SWGDRUG Identification Tiers and the UV / TLC / GC-MS / LC-MS/MS Workflow

SWGDRUG's Recommendations (Part I, Document 1, revised 2019) classify analytical methods for drug identification into three categories based on the degree of discriminating power they provide.

Category A contains methods with the highest discriminating power: those that provide structural information sufficient to distinguish between chemically similar compounds. The Category A methods are infrared spectroscopy (IR, including FTIR and ATR-FTIR), mass spectrometry (MS) when coupled with a separation technique (GC-MS, LC-MS, LC-MS/MS), nuclear magnetic resonance spectroscopy (NMR), Raman spectroscopy, and X-ray diffraction (XRD). All of these methods depend on validated calibration against certified reference materials traceable through ISO 17034. The defining feature of Category A methods is that the result is a molecular or structural fingerprint, not a global property. A mass spectrum with characteristic fragment ions at specific m/z values, or an IR spectrum with absorption bands at specific wavenumbers, identifies a compound in a way that is very difficult to replicate with a chemically distinct structure.

Category B contains methods with intermediate discriminating power: those that provide characteristic information but are less specific than Category A because the measured property (spectral or chromatographic) is shared by a broader range of compounds. Category B methods include UV-Vis spectroscopy (when a library match or characteristic absorption maximum is used), thin-layer chromatography (TLC, when run with a reference standard and a validated solvent system), gas chromatography alone (GC-FID, without mass spectrometry detection), high-performance liquid chromatography alone (HPLC-UV, HPLC-DAD, without MS detection), capillary electrophoresis, and immunoassay when used in the laboratory (not field) setting.

Category C contains methods with the lowest discriminating power. These are the colour tests covered in the preceding topic (Marquis, Mecke, Duquenois-Levine, Scott), microcrystal tests (where the morphology of a crystalline reaction product is observed under a microscope), and fluorescence or luminescence tests. Category C methods are rapid, cheap and useful for orienting the analysis, but they do not provide structural information.

The SWGDRUG identification rule is: a drug identification requires at least one Category A method, combined with at least one additional method from any category (A, B, or C). The two methods must be orthogonal: they must measure different physical or chemical properties, so that the probability of a false positive in both is the product of the individual false-positive probabilities rather than merely the higher of the two.

The orthogonality requirement excludes certain intuitive but incorrect combinations. Two mass spectrometric methods (for example, GC-MS and LC-MS/MS on the same extract) are orthogonal at the separation level (different stationary phase, different mobile phase) and at the ionisation level (electron ionisation vs electrospray), so they are acceptable. But two GC separations using the same stationary phase are not orthogonal at the separation step even if they use different detectors; the stationary phase chemistry is repeated. Similarly, FTIR and ATR-FTIR are not orthogonal because both measure the same IR absorption spectrum; they are the same technique in different sampling geometries.

Infrared spectroscopy detects the absorption of mid-infrared radiation (typically 4,000 to 400 wavenumbers, cm-1) by molecular bonds undergoing vibrational transitions. The pattern of absorption bands across the entire mid-IR range constitutes a molecular fingerprint that is unique to a compound's specific combination of bond types, geometry, and functional groups. Two structurally distinct molecules cannot have identical IR spectra. The practical limit is sample preparation: a mixture of two compounds gives a spectrum that is the sum of contributions from both, which a pure reference spectrum cannot straightforwardly match. Most forensic drug IR workflows require a minimum sample purity or use chemometric deconvolution tools to extract component spectra from a mixture.

The portable FTIR instruments that have expanded into field laboratory and airport screening settings (the Smiths Detection HazMatID Elite and the Agilent 4300 FTIR) provide Category A data in principle, but their library databases are limited compared to the NIST/EPA/NIH Mass Spectral Library, and their performance on highly diluted samples or complex mixtures is inferior to laboratory FTIR with a dedicated forensic drug library.

GC-MS combines chromatographic separation with mass spectrometric detection. The GC separates the compounds in an extract by their vapour pressure and interaction with the stationary phase. The Agilent 7890B GC, commonly coupled with the 5977B Mass Selective Detector (MSD), is the standard instrument configuration in most accredited drug chemistry laboratories in North America, the UK and Australia. The separated compounds exit the GC column through a heated transfer line and enter the ion source of the mass spectrometer, where they are ionised by electron ionisation (EI, 70 eV). The resulting fragment ions are separated by the quadrupole mass filter and detected to produce a mass spectrum with characteristic m/z values and relative abundances that together identify the compound.

The mass spectrum is compared against the NIST/EPA/NIH Mass Spectral Library (currently containing over 430,000 EI spectra as of the NIST 2026 release) or the DEA Special Testing and Research Laboratory library (approximately 60,000 drug-specific spectra). A library match score above 800 (out of 1000) combined with a retention time match within validated tolerance windows constitutes the core GC-MS identification. The retention time match provides the orthogonal dimension within the single run: the spectral match and the chromatographic retention are two independent pieces of information that together reduce the probability of misidentification dramatically.

LC-MS/MS (liquid chromatography tandem mass spectrometry) is the preferred instrument for thermally labile compounds, highly polar drugs, and the confirmatory quantification of drugs in complex matrices. The Waters Xevo TQ-S triple quadrupole mass spectrometer, coupled with a Shimadzu Nexera UHPLC system (or the equivalent Waters ACQUITY UPLC), represents the current operational standard in many EU national laboratories and specialist UK, Australian and US laboratories. In LC-MS/MS, the compound of interest is selected by the first quadrupole (Q1) as the precursor ion, fragmented in a collision cell (Q2) filled with argon or nitrogen gas, and the product ions are separated by the third quadrupole (Q3) and detected. The precursor-to-product ion transition (the Multiple Reaction Monitoring or MRM transition) is highly specific to the compound: Q1 selects for the molecular mass (or a characteristic adduct), and Q3 selects for a specific fragmentation product that is characteristic of the compound's substructure.

NMR (nuclear magnetic resonance spectroscopy), specifically 1H-NMR and 13C-NMR, provides unambiguous constitutional structure determination and is used for confirmatory identification of novel psychoactive substances (NPS) where mass spectral library matches are unavailable. The DEA's NPS Discovery programme and the EMCDDA's European Monitoring Centre drug early-warning system both use NMR as part of the characterisation workflow for new substances appearing on the illicit market. NMR is not routinely used in operational drug chemistry for established drug classes because GC-MS is faster, less expensive per sample and sufficient for scheduled drugs with well-characterised mass spectra.

UV-Vis spectroscopy measures the absorption of ultraviolet and visible radiation by electronic transitions in conjugated pi-systems and heteroatom lone pairs. The absorption maximum (lambda-max) and the shape of the absorption spectrum are characteristic of a compound's chromophore: heroin typically shows lambda-max at approximately 277 nm in methanol; cocaine is essentially UV-transparent at analytical wavelengths above 250 nm (which is why HPLC-UV at 210-220 nm is used for cocaine quantification); THC shows lambda-max at approximately 220 nm with a shoulder at 280 nm.

A UV-Vis spectrum, when compared against a validated library at the same concentration and solvent, provides a characteristic fingerprint that is less discriminating than GC-MS or IR (many structurally distinct molecules with similar chromophores give overlapping UV spectra) but is genuinely orthogonal to TLC, GC-FID and colour tests. The instrument is a Category B method; in combination with GC-MS as the Category A method, it satisfies the SWGDRUG identification requirement.

HPLC-DAD (high-performance liquid chromatography with diode-array detector) provides both chromatographic retention and full UV-Vis spectral data at each peak. Because it provides two data dimensions (retention and spectrum), it is classified as Category B rather than being split across categories: the spectral data alone would be Category B UV-Vis, but the chromatographic separation increases specificity substantially. HPLC-DAD is widely used in European national drug laboratories for operational identification and quantification of established drug classes including benzodiazepines (which are poorly ionised in standard GC-MS conditions due to thermolability) and cannabinoids.

TLC (thin-layer chromatography) separates compounds based on their partition between a mobile phase (a solvent or solvent mixture) and a stationary phase (typically silica gel on an aluminium or glass plate). A drug-containing extract is spotted alongside reference standards of the suspected drug class. After development and visualisation (UV lamp at 254 or 365 nm, then chemical development with reagents such as ninhydrin, iodine, or potassium permanganate), the Rf values (distance migrated divided by solvent front distance) are compared between the sample spots and reference standards.

TLC is Category B because the Rf value is a characteristic property that depends on the compound's polarity and the specific solvent system, but it is not as structurally discriminating as a mass spectrum or IR spectrum: structurally related compounds (morphine and codeine, for example) have Rf values that are close but not always clearly separated in every solvent system. The selection of the solvent system is therefore critical. The UNODC ST/NAR/13 recommended methods document specifies validated solvent systems for each major drug class with the expected Rf values and visualisation patterns for the drug and its common adulterants.

The combination of TLC Rf in two different solvent systems is sometimes described as providing two Category B results, but SWGDRUG's guidance is that TLC in two solvent systems counts as one Category B method, not two, because both measurements are made by the same physical principle (partition chromatography). To get orthogonality credit for a second measurement, the second method must measure a genuinely different physical property.

The routine analytical escalation in an accredited drug chemistry laboratory follows a consistent pattern across jurisdictions, whether the laboratory is the DEA's North Central Laboratory in Chicago, the German BKA (Bundeskriminalamt) laboratory in Wiesbaden, or the CFSL in New Delhi. The details differ by instrument make and model, by specific solvent system and reference library, but the logic is the same.

1. Exhibit intake and preliminary examination
   The exhibit arrives with a chain-of-custody document, is weighed and photographed, and a visual description is recorded: colour, physical form (powder, solid tablet, vegetable material, liquid), tablet markings if present. No analytical reagents are applied at this step. Gross weight and net weight after packaging removal are documented. This is the point at which NDPS small quantity or commercial quantity categorisation is initially tracked; the weight here is the gross exhibit weight, not the pure drug weight calculated at quantification.
2. Colour test screening (Category C)
   A small aliquot of the exhibit (1-5 mg for powders, a corner scraping for tablets) is tested with the appropriate colour test panel for the suspected drug class based on visual appearance. Results are recorded with reagent identity, volume, colour observed, and time of observation. A positive Marquis and Mecke result for a white powder, for example, indicates the opioid class and directs the next step. A negative colour test does not end the analysis; it narrows the class.
3. UV-Vis spectroscopy (Category B)
   A small amount of the exhibit is dissolved in a validated solvent (typically methanol or acetonitrile) and the UV absorption spectrum is measured between 200 and 400 nm. The lambda-max and spectral shape are compared to a validated reference library. This step provides rapid class confirmation and identifies cases where the exhibit has no UV-active chromophore (cocaine, GABA analogues), directing those to GC-MS or alternative solvent systems.
4. TLC (Category B)
   A sample extract and reference standards are spotted on a silica gel TLC plate and developed in validated solvent systems for the suspected drug class. Rf values are measured and compared to reference standards. Visualisation under short-wave UV (254 nm) and long-wave UV (365 nm), then chemical development with the appropriate staining reagent, provides both Rf and colour response. Two TLC runs in different solvent systems are recommended for NDPS Act exhibits by CFSL Standard Operating Procedures.
5. GC-MS identification (Category A)
   A quantitative extract of the exhibit (typically 0.5 mg dissolved in an internal-standard-containing solvent such as methanol spiked with a deuterated analogue at known concentration) is injected onto the GC column. Retention time and mass spectrum are compared to validated reference standards. The Library match score and the retention time match within the validated window constitute the positive identification. The internal standard corrects for injection volume variation and is used for quantification.
6. LC-MS/MS confirmation and quantification (Category A)
   For thermally labile compounds, when GC-MS identification requires corroboration, or for high-value exhibit quantification with tight uncertainty requirements, LC-MS/MS is run. The Waters Xevo TQ-S or Shimadzu LCMS-8050 in MRM mode provides precursor-to-product ion transitions for up to 50 compounds in a single injection. Quantification uses a five-point calibration curve bracketing the expected concentration range, with 1/x weighting for linearity across a wide dynamic range. Results are reported as concentration with expanded measurement uncertainty (k=2, 95% coverage interval).

An identification result from GC-MS or LC-MS/MS is only as defensible as the calibration system beneath it. SWGDRUG's Recommendations Part IV (Quantitative Analysis) and the ENFSI DWG best-practice manual for drug analysis (Version 3, 2018) both specify requirements for internal standards, calibration curves, system suitability and measurement uncertainty that must be met before a quantitative result is reported.

An internal standard (IS) is a compound added to every sample, calibration standard and quality control sample at the same known concentration before extraction and injection. Its primary function is to correct for variation in extraction efficiency, injection volume, and instrument response that occurs from one injection to the next. For GC-MS drug quantification, isotopically labelled analogues of the target drug (deuterium or 13C labelled, purchased from Cerilliant Corporation, Cayman Chemical, or Sigma-Aldrich) are the preferred internal standards because they have essentially identical chemical behaviour to the target through extraction but are distinguished from the target by their higher mass. For example, morphine-d3 (morphine with three hydrogen atoms replaced by deuterium) is used as the IS for morphine quantification by GC-MS: it co-elutes with morphine (same retention time), has the same extraction efficiency, and is detected at m/z 289 rather than 286, so the two compounds are clearly distinguished in the mass spectrum.

The calibration curve maps instrument response (peak area ratio of target to IS) against known concentration. SWGDRUG and ENFSI DWG both require a minimum of five calibration points spanning the expected concentration range of casework samples, with the points distributed logarithmically to cover low and high concentrations. The curve must demonstrate acceptable linearity (typically r2 greater than 0.995) and precision (typically less than 10 per cent relative standard deviation at each calibration level). A model without adequate calibration cannot produce a defensible quantitative result, regardless of how accurately the identification was made.

System suitability is checked at the start of every analytical run. It involves injecting a mid-range calibration standard and verifying that the peak shape (asymmetry factor), retention time, and response ratio meet pre-specified acceptance criteria before any casework samples are analysed. In the Agilent ChemStation and OpenLAB CDS software environments used with the 7890B GC + 5977B MSD system, system suitability parameters can be set as automated pre-run gates: if a parameter fails, the batch does not proceed.

Peak purity is a concept that applies most directly to HPLC-DAD and LC-MS/MS. In HPLC-DAD, a peak that is pure (containing only one compound) shows a consistent UV-Vis spectrum across its elution profile: the spectrum at the leading edge, the apex and the trailing edge should be superimposable. Spectral mismatch across a peak indicates co-elution of two compounds with different UV spectra, which compromises both identification and quantification. Most modern DAD data systems (the Shimadzu LabSolutions and Agilent OpenLAB CDS) calculate a peak purity index automatically. In LC-MS/MS operating in MRM mode, peak purity is effectively addressed by the specificity of the precursor-to-product ion transition: a compound that co-elutes with the target but has a different molecular mass will not appear in the MRM channel unless it produces the same precursor and product ions, which is chemically very unlikely for structurally distinct compounds.

The ENFSI DWG best-practice manual specifies that quantitative results for controlled drug content must be accompanied by an expanded measurement uncertainty (MU) value at a 95 per cent coverage interval (k=2). Measurement uncertainty propagates from weighing uncertainty (balance calibration, sample preparation), calibration curve uncertainty (regression fitting), and injection uncertainty (quantified by the IS response). A typical GC-MS drug purity result might be reported as "cocaine hydrochloride content 84 per cent ± 4 per cent (k=2, 95% CI)." This means the true value lies within ±4 percentage points of 84 per cent with 95 per cent probability. The ± value is not optional: ISO 17025:2017 Clause 7.6 requires reporting of measurement uncertainty for all quantitative results in accredited laboratories.

Under India's NDPS Act 1985, the quantity of a controlled drug in a seized exhibit directly determines the severity of the charge and the severity of punishment. Small quantity and commercial quantity thresholds are specified in the NDPS Act schedules for each drug class. For morphine, the small quantity threshold is 5 grams and the commercial quantity threshold is 250 grams. For heroin, the thresholds are 5 grams (small) and 250 grams (commercial). For cocaine, 2 grams (small) and 100 grams (commercial). For cannabis (ganja), 1,000 grams (small) and 20,000 grams (commercial). An NDPS conviction at commercial quantity carries mandatory minimum imprisonment of 10 years and a fine of not less than one lakh rupees; commercial quantity charges also shift the evidentiary burden in certain circumstances under NDPS Section 35.

This means that the quantification result from the FSL report is directly connected to the penal consequence. An FSL that reports "cocaine content 45 per cent" on a 300-gram exhibit is providing data that will determine whether the defendant faces a 10-year mandatory minimum. The analytical quantification must therefore be performed with a calibrated instrument, a validated method, traceable reference standards, documented measurement uncertainty, and a quality system that satisfies judicial and regulatory scrutiny.

The US DEA laboratory system operates under the DEA Laboratory Operations Manual, which specifies minimum qualifications for drug chemistry methods, quality control procedures, proficiency testing and reviewer qualification standards. The DEA has seven regional laboratories (including the Special Testing and Research Laboratory in Dulles, Virginia, which handles novel psychoactive substances), with an eighth regional laboratory under construction in New Hampshire and is accredited by ASCLD-LAB (now ANAB, the ANSI National Accreditation Board), which is itself a signatory to the ILAC mutual recognition arrangement (MRA). The UK's official drug analysis historically flowed through the Forensic Science Service until its closure in 2012; today, drug analysis in England and Wales is distributed across private accredited suppliers (Orchid Cellmark, LGC Forensics, KEY Forensic Services) and the Metropolitan Police Forensic Services, all accredited by UKAS (United Kingdom Accreditation Service) under ISO 17025.

The ENFSI Drugs Working Group best-practice manual (most recently DWG-BPM-003, Version 3, 2018) recommends that drug analysis laboratories across EU member states adopt a minimum analytical protocol comprising: colour test or immunoassay screen, chromatographic confirmation (GC-MS or LC-MS/MS), quantification with internal standard and five-point calibration, and measurement uncertainty reporting. The DWG also operates an External Quality Assessment (EQA) proficiency testing scheme in which participating laboratories analyse blind samples and their results are compared with assigned values from independent reference methods. EQA participation is a requirement of ENFSI membership and a condition of accreditation under most EU national accreditation bodies (DAkkS in Germany, COFRAC in France, RvA in the Netherlands, INMETRO in Brazil).

ISO/IEC 17025:2017 is the international standard for testing and calibration laboratories. It specifies requirements for management (document control, non-conforming work, corrective action, internal audits) and technical competence (personnel qualifications, equipment calibration, measurement uncertainty, method validation, and results reporting). A forensic drug chemistry laboratory accredited under ISO 17025 by an ILAC-MRA signatory body (NABL in India, ANAB in the US, UKAS in the UK, DAkkS in Germany) can issue reports that are mutually recognized across all signatory countries. This is practically significant in transnational drug trafficking cases where a seized drug exhibit may be analysed by a laboratory in one country for prosecution in another.

Method validation, as required by ISO 17025 Clause 7.2, establishes the performance characteristics of each drug identification and quantification method: specificity (the method gives a correct result in the presence of expected interferents), linearity (the calibration curve is linear across the stated working range), accuracy (measured concentration matches assigned concentration within agreed limits, typically ±10 per cent for drug purity), precision (within-run and between-run RSD, typically less than 5-10 per cent), limit of detection and limit of quantification, and robustness (performance under deliberate variations in sample preparation conditions, column lot, mobile phase composition). These parameters must be documented in a method validation report before the method is released for casework use, and re-validation or verification is required when a major instrument or reagent change occurs.