Tertiary alcohols, R1 R2 R3.COH, may be synthesised by

Action of the Grignard reagent upon diethyl carbonate

Differentiation between primary, secondary, and tertiary alcohols
(Lucas test).

The test depends upon the different rates of formation of the alkyl chlorides upon treatment with a hydrochloric acid – zinc
chloride reagent * (containing 1 mole of acid to 1 mole of anhydrous zinc chloride) and with hydrochloric acid. It applies only to aliphatic and cycfoaliphatic alcohols.To 1 ml. of the alcohol in a small test-tube, add quickly 6 ml. of Lucas’
reagent at 26-27°, close the tube with a cork, shake, and allow to stand. Observe the mixture during 6 minutes. The following results may be obtained:—
* Lucas* reagent is prepared by dissolving 68 g. (0-5 mole) of anhydrous zinc chloride (fused sticks, powder, etc.) in 62-6 g. (0-6 mole) of concentrated hydrochloric acid with cooling to avoid loss of hydrogen chloride.

(a) Primary alcohols, lower than hexyl, dissolve ; there may be some darkening, but the solution remains clear.
(b) Primary alcohols, hexyl and higher, do not dissolve appreciably ; the aqueous phase remains clear.
(c) Secondary alcohols : the clear solution becomes cloudy owing to the separation of finely-divided drops of the chloride A distinct upper layer is visible after one hour except for iso-propyl alcohol .
(d) Tertiary alcohols : two phases separate almost immediately owinG to the formation of the tertiary chloride . If a turbid solution is obtained, suggesting the presence of a secondary alcohol but not excluding a tertiary alcohol, a further test with concentrated
hydrochloric acid must be made. Mix 1 ml. of the alcohol with 6 ml. of concentrated hydrochloric acid, and observe the result : —
(e) Tertiary alcohols : immediate reaction to form the insoluble chloride which rises to the surface in a few minutes.

CRYSTALLINE DERIVATIVES OF ALIPHATIC ALCOHOLS

The acid chloride is available commercially, but it is more economical to
prepare it from the acid as and when required. Furthermore, 3 : 5-dinitrobenzoyl
chloride tends to undergo hydrolysis if kept for long periods,
particularly if the stock bottle is frequently opened. The substance may,
however, be stored under light petroleum.

Method
. Mix 1-0 g. of 3 : 5-dinitrobenzoic acid with 4 ml. of thionyl chloride in a dry 50 ml. conical flask ; fit a reflux condenser, carrying a plug of cotton wool at the upper end, into the flask and heat on a water bath for 15-30 minutes, Remove the condenser and heat the flask in a boiling water bath (until the excess of thionyl chloride has evaporated. Use the resulting 3 : 5-
dinitrobenzoyl chloride (about 1-0 g.) immediately. Add 0-5-1 ml. of the alcohol, cork the flask loosely, and heat on a water bath for 10 minutes : secondary and tertiary alcohols require longer heating (up to 30 minutes). Cool the mixture, add 10 ml. of 5 per cent, (or saturated) sodium bicarbonate solution, break up the resulting solid ester with a stirring rod (alternatively, stir until crystalline), and filter at the pump ; wash with a little sodium bicarbonate solution,
followed by water, and then suck as dry as possible. Dissolve the crude

Treat 1 ml. of the alcohols enumerated
in (iv) cautiously with 0-5-0-7 ml. of acetyl chloride. Observe the
reaction which occurs. After 2-3 minutes, pour the contents of the
various test-tubes into 3 ml. portions of water, neutralise the aqueous
layer with solid sodium bicarbonate, and examine the residual liquids
for odour and density (relative to water).

Treat 2 ml. of absolute methyl alcohol with a small thin slice of dry, freshly-cut sodium Observe the result. Cool the solution when all the sodium has reacted. Add a little water and test the solution with litmus paper. Obtain five small dry test-tubes and introduce 1 ml. of the following alcohols into each : ethyl alcohol, n-butyl alcohol, sec.-butyl alcohol, q/c/ohexanol and tert.-butyl alcohol. Add a minute fragment of sodium to each and observe the rate of reaction. Arrange the alcohols in the order of decreasing reactivity towards sodium. The reaction with sodium is by no means an infallible practical test for alcohols since, strictly speaking, it is applicable only to pure anhydrous liquids. Traces of water, present as impurities, would give an initial evolution of hydrogen, but reaction would stop after a time if an alcohol is absent: furthermore, certain esters and ketones also evolve hydrogen when treated with sodium. It may, however, be assumed that if no hydrogen is evolved in the test, the substance is not an alcohol.