The Advantages of Ethanol-Cyclohexane Mixtures as Organic Reaction Solvent Media

In designing process steps for fine chemical synthesis a bias has existed against multi-component solvent systems. In the past, it was argued that using combinations of solvents meant that money would be lost because it would be necessary to separate these solvents in a recovery step. What was not then properly recognized was that recovery of solvents from mixtures for reuse was rarely undertaken in the fine chemical and pharmaceutical industries mainly because recovering and recycling solvents required extensive expensive analytical work to prove that the specifications were being consistently met.

Here I will look at combinations of cyclohexane and absolute ethanol.

There is a contradiction within the literature concerning the miscibility or immiscibility of ethanol and cyclohexane. This may be understandable because the UCST (upper critical solution temperature ) of the combination is reported to be -16℃. In the laboratory, it is difficult to maintain the contents of a separatory funnel even at a temperature less than 0℃ without special cooling but this should be much less a problem in the plant where extractions are conducted in a reactor completely surrounded by a heating/cooling jacket and the entire charge is maintained under inert gas throughout. Consequently one might expect that two phases would not be seen in the laboratory but are reasonably easily achieved with the more effective plant cooling systems.

Ethanol and cyclohexane do have an azeotrope ( bp 64.9℃ ) that boils significantly below the boiling point of either pure ethanol ( bp 78.5℃ ) or pure cyclohexane ( bp 81.4℃ ). Mixtures in any proportion of these two pure liquids will give a homogeneous reaction medium above -16℃ and fractional distillation of the reaction mixture will remove the azeotropic composition and leave the non-volatile substrates in either ethanol or cyclohexane depending upon what solvent predominated in the starting mix.

Alternatively, the reaction mixture could be cooled to below -16℃ and whether two liquid phases separate or not and if so how substrates partition between them. Unfortunately, these two have close to the same densities so separation of phases can be expected to be slow.

The Advantage of Methanol-Hydrocarbon Solvent Mixtures for Organic Synthesis

This blog article is somewhat speculative. It is not based on experimental data. However, the information about the compositions and boiling points of the hexane, cyclohexane, and heptane azeotropes with methanol are accurate as are the upper critical solution temperatures (UCSTs).

In designing process steps for fine chemical synthesis a bias has existed against multi-component solvent systems. In the past it was argued that using combinations of solvents meant that money would be lost because it would be necessary to separate these solvents in a recovery step. What was not then properly recognized was that recovery of solvents from mixtures for reuse was rare in the fine chemical and pharmaceutical industries mainly because recovering and recycling solvents required extensive expensive analytical work to prove that the specifications were being consistently met.

KiloMentor is of the opinion that using solvent mixtures has so many advantages that they should be considered more frequently than not. In particular mixtures of lower alcohols and various hydrocarbons that provide protic media of a wide range of polarities and dielectric constants should be among the first systems considered.

Utilizing multi-component solvent systems turns the solvent composition into a continuous rather than discrete variable in the process optimization. This does not rule out eventually find a process step optimum that is !00% of one of your original solvent combination.

If a lower proportion of solvent can be used in a chemical process step the the amount of material that can be processed in a single batch is increased and a higher throughput achieved for the step. This will lead to cost savings for a campaign that includes that step. 

It is often the case that a homogeneous mixture of polar and  non-polar solvent liquids that is homogeneous is better at dissolving a substrate that has distinct polar and apolar subdomains than either pure liquid alone. That would mean, mutatis mutandis, that such mixtures should increase the throughput in reactions of such substrates.

Specifically, this suggests that a homogeneous single phase mixture of a hydrocarbon and methanol could realistically be better at dissolving some reaction substrates and producing more concentrated solutions that give higher throughputs.

Hexane, cyclohexane, and heptane all give constant boiling binary azeotropes with methanol and these azeotropic compositions fulfil the criteria of having both a polar and a non-polar component and being homogeneous at the azeotropes boiling point. 

Reaction mixtures of the azeotropic compositions in each of these cases would be obtained by mixing a chosen hydrocarbon and methanol in their correct proportions but with a consistent biased slight excess of methanol, which has the lowest boiling point among these liquids. Thus, when any such mixture is brought to reflux the excess methanol will be distilled over first  before the refluxing settles at the actual constant boiling point of each particular azeotrope.

The upper critical solution temperature (UCST) designates the temperature above which the two pure solvents form a single liquid phase. In each case, whether cyclohexane, hexane, or heptane, is paired with methanol, the upper critical solution temperature  of  each azeotropic composition is below that azeotrope’s boiling point. That is to say, refluxing will maintain a single homogenous phase which can serve as a homogeneous reaction medium.

Besides potentially providing a throughput advantage these particular azeotropes offer something else. When these mixtures are cooled below their  UCSTs, as will occur when the reaction is complete and is cooled for quenching, work-up, separation and/or purification, two liquid layers are expected to separate. This might prove useful because it provides a ‘natural’ ‘free’ phase switching extraction which might simplify the work-up.

But let us not delude ourselves about how frequently this will be useful. Neither of these two phases is predominantly methanol or hydrocarbon. 

The azeotrope between hexane and methanol has bp.50℃. The composition of the upper layer will be 85% hexane and 15% methanol with specific gravity 0.675 and the composition of the lower layer will be 42% hexane and 58% methanol with specific gravity 0.724. It is decidedly not upper almost pure hexane; lower mostly methanol.

The UCST of methanol with n-hexane is only 35℃. That means that the partitioning of a substrate between these solvents can be accelerated by heating above 35℃ where a single phase can be formed, then cooled down so that the phases separate as a lot of small bubbles with lots of surface area. There are 15 Cº between the azeotropic boiling point and the UCST.

With heptane and methanol  an UCST  is reported to occur at 51℃ but handbooks do not report any separation into separate phases on cooling. This is contradictory and needs to be examined experimentally. The azeotrope is reported to have bp. 59.1℃.

With cyclohexane and methanol the UCST is 45℃. The azeotrope bp. is 45.2℃. As you would expect the azeotrope separates into two phases in the receiver. There is essentially no point where we can exploit a single homogeneous liquid phase. 

Whether two liquid phases separate upon cooling a reaction mixture using one of these azeotropic systems and whether any two phases that do separate are useful for partitioning reaction mixture components, in every case adding a bit of water will cause the compositions of the two phases to shift- the methanol phase becoming more nearly essentially methanol and the hydrocarbon phase more nearly all hydrocarbon. This more definitely will be useful!

When Scaling Up a Synthetic Organic Intermediate that is being Purified by Distillation

If an intermediate is being worked up, isolated, and purified by distillation as part of developing a chemical process for synthesis at scale that should trigger consideration in its synthesis for using a high boiling solvent which can act as a chaser during the contemplated large-scale distilling.

The volume of this ‘chaser’ phase should be sufficient to completely occupy the ‘minimal storable volume’ in the reactor contemplated in the eventual scale-up.

This will almost always involve replacing a traditional lower boiling solvent as part of the modifications of a literature model example. These chaser phases must almost always be acceptable in cost to the solvents they replace. Because product is lost whenever it is essentially the highest boiling part of the reaction mixture because some material must always remain boiling in the still pot even at the end of the fractionation, using an appropriate chaser will save money that will be partially lost from whatever extra cost is involved in using the chaser.

Mixtures of Acetonitrile, Ethylene Glycol, and Water in a Liquid-Liquid Extraction of Polar Impurities from Mixtures in Hydrocarbon Solvents

 

It is quite well known that acetonitrile and any hydrocarbon liquid upon mixing together will separate into two layers. It is very much more poorly recognized that the polarity of the acetonitrile can be tweaked by the addition of either water (5-20%) or ethylene glycol (5-40%) or a more finely refined tertiary combination of these while not disturbing the separability of the two phases, which remain essentially immiscible. Furthermore, because all of the hydrocarbon solvents have lower densities than any of acetonitrile, ethylene glycol, or water, the hydrocarbon liquid phase will consistently be the top-most layer.

How can this be practically useful? Using a small amount of an appropriately designed immiscible polar liquid phase in this way, a mixture in a hydrocarbon solution can be freed from polar impurities by multiple extractions.

This could be much more efficient, not to mention simpler and faster, than crystallizing a product from a mixture that still contains such polar impurities.

The research upon which this suggestion is based is Leshchev S.M.; Rumyantsev, I. Yu. Zh. Prikl. Khim 1992, 65(6), 1332-6 identified in CA 118: 88569f.