Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Photomask Repair Verification Method Using Image Analysis Software

IP.com Disclosure Number: IPCOM000188020D
Original Publication Date: 2009-Sep-18
Included in the Prior Art Database: 2009-Sep-18
Document File: 6 page(s) / 320K

Publishing Venue

IBM

Abstract

Disclosed is a novel procedure for using an image analysis software package in the analysis of critical dimensions and optical properties of a feature on a photomask. The unique features of this method allow the capturing of metrology data and measurements analogous to those generated by Aerial Image Measurement System (AIMS), but without the requirement of a dedicated, stand-alone AIMS tool. Images can be processed and data can be analyzed at a photomask repair system itself, thereby eliminating the need for a separate AIMS tool. The unique features of this technique include: 1) A DUV image of the photomask region to be measured is generated at the repair tool and adjusted to create a wafer simulation, 2) Critical dimensions and optical transmission are measured by analyzing the wafer simulation image using an image analysis software package, 3) Illumination conditions at the repair tool are adjusted to closely mimic those of the wafer stepper used in the photolithographic process, and 4) Contour images, intensity plots and other data are exported to Excel or other similar applications for further processing. In particular, these features enable quick and reliable defect repair verification without the need for a separate AIMS tool, thereby reducing cycle time and increasing throughput.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 53% of the total text.

Page 1 of 6

Photomask Repair Verification Method Using Image Analysis Software

Description of the Invention:

As semiconductor technology advances,

photomask features continue to decrease in size

rapidly. As a result, defect repair has become an integral part of the mask manufacturing process. The quantitative analysis of a defect, which is the verification that the error between the critical dimensions and optical properties of the defective feature and a reference does not exceed a desired specification, is crucial to maintaining repair quality control. We require a reliable method which can accurately predict the effect of a defect on the wafer. Figure 1 is an example of a current industry method, and typical analysis of a repaired defect site. This method employs the Aerial Image Measurement System (AIMS) tools and software package, which emulates the wafer stepper, creating a simulation of how photomask features will print on the wafer. Additionally, a simulated wafer print image, or contour plot, can be generated (Figure 2). This is a prediction of what the photomask features will actually look like on the wafer using the chosen illumination conditions. These plots are used extensively in the qualitative analysis of the criticality of a photomask defect.

    The AIMS process presents several drawbacks, which are improved upon by the method described. A primary disadvantage of AIMS is the fact that the throughput can be low, when measured as the number of sites able to be simulated per hour. The operator is required to load a mask to a stand-alone AIMS tool, navigate to a defect, and then run the AIMS simulation. There are two separate tools required to perform defect repair and repair verification: 1) a repair system (femtosecond pulse laser, FIM, e-beam, etc.), and 2) an AIMS tool. Clearly, if we can combine the repair and verification processes into one system, we can drastically reduce cycle time and the cost associated with the repair process.

The invention outlined in this disclosure involves the use of an image analysis software

package to analyze the effects of photomask defects on wafer printability. For the purposes of

these experiments, the program ImageJ was used; however, the technique applies to any image analysis software. This disclosure will describe the use of ImageJ to analyze repairs performed on a femtosecond pulse laser repair system (Rave FP650); other repair systems are a topic for further study.

    Figure 3 is an image obtained after performing a repair and initiating the ImageJ software. The image shows the defective region of the photomask, as seen using the 248 nm wavelength DUV optics available on the FP650. The ImageJ program accepts a multitude of image types; the user simply needs to save the DUV image as an...