Computer-aided design and analysis of circuits and semiconductor – Nanotechnology related integrated circuit design
Reexamination Certificate
2001-10-16
2003-04-01
Siek, Vuthe (Department: 2825)
Computer-aided design and analysis of circuits and semiconductor
Nanotechnology related integrated circuit design
C716S030000, C716S030000, C716S030000
Reexamination Certificate
active
06543045
ABSTRACT:
BACKGROUND OF THE INVENTION
FIELD OF THE INVENTION
The invention, in general, relates to the field of the manufacturing integrated semiconductor circuits such as VLSI and ULSI circuits using photolithographic methods. In particular, the invention relates to increasing the resolution of conventional photolithography by using alternating phase masks.
When integrated semiconductor circuits are manufactured, the mask structures that are assigned to the circuit elements are conventionally optically imaged on light-sensitive layers on the wafer. Because of reflective effects, the resolution of such an imaging system is limited and mask structures with dimensions below the reciprocal value of this resolution, also referred to as critical structures, are imaged in a smeared fashion or are out of focus. This leads to undesired strong correlations between the circuit elements and this causes the functionality of the circuits to be impaired.
These difficulties can be overcome by utilizing the destructive interference effect of two directly adjacent and coherent light beams with phases which are shifted through 180°, and by converting the conventional masks in question into alternating phase masks in which each critical structure is provided with two phase shifters for generating the necessary phase shift.
Various types of phase masks are described, for example, in the book “Technologie hochintegrierter Schaltungen” [Technology of highly integrated circuits] by D. Widmann, H. Mader and H. Friedrich, 2nd edition, Springer-Verlag [publishing house], pages 135 et seq. A detailed summary of phase mask technology is given in the publications “Improving Resolution in Photolithography with a Phase-Shifting Mask” by M. D. Levenson et al. in IEEE Trans. Electron. Devices 29 (1982), pages 1828 et seq. and “Wavefront Engineering for Photolithography” by M. D. Levenson in Physics Today, July 1993, pages 28 et seq.
The use of what are referred to as strong phase masks, which include both the alternating phase masks already mentioned and chromium-free phase masks, requires the transparent phase-shifting structures in each affected plane to be assigned to one of two phases which have a phase difference &Dgr;&phgr;=180°. Here, the following two cases must be distinguished: in what is referred to as a dark-field phase mask, transparent structures correspond to the circuit elements (for example conductor tracks) and phases can be assigned to them, while non-transparent mask fields are formed by regions covered with chromium. In contrast, in what is referred to as a bright-field phase mask, the non-transparent regions of the phase mask, which are covered with chromium and which constitute the circuit elements, and the regions lying between them are transparent. In the latter case, suitable regions in the vicinity of the non-transparent chromium regions must be defined as phase-shifting elements. The phase-shifting elements are produced in accordance with specific design rules, known per se in the prior art. The method of production is described, for example, in U.S. Pat. No. 5,537,648, which is incorporated herein by reference.
In view of the complexity of modern circuits and the requirement for two phase-shifting elements which are displaced by 180° on each critical structure, it is, however, conceivable that there will be phase conflicts. A phase conflict occurs precisely if the same phase is incorrectly assigned to the phase shifters on both sides of a critical structure or if, because of the interaction of the phase-shifting elements, the destructive interference effect occurs at an undesired point on the aforementioned light-sensitive layer. The phase assignment for the various phase-shifting elements thus constitutes a mathematical combinatorial problem which cannot be generally solved. Because the phase assignment can in principle lead to different results and different phase assignments can take place for one and the same cell of a hierarchical layout, the phase assignment must be performed last on the finished circuit layout in an automated program. For this reason, there is a need for an automated checking routine that examines a circuit layout to determine whether phase assignment is at all possible. This checking should be complete and should restrict the problem as satisfactorily as possible, i.e. should determine its actual place of origin. The latter is not self-evident; it is due to the fact that if the combinatorial function “does not go”, this is possible in various ways and the point at which it is discovered that this is the case can lie far from the actual place of origin.
After phase conflicts have been determined in an automated routine, they can be resolved in basically two different ways. First, the circuit design can be slightly changed at the points of localized phase conflicts, for example, by shifting conductor track structures so that the phase conflicts are eliminated. On the basis of this changed circuit design, a successful phase assignment can then be carried out in order to generate a phase mask. Second, the circuit design can remain unchanged and instead the phase conflicts can be resolved by assigning two different phases to individual phase-shifted elements. However, the result of this is that a dark line occurs in the exposure on the boundary between the two different phase regions, which would lead to an interruption. For this reason, in this case an additional exposure step is carried out with what is referred to as a trim mask which is used for specially exposing the dark lines which occur.
In the prior art, two different methods are known for checking a layout for phase conflicts. The publication “Heuristic Method for Phase-Conflict Minimization in Automatic Phase-Shift Mask Design” by A. Moniwa et al. in Jpn. J. Appl. Phys., Vol. 34 (1995), pp. 6584-6589, discloses an access method based on graph theory in which a set of phase-shifting elements are postulated and a planar, non-directed graph is formed from this set taking into account the technological requirements. In this method based on graph theory, graph nodes (vertices) constitute phase-shifting elements. A graph edge between two nodes means that the region between the associated phase shifters is lithographically critical. In this method phase conflicts are manifested as those cycles with an uneven number of vertices. Because of the significance of the graph edges, a break in a cycle, i.e. resolution of a phase conflict is equivalent to expanding the corresponding critical region. An efficient conflict resolving strategy in accordance with the aforesaid method should be to break the edges occurring most frequently in the uneven cycles.
U.S. Pat. No. 5,923,566 describes a computer-implemented routine which is used to verify whether an existing circuit design can be imaged onto a phase mask or whether localized phase conflicts are present. The phase conflicts are registered from the interaction between critical circuit regions and the coherent free circuit regions which are to be determined taking into account the technological requirements. Free circuit regions with an uneven number of interactions represent the phase conflicts.
Both methods described above do not, however, operate in an optimum way in registering phase conflicts. First, as is explained below with reference to examples, these two methods prove inefficient because, for example, certain phase conflicts are indicated in duplicate. Second, they prove inadequate because they cannot be used to register other phase conflicts.
Therefore, the phase conflicts cannot be correctly registered by means of the identification methods which are known from the prior art. Consequently, even a conflict elimination method which uses the results of the identification method to eliminate the identified phase conflicts cannot give optimum results.
SUMMARY OF THE INVENTION
It is accordingly an object of the invention to provide a method for detecting possible phase conflicts on alternating phase masks and for automatically eliminating the
Ludwig Burkhard
Moukara Molela
Infineon - Technologies AG
Siek Vuthe
Tat Binh
LandOfFree
Method for detecting and automatically eliminating phase... 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 for detecting and automatically eliminating phase..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for detecting and automatically eliminating phase... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3065517