On some novel organometallic compounds of magnesium and their application to synthesis of alcohols and hydrocarbons. [1]
Note by V. Grignard, presented by H. Moissan.
Originally published as "Sur quelques nouvelles combinaisons organomètalliques du magnèsium et leur application à des synthèses d’alcools et d’hydrocarbures." Comptes Rendus, (1900) v. 130, p. 1322.
Following the syntehsis of dimethylheptenol by Barbier [2], where Saytzeff's method was employed, but with magnesium replacing zinc, I proposed to study what advantages this substitution may present. In the course of this research, I discovered a series of organometallic compounds of magnesium which permitted me to notably modify the Wagner-Saytzeff method, with a significant increase in speed and regularity of the reaction and, in general, of the yield obtained.
Cold methyl iodide attacks magnesium turnings only very slowly, but if one adds a small amount of anhydrous ether, a reaction begins immediately which becomes extremely rapid without delay. It is necessary therefore to refrigerate and add an excess of ether to the reaction. Under these conditions, the dissolution of magnesium follows rapidly and eventually one obtains a very inviscid and weakly colored liquid, without an appreciable quantity of solid deposits.
If one attempts to boil away the ether, a semicrystalline grayish mass remains which absorbs humidity very rapidly and deliquesces with concomitant self-heating. But the main advantage of the obtained compound is that it is not necessary to isolate it. In fact, using the aforementioned ethereal solution (containing a precise 1:1 atom to molecule ratio of magnesium to methyl iodide), if one adds 1 molecular equivalent of aldehyde or ketone, a rapid reaction is produced, and after decomposing the mixture with acidulated water, one isolates a secondary or tertiary alcohol corresponding to a yield of around 70%.
I observed that hydrobromic and hydriodic esters [ I ] of monatomic fatty [ II ] and saturated alcohols up to C5, as well as benzyl bromide, gave the same reaction, and it is likely that the reaction must take place with many other halogenated esters.
For reasons that I shall divulge below, I believe that it is possible to attribute the formula RMgI or RMgBr to the organometallic compounds that I have obtained, where R is a fatty or aromatic alcoholic residue. The reactions that I have outlined above can be described in the following manner:
I have thus prepared a certain number of secondary and tertiary alcohols, some of which are already known, but it is interesting to prepare them anew to ensure the generality of the method. With fatty halogenated esters, the results have always been excellent; they are much less so with benzyl bromide, which is accompanied by the relatively abundant formation of bibenzyl. Among the novel alcohols that I have obtained, I will mention:
Phenylisobutylcarbinol [ III ], by the reaction of benzaldehyde with isobutylmagnesium bromide, colorless liquid, viscous, weak odor, boiling at 122° under 9 mmHg and whose analysis gives the following values:
Dimethylphenylcarbinol [ IV ], by reaction of acetophenone with methylmagnesium iodide, colorless viscous liquid, agreeable odor, boiling at 93°-95° under 10 mmHg.
Dimethylbenzylcarbinol [ V ], by reaction of acetone with benzylmagnesium bromide, colorless liquid, somewhat viscous with weak odor, boiling at 103°-105° under 10 mmHg.
When using unsaturated aldehydes and ketones where the double bond neighbors the functional group, the alcohol which is formed is often unstable; it dehydrates upon distillation, even in vacuum, and one only obtains a diene.
In this manner, by reacting mesityl oxide with methylmagnesium iodide, I obtained a hydrocarbon boiling at 92-93° at 750 mmHg and which matches the formula
that is, 2,4-dimethyl-2,4-pentadiene.
I am pursuing the study of applications of these new organometallic compounds.
Notes Translator's Notes
Originally published as "Sur quelques nouvelles combinaisons organomètalliques du magnèsium et leur application à des synthèses d’alcools et d’hydrocarbures." Comptes Rendus, (1900) v. 130, p. 1322.
Following the syntehsis of dimethylheptenol by Barbier [2], where Saytzeff's method was employed, but with magnesium replacing zinc, I proposed to study what advantages this substitution may present. In the course of this research, I discovered a series of organometallic compounds of magnesium which permitted me to notably modify the Wagner-Saytzeff method, with a significant increase in speed and regularity of the reaction and, in general, of the yield obtained.
Cold methyl iodide attacks magnesium turnings only very slowly, but if one adds a small amount of anhydrous ether, a reaction begins immediately which becomes extremely rapid without delay. It is necessary therefore to refrigerate and add an excess of ether to the reaction. Under these conditions, the dissolution of magnesium follows rapidly and eventually one obtains a very inviscid and weakly colored liquid, without an appreciable quantity of solid deposits.
If one attempts to boil away the ether, a semicrystalline grayish mass remains which absorbs humidity very rapidly and deliquesces with concomitant self-heating. But the main advantage of the obtained compound is that it is not necessary to isolate it. In fact, using the aforementioned ethereal solution (containing a precise 1:1 atom to molecule ratio of magnesium to methyl iodide), if one adds 1 molecular equivalent of aldehyde or ketone, a rapid reaction is produced, and after decomposing the mixture with acidulated water, one isolates a secondary or tertiary alcohol corresponding to a yield of around 70%.
I observed that hydrobromic and hydriodic esters [ I ] of monatomic fatty [ II ] and saturated alcohols up to C5, as well as benzyl bromide, gave the same reaction, and it is likely that the reaction must take place with many other halogenated esters.
For reasons that I shall divulge below, I believe that it is possible to attribute the formula RMgI or RMgBr to the organometallic compounds that I have obtained, where R is a fatty or aromatic alcoholic residue. The reactions that I have outlined above can be described in the following manner:
I have thus prepared a certain number of secondary and tertiary alcohols, some of which are already known, but it is interesting to prepare them anew to ensure the generality of the method. With fatty halogenated esters, the results have always been excellent; they are much less so with benzyl bromide, which is accompanied by the relatively abundant formation of bibenzyl. Among the novel alcohols that I have obtained, I will mention:
Phenylisobutylcarbinol [ III ], by the reaction of benzaldehyde with isobutylmagnesium bromide, colorless liquid, viscous, weak odor, boiling at 122° under 9 mmHg and whose analysis gives the following values:
| Found: | Theoretical for C6H5CH(OH)C4H9: |
|---|---|
| C=80.26 | C=80.49 |
| H=9.85 | H=9.76 |
Dimethylphenylcarbinol [ IV ], by reaction of acetophenone with methylmagnesium iodide, colorless viscous liquid, agreeable odor, boiling at 93°-95° under 10 mmHg.
| Found: | Theoretical for C6H5CH(OH)(CH3)2: |
|---|---|
| C=79.69 | C=79.41 |
| H=8.61 | H=8.82 |
Dimethylbenzylcarbinol [ V ], by reaction of acetone with benzylmagnesium bromide, colorless liquid, somewhat viscous with weak odor, boiling at 103°-105° under 10 mmHg.
| Found: | Theoretical for C6H5CH2C(OH)(CH3)2: |
|---|---|
| C=79.82 | C=80.00 |
| H=9.50 | H=9.33 |
When using unsaturated aldehydes and ketones where the double bond neighbors the functional group, the alcohol which is formed is often unstable; it dehydrates upon distillation, even in vacuum, and one only obtains a diene.
In this manner, by reacting mesityl oxide with methylmagnesium iodide, I obtained a hydrocarbon boiling at 92-93° at 750 mmHg and which matches the formula
that is, 2,4-dimethyl-2,4-pentadiene.
| Found: | Theoretical for C7H12: |
|---|---|
| C=87.10 | C=87.50 |
| H=12.73 | H=12.50 |
I am pursuing the study of applications of these new organometallic compounds.
Notes Translator's Notes
- Hydrobromic and hydriodic esters — An archaic term for alkyl bromides and iodides. An ester in general is the product of the condensation reaction between an acid and an alcohol; cf. sulfate and phosphate esters as well as the standard carboxylic esters.
- "Fatty" in this context refers to alkyl—or more generally, non-aromatic—hydrocarbons.
- Phenylisobutylcarbinol — This naively translates to 3-methyl-1-phenyl-1-butanol, but it is unclear whether the starting reagent isobutyl bromide refers to 1-bromo-2-methylpropane or, more likely, 2-bromo-2-methylpropane (tert-butyl bromide in modern chemical parlance). If the latter is the case, then the synthesized product of the Grignard reaction is 2,2-dimethyl-1-phenyl-1-propanol.
- Dimethylphenylcarbinol — Also known as 2-phenyl-2-propanol.
- Dimethylbenzylcarbinol — Also known as 2-methyl-1-phenyl-2-propanol.