Method for calculating thickness removal during post-dicing wafer grinding

Abstract

The application provides a method for calculating the thickness removal amount of a wafer after cutting and grinding, which comprises the following steps: collecting a plurality of wafers produced based on the same target cutting thickness as sample wafers to calculate the total thickness value X1 of the wafer surface to be removed after cutting and to calculate the thickness fluctuation value X2 of the wafer after grinding; and calculating the total removal amount X of the wafer thickness during actual grinding according to the sum of the total thickness value X1 and the thickness fluctuation value X2. That is, the application directly obtains the specific thickness removal amount of the wafer after cutting and grinding through formula calculation, thereby saving verification cost and reducing the consumption of manpower and material resources.

Description

Technical Field

[0001] This invention relates to the field of semiconductor manufacturing technology, and in particular to a method for calculating the thickness removal amount during wafer grinding after dicing. Background Technology

[0002] Semiconductor processing steps include cutting, grinding, etching, and polishing. Among these, the step after cutting is grinding. The main purpose of grinding is to remove a certain thickness from the wafer to smooth the surface of the wafer after cutting, and to remove cutting marks and damage. In industrial manufacturing, it is necessary to define the specific value of the thickness removal during grinding in order to carry out industrial production.

[0003] Currently, silicon wafer manufacturers lack an effective formula for calculating the required thickness removal amount during polishing. The common practice is to select a fixed thickness removal amount and then proceed with further processing, judging the accuracy of the initially determined removal amount based on the surface condition and yield of the processed wafers. This method is time-consuming and requires repeated verification with numerous wafers. If cutting conditions change, extensive re-experimentation is necessary, consuming significant manpower and resources. Summary of the Invention

[0004] The purpose of this invention is to provide a method for calculating the thickness removal amount during wafer grinding after dicing, so as to solve one or more problems existing in the existing methods for calculating the thickness removal amount required for wafer grinding.

[0005] To solve the above-mentioned technical problems, the present invention provides a method for calculating the thickness removal amount during wafer grinding after dicing, comprising:

[0006] Multiple wafers produced based on the same target cutting thickness were collected as sample wafers to calculate the total thickness X1 that needs to be removed from the wafer surface after cutting and the thickness fluctuation X2 of the wafer after grinding; and,

[0007] The total amount of wafer thickness removed during actual grinding is calculated based on the calculated total thickness value X1 and the thickness fluctuation value X2.

[0008] Optionally, in the method for calculating the thickness removal amount during wafer grinding after dicing, the method for calculating the total thickness value X1 includes:

[0009] Perform one or more of the following calculations on the plurality of said sample wafers to calculate the total thickness:

[0010] Calculate the thickness variance σ of all the sample wafers. S ;

[0011] Calculate the mean of the thickness ranges of all the sample wafers. and variance σTTV(S) ;

[0012] Calculate the average depth of the maximum line mark depth in each measurement area on the sample wafer. and variance σ D ;

[0013] Calculate the average depth of the maximum damage depth in each measurement region on the sample wafer. and variance σ d .

[0014] Optionally, in the method for calculating the thickness removal during wafer grinding after dicing, multiple sample wafers are produced using different dicing machines and / or different dicing guides; and / or,

[0015] The sample wafers were selected from the head, middle and tail sections of the crystal rod.

[0016] Optionally, in the method for calculating the thickness removal amount during wafer grinding after dicing, the average depth of the maximum line mark depth in each measurement area of ​​the sample wafer is calculated. and the average depth of the maximum damage depth At that time, the selected measurement areas on the sample wafer are distributed throughout the sample wafer.

[0017] Optionally, in the method for calculating the thickness removal during wafer grinding after dicing, the area of ​​each measurement region is (0.5~1.5) cm². 2 The distance between each measurement area and the edge of the wafer is less than 5 mm.

[0018] Optionally, in the method for calculating the thickness removal amount during wafer grinding after dicing, the total thickness value X1 is calculated using the following formula:

[0019]

[0020] Where m is a natural number between 3 and 6.

[0021] Optionally, in the method for calculating the thickness removal amount during wafer grinding after dicing, the method for calculating the thickness fluctuation value X2 of the wafer after grinding includes:

[0022] The thickness fluctuation value is calculated based on at least one of the following calculation results:

[0023] Calculate the thickness variance σ of all the sample wafers after grinding. G ;

[0024] Calculate the mean of the thickness range values ​​of all the sample wafers after grinding. and variance σ TTV(G) .

[0025] Optionally, in the method for calculating the thickness removal amount during wafer grinding after dicing, the thickness fluctuation value X2 is calculated using the following formula:

[0026]

[0027] Where m is a natural number between 3 and 6.

[0028] Optionally, in the method for calculating the thickness removal amount during wafer grinding after dicing, the value of m is 3 or 6.

[0029] In summary, the method for calculating the thickness removal amount during wafer grinding provided by this invention includes: collecting multiple wafers produced based on the same target cutting thickness as sample wafers to calculate the total thickness value X1 to be removed from the surface of the wafer after cutting and to calculate the thickness fluctuation value X2 of the wafer after grinding; and calculating the total thickness removal amount X of the wafer during actual grinding based on the sum of the calculated total thickness value X1 and the thickness fluctuation value X2. That is, this invention directly calculates the specific thickness removal amount required for the wafer during grinding using a formula, saving verification costs and reducing the consumption of manpower and material resources. Attached Figure Description

[0030] Figure 1 This is a schematic diagram illustrating the process of calculating the thickness removal amount during wafer grinding after dicing in an embodiment of the present invention. Detailed Implementation

[0031] The following detailed description, in conjunction with the accompanying drawings and specific embodiments, provides a more detailed explanation of the method for calculating the thickness removal during wafer grinding after dicing, as proposed in this invention. The advantages and features of this invention will become clearer from the following description and claims. It should be noted that the drawings are all in a very simplified form and use non-precise scales, intended only to facilitate and clarify the illustration of the embodiments of this invention. Furthermore, the structures shown in the drawings are often part of the actual structures. In particular, different figures may emphasize different aspects and sometimes use different scales.

[0032] Please see Figure 1 This invention provides a method for calculating the thickness removal amount during wafer grinding after dicing, comprising the following steps:

[0033] Multiple wafers produced based on the same target cutting thickness were collected as sample wafers to calculate the total thickness X1 that needs to be removed from the wafer surface after cutting and the thickness fluctuation X2 of the wafer after grinding; and,

[0034] The total amount of wafer thickness removed during actual grinding is calculated based on the calculated total thickness value X1 and the thickness fluctuation value X2.

[0035] The following provides a more detailed description of each of the above steps.

[0036] First, multiple wafers produced based on the same target cutting thickness are collected as sample wafers. Preferably, the multiple sample wafers are produced using different cutting machines and / or different cutting guides. The more sample wafers collected, the better the accuracy of subsequent calculations. Additionally, preferably, the multiple sample wafers are selected from the head, middle, and tail of the crystal ingot, and n wafers can be selected from each of the head, middle, and tail, where n can be 1, 2, 3, etc.

[0037] Next, calculate the total thickness X1 that needs to be removed from the wafer surface after dicing.

[0038] The inventors' research revealed:

[0039] a. Cutting is a mechanical process. Due to the precision of the materials of the relevant components, the thickness of the wafer may fluctuate under the same process conditions. In order to ensure that the product meets the requirements after grinding, it is better to calculate the value of this thickness difference fluctuation when grinding removal amount.

[0040] b. The thickness range (TTV, the difference between the maximum and minimum thickness values) of the wafers after cutting varies considerably. Therefore, in order to ensure that the product meets the requirements after grinding, it is best to calculate the TTV value of the wafers after cutting to determine the amount of material removed during grinding.

[0041] c. Currently, wire cutting is used for crystal rod cutting. After cutting, there may be a certain depth of wire marks on the surface of the wafer. In order to ensure that the product meets the requirements after grinding, the grinding removal amount needs to be calculated to determine the depth of the wire marks on the surface of the wafer after cutting.

[0042] d. During cutting, the steel wire and slurry particles will interact mechanically with the wafer, leaving a damage layer of a certain depth on the wafer surface. In order to ensure that the product meets the requirements after grinding, the amount of material removed by grinding should ideally be calculated based on the damage depth value of the wafer surface after cutting.

[0043] In view of the above findings, the method for calculating the total thickness value X1 in this embodiment includes: performing one or more of the following calculations on a plurality of the sample wafers to calculate the total thickness value X1:

[0044] (1) Calculate the thickness variance σ of all the sample wafers. S ;

[0045] (2) Calculate the mean of the thickness range of all the sample wafers. and variance σ TTV(S) ;

[0046] (3) Calculate the average depth of the maximum line mark depth in each measurement area on the sample wafer. and variance σ D ;

[0047] (4) Calculate the average depth of the maximum damage depth in each measurement area on the sample wafer. and variance σ d .

[0048] Specifically, the total thickness value X1 can be calculated using the following formula:

[0049]

[0050] Where m is a natural number between 3 and 6.

[0051] Preferably, m takes the value of 3 or 6. That is, the value of X1 can be managed using 3σ or 6σ.

[0052] When 3σ management is adopted

[0053] When using 6σ management

[0054] More specifically, the process of calculating the above values ​​is as follows:

[0055] The average thickness of each wafer was measured using a flatness measuring machine, and the thickness variance σ was calculated based on the measurement results. S ;

[0056] The TTV of each wafer was tested using a flatness measuring machine. S Value, the mean is calculated based on the measurement results. and variance σ TTV(S) ;

[0057] Multiple measurement areas are selected on the sample wafer. For each selected area, a profilometer is used to test the surface height undulation. During testing, the profilometer pointer moves in a direction perpendicular to the cutting line. The maximum depth of the height undulation in each area is recorded as D, with the first area as D1, the second as D2, and so on. The maximum depth of the undulation in the Nth area is recorded as D. n Calculate the average depth based on the measurement results. and variance σ D .

[0058] For each selected region, TEM (transmission electron microscopy) was used to measure the interface perpendicular to the line mark direction, and the maximum damage depth d was calculated. The first region was d1, the second region was d2, and so on, with the Nth region being d1. n Calculate the average depth based on the measurement results. and variance σ d .

[0059] When selecting measurement areas, preferably, the number of measurement areas can be 5, 9, 17, etc., distributed across the wafer surface. The area of ​​each measurement area is preferably (0.5~1.5)cm*(0.5~1.5)cm, for example, 1cm*1cm. In addition, preferably, the distance between the measurement area located in the wafer edge region and the wafer edge line is less than 5mm.

[0060] Next, the thickness fluctuation value X2 of the wafer after grinding is calculated.

[0061] In addition, the inventors discovered during their research:

[0062] a. After grinding, there is a certain variation in the thickness between wafers. To ensure that the product meets the requirements after grinding, it is necessary to calculate this variation in thickness to determine the optimal amount of material removed during grinding.

[0063] b. Although grinding can correct the TTV value after cutting, it cannot achieve absolute flatness. In order to ensure that the product meets the requirements after grinding, the amount of material removed by grinding needs to be calculated based on the TTV of the wafer surface after grinding.

[0064] In view of the above findings, in this embodiment of the invention, the method for calculating the thickness fluctuation value X2 of the wafer after polishing includes: calculating the thickness fluctuation value based on at least one of the following calculation results:

[0065] (1) Calculate the thickness variance σ of all the sample wafers after grinding. G ;

[0066] (2) Calculate the mean of the thickness range of all the sample wafers after grinding. and variance σ TTV(G) .

[0067] Specifically, the thickness fluctuation value X2 can be calculated using the following formula:

[0068]

[0069] Here, the value of m is the same as that of m when calculating X1, and preferably, the value of m is 3 or 6. That is, the value of X1 can be managed using 3σ or 6σ.

[0070] When 3σ management is adopted

[0071] When using 6σ management

[0072] Finally, the total amount of wafer thickness removed during actual grinding is calculated based on the sum of the calculated total thickness value X1 and the thickness fluctuation value X2.

[0073] That is, in this embodiment of the invention, the formula for calculating the total amount X of wafer thickness removed during actual grinding is:

[0074]

[0075] or,

[0076]

[0077]

[0078] In summary, the method for calculating the thickness removal amount during wafer grinding provided by this invention includes: collecting multiple wafers produced based on the same target cutting thickness as sample wafers to calculate the total thickness value X1 to be removed from the surface of the wafer after cutting and the thickness fluctuation value X2 of the wafer after grinding; and calculating the total thickness removal amount X of the wafer during actual grinding based on the sum of the calculated total thickness value X1 and the thickness fluctuation value X2. That is, this invention directly calculates the specific thickness removal amount required for the wafer during grinding using a formula, saving verification costs and reducing the consumption of manpower and material resources.

[0079] The above description is merely a description of preferred embodiments of the present invention and is not intended to limit the scope of the present invention in any way. Any changes or modifications made by those skilled in the art based on the above disclosure shall fall within the protection scope of the claims.

Claims

1. A method for calculating the thickness removal amount during wafer grinding after dicing, characterized in that, include: Multiple wafers produced based on the same target cutting thickness were collected as sample wafers to calculate the total thickness X1 that needs to be removed from the wafer surface after cutting and the thickness fluctuation X2 of the wafer after grinding; and, The total amount of wafer thickness removed during actual grinding is calculated based on the sum of the calculated total thickness value X1 and the thickness fluctuation value X2. The method for calculating the total thickness value X1 includes: Calculate the thickness variance σ of all the sample wafers. S ; Calculate the mean of the thickness ranges of all the sample wafers. S and variance σ TTV（S） ; Calculate the average depth of the maximum line mark depth in each measurement area on the sample wafer. and variance σ D ; Calculate the average depth of the maximum damage depth in each measurement region on the sample wafer. and variance σ d ; The total thickness value X1 is calculated using the following formula: X1=m* σ S +2*( S +m* s TTV（S） )+ 2*( +m*s D )+2*( + m*s d ); Where m is a natural number between 3 and 6; Methods for calculating the thickness fluctuation value X2 of the wafer after polishing include: Calculate the thickness variance σ of all the sample wafers after grinding. G ; Calculate the mean of the thickness range values ​​of all the sample wafers after grinding. G and variance σ TTV（G） ; The thickness fluctuation value X2 is calculated using the following formula: X2= m* σ G +2*( G +m* s TTV（G） ); Where m is a natural number between 3 and 6.

2. The method for calculating the thickness removal amount during wafer grinding after dicing as described in claim 1, characterized in that, Multiple sample wafers were produced using different dicing machines and / or different dicing guides; and / or, The sample wafers were selected from the head, middle and tail sections of the crystal rod.

3. The method for calculating the thickness removal amount during wafer grinding after dicing as described in claim 1, characterized in that, The average depth of the maximum line mark depth in each measurement area of ​​the sample wafer was calculated. and the average depth of the maximum damage depth At that time, the selected measurement areas on the sample wafer are distributed throughout the sample wafer.

4. The method for calculating the thickness removal amount during wafer grinding after dicing as described in claim 3, characterized in that, The area of ​​each measurement region is (0.5~1.5) cm * (0.5~1.5) cm.

5. The method for calculating the thickness removal amount during wafer grinding after dicing as described in claim 3, characterized in that, The distance between the measurement area located at the edge of the wafer and the edge line of the wafer is less than 5 mm.

6. The method for calculating the thickness removal amount during wafer grinding after dicing as described in claim 1, characterized in that, The value of m is either 3 or 6.

Images