Page 111 - Academic Press Encyclopedia of Physical Science and Technology 3rd Chemical Engineering

P1: ZBU 2nd Revised Pages
Encyclopedia of Physical Science and Technology EN002G-87 May 19, 2001 20:3

Catalyst Characterization 509

their weight with the weight of washcoat initially bound E. Morphology
to the monolith.
1. Surface Texture via Electron Microscopy
Monolithic materials are frequently subjected to sub-
stantial thermal gradients (thermal shocks) in startup and Electron microscopic examination of catalyst materials,
shutdown, as in auto exhaust emission control. Frequent particularly those containing natural components, permits
thermal shocking causes the washcoat to lose adherence the identification of their origin. For example, carbons
due to the expansion difference between it and the mono- utilized as supports for precious metals in a wide variety of
lith. This is most pronounced when the monolith is metal- slurry-phase and fixed-bed reactions can be derived from
lic. a large number of naturally occurring sources (Fig. 8). The
One can evaluate this phenomenon by periodically sub- shape, morphology, and composition are useful properties
jecting the catalyzed material to gaseous flows at tem- for determining their origin.
peratures anticipated in service and noting weight losses. Edges or surface irregularities on particulate catalysts
Monoliths can be mounted on a rotating “carousel” that used in fluid- or fixed-bed applications are susceptible to
moves in and out of streams of heated gas. attrition and erosion during reaction. The morphology of
a typical fluid-bed cracking catalyst is shown in Fig. 9.
The surface of spheres, extrudates, and tablets is relatively
D. Density
free of topological features, and thus physical losses are
1. Bulk or Packing usually not a serious problem.
The texture of PtRh gauzes before and after use in
Particulate catalysts are usually sold by weight but are
nitric acid production shows significant morphological
charged to a reactor by volume. Thus, the density of the
changes due to vaporization and deposition of PtO x
support has a strong impact on the economics of the pro-
species (Fig. 10C). The surface, as examined by electron
cess. Fluidization in moving-bed reactors is also depen-
microscopy, shows a swelling or sprouting effect that in-
dent on catalyst density. The density of powders affects
creasesthesurfaceareaofthegauzeduringtheearlystages
the extent to which they can be suspended and eventually
of use.
settled in slurry-phase reactors.
Samples of powders or of formed particles such as pel-
lets, spheres, or extrudates are first dried at a tempera- 2. Adhesion of Washcoats to Monoliths by Optical
ture sufficient to remove moisture or organic contami- Microscopy
◦
nants (∼400 C). The material is cooled and vibrated while
Refractory high surface area oxides are deposited from
being poured into a cylinder of known volume and weight.
slurries onto the walls of the channels that make up mono-
The volume occupied by the catalyst is noted along with
liths in order to provide an adequate surface area to sup-
its weight. The weight of the catalyst divided by its vol-
port the active catalytic species. Washcoats such as Al 2 O 3
ume is its apparent packing density. In place of a vibrator
and TiO 2 are commonly used for pollution abatement ap-
one could also use a tapping device and thus obtain the
plications (auto exhaust, stationary NO x abatement, etc.)
tapped apparent packing density.
where the monolith is usually a ceramic. Metal monoliths
are finding increasing use; however, they represent only
2. Skeletal a small percentage of the total monoliths used. Optical
microscopy enables one to see that the catalyzed wash-
The skeletal density is representative of the solid material
coat follows the contour of the ceramic surface. Figure 7
itself, excluding its porosity. The bulk volume of catalyst
shows the Al 2 O 3 washcoat–ceramic interface for a typical
minus its pore volume and the interparticle volume be-
auto exhaust catalyst. In this case, no evidence of loss of
tween discrete particles (V I ) is the true skeletal volume.
adhesion between washcoat and ceramic can be seen.
One calculates this term by
d skeletal = M/V skeletal
F. Location of Catalytic Species Within
where M is mass and V skeletal = V bulk − V pore − V I . The a Support
pore volume is determined by the mercury intrusion
1. Electron Microscopy
method (Section II.A.2); however, only pores greater than
˚
∼30 A are measured (for an instrument with a 60,000-psi For reactions that are pore diffusion controlled, it is ad-
capability). The pore volume that includes pores less than visable to locate the active catalytic species close to the
˚
∼30 A must be determined by cumbersome gas displace- fluid–solid interface in order to decrease the diffusion
ment techniques. path of reactants and products. This applies to all forms