Determining exposure time of wafer photolithography process

Details Determining exposure time of wafer photolithography process Determining exposure time of wafer photolithography process

2002-08-12

2004-06-29

Young, Christopher G. (Department: 1756)

Radiation imagery chemistry: process, composition, or product th

Including control feature responsive to a test or measurement

C716S030000

Reexamination Certificate

active

06756168

ABSTRACT:

CROSS-REFERENCES TO RELATED APPLICATIONS
This application claims priority from R.O.C. patent application Ser. No. 090120147, filed Aug. 16, 2001, the entire disclosure of which is incorporated herein by reference.
BACKGROUND OF THE INVENTION
The present invention relates to a method for determining an exposure time of a wafer photolithography process, and more particularly to a method for determining the wafer exposure time by analyzing the exposure times of at least previous three batches of production wafers, to reduce the effects of parameter perturbations of facilities and materials.
In semiconductor manufacturing, the photolithography process is an important step, which can include several procedures of dehydration baking, priming, photoresisting, soft baking, exposing, post exposure baking, developing and hard baking, etc. Among those procedures, the exposure procedure is the subject matter of this invention.
The purpose of wafer exposure in the art is to have the photoresist layered over the surface of the wafer to undergo an effective photochemical transformation after absorbing adequate energy, and thereby to enable the developed photoresist to accurately transfer the pattern on the mask onto the wafer for being ready to perform a subsequent etching process. Two of the major operation conditions for the exposure energy control are exposure intensity and exposure time (ET), both of which can directly affect the process yield after wafer etching.
In the wafer photolithography and subsequent etching processes, factors perturbing the wafer exposure energy or the exposure process yield are numerous and can include almost all the operation parameters of each procedure in the entire photolithography process. Some of the crucial factors are photoresist material, photoresist thickness, soft bake extent, development condition, tolerance error of photoresist line width after developed, batch wafer condition, parameter perturbations of machine facilities, and the like. Consequently, in the batch wafer photolithography process, the exposure condition thereof should be adjusted frequently to meet the changes of disturbances or diminutive perturbations which affect the exposure parameters.
In order to control the yield of the photolithography process, it is common to sample the production wafers after completion of one batch photolithography process in the production line for fabricating a test wafer and to take the critical dimension (CD) of the etching line width of the test wafer as a criterion for evaluating the process yield. If the CD value is specs-in (i.e., qualified), the photolithography and etching processes can follow the original exposure conditions. If the CD value is specs-out (i.e., unsatisfied), the photolithography parameters will need to be adjusted (in particular, the exposure time thereof needs to be adjusted). However, in the case that the CD value deviates significantly from the standard, it obviously indicates the need for trouble shooting of the entire photolithography process so as to locate the problem. Regardless of the cause of the perturbations of the system, the yield can usually be adjusted through controlling the exposure time.
To more clearly manifest the time sequential relationship between the inspection line for examining the test wafer and the production line of the production wafer for the photolithography process as a function of time,
FIG. 1
shows a schematic diagram of the time sequential relationship of the wafer photolithography production line and the test wafer inspection line among batches.
As shown in
FIG. 1
, the block L(N) in the production line denotes the photolithography process for the N batch wafer and the block T(N) in the inspection line denotes the examination process for the N batch wafer. The time axis extending from the left side of
FIG. 1
to the right side thereof represents the time evolution. As shown in
FIG. 1
, the exposure condition, on which the manufacture of the L(N) batch production wafer in the production line is based, comes from the T(N−2) batch test wafer. The T(N−2) batch test wafer is sampled from the L(N−2) batch production wafer.
In the above-mentioned process time sequence, while the T(N−2) batch test wafer is under process examination in the inspection line, the photolithography process of the L(N−1) batch production wafer in the production line is simultaneously progressing under the exposure condition on basis of the examination from the T(N−3) batch test wafer. Meanwhile, as seen in
FIG. 1
, the T(N−3) batch test wafer is sampled from the L(N−3) batch production wafer.
The aforesaid method for determining the wafer photolithography exposure condition is carried out in the manner of “spacing one batch production wafer.” The feature of such a method resides in that, for the system composed of the production line and the inspection line, two independent groups are formed: an odd-batch group (including the production wafers and the test wafers of each batch linking with a center line in
FIG. 1
) and an even-batch group (including the production wafers and the test wafers of each batch linking with a dotted line in FIG.
1
).
Nevertheless, this method has at least the following drawbacks. First, when the test wafer CD value is used to judge the input value of exposure time for the batch production wafers in the identical group, the problem of response delay would occur. In the operation of the two independent groups, when one batch test wafer in one group is examined and determined to require large-scale adjustments upon the exposure condition, the adjustments cannot be instantly applied to the other group. In other words, such an approach cannot immediately determine whether the batch production wafer of the other group needs the corresponding adjustments upon the exposure time so as to respond to the possible influences by perturbations of the system parameters. For example, when the L(N−2) batch production wafer in the even-batch group is found to have larger system perturbations from examining the T(N−2) batch test wafer, the adjustments to the exposure time are made on the L(N) batch production wafer. However, the L(N−1) batch production wafer in the odd-batch group can not be suitably reflected in time. Actually, it is not until after the T(N−1) batch test wafer has been determined to need adjustment that the adjusted exposure time can be reflected in the L(N+1) batch production wafer. Under such an arrangement, at least two batches of production wafers (i.e., from L(N−2) batch to L(N−1) batch in this example) will be affected when system perturbations take place.
Second, when the test wafer CD value is used to decide whether or not an adjustment upon the exposure time of the next batch production wafer is required, the problem of adjustment fluctuation of the exposure time would arise. In the operation of the two independent groups, when one batch test wafer in one group is examined to need additional adjustments to the exposure condition, the adjustments are directly made on the production line. Yet, for the other group, such adjustments could become an impulse resulting in an adverse effect upon the production in the group and, empirically, such an impulse may merely be excluded by means of system damping.
For instance, assuming that the L(N−2) batch production wafer in the even-batch group is found to have larger system perturbations from examining the T(N−2) batch test wafer, the adjustment to the exposure time on the production line (which is referred to as “a first adjustment quantity”) is primarily applied to the L(N) batch production wafer. Similarly, assuming that the T(N−1) batch test wafer is also examined to need adjustments to the exposure condition, the adjustment made on the production line (which is referred to as “a second adjustment quantity”) is related to the L(N−1) batch production wafer which does not undergo the first adjustment quantity, rather than to th

Affiliated with

Also associated with

Rating

Determining exposure time of wafer photolithography process does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Determining exposure time of wafer photolithography process, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Determining exposure time of wafer photolithography process will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3312983