JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
Optimal Fractional Distillation Drip Rate
How do you determine the best drip rate to use for fractional distillation? I usually aim for one drop per second, but I'm not sure that's really
optimal. It seems reasonable to suppose that different equipment could work better with different distillation rates. The properties of the substances
being distilled could matter also.
|
|
|
j_sum1
Administrator
Posts: 6320
Registered: 4-10-2014
Location: At home
Member Is Offline
Mood: Most of the ducks are in a row
|
|
I think a more useful parameter is the reflux ratio. Having said that, increasing the distillation rate will decrease the proportion of reflux that
occurs.
As anything, it is highly dependent on what you are distilling. These things generally get figured out empirically by testing the product.
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
The drip rate is of particular interest because for a given distillation, it is inversely proportional to the length of the distillation and a convex
function of power consumption. It's a dependent variable.
It's rather unusual to see a reference with tables of reflux ratios. Surely there are some general rules for determining an optimal reflux ratio.
|
|
|
j_sum1
Administrator
Posts: 6320
Registered: 4-10-2014
Location: At home
Member Is Offline
Mood: Most of the ducks are in a row
|
|
Well, JJay, you are undoubtedly a lot more experienced than me with these things. All I know is that if I want to get better separation from a
fraction then it helps to adjust heating and column insulation so that there is a lot of reflux and the rate of product coming over is low.
I guess it depends on what you are trying to optimise. If you introduce time and energy expenditure into the thinking then you obviously are wanting
to raise the drip rate without compromising separation too much. I expect that the exact solution is very much dependent on what you are trying to
distill. The only common denominator in my experience is that I tend to do worse whenever I am impatient.
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
One drip per second through a 30 cm Vigreux seems to lead to decent separations, but it's not fast (something like 200 mL per hour). I think it's too
slow for solvents.
|
|
|
andy1988
Hazard to Others
Posts: 135
Registered: 11-2-2018
Location: NW Americus ([i]in re[/i] Amerigo Vespucci)
Member Is Offline
Mood: No Mood
|
|
This can be determined/tuned analytically using chemical process simulation software such as DWSIM (free & open-source). The analytical result could be used as a heuristic for later empirical testing/tuning, or the results used as-is.
With DWSIM, the process appears to be to enter in the chemicals that are in what you're distilling, then calculate the:
shortcut distillation (UI widget component on the flowsheet) [tutorial 1][tutorial 2], then
rigorous distillation (UI widget component on the flowsheet) [tutorial 1][tutorial 2]
The theoretical plate # result from the simulations above could then be translated into a hempel column and packing.
You'd adjust parameters in the simulation to get a suitable outcome, like: purity, throughput, and tower height (capital cost). You can then see
tradeoffs, like increasing reflux reduces the number of stages, while decreasing reflux increases throughput requiring more stages (taller column and higher capital
costs).
As for the reflux regulator (i.e. reflux column head, still head), it sounds like you already have a variable reflux ratio head:
Here is a NIST document I wrote about earlier which describes the difference between the variable reflux ratio and
constant reflux ratio types of reflux regulator.
The Practical Organic Chemistry 3rd edition book (page 102) in the SM library talks about the partial condensation and total condensation variable take-off types
of reflux regulator.
I'm still trying to conceptualize everything. Maybe there is a literature review on reflux regulators out there somewhere, there certainly are many
variants I can see on ebay and elsewhere.
I know both myself and Sulaiman are working on this fractional distillation problem; Sulaiman is working on ethanol/kerosene/paraffin and I'm working
on turpentine.
I have plans to post a guide here after I've worked through it. But given your expertise you'd probably accomplish this much faster than I. Right now
I'm hung up on slow internet, Debian Linux packaging issues with regards to the mono dependency of DWSIM, and conceptual understanding.
[Edited on 13-8-2018 by andy1988]
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
Hmm... DWSIM looks like an interesting piece of software.
Those are some good links too. Apparently, a typical 30 cm Vigreux has only 3 theoretical plates (plus one for the flask).
This document (which seems to be incomplete) suggests that the reflux ratio should be equal to the number of theoretical plates: https://willi4nd.wordpress.com/soal-soal-kimia/fractional-di...
|
|
|
BromicAcid
International Hazard
Posts: 3247
Registered: 13-7-2003
Location: Wisconsin
Member Is Online
Mood: Rock n' Roll
|
|
It varies from compound to compound and from point to point in a distillation. Best is relative, there is usually a tradeoff between speed and
purity. Generally when I was in production we based everything off head temp. If the temp started to come up then we were taking off too
aggressively. If it started to drop we needed more heat/stirring to help push up the column. These are general guidelines. Unless you are
distilling a pure substance your mole fraction of vapors will change over the course of the distillation making separation more difficult. You could
start off stripping a solution ~90% pentane taking off like gangbusters but by the time you're at 6% pentane your head temp is coming up like crazy
and you have to back off giving a higher reflux ratio and slower takeoff. Fractioning off 0.1% pentane from 69.1% to 69.0% is no problem but going
from 0.15 to 0.05% even with something boiling 120C above it can be a bear. You get a feel for it the more you do it. I know people that will GC
their takeoff every time they make and adjustment and others that will collect everything by 'feel'. It's a experience/comfort level thing.
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
I usually monitor head temperature while trying to keep the distillation rate constant. It often becomes necessary to accept a slower distillation
rate towards the end. If the reflux ratio is low but the head temperature is fine, I don't worry about it. I'm not sure if I should or not....
I used to actually check the temperature at intervals and record it with the time and power setting in a table, but I rarely do that anymore.
|
|
|
Sulaiman
International Hazard
Posts: 3695
Registered: 8-2-2015
Location: 3rd rock from the sun
Member Is Offline
|
|
Quote: Originally posted by andy1988  | This can be determined/tuned analytically using chemical process simulation software such as DWSIM (free & open-source). The analytical result could be used as a heuristic for later empirical testing/tuning, or the results used as-is.
|
I just installed the 64-bit Windows version of DWSIM,
does anyone have a simulation model that I can copy ?
Very steep learning curve 
|
|
|
Corrosive Joeseph
National Hazard
Posts: 915
Registered: 17-5-2015
Location: The Other Place
Member Is Offline
Mood: Cyclic
|
|
Sincerely.......I wouldn't think so....... After reading your postings in this thread.
https://www.sciencemadness.org/whisper/viewthread.php?tid=69...
/CJ
[Edited on 13-8-2018 by Corrosive Joeseph]
Being well adjusted to a sick society is no measure of one's mental health
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
