ON THE APPLICATIONS OF GRIGNARD’S REACTION, 291 
investigators. The filings provided by the Aluminium- and Magnesium- 
Fabrik, Hemelingen, bei Bremen, are very suitable ; according to Klages 
they contain 99-7—99'8 per cent. of magnesium. Evidence is not wanting 
to show that impurities in the magnesium sometimes affect the reaction. 
Beckmann ! shows that calcium may be substituted in certain cases for 
magnesium, but the method does not appear to present any advantages 
over the usual one. 
(2) Ether.—The ether should be thoroughly dried by the following 
method, for example. Wash with water, dry with calcium chloride, filter, 
add sodium wire, distil over sodium wire and then over phosphorus 
pentoxide.? Ullmann and Miinzhuber ® prepare their anhydrous ether by 
shaking 300 c.c. of ordinary ether (sp. gr. 0°722) with a mixture of 25 c.c. 
of concentrated sulphuric acid and 25 c.c. of water and then allowing the 
product to remain for one to two hours over a little sodium wire. 
The presence of the merest trace of water is shown up, in the carrying 
out of a Grignard action, by the slight turbidity which appears when the 
action starts, and which thereafter disappears. 
That the method of drying the ether may have an important influence 
on the progress of the action is indicated by the experience of Ahrens and 
Stapler.‘ In their first communication on the use of dihalogenides those 
authors showed that ethylene dibromide readily interacts with magnesium 
and ether to form the compound MgBr. CH,.CH,Br,(C,H;),0. Since, 
however, on repetition of this experiment with the same samples of mag- 
nesium and ethylene dibromide the chief product was magnesium bromide, . 
it was obvious that the action depended on the state of the ether. Experi- 
ments showed how varying the results were, according to the manner in 
which the ether was dried. 
It should also be remembered that the ether may contain negative 
catalysers, which may either retard the action or prevent it. Bischoff? 
points out, for example, that magnesium is not dissolved by a mixture of 
ethylene dibromide and ether in the presence of a little acetone or aceto- 
phenone when the mixture is allowed to remain for a night at the ordi- 
nary temperature or when it is boiled. This action of a negative catalyser 
very likely accounts for the fact that occasionally one investigator finds 
the preparation of some particular Grignard reagent to proceed with ease, 
whilst another encounters difficulties. 
(3) Halogen compound.—This ought also to be as free from moisture 
as possible. It does no harm to distil before use. 
The ease with which the Grignard reagent is made depends on the par- 
ticular halide used. Sometimes the action begins of its own accord with 
readiness, as in the case of magnesium methyl iodide. Sometimes the 
mixture requires to be heated, but when once started the action proceeds 
to a completion after the source of heat has been withdrawn. In other 
cases the mixture must be continuously heated. Occasionally a catalyser 
is necessary to start the action vigorously. 
It frequently happens that, in addition to the formation of the 
Grignard reagent, a secondary reaction takes place,® namely, the formation 
of saturated hydrocarbons by the action of magnesium on the halide, 
' Ber , 1905, 38, 904. 
* Compare Perkin and Pickles, Zrans. Chem. Soc., 1905, 87, 647. 
8 Ber., 1903, 36, 406. ‘ Thid., 1905, 38, 1296, 3259. 5 Tbid., 1905, 38, 2078. 
® Tissier and Grignard, Compt. rend., 1901, 182, 835; Tschelinzeff, J. Russ. Phys. 
Chem. Soc., 1904, 36, 549; Tiffeneau, Compt. vend., 1904, 189, 481. 
u2 
