New to the forums and biodiesel - coworker and I are going to build a plant together and have settled on the GL-1 method. I have read the website and the forum thread and have a question about methanol recovery from the glycerol drained off after the reaction, specifically, how are people accomplishing it?

1) Is the glycerol pumped back into the reactor for methanol extraction? If so, how - the glycerol using NaOH gets pretty thick, does it not? And how does the glycerol get removed from the reactor? My understanding is that it gets even thicker after the methanol is removed.

2) Same questions for using another container - what type and how?

3) I had an idea for recovery adapting something I saw elsewhere (http://home.swbell.net/scrof/procdraw.htm) using a paint pot and a vacuum pump. I've sketched it on the system diagram - basically, it does the extraction from the glycerol at the same time as the BD, using the venturi to draw air through the glycerol container then inject the methanol laden air into the reactor. Alternately, eliminate the vent to the glycerol container and just pull a hard vacuum for a while, then valve it out. When the glycerol is cool and hard, pop the top off the paint pot and scoop it out. Comments?

TIA
R2.0

ecosystem_venturi_injection_mod.gif (10 Kb, 80 downloads) Glycerol Methanol recovery

I use GL's plumbers delight condenser and a venturi from Jameserl.

Yesterday I put 20 gallons of glycerol back into my processor and heated to 182. Why 182? Because the mixture started boiling at that temp and wouldn't go any higher. The methanol recovery was happening so fast that it couldn't fall out of the pipework fast enough. Below 182, it was so slow. I recovered nearly 4 gallons. I don't have a method for checking purity right now but I will.

My $0.02.

How hard was it to pump the glycerol back in? And was it NaOH or KOH?

I was just looking at the GL processor diagrams (because of R2.0 post), and an idea occurred to me:

What would happen if, right after the reaction stage is finished, and before the prewash, we'd kept on circulating and heating the biodiesel with the still not settled glycerol, and opened V7 to distill the methanol off the biodiesel and the glycerol at the same time? Would that cause any problems later?

It seems to me it might be a way to simplify the methanol recovery process, but then again I've never produced any biodiesel yet, so maybe I'm talking pure nonsense...

It can cause reverse reaction if you do it slowly enough. I've actually had biodiesel 'unmake itself' (ie the glycerine disappeared and I got a weird in-between oily material that was neither biodiesel or oil) doing this.

quote:

It can cause reverse reaction if you do it slowly enough

Hmmm. So I guess I WAS talking pure nonsense. Too bad.
Thanks Girl Mark.

Considering about 5 threads below this there is a 5 page thread discussing justyour question, I think "nonsense" is a bit harsh. Apparently they are having some luck with it, but haven't gotten the bugs out.

I'm more interested in doing the steps simultaneously, not combining them. It seems a shame to take that nice warm glycerol, let it cool down (and then heat it up again later) when you are going to be pulling a vacuum ANYWAY as part of the regular GL process.

quote:

I'm more interested in doing the steps simultaneously, not combining them. It seems a shame to take that nice warm glycerol, let it cool down (and then heat it up again later) when you are going to be pulling a vacuum ANYWAY as part of the regular GL process.

Now that you mention it, it looks like I didn't read your first post as carefully as I should have. Yep, your idea makes sense.
Two things I would change in your design though: I would run the hot biodiesel through the glycerol, between the reactor and the pump (through maybe a 2 or 3 loop copper serpentine) to heat up the glycerol and keep it hot without using a second heater (I don't feel comfortable heating the glycerol directly), and I would keep the vent, so the vacuum created by the venturi can pull air to carry the methanol fumes with it.
I think with those mods, it should work perfectly. What do you think?

Time...

Say, an hour to set up and mix methoxide and dewater. 20 min to mix in methoxide to oil (assuming oil is below 130deg F.) No? so wait for oil to cool. We now have over 2 hours in. Process for 2 hours. 5% prewash 20 min. Settle 1 hr. drain gly. Settle, drain bio-d. So lets see we are up to about 6 hours already. No?

Then, so as to not have too much head room in the processor, there should be at least 2 batches of gly in the processor.

If, you are trying to tie in the gly container to the venturi, I can see that, but, how do you deal with the fact the gly isn't being circulated?

I am all for energy conservation. Electric here in New England is stupidly expensive.

Here's the deal, you only loose the first batch's heat. Batch #2 is emptied into same barrel as batch #1 was. It helps heat the first batch and makes it flow better. Pump both batches into the processor and demeth. But here again, we are talking a full day of playing withthe processor, no?

Thoughts?

C.

quote:

Originally posted by RNCarl:


If, you are trying to tie in the gly container to the venturi, I can see that, but, how do you deal with the fact the gly isn't being circulated?

I am all for energy conservation. Electric here in New England is stupidly expensive.

Here's the deal, you only loose the first batch's heat. Batch #2 is emptied into same barrel as batch #1 was. It helps heat the first batch and makes it flow better. Pump both batches into the processor and demeth. But here again, we are talking a full day of playing withthe processor, no?

Thoughts?

C.

I was thinking abouit the way that guy does it in the original link with the pait pot - he just applies a vacuum, I think. "The pump is capable of about 27" hg. The flow rate is outstanding if the pumps are hooked in parallel or series. Only about 22"-25" is really necessary for methanol recovery. The methanol will begin boiling at 22" and 140F, and will be done at about 25".

Another reason I'm interested in simultaneous operations is probably a personal quirk, but when I'm done with something, I like to be done. Storing glycerol for later processing would feel like procrastination. I'd like to be in a situation where I go up to my friends house in WV on a Saturday morning and, when we're done, have an empty processor, fully recovered methanol, glycerine ready for dispositioning, and fuel ready to pump into a tank. The settling process may preclude the last part, , but if I can de-meth the glycerol while other processes are going, I'll be happier.

quote:

Originally posted by R2.0:
I was thinking abouit the way that guy does it in the original link with the pait pot - he just applies a vacuum, I think. "The pump is capable of about 27" hg. The flow rate is outstanding if the pumps are hooked in parallel or series. Only about 22"-25" is really necessary for methanol recovery. The methanol will begin boiling at 22" and 140F, and will be done at about 25".

Another reason I'm interested in simultaneous operations is probably a personal quirk, but when I'm done with something, I like to be done. Storing glycerol for later processing would feel like procrastination. I'd like to be in a situation where I go up to my friends house in WV on a Saturday morning and, when we're done, have an empty processor, fully recovered methanol, glycerine ready for dispositioning, and fuel ready to pump into a tank. The settling process may preclude the last part, , but if I can de-meth the glycerol while other processes are going, I'll be happier.

Ahhh,

I see. I am not sure how much head the venturi could pull in this set-up as you describe it. It is worth a try. I understand that if you pull a vacuum a lower temp would require less heat. Contact Grahmn Lamming for high draw venturis.

Some more thoughts, is that paint pot heated? There is a post on this forum for a small paint pot type set up but I believe its fitted with a heater element. Better idea may be to loop the bio-d through the pot as suggested. I would wye it off.

I have not considered that set-up because the gly is static and I wonder how efficient the removal is since the gly isnt mixed somehow.

I almost forgot. Insulate well. especially the line from the gly pot to the venturi, otherwise all you have is a huge reflux condenser and the cooling meth will drop back into the pot.

It is a complex set-up but could work. Remember SAFETY FIRST!

Let us know how it goes.
C.

Remember if you lower the boiling point with vacuum you will also lower the condensation point as well. I distill with the venturi both from the BD as well as from Glycerin both in the basic GL processor. It works well with no additional vacuum required.

Tony

quote:

It can cause reverse reaction if you do it slowly enough. I've actually had biodiesel 'unmake itself' (ie the glycerine disappeared and I got a weird in-between oily material that was neither biodiesel or oil) doing this.

And how slowly was that Maria? Would it apply in this case?

quote:

Originally posted by tgomes:
Remember if you lower the boiling point with vacuum you will also lower the condensation point as well.
Tony

Not quite. My idea is that the venturi pulls the methanol vapors into the BD flow, which are then released into the processor to mix with the vapors from the BD undergoing recovery. So at this point, it would be atmospheric.

Another question raised was mixing. I checked around and there ARE agitating paint pots available, with a little prop mixer driven electrically or by air pressure. A little expensive, but I bet one of these would make a cool small-batch processor as well.

quote:

Originally posted by tgomes:
Remember if you lower the boiling point with vacuum you will also lower the condensation point as well. I distill with the venturi both from the BD as well as from Glycerin both in the basic GL processor. It works well with no additional vacuum required.

Tony

I've been using vacuum distillation for a while, and have been considering the venturi. Can you tell me how long it takes to distill off the methanol (after draining glycerin) using the venturi? I guess it would be helpful to reference some temps and the batch size.

I run a 150l batch with 33l methanol. After seperation I distill at 190*f and the show is over in about 2.5 hours. I usually get about 4l back. I then bubble in the settling tank for a few hours more to insure the last bit is out. The last batch of glycerin I distilled gave 30l back out of 150l. That took about 8 hours with the condensor head controlled at 154*f. Purity runs about 98% as measured with the hydrometer at 70*f.

Tony

tgomes,

Your recovery seems just a bit on the low side. Any ideas where the missing methanol went? I would have guessed that out of a 150L batch of glycerin by-product from a 22% BD recipe, you would have been able to recover 45-50L with the remaining unaccounted for methanol in the biodiesel.