Mercury intrusion/retraction

The isotherms of N2 adsorption on MCM-41 and two samples of SBA-15 synthesized at different temperatures are reported in Fig. 1. The curves of mercury intrusion-retraction... 

Figure 2. Fligh-pressure part of the curves of intrusion-retraction of mercury on (squares) SBA-15 synthesized at 403 K, (lozenges) SBA-15 synthesized at 343 K, (triangles) MCM-41. Void symbols first cycle. Filled symbols second cycle. The curves have been shifted along the y axis.

Two successive cycles of mercury intrusion in SBA-15 samples show an excellent reproducibility of the intrusion and retraction pressure. A limited decrease of the... 

Simulations of mercury intrusion and retraction, according to the mechanisms described above, were performed on 2D structural models constructed from spin density and Tj-weighted images of central slices through pellets taken from batch G2. An example of a typical set of data... 

Figure 5 shows the mercury intrusion and retraction curves, analysed using Bqs. (4) and (S), respectively, for a sample of pellets from batch G2. It can be seer that at smaller pore sizes, widiin the error in eqs. (4) and (S), the intrusion and extrusion curves overlay each otoer, while the point of separation of the intrusion and retraction curves occurs at a fractional occupied volume of 0.67, and that die final level of mercury entrapment is 40 %. Table 1 shows the mmeuty entrapment levels for several samples taken from bmch G2. From Table 1, it can be seen that the experiments are rep tabl and toe value of entr ped mercury agre well with that predicted from toe models derived from MR images above. 

It has been found that models for the structure of sol-gel silica spheres, constructed foom spin density and spin-spin relaxation time images, give rise to good predictions for tiie point of separation of mercury intrusion and retraction curves, and the level of mercuiy entrapment, found for experimental porosimetry data for the same material. This finding suggests that the MR images contain sufficient information to determine the level of mercury entrapment. Hence, this result supports the view that mercury intrusion and retraction within this material is determined by macroscopic (0.01-1 mm) heterogeneities in the spatial distribution of porosity and pore size. 

The experimental technique is straightforward but there are many difficulties in interpreting the results of mercury intrusion. The contact angle 0 is usually not known a priori and may depend on impurities on the solid surface or in the mercury. Most workers assume a contact angle of MG ". The error introduced by this assumption is nearly always small compared with other errors involved in interpretation. Ink bottle pores shown in Fig. 6.5, fill only at a pressure corresponding to the diameter of the neck, and the volume of the pores is thus erroneously assigned too small a pore diameter. Such ink bottle pores will not completely empty as the mercury pressure is lowered and therefore the retraction (mercury volume versus pressure) curve will not coincide with the penetration curve. 

Experimental studies in planar chamber-and-throat networks etched in glass plates have provided information about the mechanisms of mercury intrusion and retraction, and about the... 

In (ref. 37) and in the present work a new simulator of mercury intrusion into and retraction from a three-dimensional chamber-and-throat network is developed. The capillary resistance encountered at entrances to chambers under certain conditions during mercury intrusion, and the snap-off in throats during mercury retraction are taken into account. The effects of geometrical, topological and statistical parameters and of the intrusion and reoiiction contact angles on the form of capillary pressure curves are studied. Comparisons between the actual throat and chamber size distributions and the measured "pore size distributions" (by the convetional method of analysis) are also made. 

SIMULATION OF MERCURY INTRUSION AND RETRACTION 1) Simulation of mercury intrusion... 

A new theoretical simulator of mercury intrusion in and retraction from three-dimensional chamber-and-throat networks is developed. Stepwise porosimetry is modelled as a sequence of flow events occurring at each new external pressure value. The main conclusions resulting from the study of the effects of geometrical, topological and statistical parameters on the capillary pressure curves are listed below. 

Being able to identify the necessary properties to define a porous medium in no way implies that there are analytical methods to evaluate them. Using porosimetry and sorption, it is possible to measure 6 properties of the porous medium structure (refe. 5-7). It is not clear that these measures are independent. These properties are obtained by analyzing both mercury intrusion and retraction profiles (i.e., the nature of the hysteresis commonly found with most porous materials). These properties are enumerated below ... 

Several of the more popular models for deactivation involve "pore mouth plugging" wherein the transport limiting constrictions within the pore network are selectively reduced in dimension. If one realizes that intrusion mercury porosimetry and desorption measurements specifically characterize the constriction ("throat") dimensions then decreases in these dimensions would be greater than the changes found in the retraction porosimetry or in the desorption (which measure the opening dimensions). To understand the changes in network structure on the deactivation process it seems necessary to measure and analyze each aspect of the porous structure. 

On initial inspection the results obtained from serial sectioning of LMPA intruded samples appear at odds with the principle theory behind intrusion and retraction as predicted by the Washburn equation. But further inspection shows it is not the Washburn equation, but mercury porosimetry that is at fault. Pore network models have often been used to characterise the behaviour of pore structure in relation to mercury porosimetry. But the model is only as good as the assumptions and the data that it is based iqron. Without artificially shielding the network, the model caimot propa ly detomine the correct psd and cannot derive a more spatially accurate structure that could be used for diffusion and reaction modelling. In order to characterise the pore structure more accurately, we need to introduce some of the elements usually revealed by LMPA intrusion tests. 

In recent years it has been recognized that a more accurate method of pore analysis should consist of an appropriate combination of techniques of which mercury porosimetry is but one of the components (refs. 18,30,31). First, serial sectioning analysis of pore casts (refs. 32-35) can be used to determine the chamber-size distribution, the correlation between the sizes of adjacent chambers, and information pertaining to the interconnectivity of the network (e.g. specific genus and coordination number). Then, the capillary pressure curves can be used to determine the throat-size distribution, and the correlation between the sizes of contiguous throats and chambers. In order to deconvolve these curves a reliable simulator of intrusion and retraction of mercury in evacuated chamber-and-throat networks must be developed. 

Simulated capillary pressure curves for networks with die same CSD and two different bimodal TSD s having the same

and c, values are shown in Fig. 3a. Since intrasion is controlled mainly by the large throats, the intrusion curve widens and extends to a higher pressure range as the fraction of large throats decreases. In these networks, which have high ratio /, mercury retraction is controlled by snap-off events, and as the frequency of narrow throats increases, snap-off occurs over a wider pressure range with the result that the retraction curve widens and the residual mercury saturation increases. 

As it can be seen in Fig.3b,c, the large sizes of throats and chambo are not reflected in PSDl and PSD2 because of the shadowing effect during intrusion, and the entrapment of mercury in a large numbo of them during retraction (/=4.0). 

The intrusion curve moves to lower pressure ranges as the frequency of wide throats increases. Since mercury retains its continuity to the external mercury sink through large throats (which become disconnected by snap-off at lower pressures) the possibility of mercury retraction from chambers increases, and the residual mercury saturation decreases as the frequency of wide throats increases. In these cases where the fractions of large and small sizes of throats are comparable, neither very large sizes nor very small ones are reflected in PSDl (Fig.4b,c,d). It must be noted that the shape of PSD2 is affected by the shape of TSD as snap-off in throats intensifies during mercury retraction ((/=4.0). 

Comparison between the capillary pressure curves of an uncorrelated and two correlated networks with the same TSD and CSD is made in Fig. 5a. As the degree of coirelation becomes stronger the intrusion curve widens in both directions. The retraction curve is affected mainly in the portion near the end the residual mercury saturation decreases as the correlation increases. 

As the mean coordinaticm number decreases the intrusion and retraction curves widen, the degree of hysteresis increases and the residual mercury saturation increases. 

Throat size distribution is given from the intrusion profile. Pore body size distribution is given from the retraction profile (refs. 8-9). The ratio of the pore size to throat size at a given fraction of pore volume filled provides an indication of the shape or dimension of the void space. The amount of retained mercury (difference between intruded volume and extruded volume) gives some indication about the average connectivity and network size of the void space. Finally, total porosity in the range of 0.(X)5-0.5 microns in opening diameter is obtained from the total amount of mercury intmded and the solid densities from mercury displacement at 1 Bar. 

Mercury porosimetry intrusion-extrusion curves were obtained using Quantachrortte High (2(X)-50,000 psi) and Low (1-500 psi) Pressure Autoscan Porosimeters. An Analog Devices Macsym 150 computer was used to collea and analyze the data. Typical run times for intmsion and retraction was 7-8 minutes. 480 dyne cm" was used for the surface tension of the mercury with a contact angle of 140 degrees in the calculation pore dimensions. Total pore volume (maximum mercury intmded to 50,000... 

Conner and coworkers (refs. 7,8) have recently utilized a pore/throat network model to obtain information about the morphology of materials from mercury penetration data. The void/solid structure is viewed as an interconnected network so that adsorption/desorption and retraction/intrusion can be associated with the openings and constrictions within the void network. These latter investigators analyzed the data as if the materials consisted of agglomerated microspheres. The measured ratio of the most probable radii of intrusion to those of retraction seemed to be characteristic of the void structure and pore shape. Conner et al. (ref. 8) developed a heuristic diagram for the classification of void/solid morphologies from a... 

---