Fig. 1. (a) Sketch of the DC-MOKE setup. HWP1 and HWP2 stand for half-wave plates, and P is the polarizer. The arrows illustrate the polarization state after the optics. (b) Fluctuation of MOKE signal in 1 h after control of the temperature within ±1 mK for the laser tube and polarizing optics in a sealed environment. (c) Statistic analysis of the MOKE noise in (b) using Gaussian distribution.

Fig. 2. (a) MOKE signal (red line) fluctuates along with the laser intensity (blue line) as the laser tube temperature is drifting. The fine spectral feature is the fingerprints of the gain medium. (b) Top, mode structure of a red (632.8 nm) He–Ne laser. The adjacent longitudinal modes, labeled as s-mode (blue line) and p-mode (red line), are orthogonally polarized. Bottom shows that the measured intensity variances of the s-mode (blue) and p-mode (red) are out of phase in a He–Ne laser with cavity length of 25 cm. (c) The fluctuation of laser intensity after temperature control of ±1 mK for the laser tube (inset). (d) Comparison of the polarization noise with (red) and without (blue) the Brewster window.

Fig. 3. (a) and (b) Variation of DC-MOKE signal (red line) when modulating the temperature (blue line) of (a) the polarizer and (b) Wollaston prism, respectively. (c) Comparison of MOKE noise in sealed and unsealed condition after subtracting the drifting background.

Fig. 4. (a) Hysteresis loops at five different positions of a wedge-shaped Ni thin film on SiO₂ substrate. (b) Noise measured at the bare SiO₂ substrate.

Fig. 5. Noise spectrum of our MOKE apparatus measured by an SR830 lock-in amplifier.