Semiconductor device manufacturing: process – Introduction of conductivity modifying dopant into... – Ion implantation of dopant into semiconductor region
Reexamination Certificate
1999-12-16
2001-02-20
Nelms, David C. (Department: 2818)
Semiconductor device manufacturing: process
Introduction of conductivity modifying dopant into...
Ion implantation of dopant into semiconductor region
C438S396000
Reexamination Certificate
active
06191013
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to solid electrolytic capacitors and methods for fabricating the same, and more particularly to solid electrolytic capacitors in which conductive polymers are used as solid electrolytes and which have a low equivalent series resistance.
BACKGROUND OF THE INVENTION
A solid state electrolytic capacitor is made from a porous pellet of sintered tantalum powder, a dielectric tantalum oxide layer formed on the surface of the sintered tantalum powder, a solid-state conductor impregnated into the volume of the pellet, and external connections such as silver paint, etc. The tantalum forms the positive electrode of the capacitor, and the solid-state conductor forms the negative electrode (also called the cathode or counter-electrode).
Manganese dioxide has been utilized as the cathode of choice for solid tantalum capacitors since the commercial introduction of this style of capacitor in the early 1950's. A key property of manganese dioxide is its self-healing ability. At defective portions of the dielectric film, the manganese dioxide becomes non-conductive. This is due to the manganese dioxide transforming to a lower manganese oxide because of joule heating at the defect site. This mechanism allows capacitors with low leakage currents to be produced. It also allows small dielectric defects that occur during manufacture and use to be isolated. However, if the dielectric defect is too large, the dielectric can crack. Manganese dioxide is a powerful oxidizing agent. When it comes in direct contact with tantalum through a crack in the oxide, the capacitor can ignite, leading to destruction of the capacitor and possible destruction of other components in the circuit. It is desirable to replace the manganese dioxide with a solid-state conductor that does not cause the tantalum to ignite while maintaining the self-healing ability.
The use of tantalum capacitors in high frequency circuits has become more important. This has led to the need for tantalum capacitors having low equivalent series resistance (ESR). The best manganese dioxide has a resistivity of 0.5 to 10 ohm-cm. It is desirable to replace the manganese dioxide with a solid-state conductor that has a lower resistivity. However, many highly conductive metals and oxides do not have a self-healing ability and thus are not suitable for solid-state tantalum capacitors.
Conductive polymers such as polypyrroles, polyanilines, and polythiophenes have resistivities 10 to 100 times less than that of manganese dioxide. Since they are much less powerful oxidizing agents than manganese dioxide, these materials do not cause the capacitor to ignite upon failure. Polypyrrole was shown to have a self-healing mechanism (Harada, NEC Technical Journal, 1996). Due to these favorable properties of conductive polymer compounds, these compounds are being investigated as possible replacement materials for manganese dioxide in solid-state tantalum capacitors.
Three methods have been used to deposit the conductive polymer in the porous tantalum pellet:
1. Chemical oxidative polymerization;
2. Electrolytic oxidative polymerization; and
3. Deposition of a polymer from solution followed by oxidation and/or doping.
In chemical oxidative polymerization, a monomer, an oxidizing agent, and a dopant are reacted inside the porous pellet to form the conductive polymer. Monomers include pyrrole, aniline, thiophene, and various derivatives of these compounds. The oxidizing agent can be either an anion or a cation. Typical anion oxidizers are persulfate, chromate, and permanganate. Typical cations are Fe(III) and Ce(IV). The best dopants are anions of strong acids such as perchlorate, toluenesulfonate, dodecylbenzenesulfonate, etc. The reaction between monomer, oxidizing agent, and dopant can take place in a solvent such as water, an alcohol, a nitrile, or an ether.
Several methods have been used to get the monomer, oxidizing agent, and dopant into the porous pellet and carry out the conversion to conductive polymer. In one method, the pellet is first dipped in a solution of the oxidizing agent and dopant, dried, and then dipped in a solution of the monomer. After the reaction is carried out, the pellet is washed and then the process is repeated until the desired amount of polymer is deposited in the pellet. In this method, it is difficult to control the morphology of the final polymer. It is also difficult to control the exact reaction stoichiometry between the monomer and the oxidizing agent. Control of this stoichiometry is critical to achieve the highest conductivity polymer (Satoh et al., Synthetic Metals, 1994). Cross contamination of the dipping solutions is a problem. Since the pellet must be dipped twice for each polymerization, the number of process steps is greatly increased. The excess reactants and the reduced form of the oxidizing agent need to be washed out of the part. This adds even more process steps and complexity to the process.
In a related method, the sequence is reversed so that the pellet is dipped in the monomer solution first and the solvent is evaporated away. The pellet is then dipped in the oxidizing agent/dopant solution and the reaction is carried out. This method suffers from all the disadvantages of the previous method. In addition, some monomer may be lost in the solvent evaporation step.
In yet another method, all components are mixed together and the pellet is dipped in the combined solution. This method reduces the number of dips and allows more precise control over the reaction stoichiometry. However, the monomer and oxidizing agent can react in the dipping bath, causing premature polymerization and loss of reactants. This adds some additional complexity and cost to the process. This is especially a problem with pyrrole monomer and Fe(III) oxidizing agents. To overcome this problem to some extent, the dipping bath can be kept at cryogenic temperature (Nishiyama et al., U.S. Pat. No. 5,455,736). However, use of cryogenic temperatures adds considerable equipment and operational complexity to the process. The pyrrole/Fe(III) can be replaced with a monomer/oxidizing agent combination that is less reactive; for example, 3,4-ethylenedioxythiophene and an Fe(III) salt of an organic acid (Jonas et al., U.S. Pat. No. 4,910,645).
In electrolytic oxidative polymerization, the monomer is oxidized to polymer at an electrode and the dopant is incorporated from the electrolyte. This polymerization method produces high conductivity polymer films. There is no chemical oxidizer to wash out of the film after polymerization.
Direct electrolytic oxidation of monomer to polymer is difficult because of the high resistance dielectric oxide layer. Various methods have been proposed to circumvent this problem. One method is to form the polymer on the tantalum and then to form the oxide layer (Saiki et al., U.S. Pat. No. 5,135,618). In another method, the polymer and the oxide layer are formed at the same time (Saiki et al., European Patent Application 0 501 805 A1). However, the electrolytes best suited for depositing conductive polymer and tantalum oxide films are quite different; therefore, these methods produce neither an optimum polymer nor an optimum oxide.
Another method is to deposit a thin film of conductive material by chemical methods, followed by contacting this layer with an electrode to carry out the electrolytic oxidative polymerization. Manganese dioxide prepared by pyrolysis of manganese nitrate (Tsuchiya et al., U.S. Pat. No. 4,943,892), manganese dioxide prepared by pyrolysis of permanganate (Kudoh et al., J. Power Sources, 1996), and conductive polymer prepared by chemical oxidative polymerization (Yamamoto et al., Electronics and Communications in Japan, 1993) have been used for this thin layer. Contacting this thin layer of conductive material with an auxiliary electrode is difficult to achieve in practice. Thus, Tsuchiya et al. propose bridging the anode lead to the conductive layer. This bridging layer must be removed after depositing the polymer by electrolytic oxidative polymerization
Hahn Randolph S.
Lessner Philip M.
Rajasekaran Veeriya
Banner & Witcoff , Ltd.
Dang Phuc T.
Kemet Electronics Corporation
Nelms David C.
LandOfFree
Process for improving leakage and dissipation factor of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process for improving leakage and dissipation factor of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process for improving leakage and dissipation factor of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2567575