The screening and analysis of drugs of abuse utilizing a direct probe introduction system on a mass spectrometer has been used by forensic laboratories for more than 20 years. The technique was popular at the New Jersey State Police Crime Lab in the 70's as a general screening tool for street drug samples as well as toxicology extractions of blood and urine. This lab used the direct probe introduction method on a magnetic scanning mass spectrometer utilizing isobutane chemical ionization (CI) techniques (1 - 5). The direct probe permitted the analysis of more than 50 samples per day for subsequent verification of the analysis by other analytical techniques.
Figure # 1 - Components of the HP 5973 Probe Inlet System
The introduction of the new Hewlett Packard 5973 MSD in conjunction with the new S.I.S. direct insertion probe expands the capability of the mass spec direct probe technique in the forensic and pharmaceutical laboratory. This new mass spectrometer is several decades more sensitive than the older magnetic scanning instruments and the new software capabilities permit the averaging of mass spectral data as well as selecting the optimum region of the total ion chromatogram to minimize interference by other analytes. In addition, software techniques, as background subtraction, permit the cleaning up of the mass spectrum for identification of unknown drugs via library search routines using published mass spectral libraries. This cleaning up of the data enables the analysis of samples in the electron impact (EI) mode. This will permit a more positive identification than the CI method, eliminating the need for further analytical methodologies for the identification of the drug samples.
Figure # 2 - HP Probe Inlet System Attached To the HP 5973 MSD
The purpose of this paper is to demonstrate the use of the direct probe introduction technique on the Hewlett Packard 5973 MSD. Pure cocaine was selected as the sample to be analyzed. This paper will compare the ballistic versus the temperature programmed modes of operation of the probe and will determine the sensitivity limitations of the direct probe technique for the identification of cocaine and other drugs of abuse.
The SIS Direct Probe Inlet and Probe was attached to the GC/MS interface port on the side of the H.P. 5973 MSD (Figure1 and 2). The probe inlet system (Figure 1) consists of a ¼ turn ball isolation valve to permit the introduction of the probe into the mass spec, a ¼ turn rough pumping valve attached to an Edwards 1.5 rough pump, a dual PTFE sealing system to provide long life vacuum seals to the probe and an indexed dual guide rod introduction system. This guide rod system is indexed at three positions to permit the easy introduction of the probe into the vacuum system with automatic stops at the first seal position, the second seal position and 1 cm distance from the source (cooling position). These automatic stops enable the accurate positioning of the probe, eliminates scoring of the probe rod due to closing the ball valve on the probe rod and eliminates the accidental venting of the mass spec by inadvertently pulling the probe out too far during probe removal. The mass spec probe can be heated ballistically up to 450 degrees C at a ramp rate in excess of 500 degrees per minute. The probe can also be temperature programmed for slower heating applications. The probe uses flared glass sample vials 1.7mm diameter by 12.75 mm long into which the sample is inserted. The sample vial is then inserted into the tip of the probe where it is held in place by a small spring.
A sample of pure cocaine hydrochloride in methanol solution (Sigma Chemical Company, St. Louis, MO) was used for the analysis. This solution of cocaine at a concentration of 1.0 mg/ml was diluted to obtain a solution of cocaine at a concentration of 10 ng/ul (10 mg/liter) in methanol. This solution was injected into the direct probe sample vials using a GC syringe to prepare sample vials with the following amounts of cocaine:
Sample No. Sample Volume Amount of Cocaine
1 1.0 ul 10.0 ng
2 2.0 ul 20.0 ng
3 3.0 ul 30.0 ngAfter sample injection, the methanol solvent in each of the sample vials was allowed to evaporate to dryness before analyzing any of the samples. This was accomplished in about 15 minutes by heating the sample vials in a sample vial block to 60 degrees C.
The HP 5973 MSD was operated in the EI mode and was scanned from mass 50 to 350 daltons at 1 scan per second. The mass spec source temperature was set to 250 and the quads to 150 degrees C. Each sample was then inserted into the mass spec probe and inserted through the vacuum lock system of the probe inlet into the mass spec source of the HP 5973 MSD. Each sample was then heated ballistically to 300 degrees (ramp rate > 500 degrees per minute), and the mass spec was scanned to analyze the thermally extracted sample. The total analysis time was 1.5 minutes per sample. As an alternative technique, the direct probe was heated from 25 degrees C to 300 degrees C at a temperature ramp rate of 100 degrees C per minute. The total analysis time for this temperature programmed analysis was 3.0 minutes.
After the sample analysis was complete, the total ion chromatogram was viewed using the ChemStation data analysis module and the sample was then further analyzed using the selected ion extraction function of the ChemStation software for the 303 ion of cocaine. The mass spectrum was then taken over a 0.10 minute time range at the highest intensity range of the resulting peak and the background from the front of this broad peak was subtracted to clean up the mass spectrum. The resulting mass spectrum was recorded.
Figure # 3 - Direct Probe Analysis of 30 ng Cocaine
The total ion chromatogram for a 30 ng sample of cocaine is shown in Figure 3 - top. This chromatogram consists of several peaks or humps which are probably due to one or more impurity components in the drug sample. The total ion chromatogram was analyzed using the selected ion extraction capabilities of the HP Chem Station software to extract only the mass 303 molecular ion of cocaine (Figure 3 - bottom). The resulting extracted ion chromatogram shows the area of the total chromatogram which would contain the highest concentration of cocaine. An average of the mass spectra over the time range 0.35 to 0.45 minutes was taken and the background at the front of the peak (0.1 minutes) was subtracted from the average mass spectrum to obtain the final mass spectrum.
Figure # 4 - Direct Probe Analysis of 30 ng Cocaine Comparison of Heating Methods
Ballistic heating of the probe as shown in Figure 3 heats the probe tip and sample up rapidly in excess of 500 degrees per minute. Ballistic heating provides for the fastest sample analysis and the bet sensitivity of analysis. However, it is often desirable to heat the probe up more slowly. This slower temperature programming technique can remove unwanted solvents or permit the thermal separation of mixtures of analytes in a sample. The slower heating of the probe will increase the time of analysis and cause the analytes to be eluted later in the total ion chromatograph, but it will also result in a decrease in sensitivity of the direct probe technique. This is demonstrated in Figure 4 in which 30 nanograms of cocaine was analyzed via the two methods. The ballistic heating method analyzed the cocaine sample in less than 1 minute with a peak half width of about 0.25 minutes (Figure 4 - bottom). The programmed heating of the probe at 100 degrees per minute resulted in a three minute analysis with a peak half width of about 0.5 minutes for the cocaine. Since the total sample size is proportionately related to the area of the peak, it can easily be seen why this slower heating cycle results in the loss of sensitivity. In most instances, it is preferable to heat the sample as quickly as possible. However, there are samples (especially mixtures of drugs) which are best analyzed using a slower temperature program method.
Figure # 5 - Direct Probe Analysis of Cocaine
The results of the selected ion chromatograms for each of the 3 sample sizes of the cocaine samples analyzed are shown in Figure 5. These chromatograms were produced by the selected ion extraction procedure for the cocaine 303 ion as described above. Sample sizes down to 10.0 ng of cocaine were easily detected and identified using the direct probe technique. However, at the lower concentrations, the signal became quite weak and the noise level or background increased. The mass spectrum was taken by analyzing the average of a 0.10 minute time at the top of each peak in Figure 5, and the background was subtracted from the front of each peak utilizing the Chem Station software. The results are shown in Figure 6. Cocaine has a molecular weight of 303 and produces a molecular ion at mass 303 and a major fragment ion which is the base peak for cocaine at 182 daltons. The peaks at 198, 105, and 82 are also from cocaine. The detection limit for this ballistic heating probe method appears to be about 10 ng. However, at this concentration, the appearance of many other background peaks are apparent in the mass spectrum which could interfere with the positive identification of the cocaine sample if it were a true unknown. At the higher concentrations of 20 and 30 ng, the specra is much cleaner and the cocaine readily identified. With the temperature programmed method shown in Figure 4 - top the limits of sensitivity are at least 30 ng.
Figure # 6 - Mass Spectra of Cocaine Via Direct Probe
The sensitivity of the technique could be further increased by using the mass spec in the selected ion mode (SIM) to scan only the 182 and 303 cocaine ions. Although this would produce a more sensitive method of analysis, the analysis would be a less definitive method of identification. It is always preferable to identify 4 or 5 ions of the correct relative intensity to make a positive identification of an unknown drug.
The heated direct probe and sample introduction system has proved to be a useful addition to the new Hewlett Packard 5973 MSD for the analysis of drugs of abuse. The probe inlet system permits the direct introduction of the probe through what is normally the GC/MS interface line on the MSD. The indexed probe introduction system permits the easy insertion of the probe without the chance of inadvertent venting of the mass spec or scoring of the mass spec probe shaft. Samples were analyzed in less than 1.5 minutes using the ballistic heating probe method. However, the probe can be temperature programmed at slower rates for the purpose of thermally separating mixtures of drugs in a single sample.
The direct probe analysis of cocaine has been shown to be a very sensitive technique for solid sample analysis with detection levels down to 10 ng. The broad peak width (0.25 minute peak half width) is quite large as compared to a GC analysis and explains the loss in sensitivity as compared to conventional GC analysis where peak widths are normally less than 5 seconds. However, the outstanding sensitivity of the new HP 5973 MSD enables the direct probe analysis of sample levels comparable to many GC samples analyzed via the older 5971 MSD instruments. The highly sensitive HP 5973 MSD and the versatile ChemStation software permit the analysis of pure and mixed drug samples via the EI mass spec probe technique to achieve a fast and definitive method of analysis. The direct probe introduction system interfaced to the HP 5973 will be a valuable technique for the analysis of many drug and pharmaceutical samples, where fast sample analysis is required.
(1) Identification of Drugs by chemical Ionization Mass Spectroscopy, Part II, Richard Saferstein, Jew Ming Chao and John Manura, J. of Forensic Science, Vol 19, No.3, 1974, pp 463-485.
(2) Identification of Heroin and its Diluents by Chemical Ionization Mass Spectroscopy, Jew Ming Chao, Richard Saferstein and John Manura, Analytical Chemistry, Vol 46, No. 2, Feb. 1974, pp 296 - 298.
(3) Drug Detection in Urine by Chemical Ionization Mass Spectroscopy, Richard Saferstein, John Manura. and P.K. De, J. of Forensic Science, Vol 23, No. 1, 1978, pp 29 - 36.
(4) Drug Detection in Urine by Chemical Ionization Mass Spectroscopy, Part II, Richard Saferstein, John Manura, Thomas Brettell and P.K.De, J. of Analytical Toxicology, Vol. 2, Nov-Dec 1978, pp 245 - 249.
(5) Chemical Ionization Mass Spectrometry of Morphine Derivatives, Richard Saferstein, John Manura and Thomas Brettell, J. of Forensic Science, Vol 24, 1979, pp 312 - 316.