Heroin in the Florence area, Italy


Results and discussion


Author: F. MARI, E. BERTOL, , M. TOSTI
Pages: 37 to 44
Creation Date: 1982/01/01

Heroin in the Florence area, Italy

Professor of Toxicological Chemistry, E. BERTOL,
Forensic Medicine Department, University of Florence, Florence, Italy


The authors report data on the analysis of heroin samples seized in the Florence area between 1975 and the first half of 1981. They consider that the Florentine situation, and the Italian market in general, is different from other non-Italian markets with respect to heroin content and to types of adulterant. No dangerous substances were found in the samples and strychnine, if present, was found in very low concentrations. They conclude that the cause of death in heroin addicts or users is due to the amount of heroin injected (overdose).


All heroin samples confiscated by the judicial authorities in the Florence area are analysed in the official laboratory for forensic analysis, the Toxicological Chemistry Laboratory of the Forensic Medicine Department, University of Florence. Since 1975, traffic in heroin and the number of seizures have increased dramatically. The heroin content of seized samples has changed as have the various diluents and active adulterants found in the samples. Many authors have reported on other drugs found in street heroin in their countries [ 1-7] . In this paper, data on the analysis of all illicit heroin seizures in the Florence area since 1975 are reported in order to identify changes in heroin content, inert diluents, and active adulterants. Impurities associated with the heroin manufacturing process, 0 6-acetylmorphine and acetylcodeine, have also been investigated.


Thin-layer chromatography (TLC)

All samples were tested by TLC for the qualitative determination of inert diluents (lactose, fructose, sucrose, mannitol etc.), active adulterants, acetylcodeine and 0 6-acetylmorphine, and any other substance that might indicate the origin of the heroin. To test inert diluents 10 mg of the heroin mixture were dissolved in water (1.0 ml) and 10 μl were spotted and dried on each of two precoated silica gel plates (Merck). Prior to use each plate was sprayed with an aqueous 0.1 N boric acid solution. After saturating each plate with the boric acid solution, the plate was dried for 6-7 hours. The developing solvent was benzene-acetic acid-methanol (20:20:60). After a 10 cm development both plates were dried and one was sprayed with metaperiodate-benzidine [ 8] for the detection of mannitol. The other was sprayed with aniline-diphenylamine-phosphoric acid [ 8] for the detection of sugars. To test for other substances another 10 mg of the heroin mixture were dissolved in ethanol (1.0 ml) and 10 μl aliquots were spotted on each of four precoated silica gel plates (Merck). These plates (1-4) were developed in four different solvents: (a) methanol-ammonium (100:1.5), (b) cyclohexane-benzene-diethylamine (70:25:10), (c) toluene-acetone-ethanol-ammonium (45:45:7:3), (d) chloroform-acetone (90:10) for 10 cm. They were dried and sprayed with : (a and b) acidified iodoplatinate [ 9] , (c) Marquis [ 9] (for basic substances), (d) mercurous nitrate [ 9] (for barbiturates).

Gas-liquid chromatography (GLC)

Apparatus: Carlo Erba gas chromatograph series 4200 equipped with a flame ionization detector and a glass column (2 m x 3 mm I.D.) packed with 2% OV-17 on Anakrom ABS 90-100 mesh. Before use the column was silanized with Silyl-8 (Pierce Chemical Co.) at a low nitrogen flow for 24 hours. Detector and injector temperature: 320°C; oven temperature: initial 200°C (6 min), 10°C/min increase (6 min), final 260°C (6 min); nitrogen flow rate: 50 ml/min.

Stock solution (100 mg/dl): dissolve 10 mg of pure heroin, procaine, caffeine, lidocaine, acetylcodeine, 0 6-acetylmorphine, benzoyltropeine in 10 ml of ethanol.

Standard solutions (0.1 mg/dl, 0.2 mg/dl, 0.4 mg/dl, 0.8 mg/dl): dilute 1, 2, 4, 8 ml of stock solution with ethanol to 10 ml.

Internal standard, dieldrin solution (50 mg/dl): dissolve 0.5 g of dieldrin in 1 litre of ethanol.

Reference solutions: 1 ml for each standard solution was added to 1 ml of the internal standard solution.

Calibration curves: 1-2 μl of each reference solution was injected into the gas chromatograph and the detector's response was checked 10 times at each concentration. The peak areas were measured by a disc integrator. The response was found to be linear for the range of concentrations used. Figure 1 shows the typical chromatogram under the described conditions.

Procedure: 10 mg of each homogenized heroin sample were dissolved in 10 ml of ethanol; 10 ml of internal standard solution were added; 1-2 μl of this solution were injected into the gas chromatograph under the same conditions as were used to obtain the calibration curves.

Results and discussion

Heroin content

Table 1 shows the distribution of the heroin content of the samples seized from 1975 to the first half of 1981, classified into three groups on the basis of their heroin content. Most of the samples contained between 20 per cent and 70 per cent heroin and represented 39.1 per cent to 78.5 per cent (mean 60.1 per cent) of all the cases. Samples whose heroin content were less than 20 per cent represented zero to 10.5 per cent of all cases from 1975 to 1979. However, in 1980 and the first half of 1981, this increased remarkably to 20.5 per cent and 32.9 per cent respectively. In contrast the percentage of samples whose heroin content was greater than 70 per cent decreased from 60.9 per cent in 1975 to 4.3 per cent in the first half of 1981.

Figure I - Chromatogram of mixed acetylated derivatives from opium, active adulterants and dieldrin (internal standard). Chromatographic conditions are given in the text

Full size image: 34 kB, Figure I - Chromatogram of mixed acetylated derivatives from opium, active adulterants and dieldrin (internal standard). Chromatographic conditions are given in the text

Figure II - Chromatogram of a heroin sample seized in 1976

Full size image: 28 kB, Figure II - Chromatogram of a heroin sample seized in 1976

Figure III - Chromatogram of a heroin sample seized in 1981

Full size image: 26 kB, Figure III - Chromatogram of a heroin sample seized in 1981

Table 1

Distribution of samples according to heroin content


Less than 20% heroin

20% to 70% heroin

More than 70% heroin

39.1 60.9
1976 9.4 68.7 21.9
1977 10.5 57.9 31.6
1978 9.6 52.4 38.0
1979 5.6 78.5 15.9
1980 20.5 61.5 18.0
1981 (first half)
32.9 62.8 4.3


Impurities were found due to the manufacturing process or the origin of the heroin. These mostly included 0 6-acetylmorphine and acetylcodeine.

Acetylcodeine was found in all samples in a proportion ranging from 0.6 per cent to 12.0 per cent (mean 5.3 per cent). It represented more than 3.0 per cent in 84.6 per cent of the samples.

0 6-acetylmorphine was detected in 87.2 per cent of the samples and was quantified in 66.6 per cent of them at between 0.8 per cent and 20.9 per cent.

Papaverine, an impurity of origin, was present in about 90 per cent to 95-per cent of all samples, but its concentration was always low.

Narcotine and thebaine, other impurities of origin, were detected in traces in some cases.

Adulterants and diluents

From 1975 to 1977, 75.6 percent of the illicit heroin seized in the Florence area was of the so-called No. 3 type, or "brown sugar", a grey-brown granular product, characterized by the presence of heroin (with impurities from the manufacturing process such as acetyl-codeine and 0 6-acetylmorphine) and caffeine. Most samples (64.0 per cent) also contained strychnine. The caffeine content ranged between 18.7 per cent and 44.5 per cent (mean 29.6 per cent). The strychnine content 1was always less than 1.0 per cent and this concentration was not considered dangerous. Since 1977 there have been few seizures of heroin No. 3 in the Florence area. Inert diluents such as sugars or mannitol have been found in 92.0 per cent of the 1978 - 1980 heroin seizures. The remaining 8.0 per cent contained uncut heroin. In the first half of 1981 seizures of uncut heroin fell to 3.0 per cent and active adulterants such as procaine or lidocaine and diluents were found in 28.0 per cent of the samples (see table 2). Procaine concentrations ranged between 2.2 per cent and 27.0 per cent and lidocaine between 4.8 per cent and 11.2 per cent.


The strychnine amount was estimated by TLC using strychnine reference solutions (5 mg/dl, 10 mg/dl, 20 mg/dl).

Table 2

Distribution of samples with respect to cutting agents


Heroin and caffeine (%)

Heroin, caffeine, strychnine (%)

Heroin and diluents (%)

Heroin, diluents, adulterants (%)

Uncut heroin (%)

75.6 64.0     24.4
    92.0   8.0
1981 (first half)
    97.0 28.0 3.0

In the first half of 1981, for the first time in Italy, an unusual substance, benzoyltropeine 2,was identified in five samples. One of these (3.0 kg of powder) contained 41.3 per cent benzoyltropeine and 58.7 per cent acetylated derivatives from opium (50.7 per cent heroin, 2.3 per cent 0 6-acetylmorphine, 5.7 per cent acetylcodeine). In the remaining four samples (each weighing less than 10 grams) the benzoyltropeine content ranged from 4.9 per cent to 37.1 per cent and lactose was also present. Because the ratio of the benzoyltropeine and the acetylated derivatives from opium was 0.7 in all five cases, the assumption was that they had a common origin. The United States Drug Enforcement Administration has indicated that their Mid-Atlantic and South-eastern Regional Laboratories have reported some cases of benzoyltropeine contamination of heroin samples in recent years. Other diluents or adulterants included starch, acetylsalicylic acid, barbiturates, boric acid and amphetamines and were present in approximately 0.5 per cent of all cases.


The heroin in street samples confiscated in the Florence market from 1975 contained high concentrations of heroin in comparison with European and American markets. In 1980 and the first half of 1981, the heroin content decreased considerably and uncut heroin exhibits were sporadic. Heroin in the Florence area then resembled other non-Italian markets. The active adulterants found in heroin mixtures in the Florence area were also different from other European countries (Federal Republic of Germany and the Netherlands) and the United States of America; quinine, methapyrilene, antipyrine etc. never having been identified in the Florence samples.

In contrast, no difference was found in the composition of No. 3 heroin as the original product from Asia is not cut by the dealers. Very toxic adulterants were not found; strychnine, if present, was always in very low concentrations. Therefore, the cause of death in heroin addicts or users can be attributed to the quantity of heroin injected and not to the presence of other dangerous substances. The composition of the samples varied considerably so that drug users could not really know the percentage of heroin in the sample they used. Ignorance may result in a fatal overdose.


The authors wish to acknowledge the assistance of the GC/MS Service of the School of Medicine, Florence.


Benzoyltropeine was identified by mass-spectrometry. An LKB series 2091 was employed. Temperatures of ion source and separator were 250°C and 230°C respectively; electron energy, 20 eV.



D. W. Johnson, A. B. and J. W. Gunn, "Dangerous drugs: adulterants, diluents and deception in street samples", Forensic Science , vol. 17, 1972, pp. 629 - 639.


D. Eskes and J. K. Brown, "Heroin-caffeine-strychnine mixtures - Where and why?" Bulletin on Narcotics (United Nations publication), vol. 27, No.1 (1975), pp. 67-69.


E. P. J. Van der Slooten and H. J. Van der Helm, "Analysis of heroin in relation to illicit drug traffic", Forensic Science , vol. 6, 1975, pp. 83-88.


H. Huizer, H. Logtenberg and A. J. Steenstra, "Heroin in the Netherlands", Bulletin on Narcotics (United Nations publication), vol. 29, No. 4 (1977), pp. 65 - 74.


G. Bohn, E. Schulte and W. Audick, "Gaschromatographische und massenspektrometrische Untersuchungen an im Münsterland in den Jahren 1972-1976 sichergestellten Heroinproben", Archiv für Kriminologie , vol. 160, 1977, pp. 27-33.


J. Albanbauer and others, "Quantitative Hochdruck-Flüssigkeitschromatographie von Rauschgiften", Archiv für Kriminologie , vol. 162, 1978, pp. 103-107.


M. Gloger and H. Neumann, "Untersuchungen zur Stabilität illegaler Heroinproben", Archiv für Kriminologie , vol. 166, 1980, pp. 89-97.


E. Stahl, Thin Layer Chromatography , 2nd ed. (New York, Springer-Verlag, 1969).


E. G. C. Clarke, Isolation and Identification of Drugs (London, Pharmaceutical Press, 1975), vol. II.