JP4816390B2 - Semiconductor chip manufacturing method and semiconductor chip - Google Patents

Description

The invention includes a method of manufacturing a semiconductor chip of manufacturing a semiconductor chip by dividing the semiconductor substrate in the thickness direction, and to a semiconductor chip manufactured by the manufacturing method of the semiconductor chip.

Conventionally, in the manufacture of semiconductor chips, a semiconductor substrate that has been processed into a line to be divided and bonded to a sheet is expanded by stretching the sheet and applying stress in the surface direction of the semiconductor substrate. A method of manufacturing a semiconductor chip that is divided into semiconductor chips is used.
In recent years, as a processing method of a line to be divided, a study and research on a processing method using laser light (laser dicing) has been advanced. For example, Patent Document 1 discloses a processing technique for a semiconductor substrate using a laser. Has been. FIG. 10 is an explanatory view showing a dicing process using a conventional laser beam. FIG. 10A is an explanatory diagram of a modified region forming process by laser light irradiation, and FIG. 10B is an explanatory diagram of a semiconductor substrate dividing process.

As shown in FIG. 10A, a semiconductor substrate W made of a semiconductor such as silicon and having a semiconductor element D formed on the substrate surface is prepared, and the back surface of the substrate surface is bonded to a stretchable resin sheet S. To do. An adhesive layer B coated with an ultraviolet curable adhesive or the like is formed on the entire surface of the sheet S to which the semiconductor substrate W is adhered, and the entire back surface of the semiconductor substrate W is adhered to the adhesive layer B.
The laser head H that irradiates the laser light L includes a condenser lens CV that condenses the laser light L, and condenses the laser light L at a predetermined focal length. In the modified region forming step, the planned dividing line for dividing the semiconductor substrate W under the laser light irradiation conditions set so that the condensing point F of the laser light L is formed at a depth d from the surface of the semiconductor substrate W. The laser head H is moved along the DL (frontward in the figure), and the laser beam L is irradiated from the surface of the semiconductor substrate W. As a result, a modified region K by multiphoton absorption is formed in a path of depth d where the condensing point F of the laser beam L is scanned.

In the modified region K, the depth d of the condensing point F is adjusted along the planned dividing line DL, and the condensing point F is moved in the thickness direction of the semiconductor substrate W, thereby adjusting the thickness of the semiconductor substrate W. It is formed at a plurality of locations with a predetermined depth within the range.
Here, multiphoton absorption means that a substance absorbs a plurality of the same or different photons. Due to the multiphoton absorption, a phenomenon called optical damage occurs at the condensing point F of the semiconductor substrate W and in the vicinity thereof, thereby inducing a thermal strain, and a crack is generated in that portion, and the layer in which the crack is assembled. That is, the modified region K is formed.
Subsequently, as shown in FIG. 10B, by applying stress in the in-plane direction of the semiconductor substrate W (directions indicated by arrows F1 and F2 in the figure), the thickness of the substrate starts from the modified region K. Cracks are propagated in the vertical direction, and the semiconductor substrate W is divided along the division lines DL, thereby obtaining a semiconductor chip C (Patent Document 1).
JP 2002-205180 A

However, the conventional method causes the following problems when a thick semiconductor substrate is divided.
That is, as the depth d of the condensing point F is increased, the intensity of the laser beam condensed at the condensing point F is attenuated, so that the extent of the formed modified region K from the condensing point F is small. Become. In other words, the size of the region where the modified region K is formed is reduced. Since the modified region K formed in the vicinity of the back surface of the semiconductor substrate W serves as a starting point of division, a large force is required for dividing unless a sufficiently large modified region K is formed at this position. For this reason, cracks may not be allowed to progress from the modified region K, which causes the semiconductor substrate W to be left behind, resulting in a problem that the yield of the semiconductor chip is reduced.

Accordingly, an object of the present invention is to realize a method for manufacturing a semiconductor chip that can improve the yield of the semiconductor chip.
Another object of the present invention is to obtain a semiconductor chip manufactured by the semiconductor chip manufacturing method.

In order to achieve the above object, the invention according to claim 1 provides:
While moving the laser beam relative to the semiconductor substrate along the planned dividing line for dividing the semiconductor substrate in the thickness direction, the semiconductor substrate is irradiated with the converging point aligned and irradiated. A modified region forming step of forming a modified region by multiphoton absorption at the light spot;
The semiconductor substrate that has undergone this modified region forming step is expanded by expanding the sheet to which one of the substrate surfaces is bonded, so that the semiconductor substrate has a thickness along the planned dividing line starting from the modified region. In a method for manufacturing a semiconductor chip, comprising a dividing step of obtaining a semiconductor chip by dividing in a direction,
The extent of the modified region formed near the one substrate surface from the condensing point is larger than the extent of the modified region formed near the other substrate surface from the condensing point. And controlling the intensity of the laser light,
By controlling the intensity of the laser beam such that the closer to the one substrate surface, the larger the spread from the condensing point,
The technical means of forming the modified region at a plurality of locations in the thickness direction of the semiconductor substrate is used.

In the invention according to claim 2, technical semiconductor chip manufacturing method according to claim 1, wherein the modified region between the adjacent controls the intensity of the laser beam so as not to share some, that Use means.

The invention according to claim 3 uses a technical means that the semiconductor chip is a semiconductor chip manufactured by the method for manufacturing a semiconductor chip according to claim 1 or claim 2 .

According to the first aspect of the present invention, the intensity of the laser beam for forming the modified region is increased from the condensing point of the modified region and the position in the thickness direction for forming the modified region of the semiconductor substrate. Therefore, the modified region having a desired spread suitable for dividing the semiconductor substrate reliably can be formed at a predetermined depth in the thickness direction.
In other words, a modified region capable of reliably developing cracks with a small force can be formed, so that the semiconductor substrate can be reliably divided and the yield of the semiconductor chip can be improved. A method can be realized.

According to the first aspect of the present invention, the extent of the modified region formed near the one substrate surface from the condensing point is the concentration of the modified region formed near the other substrate surface. In order to control the intensity of the laser beam so as to be larger than the spread from the light spot, a modified region having a large spread is formed on the substrate surface side, which is the starting point of the crack propagation when the semiconductor substrate is divided, so that a small force Thus, cracks can be developed and the semiconductor substrate can be reliably divided.

According to the invention described in claim 1 , since the intensity of the laser beam is controlled so that the spread from the light condensing point of the modified region is closer to the one substrate surface, the starting point of crack propagation As the modified region closer to the substrate surface becomes larger, the spread becomes larger, and as the modified region closer to one substrate surface, the force required for the division becomes smaller. For this reason, cracks can be developed in order from the modified region in the vicinity of the substrate surface, which is the starting point of crack propagation during the division of the semiconductor substrate, toward the modified region in the vicinity of the other substrate surface. There is no risk of division failure due to deflection or the like.

According to the second aspect of the present invention, in order to control the intensity of the laser beam so that the adjacent modified regions do not share a part, the portion shared by the adjacent modified regions is recrystallized or It is possible to prevent the semiconductor substrate from being hard to be divided due to strong bonding due to remelting or the like.

When the modified region appearing on the dividing surface of the semiconductor chip is formed in the form of the modified region according to claim 1 or 2 , for example, the modified region The semiconductor chip is produced by the method for manufacturing a semiconductor chip according to claim 1 or 2 when the spread from the light condensing point becomes larger as the distance from the condensing point is closer to one substrate surface. It can be estimated that the semiconductor chip is manufactured.

<First Embodiment>
A first embodiment of a semiconductor chip manufacturing method according to the present invention will be described with reference to the drawings. FIG. 1 is an explanatory diagram showing a configuration of a semiconductor substrate. 1A is an explanatory plan view of a semiconductor substrate, and FIG. 1B is a cross-sectional view taken along the line 1A-1A in FIG. 1A. FIG. 2 is an explanatory diagram showing a method for irradiating a semiconductor substrate with laser light. FIG. 3 is a schematic view of a modified region formed by the semiconductor chip manufacturing method according to the first embodiment. FIG. 4 is an explanatory diagram of the relationship between the laser power obtained by simulation, the depth for forming the modified region, and the temperature of the focal point.
In each figure, a part is enlarged and exaggerated for explanation.

(Structure of semiconductor substrate)
As shown in FIG. 1A, a thin disk-shaped semiconductor substrate 21 made of silicon is prepared. As shown in FIG. 1B, the back surface 21 b of the semiconductor substrate 21 is bonded to a resin sheet 41 through an adhesive layer 52. The sheet 41 has stretchability, and the adhesive layer 52 is formed on the entire surface of the sheet 41 with an adhesive or the like. The outer periphery of the sheet 41 is held by an annular frame 42 so that the sheet 41 is stretched.
An orientation flat OF showing a crystal orientation is formed on a part of the outer periphery of the semiconductor substrate 21. On the substrate surface 21a of the semiconductor substrate 21, semiconductor elements 24 formed through a diffusion process or the like are arranged and arranged like a grid.
On the substrate surface 21a between the semiconductor elements 24, planned division lines DL1 to DL14, which are lines scheduled to divide the semiconductor substrate 21 in the thickness direction, are set in the thickness direction of the semiconductor substrate 21 toward the back surface 21b. ing. The division lines DL1 to DL7 are provided in a direction substantially perpendicular to the orientation flat OF and are set to be parallel to each other. The division lines DL8 to DL14 are provided in a direction substantially parallel to the orientation flat OF and are set to be parallel to each other. That is, the planned division lines DL1 to DL7 and the planned division lines DL8 to DL14 intersect each other vertically.

Each semiconductor element 24 is surrounded on its four sides by a planned division line DL. The semiconductor substrate 21 is divided in the thickness direction along the division lines DL, and a plurality of semiconductor chips 22 having the semiconductor elements 24 are obtained.
In the following description, a portion that is not divided from the semiconductor substrate 21 and that originally becomes a semiconductor chip after being divided is also referred to as a semiconductor chip. These semiconductor chips 22 are divided into a thickness direction along the division line DL by a dicing process, and then packaged by performing various processes such as a mounting process, a bonding process, and an encapsulating process. Complete.

  As shown in FIG. 1B, six semiconductor chips 22 a to 22 f are formed on the 1A-1A line of the semiconductor substrate 21. In order to divide these semiconductor chips 22a to 22f, seven division lines DL1 to DL7 and division lines DL11 and DL12 (FIG. 1A) (not shown in FIG. 1B) are set. In addition, in the planned division lines DL1 to DL7, DL11, and DL12, a modified region K (FIG. 3) serving as a starting point of division is formed in the thickness direction of the semiconductor substrate 21 by a method described later.

(Formation of modified regions by laser light irradiation)
As shown in FIG. 2, the semiconductor chip manufacturing apparatus 1 is provided with a laser head 31 that irradiates a laser beam L. The laser head 31 includes a condensing lens 32 that condenses the laser light L, and can condense the laser light L at a predetermined focal length. Here, the condensing point F of the laser light L is set so as to be formed at a depth d from the substrate surface 21 a of the semiconductor substrate 21.
Here, as the laser beam L to be irradiated, an appropriate laser type and wavelength can be selected in accordance with the structure and material of the semiconductor substrate 21. For example, a YAG laser, a carbon dioxide laser, a semiconductor laser, or the like is used. it can.

In order to form the modified region K in the semiconductor substrate 21, first, one of the division lines DL shown in FIG. 1A is scanned with the laser light for detecting the semiconductor substrate, and the irradiation range of the laser light L Set. Here, a case where the modified region K is formed in the planned division line DL4 will be described.
Subsequently, as shown in FIG. 2, the laser head 31 is scanned along the planned division line DL4 (in the direction F4 in the figure), and the laser beam L is irradiated from the substrate surface 21a, thereby condensing the laser beam L. A modified region K by multiphoton absorption is appropriately formed in the path of depth d scanned by F.

  Next, as shown in FIG. 3, the depth d of the condensing point F of the laser beam L is adjusted, and the condensing point F is moved in the thickness direction of the semiconductor substrate 21 to divide the modified region K. It is formed at a plurality of locations on the planned line DL4. Usually, in order to divide the semiconductor substrate 21 having a thickness of about 600 μm, about 30 modified regions K are formed in the thickness direction. In FIG. A case where the quality regions K1 to K8 are formed will be described.

The energy of the irradiated laser beam L is absorbed by the condensing point F, and a modified region K extending in the thickness direction and the surface direction of the semiconductor substrate 21 around the condensing point F is formed. The modified region K is formed in the shape of an elliptical rotary body having a semiconductor substrate 21 extending in the thickness direction larger than the surface direction expansion and having a vertically long longitudinal section.
In the following description, the extension of the modified region K in the thickness direction of the semiconductor substrate 21 is referred to as “vertical extension R1”, and the extension in the surface direction is used as “lateral extension R2”. When the size of the modified region K is indicated, the vertical spread R1 and the horizontal spread R2 are collectively referred to as “spread R” in the description.

Here, in the case of introducing a plurality of modified regions K in the thickness direction of the semiconductor substrate 21, if the modified region K is formed from the side closer to the substrate surface 21a, the modified laser beam L is formed first. Since the light is scattered when passing through the region K and the condensing point F becomes difficult to match, the modified region K having a sufficient size may not be formed. Therefore, the modified region K is preferably formed in order from the far side from the substrate surface 21a.
Therefore, the laser beam L is irradiated by controlling the distance M (FIG. 2) from the laser beam L emission surface of the laser head 31 to the substrate surface 21a so that the modified region K is formed in the order of K1 to K8. To do.

Control of the intensity of the laser beam L applied to the semiconductor substrate 21 is performed by forming a modified region K and a laser power P that is an input value of an apparatus for setting the intensity of the laser beam L obtained by simulation by the present inventors. This is performed based on the relationship between the depth d and the temperature T of the condensing point F. FIG. 4 shows the result of the simulation. “Focus” in the figure is the same as the depth d (FIG. 3). In the simulation, it is assumed that the laser power P and the intensity of the laser light L are proportional, and that the energy of the laser light L irradiated at the condensing point F is all consumed for heat generation.
The temperature T at the condensing point F rises in proportion to the laser power P, and the inclination thereof becomes gentler as the focal point (depth d) increases. That is, the larger the focal point (depth d), the larger the laser power P is required to raise the temperature T.

The modified region K is formed when the temperature T exceeds the melting point (1693 K) of silicon forming the semiconductor substrate 21. The vertical expansion R1 of the modified region K formed when the temperature T reaches just above the melting point is about 18 μm, and the horizontal expansion R2 is about 2 to 3 μm.
When the focus (depth d) is 620 μm, the laser power P needs to be about 0.9 W in order for the temperature T to reach the melting point of silicon. Similarly, when the focus (depth d) is 410 μm, the laser power P needs about 0.7 W, and when the focus (depth d) is 140 μm, the laser power P needs about 0.5 W. It is.
From these, when the temperature T reaches the melting point, the relationship of the following equation obtained by the method of least squares is obtained between the laser power P and the depth d.

P = 0.001 × d + 0.355 (1)
P: Laser power (W), d: Depth of condensing point F (μm)

From the above equation (1), the laser power P required to form the modified region K increases in proportion to the depth d. In other words, the intensity of the laser light L irradiated to the condensing point F is attenuated in proportion to the depth d, and the temperature T is also lowered.
Since the extent R of the modified region K changes according to the temperature T, when the laser beam P is irradiated with the laser power P constant, the extent R of the modified region K decreases as the depth d increases. Become.
For example, when the laser power P is set to 1.2 W and the laser beam L is irradiated, the modified region K8 in the vicinity of the substrate surface 21a is formed with a vertical expansion R1 of about 40 μm and a horizontal expansion R2 of about 4 to 6 μm. On the other hand, the modified region K1 in the vicinity of the back surface 21b is formed with a vertical expansion R1 of about 20 μm and a horizontal expansion R2 of about 2 to 3 μm. That is, under this condition, the modified region K1 has a similar shape in which the dimension of the modified region K8 is ½ times.

  Thus, the temperature T rises in proportion to the laser power P, and the expansion R of the reforming region K increases as the temperature T rises. Therefore, the relationship of the above equation (1), the temperature T and the reforming region By controlling the laser power P using the relationship with the spread R of K, the modified region K having the desired spread R at a predetermined depth d can be formed.

In the present embodiment, using the method for controlling the intensity of the laser light L as described above, as shown in FIG. 3, the closer to the back surface 21 b of the semiconductor substrate 21, the larger the spread R from the condensing point F. Then, the modified region K is formed. That is, the intensity of the laser beam L is controlled to be increased so that the modified region K closer to the back surface 21b has a larger size.
First, the modified region K1 closest to the back surface 21b is formed, and subsequently, the modified regions K2 to K8 are sequentially formed toward the substrate surface 21a. The reformed regions K1 to K8 are formed so as to expand and decrease in this order.
For example, the intensity of the laser beam L to be irradiated is set so that the vertical spread R1 of the modified region K1 is 40 μm and the vertical spread R1 of the modified region K8 is 20 μm, and the spread R of the modified regions K1 to K8 is set. The intensity of the laser beam L is controlled so that is proportional to the depth d. At this time, since the modified region K1 is formed at a position where the depth d is the largest and the spread R is the largest, the intensity of the irradiated laser beam L is maximized.
For other scheduled division lines DL, the reformed regions K1 to K8 are formed in the same manner as the division planned line DL4.

  Here, since the modified regions K1 to K8 are formed so that the spread R becomes smaller in this order, the intensity of the laser beam L applied to the modified region K in the vicinity of the substrate surface 21a is low. The temperature rise in the vicinity is small and there is no possibility of affecting the semiconductor element 24.

  In the modified region K, microcracks are introduced by being induced by a phase transformation of silicon. The amount of microcracks increases with the intensity of the laser beam L, and microcracks are introduced into the modified region K1 at a higher density than the modified region K8. That is, by controlling the intensity of the laser light L, the amount of microcracks introduced into the modified region K can also be controlled.

(Division of the semiconductor substrate 21)
Subsequently, by expanding the sheet 41 in the surface direction, stress is applied to the semiconductor substrate 21, cracks are developed starting from the modified region K, and the thickness of the semiconductor substrate 21 is increased along the division line DL. Split in direction.
As a method of expanding the sheet 41, for example, a semiconductor substrate is mounted from the back side of the sheet 41 using a pressing device (not shown) having a flat surface substantially the same size as the back surface 21b of the semiconductor substrate 21 with the frame 42 fixed. A known method of expanding the sheet 41 in the surface direction and applying a stress in the in-plane direction of the semiconductor substrate 21 can be used by pressing the substrate 21 so as to push it up.

  Here, the modified region K1 formed closest to the back surface 21b is formed so that the spread R is the largest among the modified regions K1 to K8. This effectively acts as a starting point of crack generation when 21 is divided. Further, since microcracks are introduced at high density in the modified region K1, the cracks are developed with a small force, and the semiconductor substrate 21 is reliably divided.

  In addition, since the modified regions K1 to K8 are formed so that the extent R is larger as they are closer to the back surface 21b, the force required for division in the order of the modified regions K1 to K8 is small. To the modified region K8, the cracks progress in order and the cracks are not deflected.

[Effect of the first embodiment]
(1) In order to control the intensity of the laser beam L according to the extent R of the modified region K from the condensing point F and the depth d of the semiconductor substrate 21 forming the modified region K, a predetermined depth d In addition, a modified region K having a desired spread R suitable for reliably dividing the semiconductor substrate 21 can be formed.
That is, since the modified region K that effectively acts for the division of the semiconductor substrate 21 can be formed, a method for manufacturing the semiconductor chip 22 that can improve the yield of the semiconductor chip 22 can be realized.

(2) The spread R from the condensing point F of the modified region K formed in the vicinity of the back surface 21b of the semiconductor substrate 21 extends from the condensing point F of the modified region K formed in the vicinity of the substrate surface 21a. In order to control the intensity of the laser beam L so as to be larger than R, the modified region K having a large size is formed on the back surface 21b side of the semiconductor substrate 21 that is the starting point of crack propagation, so that the crack is advanced with a small force. The semiconductor substrate 21 can be reliably divided.
That is, since the modified region K that effectively acts for the division of the semiconductor substrate 21 can be formed, a method for manufacturing the semiconductor chip 22 that can improve the yield of the semiconductor chip 22 can be realized.

(3) In order to control the intensity of the laser light L so that the extent R from the condensing point F becomes larger in the modified region K closer to the back surface 21b, the modified region K closer to the back surface 21b that is the starting point of crack propagation. As the spread R increases, the force required for the division decreases. Therefore, cracks can be developed and divided in order from the modified region K1 formed in the vicinity of the back surface 21b, which is the starting point of crack propagation, to the modified region K8 formed in the vicinity of the substrate surface 21a. Therefore, there is no possibility that the semiconductor substrate 21 may be poorly divided due to the deflection.

Second Embodiment
A second embodiment of a semiconductor chip manufacturing method according to the present invention will be described with reference to the drawings. FIG. 5A is a schematic diagram of a modified region in the case where adjacent modified regions share a part. FIG. 5B is a schematic view of a modified region formed by the semiconductor chip manufacturing method according to the second embodiment.
In addition, about the structure similar to 1st Embodiment, while using the same code | symbol, description is abbreviate | omitted.

  As shown in FIG. 2, when the modified region K is continuously formed at the same depth in the plane direction of the semiconductor substrate 21 along the planned division line DL4, if the modified region K is too wide, As shown in FIG. 5A, the modified region K may partially overlap in the thickness direction or the surface direction of the semiconductor substrate 21 and overlap. In this case, recrystallization, remelting, and the like occur in the overlapped portion Kw and are firmly bonded, and a large force may be required to divide the semiconductor substrate 21.

In order to avoid the above-described phenomenon, the vertical spread R1 and the horizontal spread R2 that are allowable because the modified region K does not share a part thereof are set, and the intensity of the laser beam L is controlled, as shown in FIG. As described above, it is possible to prevent the adjacent modified regions K from overlapping, and to ensure the expansion R of the modified region K that effectively acts for the division of the semiconductor substrate 21.
Here, FIG. 5B illustrates the case where the extent R of the modified region K is substantially the same, but a configuration in which the extent R is formed larger in the modified region K closer to the back surface 21b may be used.
Further, by controlling the irradiation time (speed, frequency) of the laser beam L, it is possible to prevent the adjacent modified regions K from overlapping.

[Effects of Second Embodiment]
(1) Since the intensity of the laser beam L is controlled so that adjacent modified regions K do not share a part, the portion Kw where the adjacent modified regions K overlap is recrystallized or remelted. It is possible to prevent the semiconductor substrate 21 from becoming difficult to be divided due to the strong bonding. Further, by controlling the irradiation time (speed, frequency) of the laser beam L, it is possible to prevent the adjacent modified regions K from overlapping.

<Other embodiments>
(1) The modified regions K1 to K8 are modified regions in which the spread R from the condensing point F of the modified region K formed in the vicinity of the back surface 21b of the semiconductor substrate 21 is formed in the vicinity of the substrate surface 21a. The intensity of the laser beam L may be controlled so as to be larger than the spread R of K from the condensing point F. In the first embodiment, the intensity of the laser light L is controlled so that the extent R from the condensing point F becomes larger in the modified region K closer to the back surface 21b, whereas, for example, as shown in FIG. The intensity of the laser beam L may be controlled so that the vertical expansion R1 of the modified regions K1 to K3 formed in the vicinity of the back surface 21b is 40 μm and the vertical expansion R1 of the modified regions K4 to K8 is 20 μm.
Further, the modified regions K4 to K8 may be formed by the laser beam L having the same intensity so that the spread R of the modified regions K4 to K8 increases in the order of K4 to K8.
Even when this configuration is used, since the modified region K that effectively acts on the division is formed in the vicinity of the back surface 21b of the semiconductor substrate 21, the effects (1) and (2) of the first embodiment can be obtained. Can play.

(2) The modified regions K1 to K8 may be formed by controlling the intensity of the laser light L so that the spread R from each condensing point F is substantially the same. That is, as shown in FIG. 7, the intensity of the laser beam L may be controlled so that the modified regions K1 to K8 have substantially the same spread R, for example, the vertical spread R1 is 40 μm.
Even when this configuration is used, since the modified region K that effectively acts on the division is formed in the vicinity of the back surface 21b of the semiconductor substrate 21, the effect (1) of the first embodiment can be achieved. .

(3) The modified region K can be formed by irradiating the laser beam L from the sheet 41 side.
For example, as shown in FIG. 8, the semiconductor substrate 21 is bonded using a sheet 41 formed of a material that can transmit the laser light L, and the laser light L is transmitted from the back surface 21 b side of the semiconductor substrate 21 through the sheet 41. Irradiate. Here, in order to avoid the influence of scattering of the laser beam L in the modified region K, the position of the laser head 31 is controlled to be away from the back surface 21b so that the modified region K is formed in order from K8 to K1. Then, the laser beam L is irradiated. By controlling the laser power P, a modified region K having a desired spread R at a predetermined depth d can be formed. In FIG. 8, the modified regions K <b> 1 to K <b> 8 are formed so that the spread R from each condensing point F is substantially the same, but the closer to the back surface 21 b of the semiconductor substrate 21, the farther from the condensing point F. The modified region K may be formed so that the spread R becomes large.

The modified regions K1 to K8 may be formed by combining the irradiation with the laser light L from the substrate surface 21a side of the semiconductor substrate 21 and the irradiation with the laser light L from the back surface 21b side.
For example, as shown in FIG. 9A, the modified regions K5 to K8 are formed in this order so that the laser beam L is irradiated from the substrate surface 21a side and the spread R from each condensing point F is substantially the same. To do. Next, as shown in FIG. 9B, while the semiconductor substrate 21 is adhered to the sheet 41, it is inverted with respect to the laser head 31 and irradiated with the laser light L from the back surface 21b side. The reformed regions K4 to K1 may be formed in this order so that the spread R of the two becomes substantially the same.
When this configuration is used, the depth d irradiated with the laser beam L can be reduced, so that it is not necessary to control the laser power P in a wide range.
It should be noted that any one of the modified regions K1 to K4 and the modified regions K5 to K8 may be formed first. Further, the modified region K may be formed so that the spread R from the condensing point F becomes larger as the back surface 21b of the semiconductor substrate 21 is closer.

(4) The laser light L is attenuated depending on the impurity concentration in the semiconductor substrate 21, and the attenuation rate increases as the impurity amount increases. Therefore, when the impurity concentration changes in the thickness direction of the semiconductor substrate 21, there is a possibility that the modified region K having the desired spread R is not formed in the region where the impurity concentration is high.
Therefore, a modified layer having a desired dimension can be obtained by controlling the laser power P in accordance with the impurity concentration profile with respect to the depth d in the division line DL. For example, in a region where the impurity concentration is high and the attenuation of the laser beam L is large, the laser power P may be controlled so that the intensity of the irradiated laser beam L is increased.
When this configuration is used, the modified region K having the desired spread R can be formed at a predetermined depth d even for the semiconductor substrate 21 whose impurity concentration varies in the thickness direction.
Also in the gettering layer that is an impurity layer, the modified region K can be appropriately formed by controlling the laser power P in accordance with the impurity concentration profile.

(5) The modified region K appearing on the dividing surface of the semiconductor chip 22 is formed such that the spread R from the condensing point F of the modified region K formed in the vicinity of the back surface 21b of the semiconductor substrate 21 is the surface of the substrate surface 21a. When the modified region K formed in the vicinity is formed so as to be larger than the spread R from the condensing point F, the semiconductor chip 22 is the semiconductor chip according to (1) of <Other Embodiments>. It can be estimated that the semiconductor chip is manufactured by the manufacturing method.
In the case where the modified region K appearing on the divided surface of the semiconductor chip 22 is formed so that the spread R from the condensing point F increases as the distance from the rear surface 21b increases, the semiconductor chip 22 is claimed in claim 1. It can be estimated that the semiconductor chip is manufactured by the method for manufacturing a semiconductor chip described in 1. above.
When the modified region K appearing on the dividing surface of the semiconductor chip 22 is formed so that the spread R from each condensing point F is substantially the same, the semiconductor chip 22 is <another embodiment>. It can be estimated that the semiconductor chip is manufactured by the method for manufacturing a semiconductor chip described in (2) .
When the modified region K appearing on the dividing surface of the semiconductor chip 22 is formed so that adjacent modified regions K do not share a part, the semiconductor chip 22 is described in claim 2 . It can be estimated that the semiconductor chip is manufactured by the semiconductor chip manufacturing method.
As described above, when the modified region K appearing on the dividing surface of the semiconductor chip 22 is formed by controlling the spread R from the condensing point F, the semiconductor chip 22 is described in claim 1. It can be estimated that the semiconductor chip is manufactured by the semiconductor chip manufacturing method.

(6) Although the semiconductor substrate made of only silicon is used as the semiconductor substrate 21, the application of the present invention is not limited to this. For example, an oxide film made of silicon oxide is used as the substrate surface of the semiconductor substrate 21. It is also possible to apply to a substrate formed on 21a, an SOI (Silicon On Insulator) wafer, or a substrate on which the modified region K can be formed by laser light irradiation.

[Correspondence between each claim and embodiment]
The back surface 21b corresponds to one substrate surface according to claim 1, and the spread R, the vertical spread R1, and the horizontal spread R2 correspond to the spread from the condensing point, respectively. Further, the substrate surface 21a corresponds to the other substrate surface according to claim 1 , respectively.

FIG. 1 is an explanatory diagram showing a configuration of a semiconductor substrate. 1A is an explanatory plan view of a semiconductor substrate, and FIG. 1B is a cross-sectional view taken along the line 1A-1A in FIG. 1A. It is explanatory drawing which shows the method of irradiating a semiconductor substrate with a laser beam. It is a schematic diagram of the modified region formed by the semiconductor chip manufacturing method according to the first embodiment. It is explanatory drawing of the relationship between the laser power calculated | required by simulation, the depth which forms a modification area | region, and the temperature of a condensing point. FIG. 5A is a schematic diagram of a modified region in the case where adjacent modified regions share a part. FIG. 5B is a schematic view of a modified region formed by the semiconductor chip manufacturing method according to the second embodiment. It is a schematic diagram of the modified area | region formed by the manufacturing method of the semiconductor chip which concerns on other embodiment. It is a schematic diagram of the modified area | region formed by the manufacturing method of the semiconductor chip which concerns on other embodiment. It is explanatory drawing which shows the process of irradiating a laser beam from the back surface of a semiconductor substrate, and forming a modified region. It is explanatory drawing which shows the example of a change of the process of irradiating a laser beam from the back surface of a semiconductor substrate, and forming a modified region. FIG. 9A is an explanatory diagram of a process of forming a modified region by irradiating a laser beam from the substrate surface of the semiconductor substrate, and FIG. 9B irradiates a laser beam from the back surface of the subsequent semiconductor substrate. It is explanatory drawing of the process of forming a modification area | region. It is explanatory drawing which shows the dicing process using the conventional laser beam. FIG. 10A is an explanatory diagram of a modified region forming process by laser light irradiation, and FIG. 10B is an explanatory diagram of a semiconductor substrate dividing process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Semiconductor chip manufacturing apparatus 21 Semiconductor substrate 21a Substrate surface (the other substrate surface)
21b Back surface (one substrate surface)
22, 22a-22f Semiconductor chip 24 Semiconductor element DL, DL1-DL14 Scheduled division lines K, K1-K8 Modified region L Laser light F Condensing point R Spreading (spreading from the condensing point)
R1 Vertical spread (spread from the focal point)
R2 Horizontal spread (spread from the focal point)

Claims (3)

While moving the laser beam relative to the semiconductor substrate along the planned dividing line for dividing the semiconductor substrate in the thickness direction, the semiconductor substrate is irradiated with the converging point aligned and irradiated. A modified region forming step of forming a modified region by multiphoton absorption at the light spot;
The semiconductor substrate that has undergone this modified region forming step is expanded by expanding the sheet to which one of the substrate surfaces is bonded, so that the semiconductor substrate has a thickness along the planned dividing line starting from the modified region. In a method for manufacturing a semiconductor chip, comprising a dividing step of obtaining a semiconductor chip by dividing in a direction,
The extent of the modified region formed near the one substrate surface from the condensing point is larger than the extent of the modified region formed near the other substrate surface from the condensing point. And controlling the intensity of the laser light,
By controlling the intensity of the laser beam such that the closer to the one substrate surface, the larger the spread from the condensing point,
A method of manufacturing a semiconductor chip, wherein the modified region is formed at a plurality of locations in the thickness direction of the semiconductor substrate . The method of manufacturing a semiconductor chip according to claim 1 , wherein the intensity of the laser beam is controlled so that adjacent modified regions do not share a part. A semiconductor chip manufactured by the method for manufacturing a semiconductor chip according to claim 1 .

Images