Ink Analysis Methods: TLC, HPLC, Raman, FTIR and Mass Spectrometry

The Video Spectral Comparator, manufactured primarily by Foster and Freeman (UK) in the VSC series (VSC6000, VSC6000/HS), is a forensic optical instrument that illuminates a document under multiple wavebands of light (ultraviolet at 254 and 365 nm, visible blue through red at 450 to 700 nm in stepped bands, and near-infrared at 700 to 1000 nm) and captures the reflected or fluorescent image through matched optical filters. Two inks that appear identical to the naked eye under visible daylight may show different levels of reflectance or fluorescence when the illumination waveband is stepped across the visible and near-infrared range.

The forensic application is differentiation: if a questioned entry and a baseline entry are both present on the same page and both appear the same colour, VSC examination may reveal that one ink fluoresces under UV and the other does not, or that one absorbs strongly in the near-infrared (appearing dark) while the other reflects (appearing lighter). This differentiation of inks on a single document is often the starting point for an alteration finding.

VSC examination is non-destructive: no sampling, no extraction, no damage. This makes it the preferred first step in document examination workflows at the USSS, the FBI Forensic Document Laboratory in Quantico, the Dutch NFI, the BKA Document Examination Unit, and the CFSL laboratories in India. In UK Crown Court proceedings, VSC-derived images are routinely adduced as demonstrative evidence, often alongside light microscopy images of the paper surface.

The limitation of VSC is sensitivity: it differentiates inks by their bulk optical response, but it cannot identify colourant composition and it cannot reliably differentiate two inks from the same manufacturer's product family. If the alteration was made with a pen from the same brand family as the original (for example, two Staedtler triplus ballpoints from the same year's batch), VSC may not reveal the difference, and chemical analysis must follow.

Thin-layer chromatography separates the dye components of an ink extract by differential migration through a stationary phase (typically silica gel on aluminium or glass backing) driven by capillary action of a mobile-phase solvent mixture. The dissolved ink extract is spotted at the origin, the plate is placed in a developing chamber, and the solvent front migrates up the plate. Different dye molecules travel at characteristic rates relative to the solvent front (Rf values), producing a pattern of coloured bands that identifies the dye composition.

For forensic ink analysis, the standard USSS/BKA method uses extraction with 3 to 5 microlitres of solvent (methanol, ethanol, or pyridine, depending on the dye class and vehicle) applied to a 1 to 2 mm section of ink stroke. The extract is concentrated and spotted. Developing solvents for ballpoint ink dyes typically use a combination of an organic solvent and a base, such as ethyl acetate:methanol:ammonia in various ratios. Retardation factors for the major blue ballpoint dye classes are well established: crystal violet migrates to a characteristic Rf, Victoria blue B to another, and the separation between them is the fingerprint of that ink's dye system.

TLC is a screening tool rather than a definitive identification method. Its advantages are speed (a plate develops in 15 to 30 minutes), low cost, minimal equipment requirements, and intuitive visual output accessible to the court. Its limitations are resolution (poorly resolved dye mixtures produce overlapping bands), sensitivity (very aged or very small samples may yield insufficient extract), and reproducibility (Rf values vary with temperature, humidity, and solvent batch). The SWGMAT (Scientific Working Group for Materials Analysis, US) and ENFSI (European Network of Forensic Science Institutes) forensic document examination guidelines recommend TLC as a first-line screen, with HPLC as confirmatory analysis when TLC gives ambiguous results.

In practice across European laboratories (BKA, NFI, Forensic Science Laboratory Ireland, LKA laboratories in German federal states), TLC plates from questioned ink samples are documented photographically and compared visually with reference plates prepared from the library entries. A match is recorded as "consistent with" rather than "identical to" unless the full dye pattern, solvent, and HPLC profile are all concordant.

High-performance liquid chromatography separates the components of a solvent extract by differential partitioning between a mobile phase and a stationary phase in a pressurised column. For ink dye analysis, reversed-phase columns (C18 bonded silica, typically 150 x 4.6 mm, 5 micrometer particle size) are standard. The mobile phase is a gradient of water and acetonitrile (or methanol), both with an acidic or basic modifier, programmed to step the polarity over a 20 to 40 minute run. Each dye component elutes at a characteristic retention time determined by its hydrophobicity and charge.

The photodiode-array (PDA) detector captures a full UV-Vis absorption spectrum (200 to 700 nm or wider) at every elution time point, producing a three-dimensional data matrix: retention time, wavelength, and absorbance. For each peak in the chromatogram, the examiner can extract the absorption spectrum and compare it against reference library spectra. A ballpoint ink extract typically shows a series of peaks corresponding to the primary triarylmethane dyes and their degradation products, with the ratio of peak areas being characteristic of the formulation.

For ink dating by dye decay (the LaPorte method), HPLC-PDA is the primary instrument. The decay of specific dye components relative to stable reference components over time produces a ratio that changes predictably with age, allowing an estimate of when the ink was deposited. This application is discussed in the third topic in this module. For ink classification, the retention time plus spectral identity of each peak is compared against the USSS or BKA database entry for the candidate manufacturer.

Coupling HPLC to electrospray ionisation mass spectrometry (ESI-MS or LC-MS) adds a third dimension of identity: molecular mass. For dye identification, molecular ion peaks and characteristic fragmentation patterns allow confident structural assignment even for minor components present below the UV detection limit. The USSS began incorporating LC-MS into ink examination workflows in the 1990s, and it is now considered the reference-standard method for complete dye identification in US federal casework. European laboratories including the BKA and NFI use LC-MS as a confirmatory method for novel or previously unclassified formulations.

Raman spectroscopy measures the inelastic scattering of monochromatic laser light by molecular bonds. When a photon from the laser excites a molecule, most photons are scattered elastically (Rayleigh scattering, same wavelength as incident light). A small fraction interact with molecular vibrations and scatter at shifted frequencies: the difference in wavenumber between the incident and scattered photon equals the energy of the molecular vibration. A Raman spectrum is thus a fingerprint of the molecular vibrations present in the sample, providing structural information comparable to infrared spectroscopy but with complementary band intensities.

For forensic ink analysis, Raman microscopy (using a confocal microscope to focus the laser on a spot of approximately 1 to 2 micrometres diameter) provides several advantages. First, it is non-destructive: no extraction, no solvent, no damage to the document. Second, it can interrogate a spot smaller than a single line stroke width, allowing analysis of individual components on a multi-ink document surface. Third, Raman is highly sensitive to pigments with low fluorescence: carbon black, copper phthalocyanine pigment, and other graphitic or aromatic pigment structures produce intense, characteristic Raman spectra.

Practical limitations include fluorescence interference: many paper substrates and some dye classes exhibit strong broadband fluorescence when irradiated by visible lasers (532 nm green, 633 nm red), which overwhelms the Raman signal. This is managed by switching to a near-infrared laser (785 nm or 1064 nm), at which fewer molecules fluoresce, or by photobleaching the sample before measurement. Some blue ballpoint dyes (triarylmethane dyes such as crystal violet) show sufficient Raman activity for identification; others are better analysed by HPLC-PDA.

The published Raman spectral database for forensic ink components includes entries from work at NIST (National Institute of Standards and Technology, US), the FBI Laboratory, the University of Lausanne IPSC (Switzerland), and research groups at the BKA. In the UK, the University of Strathclyde has published extensively on Raman analysis of historical manuscripts and modern forensic inks. The technique has been applied in Indian CFSL laboratories for ink analysis in manuscript examination and questioned-cheque investigations.

Fourier-transform infrared spectroscopy measures the absorption of infrared radiation by molecular bonds at frequencies corresponding to bond stretching and bending vibrations. An FTIR spectrum displays absorbance as a function of wavenumber (typically 4000 to 400 cm-1). Each class of chemical bond produces characteristic absorptions: carbonyl groups (C=O) absorb near 1700 to 1750 cm-1; aromatic C-C bonds absorb near 1480 to 1600 cm-1; ether C-O-C bonds absorb near 1050 to 1260 cm-1; alcohol O-H stretching absorbs near 3200 to 3600 cm-1. Complex materials such as ink formulations produce complex spectra, but distinctive features of major components can be identified.

For forensic ink analysis, FTIR is most useful for identifying the resin and vehicle components that HPLC-dye analysis does not capture. A phenolic resin binder produces characteristic C=O absorption at 1700 to 1720 cm-1 and ether bands at 1240 cm-1, differing from an alkyd resin (lower carbonyl at 1730 cm-1 with ester-band fine structure) or a polyurethane. Because ink manufacturers change resin suppliers and switch between resin classes for cost, performance, or regulatory reasons, FTIR of the resin component can distinguish formulations that have identical dye compositions, or flag a reformulation event.

The practical challenge with FTIR on ink strokes is that the paper substrate contributes strongly to the spectrum (cellulose has major absorptions at 1050 cm-1 and 3300 cm-1 that overlap with ink features). Two approaches address this: attenuated total reflectance (ATR-FTIR), which probes only the outermost microns of the ink surface without paper penetration, and micro-FTIR with spectral subtraction of a blank paper reference. The ATR method is effectively non-destructive; the subtraction method works best when a clean paper area adjacent to the ink stroke is available.

FTIR data from the ink surface can also detect added components: optical brighteners in the paper, coatings, or resins applied after the document was prepared, all of which may leave FTIR-detectable signatures. In historical document analysis, FTIR is the primary method for characterising iron gall ink degradation, identifying binder components in medieval manuscripts, and detecting modern conservation interventions in antique documents.

Gas chromatography-mass spectrometry (GC-MS) separates and identifies volatile and semi-volatile organic compounds from a heated solvent extract of the ink. For forensic ink analysis, GC-MS is particularly valuable for quantifying the residual semi-volatile solvents (phenoxyethanol, benzyl alcohol, glycerol esters) in a ballpoint ink stroke, because the concentration and relative ratio of these components change with age and are the basis of solvent-loss ink dating methods. A fresh ballpoint stroke contains high concentrations of phenoxyethanol relative to its non-volatile reference; an aged stroke has lower phenoxyethanol because it has evaporated or migrated into the paper fibres. GC-MS measures this ratio quantitatively.

GC-MS for solvent analysis requires thermal desorption or solvent extraction of a 1 to 5 mm section of ink stroke, which is irreversible. The method is therefore placed late in the analytical workflow, after non-destructive and low-impact methods have been exhausted. The USSS Standard Operating Procedures for ink analysis (declassified portions published in the journal Analytical Chemistry and the Journal of Forensic Sciences) specify a tiered approach: VSC and light microscopy first, then TLC (low-volume extraction from a distal site on the stroke), then HPLC-PDA if TLC is ambiguous, then GC-MS or LC-MS only if the question is specifically about solvent composition or volatile components.

The decision tree for destructive versus non-destructive analysis is institutionally formalised in several jurisdictions. The ENFSI Forensic Document Examination Working Group (FIDE-WG) publishes best-practice guidelines that recommend documenting the rationale for each destructive step in the case file. The Forensic Science Regulator's Codes of Practice and Conduct (UK) require that destruction of evidential material be proportionate and that alternatives be considered first. In India, the CFSL Technical Examination Manual includes similar language about proportionality in evidential sampling.

1. VSC and light microscopy
   Illuminate the document at UV (254 nm, 365 nm), visible band steps (450-700 nm), and near-infrared (700-1000 nm). Photograph all discriminating wavebands. Document stroke overlap, paper disturbance, and any obvious entry differentiation. Decision: do optical findings resolve the question? If yes, stop here.
2. Raman and ATR-FTIR (non-destructive)
   If optical examination reveals multiple inks or suspicious features, apply Raman microscopy at relevant sites. Use 785 nm laser to minimise fluorescence. Follow with ATR-FTIR on the surface of each ink type. Document spectra. Decision: do spectral findings identify the inks and resolve the question? If yes, stop.
3. TLC (low-impact extraction)
   Extract 1-2 microlitres from a 1-2 mm section of each ink to be compared, from a non-critical site (distal stroke, lower margin). Develop TLC plate with validated mobile phase. Photograph under UV and visible light. Compare Rf values against reference library. Decision: do TLC profiles differentiate or match the inks adequately? If TLC is ambiguous, proceed to HPLC.
4. HPLC-PDA (quantitative dye profile)
   Use a fresh 1-2 mm extraction from a different site. Inject onto reversed-phase C18 column. Run validated gradient. Collect retention times and UV-Vis spectra for all peaks. Compare against USSS/BKA database. Decision: are inks identified? Is the question about age? If age question, proceed to GC-MS for solvent profile.
5. GC-MS or LC-MS (structural/quantitative)
   Reserved for questions about solvent composition (dating) or structural identity of novel dye components. Extract 3-5 mm section. Quantify phenoxyethanol and benzyl alcohol by GC-MS against internal standard. Document chain of custody for extracted material. Report residual solvent ratios as input to dating calculation.