Biodiesel

Biodiesel Manufacture – the Imisides Method

Introduction:

This method describes a foolproof method of making biodiesel that doesn’t require any heating or titrations. It is the same as published on the infopop forum a few years ago, with one modification. This is a more extensive explanation of the method, as the infopop forum placed a restriction on the length of the post so I had to restrict it somewhat.

The advantages of this method over any method that requires heating are obvious – you can make as big a batch as your equipment and shed size allows.

It is divided into two sections. The first explains the chemistry and the second is a step-by-step guide to the method. You don’t have to understand the chemistry to follow the method.

1.0 Chemistry:

A triglyceride molecule may be converted into one glycerol molecule and three methyl esters by base-catalysed transesterification. A methoxide ion attacks the ester linkage of the triglyceride (via an SN2 mechanism) and cleaves the molecule into these four segments.

The methoxide is generated by reaction of NaOH or KOH with MeOH as follows:

Equation 1:     KOH + MeOH  = K+MeO + H2O

One thing that may not be obvious from this reaction (due to the formatting limitations of WordPress) is that it is a dynamic equilibrium.

That is, if we consider the reaction of caustic soda with hydrochloric acid we have

Equation 2:     NaOH + HCl = NaCl + H2O

In this reaction, all the NaOH and HCl have been consumed and have all been converted to salt and water.

For the methoxide reaction above, however, the reaction is proceeding in both directions simultaneously. That is, if we took a snapshot of the solution at any point in time we would find significant concentrations of KOH, MeOH, methoxide and water.

The problem with the conventional method (the so-called Dr Pepper method) is that both the methoxide and water can act as reactive intermediates – the methoxide for transesterification (biodiesel manufacture) and water for saponification (soap manufacture) respectively.

Fortunately, however, the transesterification process is more facile than the saponification reaction, so if the concentration of the base (KOH or NaOH) is carefully controlled, it is possible to facilitate the transesterification reaction without allowing the saponification reaction to proceed.

This means that a titration is required to determine the concentration of free fatty acids, as the base that is added will react with (neutralise) them first. So an exact amount has to be added, to neutralise the FFAs, and then leave enough for generation of sufficient methoxide to enable the transesterification without enabling the saponification.

This does, however, require that the solution be heated (to 55 deg C) to overcome the activation energy for the transesterification process.

When I looked at this process, I saw a modification that was so obvious that I’m frankly astonished that no one thought of it before I did.

Before I describe it, let’s first consider the role of the methoxide:

Image result for transesterification

If you look at this figure you will note that the base (NaOH in this case) does not appear as a product or reactant. That is, the reactants are triglyceride and methanol and the products are methyl esters and glycerin.

The base is written on top of the arrow which is, by convention, where catalysts appear in chemical equations. This is telling us that it participates in the reaction as an intermediate, but is not consumed.

This is the definition of a catalyst:

  1. It lowers the activation energy of a reaction
  2. It is not consumed. That is, it participates as an intermediate but is regenerated afterwards.

In other words, the fact that the NaOH doesn’t appear as a reactant tells us that the reaction will, in theory, proceed without it. In other words, it is telling us that the Free Energy change of the process is negative – in other words, it is a downhill process.

But the fact that it is thermodynamically favourable doesn’t mean much in real terms, as the kinetics of the process must also be taken into account. In other words, before it can proceed, the activation energy must be overcome. That is, enough energy must be supplied to the reactants to allow the energy hump to be overcome, and the reaction to proceed downhill. If this is not supplied, no matter how long you mix the reactants, the reaction will not proceed until all your methanol evaporates.

So, we either supply energy to the system, or add a catalyst to lower the activation energy (or both).

The Dr Pepper method is constrained in the amount of catalyst it can add because of the danger of promoting saponification. So the mixture needs to be heated (to 55 deg C).

When I looked at this method there was an obvious modification to the method that would kill two birds with one stone. This modification was so obvious that I am still scratching my head as to why no one thought of it before me.

Put simply, we dry the methoxide solution. This kills two birds with one stone. Firstly, it removes water from the equation (literally) thereby eliminating the interfering saponification reaction. Secondly, it pulls the equilibrium to the right (Le Chateleir’s principle), thus generating a far higher concentration of catalyst.

So with a higher concentration of catalyst, we lower the activation energy, allowing the reaction to proceed at room temperature, and we don’t need to worry about the saponification process.

The chemical of choice to dry any alcohol is quicklime (calcium oxide):

Equation 3:    CaO + H2O = Ca(OH)2

If we now add Equation 1 and 3 together we get:

Equation 4     KOH + MeOH + CaO = K+MeO + Ca(OH)2

When I first developed this method I was living in Melbourne, and had no trouble buying quicklime in bags from the cement factory in Lilydale. When I moved to Perth, however, I found it was not quite so easy to get – and this mirrors feedback I have received from people all over the world.

So I had to find an alternative drying agent, and it’s cement – just the cheap, $6 per bag General Purpose Grey Cement from the hardware store. Note – it’s the cement powder, not the premix concrete containing sand and aggregate.

So then all you do is add the cement to the methoxide solution, allow it to settle, and then it’s right to use. Note – the methoxide solution is unstable as its highly alkaline nature means that it will suck CO2 out of the air at a rapid rate and convert it to CaCO3. So get yourself organised so that you make your biodiesel batch two days after making the methoxide solution (two days is how long it takes for the fines to settle).

2.0 Manufacturing Process:

This describes how to manufacture a 1000L batch of biodiesel. You do not have to have read or understood the chemistry of the process. The process can be scaled up or down proportionately as required

2.1: Ingredients:

870L Waste vegetable oil
130L Methanol
13kg KOH (potassium hydroxide or caustic potash)
20kg General Purpose Grey Cement

2.2. Equipment

1 x IBC (1000L)

5 x 200L open-top settling tanks. Each to have a drain valve at the lowest point, and one about 6 inches above it.

2.3 Manufacture of raw product

  1. Set up an open top 200L plastic drum with a valve about 5 inches from the bottom. Position the drum in a well-ventilated area with no naked flames anywhere nearby, preferably near the open door of a garage. Put on long sleeves, gloves, and a repirator (although not strictly necessary if the area is well-ventilated).
  2. Add 130L MeOH to the drum. Add 13kg KOH to the methanol – it doesn’t have to be stirred while you do it as it will not clump. Begin mixing (a shovel will do). The MeOH will get hot and may even boil – if it does step back and wait for the effervescence to die down. The heat helps the KOH dissolve and it will dissolve in a couple of minutes (but there may still be some undissolved impurities in the bottom).
  3. Add the cement. Stir with the shovel for about 5 minutes, then leave it to settle for two days. Put a lid on the drum to minimise CO2 exposure and evaporation.
  4. Add 870L WVO to IBC. Make sure there is no free-phase water present.
  5. Add methoxide solution to the oil. I just do it manually by draining it into plastic water cans and tipping it in. You could use a small pump, but it’s not going to be easy to find a pump that will last long pumping highly alkaline methanol. After this step the methoxide solution will be sitting in a lightly coloured free phase on top of the oil.
  6. Open the bottom valve, turn on the pool pump, and pump the mixture back up into the top of the IBC. I used 2 inch hose and fittings. Hold the hose out of the mixture at the top of the IBC so it cascades out of the house and into the free phase methoxide solution at the top of the IBC. This provides the shear required to initiate the reaction.
  7. Within a couple of minutes, the free phase will disappear, and the liquid cascading out of the hose will begin to change colour from light murky brown to a dark chocolate brown. About this time you will smell the characteristic smell of the glycerol, and hear the pump change pitch as the viscosity of the solution changes.
  8. 8. Once the reaction has commenced, insert the hose into the IBC, and tie it in place somehow, or put a weight in top of it so that it won’t come out. The temp of the mixture will increase by about 6 degrees over a period of 15 to 20 min.
  9. Allow the reaction to proceed a for an hour and then pump into 5 x 200L settling tanks

3.0 Polishing of Raw Product

3.1: Clarification:

The raw product will settle into two layers – the lighter brown biodiesel on top and the heavier, thick black syrup that settles out at the bottom. In principle, you can then drain the black syrup from the bottom valve until biodiesel begins to flow and then tap the biodiesel from the upper valve to go into your car.

But before you can do that, the upper layer must be clarified. Immediately after manufacture, residual amounts of methanol remain in the biodiesel, and this retains residual amounts of glycerine in solution. As methanol is volatile, it will eventually evaporate, thus precipitating out the last of the glycerol.

The best way to do this is to get an aquarium bubbler, attach a tube to it with a weight, and sink it into your clarifier drum. I have found that after about two weeks of bubbling the residual methanol smell disappears, Leave it to bubble for a further two weeks, then allow it to settle for a couple of days and it should be right to go.

3.2: Quality Control

Get your hands on some beakers from somewhere – I use 500mL tall form beakers. At a pinch, glasses from the kitchen may be used, but the only proviso is that the bottom of the vessel is optically transparent (which is why beakers are good).

Fill the beaker from the upper valve and leave it overnight. The next day, tip the contents out and inspect the bottom of the beaker (hold it up and look through it). If there is any residue evident (other than a thin film of biodiesel) it’s not right to go and further clarification is required. In other words, if you can’t see straight through the bottom of the beaker – if there is any refraction of the light at all – it is not right to go and further clarification is required.

 

 

7 thoughts on “Biodiesel

    • Intermediate Bulk Container – it’s the standard industrial container for holding 1000L. You can normally get them on Gumtree for around $50

  1. Hi Mark, when i tried your method some years ago I started with oil that had been glycerol pre treated, leaving the oil at about 350ppm water. As this is predominantly an anhydrous process I ended up with a yield approaching 104%

    The glycerol produced was only about 2/3 of the methoxide volume originally added. When I water washed the crude product it was evident that the amount of potassium soaps formed was minimal, the wash water after the 1st wash was virtually as clear as the last (4th or 5th) wash on a normally produced batch.

    I did have a problem however with some calcium soap formed but I think this was my own mistake as some fines (unreacted CaO) found their way into the mix when I was decanting the cubie i had used for mixing.

    • I’m not sure what you mean by “glycerol pretreatment” but water is the enemy of this process. But if I understand your comment, you are saying that the method still worked at 350ppm water. I’ve never tested the water content of the oil that I use – the only stipulation is that there is no free-phase water in the reaction mix. Also, remember the drying process involves conversion of the CaO to Ca(OH)2, and this could certainly contribute to soap formation in the presence of water.

  2. Hey,

    if the cement is added to absorb water, would it not react and set in the bottom of the drum? if so this would be a difficult waste product to get rid of? I guess you could just put in the rubbish bin, but it would be bloody messy? I found your read quite interesting and it certainly has me thinking, but in basic cost terms you are trading a couple of hours of an electric heater for a $6 – $8 bag of cement, and you now have 2 messy byproducts to get rid of. I think the biggest advantage though is the saving in methanol, compared with the usual method which would have required 175 L. How much bio will this reaction provide? what is the approximate glycerine / bio ratio? I am a home producer and have been so for about 12 years now, I have a pretty streamlined process. I am always interested in improvements though. I look forward to your reply.

    • The cement doesn’t set hard – kind of semirigid. The reason is that it hasn’t had a chance to dry as it has the methanol sitting on top of it. I remove it by taking the drum outside, tipping it upside down and tapping it, and it all comes away and can be shovelled up. But yes – it is something that you have to dispose of which the other method doesn’t have. The advantage of this method for me is the ability to easily scale up. If you tried to heat 1000L in an IBC it would be a little impractical. In terms of the ratios, I’ve never actually measured it, but I finish up with a little more than 130L – maybe 170-180L of glycerol I think

Leave a Reply

Your email address will not be published. Required fields are marked *