PART IA
Author: Charles G. Farmilo, Leo Levi
Pages: 20 to 31
Creation Date: 1953/01/01
PREFACE
The study of narcotic properties, analgesia and addiction liability has led scientists to the discovery of new narcotics with more or less addiction liability. The numbers of narcotics have thus increased far beyond the original natural opiates. The need for strict control and wider investigation of potentialities of the increasing numbers of narcotics makes it imperative, therefore, to have up-to-date means of identification for forensic purposes, and for scientific investigation.
In a study of the identification problem over the past five years, it became apparent that along with microchemical data the physical data about narcotics were required for complete identification. It was also apparent that there is a paucity of narcotic physical data. It was decided to orient our efforts toward the collection and co-ordination of both types of data. In contrast to the lack of physical data, a great number and variety of microchemical crystal and colour tests are scattered in the chemical literature. We hope eventually to assemble all the pertinent analytical knowledge in chemistry on the subject of narcotic identification. It is obvious that the achievement of this aim can be brought about only by the collaboration of many authorities. It is hoped that these papers will begin by helping our colleagues in their research and solution of practical identification problems and that they will collaborate with us by helping to keep the material up to date in subsequent issues of the Bulletin on Narcotics.
The authors wish to express their appreciation to the many people who have helped in beginning this project. The whole-hearted co-operation of Dr. L. I. Pugsley is appreciated in generally guiding with his extensive technical knowledge the over-all task. Dr. W. Barnes, National Research Council, collected the X-ray diffraction powder photographs and data. Defence Research Chemical Laboratories made a substantial contribution by providing us with the infra-red spectra, obtained under the supervision of Dr. Charles Hubley. Mr. Charles Fulton has contributed his help on various problems.
The various drug companies have been extremely generous in their gifts of valuable and sometimes rare materials. Drs. Reti and Castrillon, of the South American firm, Atanor, contributed the newly discovered trichocereine. Dr. Léo Marion of the National Research Council, Canada, supplied the authentic protopine used in these studies.
To the typists, draughtsmen and photographers of the National Research Council, Canada, and National Health and Welfare we owe thanks for cheerful and optimistic co-operation.
Acknowledgment is also made to the officials of the Department of National Health and Welfare who gave us permission to publish these data.
Following is the general plan for the present series:
Introduction to Identification of Narcotics by Physical Methods by Charles G. Farmilo and Leo Levi, Organic Chemistry Section, Food and Drug Laboratories, Department of National Health and Welfare, Ottawa
The Common Physical Constants for Identification of Ninety-five Narcotics and Related Compounds by Charles G. Farmilo, P. M. Oestreicher and Leo Levi, Organic Chemistry Section, Food and Drug Laboratories, Department of National Health, and Welfare, Ottawa
The X-ray Diffraction Method on Powder by W. H. Barnes, Division of Physics, National Research Council, Ottawa
X-ray Diffraction Data on Powder for Eighty-three Narcotics by W. H. Barnes and Helen M. Sheppard, Division of Physics, National Research Council, Ottawa
The Ultra-violet Spectrophotometric Method by Charles G. Farmilo, Food and Drug Laboratories, Ottawa
Ultra-violet Spectral Data for Eighty-six Narcotics by P. M. Oestreicher, Charles G. Farmilo and Leo Levi, Food and Drug Laboratories, Ottawa
The Infra-red Spectrographic Method by Charles Hubley and Leo Levi, Defence Research Chemical Laboratories, and Food and Drug Laboratories, Ottawa
Infra-red Spectra of Narcotics by Charles Hubley, Leo Levi and Moira Smith, Defence Research Chemical Laboratories, and Food and Drug Laboratories, Ottawa
By Charles G. Farmilo and Leo Levi
SECTION |
Page | |
---|---|---|
1. |
Object and trend in recent research on narcotic identification |
21 |
2. |
General problem of narcotic analysis and relation to organic analysis |
21 |
3. |
Crystallographic methods |
24 |
4. |
Optical crystallography |
24 |
5. |
X-ray diffraction |
24 |
6. |
Optical rotation |
24 |
7. |
Methods based on chemical groups in narcotics |
24 |
|
(a)Spectrographic methods of determining chemical groups |
25 |
|
( b) Infra-red and Raman spectra |
25 |
|
( c) Ultra-violet spectra |
25 |
|
( d) Luminescence spectra |
26 |
8. |
Electromotive force measurements on narcotics |
|
|
( a) Electrometric titrations |
26 |
|
( b) Polarography |
27 |
9. |
Conclusions |
27 |
TABLE |
|
|
I. |
Chemical groups present in narcotics and related compounds detectable by physical methods of identification |
28 |
II. |
Phenanthrene type narcotics and related compounds, ultra-violet chromophores and detailed structure |
30 |
REFERENCES |
|
31 |
The object and characteristic trend of recent research into methods of narcotic identification for forensic purposes has been that of obtaining definite identifications on minimal quantities of generally impure substances, without destroying or altering the whole sample.
There are two practical reasons for this trend in forensic research. First, seizures of drugs by the police are often small and so heavily adulterated that they contain only small amounts of active components; and second, to facilitate court procedures rapid accurate methods of analysis are required which leave a certain amount of sample for a court exhibit and yield specific identifications. As early as 1880 it was realized [1] 1 that drug identification for court purposes could be considerably aided by instrumental methods (e.g., spectroscopy). It was not until about 1920 that in Europe forensic scientists in universities began to advocate the extended use of physico-chemical methods of identification in medico-legal and toxicological cases [2] . This development ran concurrently with the growth of theory and practice of physical chemistry; and about 1940 with the introduction of electronic circuits and recorders, instrumentation developed to the routine stage of usefulness [3] . Some delay in general use of instruments is probably caused by the lag in applications of instruments behind theoretical development, and delay in co-ordination of the theory and practice of instrumentation with chemical findings. It is hoped as a general objective of the present series in regard to narcotics to alleviate this situation. It is the object of this paper first to point out the usefulness in narcotic analysis of a number of physical methods and second to introduce a series of papers which describe for narcotic analysts the principles and practice of X-ray powder diffraction, and ultra-violet and infra-red spectroscopy. These papers have been divided into an introductory and an experimental part. In the latter the common physical constants, X-ray powder diffraction photographs, and ultra-violet and infra-red spectra of about ninety-two narcotics and related compounds are collected. This introductory paper to the series tries to point out a general approach to the problem of research on methods of narcotic analysis on the basis of qualitative organic chemical analysis in order to show the importance of a combined study of physical and chemical methods.
1. Figures in parentheses throughout the text refer to the numbered publications listed at the end of this article.
Like other schemes of qualitative organic analysis [4] , [5] , analysis for narcotics in an unknown sample involves several steps beginning usually with some dividing test which puts the unknown in one or the other of two classes, on the basis of a general property such as the solubility. Also, since most narcotics are amines or their salts, the initial classification is often made by means of general amine precipitants. Assuming the possible presence of a narcotic is indicated, then very small samples are made to undergo physical and chemical changes by means of specific tests, usually microscopic. From results of these chemical tests, inferences are drawn, and on this basis the unknown is restricted to fewer and fewer subdivisions of a particular chemical class. It becomes possible, by combining all the known data, to find a narcotic that shows complete physical and chemical similarity when put through the same tests as the unknown, and therefore is identical with the unknown under investigation.
The narcotic analyst is faced with special problems which force him to evolve certain special steps, or modify existing ones, to enable him to systematically identify the more common narcotics in microquantities; at the same time he has to be prepared to eliminate or identify all drugs which are presented in the course of police investigations. The sample to be analysed for a narcotic may be in one of several forms: such as traces of drug on the addict's paraphernalia, as a spoon, medicine dropper, needle, or swab of tissue paper or cotton; a natural product such as raw opium, coca leaves, cannabis sativa or a mixture of tobacco and marihuana; a capsule, or a "deck" containing a powder; or a liniment, or elixir or other legitimate pharmaceutical preparation containing one or more active ingredients. Samples of narcotics may also be isolated from human tissue obtained after autopsy, or in urine or saliva from "doping" of race horses. In the last two toxicological types of cases, narcotic isolation and purification require procedures especially designed for the purpose [6] , [7] . In the case of pharmaceutical preparations the sample can often be dissolved in water and shaken out with immiscible solvents from acid or basic solutions. In such a case the mode of separation usually yields enough solubility data to classify the unknown compound to a certain extent. In the case of a narcotic isolated from a pharmaceutical preparation, even a boiling point or melting point may be obtained. Such physical constants are at present rarely determined in routine narcotic identification. The fact that many samples may contain only trace amounts of the drug (which cannot be completely destroyed because of court purposes) often precludes rigorous extraction and purification procedures, or determination of ordinary physical constants which involve recrystallization or distillation, for fear of destroying the whole sample. It is the condition of sample therefore to a great extent that determines the over-all and final procedures adopted in analysis.
In the following summary the steps usually followed in narcotic analysis are outlined in a form similar to the one generally used by textbooks for organic analysis. A number of these steps are explained in order to bring out the similarities and differences between narcotic analysis and classical organic procedures in use since the time of Liebig. The steps for narcotic analysis follow the outline given by Cheronis and Entrikin [5] . It will be noted that steps B 1, B 3, B 4, and D are different from ordinary organic identification procedures. They consist of examination to detect the narcotic constituent, type of physical constants, colour and classification tests, and type of derivative formation. Although in laboratories specializing in organic chemistry it is often the routine procedure to determine physical constants prior to beginning actual detection of chemical groups, the nature of narcotic analysis sometimes makes this the last step to be performed. Often a qualitative identification is the only one required by police officers to aid in investigation, or in court procedures. This qualitative identification, at present, is. made on the basis of chemical findings.
Summary of steps in narcotic analysis
A. Isolation and purification:
Separation of the drug from the adulterant.
Sublimation, distillation, recrystallization.
B. Examination and classification of the drug:
Determination of physical constants.
Solubility classification (usually indirect information obtained from step A 1).
Qualitative analysis by precipitating reagents to determine if an amine type compound is present.
Classification by means of precipitation, and colour reactions with certain types of reagents.
C. Co-ordination of data:
All data including physical constants must be carefully co-ordinated and compared with the chemical literature and collections of data and other sources. A list of possible compounds prepared and the narcotic that best fits the data is then selected as the probable one.
D. Derivatization and final proof:
By means of microchemical crystal reactions suitable derivatives of the narcotics are prepared. Reagents yielding the best crystal tests with the probable narcotics are selected. If the selected derivative has the characteristic form described in the literature for a probable narcotic, the identification is regarded as tentative. Finally, if the forms of other characteristic derivatives prepared from the unknown, and from a pure sample of the narcotic, are identical, the proof is considered conclusive.
Classification of the unknown narcotic by reference to its behaviour towards typical reagents is still the most common method of detection. Many microchemical and colour tests are available for helping to classify narcotics. These have recently been reviewed for the new synthetic narcotics by Farmilo, Levi, Oestreicher, and Ross [7] . A great deal more information regarding analysis of narcotics and other toxicologically important drugs is still scattered in the chemical literature and various laboratory files. Attempts to co-ordinate this information are being made in the Alcohol Tax Unit Laboratories of the U.S. Treasury Department, United Nations Narcotics Laboratory, New York State Racing Commission Laboratory, the Attorney General's Laboratory of the Province of Ontario, the Royal Canadian Mounted Police Crime Detection Laboratories at Ottawa and Regina, and Canadian Food and Drug Laboratories of the Department of National Health and Welfare, and by organizations such as the Forensic Society of Canada and the American Academy of Forensic Sciences. Publication of reviews of data, by such groups, will eventually simplify the search for now widely scattered information on narcotics. However, such reviews are just beginning to be made and most of these laboratories have still to rely on their own collections of data and more often on the skill and knowledge of one or two experienced analysts.
In a practical case after co-ordination of data is achieved, during analysis of a sample, there may still be a number of possibilities in addition to the particular narcotic tentatively concluded by the analyst to be present. In the classical organic procedure at this step the analyst prepares one or more suitable crystalline derivatives of the unknown compound based on the tentative identification. In forensic analysis the step of derivative formation is carried out in a special way by means of microchemical crystal reactions. The analyst identifies the substance by a microchemical reaction based on the appearance, under the microscope, of the derivative as it begins to crystallize on the microscope slide. By reference to the same test, with the known compound as the standard of comparison, and a written description or an illustration such as a line drawing or photo-micrograph of the crystal product, the analyst makes his final decision as to the identity of the unknown. He is then able to select a number of additional best crystal tests to confirm his decision.
Experience has shown [8] that when the product of the reaction of narcotic and reagent is distinctive and readily formed, the microscopic method leaves little to be desired. Usually mere inspection with the microscope is satisfactory - when the crystals are truly characteristic [9] . Sometimes, however, still further work has to be done by the analyst.
Where truly characteristic microcrystalline forms cannot be obtained in microchemical reactions, further derivitization is required by the analyst to satisfy him as to the complete identity of the unknown. It is often possible for the analyst to isolate the same crystalline derivatives as were obtained in the microcrystal tests for recrystallization and further characterization. A few of these products have already been isolated and purified for further study; for example, reineckates of quite a few amines, amino acids, and quaternary ammonium bases are derivatives having sharp melting points. Levi and Farmilo [10] showed that a number of narcotics from the diarylalkone-, arylpiperidine-, and hydro-phenanthreneamine subdivisions of the chemical narcotic classification gave crystalline reineckates. Similar studies of morphine-Marme and morphine-Mayer reactions have been made [11] . Cadmium halide complexes of a number of nitrogenous bases including narcotics have been prepared and quantitatively estimated by means of non-aqueous titration of the complexes in glacial acetic acid [12] . The foregoing examples serve to illustrate some of a large number of derivatives which can be employed and further investigated for use in identifying narcotics. Many more reagents remain to be studied.
Ordinarily for purposes of organic chemistry in dealing with pure compounds the readily determinable, more common physical properties enable one to exclude from consideration all but a very few of the known compounds. This is not the case in narcotic identification, which has become more complicated as new products are marketed. The recent discovery [13] of the relationship between structural and optical isomers with analgesic and addiction potentialities has resulted in commercial production of diasteroisomers, such as d-α-, l-α-, dl-α-, and d- β-, l-β-, dl-β- methadols and acetylmethadols [14] and dl-α-and β- prodines [15] . Such diasteroisomers differ from each other only in physical properties such as optical rotation, melting point and solubility. Their chemical reactions are of the same type for all forms since they possess the same functional groups, but their rates of chemical reaction and biochemical activity in man [16] and animal [17] are different. In ordinary organic chemistry a sharp melting point or boiling point, together with qualitative and quantitative analysis for elements, refractive index, optical rotation, and chemical reactivities suffice to identify the organic substance. However, the commonly used physical properties are found by experience to be most often of little use in narcotic analysis. Narcotic samples, especially salts, rarely display sharp and reproducible melting points; decomposition over a temperature range often is observed. The molecular weight and empirical formula cannot be readily determined with sufficient accuracy to remove ambiguity, due to uncertainty of the hydrate forms and high molecular weights. Refractive indices for crystalline narcotics are not as commonly given as they are for liquids. These various factors plus the rapid development of the number of commercially available narcotics make it extremely important to investigate other well established physical methods of organic chemistry for use in narcotic identification.
For this purpose it will be useful to refer back to the summary of steps in narcotic analysis to determine where further information is required.
Review of the steps of identification shows that three types of properties are used in narcotic analysis: physical constants based on colligative properties depending on number of molecules; constitutive properties, depending on the size and shape of the atoms and molecule; and chemical properties governed by the functional or chemical groups in the narcotic. Accordingly the physical methods of organic chemistry which we propose to study in connexion with narcotic identification are shown in the following classification according to the molecular property which each measures:
Classification of physical methods of organic chemistry useful in narcotic identification
A. Methods based on size, shape and atomic configuration of the molecule
Optical crystallography
X-ray crystallography
Polarimetry
B. Methods based on chemical groups present in the molecule; spectrographic methods
Ultra-violet
Infra-red
Raman
Luminesence
C. Methods based on chemical groups and numbers of molecules
Electromotive force measurements. Electrometric titrations in aqueous and non-aqueous solutions
Polarography
All the methods under A and B and C in this classification yield physical constants; these, coordinated with results from chemical tests, lead to complete identifications. By itself a single method, chemical or physical, is not sufficient as a basis for determining identity; for example, in the cases of A3 and B1 a specific rotation or ultra-violet spectrum alone would not be sufficient. The optical and X-ray crystallographic methods A1 and A2 are the ones most likely to be self-sufficient in a given case. A certain judicious choice of method for completing the identification is therefore necessary. This choice has to be based on a knowledge of the principles and limitations of each of the methods listed.
The physical methods listed in the above classification will be discussed briefly in the following text, with statements of the main principles and limitations of each method, the object being to show how specific methods in two main groups of physical methods overlap in effect and results. Thus a guide to those planning the use of one or other of these methods in organic identification will be provided.
A crystallographic description (which ideally includes optical and X-ray crystallography) is a unique set of physical characteristics which, when specified, would completely determine the crystalline phase, with respect to size, shape and atomic configurations. The important directional properties of crystals brought about by virtue of the orderly arrangement of their constituent atoms and molecules provide the basis for such crystallographic descriptions. Unfortunately facilities for complete crystallographic identification are not available to most investigators who isolate new compounds, nor to analysts who are required to detect drugs for law enforcement purposes. Therefore crystallographic descriptions for only a few narcotics are available. The value of crystallography in drug identification is, however, so obvious that a number of laboratories doing forensic work are using these methods to obtain identification data.
More than thirty years ago Wright [18] and Wherry [19] working in United States Government laboratories applied the immersion method of determining the refractive indices of crystals with the polarizing microscope in chemical analysis. Chamot and Mason [26] developed the technique of chemical microscopy in this connexion. Keenan [20] and others [21] in U.S. Food and Drug laboratories and elsewhere have described the use of optical crystallography for routine identification of alkaloids and synthetic drugs [22] , [23] , [24] and [25] .
Optical data readily obtainable include the refractive indices, and sometimes the interference figure, the optic sign, optical dispersion, pleochroism, and the rotation of the plane of polarization. Textbook descriptions of the techniques are given in references 26, 27 and 28 for those interested in further details on the subject.
The constituent atoms of crystals are arranged in regular three-dimensional structures and their centres are separated by distances of the same order of magnitude as the wave-lengths of X-rays. Finely powdered crystalline materials give X-ray spectra that are eminently suitable for identification purposes. In fact an X-ray diffraction photograph obtained on the powdered substance is by itself usually sufficient for unequivocal identification, except in the particular case of distinction between d-, and l optical isomers. This method of identification together with experimental data for eighty-three narcotics is discussed in parts IIA and IIB of this series [29] and [30] .
Optical rotation has long been used as a physical constant in organic characterizations. The organic chemist for about 100 years has defined polarimetry in terms of investigation of the quantitative changes of direction of the plane of vibration of linearly polarized light during its passage through optically anisotropic liquids or solutions. The specific rotation of a substance in solution is calculated by the formula: [ a ] D 25 degrees = 100a/lc, where [ a ] D25 degrees = specific rotation at 25 degrees C; a = observed rotation; 1 = length of tube (decimeters); d = density; c = g in 100 cc of solution.
The subject of polarimetry has been reviewed recently by W. Heller [40] . The specific rotations for about thirty narcotics will be given in part IB of this series [41] .
A chemical classification of narcotics and some related substances is shown below. This divides the substances into four main families and a number of subdivisions in each family, which depend on the main structure. In table I some chemical groups of the molecules of narcotics and related compounds are listed for the purpose of specifying the methods by which they are most likely to be detected. The list shows how physical methods may be applied to the problem of narcotic identification.
CHEMICAL CLASSIFICATION OF NARCOTICS AND RELATED COMPOUNDS FOR IDENTIFICATION
1. PYRANS
meconic acid
pyrahexyl
2. ARYLS
(a) Arylcarboxylic acids
benzoic acid
opianic acid
(b) Arylpiperidines
cocaine
ecgonine
pethidine
ethylpethidine
α-prodine
β-prodine
hydroxypethidine
methylketobemidone
ketobemidone
propylketobemidone
acetoxyketobemidone
(c) Diarylalkoneamines
isomethadone
pipidone
methadone
phenadoxone
α-methadylacetate
(d) Arylethylamines
mescaline
trichocereine
narceine
ethylnarceine
3. ISOQUINOLINES
(a) Dihydroisoquinolines
cotarnine
hydrastinine
(b) Phthalideisoquinolines
narcotine
(c) Benzylisoquinolines
papaverine
dioxyline phosphate
laudanine
cryptopine
protopine
4. PHENANTHRENES
(a) Phenanthroisoquinolines
apomorphine
morphothebaine
(b) Phenanthroethanoamines
thebenine
(c) Iminoethanophenanthrenes
3-hydroxy-N-methylmorphinan
3-methoxy-N-methylmorphinan sinomenine
sinomenine
(d) I minoelhanophenonthrofurans
morphine
codeine
neopine
thebaine
dihydromorphine
dihydrocodeine
dihydromorphinone
dihydrocodeinone
methyldihydromorphinone
dihydrohydroxycodeinone
ethylmorphine
benzylmorphine
α-monoacetylmorphine
diamorphine
acetyldemethylodihydrothebaine
morphine-N-oxide
N-allyl-normorphine
pseudomorphine
Spectrographic methods of determining chemical groups
The majority of chemical groups in table I show their effects in particular regions of the spectrum in the ultra-violet, infra-red, and Raman spectra. Since we are limited, in this project to compounds in only four major chemical homologous series, it is possible by detecting the absence of a particular group to eliminate by means of a single criterion a number of compounds. The reverse is not generally true; the application of a single criterion does not establish the presence of a chemical group, nor indicate a particular compound. The study of many members of a homologous series is necessary before empirical rules can be established by which the spectral region for a chemical group can be determined. The prediction of frequencies for groups, and identifications therefrom, will become more exact as the methods are further investigated. For present purposes a statement in very general terms of the main principles and applicability to narcotic identification will be given.
Infra-red and Raman spectra
In a molecule, the size of the atoms and magnitude of the forces holding these atoms together affect its vibration frequencies. Both Raman and infra-red methods detect these vibration effects. The atomic and molecular vibrations correspond exactly to the frequencies of the infra-red vibration absorbed by the atoms. The infra-red vibration frequencies most useful to the chemist are located in the range from 4,000 to 100 wave numbers, corresponding to 2.5 and 100 microns respectively. The Raman effect is revealed as a line spectrum in the same region as the infra-red; for example, the C-H bond has one Raman frequency between 2,700 and 3,000 wave numbers, the exact frequency being dependent on the chemical environment of the group. Fine distinctions in structure are thus theoretically detectable; for example, the C-H bond in the methane, methyl, methylene, and methylidene groups have different frequencies in this range.
The Raman effect is brought about as follows: in general, a transparent medium illuminated by monochromatic light scatters light at the same wave-length as that of the incident light, and light of modified wave-lengths, the frequency difference between the incident and modified light being related to vibrational and rotational frequencies in the molecule.
The theory of the Raman effect shows that Raman frequencies are closely related to frequencies of vibration within the molecule. A Raman spectrum is a manifestation, therefore, in visible or ultra-violet light, of a molecular process which gives rise to absorption in the infra-red, but it should be noted that for molecules of high symmetry not all infra-red frequencies have Raman counterparts or vice versa. The determination of Raman spectra consists of measuring the frequencies or wave-lengths of the lines, their intensities, and states of polarization (a line produced by unpolarized light may be more or less polarized depending on the symmetry of the molecular vibration). The measurement of polarization state of a line is difficult and requires extra experimental precautions and usually is dispensed with in applications of Raman spectra to analysis [31] .
With regard to obtaining Raman spectra of narcotics, it should be noted that solubility of the molecule will be important. In principle, Raman spectra (with one limitation) should be obtainable; however, in practice it is not possible to obtain Raman spectra in water solution unless a solubility of 1:5 in terms of mole fraction is available. This means, in other words, that one molecule of narcotic has to be present for every five of water, and such molar solubilities are not the usual ones encountered with narcotics, which presents a serious limitation for using Raman spectroscopy in routine analysis. It should be noted that organic solvents with higher molecular weights than water may be usable as solvents for narcotic bases, thus reducing the molar ratio of absorbent to solvent to a more favorable fraction, say 1:1. Another possibility does exist, however, and this is with regard to obtaining Raman spectra of solid materials. The experimental difficulties in the handling of solids for obtaining Raman spectra are greater than for solutions and present a practical limitation. Another theoretical limitation may be encountered with certain compounds. Those molecules whose electrons can be excited to certain electronic states from which the electrons return to the ground states at a relatively slow rate will give rise to emission fluorescence spectra, which masks any Raman effect. However, these effects give rise to "decay" or luminescence spectra which may be important since luminescence spectra are in themselves useful for identification. Such spectra will be referred to again.
Ultra-violet spectra
Ultra-violet and visible spectra involve changes in electronic energy of the groups in contrast to infrared and Raman effects. The region of the spectrum with which we are concerned ranges from 50,000 to 100,000 wave numbers. Groups of atoms which have electronic structures which interact with the electromagnetic wave and involve the structure in energy changes can absorb energy. For example, groups containing multiple bonds which are fundamentally responsible for the colour of organic substances, and are called chromophores, also show absorption of light in the ultra-violet. A number of such chromophores is shown in table II. Numbered formulae of phenanthrene and morphine are shown in table II as a basis of comparison for other phenanthrene-like narcotics listed in column 2 containing the related chromophores.
The use of ultra-violet spectra for narcotic qualitative analysis depends on interpretation of characteristics of the spectra. The characteristic spectral curves cannot yet be explained in simple terms, since no apparent direct relationship appears to link the frequencies in this range with the size or binding force of the atoms (composition of the chromophore) as it does in the case of atomic vibration frequencies. The frequency decreases (wave-length increases) as the size of the conjugated system increases. Also in some cases two chromophores occur in molecules, and are separated by a methylene group and the total absorption is the simple sum of the absorption of the groups as if they were in separate molecules. Such generalizations, although of theoretical value, do not help to identify a narcotic. The value of ultra-violet absorption spectra in analysis rests wholly upon the identity qualitatively and quantitatively of the ultra-violet absorption curve of an unknown narcotic, with that of a known narcotic having a known chromophore or absorbing group.
Luminescence spectra
Certain emission spectra are of interest in drug analysis either for the purpose of identification or estimation; these are classed as luminescence spectra [31] . Fluorescence, phosphorescence, chemi-and electroluminescence are examples. Fluorescence involves electronic excitation of a molecule by absorption of light, and the principles of intense illumination and of avoiding spurious radiation in the emitted light apply, and in general the typical Raman arrangement is used [32] . Fluorescence phenomena have been used in alkaloid analysis since 1885 according to Fisher in his review of physical methods applied to forensic medicine [2] . A recent review of fluorimetry by White [33] indicates the extensive application to analysis of organic materials. As far as can be determined no fluorometric procedures for narcotics have been studied. A general description of means of obtaining and use of fluorescence spectra as a tool for organic research was given by Kasha [42] .
Electrometric titrations
Certain acidic and basic groups commonly present in narcotics may ionize in water or other polar solvents and are characterized by dissociation constants. Dissociation constants are a measure of acidity or basicity of narcotics and can be obtained by electrometric titrations. Those groups in narcotic molecules which respond to such measurements are shown in table I in column 3, designated by means of theletter E.
The acidic dissociation constant exponent, pK Aof alkaloids in various solvents is found from the titration values of the pH1/3, pH1/2 and pH2/3, read on a pH meter where the fractions 1/3,1/2, and 2/3 represent degrees of neutralization. The pK Avalues are derived from the dissociation constant as follows.
or pK B = pK w- pK A, where pK w for pure water is 14.2 at 20 degrees C. In mixed solvents such as alcohol and water, the value of pK w will be higher, 14.5 and 14.8 for 50 and 75 per cent alcohol, respectively [35] .
The acid dissociation constants are employed in describing the "strength" of bases, which are represented by Q as part of the "acid" [QH +].
The weaker the base the smaller is pK A and the larger is pK B.
The method of carrying out electrometric aqueous titrations is simplified by the glass electrode for H+-ion concentration measurements. Details for conducting such titrations are given by Saunders and Srivastava [34] and by Michaelis [35] . The range of narcotic pK A values is 8.97 - 3.76. A number of pK's for narcotics have been published by Levi and Farmilo [37] , and Kolthoff [36] .
Electrometric titrations in glacial acetic acid comprise another extension of the method which is extremely useful for titrating bases which are very weak in aqueous solutions. A recent review of the methods of non-aqueous titrations applied to narcotics and alkaloids [36] showed that papaverine, phenadoxone, pipidone, and narcotine, with. pK A values in water less than 7, did not have sharp inflection points in their titration curves in water, while in glacial acetic acid, titration with perchloric acid yielded sharp end points.
Polarography[38]
Polarography is another method dependent upon the measurement of electromotive force and is useful in quantitative and qualitative analysis of narcotics. Polarography detects the presence of certain functional groups (shown in table I, column 3, by label P) which can be either oxidized or reduced according to the following equation:
oxidant + n e -> reductant
Polarography combines electrolysis with the measurement of electromotive force. The electrolysis is always of short duration and never complete, if the electric current is prevented from altering by use of a microelectrode in a large volume of solution, the other electrode being kept large and unpolarizable. Polarography is concerned with electrode reactions, involving transfer of electrons. The electrode is the cathode when reduction occurs, and if the electrons leave the cathode to reduce the oxidant; and is the anode when oxidation occurs, i.e., electrons leave the solution to enter the anode and leave behind an equivalent amount of the oxidant. During the reactions at the microelectrode the macroelectrode becomes alternately the anode and cathode respectively. The narcotic in solution either is oxidant or reductant, and subsequent products from electrode reaction form an oxidation-reduction system.
The use of polarography in narcotic analysis has been limited although several compounds have chemical groups which are ordinarily capable of entering into reduction-oxidation reactions. These narcotics are shown in table I as being potentially analysable by means of polarography. Berberine, cotarnine, and hydrastinine in pharmaceutical preparations have been determined polarographically by Sentavy [39] .
CONCLUSIONS
Narcotic identifications are usually carried out by means of color and crystal tests. A closer adherence to the classical schemes of organic analysis may be warranted for qualitative narcotic identification. A general scheme of analysis has been proposed in order to introduce the necessary modifications in current methods of narcotic analysis. One short-coming of current chemical methods is that they change the original narcotic to such an extent that it is not available for further examination. Since some common physical properties are not apparently too useful in the classification step of the proposed analytical procedure, other physical data must be obtained, for complete identification in the classical organic sense. A number of methods based on size, shape, atomic configuration and functional chemical groups of narcotics have been studied and assessed in order to determine their applicability to the special problem of narcotic analysis. Three methods, X-ray powder diffraction, and ultra-violet and infrared spectrometry, were selected from a larger list of physical methods and have been studied by us extensively, in application to narcotics. These methods will be reviewed in detail in subsequent papers. A number of other physical methods are apparently also useful and will be the subject of still further work.
With regard to specific methods there are two broad, similar uses for ultra-violet spectrophotometry, electrometric titrations and polarographic determinations in narcotic analysis. First, to classify a narcotic into one of four large chemical families, and sometimes restricting it into a smaller subdivision of these families. Second, the methods may be used to determine accurately the amount of the substance present.
Infra-red is most useful in determination of specific chemical functional groups, and with additional data we may be able to eliminate numbers of possible materials, and to select the most likely compound from the list of possibilities.
Raman and fluorescence spectra need to be obtained before the usefulness of these methods can be assessed practically.
A complete crystallographic description appears to be the most desirable way of identifying completely a pure crystalline drug. The most useful single physical method that we have investigated for purposes of narcotic identification is the X-ray diffraction powder method. It has been tested by us successfully in the detection of drugs of doubtful purity obtained from toxicological cases. Optical crystallography is needed in addition to X-ray diffraction for establishing complete and unequivocal identity of a pure organic crystalline phase.
Hartley, Wm. N. "The Absorption Spectra of Alkaloids", Philosophical Transactions of the Royal Society of London, 176, II, 471 (1885).
002Fischer, Hans. Die physikalische Chemie in der gerichtlichen Medizin und in der Toxicologie mit spezieller Berücksichligung der Spektrographie und der Fluoreszenzmethoden, Arbeiten aus dem Institut fur physikalische Chemie der Universität Zurich, Direktor: Prof. Dr. V. Henri, aus dem gerichtlich medizinischen Institut der Universität Zurich, Direktor: Prof. Dr. N. Zangger. Zurich 1925, Kommissionsverlag der Buchhandlung A. Rudolf, Part VII, "Die Anwendung der Spektroskopie in der gerichtlichen Medizin als Feststellung Methodik speziell zum toxikologischen Nachweis von Alkaloiden". Separatdruck aus H. Fischer. Buchdruckerei Stafa A. G. vorm. E. Gull, Zurich, 1925.
003Hochgesang, F. P. and Schlesman, C. H. " Electronics in Petroleum Plants", Electronics, 17,116, 308 (1944).
004Shriner, R. L. and Fuson, R. C. The Systematic Identification of Organic Compounds, second edition, 1940, John Wiley and Sons, Inc., chap. 1, p. 142.
005Cheronis, N. D. and Entrikin, J. B. Semimicro Qualitative Organic Analysis, 1947, Thomas Y. Crowell Company, New York. Chap. 1, pp. 3-9.
006Morgan, C. E. Methods for the Collection and Analysis of Horse Saliva and Urine for the Detection of Drugs, 7th revision, 2nd printing, 1951. New York State Racing Commission Laboratory, Jamaica, New York.
007Farmilo, C. G., Levi, L., Oestreicher, P. M., and Ross, R. J. "Studies on the Identification of Narcotics, III. Microcrystal and Colour Tests for the New Synthetic Narcotics", Bulletin on Narcotics, IV, No. 4, 16-41 (1952).
008Keenan, G. L. "Notes on the Microscopy of some Important Alkaloids", The Chemist Analyst, 39, 33, 52, 79 (1950); ibid., 40, 4, 28 (1951).
009Fulton, C. C. "Alkaloids and Their Reagents", American Journal of Pharmacy, III, No. 5, 183-191 (1939).
010Levi, L., and Farmilo; C. G. "The Characterization of Narcotics as Reineckates", Canadian Journal of Chemistry, 30, 783-792 (1952).
011Levi, L., Oestreicher, P. M. L., Farmilo, C. G., and Pugsley, L. I. The Reaction of Morphine with Potassium Mercuric Iodide, paper presented at 124th National American Chemical Society Meeting, Chicago, Sept. 8, 1953.
012Levi, L., and Farmilo, C. G. "Quantitative Estimation of Amine-Cadmium Halide Complexes, by Titration in Non-aqueous Solvents", Anal. Chemistry, 25, 409-913 (1953).
013Lilly, E. and Co. "Analgesics", Research Today, IX, No. 1, 1-27 (1953).
014Eddy, N. B., May, E. L., and Mosettig, E. "Chemistry and Pharmacology of the Methadols and Acetylmethadols", Journal of Organic Chemistry, 17, 321-326 (1952).
015Ziering, Albert, and Lee, J. "Piperidine derivatives V. 1, 3-dialkyl-4-aryl-4-acyloxypiperidines", Journal of Organic Chemistry, XII, 911-914 (1947).
016Fraser, H. F., and Isbell, H. "Actions and Addiction Liabilities of Alphaacetylmethadols in Man", The Journal of Pharmacology and Experimenlal Therapeutics, 105, No. 4, 458-465 (1952).
017Eddy, N. B., Touchberry, C. F., and Lieberman, J. E. "Synthetic Analgesics. I. Methadone Isomers and Derivatives'', ibid., 98, 121-137 (1950).
018Wright, F. E. "The Petrographic Microscope in Analysis", contribution from the Geophysical Laboratory in Washington, D.C., Journal of the American Chemical Society, 38, 1647-1658 (1916).
019Wherry, E. T. "The Application of Optical Methods of Identification to Alkaloids and Other Organic Compounds", United States Dept. of Agriculture Bulletin, No. 679, 9 pp., 1918.
020Keenan, G. L. "The Microscopic Identification of Demerol", Journal of the American Pharmaceutical Association, scientific edition, 35, 338-339 (1946).
021Food and Drug Circular No. 1. Microanalysis of Food and Drug Products, Federal Security Agency, Food and Drug Administration, prepared by the Microanalytical Division, 1951.
022Wickstrom, A. "The Optical Crystallographic Properties of' Crystals Formed with Nitric Acid from Pethidine, Nicotinamide and Cincophen", Journal of Pharmacy and Pharmacology, 2, 444-445 (1950).
023Huback, C. E. and Jones, F. T. "Methadon Hydrochloride, Optical Properties, MicrochemicaI Reactions and X-ray Diffraction Data", Analytical Chemistry, 22, 595-598 (1950).
024Keenan, G. L. "Microscopic Characters of Some of the Crystalline Sympathomimetic Amines", Journal of the American Pharmaceutical Association, scientific edition, 38, No. 6, 1949
025Haley, T. J. and Keenan, G. L. "Microtoxicology. IV. The Identification of Antihistaminic Drugs of the Thenyl Series", The Journal of the American Pharmaceutical Association, scientific edition, 38, No. 2, 85-88 (1949). References to other articles; parts I to IV are given in this paper.
026Chamot, E. M. and Mason, C. W. Handbook of Chemical Microscopy, vol. 1, "Principles and Use of Microscopes and Accessories, Physical Methods for the Study of Chemical Problems", chapter X, p. 332, first edition, 1930.
027Lilly, E. Research Today, "Physical Methods Useful in the Identification and Structure Study of Organic Chemicals", vol. IV, No. 3, pp. 96-125 (1947).
028Rogers, A. F. and Kerr, P. F. Optical Mineralogy, second edition, 1942. Chap. VIII, p. 135 and others.
029Barnes, W. H. ''Physical Methods for the Identification of Narcotics, Part IIA. The X-ray Diffraction Powder Method", to be published.
030Barnes, W. H. and Shepard, Helen, M. "Physical Methods for the Identification of Narcotics, Part IIB. X-ray Diffraction Powder Data for Eighty-three Narcotics", to be published.
031West, W. "Spectroscopy and Spectrophotometry", chap. XVII, Physical Methods of Organic Chemistry, edited by A. Weissberger, vol. 11, pp. 806-815 (1946).
032Ibid., p. 794.
033White, C. E. "Fluorometric Analysis", Analytical Chemistry, 24, 85-90 (1952).
034Saunders, L. and Srivastava, R. S. "The Potentiometric Titration of Alkaloidal Salts", Journal of Pharmacy and Pharmacology, 3, 78-86 (1951).
035Michaelis, L. "Potentiometry", chap. XXII, Physical Methods .for Organic Chemistry, edited by Weissberger, A., published by Interscience Publishers, Inc., New York, 1946.
036Kolthoff, I. M. "Die Dissoziationskonstante, das Löslichkeitsprodukt und die Titrierbarkeit yon Alkaloiden", Bitchemische Zeitschrift, 162, 289-353 (1925).
037Levi, L., Oestreicher, P. M., and Farmilo, C. G. "Non-aqueous Titration of Narcotics and Alkaloids", Bulletin on Narcotics, V, No. 1, Jan.-Mar., 15-25 (1953).
038Muller, O. H. "Polarography", chapter XXIII, Physical Methods of Organic Chemistry, see reference 35 for details. p. 1115 ff.
039S?ntav?, F. "The Polarography of Berberine, Hydrastinine and Cotarnine", contributed from Research Dept. of Inst. for Radiotherapy, Praha Bulorka. Collection of Czechoslovakian Chemical Communications, 14, 377 (1949).
040Heller, W. "Polarimetry", chapter XIX, Physical Methods of Organic Chemistry (see reference 35 for details), p. 869 ff.
041Farmilo, C. G., Oestreicher, P., and Levi, L. "Physical Methods for the Identification of Narcotics. Part IB. Physical Data for the Identification of Ninety-four Narcotics", to be published.
042Kasha, M. Luminescence Spectroscopy of Molecules, Florida State University, U.S.A., paper 68, Symposium on Modern Tools for Organic Research. 124th meeting, American Chemical Society, Chicago, Illinois, September 6 to 11, 1953.