ABSTRACT
Background e reduction of functional groups such as S-alkyl-dithiocarbonates (S-xanthates), iodides, O-alkyl-dithiocarbonates (O-xanthates) and related compounds to the
corresponding alkanes is very important in organic synthesis, especially in natural products chemistry.[1,2] Deoxygenation (Barton-McCombie reaction) has been largely used to handle sensitive compounds such as sugars.[3] On the other hand, iodides and S-alkylxanthates are useful compounds that easily produce carbon radicals involved in radical chain reactions (cyclisation, intermolecular addition onto an olen, etc) in which trapping of the nal radical results in the transfer of the iodine atom or of the xanthate moiety.[4,5] To date, the most widely used reductive method to remove these functional groups that become superuous at the end of the reaction process, is based on the Bu3SnH/AIBN combination that operates at 80°C or above. e main virtue of this method relies on its versatility and its eciency. However, this procedure suers from crippling drawbacks in terms of toxicity, cost, disposal, and tedious purication to remove tin residues. [6-10] However, we recently faced limits to the use of some of these methods for reductive removal of S-alkyl O-ethyl dithiocarbonates and we therefore proposed diethyl phosphite/DLP and H3PO2/Et3N/AIBN as attractive reagents for this purpose that circumvent these impediments.[11]
e core of the results reported in this article and the following parts,[12,13] has been presented as communications at the Xth Symposium of the « Institut de Chimie des Substances Naturelles », 1-3 June 2005, Gif-sur-Yvette, France and at the “1st German-French Congress in Organic Chemistry”, Goslar, September 7th – 11th, 2005. e recent observation made by Wood and his colleagues that O-alkylxanthates may be deoxygenated by the combination Et3B/air/H2O under similar conditions (published on the Web on 08/18/2005),[14] the subsequent work by Renaud and coworkers who studied the reduction of B-alkylcatecholboranes under very similar conditions,[15] as well as the very recent kinetic study published by Newcomb,[16] prompt us to report in detail our ndings in this area as a series of three articles. In this rst article we report that trialkylboranes are useful reagents to achieve reduction of several radicophilic groups. A careful investigation of the process led us to develop a new method for the reduction of S-alkylxanthates, iodides and similar compounds to the corresponding hydrocarbons that requires a trialkylborane in excess and air as initiator, at 20°C, or even at lower temperature.[17,18]
Results and Discussion During the last two decades, the number of applications of Et3B as radical initiator has increased tremendously. As Et3B is known to produce radicals at 20°C, or even much lower temperature (-78°C), we thought that trialkylboranes in the presence of dioxygen would be good candidates to trigger the radical addition of α-acyl Sxanthates onto olens. When testing this hypothesis, we observed that compound
1a (Figure 1) underwent addition to decene in 40% yield when Et3B (0.1 equiv.)/ air was used to initiate the reaction. Under these conditions, traces of alkane 1b (< 5%) were also isolated. When 2.5 equiv. of Et3B were used, compound 1a failed to yield adduct 1c but instead was reduced to the corresponding alkane 1b in 63% yield. No other identiable product could be isolated. Zard and Nozaki mentioned similar reductions from α-acylxanthate[19] and α-acyliodide[20,21] respectively. ese authors proposed the formation of an intermediate boron enolate that hydrolyses on work-up. A detailed report of our results concerning the addition process is given in a subsequent article.[13]
As the absence or the low yield of the addition product 1c might be due to its consumption by an unanticipated reaction, we decided to examine the reactivity of S-alkylxanthates that do not bear a carbonyl function in the adjacent position, under similar conditions, but without added olen. In a rst set of experiments [Figure 2, Table 1, Method A], we showed that the primary and secondary 2-oxoalkyl xanthates 1a and 2a behaved similarly (entries 1 and 2), as expected, giving high yields of compounds 1b and 2b, respectively. Interestingly, secondary S-alkylxanthates 3a-5a were cleanly reduced to the corresponding alkanes in excellent yields (up to 80%) (Table 1, entries 3-5). When the starting material was soluble enough, there was no need for an extra solvent other than the mixture of hexanes present in the commercial Et3B solution (entries 3 and 6). Experimentally, in method A, a solution of xanthate, Et3B (5 equiv., 1M solution in hexanes), in the given solvent (if needed) was simply stirred for 2 h in the presence of air under anhydrous conditions.
