JP6690204B2 - Wiring board manufacturing method, data correction device, wiring pattern forming system, and data correction method - Google Patents

Description

  The present invention relates to a method for manufacturing a wiring board, a data correction device, a wiring pattern forming system, and a data correction method, and in particular, a method for manufacturing a wiring board used for manufacturing a wiring board having a fine circuit used in electronic devices and the like. The present invention relates to a method, a data correction device, a wiring pattern forming system, and a data correction method.

  Due to the trend toward higher functionality and smaller size of electronic devices, wiring boards used in electronic devices are also required to have high density by thinning wiring patterns.

  As a wiring pattern forming method for coping with such a high density due to the thinning of the wiring pattern, a direct drawing type exposure apparatus for directly irradiating a photosensitive resist with an exposure pattern by laser light or UV-LED light ( DI: Direct Imaging) and an optical inspection device (AOI :: Automatic Optical Inspection) that reads the actual pattern actually formed by etching with reflected light and compares it with the original data (design data). A method is proposed in which actual finish data after etching is taken into an inspection apparatus (AOI) and fed back to an exposure apparatus (DI) (Patent Documents 1 to 3).

JP, 2005-116929, A JP, 2006-303229, A JP, 2007-033764, A

  However, there are many factors that cause the pattern width to change, and even if there are changes in the manufacturing lot, such as due to changes in the pattern shape, density, etc., some products that have the same wiring pattern may be constant to some extent. There is also a factor such as a fluctuation in the same product that varies from manufacturing lot to manufacturing lot due to changes in the manufacturing process. For this reason, the conventional wiring pattern forming methods as disclosed in Patent Documents 1 to 3 may not be able to suppress the variation of the wiring pattern width. Therefore, the conventional method of feeding back the data of the inspection apparatus (AOI) has not been so effective in improving the precision of the fine circuit.

  For this, for example, a method of feeding back the inspection data of the inspection device (AOI) of the actual pattern substrate which is produced in advance for each manufacturing lot can be considered, but the inspection data of the inspection device (AOI) is enormous. Therefore, there is a problem that it takes a long time to process the data to be fed back. In addition, it is possible to reduce the inspection data of the actual pattern substrate that was produced in advance for each manufacturing lot to speed up the processing of the data. However, if the inspection data is reduced, the effect will occur when irregular data is included. However, there is a problem in that the state of the manufacturing process at the time of manufacturing the actual pattern substrate cannot be reflected correctly and the accuracy is rather lowered.

  The present invention has been made in view of the above problems, and in addition to variations in the line width of the wiring pattern depending on the wiring pattern specifications such as pattern gap, pattern size, pattern thickness, and pattern position, a photosensitive resist, Even if the line width of the wiring pattern changes due to changes in the manufacturing process conditions such as the etching liquid and etching liquid, the exposure data can be corrected with higher accuracy so that the line width accuracy during fine circuit formation can be improved. The purpose is to improve.

The present invention relates to the following.
(1) A step (A) of creating exposure data based on original data of a target wiring pattern, a step (B) of creating actual pattern data from an actual pattern substrate formed using this exposure data, A step (C) of creating correction data of the original data or the exposure data based on a difference between the original data and the actual pattern data, wherein the step (C) of creating the correction data comprises the actual pattern From the relationship between the step (C-1) of creating difference data from the difference between the data and the original data or the exposure data and the relationship between the difference data and the factor causing the difference, the factor causing the difference and the difference are suppressed. Step (C-2) of creating a plurality of correction functions that define the relationship between the correction amount of the original data or the correction amount of the exposure data, and a correction function that combines the plurality of correction functions. A method of manufacturing a wiring board, which includes a step (C-3) of creating a number and a step (C-4) of correcting the original data or the exposure data of the wiring pattern using the combined correction function.
(2) In the step (C-2) of creating the plurality of correction functions, a primary correction function created using an actual pattern substrate and a secondary correction function created using another actual pattern substrate are used. In the step (C-3) of creating and synthesizing the combined correction function, a step of creating a third-order correction function that combines the first-order correction function and the second-order correction function and correcting the original data or the exposure data (C In -4), the method for manufacturing a wiring board according to Item 1, wherein the third-order correction function is used to correct the original data or the exposure data of the wiring pattern of the actual pattern board.
(3) In the step (C-2) of creating the plurality of correction functions, in addition to the secondary correction function, another secondary correction function created by using another actual pattern substrate is created, Item 3. In the step (C-3) of creating the combined correction function, the third-order correction function and the further created second-order correction function are combined to create another third-order correction function. Wiring board manufacturing method.
(4) In the step (C-2) of creating the plurality of correction functions, the actual pattern substrate used to create the primary correction function and the other actual pattern substrate used to create another secondary correction function. Item 4. The method of manufacturing a wiring board according to Item 2 or 3, wherein the pattern board or another actual pattern board has the same wiring pattern.
(5) In the step (C-2) of creating the plurality of correction functions, another secondary correction function created after the secondary correction function is used for each manufacturing lot of the actual pattern substrate or each actual pattern substrate. Item 5. The method for producing a wiring board according to any one of Items 2 to 4, which is created.
(6) Factors that cause a difference between the actual pattern data and the original data or exposure data are the pattern gap, the pattern size, the pattern thickness, and the pattern position of the original data or the exposure data of the wiring pattern of the actual pattern substrate. Item 6. The method for producing a wiring board according to any one of Items 1 to 5, which is a combination of any two or more.
(7) Factors that cause a difference between the actual pattern data and the original data or the exposure data include the pattern gap, the pattern size, the pattern thickness, and the pattern position of the original data or the exposure data of the wiring pattern of the actual pattern substrate. 7. The method for manufacturing a wiring board according to item 6, wherein actual pattern data corresponding to any one or a combination of any two or more is used.
(8) A data correction device for original data of a wiring pattern or exposure data, which is used in the method for manufacturing a wiring board according to any one of items 1 to 7, wherein the original data of a target wiring pattern or this original data is used. Difference data is created from the difference between the exposure data created based on the data and the actual pattern data created from the actual pattern substrate formed using the exposure data (C-1). A plurality of correction functions that define the relationship between the factor that causes the difference and the correction amount of the original data or the correction amount of the exposure data for suppressing the difference are created from the relationship between the factor that causes the difference (C -2), a correction function is created by combining the plurality of correction functions (C-3), and the original data of the wiring pattern is corrected by using the correction function created by combining the plurality of correction functions. Alternatively, a data correction device that creates correction data for exposure data (C-4).
(9) When creating a plurality of correction functions (C-2), a primary correction function created using an actual pattern substrate and a secondary correction function created using another actual pattern substrate are used. When the created correction function is created (C-3), a tertiary correction function that combines the primary correction function and the secondary correction function is created, and the original data or the exposure data is corrected (C Item 4) is the data correction apparatus according to Item 8, wherein the third-order correction function is used to correct the original data or the exposure data of the wiring pattern of the actual pattern substrate.
(10) When creating the plurality of correction functions (C-2), in addition to the quadratic correction function, another quadratic correction function created by using another actual pattern substrate is created. When creating the combined correction function (C-3), the third-order correction function and the further created second-order correction function are combined to create another third-order correction function. The data correction device described.
(11) When creating the plurality of correction functions (C-2), the actual pattern substrate used to create the primary correction function and the other used to create another secondary correction function Item 11. The data correction device according to Item 9 or 10, wherein the real pattern substrate or another real pattern substrate has the same wiring pattern.
(12) When creating the plurality of correction functions (C-2), another secondary correction function created after the secondary correction function is used for each manufacturing lot of the actual pattern substrate or each actual pattern substrate. The data correction neglected according to any one of Items 9 to 11, which is created in 1.
(13) Factors that cause a difference between the actual pattern data and the original data or the exposure data are the pattern gap, the pattern size, the pattern thickness, and the pattern position of the original data or the exposure data of the wiring pattern of the actual pattern substrate. 13. The method for manufacturing a wiring board according to any one of items 8 to 12, which is any one or a combination of any two or more.
(14) Factors that cause a difference between the actual pattern data and the original data or exposure data include the pattern gap, pattern size, pattern thickness, and pattern position of the original data or exposure data of the wiring pattern of the actual pattern substrate. Item 14. The method for manufacturing a wiring board according to Item 13, which uses actual pattern data corresponding to any one or a combination of any two or more.
(15) Based on the data correction device according to any one of items 8 to 14 and the exposure data created from the original data corrected by the data correction device or the exposure data corrected by the data correction device A pattern exposure device that exposes an exposure pattern on a photosensitive resist disposed on a substrate; a development pattern forming device that develops the photosensitive resist on which the exposure pattern is exposed to form a development pattern; and the development pattern A wiring pattern forming system, comprising: an actual pattern forming device that forms an actual pattern by performing circuit processing on a substrate on which the pattern is formed; and an actual pattern data creating device that creates actual pattern data from the actual pattern.
(16) A data correction method of original data of a wiring pattern or exposure data using the data correction device according to any one of items 8 to 14, wherein the original data of a target wiring pattern or this original data is A step (C-1) of creating difference data from a difference between the exposure data created based on the exposure data and the actual pattern data created from the actual pattern substrate formed using the exposure data; and the difference data and the difference. Creating a plurality of correction functions that define a relationship between a factor that causes the difference and a correction amount of the original data or a correction amount of the exposure data for suppressing the difference, based on the relationship with the factor that causes the difference ( C-2), a step (C-3) of creating a correction function by synthesizing the plurality of correction functions, and a step of performing correction using the correction function created by synthesizing the plurality of correction functions. A step (C-4) of creating original data of the wiring pattern or correction data of the exposure data.

  According to the present invention, in addition to the fluctuation of the line width of the wiring pattern due to the wiring pattern specifications such as the pattern gap, the pattern size, the pattern thickness, and the pattern position, the state of the manufacturing process of the photosensitive resist, the developing solution, the etching solution, etc. Even if the line width of the wiring pattern is changed due to the change, the exposure data can be corrected with higher accuracy so that the line width accuracy at the time of forming a fine circuit can be improved.

1 shows a method for manufacturing a wiring board according to an embodiment of the present invention. 1 shows a wiring pattern (test pattern) used in an embodiment of the present invention. The schematic diagram of the real pattern substrate A used by one Embodiment of this invention is shown. The schematic diagram of the real pattern substrate B used by one Embodiment of this invention is shown. The schematic diagram of the real pattern substrate C used by one Embodiment of this invention is shown. 3 shows a primary correction function created in an embodiment of the present invention. 7 shows a quadratic correction function created in an embodiment of the present invention. 3 shows a third-order correction function created in an embodiment of the present invention. 7 shows another quadratic correction function and another cubic correction function created in an embodiment of the present invention. The block diagram of the data correction | amendment apparatus of one Embodiment of this invention is shown. 1 is a schematic diagram of a wiring pattern forming system according to an embodiment of the present invention. The primary correction function of Examples 1 and 2 is shown. The quadratic correction function of Examples 1 and 2 is shown. 3 shows a third-order correction function of Examples 1 and 2. 7 shows another quadratic correction function of Examples 1 and 2. The other 3rd-order correction function of Example 1 is shown. 9 shows another third-order correction function of the second embodiment.

(Wiring Board Manufacturing Method: First Embodiment)
<< Process (A) >>
A method of manufacturing the wiring board according to the first embodiment of the present invention will be described. As shown in FIG. 1, the method for manufacturing a wiring board according to the present embodiment first has a step (A) of creating exposure data based on original data of a target wiring pattern. The target wiring pattern is a wiring pattern to be formed as a real pattern after circuit processing, and includes a product pattern for functioning as a wiring board and a test pattern for creating a correction function described later. Further, the actual pattern means a wiring pattern of the actual pattern substrate which is actually formed by performing circuit processing. The target wiring pattern is not particularly limited, and any wiring pattern can be used.

  The original data of the wiring pattern is the design data of the target wiring pattern, and is the one that digitizes the target wiring pattern to be formed, and is represented by the numerical values of the coordinate and the pattern width, the coordinate and the pattern gap, for example. is there. It may have data to which information necessary for exposure is added. The original data is created using a device (CAD: Computer Aided Design) or the like that creates design data.

  The exposure data refers to data for forming an exposure pattern corresponding to a wiring pattern by exposing a photosensitive resist to light with a pattern exposure unit such as a linear drawing device using laser light or UV light. The exposure data is created using a device (CAM: Computer Aided Manufacturing) that creates the exposure data based on the original data.

<< Process (B) >>
Next, as shown in FIG. 1, the method for manufacturing a wiring board according to the present embodiment has a step (B) of creating actual pattern data from an actual pattern board formed using this exposure data, The step (B) of creating actual pattern data includes a pattern exposure step (B-1), a development pattern forming step (B-2), an actual pattern forming step (B-3), and a pattern inspection step (B-4). Have

<Process (B-1)>
In the pattern exposure step (B-1), the exposure data is used to expose an exposure pattern corresponding to the wiring pattern to a photosensitive resist by a pattern exposure device such as a linear drawing device using laser light or UV light. To form. Here, the pattern exposure apparatus refers to an exposure apparatus that exposes a photosensitive resist arranged on a substrate with an exposure pattern based on exposure data. Examples of the pattern exposure device include a direct writing device (DI: Direct Imaging) that directly exposes a photosensitive resist with an exposure pattern using laser light or UV-LED light. The photosensitive resist refers to an etching resist or a plating resist used when forming a wiring pattern by etching a metal foil such as a copper foil or plating a metal such as copper by a photolithography method. . The exposure pattern refers to a pattern exposed on the photosensitive resist based on the exposure data, and corresponds to a development pattern formed by subsequent development.

<Process (B-2)>
In the development pattern forming step (B-2), the photosensitive resist is removed, leaving an exposure pattern formed by pattern exposure and necessary for forming an actual pattern. Here, the development pattern means a pattern that appears when the photosensitive resist after exposure is developed. It can be formed by a development pattern forming device, and the development pattern forming device includes a developing device that develops a photosensitive resist having an exposed exposure pattern to form a development pattern.

<Process (B-3)>
In the actual pattern forming step (B-3), circuit processing is performed to produce an actual pattern substrate having an actual pattern. Here, the circuit processing means forming an actual pattern, and examples thereof include forming a conductor pattern by etching a metal foil by a subtract method. The actual pattern refers to a conductor pattern that is actually formed by performing circuit processing, and can be formed by an actual pattern forming device. The actual pattern forming apparatus is an apparatus for forming an actual pattern by performing circuit processing on a substrate on which a developed pattern is formed, and an etching apparatus can be used.

<Process (B-4)>
In the pattern inspection step (B-4), the real pattern data is acquired from the real pattern of the real pattern substrate. Here, the actual pattern substrate refers to a substrate having a conductor pattern (actual pattern) actually formed by circuit processing, for example, a substrate having a conductor pattern obtained by etching a metal foil by the subtract method. Is mentioned. Further, the actual pattern data refers to finish data obtained from the actual pattern by using an optical visual inspection apparatus (AOI: Automatic Optical Inspection), a measuring microscope, or the like. The optical appearance inspection device is generally a device that detects light reflected from the upper surface (top) of an actual pattern and digitizes the pattern to obtain data represented by numerical values such as coordinates, pattern width, and pattern gap. . On the other hand, in the present embodiment, the measuring microscope can be used to measure the line widths of both the top surface (top) of the actual pattern and the bottom surface (bottom) of the actual pattern and convert them into data. .

<< Step (C) >>
Next, the wiring pattern forming method of the present embodiment has a step (C) of creating correction data of the original data or exposure data based on the difference between the original data or exposure data and the actual pattern data. ing. Here, the correction data of the original data or the exposure data means the original data or the exposure data after the correction, that is, the corrected original data or the exposure data. Further, as shown in FIG. 1, in the step (C) of creating the correction data, the step (C-1) of creating difference data from the difference between the actual pattern data and the original data or the exposure data, and the difference A step of creating a plurality of correction functions that define the relationship between the factor that causes the difference and the correction amount of the original data or the correction amount of the exposure data for suppressing the difference from the relationship between the data and the factor that causes the difference ( C-2), a step of creating a correction function that combines the plurality of correction functions (C-3), and a step of correcting the original data or the exposure data of the wiring pattern using the combined correction function (C-4). )have.

  In circuit processing by etching, since the etching solution is applied differently to the front and back surfaces of the actual pattern substrate, the tendency of the etching process changes. Therefore, it is desirable to create a correction function for each of the front and back surfaces of the actual pattern substrate. As a result, it is possible to suppress variations in the finish value such as the line width of the actual pattern due to the difference in the front and back surfaces of the actual pattern substrate.

  Further, in the circuit processing by etching, even in one of the front and back surfaces of the actual pattern substrate, the direction of the wiring pattern changes the relationship with the transport direction of the actual pattern substrate, the direction in which the etching liquid strikes, etc. The etching process tends to have unevenness in directionality and etching amount. Therefore, the tendency of the etching process also changes depending on the direction of the wiring pattern arranged on the actual pattern substrate. Therefore, it is desirable to create a correction function for each of the vertical and horizontal wiring patterns arranged on the actual pattern substrate. As a result, it is possible to suppress variations in the finish value such as the line width of the actual pattern depending on the direction of the wiring pattern.

  Also, in circuit processing by etching, the relationship between the transfer direction of the actual pattern substrate and the direction in which the etching liquid hits changes depending on the position where the wiring pattern is placed, even on one of the front and back surfaces of the actual pattern substrate. Therefore, the etching process tends to have unevenness in directionality and etching amount. Therefore, it is desirable to divide the in-plane of the actual pattern substrate into a plurality of areas and create a correction function for each of these areas. In addition, especially in the peripheral area of the edge of the actual pattern board, in addition to the unevenness of the etching process, the unevenness of the copper foil thickness due to the variation of the plating thickness is also added. It is desirable that the section further divides the area into smaller pieces and creates a correction function for each area. As a result, it becomes possible to suppress the variation of the finished value such as the line width of the actual pattern due to the in-plane position of the actual pattern substrate.

<Step (C-1)>
In the step (C-1) of creating difference data from the difference between the actual pattern data and the original data or the exposure data, the actual pattern data obtained in the pattern inspection step (B-4) is compared with the original data or the exposure data. Then, the difference data is created. Here, the difference between the actual pattern data and the original data or exposure data of the actual pattern is, specifically, the pattern gap between the actual pattern data and the original data or exposure data of the actual pattern at the same coordinates, the pattern width, etc. Is the difference. The same coordinates are the same coordinates in the original data (design data), and the pattern gaps and the like have the same design values in the same coordinates. Therefore, it is indicated that the difference in the pattern gap or the like at the same coordinate is the difference obtained by comparing the portions having the same design value in the wiring pattern. The difference data refers to data in which this difference is represented by coordinates, pattern gaps, pattern widths, and the like. The difference data can be created using a computer from the difference between the actual pattern data and the original data or exposure data of the actual pattern.

<Step (C-2)>
In the step (C-2) of creating a plurality of correction functions, from the relationship between the difference data and the factor causing the difference, the factor causing the difference and the correction amount of the original data for suppressing the difference or the exposure data Create a plurality of correction functions that define the relationship with the correction amount. Here, the factor that causes the difference between the actual pattern data and the original data or the exposure data is the variation in the wiring pattern specifications of the original data or the exposure data, which causes the difference between the actual pattern data and the original data or the exposure data. A factor that causes a change in the difference from the data. Examples of such factors include any one or a combination of two or more of the pattern gap of the original data of the actual pattern or the exposure data, the pattern width, the pattern size, the pattern thickness, and the pattern position. Further, as the correction amount of the original data or the correction amount of the exposure data for suppressing the difference, for example, the difference itself between the actual pattern data and the original data or the exposure data of the actual pattern can be used. This is because if the difference between the actual pattern data and the original data or exposure data of the actual pattern is corrected by adding to or subtracting from the current original data or exposure data, the actual pattern data approaches the numerical value of the original data or exposure data. This is due to the fact.

  The correction function defines the relationship between the factor causing the difference and the difference data and the correction amount of the original data or the exposure data for suppressing the difference from the relation between the factor causing the difference and the difference data. Further, the correction function is a computer having a calculation function for obtaining the relationship between the factor causing the difference and the correction amount of the original data or the exposure data for suppressing the difference from the relation between the factor causing the difference and the difference data. Can be created by. As an example of the correction function, FIG. 6 shows an example in which the correction function is created using the pattern gap of the wiring pattern of the actual pattern substrate as a factor that causes the difference between the actual pattern data and the original data or the exposure data. That is, in the correction function of FIG. 6, the horizontal axis is the pattern gap of the wiring pattern of the actual pattern substrate, and the vertical axis is the correction amount.

  As a factor that causes a difference between the actual pattern data and the original data or the exposure data in the correction function, any one of the pattern gap, the pattern size, the pattern thickness, and the pattern position of the original data or the exposure data of the wiring pattern of the actual pattern substrate Alternatively, it is desirable to use actual pattern data corresponding to any two or more combinations.

  This will be described below using the example of the correction function in FIG. In the correction function of FIG. 6, the horizontal axis is the pattern gap of the wiring pattern of the actual pattern substrate, but as an example of data used as the numerical value of this pattern gap, original data of the wiring pattern of the actual pattern substrate, exposure data, Actual pattern data is possible. When the original data or the exposure data is used among these, the correction function indicates the relationship between the original data or the exposure data of the pattern gap and the correction amount for the original data or the exposure data, and the correction amount is the actual pattern data and the original data. Alternatively, since it is based on the difference data from the exposure data, even in this case, it is possible to obtain the correction amount to some extent with respect to the original data of the pattern gap or the exposure data.

  However, in this case, the correction amount for the original data or the exposure data is based on the difference data when the actual pattern data (finished value) is different from the original data (design value), and is actually obtained. Since the actual pattern data (finished value) is finished according to the original data (designed value), that is, the correction amount for matching the original data (designed value) and the actual pattern data (finished value), It is conceivable that there is a deviation from the correction amount based on the difference data when the actual pattern data (finished value) is different from the original data (design value).

  On the other hand, when the actual pattern data (measured value) is used as the numerical value of the pattern gap of the wiring pattern of the actual pattern substrate used on the horizontal axis of the correction function of FIG. 6, the correction function uses the actual pattern data (measured value) of the pattern gap. And the correction amount for the original data or the exposure data will be shown. The original data (design value) on the horizontal axis or the actual pattern data (actual measurement value) corresponding to the exposure data can be regarded as the target original data (design value) or the exposure data. The correction amount of (value) is a correction amount for obtaining target original data (design value) or exposure data. Thereby, the correction amount of the original data (design value) or the exposure data necessary when the actual pattern data (actual measurement value) is finished as the target original data (design value) or the exposure data is set more accurately. It will be possible.

  In addition, the correction function is a factor that causes a change in the difference between the actual pattern data and the original data or the exposure data, that is, a wiring pattern in which the pattern gap, the pattern width, the pattern size, the pattern thickness, the pattern position, etc. are changed. It can be created by using a plurality of actual pattern substrates. As the wiring pattern used to create the correction function, a wiring pattern used as a product of a wiring board can be used, but a factor that causes a difference as shown in FIG. 2 (pattern gap, pattern in FIG. 2). It is desirable to use a test pattern in which the width, the pattern shape, and the pattern size) are changed, because necessary data can be easily obtained and a more accurate correction function can be created.

  Further, as shown in FIG. 3, an actual pattern substrate (hereinafter referred to as a test pattern substrate) is used in which a large number of one unit of a relatively small area test pattern having the same wiring pattern is arranged in the actual pattern substrate. It is desirable that the data regarding the same wiring pattern specifications can be obtained over the entire real pattern substrate, and the data regarding the variation within the real pattern substrate can be obtained.

  In the present embodiment, in the step (C-2) of creating a plurality of correction functions as the plurality of correction functions, the primary correction function created using the actual pattern substrate and another actual pattern substrate are used. Create the created secondary correction function and. Here, the other actual pattern board means an actual pattern board different from the actual pattern board when the primary correction function is created, and the wiring patterns themselves may be the same or different. Absent. In this way, by separately creating the primary correction function and the secondary correction function, for example, using the actual pattern substrate on which the test pattern is arranged, the reference primary correction function is created and actually created. At the timing of manufacturing the actual pattern board on which the product pattern is placed, another primary pattern board is used again to create a secondary correction function that reflects the state of the production line at that time, and then these primary corrections are performed. The function and the quadratic correction function can be combined. That is, it becomes possible to correct the correction function to a more appropriate correction function according to the situation of the production line by using the secondary correction function based on the reference primary correction function.

  It is desirable that the actual pattern substrate used to create the primary correction function and the other actual pattern substrate used to create another secondary correction function have the same wiring pattern. The plurality of correction functions may be created using a plurality of real pattern boards having the same wiring pattern or a plurality of real pattern boards having different wiring patterns. It is desirable because it can improve the accuracy of the function. For example, as shown in FIGS. 3, 4, and 5, the real pattern substrate A (FIG. 3), the real pattern substrate B (FIG. 4), and the real pattern substrate C (FIG. 5) have the same test pattern. Thus, even if the wiring pattern of another portion changes, it is possible to always create a correction function using the same test pattern portion.

  In the step (C-2) of creating a plurality of correction functions, it is preferable that the secondary correction function is created for each manufacturing lot of real pattern substrates or for each real pattern substrate. This makes it possible to create a correction function that reflects the state of the manufacturing process when manufacturing the actual pattern substrate.

<Step (C-3)>
In the step (C-3) of creating a correction function that combines a plurality of correction functions, the plurality of correction functions created in the step (C-2) of creating a plurality of correction functions are combined to create one correction function. To do.

  In the present embodiment, a third-order correction function is created by combining the first-order correction function and the second-order correction function. For example, as shown in FIG. 6, first, a reference primary correction function is created using an actual pattern substrate, and then, as shown in FIG. 7, an actual pattern substrate different from this primary correction function is created. A second-order correction function is created using this, and as shown in FIG. 8, a first-order correction function and a second-order correction function are combined to create a third-order correction function. In other words, since the reference first-order correction function is always fixed, a stable third-order correction function can be obtained, so that the second-order correction function produced for each manufacturing lot of actual pattern boards or each actual pattern board is measured. Even if it includes an error or irregular measurement value, the influence can be reduced.

  As a method of synthesizing a plurality of correction functions, if the manufacturing process is relatively stable, using a weighted average with a large synthesizing ratio of the correction functions created using more data is Desirable for improving accuracy. For example, a primary correction function that uses a lot of difference data is created as a reference correction function, and then, as a secondary correction function that reflects the state of the manufacturing process, a comparison is made when an actual pattern substrate is manufactured. If you create a secondary correction function that uses less differential data and combine these primary and secondary correction functions, increase the combination ratio of the primary correction function that serves as the reference to increase the measurement error. It is possible to suppress the influence of irregular data. On the other hand, when the state of the manufacturing process is relatively unstable, in the above example, it is possible to obtain an appropriate correction function that reflects the state of the manufacturing process more by increasing the synthesis ratio of the secondary correction function. it can. In this way, the synthesis ratio in the weighted average may be set arbitrarily according to the number of difference data when creating the correction function, the state of the manufacturing process, etc., for example, in a relatively stable manufacturing process. On the other hand, it is desirable to set in the range of 0.1 to 0.4 as long as it is a composite ratio of the primary correction function serving as a reference and the secondary correction function created when the actual pattern substrate is manufactured. On the other hand, for a relatively unstable manufacturing process, the composite ratio of the reference primary correction function and the secondary correction function created at the time of manufacturing the actual pattern substrate is set within the range of 0.3 to 0.7. It is desirable to do. The term "composite ratio" here means the ratio of the secondary correction function when the total of the primary correction function and the secondary correction function is 1.

<Step (C-4)>
In the step (C-4) of correcting the original data or the exposure data of the wiring pattern using the combined correction function, the difference between the actual pattern data and the original data or the exposure data is changed by using the combined correction function. The original data of the wiring pattern or the exposure data is calculated by calculating the correction amount corresponding to the pattern gap, the pattern width, etc., which are the factors to be added, and adding or subtracting the obtained correction amount to the original data or the exposure data. to correct.

  In the present embodiment, in the step (C-4) of correcting the original data or the exposure data, the third-order correction function is used to correct the original data or the exposure data of the wiring pattern of the actual pattern substrate.

(Action / effect)
According to the method for manufacturing a wiring board of the present embodiment, a basic primary correction function is created by synthesizing a plurality of correction functions in this way, and in order to cope with variations in the manufacturing process, Even when the secondary correction function is created with less data than the one correction function at the timing of manufacturing the patterned substrate, the effects of measurement errors and irregular measured values can be reduced, and the actual manufacturing process An appropriate correction function that reflects the situation can be obtained. Therefore, it is suitable when the state of the production process is unlikely to change even if the number of productions increases, and is relatively stable. Therefore, in addition to fluctuations in the line width of the wiring pattern due to the wiring pattern specifications such as pattern gap, pattern size, pattern thickness, pattern position, etc., the wiring pattern due to changes in the manufacturing process status of photosensitive resist, developer, etching solution, etc. The line width accuracy at the time of forming a fine circuit can be improved by correcting the exposure data with higher accuracy to cope with the fluctuation of the line width.

(Wiring Board Manufacturing Method: Second Embodiment)
A method of manufacturing the wiring board according to the second embodiment of the present invention will be described. It should be noted that points of the present embodiment that are different from the first embodiment will be mainly described.

<Steps (A), (B), (C-1)>
In this embodiment, similarly to the first embodiment, first, steps (A), (B), and (C-1) are performed, and difference data is created from the difference between the actual pattern data and the original data. To do.

<Step (C-2)>
In the step (C-2) of creating a plurality of correction functions, a primary correction function created using an actual pattern substrate and a secondary correction function created using another actual pattern substrate are created. Then, in addition to this quadratic correction function, another quadratic correction function created using another real pattern substrate is created.

<Step (C-3)>
Next, in a step (C-3) of creating a correction function that combines a plurality of correction functions, a third-order correction function that combines the first-order correction function and the second-order correction function is created. After that, this third-order correction function is combined with another second-order correction function further created in the step (C-2) to create another third-order correction function.

  Specifically, for example, as shown in FIG. 6, first, a primary correction function is created using an actual pattern substrate, and then, as shown in FIG. 7, another actual pattern substrate different from this primary correction function is created. A second-order correction function is created using, and a third-order correction function is created by synthesizing the first-order correction function and the second-order correction function as shown in FIG. Further, as shown in FIG. 9, a second-order correction function is further created and combined with the third-order correction function to create another third-order correction function. In other words, since the third-order correction function is a cumulative value obtained by repeatedly combining the second-order correction function and the second-order correction function, the second-order correction function produced for each manufacturing lot of the actual pattern substrate or each actual pattern substrate causes a measurement error or irregularity. Even if the measured value is included, the influence can be reduced.

(Action / effect)
According to the method for manufacturing a wiring board of the present embodiment, a secondary correction function that is newly created each time an actual pattern board is manufactured is used for a third-order correction function that is a combination of a first-order correction function and a second-order correction function. Another third-order correction function is created by combining the functions. In other words, since the third-order correction function used to correct the original data or the exposure data is an accumulation of the second-order correction function repeatedly, the manufacturing process situation tends to change toward a specific direction due to the accumulation of the number of productions. It is suitable when there is. Therefore, it is possible to obtain the same actions and effects as those of the first embodiment, and in such a case, more stable correction can be performed.

(Data correction device: third embodiment)
A data correction apparatus according to the third embodiment of the present invention will be described. As shown in FIG. 10, the data correction apparatus of this embodiment uses a computer, and the computer has a processing unit (processor), a display unit, an input unit, a storage unit, a communication unit, and each of these components. It has a bus to connect the parts. The display unit displays an image output by the program executed by the computer. The input unit receives an input, and is, for example, a keyboard or a mouse. The storage unit can store information such as a non-volatile memory, a volatile memory, and a hard disk. The storage unit stores data such as original data, exposure data, actual pattern data, difference data, correction function, and correction data, and a process until the original data or the exposure data is corrected (correction data creation process (C) in FIG. 1). ) Is stored. The communication unit performs wireless communication or wired communication using a USB cable or the like. The original data, the exposure data, the actual pattern data, the difference data and the like may be acquired via the communication unit. When the correction program is executed, the processing unit (processor) executes the process for correcting the original data or the exposure data (correction data creating process (C) in FIG. 1). The data correction device does not have to be a general-purpose computer, and may be realized by hardware for executing all or part of each process and software that operates in cooperation with the hardware.

  The data correction apparatus according to the present embodiment, as shown in (C-1) of the (C) data creation process of FIG. 1C, is original data of a target wiring pattern or exposure data created based on this original data. And difference data with the actual pattern data created from the actual pattern substrate formed using the exposure data.

  Next, as shown in (C-2) of the (C) data creation process of FIG. 1, from the relationship between the difference data and the factor that causes the difference, the factor that causes the difference and the original data for suppressing the difference. A plurality of correction functions that define the relationship between the correction amount and the exposure data correction amount are created. As an example of the correction function, FIG. 6 shows an example in which the correction function is created using the pattern gap of the wiring pattern as a factor that causes a difference between the actual pattern data and the original data or the exposure data. That is, in the correction function of FIG. 6, the horizontal axis is the pattern gap of the wiring pattern, and the vertical axis is the correction amount.

  In the present embodiment, when creating a plurality of correction functions (C-2), a primary correction function created using an actual pattern board and another actual pattern board are used as the plurality of correction functions. And a secondary correction function created by In this way, by separately creating the primary correction function and the secondary correction function, for example, using the actual pattern substrate on which the test pattern is arranged, the reference primary correction function is created and actually created. At the timing of manufacturing the actual pattern board on which the product pattern is placed, another primary pattern board is used again to create a quadratic correction function that reflects the situation of the production line at that time. The function and the quadratic correction function can be combined. That is, it becomes possible to correct the correction function to a more appropriate correction function according to the situation of the production line by using the secondary correction function based on the reference primary correction function.

  Next, as shown in (C-3) of the (C) data creation process of FIG. 1, a correction function that combines the plurality of correction functions is created.

  In the present embodiment, a third-order correction function is created by combining the first-order correction function and the second-order correction function. For example, as shown in FIG. 6, first, a reference primary correction function is created using an actual pattern substrate, and then, as shown in FIG. 7, an actual pattern substrate different from this primary correction function is created. A second-order correction function is created using this, and as shown in FIG. 8, a first-order correction function and a second-order correction function are combined to create a third-order correction function. As a result, since the reference first-order correction function is always fixed, a stable third-order correction function can be obtained, so that the second-order correction function produced for each manufacturing lot of real pattern substrates or each real pattern substrate is Even if the measurement error or irregular measurement value is included, the influence can be reduced.

  Next, as shown in (C-4) of the (C) data creation process of FIG. 1, correction of the original data of the wiring pattern or the exposure data corrected using the correction function created by combining a plurality of correction functions. Create the data.

  In the present embodiment, when correcting the original data or the exposure data (C-4), the original data or the exposure data of the wiring pattern of the actual pattern substrate is corrected by using the cubic correction function.

  When creating a plurality of correction functions (C-2), the actual pattern board used to create the primary correction function and another actual pattern board used to create another secondary correction function, or It is desirable that the other real pattern substrate has the same wiring pattern. The plurality of correction functions may be created using a plurality of real pattern boards having the same wiring pattern or a plurality of real pattern boards having different wiring patterns. It is desirable because it can improve the accuracy of the function. For example, as shown in FIGS. 3, 4, and 5, the real pattern substrate A (FIG. 3), the real pattern substrate B (FIG. 4), and the real pattern substrate C (FIG. 5) have the same test pattern. Thus, even if the wiring pattern of another portion changes, it is possible to always create a correction function using the same test pattern portion.

  When creating a plurality of correction functions (C-2), another secondary correction function created after the secondary correction function is created for each manufacturing lot of the actual pattern substrate or for each actual pattern substrate. Is desirable. This makes it possible to create a correction function that reflects the state of the manufacturing process when manufacturing the actual pattern substrate.

(Action / effect)
According to the data correction apparatus of the present embodiment, similarly to the first embodiment, in addition to the fluctuation of the line width of the wiring pattern depending on the wiring pattern specifications such as the pattern gap, the pattern size, the pattern thickness, the pattern position, etc. Even when the line width of the wiring pattern changes due to changes in the manufacturing process such as resist, developing solution, etching solution, etc., the exposure data can be corrected with higher accuracy so that the line can be used during fine circuit formation. The width accuracy can be improved.

(Data Correction Device: Fourth Embodiment)
A data correction device according to a fourth embodiment of the present invention will be described. It should be noted that points of the present embodiment that are different from the first embodiment will be mainly described.

  The data correction apparatus according to the present embodiment has the configuration shown in FIG. 10 similarly to the data correction apparatus according to the third embodiment described above.

  When creating a plurality of correction functions (C-2), in addition to the quadratic correction function, another quadratic correction function created using another actual pattern substrate is created and synthesized. When creating a correction function (C-3), another tertiary correction function is created by synthesizing the third-order correction function and the further created second-order correction function.

  Specifically, for example, as shown in FIG. 6, first, a primary correction function is created using an actual pattern substrate, and then, as shown in FIG. 7, another actual pattern substrate different from this primary correction function is created. A second-order correction function is created using, and a third-order correction function is created by synthesizing the first-order correction function and the second-order correction function as shown in FIG. Further, as shown in FIG. 9, a second-order correction function is further created and combined with the third-order correction function to create another third-order correction function. In other words, since the third-order correction function is a cumulative value obtained by repeatedly combining the second-order correction function and the second-order correction function, the second-order correction function produced for each manufacturing lot of the actual pattern substrate or each actual pattern substrate has a measurement error or irregular Even if the measured value is included, the influence can be reduced.

(Action / effect)
According to the data correction apparatus of the present embodiment, it is possible to obtain the same actions and effects as those of the third embodiment, and the third-order correction function used for correcting the original data or the exposure data is obtained by repeatedly accumulating the second-order correction function. Therefore, when the situation of the manufacturing process tends to change in a specific direction due to the accumulation of the number of productions or the like, more stable correction can be performed.

(Wiring pattern forming system: fifth embodiment)
A wiring pattern forming system according to a fifth embodiment of the present invention will be described. As shown in FIG. 11, the wiring pattern forming system of the present embodiment first has a data correction device. As in the third or fourth embodiment, a computer is used as the data correction device, and the computer includes a processing unit (processor), a display unit, an input unit, a storage unit, a communication unit, and each of these components. It has a bus to connect to.

  Next, based on the exposure data created from the original data corrected by the data correction device or the exposure data corrected by the data correction device, the pattern exposure is performed to expose the exposure pattern on the photosensitive resist arranged on the substrate. Have a device. In the present embodiment, the substrate refers to one having a copper foil or a metal foil such as copper plating on an insulating layer such as glass epoxy, and examples thereof include a copper clad laminate. The pattern exposure apparatus refers to an exposure apparatus that exposes a photosensitive resist arranged on a substrate with an exposure pattern based on exposure data. A direct drawing device (DI: Direct Imaging) that directly exposes an exposure pattern on a photosensitive resist using laser light or UV-LED light can be used.

  Next, it has a development pattern formation device which develops the photosensitive resist which the exposure pattern was exposed to, and forms a development pattern. Examples of the development pattern forming device include a development device used in photolithography.

  Next, there is an actual pattern forming device for forming an actual pattern by performing circuit processing on the substrate on which the developed pattern is formed. Examples of the actual pattern forming device include an etching device and a plating device used in circuit processing of a wiring board.

Next, it has a real pattern data creation device for creating real pattern data from a real pattern.
Examples of the actual pattern data creation device include an optical appearance inspection device (AOI: Automatic Optical Inspection), a measuring microscope, and the like. The optical appearance inspection device detects light reflected from the upper surface (top) of an actual pattern and digitizes the pattern to obtain data represented by numerical values such as coordinates and pattern width and pattern gap. On the other hand, the measuring microscope can be used to measure the line width and convert it into data on either the upper surface (top) or the lower surface (bottom) of the actual pattern.

  In the wiring pattern forming system according to the present embodiment, the data correction device receives the actual pattern data from the actual pattern data creating device and, as shown in FIG. 1, calculates the difference between the actual pattern data and the original data or the exposure data. Difference data is created (C-1), and from the relationship between the difference and the factor that causes this difference, the relationship between the factor that causes the difference and the correction amount of the original data or the correction amount of the exposure data for suppressing the difference. Is created (C-2), a correction function that combines the plurality of correction functions is created (C-3), and the original data or exposure data of the wiring pattern is created using the combined correction function. Is corrected (C-4).

(Action / effect)
According to the wiring pattern forming system of the present embodiment, similarly to the first embodiment, in addition to the fluctuation of the line width of the wiring pattern due to the wiring pattern specifications such as the pattern gap, the pattern size, the pattern thickness, the pattern position, etc. Even if the line width of the wiring pattern changes due to changes in the manufacturing process such as a conductive resist, a developing solution, an etching solution, etc. The line width accuracy can be improved.

(Data correction method: sixth embodiment)
A data correction method according to the sixth embodiment of the present invention will be described. The data correction method of the present embodiment uses the data correction neglected state shown in FIG. 10, and as shown in (C) correction data creation process of FIG. 1, the original data of the target wiring pattern or this original data. A step (C-1) of creating difference data from a difference between the exposure data created based on the exposure data and the actual pattern data created from the actual pattern substrate formed using the exposure data, and the difference data and the difference. Creating a plurality of correction functions that define the relationship between the factor that causes the difference and the correction amount of the original data or the correction amount of the exposure data for suppressing the difference, from the relationship with the factor that causes the difference. (C-2) and a step (C-3) of creating a correction function by synthesizing the plurality of correction functions, and the arrangement corrected by using the correction function created by synthesizing the plurality of correction functions. It has a step (C-4) for creating correction data of the original data or exposure data pattern.

(Action / effect)
According to the wiring pattern forming system of the present embodiment, similarly to the first embodiment, in addition to the fluctuation of the line width of the wiring pattern due to the wiring pattern specifications such as the pattern gap, the pattern size, the pattern thickness, the pattern position, etc. Even if the line width of the wiring pattern changes due to changes in the manufacturing process such as a conductive resist, a developing solution, an etching solution, etc. The line width accuracy can be improved.

  The present invention will be described below based on examples, but the present invention is not limited to the examples.

(Example 1)
<Creation of primary correction function>
First, as a substrate for producing the actual pattern substrate C4c on which the test pattern 1 and the product pattern 7 are arranged in order to create the first-order correction function, a thickness of 0.22 mm having a copper foil of 5 μm on the front and back of the insulating layer The copper foil of the copper-clad laminate of MCL-E-700G (trade name, manufactured by Hitachi Chemical Co., Ltd., "MCL" is a registered trademark) of 440 mm in length and 510 mm in width is plated with about 9 μm by electrolytic copper plating, A copper plate having a total copper thickness of about 14 μm on each of the front and back surfaces was prepared. Next, a photosensitive etching resist is attached to the copper foils on the front and back surfaces, the product pattern 7 and the test pattern 1 in FIG. 2 are exposed, and a circuit is formed by etching using a horizontal transfer type etching device. Nine real pattern substrates C4c as shown in FIG. The original data (design value) of the test pattern shown in FIG. 2 is a linear wiring in which the pattern width (width of the pattern 3) is fixed at 100 μm and the pattern gap 2 is changed in 24 steps in the range of 14 to 150 μm. The patterns are arranged in the vertical direction and the horizontal direction on each of the front and back surfaces, and the initial exposure data created based on this original data is also the same as the original data (design value). FIG. 5 shows an outline of the actual pattern substrate C4c. In this embodiment, in addition to the product pattern 7, actually, there are 11 test patterns 1 in the vertical direction, 11 in the horizontal direction, and 121 in total. Were evenly arranged on the respective front and back surfaces of the actual pattern substrate C4c.

  The actual pattern data (finished value) was acquired for the actual pattern substrate C4c using an optical automatic visual inspection apparatus. The number of measurement points at this time is one for each of the vertical and horizontal linear wiring patterns for each test pattern. Therefore, in each of the front and back surfaces of the actual pattern substrate C4c, Each of the vertical and horizontal linear wiring patterns has 121 points, and a total of 9 sheets has 1089 points for each of the vertical and horizontal linear wiring patterns. Next, at the same coordinates of the test pattern 1, that is, where the pattern gaps have the same design value, the actual pattern data (finishing value) and the original data (design value) are obtained for each pattern gap at each design value. Difference data was created. Here, as for the actual pattern data (finish value), for each of the 24-step pattern gaps, 1089 points (121 points per sheet) of actual pattern data (finish value) are acquired for each surface, and for each point at the same position. Is an average value (every 121 points).

  In circuit processing by etching, even in one of the front and back surfaces of the actual pattern substrate, the relationship with the transfer direction of the actual pattern substrate, the direction in which the etching liquid strikes, etc. changes depending on the position where the wiring pattern is arranged. However, the etching process tends to have unevenness in directionality and etching amount. In addition, in particular, in the peripheral portion of the end portion of the actual pattern substrate, in addition to the unevenness of the etching process, the unevenness of the thickness of the copper foil due to the variation of the plating thickness and the like are added. For this reason, in this embodiment, the in-plane of the actual pattern substrate is divided into a plurality of regions, and moreover, the peripheral region of the end is divided into smaller regions than the central region of the actual pattern substrate, and each of these regions is divided. A correction function was created for. Specifically, from 121 points of actual pattern data evenly arranged in the plane of the actual pattern substrate, 2 points from the central portion of the actual pattern substrate and 40 points from the edge peripheral portion are selected for a total of 42 points. Then, a correction function was created for each of these points. In the following, one of the 42 points at the center of the actual pattern substrate will be used for description.

  Further, in the circuit processing by etching, since the etching liquid is applied differently on the front and back surfaces of the actual pattern substrate, the tendency of the etching process changes. Therefore, in this embodiment, the primary correction function is created for each of the front and back surfaces of the actual pattern substrate, but in the following, only the front surface side (the surface that is the upper side when the circuit is formed by etching during horizontal transfer) The description will be omitted for the back side.

  Further, in circuit processing by etching, even in one of the front and back surfaces of the actual pattern substrate, the direction of the wiring pattern changes the relationship with the transport direction of the actual pattern substrate, the direction in which the etching liquid strikes, etc. The etching process tends to have directionality and unevenness. Therefore, in this embodiment, a primary correction function is created for each of the vertical wiring pattern and the horizontal wiring pattern of the test pattern. However, in the following, only the horizontal direction will be described and the vertical direction will be omitted.

  Next, the relationship between the difference data and the pattern gap, which is a factor that causes the difference, is examined, and as the relationship between the pattern gap and the correction amount of the original data for suppressing the difference, the primary correction as shown in FIG. I created a function.

The design value (unit: μm) of the pattern gap shown in the table of FIG. 12 is a numerical value of the pattern gap (factor that causes a difference) of the original data of the test pattern. The finish value (unit: μm) of the pattern gap is an actual measurement value of the pattern gap acquired from the actual pattern data using the optical automatic appearance inspection device. The one-sided correction amount (unit: μm) is the correction amount of the exposure data for suppressing the difference between the finish value of the pattern gap and the original data, and was calculated by the following formula (1). Here, the one-sided correction amount is used to correct the same amount on both sides of the pattern gap when correcting the exposure data.
One-sided correction amount (μm) = (finish value of pattern gap (μm) -design value (μm)) / 2 (1)

<Correction of exposure data using primary correction function>
The exposure data of the product pattern 7 is corrected by using the primary correction function. For example, in the exposure data of the product pattern 7, the finish value of the pattern gap in the table of FIG. From the data of 28.5 μm (one-sided correction amount is 4.3 μm) and finish value 33.2 μm (one-sided correction amount is 5.6 μm), the value of the one-sided correction amount when the finish value is 30 μm is It was obtained by calculation assuming that it was on a straight line connecting the data of 1. In this way, the one-sided correction amount when the finished value of the pattern gap obtained by the calculation is 30 μm is 4.7 μm, and it corresponds to the exposure data (here, the initial exposure data means the same as the original data). A correction process was performed to narrow the pattern gap by 4.7 μm on each side. That is, the exposure data after the correction process of the pattern gap is set to 20.6 μm at the portion where the design data (design value) of the pattern gap of the product pattern 7 is 30 μm.

  At this time, the exposure data of the product pattern 7 (here, the initial exposure data is meant and is the same as the original data) is subjected to correction processing using the primary correction function created above, and the exposure data of the test pattern 1 is obtained. No correction processing was applied to the. This is because for the test pattern 1, a real pattern is always created using the same exposure data as when creating the primary correction function, and real pattern data for creating the secondary correction function is sampled.

<Creation of secondary correction function>
Next, in order to create a secondary correction function, a substrate similar to the one used to create the primary correction function is prepared, and the actual pattern substrate C4c in which the test pattern 1 and the product pattern 7 as shown in FIG. 5 are mixed is mixed. Was prepared. The test pattern 1 uses the same design data as when creating the first-order correction function, and the product pattern 7 has design data having a portion where the design value of the pattern width / pattern gap is 30 μm / 30 μm. Is. FIG. 5 shows an outline of the actual pattern substrate C4c. In this embodiment, 42 test patterns 1 in total are actually arranged in the actual pattern substrate C4c. The positions of the 42 test patterns 1 arranged in the real pattern board C4c in the real pattern board C4c are 42 points in the real pattern board C4c used to create the primary correction function shown in FIG. It corresponds to the position of test pattern 1.

  Next, real pattern data (finished value) was acquired for the real pattern substrate C4c using an optical automatic visual inspection apparatus. The number of measurement points at this time is 42 points. Further, here, among the linear wiring patterns in which the pattern gap 2 of the test pattern of FIG. 2 is changed in 24 steps in the range of 14 to 150 μm, as shown in the table of FIG. The actual pattern data was acquired only for 6 steps of 60, 80 and 130 μm. Next, with respect to each of the 6-step pattern gaps, the actual pattern data is averaged for every 42 points, and the averaged actual pattern data is used to create a primary correction function in the same manner as in FIG. A quadratic correction function as shown in was created. The pattern gaps 2 at other stages are obtained by calculation on the assumption that they are on a straight line connecting the actual pattern data of the pattern gaps 2 at the stage of obtaining the actual pattern data. That is, although not shown in the table of FIG. 13, actual pattern data was actually acquired only for 6 steps, and all 24 steps of actual pattern data were estimated based on this. As a result, the actual pattern data to be acquired is reduced and the load of data processing is reduced.

<Creation of cubic correction function>
Next, the composite ratio (ratio) of the correction amount by the primary correction function in the entire correction amount by the primary correction function and the correction amount by the secondary correction function is set to 0.5, 0.7, 0.3, and the tertiary correction function is set. It was created. FIG. 14 shows a case where the combined ratio (ratio) of the correction amounts by the primary correction function is 0.5. Specifically, when the composite ratio (ratio) of the correction amounts by the primary correction function is 0.5, the composite ratio (ratio) of the correction amounts by the secondary correction function is 0.5. The one-sided correction amount of was synthesized by the following equation (2).
One-sided correction amount of the third-order correction function (μm) = One-sided correction amount of the first-order correction function (μm) × 0.5
+ One-sided correction amount (μm) of secondary correction function × 0.5 (2)
Here, the data of the one-sided correction amount when the secondary correction function is created is acquired only for the 6-step pattern gaps, which is less than the 24 steps when the primary correction function is created, but as described above, It was estimated by calculation so as to obtain the same data for 24 steps as when the primary correction function was created.

<Correction of exposure data using a cubic correction function>
The correction process for the exposure data of the product pattern 7 using the third-order correction function is performed by, for example, calculating the combined ratio (ratio) of the correction amount by the first-order correction function and the correction amount by the second-order correction function with respect to the entire correction amount. When the value is 0.5, in the exposure data of the product pattern 7, the finished value of the pattern gap in the table of FIG. 14 is 26.8 μm (the one-sided correction amount is 3.4 μm) for the portion having the design value of 30 μm. From the data of the finish value of 31.0 μm (the one-sided correction amount is 4.5 μm), it is assumed that the value of the one-sided correction amount when the finish value is 30 μm is on the straight line connecting these data. It was obtained by calculation. The one-sided correction amount when the finished value of the pattern gap thus obtained by calculation is 30 μm is 4.2 μm, and the exposure data (in this case, means the initial exposure data, not the corrected exposure data. The same as the data) was corrected by narrowing the pattern gap by 4.2 μm on each side. That is, the exposure data after the correction process of the pattern gap is set to 21.6 μm at the portion where the design data (design value) of the pattern gap of the product pattern 7 is 30 μm. Further, the third-order correction function used to correct the original data or the exposure data of the product pattern (not shown) arranged in the actual pattern substrate C4c is created by using the test pattern 1 having the closest positional relationship with each product pattern. The third-order correction function was used.

  At this time, as in the case of creating the second-order correction function, the third-order correction function created above is used for the exposure data of the product pattern 7 (here, it means the initial exposure data and is the same as the original data). The exposure data of the test pattern 1 is not subjected to the correction process. This is because for the test pattern 1, an actual pattern is always created using the same exposure data as when the primary correction function was created, and actual pattern data for creating subsequent secondary correction functions is sampled.

<Creation of actual pattern board for evaluation>
Next, in order to create a real pattern substrate for evaluation, a real pattern in which the test pattern 1 and the product pattern 7 as shown in FIG. Three substrates C4c were produced. It should be noted that the exposure data of the product pattern 7 (which means the initial exposure data, not the corrected exposure data here, and is the same as the original data) is subjected to correction processing using the third-order correction function created above. The exposure data of the test pattern 1 was subjected to the exposure by creating the exposure data which is not subjected to the correction process.

(Comparative Example 1)
In the same manner as when the above-described secondary correction function was created, three actual pattern substrates C4c in which the test pattern 1 and the product pattern 7 shown in FIG. 5 were mixed were prepared. For the exposure data of the product pattern 7 (here, not the corrected exposure data, but the initial exposure data, which is the same as the original data), correction processing is performed using only the primary correction function created above. Then, the exposure data of the test pattern 1 was subjected to the exposure process without being subjected to the correction process, and the exposure was performed. Other than that, it is the same as that of the first embodiment.

(Comparative example 2)
In the same manner as when the above-described secondary correction function was created, three actual pattern substrates C4c in which the test pattern 1 and the product pattern 7 shown in FIG. 5 were mixed were prepared. It should be noted that the exposure data of the product pattern 7 (which means the initial exposure data, not the corrected exposure data here, and is the same as the original data) is corrected using only the secondary correction function created above. Then, the exposure data of the test pattern 1 was subjected to exposure processing without correction processing, and then exposure was performed. Other than that, it is the same as that of the first embodiment.

<Evaluation of process capability>
Regarding the actual pattern substrates for evaluation produced in Example 1 and Comparative Examples 1 and 2, the product pattern 7 has a design width / pattern gap design value of 30 μm / 30 μm (121 points / sheet × 3 sheets). ), The pattern width was measured at a magnification of 200 times using a measuring microscope, and the process capability was investigated from the difference from the original data (design value). The results are shown in Table 1. The ± 10 μm and ± 12 μm described in the right column of the process capability indicate the allowable range of variation with respect to the design value (30 μm / 30 μm) of the pattern width / pattern gap, and Cp and Cpk represent the process capability index.

  As shown in Table 1, in the present embodiment, the combined ratio (ratio) of the correction amount of the primary correction function and the correction amount of the secondary correction function is 0.5, 0. In the examples in which the third-order correction function was created as 7 and 0.3, sufficient process capability was obtained in all cases, and the best result was obtained when the synthesis ratio of the first-order correction function was 0.5. On the other hand, in Comparative Examples 1 and 2, the process capability tends to be lower than that in Examples.

(Example 2)
Similar to the first embodiment, <creation of primary correction function>, <correction of exposure data using primary correction function>, <creation of secondary correction function>, <creation of tertiary correction function>, <tertiary correction function> Correction of exposure data using.

<Creation of other secondary correction functions>
Next, in order to create another quadratic correction function, the actual pattern in which the test pattern 1 and the product pattern 7 as shown in FIG. One substrate C4c was produced. It should be noted that the exposure data of the product pattern 7 (which means the initial exposure data, not the corrected exposure data here, and is the same as the original data) is subjected to correction processing using the third-order correction function created above. The exposure data of the test pattern 1 was subjected to the exposure by creating the exposure data which is not subjected to the correction process.

  Next, with respect to the actual pattern substrate C4c, the actual pattern data (finishing value) is acquired by using the optical automatic appearance inspection device, and the same as when the above-described secondary correction function is created, as shown in FIG. A quadratic correction function like this was created. Further, regarding the pattern gaps 2 at the other stages, in the same manner as when the above-described secondary correction function is created, the pattern gaps 2 at the stage where the actual pattern data is acquired are placed on the straight line connecting the real pattern data. It was obtained by calculation assuming that there is. That is, although not shown in the table of FIG. 15, the actual pattern data was actually acquired only for 6 steps, and all 24 steps of the actual pattern data were estimated based on this. As a result, the actual pattern data to be acquired is reduced and the load of data processing is reduced.

<Creation of other cubic correction function>
Next, as in the case of creating the above-described third-order correction function, as shown in FIG. 16, the combined ratio (rate) of the correction amount by the first-order correction function and the correction amount by the other second-order correction function is set to 0. 5, another third-order correction function was created. Specifically, the one-sided correction amount for each pattern gap is combined by the following equation (3).
One-sided correction amount (μm) of other cubic correction function = (One-sided correction amount (μm) of primary correction function)
+ One-sided correction amount (μm) of other secondary correction function) / 2 (3)
That is, in the present embodiment, the primary correction function is fixed, and another secondary correction function newly created each time an actual pattern substrate is manufactured is combined with this same primary correction function to obtain another Create a cubic correction function. Therefore, it is suitable when the situation of the production process is unlikely to change even if the number of productions increases, and is relatively stable. Here, the data of the one-sided correction amount when the other secondary correction function is created is acquired only for the pattern gap of 6 steps which is less than the 24 steps when the primary correction function is created, but as described above. In addition, it was estimated by calculation so as to obtain the same data for 24 steps as when the primary correction function was created.

<Correction of exposure data using another cubic correction function>
The correction process for the exposure data of the product pattern 7 using another third-order correction function is, for example, a combination ratio of the correction amount by the first-order correction function and the correction amount by the first-order correction function to the whole correction amount by the second-order correction function. When the (ratio) is 0.5, in the exposure data of the product pattern 7, the finished value of the pattern gap in the table of FIG. 16 is 28.6 μm (the one-sided correction amount is 4 when the design value is 30 μm). .3 μm) and the finished value of 32.9 μm (the one-sided correction amount is 5.4 μm), the value of the one-sided correction amount when the finished value is 30 μm is on the straight line connecting these data. It was obtained by calculation assuming that there is. When the finished value of the pattern gap thus obtained by calculation is 30 μm, the one-sided correction amount is 4.7 μm, and the exposure data (here, not the corrected exposure data, but the initial exposure data is used. The same as the data) was corrected by narrowing the pattern gap by 4.7 μm on each side. That is, the exposure data after the correction process of the pattern gap is set to 20.6 μm at the portion where the design data (design value) of the pattern gap of the product pattern 7 is 30 μm.

  At this time, the exposure data of the product pattern 7 (here, the initial exposure data is meant and is the same as the original data) is subjected to correction processing using the other cubic correction function created above, and the test pattern 1 The exposure data was not corrected. This is because for the test pattern 1, an actual pattern is always created using the same exposure data as when the primary correction function was created, and actual pattern data for creating subsequent secondary correction functions is sampled.

(Example 3)
Similar to the second embodiment, <creation of primary correction function>, <correction of exposure data using primary correction function>, <creation of secondary correction function>, <creation of cubic correction function>, <cubic correction function> Correction of exposure data using <> and <Creation of other secondary correction function> were performed.

<Creation of other cubic correction function>
Next, the combined ratio (ratio) of the correction amount by the third-order correction function to the entire correction amount by the third-order correction function and the correction amount by the other second-order correction function created above is 0.5, 0.7, 0. As for 3, another third-order correction function was created. FIG. 17 shows a case where the composite ratio (ratio) of the correction amounts by the tertiary correction function is 0.5. Specifically, when the synthesis ratio (ratio) of the correction amounts by the third-order correction function is set to 0.5, the synthesis ratio (ratio) of the correction amounts by the other secondary correction functions is 0.5, so that each pattern The one-sided correction amount for each gap was synthesized by the following equation (4).
One-sided correction amount (μm) of another cubic correction function = One-sided correction amount (μm) of the third-order correction function × 0.5
+ One-sided correction amount (μm) of another secondary correction function × 0.5 (4)
That is, in the present embodiment, by synthesizing another secondary correction function newly created every time the actual pattern substrate is manufactured, with respect to the tertiary correction function that is a combination of the primary correction function and the secondary correction function, Create another cubic correction function. Therefore, the third-order correction function used for correcting the original data or the exposure data is a cumulative result of the second-order correction function repeatedly, so that the situation of the manufacturing process fluctuates in a specific direction due to the accumulation of the number of productions. It is suitable when there is a tendency. Here, the data of the one-sided correction amount when the other secondary correction function is created is acquired only for the pattern gaps of 6 steps, which is less than the 24 steps when the tertiary correction function is created, but as described above. In addition, it was estimated by calculation so as to obtain the same data for 24 steps as when the third-order correction function was created.

<Correction of exposure data using another cubic correction function>
The correction process for the exposure data of the product pattern 7 using another third-order correction function is, for example, a combination ratio of the correction amount by the third-order correction function to the total correction amount by the third-order correction function and the correction amount by another second-order correction function. When the (ratio) is 0.5, in the exposure data of the product pattern 7, the finished value of the pattern gap in the table of FIG. 17 is 27.8 μm (the one-sided correction amount is 3 when the design value is 30 μm). .9 μm) and the finished value is 31.8 μm (the one-sided correction amount is 4.9 μm), the one-sided correction amount value when the finished value is 30 μm is on a straight line connecting these data. It was obtained by calculation assuming that there is. The one-sided correction amount is 4.5 μm when the finished value of the pattern gap obtained by the calculation is 30 μm, and the exposure data (in this case, the initial exposure data is not the corrected exposure data, The same as the data) was subjected to a correction process for narrowing the pattern gap by 4.5 μm on each side. That is, the exposure data after the correction processing of the pattern gap is set to 21.0 μm at the portion where the design data (design value) of the pattern gap of the product pattern 7 is 30 μm.

  At this time, the exposure data of the product pattern 7 (here, the initial exposure data is meant and is the same as the original data) is subjected to correction processing using the other cubic correction function created above, and the test pattern 1 The exposure data was not corrected. This is because for the test pattern 1, an actual pattern is always created using the same exposure data as when the primary correction function was created, and actual pattern data for creating subsequent secondary correction functions is sampled.

1: test pattern 2: pattern gap 3: pattern 4: actual pattern substrate 4a: actual pattern substrate A (test substrate)
4b: Actual pattern board B (test board)
4c: Actual pattern board C (product board)
5: Base material 6: Product pattern area 7: Product pattern 8: Data correction device 9: Bus 10: Wiring pattern forming system

Claims (14)

A step (A) of creating exposure data based on the original data of the target wiring pattern,
A step (B) of creating real pattern data representing a wiring pattern actually processed on the real pattern substrate from the real pattern substrate formed using the exposure data;
A step (C) of creating correction data of the original data or the exposure data based on a difference between the original data and the actual pattern data,
The step (C) of creating the correction data includes
Creating difference data from the difference between the actual pattern data and the original data or exposure data (C-1);
From the relationship between the difference data and the factor causing the difference related to the specification of the wiring pattern specified from the actual pattern data , the factor causing the difference and the correction amount of the original data for suppressing the difference Or a step (C-2) of creating a plurality of correction functions that define a relationship with the correction amount of the exposure data,
A step (C-3) of creating a correction function that combines the plurality of correction functions,
Calculates a correction amount corresponding to the value corresponding to the factors of the original data or the exposure in data by using the correction function obtained by the synthesis, the original data or the exposure data of the wiring pattern by using the correction amount Compensating step (C-4),
A method of manufacturing a wiring board having the following. In the step (C-2) of creating the plurality of correction functions, a primary correction function created using an actual pattern substrate and a secondary correction function created using another actual pattern substrate are created,
In the step (C-3) of creating the combined correction function, a tertiary correction function is created by combining the primary correction function and the secondary correction function,
The wiring board according to claim 1, wherein in the step (C-4) of correcting the original data or the exposure data, the original data or the exposure data of the wiring pattern of the actual pattern board is corrected using the cubic correction function. Manufacturing method. In the step (C-2) of creating the plurality of correction functions, in addition to the quadratic correction function, another quadratic correction function created using another real pattern substrate is created.
3. The step (C-3) of creating the combined correction function creates the other tertiary correction function by combining the tertiary correction function and the further created secondary correction function. Manufacturing method of wiring board. In the step (C-2) of creating the plurality of correction functions, the actual pattern substrate used to create the primary correction function and another actual pattern substrate used to create another secondary correction function, or The method for manufacturing a wiring board according to claim 3 , wherein the other actual pattern board has the same wiring pattern. Wherein the plurality of steps to create a correction function (C-2), the other secondary correction function that is created after the secondary correction function is created produced each lot or the actual pattern for each substrate of the actual pattern substrate The method for manufacturing a wiring board according to claim 3, wherein As a factor that causes a difference between the actual pattern data and the original data or exposure data, any one of the pattern gap, pattern size, pattern thickness, and pattern position of the original data or exposure data of the wiring pattern of the actual pattern substrate, or The method for manufacturing a wiring board according to claim 1, wherein actual pattern data corresponding to a combination of any two or more is used. A data correction device for original data of a wiring pattern or exposure data, which is used for manufacturing a wiring board ,
Original data of the target wiring pattern or exposure data created based on this original data, and a wiring pattern actually processed on the actual pattern substrate created from the actual pattern substrate formed using the exposure data create the differential data from the difference between the actual pattern data representing (C-1),
From the relationship between the difference data and the factor causing the difference related to the specification of the wiring pattern specified from the actual pattern data , the factor causing the difference and the correction amount of the original data for suppressing the difference. Alternatively, a plurality of correction functions that define the relationship with the correction amount of the exposure data are created (C-2),
A correction function is created by synthesizing the plurality of correction functions (C-3),
A correction amount corresponding to a value corresponding to the factor in the original data or the exposure data corrected using a correction function created by combining the plurality of correction functions is calculated, and the wiring is calculated using the correction amount. It said pattern by correcting the original data or the exposure data to create the correction data (C-4), the data correction apparatus. When creating a plurality of correction functions (C-2), a primary correction function created using an actual pattern substrate and a secondary correction function created using another actual pattern substrate are created,
When creating the combined correction function (C-3), create a tertiary correction function that combines the primary correction function and the secondary correction function,
8. The data correction according to claim 7 , wherein in correcting (C-4) the original data or the exposure data, the third-order correction function is used to correct the original data or the exposure data of the wiring pattern of the actual pattern substrate. apparatus. When creating the plurality of correction functions (C-2), in addition to the secondary correction function, another secondary correction function created by using another actual pattern substrate is created.
Wherein the time of creating the synthesized correction function (C-3), by combining the other of the secondary correction function said further created when the three-order correction function to create other tertiary correction function, to claim 8 The data correction device described. When creating the plurality of correction functions (C-2), the actual pattern board used to create the primary correction function and another actual pattern board used to create another secondary correction function 10. The data correction device according to claim 9 , wherein the other real pattern substrate has the same wiring pattern. Wherein the time of creating a plurality of correction functions (C-2), the other secondary correction function that is created after the secondary correction function is created for each manufacturing lot or every actual pattern substrate of the actual pattern substrate The data correction apparatus according to claim 9 or 10 , which is performed. As a factor that causes a difference between the actual pattern data and the original data or exposure data, any one of the pattern gap, pattern size, pattern thickness, and pattern position of the original data or exposure data of the wiring pattern of the actual pattern substrate, or The data correction apparatus according to any one of claims 7 to 11, wherein actual pattern data corresponding to any two or more combinations is used. A data correction device according to any one of claims 7 to 12 ,
A pattern exposure apparatus that exposes an exposure pattern on a photosensitive resist disposed on a substrate based on exposure data created from original data corrected by the data correction apparatus or exposure data corrected by the data correction apparatus When,
A developing pattern forming apparatus for forming a developing pattern by developing the photosensitive resist having the exposed exposure pattern;
An actual pattern forming device for forming an actual pattern by performing circuit processing on the substrate on which the development pattern is formed;
An actual pattern data creating apparatus for creating actual pattern data from the actual pattern, the wiring pattern forming system. A data correction method of original data of a wiring pattern or exposure data, which uses the data correction device according to claim 7 .
Original data of the target wiring pattern or exposure data created based on this original data, and a wiring pattern actually processed on the actual pattern substrate created from the actual pattern substrate formed using the exposure data A step (C-1) of creating difference data from the difference from the actual pattern data represented ,
From the relationship between the difference data and the factor causing the difference related to the specification of the wiring pattern specified from the actual pattern data , the factor causing the difference and the correction amount of the original data for suppressing the difference. Or a step (C-2) of creating a plurality of correction functions that define a relationship with the correction amount of the exposure data,
A step (C-3) of creating a correction function that combines the plurality of correction functions,
A correction amount corresponding to a value corresponding to the factor in the original data or the exposure data is calculated by using a correction function created by combining the plurality of correction functions, and the correction amount is used to calculate the wiring pattern of the wiring pattern. the original data or the step of creating correction data by correcting the exposure data and (C-4),
A data correction method having.

Images