The furfural test for cannabis: an evaluation and modification

Sections

Introduction
Experimental
Results and discussion

Details

Author: C.A. LAU-CAM, , J. McDONNELL
Pages: 63 to 68
Creation Date: 1978/01/01

The furfural test for cannabis: an evaluation and modification

C.A. LAU-CAM,
J. McDONNELL
Department of Pharmaceutical Sciences, St. John's University, College of Pharmacy and Allied Health Professions, Jamaica, NY 11439, USA

Introduction

Several colour reactions have been proposed for the detection and identification of cannabis. Grlic [ 1] has indicated that three of the most widely used tests, namely, the alkaline Beam, the Ghamravy, and the Duquenois-Negm tests are prone to positive interferences by non-cannabis plant samples, especially the Ghamravy test. A modification of the Duquenois test variously referred to as the extended Duquenois or Duquenois-Levine test [ 2] has proven more specific than the original version but unfortunately it may still give false positive results with materials other than cannabis [ 2] - [ 4] .

In 1970, Fulton [ 5] described a colour reaction for cannabis resin under the name of the furfural test. A systematic evaluation of this test and its application to cannabis appear to be lacking. In view of the present need for a simple and reliable laboratory procedure for the identification of this material, it was decided to conduct a detailed study of the furfural test which would encompass its suitability as well as its selectivity and sensitivity towards cannabis and a representative group of samples known to interfere with the alkaline Beam and the Duquenois tests.

Experimental

Reagents

Furfural reagent: A 1 per cent solution (w/v) of furfural in ethanol;

Acid reagent: A mixture of 55 ml of sulfuric acid and 45 ml of absolute ethanol;

Alkaline Beam reagent: A 5 per cent solution of potassium hydroxide in ethanol.

Furfural test

Place 1-2 mg of suspect material (preferably powdered) in a test tube and shake with 1 ml of petroleum ether. Transfer the extract to an evaporating dish and evaporate to dryness on a steam bath. To the residue add 3-5 drops of ethanol, 2-3 drops of furfural reagent, and 1-2 drops of hydrochloric acid. Evaporate to dryness on a steam bath and to the residue add 1-2 drops of acid reagent. Observe the appearance of an intense purplish red residue which dissolves in the acid reagent upon swirling of the dish.

Extended furfural test

Proceed as described under "Furfural test". Next, transfer the coloured acidic-ethanolic solution to a small test tube, rinse the evaporating dish with 2-3 drops of fresh acid reagent and combine all the acidic solutions. To the contents of the tube add an equal volume of chloroform and shake vigorously. Allow the phases to separate. A positive reaction is indicated by the appearance of an intense purplish red upper layer and a lighter purplish red lower layer.

Alkaline Beam test

Place 1-2 mg of suspect material (preferably powdered) in a test tube and shake with 1-2 ml of petroleum ether. Transfer the extract to an evaporating dish and evaporate to dryness on a steam bath. To the residue or to 1-2 mg of cannabis resin, add 1-2 drops of alkaline Beam reagent. Observe the development of a violet colour over a period of 30 minutes.

Meta-Duquenois test

In a test tube place 1 mg of suspect material (preferably powdered), an equivalent amount of vanillin, a trace of metaldehyde (about 1/10 the amount of vanillin), 5-6 drops of methanol, and 2-3 drops of hydrochloric acid. Shake well and observe the colour changes over a period of 5 minutes. Add a volume of chloroform equivalent to the test mixture, shake, allow the phases to separate, and record the colours that develop in the upper and lower layers. A greenish blue, blue, purple, or violet colour in the lower layer indicates a positive test.

Results and discussion

Tables I-IV summarize the results of those samples investigated by the furfural test.

Table I presents thc results for those samples which gave a positive furfural test. For comparative purposes, the results of these samples with the alkaline Beam test and the Meta-Duquenois test (a modification of the Duquenois-Levine test) are also included. The alkaline Beam test was conducted according to the procedure described by Novak et al. [ 7] . The Meta-Duquenois test described here is a modification of the procedure suggested by de Faubert Maunder [ 3] .

TABLE I

Positive results

Test

Sample

Furfural
Aspidium, Cannabis, a Cannabinoids (Cannabinol, Cannabidiol, Tetrahydrocannabinol), Cannabis Resin, a Cinchona, Cinnamon, Ergot, Gum Kino, Henna, Hops, Nutmeg, Orris, Sassafras, Tolu, Uva-Ursi, Tea,a Wild Cherry.
Extended furfural
Cannabis, a Cannabinoids, a Cannabis Resin, a Nutmeg, Tolu.
Alkaline Beam
Cannabis (Mexican), Cannabidiol, Cannabis Resin.
Meta-Duquenois
Cannabis, Cannabinoids, Cannabis Resin, Coffee, a Myrrh, Orris.

aSee table IV.

TABLE II

Negative results

Test

Sample

Furfural
Aloe, Bay Leaf, Basil Leaf, Boldo, Buchu, Cascara Sagrada, Catnip, Celery, Coffee,a Eucalyptus, Ginger, Ipecac, Licorice, Marjoram, Nutgalls, Oregano, Parsley, Peppermint, Indian Rhu- barb, Rosemary, Sage, Senna, Spearmint, Strophanthus, Thyme, Tobacco.
Alkaline Beam
Aloe, Aspidium, Cannabis (New York), Cascara Sagrada, Cinna- mon, Ginger, Henna, Hops, Licorice, Myrrh, Orris, Rosemary, Sage, Senna, Tea,a Tolu.
Meta-Duquenois
Aspidium, Cinnamon, Ergot, Eucalyptus, Ginger, Henna, Hops, Licorice, Nutmeg, Oregano, Parsley, Indian Rhubarb, Sage, Sassafras, Strophanthus, Tea,a Thyme, Tolu.

a See table IV.

Table II contains the results for those samples responding negatively to either the furfural, alkaline Beam or Meta-Duquenois tests.

The furfural test proved to be more specific than the Meta-Duquenois test under the conditions described in this report. On the other hand, the alkaline Beam test showed absolute specificity for certain samples of cannabis and its resin. Such high selectivity can be related in part to the fact that only cannabidiol gives the characteristic violet colour in this test [ 9] and in part, to the low sensitivity of the test, being negative with cannabis samples containing low concentrations of cannabidiol but possibly higher concentrations of cannabinol or the biologically active tetrahydrocannabinols [ 6] , [ 8] . Furthermore, when one drop of a cannabidiol solution containing 1 mg/ml of this compound was treated with one drop of alkaline Beam reagent, the violet colour developed very slowly and was less intense than those colours obtained with either the furfural or the Meta-Duquenois tests. Consequently, the alkaline Beam test cannot be relied upon as an ideal one for the identification of cannabis and its products.

In the original furfural test, the appearance of an intense purplish red colour was recorded as a positive reaction. However, it is not clearly specified whether such colouration appeared exclusively in the residue or it was also transferred to the acidic supernatant portion of the test. This detail is quite important and diagnostical. We have tested several non-cannabis samples whose residues responded in a manner similar to cannabis but which either remained insoluble, exhibited a limited solubility, or dissolved in the acidic liquid but rendered it of a colour different than the residue. In the case of cannabis, its resin, and the cannabinoids, the purplish red residue went into solution almost immediately and the supernatant possessed identical chromogenic properties. Therefore, we decided to consider a positive furfural test when the residue as well as the supernatant became coloured in purplish red upon mixing with each other. Interestingly, we observed differences in colour response among the pure cannabinoids listed in table I and between these compounds and crude cannabis preparations, suggesting that the colour reaction must also depend on other cannabis constituents present in the powdered plant or in the corresponding resin.

The extended furfural test was introduced in order to obtain a greater selectivity than the original method. Cannabis and its derivatives coloured both the acidic-ethanol and the chloroform layers in purplish red. In the majority of non-cannabis samples reacting positively to the original test, only the acidic layer exhibited the colours whereas the chloroform layer remained yellowish, greenish, or colourless. The only exceptions were nutmeg and tolu which gave reactions similar enough to cannabis to be interpreted as positive.

In terms of sensitivity, the furfural test appeared to be more sensitive than both the alkaline Beam and the Meta-Duquenois tests. Samples of cannabis testing negatively to either of the latter two tests yielded strong positive reactions with the furfural test. Furthermore, this test was interpreted more readily since the colour appeared immediately upon addition of the acid reagent to the residue. In our hands, the violet colour of the alkaline Beam test may require between 10 and 30 minutes to fully develop.

It has been reported that several pure organic compounds of plant origin, mostly found as components of volatile oils, can react with the alkaline Beam or the Duquenois-Negm [ 1] , or with the Duquenois-Levine [ 10] test. We tested 21 of these substances with the results presented in table III. Only anethole and thymol reacted in a way which may be confused with cannabis materials in in experienced hands. However, the typical purplish red colour of a positive extended furfural test was not evident with any of these substances.

TABLE III

Results of the furfural, extended furfural, and Meta-Duquenois tests with certain organic compounds

Compound

Furfural test

Extended furfural test

Meta-Duquenois test

Anethole
Positive a
Positive b
Negative
Benzaldehyde
Negative
Negative
Negative
Camphor
Negative
Negative
Negative
Carvacrol
Negative
Negative
Negative
Catechol
Negative
Negative
Negative
Cineole
Negative
Negative
Negative
Cinnamic aldehyde
Negative
Negative
Negative
Cinnamyl alcohol
Positive e
Negative
Negative
Citral
Negative
Negative
Negative
Citronellal
Negative
Negative
Positive
Eugenol
Negative
Negative
Negative
Geraniol
Negative
Negative
Negative
Isoborneol
Negative
Negative
Negative
Linalool
Negative
Negative
Negative
Menthol
Negative
Negative
Negative
Phloroglucinol
Negative
Negative
Positive
?-Pinene
Positive d
Negative
Negative
Pyrogallol
Positive e
Negative
Negative
Resorcinol
Negative
Negative
Positive
Thymol
Positive f
Positive g
Negative
Vanillin
Negative
Negative
Not applicable
a Orange red.
 
e Violet brown.
 
b Pink.
 
f Violet brown.
 
c Purplish red.
 
g Light violet
 
d Brownish violet.
     

TABLE IV

Results of the furfural, extended furfural, and Meta-Duquenois tests for cannabis and its derivative, coffee and tea

Sample

Furfural test

Extended Furfural test

Meta-Duquenois test

Cannabis
     
Cannabis, New York
Deep purplish red
Light purplish red
Positive
Cannabis, Mexican - I
Deep purplish red
Light purplish red
Positive
Cannabis, Mexican - II
Deep purplish red
Light purplish red
Positive
Cannabis, Resin - I
Deep purplish red
Light purplish red
Positive
Cannabis, Resin - II
Deep purplish red
Light purplish red
Positive
Cannabidiol
Deep pink
Light pink
Positive
Cannabinol
Deep purplish red
Light purplish red
Positive
Tetrahydrocannabinol
Deep pink
Light pink
Positive
Tea a
     
Chin-Chu, Black
Purple red
Negative
Negative
Lipton, Black
Brownish red
Negative
Negative
Lotus, Green
Deep purple red
Negative
Negative
Maxwell House, Black
Purplish pink
Negative
Negative
Salada, Black
Purplish pink
Negative
Negative
Tetley, Black
Light purple red
Negative
Negative
Coffee a
     
Brim, Ground
Negative
Not applicable
Positive
Brown Gold, Ground
Negative
Not applicable
Positive
Finast, Ground
Negative
Not applicable
Positive
Finast, Instant
Negative
Not applicable
Negative
Maxwell House, Ground
Negative
Not applicable
Positive
Nescafé, Decaffeinated
Negative
Not applicable
Negative
Nescafé, Instant
Negative
Not applicable
Negative
Sanka, Decaffeinated
Negative
Not applicable
Positive
Yuban, Ground
Negative
Not applicable
Negative

a Brand names as found in the United States.

Fochtman and Winek [ 4] have indicated that several types and brands of coffee may react identically to cannabis when tested with the Duquenois-Levine test. Likewise, de Faubert Maunder [ 11] reported that in addition to coffee, various types of tea can mimic cannabis in a modification of the same test. Table IV contains the results of the furfural, extended furfural, and Meta-Duquenois tests for six brands of tea and nine brands of coffee available in the United States. None of these samples behave like cannabis when examined with the extended furfural test.

In summary, the furfural test as described by Fulton is a simple and very sensitive colour reaction for the identification of cannabis and its various products. The extended furfural test significantly enhances the specificity of the original test and allows to distinguish between cannabis and other botanicals that interfered positively with the latter. The furfural reagent was stable for at least four weeks when kept refrigerated at 4 °C. The appearance of the colour was immediate and no waiting period was necessary. All these features offset the greater number of steps and manipulations required when compared with the simpler but less specific and reliable alkaline Beam and Meta-Duquenois tests.

References

001

L. Grlic. Bulletin on Narcotics, XVI: 4 , 29, 1964.

002

W.P. Butler. Journal of the A.O.A.C., 45 , 597 (1962).

003

M.J. de Faubert Maunder. Bulletin on Narcotics, XXI: 4 , 37, 1969.

004

F.W. Fochtman and C.L. Winek. Clin. Toxicol ., 4, 287 (1971).

005

C.C. Fulton. Bulletin on Narcotics, XXII: 2 , 33, 1970.

006

H.J. Wollner, J.R. Matchett, J. Levine and P. Valaer. J. Am. Pharm. Assoc ., 27, 29 (1938).

007

I. Novak, G. Buzas and L. Toth. Pharmazie, 17, 166 (1962).

008

L. Grlic. Bulletin on Narcotics, XIV: 3 , 37, 1962.

009

R. Mechoulam, Z. Ben-Zvi and Y. Gaoni. Tetrahedron, 24, 5615 (1968).

010

C.J. Pitt, R.W. Hendron and R.S. Hsia. J. Forensic Sci ., 17, 693 (1972).

011

M.J. de Faubert Maunder. J. Assoc. Publ. Analysts , 7, 24 (1969).