How Not To Test A Sanitizer

Those of you who “know” me on other forums have probably heard this rant before, but they’ve done it again, so here I go again.

Brewers, for obvious reasons, are very concerned about the quality and efficacy of the agents they use for sanitizing. A failed sanitation cycle can lead to a lost beer – which costs home brewers a bit of money and a lot of time, and which costs commercial brewers buckets of both. So there is a natural inclination among both types of brewers to test their sanitizers. Unfortunately, many of these tests are not designed around the intended use of the products. In many cases this results in the conclusion that the sanitizer doesn’t work…sometimes leading to dangerous “suggestions” to overcome the sanitizers perceived lack of efficacy.

Anyways, this weeks rant was motivated by the Master Brewers Podcast Episode 96, where some brewers from a brewery that should know better did it all wrong…and recommend something that is potentially illegal.

Sanitizers – Some Background

Most countries legally restrict the term “sanitizer” to products which will reduce the concentration of a contaminating microorganism by 10^-5 to 10^-6. I.E. to one-hundred thousandth to one-millionth of the original inoculum. Agents which do not reach this threshold cannot, in most countries, be labelled as a sanitizer. Agents which exceed this threshold are, in most countries, classified as disinfectants.

Sanitizers are then broken down into a number of different categories, based on their intended use and likelihood of human ingestion. Of those categories, food-safe hard-surface sanitizers are the sole category of sanitizers used in the brewing (and food) industry. No other class of sanitizer is approved for food-contact purposes, in order to ensure that people are not poisoned by their use during food manufacture.

There are two elements to that definition that are relevant here:

1. Food safe: To be food safe, a sanitizing agent needs to either dissipate before food contact, or be made of generally non-toxic components that are present at a minimum effective concentration. An example of the first group would be alcohol sanitizers such as 70% ethanol or 70% isopropyl. The later is quite toxic to humans, but because it evaporates quickly, can be used for food preparation purposes – with the caveat that it be allowed to dry completely before food contacts the treated surface. The second category is the sanitizers most homebrewers are familiar with – star san, iodophore, peracetic acid, etc. These are generally non-toxic to humans at the dose used, and when used as intended only small amount are present on the surface when food contacts it.
2. Hard surface: This is the part that most tests screw up. All of the sanitizers approved for food-contact use are only effective on hard surfaces – and in most cases, clean hard surfaces. Because these agents are formulated to have the minimum effective dose of the sanitizing agent, a porous surface or soiled surface will absorb or react with the sanitizer, resulting in poor or no sanitation.

In other words, the sanitizers approved for use in food production (which includes beer) are strictly for use on hard surfaces that are already clean. Use on porous surfaces, or soiled surfaces, or trying to sanitize liquids, is likely to fail. Not because the sanitizers are crap, but rather because we want them to fail on those surfaces. Because a porous, soiled surface with lots of liquid around is a pretty accurate description of your mouth…and of foods like meat and milk – other food preparation industries which often use the same or similar sanitizers.

Enter The Stupid Test

I bet at this point you already know where I’m going with this. In the aforementioned podcast, brewers at a very successful and very well known brewery “tested” iodophore and ethanol’s ability to sanitize Brettanomyces and Lactobacillus. But in a way which ignored how these sanitizers are formulated and intended to work. Even worse, they then did something which would render the iodophore no longer safe for food use, show that it “works” in their flawed test, and then recommend that approach to listeners.

So what was their test?

1. Make up iodophore or ethanol as per manufacture’s recommendation – e.g. producing a solution that would be a highly effective and food safe surface sanitizer.
2. They took 9 ml of the sanitizer solution and added to that 1 ml of a dense culture of either Brettanomyces or Lactobacillus.
3. At regular time intervals they streaked out the sanitizer:microorganism mixture onto an agar plate and measured the growth.

And, not too surprisingly, they found that the sanitizers “didn’t work”. Of course, instead of looking at their experimental design and finding the numerous flaws, they instead concluded that the lack of efficacy was because these agents were tested under clinical conditions, against clinical organisms, and that brewery organisms are just too tough for the sanitizer.

Which, for anyone with a microbiology background, made your eyes roll so hard that they nearly fell out of your head. The standard panel of microorganisms used to test sanitizers were selected because they are tough as nails.

In case its not obvious, here is why their test failed

They diluted their sanitizer to 90% of its effective concentration. Keep in mind, because these are intended to be food safe, they are formulated to use a minimal dose of the active agent in order to limit the risk to people who are exposed. Which means that diluting the sanitizer will very quickly reduce its efficacy. Now diluting to 90% is unlikely to render it ineffective, but it will lessen the amount of incorrect use it can tolerate before failing.

The liquid they added was a mixture of the test organism and growth medium. Which, for the record, means they added a solution that was 5-10% organic material. Iodophore works by oxidizing organic materials – proteins and lipids in particular. Guess what happens if you add a solution containing a lot of protein and lipid? Yep, it reacts with the iodophore, leaving nothing to react with the bacteria/yeast.

They used dense cultures. Again, remember the definition and intended use of these sanitizers – reduce the concentration of a contaminating microorganism 10^-5 to 10^-6, on a clean surface. A culture of Lactobacillus can contain up to 1 x 10¹⁰ bacteria (10 billion bacteria) per mL. So even if the sanitizers worked under these conditions, a reduction of 10^-5 to 10^-6 would still leave 10⁴ to 10⁵ bacteria (10,000 to 100,000 bacteria) per ml in the solution.

Correction: the math above is off by a factor of 10, as they diluted the cultures 1:10 into sanitizer. So there could be from 1,000 to 10,000 live bacteria/yeast per ml in the samples they plated…although in all likelihood there are far more bacteria/yeast present then that.

But Wait, It Gets Worse!

Not content to stop there, the brewers then attempted to “fix” the iodophore by increasing its concentration.

Just as a reminder, to be food safe, the concentration of the active agents are kept low. This way any incidental contamination of the food (beer, in this case) with the sanitizer is safe, as the active agent is diluted to near non-existence. It also renders the agent safe for the brewer to use, as even direct exposure to the sanitizer will likely be harmless. As one example, properly prepared star san can be chugged by the pint, with the only effect on you being a bit of tooth decay (thanks to the phosphoric acid, which is at roughly the same concentration that you’ll find it in soda’s such as Pepsi), and the runs (as the surfactant is also a laxative).

If you double, or triple the concentration of the agent, you’re no longer working with the agent at a concentration safe for food, and you’re likely entering the territory where the agent isn’t safe to be used without dedicated protection (usually a respirator and eye protection).

A Week in the Lab Can Save an Hour in the Library

If you made it this far, sorry for the curmudgeonly rant, but this stuff pisses me off. There is nothing wrong with amateur science. In fact, I hate calling it “amateur” science as a lot of really important scientific discoveries have been made, and continue to be made, by so-called “amateurs”.

But for the love of god, spend a bit of time reading before you design your home experiments. Otherwise, you’ll just end up doing stupid things like this. This mistake (and to be clear, this isn’t the first brewery to make this error) could easily have been avoided by reading the sanitation chapter in any introductory book on food manufacturing – or by reading their countries legal standards for certifying food-safe sanitizers.

But now, its been featured in a respectable podcast, and apparently presented at a professional meeting. Mis-information at its finest – and mis-information with the potential to get brewers into legal hot water for using inappropriate agents for food-contact purposes…and potentially putting the brewers at physical risk themselves.