Method for forming hybrid bonding with through substrate via (TSV)

Abstract

Method for forming a semiconductor device structure is provided. The semiconductor device structure includes a first semiconductor wafer and a second semiconductor wafer bonded via a hybrid bonding structure, and the hybrid bonding structure includes a first conductive material embedded in a first polymer material and a second conductive material embedded in a second polymer material. The first conductive material is bonded to the second conductive material and the first polymer material is bonded to the second polymer material. The semiconductor device also includes at least one through silicon via (TSV) extending from a bottom surface of the first semiconductor wafer to a metallization structure of the first semiconductor wafer. The semiconductor device structure also includes an interconnect structure formed over the bottom surface of the first semiconductor wafer, and the interconnect structure is electrically connected to the metallization structure via the TSV.

Claims

1. A method for forming a semiconductor device structure, comprising: forming at least one first TSV in a first semiconductor wafer; forming a metallization structure on the first TSV, wherein the metallization structure is in direct contact with the first TSV; forming a first bonding structure on the metallization structure, wherein the first bonding structure comprise a first diffusion barrier layer in a first polymer material, and the metallization structure is between the first diffusion barrier layer and the first TSV; forming at least one second TSV in a second semiconductor wafer; hybrid bonding the first semiconductor wafer and the second semiconductor wafer, wherein the step of forming at least one second TSV is performed before the step of hybrid bonding the first semiconductor wafer and the second semiconductor wafer, wherein the step of hybrid bonding the first semiconductor wafer and the second semiconductor wafer further comprises: bonding the first polymer material of the first semiconductor wafer to a second polymer material of the second semiconductor wafer; and thinning the first semiconductor wafer from a bottom surface of the first semiconductor wafer to expose the first TSV.

2. The method as claimed in claim 1, further comprising: forming an interconnect structure over the bottom surface of the first semiconductor wafer after thinning the first semiconductor wafer, wherein the interconnect structure is electrically connected to the metallization structure of the first semiconductor wafer through the first TSV.

3. The method as claimed in claim 1, further comprising: forming a redistribution (RDL) structure over the bottom surface of the second semiconductor wafer, wherein the second TSV is electrically connected to the RDL structure.

4. The method as claimed in claim 1, wherein a width of the first TSV is smaller than a width of the second TSV.

5. The method as claimed in claim 1, further comprising: forming a transistor in the second semiconductor wafer, wherein the transistor is electrically connected to a second metallization structure of the second semiconductor wafer.

6. The method as claimed in claim 1, wherein hybrid bonding the first semiconductor wafer and the second semiconductor wafer further comprises: bonding a first conductive material embedded in the first polymer and a second conductive material embedded in the second polymer, wherein the first conductive material is surrounded by the first diffusion barrier layer.

7. The method as claimed in claim 1, further comprising: forming a transistor in the first semiconductor wafer, wherein the transistor is electrically connected to the metallization structure; and forming a contact plug on the transistor, wherein the metallization structure is in direct contact with the first TSV and the contact plug.

8. A method for forming a semiconductor device structure, comprising: providing a first semiconductor wafer and a second semiconductor wafer, wherein a first TSV is formed in the first semiconductor wafer; forming a conductive structure on the first TSV; forming a first polymer material over the conductive structure, wherein the conductive structure is between the first TSV and the first polymer material; forming a second TSV in the second semiconductor wafer, wherein a width of the first TSV is smaller than a width of the second TSV; hybrid bonding the first semiconductor wafer and the second semiconductor wafer, wherein the step of hybrid bonding the first semiconductor wafer and the second semiconductor wafer further comprises: bonding the first polymer material of the first semiconductor wafer to a second polymer material of the second semiconductor wafer; and thinning the first semiconductor wafer from a bottom surface of the first semiconductor wafer to expose the first TSV.

9. The method as claimed in claim 8, wherein the first TSV extends from the bottom surface of the first semiconductor wafer to the conductive structure of the first semiconductor wafer, and the step of forming the conductive structure is before the step of hybrid bonding the first semiconductor wafer and the second semiconductor wafer.

10. The method as claimed in claim 8, wherein the second TSV extending from a bottom surface of the second semiconductor wafer to a second metallization structure of the second semiconductor wafer, and the second metallization structure is between the TSV and a second conductive material embedded in the second polymer material.

11. The method as claimed in claim 8, further comprising: forming a redistribution (RDL) structure over the bottom surface of the second semiconductor wafer, wherein the second TSV is electrically connected to the RDL structure.

12. A method for forming a semiconductor device structure, comprising: providing a first semiconductor wafer and a second semiconductor wafer, wherein a first TSV is formed in the first semiconductor wafer; forming a transistor in the first semiconductor wafer; forming a contact plug on the transistor; forming a conductive feature on the first TSV and the contact plug, wherein the conductive feature is in direct contact with the first TSV and the contact plug; forming a first bonding structure over the conductive feature, wherein the first bonding structure comprises a first conductive material embedded in a first polymer, and the conductive feature is between the first TSV and the first conductive material; forming a second TSV in the second semiconductor wafer; hybrid bonding the first semiconductor wafer and the second semiconductor wafer, wherein the step of forming the second TSV in the second semiconductor wafer is performed before the step of hybrid bonding the first semiconductor wafer and the second semiconductor wafer, wherein the step of hybrid bonding the first semiconductor wafer and the second semiconductor wafer further comprises: bonding the first polymer material of the first semiconductor wafer to a second polymer material of the second semiconductor wafer; thinning the second semiconductor wafer from a bottom surface of the second semiconductor wafer to expose the second TSV; and forming a redistribution (RDL) structure over the bottom surface of the second semiconductor wafer, wherein the second TSV is electrically connected to the RDL structure.

13. The method as claimed in claim 12, wherein hybrid bonding the first semiconductor wafer and the second semiconductor wafer further comprises: bonding the first conductive material and a second conductive material embedded in the second polymer.

14. The method as claimed in claim 12, further comprising: thinning the first semiconductor wafer from a bottom surface of the first semiconductor wafer to expose the first TSV before thinning the second semiconductor wafer.

15. The method as claimed in claim 14, further comprising: forming an interconnect structure over the bottom surface of the first semiconductor wafer after thinning the first semiconductor wafer, wherein the interconnect structure is electrically connected to the conductive feature of the first semiconductor wafer through the first TSV.

16. The method as claimed in claim 6, wherein the first conductive material is in direct contact with the metallization structure of the first semiconductor wafer, and the metallization structure is between the first conductive material and the first TSV.

17. The method as claimed in claim 8, wherein the step of forming the second TSV in the second semiconductor wafer is performed before the step of hybrid bonding the first semiconductor wafer and the second semiconductor wafer.

18. The method as claimed in claim 1, wherein the number of the first TSV is greater than the number of the second TSV.

19. The method as claimed in claim 12, wherein the conductive feature is a single metal layer.

20. The method as claimed in claim 12, further comprising: forming an interconnect structure over the bottom surface of the first semiconductor wafer after thinning the first semiconductor wafer; forming an under bump metallization (UBM) layer over the RDL structure; and forming a conductive element over the UBM layer, wherein the interconnect structure and the conductive element are oppositely formed.

Description

BRIEF DESCRIPTION OF THE DRAWING

(1) For a more complete understanding of the present disclosure, and the advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:

(2) FIGS. 1A-1D show cross-sectional representations of various stages of forming a semiconductor device, in accordance with some embodiments of the disclosure.

(3) FIGS. 2A-2D show cross-sectional representations of various stages of forming a semiconductor device, in accordance with some embodiments of the disclosure.

DETAILED DESCRIPTION

(4) It is to be understood that the following disclosure provides many different embodiments, or examples, for implementing different features of the disclosure. Specific examples of components and arrangements are described below to simplify the present disclosure. These are, of course, merely examples and are not intended to be limiting. Moreover, the performance of a first process before a second process in the description that follows may include embodiments in which the second process is performed immediately after the first process, and may also include embodiments in which additional processes may be performed between the first and second processes. Various features may be arbitrarily drawn in different scales for the sake of simplicity and clarity. Furthermore, the formation of a first feature over or on a second feature in the description may include embodiments in which the first and second features are formed in direct or indirect contact.

(5) FIGS. 1A-1D show cross-sectional representations of various stages of forming semiconductor devices in accordance with some embodiments. Referring to FIG. 1A, a cross-sectional representation of a portion of semiconductor wafer 100 and a portion of semiconductor wafer 200 are shown in accordance with some embodiments.

(6) Semiconductor wafer 100 includes a semiconductor substrate 104, which may be made of silicon or other semiconductor materials. Alternatively or additionally, semiconductor substrate 104 may include other elementary semiconductor materials such as germanium. In some embodiments, semiconductor substrate 104 is made of a compound semiconductor such as silicon carbide, gallium arsenic, indium arsenide, or indium phosphide. In some embodiments, semiconductor substrate 104 is made of an alloy semiconductor such as silicon germanium, silicon germanium carbide, gallium arsenic phosphide, or gallium indium phosphide. In some embodiments, semiconductor substrate 104 includes an epitaxial layer. For example, semiconductor substrate 104 has an epitaxial layer overlying a bulk semiconductor.

(7) Referring to FIG. 1A, device regions 103 are formed in semiconductor wafer 100 in a front-end-of-line (FEOL) process in some embodiments. Each device regions 103 includes a gate structure 109 embedded in a dielectric layer 107, source/drain regions 110, and isolation structures 112, such as shallow trench isolation (STI) structures. Gate structure 109 includes a gate dielectric layer 106, a gate electrode 108, and possibly spacers (not shown). Device regions 103 shown in FIG. 1A are merely examples, and other structures may be formed in device regions 103.

(8) Device regions 103 may form various N-type metal-oxide semiconductor (NMOS) and/or P-type metal-oxide semiconductor (PMOS) devices, such as transistors or memories, and the like, which are interconnected to perform one or more functions. Other devices, such as capacitors, resistors, diodes, photo-diodes, fuses, and the like may also be formed on substrate 104.

(9) As shown in FIG. 1A, through-substrate vias (TSVs) 400 are formed between two adjacent device regions 103. TSVs 400 extend into substrate 104. TSVs 400 are used to provide electrical connections and for heat dissipation for 3DICs. Although FIG. 1A shows four TSVs, the number of TSVs may be adjusted according to actual application.

(10) Each TSVs 400 includes a liner 410, a diffusion barrier layer 420, and a conductive material 430 in accordance with some embodiments. Liner 410 is made of an insulating material, such as oxides or nitrides. Liner 410 may be formed by using a plasma enhanced chemical vapor deposition (PECVD) process or other applicable processes. Liner 410 may be a single layer or multi-layers. In some embodiments, liner 410 has a thickness in a range from about 100 to about 5000 .

(11) Diffusion Barrier layer 420 is made of Ta, TaN, Ti, TiN, or CoW. In some embodiments, diffusion barrier layer 420 is formed by a physically vapor deposition (PVD) process. Conductive material 430 is made of copper (Cu), copper alloy, aluminum (Al), aluminum alloys, or combinations thereof. Alternatively, other applicable materials may be used. In some embodiments, conductive material 430 is formed by plating.

(12) With high aspect ratio, filling materials into the TSV opening becomes challenging. Voids may form in a TSV opening. In addition, due to insufficient sidewall coverage of liner 410 or diffusion barrier layer 420, some extrusion or diffusion problems related to conductive via material 430 may occur. In contrast, as shown in FIG. 1A, TSVs 400 are designed to have a smaller depth D.sub.1 than height H.sub.1 of semiconductor wafer 100. Therefore, the void problems and the extrusion or diffusion problems related to the conductive material 430 are resolved or greatly reduced.

(13) In addition, width W.sub.1 of TSVs 400 is reduced along with a decrease of depth D.sub.1 of TSVs 400. When distance W.sub.2 is smaller, a larger area of device regions 103 may be used. As a result, integrated intensity of the devices in device regions 103 is further improved.

(14) In some embodiments, semiconductor wafer 100 has a height H.sub.1 from a bottom surface 104b of semiconductor substrate 104 to a top surface of gate structure 109 in a range from about 1 m to about 20 m. In some embodiments, TSVs 400 have a width W.sub.1 in a range from about 0.025 m to about 2 m. In some embodiments, TSVs 400 have a depth D.sub.1 in a range from about 0.2 m to about 10 m. In some embodiments, TSVs 400 have an aspect ratio (D.sub.1/W.sub.1) in a range from about 2 to about 15.

(15) In addition, devices in the vicinity of the TSV suffer from serious performance degradation due to the stress induced by the TSV. A keep-out zone (KOZ) is used to define a region where no devices could be placed within. In some embodiments, keep-out zone (KOZ) is defined by a distance W.sub.2, which is measured from a sidewall 400a of TSV 400 to a nearest gate structure 209. Since the depth D.sub.1 of TSVs 400 is made smaller, a smaller width W.sub.1 is achieved. Therefore, overall stress induced by TSVs 400 is reduced. In some embodiments, distance W.sub.2 is in a range from 0.01 m to about 3 m. In some embodiments, when width W.sub.1 of TSVs 400 is reduced to a range from about 2 m to about 3 m, the stress induced by TSV can almost be ignored.

(16) As shown in FIG. 1A, a metallization structure 122 is formed over TSVs 400 and device regions 103 to individually connect to TSVs 400 and device regions 103. In some embodiments, metallization structure 122 includes interconnect structure, such as contact plugs 114 and conductive features 124. Conductive features 124 are embedded in an insulating material 126. In some embodiments, insulating material 126 is made of silicon oxide. In some embodiments, insulating material 126 includes multiple dielectric layers of dielectric materials. Metallization structure 122 shown is merely for illustrative purposes. Metallization structure 122 may include other configurations and may include one or more conductive lines and via layers.

(17) A bonding structure 142 is formed over metallization structure 122. Bonding structure 142 includes a conductive material 144 embedded in a polymer material 146. Conductive material 144 is contact pad (or bond pad) formed on a top surface of semiconductor wafer 100. Conductive features 124 are connected to conductive material 144. Conductive material 144 may be made of conductive materials, such as copper (Cu), copper alloy, aluminum (Al), aluminum alloy, or combinations thereof. Other applicable materials may be used as conductive material 144.

(18) In some embodiments, if conductive material 144 is made of a metal, such as copper, which is easy to diffuse, a diffusion barrier layer 143 is needed. Diffusion barrier layer 143 may be made of silicon nitride (SiN), silicon oxynitride (SiON), titanium nitride (TiN), tantalum nitride (TaN), or aluminum nitride (AlN). In some embodiments, conductive material 144 is made of copper, and diffusion barrier layer 143 is made of Ti, TiN, Ta, TaN, Ta/TaN, CoP or CoW. In some embodiments, diffusion barrier layer 143 has a thickness in a range from about 5 to about 1000 .

(19) In some embodiments, polymer material 146 is benzocyclobutene (BCB) polymer, polyimide (PI), or polybenzoxazole (PBO). In some embodiments, polymer material 146 is made of benzocyclobutene (BCB) polymer and is applied to metallization structure 122 by spin coating. Since benzocyclobutene polymer is a soft material, it can tolerant more stress resulting from the TSV formed in the subsequent processes, compared to other dielectric materials such as silicon dioxide.

(20) As shown in FIG. 1A, semiconductor wafer 200 is similar to semiconductor wafer 100. Semiconductor wafer 200 includes a substrate 204 and device regions 203. Substrate 204 is similar to substrate 104. Device regions 203 are similar to device regions 103 and include a gate structure 209, source/drain regions 210, and isolation structures 212. Gate structure 209 is similar to gate structure 109 and includes a gate dielectric layer 206, a gate electrode 208, and possibly spacers (not shown). Gate dielectric layer 206 is similar to gate dielectric layer 106, and gate electrode 208 is similar to gate electrode 108. In addition, source/drain regions 210 in devices 203 are similar to source/drain regions 110, and isolation structures 212 in devices 203 are similar to isolation structures 112.

(21) Semiconductor wafer 200 further includes a metallization structure 222 and a bonding structure 242. Metallization structure 222 is similar to metallization structure 122 and includes a contact plug 214 embedded in a dielectric layer 207 and conductive features 224 embedded in an insulating material 226. Contact plug 214 is similar to contact plug 114, and dielectric layer 207 is similar to dielectric layer 107. Conductive features 224 are similar to conductive features 124, and insulating material 226 is similar to insulating material 126. Bonding structure 242 is similar to bonding structure 142 and includes a conductive material 244 and a polymer material 246. Conductive material 244 is similar to conductive material 144, and polymer material 246 is similar to polymer material 146. Metallization structure 222 may further include a diffusion barrier layer 243 which is similar to diffusion barrier layer 143.

(22) Before semiconductor wafer 100 is bonded to semiconductor 200, semiconductor wafers 100 and 200 are aligned, such that conductive material 144 on semiconductor wafer 100 can be bonded to conductive material 244 on semiconductor wafer 200 and polymer material 146 on semiconductor wafer 100 can be bonded to polymer material 246 on semiconductor wafer 200. In some embodiments, the alignment of semiconductor wafers 100 and 200 may be achieved by using an optical sensing method.

(23) Referring to FIG. 1B, after the alignment is performed, semiconductor wafers 100 and 200 are bonded together by hybrid bonding to form a 3DIC stacking structure 300 (die stack). Semiconductor wafers 100 and 200 are hybrid bonded together by the application of pressure and heat. In some embodiments, during hybrid bonding, stacking structure 300 is heated to a temperature in a range from about 100 C. to about 200 C., such that polymer materials 146 and 246 become a non-confined viscous liquid and are reflowed. By reflowing polymer materials 146 and 246, voids therein are eliminated.

(24) Afterwards, stacking structure 300 is further heated to a higher temperature in a range from about 220 C. to about 380 C., such that conductive materials 144 and 244 are interconnected by thermocompression bonding and polymer materials 146 and 246 are fully cured. In some embodiments, the pressure for hybrid bonding is in a range from about 0.7 bar to about 10 bar. The hybrid bonding process may be performed in an inert environment, such as an environment filled with inert gas including N.sub.2, Ar, He, or combinations thereof.

(25) As shown in FIG. 1B, hybrid bonding involves at least two types of bondings, including metal-to-metal bonding and non-metal-to-non-metal bonding. As shown in FIG. 1B, a hybrid bonding structure 150 is formed between semiconductor wafers 100 and 200. Hybrid bonding structure 150 includes conductive materials 144 and 244 bonded by metal-to-metal bonding and polymer materials 146 and 246 bonded by non-metal-to-non-metal bonding. As shown in FIG. 1B, bonding structure 150 has a metallic bonding interface 150a between conductive materials 144 and 244 but may not have a clear non-metallic interface between polymer materials 146 and 246 due to the reflowing process.

(26) Compared to hybrid bonding involving other dielectric layers, semiconductor wafers 100 and 200 are bonded through polymer materials 146 and 246. Since the bonding of polymer materials 146 and 246 involves the reflowing of polymer materials 146 and 246, voids in polymer materials 146 and 246 are eliminated and bonding strength of semiconductor wafers 100 and 200 is improved.

(27) Referring to FIG. 1C, after hybrid bonding, stacking structure 300 is put on a tape (not shown) and a thinning process 11 is performed on a bottom surface 104b of semiconductor wafer 100. After thinning process 11, TSVs 400 are exposed. Thinning process 11 may include a grinding operation and a polishing operation (such as chemical mechanical polishing (CMP)). After thinning process 11, a wet etching operation is performed to remove the defects formed on bottom surface 104b of semiconductor wafer 100. In some embodiments, bottom portion of diffusion barrier layer 420 is removed to expose conductive material 430 by thinning process 11. Therefore, after thinning process 11, height H.sub.2 is smaller than depth D.sub.1. In some embodiments, TSVs 400 have an aspect ratio (H.sub.2/W.sub.2) in a range from about 2 to about 15. In some other embodiments, height H.sub.2 is equal to depth D.sub.1. After thinning process 11, semiconductor wafer 100 has a height H.sub.2 from bottom surface 104b of semiconductor substrate 104 to a top surface of gate structure 109 in a range from about 0.2 m to about 10 m. Height H.sub.2 is smaller than height H.sub.1. In some embodiments, height H.sub.2 is in a range from about 0.01 to about 0.99 of height H.sub.1.

(28) If thinning process 11 is performed before semiconductor wafers 100 and 200 are bonded, the thin semiconductor wafer 100 is easy to break during subsequent processes. However, if semiconductor wafers 100 and 200 are bonded firstly, as shown in FIG. 1C, such bonding enables the thinning of semiconductor wafer 100. As a result, semiconductor wafer 100 could be thinned to a relatively smaller height H.sub.2 compared to semiconductor wafer 100 is directly thinned before bonding.

(29) Referring to FIG. 1D, an interconnect structure 500 is formed on bottom surface 104b of semiconductor wafer 100. Interconnect structure 500 is electrically connected to conductive features 124 over substrate 104 via TSVs 400. Interconnect structure 500 includes conductive features 510, such as conductive lines, vias, or conductive pads, formed in an insulating material 520. The metal routing of the conductive features shown in FIG. 1D is merely an example. Alternatively, other metal routing designs of conductive features may be used according to actual application.

(30) After forming interconnect structure 500, one or more redistribution layers (RDLs) (not shown) may be formed over interconnect structure 500. For example, redistribution layers (RDLs) are embedded in a passivation layer. Interconnect structure 500, redistribution layers (RDLs) and TSVs 400 provide electrical interconnection. In addition, since TSVs 400 have a relatively low resistance, RC delay is reduced.

(31) In addition, other processes may also be performed to 3DIC stacking structure 300, and 3DIC stacking structure 300 may be diced to form individual chips afterwards.

(32) FIGS. 2A-2D show cross-sectional schematic representations of various stages of forming a semiconductor device in accordance with some embodiments. Referring to FIG. 2A, a 3DIC stacking structure 700 is formed by bonding semiconductor wafers 100 and 200. As shown in FIG. 2A, semiconductor wafer 200 is almost the same with semiconductor wafer 200 except that TSVs 600 are formed in semiconductor wafer 200. In some embodiments, TSVs 400 have a diameter different from that of TSVs 600. In some embodiments, the formation of TSVs 600 involves additional patterning and etching processes. In some embodiments, number of the TSVs 400 is larger than that of TSVs 600. In some other embodiments, number of the TSVs 400 is smaller than that of TSVs 600.

(33) As shown in FIG. 2A, each TSVs 600 includes liner 610, which is similar to liner 410, diffusion barrier layer 620, which is similar to diffusion layer 420, and conductive material 630, which is similar to conductive material 430. TSVs 600 extend to metal pad 224 of semiconductor wafer 200.

(34) Because semiconductor wafer 200 will be thinned later to expose TSVs 600, TSVs 600 do not extend through the whole substrate 204 of semiconductor wafer 200. Therefore, TSVs 600 are designed to have a depth D.sub.2 which is smaller than original height H.sub.3 of semiconductor wafer 100.

(35) In some embodiments, semiconductor wafer 200 has a height H.sub.3 from a bottom surface 204b of semiconductor substrate 204 to a top surface of gate structure 209 in a range from about 17 m to about 100 m. In some embodiments, TSVs 600 have a width W.sub.3 in a range from about 0.3 m to about 10 m. In some embodiments, TSVs 600 have a depth D.sub.2 in a range from about 15 m to about 100 m. In some embodiments, TSVs 600 have an aspect ratio (D.sub.2/W.sub.3) in a range from about 5 to about 15. In some embodiments, depth D.sub.1 is the same as depth D.sub.2. In some other embodiments, depth D.sub.1 is different with depth D.sub.2.

(36) Referring to FIG. 2B, semiconductor wafer 100 is thinned to height H.sub.2 and interconnect structure 500 is formed on bottom surface 104b of semiconductor wafer 100. Interconnect structure 500 is similar to interconnect structure 500 in FIG. 1D, and detail description of interconnect structure 500 will not be repeated again for brevity.

(37) Referring to FIG. 2C, after forming interconnect structure 500, stacking structure 700 is put on a tape and a thinning process 11 is performed on a bottom surface 204b of semiconductor wafer 200. After thinning process 11, TSVs 600 are exposed. Thinning process 11 is described above and will not be repeated for brevity. After thinning process 11, semiconductor wafer 100 has a height H.sub.4 from bottom surface 204b of semiconductor substrate 204 to a top surface of gate structure 109 in a range from about 15 m to about 100 m. Height H.sub.4 is smaller than height H.sub.3. In some embodiments, height H.sub.4 is equal to or smaller than depth D.sub.2.

(38) Referring to FIG. 2D, redistribution (RDL) structure 160 is formed on bottom surface 104b of substrate 104. RDL structure 160 include a metal pad 162 formed in a passivation layer 164. Metal pad 162 is electrically connected to exposed TSVs 600. Metal pad 162 is made of conductive materials with low resistivity, such as copper (Cu), aluminum (Al), Cu alloys, Al alloys, or other applicable materials. Although FIG. 2D only shows one RDL structure 160, more than one RDL structures may be formed.

(39) An under bump metallization (UBM) layer 165 is formed on metal pad 162, and conductive element 166 (such as solder ball) is formed over UBM layer 165. UBM layer 165 may contain an adhesion layer and/or a wetting layer. In some embodiments, UBM layer 165 is made of titanium (Ti), titanium nitride (TiN), tantalum nitride (TaN), tantalum (Ta), or the like. In some embodiments, UBM layer 165 further includes a copper seed layer. In some embodiments, conductive element 166 is made of conductive materials with low resistivity, such as solder or solder alloy. Exemplary elements included in the solder alloy include Sn, Pb, Ag, Cu, Ni, Bi or combinations thereof.

(40) Interconnect structure 500 is electrically connected to another package (not shown) on the backside of semiconductor wafer 200 via TSVs 600, RDL structure 160 and conductive element 166.

(41) TSVs 400 and 600 individually provide different functions. Semiconductor wafer 100 is electrically connected to another package structure (not shown) via interconnect structure 500 and TSVs 400. Semiconductor wafer 200 is electrically connected to another package structure (not shown) via interconnect structure 500 and TSVs 600. TSVs 400 and 600 provide a fast conductive path to connect semiconductor wafer 100, semiconductor wafer 200, and/or other package structures, without the formation of complicated metal routings.

(42) Embodiments of mechanisms for forming a die stack are provided. Two semiconductor wafers are bonded together by hybrid bonding with metal-to-meal bonding and polymer-to-polymer bonding to form the die stack. TSVs are formed before the two semiconductor wafers are bonded. One or both of the semiconductor wafers are thinned to expose the TSVs after the semiconductor wafers are bonded. The bonded semiconductor wafers provides extra strength to reduce the risk of wafer breaking during a thinning process. In addition, polymer-to-polymer bonding is strong, and the polymer material is soft to provide a cushion to absorb the stress resulting from the TSVs.

(43) In some embodiments, a semiconductor device structure is provided. The semiconductor device includes a first semiconductor wafer and a second semiconductor wafer bonded via a hybrid bonding structure, and the hybrid bonding structure includes a first conductive material embedded in a first polymer material; and a second conductive material embedded in a second polymer material, and the first conductive material is bonded to the second conductive material and the first polymer material is bonded to the second polymer material. The semiconductor device also includes at least one through silicon via (TSV) extending from a bottom surface of the first semiconductor wafer to a metallization structure of the first semiconductor wafer. The semiconductor device structure also includes an interconnect structure formed over the bottom surface of the first semiconductor wafer, and the interconnect structure is electrically connected to the metallization structure via the TSV.

(44) In some embodiments, a semiconductor device structure is provided. The semiconductor device includes a first semiconductor wafer and a second semiconductor wafer bonded via a hybrid bonding structure. The semiconductor device also includes at least one first TSV extending from a bottom surface of the first semiconductor wafer to a metallization structure of the first semiconductor wafer. The semiconductor device further includes at least one second TSV extending from a bottom surface of the second semiconductor wafer to a metallization structure of the second semiconductor wafer; and a width of the first TSV is different with that of the second TSV.

(45) In some embodiments, a method for forming a semiconductor device structure is provided. The method includes providing a first semiconductor wafer and a second semiconductor wafer, and at least one through substrate via (TSV) is formed in the first semiconductor wafer, and the TSV extends from the bottom surface of the first semiconductor wafer to a metallization structure of the first semiconductor wafer. The method also includes hybrid bonding the first semiconductor wafer and the second semiconductor wafer. The method further includes thinning the first semiconductor wafer from a bottom surface of the first semiconductor wafer to expose the TSV.

(46) Although embodiments of the present disclosure and their advantages have been described in detail, it should be understood that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the disclosure as defined by the appended claims. For example, it will be readily understood by those skilled in the art that many of the features, functions, processes, and materials described herein may be varied while remaining within the scope of the present disclosure. Moreover, the scope of the present application is not intended to be limited to the particular embodiments of the process, machine, manufacture, composition of matter, means, methods and steps described in the specification. As one of ordinary skill in the art will readily appreciate from the disclosure of the present disclosure, processes, machines, manufacture, compositions of matter, means, methods, or steps, presently existing or later to be developed, that perform substantially the same function or achieve substantially the same result as the corresponding embodiments described herein may be utilized according to the present disclosure. Accordingly, the appended claims are intended to include within their scope such processes, machines, manufacture, compositions of matter, means, methods, or steps.