This page looks plain and unstyled because you're using a non-standard compliant browser. To see it in its best form, please upgrade to a browser that supports web standards. It's free and painless.
kilomentor | 14 May, 2007 09:42
Hydrocarbon solvents like hexane, cyclohexane, heptane and methylcyclohexane are special because they form biphasic mixtures with both methanol and acetonitrile. In addition, cyclohexane, at least, and perhaps the others, also form thermomorphic systems with each of:
Kazuhi Chiba, Yusuke Kono, Shokaku Kim, Kohsuke Nishimoto, Yoshihazu Kitano and Masahiro Tada, teach this in Chem.Commun. 2002 1766-1767.
A thermomorphic system is a combination of liquids that in one temperature range is a homogeneous single phase, but in another range is two immiscible liquids that can be separate by simple extraction. The significance that is the focus of this blog discussion is not that the systems are thermomorphic but only that there is a temperature range where two organic phases that can be used for partition extraction safely coexist. It is also important for this proposed application that all these solvents are aprotic.
The Extraction Difficulty in Organic Biphasic Systems
Biphasic organic solvent systems can in principle be very useful for the simple extractive separation of components of a reaction mixture.
The Kilomentor blog emphasizes the significance to chemical and pharmaceutical process development of simple robust scaleable methods of separation and the significance these can have in the teaching of organic chemical process development.
The shortcoming of these biphasic systems is that for most reaction mixtures very components are poorly soluble in the hydrocarbon phase. What is needed is some means to decrease the overall polarity of all or nearly all the components of the target mixture.
Kilomentor proposes here that because the biphasic organic solvent mixtures enumerated above are all aprotic silylation of all the components of the mixture to be separated should decrease their polar and make them partition more competitively into the hydrocarbon phase. Furthermore, because all the solvents in these systems are aprotic the solvents themselves will not interfere with the silylating procedure or the silylating agent.
Disclaimer
Please be warned that this methodology has not been experimentally verified in any situation that I know about. What I can say is it is simple enough to work and I cannot see any particular difficulty.
I have always urged my coworkers to make a clear distinction between facts and theory and this is my effort to do the same.
Making the Silylation Facile
A necessary capability to proceed in this way is a practice method to persilylate all the functional groups in a all the components in a reaction mixture. Another practically important consideration is that the silylation procedure must be inexpensive otherwise the additional reagent cost will make the procedure uncompetitive with more traditional alternatives. Fortunately it has long been known that there are catalysts for silylation, which allow chemists to use the convenient and inexpensive hexamethyldisilazane reagent for effectively all functional groups. Although this has been in the literature many years, it is infrequently used and seems to have today vanished from our chemical toolboxes.
Cornelis A. Bruynes and Theodorus K. Jurriens, then scientists at Gist-Brocades in Delft, The Netherlands published a paper called Catalysts for Silylations with 1,1,1,3,3,3-hexamethyldisilazane in J. Org. Chem. 1982, 47, 3966-3969. They reported that the following compound types could be trimethyl silylated using the title reagent and an appropriate on of their catalysts with yields of typically more than 90%:
Alcohols, phenols, carboxylic acids, hydroxamic acids, carboxylic amides, and thioamides, sulfonamides, phosphoric amides, mon and dialkyl phosphates, mercaptans, hydrazines, amines, NH groups in heterocyclic rings, and enolizable beta diketones
The silylation times were in all cases no more than two hours and the catalyst concentration can be from 0.001-10.0 mole percent.
Catalyst Structures
Although many catalysts are claimed (there is a corresponding patent EP81200771.4 now expired) five were used in the most examples:
· Saccharin [81-07-2]
· Sodium saccharin [128-44-9]
· Bis(4-nitrophenyl)N-(4-toluenesulfonyl)phosphoramidate [81589-21-`]
· Tetraphenylimidodiphasphate [3848-53-1]
· Bis-(4-nitrophenyl)N-trichloroacetyl)phosphoramidate [38187-67-6]
The registry numbers for these catalysts are given in square brackets.
Method of Application
To use this method of separation al that ought to be necessary after reaction is complete would be to
Potential Problems
It will only be determined by actual experiment with a particular mixture of solutes to determine how high a relative concentration of the solutes can be worked with before the biphasic solvent mixture goes homogeneous. Obviously there is some point where the concentration of the solutes will wreck the balance of solvent properties that allows the two phases to coexist
As is always the case if one adds something to promote a separation that facilitating agent must itself be separated in the end. So it is with the catalyst, which must remain in one or the other phase along with some elements of the mixture being separated.
kilomentor | 06 May, 2007 11:37
An important element of the Kilomentor strategy for synthesis and scale up is to enable the separation of crystallisable derivatives that are readily reversible. The uncertainty in a paper synthesis centres upon the simplicity of the work up of the intermediate steps. The importance of intermediates, which are carboxylic acids, amine bases, phenols or other ionizable substances has been stressed. The reversible conversion of alcohols into O-sulfonic acids or phthalate half-ester acids was reviewed in Kilomentor blogs. The formation of complexes of several functional groups, including alcohols, with inorganic salts such as lithium bromide, calcium bromide and calcium chloride was also reviewed.
Carbonyl compounds also form commonly reversible derivatives (oximes and phenyl hydazones for example), which are usually solids, but these derivatives do not have the overwhelming propensity to form that makes them consistently crash out of solution quantitatively and their reversible hydrolysis is something to be worked out rather than a slam dunk.
Aldehydes and ketones do form one type of ionic addition product that seems to crystallize out quickly and dependably but it is sparsely treated in the literature. In 1963, Nelson J. Leonard and Joseph V. Paukstelis reported that treatment of an aldehyde or ketone with the perchlorate salt of a secondary amine led rapidly to the crystallization of tertiary imminium perchlorate salts and the formation of a mole of water. This water could either be left behind at the stage of salt filtration or could be removed azeotropically before the filtration. These authors recognized the reluctance that many would feel to using perchlorate salts and made some tetrafluoroborates but these they found functioned “less efficiently,” Both were “far superior” to other simple anions like chloride, bromide, sulfate or nitrate [J. Org. Chem. 28, 3021 (1963)]. These salts had mps all greater than 99 C with a median mp of 238 C (15 compounds).
Two procedures where provided in the paper and these are repeated here.
A. “To 17.2 g. (0.100 moles ) of pyrrolidine perchlorate in an Erlenmeyer flask was added 11.6 g (0.200 moles) of anhydrous acetone. The pyrollidine perchlorate dissolved immediately and, on swirling. crystals separated with the evolution of heat. After a few minutes the crystals were washed with ether and recrystallized from 2-propanol yielding 20.3 g. (96%) of N-isopropylidenepyrrolidinium perchloriate, m.p. 232-233 C.
Minor variations (note: the acetone was used in 100% excess!) in procedure A included heating the combination of secondary amine salt and carbonyl compound when necessary and using ethanol as solvent to dissolve the secondary amine salt before adding the carbonyl compound. The reaction could be speeded, where necessary, by addition of a few drops of the secondary amine or of a tertiary amine such as triethylamine or pyridine”.
B. “To 18.8 g. (0.100 moles) of morpholine perchlorate were added 19.2 g. (0.200moles) of cyclohexanone (note again 100% excess) and 2 to 3 drops of morpholine. When no reaction was observed, 200 ml of benzene was added and the heterogeneous mixture was heated overnight under reflux, with stirring, while removing water continuously by means of a Dean-Stark trap. The separated solid was collected by filtration, washed with ethanol and ether and dried in vacuo. The product, N-cyclohexylidenemorpholinium perchlorate, 25.2 gm (94%) melted at 237-239 C. Recrystallization from acetonitrile-ether raised the melting point to 239-241 C Ihe use of a Soxhlet extractor containing molecular sieves and a solvent such as chloroform for azeotroping constituted a modification of procedure B, which was successful, for example in combination of pyrrolidonine perchlorate and diethylketone giving the imminium product in 86% yield”.
As I have indicated with my italics, the actual stoichiometry that is essential is not clear from the paper. Although the equation only requires a 1:1 ratio of secondary amine- perchlorate to carbonyl, the general procedures of the examples use two equivalents of carbonyl. Although the authors fail to comment on this, there is a good chance that this stoichiometry was used to drive the reactions rapidly to a 100% conversion. Clearly an excess of the carbonyl is going to be much easier to remove in the crystallization than an excess of the secondary amine perchlorate. It would be very interesting from our perspective to know whether the same fast, high yields can be obtained using some excess of say pyrrolidine tetrafloroborate. As I envision using the precipitation, the formation even of a crude solid mixture of the imminium salt with excess secondary amine salt will allow the filtration and washing away of non-carbonyls. The mixture can then be decomposed by the addition of a tertiary base to set the carbonyls free again. The inorganic salts will dissolve in water and the secondary amine can be extracted from the organic solution with an aqueous acid.
These experiments have not been tried. I commend someone to attempt making the tetrafluoroborate salts with one to one equivalents of a secondary amine and carbonyl but using an excess of amine salt if necessary. This would be useful and well within the technical skill of a beginning undergraduate chemist. (For reasons of insurance do not use perchloric acid!). It would also seem that mixtures of carbonyls might be separable using this method and a proper insufficiency of amine salt as well as mixtures of carbonyl and non-carbonyls.
If someone undertakes this, I would be really interested.
kilomentor | 03 May, 2007 18:47
There really is something addictive about writing the Kilomentor blog and it is not what I expected it to be. Satisfaction comes disproportionately from the occasional feed-back comment I receive. Amazingly it causes me to chack the site every day. I guess I am satisfying a desire to mentor that has been pent up throughout a career spent predominantly in industry. And it isn't just positive feedback I respond to. One reader criticized me about the quality of one posting and on thinking about the comment I judged the statement was fair, so I withdrew that particular piece of writing.
In any case I thought you all should know, I am almost irrationally interested in you reactions. If you have thoughts about subjects I should try to address let me know, and thanks for reading.
Regards,
Kilomentor
| « | May 2007 | » | ||||
|---|---|---|---|---|---|---|
| Su | Mo | Tu | We | Th | Fr | Sa |
| 1 | 2 | 3 | 4 | 5 | ||
| 6 | 7 | 8 | 9 | 10 | 11 | 12 |
| 13 | 14 | 15 | 16 | 17 | 18 | 19 |
| 20 | 21 | 22 | 23 | 24 | 25 | 26 |
| 27 | 28 | 29 | 30 | 31 | ||