The main purpose of the present work is to increase the fundamental understanding of the particle transport behavior in an enclosed environment and to provide knowledge to the estimate and measure the particle emission from pellets during a steel production process.
A laboratory study focused on the effect of the high sliding velocity on the particle generation from dry sliding wheel-rail contacts has been conducted. The particle concentration and the size distribution were acquired online by using particle number counters during the tests. After the completion of each test, the characteristics of pin worn surfaces and collected particles were analyzed with the aid of SEM (scanning electron microscopy) combined with EDS (energy disperse X-ray analysis). The results show that the amount of the particle generation increases significantly as the sliding velocity increases from 0.1 to 3.4 m/s. Moreover, the particle size distribution results indicate that the majority of the generated particles are submicron (ultrafine and fine) particles in the case of a high sliding velocity (1.2 and 3.4 m/s). The observations of iron oxide layers within the pin worn surface and the collected iron-oxide containing particles reveal that these substantial small particles can be attributed to an oxidative wear between the dry sliding wheel-rail contacts under high sliding velocities.
The effect of the particle transport behavior with respect to submicron particles in the test chamber on the measurements taken at the outlet was studied by a three dimensional mathematical model. With the assistance of CFD (computational fluid dynamics) simulations, the airflow pattern was found to have a major effect on the particle transport during the tests. By estimating the particle loss rate, 30% of generated particles failed to be captured at the outlet. The reason for that could be a temporary suspension and a deposition onto the surfaces. It should be noted that the particles were assumed to follow the air stream as a result of the small particle size. In addition, the Lagrangian tracking results reveal that the limiting size for particles to become airborne during tests is around 10 µm. However, the computational cost is found to be significant high when the Lagrangian method is adopted.
To consider the measurements of micron particles and to reduce the computational time, a coupled drift flux and Eulerian deposition model was developed. In this model, the effects of the gravitational sedimentation and deposition on the particle dispersion were included. The simulation results are in a good agreement with the available experimental data. The value of APD (average percentage deviation) is in the range of 7.7% to 21.2%. Therefore, a set of simulation cases have been carried out to investigate the influential factors (particle size, wall roughness, source location and duration). The results show that the homogeneity of the particle concentration distribution in the model room declines with an increased particle size (0.01 to 10 µm). An almost uniform particle concentration field is formed for submicron particles (0.01 and 0.1 µm) and for fine particles (1 and 2 µm). However, a clear concentration gradient is obtained for coarse particles (4, 6, 8 and 10 µm). This is due to that the gravitational settling dominates the motion of coarse particles. As a result, a large deposited amount and a high deposition fraction was predicted for coarse particles. Moreover, the surface roughness was found to enhance the deposition of submicron particles (0.1 and 0.01 µm) for a given friction velocity. On the contrary, the deposition of micron particles is much less sensitive to the variation of the surface roughness. For a case of an internal source in the room, where a release over a long duration is considered, the particle dispersion strongly depends on the release location. However, this is not the case for a short release time.
The dispersions and depositions of micron particles were explored in a laboratory test focused on the particle emission from the wear between the pellets. The simulation results were compared to the measured data with respect to the particle flux at the outlet. A good agreement (4.92% < APD < 12.02%) is obtained. In addition, the influence of the air flow rate at the inlet and the particle size on the sampling results at the outlet was investigated carefully. The results show that a stronger air supply at the inlet can push more particles to the outlet for any given particle sizes. However, the resulted increase of the measurable fraction is more significant for 4, 6, 8 10 µm particles compared to 1, 2 and 20 µm particles. Moreover, it is apparent that 20 µm particles are unable to be measured in such a measurement system.
KTH Royal Institute of Technology, 2016. , 45 p.