Optical: systems and elements – Deflection using a moving element – Using a periodically moving element
Reexamination Certificate
1999-04-27
2002-01-22
Phan, James (Department: 2872)
Optical: systems and elements
Deflection using a moving element
Using a periodically moving element
C359S204200, C359S213100, C359S214100, C359S305000
Reexamination Certificate
active
06341029
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention is directed to optical-radiation patterning devices and in particular to dithering the positions of the light spots with which they write.
Photolithography for small integrated-circuit chips is ordinarily performed by employing a fixed-position light source whose output shines through a mask onto a substrate. The substrate thereby receives a light pattern that selectively causes reactions in light-sensitive chemicals that coat the substrate's surface. Subsequent process steps rely on the resultant chemical patterns to etch the substrate selectively and produce the desired features.
But mask use does not scale well to large patterns, such as those needed for certain types of advanced computer monitors and television screens. For these applications, so-called direct-write patterning apparatus has been proposed. The pattern produced by such apparatus results from moving a light spot about the substrate surface. Typically, laser light is deflected by scanning mirrors to cause the desired light-spot motion.
One of the direct-write approach's difficulties is that it tends to be sensitive to laser-beam variations that cannot always be prevented.
FIG. 1
illustrates the resultant problem. Plot
12
represents the typical Gaussian laser-beam intensity profile, while line
14
represents a threshold intensity below which the required degree of chemical reaction or other effect does not occur. A beam having that intensity profile and scanned in the direction represented by arrow
16
results in a line whose width is w
1
. If w
1
is the desired line width, the intended result is achieved so long as the laser beam's intensity profile remains that of plot
12
. But that profile can change if, say, some loss of focus occurs or there is an unintended change in the laser's power. For example, suppose that an unintended drop in laser power changes the intensity profile to that of plot
18
The resultant line width will then be w
2
, which is not intended.
Designers of direct-write systems have reduced this sensitivity of feature size to beam intensity by taking advantage of the fact that, for a given beam-intensity the change, the absolute size of an edge-position variation is proportional to the spot size. In
FIG. 2
, for instance, the edge-position difference d resulting from a given intensity variation when the spot size is small is less than the edge-position difference d′ (
FIG. 1
) that results when the spot size is larger.
Without more, of course, this smaller spot size also results in lines that are narrower than desired. To obtain the desired line width but keep the lower sensitivity to intensity variation, workers in this field have “dithered” the spot position with respect to the major, scanning motion. That is, they have superimposed on the scan motion a small cyclical motion orthogonal to the scan motion so as to “paint” a wider line. The small cyclical motion must be fast enough in comparison with the scanning motion that half cycles of the dither motion overlap each other and result in an effective light-spot width that is greater than the width of a stationary spot. If dither motion orthogonal to the scanning direction is so superimposed on the scanning motion as to result in a track like that of
FIG. 3
, for instance, the resultant effective intensity profile is that shown in FIG.
4
. As that drawing shows, the line width is the same as in
FIG. 1
, but its edge-position variation d″ is much less.
Although dithering contributes significantly to repeatability, it does exact a certain speed penalty: the scanning speed must be low enough to permit the dither-deflection system actually to perform dithering, i.e., to permit half cycles actually to overlap and thereby result in a continuous feature line.
SUMMARY OF THE INVENTION
We have found two ways of reducing the speed penalty that dithering imposes.
The first way is to employ an acousto-optical device to perform the dither deflection. Whereas the operation of devices previously employed for dither deflection involved gross movement of the deflecting element, the degree of deflection introduced by acousto-optical devices is set by the frequency of the acoustic waves with which they are driven. Since acousto-optical devices thereby avoid the gross motion required of conventional dither-deflection devices, they are capable of much faster deflection, so they can permit the associated scanning motion to be much faster. As will be explained below, moreover, the resultant absence of any gross inertial effects permits the system to exercise greater control over the dither-deflection pattern and thus over the effective intensity profile.
A second way that we have devised for reducing dither-imposed speed penalties is to employ an anamorphic light spot, i.e., a light spot that is wider in a major-axis dimension than in a minor-axis dimension. The advantage afforded by the use of an anamorphic light spot results from the fact that writing speed is also limited by the average power that the light source delivers to the substrate. The smaller spot size used in dithering conventionally tends to reduce the total power that the light source can deliver to the substrate. This is because there typically are process limitations on the amount of power per unit area the substrate can safely absorb. So the light beam's total power in conventional approaches must be reduced as the light spot's size is.
Using an anamorphic spot avoids this problem. Specifically, as the spot width in the dither direction decreases to reduce edge-position variation, the spot width in the direction orthogonal to the dither direction can be increased so as to maintain the desired total power without exceeding process limits on power per unit substrate area.
REFERENCES:
patent: 3848104 (1974-11-01), Locke
patent: 4532402 (1985-07-01), Overbeck
patent: 4775205 (1988-10-01), Muramatsu
patent: 5080474 (1992-01-01), Miyamoto
patent: 5133602 (1992-07-01), Batchelder et al.
patent: 5302547 (1994-04-01), Wojnarowski et al.
patent: 5495279 (1996-02-01), Sandstrom
patent: 5635976 (1997-06-01), Thuren et al.
patent: 5837962 (1998-11-01), Overbeck
“Company A” and “GSI”, “The Alpha Agreement”, pp 1-28, pp A-1-D-1, Nov. 26, 1996.*
“Company B” and “GSI”, “The Beta Agreement”, pp 1-32 Jun. 5, 1997.
Ehrmann Jonathan S.
Fillion Timothy
Savikovsky Arkady
Cesari and McKenna LLP
GSI Lumonics Inc.
Phan James
LandOfFree
Method and apparatus for shaping a laser-beam intensity... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for shaping a laser-beam intensity..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for shaping a laser-beam intensity... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2857181