Production method for semiconductor substrate and semiconductor element
    41.
    发明授权
    Production method for semiconductor substrate and semiconductor element 有权
    半导体衬底和半导体元件的制造方法

    公开(公告)号:US07011707B2

    公开(公告)日:2006-03-14

    申请号:US10473074

    申请日:2002-03-27

    IPC分类号: C30B25/22

    摘要: A reaction prevention layer is formed to prevent Si from reacting with a gallium nitride group semiconductor (semiconductor crystal A) which is deposited after the reaction prevention layer is formed. By forming a reaction prevention layer comprising a material whose melting point or thermal stability is higher than that of a gallium nitride group semiconductor, e.g., AlN, on a sacrifice layer, a reaction part is not formed in the semiconductor substrate deposited on the reaction prevention layer when the gallium nitride group semiconductor is grown by crystal growth for a long time. In short, owing to the effect that the reaction prevention layer prevents silicon (Si) from diffusing, the reaction part is generated only in the sacrifice layer and it is never formed at the upper portion of the reaction prevention layer even by growing the semiconductor crystal A at a high temperature for a long time.

    摘要翻译: 形成反应防止层,以防止Si与形成反应防止层之后沉积的氮化镓基半导体(半导体晶体A)发生反应。 通过在牺牲层上形成包含熔点或热稳定性高于氮化镓基半导体(例如AlN)的材料的反应防止层,在沉积在反应预防的半导体衬底中不形成反应部分 通过长时间的晶体生长生长氮化镓族半导体时的层。 简而言之,由于防反射层防止硅(Si)扩散,反应部分仅在牺牲层中产生,并且即使通过生长半导体晶体也不会形成在反应防止层的上部 A在高温下长时间。

    Group III nitride compound semiconductor element and method for producing the same
    42.
    发明授权
    Group III nitride compound semiconductor element and method for producing the same 失效
    III族氮化物化合物半导体元件及其制造方法

    公开(公告)号:US06716655B2

    公开(公告)日:2004-04-06

    申请号:US10160288

    申请日:2002-06-04

    IPC分类号: H01L3304

    摘要: An object of the invention is to produce, at high efficiency, semiconductor elements which are formed of a high-quality crystalline semiconductor having no cracks and a low dislocation density and which have excellent characteristics. Specifically, a mask formed from SiO2 film is provided on the Si(111) plane of an n-type silicon substrate, and a window portion (crystal growth region) in the shape of an equilateral triangle having a side of approximately 300 &mgr;m is formed through the mask. The three sides of the equilateral triangle are composed of three edges; each edge defined by the (111) plane and another crystal plane that is cleavable. Subsequently, a multi-layer structure of semiconductor crystals in an LED is formed through crystal growth of a Group III nitride compound semiconductor. Thus, limiting the area of one crystal growth region to a considerably small area weakens a stress applied to a semiconductor layer, thereby readily producing semiconductor elements having excellent crystallinity. In addition, semiconductor elements can be arranged in a semiconductor wafer at high packing density without loss, and each side of these semiconductor elements can be readily arranged in a line on a semiconductor wafer, thereby enhancing quality, yield, productivity, etc. of semiconductor elements.

    摘要翻译: 本发明的目的是高效率地制造由没有裂纹和位错密度低且具有优异特性的高品质结晶半导体形成的半导体元件。 具体地,在n型硅衬底的Si(111)面上设置由SiO 2膜形成的掩模,并且形成具有约300μm侧面的等边三角形形状的窗口部分(晶体生长区域) 通过面具。 等边三角形的三面由三边组成, 每个边缘由(111)面和另一个可切割的晶体平面限定。 随后,通过III族氮化物化合物半导体的晶体生长,形成LED中的半导体晶体的多层结构。 因此,将一个晶体生长区域的面积限制在相当小的面积上,削弱施加到半导体层的应力,从而容易地制造具有优异结晶度的半导体元件。 此外,半导体元件可以以高封装密度无损耗地布置在半导体晶片中,并且这些半导体元件的每一侧可以容易地布置在半导体晶片上的线上,从而提高半导体的质量,产量,生产率等 元素。

    Method for fetching plural instructions using single fetch request in
accordance with empty state of instruction buffers and setting of flag
latches
    47.
    发明授权
    Method for fetching plural instructions using single fetch request in accordance with empty state of instruction buffers and setting of flag latches 失效
    根据指令缓冲区的空状态和标志锁存器的设置,使用单次提取请求取出多个指令的方法

    公开(公告)号:US5267350A

    公开(公告)日:1993-11-30

    申请号:US603998

    申请日:1990-10-25

    CPC分类号: G06F9/3814 G06F9/3802

    摘要: An instruction fetch control method is generally arranged so as to have a plurality of instruction buffers, issue an instruction read request to a memory when a part of the instruction buffers becomes in an empty state and store a fetched instruction in the instruction buffer in an empty state. A flag is provided for specifying another instruction buffer which becomes in an empty state after the instruction stored in the instruction buffer is transmitted to a decoder. The quantity of instructions to be stored in the instruction buffer is made variable in accordance with output of the flag latch, and the fetched instruction is stored in the instruction buffer in an empty state. This arrangement enables a plurality of instructions fetched upon an instruction read request to be stored in the empty instruction buffers, thereby reducing the number of read requests issued.

    摘要翻译: 指令获取控制方法通常被布置为具有多个指令缓冲器,当指令缓冲器的一部分变为空状态时,向存储器发出指令读取请求,并将指令缓冲器中的取指令存储在空 州。 提供了一种标志,用于指定在存储在指令缓冲器中的指令被发送到解码器之后变为空状态的另一指令缓冲器。 存储在指令缓冲器中的指令量根据标志锁存器的输出而变化,并且取出的指令以空状态存储在指令缓冲器中。 这种布置使得可以在指令读取请求中取出的多个指令被存储在空指令缓冲器中,从而减少发出的读请求数。

    Method and apparatus for casting an inner lining amorphous refractory
into a molten metal vessel
    48.
    发明授权
    Method and apparatus for casting an inner lining amorphous refractory into a molten metal vessel 失效
    将内衬无定形耐火材料浇注到熔融金属容器中的方法和装置

    公开(公告)号:US4421697A

    公开(公告)日:1983-12-20

    申请号:US375716

    申请日:1982-05-07

    IPC分类号: B22D41/02 F27D1/16 F27D3/00

    摘要: There is disclosed the distribution of an amorphous refractory in a case where a lining frame is positioned within a molten metal vessel preliminarily lined with a permanent lining refractory and the amorphous refractory is cast into the space between the permanent lining refractory and the lining frame thereby applying a lining to the inner surface of the vessel.A conical distributor is arranged above the lining frame and the amorphous refractory is continuously delivered from a rotary chute onto the distributor so as to describe a concentric path of moving falling points.

    摘要翻译: 公开了在衬里框架位于预先排列有永久性衬里耐火材料的熔融金属容器内的情况下的非晶态耐火材料的分布,并且将非晶态耐火材料浇注到永久性衬里耐火材料和衬里框架之间的空间中,从而施加 衬里到容器的内表面。 锥形分配器布置在衬里框架的上方,非晶态耐火材料从旋转滑道连续地传送到分配器上,以便描述移动落点的同心路径。