Browse Prior Art Database

Automated Recipe Parameter Checking for Semiconductor Processing Equipment. Disclosure Number: IPCOM000004617D
Original Publication Date: 2001-Feb-28
Included in the Prior Art Database: 2001-Feb-28
Document File: 4 page(s) / 29K

Publishing Venue


Related People

David Benoit: AUTHOR


Automated Recipe Parameter Checking for Semiconductor Processing Equipment.

This text was extracted from a Microsoft Word 97 document.
This is the abbreviated version, containing approximately 67% of the total text.

Automated Recipe Parameter Checking for Semiconductor Processing Equipment.

David Benoit


This paper relates to software verification of tool recipes in the semiconductor field. This could easily be applied to other manufacturing equipment in other fields.


Semiconductor production equipment typically run script files that contain the step by step sequence of operations the tool needs to perform on each device produced. These script files, also known as job plans, jobs, or recipes, can have 100's of internal parameters, each with numerous possible settings. While the parameter settings may be within the allowable operational range of the production equipment, there are specific parameter settings that will cause one device produced by the equipment to work properly, while causing a different device produced, to fail to operate properly. Typically this is managed by creating a separate job plan for each part with different parameter settings in each job. Verifying that these internal job parameter settings are correct for each part produced on the tool, is not an error free process, due to the number and complexity of these job plans on each production tool.

There have been many solutions to creating and distributing error-free job plans to the production equipment, such as:

Offline job managers to copy jobs between production equipment over a network.

Offline editors that allow a user to view or edit individual job plans.

Automatic job plan generators to create the individual jobs from a database.

Automatic recipe selection from a host computer system

Each of these methods has advantages and disadvantages.


There is not a method to guarantee a job plan has the correct parameter settings for a given device, each time the job is run on a production tool.


An effective solution would be a runtime compiler built into the software of the production equipment to check the job for errors before executing the job's instructions.

The rules for the compiler would be defined by the user, either in the software itself, or better still, a lookup table similar to the one shown in TABLE 1 that could be read by the compiler.

TABLE 1: Rule look up table to be used by Runtime Job parameter checking software

Check Jobs with following

Parameter name to check

Correct param. Settings

Incorrect param. Settings

Action to take

Name contains "XYZ"



Halt tool Inform user

layer 15



Halt tool

Inform user




Inform user




Halt tool inform user

A job with severe parameter errors would automatically halt the equipment and flag the user with a message indicating the parameter name and incorrect setting.

This runtime compiler method of error checking provides the following advantages:

1. The ability to group jobs. This is a powerful way to guarantee specific parameter settings for a family of parts are set the same way.

2. Each job is checked every time it is run. A job created next week/month/y...