This question is mostly about the human body, but it's also about life in general.

It's not difficult to notice a big different between the abundance of elements on Earth & that in the body. I've been pondering why. Such a huge question, but here I choose to focus on the big 4. Below is my current understanding, please correct me if I'm wrong or add things that I miss.

An umbrella reason for why $\ce{O}$ & $\ce{H}$ are prevalent is because $\ce{H2O}$ is such a wonderful solvent. Life generally depends on reactions to function, and they happen best when immersed in a liquid. Water is good and copious, hence life use it.

  • Another reason for $\ce{O}$ to be the most common element in our body is that it can oxidize things, which can become handy.

  • For $\ce{H}$ is that it's basically a proton, thus can be a "filler" in most situations.

  • $\ce{C}$ is the backbone of life because it's the element with the best bond force. In other words, not a goldilock but the GOLDilock. Life loves to handle such flexible tool, thus it uses carbon.

  • That leaves the question for $\ce{N}$. Why do we even use nitrogen? It's not anywhere close to the top of abundance in the Earth's crust. It does not have attractive characteristics like $\ce{O, C or H}$. One may say that it's dominant in the air, but just because there's a lot of $\ce{N}$ doesn't mean we evolve to use it. We can always eat dirt to get some hypothetical elementary replacement for $\ce{N}$, for example. Besides, life began in the ocean and there's little $\ce{N}$ there. Do fish have different big 4? I haven't looked into it, but I doubt it.

Both $\ce{H2O}$ and $\ce{C}$ have something "best" in their belt. I guess $\ce{N}$ must have some stuff of highest quality, too, but I have no clue what it is. So, even though my knowledge of the other 3 elements is crude, I'm bugged by nitrogen the most.

  • 8
    $\begingroup$ Basicity. Amines are probably the best proton acceptors that can exist in water solutions. Also, nitrogen is much better in forming metallocomplexes, again, one of the best donor atoms in water solutions, second only to cyanide-ion, which simply cannot be easily tied to an organic skeleton. $\endgroup$ – permeakra Jan 23 at 5:24
  • 1
    $\begingroup$ Your first point may not be very relevant. $\ce{Cl}$ and $\ce{F}$ can also oxidize more things than $\ce{O}$, yet they are not as prominent as $\ce{O}$. Also, one of the reasons of using $\ce{N}$ is it's very high bond enthalpy: N can form very strong (mostly triple) bonds with stuff, which is something none of the other 3 can do. $\endgroup$ – Aniruddha Deb Jan 23 at 7:03
  • 1
    $\begingroup$ Note that the composition of living things is very close to the composition of non-living things (though it only slowly adapts to changes in availability, so there may be things we've been stuck with for billions of years). You could build better biochemistry with different proportions, but unless the advantage is great enough to offset the lower bioavailability, life is shaped by the resources available. If the readily available resource proportions were different, so would we. It needs to be especially useless (or dangerous) not to be used. $\endgroup$ – Luaan Jan 24 at 7:20
  • 3
    $\begingroup$ I'm not at all sure why this has been closed as an "opinion-based" question. It may not be possible to provide a definitive single answer but it is clearly possible to bring many important chemical facts to bear on the question. Closing as opinion based should be reserved for questions where there are no facts and there are only opinions. $\endgroup$ – matt_black Jan 26 at 21:05
  • 1
    $\begingroup$ @matt_black I broadly agree with you. At the same time, I'd point out that 2/4 people voted to close as "opinion-based" and 2/4 voted to close as "needs more detail". SE insists on hiding this, despite it being one of the more popular feature requests for almost a decade. Mind you, I don't entirely agree with either of those close reasons... Anyway, back to the question itself: it has 2 reopen votes, so there's a reasonable chance it will be reopened soon; but if it doesn't then to meta it goes... $\endgroup$ – orthocresol Jan 27 at 9:15

Because chemistry built on carbon, hydrogen and oxygen would be very dull and building complex structures from modular components would be impossible

The key point to remember is that life depends on complex chemistry. Living things are, by definition, things that can replicate and things than can build structures that can do complex chemistry. Metabolism is the process of taking simple things from the environment and making them into complex things. This requires complex machinery, built from simpler chemicals available in the environment.

While it might, in principle, be possible to build very complex carbon-oxygen compounds that could metabillise and reproduce themselves, there is no easy way to do so from simple components. But throw nitrogen into the mix and simpler routes to complexity are available.

Take the molecules that make up most of the complex machinery of all living things: proteins. Proteins make up most of the complex machines that do the job of metabolism and they also constitute many of the structural components that make up living things. But they are built from a small number (about 20) simpler chemicals: amino acids. Amino acids are fairly simple chemicals that might be thought of as the equivalent of Lego bricks for building an incredible variety of living structures. Most consist of only carbon, hydrogen, oxygen and nitrogen. All have the core structure $\ce{NH2CH(R)CO2H}$ (where R can be a variety of organic chunks) which constitutes the equivalent of the standardised bricks in Lego allowing all pieces to fit together so more complex structures can be built. In the case of proteins, that core functionality allows a variety of different amino acids to be "strung" together end-to-end (the amino end reacts with the carboxylic acid end on each acid).

The variety of different structures than can be created by changing the sequence of amino acids is staggeringly, unimaginably large. Many proteins contain hundreds of amino acids and you only need about 60-70 for the possible number of different sequences to exceed the number of atoms in the universe. This means that, by exploiting combinations of simple chemicals (which are not that hard to create with simple chemistry), life can suddenly explore an unimaginably wide range of possible complex structures. Perhaps such structures could be built from fewer elements, but not in a modular way that uses a small range of available components and a single way of joining them together.

The availability of this level of complexity from things that (mostly) consist of C,H, N and O is why the use of nitrogen is key to life. The constraint isn't the availability of the elements (though simple nitrogen-containing compounds are fairly available), it is the accessibility of complex structures from simple components.

  • 2
    $\begingroup$ @longtry: Nitrogen is a light atom and for that reason very abundant. Sulfur is not exactly rare, but not nearly as common. Still, there are common amino acids that contain sulfur. That's why decomposing biological matter has that particular odor - those are all volatile sulfur compounds we smell, and we evolved those sulfur receptors exactly because sulfur is so biologically common. $\endgroup$ – MSalters Jan 24 at 8:01
  • 2
    $\begingroup$ I kind of doubt the statement "No other element does that" unless you can help me understand why N is uniquely so. See, in the structure of the amino acid you mentioned, NH2CH(R)CO2H, what's the most interesting and 'opens up a number of variety some 30 orders of magnitude more than our universe's total atoms' is that "R" part, and it connects to a carbon atom. What I'm trying to say is somehow, (maybe by chance?) life came up with that nitrogen in that structure, thus we say "amino acid". It can be whatever else. How about ClH2CH(R)CO2H? $\endgroup$ – longtry Jan 24 at 10:58
  • 1
    $\begingroup$ @longtry The key point about amino acids isn't the R, it is the core structure that allows long chains to be built with a variety of Rs. If it could easily be done in other ways using other elements, then we would have examples from known chemistry (chemists are pretty creative). We don't. $\endgroup$ – matt_black Jan 24 at 12:08
  • 1
    $\begingroup$ @longtry It is the easy condensation reaction between amino acid amines and carboxylic acids that is the key. $\endgroup$ – matt_black Jan 24 at 15:28
  • 1
    $\begingroup$ I feel that this answer, which is very good already, may be improved by a few words regarding why $\alpha$-hydoxy acids (think lactate) could not do the job of nitrogen in amino acids. Proline being the glaring exception. matt_black mentioned the condensation, but esters are rather stable, too. $\endgroup$ – TAR86 Jan 24 at 16:54

My take is a little bit different. Organisms evolved to adapt to the chemistry that is there. And what's there, primarily, are hydrogen, carbon, nitrogen and oxygen. Plus, as we shall see, sulfur, which cannot be ignored in proteins.

Here we find a list of the most common elements in the Universe. Although everything else us made from fusion of hydrogen, the abundance order is not based on atomic number a lobe. Nature tends to favor certain elements that form especially stable nuclei as the fusion processes succeed:

Ten most common elements in the Milky Way Galaxy estimated spectroscopically[1]

$$\begin{array}{|c|c|c|} \hline \text{Z} & \text{Element} & \text{Mass fraction (ppm)} \\ \hline 1 & \text{Hydrogen} & 739,000 \\ 2 & \text{Helium} & 240,000\\ 8 & \text{Oxygen} & 10,400\\ 6 & \text{Carbon} & 4,600\\ 10 & \text{Neon} & 1,340\\ 26 & \text{Iron} & 1,090\\ 7 & \text{Nitrogen} & 960\\ 14 & \text{Silicon} & 650\\ 12 & \text{Magnesium} & 580\\ 16 & \text{Sulfur} & 440\\ \hline \end{array}$$

These are also most common in the Solar System, accounting for 99.9% of the total (the ppm numbers quoted add up to 999,060). Of these ten, helium and neon are chemically almost inert; and magnesium, silicon and iron do not form fluid or volatile compounds under typical conditions in space and thus do not lend themselves to reactions that form complex "organic" molecules (where they do appear, it's only as very minor constituents in the presence of plenty of the other, more volatile-compound forming elements). This leaves hydrogen, carbon, nitrogen, oxygen and sulfur, the last of which also has some importance in proteins. Thus, plausibly, life found what was there and adapted the chemistry of those elements for its purposes.

Cited reference:

1. Croswell, Ken (February 1996). Alchemy of the Heavens. Anchor. ISBN 0-385-47214-5. Archived from the original on 2011-05-13.

  • 14
    $\begingroup$ The only problem with this view is that the most common elements in the universe are not the most common on earth or, more relevantly, the early earth. If "common on the earth's surface" was the criterion, we'd be based on silicon. $\endgroup$ – matt_black Jan 23 at 22:33
  • $\begingroup$ As mentioned, the elements must be available for reaction under conditions where life can evolve. That's where the "big five" (I include sulfur) win, at least for life as we know it. Might silicon become available and silicones come alive at higher temperatures? That is another question. $\endgroup$ – Oscar Lanzi Jan 23 at 23:04
  • 3
    $\begingroup$ Availability is part of the explanation, perhaps, but it is deeply insufficient to explain why current life uses C, H, N and O. Also, relatively complex silicon chemicals (clay minerals) are widely available available on the earth's surface and one theory of the origin of self-replication starts with clay minerals and only has other compounds taking over later as they can do a better job and can form more complex structures. $\endgroup$ – matt_black Jan 23 at 23:39
  • $\begingroup$ @matt_black Silicon doesn't readily dissolve in water, though. It wouldn't be available at the same conditions as early life (that is, excluding the clay minerals which are an enticing possible mechanism of the first replicators). Modern plants (especially grasses) do use silicon, but it's exactly because of how awful it is - the hard grains of silicon dioxide take great toll on grazers. Availability is a big part of the explanation, as are potential alternate biochemistries (obviously N has a huge role in our biochemistry). It doesn't explain everything, of course (e.g. iron). $\endgroup$ – Luaan Jan 24 at 7:26
  • $\begingroup$ Iron can be explained in context: in the presence of oxygen and water, it will form iron oxides. Those are ionic in nature, and relatively simple as a result. Solubility is limited, as is bioavailability, but plants and animals do use iron for oxygen-related chemistry. $\endgroup$ – MSalters Jan 24 at 13:11

Dusting off my geochemist hat here - In addition to elemental abundance, the matter of temperature & pressure is also key. The "solid" planet that we call home is undergoing solution chemistry on a vast scale, with most minerals differentiated mainly by local T/P conditions, rather than chemical composition ($\ce{SiO_4}$-based minerals being most common by far in the crust and mantle).

The narrow range of temperature & pressure conditions on Earth's surface at the cosmically "current" moment just happens to intersect with the T/P zone in which a few crust-abundant elements are able to form stable, reproducible, but also potentially-variable compounds. In our case, that means the protein-in-water "soup" that forms the basis for all life on Earth.

But it's not unreasonable to posit that a different "sweet spot" of solution chemistry and geochemical/astrophysical conditions could have occurred (or might occur elsewhere/elsewhen).

  • $\begingroup$ It might not sound unreasonable to posit alternative chemistries of life using different elements, but it is unreasonable given what we know about chemistry. The nearest anyone has come to positing a coherent theory of self-reproducing systems not based on carbon is the Cairns-Smith idea that replication started with complex silicates. But even he realised that, to get beyond a certain level of complexity, you have to switch to C H N O (and a few others) as the components. $\endgroup$ – matt_black Jan 23 at 23:50
  • $\begingroup$ It's a bit curious to me why clay didn't break through. Silicon does not have to bond with oxygen to make extended structures; we know of such structures in $\ce{SiC, Si3N4, SiS2}$. Alas, or not, the full possibilities of bonding did not materialize in clays at least on Earth. $\endgroup$ – Oscar Lanzi Jan 24 at 0:23
  • 1
    $\begingroup$ @MandisaW Science fiction writers are free to speculate that complex systems can emerge under very different physical conditions to earth. But none of them have produced convincing examples that don't violate what we know about chemistry. Hence the widely held view that life can only exist in a very narrow range of conditions. It isn't just us being parochial. $\endgroup$ – matt_black Jan 24 at 12:14
  • 2
    $\begingroup$ @MandisaW Yes, complete proof of a negative is hard. But the onus is on the speculators to demonstrate that complex alternative chemistry is possible in radically different environment. Just claiming ignorance is a very weak argument and largely untrue since we do, for example, know a lot about geological chemistry which involves extreme heat and pressure. $\endgroup$ – matt_black Jan 24 at 15:55
  • 1
    $\begingroup$ @Liam Clink - Nope, en.m.wikipedia.org/wiki/Silicon%E2%80%93oxygen_tetrahedron. It's an anionic tetrahedral structure that forms all sorts of linked superstructures, depending on the cation(s) involved, and temperature/pressure conditions (as well as the presence of water, in some notable cases). $\endgroup$ – MandisaW Jan 26 at 23:07

According to the RNA-hypothesis, life did not start with DNA and proteins (separate molecules for catalysis and storing sequence information) but RNA instead (one molecule to store sequence information, act as polymerase and catalyze other reactions).

While that hypothesis is realistic in terms of the awesome properties of RNA, it requires a lot of complexity and five different elements: C, H, N, O, P. There is a hypothesis of warm little ponds (WLPs) being fed nucleobases from meteorites and polymerizing as the years go by: https://www.pnas.org/content/114/43/11327.

The classic hypothesis of atmospheric chemistry creating proteins (Miller-Urey) has fallen out of favor because apparently the atmosphere did not contain much ammonia in most models (see e.g. Trainer (2013)).

Of course, I am going backwards in discussing these hypotheses, assuming that we need RNA. The OP was going the other way - what chemistry is possible with what was most abundant or most likely to make interesting life-supporting polymers.

Both H2O and C have something "best" in their belt. I guess N must have some stuff of highest quality, too, but I have no clue what it is.

In terms of nucleic acid chemistry, nitrogen plays the role of hydrogen bond acceptor and donor. The donor role is a unique one - there is no good way to position a hydrogen on oxygen in a certain direction, the hydroxyl group almost always rotates. In the other hand, imine functional groups (or amides or heterocycles with nitrogen) have the hydrogen pointing in a fixed direction.

Examples are Watson-Crick base pairs in nucleic acids and main chain hydrogen bonds in proteins. In both cases, all hydrogen bond donors are N-H rather than O-H.



  • $\begingroup$ That last added part is very interesting. Do you know why that imine can get the hydroxyl group into order, something apparently no element other than N can do? And what's so important of a non-rotating -OH group? $\endgroup$ – longtry Jan 25 at 5:39
  • $\begingroup$ @longtry I added some examples to illustrate what I meant by non-rotating hydrogen bond donor. Also, -OH always comes with an acceptor while the nitrogen functional groups can be pure donor, making for greater specificity. $\endgroup$ – Karsten Theis Jan 25 at 15:23

Because life uses whatever it finds useful. And organic compounds with nitrogen can be easily synthesised, metabolized and may have many useful properties. And also, various stable organic compounds with nitrogen will just come into into existence in abiogenic conditions.

Why though? Well, nitrogen-carbon bonds are fairly strong and unlikely to decompose under temperature or through reaction with other compounds. Also, nitrogen can bind to one, two or three other atoms, so nitrogen allows the compounds for more complex structure. Nitrogen atoms also can form organic aromatic compounds which are very stable.

So what about other elements? Metals (and semimetals for all pratical purposes) and noble gases go straight out of the window as they can't form stable bonds with carbon. We're mostly left with nonmetals.

The fluorides can for fairly stable bonds with carbon, but are much more likely to exist in form of anions and it's hard to oxidize those anions so they can bond with carbon. Also, they can only bond with one carbon so they can't allow for more complex compound.

The sulfur can actually form fairly stable bonds with carbon and we indeed see that sulfur is fairly common in live organisms. Selenium is similiar in that regard. Carbon-phosporus bond is not very stable and phosphorus is mostly found in organisms in form of inorganic anions. Most other nonmetals are either too rare or simply don't form stable bonds with carbon.

  • $\begingroup$ Thanks. So why did N get the edge over S & Se? I feel it has something to do with Permeakra's comment right under my question, but his/her is quite advanced and do not go into the details as your answer. $\endgroup$ – longtry Jan 24 at 16:49
  • $\begingroup$ N did notventirely get the edge over S as the latter is also in some amino acids. Among the three elements in the previous comment, abundances are given by N > S > Se. $\endgroup$ – Oscar Lanzi Jan 24 at 16:53
  • $\begingroup$ Abudance is not so important here. More so the fact that C-S bond is just much less stable than C-N bond. Silimiliary with Se, but in this cas the abudance is actually visibly much lower than S. $\endgroup$ – Jan Rzymkowski Jan 25 at 0:58

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