You're measuring a dry powder in BET and you're measuring specific surface area.. So the minimum size is what you can get for a powder (as solid-solid diffusion has small particles diffuse into one another forming solid bridges) - 0.5 to 1 um is the lower limit for a powder ('nanopowder' is an oxymoron). See:
November 3rd, 2015 Adhesion and cohesion http://tinyurl.com/zwb2wlh
To prevent (fine) dust being sucked out by the outgassing regime then it's common to use a plug of glass wool. The larger limit is set by the material and the ability to get it into the measurement zone and get realistic gas adsorption (perhaps 0.05 - 0.1 m2/g) . You'd consider Kr or Xe for low surface area material rather than the usual N2 for conventional BET. Couple of useful rules of thumb:
SSA = 6/D[3,2] where D[3,2] is the Sauter Mean Diameter so (for unit density):
Particles of 1um imply an SSA of 6 m2/g
Particles of 10 nm primary size (solid bridged) 600 m2/g
Useful texts:
S J Gregg/K S W Sing: Adsorption, surface area and porosity
Thanks Dr. Rawle. I want to analyse the pore characteristics as well as SSA for particles in the range of 300-500 micron size. Is it possible for getting them analysed in BET?
@ Pragadeesh K Sekar In theory, yes, But these are large particles pore sizes (thus very low SSA's) and you probably will need Kr or Xe physisorption. At these sizes you may consider counting and weighing and some assumption of shape (spherical) and density for a surface area. That assumes that you do not have smaller material present which will add significantly to the SSA>
@ Niranja SEM or TEM does not measure a specific surface area without some gross assumptions
Size is just something practical to the experimental procedure, there is no real limit:
As Alan above says, if particles are too small the problem might be in sample 'flying away' while degassing. This can be overcome by slowly heating/vacuum.
If particles are too large, it might not fit in your sample holder and special holders might be required.
300-500 µm should be no problem for analysis of adsorption isotherms; however, you always should be aware of slow-diffusion effects.
Thanks Dr.Rawle and Dr.Fernando for your answers. I would initially try with particles as it is. If not possible, I would go with inert gases like Ar, Kr or Xe.
One good experiment is worth a thousand Greek letters....
He is normally used to 'calibrate' the dead-space volume prior to a physisoprtion or chemisorption experiment.. He pycnometry is used to measure the density of a sample.