Imperial College London

Henry Rzepa

Chemistry gets complex very rapidly. Consider the formula CH~3~NO as the
topic for a tutorial in introductory chemistry. I challenge my group (of
about 8 students) to draw as many different molecules as they can using
exactly those atoms. I imply that perhaps each of them might find a
different structure; this normally brings disbelieving expressions to
their faces.

::: {#attachment_3113 .wp-caption .aligncenter style="width: 173px"}
![](http://www.ch.ic.ac.uk/rzepa/blog/wp-content/uploads/2010/12/ch3noa.jpg "ch3noa"){.size-full
.wp-image-3113 loading="lazy" decoding="async"
aria-describedby="caption-attachment-3113"
onclick="jmolInitialize('../Jmol/');jmolSetAppletColor('white');jmolApplet([700,494],'background image \"wp-content/uploads/2010/12/CH3NO.jpg\";');"
width="163" height="53"
srcset="https://www.ch.ic.ac.uk/rzepa/blog/wp-content/uploads/2010/12/ch3noa.jpg 326w, https://www.ch.ic.ac.uk/rzepa/blog/wp-content/uploads/2010/12/ch3noa-300x97.jpg 300w"
sizes="(max-width: 163px) 100vw, 163px"}

Click on image to see molecules constructed from these atoms. The list
is not comprehensive!
:::

Amongst the useful concepts that can be introduced are:

1.  How to determine how many *double bond equivalents* (or *degrees of
    unsaturation*) are implied by the formula.
    1.  Students spot one *dbe* in the above formula, but can take a
        little longer to notice that it can reside in a ring.
    2.  Few (and I count tutors in this) will add sub-valent atoms
        (here, the possibility of a *carbene* or a *nitrene*) to the
        list.
2.  What is meant by "different"? This can be reduced to the equations:
    Ln k/T = 23.76 -- ΔG^‡^/RT; t~1/2~ = (Ln 2)/k, where t~1/2~ is the
    half life (in seconds) of any species constrained by a free energy
    barrier of ΔG^‡^. A nice illustration of this equation is to be
    found on [Jan Jensen's
    blog](http://molecularmodelingbasics.blogspot.com/2010/07/amide-hydrolysis.html){target="_blank"}
    (and an worthwhile calculation would be to find the barrier required
    to achieve a half life based on the age of the universe). This can
    be boiled down to three ranges.
    1.  Half lives of \~10^-15^ s, or vibrations (and this includes
        transition states themselves). Arguably, *resonance isomers*,
        which involve the (nominal) motions of electrons and not nuclei,
        fall into this class as well.
    2.  Half lives of \< 10^1^ s, which would include most
        *conformational isomers* (excepting atropisomers) and highly
        *unstable isomers*, and which cannot be bottled and labelled as
        such.
    3.  Compounds with half lives \> 10^2^ s, up to of course the age of
        the universe. This would include *configurational isomers* (and
        if the students are up to it, you can ask them to identify any
        compounds constructed above which can exhibit *optical
        isomerism*).
3.  One might be inclined to (approximately) use arrows to indicate the
    timescales above. Thus electronic resonance is represented by
    double-headed arrow, conformational and E/Z isomers by an
    equilibrium arrow, and a single headed arrow indicating a reaction
    (which may in fact have a very low barrier) connecting two isomers.
4.  Its normally now time to count the electrons. This includes the
    "invisible ones", the lone pairs, and also the occasion to introduce
    the valence shell octet.
5.  Putting the appropriate charges onto any atoms which require them is
    always fun (the dative bond is avoided). The blue structure revealed
    in the click above is an extreme interpretation of this! Gernot
    Frenking has pioneered the class of compound he calls *carbones*.
    For his latest article on the theme, see DOI:
    [10.1002/anie.201002773](http://dx.doi.org/10.1002/anie.201002773){target="_blank"}.
    The green compound would belong to this class, if it did not fall
    apart (probably with no barrier) to something which is not actually
    one molecule, but two (separable) molecules (purple). This brings us
    into what a molecule actually is. Could it be two molecules
    unconected by any bonds, but nevertheless also inseparable (such as
    catenanes, rotaxanes, and many other entwined systems)? Two
    molecules can also interact weakly, which is not normally referred
    to as bonds. In this case, the two molecules would be bound by
    a *hydrogen bond*.
6.  Quite a number of the isomers can be also called *tautomers*. This
    involve the movement of one type of atom in particular, the hydrogen
    (or proton). In terms of lifetime, they would fall into **class 2**
    above (although if one takes extreme care to remove all traces of
    acids or bases, particularly from the surface of any glass
    container, one can extend the lifetimes quite considerably).
7.  The *peptide bond* is included in the isomers, and its ionic
    resonance formulation, which can lead the discussion to the
    molecular basis of life and how
    [finely-tuned](https://www.ch.ic.ac.uk/rzepa/blog/?p=1174){target="_blank"}
    this bond in fact is.
8.  One might speculate about what the most stable of all the isomers
    might be, and how many are indeed *bottleable*. One might introduce
    quantum mechanics as nowadays a very reliable way of estimating this
    (and whilst you are at it, introduce free energies, entropies etc).
    For example, which of the two red *geometrical isomers* is the more
    stable, and why? What is the best resonance representation (*i.e.*
    where does one put the charges? On this specific point, a
    CCSD/6-311G(d,p) [ELF
    calculation](http://hdl.handle.net/10042/to-6346){target="_blank"}
    does come up with a very definitive answer of on the nitrogen rather
    than the oxygen).
9.  This might be followed up by introducing *arrow pushing* as a means
    of interconverting two isomers, and with one of the pair of isomers,
    one can introduce p*ericyclic selection rules*, t*ransition state
    aromaticity* and other advanced stereochemical concepts.
10. Now we are well into to *stereoelectronics*. One can introduce
    *anomeric effects* *via* the *NBO technique*. Thus in the red
    compounds, there is an interesting interaction between the lone pair
    on carbon and the *anti* N-H bond (but, spectacularly, not the *syn*
    N-H bond). There is another particularly strong one between the
    oxygen lone pair and the C-N bond.

I dare say I have only picked at the surface, but covering the above
should be [enough for one
tutorial](http://www.ch.imperial.ac.uk/rzepa/blog/wp-content/uploads/2010/12/CH3NO.svg) I
should imagine 🙂

------------------------------------------------------------------------

PS For the (calculated) relative energies of some of these isomers, see
DOI: [10.1021/jo010671v](http://dx.doi.org/10.1021/jo010671v){target="_blank"}

::: {.wp_orcid_field}
[https://orcid.org/0000-0002-8635-8390](https://orcid.org/0000-0002-8635-8390){target="_blank"
rel="author"}
:::


Chemistry gets complex very rapidly. Consider the formula CH
<sub>
 3
</sub>
NO as the topic for a tutorial in introductory chemistry. I challenge my group (of about 8 students) to draw as many different molecules as they can using exactly those atoms. I imply that perhaps each of them might find a different structure; this normally brings disbelieving expressions to their faces.

Messaggi di Rogue Scholar

Combichem: an introductory example of the complexity of chemistry