i did a batch of 120 gallons and used 10koh/l
well, i put 20 gal of meth in the meth tank and then thought i would try putting the cat in small amounts of alchy. i was putting 500 gms to a half gallon and it of course heated and dissolved rapidly.
the meth would start to boil and as i stirred the koh dissolved in about 2 mins. that was nice.
the oil separated ok, but i am concerned that i might have degraded the methoxide or caused extra water, this chem question is out of my league. if it is no problem, great, but if so, i hate to spol large batches and this b.d. seems a little thicker and darker than the test batch.
any answers???
bio lord tilly....where are you oh great one!

Skip you will do no harm to the KOH or the methanol by getting them hot. It could be however that you lost some methanol by boiling it off. If so, your conversion would have been a bit less, and this would account for it being a bit less easy than usual.

Skip I am just a teacher and oricle. I leave the chemestry to people such as 007 who actually know what they are talking about.

Tilly SBC/IBA Oricle AND greatness div

The journey of a thousand miles begins with a broken fan belt and a leaky tire.

Just a suggestion, but have you titrated the end product to see if it's completed the reaction?
Cheers,
JohnO

Good thought Johno.

Titration of the product will show up a shortage of NaOH in the original reaction. It won't show up a shortage of methanol however.

All of my batches of BD have tested slightly acid.
Titrating the final BD would probably give very little indication of reaction completeness.
Any FFA present would have been neutralised by the NaOH in preference to catalysing the reaction.
Titrating the glycerine would give an indication of the presence of excess NaOH. At least 3.5g/L related to the oil volume should still be present as unconsumed catalyst.
If your feedstock is consistent, test results from subsequent batches/trials for SG and viscosity will give a baseline to aim for.
Regards Squarepeg.

PaulG

Some points:

"All of my batches of BD have tested slightly acid."

Then you must have added acid or didn't add enough NaOH. If you added acid, titrating the finished BD will tell you nothing. If you take a sample before you add acid and titrate it you may learn whether you added enough NaOH. If the titration is zero you may have added enough. If the titration is more than zero you certainly have not added enough.

"Titrating the final BD would probably give very little indication of reaction completeness."

Yes it is possible that if you added nearly enough NaOH it would titrate zero but still not be complete. This would happen if you added enough NaOH to neutralize the FFA but not enough to act as catalyst.

"Titrating the glycerine would give an indication of the presence of excess NaOH. At least 3.5g/L related to the oil volume should still be present as unconsumed catalyst."

Not so. The reason why you need a specific amount of catalyst is that it is consumed in an unwanted side reaction, making more soap.

"If your feedstock is consistent, test results from subsequent batches/trials for SG and viscosity will give a baseline to aim for."

About the worst way to test for completeness of reaction is to use SG. I know it has been suggested by Tickel but the world has moved on. There is not much difference between the SG of BD and the original oil. There is much more difference between the viscosities so it is a better test but must be carefully done. Both tests are completely useless if an unknown amount of methanol is present.

[This message was edited by neutral on 28 March 2002 at 06:24 AM.]

PaulG

In my first paragraph above the sentence "If the titration is more than zero you certainly have not added enough" applies only before the BD is washed. After washing a small amount of FFA may appear over time.

Dualfuel

NaOH is obviously a reagent in the reaction in which the FFA are neutralized to make soap. But it is truly also a catalyst because without a surplus the reaction won't go. The interesting thing is that the surplus has to be quite substantial, which shows that it must be being consumed in another reaction. That reaction must be the making of soap directly from the triglyceride. You are therefore right in suggesting that it is a reagent in the soap reaction.

[This message was edited by neutral on 28 March 2002 at 10:35 AM.]

Neutral,
Acidity.
Some of my early batches of BD and my recently rescued batch were washed with acetic acid in first wash. The acidity of these batches was as expected higher than untreated batches.
All washed batches of BD have tested slightly acid over a wide range of NaOH levels. In fact I have never had a batch test alkaline.
As you say testing the unwashed BD for acidity would give more useful information.
SG/Viscosity.
While there is a large difference between viscosity of origional oil/fat and BD produced from it, the viscosity difference between well processed BD and not so well processed BD is of a lower order of magnitude.

Example. Some figures from #1004

Feedstock: Waste Animal Fat (Frytol)
Titration: 2.2ml (indicates 5.7g/l NaOH)
Methanol level: 225ml for every 1L fat.
Reaction Temperature: 55 deg C
Mixing: Vigorous shaking for 15 sec in plastic bottle.

Results, washed BD.

NaOH, Acidity SG Viscosity.

4.3, 7, 0.8798, 37.96
4.7, 7, 0.8774, 36.89
5.2, 8, 0.8735, 35.68
5.7, 8, 0.8722, 35.9
6.3, 4, 0.8708, 34.86
6.8, 6, 0.8695, 34.73
7.3, 5, 0.8683, 34.3
7.8, 6, 0.8711*, 34.48*,expt error?
8.3, 8, 0.8685, 34.11
8.8, set solid, no BD

NaOH in grams per litre WAF
Acidity in drops 0.1% NaOH
Viscosity in seconds drain time for 100ml pipette.
Regards Squarepeg

[This message was edited by PaulG on 29 March 2002 at 06:38 PM.]

i have never seen it put so well....

NaOH, Acidity SG Viscosity.

4.3, 7, 0.8798, 37.96
4.7, 7, 0.8774, 36.89
5.2, 8, 0.8735, 35.68
5.7, 8, 0.8722, 35.9
6.3, 4, 0.8708, 34.86
6.8, 6, 0.8695, 34.73
7.3, 5, 0.8683, 34.3
7.8, 6, 0.8711*, 34.48*,expt error?
8.3, 8, 0.8685, 34.11
8.8, set solid, no BD

so you are showing that right after 5.7 is a breakpoint. funny how more lye makes the ph drop??? or was that a titration? either way, it shows a changeover. i have always run a little hot with the cat,(thinking i would grab a little more ester) and the water wash shows it, now i know why.
an excellent post and i wont run hot anymore! should change my water wash losses.

Yes Skip the figures for acidity represent the number of drops of 0.1% NaOH solution required to produce a permanent pink colour in 1ml of finished BD dissolved in 10ml of isopropyl alcohol.
Higher drop numbers would relate to higher acidity which is lower pH.

I favor high NaOH levels for better conversion but the glycerine must be drained off soon after it separates. Prolonged (ie overnight) contact between the BD and the alkaline glycerine produces jelly at the interface which can introduce wash problems.

Regards, Squarepeg.

Skip the above figures only tell part of the story.

a)4.2g/L,b)4.7g/L,c)5.2g/L,d)5.7g/L*,e)6.3g/L,f)6.8g/L,g)7.3g/L,h)7.8g/L,i)8.3g/L,j)8.8g/L.

Settling of Raw BD.
16 hours settling (glycerine not removed)
a) Indistinct, a lot of mealy white material.
b) slight ammount of meal at interface.
c) small ammount of meal at interface.
d)* good separation.
e) good separation,10% jelly at interface.
f) ... .. .. .. 25% .. .. .. .. ..
g) .. .. .. .. 33% .. .. .. .. ..
h) .. .. .. .. 50% .. .. .. .. ..
i) .. .. .. .. 100% .. .. .. .. ..
j) set solid

Levels below titrated ammount gave poor separation.
Levels above titration ammount produced jelly at the glycerine BD interface upon standing 16 hours.
This jelly can be avoided by draining the glycerine at 1 hour and then again at 12 hours.

Insoluble material.
Samples of BD chilled to 9 deg C and held for two days. Thawed, white deposits in all samples.
Vacuum filtered through No 1 Whatman filter paper in buchner funnel.
Ease of filtering.
a). Medium,gelatinous + whitish deposits.
b). medium,gelatinous + crumbly whitish deposit.
c). fair, slight gelatinous+crumbly whitish deposit.
d)*. good
e). good
f). difficult,a lot of jelly like material
g).
h). medium, a lot of jelly like material.
i). easy, a lot of jelly like material.

Further settling of insoluble material.
Stood at room temp for 5 days.
a). BD fairly clear, 25% white sludge = 1/4 cm dark glycerine.
b). BD fairly clear, 33% white sludge +1/2 cm of dark glycerine.
c).BD murky, 50% white sludge+1 cm of dark glycerine
d)*. BD very clear, settled and washed to a clear stable product.
e).BD murky, gelatinous flock + slurry.
f).BD murky, gelatinous flock + slurry (less than for e).)
g). BD murky, gelatinous flock.
h). BD fairly clear, 50% flock + some dark glycerine.
i). BD fairly clear, 33% flock,slight ammount of glycerine.

Bubble wash.
Wash vessel 2l cylindrical separating "funnel", 95mm diameter. About 1L of raw BD washed with equal volumes of water.
Wash at ambient temperature (26 deg C).
a). A lot of persistent foam, at finish of washes white meally material present in BD.
b). Persistent foam, intermittent bubble washing required.
c).Washed easily.
d)*.Washed easily, little foam,separated well, decanted easily.
e).Washed easily, Some meally solid at surface of BD.
f). Washed easily.
g). Washed easily very small ammount of foam, Bd separated from water well as large discrete globules.
h). Washed easily, very small ammount of foam, BD separated from water well as large discrete globules.

I will leave you to draw conclusions from the above.This data is presented as a guide only and any conclusions should be tested using the feedstock and conditions appropriate to your situation. NaOH was used in the form of conc aqueous solution in these trials.

Regards Squarepeg.

PaulG

"All of my batches of BD have tested slightly acid."

It is impossible for unwashed BD to be acidic if acid has not been added and enough NaOH has been added to neutralize the FAA in the oil and to fully complete the reaction.

There is a point that may be relevant here that hasn't been mentioned on this forum as far as I know. It is possible for alcohols to become acidic over time. In the test for FFA it is standard procedure to do a "blank". This means repeating the test using the same amount of alcohol without putting any sample in it. The amount of titration you get to the usual colour with the blank is subtracted from the titration with the sample. Thus any acid in the alcohol is allowed for. Some of the samples you thought were acidic may not have been.

After washing a small amount of FFA may be present in the BD but if no acid was used in preparation it would have to be very small. If you find appreciable acidity your alcohol is probably acidic. I found my own isopropyl alcohol had become quite acidic recently. It was originally AR with negligible acidity. Also have you checked the pH of the water you use for washing?


I note from your reports that sample (d) handled the best. Did the amount of NaOH used, 5.7 g, correspond with the amount suggested by titrating the original oil?
.

[This message was edited by neutral on 30 March 2002 at 04:52 AM.]

Neutral,
There could be a problem with acid development in the BD but unlikely. Test results have spanned a couple of bottles of AR grade. One part full and another freshly opened.
Yes the 5.7g/L figure was what the titration suggested. Because the posting of results for this test batch were spread over two posts it wasn't obvious. I marked the 5.7g/L figures in the second post to indicate that they were as suggested by titration, not because I thought that these gave the optimum results.
Perhaps the best way to resolve the acidity of BD problem will with a batch of BD which is currently being tested through BAA. Will see if my results agree with the official results.
(Feel free to contact me off board if you wish further information on this batch).

Regards,
Squarepeg.

PaulG et al
At last a worrying point has been cleared up for me. I have been having nothing but
trouble with my titrations. And it is all down to a misunderstanding. I have followed
religeously Mike Pelly’s notes, and he has led me (unintentionally no doute) up a
blind ally. Part of his Updated Method notes has a picture of a cylinder containing a
titrated sample of wvo. The caption is something like “picture showing the correct
colour of the titration” and it is PURPLE !! Damn near blue. And all this time I have
been wondering why mine would only go pink (tending to majenta). Yet I know that
phenolphalein goes pink because I use it every day at work. What is this revelation ??

From PaulG
Yes Skip the figures for acidity represent the number of drops of 0.1% NaOH solution
required to produce a permanent pink colour in 1ml of finished BD dissolved in 10ml
of isopropyl alcohol

A Permanent PINK....

Thanks PaulG, you’ve saved my sanity. I wonder how many others have been totaly
thrown by this picture.
Regards all,
A very much happier dva

PaulG

Please do a titration on your alcohol by itself and let us know your result.

I note also that you are not reporting the pH of water shaken with the BD. That would be interesting as it leaves out the alcohol and extracts most of any remaining soap. It could be alkaline.

There is a fine distinction here. While pH measures acidity or alkalinity, what you are reporting is the amount of alkali you have to add to move the pH to about 8. You are reporting a "capacity", not a pH.

It is quite possible for a material to have both a capacity to absorb alkali and a capacity to absorb acid at the same time.

If your material has some capacity to absorb acid as well as alkali it is quite possible that the pH of water shaken with it could be above 7. If this is the case you would not describe your BD as "acidic"

I am also curious to know whether, in your extensive experiment, you measured the final yield of BD. It seems that high levels of NaOH gave a higher conversion as judged by the viscosity, but there may have been a price paid in reduced yield, either because of more side reaction making more soap, or more emulsion causing higher losses.

[This message was edited by neutral on 30 March 2002 at 10:31 PM.]

Is the titration number the number of DROPS used or the number of milliliters used.
Ive always used the number of ml.

John

This problem arose because of the movie:

FAT OF THE LAND
"Produced by Niki Cousino, Sarah Lewison, Julie Konop, Florence Dore, and Gina Todus
56 Minutes * Ages 15 to Adult * 1996
Who would have taken a bet that anyone could drive a Chevy Van across the United States fueled by left-over vegetable oil? Five enterprising young women did just that travelling 2300 miles, using vegetable oil as an alternative fuel in their Ford van They proved there is a sustainable and renewable alternative to petroleum."

The women involved then wrote a book about biodiesel and in it when they talk about titration they confusingly say:

"To determine your proportions, figure:
The number of drops of NaOH needed for the WVO solution to reach a ph of 8-9 1 ml. plus 3.5 g. NaOH to catalyze the oil"

Many other sites just coppied these instructions exactly about titration which continued the confussion.

Tilly SBC/IBA

It's morally wrong to allow suckers to keep their money.

Neutral,
Did not include pH measurements as I wasn't confident of their accuracy. BD + water + universal indicator resulted in dissolution of the indicators.
Resorted to using 1ml BD + 1ml isopropanol + univ indicator.
Results, again from #1004.
"acidity" drops 0.1% NaOH.pH following.
4.2g/L-7 drops-pH5.5
4.7 7 5.5
5.2 8 5.5
5.7* 8 5.5
6.3 4 5.5
6.8 6 5.5
7.3 5 5.5
7.8 6 5.5
8.3 8 5.5

Not confidant that the test represents a true indication of the pH.
Any acidity of isoproply alcohol would have influenced these results.
will not be back in lab for a week or so but will test isopropanol and report result.

Had not considered the capacity angle. A bit like buffering?

Most standards for BD quote a figure for Neutral No. this seems to range from <0.5 to <0.8 mgKOH/g.
This I guess would be to test the capacity you mentioned more than the acidity of the BD as I had assumed.

All the batches of BD in the above investigation were washed until a clear wash water with a pH of 7 was obtained.

Yields. I think I quoted yields somewhere above.
For these tests the reaction mixture was allowed to settle out overnight in 1L measuring cylinders.
Contact of the BD with the alkaline glycerine produced a jelly like layer above the glycerine.
I thought that this might have been ploymerised BD but you suggested it was a combination of soap and BD. The figures below give yield volumes of raw BD as a % of origional WAF volume.

4.2g/L 72%
4.7g/L 99%
5.2g/L 98%
5.7g/L* 95%
6.3g/L 95%
6.8g/L 92.5%
7.3g/L 84.5%
7.8g/L 66%
8.3g/L 61%

I suspect that if the glycerine is drained at 1 hour to avoid overnight contact between BD and high alkalinity glycerine, the yield for the higher levels of NaOH would be much improved, perhaps in the low 90s. Yet to confirm with tests, but am currently mist/bubble washing a 40L batch which used a methoxide of 10.25ml conc aqueous NaOH and 250ml methanol per litre.
(shaken in 10L batches in 15L square plastic bottle, process too slow going over to pump and static mixer).
This very high (too high) level of NaOH gave good conversion and the raw yield appeares to be quite close to 40L. Will advise on final volume and basic tests.

JohnH. Yes mls are the usual units used to calculate the additional NaOH required to neutralise the Free fatty acids.
In this case I quoted drops as they seemed to make more sense than decimals of a ml.
The number of drops per ml for a given liquid depend upon the tip area and shape of the device used to produce the drops. A burette generally gives 20 drops per ml. The 50ml Sthul capped bottle I use gives 16. Test your dropper either by drops per delivered volume or drops per gram with water.

Regards, Squarepeg.