Microchemical Journal 127 (2016) 36–45
Contents lists available at ScienceDirect
Microchemical Journal
j ourna l homepage: www.e lsev ie r .com/ locate /mic rocVibrational spectroscopic study on degradation of alizarin carmine☆Lea Legan a,⁎, Klara Retko a,b, Polonca Ropret a,c
a Research Institute, Conservation Centre, Institute for the Protection of Cultural Heritage of Slovenia, Poljanska 40, 1000 Ljubljana, Slovenia
b Faculty of Chemistry and Chemical Technology, University of Ljubljana, Večna pot 113, SI-1000 Ljubljana, Slovenia
c Museum Conservation Institute, Smithsonian Institution, 4210 Silver Hill Rd., Suitland, MD 20746, USA☆ Selected papers presented at TECHNART 2015 Con
27–30, 2015.
⁎ Corresponding author.
E-mail addresses: lea.legan@zvkds.si (L. Legan), klara.r
polona.ropret@zvkds.si (P. Ropret).
http://dx.doi.org/10.1016/j.microc.2016.02.002
0026-265X/© 2016 Elsevier B.V. All rights reserved.a b s t r a c ta r t i c l e i n f oArticle history:
Received 11 January 2016
Accepted 3 February 2016
Available online 11 February 2016Identification of organic dyes togetherwith possible degradation products is often complicated in the field of cul-
tural heritage. Themain focus of thiswork is utilization and comparison between non-invasive reflection infrared
spectroscopy, transmission infrared spectroscopy, and surface-enhancedRaman spectroscopy (SERS), aminimal-
ly invasive technique in order to understand the ageing behaviour of paint layers containing organic dye alizarin
carmine (alizarin red S, sodium alizarin sulfonate, ARS) in lipid and proteinaceous binder. The results obtained
with both Fourier transform infrared (FTIR) spectroscopies highlight a significant degradation of ARS paint layers
where changes in hydroxyl, sulfonate, and carbonyl groups of alizarin red S were observed. In most of the cases,
the intensity of IR bands belonging to the degraded alizarin red Swas soweak or shifted thatwouldmake its pos-
itive identification questionable in an unknown sample. After all, it was apparent that SERS is a powerful tech-
nique to detect even the trace ARS molecules. Furthermore, degradation was even more pronounced for the
paint layers of ARS combined with linseed oil. Namely, a transparent white layer of sodium sulfate was formed
above the aged ARS glaze layer.
© 2016 Elsevier B.V. All rights reserved.Keywords:
Alizarin carmine
Alizarin red S
Degradation
Reflection FTIR
SERS
Artworks1. Introduction
Organic colorants have been an important part of artist's pallet since
ancient times. They could be extracted from natural sources such as
plants (e.g., madder roots) or animals (e.g., cochineal) [1,2]. Anthraqui-
nones have been frequently used as important coloring materials for
different artistic objects, such as archaeological textiles [3,4], manu-
scripts, drawings, or paintings [5]. For glaze layers, they were used for
giving depth, to enrich shadows, or to make dull colors more vivid.
After the synthesis of 1,2-dihydroxyanthraquinone (alizarin), synthetic
materials in many aspects substituted the natural ones. Alizarin carmine
(1,2-dihydroxyanthraqinone-3-sulfonate; alizarin red S, ARS) (see Fig. 1)
was discovered in 1871 byGraebe and Libermann as awater-soluble ver-
sion of naturally occurring reddish dye alizarin [6]. ARS as awater-soluble
dye can exist in aqueous solutions in three different forms: in neutral
form, anionic, and dianionic form [7]. Hydroxyanthraquinones have a
very wide range of applications. Beside dye industry, they are also used
in histochemistry [8], medicine, food industry [9], etc. Alizarin red S
shows a potential to be used as an organic analytical reagent for the de-
termination of inorganic substances (such as Ca, Al) [10,11], biological ac-
tive compounds, photosentisizer [12], and selective depressant in theference, Catania (Italy), April
etko@zvkds.si (K. Retko),separation of minerals [7,13]. Samples of cultural heritage show great
heterogeneity and several different microscopic, spectroscopic, chro-
matographic, etc., methods have been developed for their characteriza-
tion, which is together with the detection of their degradation products
very challenging. Regarding the priceless value of cultural heritage ob-
jects and therefore limited sampling, there is a demanding task in conser-
vation science in development of scientificmethods that could be applied
in minimally invasive and destructive manner with no or minimal inter-
ventions to keep the integrity of awork of art. Vibrational spectroscopy is
frequently used for the spectral characterization of artistmaterials [14,15]
as well as for monitoring of degradation processes of different organic
colorants [16–18]. It was found that organic pigments, namely, alizarin,
permanent red, and phthalocyanine green mixed with linseed oil, de-
graded under the UV light exposure, specifically breaking both aromatic
rings in the organic pigments' and binder's unsaturated fatty acid C = C
bonds [16]. The assessment of ageing processes occurring in natural
dyes (indigo, madder, curcumin, carmine acid, and dragon's blood)
used in oil and watercolor paintings was made by Koperska et al. [17].
It was shown that the degradation of dyes is dependent on the molecule
symmetry. The more is a molecule symmetrical, the less is prone to deg-
radation [17]. Irreversible photoassisted degradation of dye, alizarin red S,
under UV light radiation and the presence of photocatalytic TiO2were ex-
amined by Liu et al. [18]. Degradation products, such as phthalic acid and
other carbonyl species, suggest the decay of C-C bond near the C = O
group of ARS molecule (see Fig. 1), while remaining multihydroxylated
species decompose further to smaller fragments (e.g., CO2 and SO42−
ions) [18]. Furthermore, utilizing a conventional transmission FTIR
Fig. 1. Chemical structure of alizarin red S molecule.
37L. Legan et al. / Microchemical Journal 127 (2016) 36–45analysis numerous studies have been performed concerning themolecu-
lar information of organic dyes [14,19–21], but the method demands
sampling, which can result in the loss of information from the surface.
Non-invasive reflection FTIR measurements provide valuable informa-
tion about the molecular composition as well as degradation processes
of artworks' surfaces, although the interpretation of spectra is often
very challenging [22–24]. To our knowledge, the non-invasive reflection
FTIR spectroscopy has not been used for degradation studies of organic
glaze layers. On the other hand, the problemwith using FTIR spectrosco-
py for analysis of paint layers consisting of organic colorants derives from
the interfering signal of the binders and/or other inorganic components,
which could be present in other layers (for example, calcium carbonate
in the ground layer). In such cases, surface-enhanced Raman spectrosco-
py (SERS) can be utilized in order to gainmore information on the classi-
fication of the organic dyestuffs. It has already proved to be effective in
the investigations in the field of cultural heritage [25–27] and in charac-
terization of anthraquinones and its derivatives [28–35]. SERS is a sensi-
tive analytical technique as the signal of compounds with weak Raman
scattering properties in the vicinity of certain nanostructures is enhanced
in addition to reduction of fluorescence effects [36]. Several authors also
report on single-molecule detection [37,38].
This work focuses on the application of non-invasive reflection Fou-
rier transform infrared (FTIR) spectroscopy, transmission FTIR spectros-
copy, and minimally invasive surface-enhanced Raman spectroscopy
(SERS) for assessing degradation processes of paint layers of alizarin
carminebound in two binderswith different chemical composition (lin-
seed oil and egg tempera). In order to better understand the reflection
spectral anomalies of aged ARS paint layers, several techniques such
as conventional Raman, infrared spectroscopy, and electronmicroscopy
with energy dispersive spectroscopy (SEM /EDS) were complemented.
2. Experimental
2.1. Materials
Animal glue (No. 63028), chalk fromBologna, light (No. 58150), pig-
ment cinnabar (HgS; No. 42000) and organic dye alizarin carmine (aliz-
arin red s, sodium salt of 1,2-dihydroxyanthraquinone-3-sulfonate,
ARS; No. 94,150) were supplied by Kremer Pigmente GmbH & Co. KG.
Sun-bleached linseed oil was purchased from Lefrance & Bourgeois.
Eggs were bought from the local market. Other chemicals, which served
in preparation of HPC-based photoreduced silver colloid for SERS
analysis, namely, silver nitrate (AgNO3; No. 209139) and hydroxypropyl
cellulose (HPC; No. 191906) were provided by Sigma-Aldrich.
2.2. Preparation of model painting and artificial ageing
Two model paintings were prepared according to art technology
[39,40] in order to imitate the structure of a canvas painting. Linen can-
vaswas attached to awooden stretcher, sizedwith animal glue solution,
and impregnated with gesso ground. Pigment cinnabar was groundwith linseed oil and non-fatty egg tempera with pigment to binder
weight ratio of about 5:1 and 2:1, respectively, and applied on top of
the ground layer. After drying, a thin layer of colorant alizarin carmine,
mixed with the same binders as the cinnabar containing layer, was ap-
plied on top. Colorant alizarin carmine was mixed with linseed oil in
weight ratio of about 1:6 and with egg yolk binder in weight ratio 1:4.
One of the model paintings was then exposed to accelerated artificial
ageing in climatic chamberswith well-defined and controlled tempera-
ture, relative humidity, and light conditions. A metal halide lamp
(2000 W, OSRAM HQI-TS 2000 WDS; Osram GmbH, Germany),
equipped with window glass filters, was used for the simulation of
light radiation,whichwas a good simulation of daylight filtered through
glass. Light intensity was near zero in the λ range of 300–400 nm and
approximately 800 Wm−2 in the λ range of 400–800 nm.
Model painting was exposed to a metal halide lamp for a period of
30 days and followed by exposure in a climatic chamber with oscilla-
tions of temperature and relative humidity for another 30 days. One
ageing cycle in the climatic chamber lasted 210 min wherein the tem-
perature varied between 5 °C and 40 °C with an average velocity
0.56 °C/min and relative humidity varied from 20% to 90%.
2.3. Instrumentations
2.3.1. FTIR investigations
2.3.1.1. Transmission mode. FTIR transmission spectra were recorded
using a Perkin Elmer Spectrum 100 FTIR spectrophotometer coupled
to a Spotlight FTIR microscope equipped with nitrogen cooled
mercury-cadmium telluride (MCT) detector. The samples taken
from the model painting were placed between the windows of a di-
amond anvil cell and examined under microscope with aperture of
50 × 50 μm. The spectra were collected in the range between 4000
and 600 cm−1, at 4 cm−1 spectral resolution and with an average
of 64 spectral scans accumulated. Additionally, the transmission
FTIR spectra of pure colorant ARS were collected using a KBr pellet
method in the range of 4000–450 cm−1.
2.3.1.2. Reflection mode. Reflection FTIR analyses were carried out with a
portable Bruker ALPHA-R spectrometer, equipped with a dedicated re-
flection module, which allows contactless and non-destructive FTIR
analysis with a room temperature DLaTGS detector. The samples were
placed in front of the instrument at a distance of about 1 cm, and the in-
tegrated video camera provided the view of the sampling area of about
28 mm2. Reflection spectra were collected in situ in the range of 5500–
600 cm−1, at the spectral resolution 4 cm−1 over 160 scans. The back-
ground was acquired using a gold mirror as a reference sample. The
Kramers–Kronig algorithm was not applied in obtained total reflection
spectra due to contribution of specular and/or diffuse components of
reflected light, which can result in irregularities in the corrected spectra.
[22–24].
With transmission and reflectionmode, three different spectra were
obtained for each sample. All the spectra were then baseline corrected
and averaged using the OPUS 7.0 data collection software package
(Bruker, Germany). FTIR spectra are presented in absorbance and
pseudo-absorbance units.
2.3.2. Raman and SERS investigations
The spectra of the samples were recorded using a 785 nm and
514 nm laser excitation lines with a Horiba Jobin Yvon LabRAM HR800
Raman spectrometer coupled to an Olympus BXFM optical microscope.
The spectra were recorded using ×100 objective lens and/or ×50 long
working distance objective lens, and a 600 grooves/mm grating. A
multi-channel, air-cooled CCD detector was used, and the spectral
range was set between 300 and 1800 cm−1 for obtaining spectra with
λ0 = 514 nm (SERS and conventional Raman) and between 400 and
1800 cm−1 for conventional Raman measurement using λ0 = 785 nm.
Fig. 2. (a) Absorption FTIR spectrum of pure ARS in KBr; (b) Normal Raman (gray line)
(λ0 = 514 nm, 0.01 mW, 2 s exposure time) and SERS (black line) spectra of ARS
powder (λ0 = 514 nm, 0.6 mW, 5 s exposure time).
38 L. Legan et al. / Microchemical Journal 127 (2016) 36–45The wave number calibration was performed using a silicon wafer. For
the SERS analysis, a drop of a SERS substrate was deposited on a small
amount of powder samples (pure dye ARS) or on the surface of the
minute samples, taken from the model panels. All the SERS spectra
were averaged (15 spectra) using the OPUS 7.0 data collection software
package (Bruker, Germany). The SERS substrate (silver nanoparticles
are spherical, not homogeneous in size, but have a most populate size
range of 20–60 nm diameter) used in this study was the HPC-based
photoreduced substrate. The characterization and the use of this sub-
strate to study cultural heritage samples was already presented in one
of our previous work [41]. Briefly, silver nitrate and HPC (hydroxypro-
pyl cellulose) were dissolved in miliQ water to achieve a weight ratio
between AgNO3 and HPC of 1:0.6. The solution was stirred for 2 days
at room temperature so that the HPC was completely dissolved and
swelled. The substrate was then exposed to ultraviolet irradiation,
which induced and/or accelerated the reduction of silver ions to silver
nanoparticles, which served for Raman signal enhancement of fluores-
cent dye alizarin carmine.
2.3.3. Electron microscopy with energy dispersive spectroscopy (SEM/EDS)
Polished cross sections were investigated using scanning electron
microscope (SEM) JEOL 5500 LV, Japan, in a rough vacuum (12 Pa),
wherein cross-sectional areas of the samples were not necessary to be
additionally coated with graphite or gold layer. The accelerating voltage
of 20 kV was applied. Energy dispersive spectroscopy (EDS), Oxford In-
struments INCA, Great Britain, was used for image capture and for qual-
itative and quantitative elemental analysis of selected areas.
3. Results and discussion
Alizarin carmine (alizarin red S, ARS) is a synthetic derivative of 1,2-
dihydroxyanthraquinone known as alizarin. Its chemical structure is
presented in Fig. 1. In comparison to alizarin, it has an additional sulfo-
nate group, which is bound to the anthracene ring (carbon at the site 3).
A systematic approach in the investigation of ARS paint layers was con-
ducted in order to understand chemical behaviour of paint layers affect-
ed by artificial ageing. Pure colorant powder was examined by different
spectroscopic methods (namely, invasive and non-invasive FTIR spec-
troscopy, Raman, and SERS spectroscopy). Results of degradation of
ARS paint layers, which were applied atop cinnabar paint layers, are
discussed in the following sections for each binder separately. Cinnabar
was selected for the preparation of the layer beneath the glaze in order
to replicate artists' glaze layers structure and as it does not have any ab-
sorption in themid IR spectral range; therefore, it does not affect the in-
vestigation of ARS glaze layers. The numerical frequencies observed
with non-invasive reflection and conventional transmission FTIR spec-
troscopy are summarized in Tables 2 and 3, together with tentative
assignments for alizarin carmine [28,29,42–45] and specific binders
[46,47].
3.1. Infrared, normal Raman, and SERS spectra of pure alizarin carmine
There are numerous reports on the identification and characteriza-
tion of alizarin carmine (ARS) by traditional FTIR spectroscopy [28,29,
42–45] and SERS using various types of substrates [31]. However, to val-
idate the detection of the dye, and to set the baseline for the demonstra-
tion of the non-invasive procedure (reflection FTIR, Section 3.2), as well
as to show superior performance of ARS by AgHPC substrate, spectra of
the pure colorant powder were recorded in ideal conditions by trans-
mission FTIR, Raman, and SERS methods and are summarily presented
in Fig. 2. ARS isfluorescent and does not produce sufficient Raman signal
at the presented conditions. In addition, the luminescence obscures the
spectrum resulting in high background (see Fig. 2b, gray line). There-
fore, SERS can offer a solution for its detection as the Raman signal is
amplified above the fluorescence signal. ARS can interact (similar as
alizarin) with the nanoparticle surface through hydroxyl and ketogroups. Corresponding SERS spectra are pH-, excitation wavelength-,
and concentration-dependent [28]. In this study, SERS offered both
Raman signal enhancement in addition to fluorescence quenching
(see Fig. 2b, black line). The transmission FTIR spectrum (using the stan-
dard KBr pelletmethod) of ARS is shown in Fig. 2a. Bandwave numbers
of pure ARS powder and their tentative assignments are reported in
Table 1.
Briefly, in SERS, and IR spectra, the main common features were
observed:
• Stretching vibrations of carbonyl group between 1630 and 1670 cm−1
• Aromatic CC stretching vibrations at ~1590 cm−1
• Several bands between 1420 and 1500 cm−1, which can be assigned
to combinations of vibrations of ether CC, COH, CO, or CH groups
• Stretching vibration of CC group at ~1330 cm−1
• Strong IR signal at ~1205 cm−1 and aweak SERS band at ~1206 cm−1,
which can be assigned as stretching and bending modes of three car-
bons (i.e., the middle carbon belongs to 9- or 10-C = O carbon (see
Fig. 1), and the other two carbons belong to both sides of the middle
carbon) in the anthracene skeleton
• Medium strong bands at ~1069 and ~1156 cm−1 correlate to sym-
metric and asymmetric stretching of sulfonate group
3.2. Paint layers of alizarin carmine in linseed oil
3.2.1. Non-invasive reflection FTIR spectroscopy
Non-invasive reflection FTIR spectra of aged and unaged ARS paint
layers in linseed oil are gathered in Fig. 3. In the total reflection spec-
trum obtained on the unaged ARS paint layers, the characteristic
bands of ARS colorant and linseed oil binder (see Fig. 3_black line and
Table 2) could be observed, while in the spectrum of the artificially
aged ARS paint layers, most of the vibrational frequencies of ARS
Table 1
Numerical frequencies observed in IR and SERS spectra of the pure ARS powder with ten-
tative assignments.
Band assignmentb
IRa (cm−1) SERSa (cm−1) Alizarin red S
[28,29,31,42–45],
325 vw skeletal vibrations
371 vw
406 vw
453 vw
480 vw
505 vw 507 vw
526 vw
546 vw
589 m
610 w
641 m 629 vw
679 w 664 vw γ (C = O)/δ (CCC)
712 m 685vw γ (C = O)/γ (C-O)
729 vw 735 vw δ (CCC)
765 w γ (C-H)/γ (C = O)/τ (CCCC)
787 w 791 vw
823 vw 820 vw γ (C-H)/γ (C-O)
868 w 874 vw
929 w 929 w
1016 m ν (CC)/δ (CCC)
1038 m 1032 w δ (CCC) //ν(CC)/δ (CH)
1069 m 1057 m υS (SO3)
1102 w
1156 m 1163 m υAS (SO3)
1180 w ν (CC)/δ (CH)/δ (CCC)
1205 s 1206 sh ν (C = O)/δ (CCC)
1236 s hydrous SO3
1247 vs ν (C = O)
1260 vs υ (1-C-O)
1289 vs υ (2-C-O)
1330 m 1326 vs ν (CC)
1356 m ν (C-C)/δ (COH)
1418 m 1424 vs
1442 m 1447 vs ν (CC) arom.
1461 m 1464 s ν (CC)/δ (COH)/δ (CH)
1481s ν (C = O)/ν (CC)/δ (CH)
1538 m ν (CC) arom.
1564 m ν (CC) arom.
1590 m 1591 w ν (CC) arom.
1635 m 1631 m ν (9-C = O)
1666 m ν (10-C = O)
3096 sh intramolecular hydrogen bonding ν (OH)
3479 m, br ν (OH)
a vw—very weak, w—weak, m—medium, s—strong, vs—very strong, sh—shoulder.
b ν—stretching, δ—in-plane bending, γ—out-of-plane bending, τ—torsion.
Fig. 3. Non-invasive reflection FTIR spectra of unaged (black spectrum) and aged (gray
spectrum) ARS linseed oil paint layers. Spectra are translated upon vertical axis for ease
of comparison. The most representative differences between spectra are labelled.
39L. Legan et al. / Microchemical Journal 127 (2016) 36–45colorant are no longer visible (see Fig. 3_gray line and Table 2). One of
the two characteristic absorption bands, namely, ν(10-C = O) frequen-
cy, in the spectrum obtained after ageing disappears, while the second
one (ν(9-C=O)) is observed as a shoulder at 1639 cm−1 on the strong
and broaden band of carbonyl-stretching mode of lipid medium. The
ageing process affects the other moieties of ARS molecule. For instance,
all the aromatic skeletal modes and residual vibrations of anthracene
ring are no longer visible after ageing. Changes in band shape are ob-
served also in the OH stretching region. Strong band of ν(OH) of ARS
molecule decreased during ageing, while the band of ν(OH), related to
lipid binder, increased in relative intensity, suggesting a strong deg-
radation of ARS paint layer as well as lipid binder. Furthermore, few
changes appeared also in the shape and position for other distinc-
tive bands of lipid binder. Indeed, the strong stretching vibration
mode of carbonyl group, observed at 1752 cm−1, is broadened in
the spectrum of the aged layer, as well as the asymmetric stretching
vibration of CH group is shifted by 6 cm−1. Downshift and broadening
is observed in the signal of asymmetric bending vibration of linseed oil
methyl group (see Table 2). All these results obtained after artificial
ageing indicating a progressive oxidation on the alkylic chains of
linseed oil [48,49].
Further information on the layers behaviour beneath the surfacewas
possible to obtain from the spectra in the near infrared range. The lipid
binder in the near-IR range of reflection spectrum shows characteristic
signals of combination bands of methylenic C-H stretching and bending
vibrations at 4340 and at 4266 cm−1 [50,51]. Those bands areweakened
in the spectrum of the aged paint layer. Furthermore, the near-IR region
of unaged model sample reveals another combination band of asym-
metric stretching and bending vibration of hydroxyl group at
5150 cm−1, which is characteristic for gypsum (ground layer). The
band is shifted toward higher wave number after ageing and is visible
as a single sharp band at 5212 cm−1, which is typical for bassanite
[52]. This observation indicates that during ageing, gypsum in the
ground layer loses part of its bondedwater, which leads to the transfor-
mation into the hemihydrate form.
Furthermore, in the total reflection spectrum of aged ARS paint
layers in linseed oil, additional spectral features were observed (see
Table 2). An inverted (reststrahlen) band with minimum at 1139 cm−1
occurred due to high absorption index k of investigated material
and belongs to SO42− asymmetric stretching vibrations [22]. The
same distortion effect goes for asymmetric bending vibration of sul-
fate group and is placed at 621 cm−1. Symmetric stretching vibration
of the same group is seen as derivative-like band in the range from
900 to 1000 cm−1.
In order to verify the sulfate stretching and bending spectral features
in the non-invasive reflection FTIR spectrum obtained on the aged ARS
paint layers, additional analyses were performed. Electron microscopy
with energy dispersive spectroscopy (SEM/EDS) was used to study the
chemical composition of the transparent white layer, which was
newly formed atop of lipid ARS glaze layer after accelerated ageing. Re-
sults obtained on the two points (marked in Fig. 4 as 1a and 2a) of the
uppermost layer on the polished cross section revealed the occurrence
of grains made of sodium, sulfur, and oxygen that can be attributed to
sodium sulfate (Na2SO4) (see Fig. 4a,b).Most likely, the oxidation of sul-
fonate group lead to formation of sodium sulfate.
In agreementwith SEM/EDS results, the formation of sodium sulfate
was confirmed also with conventional Raman spectroscopy. The Raman
spectrum (Fig. 4c) was obtained on the same point of the uppermost
layer of the polished cross section (Fig. 4a) as for SEM/EDS analysis
(Fig. 4b) in the range between 1800 and 400 cm−1. The main band
placed at 996 cm−1 corresponds to sulfate symmetric stretching vibra-
tions. A weak asymmetric stretching triplet of sulfate anion occurs be-
tween 1100 and 1160 cm−1. The asymmetric bending vibration of
SO42− is visible as very weak triplet at 646, 636, and 621 cm−1. Another
broadened and weak band, which correlates to symmetric bending vi-
bration of sulfate anion has its maximum at 460 cm−1 [53].
Table 2
Numerical frequencies observed in reflection and transmission FTIR spectra of the ARS paint layers in linseed oil before and after artificial ageing.
Transmission FTIRa (cm−1) Reflection FTIRa (cm−1) Band assignmentb
Before ageing After ageing Before ageing After ageing Alizarin red S [28,29,42,43,44,45] Linseed oil [46,47]
592 w skeletal vibrations
617 w 614 w 620 w
621reststrahlen band ν4 (SO42−)
640 m 640 w 644 m skeletal vibrations
680 w 679 m 680 w γ (C = O)/δ (CCC)
712 m 714 w 713 m γ (C = O)/γ (C-O)
729 vw 728 vw 729 vw δ (CCC)
767 w 766 sh 764 vw γ (C-H)/γ (C = O)/τ (CCCC)
780 w 779 vw
788 w 788 w
823 vw 823 vw 825 vw γ (C-H)/γ (C-O)
870 w 870 vw 869 w
930 w 930 w 929 w
990–1000 ν1 (SO42−)
1021 m 1020 m 1018 m ν (CC)/δ (CCC)
1039 m 1044 m 1040 m δ (CCC) //ν(CC)/δ (CH)
1071 m 1070 sh 1070 m υS (SO3)
1130 sh 1139reststrahlen band ν3 (SO42−)
1159 vs 1160 vs 1158 sh υAS (SO3)
1203 vs ν (C = O)/δ (CCC)
1235 s 1230–1240 hydrous SO3
1260 vs 1255 sh 1250–1276 υ (1-C-O)
1289 s 1286 s 1280–1300 υ (2-C-O)
1330 m 1330 m 1333 s ν (CC)
1344 m, sh 1344 m, sh
ν (C-C)/δ (COH)
1357 m 1360 s
1383 w δ(CH3) umbrella mode
1418 m 1418 sh
1442 m 1443 m 1445 s 1453 s ν (CC) arom.
1461 m 1460 sh 1465 s 1462 s ν (CC)/δ (COH)/δ (CH) δAS (CH3)
1591 m 1591 m 1592 s ν (CC) arom.
1636 m 1638 m 1639 s 1639 sh ν (9-C = O)
1666 m 1668 m 1668 s ν (10-C = O)
1740 s 1740 s 1752 vs 1752 vs, br ν (C = O)
2857 m 2858 m 2860 vs 2860 vs νS (CH2)
2931 s 2931 m 2940 vs 2934 vs νAS (CH2)
3074 sh 3075 sh 3098 vw 3097 vw intramolecular hydrogen bonding ν (OH)
3240 m, br 3240 w, br 3241 br 3264 s ν (OH) ν (OH)
3460 m, br 3460 w, br 3458 s 3435 br ν (OH)
a vw—very weak, w—weak, m—medium, s—strong, vs—very strong, sh—shoulder, br—broad.
b ν—stretching, δ—in-plane bending, γ—out-of-plane bending, τ—torsion.
40 L. Legan et al. / Microchemical Journal 127 (2016) 36–453.2.2. Transmission FTIR spectroscopy
Fig. 5 reports the transmission FTIR spectra of aged and unaged
paint layers of ARS in linseed oil. The spectrum of non-aged ARS
paint layer in linseed oil (Fig. 5_black line) exhibits all characteristic
vibrational bands of ARS colorant and is in agreement with results for
pure ARS powder presented in Fig. 2a. In contrast to non-invasive
FTIR spectrum, the transmission spectrum of the aged ARS paint
layer indicate less extensive changes of the bands characteristic for
the organic dye. This might arise due to a higher concentration of
the unchanged ARS in the bulk comparing to the surface. Further dif-
ferences can be a consequence of a possible change of paint morphol-
ogy at the surface due to ageing to which reflection FTIR might be
more sensitive. The most observable change in the spectrum of the
aged ARS paint layer acquired by transmission (see Fig. 5_gray
line) is seen in the absence of two strong bands dedicated to
stretching vibrations of hydrous sulfonate group and to combination
of stretching and bending modes of three carbons in the anthracene
skeleton (bands at 1235 and 1203 cm−1). Furthermore, the symmet-
ric stretching vibration of sulfonate group decreased in intensity
during ageing and is visible as a shoulder of the stronger νAS (SO3)
at 1160 cm−1. Another shoulder band, placed at 1130 cm−1, occurs as
a new formed band after ageing, and it can be attributed to asymmetric
stretching vibrations of sulfate group related to the transparent white
layer of sodium sulfate. These results suggest that the sulfonate group
is strongly affected by the ageing process. The oxidation of sulfonategroup leading to formation of sodium sulfate was therefore confirmed
also by transmission IR approach.
The very strong signal of stretching vibration of 1-CO group at
1260 cm−1 is in the spectrum obtained after ageing shifted to lower
wave numbers by ~5 cm−1 and is apparent as a shoulder band of the
stronger asymmetric stretching mode of sulfonate group.
Additional changes in band shape and position are visible in the
combination band of CC stretching and COH bending vibration. More
precisely, after the ageing, signal at 1357 cm−1 is no longer visible,
whereas the shoulder band at 1344 cm−1 increases in intensity. Two
medium strong bands placed at 1461 and 1418 cm−1 decreased in in-
tensity and are visible as shoulder bands of the stronger aromatic CC
stretching vibration at 1443 cm−1 in the spectrum obtained on the
aged ARS paint layer. Loss of intensity is detected also in combination
band of γ (C = O)/γ(C = O)/τ (CCCC) vibrations after ageing. Two sig-
nals placed at 712 (γ(C=O)/γ (C-O) and 680 cm−1 (γ(C=O)/δ (CCC))
show variations in the relative intensity after ageing. The first one is
weakened and the latter one increased in intensity.
The reduction in relative intensity is also visible in both broadened
bands of hydroxyl stretching vibrations of ARS colorant as well as of lin-
seed oil medium placed at 3460 and 3240 cm−1, respectively. More-
over, all distinctive spectral features of linseed oil are considerably
diminished (see Fig. 5 and Table 2) after ageing. These findings
concerning lipid binder indicate an occurring oxidative degradation
[48]. Photo-oxidation of linseed oil takes place in several stages and it
Fig. 4. (a) Optical and (b) SEM image of the sample taken from the aged model painting
with marked locations of EDS analysis, identifying Na2SO4. (c) Raman spectrum of
sodium sulfate obtained at the surface of aged paint layers of ARS in linseed oil (location
of Raman analysis is marked as 1a) (λ0 = 785 nm, 3 mW, 5 s exposure time).
Fig. 5. Transmission FTIR spectra of unaged (black spectrum) and aged (gray spectrum)
ARS linseed oil paint layers. Spectra are translated upon vertical axis for ease of
comparison. The most representative differences between spectra are labelled.
Scheme 1. Proposed degradation mechanism of ARS molecule in linseed oil paint layers.
41L. Legan et al. / Microchemical Journal 127 (2016) 36–45appears as continuation of progressive hardening process [48,54]. Dur-
ing oxidation processes, many different radical species are formed,
such as alkoxyl, alkyl, peroxy, and hydroxyl radical, etc., which can dur-
ing the hardening process recombine and form tridimensional network
[55]. Photo-oxidation causes partial fragmentation of tridimensional
network and evolution of free radicals, low molecular weight com-
pounds, alkylic fragments, etc. The layer of sodium sulfate above the
paint layers of ARS in linseed oil could be explained by the formation
of hydroxyl radical during the binder's photo-oxidative degradation.Attack of hydroxyl free radical on the sulfonate moiety of ARSmolecule
can lead to the formation of sulfate ions [56]. An assumed degradation
mechanism is given in Scheme 1.
3.2.3. Surface-enhanced Raman spectroscopy (SERS)
SERS spectra of the organic dye bound in linseed oil (aged and
unaged paint layers) are presented in Fig. 6. The main differences be-
tween the spectra of unaged and aged paint layers are the variations
in relative intensities of the bands between 1400 and 1450 cm−1 and
1540–1650 cm−1. SERS spectrum of powder ARS has in the region
between 1400 and 1450 cm−1 two characteristic bands (at 1424 and
1447 cm−1, the latter having higher intensity). In the unaged paint
layer, also the band at ~1442 cm−1 shows a high intensity, while the
band at 1428 cm−1 appears as a shoulder. In the aged color layers,
there is reversed band intensity—the band with higher intensity being
at 1428 cm−1. Those bands are associated with the stretching modes
of aromatic ring. Furthermore, after the artificial ageing, the weak
band at 1176 cm−1 (ν (CC)/δ (CH)/δ (CCC) [28] appears as a shoulder
in the spectrum of the aged layer, bands at 1565 and 1541 cm−1 disap-
pear, and a new broad band emerges at 1549 cm−1. The increase in the
band at 1549 cm−1,which involves either CC stretching [28] or carbonyl
stretching [7], should appear as a consequence of a change in resonance
structure. Moreover, the strongest band in the carbonyl-stretching re-
gion appears at 1634 cm−1 in the spectrum of unaged paint layer,
while in the spectrum of aged paint layer at 1619 cm−1. Similar results
were obtained by Holmgren et al. [7], when the pH dependence of ARS
in DRIFT spectra were studied and by Cañamares et al. [28], which ex-
plained the effect on SERS spectra regarding to different adsorption
modes of alizarin on silver nanoparticles' surface. Because the changes
in the SERS spectra of ARS are visible mainly in the region of vibrations
associated with different adsorption modes of ARS on nanoparticle sur-
face,we suggest that the variations are a consequence of different ARS in-
teraction with silver nanoparticles' surface and/or the changes in the
resonance structure [7,28,31]. For the aged paint layer, it could be pro-
posed, especially based on the increase of the band at 1421 cm−1
Fig. 6. SERS spectra of unaged (black spectrum) and aged (gray spectrum) ARS linseed oil
paint layers. Spectra are translated upon vertical axis for ease of comparison. (λ0 =
514 nm, 0.6 mW, 5 s exposure time).
Fig. 7. Non-invasive reflection FTIR spectra of unaged (black spectrum) and aged (gray
spectrum) ARS egg tempera paint layers. Spectra are translated upon vertical axis for
ease of comparison. The most representative differences between spectra are labelled.
42 L. Legan et al. / Microchemical Journal 127 (2016) 36–45(which is also very intense in the spectrum of alizarin using the same
substrate and associated to the dianionic form), that the adsorption
mode of ARS in the aged samples is preferentially dianionic. ARS adsorbs
stronglywith the primary adsorption sites, which further leads to depro-
tonation ofOH-group. Furthermore, this could be possibly also correlated
to different concentration of ARS dye in the unaged and aged samples,
suggesting a higher concentration of ARS in unaged paint layers. In
such cases,when the concentration of analyte is low, ARS adsorbs strong-
ly with the primary adsorption sites, which further leads to deproton-
ation of OH-group; therefore, a higher contribution of ARS in dianionic
form is visible in the SERS spectrum of the aged paint layers.
Interestingly, no significant changes of the bands assigned to sulfo-
nate group are visible as in the case of IR spectroscopy results. The ex-
planation most likely reflects the fact, that ARS adsorbs through keto
and hydroxyl sites, and the sulfonate groups is the most distant group
from the nanoparticle surface and therefore the least affected by the in-
fluence of the nanoparticle and the SERS effect. Furthermore, the bands
associated with sulfonate vibration modes could also be overlapped by
the anthracene skeletal vibration modes.
3.3. Paint layers of alizarin carmine in non-fatty egg tempera
3.3.1. Non-invasive reflection FTIR spectroscopy
The non-invasive reflection FTIR spectra of artificially aged and
unaged ARS paint layers in non-fatty egg tempera are presented in
Fig. 7. The most pronounced changes in the medium IR range during
ageing involve the absence of two strong bands placed at 1591 and
1354 cm−1 that are assigned to aromatic CC stretching vibration and
combination of CC stretching and COH bending vibrations, respectively
(see Table 3). Additionally, a decay of band intensity at 1661 cm−1 was
observed, which corresponds to carbonyl-stretching vibration of ARS
molecule. Another loss of intensity is seen in the signal of combination
of ν (CC)/δ (COH)/δ (CH) vibrations placed at 1465 cm−1. Similarly, as
for the non-invasive reflection monitoring of degradation of ARS paint
layers in lipid binder, the strong band of stretching vibrations of 1-CO
group at 1269 cm−1 alongwith the stretching vibrations of hydrous sul-
fonate group at 1239 cm−1 are no longer visible in the spectrumobtain-
ed after ageing. Asymmetric stretching vibration of sulfonate group is
seen as distorted derivative-like/inverted band in the range of 1135–
1200 cm−1 in the reference spectrum, whereas the same signal is
broadened and has its maximum at 1142 cm−1 in the spectrum of the
aged paint. In parallel, the symmetric band of aforementioned group
at 1075 cm−1 is shifted by 20 cm−1 to higher wave numbers. The
changes in the sulfonate regions are visible to a lesser extent as in the
case of ARS in linseed oil paints. Also, no signal which could becorrelated to vibrations of sulfate group could be observed. Another
derivative-like band assigned as δ (CCC) or ν(CC)/δ (CH) is placed in
the range of 1029–1050 cm−1 in the reference paint spectrum, while
the same band is broadened and has its maximum at 1043 cm−1 in
the spectrum of the aged ARS paint layer. Degradation of ARS colorant
after ageing is visible also in the absence of combination band of CC
stretching and CCC in-plane-bending vibrations placed at 1022 cm−1.
The spectral region below 1000 cm−1 shows several derivative and/or
inverted signals belonging to ARS molecule vibrations (see Table 3 and
Fig. 7). In this region, many characteristic bands of ARS molecule disap-
pear during the ageing process (i.e., δ (C=O)/δ (CCC), γ (C-H)/γ (C-O),
γ (C = O)/γ (C-O)/τ (CCCC) and γ (C = O)/γ (C-O)).
The other changes comparing both, aged and non-aged ARS spectra
in non-fatty egg tempera are related to proteinaceous binder. Shoulder
in the reference spectrum at about 1645 cm−1 correlates to fundamen-
tal stretching of the amid carbonyl group, while in the aged spectrum
this band is overlapped with the medium strong signal of carbonyl-
stretching vibrations at 1660 cm−1 of ARSmolecule. The loss of intensi-
ty as well as broadening is present in another characteristic protein
band, amid II. Moreover, the reduction of relative intensity is visible
also in signals of stretching vibrations of methylene and carbonyl
group after ageing (see Table 3).
Also in this case, bassanite in the ground layer was identified in the
NIR range (combination band at 5212 cm−1), as a consequence of dehy-
dration of gypsum during ageing.
3.3.2. Transmission FTIR spectroscopy
Fig. 8 illustrates the comparison of two transmission FTIR spectra ob-
tained on the aged and non-aged ARS paint layers in non-fatty egg tem-
pera. The spectral characterization of non-aged ARS paint layers, which
contain two strong characteristic absorption bands of C = O stretching
vibrations at 1660 cm−1 and 1635 cm−1 dedicated to ARS molecule,
have changed after ageing and appear in the spectrumas one broadened
bandwithmaximum at 1635 cm−1. Moreover, this band has a shoulder
at ~1657 cm−1, which could be attributed to one of the carbonyl-
stretching vibrations of ARSmolecule or to amid I from the egg tempera
binder (see Fig. 8 and Table 3).
Slight decrease of relative intensity is visible in the signals of protein-
aceous binder in the CH stretching region. Carbonyl-stretching vibration
of egg tempera binder is broadened after ageing, as well as shifted to
lower wave numbers for approximately 25 cm−1. The mentioned
downshift of ν(C= O) band may be present due to oxidation of the tri-
glycerides in aged egg yolk [57,58]. Another observable change takes
place in the range of bending vibrations of methyl group. This medium
strong band placed at 1486 cm−1, corresponds to proteinaceous binder,
Table 3
Numerical frequencies observed in reflection and transmission FTIR spectra of the ARS paint layers in non-fatty egg tempera before and after artificial ageing.
Transmission FTIRa (cm−1) Reflection FTIRa (cm−1) Band assignmentb
Before ageing After ageing Before ageing After ageing Alizarin red S [28,29,42,43,44,45] Egg tempera [46]
583–610 595 vw
skeletal vibrations
640 m 639 m 645 m 615–650
678 w 667 w 671–702 γ (C = O)/δ (CCC)
686 w 689 vw
712 s 712 m 700–725 715 inverted γ (C = O)/γ (C-O)
729 vw 733 vw δ (CCC)
766 w 750–770 γ (C-H)/γ (C = O)/τ (CCCC)
789 w 789 w 791 m 770–800
824 vw 825 vw 815–835 γ (C-H)/γ (C-O)
868 w 870 w 871 m
930 w 930 w 932 m
1018 s 1018 m 1022 m ν (CC)/δ (CCC)
1038 s, 1044 s 1041 s 1029–1050 1043 m, br δ (CCC) //ν(CC)/δ (CH)
1069 s 1075 m 1075 m 1090 m, br υS (SO3)
1103 m
1158 vs 1160 vs, br 1135–1200 1142 br, vw υAS (SO3)
1203 vs 1201 s 1208 m 1210 m ν (C = O)/δ (CCC)
1235 vs 1239 s hydrous SO3
1261 vs 1266 s 1269 s υ (1-C-O)
1288 s υ (2-C-O)
1330 s ν (CC)
1348 s 1344 s 1354 s ν (C-C)/δ (COH)
1417 m 1415 m 1415 m
1444 s 1455 sh ν (CC) arom.
1463 s 1468 s 1465 s 1463 m ν (CC)/δ (COH)/δ (CH)
1486 m δ (CH3)
1546 m 1548 m 1549 s 1549 m, br Amid II
1592 s 1590 s 1591 s ν (CC) arom.
1635 vs 1635 m 1645 sh ν (9-C = O) Amid I
1660 s 1657 sh 1661 vs 1660 m, br ν (10-C = O)
1740 s 1724 m, br 1741 vs 1737 s ν (C = O)
2855 s 2855 m 2858 vs 2858 s νS (CH2)
2926 vs 2924 m 2929 vs 2929 s νAS (CH2)
3070 w, sh 3070 vw 3080 m 3080 m intramolecular hydrogen bonding ν (OH)
3278 s, br 3265 w, br 3290 vs 3290 vs ν (NH)
3451 s, br 3441 w, br ν (OH)
a vw—very weak, w—weak, m—medium, s—strong, vs—very strong, sh—shoulder, br—broad.
b ν—stretching, δ—in-plane bending, γ—out-of-plane bending, τ—torsion.
43L. Legan et al. / Microchemical Journal 127 (2016) 36–45and disappears after ageing. Furthermore, a reduction in intensities,
likewise broadening, of two characteristic signals in OH/NH stretching
region placed at 3441 and 3265 cm−1, respectively, were observed.
Peaks attributed to aromatic CC stretching vibrations at 1592 cm−1
and combination of ν(CC)/δ(COH)/δ(CH) vibrations at 1463 cm−1
remained almost unchanged after ageing, whereas additional signal of
the aforementioned vibration at 1444 cm−1 is shifted for ~10 cm−1Fig. 8. Transmission FTIR spectra of unaged (black spectrum) and aged (gray spectrum)
ARS egg tempera paint layers. Spectra are translated upon vertical axis for ease of
comparison. The most representative differences between spectra are labelled.and is present as shoulder of stronger ν(CC)/δ(COH)/δ(CH) band.More-
over, medium strong signal of unassigned vibration placed at
1417 cm−1 is no longer visible after ageing. Distinctive strong doublet
at 1348 and 1330 cm−1, assigned to combination of ν(C-C)/δ(COH)
and CC stretching vibration, respectively, is present as single band after
ageing with maximum at 1344 cm−1. The C-O stretching region shows
shifting and intensity reduction of the band placed at 1261 cm−1 and
the disappearance of the signal placed at 1288 cm−1.
The transmission spectrumobtained on the agedARS painting layers
shows another abandon of hydrous sulfonate stretching vibration band,
as well as shifting and intensity decreasing of symmetric stretching vi-
bration of sulfonate group at 1069 cm−1. In comparison to ARS paint
layers in lipid binder, the changes in the spectral features of sulfonate
regions are not so pronounced.
Another unassignedmedium strong band placed at 1103 cm−1 after
ageing disappears. Double band of δ (CCC) //ν(CC)/δ (CH) in the ARS
anthracene skeleton is visible as strong single band after ageing with
maximum at 1041 cm−1. Moreover, a reduction of intensity is seen in
the absorption bands of ν (CC)/δ (CCC) at 1018 cm−1 along with the
combination signals of out-of-plane bending of C-O and C = O group
at 712 cm−1.
3.3.3. Surface-enhanced Raman spectroscopy (SERS)
Comparing the SERS spectra of ARS powder, aged and unaged spec-
tra of ARS paint layers (Fig. 9), there is a change in spectral pattern based
on the inversion of relative intensities of 1427/1443 cm−1 doublet. In
the spectrum of the unaged paint layer, a shoulder at the 1427 cm−1
could be observed, while in the spectrum of aged paint layer relative
Fig. 9. SERS spectra of unaged (black spectrum) and aged (gray spectrum) ARS egg
tempera paint layers. Spectra are translated upon vertical axis for ease of comparison.
(λ0 = 514 nm, 0.6 mW, 5 s exposure time).
44 L. Legan et al. / Microchemical Journal 127 (2016) 36–45intensity of the band at 1427 cm−1 increases. Other changes in the spec-
trum of aged layers are the weakening of the band at 1177 cm−1, the
disappearance of the bands at 1565 and 1541 cm−1, which results in
the appearance of a new, broad band at 1551 cm−1. Moreover, the
strongest band of the carbonyl-stretching region in the spectrum of
unaged paint layer appears at 1634 cm−1, while in the spectrum of
aged paint layer a band at 1619 cm−1 becomes the strongest in this re-
gion. All the changes are associated with different adsorption forms of
ARS on silver nanoparticle surface, suggesting the higher contribution
of adsorbed ARS in dianionic form in the aged paint layers Similar to re-
sults of ARS in linseed oil paint layers, no changes in the spectra related
to vibrations of sulfonate group could be observed. The spectral changes
could only reflect different adsorptionmodes ofmolecule. No significant
differences could be observed in comparison to the behaviour of ARS in
linseed oil binder.
4. Conclusion
Until this study, not much attention was dedicated to the use of
non-invasive reflection spectroscopy formonitoring of degradationpro-
cesses of organic glaze layers at the surface. As the obtained reflection
spectra of aged paint layers could be highly obscured, a careful imple-
mentation of supporting methods should be made in order to correctly
identify the present material and/or degradation products. Such
approach could serve as a platform toward minimally or even non-
invasive investigations of the organic glaze layers.
Nevertheless, the deterioration of ARS paint layer (ARS dye in lipid
or proteinaceous binder) was for the first time identified by non-
invasive reflection FTIR spectroscopy, based on characteristic changes
in the spectral regions belonging to carbonyl, hydroxyl, and sulfonate
group, as well as anthracene aromatic skeletal vibrational modes. All
these spectral differences were confirmed also by the transmission
FTIR spectroscopy, although the changes in spectral features were visi-
ble to a lesser extent by this method, probably due to sampling as a
lower concentration of degraded products are present in the bulk com-
paring to the surface. Spectral patterns of ARS in the FTIR spectra of aged
paint layers were in many cases indistinguishable; therefore, the exact
characterization of the dye in the paint layer could be questionable in
an unknown sample. In such cases, the complementation of the re-
search with SERS spectroscopy is suggested. In this study, SERS still of-
fered very good spectra of the dye, thanks to the higher sensitivity of
the technique. No significant differences between paint layers in linseed
oil and egg tempera in their spectral behaviour were detected, most
likely due to the sensitivity and selectivity of the technique for the
detection of organic dyes. Variations in relative intensities as well asshifting of the certain bands suggested different adsorption modes of
ARS with silver nanoparticles, indicating a higher contribution of ARS
in dianionic form in the aged paint layers, which could be further possi-
bly correlated to lower concentrations of the dye in the aged paint
layers.
The analysis of paint layers using FTIR spectroscopy indicates that
the degradation is binder-dependent. It seems that the degradation of
ARS in lipid binder is more pronounced. The surface of paint layers in
linseed oil was covered with additional transparent layer after ageing.
FTIR analysis showed the presence of Na2SO4, whichwas also confirmed
by electron and Ramanmicroscopies. Regarding the fact that the chang-
es in the spectral features of sulfonate regions are not so pronounced in
the egg tempera binder, it can be concluded that the formation of
sodium sulfate is favoured under impact of UV–Vis radiation and rela-
tive humidity in the lipid binder as a result of its photo-oxidative degra-
dation [48]. Indeed, it was already shown that ageing processes has
much greater impact on degradation of lipid binders as on proteina-
ceous ones [59].
Strong degradation was detected also in the other parts of multi-
layered model painting. Indeed, the UV–Vis radiation and oscillations
of temperature and humidity did not affect only the upper most layer
of the model painting, but deterioration was evident also in the specific
molecular changes of oil and egg temperamedia aswell as in the ground
layer. The latter was corroborated by the discovery of bassanite (the
hemihydrate form of gypsum) by non-invasive reflection spectroscopy
in the NIR range.
This workmight have possible applications on real painting analysis,
facilitating determination of degraded organic glaze layers with non- or
minimally invasive techniques.
Acknowledgments
The presented research was supported by the Slovenian Research
Agency (grant number L6-4217). This work is also a result of doctoral
research, in part financed by the European Union, European Social
Fund and the Republic of Slovenia, the Ministry for Education, Science
and Sport within the framework of the Operational programme for
human resources development for the period 2007–2013 (operation
no. 3330-13-500221).
References
[1] J. Kirby, M. Spring, C. Higgitt, The technology of red lake pigment manufacture:
study of the dyestuff substrate, Natl. Gall. Tech. Bull. 26 (2005) 71–87.
[2] G. Rath, M. Ndonzao, K. Hostettmann, Antifungal anthraquinones from Morinda
lucida, Int. J. Pharmacogn. (2015).
[3] J. Wouters, N. Rosario-Chirinos, Dye analysis of pre-Columbian Peruvian textiles
with high-performance liquid chromatography and diode-array detection, J. Am.
Inst. Conserv. (2013).
[4] B. Szostek, J. Orska-Gawrys, I. Surowiec, M. Trojanowicz, Investigation of natural
dyes occurring in historical Coptic textiles by high-performance liquid chromatog-
raphy with UV–vis and mass spectrometric detection, J. Chromatogr. A 1012
(2003) 179–192.
[5] H. Schweppe, J.Winter, third ed., Artists' Pigments: A Handbook of Their History and
Characteristics, vol. 3, National Gallery of Art, Oxford University Press, Oxford, New
York, 1997.
[6] E. Gurr, Synthetic Dyes in Biology, Medicine And Chemistry, Academic Press,
London, New York, 2012.
[7] A. Holmgren, L. Wu, W. Forsling, Fourier transform infrared and Raman study of
alizarin red S adsorbed at the fluorite–water interface, Spectrochim. Acta A Mol.
Biomol. Spectrosc. 55 (1999) 1721–1730, http://dx.doi.org/10.1016/S1386-
1425(98)00342-4.
[8] H. Puchtler, S.N. Meloan, M.S. Terry, On the history and mechanism of alizarin and
alizarin red S stains for calcium, J. Histochem. Cytochem. 17 (1969) 110–124.
[9] L. Jervis, Interaction of immobilised food dye residues with gut proteins, Biochem.
Soc. Trans. 24 (1996) 395S.
[10] G. Ackermann, L. Sommer, D. Thorburn Burns, Organic analytical reagents for the
determination of inorganic substances, in: R.D. Lide (Ed.), CRC Handb. Chem. Phys-
ics, 95th ed.CRC Press/Taylor and Francis, Boca Raton, Florida 2009, p. 2704.
[11] M.K. Cyrański, M.H. Jamróz, A. Rygula, J.C. Dobrowolski, Ł. Dobrzycki, M. Baranska,
On two alizarin polymorphs, CrystEngComm 14 (2012) 3667.
[12] K. Gollnick, S. Held, D.O. Mártire, S.E. Braslavsky, Hydroxyanthraquinones as sensi-
tizers of singlet oxygen reactions: quantum yields of triplet formation and singlet
45L. Legan et al. / Microchemical Journal 127 (2016) 36–45oxygen generation in acetonitrile, J. Photochem. Photobiol. A Chem. 69 (1992)
155–165.
[13] C.J.S. Ibsen, H. Birkedal, Modification of bone-like apatite nanoparticle size and
growth kinetics by alizarin red S, Nanoscale 2 (2010) 2478–2486.
[14] E. Svobodová, Z. Bosáková, M. Ohlídalová, M. Novotná, I. Němec, The use of infrared
and Raman microspectroscopy for identification of selected red organic dyes in
model colour layers of works of art, Vib. Spectrosc. 63 (2012) 380–389, http://dx.
doi.org/10.1016/j.vibspec.2012.09.003.
[15] S.A. Centeno, J. Shamir, Surface enhanced Raman scattering (SERS) and FTIR charac-
terization of the sepia melanin pigment used in works of art, J. Mol. Struct. 873
(2008) 149–159, http://dx.doi.org/10.1016/j.molstruc.2007.03.026.
[16] E. Marengo, M.C. Liparota, E. Robotti, M. Bobba, Monitoring of paintings under expo-
sure to UV light by ATR-FT-IR spectroscopy and multivariate control charts, Vib.
Spectrosc. 40 (2006) 225–234.
[17] M. Koperska, T. Łojewski, J. Łojewska, Vibrational spectroscopy to study degradation
of natural dyes. assessment of oxygen-free cassette for safe exposition of artefacts,
Anal. Bioanal. Chem. 399 (2011) 3271–3283.
[18] T. Wu, Guangming Liu, Jincai Zhao, Photoassisted degradation of dye pollutants. 8.
Irreversible degradation of alizarin red under visible light radiation in air equilibrat-
ed aqueous TiO2 dispersions, Environ. Sci. Technol. 33 (1999) 2081–2087.
[19] E. Marengo, M.C. Liparota, E. Robotti, M. Bobba, Multivariate calibration applied to
the field of cultural heritage: analysis of the pigments on the surface of a painting,
Anal. Chim. Acta 553 (2005) 111–122, http://dx.doi.org/10.1016/j.aca.2005.07.061.
[20] T. Del Giacco, L. Latterini, F. Elisei, Photophysical and photochemical properties of
1,2,4-trihydroxy-9,10-anthraquinone adsorbed on inorganic oxides, Photochem.
Photobiol. Sci. 2 (2003) 681–687.
[21] M. Hovaneissian, P. Archier, C. Vieillescazes, Influence of cetophenolic and
diphenolic intramolecular hydrogen bonding on the chromatographic and spectro-
scopic properties of hydroxyanthraquinones, Dyes Pigments 74 (2007) 706–712,
http://dx.doi.org/10.1016/j.dyepig.2006.05.003.
[22] C. Miliani, F. Rosi, A. Daveri, B.G. Brunetti, Reflection infrared spectroscopy for the
non-invasive in situ study of artists' pigments, Appl. Phys. A Mater. Sci. Process.
106 (2012) 295–307, http://dx.doi.org/10.1007/s00339–011-6708-2.
[23] L. Monico, F. Rosi, C. Miliani, A. Daveri, B.G. Brunetti, Non-invasive identification of
metal-oxalate complexes on polychrome artwork surfaces by reflection mid-
infrared spectroscopy, Spectrochim. Acta A Mol. Biomol. Spectrosc. 116 (2013)
270–280.
[24] W. Vetter, M. Schreiner, Characterization of pigment-binding media systems: com-
parison of non-invasive in-situ reflection FTIR with transmission FTIR microscopy,
E-Preserv. Sci. 8 (2011) 10–22.
[25] M. Leona, J. Stenger, E. Ferloni, Application of surface-enhanced Raman scattering
techniques to the ultrasensitive identification of natural dyes in works of art, J.
Raman Spectrosc. 37 (2006) 981–992, http://dx.doi.org/10.1002/jrs.1582.
[26] F. Casadio, M. Leona, J.R. Lombardi, R. Van Duyne, Identification of organic colorants
in fibers, paints, and glazes by surface enhanced Raman spectroscopy, Acc. Chem.
Res. 43 (2010) 782–791, http://dx.doi.org/10.1021/ar100019q.
[27] C.L. Brosseau, K.S. Rayner, F. Casadio, C.M. Grzywacz, R.P. Van Duyne, Surface-
enhanced Raman spectroscopy: a direct method to identify colorants in various
artist media, Anal. Chem. 81 (2009) 7443–7447, http://dx.doi.org/10.1021/
ac901219m.
[28] M.V. Cañamares, J.V. Garcia-Ramos, C. Domingo, S. Sanchez-Cortes, Surface-
enhanced Raman scattering study of the adsorption of the anthraquinone pigment
alizarin on Ag nanoparticles, J. Raman Spectrosc. 35 (2004) 921–927, http://dx.doi.
org/10.1002/jrs.1228.
[29] M.V. Cañamares, J.V. Garcia-Ramos, C. Domingo, S. Sanchez-Cortes, Surface-
enhanced Raman scattering study of the anthraquinone red pigment carminic
acid, Vib. Spectrosc. 40 (2006) 161–167, http://dx.doi.org/10.1016/j.vibspec.2005.
08.002.
[30] I.T. Shadi, B.Z. Chowdhry, M.J. Snowden, R. Withnall, Semi-quantitative analysis of
alizarin and purpurin by surface-enhanced resonance Raman spectroscopy(SERRS)
using silver colloids, J. Raman Spectrosc. 35 (2004) 800–807, http://dx.doi.org/10.
1002/jrs.1199.
[31] M.L. De Souza, P. Corio, Surface-enhanced Raman scattering study of alizarin red S,
Vib. Spectrosc. 54 (2010) 137–141, http://dx.doi.org/10.1016/j.vibspec.2010.07.010.
[32] B. Doherty, B.G. Brunetti, A. Sgamellotti, C. Miliani, A detachable SERS active cellu-
lose film: a minimally invasive approach to the study of painting lakes, J. Raman
Spectrosc. 42 (2011) 1932–1938, http://dx.doi.org/10.1002/jrs.2942.
[33] C. Lofrumento, M. Ricci, E. Platania, M. Becucci, E. Castellucci, SERS detection of red
organic dyes in Ag-agar gel, J. Raman Spectrosc. 44 (2013) 47–54, http://dx.doi.
org/10.1002/jrs.4162.
[34] F. Pozzi, J.R. Lombardi, S. Bruni, M. Leona, Sample treatment considerations in the
analysis of organic colorants by surface-enhanced Raman scattering, Anal. Chem.
84 (2012) 3751–3757, http://dx.doi.org/10.1021/ac300380c.
[35] A. Idone, M. Aceto, E. Diana, L. Appolonia, M. Gulmini, Surface-enhanced Raman
scattering for the analysis of red lake pigments in painting layers mounted incross sections, J. Raman Spectrosc. 45 (2014) 1127–1132, http://dx.doi.org/10.
1002/jrs.4491.
[36] A. Campion, P. Kambhampati, Surface-enhanced Raman scattering, Chem. Soc. Rev.
27 (1998) 241–250, http://dx.doi.org/10.1039/a827241z.
[37] W.E. Doering, S.M. Nie, Single-molecule and single-nanoparticle SERS: examining
the roles of surface active sites and chemical enhancement, J. Phys. Chem. B 106
(2002) 311–317, http://dx.doi.org/10.1021/jp011730b.
[38] X.-M. Qian, S.M. Nie, Single-molecule and single-nanoparticle SERS: from funda-
mental mechanisms to biomedical applications, Chem. Soc. Rev. 37 (2008)
912–920, http://dx.doi.org/10.1039/b708839f.
[39] K. Wehlte, The Materials and Techniques of Painting, Kremer Pigments Incorporat-
ed, Aichstetten, New York, 1975.
[40] R. Hudoklin, Tehnologije materialov, ki se uporabljajo v slikarstvu, Vzajemnost,
Ljubljana, 1958.
[41] K. Retko, P. Ropret, R. Cerc Korošec, Surface-enhanced Raman spectroscopy (SERS)
analysis of organic colourants utilising a new UV-photoreduced substrate, J.
Raman Spectrosc. 45 (2014) 1140–1146, http://dx.doi.org/10.1002/jrs.4533.
[42] T. Moriguchi, S. Nakagawa, F. Kaji, Adsorbability of alizarin red S on Fe(III)- and
Pb(II)-treated hydroxyapatites in water, Phosphorus Res. Bull. 24 (2010) 62–72,
http://dx.doi.org/10.3363/prb.24.62.
[43] P.M. Jayaweera, T.A.U. Jayarathne, Acid/base induced linkage isomerization of aliza-
rin red adsorbed onto nano-porous TiO2 surfaces, Surf. Sci. 600 (2006)http://dx.doi.
org/10.1016/j.susc.2006.08.027.
[44] A. Periasamy, S. Muruganand, M. Palaniswamy, Vibrational studies of Na2SO4, K2SO4,
NaHSO4 and KHSO4 crystals, Rasayan J. Chem. 2 (2009) 981–989.
[45] T. Moriguchi, K. Yano, S. Nakagawa, F. Kaji, Elucidation of adsorption mechanism of
bone-staining agent alizarin red S on hydroxyapatite by FT-IR microspectroscopy, J.
Colloid Interface Sci. 260 (2003) 19–25, http://dx.doi.org/10.1016/S0021-
9797(02)00157-1.
[46] R. Mazzeo, S. Prati, M. Quaranta, E. Joseph, E. Kendix, M. Galeotti, Attenuated total
reflection micro FTIR characterisation of pigment-binder interaction in reconstruct-
ed paint films, Anal. Bioanal. Chem. 392 (2008) 65–76, http://dx.doi.org/10.1007/
s00216-008-2126-5.
[47] L. Brambilla, C. Riedo, C. Baraldi, A. Nevin, M.C. Gamberini, C. D'Andrea, et al., Char-
acterization of fresh and aged natural ingredients used in historical ointments by
molecular spectroscopic techniques: IR, Raman and fluorescence, Anal. Bioanal.
Chem. 401 (2011) 1827–1837, http://dx.doi.org/10.1007/s00216-011-5168-z.
[48] M. Lazzari, O. Chiantore, Drying and oxidative degradation of linseed oil, Polym.
Degrad. Stab. 65 (1999) 303–313, http://dx.doi.org/10.1016/S0141-3910(99)00020-8.
[49] J. Mallégol, J.-L. Gardette, J. Lemaire, Long-term behavior of oil-based varnishes and
paints. Photo- and thermooxidation of cured linseed oil, J. Am. Oil Chem. Soc. 77
(2000) 257–263, http://dx.doi.org/10.1007/s11746–000-0042-4.
[50] M. Vagnini, C. Miliani, L. Cartechini, P. Rocchi, B.G. Brunetti, A. Sgamellotti, FT-NIR
spectroscopy for non-invasive identification of natural polymers and resins in
easel paintings, Anal. Bioanal. Chem. 395 (2009) 2107–2118, http://dx.doi.org/10.
1007/s00216-009-3145-6.
[51] J. Workman Jr., L. Weyer, Practical Guide to Interpretive Near-Infrared Spectroscopy,
CRC Press, Taylor & Francis Group, Boca Raton, London, New York, 2007.
[52] F. Rosi, A. Daveri, B. Doherty, S. Nazzareni, B.G. Brunetti, A. Sgamellotti, et al., On the
use of overtone and combination bands for the analysis of the CaSO4–H2O system by
mid-infrared reflection spectroscopy, Appl. Spectrosc. 64 (2010) 956–963, http://dx.
doi.org/10.1366/000370210792080975.
[53] K. Ben Mabrouk, T.H. Kauffmann, H. Aroui, M.D. Fontana, Raman study of cation ef-
fect on sulfate vibration modes in solid state and in aqueous solutions, J. Raman
Spectrosc. 44 (2013) 1603–1608, http://dx.doi.org/10.1002/jrs.4374.
[54] D. Erhardt, C.S. Tumosa, M.F. Mecklenburg, Natural and accelerated thermal aging of
oil paint films, Tradit. Innov. Adv. Conserv. Contrib. to IIC Melb. Congr. 10–14 Oct.
2000 2000, pp. 65–69.
[55] J. Mallégol, J.-L. Gardette, J. Lemaire, Long-term behavior of oil-based varnishes and
paints I. Spectroscopic analysis of curing drying oils, J. Am. Oil Chem. Soc. 76 (1999)
967–976, http://dx.doi.org/10.1007/s11746–999-0114-3.
[56] H. Lachheb, E. Puzenat, A. Houas, M. Ksibi, E. Elaloui, C. Guillard, et al., Photocatalytic
degradation of various types of dyes (alizarin S, crocein Orange G,methyl red, Congo
red, methylene blue) in water by UV-irradiated titania, Appl. Catal. B Environ. 39
(2002) 75–90, http://dx.doi.org/10.1016/S0926-3373(02)00078-4.
[57] R.J. Meilunas, J.G. Bentsen, A. Steinberg, Analysis of aged paint binders by FTIR spec-
troscopy, Stud. Conserv. 35 (1990) 33–51.
[58] V. Dorge, F.C. Howlett (Eds.), Painted Wood: History and Conservation, Getty Con-
servation Institute, Los Angeles, 1998.
[59] M.P. Colombini, F. Modugno, E.Menicagli, R. Fuoco, A. Giacomelli, GC–MS characteriza-
tion of proteinaceous and lipid binders in UV aged polychrome artifacts, Microchem. J.
67 (2000) 291–300, http://dx.doi.org/10.1016/S0026-265X(00)00075-8.