Abstract: A semiconductor device includes a substrate having a high electron mobility transistor (HEMT) region and a capacitor region, a first mesa isolation on the HEMT region, a HEMT on the first mesa isolation, a second mesa isolation on the capacitor region, and a capacitor on the second mesa isolation. The semiconductor device further includes buffer layer between the substrate, the first mesa isolation, and the second mesa isolation, in which bottom surfaces of the first mesa isolation and the second mesa isolation are coplanar.

Abstract: A magnetoresistive random access memory (MRAM) includes a first transistor and a second transistor on a substrate, a source line coupled to a first source/drain region of the first transistor, and a first metal interconnection coupled to a second source/drain region of the first transistor. Preferably, the first metal interconnection is extended to overlap the first transistor and the second transistor and the first metal interconnection further includes a first end coupled to the second source/drain region of the first transistor and a second end coupled to a magnetic tunneling junction (MTJ).

Abstract: A 3D semiconductor structure includes a buffer layer, a n-type high electron mobility transistor (HEMT) disposed on a first surface of the buffer layer, and a p-type high hole mobility transistor (HHMT) disposed on a second surface of the buffer layer opposite to the first surface.

Abstract: A semiconductor device includes a substrate having a magnetic tunneling junction (MTJ) region and a logic region, a magnetic tunneling junction (MTJ) on the MTJ region, and a first metal interconnection on the MTJ. Preferably, a top view of the MTJ includes a circle, a top view of the first metal interconnection includes a flat oval overlapping the circle, and the MTJ includes a bottom electrode, a fixed layer, a free layer, a capping layer, and a top electrode.

Abstract: An embedded MRAM structure includes a substrate divided into a memory cell region and a logic device region. An active area is disposed in the memory cell region. A word line is disposed on the substrate and crosses the active area. A source plug is disposed in the active area and at one side of the word line. A drain plug is disposed in the in the active area and at another side of the word line. When viewing from a direction perpendicular to the top surface of the substrate and taking the word line as a symmetric axis, the source plug is a mirror image of the drain plug.

Abstract: A semiconductor device includes: a substrate having a magnetic tunneling junction (MTJ) region and a logic region; a magnetic tunneling junction (MTJ) on the MTJ region, wherein a top view of the MTJ comprises a circle; and a first metal interconnection on the MTJ. Preferably, a top view of the first metal interconnection comprises a flat oval overlapping the circle.

Abstract: A mask includes a substrate, a main pattern, a first assist pattern, and a second assist pattern. The main pattern is disposed on the substrate. The main pattern includes a first pattern and second patterns. Two of the second patterns are disposed at two opposite sides of the first pattern in a first direction. The first assist pattern is disposed on the substrate and disposed in the main pattern. The second assist pattern is disposed on the substrate and disposed outside the main pattern. The first assist pattern disposed in the main pattern may be used to improve the pattern transferring performance in a photolithography process using the mask.

Abstract: An embedded MRAM structure includes a substrate divided into a memory cell region and a logic device region. An active area is disposed in the memory cell region. A word line is disposed on the substrate and crosses the active area. A source plug is disposed in the active area and at one side of the word line. A drain plug is disposed in the in the active area and at another side of the word line. When viewing from a direction perpendicular to the top surface of the substrate and taking the word line as a symmetric axis, the source plug is a mirror image of the drain plug.

Abstract: A 3D semiconductor structure includes a buffer layer, a n-type high electron mobility transistor (HEMT) disposed on a first surface of the buffer layer, and a p-type high hole mobility transistor (HHMT) disposed on a second surface of the buffer layer opposite to the first surface.

Abstract: A high electron mobility transistor (HEMT) includes a carrier transit layer, a carrier supply layer, a main gate, a control gate, a source electrode and a drain electrode. The carrier transit layer is on a substrate. The carrier supply layer is on the carrier transit layer. The main gate and the control gate are on the carrier supply layer. The source electrode and the drain electrode are at two opposite sides of the main gate and the control gate, wherein the source electrode is electrically connected to the control gate by a metal interconnect. The present invention also provides a method of forming a high electron mobility transistor (HEMT).

Abstract: A semiconductor device and a method of forming the same, the semiconductor device includes a substrate, a first well and a first dummy cell region. The substrate has a plurality fins disposed therein, and the fins are extended along a first direction. The first well is disposed in the substrate, and a dummy cell region is disposed at a first boundary of the first well. The first dummy cell region includes a first isolation structure and a plurality of first gate structures. The first SDB is disposed in the substrate, along a second direction perpendicular to the first direction to penetrate through one of the fins, and the first gate structures are disposed over the first SDB.

Abstract: A mark pattern includes unit cells immediately adjacent to each other and arranged in a form of dot matrix to form a register mark or an identification code, wherein each unit cell has configuration identical to functional devices of pMOS and nMOS, and each unit cell includes a first active region, a second active region isolated from the first active region, and first gate structures extending along a first direction and are arranged along a second direction perpendicular to the first direction, and the first gate structures straddling the first active region and the second active region, contact structures disposed between the first gate structures on the first active region and the second active region, and via structures disposed on the contact structures and two opposite ends of the first gate structures.

Abstract: A semiconductor device includes: a substrate having a magnetic tunneling junction (MTJ) region and a logic region; a magnetic tunneling junction (MTJ) on the MTJ region, wherein a top view of the MTJ comprises a circle; and a first metal interconnection on the MTJ. Preferably, a top view of the first metal interconnection comprises a flat oval overlapping the circle.

Abstract: A semiconductor device includes a substrate including a plurality of chip areas and a scribe line defined thereon, and a mark pattern disposed in the scribe line. The mark pattern includes a plurality of unit cells immediately adjacent to each other, and each unit cell includes a first active region, a second active region isolated from the first active region, a plurality of first gate structures extending along a first direction and arranged along a second direction perpendicular to the first direction, and a plurality of first conductive structures. The first gate structures straddle the first active region and the second active region. The first conductive structures are disposed on the first active region, the second active region, and two opposite sides of the first gate structures.

Abstract: A method for correcting a mask pattern includes: providing an original mask pattern including at least one dense pattern area and at least one isolated pattern area, and the original mask pattern being divided into a first pattern and a second pattern, wherein the first pattern is formed in the isolated pattern area and extends to the dense pattern area, and the second pattern is formed in the dense pattern area; forming at least one slot on at least one section of the first pattern, and the at least one section of the first pattern is located on at least one transition area between the at least one isolated pattern area and the at least one dense pattern area; and performing an optical proximity correction operation on the first pattern formed with at least one slot and the second pattern. Using the corrected mask pattern may avoid the occurrence of necking or breaking on portion of the post-transfer pattern.

Abstract: A magnetic random access memory (MRAM) includes device strings coupled in parallel, each comprising magnetic tunnel junctions (MTJs) coupled in serial, wherein a quantity of the MTJs of each of the device strings is equal to a quantity of the device strings, and an equivalent resistance (Req) of the MTJs is equal to an average of the sum of a high resistance of one of the MTJs and a low resistance of another MTJ.

Abstract: A semiconductor device includes: a dummy gate on a substrate; a first control gate on one side of the dummy gate and a second control gate on another side of the dummy gate; a well in the substrate, wherein the well comprises a first conductive type; a first source/drain region between the dummy gate and the first control gate, wherein the first source/drain region comprises a second conductive type; a second source/drain region between the dummy gate and the second control gate, wherein the second source/drain region comprises the second conductive type; and a doped region directly under the dummy gate, wherein the doped region comprises the first conductive type.

Abstract: A semiconductor device includes: a dummy gate on a substrate; a first control gate on one side of the dummy gate and a second control gate on another side of the dummy gate; a well in the substrate, wherein the well comprises a first conductive type; a first source/drain region between the dummy gate and the first control gate, wherein the first source/drain region comprises a second conductive type; a second source/drain region between the dummy gate and the second control gate, wherein the second source/drain region comprises the second conductive type; and a doped region directly under the dummy gate, wherein the doped region comprises the first conductive type.

Abstract: A metal interconnection includes a substrate, a first dielectric layer, metal wirings, air gaps and air gap dummies. The substrate includes an isolated area and a dense area. The first dielectric layer is disposed over the substrate. The metal wirings are embedded in the first dielectric layer, wherein the density of the metal wirings in the isolated area is less than the density of the metal wirings in the dense area. The air gaps are sandwiched by the metal wirings. The air gap dummies are disposed in the first dielectric layer without contacting the metal wirings. The present invention also provides a method of forming a metal interconnection.

Abstract: A method of forming a fin forced stack inverter includes the following steps. A substrate including a first fin, a second fin and a third fin across a first active area along a first direction is provided, wherein the first fin, the second fin and the third fin are arranged side by side. A fin remove inside active process is performed to remove at least a part of the second fin in the first active area. A first gate is formed across the first fin and the third fin in the first active area along a second direction. The present invention also provides a 1-1 fin forced fin stack inverter formed by said method.