Plasma treatment to improve DUV photoresist process

Details Plasma treatment to improve DUV photoresist process Plasma treatment to improve DUV photoresist process

1999-12-13

2001-06-05

Smith, Matthew (Department: 2825)

Semiconductor device manufacturing: process

Chemical etching

Vapor phase etching

C438S711000, C438S725000, C438S726000

Reexamination Certificate

active

06242361

ABSTRACT:

FIELD OF THE INVENTION
The invention relates to the general field of semiconductor integrated circuit manufacturing with particular reference to photoresist processing for deep UV.
BACKGROUND OF THE INVENTION
The demand for smaller device feature sizes has necessitated the move to smaller exposure wave lengths in photolithography. However, light reflection from a given substrate increases with decreasing wavelength of said light. This degrades the performance of photolithographic tools due to reflective notching and thin film interference effects. Anti-reflection layers (ARLs) have been widely used to overcome these problems through destructive phase cancellation or full absorption so as to minimize formation of standing waves in the resist layers. ARL
12
is placed between photoresist layer
13
and substrate
11
, as shown in FIG.
1
.
Organic ARLs, applied through spin-on techniques, were initially the materials of choice but, as device geometries have continued to shrink into the deep sub micron regime, inorganic materials such as silicon oxynitride, deposited by means of PECVD (plasma enhanced chemical vapor deposition), have proven to be more suitable for the shorter light wavelengths that must, of necessity, be used. The PECVD ARL does, however, have certain problems of its own. In particular, distortion of line profiles in developed photoresist images may occur.
In
FIG. 2
we show an idealized cross-section of a line
23
that was formed through proper processing of photoresist layer
13
(of FIG.
1
).
FIGS. 3 and 4
show common distortions of this line profile that can occur when layer
12
is a ARL of the PECVD type. In
FIG. 3
, a T-top of additional material
34
has grown out from the line near its top surface while in
FIG. 4
additional material
44
is seen to have grown out at the substrate level to provide the line with an added, and unwanted, footing.
It is now known that the origin of these line profile distortions is the presence on the surface of the PECVD ARL of amino (NH
2
) groups that settle there during the PECVD process. While the ARL is being made ready to receive a layer of photoresist there is often a delay. Additionally, it gets exposed to the air and acidic airborne contaminants and/or basic contaminants at the substrate surface react with the amino groups to form the outgrowths
34
and
44
shown in
FIGS. 3 and 4
respectively, leading to a spurious increase in line width within the photoresist pattern.
Various attempts to deal with this problem of photoresist profile instability have been described in the prior art. Among the techniques tried in the past, one of the more popular approaches has been to add an oxide cap over the surface of the ARL immediately following its deposition and prior to exposure to air and later application of the photoresist. Another approach has been to expose the PECVD ARL to a nitrous oxide plasma immediately after its formation. While partially successful, none of these prior art techniques has been able to reliably and completely remove the contaminating amino groups from the surface of an anti-reflective layer.
A routine search of the prior art did not uncover any teachings similar to the present invention. Several references of interest were found, however. For example, Lee (U.S. Pat. No. 5,834,372) teaches a method for pre-treating a semiconductor surface. He purges with an inert gas and then introduces tungsten fluoride gas which is exposed to the surface and improves its nucleation capability for subsequently deposited films.
Yeh et al. (U.S. Pat. No. 5,783,493) subject a BPSG (borophosphosilicate glass) layer to a plasma of argon, nitrous oxide, nitrogen, or oxygen, at a relatively low power (low density plasma), after etch back. This treatment suppresses subsequent precipitation of defects in the BPSG surface.
Yau (U.S. Pat. No. 5,716,890) uses a capping layer of silicon oxide as part of an ILD fabrication process. Said capping layer may, optionally, be exposed to a nitrogen or argon plasma which etches part of the cap away.
Huang (U.S. Pat. No. 5,679,211) uses an oxygen containing plasma as a means to clean the surface of a SOG (spin on glass) in order to remove a polymer residue formed there during a prior etch back processing step.
SUMMARY OF THE INVENTION
It has been an object of the present invention to provide a process for removing all amino groups adsorbed on a given surface.
Another object of the invention has been that said process particularly apply to an anti-reflection layer that is used to underlie a layer of photoresist.
A still further object of the invention has been that patterns formed from photoresist deposited on anti-reflective layers that have been treated according to the process of the present invention, be dimensionally stable.
These objects have been achieved by exposing a layer, particularly an anti-reflection coating that has been deposited by means are PECVD, to a high-density plasma of either argon, oxygen, or a mixture of both. When this is done according to the teachings of the invention, all amino groups are removed from the surface and further processing of the photoresist can then be initiated. Even if delays are introduced during subsequent further photoresist processing, dimensionally stable patterns are still obtained.


REFERENCES:
patent: 5312717 (1994-05-01), Sachdev et al.
patent: 5679211 (1997-10-01), Huang
patent: 5710067 (1998-01-01), Foote et al.
patent: 5716890 (1998-02-01), Yao
patent: 5783493 (1998-07-01), Yeh et al.
patent: 5834372 (1998-11-01), Lee
patent: 5910453 (1999-06-01), Gupta et al.
patent: 5968324 (1999-10-01), Cheung et al.
patent: 6066568 (2000-05-01), Kawai et al.

Affiliated with

Also associated with

Rating

Plasma treatment to improve DUV photoresist 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 Plasma treatment to improve DUV photoresist process, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Plasma treatment to improve DUV photoresist process will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2442424