Patent Application
20090218386
The present invention relates to a soldering method of soldering a semiconductor element to a circuit board, a soldering apparatus, and a method for manufacturing semiconductor devices.
A conventional semiconductor module includes a ceramic substrate, a wiring layer, which is a metal plate joined to a surface of the ceramic substrate, and a joining layer, which is a metal plate joined to a back surface of the ceramic substrate. A semiconductor element is soldered (joined) to the wiring layer. A heat radiating device, that is, a heat sink for radiating heat generated by the semiconductor element is joined to the joining layer.
At a time of soldering, it is often the case that a void is generated in the solder in the process of solidification of the solder after the solder is molten. In the case that a lot of voids are generated in the solder, resistances of electricity and heat passing through the solder become higher. Further, if the size of one void becomes equal to or more than a certain degree, the electricity and the heat flow through the wiring layer and the circuit board while bypassing the void from the semiconductor element. Accordingly, a hot spot, which is a locally high temperature region, is generated about a void of the semiconductor element. As a result, the semiconductor element can be destroyed.
Accordingly, Patent Document 1 and Patent Document 2 propose a technique of suppressing the void generation. These publications propose a technique of soldering by evacuating a chamber at a time of heating solder so as to depressurize, and melting the solder in a state in which the degree of vacuum is high.
However, as shown in
Patent Document 1: Japanese Laid-Open Patent Publication No. 2005-230830
Patent Document 2: Japanese Laid-Open Patent Publication No. 2005-271059
An objective of the present invention is to provide a soldering method and a soldering apparatus that suppress the generation of voids.
In accordance with one aspect of the present invention, there is provided a soldering method of soldering a semiconductor element to a circuit board. The soldering method includes a step of accommodating an object-to-be-soldered in a chamber. The object-to-be-soldered includes the circuit board, the semiconductor element, and solder arranged between the circuit board and the semiconductor element. The solder has a melting temperature. The soldering method includes a step of achieving a reduction state in which the chamber is filled with an atmospheric gas at least including a reducing gas. The soldering method includes a step of melting the solder by heating the solder in such a manner as to raise a temperature of the solder to the melting temperature or higher, in the chamber in the reduction state. The soldering method includes a step of soldering the semiconductor element to the circuit board by solidifying the solder by lowering the temperature of the molten solder to lower than the melting temperature. The soldering method includes a step of structuring the chamber to be sealable, and a step of raising an internal pressure of the chamber to a melting start pressure equal to or higher than a normal pressure by the atmospheric gas until the rising temperature of the solder reaches the melting temperature. The melting start pressure is the internal pressure of the chamber at the time when the solder starts melting. The soldering method includes a step of achieving a pressurization state in which the internal pressure of the chamber is set to be equal to or higher than the melting start pressure, in a solder melting period. The solder melting period corresponds to a period until the molten solder is solidified after the solder starts melting. The soldering method includes a step of soldering the semiconductor element to the circuit board in the pressurization state.
Further, in accordance with another aspect of the present invention, there is provided a method for manufacturing semiconductor devices including a circuit board and a semiconductor element soldered to the circuit board.
Further, in accordance with another aspect of the present invention, there is provided a soldering apparatus for soldering a semiconductor element to a circuit board. The soldering apparatus includes a sealable chamber. A heating apparatus heats a solder arranged between the circuit board and the semiconductor element so as to melt the solder. The circuit board, the semiconductor element and the solder construct an object-to-be-soldered. The solder has a melting temperature. A gas introduction portion introduces an atmospheric gas at least including a reducing gas to the chamber. The gas introduction portion introduces the atmospheric gas to the chamber in a state in which the object-to-be-soldered is accommodated. The heating apparatus raises a temperature of the solder, to which the atmospheric gas has been introduced, to the melting temperature or higher so as to melt the solder. The gas introduction portion raises an internal pressure of the chamber to a melting start pressure equal to or higher than the normal pressure by the atmospheric gas, until the rising temperature of the solder reaches the melting temperature. The soldering apparatus is structured such as to achieve a pressurization state in which the internal pressure of the chamber is set to a pressure equal to or higher than the melting start pressure, in a solder melting period until the molten solder is solidified after the solder starts melting. The soldering apparatus is structured such as to solder the semiconductor element to the circuit board in the pressurization state.
a) is a plan view of the jig shown in
b) is a perspective view of the weight shown in
A description will be given below of one embodiment according to the present invention with reference to
As shown in
The semiconductor elements 12 include an insulated gate bipolar transistor (IGBT) or a diode. A plurality of, four in the present embodiment, semiconductor elements 12 are joined to the circuit board 11. The bonding layer 16 joins the heat sink 13 to the ceramic substrate 14. The bonding layer 16 is formed, for example, by aluminum or a copper. The heat sink 13 is joined to the bonding layer 16.
As shown in
As shown in
The lid 19 has a size which can close the opening 18a of the box main body 18. The box main body 18 and the lid 19 define a sealed space S within the chamber 17. The lid 19 includes a glass plate 22 opposing to the sealed space S. The glass plate 22 serves as a nonmagnetic and electrically insulating material.
As shown in
Inert gas feeding portion 24 for feeding an inert gas to the chamber 17 is connected to the box main-body 18. In the present embodiment, the inert gas is nitrogen gas (N2). The inert gas feeding portion 24 is provided with a piping 24a, an on-off valve 24b of the piping 24a, and a nitrogen tank 24c. Further, a vacuum portion 25 for evacuating the inside of the chamber 17 is connected to the box main body 18. The vacuum portion 25 is provided with a piping 25a, an on-off valve 25b of the piping 25a, and a vacuum pump 25c.
Further, a gas discharge portion 26 for discharging gas filling the chamber 17 to an outside is connected to the box main body 18. The gas discharge portion 26 is provided with a piping 26a, an on-off valve 26b of the piping 26a, and a throttle valve 26c, which is a second pressure regulating portion. The gas within the chamber 17 is discharged to the outside while being regulated in the discharging amount by the throttle valve 26c. The soldering apparatus HK is structured such as to be capable of regulating the pressure in the sealed space S by being provided with the reducing gas feeding portion 23, the inert gas feeding portion 24, the vacuum portion 25 and the gas discharge portion 26. In other words, the soldering apparatus HK pressurizes or depressurizes the sealed space S on the basis of pressure regulation.
A temperature sensor 27 measuring the temperature T within the chamber 17 is arranged in the box main body 18. The temperature sensor 27 is, for example, a thermo couple. In the present embodiment, the temperature sensor 27 is arranged in such a manner as to be capable of measuring the temperature T at a joining section of a semiconductor element 12 to a wiring layer 15, that is, a section which is soldered.
A plurality of high-frequency heating coils 28 serving as a heating device are located in an upper portion of the soldering apparatus HK, that is, above the lid 19. Six high-frequency heating coils 28 are arranged over the circuit boards 11 in such a manner as to individually correspond to six circuit boards 11 shown in
As shown in
a) shows a jig 32 used for soldering.
If a magnetic flux passing through each weight 35 is changed, an electric current is generated in the weight 35. The material of the weight 35 is selected in such a manner that the weight 35 generates heat on the basis of an electric resistance of the weight 35 itself. In the present embodiment, the weight 35 is made of stainless steel. As shown in
Next, a description will be given of a method by which the soldering apparatus HK solders the semiconductor elements 12 to the circuit board 11. As shown in
At a time of carrying out the soldering, first, the lid 19 is detached from the box main body 18, and the opening 18a is opened. As shown in
As mentioned above, an object-to-be-soldered 92 is arranged within the chamber 17. The object-to-be-soldered 92 includes the joined object 93, the solder sheets 33, and the semiconductor elements 12.
Next, the opening 18a is closed by attaching the lid 19 to the box main body 18, and a sealed space S is defined within the chamber 17. As shown in
Next, the gas within the chamber 17 is replaced. First, the inside of the chamber 17 is evacuated by operating the vacuum portion 25. Nitrogen gas is fed into the chamber 17 by operating the inert gas feeding portion 24. In other words, the sealed space S is filled with inert gas. Hydrogen gas is fed into the chamber 17 by repeating the evacuation and the feed of the nitrogen gas several times, and thereafter operating the reducing gas feeding portion 23. In other words, the inside of the chamber 17 is set to a reducing gas atmosphere.
Next, a high-frequency current is circulated to each of the high-frequency heating coils 28 by actuating the high-frequency generating device 29. Then, there is generated a high-frequency magnetic flux passing through each weight 35. An eddy current is generated in the weight 35. In other words, the weight 35 exposed to the magnetic flux generated by the high-frequency heating coil 28 generates heat on the basis of an electromagnetic induction effect. The heat of the weight 35 is transmitted to the corresponding semiconductor element 12 from the pressing surface 35a. As a result, the heat of the weight 35 is intensively, that is, locally transmitted to the joined section with the solder sheet 33 in the circuit board 11. As a result, the temperature T of the solder sheet 33 becomes equal to or higher than a melting temperature Tm, and the solder sheet 33 melts. The semiconductor elements 12 are pressed toward the circuit board 11 by the weight 35. Accordingly, the semiconductor elements 12 are prevented from being lifted up or being moved by a surface tension of the melting solder.
If the solder sheets 33 completely melt, the high-frequency generating device 29 is stopped. The solder is cooled until the melting solder is solidified. The melting solder is solidified by being cooled to the temperature lower than the melting temperature Tm, and joins the semiconductor elements 12 to the wiring layers 15. If the semiconductor elements 12 are joined to the wiring layers 15, the semiconductor module 100 is finished. Thereafter, the lid 19 is taken out from the box main body 18, the jigs 32 and the weights 35 are detached, and the semiconductor module 100 is taken out from the inside of the chamber 17. At a time of taking out the semiconductor module 100 from the chamber 17, the gas discharge portion 26 releases the gas within the chamber 17 to the atmospheric air.
The internal pressure P of the chamber 17 is raised and lowered on the basis of the measured temperature of the temperature sensor 27 and an elapse of a time. Since the reducing gas feeding portion 23, the inert gas feeding portion 24, the vacuum portion 25, and the gas discharge portion 26 feed gas into the chamber 17 or discharge gas from the inside of the chamber 17, the internal pressure P of the chamber 17 is raised or lowered. The pressure reducing valve 23c of the reducing gas feeding portion 23, and the throttle valve 26c of the gas discharge portion 26 circulate the reducing gas to the inside of the chamber 17 at a time of heating and cooling the solder.
A description will be given of a manner of regulating the atmospheric air within the chamber 17 at a time of heating and cooling the solder, in a first experimental example shown in
Dimensions of each of the semiconductor modules 10 used in the first experimental example and the second experimental example were as follows.
The ceramic substrate 14 was made of aluminum nitride. The ceramic substrate 14 was a quadrangular plate of 30 mm×30 mm having a thickness 0.635 mm. Each of the wiring layer 15 and the bonding layer 16 was made of pure aluminum, for example, 1000 series aluminum, which is an industrial pure aluminum. Each of the wiring layer 15 and the bonding layer 16 was a rectangular plate of 27 mm×27 mm having a thickness 0.4 mm. The thickness of the semiconductor elements 12 was 0.35 mm. The solder sheet 33 was made of a Sn—Cu—Ni—P based lead-free solder. The thickness of the solder sheet 33 was between 0.1 mm and 0.2 mm.
First, as shown by a graph in
Since the chamber 17 was evacuated, the internal pressure P of the chamber 17 at an initial time “to” indicated a state having a high degree of vacuum. The ambient atmosphere in the chamber 17 was replaced by a reducing gas atmosphere having a set pressure P1 which was higher than a normal pressure Po, at a first point in time t1. In the present specification, the normal pressure Po, that is, the atmospheric pressure was about 0.1023 MPa. The set pressure P1 was 0.13 MPa. The set pressure P1 was the internal pressure of the chamber 17 at a time when the solder sheet 33 starts melting, that is, a melting start pressure.
Heating of the solder sheet 33 was started at a second point in time t2 after the first point in time t1. In other words, the ambient atmosphere in the chamber 17 was replaced by reducing gas atmosphere having the set pressure P1 before the second point in time t2 at which the heating of the solder sheet 33 was started.
The temperature T of the solder sheet 33 reached a melting temperature Tm at a third point in time t3. In other words, the internal pressure P of the chamber 17 was raised to the normal pressure Po or higher at the first point in time t1, before the third point in time t3 at which the temperature T of the solder sheet 33 reached the melting temperature Tm. In the present specification, the melting temperature Tm of the solder sheet 33 is 217° C.
The solder 33 was heated until the temperature T of the solder sheet 33 reaches a set temperature T1 at a fourth point in time t4. The set temperature T1 was higher than the melting temperature Tm. The set temperature T1 was 250° C. In other words, the solder sheet 33 was heated between the second point in time t2 and the fourth point in time t4, after the gas replacement of the inside of the chamber 17 at the first point in time t1. The temperature T of the solder sheet 33 was maintained to the set temperature T1 between the fourth point in time t4 and a fifth point in time t5.
The internal pressure P of the chamber 17 was regulated between the first point in time t1 and a seventh point in time t7 in such a manner as to maintain the set pressure P1. In other words, the internal pressure P of the chamber 17 in the first experimental example was maintained at the set pressure P1 without being lowered to the normal pressure Po or lower (vacuum) between the second point in time t2 and the fifth point in time t5 at which the solder sheet 33 was heated. The heating of the solder sheet 33 was finished at the fifth point in time t5. In the first experimental example, the internal pressure P of the chamber 17 was maintained at the set pressure P1 even at a time of cooling the solder between the fifth point in time t5 and the seventh point in time t7. At a sixth point in time t6, the temperature T of the solder was lowered than the melting temperature Tm. When the solder was solidified, the internal pressure P of the chamber 17 was temporarily lowered to the normal pressure Po or less at the point in time t7, whereby the reducing gas was discharged. Thereafter, at an eighth seven point in time t8, the internal pressure P of the chamber 17 was recovered to the normal pressure Po by introducing the atmospheric air to the chamber 17.
An X-ray photograph on the right side of the graph in
Next, a description will be given of a transition, that is, a regulation of the internal pressure P of the chamber 17 in the second experimental example, as shown in a graph in
As shown in
An X-ray photograph on the right side of the graph in
As shown in
However, as shown in
In other words, the internal pressure P in the first comparative example was lowered to the normal pressure Po or less before the third point in time t3 at which the temperature T of the solder reached the melting temperature Tm at a time of heating the solder. The internal pressure P of the chamber 17 in accordance with the first comparative example was equal to or less than the normal pressure Po in both of a period between the second point in time t2 and the fourth point in time t4, which is the solder heating period, and a period between the fifth point in time t5 and the eighth point in time t8, which is the solder cooling period. The seventh point in time t7 does not exist in the graph in
An X-ray photograph on the right side of the graph in
In the second comparative example shown in
However, as shown in
An X-ray photograph on the right side of the graph in
It is quite obvious that the first experimental example in
The reducing gas having the higher pressure than the set pressure P1 (0.13 MPa) was fed from the hydrogen tank 23d in the solder melting period t3 to t7 of the first experimental example and the second experimental example. The pressure reducing valve 23c kept the internal pressure P of the chamber 17 at a fixed value, that is, the set pressure P1. The throttle valve 26c of the gas discharge portion 26 discharged a fixed amount of gas to the outside of the chamber 17. The reducing gas feeding portion 23 fed the reducing gas to the chamber 17 in such a manner as to compensate for a pressure reduction component of the internal pressure P of the chamber 17 caused by the gas discharge. As a result, the internal pressure P of the chamber 17 was kept at the fixed value. Further, the gas was circulated within the chamber 17. The internal pressure P of the chamber 17 in the solder melting period t3 to t7 was kept at a fixed value while taking into consideration a rising component of the internal pressure P of the chamber 17 caused by a temperature rising in the chamber 17 by heating the solder.
A consideration will be given to causes of the generation of voids on the basis of the results of experiments.
A surface tension of the molten solder is lowered as the temperature T of the solder rises. Since oxide exists on the surface of the solder, and the surface of the joined member (the semiconductor elements 12 and the wiring layers 15), a wettability of these surfaces is not good. Three kinds of materials including the solder, the joined member, and the atmospheric gas (the reducing gas in the present embodiment) cross in an interface in which the solder gets wet. On a cross line, or the line on which three kinds of materials cross, there exist a first surface tension applied between the joined member (solid body) and the atmospheric gas (gaseous body), a second surface tension applied between the molten solder (liquid body) and the atmospheric gas (gaseous body), and a boundary tension applied between the joined member (solid body) and the molten solder (liquid body). Each of the first surface tension, the second surface tension and the boundary tension is applied toward the corresponding interface direction from the cross line.
In many cases, the second surface tension between the molten solder and the atmospheric gas is great, and the boundary tension between the molten solder and the joined member has a negative value, immediately after the solder is molten. In this case, the solder is hard to be spread. The solder rather has a tendency to conglobate so that joined areas between the solder and the joined member are reduced. In order to suppress this tendency, it is effective to solder in a state in which the solder is pressurized by the weight 35 as in the present embodiment. For example, in a state in which a soft ball is sandwiched between a pair of upper and lower plates, the ball is collapsed by mounting a weight on the upper plate. Accordingly, the theory mentioned above is easy to understand. However, it is hard to prevent the conglobation tendency molten solder, only by the pressure of the ambient atmosphere. For example, even if the pressure of the atmospheric gas of a ball filled with water is raised, the ball is hard to be deformed from the spherical shape, however, the ball is easily collapsed by mounting a weight on the ball. Accordingly, the theory mentioned above is easy to understand.
As described in the background art of the present specification, the conventional void generation countermeasure heats the solder in the state in which the internal pressure P of the chamber 17 is set to the normal pressure Po or less (vacuum). This is based on the thinking that voids are generated by the atmospheric gas, or the gas generated from the residual gas, the solder or the like. In other words, it is thought that the generation of the void can be suppressed in the vented vacuum state.
However, as shown in
On the basis of these results, the inventor of the present invention considered that the inside of a void is in a low pressure state (a state having a high degree of vacuum), and the force generating voids is surface tension. The surface tension refers to a force minimizing the surface area of the liquid. The inventor of the present invention found out that a cylinder having a diameter of 1 mm, a height of 100 μm and a surface area 1×n×0.1 mm2 is more stable than the case where a unwet portion having a diameter of 1 mm exists in a spherical state being close to a close contact without being joined, that is, the case where the spherical surface area is 0.025×n mm2. Accordingly, if the inside of the void is in a state having a high degree of vacuum, it is considered that the void will disappears by applying a pressure overcoming the surface tension of the molten solder to the molten solder. In the case where soldering is carried out in the pressurized state as shown in
If the state of the void does not depend on presence or absence of gas, but depends on surface tension, the factors such as the material of the solder, the surface state of the semiconductor elements 12 and the wiring layers 15, the temperature T, the thickness of the solder and the like are expected to dominate the state of the void. In the case where the soldering was experimented under the same conditions by using the solder sheet 33 having a thickness of 100 μm and the solder sheet 33 having a thickness of 150 μm, a better result was obtained in the solder sheet 33 having a thickness of 150 μm.
For example, if two plates are dipped into a liquid body in a state in which a gap is formed between the plates, a liquid surface rises along the gap in the case that two plates are well wetted. The smaller the gap is, the higher the liquid surface rises. If the plate repels the liquid body due to poor wetting, the liquid surface is depressed. If the gap is small, the liquid surface is depressed. However, if the gap is large, the liquid surface is not significantly depressed.
The solder sheet 33 having a thickness of 100 μm corresponds to a case where the gap between a semiconductor element 12 and a wiring layer 15, which are two plates, is small. The solder sheet 33 having a thickness of 150 μm corresponds to a case where the gap between two plates is large. Accordingly, it is clear that the solder sheet 33 having a thickness of 150 μm suppresses voids on the basis of pressure more effectively than the solder sheet 33 having a thickness of 100 Mm. The inventor of the present invention believes that the generation of voids is suppressed by increasing the internal pressure P of the chamber 17.
The present embodiment has the following advantages.
(1) In the solder melting period t3 to t7 until the solder is solidified after starting melting, the soldering is carried out in the ambient atmosphere of the set pressure P1 equal to or higher than the normal pressure Po. Accordingly, the force overcoming the surface tension of the solder is applied to the molten solder. Therefore, it is possible to suppress the influence of the surface tension which is considered as the factor of the void generation, and it is possible to inhibit voids from being generated.
(2) In order to keep the internal pressure P of the chamber 17 at the fixed value, the reducing gas feeding portion 23 has the pressure reducing valve 23c. Accordingly, in the solder melting period t3 to t7, the stable pressurized state is achieved within the chamber 17, and it is possible to reliably inhibit voids from being generated. Particularly, in the case of solidifying the molten solder, the internal pressure P of the chamber 17 is lowered in accordance with the reduction of the temperature T within the chamber 17. It is possible to keep the internal pressure P of the chamber 17 at the normal pressure Po or higher by feeding the reducing gas to the chamber 17 from the pressure reducing valve 23c.
(3) The throttle valve 26c of the gas discharge portion 26 discharges the gas within the chamber 17 to the outside. Accordingly, the reducing gas is circulated between the inside and outside of the chamber 17. As a result, the water content within the chamber 17 generated by the reducing effect is removed by discharging the gas.
(4) As shown in
(5) The weights 35 are heated by the high-frequency heating coils 28 which are away from the weights 35. Accordingly, in the case where a plurality of semiconductor elements 12 are soldered to the circuit board 11 all at once, it is not necessary to provide the high-frequency heating coil 28 per weight 35. In other words, the high-frequency heating coils 28, the number of which is less than the number of the weights 35, can heat a greater number of joined sections on the circuit board 11 all at once.
Further, since the high-frequency heating coils 28 are away from the weights 35, it is possible to handle the high-frequency heating coils 28 independently from the weights 35 and the circuit board 11, at the time of cooling the molten solder. Accordingly, for example, in the case where a plurality of semiconductor modules 100 are arranged within the chamber 17, it is possible to improve the operating efficiency of the high-frequency heating coils 28 by moving the high-frequency heating coils 28 from a certain semiconductor module 100 to another semiconductor module 100.
Further, the present embodiment heats the joined section of the circuit board 11 by heating the weights 35 pressing the semiconductor elements 12. Accordingly, it is possible to concentrically transmit the heat to the joined section. Therefore, it is possible to improve the heating efficiency, for example, in comparison with the case of heating the entire circuit board 11 or the entire chamber 17.
(6) One high-frequency heating coil 28 is arranged above a plurality of weights 35 on one circuit board 11. Accordingly, it is possible to two-dimensionally transmit heat to a plurality of joined sections in one circuit board 11. Therefore, it is possible to uniformly heat a plurality of joined sections. As a result, it is possible to approximate the melting start timings of the solder sheets 33 arranged at the respective joined sections in such a manner as to be substantially simultaneous. Further, it is possible to approximate the timings at which all the solder sheets 33 finish melting in such a manner as to be substantially simultaneous. Therefore, it is possible to make the soldering work efficient.
(7) The high-frequency heating coils 28 are arranged outside of the chamber 17. Accordingly, the high-frequency heating coils 28 do not need to be held in positions in the soldering work except for the heating period. In other words, the high-frequency heating coils 28 can be detached from the chamber 17 except for the heating period. Therefore, it is possible to improve the production efficiency of the semiconductor modules 100 by moving one high-frequency heating coil 28 to other chambers 17 one after the other.
Further, a volumetric capacity of the chamber 17 in accordance with the present embodiment, in which the high-frequency heating coils 28 are arranged outside of the chamber 17, is smaller than, for example, the case where a heating members, that is, the high-frequency heating coils 28 are arranged within the chamber 17. Accordingly, it is possible to achieve a downsizing of the chamber 17.
The regulation of the ambient atmosphere mainly includes discharging the air from the chamber 17, that is, vacuuming, and feeding and discharging the inert gas such as nitrogen gas or the like and reducing gas such as hydrogen gas or the like. Accordingly, it is possible to reduce the time and the energy consumption necessary for discharging the air by reducing the volumetric capacity of the chamber 17. For example, it is possible to reduce the operating energy of the vacuum pump 25c. Further, it is possible to reduce the time and the energy consumption necessary for feeding or discharging inert gas or reducing gas to and from the chamber 17. It is also possible to reduce the consumption of inert gas and reducing gas.
(8) The lid 19, which is a portion of the chamber 17 that faces the high-frequency heating coils 28, is formed by the glass plate 22, which is an electrically insulating material. Accordingly, it is possible to prevent the chamber 17 from generating heat. Further, since the magnetic flux passes through the chamber 17, it is possible to heat the weights 35.
(9) One pressing surface 35a of the weight 35 can come into contact with the non-joining surfaces 12a of a plurality of semiconductor elements 12. In other words, one weight 35 corresponds to an assembly obtained by collecting a plurality of sub weights each of which is provided for pressing one semiconductor element 12. Accordingly, it is possible to enlarge the pressing surface 35a of one weight 35. Therefore, in comparison with the case where the pressing surface 35a is small, the weights 35 stably press each of the semiconductor elements 12. Accordingly, each of the semiconductor elements 12 is hardly affected by the surface tension of the molten solder, and the soldering work is stably carried out.
(10) One high-frequency heating coil 28 is allocated to one circuit board 11. Accordingly, the heat generating efficiency of the weights 35 is higher than the case where one high-frequency heating coil 28 is allocated to a plurality of circuit boards 11.
The embodiment mentioned above may be modified as follows.
As shown in
The set pressure P1 may be set higher than 0.13 MPa. The second set pressure P2 may be changed from 0.2 MPa. The set pressure P1 and the second set pressure P2 are set taking the durability of the chamber 17 into consideration.
The set pressure P1 is not limited to 0.13 MPa. In correspondence to the materials of the wiring layer 15 and the semiconductor elements 12 and the condition of the surface treatment, the set pressure P1 may be set to a range between 0.11 MPa and 0.13 MPa, inclusive. The closer to the normal pressure Po outside of the chamber 17, the more advantageous the internal pressure P of the chamber 17 in terms of the durability of the chamber 17. Further, the set pressure P1 can be changed in correspondence to the wettability and the surface tension of the molten solder. The inert gas has been conventionally fed to a reflow furnace used for soldering, for preventing the atmospheric air from making an intrusion into the reflow furnace. However, the feed of the inert gas only sets the internal pressure of the reflow furnace to about the normal pressure.
The internal pressure P of the chamber 17 in the solder melting period t3 to t7 may be kept at the set pressure P1, which is a fixed value, by introducing the reducing gas having a higher pressure than the set pressure P1 to the chamber 17 on the basis of the pressure value obtained by monitoring the internal pressure P of the chamber 17. Alternatively, the internal pressure P of the chamber 17 may be gradually increased by feeding the reducing gas.
In the embodiment mentioned above, a throttle valve may be connected to the reducing gas feeding portion 23. The internal pressure P of the chamber 17 may be gradually increased by feeding the reducing gas to the chamber 17 by the throttle valve.
In the embodiment mentioned above, the gas atmosphere of the chamber 17 at the time of heating or cooling the solder is the reducing atmosphere having 100% hydrogen. This may be changed, for example, to a mixed gas atmosphere in which hydrogen gas, that is, reducing gas is set to 3%, and the remainder is nitrogen gas, that is, inert gas.
The reducing gas is not limited to hydrogen gas, but may be, for example, gas of which the composition includes formaldehyde.
The heating system of the solder is not limited to the high-frequency induction heating by the high-frequency heating coils 28, but may be structured, for example, such that a heating device is provided within the chamber 17. Further, a heat transfer medium may be circulated in the heat sink 13. The heat sink 13 may heat the solder by transferring the heat to the solder sheet 33.
The soldering apparatus HK may be provided with an ambient atmosphere regulating device regulating an internal atmosphere of the chamber 17. The ambient atmosphere regulating device is connected to each of the on-off valves 23b, 24b, 25b and 26b and the vacuum pump 25c. The ambient atmosphere regulating device controls the reducing gas feeding portion 23, the inert gas feeding portion 24, the vacuum portion 25 and the gas discharge portion 26. As a result, it is possible to feed reducing gas and inert gas to the chamber 17, and it is possible to discharge gas and air from the chamber 17.
The joined object 93, to which the semiconductor elements 12 are soldered, may be a circuit board 11 to which no heat sink 13 is joined. In this case, the semiconductor device including the circuit board 11 and the semiconductor elements 12 is manufactured in the chamber 17. The number of the circuit boards 11 included in the semiconductor module 100 is not limited to six, but may be changed.
The lid 19 may be detachably mounted to the box main body 18, or may be structured as an opening and closing type.
The portion facing the high-frequency heating coil 28 in the lid 19 may be formed by an electrically insulating material other than glass, for example, ceramics or resin. In order to ensure that the lid 19 has a strength that stands against the atmospheric pressure difference between the inside and the outside of the chamber 17, it is preferable that the lid 19 be structured by a compound material, for example, of glass fiber and resin, that is, a glass fiber reinforced plastic (GFRP). Further, the lid 19 may be formed by a nonmagnetic metal. In the case where the lid 19 is formed by magnetic metal, it is preferable to employ a material having a higher electric resistivity than the weights 35. The lid 19 may be structured by a compound material of metal and insulating material. It is preferable to arrange an electromagnetic steel plate of a ferromagnetic material in a portion immediately above the weight in order to effectively introduce magnetic flux to the weights 35.
The weights 35 are not limited to integrated parts each formed by machining a material. Each weight 35 may be one assembly formed by joining a plurality of divided bodies.
In place of the weights 35, a plurality of sub weights each corresponding to one semiconductor element 12 may be employed. Specifically, four sub weights are prepared in correspondence to four semiconductor elements 12 joined to one circuit board 11. Each of the sub weights is arranged immediately above the corresponding semiconductor element 12.
The component of the solder sheet 33 is not limited to the embodiment mentioned above. In order to inhibit voids from being generated, it is preferable to pressurize the molten solder, that is, the solder in the molten state, and the component of the solder sheet 33 is not limited.
In the embodiment mentioned above, the reducing gas feeding portion 23 connected to the gas inlet of the chamber 17 has the pressure reducing valve 23c. The gas discharge portion 26 connected to the outlet of the chamber 17 has the throttle valve 26c. However, the layout of the pressure reducing valve and the throttle valve may be changed. For example, the reducing gas feeding portion 23 may have the pressure reducing valve 23c and a throttle valve, and the gas discharge portion 26 may also have a pressure reducing valve and the throttle valve 26c. Inversely to the embodiment mentioned above, the reducing gas feeding portion 23 may have only a throttle valve, and the gas discharge portion 26 may have only a pressure reducing valve. Only the reducing gas feeding portion 23 may have one of the pressure reducing valve and the throttle valve.
In this case, the pressure reducing valve can keep the internal pressure P of the chamber 17 constant. In the case where the throttle valve 26c is provided without the provision of the pressure reducing valve 23c, it is possible to gradually increase the internal pressure P of the chamber 17 by setting the flow rate of the gas fed to the chamber 17 higher than the flow rate of the gas discharged from the chamber 17. The pressure regulating portion connected to the inlet of the chamber 17 serves as the first pressure regulating portion. The pressure regulating portion connected to the outlet of the chamber 17 serves as the second pressure regulating portion.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2005-380186 | Dec 2005 | JP | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/JP2006/326074 | 12/27/2006 | WO | 00 | 12/9/2008 |
