Photodecomposition of Pigment Yellow 74, a pigment used in tattoo inks

Article abstract of DOI:10.1562/2004-04-06-RA-136.1

Tattooing has become a popular recreational practice among younger adults over the past decade. Although some of the pigments used in tattooing have been described, very little is known concerning the toxicology, phototoxicology or photochemistry of these pigments. Seven yellow tattoo inks were obtained from commercial sources and their pigments extracted, identified and quantitatively analyzed. The mono- azo compound Pigment Yellow 74 (PY74; CI 11741) was found to be the major pigment in several of the tattoo inks. Solutions of commercial PY74 in tetrahydrofuran (THF) were deoxygenated using argon gas, and the photochemical reaction products were determined after exposure to simulated solar light generated by a filtered 6.5 kW xenon arc lamp. Spectrophotometric and high-pressure liquid chromatography (HPLC) analyses indicated that PY74 photodecomposed to multiple products that were isolated using a combination of silica chromatography and reversed-phase HPLC. Three of the major photodecomposition products were identified by nuclear magnetic resonance and mass spectrometry as N-(2-methoxyphenyl)-3-oxobutanamide (o-acetoacetanisidide), 2-(hydroxyimine)-N-(2-methoxyphenyl)-3-oxobutanamide and N,N″-bis(2- methoxyphenyl)urea. These results demonstrate that PY74 is not photostable in THF and that photochemical lysis occurs at several sites in PY74 including the hydrazone and amide groups. The data also suggest that the use of PY74 in tattoo inks could potentially result in the formation of photolysis products, resulting in toxicity at the tattoo site after irradiation with sunlight or more intense light sources.

Full text of DOI:10.1562/2004-04-06-RA-136.1

Photodecomposition of Pigment Yellow 74, a Pigment Used in Tattoo Inks
Author(s): Yanyan Cui, Andrew P. Spann, Letha H. Couch, Neera V. Gopee, Frederick E. Evans, Mona I.
Churchwell, Lee D. Williams, Daniel R. Doerge, and Paul C. Howard
Source: Photochemistry and Photobiology, 80(2):175-184.
Published By: American Society for Photobiology
DOI: http://dx.doi.org/10.1562/2004-04-06-RA-136.1
URL: http://www.bioone.org/doi/full/10.1562/2004-04-06-RA-136.1
BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and
environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published
by nonprofit societies, associations, museums, institutions, and presses.
Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of
BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries
or rights and permissions requests should be directed to the individual publisher as copyright holder.
BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research
libraries, and research funders in the common goal of maximizing access to critical research.

Photochemistry and Photobiology, 2004, 80: 175–184
Photodecomposition of Pigment Yellow 74, a Pigment Used in Tattoo Inks{
Yanyan Cui1,2, Andrew P. Spann1, Letha H. Couch1,2, Neera V. Gopee1,2, Frederick E. Evans3,
Mona I. Churchwell1, Lee D. Williams1, Daniel R. Doerge1 and Paul C. Howard*1,2
1Division of Biochemical Toxicology, National Center for Toxicological Research, U.S. Food and Drug Administration,
Jefferson, AR
2National Toxicology Program Center for Phototoxicology, National Center for Toxicological Research,
U.S. Food and Drug Administration, Jefferson, AR
3Division of Chemistry, National Center for Toxicological Research, U.S. Food and Drug Administration,
Jefferson, AR
Received 6 April 2004; accepted 3 June 2004
photolysis products, resulting in toxicity at the tattoo site after
irradiation with sunlight or more intense light sources.
ABSTRACT
Tattooing has become a popular recreational practice among
younger adults over the past decade. Although some of the
pigments used in tattooing have been described, very little is
known concerning the toxicology, phototoxicology or photo-
chemistry of these pigments. Seven yellow tattoo inks were
obtained from commercial sources and their pigments
extracted, identified and quantitatively analyzed. The mono-
azo compound Pigment Yellow 74 (PY74; CI 11741) was found
to be the major pigment in several of the tattoo inks. Solutions
of commercial PY74 in tetrahydrofuran (THF) were de-
oxygenated using argon gas, and the photochemical reaction
products were determined after exposure to simulated solar
light generated by a filtered 6.5 kW xenon arc lamp.
Spectrophotometric and high-pressure liquid chromatography
(HPLC) analyses indicated that PY74 photodecomposed to
multiple products that were isolated using a combination of
silica chromatography and reversed-phase HPLC. Three of
the major photodecomposition products were identified by
nuclear magnetic resonance and mass spectrometry as N-(2-
methoxyphenyl)-3-oxobutanamide (o-acetoacetanisidide), 2-
(hydroxyimine)-N-(2-methoxyphenyl)-3-oxobutanamide and
N,N0-bis(2-methoxyphenyl)urea. These results demonstrate
that PY74 is not photostable in THF and that photochemical
lysis occurs at several sites in PY74 including the hydrazone
and amide groups. The data also suggest that the use of PY74
in tattoo inks could potentially result in the formation of
INTRODUCTION
Tattooing is an art form that has existed for several thousand years.
References to tattooing can be found in ancient writings from
Egypt, Mesopotamia, China, Japan and Oceania. Several reasons
for voluntarily obtaining a tattoo include recreational expression,
permanent makeup and medical–reconstruction purposes. Cosmetic
reconstructive tattooing has been used to correct undesired features
after traumatic injury to the skin (e.g. after burns or surgery) or to
correct undesired skin conditions (e.g. vitiligo). In these cases,
tattooing blends different skin color pigments into the skin to mask
unwanted or missing features (1,2). Tattooing with India ink is also
used internally for marking lesions for removal or for marking the
locations of removed growths in colonoscopies, laparoscopies and
laparotomies (3–8).
The reasons for obtaining recreational tattoos are multifaceted
and include tribal or group affiliation, social identification and
personal artistic expression. Tattooing has increased in popularity
in the United States over the past decade (9–11), especially among
youth (11–16). It has been estimated that as many as 20 million
U.S. citizens have a recreational tattoo, which involves a significant
portion of the population.
Tattooing is the method of using a needle or some other sharp
object to force suspended colored solids through the epidermis into
the dermis. Dermal deposition of the materials is important for
tattoo longevity, and in humans, the tattoo pigment is typically
deposited in the upper half to one-third of the dermis. The
pigmented materials in the tattoo ink either remain in the interstitial
space or are taken up by dermal cells through phagocytosis.
Two publications stand out as pivotal investigations into the
materials used in modern tattoo inks. Lehmann and Pierchalla (17)
examined the inorganic and organic materials used as pigments in
tattoo inks. The inorganic pigments included CdSe, CdS, Cr2O3,
CoO–Al2O3, Fe2O3, Fe3O4, HgS, TiO2 (anatase and rutile) and
ZnO. The organic pigments found in tattoo inks included Pigment
Yellow 74 (PY74), Pigment Yellow 83 (PY83), Pigment Orange
16, Pigment Red 22 (PR22), Pigment Blue 15 (PB15), Pigment
Violet 19 and Pigment Green 7 (PG7). Quantitative information
{Posted on the website on 4 June 2004.
*To whom correspondence should be addressed: Division of Biochemical
Toxicology, National Center for Toxicological Research, U.S. Food and
Drug Administration, U.S. Department of Health and Human Services,
3900 NCTR Road, HFT-110, Jefferson, AR 72079, USA. Fax: 870-543-
7136; e-mail: phoward@nctr.fda.gov
Abbreviations: APCI, atmospheric pressure chemical ionization; FDA,
Food and Drug Administration; HPLC, high-pressure liquid chromatog-
raphy; MS, mass spectrometry; NMR, nuclear magnetic resonance; NOE,
nuclear Overhauser effects; PB15, Pigment Blue 15; PG7, Pigment
Green 7; PR9, Pigment Red 9; PR22, Pigment Red 22; PR170, Pigment
Red 170; PY74, Pigment Yellow 74; PY83, Pigment Yellow 83; THF,
tetrahydrofuran; TLC, thin-layer chromatography.
ꢀ 2004 American Society for Photobiology 0031-8655/04 $5.00þ0.00
175

176 Yanyan Cui et al.
Exposure of PY74 to light. PY74 was dissolved in THF and placed in
Teflon-capped cuvettes with a 2 mm light path (Starna Cells Inc.,
Atascadero, CA). Unless otherwise indicated, the cuvettes and samples
were purged with argon before exposure to light. The light source was a 6.5
kW xenon arc lamp with the emission filtered through WG320 glass filters
(0.6 mm; Schott Glass Technologies, Puryea, PA) to achieve a spectrum
with ultraviolet light content consistent with terrestrial sunlight (UV-A:UV-
B, ;21:1) (25). Cuvettes were typically placed 0.5 m from the source,
which corresponds to an irradiance approximately equal to noontime
terrestrial sunlight (summer at 348 latitude) with the following values: UV-C
(250–280 nm), 2.2310ꢀ6 W/cm2; UV-B (280–315 nm), 2.4310ꢀ4 W/cm2;
UV-A (315–400 nm), 5.2 3 10ꢀ3 W/cm2; visible (400–800 nm), 4.2 3 10ꢀ2
W/cm2.
concerning the concentrations of the pigments and diluents
comprising the inks was not provided.
The other seminal article on the presence of organic pigments in
tattoo inks was by Ba¨umler et al. (18) in 2000. Several tattoo inks
were extracted, and the organic color component was determined by
high-pressure liquid chromatography (HPLC) and mass spectrom-
etry (MS) methods. The color pigments included Pigment Yellow
14, Pigment Yellow 55, PY74, PY83, Pigment Yellow 87, Pigment
Orange 13, Pigment Red 5, Pigment Red 9, PR22, Pigment Red
112, Pigment Red 170 (PR170), Pigment Red 122, Pigment Violet
23, PB15, PG7 and Pigment Green 36.
Isolation of photoproducts. Larger scale preparations were used for the
isolation of photoproducts for identification. PY74 was dissolved in argon-
purged THF in either glass-capped cylinders or 500 mL glass bottles with
Teflon-lined caps and placed at a distance of 0.4–0.5 m from the xenon
arc light source until a color change was noted. The temperature in the
cuvettes and vessels was maintained at ,408C using fans. The photo-
decomposition products were separated by column chromatography (2.5 3
A survey by the Danish Environmental Protection Agency (19)
focused on tattoo inks (20). Several pigments that are not approved
by the Danish government for inclusion in inks were detected and
included Pigment Orange 36, PY74, PR170, Pigment Yellow 97,
Pigment Red 146, Pigment Brown 25 and Pigment Red 266.
The U.S. Food and Drug Administration (FDA) has considered
intradermal tattoos and permanent makeup to be cosmetics because
they are applied to the body for the purpose of altering appearance
or promoting attractiveness (21). There have not been any studies to
date to determine the safety of the tattoo pigments under conditions
of use, and as a result, no pigments are currently approved by the
U.S. FDA for use in tattoo inks (21,22), although many pigments
and colors have been approved by the U.S. FDA for use in foods,
drugs and medical devices (21 CFR [U.S. Code of Federal
Regulations] Parts 73, 74 and 82).
˚
30 cm silica gel, Merck grade 9385, 60A, Aldrich, with methylene
chloride as eluant), thin-layer chromatography (TLC; Alltech Associates
[Deerfield, IL], silica 20 3 20 cm TLC plates with methylene chloride as
eluant) or HPLC as described above.
Instrumentation. The HPLC systems consisted of a Varian 9012 HPLC
pump equipped with either a Varian 9020 UV–VIS or a Pro-Star 330
diode array absorbance monitor. Injections were made manually using
a Rheodyne 6125 injector. Absorption spectra of pigment solutions were
obtained using a Perkin-Elmer Lambda 45 UV–VIS spectrometer versus
a solvent blank. Mass spectra were obtained using a Quattro Ultima triple
quadrupole mass spectrometer (Micromass, Manchester, UK) with an
atmospheric pressure chemical ionization (APCI) interface, 4008C
nebulizer probe and an ion source temperature of 1208C. Product ion
spectra were obtained using a collision gas pressure of 1.8 mbar argon,
a cone voltage of 40 V and various collision energies. The spectral
irradiance of the simulated solar light source was determined using an
Optronics OL-754 spectroradiometer (250–800 nm; Optronics Laborato-
ries, Orlando, FL) equipped with a 6 inch integrating sphere and calibrated
using a NIST (National Institute for Standards and Technology)-traceable
quartz–tungsten–halogen standard (OL752-10E; Optronics Laboratories).
Nuclear magnetic resonance (NMR) spectra were obtained at 500 MHz
using a Bruker AM-500 (Billerica, MA) operating at 288C. Data were
processed using either the spectrometer software (DISNMR) or the Bruker
1D WIN-NMR software on a PC. The samples were dissolved in CD2Cl2,
and the residual CDHCl2 peak from the solvent was assigned to 5.32 ppm
as a reference. All reported NMR results are from first-order measurements.
Chemical-shift assignments are described in the text. The ‘‘x’’ in Figs. 5 and
7 denotes an unidentified impurity.
Ba¨umler et al. (23) reported that exposure of the monoazo
pigment PR22 in acetonitrile to visible Nd:YAG laser light resulted
in the formation of 2-amino-4-nitrotoluene. This has been followed
by a recent report that Pigment Red 9 (PR9) likewise photo-
decomposes at the azo group in visible laser light to form aromatic
amine compounds (24). The work of Ba¨umler suggests that
exposure of tattooed individuals to the laser light sources used for
tattoo removal may result in the formation of potentially mutagenic
aromatic amine compounds at the tattoo site.
In view of the widespread practice of tattooing in youth, the
photochemical photolysis of PR22 and PR9 with visible lasers and
the unknown fate of photoactivated tattoo pigments, we have
examined the previously unreported photochemical instability of
PY74 after exposure to simulated solar light. PY74 was chosen for
this study because of its commercial availability, simple structure
and solubility when compared with other pigments.
RESULTS
Identification of PY74 in tattoo inks
MATERIALS AND METHODS
PY74 had been reported to be present in tattoo inks obtained in
Europe (17,18), and we tested for the presence of PY74 in yellow
tattoo inks commonly used in the United States. Seven different
yellow tattoo inks identified as Yellow, Dark Yellow, Sun Yellow,
Mohawk Yellow, Canary Yellow, Tulip Yellow and Poppy were
extracted as described in the Materials and Methods and analyzed
by HPLC. PY74 was identified in six of the seven tattoo inks by
comparison of the HPLC retention times and absorbance spectra of
the eluting components with authentic PY74 (results not shown).
The contribution of PY74 to the mass of each of the following
tattoo inks was determined using PY74 standards (% wt/wt): Dark
Yellow, 0.7%; Sun Yellow, 3.6%; Mohawk Yellow, 3.6%; Canary
Yellow, 4.5%; Tulip Yellow, 10.5% and Poppy, 12.7%. PY74 was
apparently the sole pigment in several of the inks; however, in
several inks additional organic pigments were used in combination
with PY74 to achieve the desired ink color. No attempt was made
to identify inorganic pigments in the tattoo inks.
Reagents. PY74 (CAS 6358-31-2; CI 11741; 2-[(2-methoxy-4-nitrophenyl)
azo]-N-(2-methoxyphenyl)-3-oxobutanamide), o-acetoacetanisidide (CAS
92-15-9) and tetrahydrofuran (THF) were obtained from Aldrich Chemical
Company (Milwaukee, WI). The tattoo inks were obtained from
commercial vendors as liquid suspensions and are identified by color
designations. Methylene chloride was obtained from J. T. Baker
(Phillipsburg, NJ), and methanol and acetonitrile were obtained from
Fisher Scientific Co. (Pittsburgh, PA). All other reagents were of best
available commercial grade.
Analysis of pigments in tattoo inks. Approximately 25 mg of tattoo ink
was added to a 50 mL conical glass tube containing 2 mL of water. The
contents were mixed vigorously and extracted three times with 15 mL of
methylene chloride. The methylene chloride extracts were combined, dried
in vacuo and reconstituted in 5 mL of methylene chloride. The samples
were analyzed using a Varian (Walnut Creek, CA) HPLC system equipped
with a Varian Pro-Star 330 diode array detector. The stationary phase
was a Phenomenex (Torrance, CA) Prodigy 5 lm C8 HPLC column (4.6 3
150 mm with 4.6 3 30 mm guard column), and the mobile phase started at
1:1 methanol and 0.015% formic acid and changed linearly to methanol in
25 min at a flow rate of 0.9 mL/min.

Photochemistry and Photobiology, 2004, 80 177
1
Characterization of PY74 by H-NMR and MS
1
The H-NMR spectrum of PY74 in CD2Cl2 is presented in Fig. 1.
Many of the chemical shifts were assigned from the relative
intensities, the coupling patterns and several simple decoupling
experiments. The results are presented in Table 1 along with other
NMR spectral parameters. The upfield region contains resonances
from the three methyl groups. The methoxy proton resonances
were assigned based on large nuclear Overhauser effects (NOE;
23% and 18%) to the adjacent ortho-substituted proton after
saturation of each methoxy resonance. The resonance from the
acetyl group is also distinguished from the methoxy resonances by
its chemical shift. Two broadened singlets (linewidth ca 6.5 Hz) in
the most downfield region are characteristic of exchangeable
protons with strong hydrogen bonds, and the similar linewidths
suggest that they both arise from amine protons.
PY74 could exist in several tautomeric forms, including at least
two that contain an azo group and one that contains a hydrazone
group. In one azo form there would be a methine proton at C2 and
only a single strong intramolecular hydrogen bond; however, there
was no NMR evidence of a methine resonance, and the chemical
shifts greater than 11 ppm indicate two strong intramolecular
1
Figure 1. The 500 MHz H-NMR spectrum of PY74 in CD2Cl2 is shown
with resonance assignments.
PY74 was further characterized using APCI MS (Fig. 2). PY74
showed a molecular ion at m/z 387 in the positive-ion mode and
a molecular ion at m/z 386 in the negative-ion mode (not shown). In
the positive-ion mode, PY74 fragmented at the amide group,
resulting in fragment ions of m/z 264 and 124 (the latter being
identified as o-aminoanisole). Alternatively, PY74 fragmented at the
azo groþup to form the fragment ions at m/z 167 and 169 (R5NHþ or
R–NH3 of 2-amino-5-nitroanisole). Less prevalent was the
fragmentation of PY74 between the C1–C2 bond giving rise to
fragments of m/z 236 and 150. The fragment shown at m/z 369 results
from loss of H2O from PY74. Negative-ion fragmentation (not
shown) showed a preference for azo bond cleavage (m/z 167) with
further loss to m/z 152 (3-nitroanisole) or m/z 122 (nitrobenzene).
hydrogen bonds.
A more favorable azo tautomer contains
a hydroxybutene group that would be stabilized by two strong
intramolecular hydrogen bonds. One of the hydrogen bonds would
involve an amine, but the second would involve a hydroxyl group,
which would not be expected to produce the 6.5 Hz broadening
mentioned above. The presence of two strongly hydrogen-bonded
amine protons and no methine hydrogen is consistent with
a hydrazone rather than an azo form.
Several NMR experiments were conducted to characterize the
tautomeric form of PY74 in CD2Cl2. Small long-range coupling
constants of up to 0.7 Hz (Table 1) were confirmed between each of
the exchangeable protons and protons at the ortho and meta positions
of the corresponding aromatic ring in resolution-enhanced spectra.
These are analogous to benzylic coupling constants (26), except that
the magnitudes of the couplings show small but not insignificant
differences, as would be expected. They indicate that an amine
proton is adjacent to each ring. From these data it was possible to
assign the remaining resonances and to propose a predominant
hydrazone tautomeric form that is stabilized by two strong intra-
molecular hydrogen bonds. In addition, saturation of the methoxy
proton resonance at 4.09 ppm resulted in an NOE to the arylamine
proton resonance at 14.74 ppm (1.7%), whereas no NOE was
detected at the other arylamine resonance. Saturation of the methoxy
proton resonance at 3.97 ppm had the opposite effect, with only the
arylamine resonance at 11.68 ppm exhibiting a detectable NOE
(1.8%). NOE resulting from saturation of each of the arylamine
proton resonances were consistent with these results, although a
small NOE was observed between the two arylamine protons (each at
0.4%). Other evidence for the hydrazone form came from detection
of very small peaks, when recorded with higher signal to noise ratio
(number of scans516 000), which are attributed to 15N satellites. All
expected 15N satellites were detected with relative intensities
consistent with the known natural abundance of 15N. There was
a slight isotope effect on chemical shift in the usual direction (upfield
shift of 0.0006 ppm) for both 15N-coupled protons. The 15N-coupled
resonances were much narrower because of the absence of 14N
Photodecomposition of PY74
The exposure of PY74 to artificial sources of light is problematic
because of the low solubility of PY74 in many solvents. We have
found that PY74 is more soluble in THF and methylene chloride
compared with other common solvents. Argon-purged solutions of
PY74 in THF (approximately 40 lg/mL; »0.10 mM) were exposed
to simulated solar light at 0.5 m for different lengths of time in 2
mm cuvettes as described in the Materials and Methods. The
spectra of the solutions in the cuvettes were obtained and are
shown in Fig. 3. PY74 has an absorption band with a maximum at
416 nm, and exposure to the light decreased the absorbance at
a rate of approximately 0.42 AU/h under these conditions. During
photodecomposition the maximum absorption band remained at
416 nm, indicating retention of a major chromophore in the visible
portion of the spectrum. The absorption spectra did not form
isobestic points, which indicated that the PY74 was photo-
decomposing into products that were also labile. The decrease in
absorbance at 416 nm was accompanied by an increase in
absorbance at 280–295 nm.
Samples of PY74 saturated in THF (»0.25 mM) were exposed to
simulated solar light and then subjected to analysis by HPLC to
determine the amount of PY74 remaining and to investigate any
decomposition products. The chromatography was monitored at
254 nm (Fig. 4). PY74 eluted at ;40 min and was essentially the
only compound in the unexposed solution. At 0.5 and 1 h,
1
quadrupole relaxation, and the magnitude of the JN–H coupling
constants (98.4 Hz, 14.74 ppm; 90.6 Hz, 11.65 ppm) is characteristic
of sp2 arylamine protons (27).

178 Yanyan Cui et al.

Photochemistry and Photobiology, 2004, 80 179
Figure 2. Positive-ion product mass spectrum of PY74 m/z 387 at
a collision energy of 15 V.
Figure 4. THF solutions saturated with PY74 (»0.25 mM) and purged with
argon were exposed to simulated solar light for up to 5 h. The solutions
were analyzed by HPLC under conditions where PY74 eluted at ;40 min.
The photoproducts that were isolated (a–f) are indicated in the figure, and
those labeled ‘‘x’’ were unstable upon isolation.
photodecomposition products were observed and eluted with
retention times between 32 and 37 min. Further exposure of the
solution (Fig. 4) to simulated solar light resulted in the generation
of products with decreased retention (i.e. increased polarity), with
the largest HPLC peaks eluting between 13 and 18 min.
case, all resonances from the aminoanisole ring were unambigu-
ously detected and assigned. The spectral parameters are internally
consistent with those of the corresponding ring in PY74. The amine
proton resonance associated with this ring was found in all
photoproducts, except that its chemical shift varied by more than
4 ppm. In most cases the small long-range couplings with the
aromatic protons were confirmed by decoupling. No resonances
from the nitrophenyl ring were detected in any of the products.
Thus, all products likely have the same common structural element
of a single aminoanisole moiety.
Isolation and identification of photodecomposition products
Exposure of PY74 to simulated solar light produced photodecom-
position products that were separated by HPLC as shown in Fig. 4.
The products from the 5 h exposure were initially isolated from PY74
using a silica gel column, with methylene chloride as the eluant. The
fractions containing products less retentive than PY74 (the products
that eluted between 13 and 18 min on HPLC) were then subjected to
purification through repetitive injection and collection using HPLC.
A total of 11 photolysis products were isolated, purified and
subjected to NMR and MS analysis. Three of these photolysis
products were unambiguously characterized. The peaks indicated by
the ‘‘x’’ in Fig. 4 were not stable and tended to decompose primarily
to Photoproducts ‘‘a’’ and ‘‘b’’ (data not presented).
The NMR spectrum of Photoproduct ‘‘a’’ (Fig. 5) indicated the
presence of a methylene group, and the acetyl group proton
resonance was shifted upfield compared with PY74. This is
consistent with the mass spectral data (Fig. 6), which indicated that
the product was N-(2-methoxyphenyl)-3-oxobutanamide. The 2.6
ppm upfield shift for the amine proton compared with PY74
reflects a weakening of the intramolecular hydrogen bond due to
the reduction in planar character. The proposed compound is
available commercially as o-acetoacetanisidide and its HPLC
retention and NMR spectrum (not shown) matched that of
Photoproduct ‘‘a.’’
Each of the stable photoproducts (a–f) was analyzed by 1H-
NMR spectroscopy, and the results are listed in Table 1. In each
Examination of the structure of Photoproduct ‘‘a’’ by MS in
a positive-ion mode revealed a protonated molecule at m/z 208
(Fig. 6). The positive-ion fragment ion at m/z 190 is consistent with
the loss of water from o-acetoacetanisidide. Fragmentation of the
C1–C2 bond resulted in the m/z 150 fragment, whereas amide bond
breakage resulted in the m/z 124 fragment. As a result, the MS
supports that Photoproduct ‘‘a’’ is N-(2-methoxyphenyl)-3-oxobu-
tanamide (o-acetoacetanisidide).
The NMR spectrum of Photoproduct ‘‘b’’ is shown in Fig. 7. The
most downfield part of the spectrum contains two resonances that
each integrate as a single proton. The resonance assigned to C1–NH
has a chemical shift similar to that in PY74, suggesting retention of
the hydrogen bonding. The other resonance (17.01 ppm), however, is
Figure 3. PY74 (»0.1 mM) in THF (purged with argon) in 2 mm cuvettes
was exposed to simulated solar light for the times indicated. The
absorptivities of the solutions were determined from these spectra.

180 Yanyan Cui et al.
Figure 5. The 500 MHz 1H-NMR spectrum of Photoproduct ‘‘a’’ in
CD2Cl2 is shown with resonance assignments.
Figure 7. The 500 MHz 1H-NMR spectrum of Photoproduct ‘‘b’’ in
CD2Cl2 is shown with resonance assignments.
well downfield of that in PY74, is less broadened (linewidth ca 2.4
Hz) and is consistent with a hydrogen-bonded oxime.
and m/z 109 (Fig. 10). The fragmentation pattern, consistent with
the NMR analysis, supports the structure of Photoproduct ‘‘c’’ as
N,N0-bis(2-methoxyphenyl)urea as shown in Fig. 10. This structure
is additionally supported by negative-ion MS (data not shown).
Mass spectrometric examination of the final three purified
photoproducts (d–f, Fig. 4) indicates parent masses greater than
PY74 in increments consistent with the addition of THF, suggesting
the participation of THF in the photodecomposition and incorpo-
ration of all or some of the solvent molecule in the photoproduct.
NMR analysis likewise indicates the addition of aliphatic protons,
and the aminoanisole fragment was present in each case (Table 1).
We were unable to identify these secondary products.
Mass spectral analysis of Photoproduct ‘‘b’’ (Fig. 8) was used to
confirm the structure as 2-(hydroxyimine)-N-(2-methoxyphenyl)-3-
oxobutanamide. The product had a protonated molecule of m/z 237
in the positive-ion mode. The two primary fragmentation products
were from breakage of the C1–C2 bond to a product with m/z 150
or loss of the primary amine and acetyl group to yield a product at
m/z 177. Minor products included fragmentation of the acetyl
group (m/z 195) or cleavage of the amide bond (m/z 122). The mass
spectrometric data are fully consistent with 2-(hydroxyimine)-N-
(2-methoxyphenyl)-3-oxobutanamide as the structure of the second
photoproduct.
The NMR spectral analysis of Photoproduct ‘‘c’’ indicated that
the amine resonance at 7.12 ppm was well upfield of that of the other
photoproducts and of PY74 (Table 1, Fig. 9). This is consistent with
a structure that has lost the intramolecular hydrogen bond. No acetyl
protons were detected, and no other protons aside from the usual
proton resonances from the aminoanisole fragment were detected.
MS was used to assign the structure of Photoproduct ‘‘c.’’ A full-
scan mass spectrum confirmed the protonated molecule at m/z 273
because a M þ Na was detected at m/z 295. Fragmentation of the
m/z 273 protonated molecule led to products at m/z 150, m/z 124
DISCUSSION
In this report, we have shown that PY74 is present in tattoo inks
obtained from commercial sources within the United States. The
presence of PY74 in U.S. tattoo inks and the presence of PY74 in
Figure 6. Product ion mass spectrum (m/z 208) of Photoproduct ‘‘a’’ in the
positive-ion mode is shown. The product ions were obtained in positive-ion
mode at a collision energy of 10 V.
Figure 8. Product ion mass spectrum (m/z 237) of Photoproduct ‘‘b’’ in the
positive-ion mode is shown. The product ions were obtained in positive-ion
mode at a collision energy of 10 V.

Photochemistry and Photobiology, 2004, 80 181
Figure 10. Positive-ion product mass spectrum of Photoproduct ‘‘c’’ (m/z
273) with a collision energy of 20 V.
Figure 9. The 500 MHz 1H-NMR spectrum of Photoproduct ‘‘c’’ in
CD2Cl2 is shown with resonance assignments.
efficiency of this photochemical process was beyond the scope of
this study; however, our data suggest some insights into the
mechanism of photochemical decomposition of PY74.
tattoo inks obtained in Europe (17,18) suggest that PY74 is a
commonly used pigment in tattoo inks.
The excitation of an electron in PY74 results in a Type-I
photochemical reaction, resulting in either the addition of PY74 to
a hydrogen source (THF) or the photolysis of PY74. The first
products that are formed after illumination of PY74 in THF, eluting
between 32 and 39 min on HPLC (Fig. 4), decomposed upon
isolation but were examined by online HPLC/MS-MS (data not
presented) and had molecular weights equal to PY74 plus THF.
The persistence of these products with time (Fig. 4) suggests that
they are not precursors to the other photoproducts. The isolation of
o-acetoacetanisidide and 2-(hydroxyimine)-N-(2-methoxyphenyl)-
3-oxobutanamide as photodecomposition products of PY74
suggests that the bonds adjacent to the hydrazone are particularly
susceptible to photochemical photolysis. This is consistent with the
observations reported by Ba¨umler et al. (23,24), where the laser-
induced photolysis of PR22 and PR9 occurred at the azo groups,
generating 2-amino-4-nitrotoluene with PR22.
The NMR spectroscopic analyses of PY74 in CD2Cl2 indicate
that PY74 exists predominantly in a hydrazone form. Tautomerism
plays an important role in the chemical character of many monoazo
compounds (28). The physical–chemical properties of many
aromatic azo compounds have been reported and used to determine
their tautomeric equilibria in different solvents and the role of
tautomerism in excited-state decay. For instance, Sahu et al. (29)
reported that the tautomerism between the azo and hydrazone
forms of 2-[(2-hydroxy-4-methyl)azo]benzoic acid results in
a spectral shift of 100 nm (azo 5 380 nm; hydrazone 5 480 nm)
and is dependent on the solvent, pH and temperature. With 2-[(2-
hydroxy-4-methyl)azo]benzoic acid, addition of base facilitates
tautomerism between the azo form and the hydrazo form by
allowing the ortho-hydroxy hydrogen to hydrogen bond to the azo
nitrogen. In a study using 1-phenylazo-2-naphthol and 2-phenyl-
azo-1-naphthol, Joshi et al. (30) reported that photoexcitation of
the azo form results in excited-state intramolecular proton transfer
to the excited hydrazone form, which then undergoes fluorescence
decay to the ground state. Any tautomerism in PY74 between the
azo and hydrazone forms would not involve loss of aromaticity, as
with the azobenzenes or azonaphthols mentioned above, where the
azo tautomer would be favored. Other azoacetoacetanilides such as
Pigment Yellow 1 (also known as Hansa Yellow G) or bis-
azoacetoacetanilides such as Pigment Yellow 12 and 13 exist
exclusively in the hydrazone form, being stabilized through the
formation of intramolecular hydrogen bonding (28). As a result, the
observation that PY74 exists primarily in the hydrazone form is
consistent with these previous observations.
It is also possible that PY74 could undergo a Type-II
photochemical reaction where irradiation and intersystem cross-
over results in an excited triplet state, which then transfers the
energy to oxygen, generating singlet oxygen (1O2). The rate of
photodecomposition of PY74 in THF is greatly reduced in the
presence of oxygen (data not presented). This suggests that under
conditions where oxygen is present, a Type-II mechanism
competes with the Type-I photochemical lysis. Determination of
the quantum efficiency of PY74 photolysis and the kinetics of 1O2
generation was beyond the scope of this initial study; however,
1
because of the potential of O2 to cause cytotoxic and genotoxic
events in vivo, the potential of the Type-I photolysis products of
tattoo inks and Type-II–generated 1O2 to cause toxicity in the skin
of tattooed individuals should be explored.
In this study we demonstrate that PY74 dissolved in THF
photodecomposes in the presence of simulated solar light. PY74
has no appreciable absorbance between 250 and 350 nm and
a maximum absorption band at 416 nm, suggesting that either
visible light or fluorescent tubes emitting UV-A light could initiate
the photodecomposition of this pigment. This study was designed
to identify the products after photochemical decomposition of
PY74 in THF and was not designed to imitate the conditions that
would exist for a tattoo pigment in the skin where oxygen would be
present and the solid pigment could be in either an aqueous or
a lipid environment. Investigation of the precise mechanism or
The NMR analysis of the PY74 used in this study showed the
presence of about 0.4% o-acetoacetanisidide and 1% 2-(hydrox-
yimine)-N-(2-methoxyphenyl)-3-oxobutanamide (molar percen-
tages). The manufacture of PY74 would most likely occur
through diazotization of 2-amino-5-nitroanisole to the correspond-
ing diazonium salt and condensation of the diazonium salt with o-
acetoacetanisidide. This suggests that the purity of PY74 used in
the manufacture of tattoo inks is not certain, and the extent of
contamination and toxicity of the PY74 used in these and other
products should be explored. An additional complication that

182 Yanyan Cui et al.
should be considered is the presence of inorganic materials such as
titanium dioxide in many inks (17,31).
which they reported that exposure of PR22 in acetonitrile to 532
nm Nd:YAG laser light resulted in the formation of 2-methyl-5-
nitroaniline, a known carcinogen. PR22 is a monoazo aromatic
pigment with many similarities to PY74. Ba¨umler et al. (24)
recently extended this work and obtained similar results with PR9.
Their report and our studies suggest that photochemical activation
of tattoo pigments in situ during sunlight exposure, tanning or laser
removal may generate toxic products. Although the light intensity
inside tattooed skin will differ from that of the incident light on the
surface of the skin, the potential for photodecomposition of the
tattoo pigment still exists.
There has been a dramatic change in the cultural acceptance of
tattooing in the United States in recent years. In the past, tattooing
was primarily restricted to small subsets of the population.
Tattooing has moved into mainstream culture during the past
decade, with widespread acceptance and practice among U.S.
youth. Tattoos in the mid-1900s were essentially black or blue,
with the occasional color coming from inorganic compounds such
as oxides of iron and mercury. As reported by Lehmann and
Pierchalla (17), by Ba¨umler et al. (18,24) and in this article,
monoazo and bisazo dyes are being used as pigments in modern
tattoo inks. The toxicological safety of these materials in the
absence or presence of sunlight has not been evaluated.
In summary, our study suggests that the use of PY74 as
a pigment in tattoo inks may result in an increased risk of toxicity
at the tattoo site from photodecomposition products of PY74,
contaminating impurities in the pigments, or Type-II photochem-
ical generation of 1O2. Studies are in progress to address the
toxicity of PY74 in vivo in mice tattooed with PY74 and exposed to
simulated solar light.
Tattooing has not been without risk and consequences and
historically has included (1) infection or contraction of disease;
(2) dermatological reactions, some of which may be severe or
other significant allergic reactions to chemicals in the tattoo; and
(3) complications after removal of the tattoo. Perhaps the greatest
potential risk from tattooing is the spread of diseases, e.g.
bacterial infections, hepatitis and HIV infections (32), through
improperly cleaned or sterilized needles and equipment (33–39).
In the United States, the practice of tattooing is regulated in most
states through the state health departments (40), where a major
concern is to practice proper hygiene in order to minimize the
spread of diseases. The U.S. FDA and other agencies have posted
Web sites for consumers regarding the safety and regulation of
tattooing (22,41).
Acknowledgements—The contents of this article do not necessarily reflect
the views and policies of the U.S. Food and Drug Administration, and
the mention of trade names or commercial products does not constitute
endorsement or recommendation for use. We thank the Office of Food
Additive Safety, Center for Food Safety and Applied Nutrition, U.S.
Food and Drug Administration, for funding parts of this study. Y. Cui
and N. V. Gopee were supported on an appointment through the Oak
Ridge Associated Universities Research Program at National Center for
Toxicological Research. The National Toxicology Program Center for
Phototoxicology is supported in part by an interagency agreement (IAG
224-93-0001) between the U.S. Food and Drug Administration and the
U.S. National Institute for Environmental Health Sciences. We acknowl-
edge assistance by E. Willingham and L. S. Von Tungeln in the prepara-
tion of the manuscript and figures, respectively.
The second potential risk from tattooing is dermatological
reaction to ingredients in the tattoo. There have been numerous
reports of adverse reactions to tattoos (42–45), especially after
exposure of the tattoos to sunlight (32,46–50). Cadmium sulfide
has been used as a yellow-orange–colored pigment in the past in
some tattoos and has been reported to induce severe allergic
responses in skin after exposure to sunlight (47). However,
tattoos containing red colors are most frequently associated with
skin disorders (48–60). In addition, there have been reports of
basal or squamous cell carcinomas and melanomas arising within
tattoos (61–67), suggesting direct chemical or photochemical
participation in the etiology of these diseases. Sangueza et al.
(68) reported on a patient who developed multiple pseudolym-
phomas at a tattoo site, which progressed for a period of 4 years
into malignant and monoclonal B-cell large-cell lymphoma. The
possibility that tattoo pigments elicit adverse biological responses
that could lead to cancer growth is intriguing and anecdotally
supported by clinical observations; however, because no
epidemiological studies have been conducted to date to examine
the role of tattooing in skin cancer, the participation of tattooing
or tattoo pigments in the etiology of any skin disease remains
unclear and unsubstantiated.
REFERENCES
1. Mazza Jr., J. F. and C. Rager (1993) Advances in cosmetic micro-
pigmentation. Plast. Reconstr. Surg. 92, 750–751.
2. Mutalik, S. and A. Ginzburg (2000) Surgical management of stable
vitiligo: a review with personal experience. Dermatol. Surg. 26, 248–254.
3. Lane, K. L., R. Vallera, K. Washington and M. R. Gottfried (1996)
Endoscopic tattoo agents in the colon. Tissue responses and clinical
implications. Am. J. Surg. Pathol. 20, 1266–1270.
4. Nizam, R., N. Siddiqi, S. K. Landas, D. S. Kaplan and P. G. Holtzapple
(1996) Colonic tattooing with India ink: benefits, risks, and alternatives.
Am. J. Gastroenterol. 91, 1804–1808.
5. Shatz, B. A., L. B. Weinstock, P. E. Swanson and E. P. Thyssen (1997)
Long-term safety of India ink tattoos in the colon. Gastrointest. Endosc.
45, 153–156.
6. McArthur, C. S., S. Roayaie and J. D. Waye (1999) Safety of
preoperation endoscopic tattoo with India ink for identification of
colonic lesions. Surg. Endosc. 13, 397–400.
7. Lee, J. G., A. H. Low and J. W. Leung (2000) Randomized comparative
study of indocyanine green and India ink for colonic tattooing: an
animal survival study. J. Clin. Gastroenterol. 31, 233–236.
8. Price, N., M. R. Gottfried, E. Clary, D. C. Lawson, J. Baillie, K.
Mergener, C. Westcott, S. Eubanks and T. N. Pappas (2000) Safety and
efficacy of India ink and indocyanine green as colonic tattooing agents.
Gastrointest. Endosc. 51, 438–442.
The third potential risk from tattooing is the process of
removal of the tattoo. Tattoos can be removed by dermabrasion,
salabrasion, surgical removal of the skin or laser treatment
(16,69,70). Each of these processes has procedural advantages
and disadvantages including granuloma formation, keloid
formation, unwanted color change and incomplete removal
(69,71–73).
9. Brown, K. M., P. Perlmutter and R. J. McDermott (2000) Youth and
tattoos: what school health personnel should know. J. Sch. Health 70,
355–360.
10. Drews, D. R., C. K. Allison and J. R. Probst (2000) Behavioral and self-
concept differences in tattooed and nontattooed college students.
Psychol. Rep. 86, 475–481.
In this article, we present new evidence that an additional
potential risk associated with tattoos may be from the toxicity of
photochemical products of the tattoo pigments. The first report of
this potential toxicity came from the work of Ba¨umler et al. (23), in
11. Marcoux, D. (2000) Appearance, cosmetics, and body art in
adolescents. Dermatol. Clin. 18, 667–673.

Photochemistry and Photobiology, 2004, 80 183
12. Ceniceros, S. (1998) Tattooing, body piercing, and Russian roulette.
J. Nerv. Ment. Dis. 186, 503–504.
13. Greif, J. and W. Hewitt (1998) The living canvas. Adv. Nurse Pract. 6,
26–31.
14. Armstrong, M. L., Y. Masten and R. Martin (2000) Adolescent
pregnancy, tattooing, and risk taking. MCN Am. J. Matern. Child Nurs.
25, 258–261.
15. Armstrong, M. L., K. P. Murphy, A. Sallee and G. Watson (2000)
Tattooed army soldiers: examining the incidence, behavior, and risk.
Mil. Med. 165, 135–141.
medium-security federal penitentiary. Queen’s University HIV Prison
Study Group. Queen’s J. Med. 93, 113–119.
37. Silverman, A. L., J. S. Sekhon, S. J. Saginaw, D. Wiedbrauk, M.
Balasubramaniam and S. C. Gordon (2000) Tattoo application is not
associated with an increased risk for chronic viral hepatitis. Am. J.
Gastroenterol. 95, 1312–1315.
38. Haley, R. W. and R. P. Fischer (2001) Commercial tattooing as
a potentially important source of hepatitis C infection. Clinical
epidemiology of 626 consecutive patients unaware of their hepatitis C
serologic status. Medicine 80, 134–151.
16. Anderson, R. R. (2001) Regarding tattoos: is that sunlight, or an
oncoming train at the end of the tunnel? Arch. Dermatol. 137, 210–212.
17. Lehmann, G. von and P. Pierchalla (1988) Ta¨towierungsfarbstoffe.
Derm. Beruf Umwelt 36, 152–156.
39. Post, J. J., K. A. Dolan, L. R. Whybin, I. W. Carter, P. S. Haber and
A. R. Lloyd (2001) Acute hepatitis C virus infection in an Australian
prison inmate: tattooing as a possible transmission route. Med. J. Aust.
174, 183–184.
18. Ba¨umler, W., E. T. Eibler, U. Hohenleutner, B. Sens, J. Sauer and
M. Landthaler, (2000) Q-switched laser and tattoo pigment: first results
of the chemical and photophysical analysis of 41 compounds. Lasers
Surg. Med. 26, 13–21.
19. Danish Environmental Protection Agency (2003) Chemicals in
Consumer Products. Available at: http://www.mst.dk/chemi/
01080000.htm. Accessed on 1 June 2004.
20. Danish Environmental Protection Agency (2003) Survey No. 2:
Investigation of Pigments in Tattoo Colours, January 2002. Available
at: http://www.mst.dk/chemi/01080200.htm. Accessed on 1 June 2004.
21. Food and Drug Administration (1994) Tattoos. FDA Medical Bulletin.
24, 8. Published 8 May 1994, by the U.S. Department of Health and
Human Services, Public Health Service, Food & Drug Administration,
Rockville, MD.
40. Braithwaite, R. L., T. Stephens, C. Sterk and K. Braithwaite (1999)
Risks associated with tattooing and body piercing. J. Public Health
Policy 20, 459–470.
41. Center for Disease Control and Prevention (2004) Can I get HIV from
getting a tattoo or through body piercing? Available at: http://www.
cdc.gov/hiv/pubs/faq/faq27.htm. Accessed 1 June 2004.
42. Tazelaar, D. J. (1970) Hypersensitivity to chromium in a light-blue
tattoo. Dermatologica 141, 282–287.
43. Rubegni, P., M. Fimiani, G. de Aloe and L. Andreassi (2000) Lichenoid
reaction to temporary tattoos. Contact Dermatitis 42, 117–118.
44. Zirkin, H. J., I. Avinoach and P. Edelwitz (2001) A tattoo and localized
lymphadenopathy: a case report. Cutis 67, 471–472.
45. Mahalingam, M., E. Kim and J. Bhawan (2002) Morphea-like tattoo
reaction. Am. J. Dermatopathol. 24, 392–395.
22. FDA Center for Food Safety and Applied Nutrition (2004) Tattoos and
Permanent Makeup. Available at: http://www.cfsan.fda.gov/;dms/
cos-204.html. Accessed on 1 June 2004.
23. Ba¨umler, W., R. Vasold, N. Naarmann, D. Fischer, B. Sens, B. Ko¨nig
and M. Landthaler (2002) Q-switch laser and tattoo pigments—
a chemical analysis of laser induced decomposition products. Am. Soc.
Laser Med. Surg. Abstr. 2002, 78.
24. Ba¨umler, W., R. Vasold, B. Ko¨nig and M. Landthaler (2004) Q-
switched laser and tattoo pigments, a chemical analysis of laser induced
decomposition compounds. Toxicologist 78(S1), 1079.
25. Howard, P. C., R. L. Sams II, J. R. Bucher and W. T. Allaben (2002)
Phototoxicology and photocarcinogenesis at the U. S. Food and Drug
Administration’s National Center for Toxicological Research. J. Food
Drug Anal. 10, 252–257.
46. Goldstein, N. (1967) Mercury-cadmium sensitivity in tattoos. A
photoallergic reaction in red pigments. Ann. Intern. Med. 67, 984–989.
47. Nguyen, L. Q. and H. B. Allen (1979) Reactions to manganese and
cadmium in tattoos. Cutis 23, 71–72.
48. Verdich, J. (1981) Granulomatous reaction in a red tattoo. Acta
Dermato-Venereol. 61, 176–177.
49. Bendsoe, N., C. Hansson and O. Sterner (1991) Inflammatory reactions
from organic pigments in red tattoos. Acta Dermato-Venereol. 71, 70–73.
50. Duke, D., S. S. Urioste, J. S. Dover and R. R. Anderson (1998) A reaction
to a red lip cosmetic tattoo. J. Am. Acad. Dermatol. 39, 488–490.
51. Juhlin, L. and S. Ohman (1968) Allergic reactions to mercury in red
tattoos and in mucosa adjacent to analgam. Acta Dermato-Venereol. 48,
103–105.
52. Fields, J. P., W. D. Little Jr. and P. E. Watson (1968) Discoid lupus
erythematosus in red tattoos. Arch. Dermatol. 98, 667–669.
53. Taaffe, A., A. G. Knight and R. Marks (1978) Lichenoid tattoo
hypersensitivity. Br. Med. J. 1(6113), 616–618.
26. Evans, F. E., P. P. Fu, T. Cairns and S. K. Yang (1981) Long-range
coupling constants for structural analysis of complex polycyclic
aromatic hydrocarbons by high-field proton magnetic resonance
spectroscopy. Anal. Chem. 53, 558–560.
54. Clarke, J. and M. M. Black (1979) Lichenoid tattoo reactions. Br. J.
27. Evans, F. E. and R. A. Levine (1986) Conformation and configuration
at the central amine nitrogen of a nucleotide adduct of the carcinogen
2-(acetylamino)flourene as studied by 13C and 15N NMR spectroscopy.
J. Biomol. Struct. Dyn. 3, 923–934.
Dermatol. 100, 451–454.
55. Taaffe, A. and E. H. Wyatt (1980) The red tattoo and lichen planus. Int.
J. Dermatol. 19, 394–396.
56. Bagley, M. P., R. A. Schwartz and W. C. Lambert (1987) Hyperplastic
reaction developing within a tattoo. Granulomatous tattoo reaction, prob-
ably to mercuric sulfide (cinnabar). Arch. Dermatol. 123, 1557–1561.
57. Sowden, J. M., J. P. Byrne, A. G. Smith, C. Hiley, V. Suarez, B.
Wagner and D. N. Slater (1991) Red tattoo reactions: X-ray
microanalysis and patch-test studies. Br. J. Dermatol. 124, 576–580.
58. Sowden, J. M., P. H. Cartwright, A. G. Smith, C. Hiley and D. N. Slater
(1992) Sarcoidosis presenting with a granulomatous reaction confined
to red tattoos. Clin. Exp. Dermatol. 17, 446–448.
59. Tresukosol, P., S. Ophaswongse and P. Kullavanijaya (1997) Cutaneous
reaction to cosmetic lip tattooing. Contact Dermatitis 36, 176–177.
60. Pauluzzi, P., M. Giordani, G. F. Guarneri and M. Pascone (1998)
Chronic eczematous reaction to red tattoo. J. Eur. Acad. Dermatol.
Venereol. 11, 187–188.
28. Christie, R. M. (2000) Colour Chemistry, 205 p. Royal Society of
Chemistry, Cambridge, UK.
29. Sahu, M., M. B. Saha and S. C. Bera (1995) Azo-hydrazo
tautomerization and cis-trans isomerization of 2-carboxy-29-hydroxy-
49-methyl azobenzene. J. Photochem. Photobiol. A: Chem. 89, 19–24.
30. Joshi, H., F. S. Kamounah, C. Gooijer, G. van der Zwan and L.
Antonov (2002) Excited state intramolecular proton transfer in some
tautomeric azo dyes and Schiff bases containing an intramolecular
hydrogen bond. J. Photochem. Photobiol. A: Chem. 152, 183–191.
31. Timko, A. L., C. H. Miller, F. B. Johnson and E. Ross (2001) In vitro
quantitative chemical analysis of tattoo pigments. Arch. Dermatol. 137,
143–147.
32. Wilkes, T. D. I. (1986) The complications of dermal tattooing.
Ophthalmic Plast. Reconstr. Surg. 2, 1–6.
61. Kirsch, N. (1972) Malignant melanoma developing in a tattoo. Int. J.
33. Long, G. E. and L. S. Rickman (1994) Infectious complications of
tattoos. Clin. Infect. Dis. 18, 610–619.
Dermatol. 11, 16–20.
62. Wolfort, F. C., J. E. Hoopes, H. S. Filtzer and T. C. Cochran (1974)
Superficial melanoma in a tattoo. Br. J. Plast. Surg. 27, 303–304.
63. Bashir, A. H. (1976) Basal cell carcinoma in tattoos: report of two
cases. Br. J. Plast. Surg. 29, 288–290.
64. Bartal, A. H., Y. Cohen and E. Robinson (1980) Malignant melanoma
arising at tattoo sites used for radiotherapy field marking. Br. J. Radiol.
56, 913–914.
34. Satchithananda, D. K., J. Walsh and P. M. Schofield (2001) Bacterial
endocarditis following repeated tattooing. Heart 85, 11–12.
35. Chunlertrith, K., W. Sukeepaisarnjaroen, P. Mairiang, Y. Urwijitaroon,
K. Takase, T. Yamauchi, H. Yoshimura and Y. Tameda (2000) Clinico-
epidemiology of hepatitis C viral infection in northeastern Thailand.
Southeast Asian J. Trop. Med. Public Health 31, 273–276.
36. Ford, P. M., M. Pearson, P. Sankar-Mistry, T. Stevenson, D. Bell and
J. Austin (2000) HIV, hepatitis C and risk behaviour in a Canadian
65. Earley, M. J. (1983) Basal cell carcinoma arising in tattoos: a clinical
report of two cases. Br. J. Plast. Surg. 36, 258–259.

184 Yanyan Cui et al.
66. Weiner, D. A. and R. K. Scher (1987) Basal cell carcinoma arising in
70. Hu, X. H., W. A. Wooden, S. J. Vore, M. J. Cariveau, Q. Fang and
G. W. Kalmus (2002) In vivo study of intradermal focusing for tattoo
removal. Lasers Med. Surg. 17, 154–164.
a tattoo. Cutis 39, 125–126.
67. Soroush, V., A. W. Gurevitch and S. K. Peng (1997) Malignant
71. Peach, A. H. S., K. Thomas and J. Kenealy (1999) Colour shift
following tattoo removal with Q-switched Nd-YAG laser (1064/532).
Br. J. Plast. Surg. 52, 482–487.
72. Ross, E. V., S. Yashar, N. Michaud, R. Fitzpatrick, R. Geronemus, W.
D. Tope and R. R. Anderson (2001) Tattoo darkening and nonresponse
after laser treatment. Arch. Dermatol. 137, 33–37.
73. Varma, S., N. A. Swanson and K. K. Lee (2002) Tattoo ink darkening
of a yellow tattoo after Q-switched laser treatment. Clin. Exp. Dermatol.
27, 461–463.
melanoma in a tattoo: case report and review of literature. Cutis 59,
111–112.
68. Sangueza, O. P., S. Yadav, C. R. White Jr. and R. M. Braziel (1992)
Evolution of B-cell lymphoma from pseudolymphoma. A multidisci-
plinary approach using histology, immunohistochemistry, and Southern
blot analysis. Am. J. Dermatopathol. 14, 408–413.
69. Kuperman-Beade, M., V. J. Levine and R. Ashinoff (2001) Laser
removal of tattoos. Am. J. Clin. Dermatol. 2, 21–25.