International Business Machines Corporation

Stmicroelectronics Gmbh

Globalfoundries Inc.

Globalfoundries Singapore Pte. Ltd.

Richard Stephen Wise

Lawrence Alfred Clevenger

Carl John Radens

John H Zhang

Yiheng Xu

Kaushik Arun Kumar

Michael Ray Sievers

Andres Fernando Munoz

Kangguo Cheng

Bruce Bennett Doris

Hsueh-Chung Chen

Yann Mignot

Terry Allen Spooner

Nicole A Saulnier

William K Henson

Hongwen Yan

Xiang Hu

Xi Li

Habib Hichri

Catherine B Labelle

Yannick Loquet

André P Labonté

Effendi Leobandung

Ying Li

Brett H Engel

Akil Khamisi Sutton

Ultra-low-k dielectric materials used as inter-layer dielectrics in high-performance integrated circuits are prone to be structurally unstable. The Young's modulus of such materials is decreased, resulting in porosity, poor film strength, cracking, and voids. An alternative dual damascene interconnect process incorporates air gaps into a high modulus dielectric material to maintain structural stability while reducing capacitance between adjacent nanowires. Incorporation of an air gap having k=1.0 compensates for the use of a higher modulus film having a dielectric constant greater than the typical ultra-low-k (ULK) dielectric value of about 2.2. The higher modulus film containing the air gap is used as an insulator between adjacent metal lines, while a ULK film is retained to insulate vias. The dielectric layer between two adjacent metal lines thus forms a ULK/high-modulus dielectric bi-layer.

Advanced interconnect with air gap

An integrated circuit includes a substrate with an interlevel dielectric layer positioned above the substrate. First trenches having a first depth are formed in the interlevel dielectric layer and a metal material fills the first trenches to form first interconnection lines. Second trenches having a second depth are also formed in the interlevel dielectric layer and filled with a metal material to form second interconnection lines. The first and second interconnection lines have a substantially equal pitch, which in a preferred implementation is a sub-lithographic pitch, and different resistivities due to the difference in trench depth. The first and second trenches are formed with an etching process through a hard mask having corresponding first and second openings of different depths. A sidewall image transfer process is used to define sub-lithographic structures for forming the first and second openings in the hard mask.

Trench structure for high performance interconnection lines of different resistivity and method of making same

A wavy line interconnect structure that accommodates small metal lines and large vias is disclosed. A lithography mask design used to pattern metal line trenches uses optical proximity correction (OPC) techniques to approximate wavy lines using rectangular opaque features. The large vias can be formed using a self-aligned dual damascene process without the need for a separate via lithography mask. Instead, a sacrificial layer allows etching of an underlying thick dielectric block, while protecting narrow features of the trenches that correspond to the metal line interconnects. The resulting vias have an aspect ratio that is relatively easy to fill, while the larger via footprint provides low via resistance. By lifting the shrink constraint for vias, thereby allowing the via footprint to exceed the minimum size of the metal line width, a path is cleared for further process generations to continue shrinking metal lines to dimensions below 10 nm.

Interconnect structure having large self-aligned vias

Forming a field effect transistor device includes forming first and second semiconductor fins on a semiconductor substrate. The first and second semiconductor fins are separated by a trench region. The trench region has a first sidewall corresponding to a sidewall of the first semiconductor fin and a second sidewall corresponding to a sidewall of the second semiconductor fin. A gate stack is arranged over respective channel regions of the first and semiconductor fins. A first sidewall of the gate stack corresponds to a third sidewall of the trench region. A protective layer is formed only on a bottom portion of the trench region and along the first sidewall of the gate stack. The protective layer along the first sidewall of the gate stack defines a gate spacer.

Protective trench layer and gate spacer in finFET devices

A wavy line interconnect structure that accommodates small metal lines and enlarged diameter vias is disclosed. The enlarged diameter vias can be formed using a self-aligned dual damascene process without the need for a separate via lithography mask. The enlarged diameter vias make direct contact with at least three sides of the underlying metal lines, and can be aligned asymmetrically with respect to the metal line to increase the packing density of the metal pattern. The resulting vias have an aspect ratio that is relatively easy to fill, while the larger via footprint provides low via resistance. An interconnect structure having enlarged diameter vias can also feature air gaps to reduce the chance of dielectric breakdown. By allowing the via footprint to exceed the minimum size of the metal line width, a path is cleared for further process generations to continue shrinking metal lines to dimensions below 10 nm.

Ultra-low-k dielectric materials used as inter-layer dielectrics in high-performance integrated circuits are prone to be structurally unstable. The Young's modulus of such materials is decreased, resulting in porosity, poor film strength, cracking, and voids. An alternative dual damascene interconnect structure incorporates deep air gaps into a high modulus dielectric material to maintain structural stability while reducing capacitance between adjacent nanowires. Incorporation of a deep air gap having k=1.0 compensates for the use of a higher modulus film having a dielectric constant greater than the typical ultra-low-k (ULK) dielectric value of about 2.2. The higher modulus film containing the deep air gap is used as an insulator and a means of reducing fringe capacitance between adjacent metal lines. The dielectric layer between two adjacent metal lines thus forms a ULK/high-modulus dielectric bi-layer.

Trench interconnect having reduced fringe capacitance

Antireflective residues during pattern transfer and consequential short circuiting are eliminated by employing an underlying sacrificial layer to ensure complete removal of the antireflective layer. Embodiments include forming a hard mask layer over a conductive layer, e.g., a silicon substrate, forming the sacrificial layer over the hard mask layer, forming an optical dispersive layer over the sacrificial layer, forming a silicon anti-reflective coating layer over the optical dispersive layer, forming a photoresist layer over the silicon anti-reflective coating layer, where the photoresist layer defines a pattern, etching to transfer the pattern to the hard mask layer, and stripping at least the optical dispersive layer and the sacrificial layer.

ARC residue-free etching

A method of fabricating asymmetrical spacers, structures fabricated using asymmetrical spacers and an apparatus for fabricating asymmetrical spacers. The method includes: forming on a substrate, a structure having a top surface and opposite first and second sidewalls and having a longitudinal axis parallel to the sidewalls; forming a conformal layer on the top surface of the substrate, the top surface of the structure and the sidewalls of the structure; tilting the substrate about a longitudinal axis relative to a flux of reactive ions, the flux of reactive ions striking the conformal layer at acute angle; and exposing the conformal layer to the flux of reactive ions until the conformal layer is removed from the top surface of the structure and the top surface of the substrate leaving a first spacer on the first sidewall and a second spacer on the second sidewall, the first spacer thinner than the second spacer.

Method of forming asymmetric spacers and methods of fabricating semiconductor device using asymmetric spacers

An integrated circuit having a mis-alignment tolerant electrical contact is formed by providing a semiconductor containing substrate over which is a first FET gate laterally bounded by a first dielectric region, replacing an upper portion of the first FET gate with a second dielectric region, applying a mask having an opening extending partly over an adjacent source or drain contact region of the substrate and over a part of the second dielectric region above the first FET gate, forming an opening through the first dielectric region extending to the contact region and the part of the second dielectric region, and filling the opening with a conductor making electrical connection with the contact region but electrically insulated from the first FET gate by the second dielectric region. A further FET gate may also be provided having an electrical contact thereto formed separately from the source-drain contact.

Integrated circuit contact structure and method

An alignment tolerant electrical contact is formed by providing a substrate on which is a first electrically conductive region (e.g., a MOSFET gate) having an upper surface, the first electrically conductive region being laterally bounded by a first dielectric region, applying a mask having an opening extending partly over a contact region (e.g., for the MOSFET source or drain) on the substrate and over a part of the upper surface, forming a passage through the first dielectric region extending to the contact region and the part of the upper surface, thereby exposing the contact region and the part of the upper surface, converting the part of the upper surface to a second dielectric region and filling the opening with a conductor making electrical contact with the contact region but electrically insulated from the electrically conductive region by the second dielectric region.

Alignment tolerant semiconductor contact and method

A semiconductor structure fabrication method. A provided structure includes: a semiconductor substrate, a transistor on the semiconductor substrate, N interconnect layers on the semiconductor substrate, and a temporary filling region within the N layers. N is at least 2. The temporary filling region is heated at a high temperature sufficiently high to result in the temporary filling material being replaced by a cooling pipes system that does not include any solid or liquid material. A first portion and a second portion of the cooling pipes system are each in direct physical contact with a surrounding ambient at a first interface and a second interface respectively such that a first direction perpendicular to the first interface is perpendicular to a second direction perpendicular to the second interface. A totality of interfaces between the cooling pipes system and the ambient consists of the first interface and the second interface.

On-chip cooling for integrated circuits

Structures and methods for forming the same. A semiconductor chip includes a substrate and a transistor. The chip includes N interconnect layers on the substrate, N being a positive integer. The chip includes a cooling pipes system inside the N interconnect layers. The cooling pipes system does not include any solid or liquid material. Given any first point and any second point in the cooling pipes system, there exists a continuous path which connects the first and second points and which is totally within the cooling pipes system. A first portion of the cooling pipes system overlaps the transistor. A second portion of the cooling pipes system is higher than the substrate and lower than a top interconnect layer. The second portion is in direct physical contact with a surrounding ambient.

On-chip cooling systems for integrated circuits

A semiconductor structure. The semiconductor structure includes: a first semiconductor region and a second semiconductor region; a first gate dielectric region on the first semiconductor region; a second gate dielectric region on the second semiconductor region, wherein the second semiconductor region includes a first top surface shared by the second semiconductor region and the second gate dielectric region, and wherein the first top surface defines a reference direction perpendicular to the first top surface and pointing from inside to outside of the second semiconductor region; an electrically conductive layer on the first gate dielectric region; a first poly-silicon region on the electrically conductive layer; a second poly-silicon region on the second gate dielectric region; a first hard mask region on the first poly-silicon region; and a second hard mask region on the second poly-silicon region.

Growing community of inventors

Ridgefield, CT, United States of America

Richard Stephen Wise