Hi all.
Recently we have been running very successful experiments using magnesium silicate to remove soap before washing BD with water. The results are nothing short of astounding.
Use 2% (by weight) magnesium silicate powder mixed for 20-30 min in biodiesel, filter through 1 micron filter, wash with water, no emulsion.
The wash water is so clear, one wonders whether the final wash/dry is even necessary.
This is a very inexpensive product with one of the friendliest MSDS sheets you will ever read.
Magnesium silicate is available commercially from Dallas Group. Their product name is Magnasol R60.
http://www.dallasgrp.com/.
Diff

Diff

sounds interesting, did you get much soapy looking stuff showing up in the filtering, and does your biodiesel normally require several washes to get clear water, ie did you have a comparison to the magnesium silicate batch?

Chug

Please explain '2 % by weight' (sometimes you have to draw me a picture). ie 25 gallon final product.

And - what can be done with the end waist product (safely)?

interesting, their web site says nothing about biodiesel, how did you find out about it? what does it cost?

What about all the other material in unwashed bio, ie partially unreacted oil, methanol, glycerol etc?

More on Magnasol
In their May 3, 2004 paper titled:
Evaluation of Magnasol R60 as an Alternative to Water Washing During Biodiesel Production
Jon Van Gerpen and Kirk Memges from Iowa State U, Posted the following table:
(hope this transfer works)
ASTM Specification ASTM D6751 ME treated
Specification 2% Magnasol
Free Glycerin, % 0.020 Max 0.004
Total Glycerine, % 0.24 Max 0.147
Flash Point. Deg. C 130 min 168
Water and Sediment. Vol% 0.050 Max 0.005
Carbon Residue, % 0.050 Max 0
Sulfated Ash, mass% 0.02 Max 0.002
Kinematic Viscosity, C 1.9-6.0 5.06
Total Sulphur, mass% 0.05 Max 0.00133
Cetane Number 47 Min. 57.4
Cloud Point, Deg C Report 9
Copper Corrossion No. 3 Max. 1a
Acid Number, Mg KOH/Gram 0.8 Max 0.32
Phosphorous, mass % 0.001 Max 0.0008
ppm Soap No Spec. 4
This table was from BD made from Yellow Grease, called WVO on this forum.
I'll try to get a URL so that you can read the complete paper.
As the title suggests, this may be the waterless wash for which we have searched.
One small technical issue. When I alluded to a 1 Mu filter, it is imperative that there be a layer of inert diatomacious earth on top of this filter, otherwise, the magnesium silicate will not "cake".
Diff
Senior: 2% by weight.
IE 100 kgs of BD requires 2 Kgs of Magnesium Silicate
PS Tried 3 times to get the above data to post as a table. No luck so far, but will follow up.

Hi Diff,

I was hoping to experiment with magnesium silicate soon, as I've read much about it. Thanks for paving the way for other home-brewers!

A few questions:

How have you developed your method, ie: 2% by weight, mixed for 20-30min, etc...?

Have you seen a difference in longer or shorter mixing times, and possibly more or less of a percentage of Magnasol?

Please describe you filtration setup. How can you achieve this layer of diatomacious earth on top of your filter? What type of filter/housing are you using?

Thanks Again!
-Nick

Why did this topic just stop? I'm really interested in what others have found out about magnesium silicate.

Hello Again
A number of you have asked questions which I hope to adress this evening. However, since last August I have been working essentially 100 hrs/wk. However, I too feel that the magnesium silicate issue is so important to this group, I thought it would be churlish not to share it. We are currently working on a set of tests, to determine if by using magnesium silicate, we can eliminate the use of water completely. They look promising. Our group has just been joined by a chemist.

Now I will try to attach the Van Gerpen/Menges Paper (referenced above) as a PDF file. If I fail, any suggestions are welcome. The PDF is on my computer desktop.
Well, as you can see, that did not work.
I got a pop up that said, I cannot post that kind of file. Only text and photo files can be posted.
Any suggestions?
Diff

If it's not already online somewhere, and Infopop doesn't allow us to attach pdf's, I"m pretty sure that you could attach it at localb100.com/forum in the 'links and other resources' forum: http://www.localb100.com/forum/viewforum.php?f=4
and then point to it from here.

Mark

I just want to say say thanks Diff.

Here is the best I can do in the few minutes I have. Once again, the tables did not transfer as tables, but the text is clear and the data in the tables are discernable with some effort on the part of the reader.
IOWA STATE UNIVERSITY
OF SCIENCE AND TECHNOLOGY College of Engineering
Mechanical Engineering Department
2025 Black Engineering Building
Ames, Iowa 50011-2161
515/294-1423
FAX 515/294-3261
http://www.me.iastate.edu/
Evaluation of Magnesol R60 as an
Alternative to Water Washing
During Biodiesel Production
Final Report
to the
Dallas Group of America
Jon Van Gerpen
Kirk Menges
Department of Mechanical Engineering
Iowa State University
May 3, 2004
1
Objective
The purpose of this project was to compare a commercial product called Magnesol R60, produced by The Dallas Group of America, Inc., to water washing for removal of soap, free glycerin, and total glycerin from biodiesel produced from soybean oil and yellow grease. The biodiesel pilot plant located at the Biomass Energy Conversion Center (BECON) Facility in Nevada, Iowa was used to produce the biodiesel and conduct the experiments.
Biodiesel Production
Biodiesel is an alternative fuel for compression ignition engines that has many benefits over petroleum-based diesel fuels including lower emissions and the potential for reducing global warming. It can be made from many different oilseed feedstocks and animal greases along with different alcohols and catalysts. The pilot plant at the BECON facility can produce biodiesel from soybean oil, yellow grease, and restaurant grease, using methanol and sodium methoxide as the other two ingredients of the reaction.
Soybean oil or pre-treated grease (grease from the esterification process described below) is reacted for 1-2 hours with the appropriate amount of methanol and catalyst (sodium methoxide) to produce biodiesel in a process called transesterification. Heat and agitation help speed the process of transesterification, so the temperature of the reactor is held at 140ºF. This temperature is slightly below the boiling point of the methanol. A condenser is located on the top of the sealed reactor for condensing methanol, which returns to the reaction. Agitation is provided using a constant speed mixing motor with a propeller on the top of the mixing shaft and a paddle on the bottom. Fixed baffles inside the reactor assist with mixing of the fluids. Sometimes the reaction is conducted in two steps. In the first step, approximately 80% of the methanol and catalyst are added and the mixture is agitated for an hour. Then, the free glycerin that has formed is allowed to settle and is removed. Then, the remaining 20% of the methanol and catalyst is added and the reaction is continued for another hour. Finally, additional glycerol that has formed is removed. This approach provides a very complete reaction while potentially using less methanol than the single step reaction. Upon completion of the reaction, the mixture is pumped from the reactor into the methanol recovery system for methanol removal from the fluid.
Biodiesel enters the methanol recovery system where it is pre-heated by an economizer heat exchanger before entering the main heater exchanger, which heats the fluid to 240ºF. The fluid then enters the flash vessel where methanol is flashed from the fluid and drawn through a condenser. The liquid methanol is collected in a storage tank where it will be sent to a methanol purification system for reuse in the biodiesel production process. Biodiesel exiting the methanol recovery system is pumped into the separation tank where free glycerin settles to the bottom of the cone bottom tank. The glycerin is pumped from the separation tank into a storage tank and will eventually be sent to a glycerin purification system. The by-products of biodiesel production, soap and residual methanol and free glycerin, are removed from the fuel at the biodiesel pilot plant by washing with hot, softened water.
The washing process in the biodiesel pilot plant involves spraying hot, softened water through spray nozzles on to the top of the biodiesel so the droplets fall through the fuel until they collect at the bottom of the tank with the soap that each droplet was able to extract from the
1
biodiesel. After dispensing an amount of water equal to 50 percent of the weight of the biodiesel fuel, the water, soap, and biodiesel mixture are recirculated in the tank by pumping the fluid from the bottom of the tank and spraying it through the nozzles in the top of the tank. This typically takes 15 minutes. Then 30-45 minutes is allowed for settling, which allows the separation of the water/soap (and some glycerin) mixture from the biodiesel. The water/soap mixture is pulled off the bottom of the tank and a fresh batch of hot, softened water is sprayed into the tank and the process of mixing and settling is repeated again for a total of 4 to 5 washes. The last step in the biodiesel production process involves the drying of the fuel.
The drying system consists of an economizer heat exchanger that preheats the incoming biodiesel before being heated in the main heat exchanger to 240ºF. The biodiesel is then sprayed through a nozzle into a flash tank under vacuum (around 28 inches of mercury) where the water is flashed off while the dried biodiesel is pumped from the flash tank into a storage tank. It is then analyzed in order to determine if it meets the fuel specifications for use in a compression ignition engine (ASTM D 6751).
The esterification process, or the process for pretreating grease prior to making biodiesel, is needed for greases or oils containing free fatty acid (FFA) amounts that are higher than 1.5 to 2 percent. The yellow grease used at the pilot plant is usually between 8 and 10% FFA. The pretreatment process is needed to reduce the FFA and prepare the grease for the base catalyzed process (transesterification). Without this process, soap would be produced when adding the base catalyst to the high free fatty acid grease in the biodiesel production process.
Pretreatment consists of determining the FFA of the oil or grease and then adding an acid catalyst and alcohol, in amounts based on the FFA, and heating and agitating the mixture for 1-2 hours before allowing it to separate into two phases. The acid and alcohol used in the pretreatment process at the pilot plant are sulfuric acid and methanol. The upper phase consists mainly of a water/methanol mixture and some free fatty acids, while the bottom layer is made up mostly of lower FFA grease or oil, some esters, and methanol. The two phases must be kept warm in order to prevent the grease or oil from congealing and trapping the methanol/water mixture. The mixture is kept at a temperature of 140ºF by heating coils inside the tank. The upper phase is removed and the lower phase is supplied to the main reactor for the transesterification reaction.
Experiment
In preparation for the comparison, biodiesel was made from soybean oil and yellow grease using the biodiesel pilot plant. The amount of soybean oil, methanol, and catalyst (sodium methoxide) used in the experiments is given in Table 1. The amount of yellow grease, methanol, and sulfuric acid for the pre-treatment (esterification process) along with the amount of pre-treated grease, methanol, and catalyst (sodium methoxide) used for the transesterification process is also given in Table 1. Note that a two-step transesterfication reaction was used for both feedstocks, but only the yellow grease required pretreatment.
2
Table 1: Biodiesel Reaction Material and Amounts Soybean Oil 1st Reaction2nd Reaction(80%)(20%)Biodiesel ReactionAmount ( lbs.)Amount ( lbs.)Soybean Oil @ 0.70% FFA520520Methanol95.6823.92Sodium Methoxide6.371.59Yellow GreasePre-treatment ReactionAmount ( lbs.)Yellow Grease @ 11.6% FFA480Methanol125.04Sulfuric Acid2.7880%20%Biodiesel ReactionAmount ( lbs.)Amount ( lbs.)Yellow Grease @ 1.62% FFA460450Methanol84.6421.16Sodium Methoxide8.212.05
For all reactions the temperature of the mixture of fluids was kept at 140ºF with constant agitation in order to achieve as complete a reaction as possible. Each reaction was timed for an hour and a half after the reaction reached 140ºF. After the reaction period, each batch of biodiesel was sent through the methanol recovery system, so methanol could be removed from the fuel before separation. Methanol must be removed from the fuel because it reduces the effectiveness of the Magnesol. The methanol also contributes to a lower flash point, which can affect the classification of the fuel as flammable or non-flammable. Each batch of biodiesel was tested in a Marten-Pensky Flashpoint apparatus in order to make sure the flashpoint of the fuel was well above 150ºF. In each case, as the flashpoint of each biodiesel batch was determined, a graph was used to determine the percent of methanol in the fuel based on the flashpoint. For each case, the methanol percentage in the fuel was determined to be low enough that it would not skew the results of the experiment. Once it was determined that the methanol percentage was low enough in each batch of fuel, half was set aside for water-washing while the rest was used in the Magnesol experiment.
The batches of biodiesel that were set aside for washing were washed according to the process used at the biodiesel pilot plant. This involved 4 washes using hot, softened water along with agitation, with fresh water used for each wash. After separation, the biodiesel fuel was dried using a flash vacuum drying process. All batches sent through the drying system were heated to 240ºF and then throttled from a pressure of 25 psig to a vacuum of 27-28 in Hg. A sample of fuel from each batch was tested for moisture in a Karl-Fischer Moisture Titrator to make sure the moisture level met the fuel specification required for biodiesel. Upon completion of the drying process and moisture determination, each batch of biodiesel was put in a 55 gallon barrel and labeled according to the feedstock that was used to produce the fuel.
To conduct the Magnesol tests, an agitated tank and its corresponding piping system were washed with hot water and caustic. As the hot water and caustic were being recirculated through
3
the degumming piping system, heat was supplied to the fluid to help with the cleaning process. This involved multiple washes and disposal of the water caustic mixture. Then, the unwashed and methanol-free soybean oil and yellow grease biodiesel were each added separately to the tank along with the corresponding Magnesol for each separate experiment. The Magnesol and biodiesel mixture was heated to 180ºF and recirculated for approximately 30 minutes before being filtered using a 5 micron sock-type filter. It was determined that diatomaceous earth needed to be added to the Magnesol and biodiesel mixture in order to assist in the development of a layer of “cake” on the filter. Without a layer of “cake”, the finer Magnesol particles tended to avoid being trapped in the filter and remained in the biodiesel. After each experiment the biodiesel was tested for soap level.
Results
Table 2 shows the properties of the biodiesel after being water-washed and dried compared with the original unwashed biodiesel. While the washing process reduced the soap level from 651 ppm to 13 ppm, the data show that the amount of free and total glycerin actually increased. The increase in free and total glycerin is unlikely to be correct and may indicate non-representative sampling of the biodiesel. The data also show a slight increase in the cetane number from 51.0 to 54.1. This is within the variability of the cetane test procedure.
Table 3 shows the effect of two different adsorbent treatments, Magnesol R60 and a compound identified as Sample B. Comparison to the unwashed biodiesel described in Table 2 shows that the Magnesol reduced the soap level from 651 ppm to 4 and 5 ppm,
Table 2. ASTM D6751 Results for Water-washed Soybean Biodiesel
ASTM Specification ASTM D6751 specification Unwashed, Untreated M.E.Washed And Dried M.E. Free Glycerin, % 0.020 maximum 0.033 0.084 Total Glycerin, % 0.240 maximum 0.209 0.282 Flash Point, °C 130 minimum >190 170 Water and Sediment, vol. % 0.050 maximum 0.10 0.15 Carbon Residue, % 0.050 maximum <0.010 0.050 Sulfated Ash, mass % 0.020 maximum 0.000 0.005 Kinematic Viscosity, cSt@40°C 1.9 - 6.0 4.127 4.207 Total Sulfur, mass % 0.05 maximum 0.00006 0.00014 Cetane Number 47 minimum 51.0 54.1 Cloud Point, °C Report 0.0 0.0 Copper Corrosion No. 3 maximum 1a 1a Acid Number, mg KOH/gram 0.80 maximum 0.32 0.31 Phosphorus, mass % 0.001 maximum 0.0007 0.0006 ppm Soap No specification 651 13
4
Table 3. ASTM D6751 Results for Adsorbent Treated Soybean Biodiesel
ASTM Specification ASTM D6751 specification 1% MAGNESOL R601% DALLAS SAMPLE B Free Glycerin, % 0.020 maximum 0.005 0.003 Total Glycerin, % 0.240 maximum 0.191 0.168 Flash Point, °C 130 minimum 200 189 Water and Sediment, vol. % 0.050 maximum 0.04 0.10 Carbon Residue, % 0.050 maximum <0.010 0.013 Sulfated Ash, mass % 0.020 maximum 0.000 0.003 Kinematic Viscosity, cSt@40°C 1.9 - 6.0 4.097 4.109 Total Sulfur, mass % 0.05 maximum 0.00002 0.00006 Cetane Number 47 minimum 51.3 50.9 Cloud Point, °C Report 0.0 0.0 Copper Corrosion No. 3 maximum 1a 1a Acid Number, mg KOH/gram 0.80 maximum 0.27 0.38 Phosphorus, mass % 0.001 maximum 0.0005 0.0000 ppm Soap No specification 4 5
and also reduced the free and total glycerol level. The acid value was decreased for the Magnesol R60 and increased for the Sample B. These changes in acid value are not believed to be significant and are probably within the range of error for the measurement.
Table 4 shows the corresponding results for the yellow grease-based biodiesel. Note that an important difference between the biodiesel from this feedstock and the soybean-based biodiesel is the high initial soap level of 2458 ppm. When this biodiesel was passed through the methanol recovery system, the removal of the methanol caused the soap to come out of solution and plug the screens used to protect the pumps. This prevented the methanol recovery system from working properly. To ensure that the methanol was removed, a portion of the biodiesel was processed through a batch-type solvent recovery still. Thus, the results presented in Tables 4 and 5 show data for samples of biodiesel that had not been through the still and a sample that had gone through the still.
In Table 4, the water-washed biodiesel showed a reduction in soap from 2458 to 91 ppm. There was also a reduction in free glycerin although still not to the level required by the biodiesel standard.
Table 5 shows the properties of the biodiesel samples that had and had not been treated with the still and then processed with the Magnesol R60 The sample that had not gone through the still was treated with 3% Magnesol R60 and the sample that had gone through the still was treated with 2% Magnesol R60. The results for the two cases are very similar with equal flash points, indicating similar methanol contents, and reductions in soap to 14 ppm for the sample that had not gone through the still and 4 ppm for the sample that had gone through the still. Both sample met the requirements for ASTM D 6751 except that the sample that had not been through the still was a little high on water and sediment. This may have been caused by Magnesol breaking through the filter.
5
Table 4. ASTM D6751 Results for Water washed Yellow Grease Methyl Ester
ASTM Specification ASTM D6751 specification Unwashed, Untreated M.E.Washed And Dried M.E. (not through still) Free Glycerin, % 0.020 maximum 0.063 0.037 Total Glycerin, % 0.240 maximum 0.220 0.185 Flash Point, °C 130 minimum 179 >190 Water and Sediment, vol. % 0.050 maximum 0.70 0.06 Carbon Residue, % 0.050 maximum 0.060 0.013 Sulfated Ash, mass % 0.020 maximum 0.007 0.004 Kinematic Viscosity, cSt@40°C 1.9 - 6.0 5.095 5.107 Total Sulfur, mass % 0.05 maximum 0.00146 0.00139 Cetane Number 47 minimum 57.8 60.3 Cloud Point, °C Report 10.0 9.0 Copper Corrosion No. 3 maximum 1a 1a Acid Number, mg KOH/gram 0.80 maximum 0.21 0.27 Phosphorus, mass % 0.001 maximum 0.0009 0.0008 ppm Soap No specification 2458 91
Table 5. ASTM D6751 Results for Adsorbent treated Yellow Grease Methyl Ester
ASTM Specification ASTM D6751 specification 3% Magnesol R60 (ME not through still)2% Magnesol R60 (ME through still) Free Glycerin, % 0.020 maximum 0.009 0.004 Total Glycerin, % 0.240 maximum 0.151 0.147 Flash Point, °C 130 minimum 168 168 Water and Sediment, vol. % 0.050 maximum 0.10 0.005 Carbon Residue, % 0.050 maximum 0.000 0.000 Sulfated Ash, mass % 0.020 maximum 0.003 0.002 Kinematic Viscosity, cSt@40°C 1.9 - 6.0 5.089 5.060 Total Sulfur, mass % 0.05 maximum 0.00129 0.00133 Cetane Number 47 minimum 57.1 57.4 Cloud Point, °C Report 10.0 9.0 Copper Corrosion No. 3 maximum 1a 1a Acid Number, mg KOH/gram 0.80 maximum 0.33 0.32 Phosphorus, mass % 0.001 maximum 0.0008 0.0008 ppm Soap No specification 14 4
6
Conclusions
The Magnesol R60 provided a greatly improved reduction in soap and free and bound glycerin when compared to the traditional water washing process. Final soap and free and total glycerin levels were less for the Magnesol than for water washing. While the biodiesel producer has the additional costs of the Magnesol and disposal of the spent material, this should be justified by the reduction in water usage, wastewater treatment, and the additional processing steps needed to separate the emulsions that often form when water washing biodiesel containing high soap levels.
7Return to TOC

I've got to try this sometime.

Thanks for bringing this to our attention Diff!

This magnesol stuff....we ARE talking about "Fryer Powder", right? The stuff you add to fryer oil to get rid of bad taste and bits of food??

It would be interesting to pretreat the wvo with magenesol(talc?), silica, and/or other silicon oxides and determine how much the free fatty acid levels are reduced.

I am share that these tests have already been conducted because veggie oil refiners use clays to remove free fatty acids from virgin oils. So maybe it is just a matter of acquiring the information and adjusting for batch use of homebrew.

Hello everyone,

Please pardon me if I am making a mistake. This is my first time on this forum and I would welcome any help on doing things correctly. My question involves palm oil. Does anyone know if it is better to make biodiesel from palm kernel oil or from palm oil? The kernel oil is oil ground from the nut and the palm oil is ground from the pulp around the nut.

By the way, I did try to go to a different topic on this, the one entitle Palm Oil, but it looked like it was inactive....no posts since 2004.

Thanks for any help.


mike

Diff,

I could not figure out how to PM you. The full version of acrobat can save that pdf in other formats that can be uploaded here. If you like, you can email the pdf to me and I can work on that.

-steve
smurfy@budweiser.com

Ok question : what is and where do we get "diatomaceous earth " ?? in easy terms so a simpleton like me can understand - is it like "Fullers Earth" used in the jewellery trade?

And thanks for such an interesting article and wonderful idea.
Rob

quote:

Nothing is true - Everything is Permitted.

quote:

Originally posted by Mike Lasche:
Hello everyone,

Please pardon me if I am making a mistake. This is my first time on this forum and I would welcome any help on doing things correctly. My question involves palm oil. Does anyone know if it is better to make biodiesel from palm kernel oil or from palm oil? The kernel oil is oil ground from the nut and the palm oil is ground from the pulp around the nut.

By the way, I did try to go to a different topic on this, the one entitle Palm Oil, but it looked like it was inactive....no posts since 2004.

Thanks for any help.


mike

Hi mike and welcome to the infopop forum
it doesn't matter if a thread has been inactive for a while, just post a reply after the last message and it will revive the thread, regarding your palm oil question, a quick search brought up these.palm oil search

Chug