Method for producing a semiconductor   

CROSS-REFERENCE TO RELATED APPLICATIONS

This Utility patent application claims priority to German Patent Application No. DE 10 2008 025 733.8 filed on May 29, 2008, which is incorporated herein by reference.

One embodiment of the invention relates to a method for producing a thin semiconductor body and to the use of the method for producing a power semiconductor component.

For a multiplicity of applications of electronic semiconductor components and integrated circuits (IC), it is advantageous to restrict the total thickness of the semiconductor components and of the integrated circuits. Thus, for example, in disposable electronics and for chip cards and smart cards, a very small mass and a very small structural height are of importance. By using targeted settings of the thickness of the semiconductor body used, the electrical properties of e.g., vertical power semiconductor components can be improved by adapting the thickness of the semiconductor body to the voltage class of the respective power semiconductor component, in order to avoid unnecessary electrical resistance through over-dimensioned semiconductor bodies.

However, this necessitates a very precise and reproducible thickness setting over the entire area of the semiconductor body used, in order to avoid losses of yield in production and in order to ensure reliable electrical properties of the semiconductor component and of the integrated circuit.

For these and other reasons, there is a need for the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings are included to provide a further understanding of embodiments and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments and together with the description serve to explain principles of embodiments. Other embodiments and many of the intended advantages of embodiments will be readily appreciated as they become better understood by reference to the following detailed description. The elements of the drawings are not necessarily to scale relative to each other. Like reference numerals designate corresponding similar parts.

FIGS. 1A and 1B schematically illustrate one embodiment of a method for thinning a semiconductor body.

FIG. 2 illustrates one embodiment of a doping profile of an n-type dopant after one proton implantation with subsequent heating in a p-doped semiconductor body.

FIG. 3 illustrates one embodiment of a doping profile of an n-type dopant after two proton implantations with subsequent heating in a p-doped semiconductor body.

FIG. 4 illustrates one embodiment of a doping profile of an n-type dopant after two proton implantations with subsequent heating in a p-doped semiconductor body.

FIG. 5 illustrates one embodiment of a doping profile of a semiconductor body for a power semiconductor component.

FIG. 6 illustrates a schematic cross-sectional view of a power semiconductor component with a semiconductor body.

FIG. 7 illustrates a schematic arrangement for the electrochemical etching of a semiconductor body with a pn junction.

DETAILED DESCRIPTION

In the following Detailed Description, reference is made to the accompanying drawings, which form a part hereof, and in which is shown by way of illustration specific embodiments in which the invention may be practiced. In this regard, directional terminology, such as “top,” “bottom,” “front,” “back,” “leading,” “trailing,” etc., is used with reference to the orientation of the Figure(s) being described. Because components of embodiments can be positioned in a number of different orientations, the directional terminology is used for purposes of illustration and is in no way limiting. It is to be understood that other embodiments may be utilized and structural or logical changes may be made without departing from the scope of the present invention. The following detailed description, therefore, is not to be taken in a limiting sense, and the scope of the present invention is defined by the appended claims.

It is to be understood that the features of the various exemplary embodiments described herein may be combined with each other, unless specifically noted otherwise.

Before the exemplary embodiments are explained in more detail below with reference to the figures, it is pointed out that identical elements are provided with the same or similar reference symbols in the Figures and that a repeated description of these elements is omitted. Furthermore, the Figures are not necessarily true to scale; rather, the main emphasis is on elucidating the basic principle.

The term pn junction is defined hereinafter as the location in a semiconductor body at which an n-type dopant concentration ND of the semiconductor body falls below a p-type dopant concentration NA of the semiconductor body or a p-type dopant concentration NA falls below an n-type dopant concentration ND of the semiconductor body.

One or more embodiments provide a method for producing a semiconductor body which permits an exact and reproducible thinning of the semiconductor body, and the use of this method for producing power semiconductor components.

FIG. 1A illustrates a semiconductor body 10 having a first side 11 and a second side 12, wherein the direction leading from the first side 11 to the second side 12 is designated as the y-direction. The semiconductor body 10 is typically a semiconductor wafer. Such wafers are normally starting products for the mass production of semiconductor components and are available in sizes of currently approximately 750 μm thickness (y-direction) and up to 300 mm diameter (x-direction). The starting semiconductor body 10 used here is doped with a p-type dopant, such that the semiconductor body has a p-type basic doping. The electrical conductivity of the semiconductor body 10 is therefore initially determined by “holes” (p-type charge carriers) as majority carriers. To a certain extent, but with a significantly lower concentration, however, the semiconductor body also already contains conduction electrons (n-type charge carriers) as minority carriers.

Silicon is principally suitable as semiconductor material for the semiconductor body 10. The sheet resistance of the p-doped semiconductor body lies between 100 Ohm cm and 5000 Ohm cm, for example. As illustrated in subprocess a) in FIG. 1, protons 14 are implanted into the semiconductor body 10 at the first side 11. Afterward, at least that region of the semiconductor body 10 which has been implanted with protons 14 is subjected to heat treatment, that is to say heated and held at this temperature level for a specific time. This gives rise to an n-doped zone 10a and accordingly a pn junction 13 in the p-doped semiconductor body 10. In one embodiment, the n-doped zone 10a is produced with a proton irradiation dose of between 1014 cm−2 and 1015 cm−2.

Embodiments relate generally to a method for producing a semiconductor body, wherein a p-doped semiconductor body having a first side and a second side is provided, an n-doped zone is formed in the semiconductor body by implantation of protons into the semiconductor body via the first side down to a specific depth of the semiconductor body and by subsequent heating at least of the proton-implanted region of the semiconductor body, such that a pn junction arises in the semiconductor body, and the second side of the semiconductor body is removed at least as far as a space charge zone spanned at the pn junction.

One or more embodiments provide a use of the method for producing the semiconductor body for producing a power semiconductor component, wherein in each case at least one electrode is fitted to the first and to the second side of the semiconductor body produced.

The proton irradiation with subsequent heating makes it possible to produce an n-doped zone and thus a pn junction at an exactly predefined location homogeneously also over a large area in the semiconductor body. As a result of the high penetration depth of protons into a semiconductor body, the pn junction can also be produced at a large depth in the semiconductor body, which cannot be realized by conventional implantation or diffusion techniques. The space charge zone spanned at the pn junction or the pn junction itself can be used for a precise ending of the rear-side removal of the original semiconductor body, thereby enabling an exact and reproducible thinning of the semiconductor body to a desired and predetermined final thickness.

FIG. 1B illustrates an n-doped zone 10a formed in this way and the pn junction 13 in the semiconductor body 10 in subprocess b). The energy of the proton radiation 14 is set in such a way that the pn junction 13 is produced at a predetermined location in the semiconductor body 10. In one embodiment, the position of the pn junction is determined according to the desired final thickness of the semiconductor body 10. Embodiments of the invention provide for the pn junction 13 to be produced at a depth T of 20 μm to 350 μm, measured from the surface of the first side 11. In the case of a silicon semiconductor body, the proton radiation energy here lies in the range of between 1 MeV and 8 MeV. In principle the proton radiation energy correlates with the penetration depth of the protons into the semiconductor material used. The higher the proton radiation energy used, the larger the penetration depth of the protons into the semiconductor material used.

The heating (heat treatment) of the semiconductor body 10 that is effected after the proton irradiation in one or more embodiments includes an annealing phase in the range of between 350° C. and 550° C., since the hydrogen-induced n-type doping forms in this temperature range. In one embodiment, the annealing phase is effected at temperatures of between 450° C. and 520° C. In one embodiment, the annealing phase is carried out for at least 15 minutes and can also take place over a number of hours.

After the formation of the n-doped zone 10a and the pn junction 13, in thermodynamic equilibrium as a result of diffusion of charge carriers across the pn junction 13, a space charge zone 15 forms as far as a boundary 15″ in the n-doped zone 10a and as far as a boundary 15′ in the remaining p-doped semiconductor body 10b, since a high concentration difference in the carrier densities exists between the n-doped zone 10a and the remaining p-doped semiconductor body 10b. As a result of the fixed charges remaining, the previously electrically neutral crystals have now acquired a space charge that charges the p-type crystal negatively and the n-type crystal positively. The resultant electrical voltage is called the diffusion voltage UD.

By applying an external bias voltage across the pn junction 13, the boundaries 15′ and 15″ can be displaced and the width W of the space charge zone can thus be controlled. By applying the external bias voltage in the reverse direction of the pn junction 13 (+ at the n-doped zone 10a, − at the remaining p-doped semiconductor body 10b), by way of example, the width W of the space charge zone 15 is increased. If the external bias voltage is polarized in the forward direction of the pn junction 13 (−at the n-doped zone 10a, + at the remaining p-doped semiconductor body 10b), the width W of the space charge zone 15 is decreased. As soon as the external bias voltage with polarization in the forward direction is greater than or equal to the diffusion voltage, the space charge zone 15 is dissolved, that is to say that the boundaries 15′ and 15″ fall on top of one another and the width W of the space charge zone is equal to zero.

After the formation of pn junction 13 and a space charge zone 15 that possibly occurs, the semiconductor body 10 is thinned. In the case where a space charge zone 15 is spanned at the pn junction 13, this takes place by removing the second side 12 of the semiconductor body 10 in the negative y-direction as far as the space charge zone 15, that is to say as far as the boundary 15′ of the space charge zone 15 that is situated in the residual p-doped semiconductor body 10b, as is illustrated by arrows 16 in FIG. 1B. Consequently, a region 10b″ of the semiconductor body is removed, while a residue 10b′ of the p-doped semiconductor body 10 remains in the thinned state at the second side. By way of example, by using suitable setting of the bias voltage value across the pn junction, the space charge zone can be extended up to 5 μm into the residual p-doped semiconductor body, which results in a correspondingly thick p-doped residual layer 10b′ after the thinning. In the case of a dissolved space charge zone 15, the second side is removed as far as the pn junction 13, that is to say that the residue 10b of the p-doped semiconductor body that remained from the original semiconductor body 10 after the production of the n-doped zone 10a is completely removed. The removal of the second side 12 can in any case be effected in a locally delimited manner by using masks, for example, or else over the whole area over the entire semiconductor body 10. The removal is generally effected at least in part by using an electrochemical etching method wherein the boundary 15″ of the space charge zone or, in the absence of a space charge zone, the pn junction is used as an “etching stop” for ending the etching process. When this “etching stop” is reached, the etching process automatically terminates; in other words, the etching stop is effected in a self-aligned manner in this way. A very exact removal of the second side 12 of the semiconductor body 10 is thus possible. By way of example, a characteristic change in a current flowing within the electrochemical etching apparatus is measured when the “etching shop” is reached, which is used for ending the etching process. Mechanical removal methods can also be used at the beginning of the removal of the second side 12 of the semiconductor body 10.

FIG. 2 illustrates one embodiment of a doping profile of an n-doped zone 10a produced according to the method described above in a p-doped semiconductor body 10. The n-doped zone 10a extends from the surface of the first side 11 of the semiconductor body 10 as far as the pn junction 13 at a depth T into the semiconductor body 10. The n-doped zone 10a has a region N1 having a virtually constant n-type dopant concentration ND, that is to say that the n-doped dopant concentration ND changes typically by at most a factor of 3 in the region N1. The region N1 extends within the n-doped zone 10a from the surface of the first side 11 as far as the depth A. Typical values for A are in this case 15 μm<A<300 μm. Between the region N1 and the pn junction 13, a region N2 extends in the n-doped zone 10a, the region N2 having a n-type dopant maximum NDmax at the depth B in the semiconductor body 10. In this case, the n-type dopant maximum NDmax is produced at the location of the highest proton density originating from the proton implantation. On account of the—in the case of protons—very small variation range of the penetration depth into the semiconductor body 10, the location having the highest proton density is situated virtually at the pn junction 13 (“End-of-Range”). The n-type dopant concentration ND thus falls very steeply from the n-type dopant maximum NDmax toward the pn junction 13. The “End-of-Range” region forms the end of the region which is irradiated by the proton implantation and in which the majority of the protons is incorporated during the implantation. On account of the annealing process, a large portion of the protons diffuses in the negative y-direction toward the first side 11, which results in the doping ND in the region through which the protons are radiated. The protons that diffuse into the depth of the p-type semiconductor body 10 in the positive y-direction toward the second side 12 do not lead to the formation of donors in this region, since implantation-induced crystal defects required therefor are not present there. The difference between the maximum doping concentration NDmax in the “End-of-Range” region and the doping concentration ND in the n-doped zone 10a is dependent on the temperature during the thermal process and the duration of the thermal process. It holds true here that for the same duration of the thermal process, the difference is all the smaller, the higher the temperature during the thermal process, and that for a given temperature during the thermal process, the difference is all the smaller, the longer the duration of the thermal process. The energy of the proton irradiation, the proton dose and the annealing temperature and annealing time are chosen so as to produce a sufficient n-type doping maximum NDmax e.g., for a field stop zone and for forming a pn junction, on the one hand, and for forming a suitable basic doping ND of the n-doped zone 10a, on the other hand. The n-type basic doping ND of the n-doped zone 10a can be produced without additional outlay from the production of the pn junction required for the thinning of the semiconductor body 10. In a form that is not illustrated, the n-type dopant maximum NDmax can be completely reduced by a sufficiently long annealing phase of the proton-implanted region over e.g., a number of hours and can be converted into a virtually constant n-type dopant concentration ND, with the result that the homogeneous region N1 extends from the surface of the first side 11 to shortly before the pn junction 13. However, the n-doped zone 10a can also be produced with an n-type dopant concentration ND that falls from the n-type dopant maximum NDmax in the direction toward the first side 11 of the semiconductor body 10.

The p-doped residue 10b of the semiconductor body 10 that remains from the original semiconductor body 10 extends from the pn junction 13 as far as the surface of the second side 12 and has a largely constant p-type dopant concentration NA.

FIG. 3 illustrates one embodiment of a doping profile of an n-doped zone 10a produced according to one embodiment in a p-doped semiconductor body 10. In contrast to the doping profile illustrated in FIG. 2, the depth B of an n-type dopant maximum NDmax produced in the n-doped zone 10a is spaced apart further from the pn junction 13. However, as in the exemplary embodiment regarding FIG. 2 as well, the n-type dopant maximum NDmax is still situated nearer to pn junction 13 than to the surface of the first side 11 of the semiconductor body 10 in the region N2. A region N3 having a reduced n-type dopant concentration ND is additionally formed between the n-type dopant maximum NDmax and the pn junction 13, wherein the region N3 as illustrated, can have a “shoulder” 30 having an n-type dopant concentration ND that is virtually constant over a length 1. This “shoulder” 30 produced over the length 1 in the region N3 has, for example, a higher n-type dopant concentration ND than the region N1. The “shoulder” 30 is defined as the region in which the n-type dopant concentration changes (ΔN) at most by a factor of 3. The n-type dopant concentration ND in the region of the shoulder 30 should, but need not necessarily, be lower than that of the n-type dopant maximum NDmax at least by a factor of 5. The length 1 of the shoulder can be, for example, between 1 μm and 20 μm.

As illustrated in FIG. 4, a region having an n-type dopant minimum NDmin can also be present between the shoulder 30 having an approximately constant n-type dopant concentration ND and the region N2, wherein the concentration of the n-type dopant minimum NDmin can be lower than the dopant concentration ND in the region of the shoulder 30, for example, by more than a factor of 3.

The region N3 that has a shoulder and is illustrated by way of example in FIG. 3 and FIG. 4 makes it possible to set a well-defined distance between the n-type dopant maximum NDmax in the region N2 and the pn junction 13. Consequently, it is possible to set e.g., a very readily reproducible efficiency of a rear-side emitter introduced into the wafer surface, which, moreover, can have a relatively small current dependence.

The doping profile illustrated in FIG. 3 and FIG. 4 can be produced, for example, by using a first proton implantation into the p-doped semiconductor body 10 with a subsequent annealing phase over a long time period and a succeeding further proton implantation with an increased implantation energy and decreased implantation dose relative to the first proton implantation. In this case, the duration and the temperature of the annealing phase of the first proton implantation is still chosen to be sufficiently short and low, respectively, that the n-type dopant maximum NDmax remains in the region N2 of the n-doped zone 10a. The second proton implantation is likewise carried out via the first side 11 of the semiconductor body 10 and annealed, wherein the duration and temperature of this annealing phase are chosen such that the “shoulder” 30 is formed. The n-type dopant minimum NDmin in FIG. 4 can be produced by short annealing times of the second proton implantation. The diffusion of the implanted protons in the direction of the n-type dopant maximum NDmax then extends only as far as the n-type dopant minimum NDmin.

FIG. 5 illustrates one embodiment of a doping profile of a semiconductor body for a power semiconductor component. The semiconductor body is produced by forming an n-doped zone 10a by using proton implantation and subsequent thinning, as already explained regarding FIG. 1. In this case, the second side 12 is removed as far as the boundary 15′ of a space charge zone at the pn junction 13, which is situated, for example, approximately 5 μm in the residual p-doped residue 10b of the semiconductor body as a result of the application of a bias voltage polarized in the reverse direction. As a result of the removal of the second side 12, one part 10b″ of the p-doped residue 10b of the semiconductor body 10 is removed, while another part 10b′ of the p-doped residue 10b remains. The part 10b′ has a thickness of approximately 5 μm corresponding to the position of the boundary 15′.

At the thinned second side 12 of the semiconductor body 10, an additional p-type doping near the surface is then carried out by indiffusion or implantation in combination with a thermal activation or indiffusion of the implanted atoms of p-type dopant. The resultant “rear-side” highly doped P-type zone 40 can be used, for example, as a p-type emitter for an IGBT.

At the first side 11 of the semiconductor body 10, a p-type doping into the n-doped zone 10a—which initially reaches as far as the surface of the first side 11—is likewise carried out by using diffusion or implantation of p-type dopant, to be precise, for example, before the proton implantation described above is carried out and before the thinning process described. This gives rise to a p-doped zone 41 which, at the surface of the first side 11, reaches down to a depth C and which, together with the n-doped zone 10a, forms a further pn junction at a depth C near the surface at the first side of the semiconductor body.

FIG. 6 illustrates a power semiconductor component 200. The semiconductor component 200 has a thinned semiconductor body 10 produced according to an embodiment of the method described above. An electrode 20 is fitted to the surface in each case of the first side 11 and of the second side 12 of the semiconductor body 10.

The semiconductor component 200 can be any vertical semiconductor component, that is to say a semiconductor component whose current conduction takes place from the first side transversely through the thickness of the semiconductor body 10 to the second side in the y-direction. In one embodiment, it can be an IGBT, a diode or a thyristor.

The region N2 having the n-type doping maximum NDmax described in the previous exemplary embodiments can be used as a field stop zone, for example, in the case of an IGBT. The n-type doping maximum NDmax produced by proton irradiation with subsequent heat treatment can be aligned very exactly to the surface of the second side 12. By virtue of the high proton radiation energy required for producing the n-type dopant maximum NDmax near to the surface of the second side 12 of the semiconductor body 10 via the first side 11 of the semiconductor body 10, it is possible to radiate through any defects or impurities in or on the semiconductor body without any problems. Consequently, it is possible to set the position of the field stop exactly with respect to the surface of the second side 12 of the semiconductor body 10, which enables a precise setability and high reproducibility of the gain factor αpnp of the collector-side partial transistor of the IGBT. A great variation of the electrical properties of the semiconductor component 200 is thus avoided. Moreover, the very small and well-defined distance—thus possible—of below 5 μm between the field stop zone and the rear-side emitter at the second side 12 ensures a high and reproducible short-circuit strength, a desirable softness when turning off the semiconductor component 200 and avoidance of particle-governed leakage current problems.

The n-doped zone 10a which is produced by using the method explained above and has an n-type doping with hydrogen-induced donors is suitable in one embodiment for realizing a semiconductor zone of a power semiconductor component that takes up a reverse voltage. Such a zone is, for example, the drift zone of a MOSFET, the drift zone or n-type base of an IGBT, or the drift zone or n-type base of a diode.

FIG. 7 is a schematic illustration of a setup for the electrochemical etching of a p-doped semiconductor body 10 having an n-doped zone 10a and a pn junction 13. An etching cell 600 of the setup includes at least an anode, a cathode 61 and an aqueous alkaline solution 60. The anode is formed by the n-doped zone 10a of the semiconductor body 10 to be thinned, at which a positive potential is present. The cathode 61 is a platinum electrode, for example, at which a negative potential is present. Both the anode and thus the semiconductor body 10 having the n-doped zone 10a and the pn junction 13 and the cathode 61 are immersed in the aqueous alkaline solution 60.

Further embodiments of the etching cell 600 can, for example, also have other etching solutions such as, for example, a pure potassium hydroxide solution (KOH), ethylenediamine (EDP) or hydrazine-water solutions. Further embodiments can also have three or more electrodes. The positive potential can be applied to the contact-connection of the anode, in the present case of the n-doped zone 10a, by using suitable contact structures such as, for example, a net-like contact structure.

Although specific embodiments have been illustrated and described herein, it will be appreciated by those of ordinary skill in the art that a variety of alternate and/or equivalent implementations may be substituted for the specific embodiments shown and described without departing from the scope of the present invention. This application is intended to cover any adaptations or variations of the specific embodiments discussed herein. Therefore, it is intended that this invention be limited only by the claims and the equivalents thereof.