HPLC, UHPLC and HPTLC for Forensic Analysis

A standard HPLC system on the toxicology bench at FSL Sector 14 Madhuban or the drug-analysis bench at CFSL Chandigarh consists of seven discrete subassemblies. Two solvent reservoirs (water with a buffer modifier in bottle A, acetonitrile or methanol in bottle B) feed an inline vacuum or membrane degasser that strips dissolved gases, because air bubbles in the pump head cause baseline noise and pressure drops. The high-pressure pump, either binary (two channels for A and B) or quaternary (four channels for A, B, C and D), proportions the two solvents and pushes the blend at a typical 0.5 to 2.0 mL/min flow rate against 200 to 400 bar of column back-pressure. A binary pump on a Shimadzu Prominence LC-2030 or Agilent 1260 is the most common configuration in Indian labs.

The autosampler picks 0.1 to 100 µL of sample from a vial in a tray and injects it into the flowing mobile phase via a six-port rotary valve. The column oven holds the column at a stable 25 to 60 °C, because retention shifts measurably with temperature and reproducibility depends on holding it constant. The column is the chemistry: a 250 × 4.6 mm tube packed with 5 µm spherical silica particles, surface-modified with octadecyl chains for reverse-phase work. The detector reads the eluting peaks (UV-Vis at a single wavelength, DAD over the full range, fluorescence for fluorescent analytes, refractive-index for non-absorbing solutes, or a mass spectrometer for confirmatory work). The data system on the attached PC integrates each peak, applies the calibration curve and prints the result.

The column is the part most worth understanding in detail. Modern reverse-phase silica is hybrid silica or fully end-capped to reduce silanol activity that otherwise tails basic drugs. Pore size is typically 100 Å for small-molecule work and 300 Å for peptides and proteins. The C18 ligand is the most common surface chemistry, but C8 (shorter chain, slightly less retentive) and phenyl (π-π interaction with aromatic analytes) are common second choices when C18 gives poor selectivity. Selecting the wrong column for the analyte chemistry is the single most common reason a method-development run fails, and the Waters, Phenomenex and Agilent column-screening kits on most Indian R&D benches exist precisely to remove this trial-and-error.

Reverse-phase chromatography (RP-LC) is the dominant mode in forensic work. The stationary phase is non-polar (C18, C8 or phenyl bonded to silica), the mobile phase is polar (water plus methanol or acetonitrile, often with a buffer modifier), and analytes elute in order of increasing hydrophobicity. The drugs-of-abuse panel at most state SFSLs runs on a 150 × 4.6 mm C18 column with a gradient from 5 percent acetonitrile in 25 mM ammonium acetate buffer at pH 4.5 to 95 percent acetonitrile over 20 minutes, with DAD scanning from 200 to 400 nm. Morphine, codeine, methamphetamine, MDMA, cocaine, benzodiazepines and the common adulterants resolve in the same chromatogram, each peak identified by retention time plus DAD library match.

Normal-phase chromatography flips the polarities. The stationary phase is bare silica or amino-bonded silica, and the mobile phase is non-polar (hexane plus a small percentage of isopropanol or ethyl acetate). Analytes elute in order of decreasing polarity, and the technique is best for separating closely related lipids, sugars and chiral isomers on dedicated chiral columns. HILIC (hydrophilic interaction liquid chromatography) sits between the two: a polar stationary phase (bare silica, amide or zwitterionic) with a high-organic mobile phase (typically 80 to 95 percent acetonitrile in aqueous buffer). HILIC retains polar metabolites, polar drugs (creatinine, metformin), small organic acids and underivatised amino acids that simply do not retain on C18. Toxicology screens that need to catch both lipophilic drugs and polar metabolites in the same run sometimes use a serial HILIC-RP combination on a column-switching system.

Ion-exchange chromatography uses a charged stationary phase (sulphonate for cation exchange, quaternary amine for anion exchange) and elutes analytes by competing-ion gradient. It is the workhorse for inorganic-ion analysis (chloride, nitrate, perchlorate post-blast residues) and for protein purification at preparative scale. Size-exclusion chromatography (SEC) sieves analytes by hydrodynamic radius through a porous gel; large molecules elute first because they cannot enter the pores. SEC is rarely a forensic workhorse but is the standard for sizing protein and polymer fragments in trace-evidence work.

Mobile-phase modifiers matter as much as the organic solvent. Formic acid at 0.1 percent volume-by-volume is the LC-MS friendly default because it ionises basic analytes well in positive-mode electrospray and evaporates cleanly in the source. Ammonium acetate or ammonium formate buffer at 5 to 25 mM in the pH 3 to 7 window controls peak shape for ionisable drugs by holding their ionisation state constant. Trifluoroacetic acid (TFA), once popular for sharpening basic peaks, has fallen out of favour for LC-MS work because it ion-suppresses the electrospray. Phosphate buffers are still used for UV-only methods but are completely incompatible with mass spectrometry and are now reserved for legacy methods that have not been re-validated.

A single-wavelength UV-Vis detector reads absorbance at one wavelength chosen at method-development time (245 nm for paracetamol, 230 nm for morphine, 254 nm for many aromatic drugs). It is the cheapest detector and is the first choice for routine quantitation when the analyte and its chromophore are well characterised. Limit of detection on a typical UV detector for a strong absorber is in the low nanograms-on-column range, which is enough for the 1 to 100 µg/mL working concentrations of most toxicology methods.

A diode-array detector records the full 190 to 800 nm absorbance spectrum of every peak as it elutes. The two operational benefits over single-wavelength UV are large. Peak-purity analysis compares the spectrum at the front, apex and tail of each peak; a co-eluting interferent shows up as a spectral mismatch that single-wavelength UV cannot see. Library matching compares the apex spectrum against a stored library of reference spectra for hundreds of drugs and metabolites, which gives a second identification point on top of retention time. Most toxicology and drugs-of-abuse methods at CFSL Chandigarh, FSL Madhuban and the state SFSL benches now run with DAD as the default UV detector.

Fluorescence detectors (FLD) are 100 to 1000 times more sensitive than UV for fluorescent analytes. They excite the eluting peak at one wavelength and read the emission at a longer wavelength, so non-fluorescent matrix interferents simply never show up in the chromatogram. Quinine, polycyclic aromatic hydrocarbons (PAHs), aflatoxins B1, B2, G1 and G2 in food samples, and OPA-derivatised primary amino acids are the standard fluorescence applications. The FSSAI mycotoxin laboratories use HPLC-FLD with post-column iodine derivatisation for aflatoxin quantitation in groundnut, maize and chilli samples down to 0.1 µg/kg, well below the 15 µg/kg total-aflatoxin limit in the Food Safety and Standards Regulations.

Electrochemical detection (ECD) reads the current generated when the eluting analyte is oxidised or reduced at a working electrode held at a fixed potential. It is highly selective for catecholamines (adrenaline, noradrenaline, dopamine), phenols, thiols and a handful of nitro-reducible compounds. Refractive-index detection (RID) is universal but insensitive; sugars, polymers, polyols and underivatised lipids are its routine analytes. Mass spectrometry (single-quad, triple-quad MS/MS, high-resolution Q-TOF or Orbitrap) is the gold standard and is covered in the next module of this curriculum; in routine forensic practice, LC-MS/MS is the technique that turns a presumptive HPLC-DAD identification into a confirmatory result that survives in court.

UHPLC (ultra-high-performance liquid chromatography) is HPLC done on sub-2 µm particle columns at pressures the original HPLC pumps cannot generate. The physics behind the speed-up is captured by the Van Deemter equation: smaller particles reduce the eddy-diffusion and mass-transfer terms, so the optimum linear flow velocity rises and the height-equivalent of a theoretical plate (HETP) drops. In practical terms, a 100 × 2.1 mm column packed with 1.7 µm Acquity BEH or 1.6 µm Kinetex particles delivers the same plate count as a 250 × 4.6 mm column packed with 5 µm particles, but it does so in three to five minutes instead of 25 to 30 minutes, and it consumes about a tenth of the mobile phase per injection.

The pressure budget is the cost. Sub-2 µm particles at 0.4 to 0.6 mL/min through a 100 × 2.1 mm column generate 600 to 1500 bar of steady-state back-pressure. The pump pistons, the inline filters, the column hardware and the connecting capillaries all need to be rated for that pressure, which is why a Waters Acquity, Agilent 1290 Infinity II or Shimadzu Nexera UHPLC stack costs roughly twice what an equivalent classical HPLC stack costs. Detector sampling rates also have to keep up with the narrower peaks; a 1-second peak width that a 5 Hz UV detector handled cleanly on a 5 µm column becomes a 0.1-second peak that needs a 40 Hz detector to integrate without distortion.

The labs that have most to gain from the change are the high-throughput ones. NDTL Delhi (the National Dope Testing Laboratory) runs UHPLC-MS/MS for the full WADA prohibited-substance list across a few thousand athlete samples a year; the seven-fold cut in run time is what makes the workflow feasible on a single instrument. The CDSCO bulk-drug-assay labs and the FSSAI multi-residue pesticide screens run UHPLC-MS/MS for the same throughput reason. NIPER Mohali uses UHPLC for both pharmaceutical impurity profiling and ayurvedic phytochemical mapping; the resolving power is what lets a five-minute method baseline-resolve eight or nine related impurities that classical HPLC partially co-elutes.

The flip side: not every method needs UHPLC. A routine paracetamol or salicylate quantitation in serum runs perfectly on a 150 × 4.6 mm C18 column with a 10-minute isocratic method, and converting it to UHPLC adds capital cost without adding analytical value. The decision rule is straightforward: if the bench is throughput-bound, if the matrix is complex enough that classical HPLC cannot resolve it, or if the method is being coupled to an LC-MS/MS that benefits from sharper peaks, UHPLC is worth the spend. Otherwise the existing HPLC stack is fine, and the budget is better spent on a better column or a DAD upgrade.

HPTLC (high-performance thin-layer chromatography) is the modern, instrumented version of the classical TLC plate. The silica is a finer 3 to 5 µm particle (versus 10 to 15 µm for ordinary TLC), the plates are 10 × 10 cm or 20 × 10 cm aluminium-backed glass-fibre or pre-coated silica, and every step that used to be done by hand on classical TLC is now automated. Sample application is done by a Camag Linomat 5 that sprays a precise band of sample (typically 5 to 10 mm long, 1 to 50 µL volume) through a 100 µm capillary, which sharpens the start band and removes the variability of manual spotting. Plate development happens in an automated developing chamber (Camag ADC2) that controls saturation atmosphere, solvent front position and developing time. The dried plate is then scanned by a Camag TLC Scanner 4 that runs a narrow slit of UV or visible light across each track and records the densitogram (absorbance vs distance) for quantitation.

The forensic and quality-control questions HPTLC answers best are the ones HPLC answers awkwardly. The Ayurvedic Pharmacopoeia of India and the Indian Pharmacopoeia herbal monographs prescribe HPTLC fingerprint patterns for hundreds of single-herb and multi-herb preparations; the comparison is visual track-to-track on the same developed plate, with the densitogram providing the quantitative back-up. Ashwagandha root powder spotted next to a reference Ashwagandha extract on the same plate, developed in a toluene-ethyl-acetate-formic-acid mobile phase, scanned at 254 nm and 366 nm under UV and after derivatisation with anisaldehyde-sulphuric-acid spray, gives a fingerprint that immediately flags adulteration with cheaper bulking herbs. NIN Hyderabad (the National Institute of Nutrition) and several Council of Scientific and Industrial Research (CSIR) labs run this exact workflow for the FSSAI adulteration-screening programmes on turmeric (curcumin and lead-chromate adulterant), saffron (synthetic dye adulterants), asafoetida (rosin and starch adulterants) and chilli powder (Sudan dyes).

NIPER Mohali, the National Institute of Pharmaceutical Education and Research, runs an integrated HPTLC-UHPLC workflow that uses HPTLC for high-sample-count screening and fingerprinting and UHPLC for the targeted impurity profiling of confirmed problem batches. The reasoning is operational: an HPTLC plate runs 15 to 20 samples in parallel under identical chromatographic conditions, which is what a comparative fingerprint demands. An HPLC system runs samples sequentially, and inter-run drift in column temperature, mobile-phase composition and detector lamp intensity makes track-to-track comparison shaky. Forensic ink comparison in questioned-document work uses the same logic: 12 to 15 ink samples spotted on the same HPTLC plate, developed in a butanol-ethanol-water mobile phase, scanned in visible-light reflectance mode, give a comparative dye-pattern map that no sequential HPLC injection sequence will reproduce.

The other forensic teeth of HPTLC are in mycotoxin screening, drug-impurity profiling at the pre-clinical R&D bench, and a handful of doping analyses that pre-date the LC-MS/MS conversion. Aflatoxins B1, B2, G1 and G2 in groundnut and maize, ochratoxin A in coffee and wine, and zearalenone in maize all run on HPTLC with fluorescence-mode densitometry at 366 nm; the FSSAI Reference Laboratory in Ghaziabad and the regional labs in Kolkata, Chennai and Mumbai all keep an HPTLC stack on the bench for screening before any LC-MS confirmation. The cost is the other reason: a complete Camag Linomat-ADC2-Scanner 4 stack runs at roughly the cost of a single mid-range HPLC, and the per-sample cost on plate-based screening is a small fraction of the per-injection cost on HPLC-MS.

A forensic chromatographic method is not validated until six numbers are on paper. Linearity is the calibration-curve correlation, with R² > 0.999 over at least five concentration points across the working range. Limit of detection (LOD) is the lowest concentration that gives a peak with signal-to-noise of three; limit of quantitation (LOQ) is the lowest concentration that gives signal-to-noise of ten and an RSD below ten percent. Intra-day precision (six replicate injections of the same sample on the same day) and inter-day precision (the same exercise on three different days) both need RSD below two percent for pure standards and below five percent for real biological matrices. Accuracy or recovery (analyte spiked into blank matrix, extracted and re-quantified) needs to land between 90 and 110 percent for most regulatory applications. Robustness checks how much the result moves when the mobile-phase composition, temperature, pH or flow rate is deliberately perturbed by a few percent; a robust method is one whose result does not move with these small disturbances.

Indian forensic and quality-control labs run a fairly consistent instrument list. The Shimadzu Prominence LC-2030 and the Nexera UHPLC sit on most state SFSL and CDSCO benches alongside the Agilent 1100, 1260 and 1290 series. Waters Acquity UPLC is the standard at NDTL Delhi, the higher-end CFSL benches and several IIT and NIPER central facilities. The HPTLC stack is almost universally Camag (Linomat 5 sample applicator, ADC2 developing chamber, TLC Scanner 4 densitometer, winCATS or visionCATS software), with a handful of Anchrom-supplied alternatives in regional labs. The column manufacturers are Waters (Acquity BEH, Symmetry, XBridge), Agilent (ZORBAX, Poroshell), Phenomenex (Kinetex, Luna) and Thermo (Hypersil, Accucore), with Indian distributors handling the supply chain to most labs.

The principal lab-to-application mapping in Indian forensic and quality-control practice is as follows. CFSL Chandigarh runs HPLC-DAD plus LC-MS/MS for the toxicology and drugs-of-abuse panel and for the seized-narcotics quantitation that the Narcotics Control Bureau refers in. FSL Sector 14 Madhuban runs the same panel for Haryana state cases. NIPER Mohali runs HPTLC plus UHPLC for pharmaceutical impurity profiling and herbal-product fingerprinting and supports the Ayush ministry's monograph development. NIN Hyderabad runs HPTLC for food-adulterant screening and for nutritional-component assays in the FSSAI surveillance programmes. The FSSAI national reference and regional labs run HPLC-FLD and HPLC-MS for vitamin assays, mycotoxin screening and the multi-residue pesticide programmes. NDTL Delhi runs UHPLC-MS/MS for the WADA anti-doping panel across the BCCI, AIFF and Sports Authority of India sample stream. CDSCO labs at Ghaziabad, Kolkata, Chennai and Mumbai run HPLC for bulk-drug assay, dissolution testing and counterfeit-pharmaceutical investigation under the Drugs and Cosmetics Act.

In practice, the validation file for every method is what cross-examination focuses on first. Routine measurement and documentation of those six numbers over the lifetime of a method is what qualifies a chromatography bench as a forensic instrument fit for medicolegal reporting.