Wire bonding is an important process in semiconductor manufacturing. The quality of wire bonding is guaranteed by destructive sampling inspection. However, sampling inspection is not able to detect defective products that occur non-systematically. Therefore, it is desirable to shift to inspection of each part. In this paper, we propose an inspection method based on the wire bonding state using a Thin-film AE sensor and transfer learning. We place the thin film AE sensor in the immediate vicinity of the bonding junction and measure the applied AE wave. We improve the quality determination performance in the sampling inspection by determining the quality from the measured AE waves using the transfer learning method we developed. We applied the proposed method to experiment using actual bonding samples and confirmed that it showed a 10.7% higher quality determination ability than sampling inspection.

In order to reduce electromagnetic interference at low cost, we propose a method to estimate the intensity of the electromagnetic emission for each IC. In this method, the intensity is estimated based on the ratio of noise source levels obtained by the temporal change in EM emissions when the switching currents generated in potential noise sources are intentionally modulated. In this paper, we applied the proposed method to a printed circuit board with multiple ICs and estimated the intensity of electromagnetic emission originating from individual ICs. Moreover, the accuracy of the estimation was evaluated by comparison with measurements made when the IC was individually driven. The results indicated that the emission intensity can be estimated with high accuracy when all the emissions caused by each IC are in phase.

Conventional Signal Integrity (SI) simulations, which are used to analyze the quality of the signal of differential lines, employ an analysis method to adjust the relative permittivity and the loss tangent using S_dd21 of the S parameter to the measurement result, and the S_dd21 of the S parameter can confirm the characteristics of the transmission loss to create a differential line model. However, in this technique, the precision of the phase delay time is not sufficient to analyze high-speed differential lines. The analysis result using S_dd21 which considered copper foil surface roughness agreed with the measurement of the transmission-loss and the phase delay time.

Conductive composites typically have low elastic moduli for mitigating stresses that develop during use, allowing a wide range of applications. Such properties have depended on the mechanical properties of the matrix as the filler has a comparatively high elastic modulus. In this research, Ag-coated fillers with low elastic modulus were developed with emphasis on controlling the mechanical properties of the composite while maintaining its conductive properties.

The two types of developed fillers, Ag-coated resin and Ag-coated rubber particles, were found to have elastic moduli less than 1.8 × 10⁹ Pa, considerably lower than the 8.0 × 10¹⁰ Pa of conventional Ag particles. Composite samples of both particles formed with an epoxy resin showed corresponding results; in particular, the sample with Ag-coated rubber particles was found to have 1/5 of the elastic modulus, yet with comparative conductive properties.