1. Field of the Invention
The present invention relates to a technique for correcting registration of a pattern formed on a wafer, and in particular, to a registration correction amount calculating apparatus calculating an amount of correction for a registration inspection apparatus from registration inspection data.
2. Description of the Background Art
Registration of a pattern already formed on a wafer and a pattern on a mask to be transferred has conventionally been performed in a process of manufacturing a semiconductor. In general, registration of a pattern formed on a wafer has been controlled solely based on a condition variable in a photolithography process (a series of processes including exposure and development after a sensitizer is applied to the wafer).
In inspecting registration of the formed pattern, the pattern has been recognized as excellent in quality if a measurement result is closer to a target value, which is set in advance along with a standard value. Techniques related thereto are disclosed in Japanese Patent Laying-Open No. 11-260683, and No. 2000-323383.
An apparatus for manufacturing a semiconductor material disclosed in Japanese Patent Laying-Open No. 11-260683 uses measurement data processed previously of a step performed after an exposure step, in determining an exposure condition.
According to a method of manufacturing a semiconductor disclosed in Japanese Patent Laying-Open No. 2000-323383, alignment correction is attained in the following manner. That is, registration error in a wafer that has been subjected to exposure and development process is measured. Then, an alignment correction value is calculated from the measured registration error for storage. Among the stored alignment correction values, based on an alignment correction value of a condition common to a process condition for a wafer to be subjected to exposure and development process in the next place, alignment correction in exposing a next wafer is achieved.
In manufacturing a semiconductor, a registration position in photolithography process is displaced also by an influence other than that of the photolithography process, such as displacement of an underlying layer through CMP (Chemical Mechanical Polishing), for example. Therefore, if registration is controlled solely based on a condition variable in the photolithography process, registration of the finished pattern may significantly deviate from an ideal, target value.
In addition, registration is performed based on a projected portion and a depressed portion regarded as a marker. If this marker is not correct, for example, if the markers are not symmetrically positioned, positions of the registration marks are not correctly recognized, resulting in a significant error in registration measurement. The error may even account for 50% of allowance in process accuracy. Consequently, a pattern generated through the photolithography process deviates from the target value. In many cases, this is a problem in the photolithography process caused by a specific wafer process.
Moreover, the error, that is, deviation from the target value, is revealed only after a process that cannot be restored, such as etching. These problems cannot be solved even with the above-mentioned conventional art.
An object of the present invention is to provide a registration correction amount calculating apparatus capable of improving process accuracy in terms of attaining a target value even if an underlying layer is displaced.
According to one aspect of the present invention, a registration correction amount calculating apparatus includes a data storage portion storing a measurement value of registration accuracy after a process in a first step and a measurement value of registration accuracy after a process in a second step measured by a registration inspection apparatus; and an operation portion calculating an amount of correction for the registration inspection apparatus from the measurement value of registration accuracy after the process in the first step and the measurement value of registration accuracy after the process in the second step stored by the data storage portion. Therefore, the registration inspection apparatus can be controlled so as to improve process accuracy in terms of attaining a target value, even if an underlying layer is displaced through the first step.
According to another aspect of the present invention, a registration correction amount calculating apparatus includes a data storage portion storing a measurement value of registration accuracy after a process in a first step and a measurement value of registration accuracy after a process in a second step measured by a registration inspection apparatus; and an operation portion calculating an amount of correction for an exposure apparatus from the measurement value of registration accuracy after the process in the first step and the measurement value of registration accuracy after the process in the second step stored by the data storage portion. Therefore, the exposure apparatus can be controlled so as to improve process accuracy in terms of attaining a target value, even if an underlying layer is displaced through the first step.
The foregoing and other objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings.
After the wafer is subjected to the etching process, a resist is removed. Then, registration accuracy of the pattern formed on the wafer is measured again (step 22).
Step 21 is performed for registration inspection after the pattern is subjected to the photolithography process, and step 22 is performed for inspecting finish after the series of process steps.
Next, a lot b is subjected to the photolithography process (S5). Then, an optimal amount of registration correction 33a for a next lot (lot b) is calculated from a measurement value 31a of registration accuracy after the photolithography process and a measurement value 32a of registration accuracy after the etching process of lot a (S6), using the following equation.
(optimal amount of registration correction 33)=(measurement value 31)−(measurement value 32). (1)
Thereafter, registration accuracy of lot b is corrected, using calculated optimal amount of registration correction 33a, and lot b is subjected to registration inspection (step 21) after the photolithography process (S7). Then, lot b is subjected to the etching process (S8), and to registration inspection (step 22) after the etching process (S9).
Next, a lot c is subjected to the photolithography process (S10). Then, an optimal amount of registration correction 33b for a next lot (lot c) is calculated from a measurement value 31b of registration accuracy after the photolithography process and a measurement value 32b of registration 10 accuracy after the etching process of lot b, using equation (1). Similar steps will be repeated for a subsequent process.
In this manner, amount of registration correction 33 for a lot to be processed in the next place is calculated using previous measurement values 31 and 32 of registration accuracy for feedback. Thus, registration inspection taking into account displacement of the underlying layer or the apparatus can be attained.
Registration correction amount calculating apparatus 3 includes: a data storage portion 11 storing measurement value 31 of registration accuracy after the photolithography process detected by post-photolithography registration accuracy detecting portion 1 and measurement value 32 of registration accuracy after the etching process detected by post-etching registration accuracy detecting portion 2; and an operation portion 12 calculating optimal amount of registration correction 33 from measurement value 31 of registration accuracy after the photolithography process and measurement value 32 of registration accuracy after the etching process stored in data storage portion 11.
A program for calculating an amount of registration correction is supplied by a medium such as FD 34 or CD-ROM 38. When the program for calculating an amount of registration correction is executed by computer unit 31, the optimal amount of registration correction is calculated. Alternatively, the program for calculating an amount of registration correction may be supplied to computer unit 31 via network communication unit 39 from other computers.
Computer unit 31 shown in
The program for calculating an amount of registration correction recorded in FD 34 or CD-ROM 38 is once stored in hard disk 43 via FD drive 33 or CD-ROM unit 37 by CPU 40. CPU 40 loads the program for calculating an amount of registration correction from hard disk 43 to RAM 42 for execution as required, whereby calculation of the amount of registration correction is attained.
Then, referring to the recipe for the lot process, operation portion 12 reads measurement value 31 of registration accuracy after the photolithography process and measurement value 32 of registration accuracy after etching from data storage portion 11 detected through the previous lot process, and calculates an optimal amount of registration correction, using equation (1) (S22).
Next, measurement correction portion 5 instructs the optimal amount of registration correction calculated by operation portion 12 to registration inspection apparatus 4 (S23). Registration inspection apparatus 4 performs registration inspection after the photolithography process using the optimal amount of registration correction instructed by measurement correction portion 5 (S24), and reports a measurement result of registration accuracy after the photolithography process to post-photolithography registration accuracy detecting portion 1 (S25). Data storage portion 11 obtains and stores measurement value 31 of registration accuracy after photolithography process from post-photolithography registration accuracy detecting portion 1 (S26).
Then, registration inspection apparatus 4 performs registration inspection after etching (S27), and reports a measurement result of registration accuracy after etching to post-etching registration accuracy detecting portion 2 (S28). Data storage portion 11 obtains and stores measurement value 32 of registration accuracy after etching from post-etching registration accuracy detecting portion 2 (S29).
The series of processes as above are performed with respect to each lot, each step and each apparatus, and displacement of the underlying layer or the like is continuously corrected. Therefore, registration accuracy in terms of attaining the target value can be enhanced, and process quality in a product can be improved.
In addition, when there are a plurality of lots that went through measurement of registration accuracy after the photolithography process and registration accuracy after etching, for example, when lot a and lot b are present, an optimal amount of registration correction for lot c may be calculated using the following equation.
(optimal amount of registration correction for lot c)=operation (optimal amount of registration correction for lot a 33a, optimal amount of registration correction for lot b 33b). (2)
Here, as an operation, average or center is used, for example, and the number of pieces of data to be used for the operation is separately specified in advance.
Corrected registration accuracy detecting portion 6 detects a corrected registration result measured by registration inspection apparatus 4. The corrected registration result measured by registration inspection apparatus 4 is expressed in an equation below.
(corrected registration result)=(registration measurement result after the photolithography process)−(optimal amount of registration correction). (3)
Corrected registration accuracy detecting portion 6 corrects exposure apparatus 7, using the corrected registration result measured by registration inspection apparatus 4.
As described above, according to a registration correction amount calculating apparatus in the present embodiment, an optimal amount of registration correction is calculated from the previous measurement value of registration accuracy after photolithography process and the previous measurement value of registration accuracy after the etching process. Therefore, even if displacement of the underlying layer or the like takes place, process accuracy in terms of attaining the target value can be improved.
In addition, conditioning for re-determining an optimal amount of registration correction (a preceding pilot process before a process) is not necessary, and an optimal amount of registration correction can be obtained without causing delay in manufacture, thereby reducing a manufacturing time. Further, a wafer for conditioning (a wafer to be used for the preceding pilot process) is not needed, and waste of the wafer can be avoided.
In the first embodiment, registration inspection apparatus 4 uses an optimal amount of registration correction to measure registration accuracy after the photolithography process, and outputs the measurement value of the registration accuracy (corrected registration measurement result) to corrected registration accuracy detecting portion 6. On the other hand, in the present embodiment, an operation portion 12′ calculates a corrected registration result from registration measurement value 31 after the photolithography process and optimal amount of registration correction 33, using the next equation.
(corrected registration result)=(registration measurement result 31 after the photolithography process of a currently-measured lot)−(optimal amount of registration correction 33 of a previous lot). (4)
Measurement correction portion 5 uses the corrected registration result calculated by operation portion 12′ to correct exposure apparatus 7.
As described above, according to the registration correction amount calculating apparatus in the present embodiment, operation portion 12′ calculates a corrected registration result from registration measurement value 31 after the photolithography process and optimal amount of registration correction 33. Therefore; even if registration inspection apparatus 4 without a function to use optimal amount of registration correction 33 for correction is employed, an effect as described in the first embodiment can be obtained.
Brief configuration of a semiconductor manufacturing system in the third embodiment of the present invention is similar to those in the first and second embodiments shown in
There exist a plurality of categories required in forming a pattern on a wafer, such as wafer offset, wafer rotation, wafer scaling or the like in registration. In the present embodiment, a category for calculating an optimal amount of registration correction is selected. A plurality of categories may be selected. For example, if wafer offset and wafer rotation are selected, the correction process described in the first or second embodiment is performed in parallel to respective categories. Each category is independently handled.
As described above, according to the semiconductor manufacturing system in the present embodiment, a category for calculating an optimal amount of registration correction is selected. Therefore, the correction process can be performed solely on a category which is desirably brought closer to the target value.
Brief configuration of a semiconductor manufacturing system in the fourth embodiment of the present invention is similar to those in the first and second embodiments shown in
In the semiconductor manufacturing system in the present embodiment, measurement value for each category is not limited to one from a lot. On the other hand, registration accuracy is measured at each point on a wafer and a shot respectively, and results are reported to registration correction amount calculating apparatus 3.
As described above, according to the semiconductor manufacturing system in the present embodiment, an optimal amount of registration correction is calculated for each lot, each wafer, or each shot. Therefore, a point which is desirably brought closer to the target value can be set finely, and correction process can be performed solely on that point.
Although the present invention has been described and illustrated in detail, it is clearly understood that the same is by way of illustration and example only and is not to be taken by way of limitation, the spirit and scope of the present invention being limited only by the terms of the appended claims.
Number | Date | Country | Kind |
---|---|---|---|
2003-100606(P) | Apr 2003 | JP | national |