effect of seeding in crystallysers 1

[wati]

Dear all,

Can anyone tell me more about seeding techniques and the effect in evaporative crystallizers ?

Thanks

 

[m777182]

By seeding a supersaturated solution with fines you trigger nucleation at a lower supersaturation level compared to a non seeded operation. The consequence is that you have less nuclei in your solution and more supersaturation is turned to the growth of crystals.From the population balance it results that you have by the same mass of solid phase smaller number of crystals in your system which means they are bigger.
Calculations are not so simple, however you have to solve a system of differential equations form mass, energy and class number of crystals including nucleation rate, growth rate, agglomeration and attrition rate as well.
m777182

 

[wati]

Thanks m777

But still its unclear to me:

"The consequence is that you have less nuclei in your solution and more supersaturation is turned to the growth of crystals".

How come that less nuclei results in crystal growth (volume increasing).Does seeding acts as an actuator? Can you give me a specific calculation eg. ammonium-sulphate/water? How can we investigate and monitored the seeding by experiment especially for ammonium-sulphate crystalization?

Thanks

 

[m777182]

There is a long story behind.About nucleation: in pure solutions you can increase the supersaturation to a high level without any phase transfer; it is entirely a metastable state.The more fare you go from the solubility line the higher is the probability that the nucleation will occur like an explosion-in an instant. If there are already solid particles in your solution, if you stirr it or anyhow disturb it, these paricles or small bubbles of gas will act as new nuclea so you cannot increase the supersturation deliberately. New nuclea will be formed near to the saturation line and they will start growing immediately. So whenever you look to your solution you will have there some nuclea that will be created in that particular instant of time and a population of crystalls, each one with his individual history, some of them beeing there from the very begining and some of them just about starting their growth.Because in average they have been growing from the start on they are in average larger and because the process is using nuecla for growth their number is smaller.
A seed acts apparently as a mean to lower the activation energy for a nucleation: the solubility of a particle depends (besides on T) on its radius; the smaller it is the more is soluble. Consequently smaller particles need higher supesaturation for phase change. Adding seeds means that you start actual growth immediatelly.
Regretfully I cannot give you calculations here because it is not a one-two-three task, but the ammonium sulphate is a system that has been published in many papers.
The most simple experiment would be to track the crystall size distribution with adding different amounts of seeds preferentially of equal size- a dust or fines of your produst will be fine, but keep temperature profiles, supersaturation profiles and aggitation constant!
m777182

 

[bewdley]

I can't add a comma to the good description of m777182. I would just remark that, if excessive nucleation is the problem (as it happens in manganese sulphate, for example) the solution might be of looking fromn the other side of the problem : since nucleation can only occur when supersaturation is present, the more so if it is very high, and that statistical considerations (swirls, bubbles et similia) increase the nucleation rate, you could rather try to reduce the volume of the supersaturation zone and try to define the optimum surface to volume ratio (in terms of m2 of xls surface per m3 of slurry). This is the concept of the draft tube and baffle (DTB) crystallizer.

 

[m777182]

To bewdley:
I agree that swirling and high turbulence increase the secondary nucleation so it should be kept on minimum by using efficient aggitators with low power input; however you should keep some mixing to allow the heat transfer and homogeinity of the suspension. Operating at low supersaturation makes your production rate low and by low supersaturation you can get a different crystal shape, mostly more perfect shape. The trend today is towards spherical dense crystalls with high bulk density.So the good policy might be a fines trap:separating and dissolving fines and recirculating slightly undersaturated solution.Interesting topics,isn't it?
m777182

 

[bewdley]

m777182,
Basically, we are looking at the same crystallizer... external recirculation - and thermal destruction - of fines you say, and I agree. Adding that the unsaturated solution should be injected in the best possible location, generally deep down in the vessel, since it is hot. The problem is, in real life you cannot always thermally destroy the fines : think to anhydrous sodium sulphate, for example. So you have to work on both issues. BTW, I meant reducing the volume of the supersaturation zone, not reducing the supersaturation in terms of % excess solute - of course you have to compromise, somehow.