Sciencemadness Discussion Board - How to adjust reflux ratio ?

How to adjust reflux ratio ?

Sulaiman - 21-9-2016 at 09:44

I have one of the typical Chinese '500 ml distillation kit'
http://i.ebayimg.com/images/g/w9MAAOSwLN5Wjjaf/s-l1600.jpg
plus an extra Leibig condenser (200 mm eff.) and two fractionating columns (250 mm etff.),

what do I need to adapt my glassware so that I can adjust the take off rate fron the fractionating column - reflux condenser junction?
Cheaper = better

Praxichys - 21-9-2016 at 10:23

The good news is that you don't need anything. You adjust the take off rate by adjusting the heat applied to the boiling flask or the heat lost by the column. The idea is to set up a thermal gradient.

More heat to boiling flask and thermal insulation around the column = higher take off rate but less separation

Less heat to the boiling flask and more cooling of the column = lower takeoff rate but better separation

Sulaiman - 21-9-2016 at 10:34

I thought that the idea is to set up full reflux in the column and the overhead reflux condenser,
with minimal temperature gradient in the column, using good insulation or an external heater
then take a small portion to the side condenser for product ?

P.S. I intend to do this to see how well I can do fractional distillation,
how many theoretical plates can I actually achieve vs. theory ?
An exercise in distillation more than general chemistry.

[Edited on 21-9-2016 by Sulaiman]

Maroboduus - 21-9-2016 at 10:56

I sometimes add a Claisen adaptor between the column and the stillhead(stillhead goes on the curved side) just to get a little extra condensation returned to the column. I've even gone so far as to wrap some tubing around the straight upright on the Claisen so I can get extra condensation which can be closely and quickly regulated with a valve between the tubing and the coolant pump. I know a cold finger condenser could be used there, but my cold finger is the wrong size for my still setup. Also, I'm not sure how easy it is to regulate the condensation rate with a cold finger.
Prachys, I'm glad to hear somebody say the condensation from an un-lagged column can improve the separation.
I had suspected that, but it seems to go against what it says in most of my lab manuals, which prefer insulating the column.
Perhaps insulation gets more important with really long columns or really slow distillations?

Maroboduus - 21-9-2016 at 11:28

Oh, I see. You mean like a partial take off head. I read the post too quickly.

The PTOs can be expensive, and I've wondered about rigging one up from standard parts as well.
Here's an idea I haven't tried yet: Put the flask in the mantle at an angle of about 30 degrees off horizontal. Put the column on it with a Claisen on top and the stillhead on the curved portion of that. put the thermometer adaptor on the straight portion of the Claisen and put the condenser on top of the stillhead where the thermometer adaptor normally goes. Put an addition funnel on the output of the stillhead where you normally put a condenser.

Now the condensate when running down the slanted condenser can either enter the collection flask or run back to the fractionating column depending on how the stillhead is rotated. Since rotating the stillhead so the condensate returns to the column will turn the addition funnel used for collecting the distillate on it's side you'll need to empty it before rotating the stillhead. (Maybe you could use a large flask instead of the funnel if you're not collecting too much distillate.)

This won't work with a vacuum unless you can drain the funnel into something like a Schlenk flask that you can evacuate to drain the distillate into and then break the vacuum on so you can remove it before trying to rotate the stillhead. But of course, using a large flask instead of the funnel, if possible, would solve that problem.

I hope this is clearer than it seemed to me while I was writing it!

EDIT: writing this out and reflecting on it a little made me realize this glassware setup idea has several real problems with it.

[Edited on 21-9-2016 by Maroboduus]

wg48 - 21-9-2016 at 13:32

Quote: Originally posted by Maroboduus

snip
Prachys, I'm glad to hear somebody say the condensation from an un-lagged column can improve the separation.
I had suspected that, but it seems to go against what it says in most of my lab manuals, which prefer insulating the column.
Perhaps insulation gets more important with really long columns or really slow distillations?

My understanding is if the column is insulated and the take off is 100% there may be very little condensate flow in column hence very little fractionation. If the column is then uninsulated and the heat increased to the flask to achieve the same level of take off, there will now be significant reflux in the column (due to condensation) and hence better fractionation.

Putting it an other way if the take off 100% (no reflux) and the column is perfectly insulated the column is not doing anything.

For the column to work you need a temperature gradient and condensate flow. Those can be provides by walls of the column and or the reflux.

In the Thermos insulated column the reflux provides most of the condensate flow and temperature gradient. Because of the higher condensate flow through out the whole column the separation is better than the no reflux case. But then how do you arrange the reflux ratio. The tilted adapter seems the simplest.

Magpie - 21-9-2016 at 14:01

A distillation column can have trays, sieve plates, or packing. For trays it is easy to visualize a series of equilibrium stages but the other types are equivalent. On each tray equilibrium is approached by some of the heavier component condensing and giving up enthalpy in evaporating some of the lighter component. Losing heat to the environment through the column wall decreases the efficiency of this process.

In the "Laboratory Text in Organic Chemistry," by Cason & Rapoport, (1950) the authors show how to build columns for difficult separations. These columns are not only insulated but have external heaters for close temperature control.

IIRC most all the outdoor industrial columns I have seen, such as at oil refineries are insulated.

Perry's "Chemical Engineers' Handbook," 3rd ed, p. 13-53 states "Design of small laboratory columns often presents a number of problems not encountered in the design of larger units. Enthalpy losses from a small-diameter column can represent appreciable percentages of the enthalpy in the vapor stream to the column. In such instances vapor and liquid rates are sharply reduced in proceeding up the column. Good practice requires prevention of such heat losses by the use of electric windings or vacuum jackets."

I normally don't insulate my short (200mm) fractionating column as long as an acceptable separation is achieved. However, I have done a few really tough separations (bp difference <10°C). In those cases I have used my 600mm column, insulated. I control reflux at the still head with a condenser/splitter.

[Edited on 21-9-2016 by Magpie]

[Edited on 21-9-2016 by Magpie]

Sulaiman - 21-9-2016 at 15:10

magpie, thanks,
I thought I understood what is required then got confused.
Following up I found this SM thread
http://www.sciencemadness.org/talk/viewthread.php?tid=10747
sure helps to find stuff when you know what it's called

The first unit I found cost USD2171.23 ... plus taxes and shipping

It does seem as if one of the Dean-Stark variations may work ... at my kind of price.

more reading and searching to do ........

Jekyll - 22-9-2016 at 21:07

One option to expensive partial-takeoff still heads that I’ve seen recommended is a short condenser, such as a 100 mm liebig, above the column with a small air or water pump and an adjustable pinch clamp to vary the flow rate and provide provide partial condensation (a water pump also lets you vary the water temperature in your reservoir). You could try using your 200 mm condenser above one or both of your columns, that would be the cheapest. A condenser with a drip tip joint would probably make it easier to calculate reflux to takeoff ratios.

Here's a fractional distillation summary and discussion of the technique. I wouldn't agree that Vigreux columns are "virtually useless," but packed columns do have their virtues.
http://www.q-re-s.com/chem/fractional_distillation.htm

[Edited on 23-9-2016 by Jekyll]

[Edited on 23-9-2016 by Jekyll]

Magpie - 23-9-2016 at 08:20

Than you Jekyll! That is the best write-up on fractional distillation that I have seen. What's the source. When I toggle Home I get an error message.

Maroboduus - 23-9-2016 at 09:45

I just wish that page said exactly WHAT aspirator that is. Did I miss it?
I get some decent results with my old one, but It'd be great to have some good info on which ones work best.
They're so reasonable price-wise for what they do that It's really worth it to spend a little more for the best.
I know about recirculating icewater for the aspirator, and plan on trying recirculating radiator coolant (maybe with ice or blue ice packs), but a good aspirator in the first place will mean I don't have to resort to those things as often, and when I do, I'll be getting the most out of them.

Great info though. I often need to see things explained a few different ways before I feel like I've got a handle on them.

EDIT And that link has a section on pyrolysis too! Cool ! (figuratively, of course)

[Edited on 23-9-2016 by Maroboduus]

Jekyll

Sulaiman - 23-9-2016 at 11:17

Thank You ! ... exactly what I needed.
My intended setup that requires minimum new glassware;

Thermometer
Still Head Adapter - product Leibig condenser - receiver/take off adapter - receiver
Reflux West Condenser
Insulated Fractionating Column(s)
Claisen adapter (just ordered one, £5.75)
Boiling Pot

Very tall in 24/29 !
(why did I buy such short cheap stands ? ...)

Waiting for 3mm glass spheres (not as optimistic about performance now)
and I guess some glass tubing chopping is comming soon .....

[Edited on 23-9-2016 by Sulaiman]

Jekyll - 23-9-2016 at 16:12

Quote: Originally posted by Magpie

What's the source. When I toggle Home I get an error message.

I'm also getting an error when I look for the /chem/ page, which is odd since it was there when I posted the fractional distillation link.

This organic chemistry primer refers to a "student of organic chemistry." However, I'm not sure if it's from the same source as the fractional distillation page.
http://www.q-re-s.com/chem/primer.htm

Edit: Here's the link I found earlier:
http://www.q-re-s.com/chem/chem.htm

[Edited on 24-9-2016 by Jekyll]

careysub - 23-9-2016 at 16:39

Quote: Originally posted by Jekyll

...
Here's a fractional distillation summary and discussion of the technique. I wouldn't agree that Vigreux columns are "virtually useless," but packed columns do have their virtues.
http://www.q-re-s.com/chem/fractional_distillation.htm

More puzzling than his complete dismissal of Vigreux columns is this:
Packing the column with coarse stainless steel 'pot-scrubber' material comes close to Pro-Pak, but the other popular alternatives, glass rings or short pieces of glass tubing (Raschig rings) will produce compromised results.

Really? Raschig rings are also bad? Also his enthusiasm for the metal "Pro-Pak" defies other guidance here that suggests stainless steel is useful only for unreactive distillations.

Pro-Pak is just pieces of 316 SS mesh:
http://www.sigmaaldrich.com/catalog/product/aldrich/z210536?...

316 SS mesh is readily available at low cost:
http://www.mcmaster.com/#stainless-steel-mesh/=14asjqr

[Edited on 24-9-2016 by careysub]

Panache - 27-9-2016 at 09:39

Wow, I thought it bog standard practice to have a small cold finger at the top of your completely insulated column that has the flow to it regulated with a hose clamp.
Your reflux ratio can then be tuned from 'full noise' (no flow in the cold finger) to almost complete reflux (full flow through the cold finger)

aga - 27-9-2016 at 11:00

Hmm. I thought it was all fine and froody with just a vigreux column.

Planned Distillation kit Setup ... please criticise

Sulaiman - 28-9-2016 at 02:45

Based on my existing knowledge and glassware, this is my planned distillation rig;

The Claisen adapter and 3mm glass spheres are on order, the rest is existing.

The two condensers have independant flow control
Initially the insulation will be glass wool, I am considering column heating.

I imagine starting with both condensers at high flow
then setting the pot heater to give maximum reflux without choking/flooding
then reducing the flow to the reflux condenser until the thermometer gives a sensible reading and product (very slowly) is collected.

First test will probably be just water
then I want to try to separate Toluene (bp 110.6) and Benzene (bp 80.1)
(I need to get benzene, maybe by distillation from an OTC product ?)

ALL CONSTRUCTIVE CRITICISM WELCOME / WHAT HAVE I OVERLOOKED

EDIT: two relevant documents sent to me by DJF90 (thank you

Attachment: Packing materials for fractionating columns.pdf (1.4MB)
This file has been downloaded 1518 times
Attachment: Simple calculation of theoretical plates.pdf (686kB)
This file has been downloaded 590 times

[Edited on 28-9-2016 by Sulaiman]

Sulaiman - 5-10-2016 at 10:11

I have made a few preliminary heat calculations for my still above;

to achieve 50 ml/minute at 19:1 reflux ratio:

Methanol, 230W to pot + 13W column loss
Water, 2 kW to pot + 50W column loss
98% H2SO4, 260W to pot + 265W column loss

so I will at least insulate the column,
later I may try heating to compensate for column loss.

Is it practical to get 2 kW into a 500 ml rbf ?

wg48 - 5-10-2016 at 15:09

Quote: Originally posted by Sulaiman

I have made a few preliminary heat calculations for my still above;

to achieve 50 ml/minute at 19:1 reflux ratio:

Methanol, 230W to pot + 13W column loss
Water, 2 kW to pot + 50W column loss
98% H2SO4, 260W to pot + 265W column loss

so I will at least insulate the column,
later I may try heating to compensate for column loss.

Is it practical to get 2 kW into a 500 ml rbf ?

I have a 3kW electric kettle with a capacity of 1.7l. at full boil the water volume increases by about 50% due to the vigorous bubbling accompanied by splashing and lots of vapour. I guess that 2kw into say 250ml of water would bubble out of a 500ml flask. A liquid with a lower boiling point and lower heat of vaporization than water would be more violent. So I think 2kW into 500ml is impractical.

An other way to look at it is the volume of the vapour 50mL of liquid turned into vapor has very approximately 1000x the volume. so thats 50l/minute or approximately 1l/s pushing out of just 250ml of liquid. You may need a 5 or 10l flask or a very very effective splash head. I don't see your column or condenser handling that volume either.

Sulaiman - 5-10-2016 at 16:44

OOPS ! ... well spotted,
sorry, I meant to write 50 ml/hour product at 19:1 reflux ratio,

After re-doing my calculations, now based on predicted maximum column capacity of 15.5 ml/minute liquid boiled;

Water: 583 W heat to pot, 182 W column heat loss if un-insulated
Methanol: 215 W to pot and 96 W column loss.
So my column must be well insulated, or heated.

allowing 1.5 hours for setup, warmup and clean up,
at 19:1 reflux ratio the maximum production rate would be 1 litre per day

I now realise that I can only collect different fractions if no vacuum is involved
and I want to change the glassware at the top of the column to allow the product condenser to be vertical,
making storage and mechanical support easier.

Basically, I am learning how Vogel evolved his still ......

a little bit of knowledge can be frustrating ...
I think that I shall just use this proposed distillation setup in lash-up mode to get more experience before attempting a 'nice' build.
I forsee many thermocouples being involved.
... Mk.II sometime in the future .....

P.S. I think that I fit the 'over-optimistic noob' profile quite well

P.P.S. For column insulation I am considering this http://www.ebay.co.uk/itm/271545073821?_trksid=p2060353.m143...
any reason not to ?

[Edited on 6-10-2016 by Sulaiman]

Chemetix - 18-10-2016 at 01:15

Hope this helps, but here is my reflux controller with a cold finger and thermocouple well; in one pic you can see the control taps for selecting how much goes into the finger or the condenser. On a good day it can do over 50:1 with something like cineole at 12mmHg, the moon ascendant in sagitarius...wind from the East, that sort of magic that somehow works on the day, 10:1 with something like ethanol on average-no vacuum.

Chemetix - 18-10-2016 at 01:17

Damn page with its' two types of "post" options...I'll get the hang of it- here are the pics!

Chemetix - 18-10-2016 at 01:47

Power calculations are based on specific heat of vaporisation of the liquid condensing. For example- Say 50:1 reflux ratio. 50 drops of methanol (say 1ml each) Std enthalpy change of vaporization, ΔvapHo MeOH +38.278 kJ/mol. 50ml is approx 40g, is approx 1.24n. should be 1.24*38.278kJ= 48kJ. At 2 drops a second there should be 48kJ over 25 seconds
48kJ/25s = 1.9 kW. So your heat source and cold finger should handle at least this power requirement for this reflux condition (i've ignored the 1 ml that comes over, it's tiny by comparison). When designing distillation equipment I favour the McCabe-Thiel method. Practice has taught me that for small set ups, 19/26 joints for example, theory means very little. You have an idea of what to expect but the reality is you deal with what the set up gives you in practice.

Sulaiman - 18-10-2016 at 01:50

Thanks for the photo' and info.
but
you have given me more questions than answers

based on your actual setup vs. my untried theoretical setup;
. my Leibig has a lot more surface area than your cold finger, any problem ?
. why do you not insulate your column ?
. is your packing porous, if so why ?
. what is the steel wool at the top of the column for ?

I ordered my glass spheres via eBay 5Sep. and still not arrived, if not here by Thursday I'll start a claim

Chemetix - 18-10-2016 at 02:44

Ok, I've used an uninsulated column because it has a sort of default amount of heat loss all by itself, that means the finger doesn't have to do all the work. Your leibig by itself is a lot of cooling, maybe too much, you might just end up sending everything back down the column even with very little flow through the jacket. It presents the system with a very high gradient of heat at one section of the set up, I think I prefer a more constant heat gradient over the reflux media (in the column) and the finger controlling the last bit of vapour about to come over.

With a liebig you have to really power the flask with enough boiling to get through the huge cooling the condensor can give you. It becomes a brute force game of boiling vs cooling rather than allowing the column to experience the greatest amount of area exposed to evaporation and condensation thus giving the high plate count. Flooding the reflux column just sends rivers of liquid down the path of least resistance with little chance of taking that portion of vapour that was on the cusp of condensing back down the column. In this arrangement I've use clay balls used for grow media that have very high surface area and filled with pores. More like a HPLC particle. The stainless steel wool just keeps it in place and also stops the drips from the cold finger from finding a single path to flow down the spheres.

There's a sort of art and feel to designing glassware that the formula just can't give you.

Maroboduus - 18-10-2016 at 07:56

Perhaps you could reduce the effective condenser length in this way:

Run the coolant into the top of the condenser instead of the bottom, and put a valve on the outlet hose for the condenser.
By partially opening or closing the valve you can control how high the column of water in the condenser is(as long as the inlet flow isn't too great).

If this works you could have any given length of your stillhead condenser filled with water by manipulating the valve.

It would be a further complication to add this, but by putting another valve on the condenser inlet, you could not only control the water height in the condenser, but also control it's rate of flow independently of the height. This would allow you to control the temperature of the coolant in the condenser and monitor it at the outflow.

Although I suppose it may be simpler to just use the condenser with a conventional bottom to top water flow and to control the input of water with a valve to control the temperature.

I really don't know if the shorter effective condenser would be better than the warmer condenser. In fact perhaps the warmer condenser would be best since it would return condensate to the column at a temperature closer to the boiling point of the liquid, which would reduce heat requirements of the column.

It also might make the column more efficient if it causes less disturbance in the column's temperature gradient.

XeonTheMGPony - 18-10-2016 at 09:45

I just jumped to answer the original question with my answer here, but I do encourage people to ensure to read the whole thread as I tend to post things in the simplified concept

Distillation has been one of the more gratifying things I have don from alcohol to other solvents to sulfuric acid.

It is all about surface area, temperature gradient, driving heat.

we can control the temp gradient by controlling coolant flow in a condenser column, or by heat loss from the column when air cooled by increasing the surface area exposed to free air.

Surface area is fixed, but we can manipulate heat lose with insulation to facilitate getting the temp gradient we desire.

Driving is the heat we input into the system via the boiler, you drive it hard you get a fast distillation but then lots of overlap of your fractions, drive it gentler you get slower distillation but much better divide of you fractions.

So it comes down to 2 simple variables: Driving heat, column heat gradient.

So the real and only trick to precise fractional distillation is in the configuration of your apparatus.

So a 200mm Vigrux column, a 200mm Allihn condenser then your main condenser.

You set the temp of the Allihn just a bit higher then the fraction you are targeting so only the fraction you want can pass over tot he main one, some times you just need the one reflux, for higher boiling point substances you can just manipulate the driving heat and let the flask act as the fractional column.

So you need very little gear to achieve very good distillation quality so long as you under stand there is 2 points that you must control, driving temp, temp gradient.

For more complex you can add a third variable, Vacuum

And if you want to get serious even more, sweeping gas as the 4th variable, But for the majority of your distillations you'll mostly just have the 2 to deal with and rarely the 3rd

1: Driving heat
2: Temp gradient leading to main condenser
3: Vacuum (This just effects the two variables above)

[Edited on 18-10-2016 by XeonTheMGPony]

Ditillation theory with reflux

Chemetix - 18-10-2016 at 14:41

This PDF is part of a write up I did for ethanol distillation. But I explain the design of distillation columns with reflux and the theory behind it.

[Edited on 18-10-2016 by Chemetix]

[Edited on 18-10-2016 by Chemetix]

Attachment: DistillationTheory.pdf (963kB)
This file has been downloaded 597 times

Bugger

Sulaiman - 22-10-2016 at 03:54

Glass beads never arrived,
I got a full refund and ordered more from a different supplier today.

Balls !

Sulaiman - 25-10-2016 at 03:27

I waited a month for 3mm glass spheres that never arrived, then got a refund,
I ordered from a different Chinese supplier a few days ago,
item marked as shipped, then I get this message:
..........................................................................................
Thanks for your message we just get the response that the item BOLI00036-150G*1;
BOLI00036-150G*2 sent to your counrty have been blocked by our local customs. we are not sure how long will they let the item go. but don't worry, we promise make every customer satisfied , we know you like our items very much , so we can refund money , , is that OK for you?

Much appreciate for your understanding ,because we can't control the customs.
waiting for your reply .
....................................................................................

How can 3mm glass spheres be of interest to the Chinese export customs department ?

j_sum1 - 25-10-2016 at 03:37

Who says the message they gave you is anything closely approximating the truth.

Maroboduus - 25-10-2016 at 19:03

If you keep having trouble, Walter Stern Co sells flint and lime glass beads in various sizes on Amazon.

They ship out of New York, so no Chinese customs troubles. But they sell them by the pound for $18.

Don't know if that's a lot more than you're looking for.

I also don't know how long shipping from NY to Oz takes.

I always thought Oz was a part of Washington DC, but most of my literary friends disagree with me on that.
Political symbolism can start a lot of arguments.

EDIT: Oops! YOU'RE not in Oz, that's J_sum1.
New York to England's just a long, cold swim!

[Edited on 26-10-2016 by Maroboduus]

wg48 - 25-10-2016 at 21:30

Quote: Originally posted by Sulaiman

I waited a month for 3mm glass spheres that never arrived, then got a refund,
I ordered from a different Chinese supplier a few days ago,
item marked as shipped, then I get this message:
..........................................................................................
Thanks for your message we just get the response that the item BOLI00036-150G*1;
BOLI00036-150G*2 sent to your counrty have been blocked by our local customs. we are not sure how long will they let the item go. but don't worry, we promise make every customer satisfied , we know you like our items very much , so we can refund money , , is that OK for you?

Much appreciate for your understanding ,because we can't control the customs.
waiting for your reply .
....................................................................................

How can 3mm glass spheres be of interest to the Chinese export customs department ?

I had a similar experience with glass syringes but they are still advertising them for international sale.

I have Balls !

Sulaiman - 27-11-2016 at 03:01

300g of 3mm glass balls arrived - they really were stuck in customs !
300g occupies approximately 180 cc
so I ordered another 100g so that i can fill both columns and have a little spare.

Assuming that they would never arrive, I ordered 2-7 mm 'rose quartz' tumbled chips,
unfortunately the rose colour appears to slowly dissolve in HCl !
and the quartz begins to look a lot like tumbled broken glass
I could have used just tumbled glass chips if there was no pink shit !

I also ordered a new still head adapter to convert to a gas/vapour variable takeoff head
3W_Still_Head.jpg - 23kB

it would be between the top of the fractionating column and the reflux condenser.

I am considering a magnetic stirrer ptfe disc in the side arm, adjusted by an external magnet
.. could be frustrating to adjust
or an adjustable glass rod from above with its end flattened to obstruct the side arm entrance.
... would be frustrating to adjust as the glass rod would have to pass through the reflux condenser.

Whatever method is used, I want it to be chemically inert, vacuum-sealed, and cheap

Any suggestions ?

P.S. thanks to all who have guided me,
I am unable to stay vertical for extended periods yet, but hopefully next year I will 'publish' some results.
P.P.S. before you ask - I have asked the supplier for a new invoice.
P.P.P.S. it's all been done before - at least I'm learning

bitten off more than I can chew (afford) ?

Sulaiman - 21-12-2016 at 07:11

being physically unable to do any testing yet, I've been doodling with numbers for my proposed fractional distillation rig,
and I do not like my results,
(to avoid confusion, my column would be 2x FC7/23 columns stacked vertically (upper and lower) to form one fractionating column)
based on: pot + lower collumn + upper column = 1 + 6 + 6 theoretical plates, with a 19:1 reflux ratio at 1 atm, fractionating EtOH:H₂O

starting with anything above 2% EtOH, the top of the lower column (= bottom of upper column) should have >= 90% EtOH at <= 78.4 C
the top of the upper column should be near azeotropic, 95.5% EtOH at 75.1 C.
so
the temperature difference between the top and bottom of the upper column should be c0.3 C
which I believe would require a very precisely controlled upper column heater,
and good heating control or very efficient insulation for the lower column.

so my aspirations have dropped quite a lot ... still learning

(in this case, calculate first, buy later)

I'll start with a simpler setup then work my way upwards slowly.

at least now I know why I do not see videos of amateur/hobby fine fractionating columns.

wg48 - 21-12-2016 at 08:38

I think it depends on the level of reflux versus the heat loss of the column.

So for example if the heat loss from the column is large compared to cooling due to reflux. Then small changes in the heat loss of the column will significantly change its temperature and could have a large effect on the reflux level and could stop it completely. No vapour reaches the top.

So the simplest method is to allow the reflux to regulate the column temperature which implies the heat loss from the column must be small compared heat loss due to the reflux.
Which can be achieved with an insulated column or heated column but not temperature controlled. In effect the column heater only reduces the heat loss of the column while the boiling point and reflux determine its precise temperature.

Sulaiman - 21-12-2016 at 09:44

yes, I think that I mostly agree with you,
but if a column heating jacket is at say 5C below top of column temperature
(as recommended by Vogel, page 101, Practical Organic Chemitry, in SM library)
then draughts will change the heating jacket temperature and upset equilibrium,
so many things need to be controlled and monitored (for unattended operation) that I expect this project to be spread over a few years

meanwhile I'l just slowly develop as I go along ... no more grand projects that I abandon before finishing.

wg48 - 21-12-2016 at 11:00

I agree ideally the heated jacket temperature (not the internal column temperature) can be controlled to reduce the effects of fluctuation in ambient conditions.

The key point is the jacket does not need to be regulated to a fraction of a degree provided that it’s the boiling point and reflux that primarily control the internal column temperature.

Perhaps the best option would be an insulated column with an outer temperature controlled jacket with a view slot.. This would reduce the need for good insulation and reduce the effects of ambient conditions while allowing the column to control its temperature.

Sulaiman - 21-12-2016 at 12:03

O.K. I get you, stable, not neccesarily accurate.

I realise also that my vacuum system (cobbled together pumps, pipes, gauges etc.) and pot heating also need upgrading,
Even physical support structures have to be made, the rig would be my almost height !
I do not need a fume cupboard, I need a fume wardrobe

Sulaiman - 24-12-2016 at 08:43

sorry to double post but the surprise arrival of a reflux condenser this Christmas eve has morraly obliged me to push ahead with this project
... thanks :cool:

<<< literally and figuratively <<<

Sulaiman - 20-2-2017 at 06:22

Set up for fractional distillation of c1.65 litres of 56 %ABV EtOH

Spent about two hours setting up
another two hours balancing heat vs. column flooding
my alfresco lab is so cold that even with cooling water at 1 drop per second, I can't get a drop of distillate,
total reflux on the outer wall of the reflux condenser ... to be continued ......

[Edited on 20-2-2017 by Sulaiman]

Chemetix - 20-2-2017 at 13:53

What a set up! Ahh the giddy joy of a nice bit of glassware.

The reflux condenser, which you've already identified as the problem, is too big. It will need heavy insulation in cool weather. And the flow through it may I suggest, needs to be restricted.
A tee-piece on the inlet of the downward condensers, branching to feed the reflux condenser with an inline valve will be what you need. It keeps the condensate in contact with optimum cooling but allows the reflux rate to be controlled.

If you weren't wanting fractionation and wanted to just bring over the distillate as quick as possible I'd say the two liebigs in series would be optimal. But with slow reflux, only one would be needed. For ethanol that is, ether or petroleum spirit would need them both under any condition.

Sulaiman - 20-2-2017 at 15:58

I agree about the two downward condensers ... overkill, but why not ?
I wrapped the reflux condenser in foil after the above post and got a couple of drops in the product condenser.

The main problem has been column flooding,
not achieving reflux rates that I'd expected,

as I was pondering how there could be more refluxing at the top of the column than at the bottom,
the contents of the columns bubbled up into the reflux condenser - the first of several floodings

I think that the bottom of each column, where the glass spiral is, is initiating the problem,
I will be investigating this ...
This setup is temporary until I get my diy 500ml heating/stirring mantle done.

my secret santa reflux condenser is so much more efficient than the Leibig/West types that I am used to,
all good learning .........

I'll try a simpler fractionation using one un-insulated column and no reflux condenser in the next few days,
just to learn more about the operation of columns,
I'll come back to this setup later .......

P.S. Siphoning cooling water from my upstairs toilet cistern works really well to provide constant cooling water pressure, hence flow rate.

[Edited on 21-2-2017 by Sulaiman]

tsathoggua1 - 21-2-2017 at 14:24

Partial takeoff head...that the same piece of glasswar as is referred to as a 'cow' ?

Sulaiman - 22-2-2017 at 11:59

No,
a partial take off head is used at the top of a fractionating column
to allow/take a small quantity of product and return most of the refluxed liquid to the column.
It is used to set the reflux ratio.
Industrial systems have a reservoir of condensed vapours from the top of the column, and a little liquid is periodically removed as product.
Amateur systems tend to condense most of the vapours as reflux, allowing a little vapour to pass through the reflux condenser to the product condenser.
This is the scheme that I am attempting - so far poorly.

a pig (or cow) allows one of multiple receivers to receive distillate,
the cow (or pig) can be rotated to select the next receiver for the next fraction,
only required if the still has to exclude air or completely contain vapours and gasses. (e.g. vacuum distillation, distilation of toxins etc.)

____________________________________________________

I think that 3mm is too small for glass balls for fractionating columns = terribly easy to flood... to be continued .......

P.S. when leaving a fractional distillation to continue unattended outdoors,
check that rain is not due ......

the collection bottle is full and the funnel is filling up,
I blame the rain for the poor photograph.

[Edited on 22-2-2017 by Sulaiman]

just an update on my distillation rig

Sulaiman - 4-6-2017 at 19:24

To allow semi-unattended operation I bought six used peristaltic pumps via two eBay auctions from the same seller,

All six came from a cow-milking assembly,
the motors, gearboxes, rollers and housings seem OK

I ordered 1m of 3.2mm i.d. x 1.6mm wall silicone tubing to replace the worn and swolen existing pumping-tubing
and 3x 8A (overkill) pwm controllers for the 12 Vdc motors (£6.90 total).
With 3.2mm i.d. tubing, 1 ml per revolution is pumped
and the motor and gearbox give 125 rpm at 12 V.

I should be able to use one pump for the reflux condenser, one for the product condensor and one as a vacuum pump for reduced pressure distillation.

I was pleasantly surprised by the vacuum pump potential of peristaltic pumps like this,
not a good flow rate but quite respectable pressure reduction

The best part ... all six pumps for £14 inc. p&p

Now the main thing needed to get up and running is a tall fume-cupboard/wardrobe

gatosgr - 27-10-2018 at 21:17

I am searching for a reflux condenser for a reaction that takes 10-20 hours in alcoholic solution, what kind of reflux condenser will I need? The volume is 100ml. Reaction temp is at 70C but it may have to go 100C in the end for 30 mins. I'm thinking of buying it from ebay or aliexpress. Thanks.

How do you call the pyrex dishes for the oil bath? I can't find them on ebay.

[Edited on 28-10-2018 by gatosgr]

Sulaiman - 28-10-2018 at 01:56

This is the condenser that I got from my clever 2016 secret santa and is shown at the top of my column in the earlier photograph and drawing.
It is efficient but the name is wrong, it is more like a Dimroth than a Graham
https://www.ebay.co.uk/itm/200mm-24-29-Glass-Coiled-Reflux-C...
Slightly cheaper but similar, an actual Graham condenser
https://www.ebay.co.uk/itm/230mm-Graham-Condenser-24-29-Join...

You can use a domestic Pyrex bowl, pot or dish
(some pyrex, especially from USA, is not borosilicate glass, just toughened soda-lime glass)
a glass bowl for an oil bath is nice as you can see through to the flask contents,
but I prefer a metal bowl just because I'm less likely to break it when washing or storing it,
and there is no chance of it breaking during operation.
Any domestic cooking vessel of suitable size and shape will work.

markx - 28-10-2018 at 10:38

As far as fractionating alcohols or other noncorrosive organic solvents goes I must point out that using a glass set up (especially with a fractionating configuration) is most unpleasant, although visually appealing and will mostly end in shards, tears and sobbing. I highly reccommend to use a metal setup for these operations. It is sturdy and forgiving....under some circumstances also a lot more economical.
For example if one has access to a lathe and knows how to use it.

One can start out with scrap materials like thusly:

And create quite an appealing setup from all of this (the reflux head taking shape):

A window section shall be incorporated a the base of the reflux head (in the last picture it has been substituted with a stainless section between the flanges):

The parts dissassembled and a shot of the column packing:

It's a work in progress, but it shall be a pilot production/products develpment fractionating column once completed....

The only glass part shall be a 5cm quartz section below the reflux head. The rest is stainless 304. Sturdy and functional without the dangers of shattering on first contact

Sulaiman - 28-10-2018 at 16:31

Although the still that I am building is also a work in progress,
I chose all glass so that I could watch and learn and ponder

It turns out that ALL parts with upward flowing vapour need to be thermally insulated
so I see nothing without disturbing equilibrium a little by opening thermal insulation to have a peep

SWIM - 28-10-2018 at 19:15

Have you considered spiral prismatic packing?
I just picked up a liter of 3 mm for like $25.

It's much less likely to flood, and it's supposed to be very efficient.
There's a lot on Ebay from the Russian Federation. (Stainless steel, various sizes )

If you need something more corrosion-proof, Nanshin now sells glass coils (12 x 4 mm labeled , "glass coil spring filter use for chromatography.) ) which look identical to the ones Synthware sells as distillation packing.

They are more fragile though in my experience, but cost a bit less. I can get 200 grams here for $23.

Synthware says such coils are intermediate between Raschig rings and single turn helices in efficiency.

That sounds pretty good to me. Have you ever priced single turn helices?

A Clevenger trap can act as a partial take off head. Really more like a solvent distillation head with a tiny reservoir, but it should work.

macckone - 28-10-2018 at 20:07

A clevenger or dean-stark w/ take-off is probably the best way to go.

There is an alternative though using one of these valves:

https://www.ebay.com/i/152018907331

The valve can work with a regular distillation adapter and 15 degree adapter to make a clevenger.

markx - 29-10-2018 at 00:09

Quote: Originally posted by Sulaiman

Although the still that I am building is also a work in progress,
I chose all glass so that I could watch and learn and ponder

It turns out that ALL parts with upward flowing vapour need to be thermally insulated
so I see nothing without disturbing equilibrium a little by opening thermal insulation to have a peep

Well in essence they do not need to be thermally insulated, but you got to have enough power pumped into the boiler to overcome the the thermal losses through the outer surfaces of the system. The whole apparatus partly works like a reflux condenser due to the heat losses....if insulation is applied the effect is minimised, but still there.
So if you wish to observe the operation of the column in the exposed state you have to consider a more powerful heat input. Then again if your column is of too small a diameter you might run into trouble with excessive vapour speed disturbing the smooth operation. This problem is especially apparent with small systems that intrinsically have a bigger surface covering less volume, hence the losses have a greater effect upon operation.
The geometry and nature of column packing is also very important in terms of reaching a stable operation with acceptable output and no sharp flooding conditions. The packing shall effectively rob part of the internal column volume and hence there is less free space for vapors to travel through. This shall increase the vapour speed and alter the pressure gradient in the coulmn leading to a condition where downcoming reflux is pressed up in the column by the high speed vapour coming from below....flooding shall happen.
Spheres or cylinders are by far the most restricting packing types and will effectively take up the most amount of free volume in the column, leading to flooding conditions at even moderate heat input. In theory you would want a packing that has a lot of surface area, yet lacks volume....something like thin walled glass tube sections, metal chips/scrubbers/meshes or some of the specialised geometry packings with saddle or other intricate shapes.
For practical purposes tube sections and metallic scrubbers probably are the most accessible and economical options. The specialised packing materials usually cost quite a bit if any considerable column volume has to be filled.