Page 327 - Book Hosokawa Nanoparticle Technology Handbook

FUNDAMENTALS CH. 5 CHARACTERIZATION METHODS FOR NANOSTRUCTURE OF MATERIALS
As seen in clay materials and activated carbons, hys- permeation properties are of importance. Depending
teresis loop of type H3 and H4 often closes at the rel- on the wettability the porous material is immersed
ative pressure of 0.40–0.45. As given in detail by either in freon or water. Pressurized air or nitrogen is
literature [13, 14], this closure of hysteresis will result then introduced to one side of the material. At a pres-
from the spinodal evaporation of condensed liquid sure that corresponds to the maximum size of the per-
when exceeding its tensile limit, which is determined forating pore, the gas starts to permeate. Other than
not by the pore size but merely by the nature of liquid. the detection of maximum pore, the size distribution
This phenomenon should be considered when one can be estimated applying higher pressures, because
uses the desorption branch for characterization. smaller pores start to open with increased pressure.
Detectable pore size is usually above several tens of
nm with freon, or a few hundred nm for water.
5.4.4 Other methods of interest
5.4.4.3 Thermoporometry (freezing point depression
5.4.4.1 Mercury porosimetry (mercury intrusion
method)
method)
Based upon the Gibbs–Thomson equation, which
An evacuated sample is immersed in mercury, and
pressurization of the system gives mercury’s intrusion suggests that the freezing point depression in a pore
into the pore space. The detection and analysis of the from the freezing point in bulk phase is inversely pro-
intruded volume against applied pressure gives the portional to the pore radius, the size distribution is
pore size distribution since the pore volume directly estimated from calorimetric measurement. It is some-
equals that of the intruded mercury. Many automated times the case that the pore structure when wetted
apparatuses are commercially available. The method varies after drying because of capillary suction pres-
is suitable for macropore or mesopores above tens of sure or de-swelling of the base material. The gas
nanometers. adsorption or mercury porosimetry cannot character-
Many of automated porosimetry apparatuses ize such porous materials because both method need
declare the lower limit of the pore size to be 3–4 nm, evacuation before measurements. This method may be
which corresponds to the applied pressure of ca. used to overcome the above problem: the measure-
4,000 atm. At this level of high pressure, the porous ment goes as wetted. However, not much has been
framework might be deformed or some other influ- clarified for the freezing behavior in confined space.
ence may occur. It is also pointed out that the surface Recent study, e.g., has clarified that the freezing point
of mercury in a single-nanometer range may be dif- may even get higher than the bulk, depending on the
ferent from surface of bulk liquid, which may results physico-chemical nature of pore walls. The method is
in a hindered contact angle. Much attention, then, thus especially controversial if single-nanometer
should be paid on the reliability of the data in this range is concerned, and it would be better not to be
range, for which the gas adsorption method has far relied on for smaller size of nanopores.
superior accuracy and reliability. The upper limit of
the mercury porosimetry would be around several 5.4.4.4 Small angle X-ray scattering (SAXS)
hundred of microns: The detection of cracks or The X-ray scattering with angles smaller than 10°
supermacro pores, which is difficult to be measured can probe porous characteristics in single-nanome-
by gas adsorption, is precisely done by the intrusion ter range. Since X-ray can detect not only open
method. pores but also closed (or isolated) pores, measure-
The extrusion process by decreasing the pressure ments for low-permittivity materials are typical
generally gives different path from that for the intru- examples of application. One has to be, however,
sion process, or hysteresis. Further, it would be almost careful upon interpretation of the data, because the
always the case that a certain amount of mercury resulting space distribution or correlation length
remains in pores even after complete release of the does not necessarily mean the scale of pores but has
pressure. The extrusion process, therefore, is not suit- resulted from the electron density distribution.
able for analysis, and the intrusion branch is used in Further, while an ordered material such as MCM-41
general. The intrusion of non-wetting process corre- would give quite clear signals showing its lattice
sponds to the desorption branch in gas adsorption, size or the periodicity of the regular pores, not much
and then one should notice that the analysis gives the sensitivity can be expected for materials with disor-
neck size for aggregated or sintered bodies. dered or random nature, which needs a rather com-
plicated analysis for gradually decreasing scattering
5.4.4.2 Bubble-point method intensity. From the Guinier plot, e.g., one would be
This technique detects perforating pores while the gas able to obtain the averaged radius of scattering
adsorption method and mercury porosimetry cannot body, but it may not necessarily be related with the
distinguish those from dead-end pores. Because of pore size itself. One should understand that SAXS
this feature the method is often applied to filters, has only low quantitative precision when applied to
membranes, cloths, or those porous materials whose nanopore characterization.

302