Hi folks,

CAUTION: Electricity can kill. Do not attempt this if you are not familiar and experienced with safe working methods for high voltages. Be sure no-one else can accidentally come into contact with your experiment. Do not leave your experiment unattended.


WARNING: Electrical sparks can ignite flammable materials and gasses. I strongly recommend if you try any of these experiments, you use a power supply with current limited to no more than 0.5mA (0.0005 A).

Do not use battery chargers, welding kit etc.

---

I have been experimenting with the effects of high tension electric fields on suspended soap, glycerol etc. in biodiesel, with some interesting results.

My aim was to see if an electrostatic field within the biodiesel batch could help to attract impurities together into larger groups or clumps, to make filtration & separation more effective, efficient, faster.

I am using the GL 1 day process, whereby I remove methanol from the reacted batch and then filter out the soap. Sometimes the soap seems not to drop out of the batch as quickly as I would hope. Some people find the soap remains finely dispersed in the batch, especially if KOH is used as the reaction catalyst. This seems to make filtration less efficient.

I'm posting my early findings to encourage others to share in these experiments.

My test setup consists of a glass jar, into which I place 2 parallel metal plates, each 100mm x 30mm, spaced 30mm apart.

I fill the jar with de-meth'd unwashed biodiesel, with soaps still present. I used biodiesel made with NaOH for this test.

I then apply a DC voltage across the plates and note the effects.

Here is the setup ...





The Voltage source is set to 1000 V DC. I have not tried other voltage levels yet.

Here is a closeup of the jar before I apply the field ...




Here is a closeup of the jar after 10 minutes ...





The photos aren't very clear - I must re-run and take better quality ones, but what seems to happen is this:-

1. Soap is clearly attracted to the +ve plate and forms a thick, firm gel on it.

2. Soap in the bulk of the liquid clumps to form dense masses, which drop readily to the bottom of the jar.

No apparent separation occurs in the same timescale if the field is not applied.

I did a soap titration on a sample of the clear biodiesel phase after 30 minutes of field application and found negligible soap present, my test resolution suggests around 20ppm.

I have set up a bean can as a scale model of a 45 gallon drum, with the positive electrode being the can, the negative electrode being a carbon electrode from the centre of a 1.5V alkaline flashlight cell. I have the electrode in the centre of the bulk of liquid. The soap quickly coats the inside of the can.

The current passed thru the liquid is negligible, micro-amperes at the most, so there is very little power used in this process. Safety could be dramatically improved by using a string of high voltage resistors in series with the plates to limit the maximum available current to say 0.1mA .

20 x 470K 0.6 Watt high voltage resistors in series would seem to be a suitable current limiting combination.

Filtration note: To filter the clumps, be sure you do not pass the biodiesel through a pump before it reaches the filter, or you will most likely break up the clumps of soap into smaller pieces. Put the filter on the inlet side of the pump and set flow rate to the lowest acceptable level.

Could someone else give this a try if they have the necessary kit and experience? It would be particularly insteresting if you could use KOH as your catalyst.

+++++++++++++++++++++++++++++++++
Added 1 Sept 2006
High voltage generators:

A relatively safe source of high voltage, low current DC is an "Insulation Tester"

Search e-bay and you'll see a heap of them.
Make sure it is stated as being able to generate a high DC voltage - not AC.

Some are hand-cranked, OK for short term experiments, but a pain if you want to run an experiment over more than a few minutes.

These testers will be suitable for trying out simple tests on small lab quantities of BD, similar to what I've been doing so far.

Most have a pushbutton you'll need to keep pressed in during the test.
+++++++++++++++++++++++++++++++++

This message has been edited. Last edited by: GrahamLaming,

Wow - what an incredible experiment. I hope that others are able to reproduce this. Was the soap reduction just after the 10 minutes of electrical field being applied? Or did you filter after that?

Chris

Hi Chris

I did the soap test after 30 minutes.

I used a 1cc sringe to draw up the sample of clear biodiesel in the middle of the clearest part of the jar.

I've no idea what the soap concentration is like nearer the sides after only 30 minutes, probably very high. And of course if you disturb the jar, you can swirl up the separated soap.

Needs a lot more work to see how to use this process in a practical way, but there's the hint of possibility about it.

Graham -

I'm trying this now, but only with 60 volts DC. I'll check it and take another photo in the morning.

Your bottom photo, after 10 minutes, still looks a little cloudy. Could this be residual water from the prewash, or is it actually pretty clear looking? There can't be much water with only 20ppm soap. Very impressive testing!

Dan

Hi Dan

Yes, 10 minutes is way too short for much separation, but I thought it interesting that there is such a clear amount of separation so early.

60V may be too low to see much effect at 30mm spacing - I would guess the voltage gradient will determine the speed of separation, and I had roughly 33V per millimeter voltage gradient.

So, my guess is you will need to have only 2mm plate spacing to see the same rate of migration which I see at 30mm. and you may find that too small to determine that separation has occurred at all.

Or, put another way, at 30mm, you may need to wait 17 times longer to see the kind of separation I saw at 1000V.

I say may, because I'm not sure if there are any threshold forces you may need to overcome before migration will start. My gut feeling is you may need to exceed a certain voltage gradient before any movement will occur.

But I may be completely wrong.

I look forward to seeing how it works out, Dan, thanks for giving it a try.

By the way, my BD was at around 20C throughout the run.

This sounds pretty interesting... One thought though... What about trying it wit AC? Maybe that would cause the soap not to cover the positive electrode, but to fall out to the bottom of the jar?

Paul

Hi Paul

I'll try 1000V AC tonight and see what happens at 50Hz and 400Hz.

I've heard of electrostatic 'coalescers' used in the oil industry to help water droplets group together so they fall out of oil quicker. They run on AC. Not sure what voltage gradient they run.

Will let you know how the AC soap removal goes.

But, so far, I quite like the DC effect, because you can scrape off the soap as a nice thick gel, from the positive plate.

I will also make a fresh test batch without the 5% prewash, to see if the lower water content has a noticable effect on separation speed and quality.

Remember folks, this only works properly on biodiesel which has had its methanol completely removed. If you have any methanol in your biodiesel, some soap will remain dissolved in the methanol and won't separate.

Graham,

Dude, you are a total mad-scientist !!!

glad you're on our side...

quote:

Originally posted by DanDD:
Graham -

I'm trying this now, but only with 60 volts DC. I'll check it and take another photo in the morning.

Your bottom photo, after 10 minutes, still looks a little cloudy. Could this be residual water from the prewash, or is it actually pretty clear looking? There can't be much water with only 20ppm soap. Very impressive testing!

Dan

Independant of the soap, I believe biodiesel can hold 1500 ppm of water before it's saturated.

quote:

I'll try 1000V AC tonight and see what happens at 50Hz and 400Hz.

I don't think AC is going to produce anything. You are basically electroplating with soap, and the AC current is going to repel the soap from the (+) plate as much as it attracts (and the same for the (-) plate).

If I'm wrong...cool.

quote:

Originally posted by GrahamLaming:
Hi Paul

I'll try 1000V AC tonight and see what happens at 50Hz and 400Hz.

I've heard of electrostatic 'coalescers' used in the oil industry to help water droplets group together so they fall out of oil quicker. They run on AC. Not sure what voltage gradient they run.

Will let you know how the AC soap removal goes.

But, so far, I quite like the DC effect, because you can scrape off the soap as a nice thick gel, from the positive plate.

I will also make a fresh test batch without the 5% prewash, to see if the lower water content has a noticable effect on separation speed and quality.

Remember folks, this only works properly on biodiesel which has had its methanol completely removed. If you have any methanol in your biodiesel, some soap will remain dissolved in the methanol and won't separate.

Graham, would it be possible for this field to overcome the bond that the meth and soap have? or is that bond too strong?

Well, someone, please try this!

Freshly brewed biodiesel, no 5% prewash, not settled, full of glycerol etc.

Get 2 glass jars.

Fill both with bio and apply a field to one only and compare glycerol settling speed.

If I can only get my video camera to work, I'll make a movie of it.

The glycerine PLUMMETS to the bottom in the jar with the applied field. I'd guess 10 to 15 times faster than just sitting.

You must use much less voltage, maybe 10V/mm gradient, due to the excess Na ions. (I am using 300V DC aross plates 30mm apart. I can probably use much less voltage than that.)

But this is dramatic settling speed.

quote:

Originally posted by Ryan P.:

quote:

I'll try 1000V AC tonight and see what happens at 50Hz and 400Hz.

I don't think AC is going to produce anything. You are basically electroplating with soap, and the AC current is going to repel the soap from the (+) plate as much as it attracts (and the same for the (-) plate).

If I'm wrong...cool.

Well, my thought was that a relatively slow frequency would, as you say, both attract and repell the soap, so that it clumps together on the electrode first, and then gets repelled to settle on the bottom...

Let's see what Graham finds out...

Paul

for anyone interested... if currently truly is NOT important, you can get DC voltage multipliers that will easily produce 10kV (which is sufficient according to graham) for a 28 inch difference.

Kinda interested myself but unfortunately i have way too many irons in the fire as it is. My ears are perked up and i'm anxiously awaiting the progress of this thread.

The video is 2.4 MB so I apologise now for the size.

You can find it here ....

RIGHT-CLICK and choose "Save Target As" or "Save Link AS" to save it to your hard-drive.

Choose "Save as Type" = "ALL Files" for Windows Explorer, or "3GPP Movie Importer" if you have Firefox.

Then play it. You may need to download a freeware 3GP player if your existing media player doesn't support 3GP.

Sorry, my MPEG camera is just not playing ball today.

http://www.graham-laming.com/bd/nr.htm

This message has been edited. Last edited by: GrahamLaming,

Cant get that link to work in either IE or Firefox.

I get a london electronics site map...

I did something similar last year. This was an attempt to recover pollutants from first stage washwater. I stood a wire gauze cylinder on a plastic insulator set in the centre of a tinplate biscuit tin filled with the solution. Then applied 12 V battery charger. Lots of bubbles from the gauze (forget which was anode & cathode now). Liberated FFA and bio floated to the surface. Abandoned this approach as ion exchange with Epsom salts worked better.

Here's a Windows video version of the file.

Unfortunately only 50% of the video would convert.
Frustrating things - computers....

http://www.graham-laming.com/bd/nr.htm


Any better?

This message has been edited. Last edited by: GrahamLaming,

Graham,

Was that glycerine/bio mix de-meth'd?