Henry Rzepa's blog

The proposed identification of molecules with potential metal to carbon quadruple bonds, in which the metal exhibits trigonal bipyramidal coordination rather than the tetrahedral modes which have been proposed in the literature[1],[2],[3] leads on to asking whether simple trigonal coordination at the metal can also sustain this theme?

The rational for doing this is the observation for the trigonal bipyramidal molecules that repulsions between a non-bonding occupied d-orbital on the metal and one of the two putative metal to carbon σ-bonds resulted in the two electrons localising into a lone pair on carbon. By removing the electrons in this metal d-orbital or by increasing its size, the C σ-lone pair was encouraged to abandon some of its lone pair character and participate in a quadruple bond.

Another way of achieving this result is explored here, with the molecule shown above. Two of the trigonal carbon ligands are pinned back by the ring to reduce any potential repulsions. As before, this complex constitutes a filled 18-valence shell metal. The calculated orbitals (ωB97XD/Def2-SVPD, FAIR DOI: 10.14469/hpc/8206) are shown below.

M=Co, r = 1.493Å.
NBO 26, π bond	NBO 24, non-bonding d-orbital
NBO 23 σ bond, 0.02 au	NBO 23 σ bond, 0.01 au
NBO 22 π bond	NBO 14 σ bond

Despite the appearance of a bond between Co and C along the C⩸Co axis in the representations above (inserted off its own bat by the JSmol program), no such bond exists in the NBO list. No precedent for this kind of structure appears in the crystal structure database. As before, the NBO 23 σ bond at low isosurface thresholds expands to add a second layer along the Co⩸C axis, in which the nodal surface exists between the two layers rather than along the bond itself. It therefore constitutes a bonding orbital.

These quadruple bond motifs involving carbon are certainly starting to emerge in unexpected places and I do wonder how many more variations on this theme will be identified.

References

1. A.J. Kalita, S.S. Rohman, C. Kashyap, S.S. Ullah, and A.K. Guha, "Transition metal carbon quadruple bond: viability through single electron transmutation", Physical Chemistry Chemical Physics, vol. 22, pp. 24178-24180, 2020. https://doi.org/10.1039/d0cp03436c
2. A.J. Kalita, S.S. Rohman, C. Kashyap, S.S. Ullah, I. Baruah, L.J. Mazumder, P.P. Sahu, and A.K. Guha, "Is a transition metal–silicon quadruple bond viable?", Physical Chemistry Chemical Physics, vol. 23, pp. 9660-9662, 2021. https://doi.org/10.1039/d1cp00598g
3. L.F. Cheung, T. Chen, G.S. Kocheril, W. Chen, J. Czekner, and L. Wang, "Observation of Four-Fold Boron–Metal Bonds in RhB(BO^–) and RhB", The Journal of Physical Chemistry Letters, vol. 11, pp. 659-663, 2020. https://doi.org/10.1021/acs.jpclett.9b03484

Introductory chemistry will tell us that a triple bond between say two carbon atoms comprises just one bond of σ-axial symmetry and two of π-symmetry. Increasingly mentioned nowadays is the possibility of a quadruple bond between carbon and either itself or a transition metal, as discussed in the previous post. Such a bond comprises TWO bonds of σ-axial symmetry. Since most people are unfamiliar with such double bonds and in particular with how that second σ-bond sits with the first, I thought it would be interesting to show such an orbital. This one is a localised orbital 41, selected from the previous post for the molecule (PH₃)₂(CN)₂Mo⩸C.

NBO 41, threshold 0.040 au	NBO 41, threshold 0.018
NBO 41, threshold 0.016	NBO 41, threshold 0.014
NBO 41, threshold 0.012	NBO 41, threshold 0.007

The above shows how the orbital changes with the isosurface threshold. At high values, it looks very similar to the normal σ-bond but as the threshold gradually decreases, a second “sheath” starts to surround the inner orbital until the latter is entirely enclosed. This orbital has a node not so much along the bond itself, but between the inner and outer layers of the bond, which is how the two σ-bonds are differentiated. This effect was first noted in 2016 in terms of the compound CH₃F^2-, in which an expanded carbon valence shell creates a second σ-bond.

Certainly not a representation that has ever appeared in a text book I think! But perhaps one that chemists may have increasingly to become familiar with.

Appendix Here are some superimposed orbitals to facilitate comparisons. Firstly orbital 41 (the higher energy σ-orbital) with orbital 22 (the lower energy σ-orbital). The first has yellow/green for the two phases, the second has red/blue.

Next, σ-orbital 22 (yellow/green) with orbital 42 (red/blue) surrounding it, revealing the avoided overlaps (Pauli repulsions) between the two by virtue of having orbital 42 unoccupied.

Next, σ-orbital 41 (yellow/green) with orbital 42 (red/blue) surrounding it, revealing the reduced overlap between these two.

Appendix 2 A “pure” form of the double-layered σ-bond can be seen with the diatomic molecule Ti₂, contoured at 0.0225 au. The red phase is about to join in the middle.

The electron density from this orbital is shown below and shows clearly the two layers of density comprising the σ-bond, with the outer layer at this isosurface value (0.00052 au) about to join up in the middle to complete the outer sheath. I have left it unjoined so that you can see “inside layer”, since translucency does not always get the message across.

Deltamethin is a pyrethroid insecticide for control of malaria which has been used for a little while. Perhaps inevitably, mosquitoes are developing resistance to it. So what could be done about countering this? Well, perhaps surprisingly, form a polymorph![1] These crystal structure isomers are often highly undesirable; thus Ritonavir, which changed its polymorphic form during manufacture to become far less active (due it has to be said to insolubility). Now a polymorph of Deltamethin has been discovered, which when applied as a powder, increases its effectiveness more than 10 times against Anopheles mosquitoes and provides a potentially new affordable malaria control solution for countries that are loosing protection.

You can view this new polymorph at DOI: 10.5517/ccdc.csd.cc24mrxg and 10.5517/ccdc.csd.cc24mryh. The two structures differ in the torsion angle around the O-C(CN) bond, resulting in differing anomeric interactions between the ester oxygen and the C-CN group.

The anomeric effect can be quantified by the dihedral angle between a donor oxygen lone pair and the accepting (in this case C-CN) bond. The position of the lone pair itself can be localised using the ELF (electron localisation function) method. For one polymorph, this has a torsion angle of 127° and O-C and C-CN lengths of 1.455 and 1.480Å. The other form has an almost orthogonal (i.e. non-interacting) torsion of 79° with lengths of 1.461 and 1.477Å. The first form exhibits some electron pair donation into the C-O bond, shortening it slightly, accompanied by transfer of C-CN bond electrons onto the N, weakening the C-N bond. A very similar effect was noted for a herbicide[2] which resulted in very differing stabilities for two stereoisomers in water.

Just such aqueous hydrolytic stability was indeed reported for Deltamethrin[3] and its other 7 stereoisomers in water (3 stereocentres, 2³ = 8). Persistence of any chemical applied to the soil is a very important aspect of its properties; recollect that a now banned insecticide, DDT, suffered from very long soil persistence. The article[3] looks at what happens to Deltamethrin (labelled compound 1) in water in the dark. As it goes, a new isomer labelled 2′ appears and then too decays, all in a period of ~5 days. This new isomer was found to be inactive as an insecticide.

So what is 2′? Well here is a table of all 8 stereoisomers, where 1 is labelled as α-S,1R cis and 2′ is labelled as α-S,1S cis. Here we have only one of the three stereocentres labelled using Cahn-Ingold-Prelog (CIP) notation, the 1-R/S carbon (the convention is to use italics for CIP). The cis designation implies that the 3-position is also S and the cyano carbon remains S. This cis label however conflicts with the abstract to the article, which indicates that the dark water isomerisation is subject to cis/trans isomerisation. So we have a water-induced isomerisation which inverts the configuration of 1R,3R in molecule 1 to apparently 1S,3S in molecule 2′?  

Changes in stereochemistry in Deltamethrin are likely to be the result of forming planar enols and so one might judge the hydrolytic stability by the ease of forming such an enol. Just such a process was the topic of a post on the undesired epimerisation of thalidomide in water. Time for calculations (FAIR data doi: 10.14469/hpc/8020 ) at the B3LYP+GD3+BJ/Def2-TZVPP/SCRF=water computational level. Firstly, the free energy of the enol form on the cyclopropane by relocation of the proton at C-1 is 32.0 kcal/mol higher. However, forming the enol by relocating the proton on the carbon bearing the CN group to the nitrogen is only 20.0 kcal/mol higher in energy. With a small further barrier for the TS for proton removal expected, this latter enol is within an accessible energy range for room temperature reaction, whereas the previous enol is not. So the hypothesis is that 2′ is actually the stereoisomer labelled 2 in Table 1, ie R,R,R and not S,S,S. The free energy of this R,R,R diastereoisomer is calculated 1.1 kcal/mol higher than 1 itself, which would represent about 13% of this isomer at equilibrium.

I have shown here how uncertainty caused by how to reconcile the stereochemistry of eg  Figure 4 and Table 1[4] in modern terms can perhaps be lessened by performing “reality check” calculations on possible multiple interpretations to reduce them to the most probable. We also find that a relatively old and much used bioactive compound can have surprises lurking around the corner, in this case by a simple recrystallisation that results in a new form being discovered without having to do any other chemistry.  Still, how much longer will pyrethroid insecticides be used?

References

1. J. Yang, B. Erriah, C.T. Hu, E. Reiter, X. Zhu, V. López-Mejías, I.P. Carmona-Sepúlveda, M.D. Ward, and B. Kahr, "A deltamethrin crystal polymorph for more effective malaria control", Proceedings of the National Academy of Sciences, vol. 117, pp. 26633-26638, 2020. https://doi.org/10.1073/pnas.2013390117
2. P. Camilleri, D. Munro, K. Weaver, D.J. Williams, H.S. Rzepa, and A.M.Z. Slawin, "Isoxazolinyldioxepins. Part 1. Structure–reactivity studies of the hydrolysis of oxazolinyldioxepin derivatives", J. Chem. Soc., Perkin Trans. 2, pp. 1265-1269, 1989. https://doi.org/10.1039/p29890001265
3. R.J. Maguire, "Chemical and photochemical isomerization of deltamethrin", Journal of Agricultural and Food Chemistry, vol. 38, pp. 1613-1617, 1990. https://doi.org/10.1021/jf00097a039

For obvious reasons, anti-viral molecules are very much in the news at the moment. Thus Derek Lowe highlights Molnupiravir which is shown as a hydroxylamine, the representation originating from the Wikipedia page on the molecule.

I like stereocentres more clearly identified using eg R/S notation and so I went to another source of information, SciFinder, which represents the molecule in a different way. There you get the stereocentres unambiguously identified for you, but the hydroxylamine is now replaced by an oxime! 

The Reaxys database renders it differently again as below (note the different rotamer for the hydroxylamine):

Are they all talking about the same molecule? Well yes, since the hydroxylamine and the oxime are related by tautomerism, but it takes a bit of effort to fully reconcile these three representations with each other. So the next question is does it matter which tautomer is selected to represent a molecule? Since they differ by proton transfers between acidic atoms (N,O) the presumption is that the equilibrium between the tautomers is fast and so the predominant species is determined by the position of the equilibrium. The tautomer matters in another sense. This molecule clearly interacts with DNA, and very probably by paired hydrogen bonding. It is this hydrogen bonding that was crucial in helping Watson and Crick to postulate the first successful model of DNA itself,[1] famously enabled by them using the correct tautomeric form of the component DNA bases! So the tautomers do matter!

Time for some calculations (B3LYP+GD3BJ/Def2-TZVPP/SCRF=water, using integral=(acc2e=14,grid=superfinegrid). The FAIR DOI for the collection is 10.14469/hpc/7990

In the diagram above, the non-aromatic valence representation is shown at the top, followed by a reorganisation of the electrons into an aromatic form below. These aromatic forms are all ionic with charge separation and so the question here arises: does the aromatic stabilisation energy outweigh the destabilisation caused by charge separation? 

The calculated free energies (ΔG₂₉₈) show that the oxime is 3.2 kcal/mol more stable than the next most stable, the hydroxylamine tautomer (calculated for a water continuum solvation model).The third tautomer is not far behind, being a nitrone! A 3D model is shown below illustrating the internal hydrogen bonds (I hope it is clear why I wanted the four stereocentres unambiguously identified, and why the simple perspective diagram shown on the Wikipedia page is not definitive). This model was obtained after also experimenting with hydroxyl rotamers to ensure the lowest energy was obtained.

Click to load 3D model

So how “aromatic” are the two lowest energy species? The oxime has two sets of charge separation, whilst the hydroxylamine only one. One measure of aromaticity is the NICS magnetic index. For the oxime it is -0.1ppm, very clearly non-aromatic. Separating two sets of charges clearly is not compensated by aromatic stabilisation. The next most stable hydroxylamine has NICS -1.8ppm, which shows slightly more signs of aromaticity (benzene on this scale is ~ -10 ppm), perhaps because it only separates one pair of charges. 

What about crystal structures? NETGEY is a model which removes the sugar unit and with an N-Me replacing the NH (thus preventing tautomerism and locking the molecule into the oxime form). MHCYTC protonates the hydroxylamine on the imine nitrogen OR the oxime on its imine nitrogen thus rendering the two tautomers identical.

NETGEY	MHCYTC

The crystal structure of Molnupiravir itself has not been reported, so there is no definitive answer to the most stable tautomer in the solid state. But the calculation above does suggest that it is the oxime (and hence SciFinder’s representation) that is the probable dominant form in aqueous solutions. If one is trying to build a model to show how this small molecule interacts with DNA, this might be useful information (in the same way that picking the correct tautomer of the original DNA bases worked for Watson and Crick!). 

References

1. J.D. WATSON, and F.H.C. CRICK, "Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic Acid", Nature, vol. 171, pp. 737-738, 1953. https://doi.org/10.1038/171737a0

The last post addressed the concept of “steric clashes” in a pericyclic reaction transition state as an extension of the time honoured practice of building molecular models to analyse reaction outcomes. A modern computer generated model might express this in terms of a NCI (non-covalent-interaction) surface. A few posts ago, I had looked at some “molecules of the year” for 2020, one of which was a “figure-eight” twisted dodecaporphyrin in which an aspect of the reported[1] geometry had struck me as potentially lacking features due to the so-called non-covalent dispersion or van der Waals attractions. So I am revisiting here by adding the NCI surface for this molecule and one other.

The molecule in question has 720 atoms and can be regarded as a [162]-annulene (4n+2, n=40) with a linking number Lk =2π.[2] The NCI surface is ideally computed from a “self-consistent-field or SCF density” and so in this instance I used the PM7 SCF-density, which is derived from the valence shell only and does not include the core shells. That hardly matters since the non-covalent NCI surface does not use the core shells!

Click for 3D model

You can see from the above that the porphyrin-stacking region has a very dense NCI green surface (arrow), indicating a lot of stabilisation is originating there; something lacking in the original proposed structure. There are lots of other features and so I do encourage you to explore the 3D model.

The second (hypothetical) molecule is a simpler CH-based [144]-annulene,[3] comprising a twisted coil of 144 CH= units with a linking number Lk = 18π (the largest such ever proposed for a molecule!). The SCF-NCI surface (derived from an ωB97XD/6-31G(d,p) calculation) is contiguous all the way around the circuit and must be the ultimate π-π stacked molecule!

144-annulene. Click for 3D

153-annulene. Click for 3D

I should end with a brief tutorial on how to generate these surfaces. You need a density matrix (e.g. DOI: 10.14469/ch/16967). In programs such as Gaussian 16, this can be obtained from the checkpoint file, which contains it. A progam called cubgen is used by (e.g.) Gaussian to create a 3D cube of electron density values (as well as other interesting properties). To get good resolution (~ 0.044Å) the file will be between 500 – 800 Mbyte in size. If a resolution of ~ 0.088Å is used it will be eight times smaller. For cube files less than ~105 Mbyte in size, you can use this Web-based tool (DOI: 10.14469/hpc/7864) to get the NCI surface. For the larger files you will need the Jmol application which can sustain files up to ~1 Gbyte (or larger, but I have not tested) and where you will run the following script: 

load density.cub;isosurface parameters [0.5 1 0.0005 0.05 0.95 1.00] NCI "";color isosurface "bgyor" range -0.04 0.04;write density.xyz;write density.jvxl;

(where this script assumes that the file density.cub file is in the same folder as the Jmol.java application).

Postscript: The NCI analysis is based on computing the total density of the molecule. Close inspection of the top molecule as computed using the semi-empirical method PM7 reveals some interesting features extending beyond the C-H bonds. Analysis of this reveals it to be an artefact of the computed density, itself traced back to differences in how overlaps are handled in computing the density for this particular method. This error is not present for the MNDO semi-empirical method. When evaluated using MNDO, but at the geometry computed by PM7, these artefacts are removed. The NCI feature in the π-π stacking shown above however remains and hence is not an artefact.

References

1. M. Rickhaus, M. Jirasek, L. Tejerina, H. Gotfredsen, M.D. Peeks, R. Haver, H. Jiang, T.D.W. Claridge, and H.L. Anderson, "Global aromaticity at the nanoscale", Nature Chemistry, vol. 12, pp. 236-241, 2020. https://doi.org/10.1038/s41557-019-0398-3
2. S.M. Rappaport, and H.S. Rzepa, "Intrinsically Chiral Aromaticity. Rules Incorporating Linking Number, Twist, and Writhe for Higher-Twist Möbius Annulenes", Journal of the American Chemical Society, vol. 130, pp. 7613-7619, 2008. https://doi.org/10.1021/ja710438j
3. R.J.F. Berger, "Prediction of a Cyclic Helical Oligoacetylene Showing Anapolar Ring Currents in the Magnetic Field", Zeitschrift für Naturforschung B, vol. 67, pp. 1127-1131, 2012. https://doi.org/10.5560/znb.2012-0189