Thermal oxidation of silicon

from Wikipedia, the free encyclopedia

The thermal oxidation of silicon is in the semiconductor art , a coating process in which on a single crystalline silicon substrate (e.g. a silicon - wafer ) a thin layer of amorphous silica is deposited. It is used, among other things, in the manufacture of microelectronic circuits . The coating process is based on a chemical reaction of oxygen and silicon at temperatures above 1100  ° C . In the case of very short process times, the process is also called " Rapid Thermal Oxidation " (RTO, dt .: rapid thermal oxidation ), which is used to create very thin oxide layers (<2 nm). A similar process is the production of a thermal silicon nitride layer on a silicon substrate at high temperatures.

process description

The oxidation of silicon to silicon dioxide is a diffusion-dependent solid-state reaction. It already takes place at room temperature under laboratory conditions ( water is required), but the reaction rate is far below the requirements for technical / industrial processes. In addition, due to the diffusion limitation, only a natural oxide layer of around two nanometers is usually formed .

Consumption of silicon in thermal oxidation

There are essentially two different processes for the oxidation of silicon: dry and wet oxidation. There are also a smaller number of similar variants, such as the H 2 -O 2 combustion. The oxidation process of all variants can be summarized in three steps: (i) transport of the gaseous starting materials (e.g. oxygen or water) to the surface of the substrate, (ii) diffusion through the existing oxide layer and (iii) the oxidation reaction itself Reaction, oxygen is built into the silicon substrate, which means that in this coating process no layer is actually applied to a substrate, but the substrate is converted on the surface. In contrast to a coating, part of the later layer lies in the area of ​​the previous silicon substrate. The silicon is, so to speak, "consumed". In the case of thermal oxide, the resulting oxide layer is approx. 46% below and 54% above the starting silicon substrate.

In terms of growth rates and layer properties ( density , dielectric strength, etc.), the two main processes differ in some cases greatly. Both processes have in common that firstly doping concentrations above 10 −6 (corresponds to 10 18 atoms to approx. 6 × 10 23 silicon atoms, see Avogadro's constant ) promote the oxidation, and secondly that the oxidation depends on the crystal orientation, whereby the oxidation of {111} silicon surfaces is 30–100% faster than that of {100} silicon surfaces (the information {111} and {100} denote certain crystal surfaces or surfaces of the unit cell , cf. Miller indices and diamond structure ).

As with every high-temperature process in semiconductor technology, wafer cleaning takes place before oxidation. This serves both to improve the process itself and to prevent pipe contamination . The main purpose of cleaning is to reduce metallic contamination, which would otherwise worsen the electrical properties of the oxide layers. A typical cleaning process is RCA-2 cleaning, which uses hydrogen chloride (HCl) to bind metallic contaminants. HCl is also used to clean the furnace pipes, but nowadays organic chlorine compounds such as 1,2-dichloroethene (DCE) are used more frequently .

Dry oxidation

The reaction can be accelerated significantly by using high temperatures. At the usual temperatures between 800 and 1200 ° C, silicon already oxidizes when it is exposed to oxygen. This process, in which the oxidation is only caused by oxygen, is also called dry oxidation . The layer thickness achieved depends on the temperature and the oxidation time. Layers created with this process grow rather slowly, but have a high layer quality.

Wet oxidation

Another method utilizes water vapor as the oxidizing agent, it is therefore wet oxidation (also wet oxidation called). By passing through a carrier gas, commonly oxygen or an oxygen nitrogen - mixture before it is introduced into the oxidation furnace, a warm 90-95 ° C with deionized water -filled container (so-called bubbler ). The water molecules transported by the carrier gas then react with the silicon surface:

The oxidation reaction usually takes place at temperatures between 900 ° C and 1100 ° C. The layer growth takes place here quickly, but with a lower quality than with dry oxidation.

Further process variants

In addition to thermal oxidation with pure oxygen or steam, there are other processes in which the actual reactants for the oxidation of silicon are only formed by a reaction in the process chamber, for example hydrogen (H 2 ) and oxygen (O 2 ), trichloroacetic acid (TCA) and oxygen or hydrogen chloride (HCl) and oxygen. These are rarely used in practice, but sometimes offer better layer properties.

H 2 -O 2 combustion: In H 2 -O 2 combustion ( pyrogenic oxidation ), water is formed directly in the reaction chamber through the reaction of high-purity hydrogen and oxygen at approx. 600 ° C. For this purpose, the two starting gases (mostly together with nitrogen) are fed into the process chamber via separate feed lines. Particular attention must be paid to the mixing ratio, as there is a risk of explosion due to the formation of oxyhydrogen . The actual oxidation process corresponds to that of wet oxidation, in which silicon reacts with water to form silicon dioxide. The H 2 -O 2 combustion can produce oxide layers with a high growth rate but low levels of impurities and defects.

Deal Grove model

The Deal Grove model is a frequently used description for the diffusion-based layer growth of thermal silicon dioxide on a pure silicon surface. The time t necessary for the oxidation process , which is necessary to achieve a certain layer thickness d SiO2 , is calculated as follows:

where B denotes the parabolic and the factor B / A denotes the linear growth rate.

For a silicon substrate that already has an oxide layer, the equation must be supplemented by a term . describes the time that would be necessary to generate the existing layer under the current process parameters.

The constant can also be used to take into account the rapid initial growth during dry oxidation, which cannot be described with the deal groove model, for the calculation of the process duration.

Solving the quadratic equation for d SiO 2 one obtains:

The Deal Grove model is not suitable for thin oxide layers smaller than 30 nm, since the oxide here initially grows faster than expected. In addition, there is often a delay before the oxidation process starts. This time is longer than the time it takes to exchange the gas volume in the oxidation furnace.

Different approaches are followed in the literature for modeling the growth rates of very thin oxide layers. A frequently used approach is based on the expansion of the Deal Grove model by an additional term that can be used to describe the growth rate at the beginning of the layer growth (e.g.).

Another approach is to introduce a transition layer between silicon and silicon dioxide. In contrast to the Deal-Grove model, which assumes an abrupt transition from Si to SiO 2 , it is assumed that in this transition layer made of non-stoichiometric silicon oxide (SiO 0 <x <2 ) with a thickness of 1.5 nm to 2 nm the Oxidation reaction takes place. The presence of such a transition layer could be confirmed experimentally in XPS measurements.

Oxidation techniques and facilities

Loading side of a horizontal diffusion furnace with four furnace tubes (behind the round cover) for thermal oxidation and for diffusion processes.

Most often, thermal oxidation in the heating furnaces is carried out at temperatures between 800 ° C and 1200 ° C. A single furnace usually takes several wafers (25 to 200) in a rack . There are two main furnace designs that differ in the way the wafers are stored: horizontal and vertical furnaces. The horizontal design is mainly used in older systems or systems for wafers with diameters of 150 mm and smaller. Vertical ovens, on the other hand, are more often used in newer systems for wafers with a diameter of 200 mm or 300 mm.

Schematic structure of a horizontal oxidation furnace for the optional operation with oxygen (dry oxidation) and steam (wet oxidation).

In the case of horizontal ovens, the wafers are next to each other. Falling dust can get between the wafers and in principle contaminate every wafer. Horizontal furnaces typically use a convection current within the oxidation tube, the result of which is that it is slightly colder in the reaction chamber below than above and the oxide layers thus grow slightly more slowly on the downward-facing sides of the wafers; Uneven layer thicknesses are the result. This is no longer acceptable for larger wafers (diameter greater than 150 mm), as they are used as standard today, and the increased requirements placed on manufacturing tolerances. One advantage of the horizontal ovens is that several oven pipes can be arranged one above the other in a system, which saves some space in the clean room.

The wafers are stored one above the other in vertical ovens. With this arrangement, falling dust can only fall on the highest-placed wafer; Dust pollution is minimized or prevented in this way. Due to the horizontal storage, a more uniform temperature distribution and thus uniform layer thicknesses over the individual wafer is achieved. Due to the different temperature distribution in the furnace tube, the wafers stored below have a thinner layer than the upper ones; there are also minimal differences between the top and bottom of a wafer. These problems can be reduced by guiding the gas flow from top to bottom in the opposite direction to the convection flow.

Layer properties

Important properties of thermal SiO 2 (selection)
Density (dry; wet oxidized) 2.27 g / cm 3 ; 2.18 g / cm 3
Thermal expansion coefficient 5.6 · 10 −7 K −1
modulus of elasticity 6.6 x 10 10 N / m²
Poisson's number 0.17
Thermal conductivity 3.2 · 10 −3 W / (cm · K)
Relative dielectric constant 3.7 ... 3.9
Dielectric strength (dry; wet oxidized) ≈ 10 MV / cm; ≈ 8 MV / cm
Band gap 8.9 eV

Silicon dioxide layers produced by dry or wet oxidation are glass-like and only have a short-range order (→  amorphous ). Their properties are almost identical to those of quartz glass , which is mostly used as the material for the oxidation tubes. Furthermore, not all bonds between silicon and oxygen are fully formed, which leads to unbound, charged oxygen atoms. The molecular structure thus differs significantly from crystalline silicon dioxide ( quartz ), among other things in terms of its density (≈ 2.2 g cm −3 instead of 2.65 g cm −3 for quartz) and its modulus of elasticity (87 GPa instead of 107 GPa for quartz).

As with other coating processes, the properties of silicon dioxide layers produced by thermal oxidation vary depending on the process conditions. The most important influencing factors are the oxidation process (dry or wet) and the process temperature. With regard to the performance and reliability of the microelectronic components and circuits, the electrical properties of the thermal oxides are particularly important, such as electrical conductivity, carrier trapping and the presence of oxide charges.

When producing thick oxide layers, wet oxidation is preferred to dry oxidation because of the higher growth rate. The disadvantage of wet oxidation is the poorer layer properties (especially the electrical ones). The higher growth rate means that more dangling bonds are created at the interface with the silicon and also in the layer itself; In this context, one speaks of a higher defect density . These free bonds act as impurities or scattering centers for electrons and, among other things, allow a leakage current along the interface and cause a lower dielectric strength .

In contrast to this, layers that were produced by dry oxidation have better layer properties. However, the slow growth rate has a negative impact on process costs. In practice, there are therefore more often processes that combine both methods, so-called dry-wet-dry cycles. The rapid layer growth of wet oxidation is used to keep the process times short. By creating high quality boundary layers with dry oxidation at the beginning and end, the negative properties of wet oxidation are largely eliminated.

segregation

As already mentioned, silicon is consumed in the oxide growth. Since foreign substances have different solubilities in silicon and silicon oxide, they can either be built into the oxide layer or remain in the silicon or at the boundary layer. Depending on the solubility coefficient, there can be an enrichment ( ) or a depletion ( ) of foreign atoms in the silicon at the interface with the oxide; this separation is also called segregation . The so-called segregation coefficient k is decisive for assessing this process . This allows the proportional distribution of the foreign atoms in the oxide or in the silicon to be determined.

application

Single 200 mm silicon wafer in a horde of a horizontal oxidation furnace. Due to the interference effects of a thin layer, the wafer shows a greenish discoloration, as also occurs after thermal oxidation. You can clearly see reddish discoloration at the edge, which is caused by a different layer thickness. Such “edge effects” do not occur with the thermal oxidation of silicon.

The thermal oxidation of silicon has been one of the most important processes in the manufacture of microelectronic circuits since the mid-1950s, when the first silicon-based transistors were manufactured commercially. At that time, silicon prevailed over germanium as the preferred material in semiconductor technology. The decisive factors for this development were, among other things, the better material properties of silicon dioxide compared to germanium oxide, which had poorer adhesion properties and is not stable to water.

The process of thermal oxidation of silicon was found by chance in the 1950s at Bell Telephone Laboratories in New Jersey , where the first working transistor was discovered in 1947 - but other industrial research laboratories and universities were also involved. At that time the doping of semiconductors by diffusion of gaseous dopants ( boron , phosphorus , arsenic , antimony ) was already known. The processes were carried out at high temperatures around 1000 ° C. In 1955 Carl Frosch accidentally mixed hydrogen and oxygen in a diffusion tube. After the silicon samples were removed from the furnace, they showed a discoloration to a light green. It was found that a stable thin layer of thermal silicon dioxide formed.

Important areas in which thermally produced silicon dioxide has been used and in some cases still is used are selective doping masking, surface passivation of silicon and electrical insulation of components in planar technology. In the manufacture of modern ICs, however, this technology is only used in the first process steps, for example in the manufacture of trench insulation or gate oxides, which in CMOS transistors separates the gate from the silicon in order to form the conductive channel underneath through the resulting electrical field . The main reason why this process is not used later in the manufacturing process is the high process temperature. These lead, for example, to a shift in doping profiles. For this and other reasons, “low-temperature processes” such as chemical vapor deposition (reaction with TEOS at 600 ° C) or (rarely) sputter deposition are used in all other areas and production sections (insulation of the interconnects, etc.) . Although these produce an oxide of poor quality, they are also suitable for producing oxide layers on materials other than silicon.

Material-selective masking for diffusion doping

The property of silicon dioxide of being a material-selective mask for the diffusion of dopants in silicon was first presented in 1956 by the Bell Labs employees Frosch and Derick. They discovered that the diffusion of n-type dopants (P, As, Sb) into the silicon is hindered at temperatures above 1000 ° C in an oxidizing atmosphere. The same applies to the p-dopant boron, but in contrast to the n-dopants mentioned, boron can diffuse more quickly through the oxide and into the silicon in the presence of hydrogen and water vapor. The technique of selective masking found practical application in the manufacture of the double diffused transistor or the so-called mesa transistor , because it allowed the contact between the emitter and base to be made on one surface.

Surface passivation

The properties of the transistors of that time were unpredictable and not stable due to their unprotected surface. For this reason, a large number of research laboratories dealt with the surface passivation of germanium and silicon between 1955 and 1960. An important working group carried out research at Bell Laboratories, Martin M. Atalla and co-workers found that special cleaning and subsequent production of a thin thermal oxide (15–30 nm) resulted in a significant reduction in leakage currents at pn junctions. The cause was the binding or neutralization of surface conditions. This later made it possible to control the charge carrier mobility using an external electric field (see MOSFET ). Passivation later enabled further important developments, especially the planar process and the planar diffusion transistor as well as the integrated circuits based on them.

Thermal oxidation is still used today for the surface passivation of single-crystal and poly-crystalline silicon layers. The process is not limited to the production of microelectronic components, but can be used in almost all areas that use silicon for “electrical applications”, for example solar cells or microsystems .

Component isolation

Bird's beak after a normal LOCOS process

The components (transistors, diodes) of an integrated circuit are usually located on the surface of a wafer. In the early years of microelectronics, they were initially sufficiently far apart and the components were isolated (preventing leakage currents, etc.) using reverse- biased pn junctions . In the early 1970s, however, the power requirements on the circuits increased and the packing density of the components was increasingly increased. The isolation through pn junctions was no longer sufficient. Their place was primarily taken by oxide insulation that was manufactured using the LOCOS process or similar processes. They enabled the capacities and leakage currents between the components to be minimized, and also enabled a higher packing density and thus saved space on the wafer. In the LOCOS process, silicon is only oxidized (locally) in selected areas. The areas not to be oxidized are masked with a material that blocks the diffusion of oxygen and water required for thermal oxidation, for example silicon nitride ; The structuring of the masking layer deposited over the entire area is carried out photolithographically . As a result of diffusion under the masking layer from the side, however, this process cannot create any sharp boundaries, only layer transitions (cf. “Bird's Beak” in the LOCOS process).

In the 1990s, these "LOCOS techniques" were the grave insulation (Engl. Shallow trench isolation , STI) replaced. The reason for this was the increased demands on the packing density and the planarity of the surface, especially for the photolithographic structuring in subsequent process steps. The LOCOS techniques had decisive disadvantages due to the type of oxide growth during thermal oxidation (bird's beak etc.) and further developments of the process that minimized these disadvantages became too complex and thus too expensive.

Nevertheless, thermal oxidation is also used in the insulation production of current ICs. It is used for example in the grave isolation as part of the process for forming a thin oxide layer with good electrical properties, by TEOS - are not reached or HDP oxide (HDP stands for English high density plasma , dt. High-density plasma ). There are also processes for the production of silicon-on-insulator wafers (SOI wafers), in which a thermal oxide is first produced on a wafer and later connected to another wafer ( wafer bonding ).

Modern MOS transistors

Cross-section through MOS field effect transistors as they were current at the beginning of the 2000s.

The insulation layers produced by the thermal oxidation of silicon had a decisive influence on the realization of the first field effect transistors with insulated gate ( IGFET ). The principle of the field effect transistors was already described in the late 1920s by scientists such as JE Lilienfeld and O. Heil . Due to the lack of manufacturing processes at the time that would provide sufficiently pure semiconductor crystals or insulation layers, these ideas could not be put into practice at the time. Only in June 1960, the Bell Labs employees published Dawon Kahng and Martin Atalla first time a functioning MOSFET (Engl. Metal-oxide-semiconductor field-effect transistor , dt. MOSFET ).

A MOSFET consists of a thin layer of thermally produced silicon dioxide on a p- or n-doped silicon single crystal and a metal layer (later also doped polycrystalline silicon) over the oxide layer, the gate electrode. This metal-insulator-semiconductor capacitor is an important part of the field effect transistor, because electrons or holes can be collected at the silicon-silicon dioxide interface via the gate voltage, so that a conductive channel is formed between the source and drain electrodes . However, the first MOSFETs had poorly reproducible electrical properties and in some cases did not run stably. Despite the efforts of numerous companies, the cause of these effects was only found in 1965 by employees of the Fairchild Semiconductor company . Sodium impurities (more precisely positively charged sodium ions) in the oxide and at the interface influenced the threshold voltage and thus the electrical behavior of the transistors. After identifying the alkali ions as the source of instability, much time and effort has been devoted to analyzing, removing and controlling these ionic contaminants. These include so-called CV measurements ( capacitance-voltage measurements ), which allow statements to be made about possible charges. Numerous methods for binding ions ( gettering ) or protective masking have also been developed.

Even today (2009) thermally generated silicon dioxide is used by most manufacturers as a gate material. The layer thicknesses are now in the range of 1–3 nm and are produced by rapid thermal oxidation . With these thin layers, however, the losses due to tunnel currents increase . By switching to gate materials with a higher dielectric constant than silicon dioxide ( high-k dielectric ), the thickness of the insulation layer can be increased again and the losses due to tunnel currents can thus be reduced.

Polysilicon and metal silicides

In addition to the oxidation of single crystal silicon similar processes in the semiconductor industry are also used for thermal oxidation of polysilicon and metal silicides , z. B. tungsten disilicide (WSi 2 ), cobalt silicide (CoSi 2 ), used.

The process of thermal oxidation of polysilicon is essentially identical to that of silicon single crystals. Due to the polycrystalline structure, it is not possible to distinguish between different crystal orientations during the oxidation, and the process is influenced by the thickness of the polysilicon film itself and the size of the polysilicon grains. The oxidation rates of undoped polysilicon are generally between those of {100} - and {111} -oriented silicon single crystals. In most applications, however, the polysilicon layers are heavily doped prior to oxidation, which changes the oxidation kinetics. In the case of heavily p-doped polysilicon, the oxidation rates are significantly higher; However, this effect, which is reinforced by impurities, is less than in silicon single crystals and is most evident at low process temperatures (<1000 ° C). Thermal oxidation of polysilicon is used, among other things, for the electrical insulation of different polysilicon layers, as used in a large number of VLSI applications, for example in dynamic RAM , erasable programmable memories ( EPROM ), charge-coupled devices ( CCDs ) or switched capacitors Circuits .

Metal silicides are used in semiconductor technology because of their high electrical conductivity for contacting doped silicon areas (e.g. source and drain contact) and polysilicon (e.g. gate). The oxidation of metal silicides can be used, for example, in MOSFETs for the electrical insulation of the gate electrode from subsequent layers. The oxidation converts the metal silicide into silicon dioxide. The growth kinetics of the SiO 2 layer, similar to the oxidation of silicon, depends on the material transport and diffusion of the oxidizing agent (O 2 or H 2 O) and the reaction itself. To produce well-insulating, that is to say metal-ion-free, oxide layers, a sufficiently high supply of silicon to the silicon oxide / metal silicide interface must be ensured.

literature

  • Ulrich Hilleringmann: Silicon semiconductor technology: Basics of microelectronic integration technology . 5th edition. Vieweg + Teubner, 2008, ISBN 3-8351-0245-1 , Chapter: The thermal oxidation of silicon .
  • Dieter Sautter, Hans Weinerth: Lexicon Electronics and Microelectronics . Springer, 1997, ISBN 3-540-62131-8 , pp. 755 ( limited preview in Google Book search).
  • Jan Albers: Basics of integrated circuits: components and microstructuring . Hanser Fachbuchverlag, 2006, ISBN 3-446-40686-7 , Chapter: Thermal Oxidation , p. 84–93 ( limited preview in Google Book search).
  • Richard C. Jaeger: Introduction to Microelectronic Fabrication . Prentice Hall, Upper Saddle River 2001, ISBN 0-201-44494-1 , Chapter: Thermal Oxidation of Silicon .
  • Sami Franssila: Introduction to microfabrication . John Wiley and Sons, 2004, ISBN 0-470-85106-6 , pp. 143-147 .
  • Marc J. Madou: Fundamentals of microfabrication . CRC Press, 2002, ISBN 0-8493-0826-7 , pp. 131-134 .

Web links

Commons : Thermal Oxidation  - Collection of Images, Videos and Audio Files

Individual evidence

  1. ^ Howard R. Huff, David C. Gilmer: High dielectric constant materials . Springer, 2005, ISBN 3-540-21081-4 , pp. 67 f . ( limited preview in Google Book search).
  2. Sami Franssila: Introduction to microfabrication . John Wiley and Sons, 2004, ISBN 0-470-85106-6 , pp. 315 .
  3. Ulrich Hilleringmann: Silicon semiconductor technology: Basics of microelectronic integration technology . 5th edition. Vieweg + Teubner, 2008, ISBN 3-8351-0245-1 , p. 31 .
  4. W. Ahmed, E. Ahmed, DG Armor: A comparative investigation of the oxidation of silicon using H 2 / O 2 , TCA / O 2 and HCl / O 2 mixtures . In: Advanced Materials for Optics and Electronics . tape 2 , no. 4 , 1993, p. 165-173 , doi : 10.1002 / amo.860020403 .
  5. ^ BE Deal, AS Grove: General Relationship for the Thermal Oxidation of Silicon . In: Journal of Applied Physics . tape 36 , no. 12 , 1965, p. 3770-3779 , doi : 10.1063 / 1.1713945 .
  6. M. Liu, et al .: Two-dimensional modeling of the self-limiting oxidation in silicon and tungsten nanowires . In: Theoretical and Applied Mechanics Letters . 6, No. 5, 2016, pp. 195–199. doi : 10.1016 / j.taml.2016.08.002 .
  7. a b Alexandra Ludsteck: Optimization of gate dielectrics for MOS technology . ( [1] [accessed on May 14, 2010] Dissertation; Universität der Bundeswehr München, Faculty of Electrical Engineering and Information Technology, 2005).
  8. ^ A b Hisham Z. Massoud: Growth Kinetics and Electrical Properties of Ultrathin Silicon Dioxide Layers . In: Howard R. Huff, H. Iwai, H. Richter (ed.): Silicon Materials Science and Technology X . The Electrochemical Society, 2006, ISBN 978-1-56677-439-0 ( limited preview in Google book search).
  9. Comparison Vertical vs. Horizontal furnaces. Crystec Technology Trading GmbH, accessed on December 20, 2009 (images of vertical and horizontal furnaces from the manufacturer Koyo Thermo Systems Co., Ltd.).
  10. ^ Semiconductor and Solar Thermal Processing Solutions. (No longer available online.) Coastal Technologies LLC, archived from the original on July 15, 2010 ; Retrieved December 20, 2009 (images of vertical and horizontal furnaces from Coastal Technologies LLC). Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice. @1@ 2Template: Webachiv / IABot / www.coastal-technologies.com
  11. The information is only a guide and may vary depending on the process parameters.
  12. Badih El-Kareh: Fundamentals of semiconductor processing technology . Springer, 1994, ISBN 0-7923-9534-4 , pp. 40 .
  13. Sami Franssila: Introduction to microfabrication . John Wiley and Sons, 2004, ISBN 0-470-85106-6 , pp. 145 .
  14. Thomas Giebel: Basics of CMOS technology . Vieweg + Teubner Verlag, 2002, ISBN 3-519-00350-3 , pp. 115 ( limited preview in the Google book search - representation of the segregation behavior of boron, phosphorus and arsenic).
  15. HR Huff, U. Gosele, H. Tsuya: Semiconductor Silicon . Electrochemical Society, 1998, ISBN 1-56677-193-5 , pp. 179-189 .
  16. CJ Frosch, L Derick: Surface Protection and Selective Masking during Diffusion in Silicon . In: Journal of The Electrochemical Society . tape 104 , no. 9 , 1957, pp. 547 , doi : 10.1149 / 1.2428650 .
  17. Patent US2804405 : Manufacture of silicon devices. Registered December 24, 1954 , published April 27, 1957 , inventors: L. Derick, CJ Frosch.
  18. Chih-Tang Sah: Evolution of the MOS transistor - From conception to VLSI . In: IEEE Proceedings . tape 76 , 1988, pp. 1280-1326 .
  19. Patent US2899344 : Applied on April 30, 1958 , published on August 11, 1959 , inventors: MM Atalla, EJ Scheibner, E. Tannenbaum.
  20. D. Kahng, MM Atalla: Silicon-silicon dioxide field induced surface devices. In: IRE-AIEE Solid-state Device Res. Conf., (Carnegie Inst. Of Technol., Pittsburgh, PA). 1960.
  21. Patent US3102230 : Electric field controlled semiconductor device. Registered May 31, 1960 , published August 27, 1963 , inventor: D. Kahng.
  22. Patent US3056888 : Semiconductor triode. Registered on August 17, 1960 , published October 2, 1962 , inventor: MM Atalla.
  23. ^ EH Snow, AS Grove, BE Deal, CT Sah: Ion Transport Phenomena in Insulating Films . In: Journal of Applied Physics . tape 36 , no. 5 , April 1965, p. 1664-1673 , doi : 10.1063 / 1.1703105 .
  24. LM Terman: An investigation of surface states at a silicon / silicon oxide interface employing metal-oxide-silicon diodes . In: Solid-State Electronics . tape 5 , no. 5 , 1962, pp. 285-299 , doi : 10.1016 / 0038-1101 (62) 90111-9 .
  25. DR Kerr, JS Logan, PJ Burkhardt, WA Pliskin: Stabilization of Si02 Passivation Layers with P2O5 . In: IBM Jr. of Res. And Dev . tape 8 , no. 4 , 1964, pp. 376-384 .
  26. ^ RJ Kriegler, YC Cheng, DR Colton: The Effect of HCl and Cl 2 on the Thermal Oxidation of Silicon . In: Journal of The Electrochemical Society . tape 119 , no. 3 , February 1972, p. 388-392 , doi : 10.1149 / 1.2404208 .
  27. JV Dalton, J. Drobek: Structure and Sodium Migration in Silicon Nitride Films . In: Journal of The Electrochemical Society . tape 115 , no. 8 , 1968, p. 865-868 , doi : 10.1149 / 1.2411450 .
  28. AH Agajanian: MOSFET Technologies: A Comprehensive Bibliography. Plenary, New York 1980.
  29. ^ Mohamed Gad-el-Hak: The MEMS handbook . CRC Press, 2002, ISBN 978-0-8493-0077-6 , pp. 15-6 f .
  30. ^ Roland Albert Levy: Microelectronic materials and processes . Springer, 1989, ISBN 978-0-7923-0154-7 , pp. 97 f .
  31. ^ Friedemann Völklein, Thomas Zetterer: Practical knowledge of microsystem technology . Vieweg + Teubner, Wiesbaden 2006, ISBN 978-3-528-13891-2 , pp. 53 .
This version was added to the list of articles worth reading on June 2, 2010 .