Due to its paramagnetic controllable rheological properties, magnetic liquid can meet the requirements of the continuous improvement of the stability of the lubricating oil film of sliding bearings under extreme working conditions, and has good application prospects in bearing lubrication. In order to explore the micro-lubrication mechanism of magnetic liquid, the molecular dynamics simulation method is used to construct and optimize the micro-model of magnetic liquid lubrication at the babbitt alloy interface, and the restricted shear simulation is carried out according to the actual working conditions to study the influence of temperature and shear rate on the lubrication behavior of PAO6-based magnetic liquid at the babbitt alloy interface; by analyzing the changes in parameters such as relative concentration distribution, temperature distribution, velocity distribution, mean square displacement and interface adsorption energy during the sliding process, the mechanism of magnetic liquid micro-lubrication is revealed at the molecular level. The results show that PAO6-based magnetic liquid has good diffusivity and heat dissipation, and can adhere to the friction interface of babbitt alloy to play a good role in bearing and reducing wear; at high temperature and high shear rate, the magnetic liquid lubrication film still shows good stability, and the magnetic particles have good diffusion ability. The research results are helpful to improve the theory of nano-thin film lubrication and have practical guiding significance for the engineering application of magnetic liquid.
Friction and wear are common in mechanical structures. In order to reduce the energy loss caused by friction and the wear and failure of mechanical parts, researchers are committed to improving the lubrication performance between friction interfaces. As the core load-bearing component of mechanical equipment, oil film bearings have the advantages of smooth operation, shock absorption and noise reduction. They are often used in low-speed and heavy-load conditions. Their operating performance and service life directly determine the production efficiency and safety of the equipment. With the rapid development of science and technology and the growing demand for bearing applications under extreme conditions, higher requirements are also put forward for the lubricating film performance of oil film bearings. How to improve its load-bearing capacity and lubrication performance while ensuring the stability of the oil film is one of the keys to solving the “bottleneck” problem of oil film bearings.
Magnetic liquid is a new type of soft magnetic functional material. Compared with traditional lubricating oils, the addition of nano-magnetic particles makes it paramagnetic and controllable under the action of an external magnetic field, which can improve the lubrication and load-bearing performance of the oil film, and is not easy to leak and has no secondary pollution. Therefore, magnetic liquid has become an excellent choice for lubricating oils for oil film bearings under extreme conditions. However, the theoretical system of magnetic liquid nanofilm lubrication is not perfect enough, and the interaction between molecules is difficult to measure through macroscopic experimental equipment, which limits the application scope of magnetic liquid. To solve the above problems, researchers used molecular dynamics (MD) simulation method to study the movement of molecules inside magnetic liquid from the atomic scale, and then explained the performance of magnetic liquid and its microscopic mechanism. Li Qiang et al. used MD simulation method to study the microstructure of magnetic fluid with and without external magnetic field, and analyzed the influence of particle volume fraction, magnetic dipole potential and external magnetic field potential on the microstructure of magnetic fluid. Peng et al. simulated and studied four typical microstructural behaviors of silicone oil-based magnetorheological suspensions under different magnetic fields: (a) phase separation of structure, (b) statistics of chain and cluster size, (c) structural anisotropy of aggregates, and (d) fluctuations of suspensions. Batrudinov et al. improved the frequency-dependent magnetic susceptibility theory of magnetic fluids in static uniform magnetic fields, including the dipole interaction between the constituent particles. Liu et al. used the non-equilibrium MD simulation method to study the relationship between the dynamic macroscopic magnetization intensity and its related microstructural behavior of a dipole soft sphere magnetic fluid system containing small and large particles and a ferrofluid under the combined action of magnetic field and shear flow. ZHAO et al. studied the aggregation behavior and shear characteristics of magnetorheological fluid under different magnetic field strengths, magnetic particle volume fractions and shear rates through coarse-grained MD simulation, and determined the relationship between magnetic particle aggregation and shear stress under magnetic field-shear field at the microscopic level. IVAN OV and CAMP used Brownian dynamics simulation to study the magnetization relaxation dynamics in magnetic fluids and analyzed the influence of the interaction between magnetic particles on the magnetization relaxation dynamics in magnetic fluids. ZAPOM KL and FERFECKI established a model simulation to confirm that the increase of magnetic induction intensity in the bearing gap will lead to an increase in load-bearing capacity, providing a theoretical basis for the design of controllable bearings. Zhao Yaqi simulated the solid-liquid interface lubrication model of magnetic fluid lubricated oil film bearings through MD, and concluded that the friction coefficient of magnetic fluid lubrication decreased by about 50% compared with oil lubrication, and it has better viscosity, temperature and heat dissipation performance. Hou Jinsheng et al. used the MD simulation method to study the effect of magnetic particle content on the solid-liquid interface lubrication performance of magnetic liquids. They found that the increase in magnetic particle content would cause some molecules in the magnetic liquid to adhere to the surface of the solid layer, forming a permeation layer or diffusion layer with strong friction reduction and wear resistance, thereby improving the lubrication performance of the magnetic liquid.
At present, the nano-film lubrication theory of magnetic liquids is still in the exploratory stage. Molecular dynamics simulation can provide information that is difficult or even impossible to obtain in experiments, provide scientific guidance for subsequent experimental research, and reduce experimental costs.
In this paper, the author constructed a microscopic model of PAO6-based Fe3O4 magnetic fluid lubrication at the babbitt alloy interface by molecular dynamics simulation method, simulated the friction process by restrictive shear and analyzed the changes of parameters such as relative concentration distribution, temperature distribution, velocity distribution, mean square displacement and interface adsorption energy in steady state, explored the influence of temperature and shear rate on the lubrication performance of synthetic lubricating oil-based magnetic fluid, and provided theoretical guidance for the preparation and engineering application of synthetic lubricating oil-based magnetic fluid.
1 Construction of magnetic fluid lubrication model at babbitt alloy interface
The microscopic model of magnetic fluid lubrication at babbitt alloy interface was constructed by Materials Studio software, as shown in Figure 1. The model consists of babbitt alloy in the upper and lower layers and Fe3O4 magnetic fluid lubrication film in the middle layer. In order to ensure the rationality of molecular motion, a vacuum layer with a thickness of 5 nm is set.
The friction interface is selected from the babbitt alloy commonly used in bearing alloys. The tin-based babbitt alloy unit cell with the grade ZChSnSb8-4 is constructed by doping the metal tin unit cell. In order to ensure that the final model can contain all meaningful molecular interactions between the magnetic liquid and the babbitt alloy, the thickness of the unit cell needs to be expanded by 10 times, that is, 3.75 nm, when cutting the cell. In order to avoid artificial interactions between molecules and their periodic images during the optimization stage, the friction interface area of the babbitt alloy is increased by expanding the cell. The microscopic model of the friction interface of the babbitt alloy is shown in Figure 2. The model size is 8.26 nm×8.26 nm×3.75 nm.
In order to obtain a higher performance lubricating oil film, the lubrication model proposed by the research team was optimized. The magnetic liquid selected poly α-olefin (PAO) commonly used in synthetic lubricating oil as the base oil, and added Fe3O4 magnetic particles in a certain proportion. Poly α-olefin synthetic oil is a relatively regular long-chain alkane obtained by polymerization of α-olefin (C8 ~ C10) under the action of a catalyst (mainly trimer, tetramer, pentamer), post-processing after catalyst removal, distillation, and hydrogenation. Its structural formula is: See formula (1) in the figure. In the formula: n is 3~5; R is Cm H2m+1 (m is 6~8).
Compared with mineral base oil, poly-α-olefin synthetic oil has better viscosity-temperature performance and low-temperature fluidity, is sensitive to various additives, and has the advantages of long service life and low mechanical wear; while PAO-based magnetic fluid has better friction reduction and anti-wear properties than PAO base oil. Therefore, PAO6 lubricating oil is selected as the base carrier in this paper. It is mainly composed of trimer, tetramer and pentamer of 1-decene (C10H20). Using the Connected Layer tool of Amorphous Cell Calculation, Fe3O4 crystal cells are added in a reasonable proportion to obtain a magnetic liquid lubricating film with a mass fraction of 5.2% and a density of 0.85 g/cm3 (see Figure 3). The model size is 8.23 nm×8.23 nm×4.11 nm.
2 Dynamic simulation of lubrication process
All simulation calculations are performed in the Force module. Due to the complex material composition of the calculation model, the Universal force field is selected for all analysis and calculation, and the calculation accuracy is selected as ultrafine. First, the built model is structurally balanced: the Smart calculation method is used for geometry optimization; the Anneal annealing algorithm obtains the low-energy steady-state configuration of the system according to the system energy change diagram by first increasing the system temperature and then cooling it down, and the model is annealed at a temperature of 300-500 K; the obtained low-energy steady-state structure is subjected to Dynamics dynamics simulation, using the NVT canonical ensemble, the initial velocity is Random, the working temperature is 298 K, and the temperature control function selects Nose-Hoover to perform constant temperature calculation. After the model is balanced, a restricted shear simulation (Confined Shear) is performed. In actual working conditions, the highest working temperature of the oil film is about 65 ℃, so the oil film temperature is set to 298 ~ 338 K, the shear rate is 0. 005 ~ 0. 025 nm/ps, and the simulation time is 70 ps. The lubrication process of the magnetic liquid with Fe3O4 magnetic particles added at the Babbitt alloy interface is shown in Figure 4.
3 Results and discussion
3.1 Relative concentration distribution Relative concentration can be used to analyze the density distribution information of each particle in the box. Its definition is: see formula (2) in the figure. Where: ρr is the relative concentration of A particles at r; ρi is the number density of A particles in the area near r; ρtotal is the total number density of A particles in the box.
Figure 5 (a) shows the relative concentration distribution of magnetic liquid lubrication film at different temperatures at a shear rate of 0.025 nm/ps. The horizontal axis in the figure is the film thickness direction of the magnetic liquid lubrication model. 0-3.75 nm is the lower layer of Babbitt alloy, 3.75-8.18 nm is the magnetic liquid lubrication film, and after 8.18 nm is the upper layer of Babbitt alloy. Figure 5 (b) shows the relative concentration distribution of magnetic liquid film at different shear rates at 338 K along the film thickness direction. It can be seen that the relative concentration of magnetic liquid molecules near the friction interface is higher, indicating that the magnetic liquid adheres to the friction interface of the Babbitt alloy under shear; the relative concentration distribution of magnetic liquid molecules in the central area of the oil film is relatively uniform. As the temperature and shear rate increase, the magnetic liquid molecules inside the lubricating layer still maintain a similar adsorption layer structure. The changes in temperature and speed have no significant effect on the molecular structure of the adsorption layer, indicating that the magnetic liquid lubricating film has good stability and load-bearing capacity.
3.2 Temperature distribution
The temperature distribution of the magnetic liquid film at different temperatures under a shear rate of 0.005 nm/ps is shown in Figure 6. The friction area on both sides of the oil film and the interface of the babbitt alloy produces a local temperature rise, and the closer to the center of the oil film, the lower the temperature. When T = 298 K, the oil film temperature decreases by 6.21% from the highest temperature 355.96 K to the lowest temperature 333.84 K; when T = 338 K, the oil film temperature decreases by 6.42% from the highest temperature 382.74 K to the lowest temperature 358.16 K. This shows that the magnetic fluid has better heat dissipation performance at high temperature and meets the actual application requirements.
3.3 Velocity distribution
Figure 7 (a) shows the influence curve of magnetic fluid at different temperatures on the velocity of magnetic fluid molecules when the shear velocity is 0.025 nm/ps. The velocity of magnetic liquid molecules is 0 between 0 and 3.75 nm and above 8.18 nm, because these regions correspond to the friction interface of Babbitt alloy. After the structure is completely balanced and the atoms are evenly distributed everywhere, it can be seen from the simulation results that the velocity of magnetic liquid molecules in direct contact with the friction interface is consistent with the shear velocity, indicating that the magnetic liquid molecules at the friction interface are stably adsorbed near the friction interface; while the remaining magnetic liquid molecules show solid-liquid interface slip due to the intermolecular van der Waals force. The velocity of magnetic liquid molecules changes from a negative value at the lower friction interface to a positive value at the upper friction interface. The velocity of magnetic liquid molecules is basically 0 at the center of the lubricating film. The velocity of magnetic liquid molecules at the friction interface fluctuates greatly. The molecular velocity in the middle area of the lubricating film shows a large velocity difference, which is due to the opposite direction of the molecular velocity on both sides of the center of the magnetic liquid film. Figure 7 (b) shows the velocity distribution of magnetic liquid molecules at different shear rates at a temperature of 338 K. It can be seen that the moving speed of magnetic liquid molecules increases with the increase of shear speed; the speed of magnetic liquid molecules is stable within a certain range, and the stable speed range is smaller than the shear speed, indicating that the magnetic liquid has good viscosity and can play a good lubrication and load-bearing role.
3.4 Mean square displacement
Molecules are constantly moving and diffusing in the system. Mean square displacement (MSD) is often used to characterize the mobility of molecules. The diffusion coefficient D is 1/6 of the slope of the MSD-t curve, and the definition formula is: see formula (3) and (4) in the figure. Where: Nα is the number of diffusing atoms in the system; r(t) is the position vector of the atom at time t; r(0) is the initial position of the atom.
In actual simulation, the MSD statistics usually produce large fluctuations at the end, so the data with a time interval of 0-40 ps is selected for plotting. Figure 8 shows the curve of the MSD of magnetic particle molecules with temperature at different shear rates. It can be seen that when the shear rate is low, the diffusion coefficients of the five temperatures are not much different; as the shear rate increases, it can be clearly seen that the diffusion coefficient increases, indicating that the diffusion movement ability of magnetic particles increases with the increase of shear rate. It can also be seen from the figure that at the same shear rate, the diffusion coefficient of magnetic particles increases with the increase of temperature.
Figure 9 shows the curve of the MSD of magnetic particle molecules with shear rate at different temperatures. It can be seen that when the temperature is low, the diffusion coefficients of the four temperatures are not much different; as the temperature increases, it can be clearly seen that the diffusion coefficient increases, indicating that the diffusion movement ability of magnetic particles increases with the increase of temperature. It can also be seen from the figure that the diffusion coefficient of magnetic particles increases with the increase of shear rate at the same temperature.
3.5 Interface adsorption energy
The change of adsorption energy of magnetic liquid lubricant at the friction interface of babbitt alloy is explored from the microscopic level through restricted shear simulation to explain the mechanism of magnetic liquid lubricant improving lubrication. When calculating the adsorption energy of adjacent interfaces, it is necessary to remove excess materials from the structure after restricted shear simulation to obtain the molecular dynamics model of the adjacent two layers of materials (see Figure 10). At this time, the structure is already in a balanced and stable state.
The interface adsorption energy is defined as the interaction energy between the interfaces of two layers of materials. The calculation formula of interface adsorption energy is as follows: See formula (5) in the figure. Where: E12 is the interfacial adsorption energy between the upper Babbitt alloy and the magnetic liquid film; E23 is the interfacial adsorption energy between the magnetic liquid film and the lower Babbitt alloy; E1+2 is the total energy of the upper Babbitt alloy and the magnetic liquid film; E2+3 is the total energy of the magnetic liquid film and the lower Babbitt alloy; E1 is the energy of the upper Babbitt alloy; E2 is the energy of the magnetic liquid film; E3 is the energy of the lower Babbitt alloy.
The energy of each layer of material at different shear rates at a temperature of 338 K is calculated respectively, and the interfacial adsorption energy of the magnetic liquid molecules at the Babbitt alloy interface is shown in Figure 11. A negative adsorption energy indicates that the magnetic liquid molecules are combined with the Babbitt alloy interface. The calculation results show that the sum of the adsorption energies of the two interfaces under three different shearing speeds are 14.995 278 43, 14.874 691 44, and 14.882 372 87 kJ/mol, respectively. The sum of the interface adsorption energy is the largest when the shearing speed is the lowest. With the increase of the shearing speed, the movement speed of the molecules also increases, and the kinetic energy of the molecules increases, so the interface adsorption energy decreases accordingly, which conforms to the objective law.
The energy of each layer of materials at different temperatures at a shearing speed of 0.025 nm/ps is calculated respectively, and the interface adsorption energy of the magnetic liquid molecules at the Babbitt alloy interface is shown in Figure 12. The interface adsorption energy E12 between the upper Babbitt alloy and the magnetic liquid decreases with the increase of temperature, and the interface adsorption energy E23 between the magnetic liquid and the lower Babbitt alloy increases with the increase of temperature. The calculation results show that the sum of the adsorption energies of the two interfaces at three different temperatures are 14.712 105 33, 14.723 681 07, and 14.882 372 87 kJ/mol, respectively. The sum of the interface adsorption energies increases with the increase of temperature. The friction interface of the babbitt alloy is isolated by the magnetic liquid lubricating film, thereby reducing friction and wear.
4 Conclusion
Molecular dynamics simulation was used to study the lubrication behavior of PAO6-based magnetic liquid at the interface of babbitt alloy. The main conclusions are as follows:
(1) With the increase of temperature and shear rate, the magnetic liquid molecules inside the lubricating layer still maintain a similar adsorption layer structure, and the magnetic liquid lubricating film shows good stability; the relative concentration of magnetic liquid molecules close to the friction surface is relatively high, and the magnetic liquid adheres to the friction interface under shear, playing a good lubrication and load-bearing role.
(2) The friction area between the oil film and the surface of the babbitt alloy on both sides produces a local temperature rise, and the temperature in the center of the oil film is the lowest. The magnetic fluid has better heat dissipation performance at high temperature.
(3) The diffusion coefficient of magnetic particles increases with the increase of temperature and shear rate. At high temperature and high shear rate, magnetic particles have good diffusion movement ability.
(4) The total interface adsorption energy decreases with the increase of shear rate and increases with the increase of temperature. The friction interface of the babbitt alloy is isolated by the magnetic liquid lubrication film, which plays a role in reducing friction and wear.
This paper makes a preliminary study on the microscopic lubrication behavior of magnetic liquid. In the future, it is necessary to conduct in-depth research by adding a magnetic field and studying the interaction between the microscopic molecules of magnetic liquid through molecular dynamics simulation, so as to fundamentally grasp the essential characteristics of magnetic liquid lubrication and provide scientific theoretical guidance for the preparation and engineering application of magnetic liquid.
