Students who have followed this lecture will be able to:
We will restrict ourselves to 1st order reactions to illustrate the principles of reactor design. The same concepts are relevent to more comples mechanisms, but the mathematics is harder!
We will consider the 1st order isomerisation reaction:
As well as having a very simple kinetic equation, for this system we can always say that:
Also:
There is a fundamental difference between two ways of operating any chemical process.
Batch processes are like bench chemistry on a large scale. A reactor is filled with reactant(s) and this/these change over time into product(s). Such a process is easy to visualise, although its mathematics are harder to handle than the continuous process.
Most chemical processes are continuous : reactants are constantly fed to a reactor and products are continuously withdrawn. Such a process is said to operate at steady state since, unlike a batch process, its operating conditions, e.g. the amount of reactant in a reactor, do not change with time. The mathematics of continuous processes is generally simpler than that of batch processes.
Large scale processes, with production rates greater than, say 20,000te/yr, are invariably continuous. Small scale processes, say less than 2000te/yr are usually batch.
There are practical advantages and disadvantages to both approaches.
The reactor in a batch process is normally a tank which is filled with reactants which are allowed to react, often for several hours. The contents are then withdrawn and generally processed to separate or purify the final product. Since batch separations are difficult, the aim of a batch reactor is normally to get a high conversion of reactants into product.
There are two distinct types of reactor in continuous processes. One is called a Continuous Stirred Tank Reactor (CSTR) which is again a tank but to which reactants are continuously fed and from which product is constantly withdrawn. although it looks very like a batch reactor it behaves quite differently.
The other is effectively a pipe through which material constantly flows. This is called a Plug Flow Reactor (PFR). this has quite different characteristics from the CSTR.
Consider a tank which at an instant in time contains ma kmols of reactant A.
The reaction has rate constant k in e.g. hr -1
The instantaneous rate of reaction in e.g. kmol/hr is given by the kinetic equation:
The above is an ordinary differential equation which has a simple analytical solution obtained, e.g. by separating the variables.
Since this form of equation occurs frequently in chemical engineering, it is worth remembering the solution.
For
Hence for:
ma(t)/ma(0) = 1 - Xa
We can see that the initial rate is highest, and it gradually falls off as A is consumed. In fact we can NEVER use up all the A except when exp(-kt)=0 which only occurs at t equal to infinity, which is not operationally possible.
Let if we call the batch time or
residence time in the reactor tr
we can tabulate the product (k tr) for all 1st order reactions
and so work out the batch time required for a given conversion
for any known k.
(k tr) exp(-ktr) 100 (1 - exp(-ktr)) (%)
0.5 0.607 39.3
1 0.368 63.2
2 0.135 86.5
2.303 0.100 90
3 0.0495 95
4 0.183 98.2
5 0.0067 99.3
To check that you understand this table, try this question.
If we know how long a batch takes to reach a given conversion, then we can work out how large the reactor must be for a given production or consumption over a period of time.
A process operates 24hr/day with a required production of 48 te/day. The batch time for 90% conversion is 2 hours. How much material will the reactor have to contain for each batch?
Calculation:
We can get 24/2 = 12 batches per day. We therefore need to produce 48/12 = 4 tonne per batch. At 90% (0.9 fractional) conversion we will need to fill the reactor with: 4/0.9 = 4.4 tonne/batch. From the density of the material we can determine reactor volume required.
A batch reactor is typically a tank with a stirrer. What happens if we take such a tank and arrange to feed reactant(s) at a constant rate and to withdraw a product stream so that the amount of material in the tank stays constant?
Once such a system has settled down to its steady state there will be no change in the composition of the material in the tank either. If the mixer operates effectively then the composition everywhere in the tank will be the same.
Also note that the composition of the material leaving the tank must be exactly the same as that of the material in the tank.
In general the tank will contain both reactants A and products B. let the molar quamtities of these be ma and mb.
The extensive rate of reaction will be equal to:
kma
and will be equal to the rate of consumption of A or minus the rate of A production:i.e. it will be constant and depend only on the amount of A in the reactor.
The rate of production of B will be minus the rate of A production:
Hence the production rate in a CSTR depends on the amount of reactant ma it contains, and not directly on the size of the reactor.
The reactor volume however is important because it determines the composition of the product mixture.
What is not A is B!
A high concentration of B is desirable to reduce the cost of downstream separations.
The required size of a reactor can be calculated using:
The rate constant for a 1st order reaction is 5hr-1. A production rate of 100 kmol/hr is required. How much reactant must the reactor contain?
Solution:
-Ra = production rate = 100 kmol/hr = k ma = 5 ma
So ma = 20 kmol
The product stream must contain 60% (molar) of product. What is the
total molar holdup in the reactor?
B:A 60:40 in both reactor and product.
So B in reactor = 20 x 60 / 40 = 30 kmol
Total holdup is thus (20+30) = 50 kmols.
The reactor volume can be determined from knowledge of molecular
weight and density for a liquid system, or temperature and pressure
for an ideal gas system.
Do not attempt to memorise the following equations! All can be written down when required from an understanding of what is going on.
Assumptions:
Let -Ra kmol/s be rate at which A is converted by reaction to B
(Remember that Ra is negative!)
Let Rb be rate at which B is produced
Let pb be rate at which B leaves reactor
In the above example the feed to the reactor is pure A.
To produce 100 kmol/hr of B we had calculated that the reactor should contain 20kmol A and
30 kmol of B.
At what rate does unreacted A leave the reactor?
NB: Composition in reactor and of product stream are equal:
So here: pa/100 = 20/30 So pa = 66.7 kmol/hrWhat is the A feed rate to the reactor?
fa = pa + pb = 100 + 66.7 = 166.7 kmol/hrWhat is the reactor per pass conversion?
By definition, e.g. : Xa = pb / fa = 100/166.7 = 0.6 Conversion is 0.6 or 60%Note that if reactor feed is pure A then:
Question: Does a molecule of A in a reactor care whether it sits stationary in a tank for a time tr or whether it drifts downstream in a pipe for the same length of time?
Answer: no it doesn't.
A plug flow reactor (PFR) is another type of continuous reactor which, while being physically totally different, behaves mathematically just like a batch reactor.
The key quantity is the residence time tr,
Let total molar flowrate through reactor be F kmol/hr
Let total molar holdup in reactor be M kmol, then:
An analogous relationship holds for any consistent flow and holdup units.
For a volumetric flow rate of Fv m3/hr and a reactor volume V m3:
For a mass flowrate of G kg/s and a mass holdup of W kg: