Palladium (II - IV) catalysis

25 February 2009 - organopalladium chemistry

The Pd(0) - Pd(II) catalytic cycle is well known in organopalladium chemistry, see for instance the Heck reaction. Less familiar is the Pd(II) - Pd(IV) cycle. An update.
A very recent report (Tsujihara et al. 2009 DOI) describes an enantioselective ring-closing reaction cyclopropanation tandem reaction of an enyne by a combination of a palladium(II) catalyst, an oxidizing reagent and a BOX ligand:

Two almost identical protocols for the racemic version of this reaction (that is, without the ligand) appeared in two articles, only separated by 6 days and both in JACS, in February 2007, one by Tong/Beller/Tse(DOI):

and one by Welbes/Lyons/Cychosz/Sanford (DOI):

with catalytic palladium acetate and stoichiometric iodosobenzene diacetate.

The reaction mechanism both parties agreed upon is a sequence of acetoxypalladation, cyclization, oxidation of Pd(II) to a octahedral Pd(IV) intermediate by the periodinane, cyclopropanation and reductive elimination.

The early origins of the cyclization part of this reaction can be traced back to a Pd(II) catalysed codimerisation of an alkyne with an allyl halide (Kenada et al. 1979 DOi)

which was followed up by an intramolecular ring-closing reaction of an enyne (Lu et al. 1991 DOI)

and its enantioselective version (Lu et al. 2000 DOI)

The Pd(II/IV) cycle itself has been explored by the Sanford group starting in 2004 with a C-H activation reaction of a benzoquinoline (Dick et al. 2004 DOI)

The same group went on to study oxygenation on aliphatic substrates (Desai et al. 2004 DOI), the structure of isolated Pd(IV) complexes (Dick et al. 2005 DOI), Pd(II/IV) cycles in aryl - aryl coupling reactions (Kalyani et al. 2005 DOI) and coupling reactions using oxone (Hull et al. 2006 DOI). The distinct advantage of Pd(IV) over Pd(II) in reductive elimination is the easy of forming a new C-O or C-halide bond.

The formation of an carbon-oxygen bond by reductive elimination of oxidized palladium has a history of its own. Early work is the conversion of benzene to phenol with a palladium catalyst using plain oxygen as reported by Jintoku et al. in 1990 (DOI). The methodology is ultimately based on the Shilov system (1972) which is C-H bond activation by reductive elimination not by palladium but by platinum.

Update: For Pd(III) update see here.

Chemistry blogs

20 February 2009 - in the blogs this month

Many chemistry blogs exist on the Internet. As a special service to the readers of this blog an overview of several of this month's contributions.

Lamentations on chemistry laments on the Trost vs Djerassi row (Link) (this blog fully supports Djerassi: stop citing reviews and cite the original work), Totally synthetic (Link), The Chem Blog (Link), Corante (Link) and The Curious Wavefunction (Link) all have the latest on the hexacyclinol fraud case. The Realm of Organic Synthesis (link) speaks out in support of Wikipedia, which is great, but then goes on advocating a free-access knowledge database which should look similar to wikipedia. My advise to the Realm: why not join wikipedia!.

Several days ago The Molecule of the Day was propanolol (Link) a known drug that also appears to able to cure fright!. Mining drug space wonders if there is an alternative to CAS as a supplier of chemical information (Link). RobTheJob investigates a certain beer component (in German) (Link) and Curley Arrow (Link) has the story on a diazidomethane explosion. More nitrogen related explosive stuff by the way is brought to you by Bridgehead carbons (Link). Will and beyond suggest we drink more coffee (Link) as coffee grounds can have a second life as biofuel. Carbon Based Curiosities require assistance in identifying a certain piece of glassware (Link)

Several bloggers report on work they do in the laboratory (serious work and not so serious work): The everyday scientist investigates fluorescence in Cognac (Link), Org Prep Daily (Link) treats us to an aldehyde fluorination. Milo's Musings gives a practical overview of analytical techniques typically available to an organic chemist (TLC listed as number 2!)(Link). And finally, The F-blog offers an alternative to acetonitrile (currently scarce! blame the Chinese) in HPLC.

metal-free or almost-no-metal reactions

15 February 2009 - news

An alkyne hydration catalysed by a gold/silver system at a 10 ppm level with turnover number reaching 84,000 has recently been reported (Marion et al. DOI).

In this system the ligand is a N-heterocyclic Carbene. At the same time, reports on no-metal at all (metal-free) organic reactions are abundant which on the one hand seems even better but on the other makes you wonder if in these systems minute quantities of catalyst present are not simply overlooked.

Lets take some reports from the recent literature and check for analytical chemistry on metal trace amounts, and while we are at it, check presence of a control experiment (what happens if metal IS present), check for mechanism (how does the reaction work when metal is omitted), test for scope, test for atom economy and finally, check for efficiency (is the uncatalysed reaction worth the wait)

The metal-free Ferrier rearrangement presented by De et al. (DOI) is not truly a metal-free reaction because the element boron in boron trifluoride that is replaced by hexafluoropropanol (HFIP) is not a metal but a metalloid. On the upside, a mechanism as to why HFIP works just as well is provided.

In a metal-free Trost allylic alkylation, recently reported by Cheng et al. (DOI), palladium is replaced by (stoichiometric) oxidizing reagent DDQ . Key step in the reaction mechanism is formation of a charge-transfer complex.

A report by Tu et al. (DOI) describing a Prins-like reaction with very similar DDQ chemistry (and published only a month or so prior to the one above) does not bother to specify the metal being replaced:

Finally, a metal-free asymmetric BOX ligand is employed in the Diels-Alder reaction of anthrone with a maleimide (Akalay et al. DOI). A metal is usually required to anchor the ligand to the substrate but this report shows that hydrogen bonding will do as well.

What do the above examples have in common? No surprises there: lack of control experiments, need for specific substituents for the reaction to work, no hunt for hidden metals, not always clear what metal is being replaced. On the upside all reports are accompanied by handsome reaction mechanisms.

allene! zwitterion!

07 February 2009 - News

An interesting discussion has been taking place between Guy Bertrand & group (University of California) in one corner and Christl & Engels (University of Würzburg) in the other on a certain cyclic allene. In this type of compound a central carbon atom is connected to two other carbon atoms through a double bond. This unit is linear and forcing it in a cycle is expected to be tricky. Yet, the Bertrand group in 2008 reported one such cyclic allene with an unusual CCC bond angle of 97.5°, and stable (Lavallo et al. DOI).

In terms of resonance structures Bertrand not only considers allene A but also 4-electron donor B which can be considered a carbodicarbene with a divalent carbon(0) center.

Not so, say Christl & Engels, this new compound is not an allene but a zwitterion (C) (DOI). Their evidence: no experimental evidence on five-membered cyclic allenes exists in the literature, six-membered ring allenes found are also zwitterions (DOI) and the finding is not supported by in silico experimentation. Importantly, in the X-ray structure the OCCCO unit is found in a single plane whereas for an allene an large angle is expected.

In the Bertrand response (Lavallo et al. DOI) the semantic and scientific arguments raised by Christl & Engels are summarized as erroneous: first of all, depicting a molecule as an allene and calling it an allene does not imply that the ground state is an actual allene. Jokingly, structure D is proposed with all electron pairs omitted: readers are invited to place them back using their own judgement. A strong argument in favor of the allenic structure according to Bertrand is pyramidalization of the nitrogen atoms which would disrupt aromatic stabilization of the zwitterion.

Perhaps both teams can agree on structure B. Christl & Engels dismiss it as such because the C4 atom carries a double negative charge but nevertheless see possibilities for it as a ligand to a strongly electron-withdrawing transition metal. More experimental evidence would be welcome. Apparantly an X-ray structure is not enough and any spectroscopic method fails. Polarity, usually a telltale zwitterion signature was not part of any discussion.

Update June 2009: Here

Vincent Lavallo, C.?Adam Dyker, Bruno Donnadieu, Guy Bertrand (2008). Synthesis and Ligand Properties of Stable Five?Membered?Ring Allenes Containing Only Second?Row Elements Angewandte Chemie International Edition, 47 (29), 5411-5414 DOI: 10.1002/anie.200801176