A simple sensitive specific field test for cocaine based on the recognition of the odour of methyl benzoate as a test product


Rationale for a test for cocaine
Test description


Author: Fred W. GRANT, William C. MARTIN, Ralph W. QUACKENBUSH
Pages: 33 to 35
Creation Date: 1975/01/01

A simple sensitive specific field test for cocaine based on the recognition of the odour of methyl benzoate as a test product *

William C. MARTIN
Research Division, Marcy Psychiatric Center, Marcy, New York 13403, U.S.A.

A number of field tests for cocaine have recently been described (1, 2, 3, 4) all based on the blue colour arising from the addition of a cobalt thiocyanate reagent to the test substance. The method of Scott (3) was recently given prominence in a United States Department of Justice publication (5). This test was found to be specific to cocaine among a group of 18 commonly encountered drugs. It requires the use of three reagents in sequence, including concentrated hydrochloric acid, and the recognition of colour changes, the formation and redissolving of a precipitate, and the colouration of an immiscible organic phase. The specificity of this test has already been brought into question (4) and we suspect more instances of this type will follow.

We had reason to explore the specificity of 2 per cent aqueous cobalt thiocyanate in drug testing and found 22 of 55 representative drugs to give a pronounced blue colour indistinguishable from that derived from cocaine. Twelve drugs gave a weaker blue colour. Acetone, alcohol, and certain soaps and detergents also gave a positive test. The reagent appears to be responding to the hydrophobicity of the test substance, among other properties, since a correlation generally prevails between test result and the partition of the drug between an aqueous and immiscible organic phase. Positive responding drugs will test negative if sufficient water is present, either in the reagent or in the test specimen.

Another anticipated failing of the Scott test is in dealing with cocaine specimens containing insoluble excipients. As previously noted, the test involves the recognition of a liquid-liquid phase boundary and the formation and redissolving of a precipitate. This could prove difficult if not impossible to do with certain specimens.

For the above reasons, we feel that tests for cocaine based on cobalt thiocyanate will continue to show an unacceptable incidence of false positives and false negatives despite modifications in test conditions.

Rationale for a test for cocaine

As is typical of many drugs, cocaine's biological activity is not matched by a comparable degree of chemical reactivity. However, it is quite unique among commonly abused drugs in being a benzoate ester. While colour tests for this functional group are not available, the lower alkyl esters of benzoic acid have quite distinctive odours detectable at very low concentrations relative to the average colour test. The transfer of the benzoate function of cocaine from methyl ecgonine to methanol is readily accomplished in the presence of dry methanolic potassium or sodium hydroxide as is shown in the figure above. Evaporation of the excess methanol leaves a residue containing methyl benzoate (Oil of Niobe) readily apparent by its odour. The test will not work in the presence of water but it is sufficiently specific to be unaffected by the presence of other drugs or excipients. Since the test conditions are those of transesterification rather than esterification, benzoic acid itself does not respond.

*Presented by FWG at the Sixth Northeast Regional American Chemical Society Meeting, Burlington, Vermont, 19 August 1974.

Full size image: 8 kB

The odour of methyl benzoate arising from the test being applied to acetate esters such as aspirin or heroin might have led to ambiguities in test interpretation but for the fact that methyl acetate is more volatile (B.P. 57°) than methanol (B.P. 65°) and will dissipate during the solvent evaporation step. The odour of methyl benzoate (B.P. 198°) should persist for hours at normal temperatures.

Tests based on odour detection are generally suspect if subtle qualitative or quantitative distinctions between odours are called for. However, distinctions of this type are not required in the test described here. It is being recommended for the following reasons:

  1. Over one hundred of a wide variety of drugs (6) were tested and found to give no odour of any kind. Cocaine and piperocaine alone were benzoate esters and therefore gave a positive test. Occasionally a weak fishy odour was noted when the test reagent released a low molecular weight amine, such as amphetamine, from its salt.

  2. Test sensitivity and specificity exceed those of existing field tests for cocaine. This allows cocaine to be detected in low concentration in the presence of excipients or other drugs. Water will interfere with the test so reagent and specimen should be kept dry.

  3. The simplicity of the test makes it admirably suited to field testing.

  4. It has proven to be the test of choice for the field identification of cocaine by local County and State Police.

Test description

Reagent. One gram of sodium hydroxide or potassium hydroxide is dissolved in 20 mls. of dry methyl alcohol. This reagent should serve its purpose for many months if kept in a tightly stoppered container to minimize methyl alcohol evaporation and access of moisture.

Procedure. The dry test specimen is thoroughly moistened with the reagent on a spot plate or other non-corroding surface, a few minutes allowed for the evaporation of the excess alcohol (heating is not required), and the odour noted.


This work was supported by the New York State Department of Mental Hygiene. We wish to thank Investigator William Clark of the New York State Narcotic Control Bureau, Senior Investigator James McKaig of the New York State Police, and Lt. Robert Ingalls of the Oneida County Sheriff's Department for supporting and assisting in this work.



R. Ruybal, U.S. Justice Department's "Microgram," Volume 5, No. 3 (1972).


C.N. Ruybal, ibid., Volume 6, No. 2 (1973).


L.J. Scott, Jr., ibid., Volume 6, No. 11 (1973).


S. K. Lorch, ibid., Volume 7, No. 8 (1974).


"Drug Enforcement," Volume 1, No. 3, pp. 26-27 (1974).


These drugs were essentially those listed by Bastos, et al ., Clin. Chem ., 16, 931 (1970).