Data path for high performance pattern generator

• US 6,717,097 B1
• Filed: 02/11/2002
• Issued: 04/06/2004
• Est. Priority Date: 09/09/1999
• Status: Expired due to Fees

• Current Assignee: ASML Netherlands BV (ASML Holding NV)
• Original Assignee: Micronic Laser Systems AB (Mycronic AB)
• Inventors: Thurèn, Anders, Sandstrom, Torbjorn
• Primary Examiner(s): Heinrich, Samuel M.
• Application Number:
  US10/049,286
• Time in Patent Office: 785 Days
• Field of Search: 219/121.6, 219/121.61, 219/121.73, 219/121.77, 219/121.78, 219/121.83, 219/121.85, 430/22, 430/30, 430/311, 359/298
• US Class Current: 219/121.6
• CPC Class Codes: G03F 7/70291 Addressable masks, e.g. spatial light modulators [SLMs], digital micro-mirror devices [DMDs] or liquid crystal display [LCD] patterning devices
• CPC Class Codes: G03F 7/704 Scanned exposure beam, e.g. raster-, rotary- and vector scanning mask projection exposure involving relative movement of patterned beam and workpiece during imaging G03F7/70358
• CPC Class Codes: G03F 7/70508 Data handling in all parts of the microlithographic apparatus, e.g. handling pattern data for addressable masks or data transfer to or from different components within the exposure apparatus
• CPC Class Codes: Y10S 707/99937 Sorting
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

A high-speed datapath for a high-performance pattern generator such as an analog SLM for generating the image is disclosed. The data path has provisions for completely independent parallel data flows giving true scalability to arbitrarily high throughput. In a preferred embodiment areas on the SLM are assigned to specific rasterizing and fracturing processors. There is an overlap between fields for blending of the edges in the pattern and for computation of interaction between features in the pattern. The datapath has data integrity checks and a recovery mode when an error condition is raised, allowing it to recover from most errors without creating new errors.

Claims

1. A method for converting with high throughput and precision pattern data in a symbolic input format to a multivalued bitmap and feeding said bitmap to an analog spatial energy beam modulator (SEBM) in a pattern generator,

comprising the steps of:

accepting the input format and fracturing it to fracture fields in at least one fracture processor,

sending fractured data for a fracture field to a rasterization module with at least one rasterizing processor,

rasterizing in said rasterizing processor at least part of the data to a contiguous bitmap corresponding to an area on the SEBM, and

loading said contiguous bitmap into said area of the SEBM.

2. A method as described in claim 1 where during the loading of the bitmap data into the SEBM, a multivalued datum for a pixel is converted to an analog multivalued electromagnetic quantity.

3. A method as described in claim 2 where said electromagnetic quantity is an electric potential.

4. A method as described in claim 1 where the bitmap data is corrected for non-linearities in the response of the SEBM or for pixel-to-pixel variations.

5. A method as described in claim 4 where data for one pixel is corrected by lookup in a prestored lookup table.

6. A method as described in claim 5 where said table lookup is based on data value and pixel location in the SEBM.

7. A method as described in claim 5 where said table lookup is based on pixel value and one of a set of response functions for a pixel corresponding to a variation of at least one physical parameter.

8. A method as described in claim 7 where said physical parameter includes at least one of the following:

electronic offset

electronic gain

mechanical stiffness of a mirror

built-in stress

wavefront flatness

beam pointing

light efficiency

a memory effect.

9. A method as described in claim 4 where data for one pixel is corrected by means of an arithmetic calculation.

10. A method as described in claim 1 where there are at least two rasterizing modules.

11. A method as described in claim 10 where the surface of the SEBM is divided into subfields and one rasterizing module is permanently assigned to a subfield.

12. A method as described in claim 11 where the fracture fields correspond to the subfields of the SEBM.

13. A method according to claim 1, wherein the spatial energy beam modulator (SEBM) is a spatial light modulator (SLM).

22. A method according to claim 1, wherein the spatial energy beam modulator (SEBM) is a spatial light modulator (SLM).

14. A method for converting with high throughput and precision pattern data in a symbolic input format, to a multivalued bitmap and feeding said bitmap to an analog spatial energy beam modulator (SEBM) in a pattern generator,

comprising the steps of:

accepting the input format and fracturing it to fracture fields in at least one fracture processor,

assigning a contiguous area of the SEBM to a contiguous area in the input pattern description,

sending fractured data for said pattern area to a rasterizing module having at least one rasterizing processor,

rasterizing in said rasterizing module at least part of the data to a contiguous bitmap,

and loading said bitmap into said area of the SEBM.

15. A method as described in claim 14 where at least two contiguous areas of the SEBM are assigned to two areas in the input data and sent to two rasterizing processors.

16. A method as described in claim 14 where at least two fracture processors send data to at least two rasterizing processors.

17. A method as described in claim 14 where one fracture processor sends data to at least one rasterizing processor and another fracture processor sends data to at least one rasterizing processor not receiving data from the first fracture processor.

18. A method as described in claim 17 where there is a fixed association between fracture processors and rasterizing processors.

19. A method as described in 14 where one fracture module sends data to at least one rasterizing module containing at least two rasterizing processors.

20. A method as described in 14 where one fracture module containing at least two fracture processors send data to at least one rasterizing module containing at least two rasterizing processors.

21. A method as described in 14 where one fracture module sends data to one rasterizing module and another fracture module sends data to a different rasterizing module.

23. A method for converting with high throughput and precision pattern data in a symbolic input format to a multivalued bitmap and feeding said bitmap to an analog spatial energy beam modulator (SLM) in a pattern generator,

comprising the steps of:

accepting the input format and fracturing it to fracture fields in at least one fracture processor,

tagging the data in the fracture fields with their intended position in the pattern,

sending fractured and tagged fields to at least one rasterizing module,

rasterizing in said rasterizing module at least part of the tagged fracture data to tagged bitmaps,

and loading said bitmaps into areas of the SEBM determined by the tags.

24. A rasterizer for converting with high throughput and precision pattern data in a symbolic input format to a multivalued bitmap and feeding said bitmap to an analog spatial energy beam modulator (SEBM) in a pattern generator, comprising:

an input channel, for input pattern data,

a fracture module for fracturing said input data into fracture fields,

a rasterization module for rasterizing data for the fracture fields into multivalued bitmaps,

a transfer structure, for loading said bitmaps into said SEBM.