Biopowered - vegetable oil and biodiesel forum
Biodiesel => Chemistry and process => Topic started by: DavidA on November 23, 2021, 06:58:18 PM
-
Just been reading the above in the wiki.
Could someone elaborate on the following instructions,
Determine the amount of unprocessed oil in your batch from the dropout.
Make up the second stage methoxide by mixing "unprocessed oil volume x 20%" methanol with "unprocessed oil volume x base" catalyst.
Dave.
-
2 stage process.
Undercook stage one by using 5g per litre of oil
Allow it to settle and take a 10ml sample. Put this into 90ml of methanol and mix. See what drops out. For me typically 2ml of oil sits at the bottom. This means you have 20% unreacted oil.
Drain the glyc.
Stage 2:
Add 5g NaOH per litre unreacted plus a bit of methanol. This should take the batch to full reaction.
-
IMO the typical 80/20 two stage reaction leaves the crude bio too far from the reaction end point to get accurate process results if water washing. I now like to get about 0.3mls dropout before the penultimate reaction (even though this can mean 3 stages)
Only by having a really small final stage can the slight over-chemical addition be determined accurately. This avoids the mono's when water washing.
Remember mono's are present with a clear 10/90 if the test is just on the cusp of clear.
The 10/90 test does not work the way most people think. It is not a case of the dropout being all oil, it's more like 50/50 oil/biodiesel.
-
The solubility test should be carried out with both components at 20C for best results. The methanol should be very pure - if there is significant water present you will always get dropout even from just bio.
o
The test was never designed to be quantitative, though it is usually accurate enough for determining the quantity of reagents needed.
If using the 2 stage non titration method it is often possibe to use much less methanol, 16% for example, but it is best to make sure you have mastered the porcess before trying to optimise the use of reagents.
if your oil is very acidic, or you are unsure, then it is worth titrating first regardless. If the oil titrates above 5 for example, you are unlikely to get any conversion as all the caustic will be used in neutralising the FFA.
-
Thanks for those replies.
I'll give the two stage method a try. I don't have much trouble with my oil, it's quite clean. But always good to try something else.
At the moment I'm wrestling with these soap tests. Things are not going to plan.
Paul,
..if your oil is very acidic, or you are unsure, then it is worth titrating first regardless. If the oil titrates above 5 for example, you are unlikely to get any conversion as all the caustic will be used in neutralising the FFA...
Would that mean you will get a large drop-out on the first try, and will have to do maybe three stages ?
-
David,
It depends on how acidic the oil is, yes you would get a large dropout, but you could even need 4 or 5 stages. Some people have reported getting UCO that titrates at 20+ :o
I use the 2 stage no titration normally, but if the oil is suspect, I titrate, or try a Dr Pepper test to check.
Neutralising the FFA will produce water that will catalyse more soap formation (neutralising FFA produces soap), so you may gets lots of soap with little glycerol and find it difficut to find the glycerol as it may take a long time to drop out (days) or may not drop out at all thanks to all the soap. This can result in creating a huge mess and little to no bio..
-
The solubility test should be carried out with both components at 20C for best results.
Hi Paul, from many tests I have done the recommended temp of 20degs isn't low enough. The test can remain clear is Summer overnight (temps between 15 and 20 degs) for days, then we get a night of 12 or so and bobs your uncle, yes you got it, theres the dropout.
For more accurate results I use down to 10.
-
Dave,
I seem to remember that the test is only vaid for an hour os so after being carried out and leaving it longer can show false results (not sure where I go that from though) - so leaving it a few days might not be accurate though I don't know why.
Found a reference: https://www.utahbiodieselsupply.com/biodieseltutorial/methanoltest/
-
Thanks for those replies.
I'll give the two stage method a try. I don't have much trouble with my oil, it's quite clean. But always good to try something else.
At the moment I'm wrestling with these soap tests. Things are not going to plan.
Paul,
..if your oil is very acidic, or you are unsure, then it is worth titrating first regardless. If the oil titrates above 5 for example, you are unlikely to get any conversion as all the caustic will be used in neutralising the FFA...
Would that mean you will get a large drop-out on the first try, and will have to do maybe three stages ?
Excess amounts of soap means more glycerol, there is only about 7% glycerine in veg oil. Calling the by-product glycerol is a little misleading.
If you ever have high titrating oil there is an easy way around it. in your glycerol pre-wash add an amount of catalyst that will neutralise the ffs's. As the neutralisation is then done in the presence of glycerol any water produced is absorbed by the glycerol, making the oil dry and ready for conversion.
-
Years ago someone on here reported that after processing oil that titrated at 25 (yes, 25) the yield was 62%, After using the enhanced pre-wash method the yield went up to 82% with the same oil.
-
Dave,
I seem to remember that the test is only vaid for an hour os so after being carried out and leaving it longer can show false results (not sure where I go that from though) - so leaving it a few days might not be accurate though I don't know why.
I think the only time a false result can occur is if the test is so cold that a residue can form at the tube base. I have had this when leaving the test in the fridge for extended periods.
I did dozens of tests years ago when I was doing the dropout v oil comparison. I couldn't understand why adding 1ml of oil to a clear test gave about 2mls dropout. i will see if I can find the thread although the graphs i did at the time will have disappeared.
-
Here it is and the graph is still there.
http://biopowered.co.uk/forum/index.php?topic=2523.0
There is also a comment from Chug in the thread mentioning Neutrals findings on the 3.5 v 5gms of NaOH.
-
Been thinking about why there is a supposed time limit on the test, the one thing that has crossed my mind is whether the bio is likely to polymerise at all (since most of the oils we use are unsaturated). Even just if dimers were formed these would have a much higher MW and therefore would likely be less soluble. the results would vary from oil to oil depending on source, linseed might well go off very quickly, whereas a saturated one like stearin would not polymerise at all. The main problem with this theory is that polymerisation slows down as temperature drops which does not appear to equate with obsevations.
I can't find an original paper at present so why Jan suggested 20C for consistency is not clear to me.
I have to admit to doing the tests with methanol at whatever temperature it happens to be at (roughly 10C last week) and when that gives a pass I am quite happy.
-
Now here is something different.
I did a sample litre as per the wiki.
Here are the results.
Make your first stage methoxide solution with base amounts 3.5g/l for NaOH or 5g/l KOH x full batch size + 80% of normal methanol amount.
I normally use 240Mll/Litre. So I made up a 192 mL + 3.5g mix.
Put this first stage methoxide mix into the processor at normal reaction temperature, and mix for an hour.
Stand for 20 minutes ..
Added this to 1 litre of oil at 50C, shook vigorously, then repeated the shakes for an hour (my usual method).
..and drain off the glycerol into a clean container, and seal.
Here is where things went strange.
There was no glyc drop out. Just about 30cc of (apparently) bio sitting on top of the rest which appears to be a very cloudy (but stable) liquid.
Do your 3/27, 5/45 or 10/90 test and measure the amount of drop out.
Determine the amount of unprocessed oil in your batch from the dropout.
What happens if there is no drop-out ?
Dave.
p.s. I've made a few test samples by the old titration method over the last day or two with no problem. They are in the car tank.
NigelB,
..Undercook stage one by using 5g per litre of oil
Yes, I did read that and am going to do it. Just wanted to see what the method as per wiki worked like.
-
But if your oil titrates and you have added only 3.5gms you will have hardly processed any oil into bio. Try a 10/90 you maybe surprised.
I must admit IMO those wiki instructions are to say the least not very good. That 80/20 split of the 2 stages just isn't accurate enough if you water wash. No wonder people end up with emulsions!
-
I mixed up 30 ml Meth with about 3G NaOH and added it to the original. I see what happened tomorrow.
I did think at the time that hardly any reaction appears to have taken place.
-
I must admit IMO those wiki instructions are to say the least not very good. That 80/20 split of the 2 stages just isn't accurate enough if you water wash. No wonder people end up with emulsions!
Why dont you re-write it Dave. Nothing wrong with an update.
I dare say that some emulsions are caused by bad practice rather than bad reactions.
-
I must admit IMO those wiki instructions are to say the least not very good. That 80/20 split of the 2 stages just isn't accurate enough if you water wash. No wonder people end up with emulsions!
Why dont you re-write it Dave. Nothing wrong with an update.
I dare say that some emulsions are caused by bad practice rather than bad reactions.
Sadly, I think It's been some years since we were able to edit or create on the Wiki. I don't know exactly what the problem is but we are unable to log in anymore :(
-
I mixed up 30 ml Meth with about 3G NaOH and added it to the original. I see what happened tomorrow.
I did think at the time that hardly any reaction appears to have taken place.
A quick look this morning shows quite a lot of drop-out.
I'll check it properly when things have warmed up a bit. It's only about +5C in the shed at the moment.
-
You can actually calculate the rough titration from the dropout.
-
Sadly, I think It's been some years since we were able to edit or create on the Wiki. I don't know exactly what the problem is but we are unable to log in anymore :(
That's incorrect Keef. I've just logged into the wiki and put in a small edit on the ASM page.??
-
I mixed up 30 ml Meth with about 3G NaOH and added it to the original. I see what happened tomorrow.
I did think at the time that hardly any reaction appears to have taken place.
A quick look this morning shows quite a lot of drop-out.
I'll check it properly when things have warmed up a bit. It's only about +5C in the shed at the moment.
Has a close look this sample .
It has 1Litre of clear bio above 0.3 Litre of byproducts. At 8.2C
A 3/27 test shows no visible drop out. (I'll do another one later).
Update.
Did another drop-out test.
No visible drop out after four hours.
-
I mixed up 30 ml Meth with about 3G NaOH and added it to the original. I see what happened tomorrow.
I did think at the time that hardly any reaction appears to have taken place.
A quick look this morning shows quite a lot of drop-out.
I'll check it properly when things have warmed up a bit. It's only about +5C in the shed at the moment.
Has a close look this sample .
It has 1Litre of clear bio above 0.3 Litre of byproducts. At 8.2C
A 3/27 test shows no visible drop out. (I'll do another one later).
Update.
Did another drop-out test.
No visible drop out after four hours.
So are you saying that you have complete conversion after adding 3.5 + 3.0 gms of NaOH. Did you discard the glycerol after the 1st stage.
Why are you still doing a 3/27 when you now have the proper 10/90 tube.
-
I mixed up 30 ml Meth with about 3G NaOH and added it to the original. I see what happened tomorrow.
I did think at the time that hardly any reaction appears to have taken place.
A quick look this morning shows quite a lot of drop-out.
I'll check it properly when things have warmed up a bit. It's only about +5C in the shed at the moment.
Has a close look this sample .
It has 1Litre of clear bio above 0.3 Litre of byproducts. At 8.2C
A 3/27 test shows no visible drop out. (I'll do another one later).
Update.
Did another drop-out test.
No visible drop out after four hours.
So are you saying that you have complete conversion after adding 3.5 + 3.0 gms of NaOH. Did you discard the glycerol after the 1st stage.
Why are you still doing a 3/27 when you now have the proper 10/90 tube.
No. Because I got such a low drop-out with the first part (no drop out) I decided to write off this test.
I added the extra 3g and the bit of methanol to the whole batch, brought it back up to 50C , shook it vigorously and let it settle out as one would for a normal mixing.That way I should at least get some useable fuel from it.
My reasoning was that I may as well do what was essentially a normal mix with 6.5 g of NaOH and see if it worked. I used the 3/27 test as it was the one I normally used in earlier tests. The 10/90 tube can be used in future tests as it is indeed easier to read.
I am going to try the two stage method with 5g for the first stage.
-
Sadly, I think It's been some years since we were able to edit or create on the Wiki. I don't know exactly what the problem is but we are unable to log in anymore :(
That's incorrect Keef. I've just logged into the wiki and put in a small edit on the ASM page.??
How long has that been possible, no one tells me nothing!!!!
-
Even though I was logged in on the forum I had to log in seperately on the wiki. I then did a small edit regarding the availability of ASM.
-
DavidA, it may be helpful to you to go through my process for making bio in cubies. I know you want to make really good bio but its really hard to water wash using this method, best to bubble/settle then woodchip, you will still get to zero soap.
When you have processed to your liking (clear 10/90 of whatever) leave to stand overnight.
Pour the bio into another container, try to make sure you pour as little glyc in as possible.
Bubble for 24 hours
Leave to stand for 24 hours.
Pour through oak woodchips (make sure you pour NO glycerol into the chips) You can then use the bio. If you check for soap it should be zero.
You can gather all the residual glyc with a small amount of bio and use it to glyc wash some more oil, you can do this in a cubie as well. Everything can be done at ambient, I've made bio in cubies outside at about +5degs.
-
Dave (dgs),
I do agree with you regarding the water wash. That is why I never bothered in the past.
I am sending off for a fish tank air pump and will concentrate on bubble washing in the future.
While I am concerned for the environment, I am no hard-line greenie. Everything is related to cost. And as I am on a water meter, a 40 Watt air pump seems the way to go. Possibly cheaper to run than a water pump.
I did all my past mixing in 20 Ltr poly drums and found them quite usable. I avoided the cubies as they, or at least the ones I got, were attacked by rats who nibbled through them wasting all the oil. They don't seem interested in the poly drums.
Just need to find out where to get oak chips from. Do they have to be oak ?
More tests with things like 80/20 method still to do. I think I will use 5g/Ltr of oil rather than the 3g (was it 3 or 3.5 ?) for the 80% bit. May get some dropout that way.
-
I made a batch of 160lt on Saturday and used a base of 6gm/lt for stage one. It gave me a 20% under reaction. Just where I thought it would be. Next batch will be 7gm. I'm now getting my oil from one source. That should see somewhere close to 10% undercook.
3.5gm should be ok for brand new SVO. If the oil has been used there will be FFA in there which will need to be neutralised....hence the increase in catalyst.
-
I made a batch of 160lt on Saturday and used a base of 6gm/lt for stage one. It gave me a 20% under reaction. Just where I thought it would be. Next batch will be 7gm. I'm now getting my oil from one source. That should see somewhere close to 10% undercook.
3.5gm should be ok for brand new SVO. If the oil has been used there will be FFA in there which will need to be neutralised....hence the increase in catalyst.
I don't usually process any oil that titrates. It comes out of my glyc wash ibc. As its glyc washed about 5 times it shows a conversion of about 20% so I'm on 6.5gms K + 10% methanol.
-
Dave (dgs),
I do agree with you regarding the water wash. That is why I never bothered in the past.
I am sending off for a fish tank air pump and will concentrate on bubble washing in the future.
While I am concerned for the environment, I am no hard-line greenie. Everything is related to cost. And as I am on a water meter, a 40 Watt air pump seems the way to go. Possibly cheaper to run than a water pump.
I did all my past mixing in 20 Ltr poly drums and found them quite usable. I avoided the cubies as they, or at least the ones I got, were attacked by rats who nibbled through them wasting all the oil. They don't seem interested in the poly drums.
Just need to find out where to get oak chips from. Do they have to be oak ?
More tests with things like 80/20 method still to do. I think I will use 5g/Ltr of oil rather than the 3g (was it 3 or 3.5 ?) for the 80% bit. May get some dropout that way.
When I said cubies I meant the thick poly ones. Any hardwood chips will do. I've mentioned the processing in containers to a few people, some scoff but it works really well. If I had to choose between a comparison with a metal processor I would choose the container method anytime.
-
I was wondering about this part of the process description in the wiki.
..It seems that most of the people who have been trialling this method are getting around a 25% reduction in catalyst use and much less soap produced in the process. It has also been reported that some people are getting more biodiesel out than WVO put in...
How would that work.
..
-
How do you measure how much WVO goes in and how much Bio comes out? In most cases this is done by volume.
The density of wvo varies depending on the original veg oil used, the degree of oxidation (FFA), but usually between about 0.905 and 0.925.
The denisty of Bio varies also depending on the original veg oil etc. but again between about 0.86 and 0.90.
The density of pure glycerol is about 1.260.
The density of methanol is about 0.800
All these are at about 20C and obviously change with change in temperature (increase when colder, decrease when warmer).
Using the no titration method tends to use less methanol.
If you were to use 100L of dry WVO and 15L of methanol along with 600g of NaOH and assuming the oil titrated at 0.
On average 12.5L of methanol would actually be used.
Assuming a perfect world, with perfect conversion and no by products:
100L of oil weighing 86Kg would convert to Bio weighing 86Kg (glycerol MW about 92 and methanol about 32 so 96 equivalent when converted).
If we assume the MW of Olein is 375 and that WVO is near enough olien, then that 375 will become 379 once converted to bio so about 1% more by weight.
If we assume that the oil is 0.9 density and the bio is 0.86 density, then there will be between 4 and 5% more Bio by volume than there was WVO.
The 12.5L of methanol would weigh 10Kg and produce 96/92 times as much glycerol so under 8L of glycerol, there would also be 600g of NaOH in the glycerol.
There would also be 2.5L of methanol disolved in the glycerl and bio.
In practice using caustic to make the catalyst solution produces water which catalyses the formation of soap. Of course if there are impurities in the WVO these could skew the results in any direction. The presence of FFA for example would lead to the formation of Soap by neutralisation. The soap and water will predominately be in the glycerol, however the presence of the methanol will allow soap to remain in the Bio. Using ASM (Anhydrous Sodium Methylate) rather than caustic will not cause water to be formed and therefore shoul create little to no soap other than that of neutralising FFA.
How you "wash" the bio will also have a large effect on how much actual bio you recover. Water washing with soap present will usually result in the loss of some bio. Bubble washing to remove the methanol and precipitate out the soap shoud preserve more bio, if that is followed by water washing there should be much less loss of bio.
Other variable such as temperature will also have an influcence on how much bio you recover.
If you have glycerol washed you WVO beforehand to reduce FFA and water content then wether you measure before washing of after could make a large differnce.
Lots of variables to consider but the simplest thing is that 100L of WVO could give you about 105L of Bio.
Posted under the influence of a couple of glasses of processed grape juice ;D
-
Thanks for that very detailed run down of the process.
I will give it a try, being as accurate as I can, and using the 150mL/6g mix as used above.
But I'll be doing it on a smaller scale.
Treat yourself to another processed grape. :)
Just tried a two stage test litre. (Titrated) and got 0.5mL drop out at the first stage, zero drop out when finished.
Just got to wash it now without turning it into soap.
-
I haven't read all through your post Paul but it seems to be accurate, well done. I remember 'Neutral' years ago coming up with the figure from an anhydrous process on new oil yielding 104 litres of biodiesel.
From 200 litres of my several times glyc washed oil (so any increase in volume ignored) I can get 204 litres of finished bio by using a 50/50 mix of K and ASM.
DavidA, the more stages you use, the less chemicals you will use to fully convert, but there comes a time when more stages are inefficient. Years ago I did a 7 stage process on some glyc washed oil and used a total of 3.5gms of KOH/litre of oil for full conversion.
I posted the results on Infopop and got slated for posting results other people couldn't match. There were a lot of cheeky buggars on there at the time!