Differentiation and comparison of raw, prepared and dross opium


Methods and materials
Results and discussion


Author: Han-Yong LIM, Sau-Fong KWOK
Pages: 31 to 41
Creation Date: 1981/01/01

Differentiation and comparison of raw, prepared and dross opium

Han-Yong LIM
Sau-Fong KWOK
Department of Scientific Services, Singapore, Republic of Singapore


Comparative studies involving colour tests, thin-layer chromatography (TLC) and gas-liquid chromatography (GLC) were made on raw, prepared and dross opium with a view to establishing means for their differentiation. Analysis of a large number of samples found in south-east Asia showed the typical patterns and results to be quite distinctive in the qualitative and quantitative aspects. The relative proportions of the major alkaloids in the two chromatographic systems were used to advantage in the comparison. Prepared opium was found to have a characteristic yellow fluorescent spot on the TLC plate when viewed under long-wavelength ultraviolet light. No positive marker was found for dross opium, its unusual features being the relatively high morphine and codeine content and the absence of narcotine, meconic acid and especially thebaine. Most of the prepared opium samples were found to have been made from a mixture of raw and dross opium. The modified hallucinogen reagent of Maunder was found to be a suitable alternative chromogenic spray in the TLC analysis.


Although many papers have been written on the analysis and identification of opium (usually understood to mean raw opium) in general, little can be found in the literature on the differentiation and comparison of the three types of opium, namely, raw, prepared and dross opium. The differentiation is of interest not only because it can be of forensic importance, but also because legislation in some countries makes a distinction between raw and prepared opium (often dross is classified under prepared opium), with more stringent penalties attached to one or the other. In Singapore the present Misuse of Drugs Act, which came into force in 1973, does not make such a distinction, although the preceding Dangerous Drugs Act did. Visual appearance, odour, water solubility and personal experience have for a long time played an important part in such comparisons. In 1975, de Faubert Maunder [ 1] made use of the well-known colour reaction in which raw opium gives a red colour with warm mineral acid, to distinguish it from prepared opium, which does not give such a colour with the acid. The reaction is due to the unique red-turning alkaloid, porphyroxine, discovered by Merck [ 2] [ 3] and later used in the identification of raw opium and the determination of its origin [ 4] [ 5] [ 6] .

Prepared opium, also known as chandu in Singapore, normally means raw opium that has been treated in such a way as to make it suitable for smoking or consumption. However, to increase the bulk and to cut down on cost without resorting to adulteration with foreign substances, which would have an effect on the taste and flavour, clandestine manufacturers have for a long time been mixing raw opium with dross in making prepared opium. Dross is the residue (usually dry and granular) left in the pipe after prepared opium has been smoked. The method of smoking and the type of pipe used determine the quantity of dross produced. On the average it is about half of the amount of prepared opium smoked. Thus, with a large amount of dross available in places wbere opium smoking is practised, it is not surprising that the dross residue is retained for reuse. With raw and dross opium as the primary materials, prepared opium may be made in three ways, namely from raw opium alone, a mixture of raw and dross opium and dross opium alone, giving rise to what we will term "raw prepared opium", "mixed prepared opium" and "dross prepared opium", respectively. The preparation involves boiling or cooking the substance or mixture for several hours in water, straining to remove insoluble solids and concentrating the filtrate to a sticky consistency. ln this study, raw prepared opium was rarely encountered, while occasional instances of dross prepared opium were seen. Almost all the prepared opium samples analysed belonged to the mixed variety. Therefore the term "prepared opium" refers to the prepared opium samples as received from enforcement officers, which were invariably of the mixed variety. All varieties of prepared opium were alike in appearance, solubility and odour but distinguishable from each other when analysed. The mixing of dross and raw opium to produce prepared opium seems to be a common practice, not only in south-east Asia, but also in other parts of Asia where opium smoking is practised [ 7] [ 8] [ 9] . Another form of prepared opium, known as madhak. in which some charred vegetable matter such as barley husk or leaves of Acacia arabica is added to the paste, is also reported in some parts of India and Pakistan. The addition of such inert diluents would have no effect on the relative proportion of the major alkaloids in the prepared opium.

Methods and materials

Colour tests

Preliminary examination by means of the appearance, odour, solubility in water and colour spot tests as listed in [ 1] were carried out to indicate the presence of opium. There was little difficulty in deciding to which type of opium the sample belonged, unless it was present in small quantity or in the dry, powdered form.

A small amount, about 0. 1 g, was shaken in 1 ml of each of two solutions: (a) 5% aqueous ferric chloride, (b) warm 2N hydrochloric acid. For comparison, similar tests were also conducted on meconic acid, papaveraetum BPC and on raw, mixed and dross prepared opium as described in the introduction. The results are shown in table 1.

Table 1

Colour tests for opium samples

Nature of sample

Colour with 5% FeCl 3

Colour with 2N HCL 3

Raw opium
Brown-red a (with insoluble matter)
Red a (with insoluble matter)
Prepared opium
Brown-red a
Dross opium
Light brown (with insoluble matter)
Brown (with insoluble matter)
Meconic acid
Brown-red a
Papaveraetum BPC
Light green
Raw prepared opium
Brown-red a
Mixed prepared opium b
Brown-red a
Dross prepared opium
Light brown
a Positive reaction

aPositive reaction.

b Raw/dross ratio = 5/3

Thin-layer chromatography (TLC)

The experiment utilized the following items:

Apparatus: chromatographic glass tanks lined with Whatman filter paper

Layer: Merck TLC plastic roll, 0.2 mm, silica gel 60 F 254 or equivalent

Solvent: ethyl acetate/methanol/ammonia (17/2/1), equilibrated for 10-16 h in the tank before use

- Chromogenic sprays: (a) 6N hydrochloric acid, (b) acidified iodoplatinate (0.25 g platinic chloride, 5 g potassium iodide and 2 ml concentrated-hydro-chloric acid in 100 ml water), (c) 2.5 g 4-dimethylaminobenzaldehyde (DAB) in 100 ml of an equal mixture of methanol and phosphoric acid

Technique: ascending, saturated chamber, 17 cm; development, 45 min

For comparison, 0.1 g of the dry sample was triturated with 1 ml of ethanol, and 3 μl of the solution (= 0.3 mg of the sample) was spotted on each of two plates. The solvent system used was that of Davidow [ 10] for drugs of abuse. After solvent migration and drying, the plates were viewed under ultraviolet (uv) light at 254 nm and 366 nm. Under long-wavelength uv light, prepared opium showed a strong yellow fluorescent spot (R f = 0.12) between the origin and the morphine spot.

One plate was sprayed with reagent (a)and heated in an oven at 105°C for a few minutes. Raw opium revealed a red spot (R f = 0.66) due to the presence of porphyroxine. The plate was then sprayed with reagent (b) to reveal blue to purple spots indicating the major alkaloids of opium.

The second plate was sprayed with reagent (c), and after 10 min the red spot characteristic of raw opium appeared. The plate was then heated at 105°C for 15 min. Reddish spots were seen, including morphine and codeine which appeared as red-brown and bright-pink spots respectively. The plate was also observed under long-wavelength uv light for the presence of the yellow fluorescent spot of prepared opium. Further heating at the same temperature for another hour or so revealed thebaine and narcotine as brown-purple and yellow-orange spots respectively. Raw opium also showed minor dark spots below the morphine and above the narcotine spots which fluoresced in long uv light when viewed immediately after heating. (if the plate was allowed to stand overnight and again heated for 15 min, the thebaine spot was found to acquire a strong permanent purple-brown colour.) These coloured spots were observed to fade on cooling to room temperature but reappeared when reheated. The plate could further be oversprayed with reagent (b) and, after 20 min, dark purple spots on a greyish background were seen. Figure 1 and table 2 show the results of the TLC analysis.

Figure I TLC chromatograms of opium sales

Full size image: 2 kB, Figure ITLC chromatograms of opium sales

Table 2

TLC patterns for raw, prepared and dross opiums


Marker spot

R f


Opium type

Fluorescent spot
Yellow fluorescence (uv 366 nm)
Red a
Brown-red b
Raw, prepared, dross
Pink b
Raw, prepared, dross
Brown c
Raw, prepared
Faint purple
Not detected
Raw, prepared, dross (all very weak)
Not detected
Violet d
Yellow-orange b, c
Raw, prepared

aAfter hydrochloric acid spray followed by heating

bFades on cooling, reappears on heating

cIntensifies on prolonged heating

dAppears 10-15 min after spraying

Gas-liquid chromatography (GLQ

A Pye Unicam GCD chromatograph fitted with a flame ionization detector and a glass column (1.5 m x 4 mm ID) containing a packing of 3% OV-17 on 100-120 mesh diatomite CQ was used. The operating parameters were as follows: nitrogen flow rate 60 ml/min, air pressure 1. I kgf/cml (1.1 bar), hydrogen pressure 1.5 kgf/cm [ 2] (1.5 bar), attenuation 80, recorder chart speed 160 mm/h. The GLC unit was set to operate isothermally at 250°C for 12 min from the time of injection, then raised to 280°C at 50°C/min and held at this temperature until the end of the analysis (total time 35 min). The injection point and detector temperatures were 320°C and 300'C, respectively. Analytical grade morphine, codeine, thebaine, papaverine, narcotine and buclizine were obtained from the Singapore Pharmaceutical Department and the United Nations Narcotics Laboratory.

For qualitative comparison, about 0.5 g of the dry powdered sample was triturated with a few millilitres of methanol. The contents were then washed into a stoppered 50-mi flask with methanol and adjusted to 40 ml. The flask was agitated for 2 h, the solution filtered, the first 5 ml being discarded, and 2 μl of the filtrate was injected into the GLC (equivalent to 25 μg of the dry sample). Figure If shows the typical chromatograms of raw, prepared and dross opium and figure III those of raw, mixed and dross prepared opium.

For quantitative estimation the above procedure was followed, except that the weighing was done accurately and the sample shaken for 8 h and left to stand overnight. Calibration graphs were prepared by chromatographing standard solutions of mixtures of the five alkaloids at different concentrations, using the internal standard, buclizine. The sample was then similarly quantified by adding the internal standard to the methanolic solution of the sample. Fifteen samples each of raw, prepared and dross opium were analysed; the results are presented in table 3. Buclizine was used as the internal standard as it appeared in the clean area between the narcotine peak and peak Z (figure 11). The retention times in minutes for codeine, morphine, thebaine, papaverine, buclizine and narcotine were 6.0, 7.1, 10.0, 17.8, 26.1 and 30.2, respectively.

Table 3

Results of analysis of 15 samples each of raw, prepared and dross opiums

(Weight per cent, dry basis)












9.8 -15.0
10.5 -22.7
6.8 -15.4
1.6 - 3.2
1.8 - 4.4
0.9 - 1.7
1.8 - 4.4
1.1 - 3.4
0 - 0.1
0.02- 0.52
0.08- 0.20
0.04- 0.14
5.0 - 8.0
1-2 a
1.5 a
Not detected
H 2O


Figure II Typical chromatograms of raw, prepared and dross opium samples

Full size image: 3 kB, Figure II Typical chromatograms of raw, prepared and dross opium samples

A =codeine, B = morphine, C = thebaine, D = paperverine, E = narcotine

Figure III Typical chromatograms of raw, mixed and dross prepared opiums

Full size image: 3 kB, Figure III Typical chromatograms of raw, mixed and dross prepared opiums

A = codeine, B = morphine, C = thebaine, D = paperverine, E = narcotine

Results and discussion

Table 1 shows the results of the colour reactions of the various opium samples with the ferric chloride and hydrochloric acid solutions. Based on their negative reaction to the ferric chloride test it is possible to separate dross opium and dross prepared opium samples from others. Likewise, only raw opium reacts with warm hydrochloric acid to produce a red colour due to the presence of the alkaloid porphyroxine. But when raw opium is boiled in water for 2 h and converted to prepared opium, the same test does not give a red colour. However, heating raw opium at 105°C in an oven for 1-2 h does not seem to have any effect on the production of the red colour. Apparently porphyroxine is either altered or degraded in the process of boiling in water. Other mineral acids cause a similar reaction. Weaker acids, such as acetic acid, react to produce a lighter red colour after heating for 20 min on a water bath.

The results of the TLC analysis are shown in figure I and table 2. The nature of the opium is reflected in the pattern and relative intensities of the spots on the TLC plate, both in visible and uv light. The principal spots, morphine and codeine, could not offer much to assist in differentiation. Most of the samples examined were found to be rather low in papaverine content, which was sometimes not detectable even in raw opium. The unique red spot for porphyroxine in raw opium appears at R f = 0.66 when sprayed with dilute hydrochloric acid, while the distinguishing factor for prepared opium is a pronounced yellow fluorescent spot lying between the origin and morphine when viewed under uv light (366 nm). This yellow spot is formed when raw or dross opium or both together are boiled in water which, as started earlier, also resulted in the loss of the red colour reaction for raw opium. Thus on the TLC plate, all prepared opium samples exhibit this yellow spot (weaker for dross prepared opium) but not the porphyroxine spot. These two spots, if present together, would suggest a mixture of raw and prepared opium. No specific marker could be found for dross opium, its major characteristics being theabsence of narcotine and, more particularly, thebaine.

The solvent was allowed to equilibrate for 10-16 h in the TLC tank to mensure a satisfactory separation of the major alkaloids and the development ofthe yellow spot for prepared opium. Development in a freshly prepared solvent system would result in the yellow spot being too close to morphine and thehigher R f spots, thebaine, papaverine, porphyroxine and narcotine, clustering together. Older solvents would cause the yellow spot to be near the origin. The improved hallucinogen reagent, used by de Faubert Maunder [ 1] as a spot test for opium, was found to be a suitable additive to the alkaloid chromogeniodoplatinate. However, heating was required before the coloured spots, ranging from yellow to brown, appeared. These coloured spots faded on cooling to room temperature and would reappear when reheated. The DAB spray, unlike other chromogens, did not mask the red porphyroxine spot or the yellow fluorescent spot when the spots were observed under long-wavelength uv light. Furthermore, the plate could be oversprayed with the iodoplatinate reagent.

The GLC chromatograms shown in figures II and III provide further evidence of the characteristics of the opium samples. Using the operating conditions as described, the major peaks were eluted within half an hour. The procedures for the qualitative and quantitative estimation of the alkaloids were found to be satisfactory. Except for the extraction steps, the method is similar to that adopted by Furmanec [ 11] which involved the simultaneous estimation of the alkaloids of opium without derivative formation. Table 3 shows the results of the GLC analysis for 15 samples each of raw, prepared and dross opium. Due to the weak response of narcotine at concentrations below 2 per cent on the GLC system, approximate values were assigned to the narcotine content for prepared opiums. The moisture content was found to range from a high of about 33 per cent in raw opium to a low of 6 per cent in dross.

It is interesting to note that the codeine content and particularly the morphine content of the dross opium were still relatively high despite the process of opium smoking. Considering the amount still remaining in the dross and the losses due to other factors such as burning and vaporization, it would appear that only a small proportion of the morphine in prepared opium could have been absorbed by an addict through smoking. In sharp contrast is the virtual absence of thebaine, which is an especially important feature of prepared opium and dross prepared opium. This was shown by controlled flame heating of prepared opium, in which the resultant residue was found to contain very little thebaine. The amount of these alkaloids lost when opium is smoked to some extent related to their relative stability and volatility. Thus it is observed that thebaine is the most, and morphine the least, volatile of the five major components of opium.

The illicit samples of prepared opium received were found to contain a substantial amount of dross, varying from 30 per cent to 50 per cent. This was shown by comparing the TLC and GLC chromatograms of the illicit samples with standard mixed prepared opium made from different proportions of raw and dross opium. Raw opium consistently revealed a bunch of triplet peaks (X, Y and Z in figure II) lying between papaverine and narcotine but only two peaks (X and Y) were observed when it was converted to raw prepared opium (figure III). Dross and other prepared opium samples also showed only these two peaks.

The analytical data for raw opium listed in table 3 substantiated the information that the illicit raw opium came essentially from south-east Asia. Studies made on the origin of opium [ 12] [ 13] [ 14] showed that opium rich in thebaine (up to 5 per cent) but containing very little papaverine was typical of opium from the region formerly known as Indochina. The codeine content of the samples (1.6-3.2 per cent) were also within the limits expected of opium from this region. Raw opium (with very low thebaine content) from certain isolated regions [ 15] could nullify this feature of dross opium, thereby reducing the chemical characteristics of the latter to a less-than-satisfactory state. However, as opium smoking is essentially a socio-economic problem confined to southern and south-east Asia, such a difficulty is not expected to be of significance.

Although the present studies were confined to opium samples found in south-east Asia, similar studies could be applied to opium samples found elsewhere. Differentiation and comparison are interrelated and the absence of some of the alkaloids in dross have to be viewed in relation to their presence in the raw and prepared opium used for smoking. A sequence of tests - appearance, odour, colour and TLC - is normally sufficient to make the required differentation. Additionally, GLC and quantitative analysis may be used in difficult cases and also when a sample is found in trace amounts or in a form not easily recognizable.


The authors would like to thank the Director of Scientific Services, Mr. M. C. Dutt, for permission to publish the paper. Thanks are also due to Mr. K. L. Tham, Senior Laboratory Technician, for the drawings.



F. de Faubert Maunder, "Field and laboratory tests for raw and prepared opium", Bulletin on Narcotics . vol. 27, No. 1 (1975), pp. 71-76.


E. Merck, "über bengalisches Opium und eines bei Untersuchung desselben aufgefundenen eigentumlichen Stoffes". Annalen der Pharmacie . vol. 21, 1837, pp. 201-205.


E. Merck, "Porphyroxin", Annual Report on Progress of Physical Sciences . vol. 18, 1839, p. 379.


"The story of 'Porphyroxine-Meconidine' (The red-turning alkaloid of opium)". Bulletin an Narcotics . vol. 4, No. 1 (1952), pp. 15-25.


K. Genest and C. G. Farmilo, "Simultaneous determination of morphine, codeine and porphyroxine in opium by infrared and visible spectrometry", Analytical Chemistry . vol. 34, 1962, pp. 1464-1468.


K. Szendrei, "A method of isolating porphyroxine from opium", Bulletin on Narcotics . vol. 20, No. 1 (1968), pp. 51-54.


Sir Ram Nath Chopra and I. C. Chopra, "Quasi-medical use of opium in India and its effects", Bulletin an Narcotics . vol. 7, Nos. 3-4 (1955), pp. 1-22.


C. Suwanwela and others, "The hill tribes of Thailand, their opium use and addiction", Bulletin an Narcotics . vol. 30, No. 2 (1978), pp. 1-19.


A. Masood, "Opium smoking in the Frontier Province of Pakistan", Bulletin on Narcotics . vol. 31. No. 1 (1979), pp. 59-66.


B. Davidow, N. Li Petri and B. Quame, "Thin-layer chromatographic screening procedure for detecting drug abuse", Technical Bulletin of the Registry of Medical Technologists . vol. 38, 1968, pp. 298-303.


D. Furmanec, "Quantitative gas chromatographic determination of the major alkaloids in gum opium", Journal of Chromatography . vol. 89. 1974, pp. 76-79.


"Progress in determining the origin of illicit opium", Bulletin on Narcotics . vol. 5, No. 1 (1953), pp. 8-14.


"The opium alkaloids", Bulletin an Narcotics . vol. 5, No. 3 (1953), pp. 13-14.


"Report by the United Nations Narcotics Laboratory", Bulletin an Narcotics . vol. 19, No. 3 (1967), pp. 7-15.


K. Narayanaswami, H. C. Golani and R. D. Dua, "Assay of major and minor constituents of opium samples and studies of their origin", Forensic Science International . vol. 14, 1979, pp. 181-190.