Ketones problematic

While discussing ethers we should mention that the presence of unreacted anisoles or methyl anisoles is highly undesirable in the manufacture of phenol-formaldehyde resoles. These materials tend to be unreactive relative to phenol under normal resole conditions. They are also volatile and have odors detectable at very low concentrations. They have been the source of worker complaints and costly claims in the wood products industry. Benzophenones and methyl phenyl ketones are also common phenol contaminants that are problematic in this regard. 

Systematic studies of additions to a-amino ketones are unavailable. One reason may be that the synthesis of enantiomerically pure a-amino ketones has been problematic, however, convenient preparations of various a-amino ketones are now known74-79. 

An iron complex-catalyzed asymmetric hydrosilylation of ketones was achieved by using chiral phosphoms ligands [68]. Among various ligands, the best enantios-electivities (up to 99% ee) were obtained using a combination of Fe(OAc)2/(5,5)-Me-Duphos in THF. This hydrosilylation works smoothly in other solvents (diethylether, n-hexane, dichloromethane, and toluene), but other iron sources are not effective. Surprisingly, this Fe catalyst (45% ee) was more efficient in the asymmetric hydrosilylation of cyclohexylmethylketone, a substrate that proved to be problematic in hydrosilylations using Ru [69] or Ti [70] catalysts (43 and 23% ee, respectively). 

The ketone 15 was eventually prepared by Grignard addition to Weinreb amide 21, as shown in Scheme 5.5. The Weinreb amide 21 was prepared from p-iodobenzoic acid (20). The phenol of readily available 3-hydroxybenzaldehyde (22) was first protected with a benzyl group, then the aldehyde was converted to chloride 24 via alcohol 23 under standard conditions. Preparation of the Grignard reagent 25 from chloride 24 was initially problematic. A large proportion of the homo-coupling side product 26 was observed in THF. The use of a 3 1 mixture of toluene THF as the reaction solvent suppressed this side reaction [7]. The iodoketone 15 was isolated as a crystalline solid and this sequence was scaled up to pilot plant scale to make around 50 kg of 15. 

Challenging applications in the field of macrocyclic furans have been investigated. The major synthetic advantage is the cyclization to the furan after the macro-cyclization. This will avoid a problematic ring closure to macrocycles (to 1,3-furano-phanes) with a furan substrate ( furan latest strategy ). Test substrates demonstrated the viability of this concept [50], as shown below for the synthesis of the [8]furano-phane 91 from the macrocyclic ketone 90 (Scheme 15.22) [39]. 

Cyanation of aldehydes and ketones is an important chemical process for C C bond formation." " Trimethylsilyl cyanide and/or HCN are commonly used as cyanide sources. The intrinsic toxicity and instability of these reagents are problematic in their applications. Acetyl cyanide and cyanoformates were used as cyanide sources in the enantioselective cyanation of aldehydes catalyzed by a chiral Ti complex and Lewis base (Scheme 5.31)." The Lewis base was necessary for the good yields and selectivities of these reactions. The desired products were obtained in the presence of 10mol% triethyl amine and 5mol% chiral titanium catalyst (Figure 5.14). Various aliphatic and aromatic aldehydes could be used in these reactions. 

Trost s group reported direct catalytic enantioselective aldol reaction of unmodified ketones using dinuclear Zn complex 21 [Eq. (13.10)]. This reaction is noteworthy because products from linear aliphatic aldehydes were also obtained in reasonable chemical yields and enantioselectivity, in addition to secondary and tertiary alkyl-substituted aldehydes. Primary alkyl-substituted aldehydes are normally problematic substrates for direct aldol reaction because self-aldol condensation of the aldehydes complicates the reaction. Bifunctional Zn catalysis 22 was proposed, in which one Zn atom acts as a Lewis acid to activate an aldehyde and the other Zn-alkoxide acts as a Bronsted base to generate a Zn-enolate. The... 

Replacement of formaldehyde In amlnoplasts by substitution with higher aldehydes (or ketones) Is even more problematic the equilibrium of the condensation shifts strongly towards starting materials, except In cases where the formaldehyde replacement contains electron withdrawing substituents (e.g. with glyoxal... 

Ketone and ester enolates have historically proven problematic as nucleophiles for the transition metal-catalyzed allylic alkylation reaction, which can be attributed, at least in part, to their less stabilized and more basic nature. In Hght of these limitations, Tsuji demonstrated the first rhodium-catalyzed allylic alkylation reaction using the trimethly-silyl enol ether derived from cyclohexanone, albeit in modest yield (Eq. 4) [9]. Matsuda and co-workers also examined rhodium-catalyzed allylic alkylation, using trimethylsilyl enol ethers with a wide range of aUyhc carbonates [22]. However, this study was problematic as exemplified by the poor regio- and diastereocontrol, which clearly delineates the limitations in terms of the synthetic utihty of this particular reaction. 

Interestingly, treatment of the allylic carbonate 23, which had proven problematic in the previous study, under analogous reaction conditions with the copper enolate derived from 24 furnished the a,/9-disubstituted ketone. Subsequent ring-closing metathesis furnished the 1,2-cyclohexenes 25a/25b in 75% overall yield favoring the trans-dia-stereomer 25a (2° 1°=30 1, ds=10 l) [14]. Overall, this reaction provides an alternative approach to an exo-selective Diels-Alder cycloaddition, and indicates that a-substituted enolates are even more tolerant nucleophiles than the unsubstituted derivatives. 

There were two more stereocenters to set. It was expected that cuprates would add to the open face of the strained cyclobutene. The control of the other stereocenter was more problematic. One solution was to prepare an a-sulfonyl lactone. To this end, the ketone was converted to the secondary carbonate. As hoped, conjugate addition was followed by intramolecular acylation, but the reaction continued to full acyl transfer, to give 10. Fortunately, desilylation of 10 proceeded with concomitant lactonization. Desulfonylation then gave 2, which could be brought to high by recrystallization. 

Because the tosylate is primary, substitution should be the major pathway (although in this case elimination could be problematic because of conjugation with the phenyl ring). We note, however, that the enolate needed is the kinetic enolate of 5-hexen-2-one. This poses a regiochemical control problem which can be solved by making the 7V,7V-dimethylhydrazone of the ketone. The ketone 5-hexene-2-one is available ( 46.20/25 g) or can be made by allylation of acetone. 

For the puiposes of retrosynthetic analysis, a six-membered ring in a target can be related to a Robinson annulation of an existing ketone with an a,/3-unsamrated ketone. Normally cc,/3-unsaturated methyl ketones are used to facilitate the ring closure, but this is not an absolute requirement. Thus the target steroid S could potentially be constructed by a series of Robinson annulations as shown. The last retrosynthetic step (the first synthetic step) could be problematic as a mixture of regioisomers would be formed. 

At this point, consideration was next accorded to proper introduction of the pair of substituents as in 34. As expected, the regiocontrolled introduction of a methyl group proved not to be problematic, and lithium diisopropylamide came to be favored as the base. The p isomer 29 predominted by a factor of 5 1 over the a isomer for the usual steric reasons (Scheme 5). To reach silyl enol ether 31, it was most efficient and practical to react the 29/30 mixture with chlorotrimethylsilane under thermodynamic conditions. This step proved to be critical, as it allowed for implementation of the Lewis acid-catalyzed acetylation of 31 under conditions where the benzyloxy substituent was inert. Equally convenient was the option to transform the modest levels of enol acetate produced competitively back to starting ketone by saponification with methanolic potassium hydroxide. 

Another problematic area for CAMEO is secondary degradation. As mentioned above, CAMEO typically correctly predicts benzylic oxidation to the hydroperoxide precursor however, reactivity halts at the peroxide. In actual studies, the secondary ketone degradant is typically observed and this is not reported with CAMEO. 

Another class of problematic amines are a-amino nitriles, which are readily accessible from ketones, amines, and cyanide. Like a-amino acids, these amines are electronically deactivated and less basic and nucleophilic than purely aliphatic amines, and are therefore difficult to acylate. Some a-amino nitriles or the corresponding acylated derivatives can, furthermore, decompose into imines and cyanide if reaction temperatures are too high or if the bases used are too strong (Scheme 7.11). 

Metal-catalyzed oxidation of alcohols to aldehydes and ketones is a subject that has received significant recent attention [21,56,57]. One such method that utilizes NHC ligands is an Oppenauer-type oxidation with an Ir or Ru catalyst [58-62]. These alcohol oxidation reactions consist of an equilibrium process involving hydrogen transfer from the alcohol substrate to a ketone, such as acetone (Eq. 5), or an alkene. Because these reactions avoid the use of a strong oxidant, the potential oxidative instability of NHC ligands is less problematic. Consequently, these reactions represent an important target for future research into the utility of NHCs. 

Next, the scope of this C-glycosylation reaction was investigated. The NeuSAc chloride was found to serve as a donor in samarium mediated C-glycosylation, unfortunately, purification of C-glycoside from excess chloride donor is often problematic, making the NeuSAc phenyl sulfone the donor of choice. A variety of acceptors were also evaluated including alkenes, epoxides, vinyl esters, aldehydes and ketones. Only the aldehydes and ketones afforded the desired C-glycoside products. 

A particularly effective method for the asymmetric synthesis of both aldehyde-and ketone-derived sulfinimines recently introduced by Davis and co-workers is the condensation of (5)-(+)-p-toluenesulfinamide (63) with aldehydes and ketones using activated 4-A molecular sieves or titanium ethoxide [Ti(OEt)4].46 This procedure avoids the problem of removing the menthol by-product of the one-pot procedure (see Section II.D) which is sometimes problematic.23 Importantly, this methodology affords ketone derived sulfinimines 66 which are difficult to prepare by other means. 

Keese envisioned the use of a tandem PKR for the synthesis of fenestranes. The second cycloaddition was in principle problematic as it involved an al-kene conjugated with a ketone. They were surprised when they observed the direct formation of the tetracyclic unit 136 from the endiyne 135 although with low yield [ 148]. Further studies from this group led to a mechanistic proposal that explained this result. It was clear from the fact that compound 140 failed to react, that the second PKR had to start from an intermediate metal-lacycle rather than from the uncomplexed final cyclopentenone. Thus, cobalt complex 137 would lead to 138 were both metal clusters would interact giving intermediate 139 which would evolve in the usual way to the final product (Scheme 42) [149]. These systems have been obtained later by Chung s group using cobalt nanoparticles as commented above (Sect. 2.4) [131]. 

The use of oxoammonium ions such as those derived from TEMPO in combination with inexpensive, safe, and easy-to-handle terminal oxidants in the conversion of alcohols into aldehydes, ketones, and carboxylic acids is a significant example of how it is possible to develop safer and greener chemistry, by avoiding the use of environmentally-unfriendly or toxic metals. However, separation of the products from TEMPO can be problematic, especially when the reactions are run on... 

Unfortunately, several important classes of a-diazo ketones cannot be prepared in good yield via these standard methods. a -Diazo derivatives of a.p-unsaturated ketones, for example, have previously proved to be particularly difficult to prepare.1113 12 The acylation of diazomethane with a.p-unsaturated acid chlorides and anhydrides is generally not a successful reaction because of the facility of dipolar cycloaddition to conjugated double bonds, which leads in this case to the formation of mixtures of isomeric pyrazolines. Also problematic are diazo transfer reactions involving base-sensitive substrates such as certain a,p-enones and heteroaryl ketones. Finally, the relatively harsh conditions and lack of regioselectivity associated with the thermodynamically controlled Claisen formylation step in the "deformylative" diazo transfer procedure limit the utility of this method when applied to the synthesis of diazo derivatives of many enones and unsymmetrical saturated ketones. 

---