CHPR4501/5501: Advanced Reaction Engineering and Catalysis

Lecture Recording 14 : Catalyst Characterisation and Catalyst Deactivation/Poisoning

Prof. Mike Johns

Department of Chemical Engineering
University of Western Australia
 Typical Features of ‘Modern’ Solid Catalysts

• Surface Area: 10-500 m2/g

• Microcrystalline and often multi-component (up to 10 different elements)

• Pore apertures of 4 – 4000 angstroms.

• Actual molecular level mechanism often poorly understood for heterogeneous

catalysis.
 Characterising Catalyst Surfaces

There are numerous techniques!!!

Gas Adsorption – we have already seen that this can be used to determine
surface area. However selective adsorption can also be employed. For example, the
adsorption of CO2 on the Fe Catalyst employed in Ammonia synthesis can be
used to determine the surface coverage of the promoter K2O.

Using a range of adsorbates of different molecular dimensions, it is possible to determine

the fraction of surface area accessible by an adsorbate of a given size.
 Which Elements and Which Phases are present in my Catalyst?

Historically this question was addressed using wet chemistry. Now elemental
composition can be accessed by rapid X-ray emission techniques.

This method performs best at determining bulk composition, however surface composition
is ultimately much more relevant. For this purpose we generally employ
electron spectroscopy.
 Which Elements are present on my Catalyst surface?

Surface Crystallinity is increasingly thought to be of greater importance in Catalysis.

 Other Characterisations (during reaction)

Electron Spin Resonance (ESR) is useful to characterise the ionic form of paramagnetic
species on the surface of the catalyst. e.g. O- and O2-. It is applicable only to a limited
set of species however.

Nuclear Magnetic Resonance (NMR) is useful providing a unique ability to

distinguish, for example: alcohols from alkenes, alkanes for alkenes, ethers from acids etc.
It works best in the dispersed phase (i.e. not solids). However solid state techniques are
evolving (MAGIC ANGLE SPINNING (MAS)) and are increasing used to characterise surface
species of both catalysts and reactants/products.

For example it can be used to determine the acidity of zeolite and amorphous Silica-alumina
surfaces (non-invasively)
 Electron Microscopy
Scanning Electron Microscopy (SEM) is capable of producing images of the surface
morphology of catalysts.

SEM scans the sample with a beam of electrons in a raster scan pattern. A 2D image
results. It can reveal details at resolutions lower than 1 nm. Specimens need to be
electrically conducting on the surface. Non-conductive samples are usually coated with
Gold. A vacuum is required during imaging.
 Catalyst Deactivation / Poisoning
Up to now we have assumed that the Catalyst Effectiveness factor does not change with
time. Usually this is not true and the effectiveness factor deteriorates with time. Sometimes
this takes a year, often it occurs within seconds.

If deactivation is rapid and caused by deposition and a physical blocking of the surface,
this process is usually called fouling. Removal of this solid deposition is called catalyst
regeneration. Carbon deposition (coking) during catalytic cracking is a common example:

C10H22 → C5H12 + C4H10 + C(on catalyst) - note we end up with a spare C! ALWAYS!

If the catalytic surface is slowly modified by chemisorption on the active sites by material
then the process is called poisoning. (e.g. sulphur is a pervasive catalyst poison – it is present
in virtually all transport fuel. Base metals (e.g. Fe, Ni, Pb, Zn) are particularly susceptible)
Reactivation is restoration of the active sites. The activity generally reduces with amount
of poison in the system. Poisoning is seldom easily reversible.

If the adsorption is reversible, than a change in operating conditions might suffice for
reactivation. If it is not reversible, than typically a chemical treatment or complete
catalyst replacement will be required.
Another form of catalyst deactivation is sintering – this is a loss of surface area and occurs
via:
> Surface diffusion and clumping of catalysts on their support, and
> Collapse of support structures.
This mechanism is not reversible and is more prevalent at extreme conditions
(high temperatures)
 Deactivation Sources

The source of the poison or coke laydown (P) might be:

(i) Product of main reaction (parallel deactivation),

A R + P
(ii) Alternative reaction of feed (parallel deactivation),
R
A
P
(iii) Subsequent reaction of product (series deactivation) or

A R P
(iv) an Impurity in the feed (side-by-side deactivation).

A R P P
For (i) and (ii) the concentration of the feed is important. If the Effectiveness factor () is
small, preferential deactivation will occur toward the edge of the particle. This
will promote a more homogeneous reaction rate across the pore diameter.

For (iii) the concentration of the product is important – this will potentially be higher for low
 in the centre of the particle. Thus deactivation will be higher in this central location,
accentuating the concentration difference of reactant and hence reaction rate between
the centre and the surface of the particle.

For (iv) the Thiele modulus of the feed impurity is all important. Does it penetrate the
particle by diffusion before ‘reaction’ (surface adsorption or reaction) or not? If mass transfer
controlled, will deactivate the surface region of the catalyst pellet first.

Sintering is generally equal everywhere assuming an isothermal particle.

 We can define the time-dependant activity of a catalyst pellet as (A – feed):

𝑟𝑎𝑡𝑒 𝑎𝑡 𝑤ℎ𝑖𝑐ℎ 𝑡ℎ𝑒 𝑝𝑒𝑙𝑙𝑒𝑡 𝑐𝑜𝑛𝑣𝑒𝑟𝑡𝑠 𝑟𝑒𝑎𝑐𝑡𝑎𝑛𝑡 𝐴 −𝑟′𝐴

a= =
𝑟𝑎𝑡𝑒 𝑜𝑓 𝑐𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛 𝑜𝑓 𝐴 𝑓𝑜𝑟 𝑎 𝑓𝑟𝑒𝑠ℎ 𝑝𝑒𝑙𝑙𝑒𝑡 −𝑟′𝐴0

The rate of a nth order reaction is then (a < 1 (almost always)):

−𝐸
−𝑟′𝐴 = 𝑘𝐶𝐴𝑛 . 𝑎(𝑡) = 𝑘0 𝑒 𝑅𝑇 𝐶𝐴𝑛 . 𝑎(𝑡)

Similarly we can define the rate of deactivation as follows:

−𝑑𝑎 −𝐸𝑑
= 𝑘𝑑 𝐶𝑖 . 𝑎 = 𝑘𝑑0 𝑒 𝑅𝑇 𝐶𝑖𝑛′ . 𝑎𝑑
𝑛′ 𝑑
𝑑𝑡

d is called the order of deactivation, n’ is the concentration dependency of deactivation upon

some species i. Ed is the activation energy of the deactivation.

For (i) and (ii) i is the feed (A), for (iii) i is the product (R) , for (iv) i is the impurity (P)
 In certain reactions (e.g. isomerisation and cracking) deactivation might be caused
by both reactant and product. In this case:

−𝑑𝑎 .
= 𝑘𝑑 (𝐶𝐴 +𝐶𝑅 )𝑛′ 𝑎𝑑
𝑑𝑡

However in this case CA + CR remains approximately constant, thus:

−𝑑𝑎
= 𝐾. 𝑎𝑑
𝑑𝑡
Monitoring Deactivation
When deactivation is very fast (such as in FCC systems) a flow through system (of solids)
is required to monitor deactivation. This would be a bench-top fluidised bed or a pumped
slurry. Note the activity of Cracking catalysts can have a half-life of 0.1 s.

The Carberry Spinning Basket Reactor is used to create a well

mixed solid –fluid system. Can be completely batch or fluid
flow through. Struggles at higher pressures.
 Consider the following Simple Deactivation Expressions
.
−𝑟′𝐴 = 𝑘𝐶𝐴 . 𝑎

−𝑑𝑎
= 𝐾.a
𝑑𝑡
Integration of the 2nd expression

𝑎 = 𝑎0 𝑒 −𝐾𝑡 = 1. 𝑒 −𝐾𝑡

Assuming a constant Volume in a Batch reactor

−𝑑𝐶𝐴
= 𝑘𝐶𝐴 .a= 𝑘𝐶𝐴 . 𝑒 −𝐾𝑡
𝑑𝑡

which upon integration becomes (Batch reactor)

𝐶𝐴0 𝑘
𝑙𝑛 = (1 − 𝑒 −𝐾𝑡 )
𝐶𝐴 𝐾
 −𝑟′𝐴 = 𝑘𝐶𝐴 . 𝑎
Now consider a (Constant Volume) CSTR
−𝑑𝑎
= 𝐾.a
𝐹𝐴1 𝑋𝐴2 𝐹𝐴1 𝑋𝐴2 𝑑𝑡
𝑉= =
−𝑟𝐴2 𝑘𝑎𝐶𝐴2 𝑎 = 𝑎0 𝑒 −𝐾𝑡 = 1. 𝑒 −𝐾𝑡

which on rearranging gives:

𝐶𝐴1 𝑘𝑎𝑉𝐶𝐴1
=1+
𝐶𝐴2 𝐹𝐴1

𝐶𝐴1 𝑉
Now: 𝜏=
𝐹𝐴1

𝐶𝐴1
= 1 + 𝑘𝑎𝜏 = 1 + 𝑘𝑒 −𝐾𝑡 𝜏
𝐶𝐴2
rearranging

𝐶𝐴1
ln − 1 = ln 𝑘𝜏 − 𝐾𝑡
𝐶𝐴2
 −𝑟′𝐴 = 𝑘𝐶𝐴 . 𝑎
Now consider a (Constant Volume) PFR
−𝑑𝑎
= 𝐾.a
𝑑𝑋𝐴 𝑑𝑡
𝑉 = 𝐹𝐴1 න
−𝑟𝐴 𝑎 = 𝑎0 𝑒 −𝐾𝑡 = 1. 𝑒 −𝐾𝑡

𝑑𝑋𝐴 𝐹𝐴1 𝑑𝑋𝐴

𝑉 = 𝐹𝐴1 න = න
𝑘𝑎𝐶𝐴 𝑘𝑎 𝐶𝐴

Integration gives:

𝐶𝐴1 𝑉 1 𝐶𝐴1 1 𝐶𝐴1

𝜏= = 𝑙𝑛 = 𝑙𝑛
𝐹𝐴1 𝑘𝑎 𝐶𝐴2 𝑘. 𝑒 −𝐾𝑡 𝐶𝐴2

Rearrangement gives:

𝐶𝐴1
ln 𝑙𝑛 = ln 𝑘𝜏 − 𝐾𝑡
𝐶𝐴2