Blue and Black Ballpoint Pen Inks: a Systematic Review for Ink Characterization and Dating Analysis

. Documents are frequent targets of fraud and adulteration. Sometimes it is not enough to determine the document’s authenticity or authorship, but it is also necessary to demonstrate when the document was signed or written. Determining the age of a document, also known as ink dating analysis, is still a challenge for the forensic examiners. There are two main approaches for the dating of ballpoint pen inks: the static method, which is based on the characterization and identification of the ink composition and comparison to a reference collection, and the dynamic method, which involves the study of ink’s processes and alterations that occur with time, considering environmental aging factors such as light and humidity. This article aims to provide a comprehensive systematic review of the studies regarding ballpoint ink characterization and dating, in the last ten years. There are several methods, destructive and non-destructive, which are capable of characterizing and differentiating blue and black ballpoint pens that have shown applicability to the static approach. Regarding the dynamic dating methods, most studies quantified the loss of 2-phenoxyethanol (2-PE) solvent over time


Introduction
The Forensic Documentoscopy aims to identify information regarding the document history, as well as its authenticity and the detection of a variety of alterations. With respect to the graphoscopic confrontation, sometimes it is not enough to determine only the document's authenticity and authorship, but it is also necessary to demonstrate the potential age of a document. Pen ink dating analysis can be instrumental in those instances, but still a challenging task for forensic scientists due to the variety of factors that can influence the analysis [1][2][3] . A document's dating analysis involves the chemical, optical and physical examination of ink and paper. This review will focus on approaches for ink analysis.
There are two major groups of writing instruments: the ballpoint pens, containing oil-based inks, and the non-ballpoint pens, which are water-based inks.
Ballpoint pens are composed of dyes, vehicles, resins, and additives 1,4 . Dyes are responsible for imparting color to the ink. The vehicles are the substances responsible for the homogenization of the ink's compounds, primarily solvents. The resins confer viscosity, paper adherence, lubricant properties and durability to the ink. Finally, additives are compounds with specific characteristics, improving the ink's performance 1,4 .
The ink dating of ballpoint pen inks is performed by two main approaches: the static method and the dynamic method 1,2,5 . The static profile is based on Given the high judicial demands involving documents' age, this article aims to review the leading scientific publications regarding ballpoint ink characterization and dating, in the last ten years. This data can help forensic experts to improve their scientific reports, as well as can support new research in this area.

Methods
This systematic review was performed in August 2018 on the following databases: Google Scholar, Pubmed, Science Direct and Science.gov. The strategy for this research is presented in Figure 1. The descriptor used was "ballpoint pen inks". The inclusion criteria were: a) papers that studied blue and black ballpoint pen inks, for characterization or dating analysis; b) articles published in the last 10 years. Those criteria were chosen to review the most recent methods used for the most common pens used in all kind of documents.
The initial research identified 112 non-redundant papers publications, and 63 of those fit the inclusion criteria.

Ink characterization
Over the past few years, different methods have been developed for the ballpoint pen inks characterization regarding mainly the dyes and additives of the inks. Table 1 shows 49 recent published papers in this area  , in chronologic order. For some methods, it is necessary to cut off a piece of the paper containing the ballpoint pen ink and proceed with an extraction, and the most used solvent is methanol. Those are destructive methods and there were required for the following techniques:  18,55 , LDI-MS (Laser Desorption/Ionization Mass Spectrometry) 53 , UV-Vis (UV-Vis Spectroscopy) 17,31,[40][41][42]48,50,58,61 and Orbitrap Mass Analyzer 15,20 . The non-destructive methods, when the ink is analyzed directly from the paper, include the Luminescence Spectrometry 60 , ToF-SIMS (Time-of-Flight Secondary Ion Mass Spectrometry) 54 63 . PCA consists in a data matrix transformation, in order to represent the largest variations of the data through a small number of factors. HCA is a useful analysis to determine the object resemblance and to identify similar and non-similar samples 63 .

different classes of blue pens
Extraction from the paper with methanol Capillary electrophoresis "A 100% of discrimination was achieved between pen technologies, brands, and models, although non-reproducible zones in the electropherograms were found for blue gel pen samples. The two different batches of blue oil-based pens were also differentiated" The results showed that PLS-DA presented a high discrimination power when associated with the spectra obtained by the VSC®6000. The method enables the correct discrimination of black inks of different pen types, brands and pen models" UV-Vis-NIR iv / FTIR iii / X-ray fluorescence spectroscopy/ TLC xv "After the author knowledge, this is the first time when a comparative trichromatic analysis of some colored samples registered such as and for their extracts has been applied in order to assess the contributions of pigments and organic dyes used as chromophore agents. This original approach seems to be very efficient in discriminating various brands of ballpoint pen inks when it is sustained by complementary data provided by infrared spectroscopy, thin layer chromatography and elemental analysis performed by X-ray fluorescence spectroscopy" Direct analysis of the ink line on paper FTIR iii / Chemometrics "Infrared Spectroscopy associated with linear discriminant analysis, using SPA, GA and SW algorithms for variable selection were successful employed to classify blue ink pens by type (gel, rollerball and ballpoint) and by brands" Direct analysis of the ink line on paper ToF-SIMS xviii "Minimal sample preparation and analysis time, the simultaneous acquisition of organics and metals, and ability to analyze trace amounts gives this technique advantages over others currently utilized in the forensic field. It was indicated that pens from the same manufacturer, but discrete batches, can be significantly different" "TOF-SIMS can be used for the analysis of ballpoint pen inks producing mass spectra that were highly characteristic of the constituent dyes and inorganic substances used in their formulations. TOF-SIMS was also shown to be capable of analyzing ink containing a mixture of dyes, initially separated by thin-layer chromatography, directly on the chromatographic material" "The application to inks was evaluated on a pair-wise basis, through examination of both calibrated and handwritten ink-lines. In the former case, excellent discrimination was obtained with a discriminating power across this set of pens of 78% at the 99% confidence level and 87% at 95% confidence.
For written samples, discrimination was less good, though still better than filtered light, with discriminating powers of 60% and 64% respectively. This difference is attributed to poorer spectral quality and increased sample inhomogeneity when dealing with ink-line writing" Adam, 2008 60

black ballpoint pens of different brands
Ink line on paper extracted with ethanol UV-Vis xii / HPLC x / Chemometrics "For the UV-vis spectroscopy data, the interpretation of the loadings for the first few principal components showed that both the pen inks and the extracted ink lines may be classified in an objective manner and in agreement with the results of parallel thin layer chromatography studies"

Ink dating
Despite the static dating method, which is related to ink characterization, there are two possibilities for ink dating, considering the inks' components transformations with time 3,4 : (a) Relative dating methods, when there is a suspicion that the document was partially forged at a different date. In this case, the document must contain two questionable inks and the inks' components concentration are measured and compared. For both relative and dynamic dating methods, it is necessary to have methods that are capable to detect the ink changes over time. Table 2 shows the most recent studies involving those methods [64][65][66][67][68][69][70][71][72][73][74][75][76][77][78]    The results showed that all ageing parameters were significantly influenced by the ink composition and a wide range of values were obtained for each parameter over the studied time range. The ageing parameters using PE and artificial ageing in their calculation (PE quantity, R%, R%* and RNORM) presented a descending trend over the whole time range, demonstrating their potential to date a document created a few weeks to a few months before the analyses. The PE quantity and RNORM ageing curves distributions were highly correlated and influenced by inks containing a high concentration of PE. Thus, these two parameters would mainly work to date younger inks up to a few weeks. From the two, RNORM showed the strongest differences between young and old samples and is thus more promising to estimate the age of ink entries of unknown origin" Koenig et al., 2018 65 Total ink -48 samples of a commercial ink (Inoxcrom®) Extract of the ink from the cartridge and then spread on white paper using a different brush for each ink UV-vis-NIR ii / Chemometrics "UV-vis-NIR spectrometry combined with Partial Least-Squares (PLS) method for ink dating reported in this paper named DATUVINK demonstrates that diffuse reflectance UV-vis-NIR spectroscopy combined with chemometrics is a powerful tool for predicting the age of inks deposited on paper. In addition, it is worth noting the advantage of this methodology in this type of analysis since it would be rapid and non-invasive, needing little to no sample treatment" Ortiz "This work confirmed that ink dating methods using liquid extraction followed by GC/MS analysis were easily applicable in a laboratory, proved to be sensitive enough for dating purpose and allowed measuring a wide range of concentrations corresponding to the concentrations commonly found in ink strokes. Four different ageing parameters were evaluated: the quantity of PE, the relative peak area (RPA) between different solvents and two solvent loss ratios calculated respectively from the PE quantity (R%) and the RPA values (R%*). Among those, two particularly interesting parameters, RPA and R%* values for slow, medium, and fast drying inks were reported for the first time in the literature. In terms of variability, R% proved to be the most repeatable parameter as the maximal RSD obtained was 16%. In opposition, the R% showed the highest variability with RSD values going up to 38% for older samples. Thus, when RPA can be calculated, the R% should always be preferred to the R% in order to reduce significantly the variability for the analysis of older ink samples" Crystal Violet (dye) -3 blue ballpoint pens Compare the degradation pathways of the pure dye in water and ethanol upon exposure to xenon light UV-Vis iv / LDI ix "Significant differences have been observed in the products and the kinetics of the degradation. N-demethylation, an expected decomposition process, was found to take place only in aqueous solution and kinetics calculations showed that the degradation occurred 2.5 times faster in ethanol compared to water. The degradation of crystal violet in inks from four ballpoint pens on paper was also studied for entries made over 2-3 years. It was observed that degradation reactions were quenched by the presence of another dye due to competitive absorption. It was also observed that the thickness of a stroke (concentration of ink) influenced the degradation process. In the absence of light only one ballpoint pen showed slight degradation"

Discussion
Ballpoint pen inks' characterization and dating are frequently requested analysis in judicial cases, and most commonly for blue and black ballpoint pens. There is no standard formula for the ballpoint pens manufacturing, so the components' quantities and even solvents, dyes and resins may vary between different brands of pens, and this may vary between different regions too 4,78 . In that context, it is very important to have methods to differentiate the ballpoint pens from each other. Sometimes the fact of proving that two different pens were used in the same document may even lead to some answers in a specific case; for example, the adulteration of a document's writing part through the erasure and re-writing with a ballpoint pen of similar color 79 . It is also possible to proceed with indirect dating using the chemical analysis of an ink followed by comparison with known samples in a reference collection; some ink formulas may be discontinued or known to be characteristics of a particular sale's periodthis is the static dating method 3,80 .
This manuscript discusses technological advances and novel methods that have shown potential to improve discrimination and characterization of ink profiles.
Every one of those listed articles in Table 1 81 has compared two ink-dating methods: the first one analyzing the solvents' evaporation and the second one using the dyes' degradation, and both methods were measured with a substance extraction coefficient, as a function of inks' age. In order to extract the volatile ink components, which are fully immersed on resins, it is necessary the complete resins dissolution and they are resistant to it. The resins extractions do not vary between ink from the pen cartridge and ink from the paper, so the observed variation between old and recent pen strokes is related to the remaining solvent quantity extracted from old ink strokes' resins, correlating with the age. To extract dyes, although, it is known that the dyes composition is the same inside and outside the pen cartridge. Dyes are not bonded to resins and are easily extracted. The extraction percent obtained by Aginsky shows that the dyes extraction coefficient are always higher than the real value because its extraction is easy, unlike the solvent's extraction, and this does not correlates with the ink age. Consequently, the dyes extraction is vulnerable to unknown factors. Andrasko and Kunicki (2005) 82 also studied the differences about the ink composition, and concluded that there were no changes between dyes from the pen cartridge and dyes from a pen stroke on paper with 6 years old. Additionally, they concluded that different brands of pens have different Crystal Violet, Methyl Violet and Tetramethyl Pararosaniline quantities. Some pens presented 6% of Tetramethyl Pararosaniline, commonly used as a degradation marker, while other pens presented 15% of this substance, even though those pens had the same age. In this same paper, authors analyzed one pen from 1997 to 2004 and found no differences in Tetramethyl Pararosaniline proportion, so there was no degradation of this dye in seven years.
Lastly, another aspect to consider in ink dating is the possibility of artificial aging, which have a large number of factors that might be used to make a document looks older than it is, like the excessive light exposure and the high temperature storage 83 .

Conclusion
This systematic review presented different method approaches for ink-dating analysis. However, currently, only a few answers may be given regarding ink's age, and most of the performed exams are relative dating methods, when it is possible to evaluate two inks and compare their ages. The methods' improvement and interpretation, with research, might provide a possible working protocol in the future for Forensic Document Laboratories and its legal use. It is also essential to provide forensic experts with proper training for the analytical methods, data analysis, quality control, and interpretation of the evidence.