We are measuring GV response of MOSCAP by B1500 Semiconductor Device Analyzer. We notice that at high frequency we see negative conductance behavior. Can anyone help how to solve this problem or understand this negative conductance behavior.?
Techniques used for measuring signals with sinusoidal modulation are highly sensitive to distortions in the sinusoidal waveform. These distortions are caused by the capture and release of electrons from near-interface traps, which have time constants in the order of the signal period. The distortion can result in physically impossible readings, such as negative conductance (particularly for higher frequencies). You can go through the following paper for better understanding of the results you are observing:
M. Chaturvedi, S. Dimitrijev, H. A. Moghadam, D. Haasmann, P. Pande and U. Jadli, "Fast Near-Interface Traps in 4H-SiC MOS Capacitors Measured by an Integrated-Charge Method," IEEE Access, vol. 9, pp. 109745-109753, 2021, doi: 10.1109/ACCESS.2021.3102614.
Can it be that you have too much of parasitic inductance in the loop and it acts as series LC above resonance? Note that one may expect about 0.9 nH inductance per mm of loose wire.