Semiconductor structures and memory cells and methods for manufacturing the same
The semiconductor structure with bonded substrates and implanted hydrogen layers enables the creation of dual gate capacitor-less memory cells, enhancing DRAM performance by reducing refresh rates and power consumption while simplifying manufacturing.
Patent Information
- Application Number
- PCT/US2024/028074
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-05
- Filing Date
- 2024-05-06
- Publication Date
- 2024-11-14
AI Technical Summary
Dynamic random access memory (DRAM) cells face challenges in scaling due to difficulties in etching capacitors at high aspect ratios and are incompatible with logic circuit manufacturing processes, leading to increased power consumption and resource usage from constant refreshing.
The development of dual gate capacitor-less, single transistor memory cells using a semiconductor structure with a first and second substrate bonded by a conductive and dielectric layer, where the second substrate has an implanted hydrogen layer, allowing for the removal of a portion to create a bonded structure suitable for memory cell fabrication.
This approach enables higher space efficiency, reduced refresh rates, lower power consumption, and simpler manufacturing processes, addressing the limitations of traditional DRAM cells by eliminating the need for capacitors and improving manufacturing yield.
Smart Images

Figure US2024028074_14112024_PF_FP_ABST
Abstract
Description
[0001] SEMICONDUCTOR STRUCTURES AND MEMORY CELLS AND METHODS FOR MANUFACTURING THE SAME
[0002] BACKGROUND
[0003] Related Application
[0004] This application claims priority to PCT Application No. PCT / US23 / 21090, filed May. 05, 2023, entitled “SEMICONDUCTOR STRUCTURES AND MEMORY ARRAYS,” which is incorporated herein by reference in its entirety.
[0005] Field of the Invention
[0006] The present disclosure relates to semiconductor structures, memory cells, and methods for manufacturing the same. In particular, some embodiments of the present disclosure relate to dual gate capacitor-less, single transistor memory cells, related structures, and methods for making the same.
[0007] Description of Related Art
[0008] Dynamic random access memory (DRAM) is utilized in many electronic devices. These include computers, mobile devices, tablets, and consumer electronics, to name a few. DRAM is based on a one-transistor, one-capacitor (1T1C) memory cell architecture. The data is stored as a charge in the capacitor, which is designated as either “0” or “1”, and the transistor controls the access to the data. A continuous goal is increasing of the circuit density, thus increasing the amount of memory in a physical package.
[0009] However, it becomes more difficult to scale or shrink the DRAM cell at each node, for example, it’s difficult to etch the capacitors at high aspect ratios. Also, DRAM device is hard to be embedded the into logic circuits, since the manufacturing process thereof is incompatible with that of the CPU process. Meanwhile, the constant refreshing process of a normal DRAM device can consume additional power and resources. It is therefore desirable to have improved structures for memory devices and manufacturing methods thereof.
[0010] SUMMARY
[0011] According to the present disclosure, a semiconductor structure is provided. The semiconductor structure comprises a first substrate, a second substrate, a bonding layer, a first conductive layer, and a first dielectric layer. The second substrate is disposed on the first substrate. The second substrate is a single crystalline substrate. The bonding layer is disposed the first substrate and the second substrate. The first conductive layer is disposed the bonding layer and the second substrate. The first dielectric layer is disposed between the first conductive layer and the second substrate.
[0012] According to the present disclosure, a method for making a semiconductor structure is provided. The method comprises (a) providing a second structure. The second structure comprises a second substrate, a first conductive layer on the second substrate, and a first dielectric layer between the second substrate and the first conductive layer. The second substrate comprises an implanted hydrogen layer. The method comprises (b) providing a first structure. The first structure comprises a first substrate. The method comprises (c) bonding the second structure to the first structure by a bonding layer to form a bonded structure. The method comprises (d) removing a portion of the second substrate from approximately the implanted hydrogen layer.
[0013] According to the present disclosure, a method for making a semiconductor structure is provided. The method comprises (a) providing a first structure. The first structure comprises a first substrate, a first conductive layer on the first substrate, and a bonding layer between the first substrate and the first conductive layer. The method comprises (b) providing a second structure. The second structure comprises a second substrate. The second substrate comprises an implanted hydrogen layer. The method comprises (c) bonding the second structure to the first structure by a first dielectric layer to form a bonded structure. The method comprises (d) removing a portion of the second substrate from approximately the implanted hydrogen layer.
[0014] According to the present disclosure, a memory cell is provided. The memory cell comprises a source region and a drain region, a first gate, a first gate dielectric, a second gate and a second gate dielectric. The source region and the drain region are disposed in a second substrate. The source region and the drain region are separated by a channel region in the second substrate. The second substrate comprises single crystalline material. The first gate is overlapped with the channel region and is disposed between a bonding layer and the second substrate. The first gate dielectric is disposed between the first gate and the second substrate. The second gate is overlapped with the channel region. The second substrate is disposed between the first gate and the second gate. The second gate dielectric is disposed between the second gate and the second substrate.
[0015] BRIEF DESCRIPTION OF THE DRAWINGS
[0016] FIGS. 1 to 8 are schematic views to illustrate embodiments of semiconductor structures according to the present disclosure.
[0017] FIG. 9 is a flowchart of a method for fabricating a semiconductor structure according to an embodiment of the present disclosure.
[0018] FIGS. 10A to 10D are schematic diagrams illustrating intermediate stages in the manufacture of a semiconductor structure according to an embodiment of the present disclosure. FIG. 11 is a schematic diagram illustrating an intermediate stage in the manufacture of a semiconductor structure according to an embodiment of the present disclosure.
[0019] FIG. 12 is a flowchart of a method for fabricating a semiconductor structure according to an embodiment of the present disclosure.
[0020] FIGS. 13 A to 13D are schematic diagrams illustrating intermediate stages in the manufacture of a semiconductor structure according to an embodiment of the present disclosure.
[0021] FIG. 14 is a schematic diagram illustrating an intermediate stage in the manufacture of a semiconductor structure according to an embodiment of the present disclosure.
[0022] FIGS. 15A to 15D are schematic diagrams illustrating intermediate stages in the manufacture of a semiconductor structure according to an embodiment of the present disclosure.
[0023] FIG. 16 is a schematic diagram illustrating an intermediate stage in the manufacture of a semiconductor structure according to an embodiment of the present disclosure.
[0024] FIGS. 17A to 17D are schematic diagrams illustrating intermediate stages in the manufacture of a semiconductor structure according to an embodiment of the present disclosure.
[0025] FIG. 18 is a schematic diagram illustrating an intermediate stage in the manufacture of a semiconductor structure according to an embodiment of the present disclosure.
[0026] FIG. 19 is a flowchart of a method for fabricating a semiconductor structure according to an embodiment of the present disclosure.
[0027] FIGS. 20A to 20D are schematic diagrams illustrating intermediate stages in the manufacture of a semiconductor structure according to an embodiment of the present disclosure.
[0028] FIG. 21 is a schematic diagram illustrating an intermediate stage in the manufacture of a semiconductor structure according to an embodiment of the present disclosure.
[0029] FIGS. 22 A to 22B are schematic views to illustrate an embodiment of a memory cell according to the present disclosure.
[0030] FIGS. 23 A to 23B are schematic views to illustrate an embodiment of a memory cell according to the present disclosure.
[0031] FIGS. 24 to 35 are schematic views to illustrate embodiments of memory cells according to the present disclosure.
[0032] FIG. 36A is an electronic schematic diagram to illustrate a memory array according to an embodiment of the present disclosure.
[0033] FIG. 36B is a table of exemplified voltages provided to a memory cell array under various operation modes according to an embodiment of the present disclosure.
[0034] DETAILED DESCRIPTION OF EMBODIMENTS The terminology used in the description presented below is intended to be interpreted in its broadest reasonable manner, even though it is used in conjunction with a detailed description of certain specific embodiments of the technology. Certain terms may even be emphasized below; however, any terminology intended to be interpreted in any restricted manner will be specifically defined as such in this Detailed Description section. Components and achievement of a semiconductor structure or a memory device, according to the present disclosure may be illustrated in the following drawings and embodiments. However, the size and shape shown on drawings for the semiconductor structure or the memory device do not limit the features of the present disclosure.
[0035] The phrase “on” used in this application can mean directly on or indirectly on with intervening elements or layers. The spatially relative terms, such as “beneath,” “below,” “lower,” “above,” “upper” and the like, may be used herein for ease of description to describe one element or feature’s relationship to another element(s) or feature(s) as illustrated in the figures. The spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. For example, if the device in the figures is turned over, elements described as being “below” or “beneath” other elements or features would then be oriented “above” the other elements or features. Thus, the exemplary term “below” can encompass both an orientation of above and below. The apparatus may be otherwise oriented (rotated 90 degrees or at other orientations) and the spatially relative descriptors used herein may likewise be interpreted accordingly.
[0036] FIG. 1 is a schematic view to illustrate an embodiment of a semiconductor structure according to the present disclosure. As shown in FIG. 1, a semiconductor structure 100 comprises a first substrate 10, a second substrate 70, a bonding layer 20, a first conductive layer 30, and a first dielectric layer 40. The second substrate 70 is disposed on the first substrate 10. The bonding layer 20 is disposed between the first substrate 10 and the second substrate 70. The first conductive layer 30 is disposed between the bonding layer 20 and the second substrate 70. The first dielectric layer 40 is disposed between the first conductive layer 30 and the second substrate 70.
[0037] In one embodiment, each of the first substrate 10 and the second substrate 70 is a wafer with diameter of 6, 8, 12, or 18 inches. In this situation, the first substrate may be referred to as a handle wafer and the second substrate may be referred to as a device wafer. The first substrate 10 may be a single crystalline semiconductor substrate, for example, made of silicon, germanium, gallium arsenide (GaAs), or indium phosphide (InP), or a glass substrate. In some embodiments, the first substrate 10 comprises electronic devices. The electronic devices may be transistors, diodes, capacitors, or resistors. The second substrate 70 may be a single crystalline substrate. The second substrate 70 may comprises semiconductor materials including but not limited to silicon, germanium, gallium arsenide (GaAs), indium phosphide (InP), silicon carbon (SiC), or gallium nitride (GaN).
[0038] The first conductive layer 30 may comprise doped poly silicon, metal, or conductive metal compound. The first conductive layer 30 may include one or a plurality of conductive material layers. The conductive material, by way of example and not limitation, can include doped poly crystalline or amorphous silicon, germanium, silicon germanium, a metal (e.g., tungsten, titanium, tantalum, ruthenium, zirconium, cobalt, copper, aluminum, lead, platinum, tin, silver, gold), a conductive metal compound material (e.g., tantalum nitride, titanium nitride, tantalum carbide, titanium carbide, titanium aluminum carbide, tungsten silicide, tungsten nitride, ruthenium oxide, cobalt silicide, nickel silicide), carbon nanotube, conductive carbon, graphene, or any suitable combination of these materials. In some embodiments, the first conductive layer 30 is patterned.
[0039] The first dielectric layer 40 may comprise silicon dioxide or high-k (dielectric constant) dielectric materials, including but not limited to hafnium silicate, zirconium silicate, hafnium dioxide and zirconium dioxide. The thickness of the first dielectric layer 40 may vary widely depending on the technology node of semiconductor manufacturing and the function of the first dielectric layer 40. The first dielectric layer 40 has a thickness usually less than about 50 nm. With an advanced technology node such as 10 nm or less, the first dielectric layer 40 may have a thickness less than about 5 nm. In some embodiments, the first dielectric layer 40 is deposited directly on the second substrate 70. In some embodiments, the first conductive layer 30 is deposited directly on the first dielectric layer 40.
[0040] The bonding layer 20 may comprise silicon dioxide. In some embodiments, the bonding layer 20 comprises an oxide direct bonding interface 201 due to the manufacturing process of the semiconductor structure 100 as described below in greater detail.
[0041] The second substrate 70 of the semiconductor structure 100 may be used to fabricate various electronic devices including but not limited to transistors, diodes, capacitors, and resistors. Transistors may include bipolar transistors (bipolar junction transistors, BJTs), fieldeffect transistors (FETs), and insulated-gate bipolar transistors (IGBTs). In particular, the semiconductor structure 100 may be used to fabricate dual gate memory cells with a first gate (back gate) and a second gate (front gate) (not shown) because the first conductive layer 30 formed under the second substrate 70 may be fabricated as a back gate of a memory cell and the first dielectric layer 40 may be fabricated as a back gate dielectric. In addition, the first substrate 10 may be fabricated to contain various electronic devices including but not limited to transistors, diodes, capacitors, and resistors even before the second substrate 70 is used to fabricate electronic devices. Furthermore, an electronic device fabricated on the second substrate may be electrically connected to another electronic device fabricated on the first substrate.
[0042] FIG. 2 is a schematic view to illustrate an embodiment of a semiconductor structure according to the present disclosure. The semiconductor structure 100A may be substantially similar to the semiconductor structure 100 described above with respect to FIG. 1, where like reference numerals indicate like elements. In the embodiment shown in FIG. 2, the first dielectric layer 40 comprises a silicon oxide layer 42.
[0043] FIG. 3 is a schematic view to illustrate an embodiment of a semiconductor structure according to the present disclosure. The semiconductor structure 100B may be substantially similar to the semiconductor structure 100 described above with respect to FIG. 1, where like reference numerals indicate like elements. In the embodiment shown in FIG. 3, the first dielectric layer 40 comprises a first oxide layer 42, a second oxide layer 46 on the first oxide layer 42, and a first nitride layer 44 between the first oxide layer 42 and the second oxide layer 46. This embodiment provides a first nitride layer 44 to store charges. Whether the first nitride layer 44 stores certain amount of charges may be used to indicate a memory status of zero or one. As a result, the semiconductor structure 100B may be used to fabricate flash type of memory cells.
[0044] FIG. 4 is a schematic view to illustrate an embodiment of a semiconductor structure according to the present disclosure. The semiconductor structure 110 may be substantially similar to the semiconductor structure 100 described above with respect to FIG. 1, where like reference numerals indicate like elements. In the embodiment shown in FIG. 4, the semiconductor structure 110 further comprises a second conductive layer 50 and a second dielectric layer 60. The second conductive layer 50 is disposed between the first dielectric layer 40 and the second substrate 70. The second dielectric layer 60 is disposed between the second conductive layer 50 and the second substrate 70. The semiconductor structure 110 may be used to fabricate double back gate memory cells with a first gate (first back gate), a third gate (second back gate), and another one or two front gates. Since the second back gate may be used to store charges for indicating a memory status of zero or one, the semiconductor structure 110 may be used to fabricate flash type of memory cells.
[0045] The second conductive layer 50 may comprise doped poly silicon, metal, or conductive metal compound. The second conductive layer 50 may include one or a plurality of conductive material layers. The conductive material, by way of example and not limitation, can include doped poly crystalline or amorphous silicon, germanium, silicon germanium, a metal (e.g., tungsten, titanium, tantalum, ruthenium, zirconium, cobalt, copper, aluminum, lead, platinum, tin, silver, gold), a conductive metal compound material (e.g., tantalum nitride, titanium nitride, tantalum carbide, titanium carbide, titanium aluminum carbide, tungsten silicide, tungsten nitride, ruthenium oxide, cobalt silicide, nickel silicide), carbon nanotube, conductive carbon, graphene, or any suitable combination of these materials. In some embodiments, the second conductive layer 50 is patterned. The second dielectric layer 60 may comprise silicon dioxide or high-k dielectric. The thickness of the second dielectric layer 60 may vary widely depending on the technology node of semiconductor manufacturing and the function of the second dielectric layer 60. The second dielectric layer 60 has a thickness usually less than about 50 nm. With an advanced technology node such as 10 nm or less, the second dielectric layer 60 may have a thickness less than about 5 nm.
[0046] FIG. 5 is a schematic view to illustrate an embodiment of a semiconductor structure according to the present disclosure. The semiconductor structure 200 may be substantially similar to the semiconductor structure 100 described above with respect to FIG. 1, where like reference numerals indicate like elements. As shown in FIG. 5, the semiconductor structure 200 further comprises a charge storage layer 75 between the first dielectric layer 40 and the second substrate 70. The charge storage layer 75 is in direct contact with the second substrate 70. The charge storage layer 75 may comprise poly silicon, silicon nitride, or silicon oxynitride.
[0047] The charge storage layer 75 directly under the second substrate 70 may be able to store charges which can be holes (positive charges) or electrons (negative charges). The charge storage layer 75 may be less than 20 nm. In one embodiment, the charge storage layer 75 may be less than 5 nm. When the charge storage layer 75 is the polysilicon layer, the holes or electrons may be trapped and stored for a period of time between grain boundaries in the polysilicon layer. Smaller grains generally create more boundaries for trapping either holes or electrons. In general, if the charge storage layer 75 is thinner, the smaller the grains of poly silicon are. Smaller grains generate more surface areas of the boundary to trap electronic holes. In the application of memory cells, either holes or electrons may be driven towards the bottom of the second substrate 70 by applying an appropriate voltage to the memory cells, and then enter the charge storage layer 75. Such holes or electrons may be trapped and stored for a period of time in the charge storage layer 75. Whether the charge storage layer 75 stores certain amount of charges may be used to indicate a memory status of zero or one. As a result, the semiconductor structure 200 with the charge storage layer 75 may be used to fabricate capacitor-less type of memory cells. All other descriptions about semiconductor structure 100 may apply here.
[0048] FIG. 6 is a schematic view to illustrate an embodiment of a semiconductor structure according to the present disclosure. The semiconductor structure 200A may be substantially similar to the semiconductor structure 200 described above with respect to FIG. 5, where like reference numerals indicate like elements. As shown in FIG. 6, the first dielectric layer 40 comprises a silicon oxide layer 42. FIG. 7 is a schematic view to illustrate an embodiment of a semiconductor structure according to the present disclosure. The semiconductor structure 200B may be substantially similar to the semiconductor structure 200 described above with respect to FIG. 5, where like reference numerals indicate like elements. As shown in FIG. 7, the first dielectric layer 40 comprises a first oxide layer 42, a second oxide layer 46 on the first oxide layer 42, and a first nitride layer 44 between the first oxide layer 42 and the second oxide layer 46.
[0049] Both the semiconductor structures 200A & 200B may be used to fabricate capacitor-less memory cells because of the charge storage layer 75 as described above. In addition, the semiconductor structure 200B may at the same time be used to fabricate flash type of memory cells because whether the first nitride layer 44 stores certain amount of charges may be used to indicate a memory status of zero or one.
[0050] FIG. 8 is a schematic view to illustrate an embodiment of a semiconductor structure according to the present disclosure. The semiconductor structure 210 may be substantially similar to the semiconductor structure 200 described above with respect to FIG. 5, where like reference numerals indicate like elements. In the embodiment shown in FIG. 8, the semiconductor structure 210 further comprises a second conductive layer 50 and a second dielectric layer 60. The second conductive layer 50 is disposed between the first dielectric layer 40 and the second substrate 70. The second dielectric layer 60 is disposed between the second conductive layer 50 and the second substrate 70. The charge storage layer 75 is disposed between the second dielectric layer 60 and the second substrate 70.
[0051] The descriptions about the charge storage layer 75 in the semiconductor structure 200 applies here. Thus, the semiconductor structure 210 may be used to fabricate memory cells with two features — the charge storage layer 75 and the second back gate. For the first feature, because whether the charge storage layer 75 stores certain amount of charges may be used to indicate a memory status of zero or one, the semiconductor structure 210 may be used to fabricate capacitor-less type of memory cells. For the second feature, since the second back gate may be used to store charges for indicating a memory status of zero or one, the semiconductor structure 210 may be used to fabricate flash type of memory cells.
[0052] The semiconductor structures 100, 100A, 100B, 110, 200, 200A, 200B, and 210 may be fabricated by applying several different methods, including but not limited to smart-cut process. For the related process, U.S. Patent Numbers 5,374,564 and 5,882,987 are incorporated by reference at its entirety.
[0053] As shown in FIG. 9, a semiconductor structure, e.g., the semiconductor structure 100, is fabricated by the following steps. At step 310, a second structure is provided (step (a)). As shown in FIG. 10A, the second structure II comprises a second substrate 70, a first conductive layer 30 on the second substrate 70, and a first dielectric layer 40 between the second substrate 70 and the first conductive layer 30. The second substrate 70 comprises an implanted hydrogen layer 72. At step 320, a first structure I is provided (step (b)). As shown in FIG. 10A, the first structure I comprises a first substrate 10. At step 330, the second structure II is flipped and bonded to the first structure I by the bonding layer 20 to form a bonded structure III as shown in FIGS. IOC and 10D (step (c)). At step 340, a portion of the second substrate 70 is removed from approximately the depth of the hydrogen implanted layer 72 as shown in FIG. 10D (step (d)).
[0054] As shown in FIG. 10A, the second substrate 70 is provided (step (al)). The second substrate 70 may be a single crystalline substrate comprising silicon, germanium, gallium arsenide (GaAs), indium phosphide (InP), silicon carbon (SiC), or gallium nitride (GaN). The hydrogen layer 72 may be implanted into the second substrate 70 (step (a2)). In one embodiment, hydrogen ions are implanted into the second substrate 70 using a dosage of 1016to 2xl017ions / cm2at an implantation energy of 50 to 150 KeV. A larger dosage can be used with larger substrates. The hydrogen implanted layer 72 may be formed at a depth of about 4xl0'5to 8xl0'5inch (1 to 2 pm) from the top surface of the second substrate.
[0055] The first dielectric layer 40 may be formed on the second substrate 70 (step (a3)). The first dielectric layer 40 may comprise silicon dioxide or high-k dielectric as discussed above. In one embodiment, the first dielectric layer 40 may be an oxide layer formed by thermal oxidation or deposition such as CVD or PVD. The first dielectric layer 40 may have a thickness less than about 50 nm. In some embodiments, the first dielectric layer 40 has a thickness less than about 5 nm. In another embodiment, the first dielectric layer 40 may include multiple dielectric sublayers, e.g., a first oxide layer 42, a second oxide layer 46 on the first oxide layer 42, and a first nitride layer 44 between the first oxide layer 42 and the second oxide layer 46 as shown in FIG. 3. For example, the second oxide layer 46, the first nitride layer 44, and the first oxide layer 42 may be sequentially deposited on the second substrate 70,
[0056] Then, the first conductive layer 30 is formed on the first dielectric layer 40 by deposition such as various CVD or PVD (step (a4)). The first conductive layer 30 may comprise doped polysilicon, metal, or conductive metal compound as discussed above. In some embodiments, the first conductive layer 30 is patterned before the second structure II is bonded to the first structure I.
[0057] As shown in FIGS. 10B, in one embodiment, a portion 20b of the bonding layer 20, such as a silicon oxide layer, is formed on the second substrate 70, e.g., on the first conductive layer 30, before bonding. In one embodiment, a portion 20a of the bonding layer, such as a silicon oxide layer, is also formed on the first substrate 10 before bonding. As discussed above, the implanted hydrogen layer 72 is implanted inside the second substrate 70 at a certain depth before bonding. However, the implantation may be conducted before or after the formation of the first dielectric layer 40, the first conductive layer 30, or the portion 20b of the bonding layer 20 as long as the hydrogen implanted layer will not be damaged by the succeeding processes. Thus, if the formation of an oxide layer, such as the first dielectric layer 40 or the portion 20b of the bonding layer, causes high temperature, the hydrogen probably should be implanted after the formation of the oxide layer.
[0058] As shown in FIG. 10C, the second structure II is then flipped and bonded to the first structure I by the bonding layer 20 to form a bonded structure III. Specifically, the cleaned portion 20b of the bonding layer on the first conductive layer 30 of the second structure II is bonded with the cleaned portion 20a of the bonding layer on the first substrate 10 of the first structure I. Conventional cleaning techniques such as the RCA wafer cleaning procedure may be used. In sum, a portion of the bonding layer can be formed before bonding on either or both side of the first structure I and the second structure II to be bonded.
[0059] In one embodiment, the portion 20b of the bonding layer may be bonded to the portion 20a of bonding layer by a fusion bonding process, such as a hydrophilic fusion bonding process. In other words, the flat surfaces of the portions 20a and 20b can be bonded with each other without using adhesives. For example, hydroxyl groups (OH-) are formed on the surfaces to be bonded due to the presence of electric charges of atoms. Hydrogen bonds may be formed between the portion 20b of the bonding layer and the portion 20a of bonding layer, and then, annealing process(es) may be performed to form chemical bonds (e.g., Si-0 bond) between the surfaces of the portion 20b of the bonding layer and the portion 20a of bonding layer. In the present embodiment, a bonding layer 20 may be formed by bonding the portion 20a and the portion 20b. As such, the bonding layer 20 may comprise silicon dioxide and may comprises an oxide direct bonding interface 201. However, the present disclosure is not limited thereto.
[0060] As shown in FIG. 10D, a portion of the second substrate 70 may be removed from the bonded structure III at approximately the depth of the hydrogen implanted layer 72. The portion of the second substrate 70 may be removed by heating the bonded structure III at a first temperature. A first temperature is usually below 400 °C to avoid any damage to the semiconductor device in the first substrate 10 if there is any. In some embodiments, a portion of the second substrate 70 may be removed by other methods, as long as the portion of the second substrate 70 has been sufficiently weakened by previous hydrogen implantation and some subsequent annealing. For example, the bonded structure 100C can be cleaved by applying mechanical pressure to the second substrate 70 or by dipping and quenching the bonded structure 100C in liquid nitrogen. The portion of the second substrate remained on the bonded structure may be less than 3 pm depending on the semiconductor manufacturing technology nodes applied for fabrication of various electronic devices. After removal, the exposed surface SI of the second substrate 70 usually has a roughness on the order of a few hundred angstroms. Such exposed surface SI of the second substrate 70 may be polished by chemical mechanical polishing (CMP) to planarize and minimize non-uniformity of the second substrate, after removing a portion of the second substrate 70 (step (e)). Other approaches such as etching may be used for the same purpose. An etch stop layer may need to be deposited in advance when etching is used to planarize and minimize non-uniformity of the exposed surface of the second substrate.
[0061] In another embodiment as shown in FIG. 12, a semiconductor structure, e.g., the semiconductor structure 100 is fabricated by the following steps. At step 510, a first structure is provided (step (a)). As shown in FIG. 13 A, the first structure 1-1 includes a first substrate 10, a first conductive layer 30 on the first substrate 10, and a bonding layer 20 between the first substrate 10 and the first conductive layer 30. At step 520, a second structure is provided (step (b)). As shown in FIG. 13A, the second structure II-2 includes a second substrate 70. The second substrate 70 comprises an implanted hydrogen layer 72. At step 530, the second structure II- 2 is bonded to the first structure I- 1 by the first dielectric layer 40 to form a bonded structure III- 2 (step (c)). FIG. 13C illustrates the bonded structure III- 2 when the first dielectric layer 40 is an oxide layer 42. At step 540, a portion of the second substrate 70 is removed from approximately the depth of the implanted hydrogen layer 72. FIG. 13D illustrates the removal of a portion of the second substrate 70 when the first dielectric layer is an oxide layer 42. Unlike the previous semiconductor structure fabrication embodiment, the first conductive layer 30 is formed in the first structure 1-1, rather than in the second structure II-2, before the bonding.
[0062] As shown in FIG. 13A, a first substrate 10 is provided (step (al)). A bonding layer 20 is formed on the first substrate 10 (step (a2)). The bonding layer 20 may be a dielectric layer. The bonding layer 20 is not used for bonding here. In one embodiment, the bonding layer 20 is an oxide layer formed by thermal oxidation or deposition, such as CVD or PVD. The first conductive layer 30 is formed, by deposition such as various CVD or PVD, on the bonding layer 20 (step (a3)).
[0063] Here, the first dielectric layer 40 is used to bond the second structure II-2 on the first structure 1-1. The first dielectric layer 40 may comprise silicon dioxide or high-k dielectric. As described before, in one embodiment, the first dielectric layer 40 is a silicon oxide layer 42. As shown in FIG. 13B, a potion or whole of the silicon oxide layer 42 may be formed before bonding on either or both side of the first structure I- 1 and the second structure II-2 to be bonded. Below are some possible arrangements. First, a portion of the silicon oxide layer 42 is formed on the first conductive layer 30 in the first structure I before bonding. Second, a portion of the silicon oxide layer 42 is formed on the second substrate 70 in the second structure II before bonding. Third, a portion of the silicon oxide layer 42 is formed on both the first conductive layer 30 and the second substrate 70 before bonding as shown in FIG. 13B.
[0064] As shown in FIGS. 13C and 13D, the second structure II-2 is flipped and bonded to the first structure I- 1 by the first dielectric layer 40 (the silicon oxide layer 42) to form a bonded structure III-2. Then, a portion of the second substrate 70 may be removed from the bonded structure ni-2 at approximately the depth of the hydrogen implanted layer 72. These process(es) may be substantially similar to that described above with respect to FIGS. 10C and 10D.
[0065] In another embodiment, the first dielectric layer 40 includes three dielectric sublayers (ONO) — the first oxide layer 42, the first nitride layer 44, and the second oxide layer 46, and at least one of the first oxide layer 42 and the second oxide layer 46 may be used for bonding. If the first oxide layer 42 is used for bonding, as shown in FIG. 15A, the second oxide layer 46 is formed on the second substrate 70 and the first nitride layer 44 is formed on the second oxide layer 46. Then, the second structure II-4 is flipped. And the first nitride layer 44 in the second structure II-4 is bonded to the first conductive layer 30 on the first structure I- 1 by the first oxide layer 42 to form a bonded structure III-3 as shown in FIG. 15C. As described previously, when the first oxide layer 42 used to bond the second structure II-4 to the first structure I- 1 , a potion or whole of the first oxide layer 42 may be formed before bonding on either or both side of the first structure 1-1 and the second structure II-4 to be bonded. In one embodiment shown in FIG. 15B, a portion 42a of the first oxide layer 42 and a portion 42b of the first oxide layer are formed respectively on the first structure 1-1 and the second structure II-4 before bonding. Related details of the bonding process described above may apply here if applicable.
[0066] As shown in FIG. 15D, a portion of the second substrate 70 may be removed from the bonded structure III-3 at approximately the depth of the hydrogen implanted layer 72, by methods similar to that described above with respect to FIG. 10D.
[0067] When the second oxide layer 46 is used for bonding, as shown in FIG. 17 A, the first oxide layer 42 is formed on the first conductive layer 30 in the first structure 1-2 and the first nitride layer 44 is formed on the first oxide layer 42. Then, the second structure II-2 is flipped. The second substrate 70 in the second structure II- 2 is bonded on the first nitride layer 44 of the first structure 1-2 by the second oxide layer 46 as shown in FIG. 17C. Similarly, when the second oxide layer 46 is used to bond the second structure II- 2 to the first structure 1-2, a potion or whole of the second oxide layer 46 may be formed before bonding on either or both side of the first structure 1-2 and the second structure II-2 to be bonded. Below are some possible arrangements. First, a portion of the second oxide layer 46 is formed on the first nitride layer 44 in the first structure 1-2 before bonding. Second, a portion of the second oxide layer 46 is formed on the second substrate 70 in the second structure II- 2 before bonding. Third, a portion of the second oxide layer 46 is formed on both the first nitride layer 44 and the second substrate 70 before bonding as shown in FIG. 17B. Related details of the bonding process described above may apply here if applicable.
[0068] In the embodiment shown in FIGS. 15A to 15C and 17A to 17C, the hydrogen implanted layer 72 is implanted inside the second substrate 70 at a certain depth before bonding. However, depending on the fabrication method used, the implantation may be conducted before or after the formation of the second oxide layer 46, the first nitride layer 44, or the first oxide layer 42 as long as the hydrogen implanted layer will not be damaged by the succeeding processes. The implantation process and the related details described before may apply here.
[0069] As shown in FIG. 17D, a portion of the second substrate 70 may be removed from the bonded structure III- 3 at approximately the depth of the hydrogen implanted layer 72, by methods similar to that described above with respect to FIG. 10D.
[0070] Similar to the fabrication of the semiconductor structure 100, the semiconductor structure 110 may be fabricated by applying several different methods, including but not limited to smartcut process. In one embodiment as shown in FIG. 19, the semiconductor structure 110 is fabricated by the following steps. At step 910, a second structure II-6 is provided. The second structure II-6 may be substantially similar to the second structure II described above with respect to FIG. 10A, where like reference numerals indicate like elements. As shown in FIG. 20A, the second structure II-6 further comprises a second conductive layer 50 and a second dielectric layer 60. The second conductive layer 50 is disposed between the first dielectric layer 40 and the second substrate 70. The second dielectric layer 60 is disposed between the second conductive layer 50 and the second substrate 70. At step 920, a first structure I is provided. As shown in FIG. 20A, the first structure I includes a first substrate 10. At step 930, the second structure II-6 is flipped and bonded onto the first structure I by the bonding layer 20 to form a bonded structure III-4 as shown in FIG. 20C. At step 940, a portion of the second substrate 70 is removed from approximately the depth of the hydrogen implanted layer 72 as shown in FIG. 20D.
[0071] As shown in FIG. 20A, the second dielectric layer 60 is formed on the second substrate 70. The second conductive layer 50 is formed on the second dielectric layer 60. The first dielectric layer 40 is formed on the second conductive layer 50. And then the first conductive layer 30 is formed on the first dielectric layer 40. The second dielectric layer 60 may comprise silicon dioxide or high-k dielectric and may be formed by methods similar to that of first dielectric layer 40 described above. The second conductive layer 50 may comprise doped polysilicon, metal, or conductive metal compound as discussed above and may be formed by similar methods as that of the first conductive layer 30. In one embodiment as shown in FIG. 20B, a portion 20a of bonding layer 20 is formed on the first structure I and a portion 20b of bonding layer 20 is formed on the second structure II- 6 before bonding. In addition, the hydrogen implanted layer 72 is implanted inside the second substrate 70 at a certain depth before bonding. The implantation process and the related details described before may be applied here.
[0072] As shown in FIGS. 20C and 20D, the second structure II-6 is flipped and bonded to the first structure I by the bonding layer 20 to form a bonded structure III-4. Then, a portion of the second substrate 70 may be removed from the bonded structure III-4 at approximately the depth of the hydrogen implanted layer 72. These process(es) may be substantially similar to that described above with respect to FIGS. 10C and 10D. Related details of the bonding process described above may apply here if applicable.
[0073] In some other embodiments of methods for fabricating the semiconductor structure 110, either the first conductive layer, the first dielectric layer, the second conductive layer, or the second dielectric layer may be used to bond the respective second structure to the first structure. In either embodiment, the layers under the particular layer used for bonding in the semiconductor structure 110 shown in FIG. 4 are formed in the first structure on the first substrate 10 and the layers above the particular layer used for bonding are formed in the second structure on the second substrate 70. And a portion or the whole layer of the particular layer used for bonding may be formed before bonding on either or both side of the first structure and the second structure to be bonded. All other processes may be applied in these fabrication methods.
[0074] In addition, the semiconductor structure 200, the semiconductor structure 200A and 200B, and the semiconductor structure 210 may be fabricated by the processes similar to those used for the fabrication of the semiconductor structure 100, semiconductor structure 100A and 100B, and the semiconductor structure 110, which are described above with respect to FIGS. 10A to 10D, 13A to 13D, 15A to 15D, 17A to 17D, and 20A to 20D. One difference is that a charge storage layer 75 is formed directly on the second substrate 70 before bonding as shown in FIGS. 11, 14, 16, 18, and 21. The charge storage layer 75 may comprise polysilicon, silicon nitride, or silicon oxynitride as discussed above. The charge storage layer 75 may be formed by deposition such as CVD, PVD, or ALD.
[0075] In either embodiment described above, depending on actual applications and circuit design, either or both the first conductive layer 30 and the second conductive layer 50 formed on either the first structure or the second structure may be patterned before bonding the second structure to the first structure. Patterning the first conductive layer 30 and / or the second conductive layer 50 may include photolithography and etching process. In some embodiments, after removing a portion of the second substrate, the remained portion of the second substrate 70 can be used to manufacture any kind of MOSFET, such as planar FET, FinFET, FD-FinFET, PD-FinFET etc.. The above semiconductor structures may be used to fabricate various types of electronic devices, such as of transistors, diodes, capacitors, and resistors. Transistors may include bipolar transistors (bipolar junction transistors, BJTs), field-effect transistors (FETs), and insulated-gate bipolar transistors (IGBTs). The FETs include planar FETs and FinFETs. In particular, the above semiconductor structures may be used to fabricate memory cells with one or two back gates. The memory cell may be a dual gate capacitor-less, single transistor memory cell (the “DGCL memory cell”), which functions as a DRAM memory cell because, in addition to a selection gate, a DGCL memory cell comprises a charge control gate to attract and hold charges at top or bottom portion of its channel region. Thus, a DGCL memory cell may function as a DRAM memory cell without a capacitor.
[0076] FIGS. 22 A to 22B are schematic views to illustrate an embodiment of a memory cell according to the present disclosure. As shown in FIGS. 22A to 22B, a dual gate capacitor-less, single transistor memory cell 1100 is provided. The memory cell 1100 comprises a source region 71 and a drain region 72, a first gate 31 (first back gate), a first gate dielectric 41, a second gate 91 (first front gate), and a second gate dielectric 81. The source region 71 and the drain region 72 are disposed in a second substrate 70. The source region 71 and the drain region 72 are separated by a channel region 73 in the second substrate 70. The first gate 31 is overlapped with the channel region 73 and may be disposed between a bonding layer 20 and the second substrate 70. The first gate dielectric 41 is disposed between the first gate 31 and the second substrate 70. The second gate 91 is overlapped with the channel region 73. The second substrate 70 is disposed between the first gate 31 and the second gate 91. The second gate dielectric 81 is disposed between the second gate 91 and the second substrate 70.
[0077] The function of the first gate 31 and the second gate 91 are interchangeable depending on the connection and operation voltages applied to these two gates. In one embodiment, the first gate 31 functions as a charge control gate and the second gate 91 functions as a selection gate for memory cell selection. In another embodiment, the first gate 31 functions as a selection gate and the second gate 91 functions as a charge control gate. The selection gate is used for selection of a specific memory cell. The charge control gate is used to facilitate the charge retention for DRAM cell function. For example, the charge control gate may be provided with a negative voltage, such as -IV, to attract and hold positive charges, such as electric holes, at the bottom of the channel region 73.
[0078] The source region 71, the drain region 72, and the channel region 73 are formed in a second substrate 70. The second substrate 70 may comprise single crystalline material. In some embodiments, the second substrate 70 is a single crystalline substrate which comprises silicon, germanium, gallium arsenide (GaAs), indium phosphide (InP), silicon carbon (SiC), or gallium nitride (GaN).
[0079] The first gate 31 may comprise doped poly silicon, metal, or conductive metal compound. The first gate 31 may include one or a plurality of conductive material layers made of materials similar to that of the first conductive layer 30. The first gate dielectric 41 may comprise silicon dioxide or high-k dielectric similar to that of the first dielectric layer 40. For example, the memory cell 1100 may be formed from the semiconductor structure 100 or similar structures discussed above, wherein the first gate 31 is formed from the first conductive layer 30, and the first gate dielectric 41 is formed from the first dielectric layer 40. The second gate 91 may comprise doped poly silicon, metal, or conductive metal compound. The second gate 91 may include one or a plurality of conductive material layers made of materials similar to that of the first conductive layer 30. The second gate dielectric 81 may comprise silicon dioxide or high-k dielectric similar to that of the first dielectric layer 40. The second gate 91 and the second gate dielectric 81 may be formed on the second substrate 70 (for example, on the exposed surface SI) by any suitable method.
[0080] As shown in FIGS. 22A and 22B, the memory cell 1100 may further comprise a bonding layer 20. The bonding layer 20 comprises silicon dioxide. The first gate 31 is disposed between the bonding layer 20 and the second substrate 70. In some embodiments, if the memory cell is made from a semiconductor structure similar to that described above with respect to FIGS. 10A to 10D, the bonding layer 20 may comprise an oxide bonding interface 201. However, the present disclosure is not limited thereto.
[0081] As shown in FIGS. 22A and 22B, the bonding layer 20 is disposed between the first gate 31 and a first substrate 10. In some embodiments, the memory cell 1100 is electrically connected to at least one of electronic devices, such as transistors, diodes, capacitors, or resistors, in a first substrate 10.
[0082] FIGS. 23 A to 23B are schematic views to illustrate an embodiment of a memory cell according to the present disclosure. As shown in FIGS. 23A to 23B, a dual gate capacitor-less, single transistor memory cell 1110 is provided. The memory cell 1110 may be substantially similar to the memory cell 1100 described above with respect to FIGS. 22A to 22B, where like reference numerals indicate like elements. In the embodiment shown in FIGS. 23 A to 23B, the transistor is a FinFET transistor here. The second gate 91 and the second gate dielectric 81 surround the channel region 73 with two or more sides. Specifically, the second gate 91 (FinFET gate) and the second gate dielectric 81 surround three sides of the channel region 73. The bottom of the second gate 91 may be the same as or higher than the bottom of the channel region 73. In this embodiment, the first gate 31 functions as a charge control gate and the second gate 91 functions as a selection gate for memory cell selection. The charge control gate may be provided with a voltage to attract and hold charges at the bottom of the channel region 73. For example, the charge control gate is provided with a negative voltage, such as -IV, to attract and hold holes at the bottom of the channel region 73.
[0083] In the embodiment shown in FIGS. 23A to 23B, the bottom of the second gate 91 is higher than the bottom of the channel region 73 to provide a space at the bottom of the channel region 73 for charge storage. The difference D between the bottom of the second gate 91 and the bottom of the channel region 73 may be approximately 1 / 3 to 1 / 8 of the channel height H. In one embodiment, the difference D is about 1 / 5 of the channel height H.
[0084] Such a dual gate capacitor-less, single transistor memory cell 1100, 1110, either a planner transistor or FinFET, has several advantages, compared to a traditional DRAM cell, including but not limited to high space efficiency, slower refresh rate, power saving, and lower manufacturing cost. First, the DGCL memory cell may only occupy 6F2area. Second, the dual gate feature may substantially reduce the refresh rate because the charge control gate (either the first gate 31 or the second gate 91) is able to retain charges at the top or bottom of the channel region 73 for a longer period of time by applying an appropriate potential to the charge control gate. In one embodiment, the charge control gate is applied with -2 volts for writing stage and - 1 volt for read and hold stage to attract and retain electronic holes (positive charges). For example, a conventional DRAM has a refresh rate of 64 ms while the DGCL memory cell may have a refresh rate of about 1 second assuming that all other conditions are the same or similar. Due to substantially lower refresh rate, the power consumption is reduced, which is particular important for memories in the mobile devices. Third, the capacitor part of a DRAM cell demands complicated manufacturing processes that increase the manufacturing costs and lower the yield rates. Without a capacitor part, a DGCL memory cell may be manufactured by much simpler process to lower the manufacturing costs and increase the yield rate. One additional significant advantage of the DGCL memory cell described in this disclosure is to avoid leakage and cross talk effects between the first gate 31 of adjacent memory cells and between the first gate 31 and the layer directly under the first gate 31 because that the first gate 31 (first back gate) is formed by conductive materials, such as metal, on a dielectric layer, such as a bonding layer comprising oxide.
[0085] DGCL memory cell(s) provided herein may have dual memory functions, e.g., a nonvolatile memory cell function (such as NAND or NOR) in addition to its original DRAM memory cell function, if the DGCL memory cell comprises either an ONO structure (oxide- nitride-oxide structure) or a floating gate to store charges. FIG. 24 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1200 shown in FIG. 24 may be substantially similar to the memory cell 1100 described above with respect to FIGS. 22A to 22B, where like reference numerals indicate like elements. In the present embodiment, the first gate dielectric 41 comprises a first oxide layer 42, a second oxide layer 46 on the first oxide layer 42, and a first nitride layer 44 between the first oxide layer 42 and the second oxide layer 46. In the present embodiment shown in FIG. 24, the ONO structure below the second substrate 70 may enable charge storage in the nitride layer. Such dual function type of DGCL may function as either a DRAM memory cell or a non-volatile memory cell (such as NAND or NOR) at a specific time, depending on how the first gate 31 (first back gate), the second gate 91 (first front gate), the source region 71, and the drain region 72 are connected and operated.
[0086] The first function of the dual function type of DGCL memory cell is a DRAM cell function as described above. The second function of the dual function type of DGCL is a nonvolatile memory cell function. When the dual function memory cell 1200 is operated as a nonvolatile memory cell, the first gate 31 may be applied with an appropriate potential to cause charges to punch through the second oxide layer 46, arrive at the first nitride layer 44 and remain in the first nitride layer 44. In one embodiment, with appropriate connection and operation, on one hand, the function of a dual function type DGCL memory cell may be switched from a NAND memory cell to a DGCL memory cell when an electronic device, such as a computer or a mobile phone, is turned on. On the other hand, the function of a dual function type DGCL memory cell may be switched from a DGCL memory cell to a NAND memory cell when an electronic device, such as a computer or a mobile phone, is turned off. The description in this paragraph may be applied to all other dual function types of DGCL memory cell if there is no conflicting description in the paragraphs below for a specific type of memory cell with dual function.
[0087] FIG. 25 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1210 shown in FIG. 25 may be substantially similar to the memory cell 1100 described above with respect to FIGS. 22A to 22B, where like reference numerals indicate like elements. In the present embodiment, the second gate dielectric 81 comprises a first oxide layer 82, a second oxide layer 86 on the first oxide layer 82, and a first nitride layer 84 between the first oxide layer 82 and the second oxide layer 86. This embodiment provides the first nitride layer 84 to store charges. In the embodiment shown in FIG. 25, the memory cell 1210 may be a dual function type of DGCL memory cell which may function as either a DRAM memory cell or a non-volatile memory cell at a specific time, depending on how the first gate 31 (first back gate), the second gate 91 (first front gate), the source region 71, and the drain region 72 are connected and operated. The advantages of the dual function type of DGCL memory cell 1200 are also applied to the second dual function type of DGCL memory cell 1210.
[0088] FIG. 26 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1220 shown in FIG. 26 may be substantially similar to the memory cell 1200 described above with respect to FIG. 24, where like reference numerals indicate like elements. In the present embodiment, the second gate dielectric 81 comprises a first oxide layer 82, a second oxide layer 86 on the first oxide layer 82, and a first nitride layer 84 between the first oxide layer 82 and the second oxide layer 86. In the embodiment shown in FIG. 26, the dual function type of DGCL memory cell 1220 may have two ONO structures for storing charges in the nitride layers. The first ONO structure includes layers 42, 44, and 46 disposed under the channel region 73; the second ONO structure includes layers 82, 84, and 86 disposed above the channel region 73. In one embodiment, when the dual function type of DGCL memory cell 1220 functions as a non-volatile memory cell, it may be used to represent four statuses, like two bits information (00, 01, 10, 11), due to the two ONO structures for charge storage.
[0089] FIG. 27 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1230 shown in FIG. 27 may be substantially similar to the memory cell 1100 described above with respect to FIGS. 22A to 22B, where like reference numerals indicate like elements. In the present embodiment, the memory cell 1230 further comprises a third gate 51 and a third gate dielectric 61. The third gate 51 is disposed between the first gate 31 and the second substrate 70. The third gate dielectric 61 is disposed between the third gate 51 and the second substrate 70. The third gate 51 may comprises doped polysilicon, metal, or conductive metal compound. The third gate 51 may include one or a plurality of conductive material layers made of materials similar to that of the second conductive layer 50. The third gate dielectric 61 may comprises silicon dioxide or high-k dielectric. For example, the memory cell 1230 may be formed from the semiconductor structure 110 or similar structures discussed above, wherein the third gate 51 is formed from the second conductive layer 50, and the third gate dielectric 61 is formed from the second dielectric layer 60. In the embodiment shown in FIG. 27, the dual function type of DGCL memory cell 1230 may have a floating gate (the third gate 51) between the channel region 73 and the first gate 31, rather than an ONO structure, to store charges for the non-volatile memory cell function. As discussed above, the function of the first gate 31 and the second gate 91 are interchangeable depending on the operation voltages applied to these two gates. In one embodiment, the first gate 31 functions as a charge control gate and the second gate 91 functions as a selection gate. In another embodiment, the first gate 31 functions as a selection gate and the second gate 91 functions as a charge control gate. The second gate 51 functions as a floating gate to store charges for the non-volatile memory cell function.
[0090] FIG. 28 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1270 shown in FIG. 28 may be substantially similar to the memory cell 1230 described above with respect to FIG. 27, where like reference numerals indicate like elements. In the present embodiment, the second gate dielectric 81 comprises a first oxide layer 82, a second oxide layer 86 on the first oxide layer 82, and a first nitride layer 84 between the first oxide layer 82 and the second oxide layer 86. In the embodiment shown in FIG. 27, the dual function type of DGCL memory cell 1270 may have a floating gate (the third gate 51) disposed between the first gate 31 and the channel region 73 and an ONO structure disposed between the second gate 91 and the channel region 73, to store charges for the non-volatile memory cell function. As discussed above, the function of the first gate 31 and the second gate 91 are interchangeable depending on the operation voltages applied to these two gates. In one embodiment, the first gate 31 functions as a charge control gate and the second gate 91 functions as a selection gate. In another embodiment, the first gate 31 functions as a selection gate and the second gate 91 functions as a charge control gate. The third gate 51 functions as a floating gate to store charges for the non-volatile memory cell function.
[0091] FIG. 29 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1260 shown in FIG. 29 may be substantially similar to the memory cell 1200 described above with respect to FIG. 24, where like reference numerals indicate like elements. In the present embodiment, the memory cell 1260 further comprises a third gate 161 and a third gate dielectric 151. The third gate 161 is disposed between the second gate 91 and the second substrate 70. The third gate dielectric 151 is disposed between the third gate 161 and the second substrate 70. The third gate 161 may comprises doped poly silicon, metal, or conductive metal compound. The third gate 161 may include one or a plurality of conductive material layers. The third gate dielectric 151 may comprises silicon dioxide or high-k dielectric. The third gate 161 and the third gate dielectric 151 may be formed by any suitable method(s). In the embodiment shown in FIG. 29, the dual function type of DGCL memory cell 1260 has an ONO structure disposed between the first gate 31 and the channel region 73 and a floating gate (the third gate 161) between the channel region 73 and the second gate 91 to store charges for the non-volatile memory cell function. The function of the first gate 31 and the second gate 91 are interchangeable depending on the operation voltages applied to these two gates. In one embodiment, the first gate 31 functions as a charge control gate and the second gate 91 functions as a selection gate. In another embodiment, the first gate 31 functions as a selection gate and the second gate 91 functions as a charge control gate. The third gate 161 functions as a floating gate to store charges for the non-volatile memory cell function.
[0092] FIG. 30 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1300 shown in FIG. 30 may be substantially similar to the memory cell 1100 described above with respect to FIGS. 22A to 22B, where like reference numerals indicate like elements. In the present embodiment, the memory cell 1300 further comprises a charge storage region 76 in contact with the second substrate 70. The charge storage region 76 may be disposed between the first gate dielectric 41 and the second substrate 70. The charge storage region 76 may comprise polysilicon, silicon nitride, or silicon oxynitride. For example, the memory cell 1300 may be formed from the semiconductor structure 200 or similar structures discussed above, wherein the charge storage region 76 is formed from the charge storage layer 75. In the embodiment shown in FIG. 30, the enhanced type of DGCL memory cell 1300 may have an additional charge storage region 76 directly under the second substrate 70. The charge storage region 76 provides additional space and mechanism to retain more charges. As a result, the refresh rate and the power consumption may be further reduced. In one embodiment, the charge storage region 76 has a thickness of about 0.2 nm to about 5 nm. In a writing phase, when an appropriate voltage, such as -2 volts, is applied to the first gate 31 (first back gate), electronic holes are attracted to the bottom of the channel region 73 and further enter the charge storage region 76. Some electronic holes may be retained between boundaries of polysilicon grains. Thus, the charge storage region 76 may need to be in contact with the channel region 73 without any other layer therebetween. Other suitable materials may also be used for the charge storage region 76 to store charges such as electrons and holes. In this embodiment, the first gate 31 functions as a charge control gate and the second gate 91 functions as a selection gate. The cross-section area of charge storage region 76 may be larger than the channel region 73 to cover all or part of the source region 71 and drain region 72.
[0093] FIG. 31 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1310 shown in FIG. 31 may be substantially similar to the memory cell 1300 described above with respect to FIG. 30, where like reference numerals indicate like elements. In the present embodiment, the charge storage region 76 is disposed between the second gate dielectric 81 and the second substrate 70. In the embodiment shown in FIG. 30, the enhanced type of DGCL memory cell 1310 may have an charge storage region 76 directly above the second substrate 70, rather than directly under the second substrate 70. The charge storage region 76 in the enhanced type DGCL memory cell 1310 may function the same as the charge storage region 76 in the enhanced type DGCL memory cell 1300. The description about the charge storage region 76 above with respect to FIG, 30 may apply here. For example, in a writing phase, when an appropriate voltage, such as -2 volts, is applied to the second gate 91, electronic holes are attracted to the top of the channel region 73 and further enter the charge storage region 76. Some electronic holes may be retained between boundaries of polysilicon grains. In this embodiment, the first gate 31 functions as a selection gate and the second gate 91 functions as a charge control gate. The cross-section area of charge storage region 76 may be larger than the channel region 73 to cover all or part of the source region 71 and drain region 72.
[0094] Each of the dual function types of DGCL memory cell discussed above may further include a charge storage region 76 directly under or directly above the second substrate 70, which enhances the charge storage for DRAM cell function. The structures and the manufacture of these dual function enhanced type of DGCL memory cells are substantially similar to the memory cells discussed above with respect to FIGS. 22A to 31. The dual function enhanced types of DGCL memory cells may carry the advantages of both the dual function type memory cells and enhanced type of memory cells described above. With appropriate connections and operation voltages applied to the source region 71, the drain region 72, and various gates, a dual function enhanced type of DGCL memory cell may function as either a non-volatile memory cell or a DRAM cell at a specific time.
[0095] FIG. 32 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1400 shown in FIG. 32 may be substantially similar to the memory cell 1200 described above with respect to FIG. 24, where like reference numerals indicate like elements. In the present embodiment, the memory cell 1400 further comprises a charge storage region 76 in contact with the second substrate 70. The charge storage region 76 may be disposed between the first gate dielectric 41 (e.g., the second oxide layer 46) and the second substrate 70. In the embodiment shown in FIG. 32, the dual function enhanced type of DGCL memory cell 1400 includes both an ONO structure under the second substrate 70 to function as a non-volatile memory cell and an additional charge storage region 76 directly under the second substrate 70 to retain more charges when it functions as a DRAM cell.
[0096] FIG. 33 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1405 shown in FIG. 33 may be substantially similar to the memory cell 1400 described above with respect to FIG. 32, where like reference numerals indicate like elements. In the present embodiment, the charge storage region 76 is disposed between the second gate dielectric 81 and the second substrate 70. In the embodiment shown in FIG. 33, the dual function enhanced type of DGCL memory cell 1405 includes both an ONO structure under the second substrate 70 to function as a non-volatile memory cell and an additional charge storage region 76 directly above the second substrate 70 to retain more charges when it functions as a DRAM cell. FIG. 34 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1430 shown in FIG. 34 may be substantially similar to the memory cell 1230 described above with respect to FIG. 27, where like reference numerals indicate like elements. In the present embodiment, the memory cell 1430 further comprises a charge storage region 76 in contact with the second substrate 70. The charge storage region 76 is disposed between the first gate dielectric 41 and the second substrate 70. Specifically, the charge storage region 76 may be disposed between the third gate dielectric 61 and the second substrate 70. In the embodiment shown in FIG. 34, the dual function enhanced type of DGCL memory cell 1430 which includes both a floating gate (the third gate 51) between the channel region 73 and the first gate 31 to function as a non-volatile memory cell and an additional charge storage region 76 directly under the second substrate 70 to retain more charges when it functions as a DRAM cell.
[0097] FIG. 35 is a schematic view to illustrate an embodiment of memory cell according to the present disclosure. The memory cell 1435 shown in FIG. 35 may be substantially similar to the memory cell 1430 described above with respect to FIG. 34, where like reference numerals indicate like elements. In the present embodiment, the charge storage region 76 is disposed between the second gate dielectric 81 and the second substrate 70. In the embodiment shown in FIG. 35, the dual function enhanced type of DGCL memory cell 1435 which includes both a floating gate (the third gate 51) between the channel region 73 and the first gate 31 to function as a non-volatile memory cell and an additional charge storage region 76 directly above the second substrate 70 to retain more charges when it functions as a DRAM cell.
[0098] Each of the above described DGCL memory cells may be used to generate a memory array. One or more types of the DGCL memory cells may be built on different parts or blocks of the same second substrate 70. Different connections and operations may be applied to different parts or blocks of the same type of DGCL memory cells. For example, in a first block of the second substrate 70, first dual function type of DGCL memory cells may be connected to function as a non-volatile memory while in a second block of the second substrate 70, first dual function type of DGCL memory cells may be connected to function as DRAM.
[0099] FIG. 36A is an electronic schematic diagram to illustrate a memory array according to an embodiment of the present disclosure. As shown in FIG. 36A, a plurality of memory cells (e.g., memory cell 1100, 1110, or other similar memory cells discussed above) are connected to form a memory array Ml to function as DRAM. The second gates (the front gate, e.g., the second gate 91) of the memory cells are electrically connected to respective word lines WL1, WL2, WL3, and WL4. The drain regions (e.g., the drain region 72) of the memory cells are electrically connected to respective bit lines BL1, BL2, BL3, and BL4. The first gates (e.g., the back gate, e.g., the first gate 31) of the memory cells are electrically connected to respective back gate lines BG1, BG2, BG3, and BG4). The source regions (e.g., the source region 71) of the memory cells are electrically connected to a Vss, which is grounded GND in this embodiment. Suitable arrangement of voltages may be supplied to the word lines WL1, WL2, WL3, and WL4, bit lines BL1, BL2, BL3, and BL4, and back gate lines BG1, BG2, BG3, and BG4 respectively in write “1,” write “0,” and read operations depending on the technology nodes used to make the DGCL memory cells.
[0100] FIG. 36B is a table of exemplified voltages provided to a memory cell array (e.g., the memory array Ml shown in FIG. 36A) under various operation modes according to an embodiment of the present disclosure. FIG. 36B illustrates exemplified voltages that may be provided to word lines, bit lines, and back gate lines respectively in program, erase, and read operations for DGCL memory cells, e.g., made by 14 nm technology node. During the write “1” operation of a selected cell, the bit line is provided with approximately 2V, the word line is provided with approximately 1.5V, and the back gate line is provided with approximately -2V. During the write “0” operation of a selected cell, the bit line is provided with approximately IV, the word line is provided with approximately 1.5V, and the back gate line is provided with approximately 0V. During a read operation of a selected cell, the bit line is provided with approximately 0.7V, the word line is provided with approximately 1.0V, and the back gate line is provided with approximately -IV. During the write “1,” write “0,” and read operation of an unselected memory cell, the bit line is provided with approximately 0V, the word line is provided with approximately 0V, and the back gate line is provided with approximately -IV. During a hold state of a memory cell (either a selected cell or an unselected cell), the bit line is provided with 0V, the word line is provided with 0V, and the back gate line is provided with - IV.
[0101] The foregoing description of embodiments is provided to enable any person skilled in the art to make and use the subject matter. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the novel principles and subject matter disclosed herein may be applied to other embodiments without the use of the innovative faculty. The claimed subject matter set forth in the claims is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein. It is contemplated that additional embodiments are within the spirit and true scope of the disclosed subject matter. Thus, it is intended that the present disclosure covers modifications and variations that come within the scope of the appended claims and their equivalents.
Claims
WHAT IS CLAIMED IS:
1. A semiconductor structure, comprising: a first substrate; a second substrate on the first substrate, the second substrate being a single crystalline substrate; a bonding layer between the first substrate and the second substrate; a first conductive layer between the bonding layer and the second substrate; and a first dielectric layer between the first conductive layer and the second substrate.
2. The semiconductor structure of claim 1, wherein the second substrate comprises silicon, germanium, gallium arsenide (GaAs), indium phosphide (InP), silicon carbon (SiC), or gallium nitride (GaN).
3. The semiconductor structure of claim 1, wherein the first conductive layer comprises doped poly silicon, metal, or conductive metal compound.
4. The semiconductor structure of claim 1, wherein the first conductive layer is patterned.
5. The semiconductor structure of claim 1, wherein the bonding layer comprises silicon dioxide.
6. The semiconductor structure of claim 1, wherein the bonding layer comprises an oxide direct bonding interface.
7. The semiconductor structure of claim 1, wherein the first dielectric layer comprises silicon dioxide or high-k dielectric.
8. The semiconductor structure of claim 1, wherein the first dielectric layer has a thickness less than about 50 nm.
9. The semiconductor structure of claim 1, wherein the first dielectric layer comprises a first oxide layer, a second oxide layer on the first oxide layer, and a first nitride layer between the first oxide layer and the second oxide layer.
10. The semiconductor structure of claim 1, further comprising: a second conductive layer between the first dielectric layer and the second substrate; and a second dielectric layer between the second conductive layer and the second substrate.
11. The semiconductor structure of claim 10, wherein the second conductive layer is patterned.
12. The semiconductor structure of claim 1, wherein the first dielectric layer is deposited directly on the second substrate.
13. The semiconductor structure of claim 12, wherein the first conductive layer is deposited directly on the first dielectric layer.
14. The semiconductor structure of claim 1, wherein the first substrate comprises electronic devices, and the electronic devices are transistors, diodes, capacitors, or resistors.
15. The semiconductor structure of claim 1, further comprising: a charge storage layer between the first dielectric layer and the second substrate, wherein the charge storage layer is in direct contact with the second substrate.
16. The semiconductor structure of claim 15, wherein the charge storage layer comprises polysilicon, silicon nitride, or silicon oxynitride.
17. A method for making a semiconductor structure, comprising:(a) providing a second structure comprising a second substrate, a first conductive layer on the second substrate, and a first dielectric layer between the second substrate and the first conductive layer, wherein the second substrate comprises an implanted hydrogen layer;(b) providing a first structure comprising a first substrate;(c) bonding the second structure to the first structure by a bonding layer to form a bonded structure; and(d) removing a portion of the second substrate from approximately the implanted hydrogen layer.
18. The method of claim 17, wherein the second substrate is a single crystalline substrate comprising silicon, germanium, gallium arsenide (GaAs), indium phosphide (InP), silicon carbon (SiC), or gallium nitride (GaN).
19. The method of claim 17, wherein the first conductive layer comprises doped polysilicon, metal, or conductive metal compound.
20. The method of claim 17, wherein the first conductive layer is patterned before the step (c).
21. The method of claim 17, wherein the first dielectric layer comprises silicon dioxide or high-k dielectric.
22. The method of claim 17, wherein the bonding layer comprises silicon dioxide.
23. The method of claim 17, wherein the step (c) comprises forming a portion of the bonding layer on the first conductive layer before bonding the second structure to the first structure.
24. The method of claim 17, wherein the first dielectric layer has a thickness less than about 50 nm.
25. The method of claim 17, wherein the first dielectric layer comprises a first oxide layer, a second oxide layer on the first oxide layer, and a first nitride layer between the first oxide layer and the second oxide layer.
26. The method of claim 17, wherein the step (a) comprises:(al) providing a second substrate;(a2) implanting a hydrogen layer into the second substrate;(a3) forming a first dielectric layer on the second substrate; and(a4) forming a first conductive layer on the first dielectric layer.
27. The method of claim 17, further comprising:(e) polishing an exposed surface of the second substrate after removing a portion of the second substrate.
28. The method of claim 17, wherein the second structure further comprises a second conductive layer between the first dielectric layer and the second substrate and a second dielectric layer between the second conductive layer and the second substrate.
29. The method of claim 28, wherein the second conductive layer is patterned before the step (c).
30. The method of claim 17, wherein the first substrate comprises electronic devices, wherein the electronic devices are transistors, diodes, capacitors, or resistors before the step (c).
31. The method of claim 17, wherein the second structure further comprises a charge storage layer between the second substrate and the first dielectric layer, and the charge storage layer is formed directly on the second substrate.
32. The method of claim 31, wherein the charge storage layer comprises poly silicon, silicon nitride, or silicon oxynitride.
33. A method for making a semiconductor structure, comprising:(a) providing a first structure comprising a first substrate, a first conductive layer on the first substrate, and a bonding layer between the first substrate and the first conductive layer;(b) providing a second structure comprising a second substrate, wherein the second substrate comprises an implanted hydrogen layer;(c) bonding the second structure to the first structure by a first dielectric layer to form a bonded structure,(d) removing a portion of the second substrate from approximately the implanted hydrogen layer.
34. The method of claim 33, wherein the second substrate is a single crystalline substrate comprising silicon, germanium, gallium arsenide (GaAs), indium phosphide (InP), silicon carbon (SiC), or gallium nitride (GaN).
35. The method of claim 33, wherein the first conductive layer comprises doped polysilicon, metal, or conductive metal compound.
36. The method of claim 33, wherein the first dielectric layer comprises silicon dioxide or high-k dielectric.
37. The method of claim 33, wherein the first dielectric layer has a thickness less than about 50 nm.
38. The method of claim 33, wherein the first dielectric layer comprises a first oxide layer, a second oxide layer on the first oxide layer, and a first nitride layer between the first oxide layer and the second oxide layer.
39. The method of claim 38, wherein the second structure further comprises the second oxide layer formed on the second substrate and the first nitride layer formed on the second oxide layer before bonding, and the step (c) comprises bonding the second structure to the first structure by the first oxide layer.
40. The method of claim 38, wherein the first structure further comprises the first oxide layer formed on the first conductive layer and the first nitride layer formed on the first oxide layer before bonding, and the step (c) comprises bonding the second structure to the first structure by the second oxide layer.
41. The method of claim 33, wherein the step (a) comprises:(al) providing a first substrate;(a2) forming a bonding layer on the first substrate; and(a3) forming a first conductive layer on the bonding layer.
42. The method of claim 33, wherein the first substrate comprises electronic devices, wherein the electronic devices are transistors, diodes, capacitors, or resistors before the step (c).
43. The method of claim 33, wherein the second structure further comprises a charge storage layer directly on the second substrate.
44. The method of claim 43, wherein the charge storage layer comprises polysilicon, silicon nitride, or silicon oxynitride.
45. A memory cell, comprising: a source region and a drain region in a second substrate, the source region and the drain region separated by a channel region in the second substrate, wherein the second substrate comprises single crystalline material; a first gate overlapped with the channel region and between a bonding layer and the second substrate; a first gate dielectric between the first gate and the second substrate; a second gate overlapped with the channel region, the second substrate between the first gate and the second gate; and a second gate dielectric between the second gate and the second substrate.
46. The memory cell of claim 45, wherein the second substrate is a single crystalline substrate comprising silicon, germanium, gallium arsenide (GaAs), indium phosphide (InP), silicon carbon (SiC), or gallium nitride (GaN).
47. The memory cell of claim 45 further comprising a bonding layer, wherein the first gate is disposed between the bonding layer and the second substrate, and the bonding layer comprises silicon dioxide.
48. The memory cell of claim 47, wherein the bonding layer comprises an oxide bonding interface.
49. The memory cell of claim 47, wherein the memory cell is electrically connected to at least one of electronic devices in a first substrate, wherein the bonding layer is disposedbetween the first gate and the first substrate, and the electronic devices are transistors, diodes, capacitors, or resistors.
50. The memory cell of claim 45, wherein the first gate comprises doped polysilicon, metal, or conductive metal compound.
51. The memory cell of claim 45, wherein the first gate dielectric comprises a first oxide layer, a second oxide layer on the first oxide layer, and a first nitride layer between the first oxide layer and the second oxide layer.
52. The memory cell of claim 45, wherein the second gate dielectric comprises a first oxide layer, a second oxide layer on the first oxide layer, and a first nitride layer between the first oxide layer and the second oxide layer.
53. The memory cell of claim 45, further comprising: a third gate between the first gate and the second substrate; and a third gate dielectric between the third gate and the second substrate.
54. The memory cell of claim 53, wherein the third gate comprises doped polysilicon, metal, or conductive metal compound.
55. The memory cell of claim 45, further comprising: a third gate between the second gate and the second substrate; and a third gate dielectric between the third gate and the second substrate.
56. The memory cell of claim 55, wherein the third gate comprises doped polysilicon, metal, or conductive metal compound.
57. The memory cell of claim 45, wherein the second gate and the second gate dielectric surround the channel region with two or more sides.
58. The memory cell of claim 45, wherein the memory cell is a capacitor-less memory cell.
59. The memory cell of claim 51, wherein the first nitride layer is configured to store charges, and the memory cell is a non-volatile memory cell.
60. The memory cell of claim 45, further comprising a charge storage region in contact with the second substrate, wherein the charge storage region is disposed between the first gate dielectric and the second substrate.
61. The memory cell of claim 60, wherein the charge storage region comprises polysilicon, silicon nitride, or silicon oxynitride.
62. The memory cell of claim 45, further comprising a charge storage region in contact with the second substrate, wherein the charge storage region comprises polysilicon, silicon nitride, or silicon oxynitride and is disposed between the second gate dielectric and the second substrate.
63. The memory cell of claim 45, wherein the first gate is electrically connected to back gate line, the second gate is electrically connected to a word line, the source region is grounded, and the drain region is electrically connected to a bit line.
