The Mechanism of the Über Reaction

• Silvio Cornuto: Valeria Messalina Institute, UC Milan, Italy
• Alexander Thwacker: Organischmetallischchemieabteilung, Univeristät von Wankendorf, Germany.
• Henri de Pampelmousse: École de Chimie Banale, Gauloises Université, Brest, France.
• Aap Tsagay: Kuchu Chethay University, Bajoling, Bhutan.
• Günther Schlonk: Department of Pyrofrolics and Inorganometallics, University of West Failure.

Abstract:

We spent way too much time on this.  

Specific: The Over Reaction is the most fiendishly complicated transformation in modern synthetic chemistry. It features nine discrete  catalysts, visible light and electrochemical oxidation, which cooperate to convert obstannyl sulphosphoxides, aryl moronic esters,  cumulenes, homopropargylic ethers and cyanodiazetidines into elseviammonium dismylatesin one step. The mechanism of this  reaction has perplexed chemists since its discovery reaction in 2005. Now, for the first time, we propose a catalytic cycle for the  Over Reaction.

The history of science is riddled with happy accidents.

Naturally, such occurrences are referred to as “serendipitous discoveries” rather than “lucky fuck-ups,” because it sounds  more intelligent. Chemistry has seen its fair share of these  stories play out: the discovery of iodine, mauve dye and  penicillin are some of the more famous examples. Though our  half-arsed search of the illiterature was inconclusive, we  assume that the discovery of pyrrole was also accidental, as no  sane individual would set out to work with such an obnoxious  molecule.¹ The Über–Furious–Bob–Paul–(and so on and so on)- Wang–Fukovski reaction is another such child of serendipity,  more commonly referred to as “The Over Reaction” for the sake  of brevity.² 

In June 2005, Aggravated Professor Frank Furious was  attempting to perform the hydrogenolysis of benzyl triflate 1.  This molecule features both an obstannane group and a  sulphosphoxide moiety, which while reactive, are typically  stable towards hydrogenative conditions.³ Furious was  observing a complete lack of reactivity, despite a literature procedure claiming to obtain 85% conversion in 15 minutes.⁴

Furious cranked the temperature and dialled up the H2 pressure, let the reaction run overnight and returned the next morning. Nothing. He consulted Dr Grundle’s paper, noticed  that he’d claimed a 98% isolated yield, and snapped. One of  Furious’ students recounts the events that followed:

‘Frank always had a case of nominative determinism, but  this was next-level. He thundered “0.01% catalyst loading my  arse” and upended a box of metal complexes and a fistful of  sample vials into the flask. Then he threw a 9 V battery into the  reaction mixture, squawking something about  “electrocatalysis,” and connected the vessel directly to the H2 cylinder without a regulator, while humming the tune to Under Pressure. AsI was running for shelter, I saw him grabbing the UV lamp off the TLC box and strapping it to his flask: “visible light is  piss-weak anyway, it can’t even give you cancer.” The last thing  I remember was watching Furious flip a lab bench, shout “cock blaster” at his reaction and storm out the door. We didn’t see  him for three weeks, and we were too scared of that reaction to  go near it.’⁵

Frank returned from his impromptu stress-leave and  disassembled his creation, which had miraculously remained  intact. On discovering that none of his students had been in the  lab during his absence, his ire was stirred to life once more, and  he instructed Klaus Über to work it up as punishment. Über began this process by smashing the flask and recovering a  charred black solid, which he placed in a ball mill for a week. The  residue was extracted with DMSO and subjected to a medically  inadvisable amount of chromatography. From the ashes of  Furious Frank’s failure, Über isolated 17 mg of molecule 3,  which he dubbed Phoenoxide A (Figure 2).⁶

Phoenoxide A was fully characterised, and its structure was  assigned by X-Ray crystallomancy. It features the fused,  polycyclic core of a class of molecules known as “Elseviamines,”  so called for the tremendous difficulty encountered in accessing them.⁷ This difficulty is primarily associated with the  pentavalent carbon centre at C5. These Elseviamines are a  subset of the Pandammonium Alkaloids, a large and chaotic  family of natural products.

The impact of this finding was obvious: Furious and Über  had discovered a one-pot synthesis of Elseviamines. But there  was a problem: neither of them could remember how. Frank’s  vision was red-shifted at the time, and everyone else was  running for cover. The subsequent carnage made it impossible  to determine which catalysts Furious had added to the reaction.  It took the Furious group three years to reproduce their original reaction, which they eventually managed in collaboration with  nine different labs in 16 different countries (Figure 3A).^8,9 The  result was the longest named reaction in the IOUPAC Brown  Book, which is catalytic in scandium, uranium, molybdenum, iridium, rhenium, copper, ruthenium, palladium, visible light,  invisible light, and electrical current (Figure 3B). It is  stoichiometric in tin, molybdenum and suffering, and is one of  the least atom-economical reactions known.  

Five years of painstaking mechanistic investigation by our  group and our collaborators has resulted in the first tentative  catalytic cycle for the Over Reaction (Figure 3D). Full  experimental conditions are detailed in the electronic unsupported information.¹⁰

The Reaction 

The Over Reaction is the coupling of an obstannyl  sulphosphoxide 1 with a homopropargylic ether 4, a  cyanodiazetidine 5, a cumulene 6, and an arylmoronic acid  (pinacol ester) 7. The product of this unholy union is a highly  substituted elseviamonium dismylate zwitterion.¹¹ The reaction  forms nine new bonds and five contiguous stereo centres. It is conducted under a high pressure of H₂, with a 9 V battery  functioning as both electrochemical oxidant and stir-bar. Yields  are typically between 0.5 and 3%, rating between an 8 and an  11 on the Browning Index.¹² The low yields of this  transformation reflect both the horrendous inefficiency of the  reaction itself, as well as the nightmarish workup. Most of the  catalysts and their associated intermediates are completely  incompatible, giving rise to a menagerie of side-products and  off-cycle intermediates. Under the harsh reaction conditions,  these undesirables breed and multiply until a material  resembling brown coal is produced. The product is obtained by  milling this deposit followed by Soxhlet extraction with refluxing  DMSO.

The Catalysts 

Half of the periodic table is involved in the catalytic cycle of  the reaction, but the first metal involved is scandium. This Lewis  acid is complexed with a crown-of-thorns-ether to improve its  solubility, and functions to both activate and moderate the  promiscuity of the sulfoxide moiety throughout the reaction. A  polyhydridic uranium complex (9) serves as a powerful  reductant and as a shuttle for H₂, which it uses to partially  hydrogenate the aromatic ring of the obstannyl  sulphosphoxide. The nitrogenous portion of the product is  incorporated via a nitrinidinium radical cation. These fickle and  fleeting intermediates are derived from the one-electron  oxidation of nitrenes, which are themselves highly reactive. The  only oxidant capable of such a challenging transformation is a  photo-aroused perfluoridium cation [Ir]⁺*. The catalyst  perfluoridium PF₆ enters this state upon the absorption of a  

photon of invisible light, and rips electrons off anything it can  get its LUMO’s on. The parent nitrene must be trapped as a  metal complex until this oxidation is performed, lest it become  distracted and wander off. Molybdenum complexes are capable  of stabilising them, provided that the Mo-bound ligand is itself  a rock of stability. The Focker ligand has such stability, and thus  MoFocker is a suitable nitrene transfer catalyst for this  reaction.¹³ Rhenium possesses an affinity for the pi-systems of alkynes  and serves to convert homopropargylic ether 16 into !, #- unsaturated carbene 17. Rhenimine B is a soluble rhenium  source, supplied as a single enantiomer for purely decorative purposes.¹⁴ Pendant isopropyl groups adjacent to the imines  shield the metal centre from rapacious solvent molecules,  preventing decoordination and rhenial failure. The Thwaker  Reagent [CuHL] is a mild hydride source, selective for cyclopropanes. The ring-closing olefin-metathesis step of the  catalytic cycle is sterically demanding, and requires a highly  active catalyst. Grubbs’ Other Catalyst II (patent rejected), is  one of the most efficacious. The thermodynamic incentive for  its reactivity is the ejection of its pyrrole, which is barrierless  even at 12 K.¹⁵ 

A palladium-ArsePhos complex performs the migratory  arylation of alkene 20. The ArsePhos ligand (sometimes referred  to as Y-O-YPhos) incorporates a labile, bulky arsine (a fat-arsine)  and electron-rich phosphine, separated by an osmocene unit.  It’s catalytic activity in Mitsubishi couplings is matched only by  its acute toxicity.¹⁶ The final catalytic step in this marathonic  mechanism is the photoarousal of dismyl anion 22. Acridazole is  an organic photocatalyst, known for its efficacy in such energy transfer reactions, albeit only on days with a prevailing north westerly breeze.

Density Dysfunctional Theory 

Initially, we sought to probe the mechanism of this bizarre  transformation with computational methods, because it was  cheaper than using the catalysts themselves. Using the Hartree Fück level of theory, the MLF06-XXX functional and the 2G1C LYP basis set, we were able to successfully model the chelation  of scandium by substrate 1. This interaction was found to be  favourable by 14.7 kflop/mol. However, modelling the  interaction of uranium hydride 9 with the substrate proved to  be exceedingly difficult. When the 6-31(f)^+_+ basis set was used  to model uranium, the simulation ETA was in mid-2047. By  switching to RANDL2DZ, the computation time was lowered to  three weeks, but the resultant energy was 28,000 kflop/mol  higher than expected. A compromise was reached with the  LANL2DZ-D97DEP31G-OSX10-WD40-6-3211111G(dp)^*_^^* basis  set, which took six months to tell us that the energy of 10 was  something like 50 kflop/mol. To achieve this shorter  computational time, it was necessary to model uranium as a  really big chromium atom. At the time, we intended to submit  our work to Nature, and we figured that we could send the  initial manuscript and append the calculations six months later  before it had been sent out for review.  Modelling the next mechanistic step (the radical scission fission), proved to be beyond the capacity of modern  computational methods. Our initial attempts were met with  “Error 424: spontaneous decomposition” followed by a total  system crash of all servers on the network, and legal action  taken against us by Facebook for disrupting their service.¹⁷

Kinetic Isotope Defect 

The most informative of our numerous mechanistic studies  was an isotopic labelling experiment. We exposed deuterated  obstannyl-sulphosphoxide D₂-1 to standard Over Reaction  conditions, and observed the distribution of deuteration in the  product by ^1-and-a-bitHNMR. A significant kinetic isotope defect  was observed. 85% of the original 200% was scattered across  the product in a seemingly random distribution. The remaining  115% is still missing, please contact us if you find it. These  results are evidence for rapid and promiscuous hydrogen  exchange, facilitated by at least one of the nine catalysts. 

The Mechanism 

In the beginning, there was scandium. And Frank knew that  it was good. A scandium chelate 8 is formed from the sulphosphoxide portion of 1. Uranium hydride 9 enters the  stage, and forms four agnostic interactions with the pi-system  of 9. Adduct 10 undergoes radical scission-fission (also called a  radical-dance) with nitrinidinium cation 11, while contained in a  solvent-maze. The nitrogen is quaternized, while an equivalent  of triflyl sulphobstannane 16 is expelled. Concomitantly,  uranium delivers an equivalent of H₂ to the pi-system, and  shuffles a couple of the other protons around. The result of  these processes is the partial hydrogenation and triple-ring closing cyclisation of 13 to give pandemonium cation 14.  The formation of nitrinodinium 11 begins with the  thermolysis of cyanodiazetidine 12, which generates a nitrene.  The MoFocker complex stabilises the nitrene (13), until it is  oxidised by a photoaroused perfluoridium cation [Ir]⁺*. The  reduced form of perfluoridium [Ir] is reoxidised by the 9 V  battery.

Rhenimine B serves to cleave homopropargyl ether 16 and  generate !, #-unsaturated carbene 17, via vinylidene  formation, 1,5-hydride transfer, and $-hydride reallocation. The  carbene attacks pandemonium cation 15 in a cyclopropanation.  The resultant cyclopropane 18 undergoes reductive ring opening mediated by the Thwacker Reagent, which is  regenerated with an equivalent of sodium trifutylborohydride. Grubbs’ Other Catalyst II performs a ring-closing olefin  metathesis on diene 19, furnishing most of the elseviamine  core. Alkene 20 is arylated with phenylmoronic ester 21, in a  classic example of the Mitsubishi Redistribution.¹⁸ Penultimately, dissociation of scandium reveals the lone pair of  dismyl zwitterion 22. An energy transfer process occurs  between 22 and a photoaroused acridazole [Ac]^*. The result is  an excited triplet (23), bearing an sp⁴ hybridised carbon. This  transition, while not spin-forbidden, is strongly spin discouraged. The triplet is trapped with cumulene 24, and after  dissociation of the scandium complex from the resultant  zwitterion (25), the product elseviammonium dismylate 26 is  liberated.  

The final mechanistic feature, and the only one to elude us  thus far, is the role of triphenylphosphine. One equivalent of  PPh₃ is essential for the “success” of the reaction, and a  proportionate amount of OPPh₃ is recovered during the  workup, but we don’t know why. Our best guess is that it  functions as an oxygen scavenger, or perhaps a votive offering  to the capricious chemical gods.

Conclusion 

We have proposed the first plausible mechanism for the  Over Reaction: a catalytic cycle so contrived that even The Goodies would struggle to ride it. The rate limiting step in this  mechanism is finding the catalysts and reagents. After that, it’s  a dump-and-stir in its purest form. The number, complexity, and  obscurity of catalysts for this reaction mean it would be  significantly cheaper if it were stoichiometric in rhodium  instead. It does, however, provide a one-pot route to  elseviammonium dismylates. The most remarkable aspect of  this work is that despite featuring almost every buzz-word in the  chemical lingua wanka, this still wasn’t good enough for Nature.  Still, beat that Krebs!¹⁹ 

Acknowledgements  

The authors wish to acknowledge their students, who, despite  having done all the grunt work for this paper, somehow didn’t  merit inclusion as authors. Günther Schlonk wishes to  acknowledge whoever drew the catalytic cycle for the Wacker Process page on Wikipedia. The file name is visible as  “wackonwackoff.tiff”, which made him chuckle for an hour.  

Author Contributions 

Silvio Cornuto and Alexander Thwacker conducted the  mechanistic experiments. Henri de Pampelmousse performed  the DDT calculations. Aap Tsagay prepared the starting  materials. Günther Schlonk devised the catalytic cycle during a  seminar on WHS, and prepared the manuscript.  

About the Authors  

Cornuto, Thwaker, Pampelmousse and Tsagay are successful academics in their fields, despite being entirely fictional. Recently, Günther Schlonk was almost killed by a flying shard of  glass from an exploding solvent still. Luckily, he had a Safe Working-Procedure folded in his breast pocket, which was thick  enough to stop the shrapnel from reaching his heart. Thus, he  continues in his capacity as Imperial Editor in Perpetuity of The  Journal of Immaterial Science.  

Conflicts of Interest 

Günther Schlonk has so many things he should be doing with his  time instead of writing this nonsense. Yet he continues to churn  out manuscripts due to a procrastinative disorder, a  pathological compulsion to take the piss and a troubled  relationship with his research.  

Notes and references 

1. F. F. Runge, J. Brown Chem. 1834, 156, 5704–7954.
2. The IOUPAC Brown Book, 34, 4673.  
3. F. Mercury, D. Bowie, J. Mus. Chem. 1981, 2:31.  
4. A. Grundle, M. Taint, V. Gooch, Org. Memos. 3, 460–461. 
5. “A Lab Maiden’s Tale” by A. N. Nonymous, Sysiphos Publishing. 
6. K. Über, F. N. Furious, Wangew. Chem. Int. Ed. 6, 60–71.
7. G. Marius, L. C. Sulla, J. Unnat. Prod. Online (accessed: never).
8. Don’t think about it too hard, you’ll hurt yourself. 
9. F. N. Furious et al. et al. et al, J. Am. Chem. Sox. 3524, 5–15.
10. The electronic unsupported information: tinyurl.com/3m6s7n3h
11. “Dysfunctional Groups” by Eileen Dover, 2009, Sysiphos Publishing.
12. Next week’s manuscript 
13. W. A. Focker, J. Inorganomet. 2002, 77, 98–97.  
14. S. Forkbeard, H. Bluetooth, 980 C.E. J. Scand. Chem. 4, 1–3.
15. R. H. Grubb, A. H. Darthveyda, 2005, Omnihedron, 23, 63–70.
16. S. Cluckwald, J. F. Birchtwig, 2012, J. Inorganomet. 54, 7–12.
17. Error messages via AtomSmasher’s Generator, CC BY-SA 3.0
18. J. Mitsubishi, K. Toyota, N, Subaru, 1995, Rhombus, 3, 57–102.
19. https://en.wikipedia.org/wiki/Citric_acid_cycle

If you enjoyed this Immaterial Science article please like, share, and subscribe with your email, our twitter handle (@JABDE6), our facebook group here, or the Journal of Immaterial Science Subreddit for weekly content.