In-line focus monitor structure and method using top-down SEM

Details In-line focus monitor structure and method using top-down SEM In-line focus monitor structure and method using top-down SEM

2002-05-30

2004-08-17

Young, Christopher G. (Department: 1756)

Radiation imagery chemistry: process, composition, or product th

Including control feature responsive to a test or measurement

C430S005000, C430S296000, C430S942000, C430S967000, C257S048000

Reexamination Certificate

active

06777145

ABSTRACT:

FIELD OF THE INVENTION
The present invention relates to the manufacture of integrated circuit devices. In particular, the present invention is a structure on a reticle which is used to control a lithography process such that the best focus for imaging is constantly maintained. The present invention is also a method by which the structure is used to determine best focus.
BACKGROUND OF THE INVENTION
The manufacture of integrated circuits is a highly complex process requiring constant improvements in exposure and etching tools, reticles, photoresist, measurement tools and other processing equipment. As the size of devices shrinks and circuits become more densely packed to provide greater performance, the critical dimension of structures in state of the art devices is rapidly approaching 100 nm (0.1 microns). For some advanced products in development, the critical dimension of the smallest feature is already in the sub-100 nm region.
Improvements in exposure tools consist of shifting to shorter exposure wavelengths to enable printing of smaller features according to the Raleigh resolution equation: CD=k&lgr;/NA, where CD is the lateral dimension of the feature to be printed in the photoresist, k is a constant, &lgr; is the wavelength of exposing energy, and NA is the numerical aperture of the exposure tool. During the past 20 years, the exposure wavelength has decreased from over 400 nm to 193 nm. In the same period, the energy source has changed from a mercury lamp emitting broadband light to a laser that emits a very narrow wavelength. It should be noted that multiple feature sizes on various levels of a device are required during production. Thus, a larger dimension on one level may be printed with longer wavelength from one tool while a small feature on a different level may be printed with a shorter wavelength from a second exposure tool. In general, the longer wavelength exposure tools are more mature and more cost effective and are used unless they are not capable of printing the finer dimensions which are needed in newer technologies.
Those skilled in the art will appreciate that other energy sources are also capable of imaging photoresist in a production mode and may be included within the scope of this invention. Electron beam, X-ray, extreme UV (EUV), or ion beam sources may be employed for lithography.
U.S. Pat. No. 6,320,648 describes a method and apparatus for improving pattern fidelity in the imaging process but is directed toward transmitting a pattern from a reticle to a wafer and does not relate to process control during production of a device.
The reticle or mask is generally comprised of a quartz plate which is highly transparent to the exposing radiation and a chrome coating which has been deposited and patterned on one of the two large sides during the mask making process. The chrome is opaque to the radiation from the exposure tool during the photoresist imaging process. This property allows the pattern on the reticle to be transferred into the photoresist film on a wafer. Thus, the imaging radiation passes through areas on the reticle not covered by chrome and exposes the photoresist film. Because of the projection optics, the size of the feature on the reticle is typically 4 or 5 times larger than the same feature which is focused on the wafer and printed in the photoresist. A 1 micron by 1 micron wide opening in the chrome on the reticle would translate to a 0.2 micron by 0.2 micron wide area exposed in the photoresist in a 5× reduction system.
Reticles can be bright field or dark field. Bright field reticles typically have a large portion of the quartz not covered by chrome such that up to 45% of more of the incident light passes through the plate. On the other hand, dark field reticles have a large portion of the quartz covered by chrome so that only about 10% of the incident energy is transmitted through the plate. The type of reticle could have an effect on the imaging performance of the photoresist such as the shape of the profile produced. Recent advances in mask making involve the addition of phase shifting materials to the reticle in order to improve the imaging performance of the lithography process.
A photoresist is generally composed of a polymer, photoactive compound, solvent, and may include additives to improve performance or film quality. The photoresist is coated on a substrate such as a wafer and baked to form a film normally between 0.2 and 2 microns thick. For positive tone photoresist, the exposed area becomes soluble in aqueous base developer while the unexposed regions remain insoluble in developer. With negative tone photoresist, the exposed area becomes crosslinked or is otherwise rendered insoluble in aqueous base developer while the unexposed regions dissolve in developer and are washed away. Organic solvents could be used as developers but are not preferred because of flammability and cost concerns.
In the ideal lithography system, the pattern on the reticle is focused by the projection optics of the exposure tool at or near the surface of the photoresist film. In actual practice, the focal point may be above or below the surface of the photoresist because of temperature or pressure drifts, wafer flatness variations or other factors. Since the amount of focus shift or defocus can have a dramatic effect on the size of the printed feature, it is critical to be able to control the process such that the focus is kept within a usable range for each wafer. The usable focus range or depth of focus (DOF) is defined as the range of focus settings wherein the lateral dimension of the printed feature or the space between features lies within a specification which is typically ±10% of a targeted linewidth or CD. The usable process window is a combination of the DOF and exposure range that keeps the printed feature within ±10% of a target CD.
The CD of a printed feature on a wafer is most accurately determined by cleaving the wafer and using a scanning electron microscope (SEM) to observe the feature at an angle with a side view. The SEM instrument is calibrated so that a cursor of known length is superimposed on the printed feature to determine the width of the photoresist profile or the space between two photoresist features. This cross sectional view is valuable because it also provides a picture of the shape of the resist profile that can vary from tapered to vertical or reentrant as depicted in
FIGS. 3A-3C
. However, this method is time consuming and is normally used only for development purposes. It is not compatible with a high throughput production line where a wafer can be exposed in less than a minute and rapid analysis is necessary.
Therefore, improved methods are needed by which feature sizes on wafers can be measured with a minimum amount of disruption and cost to the process flow. An alternative to SEM is atomic force microscopy (AFM) but the cost is high since it must be used off-line and cannot be implemented in mass production. A tilted SEM method to measure sidewall angle of resist patterns and to monitor focus drift has been tested by researchers but is difficult to implement on manufacturing measurement tools. This method has been published in SPIE Proceedings, Metrology, Inspection and Process Control for Microlithography XIV, Vol. 3998, pages 232-238 (2000).
One instrument that is currently widely used in the manufacturing line is a CD-SEM. This technique is non-destructive to the wafer and involves a scanning electron microscope which takes a top-down view of the photoresist pattern. When viewed by this method, the photoresist feature has a darker center portion and a lighter outer portion at the edge of the feature. This top-down measurement of feature size must be correlated to a cross sectional SEM measurement to determine an offset of the top-down size to the actual physical size. The quality of the CD-SEM results depends on the shape of the photoresist profiles. Measurements of vertical profiles can be easily correlated to those from cross sectional views because the outer edge of the

Affiliated with

Also associated with

Rating

In-line focus monitor structure and method using top-down SEM does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with In-line focus monitor structure and method using top-down SEM, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and In-line focus monitor structure and method using top-down SEM will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3340532