JP5228282B2 - Power semiconductor device and manufacturing method thereof - Google Patents

Description

  The present invention relates to a power semiconductor device requiring a high breakdown voltage and a method for manufacturing the same.

  In a power semiconductor device such as an IGBT or a power MOSFET, it is necessary to improve a withstand voltage at the time of interruption. In order to obtain ideal withstand voltage characteristics, it is generally preferable to set the specific resistance in the base region from 10 Ωcm to 500 Ωcm.

  As a technique for controlling the specific resistance of a semiconductor, an NTD (Neutron Transferred Doping) method in which element conversion is performed by irradiating neutrons is known. In this method, a silicon substrate of a power semiconductor device is irradiated with neutrons generated in a nuclear reactor, whereby silicon is elementally converted into phosphorus to form a donor. In this technique, since the neutron permeability is high, the entire region in the depth direction of the substrate through which neutrons are transmitted can be uniformly formed into a donor.

In Patent Document 1, an FZ wafer on which an N ⁻ drift layer is formed is used, and an element active region (P ⁺ base region, P ⁺ emitter region, gate oxide film, gate electrode, etc.) and emitter electrode are provided on the surface side. A technique relating to carrier concentration control of a semiconductor device having an IGBT structure formed is disclosed. In this technology, the back side of the wafer is shaved down to a predetermined thickness, proton irradiation is performed from the back side, low temperature annealing is performed to form an N ⁺ buffer layer, and boron ion particle beam irradiation is performed from the back side. Forming a P ⁺ collector layer by performing an annealing process of irradiating the back surface of the wafer with light or laser while cooling the front surface. An N-type defect layer which is a lattice defect can be formed by proton irradiation and its low-temperature annealing treatment, and this N-type defect layer substantially functions as a high N ⁺ buffer layer.

JP 2001-160559 A

  However, since the NTD method is premised on the use of a nuclear reactor, there is a problem that the manufacturing cost of a semiconductor device increases. In addition, it is necessary to install a nuclear reactor near the semiconductor substrate manufacturing factory, which reduces the degree of freedom of factory construction and increases the cost of logistics. Furthermore, the NTD manufacturing facility cannot be easily updated in accordance with the increase in the diameter of the semiconductor substrate, and it is difficult to quickly respond to changes in the semiconductor process.

  In addition, the technique disclosed in Patent Document 1 does not allow protons to be uniformly irradiated over the entire depth direction of the semiconductor device. The specific resistance required for the device cannot be obtained.

  An object of the present invention is to provide an inexpensive power semiconductor device having a high withstand voltage and a method for manufacturing the same, in view of the above-described problems of the prior art.

The semiconductor device of the present invention, N-type semiconductor substrate to which a dopant is added, the surface of the surface layer portion of at least a part with the addition of P-type dopant in a high doping concentration than the doping concentration of the semiconductor substrate P-type region and the A first step of forming an N-type region by adding an N-type dopant having a doping concentration higher than the doping concentration of the semiconductor substrate; and a region other than the P-type region and the N-type region in the semiconductor substrate as a base region, seen containing a second step, the introduction of hydrogen at a high doping concentration than the doping concentration of the N-type dopant of the semiconductor substrate over the entire depth direction of the base region, the second step, a proton the comprising the steps of irradiating a plurality of times, and characterized by irradiating the protons by shifting the peak position of the projected range of protons distance obtained by dividing the half-width of the distribution of the projected range of protons integer It can be manufactured by the manufacturing method of that semiconductor device.

As this, by irradiating the protons in a plurality of times so as to shift the peak position of the projected range of protons can be uniformly introduced hydrogen by the base region.

In the second step, it is preferable to introduce hydrogen at least 1 × 10 ¹³ / cm ^{3 to} 1 × 10 ¹⁷ / cm ³ over the entire depth of the base region. Accordingly, in consideration of the carrier activation rate of hydrogen, the carrier density of the donor is at least 1 × 10 ¹³ / cm ^{3 and not} more than 5 × 10 ¹⁴ / cm ³ throughout the depth direction of the base region, A specific resistance of 10 Ωcm or more and 500 Ωcm or less ideal as the base region can be obtained.

It is also preferable to include a third step of introducing more than 1 × 10 ¹⁷ / cm ^{3 of} hydrogen into the base region. Thus, a buffer region in a semiconductor device having a punch-through type IGBT structure and a carrier accumulation region in a semiconductor device having a collector short type IGBT structure can be formed.

  According to the present invention, an inexpensive power semiconductor device having a high breakdown voltage and a manufacturing method thereof can be realized.

<Basic configuration of semiconductor device>
The semiconductor device 100 in the embodiment of the present invention has a cross-sectional structure shown in FIG. The semiconductor device 100 is an N-channel insulated gate bipolar transistor (IGBT) having an N-type buffer region.

  The semiconductor device 100 is formed based on a semiconductor substrate 10 to which an N-type dopant such as phosphorus (P) is added. The semiconductor substrate 10 is a silicon substrate generally manufactured by the Czochralski method (CZ method) or the floating zone method (FZ method).

The concentration of the N-type dopant added to the semiconductor substrate 10 immediately after manufacture is about 1 × 10 ¹² / cm ³ , and the specific resistance of the semiconductor substrate 10 is about 2 to ³ kΩcm. That is, the semiconductor substrate 10 is a high resistivity substrate that is almost an insulator.

A P-type base region 12 to which a P-type dopant such as boron (B) is added is formed on the surface layer of the semiconductor substrate 10 by ion implantation or diffusion. Further, the surface layer of the semiconductor substrate 10, a region shallower than the P-type base region 12 is selectively N ⁺ emitter region 14 N-type dopant is added in a high concentration and, exclusively the N ⁺ emitter region 14 A P ⁺ contact region 16 in which a P-type dopant is selectively added at a high concentration is formed in this region.

Further, a trench (groove) that penetrates the N ⁺ emitter region 14 and the P-type base region 12 and slightly reaches the semiconductor substrate 10 is formed. Silicon on the inner wall of the trench is thermally oxidized to form a gate insulating film 18 that is a thermal oxide film. After the gate insulating film 18 is formed, polycrystalline silicon is buried in the trench to form the gate electrode 20. Further, an emitter electrode 22 made of aluminum or the like is formed on the N ⁺ emitter region 14 and the P ⁺ contact region 16.

A P-type dopant is added to the surface layer on the back surface of the semiconductor substrate 10 by ion implantation or the like, and short-time heat treatment or low-temperature heat treatment such as laser annealing is performed to form a P ⁺ collector region 24. Further, a collector electrode 25 made of aluminum or the like is formed on the P ⁺ collector region 24.

A region other than the P-type base region 12, the N ⁺ emitter region 14, the P ⁺ contact region 16, and the P ⁺ collector region 24 in the semiconductor substrate 10 becomes the base region 27.

Furthermore, in the semiconductor device 100, as shown in FIG. 1B, the carrier concentration of the donor is increased by introducing hydrogen substantially uniformly over the entire depth direction of the semiconductor substrate 10. Generally, a large power element such as IGBT, so it is preferable to set the specific resistance of the base region from 10Ωcm in 500Omucm, hydrogen concentration 1 × 10 ¹³ / cm ³ or more 1 × 10 ¹⁷ / cm ³ or less Then, hydrogen is introduced so that the concentration distribution falls within the range of ± 10%. In this case, considering the carrier activation rate, the carrier density of the donor is in the range of 1 × 10 ¹³ / cm ^{3 to} 5 × 10 ¹⁴ / cm ³ . Further, by setting the donor carrier density to 1 × 10 ¹⁰ / cm ³ or more and 1 × 10 ¹⁵ / cm ³ or less by heat treatment or the like, the specific resistance in the base region can be set to 10 Ωcm to 500 Ωcm.

  In addition, as shown in the cross-sectional view of FIG. The semiconductor device 101 is a punch-through N-channel IGBT having a buffer region. In the semiconductor device 101, the same components as those of the semiconductor device 100 are denoted by the same reference numerals and description thereof is omitted.

On the back side of the semiconductor substrate 10 of the semiconductor device 10, as shown in FIG. 2 (b), several μm~ several tens μm approximately deeper than the P ⁺ collector region 24, the half-width number μm~ several tens μm a width of about A buffer region 26 into which hydrogen is introduced is provided in the region having Hydrogen introduced into the buffer region 26 increases the carrier concentration of the donor. Hydrogen is introduced into the buffer region 26 so that the peak concentration of hydrogen is about 1 × 10 ¹⁷ / cm ³ to 1 × 10 ¹⁸ / cm ³ .

In the semiconductor device 101, a region other than the P-type base region 12, the N ⁺ emitter region 14, the P ⁺ contact region 16, the P ⁺ collector region 24, and the buffer region 26 in the semiconductor substrate 10 becomes the base region 27.

<Another example of semiconductor device>
The present invention can be applied to a semiconductor device having a field effect transistor (MOSFET) structure and a semiconductor device having a collector short IGBT structure in addition to the punch-through semiconductor device 100.

A semiconductor device 102 having a MOSFET structure to which the present invention is applied has a cross-sectional structure shown in FIG. The semiconductor device 102 has substantially the same configuration as the punch-through IGBT semiconductor device 100 shown in FIG. 2A, but includes a P ⁺ collector region 24 and a buffer region 26 on the back surface side of the semiconductor substrate 10. Absent.

In the semiconductor device 102, the N ⁺ emitter region 14 in the semiconductor device 100 corresponds to an N + source region, and the N-type base region 27 corresponds to an N-type drain region. A drain electrode 25 such as aluminum is formed on the back surface of the base region 27.

  Also in the semiconductor device 102, as shown in FIG. 3B, donor carrier concentration is increased by introducing hydrogen substantially uniformly over the entire depth direction of the semiconductor substrate 10. The conditions for introducing hydrogen are preferably the same as those for the semiconductor device 100.

A semiconductor device 104 having a collector short IGBT structure to which the present invention is applied has a cross-sectional structure shown in FIG. The semiconductor device 104 has substantially the same structure as the punch-through IGBT semiconductor device 100 shown in FIG. 2A, but the N ⁺ collector short region 28 and the P-type base region are formed on the back side of the semiconductor substrate 10. The difference is that an N-type carrier accumulation region 29 is formed in a region deeper than 12, and the buffer region 26 is not provided.

As shown in FIG. 4B, the carrier accumulation region 29 is formed by introducing hydrogen into a region having a half-value width of several μm to several tens of μm so as to be in contact with the P-type base region 12. Hydrogen introduced into the carrier accumulation region 29 increases the carrier concentration of the donor. Hydrogen is introduced into the carrier accumulation region 29 so that the peak concentration of hydrogen is about 1 × 10 ¹⁷ / cm ³ to 1 × 10 ¹⁸ / cm ³ .

  Also in the semiconductor device 104, as shown in FIG. 4B, the donor carrier concentration is increased by introducing hydrogen substantially uniformly over the entire depth direction of the semiconductor substrate 10. The conditions for introducing hydrogen are preferably the same as those for the semiconductor device 100.

  In the collector short type IGBT structure semiconductor device 104, since the collector short region 28 is provided, the reduction of carriers is accelerated and the switching speed is improved. Further, the provision of the carrier accumulation region 29 makes it difficult for holes to escape, and the on-characteristics of the semiconductor device 104 are improved.

<First hydrogen introduction method>
The hydrogen introduction process is performed after the P-type base region 12, the N ⁺ emitter region 14, the P ⁺ contact region 16, the gate insulating film 18, the gate electrode 20, the emitter electrode 22, and the P ⁺ collector region 24 are formed.

  When hydrogen is introduced substantially uniformly over the entire depth direction of the semiconductor substrate 10, the semiconductor substrate 10 is irradiated with protons (protons) accelerated to a high energy of 1 MeV or more, preferably 10 MeV or more and 30 MeV or less. At this time, in order to introduce protons substantially uniformly over the entire depth direction of the semiconductor substrate 10, the proton range is adjusted to the proton incident side of the semiconductor substrate 10 as shown in FIG. 5. The absorber material 30 for this is arrange | positioned. As the absorber material 30, it is preferable to use aluminum having ion scattering characteristics comparable to that of silicon.

  For example, a proton having an energy of 17 MeV is implanted into a silicon semiconductor substrate 10 having a thickness of 150 μm, and the aluminum absorber material 30 is formed so that the average range of protons is about 75 μm from the surface of the semiconductor substrate 10. By setting the thickness to 1875 μm, hydrogen can be introduced so that the hydrogen concentration is uniform within ± 25% over the entire depth direction of the semiconductor substrate 10 by one proton irradiation.

After the introduction of hydrogen, the semiconductor substrate 10 is annealed at a temperature of 300 ° C. to 500 ° C. to activate the donor. As a result, about 1 to 1/100 of the hydrogen introduced into the semiconductor substrate 10 is activated. Accordingly, by setting the proton injection amount (or irradiation amount) to 1 × 10 ¹¹ / cm ² to 1 × 10 ¹⁵ / cm ² , the carrier density of the donor is 1 × over the entire depth direction of the semiconductor substrate 10. It can be set to 10 ¹³ / cm ³ to 1 × 10 ¹⁵ / cm ³ .

When hydrogen is introduced into the buffer region 26, the semiconductor substrate 10 is irradiated with protons accelerated to energy of several MeV from the P ⁺ collector region 24 side.

For example, by providing an absorber material of a predetermined thickness on the surface of the semiconductor substrate 10 and implanting protons having energy of 4 MeV from the P ⁺ collector region 24 side, a half width from a position approximately 10 μm deeper than the P ⁺ collector region 24 is obtained. Hydrogen can be distributed in a region of about 10 μm.

<Second hydrogen introduction method>
In the first hydrogen introduction method, hydrogen was introduced substantially uniformly over the entire depth direction of the semiconductor substrate 10 by one proton irradiation. However, in order to obtain high breakdown voltage characteristics, it is preferable to make the hydrogen concentration in the semiconductor substrate 10 more uniform. Therefore, in the second embodiment, hydrogen can be introduced with higher uniformity by proton irradiation using the absorber material 30 a plurality of times. At this time, protons are irradiated by shifting the peak position of the proton range by a distance obtained by dividing the half-value width of the proton range distribution by an integer.

  As shown in FIG. 6, the second hydrogen introduction method combines step 1 using the absorber material 30a having the first thickness and step 2 using the second absorber material 30b different from the absorber material 30a. Done.

  In Step 1, protons are simultaneously irradiated to the two semiconductor substrates 10a and 10b that are targets of hydrogen introduction. As shown in FIG. 6A, the back surface of the semiconductor substrate 10a and the front surface of the semiconductor substrate 10b face each other, and the back surface of the semiconductor substrate 10b and the absorber material 30a face each other. Then, protons are irradiated from the absorber material 30a side. At this time, the absorber material so that the peak of the concentration of hydrogen introduced (indicated by the line A in the figure) becomes the surface of the semiconductor substrate 10a and the half width of the hydrogen distribution becomes equal to the thickness of the semiconductor substrates 10a and 10b. The material and thickness of 30a and the acceleration energy of protons are set.

  In step 2, the absorber material 30a is replaced with the absorber material 30b, and protons are irradiated. As shown in FIG. 6B, the back surface of the semiconductor substrate 10a and the front surface of the semiconductor substrate 10b face each other, and the back surface of the semiconductor substrate 10b and the absorber material 30b face each other. Then, protons are irradiated from the absorber material 30b side. At this time, the peak of the concentration of hydrogen introduced (indicated by the line B in the figure) becomes the back surface of the semiconductor substrate 10a and the surface of the semiconductor substrate 10b, and the half width of hydrogen distribution is equal to the thickness of the semiconductor substrates 10a and 10b. Thus, the material and thickness of the absorber 30b and the acceleration energy of protons are set.

  The amount of hydrogen introduced in each of steps 1 and 2 is half of the amount finally introduced into the semiconductor substrate 10. By adding the two hydrogen introductions in Steps 1 and 2, hydrogen can be uniformly introduced into the entire depth direction of the semiconductor substrate 10a as shown in the hydrogen concentration profile of FIG.

  Subsequently, the semiconductor substrate 10a is removed, a new semiconductor substrate 10c is disposed, and the process of step 2 is performed. As shown in FIG. 4C, the back surface of the semiconductor substrate 10b and the front surface of the semiconductor substrate 10c face each other, and the back surface of the semiconductor substrate 10c and the absorber material 30b face each other. Then, protons are irradiated from the absorber material 30b side. At this time, the peak of the concentration of hydrogen introduced (indicated by the line C in the figure) becomes the back surface of the semiconductor substrate 10b and the surface of the semiconductor substrate 10c, and the half width of hydrogen distribution is equal to the thickness of the semiconductor substrates 10b and 10c. Thus, the material and thickness of the absorber 30b and the acceleration energy of protons are set.

  By the second step 2, hydrogen can be uniformly introduced into the entire depth direction of the semiconductor substrate 10b as shown in the hydrogen concentration profile of FIG.

  Subsequently, the semiconductor substrate 10b is removed, and a new semiconductor substrate is disposed between the semiconductor substrate 10c and the absorber material 30b, and the process of step 2 is performed. Thereafter, the semiconductor substrate subjected to proton irradiation twice is removed, and a new semiconductor substrate is placed between the remaining semiconductor substrate and the absorber material, and the process of step 2 is repeated, so that the entire region of the semiconductor substrate in the depth direction is repeated. Hydrogen can be introduced substantially uniformly over the entire area.

  For example, when the thickness of the semiconductor substrate to which hydrogen is introduced is 150 μm, in Step 1, the material of the absorber 30a is aluminum and the thickness is 1650 μm, and the acceleration energy of proton is 17 MeV. Protons can be introduced so that the concentration peak (average proton range Rp) is 1950 μm from the proton incident surface side of the absorber 30a and the half width of hydrogen distribution is 150 μm.

  In Step 2, the material of the absorber 30b is aluminum and the thickness is 1800 μm, and the acceleration energy of proton is 17 MeV, so that the peak of hydrogen concentration (average proton range Rp) is the proton of the absorber 30a. Protons can be introduced so that 1950 μm from the incident surface side and the half-value width of hydrogen distribution is 150 μm.

  In the present embodiment, two semiconductor substrates 10 are irradiated with protons simultaneously, but the present invention is not limited to this. Three or more semiconductor substrates 10 and the absorber material 30 may be superposed and irradiated with protons. It is preferable that the total thickness of the superposed semiconductor substrate 10 and the half width (FWHM) of the proton irradiation distribution coincide with each other, and the peak position (Rp) of the proton irradiation distribution is set to the center position of the superposed semiconductor substrate 10. It is.

<Third hydrogen introduction method>
In the first and second hydrogen introduction methods, hydrogen is introduced substantially uniformly over the entire depth direction of the semiconductor substrate 10 by irradiating protons using an absorber material. However, the hydrogen concentration in the semiconductor substrate 10 can be made uniform without using an absorber material.

  In the third hydrogen introduction method, as shown in FIG. 8, the front and back surfaces of a plurality of semiconductor substrates 10 are superimposed and irradiated with protons. The number of semiconductor substrates 10 is the value obtained by dividing the peak position (Rp) of the proton range by the half width (FWHM) of the proton introduction distribution under the condition of irradiating the semiconductor substrate 10 with protons having a predetermined acceleration energy. Match. In other words, protons are irradiated while shifting the peak position of the proton range by a distance obtained by dividing the half-value width of the proton range distribution by an integer.

  Subsequently, the semiconductor substrate 10 disposed on the side opposite to the proton incident side is removed, and a new semiconductor substrate 10 is disposed on the proton incident side to irradiate protons. Thereafter, the semiconductor substrate 10 opposite to the proton incident side subjected to proton irradiation twice is removed, and a new semiconductor substrate 10 is disposed on the proton incident side, and the process is repeated. As a result, as shown in the hydrogen concentration profile of FIG. 7, hydrogen can be introduced substantially uniformly over the entire depth direction of the semiconductor substrate.

  For example, when the thickness of a semiconductor substrate to which hydrogen is introduced is 150 μm, the number of semiconductor substrates 10 to be stacked is 13 and the acceleration energy of protons is 17 MeV, so that the hydrogen concentration peak (proton concentration) Protons can be introduced so that the average range Rp) is 1950 μm from the proton incident side and the half width of hydrogen distribution is 150 μm.

  In the present embodiment, hydrogen is introduced almost uniformly over the entire depth direction of the semiconductor substrate 10 by two proton irradiations, but the present invention is not limited to this, and proton irradiation is performed three or more times. Also good. Also in this case, the number of semiconductor substrates 10 is adjusted to a value obtained by dividing the average range (Rp) of protons in the semiconductor substrate 10 with respect to a predetermined proton acceleration energy by the half-value width (FWHM) of the proton range.

It is a figure which shows the structure of the semiconductor device which has IGBT structure in embodiment of this invention. It is a figure which shows the structure of the semiconductor device which has a punch through type IGBT structure in embodiment of this invention. It is a figure which shows the structure of the semiconductor device which has MOSFET structure in embodiment of this invention. It is a figure which shows the structure of the semiconductor device which has a collector short type IGBT structure in embodiment of this invention. It is a figure explaining the 1st hydrogen introduction method. It is a figure explaining the 2nd hydrogen introduction method. It is a figure which shows hydrogen concentration distribution and carrier concentration distribution with respect to the depth direction of a semiconductor substrate at the time of introduce | transducing hydrogen twice. It is a figure explaining the 3rd hydrogen introduction method.

Explanation of symbols

10 (10a, 10b, 10c) Semiconductor substrate, 12 P type base region, 14 N ⁺ emitter region, 16 P ⁺ contact region, 18 gate insulating film, 20 gate electrode, 22 emitter electrode, 24 P ⁺ collector region, 25 collector Electrode (drain electrode), 26 Buffer region, 27 Base region, 28 Collector short region, 29 Carrier accumulation region, 30 (30a, 30b) Absorber material, 100, 102, 104 Semiconductor device.

Claims (3)

A P-type region in which a P-type dopant having a doping concentration higher than that of the semiconductor substrate is added to at least a part of the surface layer portion of the surface of the semiconductor substrate to which the N-type dopant is added, and a doping higher than the doping concentration of the semiconductor substrate A first step of adding an N-type dopant at a concentration to form an N-type region ;
Hydrogen with a doping concentration higher than the doping concentration of the N-type dopant of the semiconductor substrate over the entire depth direction of the base region, with the region other than the P-type region and the N-type region in the semiconductor substrate as a base region A second step of introducing
Only including,
The second step is a step of irradiating protons a plurality of times,
A method for manufacturing a semiconductor device, comprising: irradiating protons by shifting a peak position of a proton range by a distance obtained by dividing a half-value width of a proton range distribution by an integer . In the manufacturing method of the semiconductor device according to claim 1 ,
In the semiconductor device, the second step introduces hydrogen of at least 1 × 10 ¹³ / cm ^{3 to} 1 × 10 ¹⁷ / cm ³ over the entire depth of the base region. Production method. In the manufacturing method of the semiconductor device according to claim 1 or 2 ,
The method further includes a third step of introducing more than 1 × 10 ¹⁷ / cm ^{3 of} hydrogen into the base region.

Images