JP7615874B2 - Wafer manufacturing method - Google Patents

Description

The present invention relates to a method for manufacturing a wafer. In particular, the present invention relates to a coating formation method and a surface grinding technique for manufacturing a wafer with good warpage and nanotopography.

Substrate wafers are required to have small and good warpage, especially small and good warp, and small and good short wavelength waviness called nanotopography (hereafter referred to as NT). One technique for achieving the latter is a processing method in which one side of the wafer is coated with resin and then ground (e.g., Patent Documents 1 to 4).

For example, Patent Document 2 describes a method in which liquid resin is spread over one side of a wafer held by suction with a holding means to form a film of liquid resin, and then air is ejected downward from the suction surface of the holding means, and the pressure of this air ejection presses the wafer down from the suction surface, ensuring that the wafer is released from the holding means.

Patent Document 5 also discloses that during setup before resin coating, the pressure change that occurs when the distance between the pressing part and the stage is changed while gas is ejected from the suction port is memorized.

JP 2009-148866 A JP 2020-92161 A JP 2007-134371 A JP 2010-155298 A Patent No. 5670208 specification

In the method described in Patent Document 1, the shape of the holding means (platen) during resin pressing is transferred to the resin thickness distribution, which can affect the wafer shape (warping) after processing. Patent Document 2 discloses a conventional method of restoring the deformed elements of the wafer by supplying sound waves after resin pressing (after the wafer is separated from the upper platen, which is the holding means), but there was no technology to suppress the transfer of the holding means shape during resin pressing.

The present invention has been made to solve the above problems, and aims to provide a wafer manufacturing method capable of producing wafers with good warp and good nanotopography.

In order to solve the above problems, the present invention provides a method for manufacturing a wafer, comprising the steps of:
providing a wafer having a first major surface and a second major surface opposite the first major surface;
holding the second main surface of the wafer with a holding means;
disposing a resin in a plastic state facing the first main surface of the wafer;
pressing the wafer with the holding means to bring the first main surface of the wafer into contact with the resin;
While spreading the resin on the first main surface of the wafer, air is injected from the holding means between the holding means and the second main surface, thereby spreading the resin on the first main surface of the wafer in a state where the second main surface of the wafer is released from the holding means;
forming a gap between the wafer and the holding means while performing the air injection, thereby separating the wafer from the holding means;
After the wafer is separated from the holding means, the resin is cured to form a planarized resin layer, thereby obtaining a composite including the wafer and the planarized resin layer in contact with the first main surface of the wafer;
adsorbing and holding the composite using the planarizing resin layer as a reference surface, and grinding or polishing the second main surface of the wafer in that state;
removing the planarizing resin layer from the wafer;
The present invention provides a method for manufacturing a wafer, comprising: holding the second main surface of the wafer by suction, and grinding or polishing the first main surface of the wafer in this state.

According to the wafer manufacturing method of the present invention, the resin is spread over the first main surface of the wafer while the holding means for the second main surface of the wafer is released while air is injected, the wafer is separated from the holding means while air is injected, and the resin is cured after the separation, thereby preventing the shape of the holding means from being transferred to the wafer, and as a result, a wafer with good warp and good nanotopography can be manufactured.

By waiting without changing the height of the holding means while the wafer is separated from the holding means, the resin flows to form a gap between the wafer and the holding means, and the formation of the gap can be determined from the load value applied to the holding means.

For example, the wafer can be separated from the holding means by waiting without changing the height of the holding means while injecting air.

Alternatively, the gap between the wafer and the holding means may be formed by retracting the holding means at the separation between the wafer and the holding means.

In this way, the wafer and the holding means can be separated by retracting the holding means from the wafer. In this way, a sufficient gap between the wafer and the holding means can be secured in a relatively short time.

As described above, the wafer manufacturing method of the present invention can prevent the shape of the holding means from being transferred to the wafer, and as a result, it is possible to manufacture wafers with small and good warps and small and good nanotopography.

FIG. 1 is a schematic flow diagram showing an example of a method for producing a wafer according to the present invention. FIG. 4 is a schematic flow diagram showing another example of the wafer manufacturing method of the present invention. FIG. 1 is a schematic flow diagram showing an example of a conventional wafer manufacturing method.

As mentioned above, there was a need to develop a wafer manufacturing method that could produce wafers with good warp and good nanotopography.

After extensive research into the above problem, the inventors discovered that by spraying air while releasing the holding means from holding the second main surface of the wafer, spreading the resin over the first main surface of the wafer, and then spraying air to separate the wafer from the holding means, and then curing the resin after separation, it is possible to prevent the shape of the holding means from being transferred to the wafer, and as a result, it is possible to manufacture a wafer with good warp and good nanotopography, and thus completed the present invention.

That is, the present invention is a method for manufacturing a wafer, comprising the steps of:
providing a wafer having a first major surface and a second major surface opposite the first major surface;
holding the second main surface of the wafer with a holding means;
disposing a resin in a plastic state facing the first main surface of the wafer;
pressing the wafer with the holding means to bring the first main surface of the wafer into contact with the resin;
While spreading the resin on the first main surface of the wafer, air is injected from the holding means between the holding means and the second main surface, thereby spreading the resin on the first main surface of the wafer in a state where the second main surface of the wafer is released from the holding means;
forming a gap between the wafer and the holding means while performing the air injection, thereby separating the wafer from the holding means;
After the wafer is separated from the holding means, the resin is cured to form a planarized resin layer, thereby obtaining a composite including the wafer and the planarized resin layer in contact with the first main surface of the wafer;
adsorbing and holding the composite using the planarizing resin layer as a reference surface, and grinding or polishing the second main surface of the wafer in that state;
removing the planarizing resin layer from the wafer;
The method for manufacturing a wafer includes: suction-holding the second main surface of the wafer, and grinding or polishing the first main surface of the wafer in that state.

On the other hand, Patent Documents 1 to 5 do not mention or suggest spreading the resin onto the wafer while injecting air from the holding means to separate the wafer from the holding means.

The present invention will be described in detail below with reference to the drawings, but the present invention is not limited to these. In the drawings, the unevenness of the wafer before grinding and the warpage of the holding means are exaggerated for the purpose of explanation, but in the wafer manufacturing method of the present invention, the unevenness of the wafer before grinding and the warpage of the holding means are not limited to those as large as those shown in the drawings.

Figures 1 and 2 are schematic flow diagrams showing examples of the wafer manufacturing method of the present invention. The example of Figure 2 is the same as the example of Figure 1 except that the steps of Figures 2(E) and 2(F) are different from those of Figures 1(E) and 1(F). In the following, the example of Figure 1 will be described as a representative, and for the example of Figure 2, only the steps of Figures 2(E) and 2(F) will be described.

First, as shown in FIG. 1(A), a wafer 1 is prepared that has a first main surface 1A and a second main surface 1B opposite to the first main surface 1A. There are no particular limitations on the wafer 1 to be prepared.

Next, as shown in FIG. 1(B), the second main surface 1B of the wafer 1 is held by the holding means 2. In this example, the second main surface 1B of the wafer 1 is held by vacuum adsorption to the upper surface plate 2 serving as the holding means. By applying a vacuum through the through hole 2a that penetrates the upper surface plate 2 from top to bottom, the upper surface plate 2 can vacuum adsorb the wafer 1.

On the other hand, as shown in FIG. 1(B), resin 5 in a plastic state is placed so as to face the first main surface 1A of the wafer 1. In this example, a light-transmitting film 4 is laid on a glass surface plate 3 (lower surface plate) having a flat surface, and resin 5 in a plastic state, for example, in a liquid state, is supplied onto the film. There are no particular limitations on the resin 5, so long as it can be hardened.

Next, as shown in FIG. 1(C), the wafer 1 is pressed by the holding means 2 to bring the first main surface 1A of the wafer 1 into contact with the resin 5. In this example, the upper surface plate 2 serving as the holding means is lowered to press the wafer 1 held by suction on the upper surface plate 2 against the resin 5 from above, bringing the first main surface 1A of the wafer 1 into contact with the resin 5.

After the resin 5 and the wafer 1 come into contact, air injection is started. In this example, as shown in FIG. 1(D), air injection is started from the upper platen 2 to between the upper platen 2 and the second main surface 1B of the wafer 1 at the point where the resin 5 and the first main surface 1A of the wafer 1 come into contact (the position where the load applied to the upper platen 2 exceeds the set load (A)). Air injection can be performed, for example, through the through hole 2a of the upper platen 2.

Next, while continuing the air injection, the upper platen 2 is lowered until the resin 5 is sufficiently spread. In other words, the air injection is performed while spreading the resin 5 over the first main surface 1A of the wafer 1. In this way, in the present invention, the resin 5 is spread over the first main surface 1 of the wafer 1 while the second main surface 1B of the wafer 1 is released from the holding means 2.

While air is being sprayed, a portion of the second main surface 1B of the wafer 1 may be in contact with the upper platen 2 (e.g., point contact), but the second main surface 1B of the wafer 1 remains released from the holding means 2.

By pressing the resin 5 from above while injecting air in this manner, an air cushion layer is formed between the upper platen 2 and the second main surface 1B of the wafer 1, thereby preventing the shape of the upper platen 2 from being transferred to the wafer 1.

Next, the descent of the upper platen 2 is stopped at the set stopping load (B). In this example, the height of the upper platen 2, which is the holding means, is not changed and the process waits. Air injection continues during this time. This causes the resin 5 to naturally spread, and a gap 2b is formed between the wafer 1 and the upper platen 2, as shown in Figure 1 (E). This separates the wafer 1 and the upper platen 2.

Next, as shown in FIG. 1(E), the resin 5 spreads naturally, and when a sufficient gap is secured between the upper platen 2 and the wafer 1, the air injection is stopped. For example, the load value applied to the upper platen 2 is grasped, and when the load decreases to the set load (C), the air injection is stopped.

Alternatively, as shown in FIG. 2(E), after the set stop load (B) is detected, the upper platen 2 is withdrawn (moved upward) while continuing the air injection, and when a sufficient distance is secured between the upper platen 2 and the wafer 1, the air injection is stopped to separate the upper platen 2 and the wafer 1, forming a gap 2b between them.

By stopping the air injection when a sufficient gap is maintained between the wafer 1 and the upper platen 2, it is possible to prevent the wafer 1 and the upper platen 2 from coming into contact again due to the elasticity of the wafer 1 and the resin 5. As a result, it is possible to more reliably prevent the shape of the upper platen 2 from being transferred through the wafer 1 to the thickness of the resin.

Next, after the wafer 1 and the upper platen 2 are separated, the resin 5 is cured to form a planarized resin layer 6, as shown in FIG. 1(F), to obtain a composite 10 including the wafer 1 and the planarized resin layer 6 in contact with the first main surface 1A of the wafer 1.

In this example, since the lower platen 3 is a glass platen, ultraviolet light can be irradiated from below the lower platen 3, passing through the lower platen 3 and the light-transmitting film 4 and hitting the resin 5, thereby hardening the resin 5. Note that the means for hardening the resin 5 is not limited to ultraviolet light irradiation, and can be changed as appropriate depending on the type of resin 5. For example, if the resin 5 is a thermosetting resin, it can also be hardened by an external stimulus such as heat.

In the example shown in FIG. 2, after the wafer 1 and the upper platen 2 are separated as shown in FIG. 2(E), the resin 5 is cured in the state shown in FIG. 2(F) to form the planarized resin layer 6.

Next, as shown in FIG. 1(G), the composite 10 is transferred to, for example, a grinding stage 20 using the planarizing resin layer 6 as a reference surface, and the composite 10 is adsorbed and held. The grinding stage 20 is made of, for example, porous ceramic, and can hold the composite 10 by vacuum adsorption.

Next, as shown in FIG. 1(H), the second main surface 1B of the wafer 1 is ground or polished while being held by suction on the grinding stage 20. In this example, the second main surface 1B of the wafer 1 is ground (first grinding) using a grinding wheel 30 to obtain a ground second main surface 1C. For example, grinding is performed so that the total thickness variation (TTV) after grinding is 1 μm or less.

Next, as shown in FIG. 1(I), the planarizing resin layer 6 is removed from the wafer 1. In this example, the light-transmitting film 4 is also removed at the same time.

Next, as shown in FIG. 1(J), the wafer 1 is inverted and the ground second main surface 1C of the wafer 1 is suction-held by the grinding stage 20, and in this state the first main surface 1A of the wafer 1 is ground or polished. In this example, the first main surface 1A of the wafer 1 is ground using a grinding wheel 30 to obtain the ground first main surface 1D.

The wafer 1 thus obtained has a ground first main surface 1D and a ground second main surface 1C, as shown in FIG. 1(K). This wafer 1 is pressed and spread over the first main surface 1A of the wafer 1 while releasing the holding means (upper platen) 2 of the second main surface 1B of the wafer 1 while performing air injection, and the wafer 1 and the holding means 2 are separated while performing air injection, and the resin 5 is hardened after this separation, thereby preventing the shape of the holding means (upper platen) 2 from being transferred to the wafer 1. In other words, a good wafer with small warp can be manufactured. In addition, the second main surface 1B is ground while the flattened resin layer 6 is formed in contact with the first main surface 1A of the wafer 1, and then the wafer 1 is turned over and the first main surface 1A is ground, thereby making the nanotopography less than the target value and making it good.

On the other hand, conventional wafer manufacturing methods cannot produce good wafers with small warps, as will be explained below with reference to Figure 3.

The conventional wafer manufacturing method, the schematic flow diagram of which is shown in FIG. 3, mainly differs from the wafer manufacturing method of the present invention, shown as an example in FIG. 1(D) and FIG. 2(D) in FIG. 3(D). That is, in the conventional method shown in FIG. 3, even after the resin 5 and the wafer 1 come into contact with each other, the resin 5 continues to be spread over the first main surface 1A of the wafer 1 while the holding means (upper platen) 2 holds the second main surface 1B of the wafer 1. As a result, as shown in FIG. 3(E), after the holding by the upper platen 2 is released, the wafer 1 is obtained to which the shape of the upper platen 2 has been transferred. As a result, the wafer 1 obtained after grinding the second main surface 1B and the first main surface 1A has a large warp, as shown in FIG. 3(K).

The present invention will be specifically explained below using examples and comparative examples, but the present invention is not limited to these.

[Experiment details]
(Formation of planarizing resin layer on wafer)
<Examples 1-1 to 1-4>
In Examples 1-1 to 1-4, a composite including a wafer and a planarizing resin layer in contact with a first main surface of the wafer was produced under the following conditions in accordance with a flow similar to that shown in FIGS. 1(A) to 1(F).

Conditions for forming the coating (flattening resin layer) A P-type Si single crystal wafer having a diameter of 300 mm was used.
A UV-curable resin was used as the resin, and a PET film was used as the light-transmitting film.
A PET film was placed on a flat glass platen (lower platen), and 10 ml of a UV-curable resin was dropped onto the PET film.
The second main surface of the wafer was held by suction on a ceramic platen (upper platen) and lowered toward the resin, thereby pressing the resin.

In the pressing control, the servo motor holding the upper platen was driven, and when a predetermined load (A) of 100 N was detected, the vacuum suction was stopped and air injection was started.
The upper platen was lowered while continuing the air injection, and the lowering of the upper platen was stopped at a stop load (B) of 2000 N. This caused the resin to be spread over the first main surface of the wafer while releasing the second main surface of the wafer from the upper platen.
・By waiting while the air was injected, the resin spread naturally, and the load on the upper platen was reduced. This created a gap between the wafer and the upper platen, separating them.
After the load on the upper platen was reduced to the set load (C), the air injection was stopped.
・The resin was irradiated with ultraviolet light to harden it and create a flattened resin layer. (There was a gap between the wafer and the upper platen when ultraviolet light was irradiated.)
Here, taking into consideration the effect of transfer due to the gap between the wafer and the upper platen until the air injection was stopped, four examples were carried out with the set load (C) set to 1500 N (Example 1-1), 1000 N (Example 1-2), 500 N (Example 1-3), and 100 N (Example 1-4).
A UV-LED with a wavelength of 365 nm was used as the light source for curing the resin.

<Comparative Example 2>
In Comparative Example 2, a composite including a wafer and a planarizing resin layer in contact with the first main surface of the wafer was produced according to the same flow as in Figures 3(A) to 3(F). Specifically, the control of the pressing force was changed from that in Example 1-1 as follows.

In the pressing control, the servo motor holding the upper platen was driven, and the descent of the upper platen was stopped at a stopping load (B) of 2000 N. During this time, no air was injected, and the resin was spread over the first main surface of the wafer while the wafer was held by the upper platen.
-Then, the vacuum suction was stopped and air injection was started.
The upper platen was raised and stopped, and the air injection was stopped, forming a gap between the wafer and the upper platen.
・The resin was irradiated with ultraviolet light to harden it and create a flattened resin layer. (There was a gap between the wafer and the upper platen when ultraviolet light was irradiated.)
A UV-LED with a wavelength of 365 nm was used as the light source for curing the resin.

<Examples 2-1 to 2-3>
In Examples 2-1 to 2-3, a composite including a wafer and a planarizing resin layer in contact with the first main surface of the wafer was produced according to the same flow as in Figures 2(A) to 2(F). Specifically, the control of the pressing force was changed from that in Example 1-1 as follows.

In the pressing control, the servo motor holding the upper platen was driven, and when a predetermined load (A) of 100 N was detected, the vacuum suction was stopped and air injection was started.
The upper platen was lowered while continuing the air injection, and the lowering of the upper platen was stopped at a stop load (B) of 2000 N. This caused the resin to be spread over the first main surface of the wafer while releasing the second main surface of the wafer from the upper platen.
Next, while continuing the air injection, the upper platen was raised by a predetermined distance L and then stopped, and the air injection was then stopped, thereby forming a gap between the wafer and the upper platen.
・The resin was irradiated with ultraviolet light to harden it and create a flattened resin layer. (There was a gap between the wafer and the upper platen when ultraviolet light was irradiated.)
Here, taking into consideration the influence of the gap between the wafer and the upper platen on the transfer, three examples were carried out in which the above-mentioned predetermined distance L was set to 2 μm (Example 2-1), 5 μm (Example 2-2), and 10 μm (Example 2-3).
A UV-LED with a wavelength of 365 nm was used as the light source for curing the resin.

(Grinding)
<Comparative Example 1>
In Comparative Example 1, a P-type Si single crystal wafer having a diameter of 300 mm similar to that used in Example 1-1 was prepared, and this wafer was subjected to a grinding process under the following conditions.

<Examples 1-1 to 1-4, Comparative Example 2, Examples 2-1 to 2-3>
In Examples 1-1 to 1-4, Comparative Example 2, and Examples 2-1 to 2-3, the previously prepared composites were subjected to grinding under the following conditions.

Grinding conditions: DFG8360 manufactured by Disco was used for grinding.
A grinding wheel bonded with diamond grains was used.
The grinding method thereafter will be described separately since it differs between <Comparative Example 1> and the others (<Comparative Example 2>, <Examples 1-1 to 1-4>, <Examples 2-1 to 2-3>).

<Comparative Example 1>
First grinding conditions (back surface (second main surface) grinding)
The front surface (first main surface) of the wafer was vacuum-attached to the grinding stage, and the back surface (second main surface) was ground (without a planarizing resin layer).

Second grinding conditions (surface (first main surface) grinding)
The first ground surface (ground second main surface) was vacuum-adsorbed onto the grinding stage, and the surface (first main surface) was ground.
By adjusting the axis angle of the chuck table, the wafer thickness variation was adjusted to be 1 μm or less.

<Comparative Example 2>, <Examples 1-1 to 1-4>, <Examples 2-1 to 2-3>
First grinding conditions (back surface (second main surface) grinding)
The coating (flattening resin layer) side of the composite was vacuum-adsorbed, and the back surface (second main surface) was ground.
After grinding, the planarizing resin layer was peeled off.

Second grinding conditions (surface (first main surface) grinding)
The first ground surface (ground second main surface) was vacuum-adsorbed onto the grinding stage, and the surface (first main surface) was ground.
By adjusting the axis angle of the chuck table, the wafer thickness variation was adjusted to be 1 μm or less.

(Mirror polishing)
The front and back surfaces (ground first and second main surfaces) of each wafer that had been ground as described above were subjected to mirror polishing.

[Measurement results]
The warp and nanotopography (NT) of each wafer after mirror polishing were measured as follows.

For the measurement of Warp and NT, an optical interference type flatness/NT measuring device (KLA: WaferSight2+) was used.
As an index of NT, SQMM 2 mm x 2 mm was used.
The results are shown in Table 1 below.

In Comparative Example 1, grinding was performed without forming a flattening resin layer, resulting in poor results in both Warp and NT.

In Comparative Example 2, the NT was good because a planarizing resin layer was formed and then grinding was performed, but because the resin continued to be spread over the first main surface of the wafer while still being held by the upper platen, the shape of the platen was transferred, resulting in poor Warp.

On the other hand, in Examples 1-1, 1-2, 1-3, and 1-4, the resin was spread over the first main surface of the wafer while the upper platen was released from its holding position while air was being sprayed, and the resin was cured while the upper platen and the wafer were spaced apart to form a planarized resin layer. This reduced the transfer of the platen shape, and improved the warp compared to Comparative Example 2.

In particular, in Examples 1-3 and 1-4, where the air injection stop load was set to 500 N or less, a sufficient gap was secured between the wafer and the upper platen when the air injection was stopped, further improving the warp.

In addition, in Examples 2-1, 2-2, and 2-3, the resin was spread over the first main surface of the wafer while the upper platen was released from its holding position while air was being sprayed, and the resin was cured while the upper platen and the wafer were spaced apart to form a flattened resin layer. This reduced the transfer of the platen shape, and improved the warp compared to Comparative Example 2.

In particular, in Examples 2-2 and 2-3, in which the upper platen was retracted by a distance of 5 μm or more during air injection, a sufficient gap was secured between the wafer and the upper platen when air injection was stopped, further improving warp.

These results show that when forming the planarizing resin layer, the resin was spread over the first main surface of the wafer while the upper platen was released from its holding position while air was being sprayed, and the resin was cured while the upper platen and the wafer were spaced apart to form the planarizing resin layer, thereby preventing the transfer of the platen shape.

Furthermore, when the upper platen is placed on standby to separate the upper platen from the wafer, it is more desirable to set the load at the time the air injection is stopped to 500 N or less.

In addition, from the perspective of shortening the processing time, it is better to retract the upper platen while air is being sprayed, and more preferably, the distance by which the upper platen is retracted should be 5 μm or more.

The present invention is not limited to the above-described embodiment. The above-described embodiment is merely an example, and anything that has substantially the same configuration as the technical idea described in the claims of the present invention and exhibits similar effects is included within the technical scope of the present invention.

1...wafer, 1A...first main surface, 1B...second main surface, 1C...ground second main surface, 1D...ground first main surface, 2...holding means (upper surface plate), 2a...through hole, 2b...gap, 3...lower surface plate, 4...light-transmitting film, 5...resin, 6...flattened resin layer, 10...composite, 20...grinding stage, 30...grinding wheel.

Claims (2)

A method for manufacturing a wafer, comprising the steps of:
providing a wafer having a first major surface and a second major surface opposite the first major surface;
holding the second main surface of the wafer with a holding means;
disposing a resin in a plastic state facing the first main surface of the wafer;
pressing the wafer with the holding means to bring the first main surface of the wafer into contact with the resin;
While spreading the resin on the first main surface of the wafer, air is injected from the holding means between the holding means and the second main surface, thereby spreading the resin on the first main surface of the wafer in a state where the second main surface of the wafer is released from the holding means;
forming a gap between the wafer and the holding means while performing the air injection, thereby separating the wafer from the holding means;
After the wafer is separated from the holding means, the resin is cured to form a planarized resin layer, thereby obtaining a composite including the wafer and the planarized resin layer in contact with the first main surface of the wafer;
adsorbing and holding the composite using the planarizing resin layer as a reference surface, and grinding or polishing the second main surface of the wafer in that state;
removing the planarizing resin layer from the wafer;
4. A method for manufacturing a wafer, comprising: holding the second main surface of the wafer by suction; and grinding or polishing the first main surface of the wafer in that state. 2. The method for manufacturing a wafer according to claim 1, wherein the gap between the wafer and the holding means is formed by retracting the holding means at the separation between the wafer and the holding means.

Images