Hi GCG

You have raised a valid point - samples taken after reaction and methanol removal are notoriously difficult to assure as being representative of the batch as a whole.

The glycerol drops out very readily, so the lower region of the tank is far more concentrated than the upper, even with the pump running full bore.

I took off the sample whilst mixing, from the pump's outlet.

This sample probably had more glycerol in it than the batch average, because the pump draws from the bottom.

That is why I now do a whole-batch titration now, rather than a sample titration.

quote:

Originally posted by GCG:
I really had hoped that the levels were higher than the 0.15%-.20% since these are essentially ASTM levels and require significantly more energy/time/money.

The flash points of the samples containing 0.15% to 0.20% methanol are about 116 deg C to 122 deg C. I use an older "closed cup" instrument to determine the flash point. The instrument is no longer "certifiable", but is still reasonably accurate.

Producer, I believe the new ASTM requirement mirrors the EN's at .20% methanol or Flash point >130 deg. C.

http://www.biodiesel.org/pdf_files/fuelfactsheets/BDSpec.PDF

Since Regular #2 Diesel has a flash point of ~98 deg. C which if your biodiesel contains 0.20% methanol it effectively has the same flash point.

GCG

You have to meet both and since most labs are not set up to do the EN14110 test, you might as well use D93 and shoot for 130C.

The samples I reported on in the previous posts still contained a too much methanol to meet the ASTM flash point. The flash point was reported as a reference only. The main point of the posts was the % methanol remaining when the soap/glycerol particles started to drop out of the biodiesel.

Hi,
I was stuffing around trying to save energy with my bio processing back in late 2004 by not removing the glycerol before distilation, I had some very good results,not being a chemist I got some advice from a couple of scientists with a background in the petrolium oil industry and ran my ideas past them,they also advised me on the construction of my reactor.

I was confident with my sucesses and even got some INCOURAGING results with the GC, BUT that was as far as I was able to go
the main problem was heating the batch fast enough and without stiring it up through the distilation process, The glycerol
was allowed to settle out straight after the reaction for an hour or more at 50-55c I have three temp gauges top, middle, and bottom on the reactor.
Now this is where the problems start before turning the heat up I stir the batch at 6000 rpm
using a 150mm shearing type disc impeller for a couple of minutes under vacuum to virtually homoganise the batch, stop the agitator and turn the heat up to 70c initially using three 3500w 240v elements 75mm from the bottom of the reactor and only takes ten minutes before the distilation starts but the tempratures inside the vessel vary greatly as the glycerol settles
the reactor tempratures get hotter at the bottom, stiring it up again stops the distilation and you need more heat to start it again, It's as if settled glycerol acts like an insulator slowing down the recovery if it's stired up again even at 150 rpm.

It's a process problem that can't be over come using electric elements, I added another three elements total six 3500w,
and ended up no better off except the batch got to 70c much quicker but the variation in temprature persisted. my reactor is not insulated but I used a reflective silver bubble wrap cover that's good for 100c and the reactor can be closed in so there is very little heat loss.

I have been told the only way to get a uniform heat or close to it would require a internal heating coil from top to bottom using steam , heat transfer oil, or even water or a completly Jacketed Reactor and it has to be heated as quickly as possible before the glycerol settles out, also the distilation stops when the control thermostat cuts out and has reached it's set point 70c, which means most of the distilation is happening around the heater elements at a much highre temprature while the elements are on.

The temp at the top of the reactor is 67-68c the middle is 70c where the bottom of the thermo well is located and at the bottom where the glyceroll accumilates is 72-75c under - 80 to -95Kpa Vacuum, under those conditions the methanol should continue to distill regardless of the variation in tempratures,the recovery rates on Identical titrations and feed stock can vary as much as 20% but are atill much better than the total recovery rates of seperatly distilling bio then glycerols.

I don't make bio any more, I regret now having disposed of my notes and the CG reports, my advisors have moved on to better things and I have lost contact with them, there was not much interest in a single process recovery back then and bio seems to have vanished or gon completly underground in Australia Judging by the dates and number of local posts, I glance with some interest now and again and this thread caught my eye.

The Scientests made me look at the process and processor rather than the chemestry, the chemistry is theory and should work The reactor needs more work which unfortunatly requires buckets of money most of us don't have.
Making a small test batch in glassware in a lab is not the same as making it in 20 ton batches I used to work in a chemical plant and know well the stuf ups that happen when things are scaled up just a bit.

Regards,

Graham,

Didn't you say you could prevent the reverse reaction if you did the recovery hot and fast?

Hi Fleabitten

Could it be the static pressure being higher at the bottom of the tank was enough to inhibit boiling?

If your methanol concentration had dropped enough, your methanol would not be able to boil at 70C under -95kPa AT THE TOP OF THE TANK , with the pressure being much higher at the BOTTOM of the tank due to the weight of liquid.

What depth was the tank?

It may be worth trying again, by pumping the hot lower liquid up to the top of the tank, where it tends to flash. This is especially so if you pump via an inline heater.


Hi Rick

Fast and cool is best.

Fast to not allow enough time for much back reaction, cool to minimise rate of back reaction.

By cool, less than 150C.

The venturi helps achieve this by promoting evaporative cooling / vapour liberation rather than boiling.

But the biggest aid seems to be the neutralization of the catalyst - it seems to give you much more time and allows higher temperatures before back-reaction becomes an issue.

All the best,

Graham

Hi'
Graham,
Yes I tend to go along with you it appears that the optumim recovery would be facilitated by circulating the batch back to the top of a conventional reactor, but not drawing from the bottom of the reactor rather from just above the glycerol level with some way of preventing the settled matter being stirred up by the circulation of a pump.

I am of the impression a purpose built system is the answer along the lines of a one way tube reactor,the chemistry works, the processor needs redesining so the settled glycerol can be removed as quickly as it accumilates. This type of reactor would be far more echonimacal to operate increasing yealds and possibly eliminating the washing process or atleast minamising water usage further reducing polution issues.

I certainly hope it's not just cheap fuel driving the bio diesel research and production as it has been for the petrolium industry otherwise it's a waste of time, many back yard operations are only interested in cheap fuel and realistically acheiving nothing environmentally,Evapourating to atmospuere or dumping methanol laden wash water or stuffed up batches down the sewer is not making cheap fuel , we all bear the cost in the end.
Regards.

quote:

I certainly hope it's not just cheap fuel driving the bio diesel research and production as it has been for the petrolium industry

Hi fleabitten,

I hear that ! I have not figured out why people take the methanol out of the biodiesel. I leave it in and burn it. I have been doing it for years.

It is about time. If you are not in a hurry the soap will still drop out. Water washing is not needed.

I am building a still to get the methanol out of the glycerin.

Thanks, Steve

Hi' Steved,
Do you run 100% Bio diesel all year round?
Or what percentage of bio do you run and what do you blend it with Diesoline,Kerosine,Unleaded
petrol etc?
How do you determine when all the soap has fallen out and the bio is ready to use?
Have You been using it in the same engine for years?
How many miles/km do you travel on bio or blend a year?

I can't figure it out either why all the fuss
extra expence and equipment when all you got to do is make sure your exhaust dont leak to hell with any one else I cary a gas mask incase I get caught in a jam in a free way tunnel I hear the fumes can rot your innards.
2bob.

Hi' Graham,
I had a chat to a friend over the weekend and we both decided to give it another go at stuffing up a batch by extracting the methanol in one go.
Neither of us had any Iso to titrate with so it was decided to do a simple ph test all the way through that's the 50/50 oil and water shake it up and test before it settles WITH PH STRIPS, crude isn't it.

That's of course on oil we dried at 130c under
-100 kpa for an hour on NEW SOY got 1200ml out of 260L.
pH according to three seperate tests was 5.5 near enough so to bring it up to nutral is 2.5 plus 2.5 for the reaction = 5g per litre Made up a small sample batch of 120 ml including methanol 20% settled over night drained it did a 3/27 on it later that evening checked it again perfect.

The batch was sitting at 50c put in total of 50L methanol 25L was mixed in first then while that was being mixed the rest was turned in to methodoxide after mixing and under vacuum for ten minutes the reactor was put under vac the catalist mixer/ recovery vessel was vented and the methodoxide sucked in on to the impeller over another ten minutes, vacunm held low put the insulation blanket over the reactor turned off the mixer and went for a feed.

Two hours later clear as a bell Drew off three samples in three places did the 3/27 on the samples after cooling down perfect PH 7. cranked up the heaters to 58c on full vacuum stired the living daylights out of it at 6000rpm for a couple of minutes, after an hour the sight glass in the end of the condencer was dry no more to be had it seamed and a adissapointing 14L of crystal clear Methanol and beautifully clear bio, tested the Ph again no change and again three 3/27 tests perfect, sight glass in the bottom was showing 35L of Glycerine.

Not at all happy with the recovery, stired the fear of good in to the batch again breifly cranked up the heat again to 80c full vac nothing Whats going on? checked the cooling water temprature checked the vapor trap scrubber water levels all ok checket the thermo well three temp gauges slight veration but thats expected, Drained off 100ml of glycerol at 80c to test for ph,I will get to that later on, set the temp for 95c still under full vac.

Then all of a sudden the mate said there is moisture in the condenser as it hovered near to 90c as it passed 90 the methanol was gusshing out seven condenser tubes were half filled with methanol and it poured for about a minute then just as quickly stopped the heaters were still on now at 93c lifted the insulation blanket off looked in to the top sight glass no more bubbles, checked the methanol level and only another two and a half litres total 16L not good, shut it down vented then sealed the reactor, by now it's getting dark so we decided to let it settle overnight.

Next day we checked it out SURPRISE it didn't settle out it was still at 45c a dark even amber colour, took more samples this time only the ph 7 on all samples, it occured to us later when it was to late we should have been doing a visco on the samples from the start either by timing it through a funnel or a hydrometer.

So what happened? it did not REVERSE and we can prove that or atleast try to neither of us are chemists so if it sounds far fetched or way off take a moment cause we just did it on a good sized batch.

Firstly for the reaction to reverse or partly reverse it has to g0 lower than 7 ph.now even tho our ph test was a bit simplistic it was done at the same tempratures in a cooling vat (My ultrasonic cleaner filled with water at 25c in my air conditioned Garage and the same bottled tap water used as part of the 50/50 test I have several 100 ml x 1 ml measuring flasks and none were used twice, the syringes all used once only and the methanol from the one container.

From working in labs it's essentual the standard you use is duplicated exactly for each sample you test, in this case the PH strips were used as the measure, we repeatedly got the same results, we repeatadly put vinigar in the samples in varing amounts to see if it would work and change the readings it did and the variations were convincing.

SECONDLY,we now have a batch of suspended glycerol that will not settle.

What happened to the Caustic/Lye?
Why is the sample reading 7 ?
Why is it not reading around 8-9? Caustic dose not evapourate it can only Transpire in the conditions that took place.
Why is the recovery of Methanol so low?
How can we recover the batch?

One Litre was drawn off while stiring the batch
from different sample valves five in all at different levels around the vessel.

To this sample after testing the PH again 160 ml of Methanol was added to the sample that was representative of the amount recovered, it was stired up and set aside within half an hour it started to settle out with a clean break a very well defined seperation that settled out to 137ml of dark but thin glycerol.

A 3/27 was done on this sample and let sit whilst I mowed the lawns rang the mate and told him what stage it was at looked at the 3/27 test perfect.

Now getting back to the sample taken at 80c of the glycerol, I was not ecpecting the result and was astounded, Got a glass 500ml beeker poured the 100ml of glycerol in poured in 100ml of water and had a ph strip ready to dip in to the mix I began stiring and instead of it thinning out it turned from an amber colour to a white and as i stured it thickened up, the more i stired the thicker it got until it was as stiff as set jelly and behaved just like jelly with the stirer stood up and wobbled around, I thought Fantastic this stuff could be sold no bad smell no unsightly colour, took a bit and put it in a test tube and more water shock it up the water went white and lumpy so I did a ph on it 7 did it again 7, great
so I thought I would give it a go as soap felt good just like those industrial hand moisturisers only not so greasy
WRONG I should have known better Soap is around 8 ph THIS STUFF WOULD NOT WASH OFF with water it diluted some what but stuck like s*** to a Blanket I had some glycerol soap in a bucket that did the trick.

Now had the batch reversed at this stage There would have been some sign of oil at least I would have expected, it can't have formed any soap hear (more on that later on) because the oil was dry and kept dry, you need water oil and caustic to form soap or do you? There was definatly no soap in the Glycerol sample nor did it form any after water was added, there was no sign of soap in the 3/27 tests a day later still crystal clear the bio tested nutral so what's happened to the caustic if it aint in the bio or the glycerol?

I am going to bed now it's 330am and a long day ahead I will get back to it when I get some time but think about it there is more to this than you think.

quote:

Originally posted by fleabitten:
That's of course on oil we dried at 130c under
-100 kpa for an hour on NEW SOY got 1200ml out of 260L.

My conversion calculator tells me that -100kpa is equivalent to about 29.5 inches of mercury. A "perfect vacuum" is equal to about 29.92 inches of mercury.

Assuming I got the above conversion correct, can you discuss how you achieve and maintain a vacuum of -100kpa for an hour or so?

I have a sliding vane vacuum pump that develops a maximum of about 18 to 20 inches of mercury. I would like to do better. Any suggestions, tips or insights?

Hi'Producer,
You and I both know there are factors that influence readings nonemore than on vacuum,
This test was done on a reactor out in the open on days where ambiant tempratures reached 38c.
and the readings taken on two identical guages fitted to the rig, This is NOT A GLASS WARE EXPERIMENT in a lab the gauges are checked against each other,What more do you want?

In Reality it's the same guides used in industrial processing on vessels, operators and technical staff rely on.

Perhaps next time I will get a Berometer and amercury filled and calibrated U tube With Compensation tables, I know exactly if you can trust a Altimiter and the lands Department we are 110m above sea Level, High or low tide time of year was not mentioned if you want to go that far for home brewing.

This was a spur of the moment thing I thought that may have been obvious, the testing was done in controlled tempratures but more for our own comfort and not every back yard operator has an airconditioned Garage so our testing processes were not entirely a waste of time.

The variations in Vacuum will become Much clearer later on,We could have done this at Atmospheric pressures but we both have limited time and it would certainly NOT have given an insight in to what happens or can happen even at atmospheric pressures, you would not drive your Reactor that far, there would be no logical reason to.

Note Efficency Economy except for time was not the point Three interrupted days was all we had

I have Done this before with much better recovery rates on diferent feed stock in the Garage Some years ago, I have not made any bio for near on four years So the rig was moved outside under cover where most home brewers would set up so in one way it's more in tune with most homebrewing operators and probably
more likly to experience the same results we got but atleast they will have an idea of why.

I'm off to work lunch is over.

Back Again,
The storms around Sydney last night were awsom
Nature reminding us of the power she can muster up.
Now the batch was a mess,and testing neutral as I stated before we are not Chemists, we had the impulse to add more caustic to the recovered Methanol and stir it in, but how much? as the last sample after putting a representative amount of Methanol that was recovered settled out but the Glycerol didn't look right, as I stated before in hind sight may be we should have done a SG or visco on the glycerol, I have one of those digital heated water circulators used in labs for visco testing I set it up in the 5L ultrasonic cleaner tank and works well,

We decided on 1/4 gpl to start with and if it needed more mix in more in 1/4gpl in 1L of methanol until it was right.

Mixed in the catalist under -20kpa the batch was held at 45c since the last test and sealed so no water could get in Gave it a good minute at 6000 rpm turned every thing off except the heaters, left it on low vac and sealed off No chance of getting water in the batch hear I mentioned this again for a reason.

Gave the batch a couple of hours to settle out
tho it looked fine after an hour.
Vented the reactor through the vapor trap recovery vessel and condenser, norrmal practice
it minamises atmospheric moisture getting in to the reactor, Spray painters know how quickly water gets in to their air receivers,Venting vacuum is doing the same thing.

The testing was a toss up between 7.5 or 8.PH
the 3/27 was good and is still good four days later.
the Glycerol looked much thicker so it was drained it off still at 45c, 40L but that will probably drop to around 38-39 l when it cools down the drums were sealed and set aside,I was interupted by a callout so later that evening
I started to distill of the rest of the batch
drained off another half a L of glycerine before I started.

It was still light when I started at 7.30 pm and still very hot around 28c turned the heaters up to 60c turned on the vac pump on and the pressure started dropping it reached -50 kpa and it will stay there as the heat comes up
and distillation starts, it can start at 52c but mostly around 55c without stiring, the methanol bubbles comming off the heaters are enough to keep the batch circulating.

Off track Just a little.

The secret hear is to have a vac pump that can maintain a steady negative pressure within the vessel as the Methanol boils off, -15 kpa is the lowest you would even concider but it's still faster than an Identical reactor set up for Atmospheric Distilation in the exact location at the same time run side by side doing the same job, that has to be made Perfectly clear as Atmospheric conditions tempratures and altitude effect both
processes equally at the start of the distilation process when both are first turned on and that's where it also ends at the start switch.

Now I am no expert on vacuum pumps and read what's on the gauges both gauges are linier and read exactly the same over their entire range
what they read is good enough for me and most operators.

Now back to the batch, Within 20 minutes the vac was reading -75 and going up, that happens when recovery stops, the condencer tubes were almost dry the heaters cut out, Recovery was just shy of eight L,thats not good I was expecting between ten and thirteen L from the bio alone, This is the First time I have ever used new refined 100% SOY, Previously it was mostly used canola, sunflower tallow (SOLID)
blended together and occasionaly got a bit of Hydroginated used soy and Peanut oil but only small ammounts of 20L.

Now with the Vac slowly climbing still at 60c
I set the temp for 90c and was prepared to drive it to 140c just to see what would happen
the vac started to drop slowly as the heat came up. There is a 100mm sight glass near the top and another on top 50mm, By the way the batch was slightly yellow but crystal clear as I shon a toarch through the top glass.

As the heat just passed 80c distilation started again this was roughly 15 minutes later, this is interesting because it came with a gush out of the tubes then just stopped dead I checked for any bubbles in the sight glass shining the toarch in from the top sight glass no bubbles but something was happening as I wtched, the heaters cut out it was at 90c and the vac was holding at -50,there was a web forming near the top of the vessel that was rather rappidly thickning up to stringy mass that was spreading throughout the batch right down to the bottom the rest remained just as clear as when this started,I have seen this before but never noticed it actually start, Ain't sight glasses good, now what is it? and why did it happen?

Bear in mind the 3/27 showed no contamination at all , before I started and after the first distilation and are still clear,

We have a Professor in one of our labs who's sole purpose within the company is to analise oil, Vegetable oils, mineral oils and synthetic oils, polimers and esters either to be blended with or made from oils,He also knows much about Bio Diesel but insists it's a waste of time, it's not sustainable, is no answer to any thing
long or short term it can only creat more poverty and hardship in countries that can least afford it and that's exactly what is happening right now, and won't talkabout it, he rides a well oiled push bike.

I have this guy by the short and curlies to a point,I build his brain storming contraptions
and the odd bit of gear for his own private research.

I asked him what this stuf was he said first thing is first look up partial pressures of Methanol so ok,it's all about boiling points at ambiant pressures or pressures in vessels, it Dosen't tell me what the stringy stuff is, this guy is weard so I asked him what relevance it has to the stringy stuff.A
Any how he eventually said in the absence of water the bio made from soy in this instance under the right temprature and pressure, it happens at atmospheric pressure as well dependant on the type of oil once reacted will drop of some of the methanol and form Soap completly in the absence of water but what got me in was it happens or starts to happen when the free methanol has nearly been completly drawn off, a continuance of the distilation I assume

A normal conversation is impossible with this guy and he was talking down to me I asked him about peanut oil he replied it's possible and walked off.

So I hope I got it right, It makes sence since with this batch I was more than cautious with water getting in and explains my earlier experiences with these webs, they wash out ok
but have sometimes reapeared after drying and I did notice it happend with the adition of soy and Peanut oil some years ago when I was stuffing around.

It dosen't answer what happened to the caustic and why the batch gave a nutral reading.

Just a correction to the previous reply to Producer, I said I was 110 m above sea level that should have been 11.0 or 11 m above sea level, Hope that didn't stuff up any ones calculations if they tried to figure out what was happening, I apologise.

Fleabitten.

Hi Fleabitten

The stringy stuff is soap.

As for the pH strips, they are too imprecise to tell you much - the margin for error is large, the best approach is to titrate, you'll get much more accurate results that way.

Regards

Graham

Hi' Graham,
Sure most would go along with your comment on PH strips, but in this experiment given the number of tests and the consistancy of the results with the combined results of the 3/27 they seem to have been fair enough.

Unless of course there is some masking happening the 3/27 is not showing up, It's Four and a half days now and there is no change visible in either of the last two tests I did.

Ph strips are a basic standard and widly used as a guide out in the field,you can easily check the readings of your sample against another known value either side of neutral or your reading so long as they are not at the extreme ends.
Strips are predominatly accurate unless its an operator error but that can happen with titrations.

Ph Strips if used in the same way every time are ok on the other hand it's difficult even in a titration to acheive a 1/4 reading using a dropper, In labs the environment is strictly controlled ambient tempratures influence accuracy as it dose with the ph strips.
The strips are avaliable in varing degrees of accuracy and temprature ranges are printed on the pack, if those simple instructions are followed you should be near enough ,otherwise whats the point, they would be as usefull as a glass door on a Lavatory.

Also there are differing opinions on which titration method should be used, no doubt formulated in a lab some where and more than likley giving different results in the home labs in differing locations, so what would you check those against?
Surley not a Ph strip.

Regards.

Sorry, I'm a little confused - too many ideas all at once!

Can you summarise, in 1 short sentence, your basic query regarding the process?

Then I'll try and focus on that point alone.

Thanks

Graham

Have you tried neutralising the hydroxide ions left in your processor before distilling? It sems to me that without the catalyst to attack the methonolester bond the reaction could not reverse.

I have been using 3.5 grams of NaOH per liter of oil plus the amount after titration for years. When I found this site and read that you guys use 5 grams plus the titration amount, it was a surprise. Could one of you tell me the difference? Is the lye you buy less pure, so you use more or what?
I'm going to make a small test batch but in the mean time I'd like to know your experience.

p.s. what an eye opener Graham Laming's web site is. It's got us designing a new reactor and trying out those creative ideas.