Modern FT-IR systems provide rapid analysis of samples with very little effort, putting the technique on par with the dilute and shoot simplicity of the GC/MS.

According to the Scientific Working Group for the Analysis of Seized Drugs (SWGDRUG), infrared spectroscopy is a Category A Technique for drug identification. However, it is underutilized in many forensic laboratories owing largely to the ambiguous results it can produce when analyzing seized drug samples that are mixtures. These results can be difficult for forensic scientists to confidently support when giving expert testimony. Most labs rely heavily on gas chromatography coupled to mass spectrometry (GC/MS) and are confident in the results it provides, but GC/MS may not be adequate in all cases to support a drug’s identification.

This article will present the capabilities of infrared spectroscopy for use by forensic scientists involved in seized drug analysis. It will cover the role Fourier Transform Infrared (FT-IR) spectroscopy plays in forensic drug analysis as a complementary technique to GC/MS. Of key importance to the capabilities of modern FT-IR spectrometers is the use of simple ATR sampling and mixture searching algorithms to rapidly screen and identify seized drugs.

A basic need of any law enforcement agency is to positively identify controlled substances in seized drug samples to support legal proceedings and get convictions. The mission of the Scientific Working Group for the Analysis of Seized Drugs (SWGDRUG) is to recommend minimum standards for forensic examination of seized drugs and seek their international acceptance. The recommendations are intended to assist forensic analysts and managers in the development of analytical techniques, protocols, and policies in an effort to ensure consistent professional practice among the forensic community. Recent changes to the wording of the 5th Edition of the SWGDRUG recommendations published in January 2010 will impact forensic scientists with respect to education, quality control practices, and the use of data from analytical techniques for drug identification.

Currently, there are many common material identification techniques in routine use by crime labs, which are categorized by the updated SWGDRUG recommendations according to their discriminating power (Table 1). Category A techniques have the highest discriminating power and Category C techniques have the lowest. Screening techniques fall under Category C and commonly include quick colorimetric tests, melting point analysis, and solubility in specific solvents. Category B includes intermediate techniques of analysis such as microcrystalline tests or extraction and/or derivitization for gas chromatographic analysis. Finally, Category A techniques are confirmatory techniques like MS, IR, and Raman spectroscopic analysis.

Table 1: Categories of Analytical Techniques.
Click for larger image.

According to the SWGDRUG recommendations, techniques with the highest discriminating power should be used preferentially for drug analysis. If a Category A technique is used, at least one other technique from any of the three categories shall be used as well. Many drug labs choose GC/MS since it efficiently addresses the recommendation, enabling them to get credit for two techniques, namely MS (Category A) and GC (Category B). However, many labs are finding that GC/MS is not always the best choice for the particular sample they are analyzing. Infrared surpasses GC/MS in that it can analyze drugs in salt form and uniquely identify certain stereo isomers. It is also an excellent quick-screening tool.

It is also specified in the recommendations that the classification of a technique may be lower if the sample, analyte, or mode of operation diminishes its discriminating power. Specific examples of diminished discriminating power provided in the recommendations refer to infrared and GC/MS techniques. More specifically, reference is made to an infrared spectroscopy technique applied to a mixture that produces a combined spectrum, as well as to a mass spectrometry technique which only produces molecular weight information.

In the case of hyphenated techniques like GC/MS, the combined methods can only be treated as separate if results from both are used. This means that chromatographic peak data must be provided in addition to the molecular weight data from the mass spectrum. For infrared, it is specifically stated that a mixture sample that produces a spectrum combining more than one chemical entity may not provide definitive drug identification.

A typical analysis workflow for unknown drug identification using GC/MS is shown in Figure 1. In this scenario, an unknown drug tablet is crushed and dissolved in an appropriate solvent such as dichloromethane. The sample vial is inserted into an autosampler tray for injection into the GC/MS system. For certain drugs, an acid or base extraction may be required to convert the drug from the salt form into a form that will volatilize and elute on the GC column.

After the chromatographic and mass spectral data collection cycles are finished, unknown peaks in the chromatogram (Figure 2, top) can be used to display the associated mass spectra (Figure 2, bottom). A library search is used to display the best hits based on the mass spectrum of each peak (Figure 2, right). This process is repeated for the rest of the unknown peaks to identify the remaining compounds in the sample. In this example, a cocktail of amphetamines was identified in the mixture. According to the SWGDRUG recommendations, chromatographic data must be supported by the injection of drug standards on the column to verify retention times.

Figure 1: GC/MS sampling for solid dose drugs.
Click for larger image.

Figure 2: An example of a GC/MS search on an unknown peak.

For infrared analysis, the choice of sampling technique is largely determined by the information required about the sample. For identification work, a sample analysis approach must be used that ensures that spectral peak intensities remain within a linear range. To meet this need, the traditional transmission sampling technique requires time-consuming and cumbersome sample preparation. A solid powder sample must be ground into a powder, mixed with potassium bromide (KBr) salt to dilute the sample, and then made into pellets before it can be placed in the infrared beam for analysis.

In modern FT-IR spectrometers, a single-bounce diamond ATR crystal is often used instead of transmission sampling for quick analysis and unknown identification. The nature of the ATR technique ensures that spectral intensities remain within a linear range and undiluted samples can be directly analyzed (Figure 1). The entire sample analysis is typically complete in less than a minute making the FT-IR a very rapid screening technique. Additionally, the diamond ATR crystal cannot be scratched, resulting in no performance degradation.

It is important to note that since less than a milligram of sample is needed for analysis on the ATR, the sample placed on the crystal must be representative of the entire sample being analyzed. Once the spectrum of the sample is obtained, the most common approach is to perform a simple library search comparing the unknown powdered sample against a drug spectral library. This simple search assumes that the unknown spectrum represents a single material or that the library has spectra that are representative of the components found in the sample.

The Multi-Component Search Algorithm
Spectra representing a combination of more than one chemical entity influenced updates to the 5th Edition SWGDRUG recommendations related to the “diminished discriminating power” of the infrared technique. Figure 3 shows such an example whereby the sample and mode of operation produce a result that is not visually definitive. In an attempt to resolve the ambiguity resulting from this spectrum and identify the extra component, spectral subtraction is commonly used. Figure 4 shows the subtraction result for the spectrum of a methamphetamine reference spectrum from the unknown resulting in a spectrum that was identified as dimethyl sulfone, a common cutting agent for methamphetamine. Once the cutting agent is identified, it can then be subtracted from the original unknown spectrum. Although this mathematical “clean up” of the spectrum leads to a much improved search metric for the methamphetamine, it is a subjective approach that takes a trained eye to determine the best endpoint for the subtraction factor. It also leads to spectral distortions, resulting in lower confidence for the user. In addition, spectral subtraction can be viewed as “data manipulation”, making results more difficult to defend in court.

By using a software platform with a multi-component search algorithm, the extra processing time and data manipulation of the spectral subtraction step is eliminated and the mixture is identified in a single action. This leads to faster analysis, with less subjective results and higher confidence for the analyst who can provide a much more compelling argument in court, especially when results are combined with data from a complementary technique like GC/MS. Figure 5 illustrates an example of using a multi-component search algorithm to identify methamphetamine in a sample.

In general, the optimized process using FT-IR with a single-bounce diamond ATR and a multi-component search algorithm allows the forensic analyst to rapidly screen seized drug samples and aid in their identification. The integrated method is routinely used in cases prosecuted in Federal Courts as it can identify the salt forms of the drugs and in several situations even the enantiomers and the degree of hydration. In certain cases, identifying the salt form is important in sentencing. Drug spectra can be significantly different than the spectra of the salt form of the drug. These differences are readily detectable using FT-IR while identification is not possible using GC/MS. FT-IR also allows the determination of whether the drug is in the anhydrous form or a hemihydrate, impacting the quantitative calculation. In addition, FT-IR analysis can identify very similar species that co-elute and have the same mass spectrum.

Figure 3: A powdered sample being run on the ATR and searched against a drug spectral library.

Figure 4: The substraction result for the spectrum of a methamphetamine reference spectrum from the unknown resulting in a spectrum that was identified as dimethyl sulfone.

Figure 5: OMNIC Spectra software and its patented Multi-Component Search routine results in the data manipulation from spectral subtraction being eliminated and mixture identification in a single step.

Forensic Scientists’ View
From the viewpoint of many forensic scientists, there are many benefits as well as shortcomings associated with the use of FTIR in a forensic drug lab. Most importantly, the technique is non-destructive and it produces unique spectra for many illicit drug samples. It can also be used to directly analyze and identify free base and salt or hydrate forms of drugs in addition to being complementary to GC/MS, well accepted in courts, and allowing digital data to be stored electronically for further analysis. Traditionally, KBr sample preparation is tedious, spectral searching is good for pure compounds but poor for mixtures, and it is often faster to dilute and shoot using an autosampler onto a GC column on a GC/MS. Modern FT-IR technology overcomes many of these limitations by using quick and easy diamond ATR sampling devices to reduce sample preparation, offering instrument validation and traceability features, and enabling multi-component search to analyze mixture samples reproducibly.

Discussions held with forensic scientists with significant expertise in the use of FT-IR and GC/MS for seized drug analysis have yielded several important observations. Of particular note were comments about what is required to confidently support and present the results of infrared analysis in court. Having a comfort level with the methodology being used is paramount and can be lacking among criminalists due to the complexity and ambiguity the result may present. In addition, the use of certain terms like “data manipulation” should be avoided in favor of “data processing” when discussing the analysis process. The data audit trail and the quality control procedures are also very important. Finally, using software with advanced search algorithms has been found to deliver more consistent and less subjective data processing, removing the data manipulation component, making infrared results easier to defend in court and encouraging forensic analysts to use FT-IR more frequently.

The 5th Edition SWGDRUG recommendations increase the expectations for seized drug analysis in many significant ways, not the least of which is the potential reduction in discriminating power for the use of FT-IR. The commonly used techniques of GC/MS and FT-IR provide complementary information and can support each other very efficiently. The use of modern FTIR systems with diamond ATR sampling interfaces provides rapid analysis of samples with very little effort, putting the technique on par with the dilute and shoot simplicity of the GC/MS. Beyond the sampling for FT-IR, it all comes down to the ability of the system’s software to discriminate mixture spectra. This capability is offered by using a multi-component mixture searching algorithm, thereby improving the information that the FT-IR generates and displays. This is of key importance to improved court testimony associated with infrared results.

Sanford Angelos has 30 years experience as a DEA Forensic Scientist and currently works for Aris Associates in Chicago. An expert on clandestine methamphetamine manufacture and an international forensic analysis training expert, Angelos is a former chairman of the Criminalistics Section of AAFS.

Mike Garry has 20 plus years of experience with FT-IR applications and is currently the Product Manager for FT-IR Products at Thermo Scientific in Madison, WI. He works on sampling accessories, including ATR; software products, including OMNIC Specta; and spectral databases.