Provided are a device substrate with high thermal conductivity, with high heat dissipation, and with a small loss at high frequencies, and a method of manufacturing the device substrate. A device substrate 1 of the present invention can be manufactured by: provisionally bonding a Si device layer side of an SOI device substrate 10 to a support substrate 20 using a provisional bonding adhesive 31, the SOI device substrate including a Si base substrate 11, a Box layer 12 formed on the Si base substrate, having high thermal conductivity, and being an electrical insulator, and a Si device layer 13 formed on the Box layer; removing the Si base substrate 11 of the provisionally bonded SOI device substrate until the Box layer is exposed, thereby obtaining a thinned device wafer 10a; transfer-bonding the Box layer side of the thinned device wafer and a transfer substrate 40 to each other using a transfer adhesive 32 having a heat-resistant temperature of at least 150 C. by applying heat and pressure, the transfer substrate having high thermal conductivity and being an electrical insulator; and separating the support substrate 20.

Provided is a substrate holding unit that holds a pair of substrates that are aligned and layered, comprising a first holding member that holds one of the substrates; a plurality of members to be joined that are connected to the first holding member; a second holding member that holds the other of the substrates to face the one of the substrates; a plurality of joining members that exert an adhesion force on the members to be joined and are connected to the second holding member at positions corresponding to positions of the members to be joined; and an adhesion restricting section that restricts the adhesion force until the substrates are aligned.

A method for bonding a first substrate with a second substrate at respective contact faces of the substrates with the following steps: holding the first substrate to a first sample holder surface of a first sample holder with a holding force F.sub.H1 and holding the second substrate to a second sample holder surface of a second sample holder with a holding force F.sub.H2; contacting the contact faces at a bond initiation point and heating at least the second sample holder surface to a heating temperature T.sub.H; bonding of the first substrate with the second substrate along a bonding wave running from the bond initiation point to the side edges of the substrates, wherein the heating temperature T.sub.H is reduced at the second sample holder surface during the bonding.

A semiconductor device includes a first substrate having an attaching surface on which first electrodes and a first insulating film are exposed, an insulating thin film that covers the attaching surface of the first substrate, and a second substrate which has an attaching surface on which second electrodes and a second insulating film are exposed and is attached to the first substrate in a state in which the attaching surface of the second substrate and the attaching surface of the first substrate are attached together sandwiching the insulating thin film therebetween, and the first electrodes and the second electrodes deform and break a part of the insulating thin film so as to be directly electrically connected to each other.

Provided is a method of manufacturing a nanorod. The method comprising comprises the steps of: providing a growth substrate and a support substrate; epitaxially growing a nanomaterial layer onto one surface of the growth substrate; forming a sacrificial layer on one surface of the support substrate; bonding the nanomaterial layer with the sacrificial layer; separating the growth substrate from the nanomaterial layer; flattening the nanomaterial layer; forming a nanorod by etching the nanomaterial layer; and separating the nanorod by removing the sacrificial layer.

Die stacks and methods of making die stacks with very thin dies are disclosed. The die surfaces remain flat within a 5 micron tolerance despite the thinness of the die and the process steps of making the die stack. A residual flux height is kept below 50% of the spacing distance between adjacent surfaces or structures, e.g. in the inter-die spacing.

A method for removing a portion of a crystalline material (e.g., SiC) substrate includes joining a surface of the substrate to a rigid carrier (e.g., >800 m thick), with a subsurface laser damage region provided within the substrate at a depth relative to the surface. Adhesive material having a glass transition temperature above 25 C. may bond the substrate to the carrier. The crystalline material is fractured along the subsurface laser damage region to produce a bonded assembly including the carrier and a portion of the crystalline material. Fracturing of the crystalline material may be promoted by (i) application of a mechanical force proximate to at least one carrier edge to impart a bending moment in the carrier; (ii) cooling the carrier when the carrier has a greater coefficient of thermal expansion than the crystalline material; and/or (iii) applying ultrasonic energy to the crystalline material.

A method is provided for bonding and interconnecting two semiconductor chips arranged on semiconductor substrates. HSQ (Hydrogen Silsesquioxane) or an equivalent material is used as a bonding layer and after bonding and thinning one of the wafers (or first thinning and then bonding), the bond layer is locally irradiated by an e-beam through the thinned substrate, thereby locally transforming the bonding material into silicon oxide. Then a via opening is etched through the thinned substrate and an etch process selectively removes the oxide from an area delimited by the bonding material or vice versa. The filling of the via opening establishes an electrical connection between the bonded wafers, that is equivalent to a connection obtained by hybrid bonding, but that does not suffer from the disadvantages thereof.

A method and device for bonding a first substrate with a second substrate inside a sealed bonding chamber. The method includes: a) fixing of the first and second substrates, b) arranging of the first and second substrates, c) mutual approaching of the first and second substrates, d) contacting the first and second substrates at respective bond initiation points, e) generating a bonding wave running from the bond initiation points to side edges of the substrates, and f) influencing the bonding wave during course of the bonding wave, wherein targeted influencing of the bonding wave takes place by a regulated and/or controlled change of pressure inside the bonding chamber.

Direct-bonded native interconnects and active base dies are provided. In a microelectronic architecture, active dies or chiplets connect to an active base die via their core-level conductors. These native interconnects provide short data paths, which forgo the overhead of standard interfaces. The system saves redistribution routing as the native interconnects couple in place. The base die may contain custom logic, allowing the attached dies to provide stock functions. The architecture can connect diverse interconnect types and chiplets from various process nodes, operating at different voltages. The base die may have state elements for drive. Functional blocks aboard the base die receive native signals from diverse chiplets, and communicate with all attached chiplets. The chiplets may share processing and memory resources of the base die. Routing blockages are minimal, improving signal quality and timing. The system can operate at dual or quad data rates. The architecture facilitates ASIC, ASSP, and FPGA ICs and neural networks, reducing footprint and power requirements.