Tandem Mass Spectrometry (MS/MS) and Spectral Interpretation

A single mass spectrometer answers one question. What is the m/z of the ions that survived ionisation? That flags a target but does not convict. Two co-eluting compounds with the same nominal mass produce the same single-MS peak, and isobaric matrix interferences are the rule in urine, blood and viscera digests.

Tandem MS adds a second filter. The first analyser picks a precursor. A collision cell fragments it. The second analyser reads the products. Two co-eluting isobars almost never share both the precursor and a diagnostic product, which is why MS/MS is the modern confirmation backbone.

The four classical scan modes on a triple quadrupole map onto four different forensic questions.

A product ion scan parks Q1 on a chosen precursor and scans Q3 across the full product range. The output is a fragmentation spectrum for that one precursor, which is what an analyst captures during method development and what a QTRAP records when matching against an MS/MS library.

A precursor ion scan parks Q3 on a chosen product and scans Q1 across the full precursor range. The output is every ion that fragments to give the chosen product. Park Q3 on m/z 184 (the phosphocholine fragment) and you find every phosphatidylcholine in a lipid extract.

A neutral loss scan steps Q1 and Q3 in lockstep with a fixed offset. Park the offset at 176 Da and you find every glucuronide conjugate. Park at 80 Da and you find every sulphate. The neutral-loss scan is the dirty-screen workhorse for phase II metabolites in urine.

MRM is the targeted production mode. Q1 is parked on the precursor, Q3 on the product, and the duty cycle dwells on each transition for 5 to 50 milliseconds. Sub-ng/mL detection in plasma is routine. Two transitions per analyte (the 286 to 152 and 286 to 165 pair for morphine, for example) is the BSA 2023 Section 63 minimum.

Four mass-analyser families show up in Indian forensic-toxicology labs, each with a distinct sweet spot.

The triple quadrupole (QqQ) is the targeted-quantitation workhorse. Three sets of parallel rods, Q1 and Q3 as mass filters, q2 as a collision cell filled with nitrogen or argon. Mass resolution is unit (about 0.7 Da FWHM), enough for nominal-mass MRM. The Agilent 6470 at CFSL Chandigarh and AIIMS Forensic Toxicology, the Sciex 6500+ QTRAP at CDFD Hyderabad and the Waters Xevo TQ-S at NDTL Delhi all live in this family.

The quadrupole time-of-flight (Q-TOF) is the high-resolution targeted-plus-untargeted instrument. Q1 mass filter, q collision cell, TOF third stage recording full-scan accurate-mass spectra at 30,000 to 60,000 resolution and roughly 5 ppm mass accuracy. The instrument records every product ion of every precursor, so the analyst can extract any transition or any full spectrum after the run. This is what a forensic-tox lab uses for novel psychoactive substances and post-mortem unknown identification.

The Orbitrap (in its Q-Exactive and Q-Exactive HF forms) runs at 70,000 to 240,000 resolution with sub-2 ppm accuracy. The collision cell is HCD (higher-energy collisional dissociation), more reproducible than ion-trap CID and the basis for MS/MS library matching. CFSL Hyderabad runs a Thermo Q-Exactive for NPS identification and forensic metabolomics.

The ion trap is the MS-n instrument. The trap holds ions, ejects all but a chosen precursor, fragments it, holds the products, ejects all but a chosen second-generation precursor, fragments again, and out to MS-10. The trade-off is the one-third cut-off (no retention of product ions below about one-third of the precursor mass), which costs low-mass diagnostic ions in some classes.

The collision cell is where the chemistry happens. A precursor accelerates through a potential difference (the collision energy), collides with neutral gas, and converts kinetic energy into vibrational excitation. When internal energy exceeds a bond dissociation threshold the ion fragments. Different bonds break at different energies, which is why MRM optimisation is a collision-energy ramp.

The window for small-molecule drugs in ESI-positive mode is roughly 5 to 50 eV. Below 10 eV the spectrum is dominated by the protonated molecule. Above 40 eV most analytes over-fragment. The sweet spot per transition is rarely the same for the quantifier and qualifier of the same compound, and method-development software (Sciex Analyst, Waters MassLynx, Agilent MassHunter Optimizer) ramps the energy in 1 or 2 eV steps to find each automatically.

Picking the transitions themselves follows a recipe. Start with the parent ion, almost always [M+H]+ for ESI-positive small molecules and [M-H]- for ESI-negative. Pick at least two product ions: the most intense is the quantifier, the next most intense (or most diagnostic) is the qualifier. A third is sometimes added for low-prevalence analytes.

Verify the ratio. Inject the reference standard at three concentrations, calculate the quantifier-to-qualifier intensity ratio at each, and lock the median as the reference. The same ratio in a sample must agree within plus or minus 20 percent (30 percent near the LOQ). A ratio outside tolerance flags an interference even when both transitions are present and the retention time matches.

A handful of canonical transitions used in Indian forensic-tox labs:

The deuterated internal standard corrects for matrix effects. ESI sources suffer ion suppression, where co-eluting matrix components steal charge from the analyte and depress the response invisibly. A d3-morphine spiked into every sample co-elutes with morphine, suffers the same suppression, and the analyte/IS area ratio cancels the matrix effect. Skipping the IS to save a transition slot is the most common reason a quantitation drifts out of tolerance over a long batch.

Accurate-mass interpretation is what a Q-TOF or Orbitrap brings to the bench that a triple-quad cannot. The instrument reports m/z to four decimal places at 5 ppm or better, and the decimal pattern (the mass defect) constrains the molecular formula sharply.

Take caffeine. The nominal molecular ion is 194 Da and the monoisotopic accurate mass is 194.0804. A formula-finder algorithm tries combinations of CHNOPS within an elemental constraint, ranks each candidate against the measured mass and the M+1 and M+2 isotope intensities, and returns a ranked list. At 5 ppm the top hit is usually the right one.

The isotope pattern is the second axis. Carbon-13 sits at 1.1 percent natural abundance, so a 10-carbon ion shows an M+1 at about 11 percent of M. Sulphur and chlorine push M+2 up sharply (S-34 at 4.4 percent, Cl-37 at 24 percent). One chlorine gives M+2 at about a third of M. Two chlorines push it close to 65 percent. Bromine produces a near 1:1 M:M+2 doublet.

GC-MS with 70 eV electron-impact runs on a different logic. EI fragmentation is reproducible enough that the same compound on a Shimadzu QP2020 in Mumbai, an Agilent 5977B in Delhi and a Thermo ISQ in Bengaluru produces near-identical spectra, which makes library matching the standard identification step. The NIST 2023 library carries about 394,000 EI spectra and the Wiley Registry 2023 carries about 873,000. A similarity score above 800 with a matching retention index is the working identification. Final confirmation requires injecting the reference standard on the same instrument and matching spectrum plus retention time within plus or minus 2 percent.

Spectral interpretation by hand still matters when the library returns nothing. The molecular ion (M+•) carries the molecular weight when visible, though EI is harsh and many compounds (alcohols, branched aliphatics, amines) show weak or absent M+•. The base peak, the most intense ion, is usually a stable cation from a favourable cleavage and is diagnostic on its own.

Common neutral losses from M are a fast read on functional groups. Loss of 15 is methyl, 17 is hydroxyl, 18 is water, 28 is CO or ethylene, 29 is CHO or ethyl, 31 is OCH3 or CH2OH, 43 is acetyl, and 45 is COOH or OC2H5. A strong M-15 plus M-43 cluster suggests a methyl ketone. A strong M-31 suggests a methyl ester.

Diagnostic product ions point to substructures. The tropylium cation at m/z 91 (C7H7+) is the methylbenzene tag. The benzoyl cation at m/z 105 (C6H5CO+) flags a benzoate or aroyl group. The phenyl cation at m/z 77 marks a monosubstituted benzene. The pyridyl ions at m/z 51 and 65 mark a pyridine ring. An unknown with M at 240, base peak at 91 and a strong M-15 to give 225 carries a methylbenzene, a methyl close to the core, and a molecular weight of 240. A NIST search on those constraints usually closes the identification.

The Bharatiya Sakshya Adhiniyam 2023 Section 63 sets the evidentiary frame for electronic and instrumental analytical evidence in Indian criminal proceedings. A confirmatory MS/MS report needs to satisfy a converged checklist to clear that bar.

Two MRM transitions per analyte is the minimum, one quantifier and one qualifier, with the quantifier-to-qualifier intensity ratio matching the reference standard within plus or minus 20 percent (30 percent near the LOQ).

A deuterated internal standard, ideally one per analyte, spiked into every sample and every calibrator at a fixed concentration. The calibration is built on the analyte-area-to-IS-area ratio.

Retention-time agreement within plus or minus 2 percent of the reference standard, run in the same batch on the same column.

A calibration curve with at least five points plus a reagent blank, R-squared above 0.999 across the working range, verified at every batch. LOD by the 3-sigma method on at least seven blank or low-spike replicates and LOQ by 10-sigma, both on the certificate.

Method validation parameters from the original ISO/IEC 17025 method file: precision (RSD below 15 percent at LOQ, below 10 percent above), accuracy (90 to 110 percent on a spiked matrix-matched sample), matrix effect, recovery and specificity against six blank matrices.

A certified reference material with every batch where one is available. UTAK whole-blood drug-of-abuse panels and Cerilliant single-analyte CRMs are the workhorses at Indian NABL labs. The CRM result must lie within its certificate uncertainty for the batch to be released.

1. 1. Sample preparation and IS spike
   Aliquot the matrix (whole blood, urine, viscera digest), add the deuterated internal standard at a fixed concentration before extraction, and process through protein precipitation, liquid-liquid or solid-phase extraction. The IS goes in early so it tracks every loss the analyte suffers.
2. 2. UHPLC-MS/MS injection
   Inject 5 to 20 microlitres onto a C18 UHPLC column with a methanol or acetonitrile gradient containing 0.1 percent formic acid for ESI-positive work. Source: ESI at 4 to 5 kV, 350 deg gas temperature, scheduled MRM with 20 to 50 ms dwell on each transition.
3. 3. Quantitation against the calibration curve
   Build the calibration line as analyte/IS area ratio versus concentration, with at least five points plus a reagent blank, R-squared above 0.999. Read the unknown analyte/IS ratio against the line. Reject the calibrator point if its back-calculated concentration is outside 80 to 120 percent of nominal.
4. 4. Confirmation checks
   Verify retention time within plus or minus 2 percent of the standard. Verify quantifier-to-qualifier ratio within plus or minus 20 percent. Verify CRM result within its certificate uncertainty. Verify the method blank is below the LOD. Any failure forces a re-injection or a re-extraction, not a release.
5. 5. Reporting under BSA 2023 Section 63
   Issue the certificate with matrix, method (LC-MS/MS triple-quad, ESI+, scheduled MRM), the two transitions, the IS, retention time, ratio with tolerance, calibration R-squared, LOD, LOQ, measurement uncertainty, CRM result against certified value, and the medico-legal cut-off (NDPS schedule for narcotics, DFC 5-panel for sexual-assault casework).

The capital outlay on a single MS/MS instrument maps the installed base cleanly onto casework type. Triple-quads at routine-quantitation labs, high-resolution instruments at unknown-identification labs, ion traps at academic and structural-elucidation benches.

CFSL Chandigarh runs an Agilent 6470 LC-MS/MS triple-quad as its routine MRM workhorse for NDPS Schedule I and II analytes, opioids, benzodiazepines and amphetamine-type stimulants, with about 80 routine analytes plus a designer-drug list that grows quarterly.

CDFD Hyderabad runs a Sciex 6500+ QTRAP, a triple-quad with an ion-trap third stage. The hybrid lets the lab acquire MRM-triggered enhanced product ion (EPI) spectra in the same injection for low-prevalence analytes needing library confirmation.

NDTL Delhi, the WADA-accredited dope-testing lab, runs the Waters Xevo TQ-S for athlete-urine confirmation. It reaches sub-pg/mL on selected analytes, which is what the WADA minimum required performance levels demand.

CFSL Hyderabad runs a Thermo Q-Exactive Orbitrap for NPS identification and post-mortem unknown screening. The full-scan accurate-mass workflow lets the analyst re-mine data acquired six months earlier for analytes that were not on the target list at injection time.

AIIMS Forensic Toxicology Delhi runs a second Agilent 6470 alongside its AAS and ICP-MS bench, focused on therapeutic-drug monitoring overflow and suspected-poisoning admissions from the AIIMS emergency.

NFSU Gandhinagar runs a Sciex QTRAP plus a Waters Synapt G2 Q-TOF for teaching, research and casework referral from western Indian state SFSLs.

The state SFSL network is more variable. Karnataka, Maharashtra, Tamil Nadu and Punjab run Agilent or Sciex triple-quads acquired through state-government modernisation programmes. Smaller state labs still depend on GC-MS and refer LC-MS/MS work upward to CFSL or AIIMS, which slows turnaround on time-sensitive NDPS and DFC casework.