Patent Grant
7494839
The present invention relates to a method for manufacturing a semiconductor component.
Semiconductor elements, and membrane sensors, in particular, as well as methods for producing membrane sensors on the base of semiconductor substrates such as silicon wafers are already known. For instance, flat porous membrane regions are arranged on the semiconductor substrate as substrate for sensor structures, and a cavity is produced underneath the membrane, e.g., for the thermal insulation of the membrane.
The membrane sensors currently on the market are mostly fashioned as thin film membrane sensors. For this purpose, layer systems in thicknesses of between several 10 nm and several μm are deposited on a carrier substrate, and the carrier substrate is subsequently removed in predefined areas in order to obtain self-supporting membrane areas. The structural elements of the sensor can then be arranged in the center of the membrane.
Surface micromechanics (SMM), in which a sacrificial layer is generally used that is deposited on the front side of a carrier substrate before membrane deposition, is another possibility for exposing the membrane. The sacrificial layer is later removed from the front side of the sensor through “detachment openings” in the membrane, whereby a self-supporting structure is created. These surface micromechanical methods are comparatively costly, on account of the necessity for separate sacrificial layers.
Published German patent document DE 100 32 579 discloses a method for manufacturing a semiconductor element and also a semiconductor element produced according to the method, in which, e.g., for a membrane sensor, a layer of semiconductor carrier material that was rendered porous is arranged above a cavity. Two layers having different porosities are formed to produce the cavity using appropriate etching parameters. Whereas the first layer has lower porosity and seals up during a subsequent first annealing step, the porosity of the second layer increases during the annealing step in such a way that a cavity is formed. In a second process step, at a higher annealing temperature, a relatively thick epitaxy layer as second membrane layer is grown on top of the first membrane layer formed from the first porous layer.
It may also be provided that a thin epitaxy layer be grown during the first annealing step in order to ensure complete sealing of the porous first layer, which is used as starting layer for the epitaxy growth of the thick epitaxy layer. In this context, a lower growth rate is selected at a lower temperature compared to the subsequent deposition of the thick epitaxy layer.
As a result of the mentioned measures, the construction of an SMM semiconductor element can be simplified considerably since no additionally deposited sacrificial layer is required and, furthermore, the membrane itself or an essential portion of the membrane is produced from semiconductor substrate material.
However, tests have shown that at least partially porous membrane may get damaged during manufacture, or that damage may not always be reliably prevented under normal operating conditions. To avoid damage to the membrane during manufacture or in regularly occurring application cases, published German patent document DE 101 38 759 provides a method for manufacturing a semiconductor element having a semiconductor substrate, in which the semiconductor substrate receives a different doping in the region of the porous membrane layer than the doping of the region of the subsequent cavity. After doping, the semiconductor material of the membrane layer is rendered porous, and the semiconductor material underneath the semiconductor material, rendered porous, is at least partially removed or relocated to provide a cavity.
Published German patent document DE 100 30 352 discloses a micromechanical component which has a support body made from silicon and a regionally self-supporting membrane which is joined to the support body. The membrane is regionally and superficially provided with at least one stabilizing element for support. To form the regionally self-supporting membrane, the silicon is rendered porous in a first region and is selectively removed via an etch opening once the membrane layer has been deposited.
The present invention provides a manufacturing method for a micromechanical semiconductor element as well as a semiconductor component fabricated by this method. To produce the semiconductor element, it is provided that a patterned stabilizing element having at least one opening be produced on a semiconductor substrate. The opening is arranged such that it allows access to a first region in the semiconductor substrate, this first region having a first doping.
Furthermore, at least a portion of the semiconductor material having the first doping is to be dissolved out of the first region of the semiconductor substrate. In addition, using a first epitaxy layer deposited on the stabilizing element, a membrane is produced above the first region. In a further method step, at least a portion of the first region is used to produce a cavity underneath the stabilizing element. The present invention thus provides the patterned stabilizing element by means of a second epitaxy layer, which is deposited on the semiconductor substrate.
By growing the first epitaxy layer on the stabilizing element, the produced membrane advantageously has greater stability since, first of all, the stabilizing element supports the first epitaxy layer and, secondly, a uniform growing of the first epitaxy layer on the stabilizing element is possible without crystal faults.
In accordance with the present invention, it may be further provided that the epitaxy conditions be selected such that the first epitaxy layer, starting from at least a portion of the stabilizing element, seals the at least one opening above the first region in a lateral and/or vertical direction. The epitaxy conditions are advantageously chosen such that the epitaxial growth occurs only on the stabilizing element and not on the first region. One way of achieving this is by separate passivation of the first region. Within the framework of this epitaxy process, it may be provided that the stabilizing element be largely covered.
The cavity of the semiconductor element may be produced in an additional method step by a thermally induced relocation process of the semiconductor material of the first region. It may be provided, in particular, that this relocation be implemented after the epitaxial growth of the first epitaxy layer and thus after production of the membrane.
In an example embodiment of the present invention, the stabilizing element is built from the second epitaxy layer. In doing so, the second epitaxy layer is provided with openings in such a way that the doped semiconductor material of the first region is able to be removed through these openings. One possibility for producing these openings is to expose the second epitaxy layer at least regionally to a light source and remove the areas of the second epitaxy layer exposed to light in this manner. In this context, it may be provided that the removal of the light-exposed material of the second epitaxy layer and the selective removal of the semiconductor material provided with the first doping from the first region be implemented in the same method step.
However, it may also be provided that the illuminated areas on the semiconductor substrate or on the second epitaxy layer are precisely not attacked during an etching process. As a result, the illuminated regions would remain standing in the form of webs, for instance above a first region rendered porous by etching.
The second epitaxy layer deposited on the surface of the semiconductor substrate may be patterned locally, for instance via an additionally applied passivation layer. It may be provided here that the second epitaxy layer patterned in this manner form a cohesive second region above the first region. This second region may be embodied in the form of a lattice, for instance, and/or in the form of interconnected webs.
In an example embodiment of the present invention, an SiC layer is epitaxially deposited on the semiconductor substrate and subsequently patterned. If this patterned SiC layer is not processed further, the first epitaxy layer is able to grow on the SiC layer in polycrystalline form in a further epitaxy step. In contrast, in a thermal treatment of the patterned SiC layer, it is possible for the second epitaxy layer to grow in a monocrystalline manner.
The stabilizing element advantageously has at least one web and/or lattice above the cavity. As an option, the stabilizing element may also have an edging at the rim of the cavity. When the stabilizing element is embodied above the cavity, it may be provided that the thickness of the stabilizing element vary. Furthermore, it is possible that the edging has a greater thickness compared to the lattice and/or the at least one web.
The manufacturing method according to the present invention provides the particular advantage that the rigidity of the membrane may be adjusted via the positioning of the openings on the stabilizing element, and/or that the webs may be designed by varying different manufacturing parameters. The parameters during the epitaxial growth, the geometric arrangement of the openings and/or the lateral or vertical extensions of the webs may be considered manufacturing parameters. For instance, webs having a triangular cross section or having different vertical and/or lateral extensions are able to be produced.
The lattice or the at least one web in the surface area of the semiconductor substrate may be created by combining at least two implantation steps. In the process, a second doping that is differentiable from the first doping of the first region may be produced in a second region of the semiconductor substrate. In a subsequent method step, this second region may then be further developed into the stabilizing element or into the lattice and/or a web. This may be done, for instance, in that only the semiconductor material is removed from of the first region during the following selective removal, but the material of the second region remains on or within the semiconductor substrate.
In an example embodiment of the present invention, a third region, which likewise has the first doping, is produced above the first region, having the first doping, in the semiconductor substrate. The doping of the third region has a higher doping concentration than the doping of the first region. Furthermore, it is provided that, above the first and the third regions, the second epitaxy layer be produced with a second doping that is differentiable from the first doping. By patterning the second epitaxy layer, it is possible to gain access to the third and finally to the first region through the epitaxy layer in a further method step. Afterwards an etching method is utilized to remove the semiconductor material, which method selectively dissolves out only the semiconductor material having the first doping, but will not attack the material having the second doping. By choosing the higher doping concentration in conjunction with an appropriate etching method, more semiconductor material is able to be dissolved out of the third region than is dissolved out of the first region. Such a method makes it possible to produce a virtually planar underside in the stabilizing element formed in this manner.
In another example embodiment of the present invention, the surface of the semiconductor substrate is patterned across at least a portion of the first region having the first doping. Afterwards, a second epitaxy layer having a second doping is produced on the surface of the patterned semiconductor substrate and likewise patterned. It is provided in this context that the patterned second epitaxy layer form the stabilizing element.
The semiconductor material is then selectively removed from the first region, and the first epitaxy layer is deposited on the first and second regions patterned in this manner.
It is advantageously provided that the first and the second doping regions have different dopings. For instance, the first region may have a p− doping, and the second region, the edging or the lattice and/or the at least one web may have an n− doping.
An example embodiment of the present invention may provide that both the first epitaxy layer and the stabilizing element or the lattice or the at least one web have a monocrystalline form. However, it may additionally be provided that the first epitaxy layer and the stabilizing element have a polycrystalline design.
a-1d illustrate various stages of a conventional method for manufacturing a micromechanical membrane.
a-2b illustrate the various stages of an exemplary embodiment of the method of the present invention.
a-3b illustrate the steps of a treatment of the surface to prevent the deposition of epitaxy material according to the present invention.
a-4b illustrate the steps of an additional exemplary embodiment for the manufacture of a membrane according to the present invention.
a-5f illustrate various example embodiments of the webs on which the membrane is deposited according to the present invention.
a-6c show the steps of an example embodiment of a method for producing webs or lattices above a cavity.
a-7c show the steps of another example embodiment of a method for producing webs or lattices.
a-8b show the steps of another example embodiment of a method for producing webs or lattices.
a-9b show the steps of another example embodiment of a method for producing webs or lattices.
a-10c show the steps of another example embodiment of a method for producing webs or lattices.
a-11b and 12a-12c show example embodiments of various lattice and hole geometries according to the present invention.
a-13g show various example embodiments of a method for producing a lattice having a desired cross-section profile.
a-14c show the steps of an example embodiment of a method for producing a lattice.
a-15d show the steps of an example embodiment of a method for producing a lattice and/or a membrane.
a-16c show the steps of an example embodiment of a method for producing a lattice and/or a membrane.
a-17d show the steps of an example embodiment of a method for producing a lattice and/or a membrane.
As shown in
When producing a monocrystalline epitaxy layer, the deposition of a first thin epitaxy layer (thinner than 1 μm, preferably 200 to 600 μm), during which additional semiconductor material is offered for the complete sealing of the starting layer immediately at the outset, during or at the end of the first annealing process, may remedy this situation. It may be the same semiconductor material which is already present in layer 110 and/or is used in epitaxy layer 140 that is deposited. If semiconductor substrate 100 is made of silicon, for instance, such a method gives the offered silicon atoms sufficient time to arrange themselves according to the silicon crystal of the base. A lower temperature and a reduced growth speed compared to the subsequent, thick epitaxy layer are advantageously selected during the growing of the thin epitaxy layer. As an example, the temperature during the deposition of the thin epitaxy layer is approximately 900 to 1000° C., whereas the growth speed is selected to be less than 0.5 μm/min.
After the growth of this thin epitaxy layer and a possible continuation of the annealing process, the substrate is exposed to a higher temperature (preferably 1100 to 1200° C.) so that a second, thick epitaxy layer 140 may then be grown there (a few μm). Prior to the growing of the second epitaxy layer, further annealing at the higher temperature may also be carried out in order to heal any possible crystal faults in the sealed starting layer. Because of its relatively great thickness—compared to the starting layer and the first epitaxy layer—the second epitaxy layer forms the actual membrane. This membrane may then be used for a pressure sensor, for instance or, following further patterning, for an acceleration sensor as well.
Depositing a single-crystalline, micromechanical membrane on a porous starting layer has various weak points, which may have a negative effect on the production or the service life of the membrane. For example, cracks may form in the porous starting layer, such cracks spreading to the epitaxy layer deposited thereon. Furthermore, transitions may occur from the, for instance, p+ doped starting layer 160 to an edging of the starting layer. In an overlap of both dopings, the p+ doping is reduced to a p− doping (see 130,
To produce a monocrystalline membrane according to the present invention, a first region 220 and a first portion of a second region 210 having different dopings are produced on a semiconductor substrate 200, as shown in
Using a suitable etching process, for instance anodizing or electropolishing, first region 220 is etched to be rendered porous. As already described, first region 220 may be both a portion of untreated semiconductor substrate 200 and an additionally doped region. The latter has the advantage of allowing a sharper delimitation in the production of porous region 220. Whereas first region 220 is etched to be rendered porous up to a depth of 5 to 20 μm, for instance, starting from the surface of semiconductor substrate 200, second region 210 essentially will not be changed by the etching process. In an appropriate selection of the lateral extension of second region 210 on the surface of semiconductor substrate 200, first region 220 may be rendered porous by etching underneath second region 210 as well. For the further processing, an exemplary embodiment provides that first region 220 have high porosity, e.g., having pore sizes of 1 nm up to several μm in diameter. As an alternative, high porosity may be achieved by producing very many small pores (˜5 nm) or a few very large pores (e.g., up to several 100 nm). As in the production of large pores, a lot of silicon is etched out as a result. If semiconductor substrate 200 is made of silicon, a (natural) oxide layer can be found on the surface of the untreated semiconductor substrate. For this reason, a reduction of the (natural) oxide layer on the silicon surface is obtained in a following process step by short annealing of the semiconductor substrate or the silicon substrate being carried out under a hydrogen atmosphere. As an alternative, an “HF load” method or an HF GPE (gas phase etching) method with subsequent low-temperature epitaxy may be used to achieve the same effect.
Afterwards the semiconductor substrate is heated to growth temperature (1000 to 1200° C.) before the deposition of a monocrystalline epitaxy layer 240 is initiated. The epitaxy layer grows predominantly on second regions 210 and 230. In an exemplary embodiment, regions 210 are formed as monocrystalline webs to promote the growth of the epitaxy material. The growth not only occurs vertically, i.e., perpendicular to the surface, but laterally as well, so that the regions between individual webs 210 are sealed by a monocrystalline layer.
The highly porous silicon between monocrystalline webs 210 is advantageous here since it prevents the growth of silicon inside the cavity region. If the cavity were completely free, depending on the growth conditions, silicon would be able to grow on the cavity walls from the beginning of the epitaxy growth. However, due to the initially porous surface between webs 210, which relocates over course of the epitaxy process, it is prevented that silicon grows between webs 210 or that a significant portion of the epitaxy material enters cavity 250.
Of course, the production of an epitaxy layer having a cavity situated underneath it by means of a “pororization” as it is described in the present invention may also be carried out when other materials or semiconductor materials are used and is not restricted to the use of silicon. However, this requires that these other materials or semiconductor materials are also able to be rendered porous.
In another exemplary embodiment of the present invention, the region between and underneath the webs is not etched to be rendered porous, but is completely dissolved out. This may be accomplished by electropolishing, for instance, and the porosity is increased by varying the etching parameters (such as increasing the current, reducing the HF concentration), to such a degree that it reaches 100 percent. As described earlier, the webs are now able to grow together, which entails the danger of silicon growing inside the cavity as well. To avoid this, the inside of the cavity may be protected by a layer that prevents silicon from growing there.
Such a protection is able to be achieved, for instance, by the application of a silicon nitride layer (SiN) 360 on the surface of second regions 310, as shown in
Another possibility for selectively preventing the growth of silicon is to apply a thin (<60 nm) silicon oxide layer (SiO) on those regions that are not to be electropolished prior to the application of silicon nitride layer (SiN) 360. After webs 310 have been exposed through a suitable etching process, for instance by electropolishing, an oxide may be produced on the exposed regions via thermal oxidation, as in the already discussed example. However, it must be taken into account that this oxide has to be thicker than the oxide underneath the SiN overlay-mask. The nitride may later be selectively removed with respect to the oxide. As an alternative, a dry-etching process in which the oxide and the nitride have the same etching rates may be carried out as well. By suitable selection of the etching time it may be ensured that a sufficiently thick oxide layer remains on the cavity walls once the oxide that was situated underneath the SiN has been dissolved. A selective epitaxy may then be carried out analogously to the above exemplary embodiment.
In another exemplary embodiment in which the semiconductor substrate is made of silicon, for instance, the silicon is rendered porous by etching between and underneath the webs. Afterwards a thin layer of oxide, which is a few nm thick, for example, will be produced on all silicon surfaces (i.e., wafer surface as well as on the surface of the pore walls).
This may be accomplished by thermal oxidation, for instance, or by oxygen-plasma treatment, by chemical oxidation or some other type of treatment as well. This oxide prevents the relocation of the porous silicon during additional high-temperature steps as they are required in epitaxy or other kiln processes. The oxide on the wafer surface may then be removed by means of a brief HF dip using diluted hydrofluoric acid (HF). Due to the surface tension the hydrofluoric acid does not penetrate the pores so that the oxide layer on the pore walls remains as before. A corresponding description of such a pretreatment for stabilizing the porous silicon for a subsequent epitaxy growth can be found, for instance, in the article “Low-pressure vapor-phase epitaxy of silicon on porous silicon,” Material Letters 94 (1988), by L. Vescan et al. An epitaxy layer may subsequently be grown, this layer growing predominantly on the monocrystalline webs. The surface is sealed by lateral growth. After the epitaxy an access hole may be etched through the epitaxy layer from the front side, for instance, or also from the back side of the semiconductor substrate, and the oxidized porous silicon be selectively dissolved out through the access hole. As an option, the access hole may also be sealed again after the removal process, but it must be taken into account that the pressure prevailing during sealing of the access hole defines the reference pressure in the cavity.
The membrane produced in this manner may be used, for instance, for a pressure sensor having piezoresistive resistors. For this purpose a circuit may be integrated adjacent to or on top of the membrane. If an access hole for the selective removal of the semiconductor material in the second region was produced on the backside, a backside sensor or differential-pressure sensor is obtained, which has a precisely defined membrane thickness compared to the related art. If an access hole is made from the front, this hole must be sealed again in a pressure-tight manner for the pressure-sensor application.
In another exemplary embodiment, if the surface is too rough after the holes have become overgrown, a step for smoothing the surface may generally be performed as an option. One possible method for achieving the smoothing is chemical-mechanical polishing (CMP).
In another exemplary embodiment the membrane is patterned in the form of resonator structures. The use of such structures makes is possible to realize acceleration sensors and/or yaw-rate sensors, for instance.
One possibility for stabilizing the webs before the epitaxy layer is grown is to optionally produce supports underneath the webs, which melt away during subsequent high-temperature steps (epitaxy or annealing steps (oxidation diffusion) for the production of integrated circuits) due to relocation of the silicon atoms. As can be seen from
Instead of individual webs, the second region above the cavity may also be arranged in the form of a lattice, porous regions 510 and webs 500 alternating with each other. In
Various methods are available to realize the webs or the lattice on which the monocrystalline membrane is deposited in a later method step. For instance, in one exemplary embodiment the webs or the lattice are/is not produced by local n− doping, but by local amorphization of the single-crystalline Si substrate. As shown in
In accordance with the amorphous basis, epitaxial layer 590 (
In a variant of this exemplary embodiment, additional annealing may be carried out before the epitaxy layer is grown. Due to this annealing, the amorphous webs are able to recrystalize and relocate in the form of a single-crystalline lattice. This recrystallization step makes it possible to produce a monocrystalline Si epitaxy on the lattice webs.
In another exemplary embodiment for producing an Si lattice on porous silicon, the illumination selectivity of the anodization process may be utilized. As illustrated in
To produce a holographic lattice, a laser beam 735 as shown in
In an extension of the last exemplary embodiment, it may also be provided to first produce an n− doped Si epitaxy layer 755 on a p− doped Si substrate 750, as it is shown in
Another example for producing a lattice on porous silicon is the use of a patterned, n− doped Si epitaxy layer. To produce such a patterned n− doped Si epitaxy layer 805, an n− doped Si epitaxy layer 805 is deposited on a p− doped Si substrate 800 as shown in
In a further exemplary embodiment, as shown in
An additional exemplary embodiment utilizes different porosities in the lattice and in the cavity region. Such an adaptation of the porosities in the mentioned regions allows the relocation of the porous silicon into a cavity or the growth of the epitaxial silicon membrane in a more optimal manner. For instance, it may be provided that a higher or lower porosity be produced in the cavity region than in the region of the holes. In addition to a sharp separation of the different porosity regions, a porosity gradient is also conceivable.
However, in the starting phase of the epitaxy for forming the membrane, attention must be paid that the cavity layer forms a sufficiently stable basis, so that an excessively high porosity in the cavity layer would be disadvantageous. Furthermore, it would be desirable to produce high porosity in the holes in the lattice region as well since this speeds up the formation of the holes at the beginning of the epitaxy growth. By appropriate selection of a (high) porosity in the lattice area, it may thus be prevented that the growth begins on the relocated silicon in the lattice holes. For if at least a portion of the epitaxial growth were to begin on the silicon between the lattice webs during the relocation, this might cause crystal faults, which would propagate in the membrane layer, for example as stack faults.
To optimize the mechanical properties of the epitaxial Si membrane, the lattice and hole geometries may be varied locally, as is shown in
Another alternative is to broaden the lattice webs at the edge of the membrane. In addition, however, the (hole) lattice may also be arranged across only a portion of the membrane region, so that a self-supporting membrane is produced as shown in
The mechanical properties of the membrane such as the resonant frequency and/or the rigidity may also be varied by correspondingly adapted geometries of the lattice webs and holes. One possibility for increasing the rigidity is to omit holes in the center of the membrane. However, reliable undercut-type etching of the holes must be ensured. This requirement results in a limit for the maximum number of holes that may be omitted.
If electrically active elements are applied on the membrane, it may also be useful to locally adapt the lattice and hole geometries. Piezoresistive resistors such as for a pressure sensor, for example, may be located in a region that stands out as a result of an especially defect-free epitaxy (avoidance of leakage currents and shunts via so-called diffusion pipes). This may be accomplished in that an especially low number of lattice holes, to be overgrown epitaxially, is present in these regions and/or that an especially adapted lattice and hole geometry is used, which becomes overgrown in a particular satisfactory manner.
Of course, it is also possible to superpose a plurality of different lattice geometries and lattice profiles on top of one another and combine them in this manner.
A schematic cross section through a membrane region having different lattice geometries is shown in
In addition to the arrangement of the holes in the membrane region, it is also possible, as already mentioned, to adapt the cross-section profiles of the lattice to the requirements of the membranes to be produced. In the least complicated case, as shown in
Suitable selection of the pattern of mask 910 (such as a gray-tone mask of photoresist or Si oxide) in conjunction with an appropriate implantation method 915 allows the lattice profile to be influenced in a variety of ways as is shown in
In addition to a simple implantation, a multiple implantation using different masks and/or implantation energies is conceivable as well. One possible result of a twofold implantation with an increase in the implantation energy in the second implantation step (
In general, multiple implantations using appropriate masks and varying implantation energies are able to produce virtually unlimited lattice cross-section profiles.
a and 14b illustrate another possibility for producing a lattice on or within a region rendered porous by etching. In the process (shown in
In a further exemplary embodiment, the lattice and membrane production may be accomplished via an additional p+ doping. This additional p+ doping is able to broaden and improve on the method described in published German patent document DE 10 138 759, for instance. As illustrated in
A further advantage of using an additional p+ doping results from a better adaptation of the anodization during the production of the lattice. Without the p+ doping, the p− doped substrate is etched beneath lattice 1070 in an undercut manner in the form of a beak 1080, as illustrated in
It has shown to be especially advantageous if not only semiconductor layer 1000, but p+ doped layer 1010 as well is made of p− doped silicon, and if n− doped layer 1020 is made of silicon. However, the use of other semiconductor materials is conceivable as well.
Another exemplary embodiment for producing a lattice and/or a membrane on a semiconductor substrate relates to the production of p+ doped regions in an n− doped layer. In the process a planar n− doped (monocrystalline) layer 1100 is first deposited on a p− doped substrate 1100, as shown in
The structure thus produced and illustrated in
In a further exemplary embodiment (shown in
An anodizing process follows, which once again does not attack the n− doping, but renders the p− doping of the substrate porous by etching, ultimately forming a region 1230, which etches the n− doped lattice webs in an undercut manner (
Due to the manner in which the n− doped lattice is produced in this exemplary embodiment, only geometries that form cohesive lattice webs after physical etching may be used, for instance a chessboard-type geometry (see, e.g.,
Silicon is utilized as semiconductor material in the afore-described manufacturing method for producing a lattice above a porous layer. It should be noted, however, that materials or semiconductor materials other than silicon, which are able to be rendered porous by etching via an electrochemical method, for instance, may be used in the manufacturing method.
| Number | Date | Country | Kind |
|---|---|---|---|
| 103 58 859 | Dec 2003 | DE | national |
| 10 2004 036 032 | Jul 2004 | DE | national |
| Number | Name | Date | Kind |
|---|---|---|---|
| 4766666 | Sugiyama et al. | Aug 1988 | A |
| 5242863 | Xiang-Zheng et al. | Sep 1993 | A |
| 5445991 | Lee | Aug 1995 | A |
| 5510276 | Diem et al. | Apr 1996 | A |
| 6376291 | Barlocchi et al. | Apr 2002 | B1 |
| 6743654 | Coffa et al. | Jun 2004 | B2 |
| 20030116813 | Benzel et al. | Jun 2003 | A1 |
| Number | Date | Country |
|---|---|---|
| 100 30 352 | Jan 2002 | DE |
| 100 32 579 | Jan 2002 | DE |
| 101 38 759 | Mar 2003 | DE |
| Number | Date | Country | |
|---|---|---|---|
| 20050181529 A1 | Aug 2005 | US |
