Patent Application
20120032356
The invention relates to a method for the production of at least one alignment mark in a light-opaque layer on a semiconductor wafer, there, for positioning (alignment) and arranged in at least one scribing trench area. Also concerned is an alignment mark which has been produced according to the method. It is, on the one hand, directly described as a result of the method steps and, on the other hand, it is also protected by the production and the use of the method.
The described light-opaque layer (or light-absorbing layer) is, for example, a black, light-sensitive lacquer. Such light-absorbing layer protects, therein, for example light-sensible areas of circuits with optically integrated elements against incident light, and this layer is, preferably, structured. Examples for black photo resists are carbon-based black resist or pigment-based black resist which both can be used as light-absorbing layer (light shield). This layer is applied at least at the end of the production process onto a passivation layer. It concerns a non-transparent negative photo resist having planarizing properties.
A lacquer having a defined thickness, can also equalize the existing surface topography (small recesses or cavities are filled up with this lacquer). The result will then be a planar surface such that usual alignment marks cannot be recognized anymore by alignment optics because of the substantial covering height of the surfaces of the alignment marks (or of all projections of a respective alignment mark).
Because of the application of a light-opaque layer to be structured at the end of the production process, this layer covers circuits which have previously be inserted into the active areas of the semiconductor-wafer, and can, therein, further on protect the light-sensitive areas there, where no structuring is made, or can make accessible for the light certain light-sensitive areas as optical sensors (optically active elements) in the integrated circuits after the structuring, whereas it can protect other areas against incident light because of the on-going presence of the layer.
Light-opaque, for example shielding layers, so-called black, light-sensitive lacquers or “light-shield-photo resists”, as well as optical protection layers are applied in sensors and circuits of the semiconductor industry, according to their characteristics. The light-layer protects, thereby, for example light-sensitive areas of circuits outside of windows (introduced openings) in a structured, light-absorbing layer. The circuits may respectively be provided with one or several integrated optical element(s). The areas outside of the active optical elements may be protected against incident light. The light-absorbing layer may, however, being generally associated with circuits without optical elements within which light-sensitive areas are to be protected. The protection relates, therein, for example to the suppression of parasitic influences of a light incident onto the circuits. This light protection layer is, for example, produced as originally plain metal-, photo resist, or plastic layer which has reflective and/or absorbing properties depending on the materials used, and which may be structured during the processing sequence.
U.S. Pat. No. 3,969,751 (Drukaroff, RCA) describes a light-absorbing (or light-shielding) layer, for example out of blackened photo resist, blackened metal or plastics which contains carbon particles. The use of a light-absorbing layer in sensors is described, among others in U.S. Pat. No. 4,785,338 (Kinoshita, Canon). For producing this layer, materials like epoxy resin, silicone resin, urethane resin and a kind of black colour or black colouring agent are used.
US 2003/0030055 A1 (Nakano et al.) describes a colour filter containing a light-absorbing layer out of epoxy resin with black pigments. An opaque layer out of light-absorbing, black carbon particles and out of reflective titanium dioxide particles for integrated circuits is described in U.S. Pat. No. 5,031,017 (Pernyeszi, H P).
From US 2004/0032518 A1 (Benjamin, Tower Semiconductor), a light-shielding, structured layer is known with picture sensors wherein this layer contains a light-sensitive lacquer with carbon particles.
Known alignment marks for transparent photo resists consist mostly out of a layer (single layer) for example out of oxide, oxy-nitride or nitride. However, also combinations out of these are perceivable. They are always totally covered by the transparent photo resist. The photo resist protects, thereby, also the alignment marks against etching attacks.
In many cases, the structured layers are transparent to light or they are, like the photo resist, transparent at the wavelength of the alignment optics. Thereby, the alignment marks can be optically detected through the photo resist or the structured layer. A general problem with light-absorbing layers is, however, the exact alignment of these layers with respect to prior layers since the photo resists containing carbon or pigments are not transparent in the range of wavelengths of the visible light (350 nm to 800 nm).
In case alignment marks are placed in light-opaque materials, like metal of poly silicone, they cannot be recognized by the alignment optics so that the pinpointed Alignment of these layers to the prior layer is not possible. One manages this with a corresponding surface topography for the application of the at least light-absorbing layers to be structured which is transferred in the shape of a groove in exactly this layer and remains recognizable for the alignment optics.
U.S. Pat. No. 6,153,492 (Wege/Lahnor, Infineon) describes, in this respect, a method for putting up the contrast of alignment marks by depositing and reverse planarizing a tungsten layer. Therein, the trenches around the alignment marks are etched free, for example, back-etched, in an additional step. The depressions formed in this way, are transferred into a subsequently deposited aluminium layer. A modification also stated in the U.S. Pat. No. 6,153,492 includes a back-etch of the oxide layer surrounding the trenches. The tungsten layer now projects beyond the surface of the oxide layer and forms an edge for the exposure of a subsequently deposited aluminium layer.
DE 102 59 322 B4 (Infineon) describes a method for forming an alignment mark in a light-opaque layer on a semiconductor wafer. Therein, trenches having different depths and breadths are formed in the semiconductor substrate that are filled with a first layer. Trenches having a small breadth and depths are completely filled, and trenches having a larger breadth and depth are filled only partly. The latter ones are the trenches for the alignment marks. Thereafter, a second layer is deposited with such a thickness which now also completely fills the trenches of the alignment marks. Therein, the second layer comprises a selectivity in the etching process with respect to the first layer. It follows a chemical-mechanical polishing process in order to achieve a plain surface.
Thereafter, the second layer which is now only present in the trenches with the larger breadth and depth (the trenches of the alignment marks) are etched in an etching process such that these trenches contain new cavities. Since the etching rate of the second layer is by a multiple larger than the one of the first layer, the back-etch depth of other areas of the substrate surface is comparatively small. If now a light-opaque layer is deposited, the new cavities are transferred to the light-opaque layer in case of a sufficient conformity and an adapted thickness of the layer. The position of the cavities can now be recognized by means of alignment optics as alignment mark. This method requires at least one additional layer as well as at least one additional etching step and a polishing process.
The invention is based on the objective to provide a simple and cost effective method for the production of alignment marks in an semiconductor wafer with an at least light-opaque or light-absorbing layer which can do without auxiliary layers and additional process steps, wherein alignment marks are intended to be produced which can be recognized by an exposure system (or alignment optics).
For achieving this objective, the inventive method for the production of at least one alignment mark on a semiconductor wafer by using a light-opaque layer for (later) structuring or for (permanent) protection of circuit elements is proposed. Furthermore, a semiconductor wafer is proposed having at least one alignment mark in a scribing trench area (scribe line), in particular at least one or several group(s) out of respective at least two mutually vertical pillar groups within mutually perpendicular scribing trench areas (in x- and y direction). This includes that also the two pillar groups of a group are aligned perpendicular with respect to each other.
The claimed method achieves to apply the light-absorbing layer on a wafer having a plurality of mutually delimited, integrated circuits without losing the possibility to recognise the wafer in its position by means of optical detection of alignment marks (in an exposure system or by alignment optics) and to position, thereby, the wafer, for example for depositing a further layer or for further alignment of a photo masc. The pinpointed positioning of, for example, a following layer with respect to the prior layer is possible in spite of the usage of the deposited, light-absorbing layer which covers light-sensible areas of the circuits—in the delimited, integrated circuits—, or which, as an alternative, sets free after a further structuring, light-sensitive areas in the circuits for incident light but which, in the rest of the areas outside of the light-sensitive active elements, further on covers against incident light.
For example, the light-opaque layer is applied at the end of a production process on a passivation layer, wherein this deposited layer includes the area of the scribing trenches between the integrated circuits as well as covers the active surface areas of the semiconductor wafer.
Preferably, the scribing trenches are as narrow as possible in order to be able to use the wafer surface with active area in an optimal way for circuits to be inserted therein, in the cavities of the scribing trenches, the alignment marks are placed such that an alignment mark extends along the longitudinal direction of a cavity in the scribing trench. The cavity forms the lowered portion in the scribing trench within which a respective alignment mark is arranged, and has a distance from the edges of the lowered area.
Since X as well as Y scribing trenches are present and an alignment mark is inserted into the X and Y cavity, these two alignment marks extend in two mutually perpendicular directions (x/y direction).
One alignment mark in the X direction and one alignment mark in the Y direction each belong functionally together and is here called “logical alignment mark”. It is also arranged in a spatial area and consists of two mutually perpendicular pillar rows which are, however, not arranged at the same location but in a certain, functionally delimited area.
The two mutually perpendicular pillar groups are functionally associated as a logical alignment mark for the alignment optics or the exposure system.
Also a plurality of pairs of orthogonal, functionally associated pillar groups can be formed, however, each pillar group extends along the cavity within which it is arranged.
The cavity itself is not a complete cavity anymore after application of the light-opaque layer, but it is partly filled with the deposited layer or with the material of this layer, respectively. Since the cavity, however, has a—functionally described—minimum breadth, and since the pillar group (which forms the alignment mark) extends in the longitudinal direction of the cavity, the cavity is only partially filled, and the pillar group remains optically visible, for example, for the alignment optics. This is in spite of the light-opaque layer.
The surfaces of the individual pillars of the pillar row are not covered such that these surfaces cannot be optically recognized anymore by the above-mentioned alignment optics. This can then be in the sense that the surfaces of the individual pillars of the pillar group (associated with an alignment mark) remain free, i.e. are not covered by the light-opaque layer or a portion thereof or, however, a (parasitic) coverage is only so small that an optical detectability is maintained. A thickness not surpassing 200 nm is such a thickness which, because of its transmission still allows to detect the surfaces of the pillar row. However, the surface of the pillar row remains free and uncovered by the light-opaque layer which results wherefrom that the cavity has a minimum breadth and the pillar group has an associated height which is larger than the deposited thickness of the deposited, light-opaque layer such that the surfaces of the pillar group are not covered by the light-opaque layer. By non-coverage it is understood—as previously stated—also such a minor coverage which does not make impossible the optical detectability by the alignment optics or the exposure system.
A “light-obstructing” layer as a light-absorbing or light-opaque layer can be carbon-based. This carbon containing layer is, for example, a carbon-based black resist.
At least one alignment mark is produced in a scribing trench (in the longitudinally extending cavity thereof). Wherein, it is to be read on the X direction as well as on the Y direction, each individually. It may also be read such that, in case of two cavities, one alignment mark each is arranged in a respective cavity and this respective cavity is associated to a respective scribing trench wherein both scribing trenches are adjusted mutually perpendicular to each other. The two cited alignment marks (as pillar groups) would then be functionally associated with each other and form a “logical alignment mark” (consisting out of at least two physical alignment marks and these, each, out of a pillar group). Also a plurality of pairs of logical alignment marks can be arranged, however, not in such a way next to each other that several pillar groups in the same scribing trench in the same cavity are immediately next to each other and, thereby, each have a smaller distance from the inwardly facing walls of the cavity than each individual pillar. This would be a group of logical alignment marks.
In order to generally allow an alignment, sufficient light has to penetrate to the mark (alignment mark) such that this mark can be scanned and sufficient light is reflected which can be converted to an electrical signal in the exposure system and the spatial position of the alignment mark can be evaluated as a basis for a positioning of a further layer (automatically).
The alignment marks produced according to the invention are also recognizable when using light-opaque layers like non-transparent photo resists by means of the exposure system such that a pinpointed alignment of the wafer during the exposure is made possible and, thereby, the quality of a structuring process is put up. As an alternative, the alignment marks present in the cavity (in the scribing trench) can be used for adjusting a photo mask with respect to the prior layer. By means of this photo mask, the photo resist is exposed at certain areas and, thereby, developed such that the above-mentioned structuring is resulting. This also comes under the understanding of the improvement of the quality of a structuring process which is achieved naturally according to the pinpointed alignment by means of the at least one alignment mark, the logical alignment mark or a group of logical alignment marks.
A further advantage of the invention resides in that the method may be implemented in a simple way since existing methods (etching steps) are used in the process whereby the claimed process can be carried out cost-effectively and without additional effort. For example, the etching of the cavities (in the scribing trenches) may be carried out with the etching step for the passivation layer, which passivation and/or soft etching step is anyway present in the production method of the wafer.
The passivation may be part of the layer stack into which the cavities are etched and in which the pillars remain standing up. It may, however, also be located outside of the scribing trench in the rest of the active wafer area.
The layer stack may be formed, in an embodiment, out of an individual layer dielectric (intermediate layer dielectric-ILD), an intermediate metal dielectric (inter metal dielectric-IMD) or a passivation layer (pad oxide). This sequence of steps is etched as “layer stack” with a photo mask for forming the cavity.
At least one etching step for etching the layer stack is already used in the system for producing the wafer. Preferably, this etching step is provided for the passivation layer.
Also several etching steps may be taken depending on whether it is an oxide or a combination of oxide and nitride as layer stack. At least one or several thereof are such etching steps which are also used on the production of the wafer or, to put it in another way, anyway would have been used.
The etching time is adjusted such that it is etched down to the substrate by a corresponding surplus etching of the layer stack—for forming the cavity—and such that the pillar of the pillar group remains standing up in the cavity.
In an advantageous embodiment of the claimed method, the height of the pillar group is formed reaching down to the substrate of the wafer, wherein the alignment of the wafer in the following fabrication steps is always ensured. For forming the height of the single pillars of a pillar group and, thereby, also of the whole pillar group, the cavity or each cavity is etched around the respective pillar group and also around each individual pillar down to the substrate of the semiconductor wafer whereby the group of pillars in the cavity remains standing up. Silicon may be used as substrate of the wafer. This production of the pillar groups is true for the x direction as well as the y direction which are called, in the following, X cavity and X scribing trench and Y cavity and Y scribing trench respectively.
A further advantageous embodiment of the inventive method is characterized in that, after depositing the light-opaque layer, no planarization is carried out, whereby, at the one hand, a fabrication step is saved since it is redundant, and, on the other hand, a further damage of the pillars of each pillar group is avoided.
In further advantageous embodiments of the inventive method, the pillar groups are produced with the following dimensions: height of the pillars above 1 μm to 5 μm, preferably 3 μm, and length/breadth of 2 μm to 5 μm, preferably 4 μm. Pillar-like alignment marks are produced which, at the one hand, can be well recognized by the alignment optics and, on the other hand, are not only parasitically covered with the material of the light-opaque layer on the subsequent deposition of the light-opaque layer.
If a respective cavity is formed with a breadth dimension of at least 80 ρm, it is achieved with a high security that each pillar group usable as alignment mark, is not or only slightly covered with the material of the layer portion of the light-opaque layer related to the cavity on the subsequent deposition of the light-opaque layer. As such, the pillars out of this light-opaque material would protrude with their surfaces in the cavity.
A characterization of the breadth-dimension of the cavity can also be made relative to the breadth of the pillars in the pillar group in this cavity. The height of the pillars is, therein, above 1 μm such that they can be named with the term pillars. The light-opaque layer material in each cavity within which a pillar group is standing up, is, thereby, restricted in its filling height thereby that the minimum height of the pillars is, as functionally described, also put up like the breadth of the cavity. Outside of the cavity, this light-opaque layer is positioned on the wafer in its deposited thickness. It serves there to the (permanent) protection of a circuit element or is provided for a (later) structuring. It has there a preferred thickness between 0,8 μm and 2,5 μm.
The characterization of the relative breadth of the cavity is given with reference to a multiple of the breadth of the pillar group, the preferred breadth of essentially 4 μm of the pillar group results, by means of the factor 20, in the preferred breadth of the cavity. In a range of the breadth of the pillar between 2 μm and 5 μm, a breadth range between 40 μm and 100 μm results for the cavity.
Essentially one half of thereof is located on both sides of a presumed single mark in the cavity.
Preferably, the light-opaque surface is produced with a thickness of 0,8 μm to 2,5 μm (two-dimensionally outside of the cavity of the scribing trench). Thereby, it also penetrates the cavity, however, only to such a limit that the level of the light-opaque layer remains below the level of the surfaces of the pillars, whereby it is achieved with highest probability that the upper ends of the pillars remain free and can, thereby, be detected with certainty.
A further advantageous embodiment of the inventive method is characterized in that an area saving “single mark” is formed. The scribing trench area (scribe line) of the wafer can, thereby, be kept as narrow as possible in order to be able to use the wafer surface in an optimal way for chips. Two single marks are functionally defined and comprise a pillar group in x direction and a pillar group in y direction, however, not in the same location but in two mutually perpendicular cavities of two mutually perpendicular scribing trench areas. Two single marks comprise, therein, two pillar rows and would have to be understood as “logical alignment mark” as described above. The single mark comprises, however, also the statement that to left and to the right of a pillar group within the same cavity no further pillar groups are arranged. Thereby the breadth of the cavity is divided up into the left-hand side and the right-hand side of the pillar group which is also true for the other pillar group perpendicular thereto which is, together with the first mentioned pillar group, part of the “logical single mark”.
It is a further advantageous embodiment of the inventive method to form the pillar group as stack mark. The materials of the pillar group are advantageously adapted to the fabrication steps on the wafer which are to be carried out anyway, and are formed out of oxide or nitride or oxy-nitride or any possible combination thereof. This stack mark describes the usage of several layers of a layer stack which are lying on the substrate prior to the etching of the cavity and which define, in the framework of the lowering of the cavity and the formation of the pillar groups, also the consistency and configuration of each individual pillar which remain standing up at the bottom of the cavity, preferably on the substrate. The layer stack out of several of the above mentioned layers also forms the left-hand and right-hand wall of each cavity. The breadth dimension of the cavity can be described either by means of the rim of the cavity or by means of the wall of the remaining layer stack.
Furthermore, it is advantageous when the alignment marks are produced out of at least a metal layer. At least one metal layer is provided, or the pillars of a pillar group are formed as a pure metal pillar for forming a pure metal mark or a “metallic pillar group”. Preferably, no metal layers are used in combination with oxide layers/nitride layers.
The semiconductor wafer which comprises at least two alignment marks which are running perpendicular to each other, is a result of the production process. It is, at the one hand, characterized by method steps which leave their traces in the alignment marks and, on the other hand, by structural elements. An x direction and a y direction define two mutually perpendicular alignment marks. They are each surrounded by a light-opaque layer portion of a deposited layer, and they are embedded therein, are, however, not covered such that an optical detection by an exposure system (or a alignment optics) becomes impossible.
The deposited layer is later on structured or it serves for protecting of at least one optically sensible area of a circuit element outside of the two mutually perpendicular cavities.
A light-opaque layer portion each of the deposited layer fills the mutually perpendicular cavities not completely so that the pillar grooves standing up therein, are not completely covered by the light-opaque layer portion which is allotted to the respective cavity.
A preferred minimum breadth of at least 40 pm is ensued on one side and the other side of a bigger group up to the rim of the cavity or, however, up to the walls of the remaining layer stack which walls define the respective length dimension of the cavities and their rims.
The pillars should be high above 1 μm, preferably in the order of magnitude of a height of 4 μm.
An example of an alignment mark according to the state of the art and embodiments of the invention are now describes with reference to the drawings.
a shows this alignment mark 4 as parallel groups with an explanation of the dimensions of the projections and their distances in the cavity in the photo resist 3.
a shows a new alignment mark as group 16y of pillars according to an embodiment of the invention in top view in y direction in a Y cavity of a Y scribing trench.
b shows the pillar 16 of the pillar group 16y in a section AS-AS of
a shows the structure of
b shows the section CS-CS of
a is a separated chip 100a of the wafer 100 of
The structuring of a non-transparent (light-opaque or light-reflective or light-absorbing) layer, for example out of black photo resist on carbon base or pigment base requires other alignment marks. The black photo resist is used as a “light-shielding” or at least light-obstructing layer in order to protect light-sensitive areas of circuits on a semiconductor wafer, or the black photo resist is used as a photo mask in the structuring of one or several circuit element(s). For this purpose, the black lacquer is to be adjusted with respect to the prior layer.
The black lacquer layer is deposited at the end of the fabrication process, for example on a passivation layer. It is a non-transparent, negative photo resist with planarizing properties. Lacquer having a defined thickness, may equalize the existing surface topography. Smaller depressions or cavities are filled up with this lacquer. The result then is a planar surface so that conventional alignment marks are not detected anymore by the alignment optics.
For forming alignment marks according to a first example of the invention, layers present in the process and etching steps are used in that doped or un-doped oxide layer or oxide nitrite layers or nitrite layers or a combination thereof are etched. By using a combination of layers, so-called “stack marks” are produced. Also metal layers can be used for forming the alignment marks, however, not in combination with oxide/nitride layers. For simplicity, it is taken care by means of a adequate mask layout, that no metal is present between the oxide/nitride/oxy-nitride layers.
Preferably, pillar groups of the stack marks have the above-mentioned layers. The pillars originate after an etch step, for example by anisotropic dry etching (RIE). The layer(s) is/are etched down to the silicone substrate for forming of cavities. The “uncovering” of the pillar groups in the cavities is made by an etching step already used in the process, for example, in such a step in which a passivation layer is etched.
For the differentiation of the alignment marks from the known alignment marks explained with respective
This pillar row 16y with a single-digit number of individual pillars extends in y direction in parallel to walls 15′ and 15″ of the cavity 12y. In the section BS-BS through the individual pillar 16e, the lower half picture of
Previously, a stack mark has been described, and this configuration of the stack mark is also the one of
Therein, the pillars receive a breadth b16 and a height h16, and they stand up on the bottom of the cavity 12y, which is he surface 11a of the substrate 11. The substrate can be a silicone substrate.
The running direction of the “trench” which is named here Y cavity, is shown in
In this example, the pillars are rectangular in cross section which is to be seen from
As to the surroundings, it is referred to
A constituent part of a scribing trench is also a cavity each which extends in the longitudinal direction of the scribing trench and the breadth of which is specified, with respect to its dimensions, smaller than the scribing trench. If the cavity is kept small in its breadth b12, also the breadth of the scribing trench can be dimensioned to be small. The scribing trench should be as narrow as possible in order to be able to use the wafer surface in an optimal way for chips.
The cavities in y direction explained with reference to the
Also here, stack marks are present which are formed according to the arrangement of
As an example of a possible dimensioning of the cavities 12x, 12y, it can be said that a scribing trench has presently a breadth of about more than 100 μm, and that, therein, a breadth b12 with 80 μm finds sufficient space in an embodiment. On both the sides of a s pillar row this would mean a distance from the inner wall 15′ or 15″ respectively, of the remaining layer step of at least 40 μm. The pillar row itself may have an extension in this direction of l16 which may be in the range between 2 μm to 5 μm. Preferably, the pillars have an order of magnitude of 4 μm in the breadth direction and, according to the rectangular cross sectional configuration, also an extension l16 in longitudinal direction which may comprise the same dimension as the breadth direction.
A relative consideration shows a relationship of essentially 20 between b12 and b16 or between 8 and 50 at the broadest pillars and the narrowest cavity, or vice versa, respectively.
The height of each pillar in the pillar group 16y and 16x amounts to above 1 μm and is advantageously placed in the range up to 5 μm. Preferred orders of dimensions are around 3 μm such that it is evident that these pillars have an essentially larger extension in height as compared to the projections of
The at least one pillar group 16y in y direction is supposed to be explained after deposition of the light-opaque layer 17 in more detail with reference to
For a cavity 12y or 12x, respectively, a layer portion 17y or 17x, respectively, is symbolically used which, however, is constituent part of a deposited layer 17 which is light-opaque or light-absorbing. In the preferred embodiments, this layer has a thickness of not more as 2,5 μm and should not be below a thickness of 0,8 μm.
After the deposition of the layer 17 on the wafer 100 and thereby, also on top of the cavity 12y and the pillar group 16y of
This “free standing” property of the pillar group 16y results from the breadth of the etched cavity and the pillar row left standing up within with its height dimension. The height dimension is larger than the thickness of the layer portion 17y whereby the outer etch area 12y′ near to the vertical wall 15′ on the left-hand side and 12y″ next to the vertical wall 15″ on the right-hand side are not relied upon for the thickness dimensioning of the layer potion 17y. The portions 12y′ and 12y″ at the left-hand and right-hand rim of the filled up layer portion 17y may extend up to the height of the layer or the layer stack 15 or it lies slightly below this on the one or the other side thereof. This depends on the circumstances of the fabrication process and cannot be defined in this accuracy. The consideration of the surroundings of the pillar group 16y and, in particular, in the respective central area of the left-hand portion and the right-hand portion of the cavity 12y is essential by which it is clearly evident that the deposited layer portions 17y does not reach the height measurement a h16 of the single pillars 16a to 16f of the pillar group 16y.
The layer portion 17y of the light-opaque layer 17 filled into the cavity 12y, does not reach, in the central area around the pillar group 16y and in the left-hand and right-hand central area, the height which lies between the surface of the pillars of the pillar group and the bottom 11a of the cavity 12y. Only in the outer rim area 12y′ and 12y″, the filled in portion 17y can reach such height dimension or also accede it, where this functionally does not contribute to maintaining the optical detectability of the pillar group and, respectively, does not contribute anything detrimental.
The pillar group 16y formed as a stack mark, is shown in section of the pillar 16d consisting out of two layers 15a′, 15b′ of which at least one layer is an oxide or nitride or oxy-nitride, and the other layer of which comprises a correspondingly different material which is also selected from the above-mentioned group. More than two layers are possible but are not separately shown. The man skilled in the art can easily imaging this with reference to the explanations given here. Also only one single-layer may be provided. One of the layers, for example the layer 15a, can be a passivation layer. It can also extend outside of the shown area to the rest of the active wafer area, and this is also true for the pillars of the pillar group 16x in the same way.
At least one metal layer is also possible, or the formation of the pillar group 16y out of single pillars 16a to 16f which completely consists out of metal. In case a plurality of metal layers is provided, they correspond to the shown layers 15a and 15b. In other words, each of the pillars of the pillar group may comprise at least one metal layer in one embodiment, and this metal layer can be present a plurality of times in the height direction or may extend also completely in the height direction h16. This is also true for the pillars of the pillar group 16x.
With respect to the drawings of the
Other than in
The explanation as to the y direction is comparatively to be transferred to
The pillar rows 16x and 16y are arranged at different locations. They are aligned in different directions which is easily to be seen in
Two such single marks on the semiconductor wafer in the scribing trench areas for the x and y direction would be used by an alignment optics in order to reliably achieve the positioning.
A logical alignment mark comprises, however, at least two mutually perpendicular pillar rows or to mutually perpendicular “alignment marks” one of which is adjusted in the y direction and the other one in the x direction, however, not at the same location.
The logical association of two single marks does not say anything about their physical association. By example, the y-single mark may be arranged in an y-scribing trench and the x-single mark may be arranged in the next x-scribing trench.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2008 063 156.6 | Dec 2008 | DE | national |
The application is a U.S. National Stage Application of International Application of PCT/IB2009/055935 filed Dec. 23, 2009, which claims the benefit of German Patent Application No. 10 2008 063 156.6 filed Dec. 24, 2008, the disclosures of which are herein incorporated by reference in their entireties.
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/IB09/55935 | 12/23/2009 | WO | 00 | 8/23/2011 |
