Search for content and authors
 

Dielectric permittivity mappings obtained by GHz region Microwave Microscopy using non-contact probe and their noise reduction of Fourier transform refinements

Hirofumi Kakemoto 1Takakiyo Harigai 1Jianyong Li 1Song-Min Nam 2Satoshi Wada 3Takaaki Tsurumi 1

1. Tokyo Institute of Technology (TIT), Ookayama, Meguro-ku, Tokyo, Tokyo 152-8550, Japan
2. Department of Electronic Materials Engineering, Kwangwoon University, Seoul 139-701, Korea, South
3. University of Yamanashi, Takeda, Yamanashi, Yamanashi 400-8551, Japan

Abstract

Recently, dielectric devices, such as multi - layer ceramic capacitor (MLCC) prepared by BaTiO3 (BT) based material (X7R property) have been miniaturized and used at high frequency. The high frequency dielectric measurement technique for microscopic region is useful for the designing and preparation of these dielectric devices. In experimental, the microwave reflection (r) intensity measurement using scanning non - contact state probe was carried out for measuring the dielectric distribution in the ferroelectric material and device. The dielectric measurement using non-resonant and contact probe was setup with an oscillator inserted a Gunn diode, directional coupler, detector (diode), coaxial tube and probe (6μmφ). The measurement frequency was changed from 9GHz to 11GHz using an oscillator. The r intensity was measured at room temperature as a function of distance between sample and probe. The sample was selected for BT single crystal and MLCC (cross - section). Dielectric mapping image obtained by r intensity minimum value, however image was not clear because of the noise insertion slightly to the image. Fourier transform (FT) analysis was hence carried out for reduce the noise insertion to the image as a function of space frequency. A two dimensional discrete FT was expressed as F(m1,m2)=(1/MN)ΣΣf(k1, k2)exp[-j2π(m1k1/M+m2k2/N)], where m1, m2 are discrete integers for x and y axes, and M and N are also discrete integers, and k1, k2 are space frequencies. From FT image, the noise was selected from raw data, and then inverse Fourier transformation (IFT) was carried out with filtering of the noise. As the result, noise reduction was carried out, and clear dielectric image was reconstructed by FT refinement.

 

Legal notice
  • Legal notice:
 

Related papers

Presentation: Poster at E-MRS Fall Meeting 2007, Symposium J, by Hirofumi Kakemoto
See On-line Journal of E-MRS Fall Meeting 2007

Submitted: 2007-06-21 10:34
Revised:   2009-06-07 00:44