A new way of dealing with jelly?

[therecklessengineer]

Unlikely this is anything to do with the ASM - more likely the funky things I was trying with the sludge in the bottom of my settling tank.

I'm doing a drywash process, biodiesel demeth'd in the reactor then transferred to a settling tank and circulated through a spray bar to remove the residual methanol.

However, went out this morning to check on the contents of the settling tank and I found a tank full of jelly. Bugger!

I was liquid enough to pump (barely) and I tried several processes to break it on small samples. Adding gylc/water sorted it the fastest, but I didn't really want to add water into my settling tank.

Rather surprisingly (to me anyway) I found that lots of heat caused bubbles in the jelly - which suggested a residual methanol content. Taking a small sample up to 110C suddenly made some glyc drop out and the sample go cloudy. Leaving to settle for a while resulted in some glyc, soap and crystal clear biodiesel.

So - it would seem my jelly still has some methanol in it which is holding the soap in solution and causing jelly.

Anyhow, back into the reactor, loads of heat (110C) and several litres out of the condenser has produced about 5 litres of glyc and cloudy biodiesel - the same situation as the sample.

It's now back in the settling tank and I'm pretty certain that by tomorrow I'll have clear bio and the soap will have settled out.


So this is a somewhat different approach to the normal 'add glyc and do a post-wash' and so far seems to be working. I'll update in the morning and I'm kicking myself that I didn't get photois.

[Rotary-Motion]

hmmm sounds interesting let us know...

[Julian]

Interesting sounding discovery ... sounds a bit like the bio equivalent of "hitting it with a big hammer"!

It's not jelly, but I've got some solid soap.  I can appear to recover oil from it by heating with glycerine (to about 65°C), pump mixing and settling.  It then appears to process as normal to give what I'm assuming is biodiesel.

I'll try replicating what you've done and see if the result is different.

[1958steveflying]

Unlikely this is anything to do with the ASM - more likely the funky things I was trying with the sludge in the bottom of my settling tank.

I'm doing a drywash process, biodiesel demeth'd in the reactor then transferred to a settling tank and circulated through a spray bar to remove the residual methanol.

However, went out this morning to check on the contents of the settling tank and I found a tank full of jelly. Bugger!

I was liquid enough to pump (barely) and I tried several processes to break it on small samples. Adding gylc/water sorted it the fastest, but I didn't really want to add water into my settling tank.

Rather surprisingly (to me anyway) I found that lots of heat caused bubbles in the jelly - which suggested a residual methanol content. Taking a small sample up to 110C suddenly made some glyc drop out and the sample go cloudy. Leaving to settle for a while resulted in some glyc, soap and crystal clear biodiesel.

So - it would seem my jelly still has some methanol in it which is holding the soap in solution and causing jelly.

Anyhow, back into the reactor, loads of heat (110C) and several litres out of the condenser has produced about 5 litres of glyc and cloudy biodiesel - the same situation as the sample.

It's now back in the settling tank and I'm pretty certain that by tomorrow I'll have clear bio and the soap will have settled out.


So this is a somewhat different approach to the normal 'add glyc and do a post-wash' and so far seems to be working. I'll update in the morning and I'm kicking myself that I didn't get photois.

Easy way to see into the future is take a small sample and put it in the fridge, if the batch is going to gel again it will be as it cools.

[therecklessengineer]

Update this morning - after being circulated via the spray bar overnight.

I've got a drum full of liquid biodiesel. Soap scum on the top that will be skimmed off, but otherwise all good!

[Tony]

Interesting stuff.  One to add to the list of things to try if it all goes pear-shaped

[1958steveflying]

Update this morning - after being circulated via the spray bar overnight.

I've got a drum full of liquid biodiesel. Soap scum on the top that will be skimmed off, but otherwise all good!

That's a great result to an annoying problem that can rear it's ugly head. Was your original demeth low on Methanol and glyc  recovered ?

[therecklessengineer]

I think this jelly arose because I added 40 titles of very roughly water washed Bio from another experiment. The intention was to dry it all at the same time, but I wonder if there was too much methanol/water to evaporate off in one cycle.

Anyhow, this suggests to me that jelly is caused by a residual methanol content (or water I suppose) and can be dealt with by removing it.

[1958steveflying]

I think this jelly arose because I added 40 titles of very roughly water washed Bio from another experiment. The intention was to dry it all at the same time, but I wonder if there was too much methanol/water to evaporate off in one cycle.

Anyhow, this suggests to me that jelly is caused by a residual methanol content (or water I suppose) and can be dealt with by removing it.

The instances that I had a batch gel as it cools have definitely been caused by excess Lye (glyc wash partial reaction not taken into account at that time as it had not been discovered), and were cured by adding back the glyc and 5% bio volume water. This process ended with the same amount glyc and water being retrieved from the batch and sparkling bio when dried and settled.

That is not saying anything against what you have found with your batch they all have their own peculiarities. 

[therecklessengineer]

Yes, you're right. I should re-phrase the above.

Sure - the jelly is a manifestation of the soaps produced in the biodiesel for whatever reason (excess catalyst, water etc). However, the reason why a jelly is formed and the soap doesn't precipitate out as expected is due to a residual methanol content preventing the formation of a precipitate - which would also go some way to explain why a jelly is normally observed after a de-meth cycle.

And because it's a jelly and acts like a solid, little methanol is released from the surface without encouragement - therefore keeping the jelly in jelly form.

At least - this is the theory I've formed from the observation of above.