Manufacturing adjustment during article fabrication

• US 5,283,746 A
• Filed: 02/25/1993
• Issued: 02/01/1994
• Est. Priority Date: 02/22/1990
• Status: Expired due to Term

• Current Assignee: NEC Corporation
• Original Assignee: AT&T, Inc.
• Inventors: Rietman, Edward A., Cummings, Kevin D., Frye, Robert C.
• Primary Examiner(s): RUGGIERO, JOSEPH F
• Application Number:
  US08/023,158
• Time in Patent Office: 341 Days
• Field of Search: 364/468, 364/488-491, 364/148, 364/800, 364/807, 364/600, 364/602, 364/607, 395/903, 395/904, 395/21-27, 395/921, 382/14, 382/15, 250/492.1-492.3
• US Class Current: 700/104
• CPC Class Codes: G06N 3/063 using electronic means
• CPC Class Codes: G06N 3/065 Analogue means
• CPC Class Codes: H01J 2237/30455 Correction during exposure
• CPC Class Codes: H01J 2237/31769 Proximity effect correction
• CPC Class Codes: H01J 37/3026 Patterning strategy
• CPC Class Codes: Y10S 706/904 Manufacturing or machine, e.g. agricultural machinery, machine tool

Abstract

The use of neural networks has been employed to adjust processing during the fabrication of articles. For example, in the production of photolithographic masks by electron beam irradiation of a mask blank in a desired pattern, electrons scattered from the mask substrate cause distortion of the pattern. Adjustment for such scattering is possible during the manufacturing process by employing an adjustment function determined by a neural network whose parameters are established relative to a prototypical mask pattern.

Claims

1. A process for manufacturing an article having desired specifications, said process being complex and comprising the steps of causing a neural network, having parameters determined through a prototypical article so that a solution derived from data representing a variable for said prototypical article yields in said process a resulting article corresponding to said prototypical article that essentially meets said desired specifications, to be applied to data representing a variable of said process, generating a control signal corresponding to the solution derived from said data by said neural network to effect the fabrication of said article, and adjusting said variable in conformance with said signal without adjusting said parameters based on said signal.

2. The process of claim 1 wherein said article comprises an integrated circuit.

3. The process of claim 1 wherein said article comprises a lithographic mask.

4. The process of claim 3 including the step of producing a pattern in an electron sensitive material by selectively exposing said material to electrons.

5. The process of claim 4 wherein said variable comprises the dosage of said electrons produced by control of an electron beam.

6. The process of claim 5 wherein said neural network includes active devices that perform a linear operation.

7. In a process characterized by one or more variables in which a directable energy is directed at a plurality of target locations wherein said energy impacting at one of said plurality of target locations in a body produces an effect at such target locations and also produces an effect at one or more locations in proximity to said target location, the subprocess for achieving the desired effect at each of a plurality of said locations comprising the steps of deriving a solution by applying a representation of said process to a neural network having parameters determined through a prototypical sample without adjustment of said parameters based on said solution, and adjusting for each of said target locations one or more of said variables of said directable energy in accordance with the effect created at said proximate locations.