Sie sind auf Seite 1von 8

Forensic Chemistry 3 (2017) 2835

Contents lists available at ScienceDirect

Forensic Chemistry
journal homepage:

Characterization and identification of luminescent components in inks

using various analytical techniques for the study of crossed-line
Rhett Williamson a, Daniela Djidrovska b, Andrea Ledic c, Stjepan Brzica c, Vesna Antikj d, Rolf Hofer e,
Jose Almirall a,
International Forensic Research Institute and Department of Chemistry and Biochemistry, Florida International University, Miami, FL, USA
Counterfeit Currency and Security Documents Branch, INTERPOL General Secretariat, Lyon, France
Forensic Science Centre Ivan Vucetic, Zagreb, Croatia
Ministry of the Interior, Forensic Chemistry Department, Ministry of Internal Affairs, Skopje, Macedonia
Forensic Chemistry Section, Forensic Science Institute, Zurich, Switzerland

a r t i c l e i n f o a b s t r a c t

Article history: Several analytical instrumental techniques have been tested and utilized in an effort to determine the
Received 14 October 2016 sequencing in crossed-line intersections (CLI) in order to gain insight into the eventual development of
Received in revised form 15 January 2017 a standardized method to determine temporal sequence of line writing. One important facet of this study
Accepted 25 January 2017
was to determine the chemical identity of the luminescent compounds present in the formulation of inks
Available online 28 January 2017
to better understand the interaction of different inks in crossed-line intersections. This study involved
independent analyses of a number of inks by three laboratories. A combination of Thin Layer
Chromatography (TLC), Gas Chromatography Mass Spectrometry (GC/MS), High Performance Liquid
Ink analysis
Crossed-line intersections
Chromatography (HPLC), Matrix-Assisted Laser Desorption Mass Spectrometry (MALDI-MS), Direct
Luminescent components Analysis in Real Time Mass Spectrometry (DART-MS), and Liquid Chromatography Mass Spectrometry
Questioned document examination (LC-MS) were implemented by the three laboratories in order to characterize the luminescent compo-
nents of inks present in crossed-line intersections. A combination of luminescent compounds including
Crystal Violet and Methyl Violet were characterized and identified to be present in mixtures in the ink
formulations utilizing each of the analytical techniques included in this study. However, the temporal
sequence of deposition of inks present in crossed-line intersections could not be determined. The proto-
col described here allows for the isolation and characterization of luminescent compounds present in the
formulation of inks to varying degrees, and the information presented here can be used in the future
establishment of a standard protocol for the identification of luminescent compounds in inks.
2017 Elsevier B.V. All rights reserved.

1. Introduction the multiple inks interact with one another affect the ability to
determine sequence of deposition [2]. Additional techniques that
In questioned document examinations, it is useful to be able to have been applied to the analysis of line crossings with varying
determine the sequence of deposition of ink and/or markings on a degrees of success including atomic force microscopy, attenuated
substrate. This is most commonly explored in fraud investigations total reflectance Fourier transform infrared spectral imaging, 3D
involving official documents, checks, or other documents of inter- laser profilometry, and scanning electron microscopy energy dis-
est. Though optical examination and electron microscopy have persive X-ray spectroscopy (SEM-EDX) [1,36].
been most widely applied, there is no consensus method to deter- Characterizing the fluorescent components of inks, most com-
mine the temporal sequence of ink deposition [1]. The composition monly dyes, is very useful in document examination for associa-
of each component ink of a crossed-line intersection (CLI) and how tion, discrimination, and intelligence purposes. Optical methods
such as fluorescence, infrared luminescence, and spectroscopic
methods such as Fourier transform infrared spectroscopy and
Corresponding author. UV/Vis spectroscopy rely on these properties, and useful informa-
E-mail address: (J. Almirall). tion is derived [14]. In addition, chromatographic, spectroscopic,
2468-1709/ 2017 Elsevier B.V. All rights reserved.
R. Williamson et al. / Forensic Chemistry 3 (2017) 2835 29

and mass spectrometric techniques such as High Performance Liq-  CRO: Croatia High Performance Liquid Chromatography
uid Chromatography (HPLC), Gas Chromatography (GC), paper (HPLC);
chromatography, nuclear magnetic resonance (NMR) spectroscopy,  USA: United States of America Matrix Assisted Laser Desorp-
and liquid chromatography mass spectrometry (LC-MS), and gas tion IonizationMass Spectrometry (MALDI-MS), Thin Layer
chromatography mass spectrometry (GC/MS) have been applied Chromatography (TLC), Liquid ChromatographyMass Spec-
to analysis of dyes and pigments [7]. Matrix Assisted Laser Desorp- trometry (LC-MS), and Direct Analysis in Real TimeMass Spec-
tionMass Spectrometry (MALDI-MS) is another attractive method trometry (DART-MS).
for ink analysis in that it is considered a soft ionization technique
which characterizes molecular ions with limited fragmentation as 2. Experimental section
compared to traditional electron impact ionization [810]. In
recent years, laser desorption has been applied to the analysis of 2.1. Materials and methods
inks, with success in characterizing both the dye, pigment, and
more limitedly the polymeric content of the inks on paper in com- A total of eight different inks, including two ballpoint pens, two
bination with other techniques such as TLC and Raman spec- fountain pens, three ink pads, and one felt tip pen were analyzed a
troscopy [9,11]. part of this study and their descriptions below and the analysis
After conducting non-destructive optical examinations, the performed on each writing instrument can be seen in Table 1:
chemical analysis of the ink (e.g. ballpoint pen inks, felttip pen
inks, ink pad) is useful for the identification and separation of mix- Writing Instrument A: Ballpoint pen, Stabilo, Ref n 0800M 97 3,
tures of organic compounds, including luminescent components. blue ink
For the past 50 years, optically examining documents and inks Stroke dimension: 0.3 mm
with infrared light sources to detect luminescent properties has Writing Instrument B: Ink pad, Troadat, Ref n 6/4911C, red ink
been well established and readily used for the comparison of inks Stroke dimension: 0.6 mm
[12]. However, the limitations of solely relying on optical examina- Writing Instrument C: Felt-tip pen, Paper Mate Flaire, black ink
tion has been repeatedly reported, with the majority opinion Stroke dimension: 0.7 mm
necessitating that additional examinations and testing be per- Writing Instrument L: Fountain pen, Pelikan Script, black ink
formed before any conclusion is made [13]. Thus, luminescence Stroke dimension: 0.3 mm
observation has been readily combined with other techniques such Writing Instrument M: Fountain pen, Pelikan Script, blue ink
as Thing Layer Chromatography (TLC) to improve the amount of Stroke dimension: 0.3 mm
chemical information obtained [14,15]. The biggest obstacle for Writing Instrument Q: Ink pad, Trodat Printy 4822, red ink
the utility of optical and luminescence examination for inks is that Stroke dimension: 0.6 mm
a majority of inks are mixtures of components, some of which are Writing Instrument Y: Ballpoint pen, Pilot BPA-10F, black ink
luminescent and others that are not, and the non-luminescent Stroke dimension: 0.3 mm
components have a masking effect and therefore inhibit overall Writing Instrument Z: Ink pad, Sachihata HGN-1, blue ink
luminescence [16]. Therefore, it is most useful to employ optical Stroke dimension: 0.6 mm
examination as a preliminary step in the analysis of inks, and to
combine with other techniques that can provide complimentary
information [1719]. 2.2. GC/MS and TLC analysis by the MKD laboratory
The project described here was performed in collaboration with
three [3] laboratories (in Croatia, Macedonia and USA) in an A 10 mm long ink line was cut out and put into a micro-vial
attempt to establish a standard and reliable method for the deter- with 1 ml of methanol around 15 min. Different solvent systems
mination of the sequence in crossed-line intersections. The specific were used to develop chromatograms as follows: ethyl acetate,
objective of this protocol was to identify the compounds contained ethanol, water (26:13:11); 1-butanol, ethanol, water (50:10:15)
within the inks. In particular, there was an interest to identify the and ethanol (100). The TLC plates were activated at 60 C for
luminescent compounds within the inks in order to better under- 20 min. After cooling, small amounts (210 ll) of the extracts were
stand the migration effects of these compounds that were observed spot onto the TLC plate at a distance of about 10 mm from the bot-
at the interfaces of the line-crossing sites. The analysis of multiple tom edge with the aid of a micro-pipette. The TLC plates were
inks by different laboratories has also provided some insight into developed in a horizontal developing chamber. Chromatographic
the chemical composition of the inks as well as demonstrated development of the plates was performed at room temperature
the forensic utility and complementarity of each of the analytical for 30 min. The TLC plate was placed upright into the developing
methods for the chemical analysis of inks. This information will chamber and the solvent system allowed to migrate for about
help to establish a standard protocol that could be used in the 6 cm. The plate was removed from the chamber and dried in cold
determination of sequence in crossed-line intersections in the
future. Table 1
Several chemical methods were used by participating countries Chemical Analysis of each ink independently by laboratory in each country.
including: Thin Layer Chromatography (TLC), Gas Chromatography
Countries Chemical Writing
Mass Spectrometry (GCMS), High Performance Liquid Chromatog- analysis instruments
raphy (HPLC) and Matrix Assisted Laser Desorption Ionization
Ministry of the Interior, Forensic TLC L, M and traces
Mass Spectrometry (MALDI-MS) and Liquid Chromatography Chemistry Department, Skopje, of A, B, C, Q, Y, Z
Mass Spectrometry (LC-MS). Republic of Macedonia (MKD) GC/MS L, M
To identify the luminescent compounds in inks, forensic exam- Forensic Science Centre Ivan Vucetic, HPLC Y, Z and traces
iners from the participating countries were able to use the follow- Zagreb, Croatia (CRO) of A, B, L, M
International Forensic Research MALDI-TOF, Y, Z
ing instrumental techniques:
Institute and Department of DART-MS, TLC,
Chemistry and Biochemistry, LC-MS
 MKD: Republic of Macedonia Thin Layer Chromatography Florida International University,
(TLC), Gas ChromatographyMass Spectrometry (GC/MS); Miami, FL USA
30 R. Williamson et al. / Forensic Chemistry 3 (2017) 2835

air stream. Writing instruments L and M were exposed to three age 65V, and mass range of 501000 amu. Mass Chromatogram,
separate conditions for three days prior to separation by TLC: 1) mass spectra, and data acquisition were achieved using MassHun-
in dark ambient conditions at 25 C, 2) in complete darkness at ter Workstation Software LC/MS Data Acquisition for the 6200 ser-
25 C, and 3) under direct sunlight. ies TOF/6500 series Q-TOF (v B.05.00: Agilent: Santa Clara, CA,
Each of the components of the Fountain black and blue ink sam- USA). Samples were extracted from paper with 0.1 mL methanol
ples, resolved by TLC and observed under IR-Infra red light and (HPLC grade, Thermo Fisher Scientific: Waltham, MA, USA). Sam-
spot light source 440580 nm VSC2000/HR were scraped off the ples analyzed include methanol blank, whatman paper blank, writ-
pure luminescent zone of the plate, and extracted with methanol ing instrument Y TLC extract, and writing instrument Z extract.
in an Eppendorf tube using an ultrasonic bath for 6 min. The DART-MS analyses were conducted using a 6530 Q-TOF-MS
The extract was analyzed by GC/MS and its components were (Agilent, Santa Clara, CA) coupled to a DART SVP (IonSense, Saugus,
identified. GC/MS analyses were conducted using a GC/MS-QP MA) ambient ionization source. The experimental parameters were
2010 Shimadzu (Columbia, MD, USA), software package version as follows: positive ion mode with a DART helium gas at 350C, Q-
CCSS-5000 Equipped with a ZB-5 column (Supelco, Bellefonte, PA, TOF heater gas at 350C, drying gas flow at 0.2 L/min, fragmentor
USA) 30 m, liner diameter 0.25 mm, thick film of stationary phase voltage of 175 V, skimmer voltage 65 V, and a mass range of
0.25 lm, injection temperature 270 C, injection volume 1 lL, split 1001000 amu. A 5 ppm solution of polyethylene glycol 600
ratio 20:1, split flow 19.9 mL/min, pressure-7.32 Psi, controlled lin- (Emerald Biosystems, Bainbridge Island, WA) in 50:50 water:
ear velocity 36 cm/s, total flow 23.8 mL/min, carrier gas He, col- methanol (Thermo Fisher Scientific, Waltham, MA) was used as a
umn flow 1.0 mL/min, Oven initial temperature: 45 C, hold quality control standard and internal mass calibrant. DART ion
3.70 min, Rate 10 C/min, Final temperature: 295 C, hold 10 min, source control was achieved using DART SVP software (v 3.0.x: Ion-
total run time 38.70 min. Sense, Saugus, MA). Mass spectra and data acquisition were
achieved using MassHunter Workstation Software LC/MS Data
2.3. HPLC analysis by the CRO laboratory Acquisition for the 6200 series TOF/6500 series Q-TOF (v B.05.00:
Agilent, Santa Clara, CA). Samples were analyzed directly from
HPLC analyses were conducted using LC 1200 (Agilent: Santa paper substrate. Samples analyzed include paper blank, writing
Clara, CA, USA) with Waters symmetry column, C18 3.5 lm, instrument Y, and writing instrument Z.
4.6  100 mm (Waters: Milford, MA, USA), Tcolumn = 40 C, flow The TLC separation were performed using silica gel plates
rate: 1.0 mL/min, eluent A (ACN acetonitrile), eluent B (H2OTFA), heated at 60 C for 20 min with mobile phase: ethyl acetate/etha-
Program: Start 40% A (ACN), 60% B (H2OTFA), 29 min. 95% A nol/distilled water (70:35:30, v/v/v). Ink was extracted from paper
(ACN), 5% B (H2OTFA), 30 min. 40% A (ACN), 60% B (H2OTFA), using 0.1 mL pyridine and spotted using tube capillary 1.5
injection volume = 20 lL, k1 = 591.0 nm (Ref. 360 100), 1.8  90 mm (Kimble Chase: Vineland, NJ, USA). VSC observations
k2 = 591.0 nm (Ref. 360 100) Run time = 30 min + 3 min stabiliza- were made using the VSC-200 (Foster & Freeman, UK) under spot
tion. Crystal violet in methanol was used as a standard. Ink samples light source (white light) with 440580 nm filter. Samples sepa-
were analyzed by extracting from paper with methanol. Only sam- rated via TLC include writing instrument Y, writing instrument Z,
ples that exhibited luminescent properties after visualization in paper blank, and pyridine blank.
VSC were analyzed via HPLC.

3. Results
2.4. MALDI-TOF-MS, DART-QTOF-MS, TLC, LC-MS analysis by the FIU
3.1. GC/MS and TLC analysis performed at the MKD laboratory
MALDI-MS analyses were conducted on a Voyager DE Pro TOF
Separation of pen inks was achieved, at varying degrees of suc-
mass spectrometer (Applied Biosystems: Carlsbad, CA USA). Sam-
cess, utilizing three different solvent systems. The TLC separation
ple preparation involved cutting a 2  2 mm portion of the ink
of writing instruments M and L was achieved in the 1-butanol,
on whatman 42 paper (Whatman: Maidstone, UK) and attached
ethanol, water (50:20:30) solvent system under spot light source
to a 100 well stainless steel target (JBI Scientific: Huntsville, TX,
(440580 nm filter). The TLC separation of writing instruments A
USA). The matrix used was 9-aminoacridine (9AA: Aldrich: St.
and Y was achieved using the same solvent system, and was visu-
Louis, MO, USA) dissolved in 70% methanol:water (5 mg/mL).
alized under infrared and spot (440580 nm) light sources. The
0.5 lL of the matrix solution was spotted on top of the sample
separation of writing instruments B, Q, and C viewed under spot
on the target. Instrumental parameters were as follows: 337 nm
(440580 nm) light source revealed only one luminescent compo-
Nitrogen laser, 3 ns pulses, 2 Hz, 50 shots per sample, negative
nent of writing instrument B, with little separation overall. The gas
and positive mode, mass range m/z 1001000. Data Analysis was
chromatogram of the luminescent zone of writing instruments L
performed using Data Explorer TM software (v Applied
and M can be seen in Fig. 1 in which four distinct elution peaks
Biosystems, Carlsbad, CA, USA). Samples analyzed by MALDI-MS
were separated at retention times (RT) of 11.075 min,
include writing instrument Y, writing instrument Z, and crossed-
11.102 min, 12.135 min, and 25.77 min. The mass spectra of each
line intersections with writing instrument Y on top, and writing
individual peak in which degradation products of crystal violet
instrument Y on bottom.
were confirmed with the NIST standard reference database.
The LC-MS analyses were conducted using a with a zorbax col-
umn coupled to a 6530 Q-TOF-MS (Agilent: Santa Clara, CA, USA).
Chromatography parameters were as follows: 1.0 lL injection vol- 3.2. HPLC analysis performed at the CRO laboratory
ume, Tcolumn = 40 C, flow rate: 0.400 mL/min, eluent A water,
0.1% 5 mM ammonium formate pH 4.0, formic acid, eluent B ACN Prior to HPLC analysis, all inks were visualized in the VSC under
acetonitrile, 0.1 formic acid, program: start-5 min 95.00% A(water), IR and spot (580725 nm filter) light sources to confirm lumines-
5.00% B (ACN), 25 min 5.00% A (water), 95.00% B (ACN), 30.0 min cence. Invisible migration was observed in writing instruments Y
runtime total. Electrospray ionization and mass spectrometer and Z, with a luminescent area surround the ink on paper, similar
parameters: heater gas at 350 C, drying gas flow at 7 L/min, nebu- to that observed in the USA lab, seen in Fig. 2. The liquid chro-
lizer at 35 psig, sheath gas at 375 C, sheath gas flow at 11 L/min, matogram of the standard crystal violet contained five peaks. The
cap voltage at 4000V, fragmentor voltage at 135 V, skimmer volt- chromatogram for instrument A showed the presence of crystal
R. Williamson et al. / Forensic Chemistry 3 (2017) 2835 31

Fig. 1. Total Ion Gas Chromatogram for fountain pen ink extract exhibiting four distinct peaks which correspond to Crystal Violet degradation products.

Fig. 2. Cross-line intersections of writing instruments Y and Z visualized in VSC exhibiting invisible migration (colorless) under spot light source (480620 nm) (left) color,
(right) black and white image.

violet, with five peaks at the same elution times. The chro- (Figs. 3 and 4). The series of molecular ions present in positive mode
matogram for writing instrument Z did not reveal any useful infor- at m/z 344, 358, and 372 correspond to methyl violet minus a chlo-
mation as to the identity of the compound causing the luminescent rine atom and its degradation products, which preferentially ionize
migration region, but the chromatogram for writing instrument Y in the positive mode [20]. It has also been reported that the m/z val-
contained five peaks with elution times that are consistent with ues at 358 and 344 actually correspond to distinct crystal violet
crystal violet. Therefore, it was found that crystal violet was pre- molecules with less than six methyl groups, with the m/z 358 con-
sent in both writing instrument A and writing instrument Y. taining five methyl groups (commonly referred to as methyl violet)
and the m/z 344 containing four methyl groups (referred to as
3.3. MALDI-TOF-MS, DART-QTOF-MS, TLC, LC-MS analysis performed tetramethyl pararosaniline) [21]. The intact molecule does not exist
at the FIU laboratory in ionized form in the mass spectrum, but the molecule plus a
sodium molecule does form a molecular ion which is present at
A colorless, therefore invisible migration band was observed at m/z 429. The presence of crystal violet was confirmed by analyzing
the interface of the paper and the visible inks when writing an extract of the ink via LC-MS, with the same ions present (Fig. 6).
instruments Y and Z were viewed in the VSC under spot light source In negative mode, the dominant ions present at m/z 712 correspond
(480620 nm filter), as illustrated in Fig. 2. The MALDI-MS spectra to Acid Violet 49, which is also present in the formulation of the pen
obtained of writing instrument Y contains molecular ions that cor- ink, and preferentially ionizes in the negative mode. Acid Violet 49
respond to multiple dyes present in the formulation of the ink is also present as a mixture with other dyes that are analogues of
32 R. Williamson et al. / Forensic Chemistry 3 (2017) 2835

Fig. 3. Positive mode MALDI-MS spectrum of pen ink Instrument Y containing Crystal Violet.

Fig. 4. Negative mode MALDI-MS spectrum of pen ink Instrument Y containing Acid Violet 49.

the molecule, substituted with additional sulfonated benzene In the positive ion MALDI-MS spectra obtained of writing
groups [21]. The ion at m/z 925 refers to an Acid Violet 49 analogue instrument Z, the prominent ionic species present at m/z 575 cor-
(structure seen in Fig. 4) and is present in all replicate measure- responds to copper phthalocyanine. In the negative mode, the ions
ments of the pen ink in negative mode. present at m/z 653 and 733 correspond to sulfonated copper
R. Williamson et al. / Forensic Chemistry 3 (2017) 2835 33

phthalocyanine species, which are ionized preferentially in the the blank TLC plate. DART-MS analysis of writing instrument Z also
negative mode, due to the anionic nature of the molecules. These showed the presence of polymeric content. The LC-MS of writing
molecular ions give insight into the presence of three distinct cop- instrument Z corroborated the presence of the liquid polymeric
per phthalocyanine pigments used in this ink formulation, one vehicle component.
disulfonated pigment, one monosulfonated pigment, and one with
no substitutions. Manufacturers often include these sulfonated
pigments to increase solubility and thus can be used as a point of 4. Discussion and conclusion
chemical characterization [11].
The positive ion mode spectra of the crossed-line intersections Table 2 contains a list of the chemical components of interest
(both with pen on bottom and on top of stamp ink) contained that were identified by each of the analytical techniques employed
molecular ions corresponding to both writing instrument Y and by the laboratories in this study.
writing instrument Z, with the series of m/z 429, 372, 358, 344 By placing the TLC plates in VSC 2000 under IR (Infra red light)
originating from the methyl violet present in the pen ink instru- and a spot light source, the luminescent components in ink sam-
ment Y, and the peak at m/z 575 originating from the copper ples were visualized. The most intense colours of the spots were
phthalocyanine present in the stamp ink instrument Z (Fig. 5). developed in the solvent system: ethyl acetate, ethanol, water
The negative ion mode spectra of the crossed-line intersections and plates activated at 60 C for 20 min. This solvent system was
(Fig. 5) also contained molecular ions corresponding to both inks, utilized by two separate labs and produced the best separation of
with the molecular ions at m/z 925 and 712 originating from the the inks into luminescent components. Reliable results were not
acid violet 49 and analogue present in the pen ink writing instru- obtained when the mobile phase consisted of only ethanol.
ment Y and the peak at 575 originating from the copper phthalo- Dyes are aromatic, organic components that absorb electromag-
cyanine present in the stamp ink writing instrument Z. netic radiation including visible light with wavelengths 350
The TLC analyses of writing instrument Y revealed the presence 700 nm. Fluorescence generally refers to re emission in the visible
three fluorescent bands visible under spot light source and filter of region of the spectrum 400 nm700 nm, UV Fluorescence or UV
400480 nm. The TLC of instrument Z did not reveal any separation excited visible fluorescence is referred to in ink analysis as infrared
or luminescent visualization. It was confirmed with LC-MS analy- luminescence or visible excited infrared luminescence. The prop-
ses that the fluorescent component of the pen ink is crystal violet, erties of these molecules allows for visualization in the VSC under
methyl violet, and tetramethyl pararosaniline (Fig. 6). the spot light source.
The DART-MS results did not characterize any colorant mole- An invisible migration band may be the result of and/or influ-
cules, though revealed the presence of both polypropylene glycol enced by the chemistry of the paper (such as additives), the chem-
(PPG) and polyethylene glycol (PEG) polymeric fragments in the istry of the inks (such as solvents), the chemistry of the
pen ink. The direct DART-MS analysis of the luminescent bands luminescent agent, and the stability of the components (solvent
for instrument Y did not contain any ions that did not come from evaporation rate). It was shown that the components present in

Fig. 5. Positive mode (top) and negative mode (bottom) MALDI-MS spectra of crossed-line intersections containing pen and stamp ink, illustrating diagnostic ions
corresponding to both inks when present in crossed-line intersections.
34 R. Williamson et al. / Forensic Chemistry 3 (2017) 2835

Fig. 6. Liquid chromatogram and Positive mode LC-MS spectrum of writing instrument Y pen ink exhibiting three distinct molecules: tetramethyl p-rosaniline, methyl violet
and crystal violet.

Table 2 violet naturally demethylates over time, thus a solution of crystal

Summary of components of interest identified in inks via each of the corresponding violet or an ink containing crystal violet will always have a mixture
analytical techniques.
of dyes present [20]. The HPLC of the crystal violet standard shows
Countries/Labs Chemical Components of interest multiple peaks, and these peaks can be associated to the demethy-
analysis identified lated derivatives of crystal violet. The LC-MS data confirms this, as
Ministry of the Interior, GC/MS, TLC Crystal violet, degradation three distinct peaks were resolved, and the mass spectra of each of
Forensic Chemistry products those peaks correspond to crystal violet, methyl violet (one less
Department, Skopje,
methyl group), and trimethyl pararosaniline (two less methyl
Republic of Macedonia
Forensic Science Centre Ivan HPLC Crystal violet groups) [22]. The three visible bands in the VSC of writing instru-
Vucetic, Zagreb, Croatia ment Y also confirm the presence of three distinct luminescent
International Forensic MALDI-TOF, Crystal violet, demethylated compounds. The VSC of writing instrument Y also shows a non-
Research Institute and DART-MS, derivatives (methyl violet and luminescent band. This corresponds to a black, non-luminescent
Department of Chemistry TLC, LC-MS tetramethyl pararosaniline);
and Biochemistry, Florida acid violet 49 and analogue,
dye that is present in the formulation of the ink. The inclusion of
International University, copper phthalocyanine and a non-luminescent black dye has been reported in the formulation
Miami, FL USA sulfonated analogues; of black pen inks, similar to writing instrument Y [23,24] For that
polypropylene glycol, reason, the ink itself does not seem to luminesce, but when it is
polyethylene glycol
separated into components via TLC, the luminescent dyes are no
longer inhibited by the non-luminescent black dye and lumines-
cence is observed.
this invisible migration region are crystal violet and demethylated When inks interact in crossed-line intersections, invisible
derivatives, as exhibited and corroborated by HPLC, MALDI-MS, migration may occur, as illustrated in Fig. 2. This migration is a
and LC-MS analyses. Two separate labs were able to confirm, inde- result of the interaction of the liquid inks with one another, caus-
pendently, that the colorless component of the ink, in this case ing a solvation effect. This causes the luminescent compounds, in
writing instrument Y, contained crystal violet and degradation pro- this case crystal violet and derivatives, to leach out of the ink mark-
ductions, which are luminescent compounds. The dyes themselves ing. Once this occurs, the quenching effect of the non-luminescent
are not, in fact, colorless, but because they exist at such low con- black dye is inhibited, and the crystal violet dyes are able to lumi-
centrations, they are not visible to the naked eye. nesce. This seems like an invisible migration, but in fact, the con-
Identification of these luminescent compounds was achieved centration of the violet dyes is so low that the color is not visible
via HPLC analysis of the ink extract and confirmation with a crystal to the naked eye. That being said, the identification of the lumines-
violet standard, and separately, LC-MS analysis of the luminescent cent compounds in these crossed-line intersections does not give
bands of the ink, which produced mass spectra corresponding to insight as to the sequence of deposition of the inks, as this migra-
crystal violet and demethylated derivatives. These compounds tory effect occurs regardless of which ink is deposited first.
are demethylated derivatives of the triaryl methane dye crystal While HPLC by itself provides minimally useful information
violet, which are methyl violet and trimethyl pararosaniline. The unless a comprehensive set of chemical standards representative
dyes are known to be present in a mixture in the formulation of of ink substituents is available, LC-MS is a preferred method which
writing inks [20,22]. Additionally, it has been reported that crystal allows for both separation of ink components and confirmation
R. Williamson et al. / Forensic Chemistry 3 (2017) 2835 35

with mass spectrometric analyses. The HPLC analyses conducted as [3] Y. Wang, B. Li, Determination of the sequence of intersecting lines from laser
toner and seal ink by Fourier transform infrared microspectroscopy and
a part of this study, however, provided complimentary results to
scanning electron microscope/energy dispersive X-ray mapping, Sci. Justice 52
the other analytical techniques and allowed for the identification (2) (2012) 112118.
of crystal violet and derivatives in the ink samples. [4] G.S. Spagnolo, Potentiality of 3D laser profilometry to determine the sequence
MALDI-MS is an effective tool for the characterization of inks of homogenous crossing lines on questioned documents, Forensic Sci. Int. 164
(23) (2006) 102109.
directly on paper and it could be minimally destructive to the sam- [5] K. Bojko, C. Roux, B.J. Reedy, An examination of the sequence of intersecting
ple. Positive and negative modes of MALDI can provide compli- lines using attenuated total reflectance-fourier transform infrared spectral
mentary chemical information. The addition of a MALDI matrix imaging, J Forensic Sci. 53 (6) (2008) 14581467.
[6] R. Corzo, K. Subedi, T. Trejos, J.R. Almirall, Evaluation of the forensic utility of
improves the spectral quality and reduces the background. Inks scanning electron microscopy-energy dispersive spectroscopy and laser
in layers/crossed-line intersections can be characterized using ablation-inductively coupled plasma-mass spectrometry for printing ink
MALDI-MS, allowing for the identification of the separate inks in examinations, J Forensic Sci. 61 (3) (2016) 725734.
[7] K. Venkataraman, The Analytical Chemistry of Synthetic Dyes, Wiley, New
a mixture but the layer/line sequence cannot be determined using York, 1977.
MALDI-MS. MALDI-MS, as compared to the other combination of [8] B. Matthews, G.S. Walker, H. Kobus, P. Pigou, C. Bird, G. Smith, The analysis of
chromatographic and mass spectrometric techniques included in dyes in ball point pen inks on single paper fibres using laser desorption
ionisation time of flight mass spectrometry (LDI-TOFMS), Forensic Sci. Int. 209
this study, allows for the instantaneous characterization of multi- (13) (2011) 26.
ple components present in ink, such as the crystal violet, acid violet [9] L. Heudt, D. Debois, T.A. Zimmerman, L. Kohler, F. Bano, F. Partouche, et al.,
49, and copper phthalocyanine molecules identified in this study. Raman spectroscopy and laser desorption mass spectrometry for minimal
destructive forensic analysis of black and color inkjet printed documents,
Separation of ink components via TLC prior to further analysis
Forensic Sci. Int. 219 (13) (2012) 6475.
by GC/MS, HPLC, and/or LC-MS allows for isolation and confirma- [10] J. Allison, Ink Analysis Using UV Laser Desorption Mass Spectrometry, in: R.D.
tory characterization of the luminescent component(s) of inks, use- Blackledge (Ed.), Wiley & Sons, 2007.
ful in document examination and intelligence purposes. GC/MS is [11] S. Donnelly, J.E. Marrero, T. Cornell, K. Fowler, J. Allison, Analysis of pigmented
inkjet printer inks and printed documents by laser desorption/mass
useful in determining the degradation products of luminescent spectrometry, J Forensic Sci. 55 (1) (2010) 129135.
compounds. HPLC can confirm the presence of a compound if a ref- [12] R.L. Brunelle, M.J. Pro, A systematic approach to ink identification, J. Assoc. Off.
erence standard is available. LC-MS also allows for identification of Anal. Chem. 55 (4) (1972) 823826.
[13] H. Chen, H. Meng, K. Cheng, A survey of methods used for the identification
luminescent compounds and dyes present in ink formulations. and characterization of inks, Forensic Sci. Int. 1 (2002) 114.
DART-MS is useful when characterizing the semi-volatile solvent [14] T.W. Sinor, J.P. Wilde, K.E. Everse, E.R. Menzel, Lasers and optical spectroscopy
and/or polymeric vehicle components of inks. MALDI-MS is a use- in questioned document examination, J Forensic Sci. 31 (3) (1986) 825839.
[15] A.A. Cantu, Comments on the accelerated aging of ink, J Forensic Sci. (1988).
ful technique for the rapid analysis of inks directly on paper that [16] C.A. Sensi, A.A. Cantu, Infrared luminescence: is it a valid method to
allows for instantaneous characterization of multiple colorant differentiate among inks?, J Forensic Sci. 27 (1) (1982) 196199.
and/or luminescent components that may be present in inks. Over- [17] A.A. Cantu, Analytical methods for detecting fraudulent documents, Anal
Chem. 63 (17) (1991) 854A.
all, this study demonstrates the utility of a protocol involving VSC [18] R. Williamson, A. Raeva, J.R. Almirall, Characterization of printing inks using
observation, TLC separation, and subsequent analysis by GC/MS, DART-Q-TOF-MS and attenuated total reflectance (ATR) FTIR, J. Forensic Sci. 61
HPLC, LC-MS, DART-MS, MALDI-MS and/or a combination of these (3) (2016) 706714.
[19] T. Trejos, P. Torrione, R. Corzo, A. Raeva, K. Subedi, R. Williamson, et al., A novel
methods for the identification and isolation of luminescent com-
forensic tool for the characterization and comparison of printing ink evidence:
pounds present in the composition of inks for implementation in development and evaluation of a searchable database using data fusion of
document examination. spectrochemical methods, J. Forensic Sci. 61 (3) (2016) 715724.
[20] C. Weyermann, R. Marquis, W. Mazzella, B. Spengler, Differentiation of blue
ballpoint pen inks by laser desorption ionization mass spectrometry and high-
Acknowledgements performance thin-layer chromatography, J. Forensic Sci. 52 (1) (2007) 216
Dr. Antonio Cant is acknowledged for thoughtful input and [21] J.D. Dunn, J. Allison, The detection of multiply charged dyes using matrix-
assisted laser desorption/ionization mass spectrometry for the forensic
guidance in the preparation of this manuscript. Dr. Floyd Russell examination of pen ink dyes directly from paper, J. Forensic Sci. 52 (5)
from Florida Atlantic University is acknowledged for use of the (2007) 12051211.
MALDI-TOF-MS. [22] J. Andrasko, HPLC analysis of ballpoint pen inks stored at different light
conditions, J. Forensic Sci. 46 (1) (2001) 2130.
[23] Buell BG, inventor; American Cyanamid Company, assignee, Black dyes for
References ball-point pen inks, USA patent US4224071, 1980 September 23.
[24] Eckart Mathias, inventor; Grace WR, assignee, Ink composition containing
[1] S. Kasas, A. Khanmy-Vital, G. Dietler, Examination of line crossings by atomic fluorescent component and method of tagging articles within, USA patent
force microscopy, Forensic Sci. Int. 119 (2001) 290298. US5135569, 1992 August 4.
[2] N. Ozbek, A. Braz, M. Lpez-Lpez, C. Garca-Ruiz, A study to visualize and
determine the sequencing of intersecting ink lines, Forensic Sci. Int. 234 (C)
(2014) 3944.