New in water oxidation

22 April 2010 - Catalysis

Water splitting is any reaction converting water into hydrogen and oxygen for example by electrolysis. More specifically in photocatalytic water splitting the energy source is light and in artificial photosynthesis carbon dioxide is added to water forming oxygen and a carbohydrate. Given the need for clean and cheap energy (hydrogen) the hunt is on for a suitable photocatalyst to go with it.
In a recent Science report Yin et al.(DOI) describe the cobalt cluster compound Na10(Co4(H2O)2(PW9O34)2)·27H2O. Cobalt containing catalysts have already been investigated by others but are heterogeneous and the first homogeneous catalyst ever found is based on expensive ruthenium. The homogeneous cobalt compound was rediscovered in an old 1973 article (Weakley et al. DOI) and can be made by mixing and refluxing in water sodium tungstate Na2WO4, sodium biphosphate Na2HPO4 and cobalt(II) nitrate Co(NO3)2· The resulting cluster compound can be envisioned as having a Co4O16 core lined with PO4 tetrahedra and WO6 octahedra.

In order to save lab time the catalyst was only tested with respect to water oxidation in water against the oxidized form of tris(bipyridine)ruthenium(II) chloride in this reaction:
4 (Ru(bpy)3)3+ + H2O -> 4 (Ru(bpy)3)2+ + 4H+ + O2
By measuring the amount of oxygen generated a turnover number for the catalyst was measured of 80 per second with ultimately 70% oxygen yield. A total of 7 very related catalysts did not work at all and this catalyst only worked in a narrow pH region demonstrating that finding a catalyst at all is not that easy. Contrary to many organic-based catalysts no catalyst degradation was observed. For comparison regular cobalt nitrate scores a TON of only 10 and 33% oxygen yield (converted to gram conversion it would still win out).

CuAAC mechanistically

20 April 2010 - organometallic chemistry

CuAAC is an organic reaction that stands for copper-catalysed alkyne-azide-cycloaddition (review:DOI). It is an extension of the Azide alkyne Huisgen cycloaddition with copper independently introduced by Meldal (DOI) and Sharpless (DOI) both in 2002. A regular thermal reaction results in a mixture of triazole isomers but copper forces the formation of just one. Earlier in 2010 Buckley et al. published the preparation of a dicopper compound they say plays a part in the reaction mechanism for this reaction (DOI) and in even more recent publication (DOI) they expand on it.

This polymeric ladderane ((PhCCCu)2)n can be prepared from copper acetate through copper(II) hydroxyacetate and phenylacetylene with Cu(II) reduced to Cu(I) at the expense of solvent methanol. The same ladderane can also be obtained from copper sulfate with reducing agent sodium ascorbate. ((PhCCCu)2)n is then the catalyst in the actual cycloaddition between phenylacetylene and benylazide to the triazole. You may wonder why in this reaction sodium ascorbate is again listed as an ingredient. It should not be there. In a control experiment with another substrate the authors themselves make it clear the reducing agent no longer plays a part but for some reason the ingredient stayed.
CuAAC mechanism Buckley 2010

The traceless bond construction

12 April 2010 - Organic chemistry

In a new concept called traceless bond construction Mundal, Avetta Jr & Thomson form a new carbon-carbon bond in an organic reaction with the original functional groups completely disappearing (DOI). In terms of retrosynthetic analysis it means that there may exist more strategic avenues towards a desired molecule if only you look hard enough. In their case study a boc (tBuOCO) protected allyl hydrazone (RRC=N-NRR) takes part in an triflimide-catalysed sigmatropic reaction with subsequent loss of isobutylene, nitrogen gas and carbon dioxide.


True enough, a forensic team would be clueless when presented with the raw reaction product because most of the accomplices have disappeared. Equally true, in many regular coupling reactions bond formation is also traceless and equally true as with coupling reactions the atom efficiency is just horrible. As an aside, the authors skillfully avoided an confrontation with the final part of the reaction mechanism, an acidic Wolff-Kishner reaction type removal of the dinitrogen group, This blog too gave up but if it is any consolation managed to steer away from anything carbanionic.

Target cubebol

09 April 2010 - Total synthesis

Hodgson et al. have successfully synthesised cubebol, a biomolecule and sesquiterpene of potential use as cooling reagent in culinary applications. Not that it is the first total synthesis of this compound, many methods have been published (Yoshikoshi 1972 DOI, Okamoto 1976 DOI , Fürstner 2006 DOI, Fehr 2006 DOI ) but this one starts from menthol and is particularly short. It is also solves an outstanding problem in menthone chemistry first encountered by Ludwig Claisen in 1894 (DOI).


Menthol 1 was oxidized to menthone 2 using calcium hypochlorite, formylation took place next with 2,2,2-trifluoroethyl formate and LDA (DOI) to enol 3 and reaction with lithiumaluminumhydride to exocyclic alkene 4 was a tandem of elimination and reduction. Allylic chloride 5 was obtained reaction with MsCl / triethylamine / LiCl, a Grignard reaction with chiral epichlorohydrin (Mg, CuI, THF) gave alcohol 6 , cyclopropanation to 7 was accomplished using BuLI / TMP (DOI), oxidation to ketone 8 required TPAP / NMO and finally a methyl group was introduced forming (-)-cubebol 9 using MeLi / caesium chloride.
Regular menthone formylation using an alkoxide is accompanied by epimerization of the isopropyl group and formation of the cis product (apparently already observed by Claisen with another formylation agent) which is undesirable. In a simple mechanistic picture this compound results from the thermodynamic product. The low-temperature LDA method -according to Hodgson et al. - favors the kinetic product despite methyl group interference.

Recently in DOSY NMR

07 April 2010 - Spectroscopy

Diffusion ordered NMR spectroscopy or DOSY NMR, invented by Morris & Johnson in 1992 (DOI DOI), is a 2D-NMR technique displaying regular chemical shift in one dimension and diffusion behaviour in the other. In this way each signal is enhanced with information about the size and even molar mass of the molecule under scrutiny.

In technical terms multiple NMR spectra are obtained with variation in the pulsed field-gradient only. The signal intensity of each peak in the spectrum decreases linearly with the slope proportional to the diffusion coefficient D. The Stokes-Einstein equation then relates D with the hydrodynamic radius of the analyte. This equation was originally set up for colloids and extrapolation to small molecules require correction factors. Interestingly the very shape of the molecule is taking on significance e.g. oblate or prolate. (DOI)

In recent times DOSY has been used to study host-guest chemistry with cyclodextrin(DOI), the detection of biotoxins for the military (DOI), the molar mass distribution in block copolymers contaminated with homopolymer (DOI), the analysis of nanotubes (DOI), all the way to the detection of counterfeit sildenafil pills (DOI).

Garcia-Alvarez et al. have applied DOSY NMR to turbo Grignards. In 2008 this group found LiClMgCl rings in the solid state structure of (TMP)MgCl.LiCl (turbo-TMP, see earlier blog). In their latest work (DOI) they mixed Lithium diisopropylamide and magnesium chloride in THF. The solid state structure of the resulting crystals is an assembly of centrosymmetric dimers, again with the LiClMgCl motif. This structure was ruled out in the THF solution phase because a species with that molecular weight could not be found. Instead DOSY supported the formation of solvent-separated ion pair with a completely THF solvatated lithium ion complemented with a dimer dianion and a monomer anion in equilibrium.

10-04-2010: Minor update with general DOSY
DA turbo grignard Armstrong  2010

Shvo catalyst at work

05 April 2010 - Catalysis

Bähn et al. have demonstrated a new application of the Shvo catalyst with the N-alkylation of indole by hexanol (DOI) . This reaction is a indirect reductive amination with the alcohol first oxidized to the aldehyde by transfer hydrogenation from the indole which then reacts with the amine to the enamine which isomerizes. By adding a small amount of p-toluenesulfonic acid both conversion and selectivity increases (less C-alkylation). Traditional methods for N-alkylation are messy when they involve alkyl halides.

Another recent alcohol N-alkylation but not catalytic is one based on a Mitsunobu reaction (Bombrun et al. 2002 DOI)