Chemical analysis of hazardous substances in permanent tattoo inks available on the market

(1)

(2)

Abstract

As permanent tattoos are becoming more popular and common, an increased number of allergic reactions to tattoos is reported. The purpose of this project was to analyze tattoo inks for hazardous substances, and whether they comply to current Swedish and European legislative requirements. The tattoo inks were qualitatively analyzed for pigments, and quantitatively analyzed for metals.

A total of 73 tattoo inks were collected from various sources such as a tattoo ink supplier, online retailers, and provided directly from tattoo artists. The labels of each tattoo ink bottle were inspected to investigate their compliance with the Council of Europe and the Swedish Medical Products Agency.

Matrix assisted laser desorption/ionization time of flight mass spectrometry (MALDI-ToF-MS) was used to qualitatively analyze 20 selected tattoo inks for different pigments. Inductively coupled plasma mass spectrometry (ICP-MS) was used to quantitatively analyze trace metals in 70 of the samples.

A large majority (90%) of the tested samples violated the requirements and criteria in the European resolution ResAP 2008(1), such as information on name and address of the manufacturer, minimum date of durability, sterility, batch number, and storage. Patch and allergy testing were incorrectly recommended for many samples in a way that is not accepted by dermatologists. In a worst-case scenario, this testing could be a sensitizing step. Also, it can not prevent future allergic reactions from occurring or provide any juridical insurance. Only one brand, World Famous, fulfilled the requirements for labeling for six of the seven samples (one sample failed due to a faulty declared pigment). The brands Tang Dragon and Dynamic did not fulfill any of the requirements listed in ResAP 2008(1). The list of ingredients was incorrect for all samples from Tang Dragon (ordered online in March 2019).

Also, six of the other 50 samples from different brands (World Famous, Intenze, Fusion Tattoo Ink, Eternal Ink, Solid Ink) declared at least one pigment incorrectly in their ingredients list.

25% of the declared and theoretically detectable pigments were detected by means of MALDI-ToF- MS, whereas the other pigments were either absent or below the limit of detection. Future analyses should include an MS/MS analysis. Polyethylene glycol (PEG) was identified qualatively in 15 of the 20 samples analyzed with MALDI-ToF-MS but was not listed in any of the ingredients lists. ICP-QQQ- MS is a very sensitive technique and could both detect and verify the presence of all metal-containing pigments, as well as the level of impurities. Copper was clearly more present in green and blue colors, regardless of the brand. The metal content was evidently dependent on the brand for arsenic, aluminum, bismuth, chromium, nickel, zinc, and strontium. Elevated levels of barium and strontium (partially very high levels: up to 727 mg/kg barium and up to 8.06 g/kg strontium) were found in several samples.

High amounts of aluminum (4 to 11,0 g/kg) and titanium (as judged from white precipitates and ingredients lists) were present in most samples. Nickel (0.1 to 41 mg/kg) and chromium (0.1 to 139 mg/kg) were also present in the samples. Some other impurities were also present (arsenic – 3.8 mg/kg, mercury – 1.6 mg/kg, and lead – 5.4 mg/kg for one sample, respectively). Known sensitizing pigments were declared and partially confirmed by MALDI-ToF-MS in 17 of 53 samples of the brands Radiant Colour, Eternal Ink, Fusion Tattoo Ink, and Kuro Sumi. Four samples (from Intenze, Eternal Ink, and Kuro Sumi) also declared pigments listed as non-suitable substance according to the European Commission regulation on cosmetic products from 2009.

(3)

Sammanfattning

I samband med en ökad användning av permanenta tatueringar, ökar även de rapporterade fallen av allergiska reaktioner mot tatueringar. Syftet med det här projektet var att analysera potentiellt skadliga substanser i tatueringsfärger samt om tatueringsfärgprodukter följer lagstiftning och både svenska och europeiska rekommendationer. Pigment analyserades kvalitativt och metaller kvantitativt i tatueringsfärgerna.

Totalt 73 tatueringsfärger samlades in från olika källor såsom återförsäljare av tatueringsfärger i butik och online samt direkt från tatuerare. Tatueringsfärgernas etiketter undersöktes för att kunna avgöra om de uppfyllde kraven från Europarådet och Läkemedelsverket. Matris-assisterad laser desorptions/joniserings masspektrometri med en flygtidmassanalysator (MALDI-ToF-MS) användes för kvalitativ analys av pigment i 20 utvalda tatueringsfärger och med induktivt kopplat plasma masspektrometri (ICP-MS) analyserades 70 av proverna kvantitativt på deras innehåll av spårmetaller.

En majoritet av proverna (90%) klarade inte kraven från den Europeiska resolutionen ResAP 2008(1) med avseende på korrekt information om namn och adress av tillverkaren, bäst före datum, sterilitet, LOT-nummer samt hur produkten ska lagras. Lapp- och allergitest rekommenderades felaktigt för många prover på ett sätt som inte godkänns av dermatologer. I värsta fall kan dessa rekommendationer leda till att en person sensibiliseras (får en allergi). Sådana självgjorda allergitester kan vidare inte skydda från en framtida allergi och kan inte anses utgöra en juridisk säkerhet för tillverkarna. Ett av de undersökta märken, World Famous, klarade alla märkningskrav för sex av de sju undersökta prover (ett prov hade ett pigment feldeklarerat). Märkena Tang Dragon och Dynamic klarade inte ett enda krav av ResAP 2008(1). Innehållsförteckningen för 20 prover av märket Tang Dragon (som beställdes online i mars 2019) var komplett felaktigt. Sex av de andra 50 prover från olika märken (World Famous, Intenze, Fusion Tattoo Ink, Eternal Ink, Solid Ink) hade minst ett felaktigt deklarerat pigment.

25% av de deklarerade och teoretiskt detekterbara pigmenten bekräftades med MALDI-ToF-MS, medan de andra pigmenten antingen inte fanns eller var under detektionsgränsen. Framtida analyser borde även använda sig av MS/MS analyser. Polyetylenglykol (PEG) detekterades i 15 av de 20 analyserade prover trots att det inte fanns i innehållsförteckning. ICP-MS kunde detektera och bekräfta samtliga metallinnehållande pigment samt även föroreningar. Koppar förekom oftare i gröna och blåa färger, oavsett märke. Metallinnehållet var i övrigt tydligt beroende av märke för metallerna aluminium, vismut, krom, nickel, zink samt strontium. Även innehållet av arsenik var beroende av märket. Förhöjda halter av barium och strontium, med delvis mycket höga halter (<727 mg/kg barium och <8055 mg/kg strontium) upptäcktes i flera prover. Höga halter av aluminium (4 – 11000 mg/kg) och titan (titanoxid är ett av de vanligaste pigment) förekom i nästan alla prover. Spår av nickel (0.1 – 41 mg/kg) och krom (0.1 – 139 mg/kg) fanns i alla prover. Några andra föroreningar (3.8 mg/kg arsenik, 1.6 mg/kg kvicksilver och 5.4 mg/kg bly) fanns i ett prov vardera. Redan kända allergiframkallande pigment deklarerades och bekräftades delvis med MALDI-ToF-MS för 17 av 53 prover av märkena Radiant Colour, Eternal Ink, Fusion Tattoo Ink och Kuro Sumi. Fyra prover (från Intenze, Eternal Ink och Kuro Sumi) deklarerade pigment som anses olämpligt enligt Kosmetikaförordningen från 2009.

(4)

Acknowledgements

We would like to express deep gratitude to our supervisors Dr. Yolanda Hedberg (KTH Royal Institute of Technology, Stockholm, Sweden), Dr. Leila Josefsson (KTH Royal Institute of Technology, Stockholm, Sweden), and Dr. Marie-Louise Lind (Centrum of Occupational and Environmental Medicine, CAMM, Region Stockholm, Sweden) for their guidance, encouragement and useful inputs during this research project. We would also like to thank Mag. Silvia Meschnark and Dr. Walter Gössler (University of Graz, Graz, Austria) for the performed ICP-MS analysis and the compilation of the results.

Without the help of our supervisors, Dr. Gössler and Mag. Meschnark this project would not have been possible.

(5)

1. Introduction

The art of tattooing is an ancient tradition. Over the years, tattoos have been used in several cultures around the world. It is estimated that around 12% of all Europeans and 24% of all Americans have tattoos today, and tattoos are becoming more and more popular (Piccinini et al., 2016). However, while the number of tattoos worldwide is increasing, the reports of health effects such as allergies and infections related to tattoos and tattoo ink are also increasing (Piccinini et al., 2016).

According to Sweden’s registered tattooers (SRT), there are around 9000 active and registered tattooing businesses in Sweden today. These businesses have been approved and registered by the Swedish Medical Products Agency, and can therefore only use tattoo ink from specific suppliers. An important aspect to consider when discussing the tattooing business is that there is a large tattooing market beyond this. SRT claims that there are around 30 000 people active in Sweden without any registration. These people can order tattoo ink from all over the world, that has not been approved in Sweden.

According to studies that have been made by the European Chemical Agency (ECHA), even the approved products do not comply with existing laws in many aspects. In this project, tattoo ink from both approved suppliers and non-approved suppliers will be studied and analyzed regarding their compliance with requirements, and whether they consist of any hazardous substances.

1.1. Health issues and allergies

Tattoo ink consists of pigments or other dyes, which accounts for about 50 % of the weight. The remaining content consists of adhesive substances, addition agents, thickening agents, solvents and other residues (Swedish Chemicals Agency, 2010). When getting a tattoo, tattoo ink is injected into the skin, whereby a permanent skin marking or design is being made. The tattoo ink is not specially intended for intradermal injection into the human body and often has a low purity. Several studies mentioned in a report by ECHA shows that chemicals in tattoo inks contain hazardous substances, that have significant effects on human health (European Chemicals Agency, 2018). ECHA also reports that studies have shown that colored tattoo inks are mainly responsible for the adverse skin reactions occurring after tattooing, with red colorants being responsible for the majority of the allergic reactions.

These allergic reactions can appear months or even years after the tattoo was made. The delayed response indicates that exposure to tattoo pigments through intradermal injection results in lifelong exposure and can potentially have a negative effect on human health. It is also known that pigments re- distribute in the body and pigments have been found in organs such as the lymph nodes and the liver.

Not only do local effects at the point of injection occur, there is also clear evidence of systemic exposure.

This may result in a potential risk of other health effects including cancer and reproductive health.

There are different types of allergic reactions, including immediate and delayed allergic reactions.

Tattoo inks most often cause allergic reactions of type IV, called delayed allergic reactions. These reactions require a molecule, a hapten, to bind to a protein. The protein is called a carrier and is usually a skin protein. When the hapten and the carrier bind, they form an antigen. The immune system then creates a memory of the hapten.This sensitization step, which is the step where the allergy is developed, takes from three days up to several weeks and at this step there is no sign of skin damage (Hedberg and Lind, 2020; Chen and Thyssen, 2018). Once the allergy is developed the body reacts with allergic

(7)

contact dermatitis upon re-exposure to the allergen. A developed allergy of this type is typically lifelong, and the only treatment that exists today is immunosuppressive medicine and avoiding exposure to the allergens. The latter is not possible for permanent tattoos and the immunosuppressive treatment has side effects, especially if needed during a long period of time.

1.2. Regulation of tattoo inks

According to the Swedish Medical Products Agency, tattoo inks are not classified as cosmetics or as medication (Swedish Medical Products Agency, 2017). Therefore, tattoo inks do not follow the laws that regulate pharmaceutical products, medicinal technology products and cosmetics. Because of the difficulty of classifying tattoo inks, there is no harmonized regulation of tattoo inks in the EU (Piccinini et al., 2016).

Before 2012, Sweden did not have its own legislation to regulate the production, importation, sales and usage of tattoo inks in the tattoo business. Instead, a resolution with recommendations for tattoo inks was published by the Council of Europe (CoE). The first version of the resolution was published in 2003 (CoE ResAP(2003) and was revised in 2008. The resolution urges the members of the EU to form their own legislation regarding the principles of the resolution (Piccinini et al., 2016). The resolution recommends that the product should not contain any hazardous chemicals and that products should be labeled correctly with the correct ingredients, name and other contact information of the manufacturer.

It is also recommended to do a risk evaluation of the product before releasing it for sale. For tattoo artists, the resolution recommends them to inform both the customers and the general public of the risks when getting a tattoo (Swedish Chemicals Agency, 2010). With this in mind, a Swedish regulation was published by the Ministry of the Environment and Energy in Sweden. The regulation was effective in August 2012. A year later, the Swedish Medical Products Agency, published another regulation that was effective the same year. The regulations cover both product directory, labeling, product information, importation and usage of tattoo inks. The manufacturer or the importer are responsible for registering the product in the product directory regulated by the Swedish Medical Products Agency.

The Swedish Medical Products Agency and the municipalities of Sweden are responsible to control if the tattoo business in Sweden is following these regulations (Swedish Medical Products Agency, 2017).

1.3. Project goal

In this project, different tattoo inks that are known to be used in the Stockholm region have been collected and analyzed using mass spectrometry. These samples include both tattoo inks from stores in Stockholm as well as samples ordered online. Thus, this project covers both tattoo inks that have been approved in Sweden, and tattoo inks that can be ordered and shipped to Sweden but are not necessarily approved. This project aims to identify substances that are used in different inks and to study whether they could be considered as hazardous based on existing regulations, restriction limits, and classifications. The main focus of this project was on metals and selected organic and inorganic pigments.

The purpose of this project was to qualitatively and quantitatively analyze hazardous substances in different tattoo ink pigments. Further, the labels of the tattoo products and their compliance to existing regulatory requirements were also studied.

(8)

1.4. Previous studies

According to a report compiled by the Swedish Chemicals Agency in Sweden from 2010, only five tattoo inks were free from hazardous substances out of 31 tattoo inks of different shades tested. The remaining tattoo inks contained substances such as aromatic amines, which are carcinogenic, mutagenic, and allergenic, and also different metals and polycyclic aromatic hydrocarbons (PAHs).

These substances were found at levels above the CoE’s recommended limits for tattoo inks (Swedish Chemicals Agency, 2010). To determine the presence of aromatic amines, the testing method of choice was either high performance liquid chromatography (HPLC) or gas chromatography/mass spectrometry (GC/MS). The testing method for analysis of the metals was inductively coupled plasma - optical emission spectrometry (ICP-OES) (Swedish Chemicals Agency, 2010).

The Swedish Chemicals Agency reported that the substances in tattoo inks are relatively unknown (Swedish Chemicals Agency, 2010). However, it is well known that metals and metal compounds that are toxic have been used in tattoo inks for a long time. There has also been an increased use of azo dyes in tattoo inks recently. Most azo dyes are not toxic but some, or their degradation products, can have different adverse health effects. The problem with some azo dyes occurs when they break down by UV radiation in sunlight or by laser in the process of tattoo removal, where the breakdown products are carcinogenic and mutagenic (Swedish Chemicals Agency, 2010).

Another more recent report produced by Joint Research Centre (JRC) in 2016, showed similar results as the one by the Swedish Chemicals Agency. The hazardous substances found in the analyzed samples were PAHs, (43%), primary aromatic amines, PAAs, (14%), heavy metals (9%), preservatives (6%), and microbiological contamination (11%) (Piccinini et al., 2016). These obtained numbers have been determined by using a range of different analysis methods, such as HPLC, GC/MS and ICP-OES.

1.5. Mass spectrometry

Mass spectrometry (MS) is widely used to identify components in a sample due to differences in molecular weight and chemical structure. The identification occurs according to discrepancies in the mass-to-charge ratio (m/z) of analytes in the sample. MS is a universal analytical method and can therefore be used for most analytes. The mass spectrometers commonly consist of three different components; the ionization source, the mass analyzer and the ion detection system. (Broad Institute, n.d)

In the ionization source, the analytes are turned into ions in the gas phase. Ions are able to move inside the instrument because of the external electric and/or magnetic fields applied. After ionization, the ions are separated and arranged according to their m/z-ratios with the help of the mass analyzer. Most mass spectrometers have the ability to separate isotopes. Every analyte has a unique molecular formula and every atom in the molecule can appear as one of its isotopes, resulting in different isotopes of the molecule. When the molecule consists of each atom isotopes with their respectively lowest mass, that molecule is called monoisotopic, and its mass is called monoisotopic mass. Data about the m/z-ratios along with the intensities are detected by the detection system and stored as a mass spectrum. Several signals can represent the component and these kinds of peaks are called isotopic peaks. An example of a mass spectrum can be seen in Figure 1. The height of the signals represents the relative abundance of the analyte (Broad Institute, n.d). As the analytes exist as all of its different isotopes, they will all show

(9)

in the mass spectra. There is usually a distance of 1 Da between each isotope peak given by the analyte isotopes when the charge of the ions is equal +1.

The MS instruments used in this project are matrix assisted laser desorption/ionization time of flight mass spectrometry (MALDI-ToF-MS) and inductively coupled plasma mass spectrometry (ICP-MS).

Figure 1: An example of a mass spectrum of a pigment in sample 10 with the m/z 623.570, and the isotopic peaks.

1.5.1 Tandem Mass Spectrometry

Tandem mass spectrometry (MS/MS) is used to fragment the ions of interest (precursor ion) in the primary MS (MS1). The fragmented ions, also called the product ions, can then help reveal the chemical structure of the precursor ion in the secondary MS (MS2). This is enabled by locating a collision cell between the first and second m/z-analyzer (J. Throck Watson & O David Sparkman, 2008). The basic principles of MS/MS are presented in Figure 2.

1.5.2. Matrix Assisted Laser Desorption/Ionization Time of Flight Mass Spectrometry

MALDI-ToF-MS is an analytical method widely used to detect and characterize organic compounds.

To ionize the sample in the MALDI ion source, the sample is mixed with a crystalline matrix and then irradiated with laser. Desorption of the analytes in the sample occurs in the vacuum while being ionized by the matrix. The roles of the matrix are to ionize the analytes, separate the analyte molecules in the crystal to avoid analyte-analyte molecular or ionic interaction during the ionization, and most important is to absorb some of the laser radiation to protect the analyte from radiation damage. When selecting

(10)

the most suitable matrix for the analysis, there are a few things to consider; not only does the matrix have to be able to absorb laser energy at the proper wavelength of the laser, but it also has to have characteristics similar to the analytes so that appropriate crystallization occurs. (J. Throck Watson &

O David Sparkman, 2008)

In the ToF mass analyzer, ions are accelerated, and their potential energy from the ion source is transformed into kinetic energy. The ions are separated by their m/z-ratio by simply drifting in the ToF- tube, and the time it takes for the ions to reach the first detector is determined. Ions are considered to have the same potential energy in the ion source, and ions with lower mass will reach the detector first.

(J. Throck Watson & O David Sparkman, 2008) A schematic layout of the MALDI-ToF can be seen in Figure 2.

Figure 2: The principles of MALDI-ToF-MS and MS/MS.

The result from MALDI-ToF-MS will be presented as a mass spectrum, where expected substances can be identified with the monoisotopic mass and the isotope pattern.

1.5.3. Inductively Coupled Plasma Mass Spectrometry

ICP-MS can be used for measuring elements at trace levels in samples. The sample preparation for the analysis consists of either diluting/dissolving the samples or thermally digesting them. The ICP-MS analysis is built up by a sample introduction system, ICP plasma, interface, ion optics, a mass analyzer, and a detector. The sample is nebulized in the sample introduction system where a fine aerosol is created and then transferred into the argon plasma. Because of the high temperature, the plasma ionizes the sample. The ions are extracted over the ion optics which helps the ions get into the mass analyzer. For ICP-MS, there are different types of mass analyzers that can be used. The mass analyzer used for this project is a single quadrupole mass analyzer. The quadrupole consist of a mass filter that separates ions based on their m/z-ratios. When the ions reach the detector, a mass spectrum can be obtained.

(Wilschefski & Baxter, 2019)

2. Experimental method

2.1. Materials

The aspiration of liquids was performed using automatic pipettes with disposable tips from Eppendorf, Hamburg, Germany. All liquids were stored in vials from Eppendorf. A spoon was used for weighing the amount of solid chemicals needed.

(11)

Trifluoroacetic acid (TFA, Sigma Aldrich), acetonitrile (ACN, Sigma Aldrich), sinapinic acid (SA, Bruker), ethanol (VWR), and Milli-Q water (Synergy 185, Millipore, Bedford, MA, USA) were the chemicals needed for the MALDI-ToF-MS analysis. The calibrants used was SpheriCal®-ESI dendrimers with the calibration interval m/z 270-1715 (Polymer Factory Sweden AB, Stockholm, Sweden) and Peptide Calibration Standard II with the calibration interval m/z 750-3150 (Bruker).

Isopropanol and detergent (Sigma Aldrich) were used for cleaning the plate.

The aspiration of liquids was performed using automatic pipettes with disposable tips from Eppendorf, Hamburg, Germany. All liquids were stored in vials from Eppendorf. A spoon was used for weighing the amount of solid chemicals needed.

Trifluoroacetic acid (TFA, Sigma Aldrich), acetonitrile (ACN, Sigma Aldrich), sinapinic acid (SA, Bruker), ethanol (VWR), and Milli-Q water (Synergy 185, Bedford, MA, USA) were the chemicals needed for the MALDI-ToF-MS analysis. Isopropanol and detergent (Sigma Aldrich) were used for cleaning the plate.

Nitric acid (HNO³, ≥65%, Chem-Lab NV, Zedelgem, Belgium), hydrochloric acid (HCl, 25%, Merck, Darmstadt, Germany), phosphate-buffered saline (PBS, 8.77 g/L NaCl, 1.28 g/L Na2HPO⁴, 1.36 g/L KH²PO⁴, all of analytical grade and from VWR, Sweden, adjusted with 50 % NaOH to pH 7.2–7.4), and Milli-Q water (Merck, Millipore, Darmstadt, Germany) were the chemicals needed for the ICP-MS analysis.

The Milli-Q water used in both the MALDI-ToF-MS and the ICP-MS had the specific resistivity of 18.2 MΩ at 25 °C.

2.2. Sample preparation

The tattoo inks used in this project were bought and contributed from different places. Samples 1-29 were bought online from the leading supplier in Europe called Killer Ink, which were manufactured by World Famous Tattoo Ink, Intenze Advanced Tattoo Ink, Radiant Colour, Eternal Ink and Fusion Tattoo Ink. Samples 30-36 were bought from a tattoo ink supplier in Stockholm, which were manufactured by Eternal Ink, Solid Ink and Dynamic. Samples 37-53 were contributed from a local tattoo artist, which were manufactured by Eternal Ink and Kuro Sumi Colours. The last samples, samples 54-73, were bought by a local tattoo artist from the online store Wish and manufactured by Tang Dragon, and kindly provided to this project.

The front side and the ingredients list of each of the ink bottles were photographed and then registered in an Excel-sheet, where information about brands, restrictions, shades, colors, known metal contents, certificates, manufacturers, reference numbers (denoted “ref” on the bottles), LOT-numbers, batch codes, article numbers, warnings, and ingredients were collected. An example of how the bottles were photographed is presented below in Figure 3, and the information in the Excel-sheet can be found in the Appendix. The samples were then transferred into Eppendorf tubes, two Eppendorf tubes for each sample, to facilitate their use in the different analyses. The remaining amount of tattoo inks in the original and sealed bottles were then sent to the University of Graz, Graz, Austria, for ICP-MS analysis.

Five blank samples containing phosphate-buffered saline for background contamination control were also sent to Austria.

(12)

Figure 3: The front of the bottle (right) and the ingredients list (left) of sample 54.

The Excel-sheet was used for obtaining different statistical results, such as how declared metal content differs in different inks or if known allergenic substances occur more often in specific colors or brands based on the ingredients list.

2.3. Mass spectrometry analysis

2.3.1. MALDI-ToF-MS

For the MALDI-ToF-MS analysis, 15 samples containing pigments with known allergens and five samples with no allergen pigments, according to the ingredient lists and ECHA were randomly chosen.

These 20 samples also contained pigments that are included in a MALDI-ToF-MS database for pigments used in products such as tattoo inks. The database was kindly sent by Dr. Ines Schreiver, who works for the federal institute of risk assessment in Germany (Bundesinstitut für Risikobewertung, BrF). Dr. Schreiver has conducted MALDI-analysis of different pigments and gathered the results in a database (Serup et al. 2020).

The preparations for the experimental setup of the MALDI-ToF-MS analysis include preparations of the matrix and the samples. The method used to prepare the solutions were made with the help of Schreiver’s documented procedure, but with some modifications such as different calibrants. The matrix, sinapinic acid (SA) (50 mg), was mixed with two parts ACN and one part 0.1 vol% TFA in water, to a saturated solution. The matrix solution was placed in an ultrasonic bath for 5 minutes and then centrifuged (Heraeus Biofuge pico, Hanau, Germany) at 16 000g for 10 minutes. A new matrix solution was prepared each day.

The samples were prepared by first vortexing the tattoo inks to obtain a homogenous solution in the vial. The samples were later diluted with ethanol in two dilutions, 1+9 (Batch A) and 1+100 (Batch B).

For Batch A, 10 µL of the sample was diluted with 90 µL ethanol in small vials and for Batch B, 4 µL

(13)

of the sample was diluted with 400 µL ethanol. All the diluted samples were vortexed for an additional 10 minutes. The samples that were still not fully dissolved in the solvent, were placed in an ultrasonic bath for 5 minutes.

The matrix was mixed in equal parts with the sample or a calibrant in a small vial. Equal parts of the diluted sample were mixed with Milli-Q water in a small tube. The mixtures were deposited on the MALDI-plate (MTP 384 ground steel, Bruker, Bremen, Germany) as seen in Figure 4. On each spot, 0.5 µL of the solutions were deposited twice to get a total amount of 1 µL. A spot with only the matrix, away from the samples, was also put on the plate to facilitate identification in later steps. Before inserting the plate into the instrument (UltrafleXtreme, Bruker, Bremen, Germany), all the solvents were allowed to evaporate at room temperature.

Figure 4: The pattern of the positions on the MALDI-plate for the samples.

The software used for the analysis was FlexControl. The detection range used was m/z 0 to 3000. To begin the analysis, a calibration had to be done by shooting 2000 laser shots, 1000 at a time, on the calibrant position, and assigning each of the detected peaks to a value that matched the theoretical mass of the calibrants. The intensity of the laser during the calibration was set to 80% and the frequency was 2000 Hz. After the calibration, 5000 shots, 1000 at a time, were shot randomly over the sample spot with an intensity of 96% and a frequency of 2000 Hz. After shooting on two spots, the method had to be recalibrated. The procedure continued until all the samples were analyzed. The matrix position was shot with 5000 shots, 1000 shots at a time, with a laser intensity of 75%.

To be able to identify which pigments were present in each sample, the peaks in the mass spectra were analyzed, and the isotope patterns compared to the theoretical. The theoretical isotope patterns for the hydrogen, sodium, and potassium adducts were achieved by entering the pigments’ chemical formula in enviPAT Web 2.4 (Loos et al., 2015).

2.3.2. ICP-MS analysis of trace metals

The analysis of trace metals with ICP- MS was conducted at the University of Graz, Austria. Blank samples containing PBS were diluted 1+9 with water (9 % HNO³ and 1 % HCl). The other samples (tattoo inks) were digested with the help of an Ultraclave IV microwave digestion system (MLS GmbH, Leutkirch, Germany). 0.1 g of the tattoo ink was weighed into 10 mL quartz vessels. 4.5 mL of HNO³

(14)

was added into the vessel before closing it. The vessels were then placed in an autoclave (Agilent 7700x ICPMS, Agilent Technologies, Waldbronn, Germany) with a pressure of 4·10⁶Pa of Argon (grade 5.0, Messer, Austria). The time the samples were left in the autoclave and the power and temperature of each step varied according to Table 1.

Table 1: Variation of temperature in the autoclave. Rt – room temperature.

Step No Time [min]

E [W] T1 [°C]

1 0 - 10 1000 Rt - 80

2 10 - 20 1000 80 - 150

3 20 - 30 1000 150 - 250

4 30 - 60 1000 250

5 60 - 120 0 < 80

When the samples were cooled down, the solutions were transferred into 50 mL tubes. HCl and water were added into the tubes to obtain a final concentration of 9% HNO³ and 1% HCl in the solutions. For the blank samples, the dilution factor was 10 and for the tattoo inks, the dilution factor was 500. The digestion blanks were prepared in the same way.

The total element concentrations were determined with an Agilent 7700x. The instrument was equipped with a Micro Mist nebulizer (Glass Expansion, Melbourne, Australia), a Scott type double pass spray chamber, a 2.5 mm ID quartz torch, a sample cone made from copper with a nickel tip and a nickel skimmer cone. A dilution gas was used to improve the measurements.

An external calibration solution for vanadium (V), chromium (Cr), manganese (Mn), cobalt (Co), nickel (Ni), zink (Zn), gallium (Ga), arsenic (As), strontium (Sr), molybdenum (Mo), palladium (Pd), cadmium (Cd), tin (Sn), barium (Ba), tungsten (W), gold (Au), thallium (Tl), lead (Pb), bismuth (Bi), thorium (Th), and uranium (U) was prepared, respectively, with the same concentration as the samples with the ranges of 0.01-100 µg/kg. For aluminum (Al), iron (Fe), and copper (Cu), the calibration solutions were prepared with higher ranges; 0.1-10 mg/kg. The calibration standards were prepared from single- element standards (1000 mg/kg) gravimetrically.

2.3.3. Analysis of ICP-MS results

The obtained results from the ICP-MS were analyzed in different aspects. One analysis was to compare the metal content between the different manufacturers. This was being made with a student’s t-test for

(15)

unpaired data with unequal variance and the program KaleidaGraph, version 4.0, that was used to plot graphs in section 3.3.3 and 3.3.4.

In these graphs, each box represents 50% of the data with the median value of the variable displayed as a line. The lines extending from the top and bottom of each box mark the minimum and maximum values within the data set that fall within the range R. Any values outside of this range are displayed as individual points. The range R is defined in Eq. 1:

LQ-1.5×IQD<R<UQ+1.5×IQD (Eq. 1)

where LQ is the lower quartile – the data value located halfway between the median and the smallest data value; IQD is the interquartile distance – the distance between the upper and lower quartiles (UQ – LQ); and UQ is the upper quartile – the data value located halfway between the median and the largest data value.

(16)

3. Results and discussion

3.1. Labels

3.1.1. Markings

In 2003, CoE developed a ResAP, a non-binding resolution on the safety of tattoos and PMU (permanent makeup). This resolution included chemicals, hygiene and labelling requirements (Piccicini et al. 2016). According to the ResAP, these products should have a label that consists of: name and address of the manufacturer, minimum date of durability, conditions of use and warnings, batch number, guarantee of sterility and a list of ingredients (including CAS number or CI number) (Piccicini et al.

2016). The list of requirements was then updated in 2008 with the following: period of maximum durability after opening, storage conditions, product type, health warnings and quantitative composition label (Piccicini et al. 2016).

The labels on the samples have been studied and compared to the requirements from CoE. To be able to draw conclusions, samples have been divided into groups according to manufacturer. Of all samples, excluding samples from Tang Dragon (sample 54-73), only 45% of the samples had the name and address of the manufacturer on the label. 90.5% of the samples had a mark about the maximum durability after opening. However, “Eternal Ink” had two different dates, 6 months and 365 days, which probably reflects the transition from older to newer requirements (older samples had more often 6 months duration on the label). 83.5% of the samples had a mark about sterility. Only 23% had a correct batch number (some manufacturers had the same batch number for every sample, which is wrong), and 66% had marks on storage conditions, mostly to keep the product out of the sun. The results are presented in Table 2.

Table 2: Marking on labels according to name and address of the manufacturer, minimum date of durability, sterility, batch number, and storage conditions of the samples between the different manufacturers, with a summary of the percentage within each group.

Manufacturer Name and address of the manufacturer

Minimum date of durability/period of maximum durability after opening

Sterility Batch n° Storage conditions

World Famous Tattoo Ink

(samples 1-7)

Yes Yes

“Maximum 360 days after opening”

Sterile R No Yes

“Don’t leave out in sun”

Intenze Advanced Tattoo Ink

(samples 8-17)

No Yes

Sterile Yes No

Radiant Colour (sample 18)

Yes Yes

Sterile R No Yes

Eternal Ink

(samples 19, 30-33, 37-50)

4 out of 19: yes 15 out of 19: no

16 out of 19:

“Maximum 6 months”

11 out of 19:

Sterile/

Sterilized

2 out of 19:

yes

17 out of 19:

no

16 out of 19:

yes

(17)

3 out of 19: no 8 out of 19: no 3 out of19: no

Fusion Tattoo Ink (samples 20-29)

Yes Yes

Sterile R No (Same number)

Yes

Solid Ink (samples 34-35)

Yes Yes

Sterile No (Same number)

Yes

Dynamic (sample 36)

No No No mark No No

Kuro Sumi Colours (samples 51-53)

No Yes

2 out of 3:

Sterile 1 out of 3: no

No No

Tang Dragon (samples 54-73)

No No No No No

Summary Percent 24 out of 73

(33%) have a mark with name and address of the manufacturer

48 out of 73 (66%) have an expiring date

44 out of 73 (60%) have a sterility mark

12 out of 73 (16%) have a correct batch n°

35 out of 73 (48%) have some storage conditions

Percent (excluding samples 54-73)

24 out of 53 (45%)

48 out of 53 (90.5%) 44 out of 53 (83%)

12 out of 53 (23%)

35 out of 53 (66%)

From Table 2 it can be seen that none of the nine manufacturers fully comply with the requirements of the ResAP. It is also interesting that two manufacturers, “Tang Dragon” and “Dynamic” do not comply with any of the requirements. The samples from “Tang Dragon” were bought from Wish, but the sample from “Dynamic” was bought from a professional tattoo store in Stockholm. This sample is therefore supposed to be approved in Sweden, and its deviation from the requirements is hence remarkable.

According to Table 3, all manufacturers had marks with conditions of use and/or warnings. However, several of these marks can be questioned and some of them are misleading or wrong. The mark that the consumer is supposed to read the instructions before use can be questioned since instructions rarely occurred. 2 of the 9 manufacturers were recommending a patch test, but they did not inform how or where to conduct a patch test. Another manufacturer recommended a patch test and gave instructions.

However, these instructions were not describing a correct patch test. Another critical aspect here is that a patch test should actually be performed by a dermatologist. Also, a patch test cannot guarantee that a future allergic reaction occurs. One manufacturer, “Radiant Colour”, also recommended an allergy test before use instead of a patch test, without the instructions on how to do it. “Eternal Ink” and “Tang Dragon” state on their products that they do not take any responsibility if an allergic reaction occurs.

Table 3: Conditions of use and warnings marked on the labels between the different manufacturers.

(18)

Conditions of use and warnings

World Famous Tattoo Ink (samples 1-7)

“Do not dilute with water”

“For professional use only”

“Read instructions before use”

Intenze Advanced Tattoo Ink (samples 8-17)

“Do not use in or near eyes”

“Shake well before use”

“Manufacturer recommends color and skin patch test”

Radiant Colour (sample 18)

“Application of this product can cause allergic reactions in individuals sensitive to the ingredients of the ink, it is recommended to perform an allergy test before use”

Eternal Ink

(samples 19, 30-33, 37-50)

“Application of this product on some individuals may cause an allergic reaction due to the sensitivity of the individual to the dye or pigment. We disclaim any responsibility of the allergic reactions of individuals to whom this pigment is applied. Where there is history of tolerance to this dye or pigment spot testing is advised to check for potential allergies.”

“Pigment Patch Test: 1) Using soap & water or alcohol, clean small area of skin in inner forearm. 2) apply small amount of pigment to the area and allow it to dry. 3) After 24 h, wash with soap & water. 4) if no irritation or inflammation is apparent, it may be assumed no hypersensitivity to the pigment exists. Test before application.”

Fusion Tattoo Ink (samples 20-29)

“Do not dilute with water”

“For professional use only”

Solid Ink (samples 34-35)

“Can cause allergic reactions!”

Dynamic (sample 36)

“As with any product, allergic reactions can occur. Manufacturer recommends color and skin patch test”

Kuro Sumi Colours (samples 51-53)

Tang Dragon (samples 54-73)

“Warning: This product is non-toxic, pure organic pigment. Application on certain individuals may cause allergic reaction. We disclaim any responsibility for allergic reactions of individuals to whom this pigment is applied.”

3.1.2. Ingredients

By observing the labels on each bottle containing tattoo ink, it was evident that the labels on samples 54-73 (all from “Tang Dragon”) were not correct. These labels were all exactly the same, including the same information, list of ingredients, warnings, and LOT number. The LOT number is specific to each product and it must, therefore, be different for different bottles and colors. Additionally, the labels state that the products only contain organic pigments, but the only pigments listed were two inorganic

(19)

pigments. The listed pigments were the white pigment titanium dioxide (CI77891) and the black pigment carbon (CI77266). From these two pigments, it would be impossible to create all the different shades of the inks ranging from all colors. Figure 3 illustrates the tattoo ink bottle of sample 54 with the shade yellow. Thus, the labels on the bottles associated with samples 54-73 were considered completely unreliable and the label information of these samples was hence not analyzed further.

Samples 1-53 were divided into 11 different groups depending on their different shades. The assignments to a certain color shade were made based on the name of the color and their visual appearance (confirmed by four people). The 11 groups were white, yellow, orange, red, green, blue, pink, purple, black, grey, and brown, presented in Table 4.

Table 4: Assignments of samples to different color groups.

Group: Samples:

White 1, 46

Yellow 2, 8, 35

Orange 3, 9, 10, 20*, 44

Red 4, 5, 18*, 19*, 21*, 22*, 33*, 52*

Green 6, 13, 14, 24*, 25*, 26, 51+

Blue 7, 15, 31, 32, 37, 41+, 42*, 50

Pink 11, 12, 38*, 48*, 49, 53*

Purple 16⁺, 23, 28, 29*, 34, 40*, 47

Black 17, 36

Grey 27*, 30*, 39, 45

Brown 43

*Samples containing known allergenic substances according to ECHA and their ingredients label.

+Samples containing non-suitable substances according to the EC regulation on cosmetic products (EC, 2009).

The samples marked with an asterisk (*) in Table 4 contain a known allergenic (sensitizing) pigment according to their ingredients list and their registration entry information based on ECHA. However, the samples without asterisks could still contain sensitizing substances that are either unknown or present unintentionally. The samples marked with a raised plus sign (⁺) in Table 4 declares containing pigments listed by the Swedish Medicines Agency for non-suitable substances for tattoo ink (Swedish Medical Products Agency, 2017).

The percentage of samples containing allergenic substances according to the ingredients list in each color group is presented in Figure 5. Group red contained the largest share (75%) of samples with known allergenic substances and the groups white, yellow, black, and brown contained the smallest share (0%) of samples with known allergenic substances. The groups pink and grey contained 28.6%, group orange contained 20%, and group blue contained 12.5% samples with known allergenic substances.

(20)

Figure 5: The proportion of samples containing known allergenic pigments (based on ECHA) in each color group and according to the sample ingredients list.

The proportion of known allergenic substances in Figure 5 might be misleading since the groups have different sizes and are hence differently influenced by coincidence. For example, it is problematic to compare group brown containing one sample with group red containing eight samples. To investigate any statistically significant differences and patterns, a probability test (student’s t-test) has been applied.

The t-test was conducted with a significant level of p=0.05 (5% probability that two groups are equal), and with a two-tailed hypothesis (the difference could be in both directions, so for example group 2 could be larger or smaller than group 1). A significant level of 0.05 implies that there is only a 5%

chance that the groups are exactly the same, which has been considered as a sufficiently low chance. If the p-value between two groups is below 0.05, the result shows a statistically significant difference between the two groups investigated. The calculated p-values between all the groups in terms of their known allergenic pigment contents as a function of color shade are presented in Table 5, in a matrix showing the two groups that were compared. A web-based student’s t-test calculator was used (Social Science Statistics, 2020).

Table 5: P-values between the content of known allergenic pigments in two different color groups where n represents the number of samples. Statistically significant differences are marked in green and bold.

White (n=2)

Yellow (n=3)

Orange (n=5)

Red (n=8)

Green (n=7)

Blue (n=8)

Pink (n=6)

Purple (n=7)

Black (n=2)

Grey (n=4)

Brown (n=1)

White (n=2)

(21)

Yellow (n=3)

1

Orange (n=5)

0.576 0.482

Red (n=8)

0.060 0.024 0.059

Green (n=7)

0.456 0.356 0.763 0.081

Blue (n=8)

0.645 0.568 0.742 0.009 0.474

Pink (n=6)

0.267 0.170 0.353 0.373 0.471 0.145

Purple (n=7)

0.456 0.356 0.763 0.081 1 0.474 0.471

Black (n=2)

1 1 0.576 0.060 0.456 0.645 0.267 0.456

Grey (n=4)

0.313 0.203 0.407 0.433 0.527 0.188 1 0.527 0.313

Brown (n=1)

- - - - - - - - - -

Table 5 presents the different p-values between the color groups, with the different groups in the axis where the n-value indicates the number of samples in each group. There is a significant difference between the groups yellow and red, and between the groups red and blue. This is evident since the values between these groups are below 0.05 compared to the other groups with values higher than 0.05 (Table 5). This concludes that there is a significant difference in the labeled presence of known allergenic substances in the samples of different color shades. Hence, more known allergenic pigments were found in the red colors compared to the yellow and blue colors. The difference between the remaining groups was not large enough to exclude the influence of coincidence, mainly due to a low number of samples in some groups (such as the brown, black, and white groups).

Pigments are different chemical substances that can be divided into chemical groups depending on their structure. The pigments in the collected tattoo inks are classified as monoazo, disazo, xanthene, oxazine, aminoketone, indigoid, phthalocyanine, and inorganic pigments. As shown in Figure 6, 79.2% of samples contain inorganic pigments e.g. iron oxide (Fe²O³) and titanium dioxide (TiO²).

(22)

Figure 6: Percentage of samples with pigments classified by a specific chemical group.

Some of the pigments in the samples contained metals such as copper, iron, titanium, and molybdenum.

Metals are often used in different substances to make them colored, either in inorganic pigments such as metal oxides or in metal-organic complexes. For example, blue substances often contain copper (in all cases belonging to the phthalocyanine group of pigments, CI74160, CI74260, and CI74265) and red/orange pigments often contain iron (in all cases as the inorganic pigment iron(III) oxide, CI77491).

Figure 7 illustrates the percentage of known (labeled) metal contents as a function of the color group.

The majority of samples (77.8%) containing copper are blue and green. 19 of 53 samples contained copper according to the ingredients lists. Most samples (55.6%) containing iron are purple and red, and most samples containing titanium are purple, blue, and green (52.7%). 9 of 53 samples contained iron according to the ingredients lists. Titanium was only present in the inorganic pigment CI77891 (titanium dioxide), which was very common (39 of 53 samples). Out of all collected samples, only one (sample 10 in group orange) contained molybdenum, as the pigment CI45170:2, which belongs to the xanthene group of pigments. The labeled metal content displayed in Figure 7 was found to be incorrect and will be discussed more thoroughly in section 3.3.2.

(23)

Figure 7: Percentage of known (labeled) metal content in different color shades.

(24)

3.3. Mass spectrometry

3.3.1. MALDI-ToF-MS

Analysis with MALDI-ToF-MS resulted in several mass spectra, which were then analyzed to identify pigments by identifying their corresponding peaks and examining the isotope patterns. A total of 20 samples were analyzed, of which 15 samples contained labeled allergenic pigments, and five samples did not contain any labeled allergenic pigments. The 15 samples analyzed were: 18, 19, 20, 21, 22, 24, 25, 27, 29, 30, 33, 38, 40, 42 and 48, and the control group consisted of sample: 6, 10, 16, 45 and 51.

The samples were analyzed with matrix and in ethanol only.

Since the ingredients list of the tattoo inks did not state the proportions of each ingredient, an assumption that the samples contained 50 wt% pigments was made, which, according to the Swedish Chemicals Agency, is a fair assumption (Swedish Chemicals Agency, 2010). The chosen samples mentioned above all contained either one or four pigments according to the ingredients list. In order to get the right concentration of the pigments needed to get reliable results, the sample had to be diluted in two batches;

Batch A and B. For the samples with four pigments, a 1+9 dilution could lead to a high concentration of the pigments, which would lead to difficulties for the rest of the measurements.

Batch B was used in the first set of the MALDI-ToF-MS analysis but it lead to the results being difficult to interpret due to the absence of relevant peaks. Because of that, Batch A was used.

Two examples of mass spectrums are shown in Figure 8, where the top spectrum shows the analysis in ethanol solution and the bottom spectrum shows the analysis in the matrix. The spectra look fairly similar, but analysis with matrix results in more peaks belonging to the matrix in the m/z range of 0 to 500. Additionally, other substances that did not ionize can be detected with matrix in contrast to ethanol.. This is because the matrix assists in ionization for other substances, which is not true for the pigment in ethanol. However, when studying peaks in mass spectra from analyses in matrix, one must be aware that some peaks are originating from the matrix. Using the mass spectra of only the matrix, one can clarify which peaks correspond to the matrix. The advantage of analyzing the samples in only ethanol is that all the peaks in the mass spectrum derive from the sample. The disadvantage, however, is that the mass spectrum may provide less information.

(25)

Figure 8: Mass spectra of sample 18 in ethanol (top) and matrix (bottom) from analysis with MALDI- ToF-MS.

(26)

Pigment CI21110 was listed as an ingredient in sample 10. To be able to confirm this, peaks at the expected m/z-site needed to be identified in the mass spectrum, and the isotope pattern had to match.

The pigment has a monoisotopic mass of 622.14 g/mol, and peaks around this value are found in the mass spectrum associated to sample 10 (Figure 9a). The isotope pattern of the ionized pigment with hydrogen as an adduct ion is visible in Figure 9b. By comparing the peaks in the mass spectrum with the theoretical isotope pattern of the ionized pigment, it is visible that they had the same pattern for the first five peaks, where the ratio between the intensities are the same. The theoretical isotope pattern has two more peaks than the mass spectrum of sample 10. Since these peaks have a low signal in the theoretical isotope pattern, it may result in the peaks not being visible in the mass spectrum of the sample. Since the isotope patterns match reasonably well and are located at the same site, the conclusion that pigment CI21110 was found in sample 10 can be drawn. The first peak in Figure 9a has m/z 623.57 and the first peak in Figure 9b has m/z 623.15, and the values differ by 0.42 Da. The remaining peaks differ in a similar way and this was most likely due to calibration errors.

Figure 9a: A section of the mass spectrum from analysis of sample 10 in ethanol solution between m/z 620-630.

(27)

Figure 9b: The theoretical isotope pattern of the ionized pigment CI21110 with hydrogen as an adduct ion [M+H]+.

Pigment CI51319 was present in sample 16 according to the ingredient list. The mass spectrum from the analysis of sample 16 in ethanol showed peaks at the expected m/z-values, which was around m/z 588.11, corresponding to the monoisotopic mass of the pigment (Figure 10a). By comparing the peaks in the mass spectrum of sample 16 with the expected isotope pattern of the ionized pigment (Figure 10b), it was evident that the pigment was present in the sample. This was due to the isotope pattern in Figure 10a and Figure 10b matching each other, both isotope patterns contain 7 peaks with similar ratios of the signals. To study how the isotope patterns differ, the first peaks were compared. The first peak in Figure 10a has a slightly lower mass value than the first peak in Figure 10b, differing with 0.68 Da, which was most likely caused by calibration errors.

Chemical analysis of hazardous substances in permanent tattoo inks available on the market

Abstract

Sammanfattning

Acknowledgements

Table of contents

1. Introduction

1.1. Health issues and allergies

1.2. Regulation of tattoo inks

1.3. Project goal

1.4. Previous studies

1.5. Mass spectrometry

2. Experimental method

2.1. Materials

2.2. Sample preparation

2.3. Mass spectrometry analysis

3. Results and discussion

3.1. Labels

3.3. Mass spectrometry