I'm wondering if anyone else has noticed this observation. I have three sources of waste cooking oil, two pizza shops and one fish fry shop. The pizza shop oil is generally poor quality highly degraded & overheated. The fish fry shop is by comparison almost virgin. When I make biodiesel using a high percentage of pizza shop oil I have difficulty getting complete conversion (as measured by the 3/27 test) even if I do a third base stage (sometimes even a fourth treatment). When I use a high percentage fish fry oil the second stage always goes to completion easily.
Have other people seen this? Why is this? The implication is that there is some contaminate that prevents more complete conversion. But why should that be. Each time I do a base treatment I remove more of the glycerin byproduct and the remaining material should process even better.
While I haven't measured parameters like the initial soap content titration suggests that there is comparable soap levels. Generally the titrations are 3-4 ml of NaOH for both. There is however much more "foreign" material in the pizza shop oil, flour & some kind of sticky, almost adhesive like material that coats the plastic pails.
Any suggestions or comments?

The UVO I process is from fish&chips, lightly used by high-end restaurants. No animal fats, hydrogenated oils, water, FFAs, and other contaminants from overuse. The contaminants consume reagent and make soap with NaOH that would otherwise be used for conversion. Animal fats make BD that precipitates out of the fuel at cold temperatures below 10C/50F, clogging filters and causing problems.

The UVO I collect is dewatered and cleaned with filtered cold upflow then converted to 99% BD w/ 4.5gm NaOH/Litre UVO & 18% methanol. The 'cleaner' the UVO going in ≈ cleaner BD coming out, with allmost all the chemicals used for converting UVO to BD.

Hi John Galt,

quote:

Originally posted by john galt:
The UVO I collect is dewatered and cleaned with filtered cold upflow then converted to 99% BD w/ 4.5gm NaOH/Litre UVO & 18% methanol.

On the face of it that ia pretty amazing.
I just have a few questions to nail this down a bit better.

1. Do you mean you do not titrate your WVO before reacting?

2. Do you perform a glycerine pre-wash/ neutralization/ whatever else you might want to call it?

3. If you perform a glycerine pre-wash/ etc, do you include the chemicals contained in the glycerine as part of the chemicals used in the reaction? ("4.5gm NaOH/Litre UVO & 18% methanol.")

4. How many stages of reaction, including glycerine stages, do you perform?

5. If more than one stage, how much chemicals are used in each stage?

6. Can you explain what you mean by "converted to 99% BD"

Thank you for the information in advance

Rick G, Hi.
When we do the dropout test what we are actually seeing as dropout is a mixture of oil and some biodiesel. This is roughly 40% oil and 60% bio, but this ratio alters in proportion to the amount of dropout.

So you may well ask why we then add the next stage chemicals as a result of the total dropout, and why this doesn't result in a massive chemical overdose.

The reason is that also in the dropout tube is a certain amount of di and monoglycerides, dissolved in the methanol. Both these also require more chemicals to convert them to methyl esters.

So it works out as convenient that calculating the chemical addition on the total dropout is roughly correct.(but no more accurate than roughly)

If there are a higher than normal amount of di's and mono's dissolved in the methanol phase then the correct chemical addition (worked out from the dropout amount) will be insufficient to give you a pass with no dropout. From your post I think this is the reason why you have to do multiple reactions.

Hope you follow all that, Dave.

Hi Dgs,

quote:

Originally posted by Dgs:
Rick G, Hi.
When we do the dropout test what we are actually seeing as dropout is a mixture of oil and some biodiesel. This is roughly 40% oil and 60% bio, but this ratio alters in proportion to the amount of dropout.

Please explain the testing you performed. The dropout tests I performed showed the dropout to be quite variable.

quote:

So you may well ask why we then add the next stage chemicals as a result of the total dropout, and why this doesn't result in a massive chemical overdose.

The reason is that also in the dropout tube is a certain amount of di and monoglycerides, dissolved in the methanol. Both these also require more chemicals to convert them to methyl esters.

Please describe the testing you performed and results achieved that showed this large amount of dissolved and invisible mono and diglycerides are present in the methanol.

I am particularly interested to know what amount of mono and diglycerides your testing has shown are dissolved invisibly in the methanol compared to the visible dropout and how you determined the mono and diglycerides were invisibly there.
My understanding is that it is only the monoglycerides that dissolve into the methanol and the diglycerides drop out in the test.

When I did my testing, in some cases I used extremely high conversion biodiesel made to a formula that would produce biodiesel with a purity in excess of 99.9%.
There would have been very minimal monos and diglycerides present in the biodiesel and still, in some of the fall out tests I performed adding veg oil to this biodiesel the fallout was over 3 times the amount of veg oil added to the test.

I think it is a good idea to let Rick G see the post and reply before we get into all this again Tilly.

Hi Dgs,

quote:

Originally posted by Dgs:
I think it is a good idea to let Rick G see the post and reply before we get into all this again Tilly.

Get into what?
I am assuming you have performed further testing since we last talked that supports these latest claims you make about the 3/27 test.

Dgs, I understand what you are saying about the possibility that there are unreacted mono & di glycerides in both the "dropout" and the methanol as well methyl esters (BD) dissolved in the "dropout" as well as the methanol for any 3/27 test. I don't have any information about this but my question is a little different. I'm concerned about the possibility of some "refractory" components in waste cooking oil from different sources. You might imagine that cooking oils from different sources, soybean, canola, peanut, etc. would have marginally different fatty acid profiles and thus process differently but except for the very occasional "cube" of canola the vast majority of the oil from all three shops is soybean oil. I agree that some "highly used" waste cooking oil (high soap content etc) might cause more catalyst or methanol to partion to the glycerin layer and be less available for reaction but to the extent that I can measure the amounts of glycerin layer removed there doesn't seem to be a substantial difference. I try to always fill my processing vessel to the same level for each batch and process in the same way for each batch yet I still find that I have some "dropout" that seems resistent to conversion when using high percentage pizza shop oil. Even after four treatments. For example, this past couple of months I have made two or three batchs per week. I do have the occasional emulsion form during washing due to various like too much flour contamination, incomplete glycerin drainoff, etc. but the problem of resistent "dropout" appears during the conversion steps not the wash steps.
Your suggestion that I better filter the oil prior to reaction is a good idea but my experience with more extensive filtration is that it is time consuming and somewhat impratical given the amount of solids in the oil, particularly the pizza shop oil.
Tilly, you raise good questions about the 3/27 test, the answers to which might be very interesting and helpful irrespective of my question. My own brief investigation suggests that answering your questions even if you had a gas chromatograph available would require considerable work. I've located a citation concerning a thin layer chromotography method which might just answer your questions without too much equipment and work. I've requested the paper describing the TLC analytical method and if it's practical might eventually have some answers for you.
Thanks all for your interest and comments.

quote:

more extensive filtration is that it is time consuming and somewhat impratical given the amount of solids in the oil

Not necessarily, depends on how you choose to do it. The most important contaminant to remove is water. The drier the UVO the less soap will be formed.