Sciencemadness Discussion Board - Insulation on fractional distillation?

As magpie correctly states, the column should be adiabatic. More expensive columns are vacuum jacketed and silvered for this reason. Some people here believe that this makes the column useless (I think the term used was "like a steam pipe"...). Clearly these people are unaware that for any real fractionation to occur you need a reflux ratio, provided by a splitting head.

The term 'like a steam pipe' was used by me and trust me, I'm NOT confused about the need for a splitter. But it appears others remain so, despite positive affirmations to the contrary.

So lets try an work this out for once and for all, shall we?

MOST people here, including myself, do not have a splitter head and yet use fractionating columns all the time, with great success.

Case I:

If you don't have a splitter head and run your column adiabatically you are simply NOT FRACTIONATING ANYTING, because all you do is send vapour through your perfectly insulated column, without any downward flow. Without mass exchange between upward vapour flow and downward liquid flow no fractionation can take place.

Consequence: if you don't have a splitter head, rely on NATURAL condensation of vapour, due to heat loss of the column. In plain English: in that case DON'T or POORLY insulate your column.

Case II: you do have a splitter head. Well, in that case, in order to control reflux ratio PERFECTLY also insulate your column PERFECTLY, as in that case the downward liquid flow is purely defined by the splitter head.

Conclusion: running the column adiabatically or not depends on whether you have a splitter head or not. But even with a splitter head not running perfectly adiabatically isn't necessarily a problem: it only means you won't control reflux ratio as well as you would in adiabatic conditions. And most distillers don't obsesses about controlling reflux ratio perfectly: they just want to fractionate well...

Vogel has a section on this, and I direct you to his text for details. From whatI recall the column is operated at full reflux until a stable minimum head temperature is acheived. This corresponds to the best separation possible of the most volatile component (that the column can provide...). Distillate is then collected whilst maintaining a good reflux ratio (9:1 or so...).

Vogel is not the best guide on this although what you quote is largely correct. Reflux ratios of 9:1 or so aren't practical in industry though: you'd hardly produce anything.

Instead do some searches on McCabe Thiele, to better understand how fractionating columns actually work.

A big problem here with those here who quote on adiabaticity is that they quote out of context: an adiabatically run column only works in tandem with an effective reflux controller, i.e. a splitter head.

I refer also to this recent thread where with a simple Vigreux without insulation and without splitter ethanol was rectified from less than 20 % to over 80 %, by a beginner (with my help):

http://www.sciencemadness.org/talk/viewthread.php?tid=33065#...

That result was also compared to an identical distillation but without Vigreux.

[Edited on 17-10-2014 by blogfast25]

blogfast25 - 17-10-2014 at 12:04

Quote: Originally posted by Actinium

DJF90- " The column should be equilibriated at full reflux for a period prior to collecting distillate to provide this thermal gradient." Does the length of time to reach equilibrium depend on the Column length? Diameter? Both? what you are distilling? or any of/all the above?
Its increadibly fascinating how something so ancient and simple as distillation has so many intricacies involved.

All of the above.

Today's distillation technology has evolved enormously with respect to 'ancient and simple distillations'...

[Edited on 17-10-2014 by blogfast25]

Magpie - 17-10-2014 at 12:57

A big problem here with those here who quote on adiabaticity is that they quote out of context: an adiabatically run column only works in tandem with an effective reflux controller, i.e. a splitter head.

You are probably referring to me. My quote from Cason & Rapoport indicated that the fractionating zone of the column should be as adiabatic as possible. This did assume that liquid reflux was provided from the top of the column (use of a splitter).

I agree that some condensation has to take place at the top of the column so that there is a liquid flow down through the column.

I think your Case 1/Case 2 analysis is correct - well done.

blogfast25 - 17-10-2014 at 13:07

[You are probably referring to me. My quote from Cason & Rapoport indicated that the fractionating zone of the column should be as adiabatic as possible. This did assume that liquid reflux was provided from the top of the column (use of a splitter).

I agree that some condensation has to take place at the top of the column so that there is a liquid flow down through the column.

I think your Case 1/Case 2 analysis is correct - well done.

Not only you, Magpie. In the discussion DJF90 refers to and which left a really bad taste in my mouth, others came up with the adiabatic requirement too. As said, without some source of condensate being sent down the column, perfectly insulated columns simply don't actually do anything. It's like sending steam up an insulated pipe. There are quite a few that don't seem to understand that simple fact.

Perfectly insulated columns with perfectly functioning and accurate splitters perform, all other things being equal, better than more primitive systems, I think that goes w/o saying. But look at moonshiners: from 12 % EtOH to 80 - 90 %, not too shabby, I say. Don't talk to these people about "reflux".

Magpie - 17-10-2014 at 13:41

For normal fractionations which aren't too challenging, say 30°C separation in bp or greater, I use my 200mm fractionating column, uninsulated, w/ broken glass or ss scrub pad packing.

For the challenging fractionations like carbon tet/chloroform (16°C of bp separation) I use a packed 600mm column, insulated, with a splitter head. I would likely also use this setup for separating fatty acids from a natural product like saponified coconut oil.

S.C. Wack - 17-10-2014 at 14:04

IDK what this splitter of which you speak needs to be, but the implication is that standard student glassware sets are incapable of fractionation if the column is insulated, period...Well, above the packing is a little more column and an adapter, a thermometer...and if the packing is insulated and even heated, and the bp's not particularly close, quite much of this reflux is in the column regardless of no lost heat, depending on how much of what you've got going.

It should be enough to say with ordinary glassware to maintain boiling and go for the reflux ratio dictated by the HETP of the column, insulated from the top of the flask to the top of the packing. It works. Really.

forgottenpassword - 18-10-2014 at 02:21

In practice, insulation is often needed with compounds that boil above 100C, since heat loss prevents them from reaching the still head.

blogfast25 - 18-10-2014 at 05:33

IDK what this splitter of which you speak needs to be, [...]

That explains rather a lot. Example:

http://www.sigmaaldrich.com/catalog/product/aldrich/z413097?...

I'm beginning to wonder if you know what reflux ratio actually means (see below).

but the implication is that standard student glassware sets are incapable of fractionation if the column is insulated, period...[...]

True. If nothing forms a condensate, there no liquid phase and no fractionating.

Well, above the packing is a little more column and an adapter, a thermometer...and if the packing is insulated and even heated, and the bp's not particularly close, quite much of this reflux is in the column regardless of no lost heat, depending on how much of what you've got going.

If that 'little more column' is not insulated it will act as a primitive condenser but it won't allow you to control reflux ratio with any degree of accuracy.

It should be enough to say with ordinary glassware to maintain boiling and go for the reflux ratio dictated by the HETP of the column, insulated from the top of the flask to the top of the packing. It works. Really.

Yes, if you have a splitter to send condensate back down, see top of this column:

V is the vapour flux (say mol/s) flowing through the (insulated) column. C is a condenser/splitter which splits the condensate into D (mol/s, distillate) and L (mol/s) the liquid flux you're sending down the column to cause fractionating.

R is the reflux ratio, V = D + L the mass balance at the top of the column.

The splitter is often a timed valve: for 'x' seconds it will send all condensate down the column, for 'y' seconds all condensate is collected as distillate. R = y/x.

Quote: Originally posted by forgottenpassword

In practice, insulation is often needed with compounds that boil above 100C, since heat loss prevents them from reaching the still head.

Yes but that doesn't fundamentally change anything.

[Edited on 18-10-2014 by blogfast25]

forgottenpassword - 18-10-2014 at 06:07

I dunno - it changes whether the column is insulated or not!

Magpie - 18-10-2014 at 06:48

A minor correction:

R=L/D

S.C. Wack - 18-10-2014 at 07:04

if you have a splitter to send condensate back down

In the simple glassware world, this is known as a thermometer bulb. I'm not saying this is perfect or wise, it is what can be done without whichever thingy you want to attach. The heat can be adjusted so that there is boiling and the reflux occurs, much of this at the top of the column and which you can see, which does not resemble what you have in mind or ever attain a reflux as you would like it. I'm not saying it's faster or as perfect as anyone would like, or how an engineer would design things.

[Edited on 18-10-2014 by S.C. Wack]

blogfast25 - 18-10-2014 at 11:24

A minor correction:

R=L/D

Correct, but I've seen the inverse being used too. V = D + L is the more important equation here.

blogfast25 - 18-10-2014 at 11:26

Quote: Originally posted by forgottenpassword

I dunno - it changes whether the column is insulated or not!

True but it doesn't change the reasoning here used.

S.C. Wack - 18-10-2014 at 14:57

IDK what this splitter of which you speak needs to be, [...]

That explains rather a lot.

In the simple glassware world, this is known as a thermometer bulb

I'm going to have to conditionally retract this, since instead of watching football like I should be doing, I'm writing and looking for a distillation "splitter" and still having no idea WTF. So I'm not calling anything a "splitter". Sigma-Aldrich may be the only manufacturer of what you're talking about calling it that or something, because what Ace sells as a splitter...IDK what to do with that...What I'm saying is, how much does the desired item cost? If you're running a 500 mm Vigreux, anything over a what, 5:1 ratio is useless...PS hmmm actually the reflux ratio can be changed using the top of the column and the adapter as a condenser, yes maybe not quite to the point of total reflux if you want your distillate today: a bag of water, just like how condensers used to be.

[Edited on 18-10-2014 by S.C. Wack]

Magpie - 18-10-2014 at 18:14

As blogfast shows in his sketch a splitter is nothing more than a condenser that can route part of the condensate to the top of the column. Splitters generally have the ability to vary the amount returned thereby allowing one to set the reflux ratio.

There are many different splitter configurations. I have the Hennion type as shown below:

The male end at the lower left connects to the column head. The two tublatures at the top are for the cold-finger type condenser cooling water. The male end at the lower right connects to the condensate receiver. The lower stopcock is used to set the reflux ratio. The other valves and tublatures are for use in isolation during vacuum distillations, I believe.

As to cost, yes, they are very expensive. I would not have purchased it except for the deep discount offered by Dr. Bob.

One certainly doesn't have to have one of these, as you say. But for difficult separations I would wager that most professional chemists do.

[Edited on 19-10-2014 by Magpie]

[Edited on 19-10-2014 by Magpie]

[Edited on 19-10-2014 by Magpie]

[Edited on 19-10-2014 by Magpie]

[Edited on 19-10-2014 by Magpie]

[Edited on 19-10-2014 by Magpie]

[Edited on 19-10-2014 by Magpie]

blogfast25 - 19-10-2014 at 04:39

Splitters generally have the ability to vary the amount returned thereby allowing one to set the reflux ratio.

There are many different splitter configurations. I have the Hennion type as shown below:

That's correct and lucky you to have one of those. I agree that for difficult separations of significant amounts of stuff it's more or less indispensable.

In the case of your Hennion gadget, clearly you can set reflux with it but do you actually know the value of the reflux ratio? Methinks not but please correct me.

Clearly if that bottom valve is closed you run at total reflux but if that bottom valve is fully open, does all condensate flow in to the distillate receiver? It's not clear from the photo.

As pointed out a much simpler geometry uses a timed valve. An overhead condenser on top of the column condenses all of V, then directs the condensate for 'x' seconds back into the top of the column and for 'y' seconds in to the distillate receiver. I've seen this used on a pilot plant rectifying column for high grade methanol I used to work on some 30 years ago, as a student.

The cost of these splitters is a problem (although the Sigma gadget probably isn't very representative). Someone above mentioned that eBay (UK) seller 'baffcat' sells a splitter for £99 but I could not find the item.

I'm also interested in SC's alternative that he wrote about a couple of posts ago but haven't really found anything. What he called a 'thermometer bulb'...

[Edited on 19-10-2014 by blogfast25]

Little_Ghost_again - 19-10-2014 at 06:16

I looked at vogel 3rd ed and the whole setup with variacs etc would be outrageously expensive to set up,
I did try to make my own with nichrome wire in sections wrapped around a column and temp sensors for each section. the whole thing microcontroller controlled.
The problem I had was this gave me a max temp of around 120c, also it melted parts of the insulation.
I have no idea if my other solution is considered any good or not but I have settled for the following.....
I have a clasian adapter at the top of the column, in the first port I have a small cold finger that I can lower or raise around 5cm each way, I get reflux on this and it seems to work ok.
Obviously if I could control the temp of the fluid used in the cold finger it would work alot better, but so far I havnt bothered. When I do any of my distillation I mess about until I get a fixed temp coming over, if it starts wondering slowly up now and then I lower the finger until it stays at a fixed temp.
All I can really offer is I made chloroform with IPA, I got chloroform come over at bang on its BP, then it climbed 2-3C after around 20 mins, I returned this to the flask and lowered he finger a bit, I did this several times and ended up with chloroform with a bang on BP.
I also ended up with 4 other distillates that I have no idea what they are! One day with better equipment I will analyze the buggers and see how good my method was.
A clasien adapter is pretty cheap and at a push I would use two thermometers one to reflux (ok a bit piss poor) the other to show the temp going into the condenser.

I am expecting to be told this is a bad way, but I did what I could with what I have. I have some ethanol screen wash to distill that should sort the men from the boys lol Lets see if I can get ethanol using that lol

S.C. Wack - 19-10-2014 at 06:38

I have a clasian adapter at the top of the column, in the first port I have a small cold finger

Excellent use of cheap glassware!

You would think this would be illustrated in the student lab manuals, but no, just the kind I indicated earlier...but then, they rarely mention insulation.

I'm also interested in SC's alternative that he wrote about a couple of posts ago but haven't really found anything. What he called a 'thermometer bulb'...

A more pure and true blogfast25 response could not have been composed, in response to the sentiment that distillation/column heads have traditionally been known as distillation/column heads and not as "reflux splitters", (at least outside of the arcane world of chemical engineering) until (apparently) recently.

With organic kit glass yes imperfect insulation could be desirable, but at least for the usual higher boiling distillates you're still going to need insulation. It would be wiser to have thin insulation all over instead of insulation only over part of the fractionating area.

Magpie - 19-10-2014 at 06:53

In the case of your Hennion gadget, clearly you can set reflux with it but do you actually know the value of the reflux ratio? Methinks not but please correct me.

You have to determine the reflux ratio by counting drops, both at the drip tip of the cold-finger, and at the receiver.

Clearly if that bottom valve is closed you run at total reflux but if that bottom valve is fully open, does all condensate flow in to the distillate receiver? It's not clear from the photo.

Yes. The drip-tip is directly above the opening of the tube going to the receiver.

As pointed out a much simpler geometry uses a timed valve. An overhead condenser on top of the column condenses all of V, then directs the condensate for 'x' seconds back into the top of the column and for 'y' seconds in to the distillate receiver. I've seen this used on a pilot plant rectifying column for high grade methanol I used to work on some 30 years ago, as a student.

Apparently that device works OK then? When I first read about it it didn't sound good. I thought that the constant flow interruptions would cause disturbances to the column equilibrium.

I'm also interested in SC's alternative that he wrote about a couple of posts ago but haven't really found anything. What he called a 'thermometer bulb'...

I assumed that this was SC's tongue-in-cheek description of a de facto condenser formed from the uninsulated column head/thermometer adapter/thermometer.

Little_Ghost_again - 19-10-2014 at 07:16

I often bodge glass ware to make an approximation of something I have seen in the books.
Mapgpie can you do a couple of clear pics of your please? I have an idea how I might be able to do that on the cheap

Little_Ghost_again - 19-10-2014 at 07:22

baffcat has them alot cheaper than £99 and no I am not telling until I can afford one lol.
Oh ok look up peking triangle in his shop, thats pretty close to what you want

Little_Ghost_again - 19-10-2014 at 08:03

I thought I might as well try and add something of use. I now have a real cold finger (no jokes please) but my first I made, I found a small thin test tube with a reasonable thick bottom, heated a solid glass tube and the bottom of the test tube together (like in the books) and pulled a bit, I wasnt after a cold finger but something else I had read in vogel, however the bottom of the test tube became pointed and the glass rod detached.
I was about to try again when I thought..............Hang on that looks alot like a cold finger, so I stuffed and sealed a two hole rubber bung in the top and used a pipette to draw a couple of tubes slightly thinner near the bottom. I then put one in the tube near the bottom and one near the top, the whole thing was then covered in araldite glue (no idea why it just seemed like a good idea).

blogfast25 - 19-10-2014 at 08:57

Apparently that device works OK then? When I first read about it it didn't sound good. I thought that the constant flow interruptions would cause disturbances to the column equilibrium.

Yes, it works well. I saw it used on a larger size column (about 20 cm diameter) but even on smaller columns those disturbances don't affect overall outcome much.

@LGA:

I think you mean 'Perkin (not Peking) triangle':

http://www.ebay.co.uk/itm/RA6-33-Intermediate-Perkin-Triangl...

But that's not a distillate splitter.

[Edited on 19-10-2014 by blogfast25]

Magpie - 19-10-2014 at 09:22

Here you go Little Ghost:

Little_Ghost_again - 19-10-2014 at 09:33

LMAO yeah thats what I meant! and correct after looking at it I realized I was talking shit sorry.
Magpie thanks alot, I am thinking that isnt too hard to approximate cheaply

. time to see how good my glass melting has got

DJF90 - 19-10-2014 at 23:24

@Magpie: Fractionating heads can get rather elaborate, and I've,seen some with "swinging bucketsv before. A bucket and chute is attached to the bottom of the reflux condenser, in a way that allows it to,swing back and forth. A magnetic strip is embedded in the chute, which normally sits above the column (liquid return). An electromagnetic controller is placed inline on the outside of the glassware, an operates on a x seconds on, x seconds off sequence (like PWM I guess...). When the electromagnet is on, the bucket swings across to position 2, where the chute sits over the distillate receiver. Its a very automated and accurate way of setting reflux ratio, but also an even more expensive setup. Normag make fraction dividing heads like this.

And especially for you, blogfast...111215225328

blogfast25 - 20-10-2014 at 04:58

And especially for you, blogfast...111215225328

Nice number but what does it mean?

Little_Ghost_again - 20-10-2014 at 09:09

probably a ebay number

Dr.Arz - 20-10-2014 at 10:02

Is it efficient to pack multiple glass fraction columns top by top, like two or even three 400mm? Can distilling speed be increased with keeping up the purity by increasing column height? How suitable stainless steels of 300 and 400 series are for packing columns?

blogfast25 - 20-10-2014 at 11:56

probably a ebay number

Yes!

http://www.ebay.co.uk/itm/111215225328?ru=http%3A%2F%2Fwww.e...

£97.50 and it still needs a condenser on top of the larger female ground glass joint. In fact it' not entirely clear to me how this thing's supposed to work.

blogfast25 - 20-10-2014 at 12:01

Quote: Originally posted by Dr.Arz

Is it efficient to pack multiple glass fraction columns top by top, like two or even three 400mm? Can distilling speed be increased with keeping up the purity by increasing column height? How suitable stainless steels of 300 and 400 series are for packing columns?

All other things being equal, longer columns have higher separation power. And as a consequence separation power can be sacrificed for speed with taller columns. But you need to be realistic with column diameter too: how much vapour can you send through an inch of pipe diameter without pressure build up, for instance?

Stainless steel? That will depend entirely on the stuff you're distilling. AFAIK everyone uses glass or ceramic packings.

S.C. Wack - 20-10-2014 at 12:10

AFAIK everyone uses glass or ceramic packings.

Naturally, I use SS. It may not be for the best, IIRC different materials in the same form do not separate the same.

Little_Ghost_again - 20-10-2014 at 12:54

probably a ebay number

Personally for that money I would stick to claisian adapter cold finger and the bit where the valveis I would just use a rubber bung with a glass stopcock going to another bung at the condenser if that makes sense, far far cheaper and exact same job. Only real difference is the swivel joint but I am not a fan of them

[Edited on 20-10-2014 by Little_Ghost_again]

Little_Ghost_again - 20-10-2014 at 12:59

I must spend way too long looking at porn on ebay! I need to get out more now I can recognize ebay numbers lol. If you want one message baffcat on ebay he is open to offers normally

Or use my method and pay me £5 license fee lol

[Edited on 20-10-2014 by Little_Ghost_again]

Fulmen - 20-10-2014 at 14:31

Question; What is better for a given column: A perfectly insulated column and top condenser or a perfectly tuned cooled condenser?

Magpie - 20-10-2014 at 16:04

I use broken glass packing or a stainless steel scrub pad. For the broken glass just take a thin-walled food jar and smash it to the appropriate size for your column. BTW, the ss scrub pad was first suggested by Louis Fieser I believe, as a cheap but effective column packing.

My experience is that the ss scrub pad is somewhat more efficient than the broken glass. But glass is more chemically resistant. I found out that the ss scrub pad cannot be used when distilling acetic acid due to severe corrosion.

If cost was not an object I would pack my column with this:
http://www.sigmaaldrich.com/catalog/product/aldrich/z210536?...

[Edited on 21-10-2014 by Magpie]

DJF90 - 20-10-2014 at 22:50

£97.50 and it still needs a condenser on top of the larger female ground glass joint. In fact it' not entirely clear to me how this thing's supposed to work.

Just like any other stillhead, it needs to be used with a condenser and thermometer. Receiving flask attaches to the joint the other side of the stopcock. Really not rocket science...

Little_Ghost_again - 21-10-2014 at 03:10

I use 2mm glass beads sold on ebay with a hole in the middle for packing, 20 different colours available apparently, I tried broken glass but couldnt afford the plasters, if I am lazy I use SS scrub pad thing, again watch the thicker one it slices your fingers open and sting worse than a paper cut.
I advise on plaster as some compounds make you hop about a bit if they get in a cut.

Fulmen - 21-10-2014 at 03:28

Choice of packing is usually a compromise between surface area and holdup, the perfect material should have a huge surface area but little volume to reduce holdup and risk of flooding. Raschig rings seems to be a good place to start.

blogfast25 - 21-10-2014 at 04:26

Naturally, I use SS. It may not be for the best, IIRC different materials in the same form do not separate the same.

Like Magpie I've used an SS scouring pad. Worked well but I'm not comfortable with any acidity in whatever you're distilling.

Just like any other stillhead, it needs to be used with a condenser and thermometer. Receiving flask attaches to the joint the other side of the stopcock. Really not rocket science...

I was looking at it wrongly: it looked to me as if the thermometer tube was passing behind the condenser tube but it's not.

A slight improvement would be to lower the connection between the bend in the condenser tube and the distillate tube. Also: needle value instead of T-valve, the latter are harder to adjust flow with.

[Edited on 21-10-2014 by blogfast25]

DJF90 - 21-10-2014 at 09:09

Glad you can make sense of it now. I personally think your suggested improvements are redundant: the distillate take-off tube is adequately positioned and I find the stopcock fine, although thats personal preference. The bore however is large enough to enable fairly fine adjustment, and groves can be made with a diamond file to improve control if required. The newer part by Quickfit does use a rotaflow valve, however I tend to avoid these where possible.

More beneficial would be shortening of the distillate take-off (reduce dead volume), possibly by removal of the ball joint (although this is useful for taking up some of the strain). The significnt disadvantage of this setup is the inability to operate at reduced pressure without bodging together several other components (a couple of 3 way taps, a straight vacuum receiver adaptor and a hose barb for atop the condenser), which will likely be a major ball ache. However for recycling solvents it would be more than adequate.

Considering the materials cost for the glass joints and stopcock, I'd say 100 quid is a fair price. But if you can get one like Magpie has for a similar price, or reduced pressure work is a necessity, then get that instead.

Little_Ghost_again - 21-10-2014 at 11:52

these rachel rings? they are little more than glass tube bits cut 2-3mm long arnt they?

I got a box of broken pipettes I could cut up thats got to be pretty close?

[Edited on 21-10-2014 by Little_Ghost_again]

Fulmen - 21-10-2014 at 16:10

Raschig rings are usually tubes, length rougly equal to the diameter. I've also seen rings that were split tubes and even some shaped like a "G", don't know if they are known under a different name. Something like a scouring pad can easily be packed to tight which can cause excessive hold up and flooding.

Famousroger - 21-10-2014 at 17:38

The answer is in practicality; insulation will heat column faster, or keep it from cooling quickly, although insulation almost always results in loss of observability. An appropriate set up merely needs to achieve a function as efficiently as practically possible; some procedures will benefit from insulation, some will not; IMO, qualitative observation is helpful, so if not necessary, I don't use insulation. That being said, ace wrap and a small space heater often works sufficiently nicely and the assembly is easy.

blogfast25 - 22-10-2014 at 04:21