Method for manufacturing a zero layer via
By adjusting the cleaning process sequence, removing the Co-damaged layer first and then the F-containing polymer, the problem of tungsten loss defects in tungsten selective growth was solved, achieving high-quality zero-layer through-hole manufacturing and improving product performance and yield.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANGHAI HUALI INTEGRATED CIRCUIT CORP
- Filing Date
- 2025-01-02
- Publication Date
- 2026-07-07
AI Technical Summary
In existing technologies, tungsten loss defects are easily generated when selectively growing tungsten to form the zero-layer through-hole, which affects product yield.
By changing the cleaning process sequence, wet cleaning to remove the Co-damaged layer is performed first, and the F-containing polymer is used to remove the Co-damaged layer. Then, the F-containing polymer is removed to ensure that no Co-containing residue defects are formed during the selective growth of tungsten.
It effectively prevents tungsten layer growth on surfaces with Co residue defects, avoids tungsten loss defects, ensures that the tungsten layer completely fills the through-holes, and improves product performance and yield.
Smart Images
Figure CN122349355A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a method for manufacturing a semiconductor integrated circuit, and more particularly to a method for manufacturing a zero-layer via (V0). Background Technology
[0002] To reduce contact resistance, the metal zero layer in the mid-stage process (MOL) has been upgraded from a W-bonding process to a Co-bonding process. Because Co is relatively reactive, a liner layer needs to be formed after the Co layer is formed but before the first interlayer film is formed. This liner layer covers the Co layer, thus protecting it.
[0003] A zero-layer via needs to be formed on top of the zero-layer metal layer to further connect with the upper metal layer. The zero-layer via is typically formed by selectively growing tungsten to fill the via opening. During the formation of the zero-layer via, severe tungsten loss defects often occur.
[0004] like Figures 1A to 1D The diagram shown is a schematic representation of the device structure in each step of the existing method for manufacturing a zero-layer via; the existing method for manufacturing a zero-layer via includes the following steps:
[0005] like Figure 1A As shown, an underlying structure is provided, the underlying structure having a metal zero layer 102, a first pad layer 103 and a first interlayer film 104, the metal zero layers 102 being separated by a zero interlayer film 101, the first pad layer 103 covering the top surface of the metal zero layer 102 and the top surface of the zero interlayer film 101, and the first interlayer film 104 being formed on the top surface of the first pad layer 103.
[0006] The material of the metal layer 102 includes Co; due to the reactivity of Co, after the chemical mechanical polishing (CMP) of Co is smoothed, the first pad layer 103 needs to be quickly covered on top, and the first pad layer 103 serves as a protective layer for Co to reduce the oxidation and loss of Co.
[0007] Typically, the material of the first liner layer 103 is silicon nitride or silicon carbide.
[0008] The material of the first interlayer film 104 includes silicon oxide or a low-k dielectric material.
[0009] like Figure 1A As shown, the first interlayer film 104 and the first pad layer 103 in the formation region of the zeroth layer via are etched sequentially to form the zeroth layer via opening 105.
[0010] The etching process for the first liner layer 103 is a liner removal (LRM) process. The etching process for the first liner layer 103 is a dry etching process, and the process gases include difluoromethane (CH2F2), and also include oxygen or nitrogen.
[0011] Figure 1A In the diagram, the dashed line 106 with arrows indicates the etching process of the first backing layer 103.
[0012] Because fluorine-containing gas was used in the etching process of the first liner layer 103, an F-containing polymer 107 will eventually be formed.
[0013] Meanwhile, since the etching process of the zero-layer via opening 105 includes multi-step etching, plasma is used in the multi-step etching, i.e., plasma etching is performed. That is, the etching process of the first interlayer film 104 and the first pad layer 103 will use dry plasma etching. The bombardment of plasma in plasma etching will damage the surface area of the metal zero layer 102 and form a Co damage layer 102a.
[0014] like Figure 1B As shown, a first wet cleaning is performed to remove some of the F-containing polymer 107. Typically, the cleaning solution used to remove the F-containing polymer 107 is the applicant's C1000 solution. Figure 1B The results show that the content of the F-containing polymer 107 decreases. The applicant's solution, model C1000, mainly contains some inhibitors of Co and is weakly alkaline.
[0015] like Figure 1C As shown, a second wet cleaning was performed. In the second wet cleaning, a deionized (DI) solution was first used to break open the Co damaged layer 102a, and then a C1000 solution was used to clean the Co residue.
[0016] When removing the Co-damaged layer 102a using a DI solution, it is necessary to utilize the F in the residual F-containing polymer 107 to form HF, and then use the HF to remove the Co-damaged layer 102a.
[0017] However, in actual processes, such as Figure 1C As shown, the second wet cleaning cannot completely remove Co residues; on the contrary, it is easy to form defects 108 containing Co residues.
[0018] Then, an alkaline solution is used for pretreatment, namely OH pretreatment.
[0019] Subsequently, tungsten selective growth is performed to deposit a tungsten layer 1091 from the bottom surface of the zero-layer via opening 105 upwards, and the tungsten layer 1091 filling the zero-layer via opening 105 forms the zero-layer via.
[0020] The selective tungsten growth refers to growth only on surfaces containing metal. When the side or exterior of the zero-layer via opening 105 has Co-containing residue defects 108, tungsten will also grow on the surface containing the Co-containing residue defects 108. Figure 1D In the diagram, the tungsten layer grown on the surface containing Co residue defect 108 is designated as 1092. The tungsten layer 1092 is prone to forming a seal at the top of the zero-layer through-hole opening 105. After sealing, the tungsten can no longer grow inside the zero-layer through-hole opening 105, eventually forming a void 110.
[0021] The tungsten layer 109 is planarized by a tungsten chemical mechanical polishing (CMP) process. After the CMP process, the internal voids 110 in the zero-layer through-hole opening 105 where the tungsten layer 1092 is formed may be exposed.
[0022] like Figures 2A to 2D The image shown is a top view photograph of the device in each step of the existing zero-layer through-hole manufacturing method; Figure 2A yes Figure 1C The corresponding photo after the second wet cleaning was completed. Figure 2A In the diagram, the first interlayer membrane is indicated by the designation 104a, the zero-layer through-hole opening is indicated by the designation 105a, and the Co-containing residue defect is indicated by the designation 108a.
[0023] Figure 2B The photo shows the result after OH pretreatment. It can be seen that the Co residue defect 108a cannot be completely removed.
[0024] Figure 2C yes Figure 1D The corresponding image shows the result after selective tungsten growth. Figure 2C In the diagram, the tungsten layer formed from the bottom up in the zero-layer through-hole opening 105a is indicated by the designation 1091a, and the tungsten layer grown from the surface of the Co-containing residue defect 108a is indicated by the designation 1092a.
[0025] Figure 2DThe image shows the result after the tungsten chemical mechanical polishing (CMP) process. It can be seen that when the tungsten layer 1091a completely fills the zero-layer via opening 105a, a normal zero-layer via 109 composed of tungsten layer 1091a is formed. However, in the region containing the tungsten layer 1092a, the zero-layer via opening 105a forms a via 109a with tungsten loss defects. Vias 109a with tungsten loss defects cannot achieve normal electrical connection, affecting product yield. Summary of the Invention
[0026] The technical problem to be solved by the present invention is to provide a method for manufacturing a zero-layer through-hole, which can selectively grow tungsten to form a tungsten layer for the zero-layer through-hole and prevent tungsten loss defects.
[0027] To solve the above-mentioned technical problems, the method for manufacturing the zeroth layer through-hole provided by the present invention includes the following steps:
[0028] A substructure is provided, the substructure having a metal zero layer, a first pad layer and a first interlayer film, the metal zero layers being separated by a zero interlayer film, the first pad layer covering the top surface of the metal zero layers and the top surface of the zero interlayer film, the first interlayer film being formed on the top surface of the first pad layer; the material of the metal zero layer includes Co.
[0029] The first interlayer film and the first liner layer in the formation region of the zero-layer via are etched sequentially to form the zero-layer via opening; during the etching process of the first liner layer, an F-containing polymer is generated; the plasma in the etching process of the zero-layer via opening forms a Co damage layer on the surface region of the metal zero layer.
[0030] A first wet cleaning is performed to remove the Co-damaged layer in the metal layer. In the first wet cleaning, the F in the retained F-containing polymer is used to enhance the removal effect on the Co-damaged layer and remove the Co-damaged layer.
[0031] A second wet cleaning is performed to remove the F-containing polymer, taking advantage of the fact that the Co-damaged layer has been removed beforehand, to prevent the formation of Co-containing residue defects during the second wet cleaning.
[0032] Selective tungsten growth is performed to deposit a tungsten layer from the bottom surface of the zeroth layer via opening upwards, and the tungsten layer filling the zeroth layer via opening forms the zeroth layer via.
[0033] A further improvement is that the material of the first liner layer includes silicon nitride or silicon carbide.
[0034] A further improvement is that the material of the first interlayer film includes silicon oxide or a low-k dielectric material.
[0035] A further improvement is that, in forming the zeroth layer via opening, the etching process for the first interlayer film includes a first dry etching.
[0036] A further improvement is that, in forming the zeroth layer via opening, the etching process for the first interlayer film further includes: performing a second wet etching after completing the first dry etching.
[0037] A further improvement is that, in forming the zeroth layer via opening, the etching process of the first pad layer includes a second dry etching; the process gas for the second dry etching includes a fluorine-containing gas.
[0038] A further improvement is that the process gas for the second dry etching also includes oxygen or nitrogen.
[0039] A further improvement is that the fluorine-containing gas includes difluoromethane.
[0040] A further improvement is that the cleaning solution used in the first wet cleaning process is a deionized solution.
[0041] A further improvement is that the cleaning solution used in the second wet cleaning is a solution containing a Co inhibitor and being weakly alkaline.
[0042] A further improvement is to repeat the second wet cleaning process multiple times.
[0043] A further improvement is that the second wet cleaning is repeated twice.
[0044] A further improvement is that, after the second wet cleaning is completed and before the tungsten selective growth is performed, the process further includes:
[0045] Pretreatment was performed using an alkaline solution.
[0046] A further improvement is that, after the selective tungsten growth is completed, the process further includes:
[0047] The tungsten layer is planarized by a tungsten chemical mechanical polishing process.
[0048] This invention utilizes the characteristic that both an fluorine-containing polymer and a Co-containing damage layer are generated simultaneously during the formation of the zero-layer via opening. It changes the process sequence of the prior art, which involves first removing the fluorine-containing polymer and then removing the Co-containing damage layer. Instead, the cleaning process for removing the Co-containing damage layer (the first wet cleaning) is performed first. Since the fluorine-containing polymer is not removed at this stage, the fluorine contained in the polymer can be used to remove the Co-containing damage layer. After removing the Co-containing damage layer, the cleaning process for removing the fluorine-containing polymer (the second wet cleaning) is performed. This second wet cleaning completely removes the fluorine-containing polymer. The Co layer is completely removed. Since the Co-damaged layer has been removed, and the second wet cleaning does not affect the undamaged Co layer (i.e., the fresh Co layer), no Co-containing defects will be formed during the second wet cleaning. That is, no Co-containing defects will be formed on the side or outside of the zero-layer via opening. This ensures that the tungsten layer grows completely from the top surface of the metal zero layer exposed at the bottom of the zero-layer via opening during selective tungsten growth. This prevents the formation of top sealing defects of the zero-layer via opening and tungsten loss defects on the right side when the tungsten layer grows along the Co-containing defect surface in other areas. As a result, the tungsten layer completely fills the zero-layer via opening from the bottom upwards and forms a high-quality zero-layer via, ultimately improving product performance and yield. Attached Figure Description
[0049] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:
[0050] Figures 1A-1D This is a schematic diagram of the device structure in each step of the existing zero-layer through-hole manufacturing method;
[0051] Figures 2A-2D These are top-view photographs of the devices in each step of the existing zero-layer through-hole manufacturing method;
[0052] Figure 3 This is a flowchart of the manufacturing method of the zero-layer through hole in an embodiment of the present invention;
[0053] Figures 4A-4D This is a schematic diagram of the device structure in each step of the manufacturing method of the zero-layer through-hole in an embodiment of the present invention. Detailed Implementation
[0054] like Figure 3 The diagram shown is a flowchart of the manufacturing method of the zero-layer through-hole in an embodiment of the present invention; as shown... Figures 4A to 4D The diagram shown is a schematic representation of the device structure in each step of the manufacturing method of the zero-layer via according to an embodiment of the present invention. The manufacturing method of the zero-layer via according to an embodiment of the present invention includes the following steps:
[0055] Step S101, as follows Figure 4A As shown, an underlying structure is provided, wherein a metal zero layer 202, a first pad layer 203 and a first interlayer film 204 are formed thereon, the metal zero layers 202 are separated from each other by a zero interlayer film 201, the first pad layer 203 covers the top surface of the metal zero layer 202 and the top surface of the zero interlayer film 201, and the first interlayer film 204 is formed on the top surface of the first pad layer 203.
[0056] The material of the metal layer 202 includes Co; due to the reactivity of Co, after the Co is chemically and mechanically ground and smoothed, the first pad layer 203 needs to be quickly covered on top. The first pad layer 203 serves as a protective layer for Co to reduce the oxidation and loss of Co.
[0057] In this embodiment of the invention, the first padding layer 203 is made of silicon nitride. In other embodiments, the first padding layer 203 may also be made of silicon carbide.
[0058] In this embodiment of the invention, the material of the first interlayer film 204 includes silicon oxide or a low-k dielectric material.
[0059] The metal zero layer 202 and the bottom semiconductor device's gate conductive material layer and doped regions such as source or drain regions.
[0060] Step S102, as follows Figure 4A As shown, the first interlayer film 204 and the first liner layer 203 in the formation region of the zero-layer via are etched sequentially to form the zero-layer via opening 205; during the etching process of the first liner layer 203, an F-containing polymer 207 is generated; the plasma in the etching process of the zero-layer via opening 205 forms a Co damage layer 202a on the surface region of the metal zero layer 202.
[0061] In forming the zeroth layer via opening 205, the etching process of the first interlayer film 204 includes a first dry etching. The first dry etching uses plasma and stops at the first pad layer 203. Therefore, in the first dry etching, the first pad layer 203 serves as an etching stop layer, which improves the uniformity of the etching depth.
[0062] In forming the zeroth layer via opening 205, the etching process of the first layer interlayer film 204 further includes performing a second wet etching after completing the first dry etching.
[0063] In forming the zeroth layer via opening 205, the etching process for the first pad layer 203 is a pad removal process. The etching process for the first pad layer 203 includes a second dry etching. Figure 4A In the diagram, the dashed line 206 with an arrow indicates the second dry etching process. This second dry etching process also employs plasma, and the process gas for this process includes a fluorine-containing gas.
[0064] In some embodiments, the fluorine-containing gas includes difluoromethane.
[0065] The process gas for the second dry etching also includes oxygen or nitrogen.
[0066] Because the fluorine-containing gas was used in the second dry etching process, the final product is an F-containing polymer 207.
[0067] Meanwhile, since the etching process of the zero-layer via opening 205 includes multi-step etching, plasma is used in the multi-step etching, i.e., plasma etching is performed. For example, plasma is used in the first dry etching and the second dry etching. The bombardment of plasma will damage the surface area of the metal zero layer 202 and form a Co damage layer 202a.
[0068] Step S103, as follows Figure 4B As shown, a first wet cleaning is performed to remove the Co-damaged layer 202a in the metal layer 202. In the first wet cleaning, the F in the retained F-containing polymer 207 is used to enhance the removal effect on the Co-damaged layer 202a and remove the Co-damaged layer 202a.
[0069] In this embodiment of the invention, the cleaning solution used in the first wet cleaning is a deionized solution. The DI solution and the F in the F-containing polymer 207 form HF, which can remove the Co-damaged layer 202a. It is necessary to completely remove the Co-damaged layer 202a to ensure that no defects containing Co residues will form subsequently.
[0070] In this embodiment of the invention, since the F-containing polymer 207 is not removed before the first wet cleaning, a sufficient concentration of HF can be formed to ensure that the Co-damaged layer 202a can be removed.
[0071] In existing methods, after removing the F-containing polymer, the DI solution cannot form a sufficient concentration of HF to remove the Co-damaged layer during cleaning, thus resulting in Co-containing residue defects.
[0072] Step S104, as follows Figure 4CAs shown, a second wet cleaning is performed to remove the F-containing polymer 207, taking advantage of the fact that the Co-damaged layer 202a has been removed beforehand, to prevent the formation of Co-containing residue defects during the second wet cleaning.
[0073] In this embodiment of the invention, the cleaning solution for the second wet cleaning is a weakly alkaline solution containing a Co inhibitor. Preferably, the cleaning solution for the second wet cleaning is the applicant's C1000 solution.
[0074] The C1000 solution effectively removes the F-containing polymer 207. Furthermore, since the C1000 solution does not damage the undamaged, fresh Co layer, it does not carry away Co-containing residues, thus preventing the formation of Co-containing residue defects.
[0075] In this embodiment of the invention, since the C1000 solution does not damage the fresh Co layer, the second wet cleaning can be repeated multiple times. In some embodiments, the second wet cleaning is repeated twice.
[0076] By repeating the second wet cleaning process multiple times, it is ensured that all F-containing polymer 207 is removed, as well as all newly generated residues from the first wet cleaning process. This reduces the defect sites for selectivity loss in subsequent tungsten selective growth, thereby reducing the root cause of subsequent tungsten loss defects.
[0077] In addition, in this embodiment of the invention, the first wet cleaning itself does not pose a risk of Selectivity Loss.
[0078] In some embodiments, the process includes the following steps after the second wet cleaning is completed and before subsequent tungsten selective growth:
[0079] Pretreatment is performed using an alkaline solution, i.e., OH pretreatment.
[0080] Step S105, as follows Figure 4D As shown, selective tungsten growth is performed to deposit a tungsten layer 209 from the bottom surface of the zero-layer via opening 205 upwards, and the tungsten layer 209 filling the zero-layer via opening 205 forms the zero-layer via.
[0081] The selective tungsten growth occurs only on the surface containing the metal. In this embodiment, the metal is located only at the bottom of the zero-layer via opening 205, allowing growth to proceed from the bottom upwards. This achieves filling of the zero-layer via opening 205 without the need for an adhesion barrier layer and seed layer. Furthermore, by eliminating the Co-containing residue defect, the sides and outer surface of the zero-layer via opening 205 are not at locations where tungsten grows. This ensures that the top of the zero-layer via opening 205 remains unsealed during selective tungsten growth, preventing voids and tungsten loss.
[0082] In this embodiment of the invention, after the selective tungsten growth is completed, the process further includes:
[0083] The tungsten layer 209 is planarized by a tungsten chemical mechanical polishing process.
[0084] This invention utilizes the characteristic that an F-containing polymer 207 and a Co-damaged layer 202a are simultaneously generated during the formation of the zero-layer through-hole opening 205. It changes the process sequence of the prior art, which involves first removing the F-containing polymer 207 and then removing the Co-damaged layer 202a. Instead, the first wet cleaning process (removing the Co-damaged layer 202a) is performed first. Since the F-containing polymer 207 is not yet removed, the F contained in the polymer 207 can be used to remove the Co-damaged layer 202a. After removing the Co-damaged layer 202a, a second wet cleaning process (removing the F-containing polymer 207) is performed. This second wet cleaning completely removes the F-containing polymer 207, while simultaneously removing the Co-damaged layer 202a. The Co layer has been removed, and the second wet cleaning will not affect the undamaged Co layer, i.e., the fresh Co layer. Therefore, no Co-containing defects will be formed during the second wet cleaning. That is, no Co-containing defects will be formed on the side and outside of the zero-layer through-hole opening 205. In this way, during the selective growth of tungsten, it can be ensured that the tungsten layer 209 grows completely from the top surface of the metal zero layer 202 exposed at the bottom of the zero-layer through-hole opening 205. This can prevent the tungsten layer 209 from forming the top sealing of the zero-layer through-hole opening 205 and the tungsten loss defects on the right side when it grows along the Co-containing defect surface in other areas. Thus, the tungsten layer 209 completely fills the zero-layer through-hole opening 205 from the bottom to the top and forms a high-quality zero-layer through-hole, which can ultimately improve the performance and yield of the product.
[0085] The present invention has been described in detail above through specific embodiments, but these are not intended to limit the invention. Many modifications and improvements can be made by those skilled in the art without departing from the principles of the invention, and these should also be considered within the scope of protection of the present invention.
Claims
1. A method for manufacturing a zero-layer through-hole, characterized in that, Includes the following steps: A substructure is provided, the substructure having a metal zero layer, a first pad layer, and a first interlayer film, the metal zero layers being separated by a zero interlayer film, the first pad layer covering the top surface of the metal zero layers and the top surface of the zero interlayer film, the first interlayer film being formed on the top surface of the first pad layer; the material of the metal zero layer includes Co; The first interlayer film and the first liner layer in the formation region of the zero-layer via are etched sequentially to form the zero-layer via opening; during the etching process of the first liner layer, F-containing polymer is generated; the plasma in the etching process of the zero-layer via opening forms a Co damage layer on the surface region of the metal zero layer. A first wet cleaning is performed to remove the Co damage layer in the metal layer. In the first wet cleaning, the F in the retained F-containing polymer is used to enhance the removal effect on the Co damage layer and remove the Co damage layer. A second wet cleaning is performed to remove the F-containing polymer, taking advantage of the fact that the Co-damaged layer has been removed beforehand, to prevent the formation of Co-containing residue defects during the second wet cleaning; Selective tungsten growth is performed to deposit a tungsten layer from the bottom surface of the zeroth layer via opening upwards, and the tungsten layer filling the zeroth layer via opening forms the zeroth layer via.
2. The method for manufacturing the zeroth layer through-hole as described in claim 1, characterized in that: The material of the first liner layer includes silicon nitride or silicon carbide.
3. The method for manufacturing the zeroth layer through-hole as described in claim 1, characterized in that: The material of the first interlayer film includes silicon oxide or a low-k dielectric material.
4. The method for manufacturing the zeroth layer through-hole as described in claim 3, characterized in that: In forming the zeroth layer via opening, the etching process of the first interlayer film includes a first dry etching.
5. The method for manufacturing the zeroth layer through-hole as described in claim 4, characterized in that: In forming the zeroth layer via opening, the etching process of the first interlayer film further includes performing a second wet etching after completing the first dry etching.
6. The method for manufacturing the zeroth layer through-hole as described in claim 2, characterized in that: In forming the zeroth layer via opening, the etching process of the first pad layer includes a second dry etching; The process gas used in the second dry etching includes fluorine-containing gas.
7. The method for manufacturing the zeroth layer through-hole as described in claim 6, characterized in that: The process gas for the second dry etching also includes oxygen or nitrogen.
8. The method for manufacturing the zeroth layer through-hole as described in claim 6, characterized in that: The fluorine-containing gas includes: difluoromethane.
9. The method for manufacturing the zeroth layer through-hole as described in claim 1, characterized in that: The cleaning solution used in the first wet cleaning process is a deionized solution.
10. The method for manufacturing the zeroth layer through-hole as described in claim 1, characterized in that: The cleaning solution used in the second wet cleaning process is a weakly alkaline solution containing a Co inhibitor.
11. The method for manufacturing the zeroth layer through-hole as described in claim 10, characterized in that: Repeat the second wet cleaning process multiple times.
12. The method for manufacturing the zeroth layer through-hole as described in claim 11, characterized in that: The second wet cleaning process is repeated twice.
13. The method for manufacturing the zeroth layer through-hole as described in claim 1, characterized in that: The process includes, after the second wet cleaning is completed and before the tungsten selective growth is performed: Pretreatment was performed using an alkaline solution.
14. The method for manufacturing the zeroth layer through-hole as described in claim 1, characterized in that: After the selective tungsten growth is completed, the process also includes: The tungsten layer is planarized by a tungsten chemical mechanical polishing process.