Hisafumi Iwata

Repair yield simulation with iterative critical area analysis for different types of failure

We propose a general method for repair yield estimation based on critical area analysis using a commercial Monte-Carlo simulator. We classify failures into several types according to the repair rules and use iterative critical area analysis for each type of failure (ICAA-ETF) to calculate the repair yield. Our proposed method makes it possible to accurately estimate within a few hours the repair yield of a memory product. An example of application to an actual SRAM product is discussed to illustrate in detail how our method can be used for critical area calculation and repair yield modeling.

In-line wafer inspection data warehouse for automated defect limited yield analysis

A data warehouse approach for the automation of process zone-by-zone defect limited yield analysis is presented in this paper. The system employs pre-calculation of adder defects extraction and clustered defects recognition, a newly developed wafer-wise defect record structure, and a graphical user interface purpose-designed for data selection navigation. Analysis time can be reduced to less than 1% of that of benchmarked conventional procedures.

Precise Visual Inspection of LSI Wafer Patterns by Local Perturbation Pattern Matching Algorithm

A new defect detection algorithm that compares grayscale images of actual patterns and an automatic visual inspection system implementing this algorithm have been developed. The objective is to detect defects reliably down to 0.3 μm in LSI photoresist patterns on a silicon wafer. To detect defects reliably while remaining uninfluenced by tiny differences between two images of satisfactory patterns, the images are matched in local windows by perturbing one image in the x-y plane and in the brightness direction against the other image. The resulting unmatched regions are recognized as defects. One unique feature of the algorithm is its utilization of polarity changes in the subtracted images during the perturbation for deleting tiny differences between two images. All processing can be done in real time by local, one-pass operators. The developed automatic visual inspection system has achieved a 100 percent detection rate for defects down to 0.3 μm.

Accuracy of yield impact calculation based on kill ratio

We evaluated the accuracy of yield impact calculations based on kill ratio analysis. The accuracy was calculated using computer simulated defect maps and bin maps. The results show that the yield impact was inaccurate when parametric faults caused low yield or a large number of non-killer defects were included in inspection reports. It is therefore recommended to evaluate bin maps and reduce the non-killer defects before calculating the yield impact.

Automatic visual inspection system for thin film magnetic head wafer using optical enhancement and image processing techniques

An automatic visual inspection system has been developed for Thin Film magnetic Head (TFH) wafers. Although there are only a few classes of defects to be detected, the difficulty of defect detection varies drastically depending on the location of the defect. When the optical characteristics of a defect and the underlying element pattern are similar, the defect becomes difficult to detect. To detect all defects reliably, we developed the following new techniques. (1) Optical enhancement: The wafer is illuminated by a slit-shaped light source from an oblique direction, and the scattered light is used for detecting flakes in the transparent protection layer. Reflected light from the surface is also used for detecting surface defects. Defects are easily extracted by thresholding the detected image. (2) Image processing : An element-to-element comparison method is employed to detect defects that cannot be enhanced optically. Many bright spots within the ceramic substrate cause discrepancies when compared. Local minimum processing is used for eliminating these and stabilizing defect detection. The system has been evaluated in an actual production line and the defect detection rate achieved is approximately 13% higher than human performance.

Precise alignment technique for TAB inner-lead bonding

A precise alignment technique for inner-lead bonding of TAB (tape automated bonding), called direct alignment, is described. In this technique, inner leads on a tape and bumps on a LSI (large-scale integration) chip are detected at the bonding station. The position of leads is detected from the projection waveform of a dark-field illuminated image, where complicated background information of LSI patterns is eliminated. The position of bumps, which are partially concealed by the leads, is detected from the projection waveform of a bright-field illuminated image by utilizing the lead position information detected before. This technique has been applied to an automatic TAB inner-lead bonder, and alignment accuracy after bonding of less than +5 mu m is achieved.<<ETX>>