DieselDan,

I have yet to try the soap titration method so I can't report on soap conc. I knew that there was a method but never got around to it until now. After thinking through the implications of having such a powerful tool in my arsenal, I have to wonder why I never got up to speed on it sooner... My measurement methods thus ar have been Acid Values, pH, mass, volume, temperature, density and sending away for total and free glycerine wt%.

Now your questions...

For the yellow grease w/FFA <15% I use the method I learned from Van Gerpen at the U of Idaho in March. Acid pretreatment with H2SO4 and MeOH, phase split, neutralization with NaOCH3, Transesterification with NaOCH3 and MeOH in 80% and 20% 2-step, glycerol splits each time (I rarely, if ever, see glycerol the second time but there is a slightly browninsh color in the wash that I attribute to residual glycerol). I then boil off and recover MeOH from the BD, glycerol and acidified MeOH. Is this process called FATTA? What does FATTA stand for???

I've also thought that the excess MeOH from the 80/20 base step causes some of the glycerol to stay in solution with the BD and minimizes the phase split.

I've had some gelling occur several times after boiling off the MeOH as well. Sometimes it's more of a gel/skin on top of the batch. It will wash out with lots of water & time. I think that this occurs because the catalyst is possibly grabbing onto the BD and water or soap. I've since cut back on the catalyst and haven't seen a significant amount of gel since then.

I'm waiting to follow the process through while monitoring soap content. I have to use ~300L of soft water to bubble wash 100L of BD until the wash water is fairly clear. I use ~50L of water per 100L of BD. This is too much water & time.

I've made ASTM spec at the 100-gal scale. The 80%/20% base step really does drop the total glycerine an order of magnitude. I did a straight single base step and had 0.221%wt glyceride and got 0.021 wt% glyceride using 80/20 on the same batch of oil (ASTM Spec is 0.24max) - This has pretty much convinced me that this is a useful process - maybe it needs to be balanced against additional soap formation. I had one batch that didn't make ASTM spec for 0.261Wt% Glyceride and it also had an elevated ash content that exceeded spec.

I'm going to the Renewable Energy Workshop in Prince Edward Island on Thursday and I'll try to post any promising results late next week. Right now I figure that Magnesol would cost $0.04/L if used at 2% by wt. Anything in this ball park that saves me the agro of slowing everything down to wash will be cheap at twice the price.

Cheers.

foxcreek, what molar ratio of methanol are you using?

Dropout

Molar Ratio for the Transesterification step is 6:1 MeOH:TG. I've never played with this ratio but I do account for methanol that is unrecovered from the acid esterification phase split and carry it as inventory into the Transester step.

I'll add something on-topic so I might be permitted to ask an off-topic question.

I brew beer for a living and we filter it with Diatomaceous Earth (DE). Your local microbrewery would probably be able to get you some no problem. ($0.50 a lb. roughly) The stuff we use is nominally rated to 3 microns all by itself. For beer a pressure leaf filter is used because of beer's carbonation. It relies on a depth of DE to build up so as not the blind the filter. Which, I assume, is why you need the volume of a sock filter.

I'm really interested in the magnesol but I'm currently not doing any methanol recovery or removal (other than water wash) which sounds imperative. Then I'll be looking at it for sure. I'll just have to burn these bridges as I come to them .

Now what I'd like to know (foxcreek[?]) is the actual procedure/technique for the 80%-20% base transesterification. What are the mixing times, settling times, next stage mixing times and settling times, etc. It sounds like the easiest way to get a high conversion fuel.

Thanks in advance!

Chris E.

OK, I finally figured out the soap test. I am confident that my above batch of biodiesel made from oil that titrated at 3 ended up with 1310ppm soap.

At 1310ppm soap you need .0949 pounds per gallon of Magnesol R60 if you are going by the 1% by weight per 1000ppm soap. If you are going by the 1% by weight per 800ppm soap you will need .119 pounds per gallon.

What remains to be seen is how low can we go and still get a good clean fuel.

My first though is to run .119 pounds per gallon for 20 minutes and then filter it out.
Then do a soap test on the finished product.

Then I want to soap test my next batch of biodiesel and take the Magnesol R60 that was used on the above batch and run my next batch through it. Then soap test the biodiesel that was washed using "spent" magnesol. If the soap level stays the same then the magnesol was truely spent. If the soap level goes down, then it still had some life left in it. Then you use what you found out to, hopefully, decrease the amount of magnesol needed to treat a specific soap level.

It really comes down to what is it worth to you to be able to skip the pain of water washing.

-Jim

Chris Erickson

Do a search or the thread "Jon van Gerpen Two-Stage Base process!!!!!" as well as 2-step base process on this site and you'll find more than you want to know about the process!

Jim D - I like your experimient, I will try my own on 1-L scale next week. While 'playing' with Magnesol I've noticed that it really seems to drop out quite readily. I'm thinking that if I decant a sample from the BD layer that I won't have to filter to obtain soap results.

Is it the active adsorption while mixing that gives the soap removal or is it the filtering with the pre-coating of magnesol that removes the soap - or both? Any thoughts about this out there?

Thanks a bunch Foxcreek. Tons of good reading there. In my feeble searches I never quite get the right phrase for where I really want to go.

Was there ever consensus by those using the method whether removing too much glycerin early also took away too much methanol?

Chris E.

I haven't been following the discussion that closely but if you assume that the reaction is 80% complete with 80% of the methanol, then even if all the methanol from the 80% addition left with the glycerol, it would be replaced with the remaining 20% methanol necessary to compete the remaining 20% of the conversion. The argument is favor of this process is that the reaction equilibrium moves to a more complete conversion by removing the glycerol resulting from the 80% addition of the reactants. Some people try to remove the glycerol as it forms during the reaction by stopping mixing for a brief time, dropping the glycerol and continuing on. I don't do this; the reaction is the shortest part of the total process of making BD.

Cheers.

After you have boiled off the Methanol from your biodiesel you will need to do a soap test as described here:
http://www.biodieselcommunity.org/testingforsoap/
I use 15-20 drops of Bromophenol Blue and 10 grams of biodiesel. (I may run 100 grams in the future when testing the final product, but for now I'm having trouble because of the size of the Erlenmeyer flask that I use (125mL). It is hard to get 100mL of IsoHEET, 15-20 drops of Bromophenol Blue, and 100g of biodiesel in there and have room to put in your .01N HCL and still be able to swirl the contents. I think I am going to get some 250mL Erlenmeyer flasks.)

This will tell you how much soap is in your biodiesel so you can determine how much Magnesol R60 to use. My last batch was from oil that titrated at 3. I used 5g of NaOH for base and 3g of NaOH for titration per liter. I used 23 percent methanol. When I soap tested my dry, demethanoled, unwashed biodiesel it came out at 1310ppm. A few weeks ago I got some documentation from the manufacturer that states to use 1 percent by weight per 800ppm soap. That means that I needed to add between 1.64 percent by weight Magnesol R60 to my biodiesel. I tried 1.31 percent to start with (about 1% per 1000ppm.) I put a 400g sample of my unwashed, demethanoled biodiesel from above and put it in a mason jar. I heated it to 170 degrees F in a water bath. I added 5.24g of Magnesol R60 to it and capped the jar. I swirled the jar for 5 minutes to start with and then once every minute or two for a total of 20 minutes. Then I let it settle for 20 minutes until it looked clear. Then I pulled a 10g sample off of the top and did the Soap Test on it. It tested at 350ppm soap. So it looks like their guidance is right on. The best bet is to use 1 percent by weight per 800 ppm, or in this case 1.64 percent by weight Magnesol R60 to my biodiesel. To verify this I added 1.3g of Magnesol R60 to the remaining 390 grams of biodiesel. This time to oil was only about 110 degrees F to start with. After the same mixing and settling described above I again took a 10g sample from the top. It turned before I got the first 1/4 mL in. 1/4 mL of .01N HCL in a 10g sample works out to 82 ppm. So there was some soap left, but it was less than what I could accurately test without going to a larger sample. Call it >0, but <82ppm.
Unfortunately this means that we are not going to be able to go as low in the percentage by weight as we had hoped. It looks like 1 percent by weight per 800ppm soap is about right. The trick to keeping down the cost of the Magnesol R60 that you use is to minimize the amount of soap produced while still achieving good conversion. This will take some experimentation, but if you are making biodiesel at home, then you are probably not afraid to experiment. I will be testing and will post on my website, http://www.FryPower.com, and here what I find.

As it is now, at 1.64 percent by weight Magnesol R60 will cost you from 21.4 cents per gallon to 27.4 cents per gallon to dry wash your biodiesel, depending on how much it cost to ship a 50-pound bag to your location. ($75 per 50-pound bag at http://www.FryerPower.com as of tonight, plus shipping. It works out to 21.4 cents to a location about 150 miles from Fayetteville, TN, 27.4 cents per gallon to the west coast.) The question you have to ask yourself is are you willing to pay 20 something cents to never have to water wash and then dry your biodiesel again? I am.

Based on the testing done above, the 9-pound box of Magnesol R60 will treat around 76 gallons of average biodiesel. A 50-pound bag will treat around 420 gallons of average biodiesel.

I'm kind of bummed out that their guidance is right. I was really hoping that we could treat a bunch more biodiesel for a given amount of magnesol.

Anyone have any ideas on where to start on lowering the soap content? Do we go Acid/Base no matter what and get the titration as LOW as possible, then dewater before going to the base method?

-Jim
(PS I have not forgot about those of you that bought whole bags early. Contact me directly for your better pricing when you are ready to order more.)

foxcreek

quote:

Is this process called FATTA? What does FATTA stand for???

The FATTA method is where you add 100 mL MeOH and 1-2 mL H2SO4 and process till the ffa% drops low enough. I also use the ISU acid/base method, but I do not neutralization with NaOCH3. I recently got some lab results back and the total glycerine was the same, in fact everything was pretty close. Thanks for the info.

Jim D
In Girl Mark's soap test procedure, the 30.44 needs to be 3044 to get ppm (if using 0.01N HCl). I have been looking at the soap test forum.

quote:

I get:
(4*.01*40)/(1000*10) = .00016
Even if I multiply this times 1 million I only end up with 160 ppm soap.

You should have put in 304.4 for the MW of the soap, not 40 the MW of NaOH. So you should have gotten (4*0.01*304.4)/(1000*10)=0.0012176
0.0012176*10^6=1217.6 ppm soap.

I also constantly stir the Magnesol/BD mixture at 77 C for about 30 min. The highest amount of soap content in the BD after Magnesol was 147.

quote:

Anyone have any ideas on where to start on lowering the soap content? Do we go Acid/Base no matter what and get the titration as LOW as possible, then dewater before going to the base method?

I get about the same amount of soap formation with base method as with the acid/base method. If you are getting 1200ppm, that is good. I usually get 2000 ppm for the base and 2300 ppm for acid/base. I am wondering though if maybe if we washed the spent Magnesol with acid, maybe it will strip the soap, so that we can reuse it.

I'd be concerned that the Magnesol would be consumed, at least in part, by the acid... I just started doing the soap test and I get ~2800ppm for the unwashed BD. I'm sitting on my first 1000gal batch of this BD. I've been playing with different washing approaches on 200L at a time and trying to clean up the soap. Too much shear and it's emulsion time. I can break the emulsion with salt and continue washing but there's got to be a better way... Bubble washing works - as always but takes too long! Any thoughts?

The quickest and easiest method is Magnesol. No emulsion, less than a day to clean the bd, and no water wasted. I sent some samples out to be tested and find that most of the samples did not meet ASTM standards in sulfated ash and carbon residue, but they are just above the max. Comparing Magnesol with bubblewash, I found that the results are very close. I think that if I filtered better, I could get ASTM standard bd.

Maybe vinegar can be used to remove the soap from the Magnesol. It is weak, so I don't think that it would consume too much of it. I might be worth a shot.

If I remove all my glycerin and then let my biodeisel "breathe" for a time, let's say a week, will enough methanol evaporate to not interfere with the magnesol?

I don't have a set-up for methanol recovery but I understand it is mostly done to the byproduct, not so much on the biodiesel itself?

Thanks,

Chris E.

The MeOH should evaporate with time, but I would be cautious. Here is the MSDS. MSDS Methanol

When treating with Magnesol, the treatment temperature should be between 21 - 120°C. I usually process mine at 77°C. MeOH has a boiling point of 64.7°C, so any left over MeOH should be evaporate at that temperature. You really should look into MeOH recovery. I recover about 40%. Legal Eagle has a nice web page on his condenser. The B100-WH Recovery Condenser

I bought a 50 pound bag from Jim that now I will not use. Nothing wrong with the Magnesol solution, I just can not invest in the required filtration right now. So if you want a cheap 50 pound bag, please send me a private message.

txcas,
I've been messing with the filtration a little bit in my (limited) spare time. I got slow, but good results from a simple 1 micron felt sock filter. I cut a 6 inch hole in the top of a 5 gallon bucket lid with a sharp knife and dropped the sock filter into the hole. The filter has a metal 7 inch diameter ring in the top to it will not fall through the 6 inch hole. I put a grate in the bottom of the bucket and a spigot in the side of the bucket below the grate. The sock only holds about 1 to 1-1/2 gallons at a time. It filters kind of slow, but it does work. The biodiesel coming out the other side is crystal clear.

(No, I have not sent it out for testing yet. Been too busy and I can't find those stupid little sample bottles. They are here somewhere. As I said that I turned around and dug under a pile of books and found one!!!! OK, so now I can send out a sample to get it particle counted. I would like to see how clean it is with just the most simple filter out there. A simple $5-$10 felt sock filter.)

My point is, you don't need an expensive set up. A $3 bucket, a $1 lid, a $5-$10 sock filter, and a $5 spigot. Maybe a $12 grate for the bottom. It will be SLOW, but it seems to work.

-Jim

I filtered with the sock and found that the sock absorbed a lot of the biodiesel. When I vacuum filtered, I don't lose product. I have tested the bd filtered with a sock and it was just over ASTM levels in carbon residue and sulfated ash.

I put a lid over the lid with a hole in it. I did that to keep the dust/dirt out and to slow down the evaporation of BD from the sock. I can't imagine it holding more than 500 mL in the fabric. If you are filtering small batches, that is a lot. If you are doing 50 gallons it is not so much.
What I liked about it most is that I can dump the Magnesol out very easily. Once I get the results back from the particle count I will know more about the need to recirculate it through a 1 micron home water filter.
If that doesn't do good enough of a job on the really small stuff you could always add a little bit of the magnesol you sock filtered out back in to build up a cake on the water filter. You just don't want to send all of the magnesol to the water filter. It will fill the housing and the pipe leading to the housing and kill your pump. IMHO only on that.

-Jim

I am thinking a 30-10-1 micron filter setup using spin on filters. Anyone know how many PSI the Harbor freight "blue" water pump puts out. Do I need a flow guage/PSI meter in my system. Is there a better way to filter out the Magnasol. How do you recover the Methanol from the processor, do yo use the appleseed processor with high heat?

Ian