Removal of Inclusions from Melts

2021 ◽  
pp. 303-364
Author(s):  
Thorvald Abel Engh ◽  
Geoffrey K. Sigworth ◽  
Anne Kvithyld

Inclusion origins and the methods for determining the content of inclusions in a melt are described. Removal of inclusions by flotation/settling is demonstrated. The method for removing inclusions from molten metals by bubbling is described in detail with attachment mechanism to bubbles. Use of microbubbles are included. Filtration capture mechanisms of inclusions, cake and deep bed mode, are derived. A model for removal of inclusions by ceramic foam filters is introduced. Re-entrainment of inclusions are examined. In addition the use of rotational and electromagnetic forces to remove inclusions is explained. The number size distribution of inclusions is taken into account, both the change in the distribution function and the growth of inclusions with time. In the end the interaction of dissolved elements and inclusions is described.

2003 ◽  
Vol 37 (37) ◽  
pp. 5247-5259 ◽  
Author(s):  
Urs Lehmann ◽  
Martin Mohr ◽  
Thomas Schweizer ◽  
Josef Rütter

2011 ◽  
Vol 11 (8) ◽  
pp. 3835-3846 ◽  
Author(s):  
Z. Z. Deng ◽  
C. S. Zhao ◽  
N. Ma ◽  
P. F. Liu ◽  
L. Ran ◽  
...  

Abstract. Size-resolved and bulk activation properties of aerosols were measured at a regional/suburban site in the North China Plain (NCP), which is occasionally heavily polluted by anthropogenic aerosol particles and gases. A Cloud Condensation Nuclei (CCN) closure study is conducted with bulk CCN number concentration (NCCN) and calculated CCN number concentration based on the aerosol number size distribution and size-resolved activation properties. The observed CCN number concentration (NCCN-obs) are higher than those observed in other locations than China, with average NCCN-obs of roughly 2000, 3000, 6000, 10 000 and 13 000 cm−3 at supersaturations of 0.056, 0.083, 0.17, 0.35 and 0.70%, respectively. An inferred critical dry diameter (Dm) is calculated based on the NCCN-obs and aerosol number size distribution assuming homogeneous chemical composition. The inferred cut-off diameters are in the ranges of 190–280, 160–260, 95–180, 65–120 and 50–100 nm at supersaturations of 0.056, 0.083, 0.17, 0.35 and 0.7%, with their mean values 230.1, 198.4, 128.4, 86.4 and 69.2 nm, respectively. Size-resolved activation measurements show that most of the 300 nm particles are activated at the investigated supersaturations, while almost no particles of 30 nm are activated even at the highest supersaturation of 0.72%. The activation ratio increases with increasing supersaturation and particle size. The slopes of the activation curves for ambient aerosols are not as steep as those observed in calibrations with ammonium sulfate suggesting that the observed aerosols is an external mixture of more hygroscopic and hydrophobic particles. The calculated CCN number concentrations (NCCN-calc) based on the size-resolved activation ratio and aerosol number size distribution correlate well with the NCCN-obs, and show an average overestimation of 19%. Sensitivity studies of the CCN closure show that the NCCN at each supersaturation is well predicted with the campaign average of size-resolved activation curves. These results indicate that the aerosol number size distribution is critical in the prediction of possible CCN. The CCN number concentration can be reliably estimated using time-averaged, size-resolved activation efficiencies without accounting for the temporal variations.


2021 ◽  
Vol 1031 ◽  
pp. 58-66
Author(s):  
Vitaly Polosin

For the particle size distribution function various forms of exponential models are used to construct models of the properties of dispersed substance. The most difficult stage of applied research is to determine the shape of the particle distribution model. For the particle size distribution function various forms of exponential models are used to construct models of the properties of dispersed substance. The most difficult stage of applied research is to determine the shape of the particle distribution model. The article proposes a uniform model for setting the interval of information uncertainty of non-symmetric particle size distributions. Based on the analysis of statistical and information uncertainty intervals, new shape coefficients of distribution models are constructed, these are the entropy coefficients for shifted and non shifted distributions of the Amoroso family. Graphics of dependence of entropy coefficients of non-symmetrical distributions show that distributions well-known are distinguish at small of the shapes parameters. Also it is illustrated for parameters of the form more than 2 that it is preferable to use the entropy coefficients for the unshifted distributions.The material contains also information measures for the well-known logarithmic normal distribution which is a limiting case of distribution Amorozo.


Sign in / Sign up

Export Citation Format

Share Document