Written by Brad Wilder

The 80/20 method as it is called is simply the practice of breaking the biodiesel reaction process into two, and sometimes three (77/22/1), parts. Specifically, this is done to reduce soap counts, counteract higher water content, and increase yields from high water oils. Testing during each stage of the process can also be used to help troubleshoot recipe quantities and bad batches.

Basic Recipe

I generally use KOH and a basic recipe as follows:

• 7 + Acid # grams KOH per kg oil (adjusted for purity)
• 180g Methanol per kg oil
• 1.5 hours of mixing at 55C

Once mixed I simply add 80% of the methoxide mass to the oil as the first reaction and then add the remainder for the second reaction. One can multiply the total mass of each chemical by the appropriate amount if the two stages are created at different times. I generally just use a big mixing tank and seal it while I wait for the 80% stage to finish.

Advantages

The main advantage of the 80/20 method is a reduced final soap count. This is done primarily due to the bulk of the soaps being created in the first (80%) reaction. Once the glycerin has been removed the intermediate oil/biodiesel mixture will have a titration of essentially zero (dependant upon initial water content), and will therefore create much less soaps in the second (20%) stage. The three main reactions in a biodiesel reactor are shown below.

1. FFA + KOH = Soap + water                                                (BAD)
2. Water + Glyceride = FFA + Glyceride or Glycerin                    (BAD)
3. Methanol + Glyceride = Biodiesel + Glyceride or Glycerin          (GOOD)

During the biodiesel reaction all three of these will occur. Also of note is that reaction 1 is very fast when compared to reaction 2 and 3. FFA and KOH will react almost instantaneously, while at process temperatures water and glycerides (reaction 2) will react much slower than reaction three. What this means is that during the first (80%) reaction almost all of the FFA will react with KOH. The glycerin created will also absorb much of the water either already present in the oil or created by reaction one. All subsequent reactions will thus have much lower FFA and water content than the first reaction and should logically have lower soaps.

Method as a Quality Control Test

When I perform the 80/20 method I will take a soap count after the first stage. The number I get will vary depending on acid number and water content. Then I will take a soap count again after the second reaction. If the second soap count is higher than the first I have used too much KOH, I will calculate the mass of soaps in the oil and convert this to mass KOH, I will then lower the base amount of KOH to fit 80% of this for subsequent batches. If it is equal I have used ‘slightly’ too much KOH and will lower the amount by the same as above but for ½ the soap content. This procedure hasn’t been optimized but it works for me. Also note that this only works for dry to ‘moderately’ wet oil. Oil that is between 0.1 and 0.6% water can be fully reacted this way. However, the final soap counts are going to be higher with higher water. Using this method the final soap count of the oil has more to do with the water content than the initial acid #.

Note that this isn’t a substitute for a GC, SAFTEST, or even 3/27 test. This is simply a double check that can be used to optimize your own homebrew recipes. Most home brewers don’t have access to a GC or SAFTEST machine. Test batches in bottles or blenders may also not react exactly as full scale batches. I have used this to fine tune KOH amounts to minimize soap counts below 2000ppm after the second reaction no matter what the initial titration was. Yields obviously still vary though. Using exceptionally dry oil (under 1000ppm or 0.1%) final soap counts in the range of 600ppm have been possible.

Examples and Explanations:

Batch 1: Typical results of a passed batch, note that this batch had very good initial water. Soaps drop from first to second reaction

Batch 2: Higher acid #, higher water batch. Note that even with higher titrating oils final soap counts are comparable with first batch. Second stage soaps again lower than first stage

Batch 3: Higher acid # oil that required a reprocess, second stage oil still lower than first stage signifying not enough catalyst was used. Reprocess was successful and not too much catalyst used as displayed by soap counts

Batch 4: Too much catalyst was used this batch. Soap counts between stages did not drop. Could have completed reaction with less KOH.

Batch 5: Very high water batch, note that even with the high water final soap counts are still relatively low compared to what would have been obtained from a single pass reaction. Also note increase in acid # across stages due to water breaking glycerides into FFA.

Batch 6: Special batch. Some glycerine was accidentally left in the oil during the soap test. Therefore, soap test showed erroneously high readings since glycerine holds the majority of the soaps created.