According to this answer, heating potassium ferrocyanide will decompose it into potassium cyanide.

Now, potassium ferrocyanide is an approved anti-caking agent, and hence used in table salt, for example.

Table salt is often heated as part of cooking. So, isn't there a chance for trace quantities of potassium cyanide entering the body?

NB: I understand that the dosage here, if any, might not be lethal; am more concerned about trace amounts causing inefficiencies in bodily functions.

  • 6
    $\begingroup$ That reaction needs very high temperature (you need melt it) and high concentration. So in short,no. $\endgroup$
    – Greg
    Nov 6, 2019 at 3:33
  • 13
    $\begingroup$ The same question: why are peptides considered safe for consumption when they are just one reaction away from highly carcinogenic nitrosoamines? I hope you see the parallels. $\endgroup$
    – Jan
    Nov 6, 2019 at 3:54
  • 4
    $\begingroup$ Consider that when heated to 600 Celsius or more a lot of things might decompose and release toxic compounds. This is a major problem in caser of fire in closed environment, when gases can be as harmful as the fire itself. This is to put the example you are giving in place. $\endgroup$
    – Alchimista
    Nov 6, 2019 at 9:12
  • 2
    $\begingroup$ I remember my chemistry teacher in the high school in early 80s, who was approaching then his retirement. He told us, he had once been a crazy young teacher, who had demonstrated this safety by eating potassium ferrocyanide in front of his class. $\endgroup$
    – Poutnik
    Nov 6, 2019 at 12:33
  • 2
    $\begingroup$ The reaction described in the linked answer involves producing molten KCN in a crucible. KCN melts at 634 °C. No food could survive at that temperature. $\endgroup$
    – barbecue
    Nov 6, 2019 at 17:00

2 Answers 2


Under biological conditions it is almost impossible to release HCN

Free cyanide can be released from potassium ferrocyanide by heating or by strongly acidic conditions (and some heat). Neither of these conditions is possible in a living organism.

The reason why this is so is because the cyanide ligands are very strongly bound to the iron (in slightly more technical terms the dissociation constant is very, very small (ca 10-24)). Most of the mechanisms that generate cyanide require much more violent chemical attack: waiting for the cyanide to dissociate and then protonating it to give deadly HCN will never result in a notable amount of NCH. This means that, in biological systems, ferrocyanide is effectively inert.

This explains why the LD50 in mice, for example, is about 6g/kg which is a lot ([see the safety info here]2 for both ferri- and ferrocyanides).

It is also worth noting that food safety authorities have studied the compounds and agree that:

The Panel concluded that ferrocyanides (E 535–538) are of no safety concern in these current authorised use and use levels.

So, while it might look like you are "one reaction away from cyanide" in practical terms this is not remotely true.

  • 4
    $\begingroup$ @OscarLanzi yes, typo. But the central issue that Fe(CN)6 complexes are very strongly bound is the same for both oxidation states. $\endgroup$
    – matt_black
    Nov 6, 2019 at 12:33
  • 13
    $\begingroup$ Worth pointing out that cyanide is toxic because it binds so tightly to iron (it prevents hemoglobin from carrying oxygen). If it's already bound to iron, it won't affect your hemoglobin. $\endgroup$
    – chepner
    Nov 6, 2019 at 13:44
  • 5
    $\begingroup$ I was asking about the decomposition that might happen during cooking, not inside a living organism. Are cooking acids like vinegar and typical oven temperatures (260 centigrade) strong enough to trigger the decomposition, partially or otherwise? $\endgroup$
    – HRJ
    Nov 6, 2019 at 14:00
  • 4
    $\begingroup$ @HRJ it is also worth noting that you need about 0.25g of pure potassium cyanide to kill an adult. This is a lot. Typical table salt contain maybe 10 micro grams of the ferrocyanide per gram. You would die far faster from excess table salt even if all the cyanide were released. $\endgroup$
    – matt_black
    Nov 6, 2019 at 14:26
  • 2
    $\begingroup$ @nick012000 Citric acid (principal acid in lemon juice) isn't notably stronger than acetic acid. Besides, how are you going to get enough ferrocyanide to notice anything given how little there is in table salt? $\endgroup$
    – matt_black
    Nov 6, 2019 at 16:28

Because like so much more, in small doses it's fine. Unlike the previous answer though, I think you may be on to something when you say

I understand that the dosage here, if any, might not be lethal; am more concerned about trace amounts causing inefficiencies in bodily functions.

I think the only concern being about the conversion to cyanide is a little misled though.

See here for the first case of poisoning by potassium ferrocyanide. The guy drank two glasses of it of course, so what's in your salt will never kill you.

Please don't let me speak to any health implications though, I'm not a chemist or biologist. Just providing a source I found and adding my own two cents.

For why it's allowed, there's no evidence it hurts people, and certainly isn't lethal when used as allowed. whether or not it causes slight inefficiencies in bodily functions... I think it's likely it does. But I have no basis for saying that, and it certainly won't stop me from eating table salt with potassium ferrocyanide. Again, fully an assumption, but I assume the 'harm' is negligible. Also, peppers can be poisonous in large doses and have a host of health benefits when eaten regularly and in small amounts. I think we need some serious funding and studies on it's effect on the human body over a real sample of the population to give a definitive answer.

TL;DR; I don't really know, but it's clearly poisonous in large doses, so as far as I can see small doses have to have some, maybe negative, effect.

  • 1
    $\begingroup$ This is an attempt to answer what I think might be more the spirit of the question than literal translation, because I think @Jan's comment (2nd on OP) answers the exact question very well $\endgroup$
    – TCooper
    Nov 7, 2019 at 19:24
  • $\begingroup$ Thanks for reading the question fully and taking the time to answer! Since it is not conclusive I won't mark it as an accepted answer. In my naive opinion, anti-caking property seems like a small benefit compared to the potential downside. BTW, I can't access the linked article (some problem with cookies / javascript). $\endgroup$
    – HRJ
    Nov 8, 2019 at 16:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.