Hello All, I have not posted alot on the board but have an operational GL1 appleseed with custom made purolite towers.

My question for all those out there is:

At what point do you say that all of the methanol is out of the biodiesel?

I have had alot of problems getting the soap to settle out fast enough (Using KOH) Which is why I installed the Purolite towers.

Now, I'm left wondering if I'm getting all of the methanol out of the biodiesel? Am I reversing the reaction? 3/27 test passes with flying colors. It goes clear instantly and even after 1 week there is no drop out.

I have temperature guages in the oil recirc pipe at the bottom of the tank and one at the top of my tank for indicating the temperature of the vapor comming off and being recirculated through the condenser.

I am using a mazzei injector with the check valve removed from the suction port.

How high should I have to drive the oil temperature? I normally take it to ~90C - 94C
and ensure that the vapor temperature starts to drop before sutting everything down.

Is there a quick way to test my biodiesel for methanol content?

In my last batch, I took samples every .25 gallons of recovered methanol. By 90C I had recoverd~ .5 Gallons in ~ 1 hour and let it sit at this temperature for bit and recovered another .25 (total of .75) after another 2 hours or so. Then, I turned off the heaters and let it continue to run in recirc until the oil had cooled and the temp of the vapor was about 120 or so. At this point I was sure i had all the methanol out as well as some water..

On these samples that I took at every .25 gal of methanol recovered, I ran wash tests on them every day for 4 days.

Day 0.
.25 Sample smells of methanol very strongly. but cleared after ~ 5 min of settling.

.5 sample has little to no methanol smell (Measure using my grossly uncalibrated nostrils) Sample was clear after ~ 1 hour of settling

.75 Sample has no methanol smell (Again using nostrils) Sample was a bit foggier than the .5 gallon sample.

1.0 Sample has no methanol oder. Very foggy

1 Day settling

.25 wash is very white
.5 sample wash white but clearer than the .25
.75 wash water is ~ the same as .5
1 Very white and sample is still very foggy

2 days of settling
.25 wash water is white
.5 wash water is clearer than day 1
.75 is ~ same as day .5 sample
1 wash is very white and sample is still very foggy.

What I have noticed is that the .5 and .75 samples have consistently clearer wash water than any of the other samples even on Day 5 of settling.

All samples pass the 3/27 test (I wanted to see if I was reversing the reaction at all

On day 6 the 1 gallon sample finaly started to clear but still has white wash water. The .5 and .75 samples are still clearer.

So, How do I know when I've got all the methanol out?

My last batch I took the temp all the way up to 106C watching the vapor temp and shut every thing down when the vapor outlet temp started to climb past 160F (~165F). I know that there is water in the methanol recovery but at this point I'm trying to dial in the biodiesel making portion of this thing.

Any numbers/temps that other users use would be greatly appreciated.

Also a test to tell if there is methanol (possible qty) in the biodiesel.

The only performance issue with my system that I have noticed, I'm sure is probably an issue that other users would run into as well. As I am boiling and distilling be it water or methanol, my pump's performance starts to drop and therefore i have less recirculation through the venturi suction. After I have reached a certain temp and shut off the heaters, the suction of the pump begins to return as there is less boiling taking place.


Sorry, This post ended up being alot longer than I expected.

quote:

Originally posted by cntsnap:

At what point do you say that all of the methanol is out of the biodiesel?

This information regarding your question was posted a few weeks ago in response to a similar question. You might have seen this already. If not, then you will probably want to read the entire thread as it is relevant to your inquiry.

My experience has been that when the methanol has been removed the soap drops out quite dramatically. If you are not getting adequate soap drop out, then maybe you still have too much methanol in your biodiesel.

quote:

when the methanol has been removed the soap drops out quite dramatically. If you are not getting adequate soap drop out, then maybe you still have too much methanol in your biodiesel.

Thanks! I'm going to give it a read. I did search prior to posting but did not find the info.

When you say the soap drops out quite dramatically, what time frame and is it the same with KOH (Thats what I use)

Thanks,

Air was bubbled through my biodiesel to effect methanol removal. A vacuum pump pulled a vacuum in the head space above the BD. A drop tube, open to the air on top and extending to the bottom of the flask, was used to introduce the air that bubbled through the BD. This arrangement was used to eliminate the chance of contaminating the BD with vacuum pump lubricating oil.

The air bubbled gently through the BD for about 35 to 40 minutes. No noticeable changes were observed in the BD during this time. Then almost like magic a soap scum started to form on the inside of the glass at the surface of the BD. After a few minutes the vacuum pump was turned off and the BD was allowed to settle down for observation. Not only was there a very noticeable soap scum on the surface, but soap particles were observed throughout the biodiesel. The soap particles started to settle as soon as the air bubbling stopped.

I don't think more than 10 minutes passed from the time the first soap scum was observed until the bubbler was shut off.

A soap test, using the AOCS method, was performed on the biodiesel after the soap particles settled out. The soap settled for about an hour. If I remember correctly the soap test showed the soap to be less about about 50 PPM. But, this is from memory. I will look around to find my notes.

What is interesting to note is the soap seamed to drop out of solution very, very quickly once the methanol concentration dropped below a critical point. It just seamed to happen "all at once" instead of gradually or proportionally to the time of bubbling.

quote:

ting to note is the soap seamed to drop out of solution very, very quickly once the methanol concentration dropped below a critical point. It just seamed to happen "all at once" instead of gradually or proportionally to the time of bubbling.

And this is with NAOH right? It should work the same with KOH?(I use KOH) Anyone out there very successfull at this with KOH?

I purchased a breathalizer to setup in my recirc tube after the condenser. I'm going to order a vacuum pump as well. Hopefully I will be able to relate a digital number readout with your done or something to that effect.

Now for a bunch of questions.

What is your procedure with temps etc?

Heat and distill until oil temp reaches a maximum of? Then draw vacuum? Or do you do the entire procedure under vacuum? Still, what temp do you go to?

I'm still confused as to why my samples with less methanol extracted (.5 and .75) have clearer wash water than than my 1 gal and 1.25 samples. This is what made me think that I was reversing the reaction but everything still checks out on the 3/27 tests.

Also, How deep of a vacuum are you pulling? I'm debating weather to place mine before or after the condenser. I'm leaning toward after...

And a ton of thanks for all your feedback.

I use sodium based catalyst. I have not used "K" based catalyst, so can offer no meaningful information of K based soap behavior.

The biodiesel in the above described test was at room temperature. In this case about 68 deg F.

The only vacuum needed in the above described test was the amount needed to overcome the "pressure" of a column of biodiesel about 8 to 10 inches (3 to 4 cm) high.

I don't have anything to offer as to why some of your samples have clearer wash water then other samples.

I was not trying to recover methanol in this test, just remove it from the biodiesel and observe how the soap behaved.

Producer: just a little correction. 8 to 10 inches is approximately 20 to 25 cm.
Did you mean that, or 8 to 10 cm (which is actually 3 to 4 inches approx.)?

quote:

Originally posted by bernyjb:
Producer: just a little correction. 8 to 10 inches is approximately 20 to 25 cm.
Did you mean that, or 8 to 10 cm (which is actually 3 to 4 inches approx.)?

berbyjb, Good catch.

8 to 10 inches.

20 to 25 cm.

I divided by 2.54 cm/in instead of multiplying.

Maybe I should just say "Hell it" and switch to NA based?

I heat all the way to 210F and recirc and inject fresh air for 30 min and still have issues. Maybe I am causing some weird stuff to happen by overdoing things.

If I am reversing the reaction, the only thing I can think of is that maybe they are all monoglycerides. The 3/27 test only tests for di and triglycerides right? Anyone know anything about the reverse reaction behavior?

Another way that I could tell is by a phlip test I guess. But, Where do i purchase those test kits from?

I did wave a breathalyzer over my settling barrel and got a .02 reading. So I'm leaning toward there still being methanol in the mix.

If I just setup an aquarium air pump to buble air into the settling tank, maybe this could get the remaining methanol to come out.

Any examples of a good bubbler for a 55gal drum?
I need something I can throw together fast and hopefully cheap.


I think I will try it in a 500ml jar first. Maybe I can catch all of the air output stream and relate a BAC reading with methanol content.

I said before that I waved the tester over the settling tank and got a .02. This was without bubbling air and is not an acuurate reading.

I know it wont be exact but for the human body,

.01BAC = 1 cg of alcohol /mL of blood.

This has to do with Liquid Vapor Equalibrium as well as Daltons Law, Ideal Gas Law and Raoult's Law. If I can get the proper ratios and values for equilibrium of methanol and bio then we can get a real conversion with more acurracy but until then..

1 cg/mL * 1g/100cg * 1000mL/L * 1cm3/.7918g * 1L/1000cm3
Density of Methanol
After conversion factors:

.01BAC = .01263 L of meth/ Liter of solution

so the forumula for toal methanol content:

Methanol = (BAC Reading) * Volume of BD * .01263

My very Inacurate .02 reading would indicate that there is .0253 L left. I'm sure that the only way to get an acurrate reading is to bubble air through the BD and force all air through the detector.


Now, Lets throw all of the above out as the only useful thing there is the conversion factor.

ASTM is <.2% by volume right?

Lets do away with the actual volume number and go strictly for content percentage.

If there was an exact relationship between Blood Alcohol Content and Biodiesel Alcohol content then:

Meth% = BAC * Unit Conversion Factor * 100

Meth% = BAC * 1.263

So:

BAC = meth%/1.263

I wish i could test this as it would mean that a fresh mix of oil and 22% meth would blow a 17.42!

My meter only goes to .40 so no luck there.

However, If we want to target .2% for ASTM then we need to know what BAC to shoot for.

BAC = .2/1.263

BAC = .158 My detector will detect this.


Now Producer has been gracious enough to give us some stats that the soap dropout happens when methanol is between .15 and .10 so we can solve for BAC

BAC reading = .15/1.263 = .119

BAC Reading = .10/1.263 = .079

So, By my calculations if my biodiesel can drive then the soap should drop out! That is of course assuming that there is a relationship here in which case I want credit for converting Blood Alcohol Content to Biodiesel Alcohol Content.

I will test this weekend and report back. Let me know if I'm wasting everyones time here.

quote:

I divided by 2.54 cm/in instead of multiplying.

Don't worry, happens to the best of us...
I didn't want to be a nitpicker, but both results seemed plausible to me, so I wanted to know which one was right.

quote:

Maybe I should just say "Hell it" and switch to NA based?

Cntsnap: nobody can tell you what you should do.
Of course, you CAN switch to NaOH, and forget about the problem. Or you can keep going at it until you can fine tune your process, and then share your acquired knowledge with others. I wish I could be of some help, but so far, my experience with biodiesel is limited to 2 Dr. Peppers, both with NaOH.
For what I see in your post (the last one), you seem to know a lot more than what you give yourself credit for (or at least, a lot more than me).
And no, you're not wasting anybody's time.

I'm currently getting some mixed results. It seems to have worked on the small scale but I'm working through a few bugs. I have some great pics that I will be adding as well

I started a topic on this test here:

Methanol test?