Abstract: Abstract: Ultrasound can promote nucleation at higher temperatures, resulting in larger ice crystals. Thereby the sublimation time is significantly reduced. From all the researches, either ultrasound is applied in ultrasonic water bath or sample itself is liquid form since the low acoustic impedance between liquid and acoustic waves favors ultrasound transfer. However solid samples have barely been taken into account due to the poor transfer efficiency of ultrasound in food matrix. During the freeze drying process, the freezing is usually realized by shelf cooling in freeze-drier. Thus, using liquid as ultrasound applying media is not applicable. The technology currently is for liquid samples or adopts the immersion freezing process, but the plate freezing process for solid samples (fruits/vegetables) was barely taken into account. In this study, contact ultrasound-assisted plate freezing process was achieved. Firstly, the effect of direct contact ultrasound on inducing nucleation of solid samples was observed in the micro environment. Then a microscopic observation was carried out to examine the effectiveness of direct contact ultrasound on inducing nucleation for solid samples at different supercooling degree. The results showed that crystal morphology and nucleation temperature had significant correlation, namely the ice crystal size increased with the nucleation temperature. For example, under the nucleation temperature of -6 ℃, the average ice diameter was 20.20 μm, while the mean diameter increased up to 33.16 μm when the nucleation temperature was raised to -3 ℃. In addition, a phenomenon was found that there existed a postponement of nucleation after the onset of ultrasound application, in other words, nucleation did not follow the ultrasound triggering immediately. Secondly, the influences of ultrasonic parameters (power, exposure time) on the ice crystal size and freeze drying rate of carrot were investigated. The samples thickness was 5 mm. The results showed that applying 178.7 W power ultrasonic for 10 s at -1 ℃ could significantly increase the nucleation temperature, and make the void size of freeze-dried carrot increase from （66.29±3.58） to (80.81±3.03) μm, and the sublimation time was significantly reduced by 29.1% when the residual moisture was 10%. But it used the appropriate ultrasound irradiation condition to avoid the heat accumulation inside the food matrix. In addition, it was further verified that the primary drying rate was significantly reduced since the bigger crystals were formed in the presence of ultrasound. Therefore, the direct contact ultrasound is a promising technology in solid sample freeze-drying process. Finally, the influence of sample thickness on inducing nucleation was investigated. Under the sample thickness of 5 mm in this experiment, there was no significant difference between the void size of upper and lower part of the freeze-dried cake. But if larger thickness was considered, the freezing rate gradient within the sample may result in crystal size difference in axial direction. Smaller ice crystals may form at the region closer to the cooling plate since it was cooled faster than the interior. Furthermore, with the increase of thickness, the region far away from the vibrating plate may not be subjected to enough ultrasound radiation due to the acoustic attenuation within the sample and thus inducing nucleation would not be effective. Considering both inducing nucleation effectiveness and freeze drying efficacy, a maximum product thickness should not be too thick. These studies can provide a reference for the freeze drying process of samples (fruits/vegetables) by contact ultrasonic vibrating plate freezing.