Fig. As shown in Figure 24B, the logic gate hard mask 2102 is used (see, for example, Figure 24B). Fig. 23B) from the logic gate electrode 458. Such removal can be carried out, for example, by an etching process or one or more other suitable removal methods.

[0118] As can also be seen from the cross-sectional views 2400A, 2400B of the Fig. 24A and Fig. As shown in Figure 24B, silicide pads 428 are formed on the upper surfaces of the test device selection gate electrodes 108, the upper surfaces of the test device floating gate electrode 216 exposed through the contact openings 2202, the upper surfaces of the individual memory source / drain regions 430i, the upper surfaces of the memory selection gate electrodes 210, one upper surface of the memory erase gate electrode 212, the upper surfaces of the logic source / drain regions 450, and one upper surface of the logic gate electrode 458. In some embodiments where the test device source / drain regions 427 are present, the silicide pads 428 are also formed on the upper surfaces of the test device source / drain regions 427 between the test device selection gate electrodes 108. For the sake of simplicity, only some of the silicide pads 428 are labelled 428.The silicide pads 428 can, for example, consist of nickel silicide or one or more other suitable silicides and / or can be formed, for example, by a silicide process or one or more other suitable processes for forming silicide.

[0119] As can be seen from the cross-sectional views 2500A, 2500B of the Fig. 25A and Fig. As shown in 25B, a contact etch stop layer 422 is formed, which forms the structure of the Fig. 24A and Fig. 24B covers and adapts to some embodiments. The contact etch stop layer 422 can, for example, consist of silicon nitride, silicon oxynitride, silicon carbide, one or more other suitable dielectrics, or any combination thereof. Furthermore, the contact etch stop layer 422 can, for example, be formed by conformal deposition and / or can be formed, for example, by CVD, PVD, one or more other suitable deposition methods, or any combination thereof.

[0120] As can also be seen from the cross-sectional views 2500A, 2500B of the Fig. 25A and Fig. As shown in Figure 25B, an underwire dielectric layer 420a is formed, covering the contact etch stop layer 422 and having a top surface 420t that is planar or substantially planar. The underwire dielectric layer 420a can, for example, consist of or comprise silicon oxide, a low-k dielectric, one or more other suitable dielectrics, or any combination thereof. In some embodiments, a method for forming the underwire dielectric layer 420a comprises depositing the underwire dielectric layer 420a and subsequently performing planarization on the top surface 420t of the underwire dielectric layer 420a. The deposition can be carried out, for example, by CVD, PVD, or one or more other suitable deposition methods. Planarization can be carried out, for example, by a CMP or one or more other suitable planarization methods.

[0121] As shown in the cross-sectional views 2600A, 2600B of the Fig. 26A and Fig. As shown in Figure 26B, vias 426 are formed which extend from an upper surface of the under-wire dielectric layer 420a through the under-wire dielectric layer 420a such that they contact the test device floating-gate electrode 216. See Fig. 26A. Furthermore, additional vias 464 are formed, extending from the upper surface of the under-wire dielectric layer 420a through the under-wire dielectric layer 420a to the individual memory source / drain regions 430i, the memory selection gate electrodes 210 and the memory erase gate electrode 212, and the logic source / drain regions 450. See Fig. 26B. The vias 426 and the additional vias 464 may, for example, consist of or comprise tungsten, aluminium-copper, copper, aluminium or one or more other suitable conductors.

[0122] In some embodiments, a method for forming the vias 426 and the additional vias 464 comprises forming a photoresist layer covering the underwire dielectric layer 420a. The photoresist layer is structured with a layout of the vias 426 and the additional vias 464, and etching is performed in place into the underwire dielectric layer 420a with the structured photoresist layer to form contact openings corresponding to the vias 426 and the additional vias 464. The structuring can be carried out, for example, by photolithography or one or more other suitable structuring methods.A conductive layer is deposited, covering the underwire dielectric layer 420a and filling the contact holes, and planarization is performed on the conductive layer until the underwire dielectric layer 420a is reached. The conductive layer can be deposited, for example, by CVD, PVD, electroless plating, electroplating, or one or more other suitable deposition or plating processes. Planarization can be performed, for example, by CMP or one or more other suitable planarization processes.

[0123] As shown in the cross-sectional views 2700A, 2700B of the Fig. 27A and Fig. As shown in Figure 27B, a wire 424 is formed which overlaps and touches the vias 426. See Fig. 27A. Furthermore, several additional wires 462 are each configured to overlap and touch the additional vias 464. See Fig. 27B. For the sake of simplicity, only some of the additional wires 462 are labelled 462. The wire 424 and the additional wires 462 may, for example, consist of or comprise aluminum-copper, copper, aluminum, or one or more other suitable conductors.

[0124] In some embodiments, a method for forming the wire 424 and the additional wires 462 comprises depositing a first ILD layer 420b covering the underwire dielectric layer 420a, the vias 426, and the additional vias 464. The deposition of the first ILD layer 420b can be performed, for example, by CVD, PVD, or one or more other suitable deposition methods. The first ILD layer 420b is planarized to create a flat top surface, and the first ILD layer 420b is patterned to form a plurality of wire openings with a layout for the wire 424 and the additional wires 462. The patterning can be performed, for example, by a photolithography / etching process or one or more other suitable patterning methods.A conductive layer is formed, covering the first ILD layer 420b and filling the wire openings. Further planarization is performed on the conductive layer to form the wire 424 and the additional wires 462 within the wire openings. The conductive layer can be deposited, for example, by CVD, PVD, electroless plating, electroplating, or one or more other suitable deposition or plating processes. The planarization of the first ILD layer 420b and the conductive layer can be performed, for example, by CMP or another suitable planarization process.

[0125] As can also be seen from the cross-sectional views 2700A, 2700B of the Fig. 27A and Fig. As shown in Figure 27B, a second ILD layer 420c is formed, covering the first ILD layer 420b, the wire 424, and the additional wires 462. Furthermore, the second ILD layer 420c is formed with a top surface 42ot' that is planar or substantially planar. The second ILD layer 420c can, for example, consist of or comprise silicon oxide, a low-k dielectric, one or more other suitable dielectrics, or any combination thereof. In some embodiments, a method for forming the second ILD layer 420c comprises depositing the second ILD layer 420c and subsequently performing planarization on the top surface 420t' of the second ILD layer 420c. The deposition can be carried out, for example, by CVD, PVD, or one or more other suitable deposition methods.Planarization can be carried out, for example, by a CMP or one or more other suitable planarization methods.

[0126] With reference to Fig. 28 is a flowchart 2800 of some embodiments of the method of Fig. 5A and 5B through 27A and 27B are provided. The method can be used, for example, to remove the ICs from any of the Fig. 1A, 1B, 2A, 2B, 3A, 3B and 4A - 4D.

[0127] In the 2802, a floating-gate test fixture structure and a memory cell structure are formed on a semiconductor memory region of a semiconductor substrate. The floating-gate test fixture structure has a cell-like top layout, examples of which are shown in the Fig. 2A, Fig. 2B, Fig. 3A and Fig. Figure 3B shows this. Furthermore, a multilayer film is formed on a semiconductor logic region of the semiconductor substrate. See, for example, the Fig. 5A and 5B to 15A and 15B.

[0128] In version 2804, the heights of the floating-gate test fixture structure and the memory cell structure are reduced. See, for example, the Fig. 16A and 16B to 17A and 17B.

[0129] In the 2806, a memory hard mask is formed that covers the floating-gate test fixture structure and the memory cell structure, but not a semiconductor logic area. See, for example, the Fig. 18A and Fig. 18B and the Fig. 19A and Fig. 19B.

[0130] In the 2808, the multilayer film is removed from the semiconductor logic area by an etching process, and the memory hard mask protects the floating-gate test fixture structure and the memory cell structure during removal. See, for example, the Fig. 19A and Fig. 19B.

[0131] In the 2810, a logic device structure is formed on the semiconductor logic area, and the memory hard mask protects the floating-gate test device structure and the memory cell structure during formation. See, for example, the Fig. 20A and 20B to 21A and 21B.

[0132] At 2812, contact openings are formed that expose a floating-gate electrode of the floating-gate test fixture structure. See, for example, the Fig. 22A and Fig. 22B.

[0133] In the 2814 architecture, source / drain regions are formed along the sidewalls of the memory cell structure and the logic device structure. See, for example, the Fig. 23A and Fig. 23B.

[0134] In 2816, silicide pads are formed on the source / drain regions, gate electrodes of the memory cell structure, and gate electrodes of the floating-gate test fixture structure. See, for example, the Fig. 24A and Fig. 24B.

[0135] At 2818, a BEOL interconnect structure is formed that covers the logic device structure, the memory cell structure, and the floating-gate test device structure. See, for example, the Fig. 25A and 25B to 27A and 27B.

[0136] While the flowchart 2800 of Fig. 28. Where a sequence of operations or events is shown and described herein, it is understood that the sequence of such operations or events shown should not be interpreted in a restrictive sense. For example, some operations may occur in a different order and / or concurrently with other operations or events than those shown and / or described herein. Furthermore, not all of the operations shown need be required to implement one or more aspects or embodiments of the present description, and one or more of the operations shown herein may be performed in one or more separate steps and / or phases.

[0137] With regard to the Fig. Figures 29A and 29B to 41A and 41B are a series of cross-sectional views. Figures 2900A, 2900B to 4100A, 4100B show some alternative embodiments of the method. Fig. Figures 5A and 5B through 27A and 27B are provided, in which the procedure features a gate replacement method. Figures with a suffix of "A" are located along line AA' in the Fig. 1A, Fig. 1B, Fig. 2A, Fig. 2B, Fig. 3A, Fig. 3B or any combination of the above. Figures with a suffix of "B" are taken along line CC' in the Fig. 2A, Fig. 2B, Fig. 3A, Fig. 3B or any combination of the foregoing.

[0138] As shown in the cross-sectional views 2900A, 2900B of the Fig. 29A and Fig. As shown in Figure 29B, an upper surface section of the semiconductor substrate 402 in the semiconductor memory region 104m is recessed beneath an upper surface section of the semiconductor substrate 402 in the semiconductor logic region 104l by a distance D. In some embodiments, a method for performing the recession comprises forming a mask covering the semiconductor logic region 104l but not the semiconductor memory region 104m, and subsequently performing oxidation with the mask in place. The oxidation consumes the upper surface section of the semiconductor substrate 402 in the semiconductor memory region 104m, thereby recessing the upper surface section of the semiconductor substrate 402 in the semiconductor memory region 104m. The mask and the oxide formed by the oxidation are then removed by an etching process or one or more other suitable removal methods.The mask can, for example, consist of or include silicon nitride or one or more other materials suitable for protecting the semiconductor logic area 1041 from oxidation.

[0139] As shown in the cross-sectional views 3000A, 3000B of the Fig. 30A and Fig. As shown in 30B, the processes in the Fig. Steps 5A and 5B through 13A and 13B are performed. By performing these steps, the following features are formed on the semiconductor substrate 402. The test fixture floating-gate dielectric layer 406, the test fixture floating-gate electrode 216, the test fixture control-gate dielectric layer 408, the test fixture control-gate electrode 202, and the test fixture control-gate hard mask 410 are stacked and formed on the semiconductor memory area 104m. See Fig. 30A. The memory floating-gate dielectric layers 434, the memory floating-gate electrodes 218, the memory control-gate dielectric layers 436, the memory control-gate electrodes 208, and the memory control-gate hard masks 438 are stacked on the semiconductor memory area 104m. See Fig. 30B. The isolation structure 404, the common memory source / drain area 430c, the test fixture control gate spacer 412, the test fixture floating gate spacer 414, the memory control gate spacer 440, the memory erase gate dielectric layer 442, and the memory floating gate spacer 444 are formed on the semiconductor memory area 104m. The lower filler layer 502, the upper filler layer 504, the control gate dielectric layer 902, the control gate electrode layer 904, and the control gate hard mask layer 906 are formed on the semiconductor logic area 104l.

[0140] As also shown by the cross-sectional views 3000A, 3000B of the Fig. 30A and Fig. As shown in Figure 30B, the selection gate dielectric layer 1402 is formed on the semiconductor substrate 402 laterally by the test fixture floating gate electrode 216 and laterally by the storage floating gate electrode 218. The selection gate dielectric layer 1402 can, for example, consist of or comprise silicon oxide or one or more other suitable dielectrics. Furthermore, the selection gate dielectric layer 1402 can be formed, for example, by thermal oxidation or one or more other suitable deposition methods.

[0141] As also shown by the cross-sectional views 3000A, 3000B of the Fig. 30A and Fig. As shown in Figure 30B, a gate electrode layer 3002 and a back-etch layer 3004 are stacked over the selection gate dielectric layer 1402, the test device control gate hardmask 410, the memory control gate hardmasks 438, and the control gate hardmask layer 906. The gate electrode layer 3002 can, for example, consist of or comprise metal, doped polysilicon, or one or more other suitable conductors. The back-etch layer 3004 can, for example, consist of a flowable material, an organic material, an arc reagent (ARC), one or more other suitable materials, or any combination thereof.

[0142] In some embodiments, a method for forming the gate electrode layer 3002 comprises depositing the gate electrode layer 3002 and subsequently performing planarization on an upper surface of the gate electrode layer 3002. The gate electrode layer 3002 can be deposited, for example, by CVD, PVD, electroless plating, electroplating, or one or more other suitable deposition or plating processes. The gate electrode layer 3002 can be planarized, for example, by CMP or one or more other suitable planarization processes. In some embodiments, a method for forming the back-etch layer 3004 comprises depositing the back-etch layer 3004, for example, by rotational plating or one or more other suitable deposition processes.

[0143] As shown in the cross-sectional views 3100A, 3100B of the Fig. 31A and Fig. As shown in Figure 31B, back-etching is performed into the back-etch layer 3004 and the gate electrode layer 3002. The back-etching process etches the back-etch layer 3004 until the gate electrode layer 3002 is reached. Then, the back-etching process etches the back-etch layer 3004 and the gate electrode layer 3002 in parallel until the back-etch layer 3004 is removed. The back-etching process then continues until an upper surface of the gate electrode layer 3002 lies beneath an upper surface of the test device control gate hardmask 410 and / or upper surfaces of the memory control gate hardmasks 438. In some embodiments, the back-etching process continues until the upper surface of the gate electrode layer 3002 is approximately planar with an upper surface of the test device control gate electrode 202 and / or the upper surface of the memory control gate electrodes 208.

[0144] As shown in the cross-sectional views 3200A, 3200B of the Fig. 32A and Fig. As shown in Figure 32B, a selection gate hard mask layer 3202 is formed over the gate electrode layer 3002, which forms the structure of the Fig. 31A and Fig. 31B covers and lines. The selection gate hard mask layer 3202 can, for example, consist of or comprise silicon nitride, silicon oxynitride, silicon oxide, one or more other suitable dielectrics, or any combination thereof. The selection gate hard mask layer 3202 can be formed, for example, by CVD, PVD, electroless plating, electroplating, or one or more other suitable deposition or plating processes.

[0145] As shown in the cross-sectional views 3300A, 3300B of the Fig. 33A and Fig. Figure 33B shows the selection gate dielectric layer 1402 (see the Fig. 32A and Fig. 32B), the gate electrode layer 3002 (see the Fig. 32A and Fig. 32B) and the selection gate hard mask layer 3202 (see the Fig. 32A and Fig. 32B) structured to define test fixture selection gate dielectric layers 416, test fixture selection gate electrodes 108, and test fixture selection gate hard masks 3302. See Fig. 33A. The structuring further defines memory selection gate dielectric layers 446, memory selection gate electrodes 210, memory selection gate hardmasks 3304, a memory erase gate electrode 212 and a memory erase gate hardmask 3306. See Fig. 33B. Structuring can be carried out, for example, by a photolithography / etching process or one or more other suitable structuring methods.

[0146] The test device selection gate electrodes 108 are each located above the test device selection gate dielectric layers 416 and are adjacent to the test device floating gate spacers 414. Furthermore, the test device selection gate hard masks 3302 are each located above the test device selection gate electrodes 108. The storage selection gate electrodes 210 are each located above the storage selection gate dielectric layers 446 and are adjacent to the storage floating gate spacers 444. Furthermore, the memory selection hardmasks 3304 are each located above the memory selection gate electrodes 210. The memory erase gate electrode 212 is located above the common memory source / drain area 430c and the memory erase gate hardmask 3306 is located above the memory erase gate electrode 212.

[0147] As shown in the cross-sectional views 3400A, 3400B of the Fig. 34A and Fig. As shown in 34B, the processes in the Fig. 16A and 16B through 24A and 24B were carried out. By performing the steps in the Fig. Steps 16A and 16B through 24A and 24B flatten the upper surfaces of the various hard masks (e.g., the test fixture selection gate hard mask 3302). By performing the steps in the Fig. In embodiments 16A and 16B through 24A and 24B, the following features are formed. The contact openings 2202 are formed in the test device control gate hard mask 410, the test device control gate electrode 202, and the test device control gate dielectric layer 408. The logic gate dielectric layer 456, the logic gate electrode 458, and the logic gate hard mask 2102 are stacked on the semiconductor logic area 104l. The individual memory source / drain areas 430i, the logic source / drain areas 450, and the silicide pads 428 are formed. In some embodiments, the test device source / drain areas 427 are formed. In other embodiments, the test device source / drain areas 427 are omitted.

[0148] As can be seen from the cross-sectional views 3500A, 3500B of the Fig. 35A and Fig. As shown in Figure 35B, the various hard masks are removed. These hard masks include, for example, the logic gate hard mask 2102, the test fixture control gate hard mask 410, and the test fixture selection gate hard mask 3302. See, for example, the Fig. 34A and Fig. 34B. In some embodiments, a method for carrying out the removal includes forming a re-etch layer which restores the structure of the Fig. 34A and Fig. 34B covers and adapts to it. The back-etch layer can consist of, for example, a flowable material, an organic material, an ARC, one or more other suitable materials, or any combination thereof. Back-etching is performed into the back-etch layer and the various hard masks. The back-etching process etches the back-etch layer until the various hard masks are reached. Then, the back-etching process etches the back-etch layer and the various hard masks in parallel until the various hard masks are removed. When the various hard masks are removed, the back-etch layer is removed. Removal of the back-etch layer can be carried out, for example, by an etching process or one or more other suitable removal methods.

[0149] As can also be seen from the cross-sectional views 3500A, 3500B of the Fig. 35A and Fig. As shown in Figure 35B, the contact etch stop layer 422 is formed such that it conforms to the structure of the Fig. 34A and Fig. 34B, except for the various hard masks, covers and conforms to it. The contact etch stop layer 422 can consist, for example, of silicon nitride, silicon oxynitride, silicon carbide, one or more other suitable dielectrics, or any combination thereof. Furthermore, the contact etch stop layer 422 can be formed, for example, by conformal deposition and / or can be formed, for example, by CVD, PVD, one or more other suitable deposition methods, or any combination thereof.

[0150] As can also be seen from the cross-sectional views 3500A, 3500B of the Fig. 35A and Fig. As shown in Figure 35B, a first underwire dielectric layer 420a' is formed, covering the contact etch stop layer 422. The first underwire dielectric layer 420a' can, for example, consist of or comprise silicon oxide, a low-k dielectric, one or more other suitable dielectrics, or any combination thereof. The first underwire dielectric layer 420a' can, for example, be formed by depositing the first underwire dielectric layer 420a' by CVD, PVD, or one or more other suitable deposition methods.

[0151] As shown in the cross-sectional views 3600A, 3600B of the Fig. 36A and Fig. As shown in Figure 36B, planarization is performed on the first underwire dielectric layer 420a' and the contact etch stop layer 422 to expose the various gate electrodes. These various gate electrodes include, for example, the test fixture control gate electrode 202, the memory select gate electrodes 210, and the logic gate electrode 458. The planarization can be performed, for example, by a CMP and / or one or more other suitable planarization methods.

[0152] As shown in the cross-sectional views 3700A, 3700B of the Fig. 37A and Fig. As shown in Figure 37B, the logic gate electrode 458 (see Fig. 36B) is removed, thereby defining a logic gate opening 3702 in place of the logic gate electrode 458. The removal can be performed, for example, by a photolithography / etching process or another removal method.

[0153] As shown in the cross-sectional views 3800A, 3800B of the Fig. 38A and Fig. As shown in Figure 38B, a gate replacement layer 3802 is formed, which has the structure of Fig. 37A and Fig. 37B covers and fills the logic gate opening 3702 (see Fig. 37B). The gate replacement layer 3802 may, for example, consist of or comprise metal or other suitable conductive material and may be formed, for example, by CVD, PVD, electroless plating, electroplating, one or more other suitable deposition or plating processes, or any combination of the foregoing.

[0154] As shown in the cross-sectional views 3900A, 3900B of the Fig. 39A and Fig. Figure 39B shows a planarization on a top side of the gate replacement layer 3802 (see the Fig. 38A and Fig. 38B) is carried out until it is approximately planar with a top surface of the first under-wire dielectric layer 420a', thereby forming a logic replacement gate electrode 3902 on the semiconductor logic area 104l. The planarization can be carried out, for example, by a CMP and / or one or more other suitable planarization methods.

[0155] As also shown by the cross-sectional views 3900A, 3900B of the Fig. 39A and Fig. As shown in Figure 39B, several of the silicide pads 428 are formed on upper surfaces of the various gate electrodes. In some embodiments, the various gate electrodes include the test device selection gate electrodes 108, the memory selection and erase gate electrodes 210, 212, the backup logic gate electrode 3902, or any combination thereof. In some embodiments, the silicide pads 428 are not formed on an upper surface of the backup logic gate electrode 3902. In some embodiments, the silicide pads 428 are formed on upper surfaces of the memory control gate electrodes 208 and / or an upper surface of the test device control gate electrode 202. The silicide pads 428 may, for example, consist of nickel silicide or one or more other suitable silicides and / or may be formed, for example, by a silicide process or one or more other suitable processes for forming silicide.

[0156] As can be seen from the cross-sectional views 4000A, 4000B of the Fig. 40A and Fig. As shown in Figure 40B, a second underwire dielectric layer 420a'', covering the first underwire dielectric layer 420a', is formed with a top surface that is planar or substantially planar. The second underwire dielectric layer 420a'' can, for example, consist of or comprise silicon oxide, a low-k dielectric, one or more other suitable dielectrics, or any combination thereof. In some embodiments, a method for forming the second underwire dielectric layer 420a'' comprises depositing the second underwire dielectric layer 420a'' and subsequently performing planarization on the top surface of the second underwire dielectric layer 420a''. The deposition can be carried out, for example, by CVD, PVD, or one or more other suitable deposition methods.Planarization can be carried out, for example, by a CMP or one or more other suitable planarization methods.

[0157] As also shown by the cross-sectional views 4000A, 4000B of the Fig. 40A and Fig. As shown in Figure 40B, vias 426 are formed, extending from an upper surface of the second underwire dielectric layer 420a'' through the first and second underwire dielectric layers 420a', 420a'' such that they contact the test device floating gate electrode 216. Furthermore, additional vias 464 are formed, extending from the upper surface of the second underwire dielectric layer 420a'' through the first and second underwire dielectric layers 420a', 420a'' such that they contact the individual memory source / drain regions 430i, the memory select gate electrodes 210, the memory erase gate electrode 212, the logic source / drain regions 450, or any combination thereof.The vias 426 and the additional vias 464 may, for example, consist of or comprise tungsten, aluminum-copper, copper, aluminum or one or more other suitable conductors and / or may, for example, be as with reference to the . Fig. 26A and Fig. 26B described as being trained.

[0158] As shown in the cross-sectional views 4100A, 4100B of the Fig. 41A and Fig. As shown in Figure 41B, a wire 424 is formed that overlaps and contacts the vias 426. Furthermore, several additional wires 462 are each formed such that they overlap and contact the additional vias 464. The wire 424 and the additional wires 462 are laterally surrounded by the first ILD layer 420b and covered by the second ILD layer 420c. The wire 424, the additional wires 462, the first ILD layer 420b, and the second ILD layer 420c are formed as shown in the Fig. 27A and Fig. 27B is shown and described. The invention is defined by the main claim and the dependent claims. The subclaims describe further embodiments of the invention.

Claims

[1] A method (2800) for forming an integrated circuit (IC), the method comprising: Forming a floating gate test device structure (102) on a semiconductor substrate (402), the floating gate test device structure comprising a first floating gate electrode (216) and a first control gate electrode (202) overlying the first floating gate electrode, the first floating gate electrode and the first control gate electrode partially defining an array of islands (106i) and further partially defining a plurality of bridges (106b) connecting the islands; Forming a memory cell structure (204) on the semiconductor substrate (402), the memory cell structure comprising a second floating gate electrode (218) and a second control gate electrode (208) overlying the second floating gate electrode; and Depositing an etch-back layer (1602) covering the memory cell structure (204) and the floating gate test device structure (102) with the array of islands and the plurality of bridges, wherein the etch-back layer has a first thickness (T1) directly above the first control gate electrode (202) and a second thickness (T2) directly above the second control gate electrode (208), and wherein the first and second thicknesses are equal or substantially equal. [2] The method (2800) of claim 1, further comprising: Performing an etch into the etch-back layer (1602), the floating gate test device structure (102), and the memory cell structure (204) to uniformly or substantially uniformly reduce the respective heights of the floating gate test device structure and the memory cell structure; and Removing the etch-back layer (1602). [3] The method (2800) of claim 1 or 2, further comprising: Forming a memory hard mask (1902) covering the floating gate test device structure (102) and the memory cell structure (204), but not a semiconductor logic region (104l) of the semiconductor substrate (402); and Forming a logic device structure of the semiconductor logic region (104l), wherein forming the logic device structure partially etches the memory hard mask (1902) but not the memory cell structure (204) and the floating gate test device structure (102). [4] The method (2800) of claim 3, wherein forming the logic device structure comprises: Forming a plurality of logic device layers (2002, 2004, 2006) covering the memory hard mask (1902) and the semiconductor logic region (104l); Structuring the logic device layers (2002, 2004, 2006) to define the logic device structure on the semiconductor logic area (104l); and Removing the memory hard mask (1902). [5] The method (2800) of claim 3 or 4, wherein forming the memory hard mask (1902) comprises: Forming a hard mask layer (1802) covering the floating gate test device structure (102), the memory cell structure (204) and the semiconductor logic region (104l); performing a planarization on an upper surface (1802t) of the hard mask layer (1802) to flatten the upper surface of the hard mask layer; and Patterning the hard mask layer (1802) to remove the hard mask layer from the semiconductor logic region (104l) without removing the hard mask layer from the floating gate test device structure (102) and the memory cell structure (204). [6] The method (2800) of any preceding claim, wherein forming the floating gate test device structure (102) comprises: Forming the first floating gate electrode (216) comprising a floating gate array of floating gate islands (216i) and further comprising a plurality of floating gate bridges (216b), the floating gate bridges interconnecting the floating gate islands; and Forming the first control gate electrode (202) comprising a control gate array of control gate islands (202i) and further comprising a plurality of control gate bridges (202b), wherein the control gate islands are each located above the floating gate islands (216i), and wherein the control gate bridges connect the control gate islands. [7] The method (2800) of claim 6, further comprising: Forming an interlayer dielectric (ILD) layer (420a) covering the floating gate test device structure (102) and the memory cell structure (204); Forming a plurality of vias (426) extending through the ILD layer (420a), wherein the vias each lie over the floating gate islands (216i) and each lie over the control gate islands (202i), and wherein the vias each extend through the control gate islands and each directly touch the floating gate islands; and Forming a conductive wire (424) covering and directly contacting the vias (426). [8] Integrated circuit (IC) comprising: a semiconductor substrate (402); and a floating gate test device (102) on the semiconductor substrate (402), the floating gate test device comprising a floating gate electrode (216) and a control gate electrode (202) lying above the floating gate electrode, the floating gate electrode and the control gate electrode partially defining an array of islands (106i) and further partially defining a plurality of bridges (106b), and the bridges connecting the islands to one another, wherein the plurality of bridges (106b) comprises one bridge for each pair of adjacent islands (106i) in the array, and wherein the bridge extends from direct contact with a first island to a second island of the pair. [9] The IC of claim 8, wherein the array is limited to a single row or column. [10] The IC of claim 8, wherein the array comprises a plurality of rows and a plurality of columns. [11] The IC of any one of the preceding claims 8 to 10, wherein the islands (106i) have substantially the same upper island layout and wherein the bridges (106b) have substantially the same upper bridge layout. [12] The IC of any one of the preceding claims 8 to 11, wherein the floating gate electrode (216) comprises a floating gate array of floating gate islands (216i), the floating gate electrode further comprising a plurality of floating gate bridges (216b), the floating gate islands each partially defining the islands (106i), the floating gate bridges each partially defining the bridges (106b), and the floating gate bridges interconnecting the floating gate islands. [13] IC according to one of the preceding claims 8 to 12, further comprising: a back-end-of-line (BEOL) interconnect structure (418) covering the semiconductor substrate (402) and the floating gate test device (102), the BEOL interconnect structure comprising an interlayer dielectric (ILD) layer (420a), a plurality of vias (426), and a wire (424), the wire overlying the floating gate test device, the vias each overlying the islands (106i) and extending from direct contact with the wire through the control gate electrode (202) to direct contact with the floating gate electrode (216). [14] The IC of any one of the preceding claims 8 to 13, wherein the control gate electrode (202) comprises a control gate array of control gate islands (202i), the control gate electrode comprising a plurality of control gate bridges (202b) connecting the control gate islands, the control gate islands each partially defining the islands (106i), the control gate bridges each partially defining the bridges (106b), and the control gate bridges connecting the control gate islands. [15] IC according to one of the preceding claims 8 to 14, further comprising: a memory cell (204c) on the semiconductor substrate (402), the memory cell having a second floating gate electrode (218), a second control gate electrode (208), an erase gate electrode (212), and a select gate electrode (210), the second control gate electrode overlying the second floating gate electrode, and the erase gate electrode and the select gate electrode each adjacent to opposite sidewalls of the second control gate electrode. [16] The IC of claim 15, wherein the floating gate electrode (216) is separate from the second floating gate electrode (218), and wherein the IC further comprises: an inter-device line (214) extending continuously from direct contact with the control gate electrode (202) to direct contact with the second control gate electrode (208). [17] A method (2800) for forming an integrated circuit (IC), the method comprising: Forming a floating gate layer (702) covering a semiconductor substrate (402); Patterning the floating gate layer (702) to define a first floating gate region and a second floating gate region separate from the first floating gate region; Forming a control gate layer (904) covering the first and second floating gate regions; Structuring the control gate layer (904) to define a first control gate electrode (202) and a second control gate electrode (208), wherein the first control gate electrode lies over the first floating gate region, wherein the first control gate electrode comprises a control gate array of control gate islands (202i) and further comprises a plurality of control gate bridges (202b) connecting the control gate islands, and wherein the second control gate electrode lies over the second floating gate region and is connected to the first control gate electrode; and Performing a first etch into the first and second floating gate regions with the first and second control gate electrodes (202, 208) in place to form a first floating gate electrode (216) and a second floating gate electrode (218) below the first and second control gate electrodes, respectively. [18] The method (2800) of claim 17, further comprising: Forming a filling layer (502, 504) covering the semiconductor substrate (402); Patterning the fill layer and the semiconductor substrate (402) to define a trench that delimits a semiconductor device region of the semiconductor substrate and extends in a closed path along an edge of the semiconductor device region; Filling the trench with a dielectric layer (404); performing a second etching to remove a portion of the fill layer (502, 504) from the semiconductor device region and to form an opening (602) in place of the removed portion of the fill layer; and Forming the floating gate layer (702) covering the dielectric layer (404) and filling the opening (602). [19] The method (2800) of claim 18, wherein patterning the floating gate layer (702) comprises: Performing planarization on the floating gate layer (702) until the dielectric layer (404) is reached to confine the floating gate layer to the opening (602), wherein the floating gate layer has the same layout as the semiconductor device region after completion of the planarization; and Performing a third etch into the floating gate layer (702) to define the first and second floating gate regions.

Citation Information

Patent Citations

  • Method of fabricating flash memory device having self-aligned floating gate

    US20070026612A1

  • Test pattern structure for endurance test of a flash memory device

    US5864501A