In high-energy laser facility, the residual nano-particles that are remained in mechanical system or produced by the
interaction of kinetic-pairs are inevitable. The generation and the propagation of particulate pollutants will seriously
reduce the performance of the laser systems. Therefore, the research about the adsorption behavior of particle
contaminants on fused silica is very important to maintain the optical components’ surface clean, reduce induced
damage, and finally prolong the life of the optical components. In this paper, the adsorption behavior between aluminum
nano-particles and fused silica was simulated by molecular dynamics method. The effect of the surface roughness of
fused silica on the state of adsorption and the state before adsorption has been studied. Then an experiment system based
on an atomic force microscope was established to measure the adsorption force and further to verify the simulated
results. Finally, the adsorption mechanism between metallic nano-particles and fused silica was revealed. The results
show that surface roughness and the size of the particles are two of the main factors to influence the adsorption force.
The rough fused silica surface can be “particle-phobic” due to the decreased contact area, which is beneficial to keep the
fused silica surface clean.
In some industrial fields, the workpiece surface need to meet not only the demand of surface roughness, but the strict requirement of multi-scale frequency domain errors. Ultra-precision machine tool is the most important carrier for the ultra-precision machining of the parts, whose errors is the key factor to influence the multi-scale frequency domain errors of the machined surface. The volumetric error modeling is the important bridge to link the relationship between the machine error and machined surface error. However, the available error modeling method from the previous research is hard to use to analyze the relationship between the dynamic errors of the machine motion components and multi-scale frequency domain errors of the machined surface, which plays the important reference role in the design and accuracy improvement of the ultra-precision machine tool. In this paper, a fourier transform based dynamic error modeling method is presented, which is also on the theoretical basis of rigid body kinematics and homogeneous transformation matrix. A case study is carried out, which shows the proposed method can successfully realize the identical and regular numerical description of the machine dynamic errors and the volumetric errors. The proposed method has strong potential for the prediction of the frequency domain errors on the machined surface, extracting of the information of multi-scale frequency domain errors, and analysis of the relationship between the machine motion components and frequency domain errors of the machined surface.