Significantly Boosts Automated AFM Throughput in Automatic Defect Review

Table of Contents

Why Automatic Defect Review (ADR) AFM?
How to Transfer Defect Map from Inspection Tool to Review AFM
Automated, Accurate Transfer of Defect Map WITHOUT Reference Marks
Conclusion: Results of 300mm Bare Wafers ADR by Park NX-Wafer


In semiconductor design the scaling trend requires tighter control over defects on wafers. To characterize these defects both review and inspection tools are required.

Inspection tools, such as surface scanning inspection systems (SSIS), are used to detect and map the sites of defects, while the review tools, such scanning electron microscopy (SEM) are used to acquire morphological information.

Large defects are visible with optical microscopy, and they can be easily detected by an inspection tool and imaged by a review tool. However, as processing conditions become increasingly stringent, the sizes of defects of interest become smaller - below the diffraction limit of optical microscopy.

In order to locate and identify small defects, the review tool should perform a large survey scan to identify the accurate location of the defects.

Defect review SEM is used to image bare wafer defects after mapping the defect sites with the laser-scattering defect inspection tools, such as KLATencor’s Surfscan. However, the images captured by defect review SEM are limited to 2D and fail to provide any 3D information, which is more important for bare wafer manufacturing control. Smaller defects of less than 50 nm are also not easily detected by the SEM-based defect review system.

Figure 1. The goal of defect review is to locate the defect site identified in a survey by an inspection tool and image the defect using a review tool

Why Automatic Defect Review (ADR) AFM?

Atomic force microscopy (AFM) is gaining much more importance as a review tool as it can directly determine the defect dimensions such as height and width as well as the physical properties of the defects of interest. However, using traditional manual AFM to identify nanoscale defects is time consuming, and has a negative impact on both productivity and throughput.

The use of manual AFM can be quite challenging for researchers and engineers working with bare wafers. Therefore, an automatic AFM solution will allow failure analysis and production engineers to identify the cause of defect areas on bare wafers and eventually reduce them.

Figure 2. The typical routine of defect review by manual AFM has a very low throughput: 10 defects per day at best. Moreover, the tip cost can run high if one uses a destructive scan mode such as tapping.

In order to address this problem Park Systems developed the Automatic Defect Review (ADR) AFM, which accelerates and improves the way defects are imaged and analyzed. Using this solution, users can now acquire additional details and height and depth information of the defects, which were otherwise too expensive or impossible to achieve with SEM-based ADR.

How to Transfer Defect Map from Inspection Tool to Review AFM

Although the general defect review process is simple, it becomes difficult when performed on the nanometer scale. It is simple because all it requires is the proper translation of the defect coordinates on a map from an inspection tool to a map for use on a review tool.

The process becomes complicated when attempts are made to go near the defect sites to image them, and to do this in automation. This is because stage mapping errors occur between the work performed with the review tool and the work performed with the inspection tool.

For a patterned wafer, alignment marks are used as reference points to map the two stages between the inspection tool and AFM, and the position error from both tools can thus be accurately predicted.

Park’s initial success in the Automated Defect Review (ADR) AFM solution was adopted in the hard disk (HDD) industry, using the reference marks inscribed by optical inspection tools like KLA-Tencor’s Candela series.

The defect maps of substrates or hard disk media can be transferred reliably and accurately using these reference marks, and automated review AFMs, such as the Park HDM Series, was able to reach close enough to the defect to perform a survey scan and then tollow up with a zoom-in scan to provide the details the user requires.

As a whole, this solution has been proven to be effective, increasing the productivity by up to 1000% in the HDD industry’s FA labs.

Automated, Accurate Transfer of Defect Map Without Reference Marks

Achievement in the HDD industry has resulted in the semiconductor industry sending similar requests. For the semiconductor industry, the required cleanliness level is much stricter compared to HDD, and so it is not possible to create markers on the bare wafers. Therefore, there is a requirement for a bare wafer ADR process without using any reference marking.

The positions of defects are roughly estimated [1, 2] without alignment marks. There is a need for an additional alignment process to reduce the positional error [3] during the transfer of the defect map. In order to handle this issue, a new process was developed that uses the notch, wafer edge, and large defects that are observed by optical microscopy as the reference points.

Combined with improved vision, the enhanced remapping method does not require any reference markers on a bare wafer (Figure 3). It also does not need separate steps to calibrate the stage of the targeted defect inspection system.

Figure 3. Wafer edge and actual defects are used to orient the wafer and translate the coordinates from the inspection tool to the Park AFM.

The success rate of the 300 mm bare wafer ADR depends on two factors - the size of initial survey scan and the accuracy of the stage mapping. With more accurate stage mapping, a smaller survey scan size can be used to increase the chances of accurately identifying the defect. The new stage mapping method greatly enhances the accuracy by adopting an enhanced vision technique and an advanced remapping algorithm.

In the new mapping method, the RMS position error of survey scans are below 5 µm regardless of wafer loading positions. Using the defect map from a defect inspection tool, all defects are located within ±5 µm.

Figure 4. The positional error of actual defect sites after the automatic stage mapping is quite small, about 5µm in RMS. The positional error remains largely the same regardless of wafer loading positions.

Conclusion: Results of 300 mm Bare Wafers ADR by Park NX-Wafer

Using the advanced coordinate translation technique, the defect map acquired from a laser-scattering defect inspection tool, such as KLA-Tencor’s Surfscan, can be effectively transferred to a 300 mm AFM system, enabling high throughput, fully automatic defect imaging.

The new 300 mm bare wafer ADR enables a typical automatic measurement to run as follows:

  • A laser-scattering defect inspection tool (e.g. Surfscan) is used to run a bare wafer
  • An operator registers the defect coordinates file, called KLARF for the Surfscan tool, to Park NX-Wafer
  • Defect map coordinates are de-skewed automatically and the connection of the two stages are enabled between the Park NX-Wafer and inspection tool
  • The Park NX-Wafer runs the ADR on the sample bare wafer

The above data illustrates the results obtained from the 300 mm bare wafer ADR using Park NX-Wafer. The typical success rate of this method is above 95%, irrespective of the orientation of a wafer.

The zoomed-in scans clearly differentiate pit, bump, and scratch, and shows the detailed morphology of defects.

Figure 5. The 300 mm bare wafer ADR AFM correctly locates all the defect sites identified from an inspection tool and automatically zooms in for the detailed imaging.

Figure 5 (a)

Figure 5 (b)

Figure 5 (c)

Figure 6. The AFM imaging shows the residual damage left during the SEM imaging.

This information has been sourced, reviewed and adapted from materials provided by Park Systems Inc.

For more information on this source, please visit Park Systems Inc.

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback