Lately I have seen a number of unrelated "scientific" debates over whether certain substances should be outlawed because they are toxic to humans. My initial, informal reaction is usually to respond that anything is toxic to humans if you give them a sufficiently large dose.

However, formally I don't know if that's really true for everything a human being could ingest in some way. I started to wonder if there were some substances that our body could handle unlimited amounts of without any negative consequences.

As this question has been (correctly) identified as a bit vague, I'll try to explain what i'm looking for. For the purposes of this question, I'm willing to ignore the limitations of actually ingesting a given substance in "the usual way". For example, if you can't physically drink enough of some liquid fast enough to kill you without your stomach filling up and vomiting, but that same liquid injected intravenously could be lethal, I could consider that toxic. I also recognize that the body can only physically contain a certain volume of stuff, after which sheer pressure would cause it to fail; I'm more interested in "biochemical toxicity" as opposed to any physical damage (I just don't know the term for what I'm looking for.)

In other words, one of my goals is to learn if, under laboratory conditions, a properly motivated researcher could always find a dose that would be toxic, regardless of the impracticality of a real person ingesting that dose under normal circumstances.

So, with that qualification, my ultimate question is:

Is there any substance we know of that is completely non-toxic to humans at arbitrarily large doses ingested over an arbitrarily short period of time?

  • 4
    Strictly speaking, I suppose the inert gasses (Helium, Neon, Argon, &c) wouldn't be toxic. Likewise metals like gold & platinum that have low reactivity. Though swallowing a bunch would probably kill you by mechanical obstruction of the digestive tract. More interesting to me are things like vitamin A, which are necessary for health at small doses, but toxic at larger ones. – jamesqf Jan 30 '15 at 18:41
  • 1
    I rather think there's a lot of things that would kill you at a high enough dose, but I don't think qualify as toxic. Like matching blood plasma clearly won't be toxic, but eventually the increased pressure would kill you. Can you clarify your question as per this? – Mooing Duck Jan 30 '15 at 22:59
  • 1
    @MooingDuck I hope that helps. I just don't know the word that means what I want; i want "toxic because it interacts biochemically with the body, not because your kidneys rupture under the pressure." – KutuluMike Jan 30 '15 at 23:16
  • 2
    @jamesqf: Gold actually is toxic at high doses. According to rarediseases.org/rare-disease-information/rare-diseases/byID/…, "Overexposure to gold (as in treatment of rheumatoid arthritis) may cause skin rashes; bone marrow depression; stomach and intestinal bleeding; headaches; vomiting; focal or generalized continuous fine vibrating muscle movements (myokymia); and yellowing of the skin, mucous membranes, and whites of the eyes (jaundice)." (The same page also mentions platinum poisoning, but gives no information about it.) – ruakh Jan 31 '15 at 4:44
  • 1
    @jamesqf Yeah, heavy metals have their toxic effects in some ionized state, in which they substitute for a lighter metal that's used as a cofactor on some critical metabolic enzyme, causing it to malfunction. If you swallow a lump of the un-ionized metal, it'll be basically a question of how much your digestive system can corrode it, which for gold = hardly at all. – zwol Feb 1 '15 at 1:51
up vote 33 down vote accepted

I’ll answer this theoretically, since that’s how it has been posed. And if we’re ignoring practicalities, we may as well posit that the substance in question will be introduced directly into the bloodstream (This is, of course, simple to do in reality, but not how most people consume their non-toxic substances.) The easiest way to show that any unspecified substance can be toxic at an unlimited volume is to invoke the human body’s mechanisms for volume homeostasis.

As mentioned in this answer, the human kidneys functioning optimally can produce up to ~ 25 L/day of urine.1 This would require complete suppression of ADH (anti-diuretic hormone, a.k.a. arginine vasopressin), which would occur only if the “toxin” load were markedly hypotonic (think water).2 There is therefore a theoretical maximum volume of any substance that can be dealt with by the body, which is something less than 25 L per day. (For any substance other than water, the maximum will be lower because ADH will not be as fully suppressed by a less hypotonic load.)

A volume of any substance introduced into the bloodstream (including a product precisely mimicking the constituents of the bloodstream itself!) will overwhelm the body’s homeostatic mechanism. This will result in edema which is unpleasant and, in the case of pulmonary edema, certainly pathologic - a “toxidrome” in your scenario. In the case of hypotonic fluids, serum osmolality will also fall causing hyponatremia with all of its consequences.

Summary: No, the human body can not tolerate an unlimited volume of anything, therefore there is no substance that is non-toxic "at any dose."


1. Christopher Lote. (2012). Principles of Renal Physiology. Springer New York.

2. No, you may not drink 25 liters of water per day. For one thing, urine can not be made with a tonicity of 0 to balance this (more like 60 mOsm/kg minimum). Additionally, ADH can rarely be completely suppressed, yielding a somewhat more concentrated urine and therefore lower tolerance for hypotonic intake before serum osmolality is compromised.

  • 1
    @BlueRaja-DannyPflughoeft - The question posited "an unlimited amount," not me. :-) This just seemed to me the easiest route to address a question that is otherwise difficult logically (to "prove" that no such thing exists). I went through the factors that dictate that physiologic limit for volume intake and the toxicity of exceeding it, which is why I considered it substantive enough to be an answer. You may disagree. – Susan Jan 30 '15 at 18:03
  • 6
    I know, I upvoted your answer. I just think it's a silly question. – BlueRaja - Danny Pflughoeft Jan 30 '15 at 18:07
  • 1
    Not a silly question at all, though perhaps needs to be re-worded to add qualifications about amounts a person could potentially consume - as for instance water being toxic in large amounts. – jamesqf Jan 30 '15 at 19:05
  • 2
    @MichaelEdenfield The question is pretty highly upvoted, so I wouldn't worry too much. :-) Your edit did clarify your intentions. I actually think I've undermined them a bit by making it about volume rather than dose in any normal sense of the term. But if you want talk about the "most benign" sorts of infusions, my mind goes to isotonic saline or some (hypothetically non-immunogenic) whole blood product, and then the limiting factor is obviously volume. Anyway, glad it was helpful. – Susan Jan 30 '15 at 22:54
  • 1
    There is an applicable norwegian proverb that says; "Health in every drop, poison in every liter". – fileunderwater Feb 2 '15 at 12:25

It depends largely on the method of administration. If you are atomizing the substance and delivering it via water vapor, many, many substances have no known LDLo (lowest dose required to kill a member of the tested population). Almost any substance in existence has the potential to kill you if it is diluting your bloodstream via direct intravenous injection or oral consumption; however, when it comes to inhalants, many substances cannot kill you.

Since your question was specific to intoxicants, here's a couple of examples: There is no LDLo level of Tetrahydrocannabinol (the active ingredient in marijuana) when delivered via atomization. There is also no known LD50 (a similar, albeit somewhat less reliable, metric) for lysergic acid diethylamide (commonly referred to as LSD). For psilocybin (the active ingredient in "magic mushrooms"), the LD50 is high enough that an average person would need to ingest around 6 pounds before cause for concern.

Other far more dangerous substances that can kill with vastly lower amounts include anything that speeds or slows the heart rate: most specifically, cocaine (including crack), opiates (including morphine, heroin, and various pain pills), and any amphetamine, methamphetamine or derivative substance, or other stimulant (crystal meth, ADHD medication, and even caffeine or ephedrine). Of course, the most common killer categorically from a historical perspective is alcohol.

  • to be honest, I was motivated by the claim that propylene glycol (what's in most e-cig liquid) was toxic. But I also know that at least one THC study claimed that THC was toxic after IV-injecting it into rats at thousands of times a normal dose, which motivated the exact wording of my question. – KutuluMike Jan 30 '15 at 15:47
  • 2
    The absence of a known LD50 doesn't mean there isn't one. It just means it's beyond what anyone has been exposed to. – Loren Pechtel Jan 31 '15 at 1:38
  • Poor rats... :( – Lightness Races in Orbit Feb 2 '15 at 0:06

There is a problem with definition of toxicity — things that are dangerous in large amounts aren't usually called toxic. In spite of this, you're right: everything can be dangerous to a human in large enough amounts, or if delivered improperly.

For example, even water can be toxic if drank too much. Also, when it gets into the lungs, it may cause drowning.

On the other hand, air, while necessary in lungs, is dangerous if present as a gas in the bloodstream.

BTW, even botox (being one of the strongest poisons) is used in medicine in very small doses.

  • True, I knew the word toxic was being used sloppily, but that's the word I see used to describe things like IV-delivery of absurd amounts of THC, so I didn't know what other word to use. – KutuluMike Jan 30 '15 at 14:21
  • English is not my native language so you probably know better :) – Mithoron Jan 30 '15 at 14:24
  • First thing I thought of when I read this... and I remember reading about this way back when w/r Water Intoxication: cbsnews.com/news/woman-dies-after-water-drinking-contest – WernerCD Jan 30 '15 at 17:57

The inert gasses Helium and Neon are non-toxic when administered through inhalation, so long as the patient's oxygen supply is sufficient. They are also non-toxic when injected, so long as the injection is slow enough for them to be dissolved in the bloodstream.

You can be killed by them through various means (asphyxiation through oxygen displacement, rapid injection causing an air embolism, rapid decompression causing decompression sickness, and so on), but since the cause of death is unrelated to the chemical properties of the substance involved, it's not accurate to call this "toxicity" (unless you're XKCD).

Other inert gasses (Argon, Krypton, Xenon) may be toxic at high pressures: although I haven't found an LD50 for any of them, they can all induce nitrogen narcosis, and Xenon is usable as a general anesthetic.

  • 1
    Wrong. All the "inert" gasses will kill in sufficient quantity. It requires considerably more than you can breathe at normal atmospheric pressure but if you run the pressure high enough they turn evil. This is what imposes the limit on scuba diving--there's no gas mixture that doesn't become toxic at a high enough pressure. – Loren Pechtel Jan 31 '15 at 1:40
  • He, he :) Indeed this question smells like XKCD - arbitrarly large numbers are quite often in What if :) – Mithoron Jan 31 '15 at 1:40
  • @Loren Pechtel: But is it the toxic effects of the gas itself, or just that it's something saturating the blood at high pressure, then coming out when the pressure is reduced - that is, 'the bends'? Maybe splitting hairs, but I see it as like the difference between the toxic effects of drinking too much water, and drowning. – jamesqf Jan 31 '15 at 5:05
  • @LorenPechtel, I've updated my answer to take into account nitrogen narcosis, since Ar, Kr, and Xe can induce it. – Mark Jan 31 '15 at 20:09
  • @Mark I doubt anyone knows the LD50 for them. Where would the data come from? They're going to stop using them at the depth where the symptoms become serious, nobody's going to take them to a lethal depth. And it's not just Ar, Kr and Xe, H and He also do it. – Loren Pechtel Feb 1 '15 at 3:09

not even AIR, because if you force too much you will explode it depends on how extreme is the "any dose" statement

water is also toxic in large Ingestible amounts

and since we go into theoretical application the answer would be dark matter

so the final answer is nothing , because the human body has evolved to exist in some equilibrium, so too much of one thing even if is harmless it itself ( like water, or proteins ) it causes an imbalance, and as a result it becomes "toxic"

Even simple water is "toxic in high amounts" as kidneys can remove 25 l per day at most. All other substances are probably even more "toxic".

  • 1
    Downvoter, please leave a comment. – h22 Feb 2 '15 at 18:29

This is not really a substance, but anyway.

Neutrino

Neutrinos are ghostly particles that barely interact with any matter. Therefore, no sufficient amount of neutrino (that human can ever collect) can kill you. To have a lethal dose of neutrino radiation, you must stand inside the outer layout of a giant star that creates a supernova.

Source: XKCD, Lethal Neutrinos

  • The source you link (and your last sentence) say there can be a lethal dose of neutrinos. The point is just that it's hard to kill anybody with neutrinos because the usual sources of lethal doses of neutrinos also produce large amounts of more lethal particles, radiation and heat and therefore a human would die from other causes long before being exposed to the lethal dose of neutrinos. Therefore, neutrinos aren't an answer to the question. – Pere Jan 20 at 9:25
  • I agree. It's just that so is the top voted answer. So I take my deliberation and post this answer – Ooker Jan 20 at 9:30

Vitamin C won't kill you no matter how much you get into your body as long as it is enough to help prevent arterial wounds and atherosclerosis(This happens in people with scurvy or vitamin C deficiency because of how cholesterol is used to help repair the wounds and this can lead to a complete blockage of the artery and thus an infarction of all the tissue that artery supplies. So if you don't want to have an MI, one of the things you have to do is get vitamin C into your system.

Another example of this is chloride ions. While having high chloride ions might cause there to be a slower reaction time(since chloride acts as an inhibitor in neurons) it itself won't kill you. Yes it might affect the muscles by not having them contract as much as they are supposed to but this is naturally cured by urinating out more chloride and thus more sodium which can lead to a sodium deficiency which is bad because your body needs sodium in order to function. With a low amount of it it you wouldn't be able to think clearly or even have a normal heart rhythm which might lead to bradycardia from a higher pottasium concentration in the heart.

  • Doesn't your second paragraph just explain that chloride is in fact toxic at some dose? (The MI/high-dose vitamin C association also needs some referencing, except it's not really relevant.) – Susan Jan 31 '15 at 1:43
  • Wikipedia indicates that it is unknown if vitamin C can be toxic or not. It cites a source (at a dead link) that apparently managed to kill rats by administering excess vitamin C, but it is uncertain whether the cause of death was chemical or mechanical. – Mark Jan 31 '15 at 2:20
  • 1
    No. The second paragraph says that while chloride is nontoxic at a high dose, high chloride results in low sodium and it is the low sodium that is toxic. as far as the association of scurvy with MIs that is actually relatively simple. When somebody has scurvy this causes them to easily get arterial wounds both external and internal. The body uses cholesterol to heal these wounds in somebody with scurvy instead of platelets. This is turn causes more cholesterol to build up in the artery causing atherosclerosis. If this happens in the heart than it can cause a heart attack or MI. – Caters Jan 31 '15 at 19:19
  • And as far as the death from excess vitamin C it was probably mechanical and not from the vitamin C itself. And even if it was from the vitamin C that does not mean that excess vitamin C in humans will cause death. In fact I have been taught that since vitamin C is water soluble and vitamin C always helps your body by strengthening your immune system and acting as an antioxidant that vitamin C toxicity does not exist, only deficiency. – Caters Jan 31 '15 at 19:25
  • 1
    high chloride results in low sodium and it is the low sodium that is toxic : you’re describing a mechanism of toxicity there, not the absence of toxicity. – Susan Jan 31 '15 at 22:22

protected by AliceD Jan 31 '15 at 14:02

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

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