F-Blog

Another interesting design of fluorosurfactants has recently been reported by Krafft et al. (Tetrahedron 64: 2008, 522-528). In the communication they described the synthesis of a triphilic, Y-shaped molecular surfactant that comprises three chains, viz., a perfluoroalkyl, a hydrocarbon and a methyl-capped polyethylene glycol. These chains have variable lengths and are interconnected in a Y-shape without any spacer, hence each unit is connected directly to the other two.

Synthetically, 2-perfluorohexyl or octyl ethanoic acid, obtained by oxidation of the commercially available corresponding 2-perfluoroalkylethanols, when treated with an excess of alcoholic KOH gave the 3-F-alkyl-3-ethoxypropenoic acids (mostly Z-isomer). The acid chlorides generated from these acids were then made to react with polyethyleneglycol monoethers to get to the final triphilic surfactants.

One major disadvantage that can be foreseen is the need to synthesize the entire molecule for every change in the perfluoro chain. Secondly, it would be difficult to predict how the microemulsion would be and to which component are the others (like the aqueous biological components) are driven to.

Considering that earlier, a triphilic polymer was found to self-assemble into micelles having three different types of compartments (one hydrophilic, one lipophilic, and one fluorophilic, the latter being both hydrophobic and lipophobic), it would definitely be interesting to see how these actually behave in a test environment. No applications utilizing these type of fluorosurfctants were actually reported except stating that “they form a highly stable Langmuir monolayers and surface hemimicelles with a hitherto unreported, facetted structure”.

The immiscibility of perfluorocarbons (PFCs) in aqueous media limits their use in humans for diagnosis and delivery. Fluoroalkylated surfactants by virtue of their ability to lower the interfacial tension between PFC and water to a near-zero value, can form transparent PFC microemulsions. For this reason development and use of fluorosurfactants has recently caught the attention of the scientific community. In a recent article, Peng et al, (J. Disp. Sci. Technol. 2008, 2, 46-51) have described the utility of a series of such fluorosurfactants.

Like any surfactant, these fluorosurfactants have a hydrophilic moiety made up of methoxy-polyethylene glycol that is end-functionalized with a perfluorocarbon containing fluorophilic moiety (R_f). The choice for the use of polyethylene glycol was made due to its biocompatibility in a variety of medical applications. Various Rf are attached to amino-MPEGs (MWs 350-750) through simple acylation. Two type of PFCs, perfluoroctyl bromide (C8F17Br) and perfluoropolyether (PEFE) were used for the study. Based on the transparency of the microemulsion formed by mixing known amounts of aqueous solutions containing fluorosurfactants with PFCs, ternary phase diagrams were plotted that differentiate phases as well as proportion of each component. When compared to the commercially available zonyl series nonionic fluorosurfactants, these appear to have low polydispersity as well as form more stable microemulsions. Based on these studies it was speculated that the molecular weights of the hydrophilic and fluorophilic groups of the fluorosurfactant and the PFCs should be comparable for it to cause a transparent microemulsion.

One of the main limitations of fluorous tags in various chemistries is poor solubility in aqueous solvents. Thus this communication would pave the way towards solving this limitation. Picking the right fluorosurfactant (with appropriate fluorophilic and hydrophilic tags) in conjunction with the PFCs one could easily be able to overcome the precipitation issues encountered, especially in biological applications.

Writing a blog post about a recent communication “Yes, it’s all fluorous…”, that is posted under the “Carbohydrates” section I realized how important this technique would be of to the synthesis of biopolymers. The described strategy involves the use mono-fluorous tagged substrates to get to a double-tagged product that could be easily separated from the rest (mono-tagged and untagged) by a simple FSPE.

While the unpredictability of the solubility of double-fluorous tagged products in organic solvents may limit its usage to the synthesis of organic compounds, it would have tremondous application to the synthesis of oligopeptides, oligonucleotides and oligosaccharides. These biopolymers being rich in polar groups could tolerate the high hydrophobicity of the two fluorous tags to an extent. This would enable the synthesis of these biopolymers to be easily carried out in solution phase without resorting to the fancy solid-phase especially when large amounts are required.

In a typical solution phase chemistry involving the formation of the amide bond, the reaction mixture after the reaction mixture is subjected to acid/base washes to eliminate the excess base/acid. This is easy up to a level of tetramer after which it starts to get complicated. For the synthesis of any peptide longer than this it is convenient to resort to “segment-condensation”, where in short fragments (up to 4-mer) are first synthesized in solution and then coupled to yield the longer sequences. Here is where the double-tagging strategy could be put to use. Each of the shorter fragments could be mono-fluorous tagged to get to the double-fluorous tagged oligopeptide that could then easily separated from the rest by FSPE. Again the presence of too many bulky (like trityl) side-chain protecting groups may complicate things a little. Maybe one-day we would have fluorous versions of these protecting groups with small fluorous tags (upto C4F9) that add up to give enough fluorophilic leverage so as to aid fluorous separation.

In the absence of a fool-proof solid-phase synthetic strategy for the oligosaccharide synthesis, as well as non-feasibility of a simple acid/base wash to clean-up, it makes even more sense. However, additional protection/deprotection steps involving the fluorous tags cannot be dispensed with.

A new J. Org. Chem ASAP article from Prof. Sanotos Fustero and colleagues describes the transesterification of 2-(TMS)ethyl α-imino esters with TBAF. α-Imino esters have been used quite a bit in the synthesis of unnatural amino acids through reduction of the imine to the amine. In this instance Prof. Fustero uses fluoro-substituted α-imino esters which can be precursors to perfluorinated amino acids, which can impart interesting properties to peptides and proteins as described in a previous post. The methodology is quite simple. By starting with a TMS-ethyl ester, TBAF mediated deprotection results in the ammonium carboxylate which can be alkylated in situ if an appropriate electrophile is present. Within this note, they report several different types of eletrophiles resulting in net tranesterification, including benzyl, primary, secondary, and even tertiary halides.

Among the notable items was the comparison of solid phase vs. fluorous phase alkylation. Since they want to eventually adopt this methodology for the construction of chemical libraries they conducted the alkylation with both solid phase and fluorous phase halides. As is usually the case, the fluorous phase reaction was much faster due to homogeneous reaction conditions. Also there was no reported yields for the solid phase reactions while the fluorous phase reaction resulted in 90% yield. They could have cleaved the new ester off the solid phase to gauge yield, but I guess they chose not to for whatever reason.

The second thing that caught my eye was that within that paper there is a very nice example of the reactivity difference between a fluorous ethyl spaced iodide (RfCH2CH2I) and a fluorous propyl spaced iodide (RfCH2CH2CH2I). When they tried the reaction with the ethyl spaced iodide they obtained no product, but the propyl spaced they obtained 90% reaction. They rightly attribute this to competing beta-elimination. The protons alpha to the fluorous chain can be quite acidic. At FTI we have observed this phenomena often and rarely do we even try to use the ethyl spaced iodide for alkylations. We strictly use the propyl spaced halides for alkylations and reserve the ethyl spaced for other chemistries, generally organometallic chemistry.

So if you’re looking to make your own fluorous reagents, tags, or compounds always keep in mind the difference between the ethyl and propyl spaced variants. Remember, one carbon is all it takes.

Continuing their efforts to find a reliable and efficient automated oligosaccharide synthesis, Pohl et al have published a double-fluorous-tagging strategy that offers a robust simple separation to eliminate failed glycosylation sequences in solution-based oligosaccharide synthesis. In this upcoming article, the authors used a light fluorous (C8F17) acetate protecting group on an oligoglucosamine glycosyl donor and a fluorous (C8F17) allyl protecting group on an acceptor to get to a double fluorous-tagged glycosylation product. This product could then easily be separated from the unreacted single-fluorous-tagged acceptor as well as donor byproducts and their decomposition products by a simple gravity filteration over a commercial FSPE catridge.

While citing the numerous advantages of using fluorous strategy (purification, invisibility in [1H]-NMR, diverse protecting groups, stability to most reaction conditions) to deal with automated oligosaccharid synthesis, the authors also highlight a potential problem with a single-fluorous-tag strategy when a reaction does not go to completion. To circumvent this problem, addition of a fluorous tag to both rather than just one of the coupling partners was envisaged. This would offer a chance to eliminate by-product build up during the synthesis without the need of an additional capping step. The doubly-fluorous tagged product having stronger affinity than a mono-tagged to fluorous phase was then demonstrated to be easily separated from the mono-tagged excess/unreacted reagents by a simple gravity FSPE using three washes instead of the usual two.  A simple method to remove the fluorous acetate byproduct from the desired fluorous-tagged sugar in the deprotection cycle is also reported.

At this stage it is too early to generalize or say if this doubly fluorous tagged strategy works in non-carbohydrate scenarios mainly owing to the unpredictability of the solubility of these heavy fluorous tags in organic solvents. Having eliminated the additional capping step and simplifying the purification during each cycle of building 1-4-linked glucosamine oligomers, the reality of a robust automated oligosaccharide synthesis is not far away.