Troubleshooting – How the Lab Can Help

Reviewing the basics of fermentation from a lab analyst’s point of view
By Sabrina Trupia | July 20, 2011

While today the trend is to get online instrumentation for fast problem solving on issues that occur at an ethanol plant, in the majority of ethanol plants it is still up to the laboratory to correctly quantify the problem and identify the issues, especially when it comes to fermentation. The lab is often the first line of defense when a problem is suspected, as it is the laboratory’s task to correctly identify what really is going on in the fermentor. Naturally, the lab plays a role in quality control for every aspect of the process (front end and back end included), but its most crucial task is to control the quality of the fermentation.

Lab Analyst, Troubleshoot Thyself

Correctly identifying that there is, in fact, real trouble is the first step in troubleshooting. Laboratory analysts should be able to troubleshoot themselves before deciding whether there is a production issue. This is possible because the lab keeps a log of previous fermentation data, and can decide whether things are within the normal range. The two most important issues that may lead to an incorrect diagnosis—trouble when there is no trouble or perhaps worse, no trouble when there is trouble—are issues in the quality control process such as calibration, sample preparation or measurement errors, or communications with plant operators that may lead to sampling errors or other issues.

So, for the first step of troubleshooting, two things are essential: that the lab communicate effectively with the plant and that the lab follow appropriate quality control guidelines. That way the lab is in a good position to provide valuable information for plant troubleshooting. To communicate effectively with plant operators, the lab analyst needs to understand the process. For quality control guidelines, a couple of basic things need to be kept in mind. First, it is important to calibrate the high performance liquid chromatography (HPLC) and gas chromatography (GC) instruments on a curve using at least three points. Second, the lab analyst needs to run “check” standards every five to 10 samples, and before and after a questionable data point. While it may seem these steps are not needed in an ethanol-plant laboratory or that they may be time consuming, they do save time and resources in the long run. (Incidentally, for both these points, the National Corn to Ethanol Research Center has a proven record of providing effective hands-on laboratory training on best ethanol-lab practices.)

Once the trouble in fermentation has been spotted unequivocally, in a corn- or starch-based fermentation there are generally three areas of troubleshooting—the enzymes, the yeast and the nutrients.

The Enzyme Alpha Amylase

We are working under the assumption that the correct dosage of all additions are known by the plant operators, but that either some amount of alpha did not get to the slurry tank or that liquefaction targets are not being reached for whatever reason. If there are questions whether the dosage of alpha used routinely is correct, the ethanol plant needs to contact the enzyme manufacturer to learn more about optimization of dosage. We are omitting from this discussion of troubleshooting of other, more specialized enzymes such as beta glucanase.

Alpha amylase is used to break down the starch and reduce viscosity. In general, issues with alpha amylase are solved relatively easily. They just require monitoring and adjusting the dose to improve liquefaction. If identified in time, alpha dosing issues usually do not stall a fermentation. The thing that needs to be checked when troubleshooting this part of the process is the dextrose equivalent (DE) levels.

Dextrose equivalent is a measure of the total amount of reducing sugars in a solution, calculated as dextrose and expressed as a percentage of the total solids in a solution. DE measurements are therefore affected by the total solids in solution, so those have to be accounted for. In liquefaction, starch is converted to soluble dextrins with chains of 1 to 50 dextrose units. In the liquefied mixture, only the terminal dextrose unit has reducing capacities, so the reducing capacity of a solution is a measure of the extent of the starch conversion to glucose. The higher the DE, the shorter the average chain length, and the more efficient the liquefaction.

In the case of ethanol plants using a jet cooker (hydro heater), usually dosing of alpha will occur both in the slurry tank (generally one-third of the total dose) and in the liquefaction tank (the remaining two-thirds of the dose, applied after jet cooking). At NCERC, the target DE levels in the liquefaction tank, at a total solids level of about 32 percent, are 11 to 13, while in the slurry tank they are lower, between 5 and 7, because there is less alpha to convert the starch into dextrins. While DE is the most common measurement performed on slurry and mash, there are other methods (HPLC, starch analyzers, etc.) that can give a picture of the state of the liquefaction. These other analyses will also have normal ranges that depend on the instrument’s specific calibrations. If there should be an anomaly in the alpha dosage, however, it would be clear from the lab results, no matter what the method of detection is.

There are several things to check if the DE readings/liquefaction levels are outside the acceptable range, including

• Has the temperature changed?
• Has the pH changed?
• Has the concentration of solids increased or decreased?
• Is the particle size of the corn different?

The above are all parameters affecting the liquefaction and dosage of alpha amylase, with liquefaction temperature and pH being the most crucial ones. If none of the above questions has been answered with a “yes,” it is possible that the batch of alpha being used has lost activity (troubleshooting tip: check enzyme storage tank temperature), or that there is an issue with dosing (troubleshooting tip: is the pump working, is it set at a correct rate, are there blockages in the delivery system?). If loss of enzyme activity is suspected, there are commercially available test kits to test the alpha amylase in the lab, although most require the use of an ultraviolet spectrophotometer. It is possible that the enzyme manufacturer can provide either the testing or a small batch for comparative purposes.

Glucoamylase: Troubleshooting Saccharification

Glucoamylase will convert dextrins to simple sugars, making them available to the yeast that will ferment them, producing ethanol. Problems with sugar can be caused by having either too little or too much sugar during fermentation. In either case, we talk about a “stuck” fermentation. In the first scenario, the glucoamylase level is too low, so that there isn’t enough sugar for the yeast to metabolize and the yeast will not produce ethanol because there isn’t enough “food” around. In the second scenario, too much sugar is liberated by the glucoamylase and it will overwhelm the yeast, stressing it and reducing the ethanol production.

What does the lab analyst see when glucoamylase is incorrectly dosed? In the chromatogram (HPLC), low-glucoamylase dosing will result in a higher concentration of oligosaccharides (especially DP4+) persisting late in the course of the fermentation instead of being broken down into glucose, resulting in lowered ethanol production. In the second scenario, we would notice an abnormally high glucose concentration early in the fermentation as well as a plateauing out of the ethanol concentration. In both low- and high-dosing scenarios then, ethanol production would be stalled, but for different reasons.

When troubleshooting a fermentation with high sugars/low ethanol, it is sometimes possible to re-start a fermentation that is “stuck” because of low glucoamylase by adding more enzyme to the fermentor. If there is too much sugar and the yeast gets overwhelmed, it is also possible that waiting it out might help the fermentation. Another, more common way of handling too much sugar in the fermentation broth would be to add fresh yeast in appropriate concentrations (usually higher than the initial dose) to consume it.

Yeast Happiness

The availability of adequate amounts of fermentable sugars is only one aspect of troubleshooting a “stuck” fermentation. Other signs can let us know if the yeast is thriving or not. Yeast “happiness” is also a function of nutrient availability, lack of contamination, adequate temperature and pH control.

Most of the parameters that need controlling during a fermentation for optimum yeast performance can be checked by looking at chromatograms (HPLC data) during the fermentation. There are two major things to watch in fermentation data: signs of nutrient limitation and contamination.

In monitoring nutrient limitation, look at the total sugars concentration over time (HPLC data). If the concentration of sugars does not decrease and we are sure there is sufficient glucoamylase in the system, then the yeast may be nutrient limited. Another sign is given by the levels of glycerol, which increase during times of yeast stress. If it can be determined that it is not the glucoamylase that is too low, adding urea or other yeast nutrients could help the yeast, if it is early in the fermentation. Re-inoculating is another solution, as the yeast that is stressed cannot easily recover. If this is a persistent problem, it might be worth analyzing the corn, the water/backset used in the preparation of the slurry, and also the fermentation broth for levels of nutrients or perhaps presence of compounds that could inhibit nutrient availability to the yeast.

Bacterial contamination is indicated when the levels of lactic and acetic acid are high (HPLC data), the ethanol level is low, as bacteria compete with yeast for the sugar substrate, and glycerol is high. Normally, in a clean fermentation the lactic acid should be below 0.1 percent (by weight), and the acetic acid below 0.05 percent. If they are not, the fermentation is contaminated. Usually, in addition to the high organic acids levels, glycerol levels are also higher than normal, indicating a stressful environment for the yeast. Bacterial contamination can also be indicated by pH measurements below 4 and by looking at the broth under a microscope. The yeast viability and cell counts will likely decrease under contamination conditions. Lactic acid bacteria can be detected under the microscope in an oil immersion—they are rod-shaped and much smaller than yeast, but still visible.

What can be done when there is bacterial contamination? The only way to fight a bacterial contamination is using more antimicrobial agents. After introducing more antimicrobial, it is possible that the yeast may still be stressed and not able to come back to produce, so more inoculum—more yeast—may be needed. Now, if the infection has been going for a long time, the bacteria will have consumed a measurable amount of the sugar that would otherwise be converted into ethanol. The ethanol yield will be reduced even if the fermentor can be recovered. That is why preventive introduction of the right amounts of antimicrobial agents is preferred, as well as thorough sanitation procedures to keep a clean system.

The fermentation troubleshooting issues discussed here are the primary ones that can present themselves in any ethanol plant. Each ethanol plant is different, however, with its own particular set of enzymes, nutrients, type of yeast and mode of operation. In general, if none of these items in troubleshooting seems to help, or if specific issues arise, it is helpful to contact the experts at the enzyme or yeast companies or the provider of antimicrobials. They usually can help. Because it is part of our mission, the NCERC is also always available for questions and to help with specific troubleshooting.

Author:  Sabrina Trupia, Ph.D.
Asstistant Director of Biological Research,
National Corn to Ethanol Research Center
(618) 659-6737