foram®3:
Discriminating Inks and Toners

Using Raman Spectroscopy to Discriminate Blue Gel Pens

Gel pens are increasingly being used in by the general public in preference to more traditional ballpoint and liquid ink pens. Gel pens have presented new challenges to document examiners; primarily because many of them are pigment based and cannot be easily extracted for analysis by thin layer chromatography (TLC). It is, therefore, useful to find other methods of analysis for these pens.

There have been several scientific studies published using Raman spectroscopy and other methods to discriminate Gel Pens. Mazella and Buzzini[1] used Raman Spectroscopy at 2 different wavelengths to give a discrimination rate of 68% for pigmented blue gel pens. Zieba-Pulus[2] et al utilized a combined Raman/µXRF instrument to analyze a range of materials of forensic interest including blue Gel Pens.

Here we present a brief study applying Raman spectroscopy to discriminate blue gel pens, using 3 different wavelengths of laser excitation; namely 532 nm, 685 nm and 785 nm

foram ink
foram ink

The inks used in this study are listed in Table 1 below. 3 spectra of each ink were recorded; each spectrum was an average of 6 readings. Integration times ranged from 0.1 to 30 seconds.

All spectra were recorded using the FORAM Raman spectrometer (Foster and Freeman Ltd, UK), running at 532, 685 and 785 nm. The lowest power setting was used for all spectra. This was 0.5 mW for 532 and 685 nm, and 0.28 mW for 785 nm.

Spectra were baseline corrected using a propriety (Foster and Freeman) fluorescence filter.

Pen NamePigment/Dye basedPen NamePigment/Dye-based
Faber CastellUnknownPaper Mate Star Gels GripUnknown
Pentel Hybrid K230PigmentPentel Hybrid KN706Pigment
Pilot G-1 0.7DyePilot G-2 10Dye
Pilot P-500PigmentStabilo PointViscoUnknown
Uniball Jetstream SX-210PigmentUniball Signo UM-120Pigment
Zebra JimniePigmentZebra J-Roller RXPigment
Zebra SarasaPigment

Table 1 Blue Gel Pens used in this study.

Figure 1 shows typical spectra obtained with the gel pens at 532 nm.

The spectra all show clear differences.  Visually comparing pairs of spectra yielded a discrimination rate of 73.1%.

Comparing pairs of spectra recorded at 685 nm and 785 nm gave discrimination rates of 70.4 % and 71.8% respectively. 

One general observation of the 685 nm and 785 nm data, was the fact that the signals were quite a lot weaker than the 532 nm data – especially the 785 nm data.  This is expected due to the variation of absolute Raman intensity with wavelength, which is proportional to (wavelength)-4.  In addition, the 685 nm data had quite a high fluorescent background compared to the 785 nm data.

foram 3 GelPens1
Figure 1: Raman spectra of Blue Gel pens recorded at 532 nm

Spectral pairs that were not discriminated by 532 nm, were then checked with the 685 nm and 785 nm pair results, to see if any further discrimination was possible by use of additional wavelengths. 

An example of such a pair is shown in Figure 2.  There is a peak at 1392 cm-1 in the spectrum of the Zebra Jimnie, which is absent in the spectrum of the Pilot P500 ink.  The spectra of these 2 inks at 532 nm only showed minor differences.

Combining the data from all three wavelengths, yielded a total discrimination rate of 75.6%.

foram3 GelPens2
Figure 2: Raman spectra of Blue Gel pens recorded at 685 nm.

We have demonstrated the ability to discriminate between different types of Blue Gel Pens, using 3 different wavelengths of Raman spectrometer.  All three wavelengths yielded a discrimination rate in excess of 70 %.  Combining the data from all 3 wavelengths gave a discrimination rate of 75.6 %.

[1] W.D. Mazzella and P.Buzzini, Forensic Science International 152 (2005), 241-247.

[2] J. Zieba-Palus, R.Borusiewicz and M.Kunicki, Forensic Science International 175 (2008), 1-10.

Using Surface-Enhanced Resonance Raman Scattering (SERRS) to Discriminate Inkjet Inks

Due to the ready availability and cheap cost of inkjet printers, printed documents produced by these machines are frequently encountered by forensic document examiners. Conventional techniques such as visible/IR absorption, so useful in ink examination are not as effective in the examination of printed documents produced by inkjet printers. Other techniques such as chromatography involve the destruction of a small portion of the document.

Littleford et al[4] have used SERRS Spectroscopy to the probe the structural changes of the chromophore present in black inkjet inks when deposited onto paper. They also give examples of the types of dye that are likely to be found in inkjet inks.

Here we present a brief study applying SERRS spectroscopy to the discrimination of black inkjet inks, and determine its potential use in the discrimination of inks produced by different manufacturers.

foram inkjets
foram inkjets

The inkjet samples used in this study are listed in Table 1 below.

All spectra were recorded using the FORAM Raman spectrometer (Foster and Freeman Ltd, UK), running at 685 nm. Each spectrum was run on full power (~5 mW). Each spectrum consisted of 6 averaged scans. The integration time for each scan was 1 minute or 30 seconds depending on the fluorescence of the sample. Poly-L-lysine (Sigma-Aldrich) and gold colloid (British Biocell) were applied to a small portion of the document as previously described [5].

All Spectra were baseline corrected using a propriety (Foster and Freeman) fluorescence removal filter.

Inkjet Printer
Canon M610Canon Pixma MP520Epson Stylu C64Epson Stylus C66
Epson Stylus DX7450HP Business Inkjet 1200HP C4180HP Deskjet 820 CXi
HP Deskjet 940CHP Deskjet 1220CHP Deskjet 6122HP Deskjet F2180
HP Photosmart 7150Lemark 4300Lexmark Black Ink #70
Uniball Jetstream SX-210PigmentUniball Signo UM-120Pigment

Table 1 Inkjet inks used in this study.

Figure 1 shows a range of SERRS spectra of the samples used in this study.   

The spectra all show clear visual differences – these are highlighted by the red asterisks. Pair-wise comparison of all the samples yielded a discrimination power of 84%.

Brunelle and Crawford[6] comprehensively describe the various types of dyes, solvents, dye complexing agents and surfactants typically found in inkjet inks. The dye component, which is the component expected to give rise to the spectra shown above, is frequently an azo dye with a very broad visible absorption profile. The differences between the dyes are often due to modification or addition of side chain groups [4] to improve properties such as light fastness or solubility. Although it has not been possible to identify the dyes giving rise to the different spectra shown in figure 1, the small spectral differences observed are consistent with the assertion that the dye molecules have a similar basic molecular structure, but have different side chain groups. Further work is needed to prove this assertion.

inkjet1
Figure 1. Raman spectra of inkjet inks using SERRS spectroscopy recorded at 685nm 

It has been shown that using Raman spectroscopy at 685 nm it is possible to discriminate toners with the application of SERRS spectroscopy.  It would be interesting to try to definitively identify the inkjet components which are giving rise to the different spectra.

[1] T.Andermann, Problems of Forensic Sciences, Volume 46 (2001), Pages 335 – 344.
[2] W.D. Mazzella and P.Buzzini, Forensic Science International 152 (2005), 241-247.
[3] M.Kunicki, Problems of Forensic Science, Volume 51 (2002), Pages 56 – 70.
[4] R.E.Littleford, M.P.Hughes, G.Dent, D.Tackley and W.E.Smith, Applied Spectroscopy, Volume 57, Number 8 (2003), pages 977 – 983.
[5] E.Wagner and S.Clement, Problems of Forensic Sciences, Volume 46 (2001), Pages 437 – 441.
[6] Advances in the forensic analysis and dating of inks, R.L.Brunelle and K.R.Crawford, published by “Charles C Thomas”, Springfield, Illinois 2003. Pages 41 – 44.

Using Raman Spectroscopy to Discriminate LaserJet and
Photocopy Toners

LaserJet and Photocopy toner are some of the more challenging materials which the document examiner is asked to examine. Conventional techniques such as visible/IR absorption, so useful in ink examination are not applicable to toners. Techniques which are used such as FTIR are either quite destructive to the document or are time-consuming and expensive to carry out.

Merrill et al[1] comprehensively describe the various FTIR techniques as applied to toners.

Here we present a brief study applying Raman spectroscopy to the discrimination of toners. Firstly we attempt to discriminate toner in-situ on the document.

toner
toner

The toner samples used in this study are listed in Table 1 below.

All spectra were recorded using the FORAM Raman spectrometer (Foster and Freeman Ltd, UK), running at 685 nm. Each spectrum was run on full power (~5 mW). 3 spectra were recorded per sample, each consisting of 6 averaged scans. The integration time for each scan was 1 minute.

For the extraction work, a small portion of the document ~5 mm2, was dropped into 2 ml of acetone (Chromasolv Plus, Sigma Aldrich 650501-1L) and left for several hours. Approximately 0.3 ml was applied to an aluminum foil coated slide and allowed to evaporate. Spectra of the remaining residue were recorded.

All Spectra were baseline corrected using a propriety (Foster and Freeman) fluorescence filter.

Toner NameColour on ExtractionToner NameColour on Extraction
HP 98XColourlessKodak 235Black
HP Color LaserJet 4600 PCL6ColourlessKonica 3340Black
HP Color LaserJet 9500 PCL6ColourlessLanier M6765 6755Grey
HP LaserJet 6PColourlessMinolta CF 900Yellow
HP LaserJet 1022GreyMita DC1860Black
HP LaserJet 1160GreyOcé 3165 BCColourless
HP LaserJet 1200GreyOcé im3511Blue
HP LaserJet 2300ColourlessOKI Data 3200nColourless
HP LaserJet 4300tnGreyRicoh 7670Colourless
HP LaserJet P3005 PCL6ColourlessSharp AR-M276Grey
Sharp SF 7800Colourless
Xerox Doc Centre 20Colourless
Xerox Vivace 330Colourless

Table 1 toners used in this study.

In-situ Examination
Figure 1 shows typical spectra of a range of toners.

The spectra all show clear differences. In an attempt to identify which components were giving rise to the observed peaks, a thorough literature and patent search was conducted on the composition of toners. Toners consist of the mixture of components, which may include a fusible resin, Fe3O4 carbon black, dyes/pigments, surfactants and charge control agents[2]. Typical resins include styrene/butadiene copolymer, polyester, styrene ethylhexylacrylate, Styrene n-butylacrylate, and various other copolymers.

The colour of the toner is often modified by the addition of dyes such as nigrosine, victoria blue, methyl violet, pthalocyanines, azo-pigments, and quinacridones. The charge control agent are often complex organometallic compounds, which also act as dyes, or quaternary ammonium salts (both aromatic and aliphatic) [3].

toner1
Figure 1 Raman spectra of toners recorded at 685 nm

We recorded spectra of range of these components. Some of these are shown in Figure 2.

Many of the other components either fluoresced or yielded no spectrum. It is interesting to compare the component spectra with those of the toners in figure 1. The Ricoh 7670 spectrum correlates well with the spectrum of amorphous carbon. However, it is surprising that no resin peaks are observed, given that the resin can be as much as 60% of the toner. It is likely that the 668 cm-1 in the spectrum of HP 6P LaserJet arise from Magnetite[4]. Currently, we have been unable to identify which toner components are giving rise to the prominent peaks in the spectra on Minolta CF900 and Lanier 6765 6755.

Comparing pairs of spectra the discriminating power was found to be 72 %.Examination of extracted residue

The aim of extraction of acetone soluble components was to concentrate those components which are expected to be the resins and dyes and remove any interference from the insoluble components like carbon black, Fe3O4.

toner2
Figure 2 Raman spectra of toner components recorded at 685 nm

Figure 3 shows spectra of the toner extracts.

In the case of the Sharp SF800 toner extract it is possible to observe peaks attributable to styrene, which presumably is part of a styrene containing copolymer. The components giving rise to the other Raman spectra shown in Figure 3 are yet to be identified. Visually comparing pairs of spectra gave a discrimination power of 84 %.

toner3
Figure 3 Raman spectra of toner extract recorded at 685 nm.

It has been shown that using Raman spectroscopy at 685 nm it is possible to discriminate toners both in situ on the document, and extracted into acetone.  The extracts yielded discrimination rate of 84 %, whilst in-situ on the document this was 72%. As Raman spectroscopy is capable of probing very tiny amounts of sample, the method for extracting and concentrating the toner extracts could be optimized further, so that a very tiny amount of the document is used

[1] R.A.Merrill, E.G.Bartick and W.D.Mazella, Journal of Forensic Sciences, Vol.41, No.2 March 1996.
[2] Advances in the forensic analysis and dating of inks, R.L.Brunelle and K.R.Crawford, published by “Charles C Thomas”, Springfield, Illinois 2003, Pages 41 – 44.
[3] Electro-photography and Development Physics, L.B.Schein, published by Laplacian Press 1996. Pages 85 – 87.
[4] A.Zoppi, C.Lofrumento, E.M.Castellucci and M.G.Migliorini, Spectroscopy Europe 14/5 (2002), Pages 16 – 20.