CMOS-based devices for harsh media

By filling metal slits in an oxide layer and covering it with a passivation layer, the problem of unstable connections in semiconductor devices in harsh media is solved, enabling the fabrication of semiconductor devices with low resistivity and high reliability, suitable for applications such as pressure sensors.

CN110610939BActive Publication Date: 2024-11-12MELEXIS TECH NV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201910522640.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2018-06-15
Filing Date
2019-06-17
Publication Date
2024-11-12
Estimated Expiration
2040-03-26

Smart Images

  • Figure CN110610939B_ABST
    Figure CN110610939B_ABST
Patent Text Reader

Abstract

The present application discloses a CMOS-based device for harsh media. A semiconductor device (100) comprises: a first doped semiconductor layer (112), a second doped semiconductor layer (122), an oxide layer (127), and an interconnection (129), the oxide layer (127) covering the first doped semiconductor layer (112) and the second doped semiconductor layer (122); the first doped semiconductor layer is electrically connected to the second doped semiconductor layer (122) by means of the interconnection (129), the interconnection (129) spans the sidewall (133) of the second doped semiconductor layer (122); the interconnection comprises a metal-filled slit (123) in the oxide layer (127); at least one electronic component (115) is formed in the first semiconductor layer (112) and / or the second semiconductor layer (122); the semiconductor device further comprises a passivation layer (128), the passivation layer (128) covering the first doped semiconductor layer (112), the second doped semiconductor layer (122) and the oxide layer (127).
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to semiconductor devices that can be deployed in harsh media. More particularly, the present invention relates to harsh media sensors including a passivation layer, and to methods including such sensors. Background Art

[0002] FIG1 shows a standard connection between polysilicon 14 and single crystal silicon 11 implemented in a CMOS process. Well doping 10 may be present in the single crystal silicon 11. Polysilicon 14 is deposited and patterned on top of field oxide 13. The sides of polysilicon 14 are covered with an insulator having a sidewall passivation 15 (such as an oxide or nitride). The single crystal silicon 11 and polysilicon 14 not covered by the field oxide are implanted simultaneously. Diffusion then occurs to form source and drain contacts 12. Metal is then deposited and dissolved into the single crystal silicon and polysilicon during an annealing step to form silicide 16. The undissolved metal is etched away while silicide 16 remains. Obtaining a first metal conductor layer 18 connected to the single crystal silicon 11 and / or polycrystalline silicon 14 is achieved by depositing a first high density plasma (HDP) oxide 19, then etching contact holes in the first HDP oxide 19 to provide silicide access to the polycrystalline silicon 14, and simultaneously etching other holes to provide silicide access to the single crystal silicon 11. The contact holes are filled with metal plugs 17, and the connections between the different plugs are made with the deposition of the first metal conductor 18. The patterning of the first metal 18 defines which contacts are connected and which contacts are not connected. Then, the process of applying another metal layer treatment can be started from depositing the second HDP oxide 20. As a result, the single crystal silicon 11 is electrically connected to the structure of the polycrystalline silicon 14 through the metal line 18 on top of the metal contact plug 17. Typically, copper or aluminum is used to form the first metal conductor 18.

[0003] The electrical connection between the polysilicon shield 14 and the single crystal silicon contact 12 shown in FIG. 1 is not suitable for use in pressure sensors operating in harsh environments.

[0004] In FIG. 1 , a second HDP oxide 20 is deposited over the first metal 18 and over the first HPD oxide 19 , and there is a PECVD nitride 21 on top of the device.

[0005] Instead of passing through the first metal layer, in principle, a single crystal silicon 11-polysilicon 14 contact can be created by sintering the edge 22 of the polysilicon 11 defined on the single crystal silicon 14. The cross section of such a contact is shown on the left side of Figure 2. During the patterning of the polysilicon 14, the gate oxide near the edge of the polysilicon in the area without field oxide is also removed. After the drain / source 12 is implanted and diffused, metal sputtering for the silicide 16 is performed, which then also covers the edge of the polysilicon. During the annealing step, the metal diffuses into the silicon, and the silicon diffuses into the metal. As a result, the silicide 16 grows on the silicon surface, so the silicide 16 grown on the single crystal silicon 12 will be connected to the silicide 16 grown at the edge of the polysilicon 14 and bridge the thin gate oxide 23, thereby forming an edge contact 22. In order to ensure that such contacts between single crystal silicon and polysilicon are not formed between the gate and source / drain diffusion of the CMOS transistor, sidewall passivation 15 is usually defined at the edge of the polysilicon 14, as described in US4622735. Typically, photolithography is not performed on the sidewall passivation 23. In this example, when the shielded piezoresistors are integrated with the CMOS transistors, the processing of the sidewall passivation 23 must be skipped so that the single crystal silicon-polysilicon contacts are etched away or must be locally etched away. Therefore, creating single crystal silicon-polysilicon contacts by creating silicide on the edge of polysilicon is a non-standard way of CMOS processing. Sputtering on the sidewall is usually not required for CMOS processing, and therefore special control is required to ensure mechanical and electrical connections at the edge of the polysilicon. Standard CMOS processes may need to be modified to ensure correct single crystal silicon-polysilicon contacts. A dedicated process control module must be developed to monitor the quality of single crystal silicon-polysilicon contacts during mass production.

[0006] If it is not possible to use copper aluminum interconnects, the alternative prior art solutions result in increased resistance of the connecting line. The following table illustrates the resistance of different types of interconnects. A top metal layer of 1 μm aluminum is very common for CMOS processing (e.g., metal 18 in Figure 1) and can be used as a benchmark for resistivity. The table shows that the sheet resistance of silicon doped with silicide has a sheet resistivity that is two orders of magnitude higher than that of aluminum:

[0007]

[0008] For harsh media, a 0.2 μm thick platinum layer is usually used, with a compromise between thickness (cost) and resistivity resulting in a resistivity that is an order of magnitude higher than that of 1 μm aluminum.

[0009] In view of this combination of good conductivity of interconnects and reliable operation in harsh media, there is still room for improvement in semiconductor devices that need to be robust to harsh media and in methods of producing these semiconductor devices. Summary of the invention

[0010] An object of embodiments of the present invention is to provide a semiconductor device that provides reliability in harsh media, and to provide a method for manufacturing such a semiconductor device.

[0011] The above objects are achieved by the method and device according to the present invention.

[0012] In a first aspect, an embodiment of the present invention relates to a semiconductor device, which includes a first doped semiconductor layer, a second doped semiconductor layer, and an oxide layer covering the first doped semiconductor layer and the second doped semiconductor layer.

[0013] The semiconductor device also includes an interconnect. A first portion of the interconnect is electrically connected to the first doped semiconductor layer, and a second portion of the interconnect is electrically connected to the second doped semiconductor layer. A portion of the interconnect between the first side and the second side spans a sidewall of the second doped semiconductor layer. The interconnect includes a metal-filled slit in the oxide layer. In an embodiment of the present invention, the metal-filled slit electrically connects the first doped semiconductor layer to the second doped semiconductor layer.

[0014] At least one electronic component is formed in the first semiconductor layer and / or the second semiconductor layer.

[0015] The semiconductor device further includes a passivation layer covering the first doped semiconductor layer and the second doped semiconductor layer, and the oxide layer.

[0016] An advantage of embodiments of the present invention is that the passivation layer provides reliability of semiconductor devices (e.g. sensors) in harsh media. An advantage of embodiments of the present invention is that all conductive paths exist below the passivation layer, as this allows avoiding leakage currents between connection paths, especially when the semiconductor device is exposed to moisture.

[0017] An advantage of embodiments of the present invention is that contact between the first doped semiconductor layer and the second doped semiconductor layer is achieved through metal-filled slots, and no metal stack is required to provide interconnection.

[0018] Advantageously, semiconductor devices according to embodiments of the present invention may be produced using standard CMOS production facilities.

[0019] In an embodiment of the present invention, the bond pad is made of a noble metal.

[0020] In an embodiment of the present invention, the semiconductor device includes a well of a second conductivity type in a deep well of the first conductivity type or in a substrate of the first conductivity type, the first conductivity type being opposite to the second conductivity type. In this case, the deep well or substrate of the first conductivity type corresponds to the first doped semiconductor layer.

[0021] At least one of the electronic components is present in a well of the second conductivity type.

[0022] A second semiconductor layer is present between the passivation layer and the electronic component, and an oxide layer is present over the second semiconductor, over the well and over the deep well or substrate of the first conductivity type.

[0023] In an embodiment of the present invention, a semiconductor device includes at least one conductive path connected between at least one electronic component and at least one through-via, the through-via passing through a passivation layer and connected to a bonding pad.

[0024] An advantage of embodiments of the present invention is that by providing a conductive path between the at least one through-via and the electronic component, the through-via and the electronic component may be spaced a distance from each other.

[0025] In an embodiment of the present invention, the semiconductor device comprises a field oxide, wherein the field oxide is present between the second doped semiconductor layer and the first doped semiconductor layer.

[0026] If a well exists in the first doped semiconductor layer, a field oxide may exist between the second doped semiconductor layer and the well.

[0027] In an embodiment of the present invention, a field oxide provides a spacer between an electronic component and the polysilicon acting as a shield on top of the electronic component, and prevents the polysilicon shield from modulating a low-doped region below the shield. In an embodiment of the present invention, the region below the field oxide has a lower doping than a region without field oxide growth. In an embodiment of the present invention, these low-doped regions are used as electronic components (e.g., varistors). An advantage of an embodiment of the present invention is that electrical insulation between the well and the polysilicon shield is provided by the field oxide.

[0028] In an embodiment of the present invention, a second semiconductor layer may exist below the bonding pad.

[0029] In an embodiment of the present invention, the passivation layer has Si 3 N 4 or diamond-like carbon or diamond or SiC or Al 2 O 3 The stoichiometric structure.

[0030] An advantage of embodiments of the present invention is that chemical and mechanical robustness is improved by providing a passivation with a stoichiometric structure.

[0031] In an embodiment of the present invention, the second doped semiconductor layer and / or the first doped semiconductor layer is at least partially covered with a metal alloy.

[0032] The second doped semiconductor layer may be, for example, a polysilicon shield, and the metal alloy may be, for example, a silicide. To avoid CTE mismatch, it is best not to have the metal alloy above the electronic component (e.g., the sensing element). Therefore, the metal alloy sometimes only partially covers the second doped semiconductor layer.

[0033] An advantage of a metal alloy is that it may result in an improved contact (lower resistivity) between the second doped semiconductor layer and the first doped semiconductor layer.

[0034] In an embodiment of the invention, the at least one electronic component is a piezo-resistor.The semiconductor devices may be configured such that they can be used as pressure sensors.

[0035] In an embodiment of the invention, the at least one conductive path comprises a highly doped path of the second conductivity type in the well.

[0036] In an embodiment of the present invention, the at least one conductive path comprises a polysilicon layer between the electronic component and the at least one through-hole.

[0037] In an embodiment of the present invention, the polysilicon layer as part of the conductive path is patterned.

[0038] An advantage of embodiments of the present invention is that the conductivity is increased by patterning the polysilicon layer. In embodiments of the present invention, the patterning results in an increased surface of the polysilicon structure perpendicular to the direction of current flow. In embodiments of the present invention, where the polysilicon structure is covered with silicide, this causes an increased amount of silicide in the direction of current flow, and thus the line conductivity can be at least twice higher than a flat polysilicon structure.

[0039] In an embodiment of the invention, at least one conductive path comprises a metal-filled slot parallel to a current flow direction when current flows in the conductive path.

[0040] In an embodiment of the present invention, the metal-filled slit is electrically connected to the first doped semiconductor layer through a highly doped contact.

[0041] In embodiments of the present invention in which the first doped semiconductor layer is a deep well of the first conductivity type, the highly doped contact may also be of the first conductivity type.

[0042] In an embodiment of the present invention, the at least one conductive path comprises a structured polycrystalline semiconductor material (e.g., silicon) covered with a metal alloy. The structured polycrystalline semiconductor material is partially transformed into a metal alloy. Thus, the metal diffuses in the polycrystalline material so that the remaining portion is covered with the metal alloy.

[0043] In an embodiment of the present invention, a metal alloy is added after the deposition of the second semiconductor layer (eg, polysilicon) and before the deposition of the oxide layer. The metal alloy may be, for example, a silicide.

[0044] An advantage of embodiments of the present invention is that by providing the metal alloy on top of the silicon as part of the path, the conductivity of the conductive path is even higher.

[0045] In a second aspect, an embodiment of the present invention relates to a method for manufacturing a semiconductor device. The method comprises:

[0046] - providing a first doped semiconductor layer and providing a second doped semiconductor layer, wherein the first doped semiconductor layer and the second doped semiconductor layer are provided such that at least one electronic component is formed in the first and / or second doped semiconductor layer,

[0047] - depositing an oxide layer over the second doped semiconductor layer and over the first doped semiconductor layer,

[0048] - removing portions of the oxide layer to obtain slits in the oxide layer above the first doped semiconductor layer, above the sidewalls of the second doped semiconductor layer, and above the second doped semiconductor layer,

[0049] - filling the slit with metal, thereby creating an electrical interconnection between the first doped semiconductor layer and the second doped semiconductor layer,

[0050] - providing a passivation layer over the first and second doped semiconductor layers and over the oxide layer.

[0051] An advantage of embodiments of the present invention is that the contact between the first doped semiconductor layer (e.g., a deep well) and the second doped semiconductor layer (e.g., a polysilicon shield) is made by making a metal-filled slit, because this allows the passivation layer to be applied at high temperatures. By increasing the temperature, dislocations and voids can be reduced and a crystalline structure can be obtained. The advantage of crystalline structures is that they etch more slowly and are therefore more robust against harsh environments than amorphous materials with the same atomic composition. In embodiments of the present invention, the passivation layer is applied using low pressure chemical vapor deposition.

[0052] In an embodiment of the present invention, the first doped semiconductor layer is a first conductivity type deep well or a first conductivity type substrate. The electronic component may exist in the deep well or in a well in the substrate.

[0053] In an embodiment of the present invention, the second doped semiconductor layer is a polysilicon shield. The shield may be provided above an electronic component without sidewall protection at the edge of the electronic component. A silicide is then formed between the edge of the polysilicon shield and the substrate in an area where there is no field oxide at the edge of the polysilicon.

[0054] In an embodiment of the present invention, the method further comprises:

[0055] - providing at least one electrically conductive path from said at least one electronic component,

[0056] - providing at least one via through the passivation layer such that the via is connected to the conductive path and to a bonding pad on the via.

[0057] In an embodiment of the invention, the conductive path is created by at least one metal-filled slot.

[0058] In an embodiment of the present invention, a metal-filled slit is created on top of the second doped semiconductor layer. It may, for example, be created in a slit in an oxide on top of the second doped semiconductor layer.

[0059] In an embodiment of the present invention, the conductive path is a doped conductive path of the second type and forms a diode with a deep well of the first type. In this case, the capacitance and leakage current are proportional to the surface of the conductive path. A trade-off between leakage current and capacitance can be made by defining the conductive path with a metal-filled slit on top of the field oxide. Subsequently, the conductivity is reduced because no conduction occurs in the silicide below the metal-filled slit. However, capacitance and leakage current are only generated in the sensing element.

[0060] Particular and preferred aspects of the invention are set out in the accompanying independent and dependent claims. Features from the dependent claims may be combined with features of the independent claims and with features of other dependent claims as appropriate and not just as explicitly set out in the claims.

[0061] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment(s) described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] FIG. 1 shows a standard connection between polysilicon and single crystal silicon implemented in a CMOS process.

[0063] FIG. 2 shows a prior art silicide contact between polysilicon and single crystal silicon.

[0064] Figure 3 Layout features of a semiconductor device according to an exemplary embodiment of the present invention are shown.

[0065] Figure 4 An exemplary embodiment of the present invention illustrating the implementation of a monocrystalline silicon-polycrystalline silicon contact is schematically shown.

[0066] Figure 5 A metal-filled slot interconnect as a low-ohmic interconnect line according to an embodiment of the present invention is shown.

[0067] Figure 6 and Figure 7 A possible implementation of a conductive path according to an embodiment of the present invention is shown.

[0068] Figure 8a and Figure 8b A conductive path including a polysilicon structure on top of a field oxide is shown in accordance with an embodiment of the present invention.

[0069] Figure 9a and Figure 9b A conductive path including tungsten-filled slots on top of a field oxide is shown in accordance with an embodiment of the present invention.

[0070] FIG. 10 shows a prior art thermocouple manufactured for an IR sensor.

[0071] Fig.11 A thermocouple for an IR sensor according to an embodiment of the present invention is shown.

[0072] Figure 12 to Figure 13 Method steps of a method according to an embodiment of the present invention are shown.

[0073] Any reference signs in the claims should not be construed as limiting the scope.The same reference signs in different drawings indicate the same or similar elements. DETAILED DESCRIPTION

[0074] The present invention will be described with respect to specific embodiments and with reference to certain drawings, but the invention is not limited thereto but only by the claims. The drawings described are only schematic and non-restrictive. In the drawings, the dimensions of some of the elements may be exaggerated and not drawn to scale for illustrative purposes. The dimensions and relative dimensions do not correspond to actual reductions to the implementation of the invention.

[0075] The terms first, second, etc. in the specification and claims are used to distinguish similar elements and not necessarily to describe a sequence in time, space, level or otherwise. It should be understood that the terms so used are interchangeable where appropriate and that the embodiments of the invention described herein are capable of operating in other sequences than described or illustrated herein.

[0076] In addition, the terms top, below and the like in the specification and claims are used for descriptive purposes and are not necessarily used to describe relative positions. It should be understood that these terms used in this way are interchangeable under appropriate circumstances, and the embodiments of the invention described herein can be operated with other orientations except those described or illustrated herein.

[0077] It should be noted that the term "comprising" used in the claims should not be interpreted as being limited to the means listed thereafter; it does not exclude other elements or steps. Therefore, the term should be interpreted as specifying the presence of the stated features, integers, steps or components, but does not exclude the presence or addition of one or more other features, integers, steps or components, or groups thereof. Therefore, the scope of the expression "a device comprising means A and B" should not be limited to devices consisting only of components A and B. For the purposes of the present invention, a device consisting only of components A and B means that the relevant components of the device are only A and B.

[0078] References throughout this specification to "one embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification do not necessarily all refer to the same embodiment, but may refer to the same embodiment. Furthermore, in one or more embodiments, as will be apparent to one of ordinary skill in the art from this disclosure, the particular features, structures, or characteristics may be combined in any suitable manner.

[0079] Similarly, it should be appreciated that in the description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, this method of disclosure should not be interpreted as reflecting an intention that the claimed invention requires more features than those expressly recited in each claim. On the contrary, as reflected in the appended claims, each inventive aspect may have fewer features than all of the features of a single preceding disclosed embodiment. Therefore, the claims appended after the detailed description are hereby expressly incorporated into the detailed description, with each claim itself representing a separate embodiment of the invention.

[0080] In addition, although some embodiments described herein include some features included in other embodiments but do not include other features included in other embodiments, the combination of features of different embodiments is intended to fall within the scope of the present invention and form different embodiments as will be understood by those skilled in the art. For example, in the appended claims, any of the claimed embodiments can be used in any combination.

[0081] In the description provided herein, numerous specific details are set forth. However, it should be understood that embodiments of the present invention may be implemented without these specific details. In other examples, well-known methods, structures, and techniques are not shown in detail to avoid obscuring the understanding of this description.

[0082] In a first aspect, embodiments of the present invention relate to a semiconductor device 100 , which includes a first doped semiconductor layer 112 , a second doped semiconductor layer 122 , an oxide layer 127 covering the first doped semiconductor layer 112 and the second doped semiconductor layer 122 , and an interconnect 129 .

[0083] The first doped semiconductor layer 11 is electrically connected to the second doped semiconductor layer 122 by means of an interconnect 129, which includes a metal-filled slit 123 in the oxide layer 127. The second doped semiconductor layer 122 partially overlaps the first doped semiconductor layer 112. A first portion of the interconnect 129 contacts the first doped semiconductor layer 112. A second portion of the interconnect contacts the second doped semiconductor layer 122. The portion of the interconnect between the first side and the second side spans a sidewall 133 of the second doped semiconductor layer. The interconnect may have multiple (e.g., alternating) contacts with the first semiconductor layer and the second semiconductor layer. Sidewall passivation 132 may be present on the sidewall 133.

[0084] In an embodiment of the present invention, at least one electronic component 115 is formed in the first semiconductor layer 112 and / or the second semiconductor layer 122 .

[0085] In addition, the semiconductor device includes a passivation layer 128 covering the first doped semiconductor layer 112 , the second doped semiconductor layer 122 , and the oxide layer 127 .

[0086] The length of the slit is so long that the metal filling the slit is in contact with the second doped semiconductor layer 122 and in contact with the first doped semiconductor layer 112 , even if there is a sidewall passivation 132 at the edge 133 of the polysilicon.

[0087] The metal in the metal filled slots may, for example, comprise tungsten. Compared to, for example, copper or aluminum interconnects, this allows the passivation layer to be applied at higher temperatures than conventional PECVD passivation layers without damaging the interconnects.

[0088] In an embodiment of the present invention, the first semiconductor layer and the second semiconductor layer may be silicon layers. The first or second semiconductor layer may be, for example, a polycrystalline silicon layer or a single crystal silicon layer.

[0089] - In an embodiment of the present invention, the semiconductor device 100 includes a well 114 of a second conductivity type (e.g., p-type) in a deep well or substrate 112 of a first conductivity type (e.g., n-type) opposite to the second conductivity type. The deep well or substrate of the first conductivity type thus corresponds to the first doped semiconductor layer. Reference to the deep well 112 is made to distinguish the well 114 from a deep well 112 that is deeper than the well 114.

[0090] In this example, at least one electronic component 115 is in the well 114 , and a second doped semiconductor layer 122 , which may be, for example, a polysilicon shield 122 , is present between the passivation layer 128 and the electronic component 115 .

[0091] In this example, the oxide layer 127 exists on the second semiconductor layer 122, on the well 114, on the deep well 112 of the first conductivity type or on the substrate of the first conductivity type. Thus, the deep well 112 of the first conductivity type or the substrate of the first conductivity type corresponds to the first doped semiconductor layer 112. The deep well 112 can be, for example, a doped single silicon layer.

[0092] For CMOS circuits, n-type deep wells 112 are created in the p-substrate 110 to have electrical isolation between different deep n-wells 112, which avoids crosstalk between sub-circuits. The deep n-well 112 can be omitted, but then an n-type substrate 110 must be used instead of a p-type substrate, and then the body of the device must be connected to the highest voltage. Typically, the body of a CMOS is p-type and electrically connected to ground. Discrete pressure sensors without CMOS circuits are typically made with an n-type substrate.

[0093] The electrical connection 123 between the first doped semiconductor layer and the second doped semiconductor layer (eg, monocrystalline silicon and polycrystalline silicon) is covered with a passivation layer 128, which provides reliability in harsh media. In an embodiment of the present invention, the passivation layer is a Si 3 N 4 , DLC, diamond, SiC or Al 2 O 3, to achieve maximum chemical and mechanical robustness. Stoichiometric passivation layers are formed at temperatures above 600 or even 800° C. and are therefore incompatible with metals commonly used in standard CMOS processing, such as platinum, aluminum, and copper. At such temperatures, aluminum melts and copper diffuses into silicon and destroys the semiconductor properties of silicon. In addition, in the case of copper metal conductors 18, the structure shown in FIG. 1 will break due to the high thermal expansion and high stiffness of copper. Silicon and silicide can be used as interconnects, but the drawback of this solution is that a lot of resistance is added to the varistor, which will reduce sensitivity and increase offset if the interconnect is not matched for all resistors. For an interconnect with a sheet resistance of 3 ohms / square, a width of 20 μm and a length of 1 mm, the varistor increases by 150 ohms. For a varistor with a typical value of 4.5 kiloohms, this will reduce sensitivity by more than 3%.

[0094] In an embodiment of the present invention, on the other hand, because the interconnect 123 between polysilicon and monocrystalline silicon is a metal-filled slit in the oxide layer, a stoichiometric passivation layer can be applied. The metal plug should be a high temperature resistant metal plug. The metal can be tungsten, for example. Tungsten has a very small expansion coefficient (5ppm / °C) and therefore does not crack the oxide providing the slit. Another advantage is that tungsten does not react with the atmosphere of silane and ammonia during the LPCVD deposition of nitrides. Ti or TiW does react and has a large thermal expansion coefficient (7-9ppm / °C). Other metals used in standard CMOS can be nickel, but due to the large CTE (>13ppm / °C), cracking will result.

[0095] By replacing the holes in the oxide layer with metal-filled slits, the metal-filled slits form interconnections between monocrystalline silicon and polycrystalline silicon, and no metal stack is required. The absence of a metal stack (including aluminum or copper) allows the passivation layer to be deposited at a higher temperature, thereby allowing the growth of a passivation layer with a stoichiometric structure. Moreover, a conductive path between at least one electronic component and at least one bonding pad can be achieved by a metal-filled slit in the oxide layer. The bonding pad is typically located at the edge of the die so that the slit covers a certain distance between the electrical component or varistor and the through hole toward the bonding pad. Preferably, the through hole is away from the varistor below the bonding pad.

[0096] By having slits instead of holes, it is not necessary to remove any sidewall passivation at the edge of the polysilicon shield to achieve contact between the polysilicon shield and the single crystal silicon. This is compatible with processes that require sidewall passivation to create CMOS transistors.

[0097] The ohmic contact between polysilicon and single crystal silicon cannot be made by simply growing polysilicon on top of single crystal silicon using standard techniques for CMOS processes (such as LPCVD) because this would result in oxidation of the single crystal silicon during loading of the wafer with exposed single crystal silicon into the deposition tool. It is also advantageous not to create silicide on the silicon before depositing the polysilicon because the sputtering process used to make silicide would degrade the oxide used for the gates of the CMOS transistors or even completely etch away the oxide.

[0098] In an embodiment of the present invention, the first conductivity type is n-type and the second conductivity type is p-type. In an embodiment of the present invention, the at least one electronic component is a varistor.

[0099] In an embodiment of the present invention, the polysilicon shield 122 over the electronic component 115 includes dopants of the first conductivity type.

[0100] Figure 3 Layout features of a semiconductor device according to an exemplary embodiment of the present invention are shown.

[0101] The upper left schematic 310 shows an intermediate processing stack, in which the polysilicon shield 122 and the contact well 116 are shown. In this stack, there is no silicide on the polysilicon shield 122 above the electronic component. In an embodiment of the present invention, the polysilicon shield can be doped to ensure conductivity sufficient to shield the electronic component. It may be advantageous not to add silicide to the polysilicon shield because the silicide may introduce a CTE (coefficient of thermal expansion) mismatch between the polysilicon shield and the area of ​​the electronic component below. In an embodiment of the present invention, silicide is present on the contact area (e.g., on the polysilicon where contact is made to the metal slit) to make proper electrical contact with the substrate. Preferably, there is no silicide above the electronic component to avoid undesirable stress in the polysilicon above the electronic component (e.g., the piezoresistive area).

[0102] In this example, all polysilicon is removed from the interconnect path and bond pad area. Because metal filled slots are used for the interconnect, polysilicon is not required.

[0103] Schematic 320 shows interconnects 129 including metal-filled slots 123 that provide substrate contact for the polysilicon shield. The schematic also shows conductive paths 121 of electronic components 115 (eg, varistors). In this example, conductive paths 121 include metal-filled slots 123.

[0104] Schematic 330 shows the bond pad area in the upper right, where the dark squares indicate bond pad metal 130 on top of stoichiometric passivation layer 128. A hole etched in the passivation layer (not shown) provides electrical connection from the noble metal top metal to the metal-filled slot below the passivation layer.

[0105] Section A shows, from left to right, a conductive path 121 with a metal-filled slot 123 on a substrate contact 116 of a second conductivity type (e.g., a p-type source / drain diffusion), an electronic component 115 (e.g., a varistor) defined by a well diffusion 114 of the second conductivity type and isolated from a polysilicon shield 122 using a field oxide 120 (the polysilicon shield 122, the field oxide 120, and the electronic component together form a shielded electronic component). Further to the right, a conductive path 121 (including silicide 124) between the electronic component 115 and a bonding pad 130 is shown. For the bonding pad, a noble metal top metal 130 is deposited on top of a stoichiometric passivation layer 128 and connected to the metal-filled slot 123 through a via 131 in the passivation layer. Preferably, the via is made away from the electronic component below the bonding pad.

[0106] Section B shows a cross section of an interconnect that includes a metal-filled slit 123 that is vertically placed across an array of polysilicon structures 122 and provides electrical contact between the single crystal silicon 112 and the polysilicon 122, but the polysilicon sidewalls 133 are isolated by the polysilicon sidewall passivation 132. The polysilicon structures 122 and the deep well 112 include dopants of the first conductivity type. The metal-filled slit 123 connects the deep well contact 125 to the polysilicon shield 122, and the deep well contact 125 is doped with dopants of the first conductivity type (e.g., n++ contacts). The metal-filled slit spans the sidewall 133 of the second semiconductor layer 122. It may be in contact with the sidewall 133. In this example, a sidewall passivation is present between the sidewall and the metal-filled slit.

[0107] Field oxide 120 may, for example, have a thickness of 450 + / - 200 nanometers. Polysilicon 122 may, for example, have a thickness of 400 nanometers + / - 150 nanometers, and HDP oxide 127 may, for example, have a thickness of 800 nanometers + / - 400 nanometers.

[0108] An advantage of embodiments of the present invention is that electronic components (e.g., varistors) can be created that are connected to a substrate and interconnected under a stoichiometric passivation layer. Thus, semiconductor devices that are protected from harsh environments can be created. An advantage of embodiments of the present invention is that the interconnection has a resistance that is not significantly higher than the resistance of a conventional interconnection. The semiconductor device can be, for example, a sensor. An increase in the interconnection resistance will result in a decrease in the sensitivity of the sensor.

[0109] An advantage of embodiments of the present invention is that it is not required to remove the sidewall passivation of polysilicon to obtain interconnection between polysilicon and single crystal silicon. Because it is not required to remove the sidewall passivation, co-integration with CMOS transistors is possible.

[0110] In an embodiment of the present invention, a highly conductive path 121 connects at least one electronic component 115 with at least one bonding pad 130 on top of the passivation. This connection between the conductive path 121 and the bonding pad can be achieved by means of a via 131 passing through the passivation layer 128. In an embodiment of the present invention, the bonding pad is made of a noble metal.

[0111] An advantage of embodiments of the present invention is that such a stack can be realized using a process flow comprising standard CMOS processing steps. Thus, when a stoichiometric passivation layer is applied, most of the flow can be performed in any CMOS wafer fabrication up to the deposition of the passivation layer.

[0112] Standard CMOS provides a (low) doped region 115 below the field oxide 120 and a (higher) doped region 116, where no field oxide grows. The surface of these latter regions can become silicide 124. In an embodiment of the invention, the (low) doped region 115 below the field oxide 120 is used as an electronic component 115 (e.g. a varistor).

[0113] In an embodiment of the present invention, the second conductivity type is p-type. By using an n-type wafer, the p-well region 115 or the varistors can be electrically isolated from each other. However, CMOS is usually implemented in p-type silicon. Therefore, before the p-well, a so-called deep n-well diffusion can be defined to provide electrical isolation between the varistors 115. Another way to isolate the resistors from each other is to use an SOI wafer with oxide-filled trenches around the resistors that reach down to the buried oxide. Such deep n-well diffusions can be defined for the entire wafer without using implanted lithography. In an embodiment of the present invention, the low-doped well 114 is made more conductive by adding a source-drain diffusion 116 with silicide, and the low-doped well 114 is used as a conductive path 121 between at least one electronic component 115 and the through-hole 131. In CMOS, such diffusions 116 are usually referred to as source / drain regions. The low doped well 114 may be, for example, a p-well 114 with a sheet resistance between 2000-5000 ohm / square, and by adding implants 116 a sheet resistance of 100-150 ohm / # without silicide and 3 ohm / # with silicide may be achieved.

[0114] An advantage of the embodiments of the present invention is that the polysilicon shield 122 can be manufactured using the same process as the gate of a CMOS transistor.

[0115] In an embodiment of the present invention, field oxide 120 provides a spacer between electronic component 115 and the polysilicon shield. The spacer prevents the shield from modulating the lowly doped region beneath the shield.

[0116] In an embodiment of the present invention, contact holes 131 are etched in the passivation layer 128 so that the bonding pads 130 on top of the passivation layer 128 are connected to the conductive paths 121 below the passivation layer 128. Preferably, the bonding pads 130 are implemented with a noble metal so that only the passivation layer 128 and the noble metal bonding pads are exposed to the medium above the sensor.

[0117] The application of passivation layers and precious metals can be done using specialized equipment in dedicated clean rooms to avoid contamination from the CMOS.

[0118] Figure 4 An exemplary embodiment of the present invention is schematically shown to illustrate the implementation of the single crystal silicon 112-polycrystalline silicon 122 contact. Figure 4 In FIG. 1 , interconnect 129 includes a metal-filled slot 123. The size of the slot allows the polysilicon shield 122 and the substrate 112 contact to be interconnected using a metal plug.

[0119] The width of the slit is, for example, between 0.8 and 1.2 times the critical dimension or even between 0.5 and 3 times (the critical dimension is the diameter of the CMOS contact plug). In an embodiment of the present invention, the width of the slit is about twice the thickness of the sputtered metal. The metal is deposited not only at the bottom of the slit, but also on the sidewalls. When the slit is too deep, the deposition at the side of the slit will close the slit at the top before the bottom of the slit is filled. When the slit is too wide, the slit will not be filled when the thickness of the metal is less than the depth of the slit. Therefore, the width of the slit is optimized for a specific technology as a function of the depth of the slit and the thickness of the sputtered layer. In an embodiment of the present invention, the depth of the slit can be greater than the width. The depth and width can be preferably selected so that the slit is filled with metal without a void at the bottom or a depression at the top.

[0120] The length of the slit may be, for example, between 1.5 and 3 times the critical dimension, or even between 3 and 10 times. The length of the slit may be, for example, at least 3 times, or even 10 times, or even 100 times, or even 5000 times longer than the slit width. Thus, the length of the metal-filled slit is measured from one outer end of the metal to the opposite outer end of the metal, the measured dimension spanning the first semiconductor layer and the second semiconductor layer.

[0121] Figure 4 The top of 124 shows how such a slit 123 filled with metal can be used as a contact between the single crystal silicon 112 and the polycrystalline silicon 122. In this example, the slit is defined perpendicular to the edge of the polycrystalline silicon. Figure 4 The bottom of shows details of some possibilities of such slits 123. Figure 4 The well doping in the contact region 125 has a first conductivity type (n-doping), and in the contact region 125, the conductivity is enhanced by a source / drain implant of the first conductivity type (also used to implant the polysilicon layer). After the implant diffusion, silicide 124 is formed simultaneously on the polysilicon 122 and the monocrystalline silicon 125. The figure shows that a slit filled with tungsten can connect the monocrystalline silicon to the polysilicon even when there are still sidewall passivation structures 132 or residues 134. This is possible due to the length of the slit.

[0122] exist Figure 4 In the embodiment of the present invention, after structuring the polysilicon 122 in a comb-like shape, an interconnection 129 between the monocrystalline silicon and the polysilicon is realized. The metal-filled slits 123 are present orthogonally to the comb teeth and provide contact between the polysilicon shield and the monocrystalline silicon. Thereafter, the interconnection and the polysilicon are covered with a passivation layer. This is preferably a stoichiometric passivation layer. Figure 4 Also shown are a first HDP oxide 127 a in which metal-filled slots are formed and a second HDP oxide 127 b located on top of the first HDP oxide and on top of the metal-filled slots.

[0123] In embodiments of the present invention, metal-filled slots may be provided as low-ohmic interconnects between electronic components and bond pads, even without forming monocrystalline silicon-polycrystalline silicon contacts.

[0124] Figure 5 Metal (e.g., tungsten) filled slit interconnects 129 as low ohmic interconnects according to an embodiment of the present invention are shown. These interconnects 129 are shown in the middle of the left picture. On the left side of the left picture, a conductive path 121 comprising a metal filled slit 123a is shown, which is defined in a path where no field oxide is grown. The width of such a path can be, for example, in the range of 1 μm to 40 μm, and can be as long as the length of the die or even longer.

[0125] In an embodiment of the present invention, the surface of the single crystal silicon is doped with source / drain implants to obtain a highly doped path 116 and covered with silicide 124. An array of metal-filled slits 123a is placed on top of the silicide 124, and these slits are implemented over the entire length of the path aligned in the direction of current flow. These metal-filled slits are part of a conductive path 121 between at least one electronic component and at least one bonding pad 130. The conductive path 121 is electrically isolated from the rest of the chip by the source / drain diffusion 116 of silicon. Therefore, the diffusion must have a doping opposite to the doping of the silicon surrounding the diffusion. For an n-type single silicon deep well 112, the source drain doping must be p-type. The voltage of the substrate must be maintained at a voltage that keeps the pn junction between the conductive path 121 and the body 112 reverse biased in order to obtain isolation. It is also possible to consider defining a p-type well over the path length, which will increase the breakdown voltage between the line and the substrate, but also increase the capacitance between the line and the body. Another effect is that there will be a small leakage current of the order of 10 to 100 picoamperes between these conductive paths 121 and the body at room temperature. Such current will not substantially affect the performance of a Wheatstone bridge with a resistance of up to 100 kilo-ohms. (Note that such leakage currents are always present for varistors or other electrical components utilizing the second type of well implants.)

[0126] A similar conductive path 121 is shown at the right side of the left picture, where a metal-filled slit is placed on a polysilicon shield 122 on a field oxide 120. The conductive path includes a metal-filled slit 123b. The drawback of this conduction path is that the slit is less deep than the slit on single crystal silicon, and therefore the resistance of such a line will be higher. However, the isolation of the line toward the body 112 is provided by the field oxide 120, rather than by the pn junction. Therefore, the leakage current of such a wire toward the body can be completely ignored. The breakdown voltage is not improved because the end of such a line is always in contact with an electronic component 115, and the electronic component 115 in the well 114 always has a breakdown voltage.

[0127] Figure 5The right picture in shows four different options by defining the metal-filled slits as conductive paths according to embodiments of the present invention. The left portion 400 shows the metal-filled slits placed on the single crystal silicon 116. This will result in the lowest resistivity of the path, but will also slightly increase the leakage current to the body. Section 410 shows a similar conductive path with polysilicon 122 between the metal-filled slits 123 and the single crystal silicon 116, resulting in a low-ohmic connection between the single crystal silicon 116 and the polysilicon 122 at the left edge of the polysilicon 122. This line will have a higher resistance than the metal-filled slit array on the single crystal silicon 116, but it will show less leakage current due to the presence of a native oxide between the polysilicon 122 and the single crystal silicon 116. The cross section 420 shows that the height of these metal-filled slits 123 is the smallest when there is polysilicon 122 between the metal-filled slits 123 and the field oxide 120. This will result in the smallest conductive path. However, the polysilicon acts as an etch stop for the list etch and makes this etch step less critical. Cross section 430 shows completely isolated conductive paths 121 on top of field oxide 120. These paths will have less good conductivity than the paths in cross section 300 at the left, but without leakage current and depletion capacitance. Therefore, this solution would be preferred for high temperature applications where leakage current must be minimized.

[0128] The following table shows a comparison of the resistance of conductive paths according to an embodiment of the present invention. The resistance of an array of polysilicon lines with a width of 10 μm and a length of 1 mm covered with silicide will be about 1.5 ohms, while a single tungsten line with a length of 1 mm will give a similar resistance of 1.32 ohms. However, a tungsten filled slot array with a width of 10 μm and a length of 1 mm will only be 0.11 ohms, about 8 times smaller than a conventional platinum interconnect with a thickness of 0.2 μm and a width of 10 μm.

[0129]

[0130] Figure 6 A possible implementation of the conductive path 121 according to an embodiment of the invention is shown. In this example, the channel is defined by the absence of field oxide 120, and implants and silicide are provided on the single crystal silicon 116 between the areas of field oxide 120. This is Figure 6 The cross section shows that the conductivity of the source / drain diffusions is enhanced by the silicide process, which results in a flat surface. The sheet resistance of such a surface is about 3 ohms / square, which is 100 times higher than the sheet resistance of a 1 μm thick aluminum layer.

[0131] In a preferred embodiment, the conductive path 121 comprises a thin line of polysilicon 122 aligned with the current direction. The second cross section shows that in this way, a large amount of silicon surface is created perpendicular to the current direction. As a result, more silicide 124 is available in the current direction, and therefore the line conductivity can be at least twice as high as the conductivity of the option on the right.

[0132] exist Figure 6 1 shows a conductive path at the bottom of the device, which includes a conductive line at the far right, which consists of a polysilicon line 122b on top of a field oxide 120, which is doped and covered with a silicide 124. The advantage of this configuration is that the electrical insulation to the substrate is provided by the field oxide rather than by the pn junction. This will significantly reduce the leakage current of the device. Moreover, polysilicon lines can be used here, but then there is no conduction between the polysilicon lines.

[0133] The least interesting option is indicated to the right from center, where unpatterned polysilicon 122a provides conduction in the region where field oxide 120 is removed. With the same diffusion and silicide 124, polysilicon will always have a higher sheet resistance than single crystal silicon.

[0134] Figure 7 An exemplary implementation of a conductive path 121 connected between at least one electronic component 115 and at least one via 131 through the passivation layer 128 connected to a bond pad 130 is shown.

[0135] In this example, region 510 shows the layout of a well 114 of the second conductivity type on a deep well 112 of the first conductivity type.

[0136] Region 520 shows the area in which field oxide 120 is grown. This region covers two rectangles, each rectangle indicating an electronic component 115 (eg, a varistor) to the bottom left of the region.

[0137] Region 530 shows where polysilicon exists as a shield 122 and where polysilicon exists to enhance conductivity in the path defined by the field oxide. The shield 122 and its contact to the substrate are of a first conductivity type (e.g., n-doped), while the bond pads and conductive paths are of a second conductivity type (e.g., p-doped).

[0138] Area 540 is similar to Figure 3 Region 320, region 540 includes the option of omitting silicide on the shield over the varistor.

[0139] Area 550 shows contact hole 131 in passivation layer 128 over the polysilicon in the bond pad area on the top right, with the dark squares indicating bond pad metal.

[0140] Section A shows from left to right a conductive path 121 covered with silicide 124, an electronic component 115 (e.g., a varistor) defined by a diffusion of a second conductivity type and isolated from a polysilicon shield 122 by a field oxide 120, the conductive path 121 comprising a polysilicon line covered with silicide between the electronic component 115 and a bonding pad 130, wherein the depth of the through hole through the passivation layer is minimized by the presence of the field oxide and the polysilicon.

[0141] Section B shows from left to right an electronic component 115 defined by diffusion of the second conductivity type and isolated from a polysilicon shield 122 with a field oxide 120, and a connection to the body 112 with doping of the first conductivity type and silicide 124. The polysilicon shield 122 has no silicide over the electronic component to ensure optimal matching of the expansion coefficients.

[0142] Figure 8a 1 shows a conductive path including a polysilicon structure 126 on top of a field oxide 120 according to an embodiment of the present invention. The polysilicon structure 126 is in contact with the p+ diffusion 116. The silicide 124 covers the polysilicon structure 126 and the p+ diffusion 116. The polysilicon structure 126 is p-doped polysilicon. Figure 7 Compared to the conductive path shown in FIG, lower leakage current and depletion capacitance can be obtained at the expense of increased resistivity.

[0143] Figure 8b Shown with Figure 8a Same configuration but now with a connection 140 between the polysilicon shield 122 and the diffusion 125 of the second conductivity type in the body, where 119 represents the sensing element and 150 represents the conductive path. Figure 9a A conductive path including a tungsten-filled slot 123 on top of a field oxide 120 is shown according to an embodiment of the present invention. The conductive path includes a p-doped polysilicon structure 126 on top of the field oxide 120, a silicide 124 on top of the polysilicon structure, and a tungsten-filled slot 123 on top of the silicide. Figure 3 Compared to the conductive path shown in FIG. , lower leakage current and depletion capacitance can be obtained at the expense of increased resistivity. In addition, it is advantageous that a more robust process can be used to obtain Figure 9a structure.

[0144] Figure 9b Shown with Figure 9a Same configuration but now with a connection 140 between the polysilicon shield 122 and the diffusion 125 of the second conductivity type in the body, where 119 represents the sensing element and 150 represents the conductive path.

[0145] In the previous example, the electronic component is a varistor. However, the present invention is not limited thereto and can be applied to various other electronic components. The present invention can also be applied, for example, in the field of IR sensors to connect p-type polysilicon to n-type polysilicon on a thin film to make a thermocouple.

[0146] An example of such a prior art thermocouple device is illustrated in FIG. 10 . It shows a silicon wafer 840 with an etched membrane. A field oxide 850 covers the silicon wafer. The first doped polysilicon layer 112 is covered with an oxide 830 grown on the first doped polysilicon layer 112, and the second doped polysilicon layer 122 covers the first doped polysilicon layer 112. Metal contacts 810 are connected to the first and second doped polysilicon layers using tungsten plugs 820. Aluminum bridges between the metal contacts provide contact between the first and second polysilicon layers. These aluminum bridges cause stress, reflections, and heat leakage.

[0147] This is Fig.11 can be solved in semiconductor devices. Fig.11 A schematic diagram of a thermocouple device according to an embodiment of the present invention is shown. In this example, the first HPD oxide 127a covers the first and second doped semiconductor layers (polysilicon). The metal-filled slots 123 in the first high-density plasma (HDP) oxide interconnect the first doped semiconductor layer 112 with the second doped semiconductor layer 122. In this example, the second HDP oxide 127b covers the first HDP oxide. Thus, an advantage of embodiments of the present invention is that the absence of metal on the film prevents surface stress (cracking).

[0148] In a second aspect, embodiments of the present invention relate to a method 200 for manufacturing a semiconductor device. Fig.12 An example of this approach is shown in Figure 2. Figure 3 , Figure 9a , Figure 9b and Fig.11 The method includes providing 210 a first doped semiconductor layer 112 and providing 220 a second doped semiconductor layer 122, wherein the first doped semiconductor layer and the second doped semiconductor layer are provided such that at least one electronic component is formed in the first and / or second doped semiconductor layer.

[0149] The method also includes depositing 230 an oxide layer 127 over the second doped semiconductor layer 122 and over the first doped semiconductor layer 112 .

[0150] Furthermore, the method comprises removing 240 portions of the oxide layer 127 to obtain a slit 123 in the oxide layer above the first doped semiconductor layer 112 and above the second doped semiconductor layer 122 , and filling 250 the slit with metal, thereby creating an electrical interconnect 129 between the first doped semiconductor layer 112 and the second doped semiconductor layer 122 .

[0151] In an embodiment of the present invention, a passivation layer 128 is provided 260 over the first doped semiconductor layer 112 and the second doped semiconductor layer 122 and over the oxide layer 127 .

[0152] In an embodiment of the present invention, providing a first semiconductor layer includes providing a substrate 110, which includes a well 114 of a second conductivity type in a deep well 112 (first doped semiconductor layer) of a first conductivity type opposite to the second conductivity type, wherein at least one electronic component is provided in the well 114.

[0153] Providing the second semiconductor layer 122 includes applying 220 a polysilicon shield 122 over the electronic component 115 .

[0154] In an embodiment of the present invention, this step may be followed by a step in which source drain implants and silicide are formed simultaneously on the polysilicon and the single crystal silicon.

[0155] An oxide layer 127 may be deposited 230 over the polysilicon shield 122 and over the deep well 112 , and portions of the oxide layer 127 are removed 240 to obtain a slot 123 in the oxide layer over the deep well 112 and the polysilicon shield 122 .

[0156] The slot is filled 250 with metal, creating an electrical interconnect 129 between the polysilicon shield and the deep well.

[0157] A passivation layer 128 is provided 260 over the well 114 and the conductive path 115 .

[0158] The oxide layer 127 may be, for example, a high density plasma (HDP) chemical vapor deposited oxide.

[0159] An advantage of embodiments of the present invention is that these steps can be applied using standard CMOS processing, followed by some special post-processing steps to achieve reliability in harsh media.

[0160] Steps 210 to 250 can be implemented on a wafer of the second conductivity type using CMOS steps, depending on the definition of the contact with the metal plug for the first interconnect metal layer. The deep well diffusion of the first conductivity type can provide electrical insulation of the electronic components, and the diffusion of the second conductivity type can define the electronic components. The field oxide isolates the polysilicon shield from the electronic components, and the source drain implant can be used to create a substrate contact and to connect the electronic components to each other. Standard silicide processes can provide contact areas and areas with high conductivity.

[0161] The HDP oxide may provide first metal isolation, and a damascene process is performed to define metal-filled slots 123 in the HDP oxide that form interconnects between the second semiconductor layer 122 (e.g., polysilicon shield) and the first semiconductor layer 112 (e.g., through-hole base contact 116). These interconnects are also referred to as metal plugs.

[0162] The advantage of this method is that standard CMOS processing including polysilicon sidewall passivation can be used. In an embodiment of the present invention, the plug is made of a metal that has the property of remaining inert during the deposition of a passivation layer that can be applied even at high temperatures. The metal can be, for example, tungsten. The passivation layer can even be an LPCVD nitride applied at a temperature of 830°C. Metals such as titanium and aluminum will react with the deposition gases silane and ammonia. Applying such heating to a CMOS circuit with copper as the first metallization to a temperature above 830°C will result in copper contamination in the p and n diffusion of the circuit below the copper, and subsequently the device will no longer work. Another advantage of a high temperature resistant metal plug (e.g. tungsten) is that it has a very small thermal expansion coefficient (4.5ppm / °C) similar to that of silicon (2.6ppm / °C). Therefore, such a material does not introduce stress gradients in the material around the metal as in the case of aluminum (CTE 24ppm / °C) or copper (CTE 17ppm / °C).

[0163] In step 260, a passivation layer is provided. This is achieved by high temperature deposition 260 of the passivation layer to ensure a stoichiometric structure with minimal dislocations and voids. This layer is essential to provide reliability in harsh media. In principle, LPCVD nitride deposition tools can be used, which also define the nitride used to mask the growth of the field oxide.

[0164] This step may be followed by a non-CMOS step of patterning the passivation layer to provide bond pads to the interconnect vias and to provide definition of bond pads covering these vias.Preferably, these bond pads are made of a noble metal to withstand harsh environments.

[0165] The advantage is that it is not necessary to develop a dedicated process control module to monitor the quality of the monocrystalline silicon-polycrystalline silicon contacts during mass production.

[0166] Fig.13 The figure illustrates exemplary method steps of a method for providing an alternative interconnection (in addition to the metal-filled slits according to an embodiment of the present invention) between a first doped semiconductor layer and a second doped semiconductor layer. The first box represents a standard CMOS step on a wafer of the second conductivity type. The deep diffusion of the first conductivity type provides electrical insulation of the electronic component 115, and the diffusion of the second conductivity type defines the electronic component. Field oxide 120 isolates the polysilicon shield from the electronic component 115, and source drain implants 116 can be used to create substrate contacts and are used to connect the electronic components 115 to each other. Figure 7 , Figure 8a and Figure 8b An example of a device fabricated using this method is shown in FIG.

[0167] In an embodiment of the present invention, a silicide contact may be provided between the monocrystalline silicon and the polycrystalline silicon. This is achieved as indicated in the second box. The second box indicates that, in this example, the CMOS process needs to be modified in such a way that there is no sidewall protection before the silicide process begins. It may be considered to omit the gate oxide process to minimize the oxide thickness between the monocrystalline silicon and the polycrystalline silicon. Fig.12 The method illustrated in can be better integrated with the process for creating MOS transistors because in this method, it is not required to skip the sidewall protection. The MOS transistor can be created on the same chip as the pressure sensor sensing element.

[0168] The third box represents the standard CMOS steps for forming silicide 124 and HDP oxide. For the silicide process, silicide is created between the single crystal silicon and the polysilicon at the edges of the polysilicon defined in areas without field oxide. The HDP oxide surrounds the edges of the polysilicon and planarizes the surface. The use of chemical mechanical polishing (CMP) allows for the manufacture of a completely flat surface, but is not a necessary step.

[0169] The fourth box represents the high temperature deposition of the passivation layer 128 to ensure a stoichiometric structure with minimal dislocations and voids. This layer is preferred for providing reliability in harsh media. In principle, an LPCVD nitride deposition tool can be used, which also defines the nitride used to mask the growth of the field oxide.

[0170] The last box represents the defined (non-CMOS) step of patterning the passivation layer 128 to provide bond pads 130 to interconnect vias 131 and to provide bond pads 130 covering these vias 131. Preferably, these bond pads 130 are made of a noble metal to withstand harsh environments. Figure 7 An example of a stack-up achieved using this method is shown in the figure.

Claims

1. A semiconductor device (100), include: A first doped semiconductor layer (112), a second doped semiconductor layer (122), an oxide layer (127), and an interconnection (129), wherein the oxide layer (127) covers the first doped semiconductor layer (112) and the second doped semiconductor layer (122), wherein a first portion of the interconnect is electrically connected to the first doped semiconductor layer (112), and a second portion of the interconnect is electrically connected to the second doped semiconductor layer (122), and wherein a portion of the interconnect between the first portion and the second portion spans a sidewall (133) of the second doped semiconductor layer (122), wherein the interconnect comprises a metal-filled slit (123) in the oxide layer (127), and wherein at least one electronic component (115) is formed in the first doped semiconductor layer (112) and / or the second doped semiconductor layer (122), The semiconductor device further comprises a stoichiometric crystalline passivation layer (128), the stoichiometric crystalline passivation layer (128) covering the first doped semiconductor layer (112), the second doped semiconductor layer (122), the oxide layer (127), and the interconnection (129).

2. The semiconductor device (100) according to claim 1, include: a deep well (112) of a first conductivity type or a well (114) of a second conductivity type in a substrate of the first conductivity type, the first conductivity type being opposite to the second conductivity type, wherein the deep well (112) of the first conductivity type or the substrate is implemented in the first doped semiconductor layer (112), wherein at least one of the electronic components (115) is present in the well (114) of the second conductivity type, The second doped semiconductor layer (122) is located between the passivation layer (128) and the electronic component (115). And wherein the oxide layer (127) exists above the second doped semiconductor layer (122), above the well (114), and above the deep well (112) or substrate of the first conductivity type.

3. The semiconductor device (100) according to claim 2, comprising at least one conductive path (121), the conductive path (121) being connected between the at least one electronic component (115) and at least one through-hole (131), the at least one through-hole (131) passing through the passivation layer (128) and connected to the bonding pad (130).

4. The semiconductor device (100) according to claim 1, comprising a field oxide (120), in, The field oxide (120) exists between the second doped semiconductor layer (122) and the first doped semiconductor layer (112).

5. The semiconductor device (100) according to claim 1, in, The stoichiometric passivation layer (128) comprises Si 3 N 4 or diamond-like carbon or diamond or SiC or Al 2 O 3 .

6. The semiconductor device (100) according to claim 1, in, The second doped semiconductor layer (122) and / or the first doped semiconductor layer (112) are at least partially covered with a metal alloy (124).

7. The semiconductor device (100) according to claim 3, in, The at least one conductive path (121) comprises a highly doped path (116) of the second conductivity type in the well (114).

8. The semiconductor device (100) according to claim 3, in, The at least one conductive path (121) includes a polysilicon layer between the electronic component (115) and the at least one through-hole (131).

9. The semiconductor device (100) according to claim 8, in, The polysilicon layer as part of the conductive path (121) is patterned.

10. The semiconductor device (100) according to claim 3, in, The at least one conductive path (121) includes a metal-filled slit (123) parallel to the direction of current when the current flows in the conductive path.

11. The semiconductor device (100) according to claim 1, in, The metal-filled slit (123) is electrically connected to the first doped semiconductor layer (112) via a highly doped contact (125).

12. The semiconductor device (100) according to claim 3, in, The at least one conductive path (121) comprises a structured polycrystalline semiconductor material covered with a metal alloy (124).

13. A method (200) for manufacturing a semiconductor device, the method include: providing (210) a first doped semiconductor layer (112), providing (220) a second doped semiconductor layer (122), wherein the first doped semiconductor layer and the second doped semiconductor layer are provided such that at least one electronic component is formed in the first doped semiconductor layer and / or the second doped semiconductor layer, depositing (230) an oxide layer (127) over the second doped semiconductor layer (122) and over the first doped semiconductor layer (112), removing (240) portions of the oxide layer (127) to obtain slits (123) in the oxide layer above the first doped semiconductor layer (112), above the sidewalls (133) of the second doped semiconductor layer (122), and above the second doped semiconductor layer (122), filling (250) the slit with metal, thereby creating an electrical interconnect (129) between the first doped semiconductor layer (112) and the second doped semiconductor layer (122), A stoichiometric crystalline passivation layer (128) is provided (260) over the first doped semiconductor layer (112) and the second doped semiconductor layer (122), over the oxide layer (127), and over the electrical interconnect (129).

14. The method according to claim 13, further comprising: include: providing at least one conductive path (121) from the at least one electronic component (115), At least one through-hole (131) is provided through the passivation layer (128) such that the through-hole is connected to the conductive path (121) and to a bonding pad on the through-hole.

15. The method according to claim 14, in, The conductive path is created by at least one metal-filled slot.