Siemens Aktiengesellschaft

Siemens Aktiengesellschaft Ag

Christoph Werner

Lothar Risch

Ulrich Schwabe

Matthias Ilg

Markus M Kirchhoff

Wolfgang Peter Mueller

Reinhard Tielert

Alfred Kersch

Ignacio Ulacia

A method and apparatus for forming a multi-constituent device layer on a wafer surface are disclosed. The multi-constituent device layer is formed by flowing a first chemistry comprising a first constituent and a second chemistry comprising a second constituent via a segmented delivery system into a reaction chamber. The reaction chamber comprises a susceptor for supporting and rotating the wafers. The segmented delivery system comprises alternating first and second segments into which the first and second chemistries, respectively, are flowed. The first segments comprise an area that is greater than an area of the second segments by an amount sufficient to effectively reduce the diffusion path of the first constituent. Reducing the diffusion path of the first constituent results in a more uniform distribution of the first constituent within the device layer.

Uniform distribution of reactants in a device layer

A method for depositing a layer on a substrate is disclosed wherein a collimator having cylindrical holes is employed to reduce the lateral component of a particle flux. The cylindrical holes are aligned to be perpendicular to a substrate wafer and have a variety of radii such that the hole radii are smaller in regions having a higher vertical component of particle flux than in regions which have a lower vertical component of the particle flux.

Method for depositing a layer on a substrate wafer with a sputtering

Method for the selective deposition of tungsten on a silicon surface. A silicon surface is provided with a masking structure of, for example, SiO.sub.2. The deposition from the vapor phase is implemented in a CVD reactor from a mixture of process gases such that the masking structure remains essentially free of tungsten. The process parameters of temperature, pressure and flow-through rates of the process gases for the selective deposition are determined for a prescribed deposition rate by solving an equation system that contains the conservation laws for the mass, the chemical constituents participating in the deposition reaction, the momentum and the energy taking the boundary conditions of the CVD reactor into consideration, and the effects of thermodiffusion and multi-constituent diffusion.

Method for the selective deposition of tungsten on a silicon surface

A voltage-stable sub-.mu.m-MOS transistor for VLSI circuits consist of a low-resistant silicon substrate of a first conductivity type with a high-resistant, thin, epitaxial layer of the first conductivity type situated thereon and on which a gate electrode consisting of polysilicon is disposed. Highly doped source/drain zones of the second conductivity type form a channel region of the first conductivity type. A doping substance concentration, rising in the direction of the substrate, is generated by means of double implantation, whereby the concentration maximum extends to behind the source/drain zones. A method for manufacturing same incorporates steps of forming the several layers, applying a mask, executing a double implantation in the channel region, and forming the gate electrode.

Voltage-stable sub-.mu.m MOS transistor for VLSI circuits

A manufacturing method for VLSI MOS field effect transistor circuits having digital and analog functions performed by short channel transistors and analog transistors integrated on one chip. An n-tube manufacture is performed wherein as soft as possible a field progression in front of a drain-side pn-junction of the analog transistor is achieved. This occurs by means of an additional drain implantation (curve II) with drive-in diffusion before the actual source/drain implantation (curve I) of the n-channel transistors. Both the additional implantation as well as the source/drain implantation are carried out with phosphorous ions. The dosage of the additional implantation lies one to two orders of magnitude below the dosage of the actual implantation, and the penetration depth x in the additional drive-in diffusion is about twice as great as the penetration depth x of the actual source/drain regions. The method is applied in the manufacture of VLSI CMOS circuits.

Growing community of inventors

Moosach, Germany

Christoph Werner