October, 19611 MORGAN 631 The Identification of Small Amounts of Bases in Urine by Infra-red Spectrophotometry BY P. J. MORGAN (Department of Pharmacology, University of Melbourne, Australia) A method is described for the rapid extraction of organic bases from urine, in amounts of the order of lOOpg, in a sufficiently pure condition for identification by infra-red absorption spectrophotometry. THE analytical significance of all organic compounds having highly characteristic absorption spectra in the wavelength range between 2.5 and 15 p has, in recent years, resulted in the availability of instruments that record spectra in this range and yet do not call for a pro- hibitively large capital outlay in view of the useful work they can perform. An objection often made when considering the application of infra-red absorption to substances extracted from biological sources is that any associated impurities will contribute absorption bands similar in intensity to those given by the substance being studied. The extraction procedure described here is sufficiently selective when applied to urine to permit this disadvantage to be overcome.The procedure is an adaptation of the acid-spot extraction method previously describedl for detecting organic bases in urine at the 1 p.p.m. level. In this method, bases are trans- ferred from a chloroform extract of the urine to spots of dilute sulphuric acid supported on strips of Whatman 3MM filter-paper and are detected by paper chromatography of the strips in acidified solvents after exposure to ammonia vapour.The procedure described here supplements this method in that a larger amount of any base detected on the paper chromatograms can be rapidly isolated in a relatively pure condition. It also facilitates the rapid detection of bases, such as morphine, not readily soluble in organic solvents, when these are present in small concentrations in the urine. APPARATUS- A modified form of the acid-spot extraction apparatus described previously is used (see Fig. 1). It consists essentially of an inner jacket conveniently made from a Liebig condenser 14 inches in diameter and cut off square near one end. It should be made as long as possible, but should still maintain reasonable rigidity of the central tube; a length of 12 inches was found to be satisfactory. The tip of the central tube is constricted in a flame to give an opening about $ inch in diameter. The open end of the Liebig jacket fits into a cylindrical beaker 18 to 12 inches in diameter and 74 inches tall, which is supported in a water bath.The outlet at the top of the Liebig jacket is connected by means of an 18-inch length of polythene tubing to a test-tube fitted with a side-arm. This serves as a trap for condensed solvent vapours and may be cooled if desired. An acidified solution of chromous sulphate, concentrated sulphuric acid and a U-tube of silica gel may be used as previously describedl to provide a dry stream of nitrogen. If the gas is passed through a U-tube of fresh self-indicating silica gel, this will change in colour from pink to blue if the gas is being satisfactorily dried.ACID-SPOT EXTRACTION PROCEDURE- A chloroform extract of the urine is prepared by gentle agitation of the specimen with 100ml of chloroform after adjustment of pH to a suitable value with 20 per cent. sodium hydroxide solution as previously described.l The chloroform extract is introduced into the beaker, which is supported in the water bath at a temperature at least 20" C above that of the room. Spots of 0.25 and 0-5 N sulphuric acid - thymol blue reagent1 are placed about 3 cm from one end of a strip of Whatman 3MM filter-paper having the same width as the Liebig jacket and of any desired length. When the spots have dried somewhat, the strip is inserted into the Liebig jacket, which is then clamped in the beaker at a slight angle to the vertical.The angle of tilt and the arrangement of the polythene tubing should be such that any solvent METHOD632 MORGAN: THE IDENTIFICATION OF SMALL AMOUNTS OF [Vol. 86 condensing in the polythene tube does not run back into the chloroform extract. The acid spots should be below the surface of the chloroform extract, which is agitated by passing a brisk stream of dry oxygen-free nitrogen through the central tube of the jacket. Two filter- paper strips may be inserted if so desired, one on each side of the central tube. Extraction of bases proceeds on to the acid spots, which are examined from time to time for any change in colour of the thymol blue indicator. If the red colour of the spot of 0.25 N acid shows any change towards neutrality (yellow) when compared with the spot of 0.5 N acid, the strip is immediately removed and replaced by a fresh strip.Generally, any early tendency to neutralisation is rapidly overcome as volatile bases, such as ammonia and amines having low molecular weights, are swept out of the chloroform extract by the stream of gas. The tendency of chloroform to emulsify in the urine generally makes a second extraction with a further portion of chloroform desirable; further, for a base not particularly soluble in chloroform, several successive extractions may be needed to extract most of the base. These portions of solvent can be added via the side of the beaker from time to time as chloroform evaporates into the stream of nitmgen and serve to replenish base that is being removed from the solution on to the acid spots.\ extract- To solvent c.- 0 - i - trap Solvent iner jacket ilter paper strip *Beaker ,Acid spots Fig. 1. Acid-spot extraction apparatus Any thick emulsion remaining after the final extraction with chloroform can be broken as follows. The emulsion is run, in small portions, from a separating funnel into a flask and then shaken vigorously. If the proportion of non-aqueous phase is sufficiently high, the shaking will break the emulsion. When this procedure is no longer effective, small amounts of cyclo- hexane are added to the emulsion, and the vigorous shaking is repeated; any further organic phase that separates is added to the main extract in the acid-spot extraction apparatus (with filtration through a dry filter-paper if it is turbid).As a last resort, vigorous shaking, together with the addition of small amounts of anhydrous sodium sulphate, will result in further breaking of the emulsion. Since cyclohexane is a much poorer solvent than chloroform and is also appreciably less volatile, its introduction into the extraction system will lead to an increase in the rate of extraction of most bases on to the acid spots in the final stages. The lack of solubility of some extremely weak bases, such as caffeine, in cyclohexane can be utilised to achieve acid- spot extraction of these bases when they occur in the urine at concentrations of about 1 p.p.m.October, 19611 BASES IN URINE BY INFRA-RED SPECTROPHOTOMETRY 633 It is preferable to pass nitrogen until the volume of chloroform has been decreased to about 50 ml and then to dilute with cyclohexane to about 200 ml.The passage of gas is continued until the chloroform content of the extract has been decreased to less than 1 per cent. by volume; this operation is conveniently carried out overnight at 45" C. The chloroform content is easily determined by recording an infra-red spectrum in a cell 0.05 mm thick, when the intense chloroform absorption band at 13-1 p can be used to determine concentrations of chloroform of less than 1 per cent., the weaker band near 8.3 p in the range 1 to 10 per cent. and the weak cyclohexane band near 8-0 p when the chloroform content is more than 10 per cent. by volume. PAPER CHROMATOGRAPHY OF ACID-SPOT EXTRACTS- Chromatography of the strips removed from the acid-spot extraction apparatus can be carried out as previously describedl by exposing them to ammonia vapour for a short time to neutralise the excess of sulphuric acid on the spots and then developing in either the n-butyl alcohol - hydrochloric acid or the isobutyl methyl ketone - acetic acid solvent. These solvents were chosen for the rapid detection of bases because they give good chromatograms without the need for any equilibration; however, for purposes of isolation, it is preferable to allow some time for equilibration, particularly if the thick Whatman 3MM filter-paper is used.Detection of typical alkaloids is achieved, as before, by spraying the chromatograms with the iodoplatinate reagent. For the detection of the weak base caffeine it is essential to add about 6 per cent.of concentrated hydrochloric acid to this reagent. Amines of lower molecular weight, which may not react satisfactorily with the iodoplatinic acid to give the coloured precipitate on the paper, are detected by spraying with bromocresol green in ethanol after development in the isobutyl methyl ketone - acetic acid solvent. A base may be recovered in good yield from chromatograms that have been sprayed with either or both of these reagents by cutting out the coloured spots, placing them in a small Erlenmeyer flask (Quickfit & Quartz Ltd.) and adding sufficient concentrated ammonia solution just to moisten the paper completely. The stoppered flask is shaken in a mechanical shaker for 30 minutes with sufficient chloroform just to cover all the pieces of paper (the shaking should not be so vigorous that the paper disintegrates). If this condition is observed, it is possible to check for completeness of extraction of base by drying the pieces of paper and re-spraying them with the iodoplatinate reagent after decantation of the chloroform extract.If any cellulose fibres are detached by over-vigorous shaking, or if too much ammonia has been used, then the extract must be filtered, and completeness of extraction checked by re-shaking the residue plus the filtering medium with a further amount of chloro- form. Prolonged shaking of the initial extract should be avoided, as it may lead only to a higher level of impurities in the extract. When this procedure is adhered to, it has been found that bases are obtained in a sufficient degree of purity to give a useful infra-red spectrum ; it is best to aim a t the collection of a few hundred micrograms of base from about ten chromatographic spots.INFRA-RED SPECTROPHOTOMETRY OF EXTRACTS- Spectrophotometry is best carried out in solution in an organic solvent in micro cells having sodium chloride windows. The chloroform extract from the chromatographic spots must be evaporated to dryness before double-beam spectrophotometry can be attempted, even if chloroform is the solvent of choice. This is conveniently carried out in a glass tube having a narrow tip, such as a centrifuge tube, by passing a stream of nitrogen through the extract at 40" C. Vapour condensing on the cooler sides of the tube eventually washes all the solute down into a small region a t the narrow tip, so that, when the amount of solvent required to fill the micro cell is added, the solute is totally covered.Solution may be assisted by the application of gentle heat to the tip of the tube if precautions are taken to avoid loss of solvent by evaporation. After a spectrum has been recorded, some sort of purification technique should be applied to the sample. This purification may take the form of a further paper-chromatographic stage if the level of impurities Appears to be high. It is preferable to use a solvent system different from that used for developing the first chromatograms of the acid spots. Infra-red absorption bands that disappear or are decreased in absorption relative to other bands in634 MORGAN: THE IDENTIFICATION OF SMALL AMOUNTS OF [Vol.86 the spectrum after the purification can then be ascribed to the impurities present. When paper chromatography is used, it is never possible to get rid of the last traces of absorption (due to organic impurities from the paper) in the carbon - hydrogen stretching and bending regions of the spectrum. Sublimation in vacuo is often a most effective final procedure for purification. The use of a double-beam spectrophotometer recording from 2-5 to at least 1 5 p is essential for the satisfactory study of the small amounts of material being examined, Instru- ments that cover this range are usually fitted with a sodium chloride prism for dispersing the radiation ; however, an instrument having a diffraction grating for dispersing the radiation is generally capable of giving better results in the shorter-wavelength portion of the spectrum.A choice of two widely different recording speeds is desirable, as this permits more careful coverage of spectral regions in which absorption from the solvent is high. To cover regions in which absorption from the solvent is extremely high, the use of an alternative solvent is necessary. A laboratory-reagent grade of tetrachloroethylene (fractionated before use, with rejection of the initial and final fractions) was found to be most satisfactory for work on the micro scale; this solvent keeps well in a dark-glass bottle. When a fairly large amount of material has been isolated in a substantially pure condition, simple washing of the sample with small portions of tetrachloroethylene may remove sufficient of the impurities present to permit detection of the absorption bands associated with the substance under investigation.I t is recommended that micro cells about 0-3 and 1.0 rrim thick should be used for work with tetra- chloroethylene. It should be noted that the extraction method described is biased in favour of substances readily soluble in organic solvents and that thicker cells may be required for some solutes. When lack of solubility in organic solvents is associated with the presence of hydroxyl groups in the molecule, it is sometimes expedient to improve the solubility by partial acetylation of the substance. For example, it was found that conversion of morphine into its monoacetyl-derivative permitted a much more detailed spectrum to be obtained in a cell 0.3 mm thick than could be obtained from morphine itself in a cell ten times as thick.The fact that hydroxyl groups can forrn intermolecular hydrogen bonds, even in fairly dilute solution, should be kept in mind when two spectra are being compared, as, when such bonds are formed, the sharp absorption band due to free hydroxyl is replaced by a much broader band at a longer wavelength. Some general aspects of infra-red solvents have been discussed by Cole.2 The use of tetrachloroethylene has been discussed elsewhere.3 DISCUSSION OF THE METHOD The extraction method described is based on a 500-ml sample of urine, but it can be applied as it stands to larger or smaller samples, within reason. Up to 5 per cent.of ethanol may be added to the chloroform used for extraction of the urine in order to improve the rate of extraction of substances, e.g., morphine, soluble only with difficulty, but it is doubtful whether this offers any real advantage, since not only are more impurities extracted from the urine, but difficulty may also be caused by emulsification of solvent in the urine. In order to minimise loss of thymol blue indicator from the acid spots, it is advisable that they should be allowed to dry in air before introdiiction into the chloroform extract ; however, it should be realised that the spots will pick up any volatile bases present in the laboratory atmosphere if they are left exposed for too long a period. Since the thymol blue indicator is retained near the centre of the spot when the acid spots are applied to the paper, it is possible to follow the progress of extraction by adding a drop of an extremely dilute solution of the basic dye methylene blue in ethanol to the chloroform extract.The extracted dye also serves as a standard of R, value when the extracts are subjected to chromatography. Since the colour of the thymol blue indicator may occasionally not be visible when the strips are removed, a fairly concentrated solution of thymol blue in ethanol should be kept at hand and applied to the spots with a loop of platinum wire if there is any doubt. Unless the spots have retained their acidity to thymol blue, there is no guarantee that weak bases initially extracted on to the spots have not been displaced by stronger bases.If long strips of paper are used in the apparatus, it is essential to keep the chloroform extract at a temperature substantially higher than that of the room so that some condensation of solvent vapour will occur on the portion of the strip not immersed in the solvent. This serves to keep the upper part of the strip washed free of solutes that may otherwise accumulateOctober, 19611 BASES IN URINE BY INFRA-RED SPECTROPHOTOMETRY 635 there. If short strips completely immersed in the extract are used, it is not necessary to apply heat, and the risk of loss of any labile material is therefore substantially decreased; however, the time needed for extraction is increased, and chromatography of short strips presents some difficulties, as it is necessary to place the spots of acid close to the end of the strip, and this results in too fast a flow of solvent over the spots when chromatography is begun.The end of the strip may be cut into a serrated pattern, with the tips of the serrations just touching the solvent when chromatography is begun, or wire staples may be used to attach another piece of paper to the end of the strip to serve as a “lead-up” for the solvent. Alternatively, if the highest possible purity is to be sought from the chromatogram, a small rectangle containing the spots This defect may be overcome in three ways, as shown in Fig. 2. Two-phase system S ingle-phase system Fig. 2. is cut out and placed centrally between two glass plates. One edge of the assembly is grasped between thumb and fore-finger so that the other edge opens out slightly, with the edge of the paper as fulcrum, and a short “lead-up” length of paper is inserted to meet the rectangle, with a slight overlap.The pressure is then transferred to the other side of the assembly to permit similar insertion of a longer piece of paper on the other side of the rectangle. The assembly is then clamped together with clothes pegs of the spring-type, and chromatography is carried out by the ascending-solvent technique. Some methods for paper chromatograms TABLE I LIMITS OF DETECTION OF VARIOUS BASES IN URINE Base Limit of detection (as concentration in original sample) in- f A \ solution after concentration cyclohexane solution initial chloroform of several extracts containing 1% v/v extract at pH 9, to 50 ml, of chloroform, p.p.m.Strychnine . . . . 2 Cocaine. . .. .. 2 Amphetamine . . . . 5 Morphine .. . . 8 Caffeine .. . . 100 p.p.m. 1 1 2 2 40 p.p.m. ( 1 1 1 2 <1 In the final stages of the acid-spot extraction from the cyclohexane-rich solvent, some solid matter may be deposited near the tip of the gas-delivery tube. This can be easily re-dissolved by placing the delivery tube for 1 or 2 minutes in a few drops of chloroform in a test-tube; the solution is then washed into the main extract with a little cyclohexane. Table I gives the approximate limits of detection for some bases at various stages of the procedure; the limits shown are based on practical experience of the method. The limit636 FEW: A VANILLIN - PERCHLORIC ACID REAGENT FOR DETECTING [Vol. 86 of detection attained depends largely on the skill and experience of the operator and involves factors such as the technique of spraying the chrornatograms. For instance, if both sides of a chromatogram are sprayed with the acidified iodoplatinate reagent, it is usually found that the sensitivity is greatly increased. Under these conditions, ammonium ion may develop as a grey spot on the pink background at a low R, value. It is sound practice initially to carry out some blank extractions in order to ascertain whether or not (a) the solvents or paper used make any contribution and (b) the ammonia introduced just before development of the chromatogram can be detected. I acknowledge that this work was made possible by grants from the Victoria Racing Club and the Anti-Cancer Council of Victoria. REFERENCIE S 1. 2. 3. Morgan, P. J., Analyst, 1959, 84, 418. Cole, A. R. H., Rev. Pure AppZ. Chem., 1954, 4, 118. Ard, J. S., Anal. Chem., 1953, 25, 1743. Received May 29th, 1961