Dear Ahmed,

BET is a theory or method to determine surface areas, it will not give you direct information about porosity. From N2 adsorption at 77K you could determine most of the parameters you need when full isotherms are measured (from about 0 to about 1 p/p0, so more than the typical BET range). BJH theory could give you pore size and area distributions, in a range from about 3 to 300 nm. Micropore sizes can be obtained from other theories (DR, DA, DFT among others) or other gasses (CO2 at 273K). Gas adsorption methods will also not give you direct information about bulk or apparent densities. Though gas pycnometry gives the skeletal density, and that is typically done at the start of a gas adsorption measurement.

Hg intrusion can be used to measure pore sizes down to about 3.5 nm, but this limit depends on the properties of your sample (compressibility), the upper limit generally is larger than the 25 µm of your particles. Hg intrusion does not give direct information about areas or morphologies. It is based on Laplace's equation for capillary pressure, and therefore theoretically gives you only the size of the largest entrance to a pore.

There are many excellent texts that compare the different methods for pore analysis, I personally like this one:

http://www.iupac.org/publications/pac/66/8/1739/

I do not know where to find these instruments in Malaysia, but could make you an offer for measurements in our lab.

regards,

Pieter

Whatever the method you use, well-explained by Pieter, you have to be aware that powders are difficult materials from this point of view, as the grains are able to move relatively to each other. It means that, depending on the pressure applied to the powder (which is obvious for Hg intrusion), the results may change. Additionally, if your grains are themselves porous, it may be extremely difficult, if not impossible, to separate the pores inside the grains from the pores betwen the grains. This can be done only if the grain size is different enough from the characteristic size of the spaces between the grains.

Alain

Both techniques are complimentary (I suppose you refer as 'BET' to the standard nitrogen adsorption at 77 K).

You should first know the magnitude of the pore size in your sample, and then choose the more appropriate. I would recommend to start with N2 adsorption (fast, cheap and readily available in any adsorption laboratory).

Gregg&Sing's "Adsorption, Surface Area and Porosity" is the old standard book for learning this.

I would suggest a mixture of methods, by using BET as well as Hg-injection. From our lab experience, BET is only feasible for the micro- to meso-porosity range, so going from a few nm to few microns. Since you do have a powder, Hg would be applicable - as stated by Pieter, from severel microns upward.

Please keep in mind, that BET determines equivalent pore diameters, whereas Hg-injection determines pore throat diameters, which is not 1:1 comparable.

All of the comments are valid. However, I would add that nitrogen adsorption will require a certain minimum amount of total surface area of your sample. This will depend on the specific instrument (for example mine requires officially 5m^2 per sample.

Having particle sizes of several micrometer will frequently not provide enough surface area to perform a BET analysis (or you might need a very large quantity of sample.

Another advantage of the mercury porosimetry analysis is the fact that you might be able to separate external from internal surface area in your particles as the technique allows one to associated surface areas with different pore size ranges.

•N2 sorption is suitable to characterize materials with pores within the range of ~20 Å to below ~1500 Å (0.15 μm).

•For materials containing larger pores, mercury porosimetry is the preferred experimental technique and spans the pore range from ~35 Å to ~500 μm.

Mercury porosimetry is paired with the gas sorption technique to obtain porosity information in the large size range (greater than about 0.3 mm diameter up to about 360 mm), which is not attainable by gas sorption (BET). When using two different techniques (two very different models), one should not expect necessarily to obtain the same results in the overlapping or common range of both instruments. One potential difficulty in comparing porosity data sets obtained in the smaller size range by mercury porosimetry with data gathered in the same range by gas sorption (BET) is that sample compression is a possibility with mercury porosimetry. If sample compression occurs, an apparent uptake of mercury is superimposed on the intrusion curve leading to an erroneous indication of pore volume, and one not reproduced by the gas adsorption analysis.

Only to add: We had a bit different materials: intact and crushed rock; we wanted to show the different results by BET and Hg-porosimetry and artefacts due to sample preparation.

Our paper we had uploaded once.

K-H. Hellmuth, P. Klobes, K. Meyer, B. Röhl-Kuhn, M. Siitari-Kauppi, J. Hartikai­nen, K. Hartikainen, J. Timonen,

Matrix retardation studies: size and structure of the accessible pore space in fresh and altered rock.

Z. geol. Wiss. 23 (5/6) (1995) 691-706