US3786249A

US3786249A - Negative ion duoplasmatron mass spectrometer for isotope ratio analysis

Info

Publication number: US3786249A
Application number: US00209528A
Authority: US
Inventors: M Anbar; W Aberth
Original assignee: Stanford Research Institute
Current assignee: SRI International Inc
Priority date: 1971-12-20
Filing date: 1971-12-20
Publication date: 1974-01-15
Anticipated expiration: 1991-01-15

Abstract

A technique for performing isotope ratio analysis is described which utilizes a negative ion mass spectrometer coupled with a negative ion duoplasmatron source. An untreated sample to be analyzed for a specific compound is diluted with a known amount of isotopic analog. The sample plus analog is volatilized if necessary, and purified by gas chromatography before entering the ion source. The extremely hot plasma of the duoplasmatron thoroughly fragments the large molecules and efficiently ionizes the input gas stream. The resulting ionized stream is analyzed in a magnetic mass spectrometer. The signals from a pair of peaks representing the common and noncommon tag isotope of the same element are converted to frequency signals and the frequency signals are converted to a ratio signal by a frequency ratio counter. The value of this ratio then indicates the precise amount of compound in the sample. The system operates very efficiently on stable, non-radioactive isotopes and is applicable to many analyses which cannot be performed with radioactive isotopes.

Description

United States Patent [191 Anbar et al.

[ 51 Jan. 15,1974

[75] Inventors: Michael Anbar; William H. Aberth,

both of Palo Alto, Calif.

[73] Assignee: Stanford Research Institute, Menlo Park, Calif.

[22] Filed: Dec. 20, 1971 [21] Appl. No.: 209,528

OTHER PUBLICATIONS Characteristics of a Low Energy Duoplasmatron Negative lon Source Aberth et a1. Rev. Sci. Inst; June 1967.

Primary Examiner.lames W. Lawrence Assistant Examiner-B. C. Anderson Attorney-Samuel Lindenberg et al.

[57] ABSTRACT A technique for performing isotope ratio analysis is described which utilizes a negative ion mass spectrometer coupled with a negative ion duoplasmatron source. An untreated sample to be analyzed for a specific compound is diluted with a known amount of isotopic analog. The sample plus analog is volatilized if necessary, and purified by gas chromatography before entering the ion source. The extremely hot plasma of the duoplasmatron thoroughly fragments the large molecules and efficiently' ionizes the input gas stream. The resulting ionized stream is analyzed in a magnetic mass spectrometer. The signals from a pairof peaks representing the common and noncommon tag isotope of the same element are converted to frequency signals and the frequency signals are converted to a ratio signal by a frequency ratio counter. The value of this ratio then indicates the precise amount of compound in the sample. The system operates very efficiently on stable, non-radioactive isotopes and is applicable to many analyses which cannot be performed with radioactive isotopes.

21 Claims, 1 Drawing Figure PLE ND SAM A VAPORIZER GAS ISOTOPIC ANALOG CH ROMATOGRAPH AND CARRIER GAS NEGATIVE ION DUOPLASMATRON MASS SPECTROMETER FOR ISOTOPE RATIO ANALYSIS BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an isotope dilution analysis technique for detecting and quantitatively determining the quantity of compounds in fluids with a sensitivity approaching grams per sample, and more particularly, to a technique for performing isotope ratio analysis by means of a negative ion mass spectrometer coupled with a negative ion duoplasma tron source.

2. Description of the Prior Art The most effective and unambiguous way to follow the fate of a given material throughout any system is the use of tags, which cannot be removed by any practical means. lsotopically-substituted molecules are such tags. Organic molecules can be isotopically labeled by substituttion of one or more of their atoms with a noncommon isotope, radioactive or stable. Such labeling makes the given molecules unique and allows their tracing through chemical, biological or even commercial systems.

One of the major problems in assessing the applicability of a drug for human consumption is to secure prior information on its biological half-life in circulation and its metabolic fate. It has been shown that it is very difficult and occasionally misleading to obtain this type of information from animal studies. The most appropriate way to obtain this information is to administer minute amounts of a labeled drug into the human system and to follow the change in its level in blood, urine, and feces. The need for a label is critical because conventional chemical analysis becomes impractical in most cases at concentrations below 10 moles l0 g). Up to date only radioactive isotopes provided a practical means for isotope dilution analysis.

The use of radioactive tracers in humans can be allowed in certain clinical situations but can hardly be justified for routine clinical analysis. Moreover, they are not generally admissible as tracers for commercial products which are likely to be consumed by humans. There is a calculated risk in the introduction of any amount of ratioisotopes into a subject system, and one should avoid the exposure of any individual to a risk. The use of radioactive tracers in chemical diagnosis suffers from three drawbacks: (a) potential radiological damage which prevents its application for routine clinical analysis and limits it to adult patients; (b) the limited halflife of the isotopes approved for clinical use, which restricts their use to short-time tests and prevents extensive loading studies; and (c) where high sensitivity is required. high specific activities have to be used. Under these conditions there appear serious uncertainties about the chemical identity of the labeled material which might readily undergo chemical changes induced by its own radiation. It may be concluded, therefore, that the only way of tracing drugs by isotopic tagging through a sizable segment of the population is by using a nonradioactive isotope.

Many chemical elements found in nature consists of one isotope that constitutes over 90 percent of all the atoms, and one or more other isotopes that constitute a few percent or a fraction thereof (See Table I). The

isotopes D and C are less abundant in nature than H and C. Thus, natural methane (CH will contain 1.108% CH and 4 X 0.0148 0. 0592% Cl-l D, but only 1.6 X l0"% "CH D and 9 X 10'6% "CH D The isotopic composition of nonradioactive compounds is assessed most accurately by mass spectrometry. A high resolution mass spectrometer (M/AM 20,000) can distinguish not only between, say CH and CH but also between CH (mol wt 17.034654) and CH D (mol wt= 17.037577); M/AM 17.038/0.0029 5,873. Likewise, CI-I D (mol wt 18.040931) may be distinguished from CH D (mol wt 18.04385); AM 0.0029. This high resolution is necessary in practice because the :mass spectrometric analysis using conventional electron bombardment ionization sources results in a complex fragmentation spectrum of the analyzed species. Taking a most elementary example: ethane, and using C H D as tracer, one can obtain, with a certain electron impact ionization source, CPL- as a predominant peak, but CH CHJ, and CH would also be present in the fragmentation spectrum to an extent depending on the geometry of the ionization source and gas pressure. Mass 16 may consist of CH D (originating from the tracer molecule) plus CH; and CH.,*.. Mass 15 will incl rde, in addition to Cl-l CH and CHD. Mass 17 will consist of CH plus CH Cl-1 D", ch D and CH D2+. Only a high resolution mass spectrometer analyzer can distinguish between these different species and allow the assessment of CH D compared with CH which gives a measure of the C H D/C H ratio in the sample.

It is extremely difficult using high resolution mass spectrometry to make the ratios between the ionic fragmetabolites which of, say, 10 carbon atoms and 20 hydrogen atoms plus oxygen, nitrogen and occasionally sulfur or phosphorus, the fragmentation pattern becomes increasingly complex, and a dozen different species may contribute to a certain mass number peak. Thus, the resolution of these samples into individual isotopic constituents may in many cases exceed the capabilities of even the best available high resolution mass spectrometer. It is for this reason that high resolution mass spectrometers, while useful in identification of unknown organic compounds through their fragmentation spectrum have a very limited capability in isotopic tracer analysis.

One possible way to overcome the difficulties in isotopic tracer analysis of organic compounds is to obtain a representative sample of the isotopic species in an ionic form which contains only one to three atoms. This can be done by various chemical reactions such as oxidation to C0, C NO, N H O etc., reduction to H or exhaustive fluorination to yield 0 N and CF...

In the case of hydrogen isotopic determination, in organic compounds, for instance, the GC effluent is oxidized to H O over CuO and reduced to H over Zn or U followed by further gas chromatograph separation before introduction into an MS. The two preliminary treatments require a minimum of 10""g to avoid dilution by trace impurities and surface memory effects. The mass spectrum is complicated by theprese nce of Hi and requires several measurements 2ft different pressures. As hydrogen constitutes only about 10 percent of most organic compositions the amount of sample required for precise measurement of H/O may be as high as 20 mg.

The present technique for determination of isotopic abundance of carbon involves oxidation of the test compound to CO following gas chromatographic purification. An additional gas chromatographic purification stage is required, however, before introduction into the mass spectrometer because N O formed in trace amounts in the oxidation stage, as well as N 0 formed in the ion source from N0 would impair the precision measurement of masses 4Tand 4K The relatively low sensitivity of the electron impact ionization source requires minimum amounts of 10 moles or 3 5 X l0 g of CO for precision determination.

isotopic oxygen in organic compounds can be determined with a sufficient precision only by exhaustive fluorination, using highly elaborate instrumentation and techniques. Such a fluorination system can handle l0' g quantities but lO' g are most desirable in practice. Extraction of the oxygen from a sample purified by a gas chromatograph takes over 2 hours. This oxygen must now be repurified by gas chromatography to remove traces of F and HCl (mass 36) prior to introduction into the mass spectrometer.

OBJECTS AND SUMMARY OF THE INVENTION It is therefore an object of this invention to provide an improved technique for isotopic tracer analysis of compounds.

A further object of the invention is to provide an accurate technique for performing isotopic ratio analysis which greatly simplifies the measuring processes and improves the overall accuracy of the ratio measurement for a number of atomic species.

Yet another object of the invention is the provision of precision isotopic ratio analysis which can be performed with a much higher sensitivity and at a considerably lower cost than previously possible.

Still another object of the invention is the provision of an isotope dilution analysis technique for detecting and quantitatively determining drug levels in body fluids with the sensitivity approaching gram per sample.

These and other objects and many attendant advantages will become apparent as the description proceeds.

The analysis technique in accordance with the invention performs isotopic ratio analysis by negative ion mass spectrometric analysis on a negative ion stream generated in a duoplasmatron source. This combination of devices greatly simplifies measurement and improves the overall accuracy of the ratio measurement for a number of species. To a sample containing an unknown concentration of a known compound, a known amount of the same compound with a different isotopic I composition is added. The combined mixture is preliminarily volatilized, if necessary, and impurities may be removed by subjecting the sample to gas chromatography before it is fed into the negative ion duoplasmatron source.

A carrier gas separator such as a ceramic disc or a organic plastic membrane permits the rejection of a major portion of the carrier gas and introduction of most of the compound and analog percent) into the ion source. The duoplasmatron has the unique combined characteristic that it utilizes an extremely hot plasma which thoroughly fragments large molecules and that it efficiently ionizes the input gas molecules yielding a net ionization efficiency of between 1 and 10 percent. The mass peak ratios of selected isotopic ions are monitored at the output of a multiple mass peak magnetic mass spectrometer. it is then possible to measure the ratio of the sample compound to the isotopic compound with an accuracy of 1 percent.

The sensitivity of the negative ion duoplasmatron mass spectrometer which is higher by over six orders of magnitude than that of conventional isotope mass spectrometers allows the determination of unknowns down to 10 gram amounts. The negative ion mass spectrum is considerably more simple than the positive ion counterpart. This is because not all atoms and molecules have an electron affinity and double negative ionization does not exist as does double, or even higher, positive ionization.

As a consequence relatively unambiguous, isotope ratios for certain atomic species can be measured in the presence of numerous other species using negative ion mass spectrometry. For example, the ratio of H'/D" can be measured unambiguously since no other negative ioris with mass numbers of l or 2 exist. H is unstable and has never been observed. The measurement of the ratios H2 IHD and ll2 /D using positive ions is complicated by the presence of H and He. To measure the isotopic ratio C/ C one must observe the ratio of CH-/( CH CD) and subtract out the contribution of CD- by knowing the H/ D abundance ratio. One could also use the CN- molecule to measure the C/ C ratio if the N/ N ratio is known. Or conversely if the C/ C ratio were known; the N/ N ratio could be measured by observing C N"/( C N"+C N-y Oxygen and sulfur isotopes could also be determined as OH- (masses 17, 18, and 19), O2- (masses 32, 33 and 34) SO- (masses 48 to 52) and S2 (masses 64 to 72%. i i The method of the invention can be used to d etermine drug concentrations in body fluids. For instance, the body fluid sample may be combined with a known amount of an isotopic analog of the drug which has its minor isotopic component of C or D altered from the naturally occuring isotopic concentration. The same technique may also enable the tracing of label drugs administered to humans when present in ppb concentrations. The technique should also make practical the labeling of commercial drugs for the identification of the manufacturer or for the date of production. Time required for complete analysis should be between 30 and 60 minutes and at moderate cost per assay.

The novel, highly sensitive mass spectrometric technique of the invention can be applied to numerous pharmacological, toxicological and physiological problems. The method can be utilized to determine the persistence and fate of drugs in humans at subpharmacological levels, to determine and assess unknown pharmacological or toxic agents in clinical samples and to determine the biological half-life and fate of physiological constituents and nutrients.

These capabilities will facilitate the testing of new drugs and will aid in clinical treatment of subjects suffering from poisoning and in forensic medicine. The method will also advance fundamental clinical studies of human nutrition and physiology and will promote better control of drug distribution and prevention of drug abuse. 1

Organic molecules labeled with stable isotopes, primarily deuterium (D) and "C can be used in the performance of the desired analysis. Only the determination of source of date of production of drugs require the administration of labeled compounds to humans. The other analyses can be performed by applying isotopic dilution analysis at the analytical stage in the laboratory.

The method of the invention can also be utilized with radioisotopic species and in nonpharmacological applications. For example, carbon dating can be performed to count very low count rates of "C by determining the ratio of CNF CH'. CN' which occurs at mass numbers 26, 27 and 28 will reflect the isotope populations of C, ""C, N, and N. CN is an extremely stable negative ion, (electron infinity of about 3.6 eV) and will present a dominant negative ion peak in discharges contaiing nitrogen and carbon. Thus, this technique will be quite superior to the classical counting techniques now utilized in carbon dating.

The invention will now become better understood by reference to the following detailed description when considered in conjunction with the accompanying drawing.

BRIEF DESCRIPTION OF THE DRAWING The FIGURE is a schematic view of a duoplasmatron mass spectrometer apparatus for producing the inven- DESCRIPTION OF THE PREFERRED EMBODIMENTS Referring now to the FIGURE, a mixture of the sam ple and a precise amount of an isotopic analog thereof are fed into the inlet 12 to the vaporizer 10, such as a retort. The vaporized mixture is dispersed in carrier gas introduced with the sample at 12. The dispersion is separated into bands within the gas chromatograph 16.

The impurity bands are vented if they are eluted first. The band containing the mixture of isotopic compounds is then separated from most of the carrier gas by a silicone rubber membrane 14 and introduced into the inlet 18 of the duoplasmatron negative ion source 20.

In the duoplasmatron an intense, high pressure are is produced by the combination of electric and magnetic fields. The intensity of this arc permits efficient extraction of negative ions from a very small aperture which is typically 0.10 inch in diameter. Many such sources have been developed for use in tandem accelerators and in negative ion collision studies. A very suitable source of negative ions for beam energies of less than 5,000 eV is disclosed in The Review of Scientific lnstruments, Vol. 38, No. 6, 745-748, June 1967.

In this source, the arc axis is offset from the ion beam axis by offsetting the anode extraction aperture a small distance, typically 0.020 inch, from the center of the are. This prevents loading of the negative ion beam with electrons from the arc. Since the strong magnetic field tends to confine the electrons to the center of the are more efficiently than the heavier negative ions, the offset aperture effectively blocks the extraction of electrons.

Referring again to the FIGURE, the source 20 comprises an arc section 22 and a focussing section 24. The source is housed in a container 26 maintained at a low pressure by a vacuum pump 28, typically about 2 to 10 X 10 Torr. The are is struck between a cathode 30 and an anode 32. The Z-electrode 34 helps to confine the arc to a narrow axial pencil. The Z-electrode is slightly displaced from the beam axis. A magnetic coil 36 coaxially surrounds the cathode 30 and Z-electrode 34. The anode 32 is suitably a disc having a central output aperture 34. The disc which is coaxial with the electrostatic focussing electrodes may be displaced from the Z-electrode axis.

As the separation band dispersed in carrier gas enters the arc, the compounds are efficiently fragmented into ionized components by the hot plasma. The negative ion stream leaves thearc section 22 through the offset aperture 34. The beam is focussed by means of

electrostatic focussing electrode

40, 42, 44 and by means of deflector shims 46, 48.

The'focussed beams then pass through the entrance slit 50 of the mass spectrometer 52. The negative ions in the beam are dispersed by the magnetic field of the mass separator and enter the slits 54 provided in the target collector 56.

The detection system of the mass spectrometer consists of a separate

electron multiplier detector

60, 62, 64, 66 coupled to ground by a high impedance resistor 68. The detectors are positioned behind each slit 54. A voltage to frequency converter 70 converts the output voltage of each multiplier to a proportional frequency signal. The signals from a pair of detectors (GO-62 and 64-66) are applied to a frequency ratio counter 72 which provides a ratio output signal which is recorded on recorder 74.

The ratio signal output is calibrated by observing the ratio produced from a natural sample and a pure isotopic tracer sample. Standard state of the art electronics enables measurement of signal ratios with a precision of one part in 10 The carrier gas can be inert or reactive with the compound and analog. If an organic compound containing hydrogen and carbon along with N as a carrier gas are introduced into a negative ion duoplasmatron source, extensive fragmenting of the drug molecule will occur and a negative ion mass analysis will yield a highly selective distribution of a few negative ions. T h us, unlike the positive ion counterpart, massesTand 2 will truly reflect the isotope population of H asap, since I-I does not exist Similarly CN- which occurs at mass numbers 26, 27, and 2. will reflect the isotope populations of "C, C, *N, and N. CN is an extremely stable negative ion (electron affinity of about 3.6 eV) and represents the dominant negative ion peak in discharges containing nitrogen and T1;- in the extracted ion be am.

carbon. Thus, when standard nitrogen gas is used as the carrier gas, the ratios of the mass peaks occurring at mass numbers 1 and 2 and 2 6'and 27, will be a sensitive indication of the relative isotopic abundance of hydrogen and carbon. Nitrogen by itself does not produce a stable negative ion and therefore will not appear as If an inert carrier gas is used, say argon, with nitro gen-containing carbon compounds, the CN peaks will reflect the relative abundance of both C and N. Oxygen isotopic analysis may be performed on OH ions (masses T7 to 19) which will be predominant in the presence ofexces hydrogen, or oiTOQ' (masses 82th R) which may, however, be contaminated with traces of H02 ions. (The latter may interfere with the determination of but not of O abundance.)

The isotopic compositions of some of the sulfur isotopes can be determined using the SO- peak (mass 48 to 4) formed in the presence of excess oxygen as the carrier gas. Owing to the possible formation of the SOH ion, only S, S, and 5 might be analyzed in this manner.

Considering the present cost of enriched stable isotopes, only D and C are available at reasonable prices, and thus most analyses will be performed with these two elements. It is conceivable, however, that as soon as a demand develops for N, '"O, or 5, these isotopes could be separated at a cost comparable with that of "C.

The technique may be practiced either by tagging the unknown compound or metabolite with the uncommon isotope or by tagging the added analog with the uncommon isotope.

There remains the question of the safety of'administration of small quantities moles) of nonradioactive labeled pharmaceuticals to human subjects. The difference in physiological or pharmacological action of the labeled compound and its natural analog (which contains the same isotopes at a lower concentration) is expected to be minimal. In no case should the D- substituted compounds carry the label on easily dissociated hydrogens (where -a considerable difference in physiological or pharmacological activity is expected). These hydrogens readily undergo isotopic exchange with water or other labile hydrogens and thus they are unsuitable as biological tracers.

Isotopic labeling of the administered pharmaceutical followed by separation and isotope ratio analysis with the addition of the normal, common isotopic form allows quantitative determinations of drug levels down to 10 g quantities. Even at assay levels of 10 g, the technique is four orders of magnitude better than conventional chemical analysis. Such a sensitivity allows the administration of quantities as low as 10 g of a drug to a human subject and still follows its fate with a fair degree ofconfidence. The method of-the invention enables isotope dilution analysis to'be accurately performed with stable, non-radioactive isotope.

The sensitivity attainable by the method of this invention is in fact higher than that obtainable by using radioactive tracers. For comparison, using tritium or "C as tracers, one can determine with a similar precision within the same time period only 3 X 10' moles ofT and 2 X 10' moles of C. Taking singly labeled molecules of mol wt 200, these quantities correspond to 6 X 10 g and 4 X 10' g, respectively. In other words, the present method using stable isotopic tracers for hydrogen has as high a sensitivity as that obtainable by the radiotracer methodology and by far a higher sensitivity than by using radiocarbon as tracers. lf one remembers that it is practically impossible to label a compound with a carrier-free tracer, whereas stable isotopes can be readily handled up to percent purity, it may be concluded that the method of the invention is in fact more sensitive. Furthermore, the present method allows the use of 0 and N as tracers with a comparable sensitivity. There is no practical radiotracer for these elements.

Another method of acquiring information on the fate of new drugs at subpharmacological doses is by the administration of the tested compound as such (with its natural isotopic composition) and using a stable,-labeled or multilabeled compound as the known amount of isotopic analog. For example, one may administer 100 mg of C H Q, (aspirin) into a human subject and recover 0.01 mg in a blood sample. Now one may add to it one mg of C H D O 98 percent D of the formula:

c1) 'coo CGHS COOH This mixture is then separated by a gas chromatograph.

ln view of the relatively high natural abundance of C, one could use C depleted carbon as a tracer in biological systems using compounds labeled with enriched C as the known analog for isotopic ratio analysis. Alternatively, the isotope-depleted compounds may be used as the known analog. For example, one may use barbital C H N O with its ethyl carbons depleted to 10 percent of its natural abundance as a tracer. This compound will contain (4 X 1.108 X 10 4 X 1.108 X l0"'')/ (8 X 98.892) 0.6170 percent C compared with 1.124 percent of its natural isomer. Taking such labeled barbital as tracer and using a 1:50 dilution factor, the addition of natural barbital as known analog would result in a change from 1.124% C to 1.1 14 percent, which can be dtermined with :1 percent precision with the duoplasmatron system. Taking natural barbital -as tracer and the depleted barbital as the known analog, the C abundance would change from 0.617 percent to 0.627 percent, which can again be determined with :1 percent precision. As the amounts necessary for analysis in the highly sensitive system of theinvention are as low as 10' g, the amount of drugs to be determined with i1 percent precision is as low as 10*g. If lower precision is required or if special care is taken in the gas chromatography, quantities down to 10 g of drugs can be determined with :10 percent precision.

No health hazard whatsoever is involved in the tagging of organic compounds intended for general human consumption by double or triple-labeled molecules at concentrations of 1:10 Let us take, for example, the case of a compound tagged by the addition of a 10 mole fraction'of molecules triple-labeled with deuterium. Let us assume that the given compound contains 12 hydrogen atoms (e.g., barbital). The total amount of hydrogen in 1 g of barbital is 12/1842 X 1,000 65mg containing lOng D. Now 1mg of the triple-labeled bartibal. which is sufficient to make lg barbital identifiable, contains 6,000/1 86.2 36mg D. In other words, by tagging the barbital, its D content has been increased by about one part in 300. The latter is considerably smaller than the inaccuracy inherent in the dosage of pharmaceutical products. In fact, no commercial drug is free of impurities at the parts-p er-m ilhon level and these impurities must have by far more pronounced side effects than any isotopically-substituted molecule of the drug itself.

The possible adverse biological effects of C, N, or O are smaller by two orders of magnitude than those of D, as can be concluded from their measured kinetic isotope effects. The former isotopes may thus be used as tracers even at carrier-free levels, and there should evidently be no doubt regarding their safety at levels of dilution.

Another application of the isotope dilution methodology is in the identification of unknown drug or poison in a clinical sample (for purposes of treatment of a poisoned subject or in forensic cases). To aliquots of the biological sample, small amounts of different suspected drugs, narcotics, or organic poisons enriched with a certain stable isotope are added. Each of these spiked samples is then chemically treated to separate a small quantity of the added constituent. This will now be subject to isotopic analysis and its isotopic compositon compared with that of the original labeled drug. Only the sample which contained the particular drug will affect the isotopic composition of the added stable isotopic compound, while those cases in which no change in isotopic composition took place on addition to the clinical sample may be excluded from the list of suspected poisons. A quantitative assessment of the concentration of the suspected poison can be obtained simultaneously in this screening test according to this invention.

The principles outlined above for testing new drugs on human subjects can be applied to testing of the persistence and fate of physiological constituents, e.g., hormones, nutrients, and coenzymes. In each case, one can use either isotope-enriched or isotope-depleted compounds, either as tracers or as carriers. It is evident that the number of applications of this methodology for clinical diagnosis and for a better understanding of physiological processes is practically unlimited.

The existence of a rapid and precise method of the invention for the determination of isotopic ratios in or ganic molecules will find application in the control of the distribution of pharmaceuticals. If all the production of a certain manufacturer is uniquely tagged, this product may be identified and distinguished from any other drug of the same composition. Different manufacturers may thus be individually identified by using different unique labeling of the same molecules.

Identification of a certain batch of drugs may be important if the expiration of drug effectiveness is critical. Each batch can be uniquely tagged by using different combinations of isotopic substitution on the molecule, such as 3D, 3D C, 3D N, C N+D, and others. D, C, and N at five different relative concentrations in the molecule may give over 100 different distinguishable combinations. This may allow changing the labeling every month for eight years of production without repetition of the code.

anol can be determined injecting C I-I OH or C D OH into rats followed by isotopic dilution analysis of blood samples with the isotopic analog at dilution ra tics of 1:10 to 1:10 The metabolization of C-labeled carboxyl-phenylacetic acid by rats may be determined by monitoring the exhaled carbon dioxide for C.

The degree of dilution of the sample with isotopic analog will vary from case ,to case and will depend on the size of the sample, the natural abundance of the noncommon isotope and the sensitivity of the detection system for the negative ions of interest. Isotopic dilution analysis in most cases can be practiced with dilution ratios from about 1:25 to about 1:10.

The sensitivity and precision of the system may be demonstrated by the following example. Let us assume that a 10 g sample of molecules of molecular weight 200 is used and that each molecules produces one CN" ion. This estimate is conservative since each molecule is likely to contain many carbon atoms. The total number of ions produced by the sample is (10 /200) X 6 X 10 X 10' 3 X 10 Of this number, about 1 percent or 3 X 10 will be due to CN and about 0.4 percent will be due to C N The noise attributed to the "CN signal due to isotope fractionation will be 3 X 10 particles and represents its dominant source of uncertainty.

Another source of uncertainty can arise from organic impurities. These must be separated from the components of interest by gas chromatography. The minimum sample size is therefore limited by the resolution of the gas chromatograph consequently, a sample of lO' g would be an average sample in the system of the invention, though samples of lO g to l0'g can be processed depending on resolution and capacity of the source and mass spectrometer. Following the same calculation, a sample of l0' g could be determined with an ultimate precision of & X 10' The effectiveness of isotopicdilution analysis can be demonstrated by the following analysis. The instrument is utilized to measure the ratios of masses l-F/D R in a mixture of an isotopically labeled isotope added to a urine sample. The diluted mixture is introduced into the gas chromatograph-mass spectrometer combination. Since the natural abundance of D is only 1.48 X 10 of H,

where .1: represents the amount of drug to be measured in the sample, y the amount of isotopically substituted drug added, and a is the percentage of H in the labeled compound. It is assumed, for simplicity, that the amount of natural D in the sample is negligible compared to the added D isotope. Solving for x, one obtains A In the case of masses 2 6 and T7, the analysis is somewhat more complicated owing to the presence of N (0.365 percent) and the higher concentration of C (1.108 percent). Assuming no change in the isotopic composition of nitrogen:

2 /27: i2 r4 13 14 IZCISN) R (98.892 x ay)/[l.l08 x (100 a)y] The accuracy of the measurement of R will depend on the accuracy in determining the values ofa and y as well as the uncertainty in the C isotopic abundance in x. Since a and R can be determined to a higher precision than the dilution factor x/y, which is approximately :1 percent, this will be the precision with which x can be determined. This estimate neglects possible background errors due to limited resolution of the gas chromatograph. The limitation in sensitivity of this technique will probably be due to the resolution of the gas chromatograph. A sensitivity of lO g with an accuracy of 10 percent is attainable with a capillary gas chromatograph.

The analysis system of the invention provides an indicative, representative sample of the isotopic species in a form containing only one to three atoms without the necessity of chemical pretreatment. Due to increased sensitivity, the amount of sample necessary for analysis will be reduced by several orders of magnitude. The technique is fast and efficient and the apparatus is more simplified.

It is to be understood that only preferred embodiments of the invention have been described and that numerous substitutions, alterations and modifications are all permissible without departing from the spirit and scope of the invention as defined in the following claims.

What is claimed is: l. A method of determining the quantity of a substance in a sample comprising the steps of:

adding a measured amount of an isotopic analog of the substance to the sample to form a mixture;

ionizing and fragmenting the mixture in a hot plasma to form ionized fragments of the substance and isotopic analog;

extracting the negative ion fragments as a beam from the hot plasma;

magnetically separating the beam into a first and second beam corresponding to negatively ionized fragments of the substance and negatively ionized fragments of the isotopic analog;

measuring the intensity of said first and second beams; and

rationing the measured intensities of said first and second beams, whereby the quantity of said substance in said sample may be determined.

2. A method according to claim 1 further including the step of vaporizing said mixture.

3. A method according to claim 1 further including the step of purifying said sample.

4. A method according to claim 3 in which the mixture is purified by passage through a gas chromatographic column.

8. A method according to claim 7 in which the carrier gas is selected from a group consisting of nitrogen, oxygen and hydrogen.

9. A method according to claim 1 in which the isotope is non-radioactive.

10. A method according to claim 1 in which the sample contains a first isotopic concentration and the analog contains a substantially different isotopic concentration.

11. A method according to claim 1 in which the isotopes are selected from a group consisting of D, C, N, O, and S.

12. A method according to claim 11 in which the substance contains a first concentration of at least one of said isotopes and the analog contains a lower concentration of said isotopes.

13. A method according to claim 11 in which the added analog contains a higher concentration of said isotopes.

14. A method according to claim 1 in which the substance and analog are organic compounds containing at least one of the atoms selected from the group consisting of C, H, O, N and S and isotopes thereof.

15. A method according to claim 14 in which the sample is a biological fluid and said isotopes are substituted on non-labile positions of said compound that are not dissociated by the biological system of the animal subject.

16. A method according to claim 1 in which the dilution ratio of said substance to said analog is from 1:25

17. A method according to claim 16 in which the sample contains from 10 grams to 10". grams of said substance.

18. A method according to claim 15 in which said sample is a pharmaceutical and further including determining the persistence level of the pharmaceutical by administering a material selected from the group consisting of said pharmaceutical and an isotopic analog thereof to a subject, recovering a sample of body'fluid of the subject containing said material, adding a known amount of said isotopic analog to said sample to form said mixture.

19. A method according to claim 1 in which said mixture is ionized and fragmented in a negative ion duoplasmatron. v

20. A method of determining the identity of an unknown first substance comprising the steps of:

adding known amounts of isotopic forms of known different second substances to different aliquots of said first substance to form different mixtures, one of said different second substances being suspected as being the same as said unknown first substance; successively, for each different mixture, ionizing and fragmenting the mixture in a hot plasma to form a beam of negatively ionized fragments; magnetically separating the beam into a first beam path corresponding to the negatively ionized fragment of the said isotope and a second beam corresponding to the negatively ionized fragments of the first substance;

measuring the intensity of the first and second beams,

and

observing the presence of a characteristic ratio of said first and second beams as an indication of the identity of said first substance.

21. A method of identifying a batch of an organic chemical comprising the steps of:

labeling the batch by adding a known amount of noncommon isotopes to the batch at non-labile, nondissociative positions of the chemical;

recording the labeling information;

forming a mixture of a known amount of a sample of the labeled compound and a known amount of a 14 common isotopic analog thereof; introducing the mixture into a hot plasma to form a beam of negatively ionized fragments; magnetically separating the beam into a first beam path of the negatively ionized fragments containing the non-common isotope and a second beam of the negatively ionized fragments containing the common isotope;

measuring the intensities of the first and second beams;

determining the isotopic content of the chemical; and

comparing it to the recorded labelinginformation.

Claims

1. A method of determining the quantity of a substance in a sample comprising the steps of: adding a measured amount of an isotopic analog of the substance to the sample to form a mixture; ionizing and fragmenting the mixture in a hot plasma to form ionized fragments of the substance and isotopic analog; extracting the negative ion fragments as a beam from the hot plasma; magnetically separating the beam into a first and second beam corresponding to negatively ionized fragments of the substance and negatively ionized fragments of the isotopic analog; measuring the intensity of said first and second beams; and ratioing the measured intensities of said first and second beams, whereby the quantity of said substance in said sample may be determined.

5. A method according to claim 1 further including the step of dispersing said mixture in a carrier gas.

6. A method according to claim 5 in which the carrier gas is unreactive with the sample.

7. A method according to claim 5 in which the carrier gas reacts with the sample to form ionized fragments containing said isotope.

9. A method according to claim 1 in which the isotope is non-radioactive.

11. A method according to claim 1 in which the isotopes are selected from a group consisting of D, 13C, 15N, 18 O, and 36S.

16. A method according to claim 1 in which the dilution ratio of said substance to said analog is from 1:25 to 1:107.

17. A method according to claim 16 in which the sample contains from 10 1 grams to 10 10 grams of said substance.

18. A method according to claim 15 in which said sample is a pharmaceutical and further including determining the persistence level of the pharmaceutical by administering a material selected from the group consisting of said pharmaceutical and an isotopic analog thereof to a subject, recovering a sample of body fluid of the subject containing said material, adding a known amount of said isotopic analog to saId sample to form said mixture.

19. A method according to claim 1 in which said mixture is ionized and fragmented in a negative ion duoplasmatron.

20. A method of determining the identity of an unknown first substance comprising the steps of: adding known amounts of isotopic forms of known different second substances to different aliquots of said first substance to form different mixtures, one of said different second substances being suspected as being the same as said unknown first substance; successively, for each different mixture, ionizing and fragmenting the mixture in a hot plasma to form a beam of negatively ionized fragments; magnetically separating the beam into a first beam path corresponding to the negatively ionized fragment of the said isotope and a second beam corresponding to the negatively ionized fragments of the first substance; measuring the intensity of the first and second beams, and observing the presence of a characteristic ratio of said first and second beams as an indication of the identity of said first substance.

21. A method of identifying a batch of an organic chemical comprising the steps of: labeling the batch by adding a known amount of non-common isotopes to the batch at non-labile, non-dissociative positions of the chemical; recording the labeling information; forming a mixture of a known amount of a sample of the labeled compound and a known amount of a common isotopic analog thereof; introducing the mixture into a hot plasma to form a beam of negatively ionized fragments; magnetically separating the beam into a first beam path of the negatively ionized fragments containing the non-common isotope and a second beam of the negatively ionized fragments containing the common isotope; measuring the intensities of the first and second beams; determining the isotopic content of the chemical; and comparing it to the recorded labeling information.