Skip to content

Preserving your product without ‘chemicals’ – a worked example.

September 15, 2017

In cosmetic science, as in all aspects of life, it is important to be careful of the stories you tell in case they come back and bite you.  The ‘chemical’ free preservative story is one such case but, as it is a story that many people want to tell (and I can understand why) it is one we should explore pro-actively, so that we might write our own product narrative on fact rather than fiction. With that in mind, this is a story about preserving your product with Radishes and Coconuts.

Leucidal:  Radish Root Ferment Filtrate.

AMTcide:Lactobacillus & Cocos Nucifera Fruit Extract.

These ingredients are both for sale here in Australia at New Directions and there is a bit more information about them on their website here and here. 

Otherwise you can read more data from the manufacturer Active Micro Technologies by following this link. 

The above preservative pairing may be attractive to those with a brief to product a product that is:

  • Natural
  • Food Inspired – If you can’t eat it, don’t put it on your face etc.
  • Palm Free
  • Biodegradable
  • Sustainable
  • Vegan
  • Approved for use in organics
  • Gentle
  • Multi-functional.

But don’t get too excited because the above pair may not be good for those with a brief to produce a product that is:

  • Cost effective
  • Clay based or made with clay
  • Containing honey or milk (honey can be OK at very high levels but is a pain when diluted)
  • Foundation/ Concealer/ BB Cream etc
  • Containing natural exfoliant particles.
  • Containing plant matter not including filtered glycerin or alcohol extracts with low initial micro count.

So, now we’ve got that cleared up let’s explain the plan.

I wanted to come up with a simple, but not too simple, cream that we could use as a base – not a ‘nothing’ base but a base that would be fairy well representative of what a customer might start out with.  A daily moisturiser with medium viscosity and intensity.  Here it is:

OK so this is a simple cream as you can see. The water phase contains a thickener (Acadia and xanthan Gum) to help build viscosity and also stability, especially freeze/thaw stability and glycerin – at high level in this cream, this is to bind moisture to the skin and also to maintain a cushiony texture over time – cetearyl alcohol heavy creams can dry out over time.  The oil phase contains just an ester (fractionated coconut oil type, light skin feeling, non greasy), Organic Camellia oil – another lovely oil for the skin and especially for facial care as it is so soft and light and then there is the emulsifier and emulsion stabiliser/ thickener (cetearyl alcohol).   As you can see from the numbers the oil phase in this cream is quite substantial (albeit simple) and so the emulsion will contain plenty of nutrition for microbes without being ridiculously impossible to preserve.   Lastly I wanted to add a liquid (as in water based) extract to just add a bit of micro-drama into the product.  This green tea extract contains Water, Alcohol, Camellia Sinensis Leaf Extract, Xanthan Gum, Potassium Sorbate, Citric Acid – so it is preserved as-is but may still increase the microbial stress on the formula, especially given it is there at 3%.  Finally the preservatives were added which, in this case was 2% of each of the Leucidal and AMT Cide.  You may notice that I have not added a chelating agent into this formula. I usually put chelating agents into everything but didn’t want to ‘help’ the preservative out in this case.  Adding something like EDTA, EDDS or Sodium Phytate would be a worthwhile thing to do, especially if you were looking to add more challenging actives into your base cream.   Finally I opted to settle on a pH of 5-5.5 for this cream so as to be skin friendly. This pH is more tricky to preserve than a pH of 4-4.5 which is slightly more extreme and unfavourable for some microbes so I could have helped myself by lowering the pH slightly – that is a strategy I could try should this product fail PET.


Good Manufacturing Practice can be the difference between a pass or fail but again, I like to operate in a ‘this world’ sphere as best as I can rather than operating in a lab that looks like it belongs on a NASSA space program.  So, to that end, I made this in my lab down the garden with the door open to the sunshine and showers.  I didn’t even wear gloves to make this and neither did I tie my hair back but I did take reasonable precautions to not contaminate my product.  The water used was demineralised, high purity water and I churn through my ingredient stock very quickly so as not to have any old and rubbish stuff going in.  But other than that it was no cleaner in my lab than it would be in the average kitchen.  Finally I packed the cream into a glass bottle that had come out of a full box – I didn’t pre-sterilise it though so it was only as good as it left the glass bottle factory.


So I sent off my sample to the test lab and waited.  For a Preservative Efficacy Test around 100g of product is needed.  The test runs for 28 days and during that time the moisturiser is ‘challenged’ by having a known amount and type of microbe added to it at. The product is then measured at  7, 14 and 28 days to see how it is fairing.  The speed at which it cleans up this micro soup mess is important and ideally the cream will meet every ‘challenge’ with glee and keep the cream and its user safe.  The idea behind this is to emulate what could happen in-use with people putting their hands into the container repeated times.  A challenge test should be challenging so it introduces quite a bit more ‘dirt’ than you might usually find during use over a typical use period.  If you want more info on PET I wrote the blurb on the New Directions site so you can check that out. 


The first thing to note with these results is the Initial TVC – this is the initial micro count.  You can send your product off for a micro count and you would get a report with just that top bit. This basically confirms whether your product starts off clean or not.  If the product is too dirty to start with it may already be ‘out of spec’ for micro and not make it to PET.  For a cosmetic of general use we need the TVC to be under 1000 and for the product to contain no pathogens.  This product meets that by a mile having counts in all categories of less than 10.

The second thing to look at now is the table with data for 0, 7, 14 and 28 days.  This is the challenge part.  As you can see different microbes are added into the cream at the start and their numbers are recorded.  Ideally a product will quickly knock out all threats leaving a very clean product almost immediately but as you can see here, while the product complied it did take a bit of a while to get to grips with the A Brasillensis and C.Albicans.   This result with regards to these microbes is why, in the beginning I cautioned people about using this in a more challenging formula.  These more natural preservatives are typically quite slow to act as they rely on either starving the microbes, crowding them out with their ‘good’ bacteria or by acidifying their environment to make it more hostile to microbes.  I like to think of these natural preservatives as being a bit passive-aggressive.   Imagine you are at a party and the host wants to go to bed but people are still drinking and partying.   The host could either a) quietly squirrel away the alcohol, turn the music down and start cleaning up  or b) grab a baseball bat, stand on the table and shout ‘if you don’t all leave now I’ll smash something’.  These preservatives are doing strategy a.   Now strategy a works well for polite and gentle people but not all microbes are polite and gentle, some are the micro equivalent of a criminal gang and violence is all they understand.  Bringing that back to a cosmetic situation a challenging formula such as one that contains clays, other minerals and plant matter probably does require a more direct and faster response save the situation turning into a nightmare.

The last thing to note here is that the above product didn’t end up as clean as it started.  We started with a TVC <10 for everything but ended up with a A brasillensis count of 550 CFU/g.  This is still OK for a body product but I’d want it to work a bit better if we wanted to market this product for babies, people with eczema or for an eye cream.  In any of these scenarios the micro result might still be appropriate if we are going to package this product in a way that avoids such an insult but if we are going for a jar, we should really look to bolster our micro protection to save the product becoming overwhelmed in a worse-case-scenario.

The take-home message.

I wanted to share this with you because I really wanted to show you that a) these ‘gentle’ and food-like preservative systems can and do work but that they aren’t as robust as some other options.  I also wanted to make you aware of some strategies for improving your success with a preservative system like this (chelating agents, good manufacturing practice, smart packaging choices, lower pH etc).  I also wanted to share this worked example with you to help you really see why people like me worry when clients ONLY want to use nice sounding, food-type preservatives across their whole range  – that strategy is unlikely to work for everything and over time it does run the risk of leaving your factory open to ‘house microbes’ that evolve strategies to resist your gentle anti-microbial strategy. This can end very badly.

Overall I love the fact that these days you can preserve your product with ‘nice’ sounding ingredients that actually work. I love the fact that we now have so many choices available to us and I also love that a simple test like this can tell you so much – albeit at a cost (which I prefer to see as an investment actually).

One test does not an experiment make but it is a starting point and with that I’ll leave you to go off and run some tests of your own.  Remember, in the world of cosmetic science, it is better and safer to weave your narrative around fact, not fiction and certainly not fear.

Amanda x

PS: I should also mention that there are different test methodologies available to the cosmetic chemist. I opt for the ISO 11930:2012 method as it was developed specifically for cosmetics rather than a pharmacopea method which, in many cases is over-the-top in terms of required endpoints.  That said, it might be worth opting for a stricter test if you are making products for vulnerable people.   I should also mention that it is highly recommended that you run your PET at the beginning and towards the end of your shelf life as a minimum. It is important that the preservative stays active for the whole shelf-life of the product and that you have evidence to prove that.  A micro count at the end of shelf-life is better than nothing and is a bare-minimum starting point. Again the risks rise relative to other choices you make including packaging, pH and pack size.



20 Comments leave one →
  1. September 15, 2017 2:39 pm

    Thanks for the great information!
    I think an even better option in the “natural” range of preservatives would be LEUCIDAL® SF COMPLETE (Lactobacillus Ferment & Lactobacillus & Cocos Nucifera (Coconut) Fruit Extract) as it is a broad spectrum preservative, or to combine AMTcide (Lactobacillus & Cocos Nucifera Fruit Extrac) with for example potassium sorbate to make it broad spectrum.
    On the other hand in PubMed is a paper which deals with Leuconostoc/radish root ferment filtrates with the conclusion that the broad spectrum antimicrobial activity doesn’t come from the peptides but rather from didecyldimethylammonium salts and salicylic acid – not a good choice for “natural” solutions.
    What do you make out of this? (Just go to PubMed and put in Leuconostoc/radish root ferment).
    Thanks again for your great source of information,

    • RealizeBeautyEd permalink*
      September 15, 2017 3:14 pm

      Hi there,

      I based my preservative selection on the scientific logic of ‘what has lovely names and might work’ rather than anything else so it is quite probable that this isn’t the best combo 🙂 I will look at the reference and see what I think from there. Thanks for the participation.

    • RealizeBeautyEd permalink*
      September 19, 2017 12:17 pm

      Hi Judith,
      I’ve had a quick look at that article and it does seem to be a worry. There was a similar study looking at grapeseed extract as a preservative a while back that found it contained anti-microbial quats and the Honeysuckle extract that worked because it was a natural source of paraben. I think what I’d like to do is this, I’d like to find a way to run an assay on a sample of the radish root ferment filtrate specifically looking for the didecyldimethylammonium salts (these are more commonly used as wood preservatives). I am less bothered by the presence of salicylic acid as that is a naturally derived chemical anyway so it is possible the plant contains some (but it might also not). The important thing here will be to test to see if the didecyldimethylammonium salt is present, if so how much and if so how come? The same study failed to find any peptides – the thing the preservative is supposed to contain – that is also interesting although there may be an explanation for that. The study you referenced doesn’t look to have made much impact in the scientific world and has only been cited 3 times so far which is not much. It might not mean much but it is also worth looking into the methods they used to make sure no cross-contamination was possible. As with anything like this, first glance looks alarming but often the reality is far less dramatic albeit it might uncover a truth that is somewhat inconvenient. What I do know though is that if this preservative doesn’t contain the active it claims it contains then that’s not good at all but it would take more than one assay to get the evidence for that. I’ll see what I can find and will probably blog about it a bit later once I’ve got some info to go on.

  2. September 19, 2017 1:11 pm

    Hi Amanda!
    Great that you will look further and thanks so much for your long reply.
    I know about the GSE and thought maybe it’s a similar situation here.
    Looking forward to keep in touch!

    • RealizeBeautyEd permalink*
      September 19, 2017 1:26 pm

      Something else to ponder – as much for the keen gardener as the chemical investigator is this. The contaminant found is a common wood preservative. What if the radishes were grown in a wooden planter or a garden bed made with wooden sleepers? This quat is known to contaminate soils and root vegetables are known to take up this contaminant via their roots (many will store it there too). It is not beyond doubt that the rogue chemical, should it be validated as being there – may have come from the soil that the ingredients were grown in. Whether it was put there consciously or not would be a further question. I think it is important to think through all avenues as if soil contamination is a factor then we can’t really exclude the notion that this chemical could be present in any vegetable based ingredient albeit to varying degrees. I am not even sure if organics would be immune given that this chemical can enter the soil via decaying wood. I wonder if the organic standard covers the wood that beds might be made out of? Interesting indeed.

    • RealizeBeautyEd permalink*
      September 20, 2017 8:11 am

      OK so I went back through to the manufacturer via their agent here in Australia. Here is what they say:


      This article was addressed over twelve months ago by Active Concepts.

      They flatly refute the article and question the methodology, the results and the interpretation of those results. They especially question the specious remarks about the presence of salicylic acid in the product.

      The following email discussed the article and the attachments confirm the contents of the email.

      Should you have any further questions please don’t hesitate to ask me.

      From the manufacturer:

      M15008 Leucidal Liquid does not contain any didecyldimethylammonium salts. Please find the below which may be used to highlight the attached official AMT response to this article, as well as the IR spectrum comparing M15008 to DDAC. You can share this with whomever is concerned

      The claims published by this paper eluding that the antimicrobial activity of Leucidal Liquid is not based on antimicrobial peptides but rather, salicylic acid and a didecyldimethylammonium salt (such as DDAC) are completely flawed and inaccurate. The salicylic acid isolation results were not entirely surprising and in fact, we have attested to this on multiple occasions both internally and publically. We fail to see why the presence of salicylates in Leucidal Liquid is something considered worthy of publication. We have been very clear in disclosing the salicylate content in this product and in fact list it on the specification (Phenolics, tested as salicylic acid: 18.0 – 22.0%). AMT can re-attest to the fact that the composition of Leucidal Liquid is unchanged from what is reported on its compositional breakdown and specification. This product contains natural salicylates and is the result of fermentation of radish root in the presence of Leuconostic kimchi. It is public knowledge that the product contains phenolic compounds, which are known to be a mixture of salicylates. Therefore the fact that these may appear as salicylic acid and the authors isolated it as such is simply an artifact of the chosen test method. Certain required conditions of some test methods (such as HPLC used to identify phenolics present in Leucidal Liquid) inevitably convert the present salicylate compounds into salicylic acid.

      The second erroneous claim is that a didecyldimethylammonium salt is present and could be isolated from Leucidal Liquid, which is completely invalid. There are no didecyldimethylammonium salts in Leucidal Liquid, so we are not sure how or why those results were even obtained. We do, however, know that interaction between radish root and rice phytochemicals such as indole-3-carbinol and glucosinolate can be responsible for the creation of aromatic polyamines that may be detected in Leucidal Liquid. However, we can undoubtedly confirm that no didecyldimethylammonium salt, such as DDAC, is present simply based off the composition, manufacturing procedure, and FT-IR comparison alone. As can clearly be observed in the FT-IR analysis, there are distinct peak differences in absorbance and intensity between DDAC and Leucidal Liquid, specifically within the 3500 – 2500 range and 2000 – 1000 cm -1 range indicating that the compound is not present at detectable levels. Lastly, carbon dating analysis must not be taken at face value without carefully examining every component of the product being tested, because in this case the fermentation media of Leucidal Liquid is the cause of the fossil-based carbon content this paper claims to be present. The study simply did not take into consideration the effect of fermentation media on carbon distribution, which is an investigative scientific failure on their part

      My Comment: I have no reason to distrust what the manufacturer is saying. Indeed I have every reason to think they have made an honest product and that should any ‘synthetic’ or ‘undesirable’ chemical be present it would have either come from a) contamination in the growing process of the material b) contamination in the supply chain C) contamination or mis-interpretation in the analytical side. It would be a crazy idea for a company this high profile to intentionally dupe the public by adding a known synthetic biocide into a natural product while trying to pass it off as natural.

      With regards to the microbial peptides from what I can gather after a very brief email from a micro lab, testing probiotics is not straightforwards and requires special handling to accurately analyse them. There might be something missing from the test method that gave that result. Then again there could be other explanations good and bad but the point is, we still can’t be sure.

      I guess the bottom line here is that science is a way of investigating things and that investigations can lead to different conclusions,the most common conclusion being ‘we need more evidence’ thus starting the merry-go-round again! As with justice and law, sometimes the simplest explanation is misleading and coloured by bias but then again sometimes we over-complicate things ‘just in case’. In this case I’d say the spotlight is on this company anyway, being a supplier of natural goods, so I’d expect them to be doing more analysis and adding to the knowledge base of what really happens chemistry wise in a fermented extract. Other people like me assaying their actives is a possibility but I, like others would have to ensure we were pursuing the assay or count in the right way and interpreting it with an open mind. I was curious of how readily that paper jumped to its conclusions. It is unusual for science to be so sure, especially when the result is unexpected or unusual.

      Anyway, I hope that helps. I’ll still be using these preservatives.

  3. September 20, 2017 10:07 am

    I’m still not convinced – why would scientists from accredited universities go on a crusade to discredit a manufacturer of cosmetic ingredients or publish content if t is not thoroughly researched?
    Claims or studies from a manufacturer’s side are always going to be in their favour, otherwise they lose their selling point. That’s why (at least) I rather go to scientific studies like PubMed, Wiley and the likes to get my informations.

    Thanks again for your great efforts!

    • RealizeBeautyEd permalink*
      September 20, 2017 10:17 am

      The note at the bottom of the research re funding and interest:
      Funding Information
      This work was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC), Griffith Laboratories Limited, Alberta Innovates Health Solutions, and the Canada Research Chair in Bioorganic and Medicinal Chemistry.

      The authors declare the following competing financial interest(s): Griffith Laboratories produces and markets food ingredient systems for enhancement of food safety and flavor.

      Of course one would expect the ingredient manufacturer to defend themselves and their product. They are not in court on the basis of this and as such are not being judged or cross examined so we can either take what they say at face value or not.

      I am naturally skeptical but also trusting by nature so my take on it all is that there is more to learn about the chemistry of these fermented extracts, especially with regards to how they actually work as preservatives. I would like to get to the bottom of that without getting hung up on this particular study although one would be foolish to discount it entirely. That’s my take on it all 🙂

      • Judith permalink
        September 20, 2017 10:25 am

        Your totally right – and I most have overseen the conflicting interests on the bottom even though these involve food additives and we’re talking cosmetic ingredients.

      • RealizeBeautyEd permalink*
        September 20, 2017 10:42 am

        They would be interested in how this company has achieved something they cannot. Science works like that and is always cross-discipline but it can lead to people jumping to conclusions that may or may not be right. I can’t say that this is what has happened here but it is important to balance the interests and risks of the investigator with the interests and the risks of the investigated 🙂 It is never easy hence why most robust sets of data are amassed from multiple independent sources in a systematic review. All very un-sexy and time consuming but interesting nevertheless x

  4. Kate permalink
    September 22, 2017 12:00 am

    Hi Amanda, I noticed some Miessence skin conditioners now have Leucidal Radish Root Ferment Filtrate as their second ingredient, on their website it says it increases the moisutrization of cream formulations. I found this quite unusual as I have only ever seen it as a preservative listed at the end of formulas, unless all the ingredients after it are in such tiny amounts there is only a small percentage of it. May I ask, my skin is very sensitive to aha/bhas and I avoid them in leave on products, should I be concerned about the salicylic acid content of the Radish Root? Any advice would be appreciated! Thank you for your informative article.

    • RealizeBeautyEd permalink*
      September 22, 2017 6:10 pm

      Hi there Kate,
      I can’t speak for Miessence but what I can say is that the Leucidal can be used at levels up to 4% so it is absolutely possible for this to appear high up on the ingredient list of products, especially those based on simple concepts with few ingredients. The manufacturer of the ingredient does mention some skin conditioning benefits which may be related to the fermented nature of the ingredient and its AHA content.

  5. Kate permalink
    September 23, 2017 2:39 am

    Thank you. Hopefully there will be more studies in the future as I don’t really know what to think about it right now. (I am suprised they used it as an active ingredient when there are more transparent, studied and risk free ingredients out there)

    • RealizeBeautyEd permalink*
      September 23, 2017 8:01 am

      I can see how this would look a bit odd but I’d caution against viewing it as a negative. Some companies like to keep their ingredient lists simple and the fewer the ingredients, the higher up the list the preservative will end up. Seeing ahead of other extracts is less surprising when you know the preservative can be safely used up 4% . Anything under 1% can be listed in any order so if this company had opted for a synthetic preservative blend that maxes out at 1% usage, the components would all be less than 1% so they could, in theory place the preservative elements at the end of the INCI thus giving the impression that the extracts are at much higher concentrations than they actually are (You would not be able to tell if the extracts were at 2%, 10% or 50% for example – extracts are most always used at below 4% each, not least because many plant extract powders are 10,20, 50 or 100 times stronger than the fresh plant so reconstituting them doesn’t take much. So you could add 1% but have it equivalent of 10, 20, 50 or 100% fresh plant. Labelling makes you list it as 1%. I hope that helps.

  6. kate permalink
    September 25, 2017 4:44 am

    Hi Amanda, thank you for all the info. Now I look at the ingredients there is just water and extracts following it, so it could well be less than 4%. Although I do think they should mention it can exfoliate in the ingredient bio not just that it moisturises, as I would never have realised this unless I googled it, as even low percentages of salicylic acid can be quite strong on sensitive skin. From a picky consumers point of view I really wish the INCI listed below 1% too, I would be interested what the amounts of the final ingredients are as there can be a huge difference between 1% and 0.01% of an extract or a preservative.

    • RealizeBeautyEd permalink*
      September 25, 2017 9:56 am

      That is true Kate and in some ways it would be useful for INCI lists to go from highest to lowest all the way to zero. I’m not sure what reasoning was behind them not. With regards to salicylic, lots of plants contain salicylates and I suspect that many plant extracts are the cause of minor irritation and discomfort for some clients.

  7. Dom permalink
    October 17, 2017 5:35 pm

    I am trying to decide with natural preservative system to use: 1) Sodium Levulinate + Sodium Anisate vs 2) Natapress + Radish Root Extract + Willow Bark + Aspen Bark.

    • Dom permalink
      October 17, 2017 5:53 pm

      I forgot to mention the formula I came up with is 9% of formula (Natapress + Radish Root Extract + Willow Bark + Aspen Bark.). ty

    • RealizeBeautyEd permalink*
      October 19, 2017 5:38 am

      It is difficult to be able to help you with that with an opinion. If I had your formula, formula pH, packaging choice and manufacturing method information I might be able to give you a heads up on whether I think it will work (but I’d charge for that sort of analysis) so that leaves us with testing. I would go get a PET (Preservative Efficacy Test) done on your creation to see how it stacks up. I assume the ingredients you have used don’t exceed their maximum level? There looks to be a heck of a lot of ‘preservative’ there. I’m not personally familiar with the Natapress or how the Aspen bark works at this stage. I’d have to look into that too.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: