Glass Marbles for Reflux Column Structured Packing

[rad14701]

Do marbles prevent channeling (reflux running down the side of the column, instead of in the packing)? I know that can be an issue with mesh/scrubbies.

There should be less chance of channeling with marbles because they all have the same relative amount of voidage between them... Scrubbers tend to compact more in some areas and less in others which is a major contributor to channeling, IMHO... And if any reflux heads to the column wall it could potentially get redirected back inward as it meets the next marble and hugs onto the whetted surface...

[thecroweater]

Understanding and acceptance of Langmuir's Equation isn't mandatory, but denial of it's importance as the necessary foundation for any reflux-based fractionator is about as useful as declaring that the Moon is made of cheese.

What? I just looked that equation up, and didn't learn anything practical.
On a side note Odin claims that he does not etch his SPP for better performance. So whatever is going on there, may also apply to marbles. I wouldn't think etched marbles would perform any better/worse without collecting real world data. You know what they say about assumptions...

glad you added some humour to the thread guys, You are kidding , right
if you want to think that a highly polished ultra non absorbent surface with almost no ability to retain a liquid surface of any significance will work the same or better than the opposing proposal go right ahead (some folk think the world is flat and have some convincing psudoscience to back it up) Langmuir's Equation is one of half a dozen varying equations relating to the adsorption of molecules on a solid surface his theory biased to having a high concentration of a given gas to a high concentration of a given surface so really of no darn use to the average home distiller in determining what packing to use. really guys you would be able to contribute must more positively to threads like this by testing ideas collecting the data and reporting the result rather than zipping around the web to see if you can find some 98 year old theory and attempt to make it fit to disparage someone's idea or method just my two bob's worth anyway

[lampshade]

There should be less chance of channeling with marbles because they all have the same relative amount of voidage between them... Scrubbers tend to compact more in some areas and less in others which is a major contributor to channeling, IMHO... And if any reflux heads to the column wall it could potentially get redirected back inward as it meets the next marble and hugs onto the whetted surface...

Could channel minimization be the reason that marbles appear to perform better than scrubbers?

I don't know how to test that theory. Maybe run a mesh column with *many* centering sections to insure that channeling is minimized and see if performance is enhanced. But that would be a lot of work. I guess it's best just to accept that marbles work wonders.

[lampshade]

Does anyone know of a lighter-weight alternative to glass marbles? Ceramic raschig rings, maybe?

[thecroweater]

Maybe expanded Perlite

[Roddy]

Roddy, have you ever run marbles in a still?

Yes. And I evaluated (for the still in question) the best feed rate/marble diameter/reflux ratios too.
My conclusion was that like all packing materials, marbles have a "sweet spot" of operating conditions which needs to be "tuned into" to get the best results. Unfortunately, the "sweet spot" is quite a sharp peak, IMHO because packed spheres have constant surface area/void-space ratio. (but only until you wet them.... )
But I wouldn't set that opinion in stone. Marbles are a fine medium for those who take the trouble to get them working properly in their own still, and stay clear of the chain-reaction choking potential of overdoing throughput.

One of the major attractions of helices and SPP's, for instance (which I've also worked with) is that the "sweet spot" is much wider and easier to safely tune into. However, the cliff-edge at the high throughput end of that spot is pronounced - especially with SPP's in my experience - and a chain reaction of choking sets in. Think of it as "living on the edge".
For those playing around with SPP's, you have probably already noticed that the column starts to "sing" or sizzle as you get close to its porosity limit. If you hear that, you are probably already close to the choking point.

Having a column choke, aka spew, ("because we can") in a controlled lab or pilot production environment is a much safer proposition than risking that in domestic premises......... one of the first checks I do now that I'm prototyping at the domestic level is to ensure that if pressure (as with choking, for example) occurs unexpectedly, anywhere, there is some SAFE exit for it. So I have simple, cheap, vapour-lock U-tubes in strategic places. And their output pipes lead to safe containment vessels, NOT the floor or the ceiling!

[Roddy]

OK, first of all, I'm WAY out of my pay grade and in over my head on this one. But, I am trying to learn more about reflux so that someday I might build a column still. This thread seems to be generating a lot of current, useful info about packing material.

Roddy, you seem to be suggesting that if we all just bow down to Irving (Nobel prizewinner), and pay attention to surface areas, that's all there is to it, no more to think about, no ifs, no buts. I think most folks here would agree that a key element of reflux is the area of contact between surface of liquid and passing vapor, but as you begin to maximize that area, then the key areas shift to other factors, such as reduced vapor passageways, rates at which you can run the still, etc. As Croweater suggests, there are many things in play when trying to maximize efficiency in a reflux column. I don't think anyone here is trying to improve on, or trying to surpass Mr. Langmuir's design ideal, we're just trying to get as close as possible to it.

Roddy suggests:
"once we learn that the molecular exchange takes place solely at the liquid/vapour contact area, we can then, and only then, start to investigate and manage all of the parameters which can spoil the ideal case of a constructed, suitable contact area."

Agreed. Even I, a reflux noob, am prepared to accept that exchange happens at liquid/vapor contact area. So, now, and only now, am I prepared to investigate all the other parameters.

That's why I started following this thread in the first place, because Rad started experimenting with a packing that seemed to have been dismissed over the years. This thread seems to be getting a lot of response on the hows and whys of how various packings work, and how the efficiencies can be improved. It has at least really helped me get a better understanding of how the reflux columns work, and I'm grateful for all the back and forth, guys, thanks!

You're absolutely in synch.
The purpose in bringing up Langmuir was to invite awareness that the fundamentals were always at the forefront of everything we try to improve our "lot": if a packing material, for example, is badly designed, or pushed so hard that the optimum surface area OR void-space (for exchange) is compromised, then we are moving away from improvement. Which may not always be that obvious.......

I'd noticed that there is lots of focus on surface area of packing materials, but much less on the OUTER surface (the important one!) of WETTED material and the consequent reduction in void-space (porosity).

To be clear:
In no way are my posts intended to undermine the standing and respect due to Rad and other distinguished and longstanding oracles!

[Roddy]

Understanding and acceptance of Langmuir's Equation isn't mandatory, but denial of it's importance as the necessary foundation for any reflux-based fractionator is about as useful as declaring that the Moon is made of cheese.

What? I just looked that equation up, and didn't learn anything practical.
On a side note Odin claims that he does not etch his SPP for better performance. So whatever is going on there, may also apply to marbles. I wouldn't think etched marbles would perform any better/worse without collecting real world data. You know what they say about assumptions...

glad you added some humour to the thread guys, You are kidding , right
if you want to think that a highly polished ultra non absorbent surface with almost no ability to retain a liquid surface of any significance will work the same or better than the opposing proposal go right ahead (some folk think the world is flat and have some convincing psudoscience to back it up) Langmuir's Equation is one of half a dozen varying equations relating to the adsorption of molecules on a solid surface his theory biased to having a high concentration of a given gas to a high concentration of a given surface so really of no darn use to the average home distiller in determining what packing to use. really guys you would be able to contribute must more positively to threads like this by testing ideas collecting the data and reporting the result rather than zipping around the web to see if you can find some 98 year old theory and attempt to make it fit to disparage someone's idea or method just my two bob's worth anyway

You certainly know a bit about " zipping around the web to see if you can find some 98 year old theory and attempt to make it fit to disparage someone's idea or method".


Langmuir did LOTS of work on surface absorption and adsorption too. Little of that work directly applies to HD.

It's his work on evaporation which makes all of our attempts at fractionation work.
Read http://bado-shanai.net/Map%20of%20Physi ... ration.htm" onclick="window.open(this.href);return false;" rel="nofollow

I repeat, it is not mandatory to follow such science. Or understand it.
I don't mind if your Moon is made of cheese.
Mine isn't.

[thecroweater]

Look champ I'm not sure where your coming from. I read that some 25 or 30 yrs ago, great old theory from a fairly cleaver chap but it has approximately zip to do with weather a strata like say marbles are better smooth or scuffed and your statement that it is the only thing that effects reflux in a distillation column no ifs or buts is complete rubbish.
Any sense of your preponderance you hold will make as little difference to that fact as your animadversions of my posts or aspersions to my aptitude of distillation. You seem to have a propensity to reading, may I suggest you could "possibly" greatly benefit from reading here and the parent site should you choose to pay them credence

[FullySilenced]

Well Croweater... take a marble a day out of your column and throw it away....

When you've lost all your marbles only then will you become the distiller you want to be....



Happy Stillin,

FS

[MichiganCornhusker]

Roddy, sorry, but your posts are coming across as arrogant and antagonistic. It's getting in the way of the premium knowledge that you are trying to share. I know that when you start throwing out axiomatic statements, and call us cheese mooners, it gets my hackles up.

BUT, I went back, and I do think I understand, now, what you were trying to clarify in the beginning of all this: that what matters when considering the surface area available to reflux, we need to be looking at the surface of the liquid, not the surface of the medium, and that the surface of the medium only needs to be roughed up enough to provide a uniform covering of liquid. I've put together a sketch to try to illustrate what I'm saying:

The drawing is meant to show marbles with different surface finishes: smooth polished, etched, and something worn, more aggressive than just etched.

As mentioned above, it seems like a smooth polished marble won't have enough "cling" to keep a continuous coating of liquid over it's entire surface. So, the idea of etching. This would provide the "cling", and allow the liquid film to cover the entire surface of the marble. I think what you were saying is that once you achieve an effective "cling" surface, and a continuous liquid film, no matter how thin that liquid film is, there is no reason to go any further with abrading the surface of the marbles. I was having trouble making the distinction between surface area of the medium, and surface area of the liquid film. My understanding, now, is that, in my drawing, the etched marble and the worn marble would be equivalent because the outside surface of the liquid is the same in each case. That even though the worn marble would have more surface area than the etched marble, that doesn't affect the ultimate surface area of the liquid film. That it is the very outside diameter of the liquid film that determines the effective diameter of the medium. It took me a few days for that distinction to seep through the thick protective "slow" meat layer wrapped around my brain. I realize this is probably reflux 101, and I should know this before commenting, but I just want to make sure I'm interpreting this correctly.

Sorry if I'm jumping in out of place here. I realize I don't understand enough about the reflux process to be engaging in the threads, yet, but I just want to make sure I'm understanding this one point. The idea of what's happening in a reflux column is fascinating to me, and I'm trying to start getting a handle on it. Thanks for the info, Roddy, it is helpful, but I think it would be more productive if you could ease up on the tone....

[FullySilenced]

My suggestion just lose your marbles....

Scoria...

FS

[emptyglass]

Roddy, sorry, but your posts are coming across as arrogant and antagonistic.

we need to be looking at the surface of the liquid, not the surface of the medium.

two good points you make Michigan. The surface tension of the liquid has more bearing on the performance of the packing.


Can we get back to what this threads about?
I had lost interest in marbles long ago, but I'm interested to see where all the experiments are going. Everything old is new again. Rad put a good bit of work into documenting and sharing his findings, very interesting stuff.

[rad14701]

I may try either using a rock tumbler or etchant to see if the roughened surface makes any significant difference... I'd ballpark the amount of whetted area to be roughly doubled, at the very least... I guess my only concern is whether all those nooks and crannies will make the marbles harder to keep clean or not... Considering how I have almost 4X as many marbles as my column requires, doing some experimentation surely won't break the bank...

My first impression is that the rough surface will slow the reflux and cause more reboiling higher in the column... Whether that theory plays out to be factual in practical application remains to be seen...

[Bushman]

My first impression is that the rough surface will slow the reflux and cause more reboiling higher in the column... Whether that theory plays out to be factual in practical application remains to be seen...

This is the reasons I etched my copper packing!

[thecroweater]

yes for sure there is that and I tend to think the depth of the liquid layer plays a fairly profound part as well, in the same way a bath depth on a plate does. Back when i was using stainless scrubbies I found shiny new scrubbies improved once they were scuffed up a bit, and so I would expect the same result from any impervious strata including marbles

[FullySilenced]

Glass bead or sand blast your marbles if you must....

[MichiganCornhusker]

This is a great conversation lets try and be civil to one another!

I had lost interest in marbles long ago, but I'm interested to see where all the experiments are going. Everything old is new again. Rad put a good bit of work into documenting and sharing his findings, very interesting stuff.

Thanks, all, I'm getting a lot out of this thread.

[Roddy]

Roddy, sorry, but your posts are coming across as arrogant and antagonistic. It's getting in the way of the premium knowledge that you are trying to share. I know that when you start throwing out axiomatic statements, and call us cheese mooners, it gets my hackles up.

Apologies; my disdain was aimed in one direction only, and not towards you!
My own hackles were raised at the suggestion that such an eminent American scientist as Langmuir was in fact somehow wrong!!!

You are - as we say in the UK - "cooking on gas" and very much in synch with the general idea.

It is obvious that most here have also grasped the point: that all the Langmuir "stuff" is merely a reminder that something fundamentally important lies underneath everything we explore in discussions like this. To make progress (and progress in the discussion does seem to be blossoming!) we only need to accept that anything and everything beyond the foundations is "up for grabs" and we can speculate & then investigate on anything we find interesting.
Which is what I believe you do, and I do too.

I very much like your sketch, which serves to highlight the fact that a rough surface can help establish and stabilise a coating of liquid - especially if that liquid has a high surface tension and doesn't spread easily onto a surface, but, thereafter it is only the outer surface of that established liquid coating which interacts with anything.

Nice one.

I tried a similar sketch using 2-D packed marble sets - one marble on the top sitting on two marbles on the bottom. I drew three sets of these.
(I'd post it but don't know how to do that here, yet.... )

Anyway, I drew a thin coating of liquid onto the second set, and a thicker coating on the third set.
The marbles themselves stay in the same respective positions.
I see a dramatic closure of the void-space between the marbles! which helps visualise the rapid rate at which "choking" begins and explains why it tends to become a chain-reaction thing too.

Do please consider adding such a drawing in your next post, if you see what I saw!

[MichiganCornhusker]

Apologies; my disdain was aimed in one direction only, and not towards you!

Anyway, I drew a thin coating of liquid onto the second set, and a thicker coating on the third set.
The marbles themselves stay in the same respective positions.
I see a dramatic closure of the void-space between the marbles! which helps visualise the rapid rate at which "choking" begins and explains why it tends to become a chain-reaction thing too.

Do please consider adding such a drawing in your next post, if you see what I saw!

No need for disdain, my experience here has been that everyone has been very helpful and encouraging. If something seems disrespectful, it is usually a misunderstanding.

Here are a few drawings showing how quickly the voids between the marbles can be filled up with liquid:

I've drawn the surface of the marbles with a lighter line weight, and the liquid film layer with a darker line weight.
My scale is just random, and I leave it up to you all to figure if the thickness of the liquid films, as they relate to the diameter of the marble (or smaller sphere), make sense.
Perhaps what I am illustrating is a thickness of liquid that would not ever happen in practice. But as the diameter of the sphere shrinks, as stated above, the liquid layer thickness becomes more of a factor.

As Croweater suggests, "I tend to think the depth of the liquid layer plays a fairly profound part as well".

I would think things like surface tension, viscosity, "cling" would all factor into how the liquid interacts with the packing medium. So my question would be: as the whole vapor/liquid environment moves up the column, how much do these qualities change? For instance, at the top of the column, because the vapor/liquid is becoming more alcohol, and less water, does the surface tension of the liquid change enough to affect how it interacts with the packing? In other words, if alcohol has less surface tension, would the layers on the packing be thinner at the top of the column, and if choking does occur, does it tend to happen near the bottom, where the liquid films might be thicker? Or is this just not at all what is happening...

[thecroweater]

Apologies; my disdain was aimed in one direction only,
My own hackles were raised at the suggestion that such an eminent American scientist as Langmuir was in fact somehow wrong!!!

Hang on there chum I never ever suggested or in anyway implied that that equation was wrong but I don't agree that really relates to scuffing marbles further more i think one would be drawing a long bow indeed to say that mathematical equation means that a thin layer is better than a deeper one

Ihad a thought about improving the performance of marbles a couple of years back but never actually tried it or asked if others have. My idea was to put the marbles in a concrete mixer with fine gravel and a bit of water and run it until such time as to scuff the surface of the marbles. I was thinking this would improve their "wettability" no end and therefore cut down the amount of reflux cycles required for the same desired result obtained by un-modified marbles , Just a thought

The key element in all of this is the AREA of contact between the surface of the liquid and the passing vapour.
Increasing the surface area of the liquid surface bearer medium (e.g. marbles) at the bottom of the liquid layer (not the surface which meets the vapour) is of no consequence.
On the other hand, using smaller diameter marbles can provide a thinner liquid layer and an increased area of liquid/vapour contact (per unit mass of marbles and unit feed of liquid), so this would indeed improve the flux of the molecular exchange - alcohol evaporating out and water condensing in.
But after a point (see many references to packed columns "choking") gravity cannot remove liquid fast enough and vapour passageways, upwards, get clogged with liquid and produce the expected Physics chaos.
This is most easily understood if you forget all about "boiling" once vapour is passed out of the pot. Fractionation is about molecular exchange. And that occurs only through the area where the liquid and vapour phases contact each other.
No boiling occurs beyond the pot.

I have no idea how my post (pretty much directed at Rad )on my idea on modifying the surface of marbles enveloped such disdain from you but this was your retort to my post posted several posts before your even mentioned any equations . perhaps lets leave the misleading personal attacks to PM's rather than choking a good thread with it

[bearriver]

This thread went a bad way really fast...???

My plan is to test the marbles, in a 12 inch section. Then perform the same test with the same marbles after I chemically etch them. My hopes are that the second test will show what everyone here seems to be agreed upon...

I'm not as brilliant as some people think they are. Which is why I didn't make any claims as to what is what with a packing I have never used... Real world testing and results are all that intrest me.

[rad14701]

One other aspect that we may or may not have touched on here is the difference in the speed of heat saturation of our structured packing... Several members are having great results with relatively low amounts of copper or stainless scrubbers and that material rises to vapor temperature almost immediately, and perhaps even faster than the column wall itself... Marbles, on the other hand, take a more significant amount of time to rise to vapor temperature but, once they do, they will tend to retain the heat and not drop as rapidly as scrubbers... So this should, theoretically, help maintain a more stable temperature gradient within the packed column... Food for thought...

[Roddy]

Back to.......Marbles.

A bit of mathematical thought.... a sphere represents a volume enclosed by the minimum surface area.
When they are closely packed, the inter-sphere gaps are pretty small too.


It seems to me that if we want to maximise both wetted surface area and void-space to allow vapour to contact that surface, we should consider other structures as we progress the train of thought.
Perhaps this was the driving thought towards using helices and then SPP's....?

Now we start to hit problems!
As MichiganCornhusker has so deftly shown, we can easily visualise both surface area and void effects as liquid is progressively applied to marbles. This will not be so easy for more complex solids used as column packing.

In evaluating the relative merits of packing materials (perhaps using marbles as an idealised reference point) we need a way to determine liquid and vapour throughputs, and gain insight into how one affects the other. Worse, irregularly-shaped packing shapes will pack irregularly most of the time, so instead of calculating surface area vs void ratios, we're going to have to measure them instead. I think!

In an "evaluation still" it is relatively easy to derive the upward vapour flow rate and mass, but getting a useful measure on the real, downward flow rate and mass of falling liquid is much more difficult, I think.
Perhaps if we discuss suggestions on how we might succeed in getting at least a "working handle" on doing that?

For example, if we were able to measure everything which falls out of the bottom of the column and compare that with everything (distillate) which we send "top downwards" as reflux liquid, we'll be moving in the right direction?

THIS is the sort of rewarding discussion which brings me to HD.

[Bagasso]

I thought the point to be taken away from this thread was that surface area isn't an indicator of how well a packing material performs.

Will scuffed marbles work better than smooth? Could be, but the point is that smooth seems to have worked surprisingly well despite their low surface area, when compared to SPP or SS scrubbies.

My layman's guess would be that it has to do with the materials mass and thermal conductivity.

[whiskymonster]

I thought the point to be taken away from this thread was that surface area isn't an indicator of how well a packing material performs.

Will scuffed marbles work better than smooth? Could be, but the point is that smooth seems to have worked surprisingly well despite their low surface area, when compared to SPP or SS scrubbies.

My layman's guess would be that it has to do with the materials mass and thermal conductivity.

+1

Surely having the high thermal mass of the glass marbles are made of would give a nice gentle heat gradient up the column. Kind of equivalent to a nice thick layer of armaflex insulation on the outside of the column?

Tried and true method of keeping the run rock solid throughout, and letting you push just that little bit harder without the drop in purity from a variable RR.

The other thing that occurs to me is the thermal lag of the packing.
The marbles would be slow rise or drop in temp, so maybe once the boiler starts to depleted itself the almost constant temp of the marbles would slow the rate of water passing up the column by keeping the vapour temp closer to that of alc than water. For a while at least.
Scribbles wouldn't do this, and would allow the temperature gradient to compress much quicker.

What could be interesting is trying this the other way around, and filling the inside of the column with scrubbies, and insulating the outside with something of a very high shc. Fire clay maybe?
Just a thought...

[rad14701]

I think we're on the same page with regard to the thermal mass and surface area, Bagasso and whiskymonster...

[guyonthecouch]

I know I am a total noob here but just thinkin out loud. Anybody ever think about doing a layer of scrubbers then marbles then scrubber's then marbles. This might be totally useless but just throwing this out there?

[rad14701]

I know I am a total noob here but just thinkin out loud. Anybody ever think about doing a layer of scrubbers then marbles then scrubber's then marbles. This might be totally useless but just throwing this out there?

The problem with that would be the disruption of the vapor density and temperature gradients through out the packed column... It would essentially be like forcing different theoretical plates based on packing density... A column "should" perform better when completely filled with the same composition and density of structured packing...

[guyonthecouch]

Sounds like a good enough reason to me lol