Introduction
Calcium aluminate cement/hydratable alumina‐bonded castables still pose a serious challenge with regard to removing the physical and chemically bound water components
from the porous network prior to their application in steel ladles. Due to the high risk of generated steam pressures destroying the material at temperatures above 100C, castables must be predried under conservative heating schedules below 1C/min to ensure safe evacuation of the moisture. In particular, the dangerous ebullition stage and hydrate phase decomposition stage pose the greatest concern during heatup. Therefore, a quantitative understanding of the transport physics in the range of first‐drying, that is from 100C to 300C, will help guide industrial producers and refractor

    Results
As an example material, RC is used to discuss the one‐dimensional, one‐sided moisture loss throughout a first‐drying experiment. Due to the relatively short sample length
(74 mm), the timescale of an experiment is around 1.5 hours. In Figure 1B, the measured temperature profiles are shown, which demonstrate the standard techniques of evaluating moisture loss in a first‐drying experiment. As can be seen, once the heat flux is applied, the temperature at the surface begins to rise and a temperature gradient develops. The sample average heating rate is 1.5C/minute. The maximum surface temperature that reached the top of the sample is around 180C‐200C. As soon as the temperature at the surface reaches values on the order of 110C, a decrease in both the heating rate and temperature gradient is observed. This indicates that around 110C a drying front should develop, however, no direct information can be taken from this temperature data.



Fig 1 The measured saturation and temperature profiles for regular castables (RC) cured for 48 hours.
The black markers indicate the position of the drying front as a function of time. The time in between each profile
for RC is 3 minutes. The dashed lines in the figures indicate thecritical moisture content.
 The solid line at the bottom of the saturation profile figures corresponds to the residual moisture background level.


In Figure 1A, the corresponding saturation profiles are shown demonstrating what additional information can be measured by NMR. Here we have plotted the saturation
profiles, that is, each profile is obtained by dividing by a reference taken at the beginning of the experiment at room temperature, corresponding to a uniform saturation of 1. This allows for sample inhomogeneities to be eliminated from the measured quantitative moisture content. Depending on the number of averages, measuring each profile for RC takes on average 1 minute. The time‐step in between subsequent measurements is 3 minutes.


    Discussion
However, although differences can be observed from one material to another, it is difficult to compare all the drying experiments for the various materials. But, as shown in Figure 2, the drying behavior can be characterized by considering both the front positions and the front temperatures as a function of time. As can be seen, there are some common features displayed by the drying behavior of most materials. Firstly, the drying front speed is approximately linear, regardless of material choice. The slope of this front is related to the high‐temperature vapor permeability, as can be shown according to the simple model.



Fig 2: The boiling front position as determined from the measured moisture profiles as a function of time for all materials measured


     


Ahmed J. Barakat, Leo Pel, Olaf Krause, Olaf C. G. Adan, Direct observation of the moisture distribution in CAC- and HA-bonded castables during first-drying: an NMR study in J Am Ceram Soc. 113. 2019


A. J. Barakat , L. Pel, O. C. G. Adan, One‑Dimensional NMR Imaging of  high‑Temperature First‑Drying in Monolithics, Appl Magn Reson 49,739-753 (2018)