BEAMETR

See frequently asked questions or references.

The performance of any e-beam system greatly depends on the beam size. Frequent monitoring the beam size is necessary. BEAMETR is a technique to automatically measure the electron-beam size. The commonly used "knife edge" technique is time consuming, inaccurate, and operator dependent. BEAMETR allows the beam measurement to be automated. The product includes the software and a piece of silicon with a fabricated nano-pattern, and uses spatial frequency analysis to extract the beam size. BEAMETR on 300 mm wafer is also available.

BEAMETR is independent of equipment vendors; vendors use different definitions of the beam size. We use the Full Width at Half Magnitude (FWHM) definition, which corresponds to 12.5% - 87.5% of the signal amplitude in the knife edge method.

Using the BEAMETR technique, the operator takes an image of the pattern, and the software determines the beam size. While the math involved in this method is fairly complex, the procedure is ultimately simple for end users. The test pattern is specially designed an fabricated, its spatial frequency spectrum is known. The SEM image of the pattern will have a changed spectra depending on the beam used to make the image. The comparison of the designed pattern to the SEM image of the pattern enables the beam size to be extracted.

The central part of the v.1 test pattern is shown in the Figure below. The minimum feature size is 35 nm.

In e-beam systems like defect inspection or electron beam lithography, 300 mm wafer with the pattern can be used; or the pattern can be installed on the stage permanently, while production wafers can go in and out. This makes beam size monitoring available at any time, as needed.

E-beam test pattern on a 300 mm wafer

The analysis uses an advanced mathematical model to examine the signal spectra. The output of the software is the beam size in the x and y direction. The task itself is an ill-posed mathematical problem. Tikhonov regularization is involved to solve the problem. The algorithm also involves

  • Automatic pattern recognition
  • Calibration of magnification
  • Distorted image compatibility
  • Noise and jitter reduction techniques

Examples of images at various beam sizes:

SEM image of the test pattern at small beam sizeSEM image of the test pattern at large beam size

The graphical user interface of BEAMETR software with an image of the test sample (v.30), and the extracted beam size is displayed below.

Electron beam size measurement using BEAMETR

See frequently asked questions or references.