ftir study of ageing of fast drying oil colour (fdoc) alkyd paint replicas

This article appeared in a journal published by Elsevier. The attached

copy is furnished to the author for internal non-commercial research

and education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling or

licensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of the

article (e.g. in Word or Tex form) to their personal website or

institutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies are

encouraged to visit:

http://www.elsevier.com/authorsrights

http://www.elsevier.com/authorsrights

Author's personal copy

FTIR study of ageing of fast drying oil colour (FDOC) alkyd paint replicas

Celia Duce a, Valentina Della Porta b, Maria Rosaria Tiné a, Alessio Spepi a, Lisa Ghezzi a,Maria Perla Colombini c, Emilia Bramanti b,⇑

aDipartimento di Chimica e Chimica Industriale, Universita’ di Pisa, Via Risorgimento 35, 56127 Pisa, ItalybNational Research Council of Italy, CNR, Istituto di Chimica dei Composti Organo Metallici-ICCOM-UOS Pisa, Area di Ricerca, Via G. Moruzzi 1, 56124 Pisa, ItalycNational Research Council of Italy, CNR, Istituto per la Conservazione e la Valorizzazione dei Beni Culturali, Via Madonna del Piano 10, 50019 Sesto Fiorentino (FI), Italy

h i g h l i g h t s

� Ageing of alkyd resin Winsor &

Newton, Fast drying oil colour studied

by ATR-FTIR.

� The increase of the absorbance at

1635 cm�1 was selected as a

parameter to follow the ageing

process.

� No extractive procedure, no

destructive technique.

� The carbonate of the filler and the

carbonyl group of the esters react

under acetic acid atmosphere.

� A sample from a hyper-realistic

artwork of Patrizia Zara was

investigated to evaluate its natural

ageing.

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:

Received 30 December 2013

Received in revised form 25 March 2014

Accepted 29 March 2014

Available online 16 April 2014

Keywords:

Alkyd paints

ATR-FTIR

Natural ageing

Artificial ageing

a b s t r a c t

We propose ATR-FTIR spectroscopy for the characterization of the spectral changes in alkyd resin from the

Griffin Alkyd Fast Drying Oil Colour range (Winsor & Newton), occurring over 550 days (�18 months) of

natural ageing and over six months of artificial ageing under an acetic acid atmosphere. Acetic acid is

one of the atmospheric pollutants found inside museums in concentrations that can have a significant

effect on the works exhibited. During natural ageing we observed an increase and broadening of the OH

group band around 3300 cm�1 and an increase in bands in the region 1730–1680 cm�1 due to carbonyl

stretching. We found a broad band around 1635 cm�1 likely due to C@O stretching vibrations of b dich-

etons. These spectral changes are the result of autooxidation reactions during natural ageing and cross-

linking, which then form f alcohols and carbonyl species. The increase in absorbance at 1635 cm�1 was

selected as a parameter to monitor the ageing process of paintings prepared with FDOC, without the need

for any extractive procedure. FTIR spectra of paint replicas kept under an acetic acid atmosphere indicated

the chemical groups involved in the reaction with acid, thus suggesting which spectral FTIR regions could

be investigated in order to follow any degradation in real paintings. A red paint sample from a hyper-

realistic artwork (‘‘Racconta storie’’, 2003) by the Italian painter Patrizia Zara was investigated by FTIR

in order to evaluate the effects of 10 years natural ageing on alkyd colours. The results obtained suggested

that after the end of chemical drying (autooxidation), alkyd colours are very stable.

� 2014 Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.saa.2014.03.123

1386-1425/� 2014 Elsevier B.V. All rights reserved.

⇑ Corresponding author. Tel.: +39 050 315 2293; fax: +39 050 315 2555.

E-mail address: [email protected] (E. Bramanti).

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 130 (2014) 214–221

Contents lists available at ScienceDirect

Spectrochimica Acta Part A: Molecular andBiomolecular Spectroscopy

journal homepage: www.elsevier .com/locate /saa


Introduction

Alkyd resins were introduced in the 1930s as paint binders.

Their compatibility with many polymers and the extremely wide

formulating latitude made them suitable for the production of a

very broad range of coating materials. They have been used in

artistic works since the 1920–1930s. To the human eye they look

like traditional oil paints, and due to their high molecular weight

they dry in only 24–48 h (touch dry) [1].

The term ‘alkyd’ is derived from the al in alcohol and the cid in

acid, and refers to polyesters modified with fatty acid polyesters

[1]. They are generally prepared by the condensation polymeriza-

tion of polyalcohols, polybasic acids and fatty acids. Fig. 1 shows

an ideal alkyd resin [2,3]. The type of fatty acid employed in the

alkyd resin determines the final drying characteristic and flexibility

of the coating.

Modern paints are a complex mixture of binders (30%), solvents

(25%), fillers (10%), additives (3%), pigments (22%), and water (10%)

[4]. Additives such as driers, anti-oxidants, anti-skinning agents,

thickeners, fungicides, emulsifiers, UV-absorbers, and matting

agents are generally added in very small amounts to paints, but

have a considerable influence on the final properties [2].

Alkyd resins are soluble in organic solvents producing

solvent-borne coatings, or can be emulsified in water, producing

water-borne coatings. The drying process includes: (1) a solvent

evaporation step (physical drying) and (2) autooxidative drying

(chemical drying), which includes an induction period, oxygen

uptake, peroxide formation, and peroxide decomposition. Radical

reactions produce a crosslinked polymer network (Fig. 2) [2,5].

Unsaturated fatty acid side chains are mainly responsible for

autooxidation reactions. The reaction is initiated by hydrogen

abstraction of the doubly-activated methylene group. The resulting

radical R� reacts with O2, leading to hydroperoxide species (ROOH).

Hydroperoxides decompose in a metal-catalyzed reaction to

alkoxy (RO�) and peroxy radicals (ROO�). The radicals recombine

and produce a three-dimensional polymer network, which is

responsible for curing the paint film [1,5–10].

Autooxidation is usually catalyzed by metal salts known as

driers, which are classified into three major types: surface, through

and auxiliary driers [2,11,12]. Driers deactivate natural anti-

oxidants, shortening the induction period, accelerating the oxygen

absorption and the peroxide production. The cobalt drier, for

example, catalyzes several beneficial reactions in the process of

autooxidation. The most important reaction is the cleavage of the

peroxides to form free radicals, which can interact within the auto-

oxidative mechanism to accelerate crosslinking. Another function

is to generate the singlet oxygen, which is more reactive than

ground state oxygen. An improvement in the drying time was

monitored with a number of micron sizeed pigments including

ZnS, V2O5, ZnO and nanosized-ZnO, compounds from the amor-

phous chalcogenides GeAS(Se) semi-conductor family, titanium

sol–gel precursors, iron catalysts combined with ascorbic acid,

reviewed in [2].

Alkyd paints are a very new material, thus little is known about

their stability and properties during ageing. However, their charac-

terization represents an important contribution to the study of the

painting techniques used in contemporary art.

Fundamental studies have been developed by Ploeger et al. by

thermal analysis and FTIR-ATR [1,7] Chiantore et al. characterized

alkyd paints from the Griffin paint colour series, introduced in

1970 by Winsor & Newton, and their ageing [1]. Cakic et al. studied

the formation of photo-oxidative species in long-oil air drying

alkyd paint by monitoring the changes by FTIR with different sta-

bilizers after UV exposure (450 h, >300 nm). They found that the

photochemical degradation of alkyd paint is associated with an

increase in the hydroxyl content and broadening of the absorption

in the carbonyl region [13]. Stamatakis et al. studied the ageing

process of polyester resins by NMR spectroscopy [14]. Phenix stud-

ied the increased sensitivity and plasticization of alkyds in water as

a result of the formation of hydrophilic groups using computer-

based image analysis [15]. Stava et al. studied the cross-linking

reactions that take place during the creation of an alkyd resin film

and the catalytic action of cobalt, manganese and their mixed salts

as driers [11]. Ouldmetidji et al. and Mallègol et al. described the

use of DSC to follow the peroxidation of polymers during ageing

[16–18]. Muizebelt et al. studied the oxidative crosslinking of alkyd

resins using mass spectrometry and NMR with model compounds

[19]. Recently, a special issue was published to present both histor-

ical reviews on the technological evolvement of alkyds, as well as

leading research papers in this field in order to show the potential

of these coatings in the 21st century [20].

The aim of this work is to investigate the natural ageing (up to

550 days) and the artificial ageing under acetic acid atmosphere of

a set of alkyd resins from Winsor & Newton’s Griffin series of fast

drying oil colours (FDOCs), using ATR-FTIR spectroscopy. Fast dry-

ing oil colours have a new unknown formulation compared to the

alkyd resins investigated by Chiantore et al. as the source of fatty

acid is a semi-siccative soy oil and a Cobalt-carboxylate drier

was added in FDOCs to accelerate the drying speed.

Our aim was thus to explore the impact of the new formulation

on the ageing process and on the stability of the paint film, by eval-

uating the spectroscopic parameters which are potentially usefulFig. 1. Chemical structure of a typical alkyd resin containing phthalic anhydride,

glycerol and linoleic acid [1,3].

Fig. 2. Oxidative process leading to a dry paint film [2,5]. The peroxide formation in

non-conjugated and conjugated double bonds are shown.

C. Duce et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 130 (2014) 214–221 215


in dating contemporary painted works of art. Blue, green, yellow

and red paint replicas obtained with FDOCs were investigated.

These colors have formulations based on inorganic or organic pig-

ments. Ivory black formulated with carbonized bones and Titanium

White were also studied. Table 1 summarizes the colour name,

code and formulation of the FDOCs studied in this work.

This work was carried out within the framework of a project

concerning the characterization of materials used in modern and

contemporary art (COPAC). In addition, we studied the same FDOC

paint replicas also using thermogravimetry (TG) and differential

scanning calorimetry (DSC) [21]. The curing of the French ultrama-

rine in the Griffin Alkyd FDOC series was also recently studied by

Bartolozzi et al. using FT-IR and NMR spectroscopies with an

extractive method [22]. La Nasa et al. also characterized, three

colours (PY43 Yellow Ochre, PB28 Cobalt Blue and PBK9 Ivory

Black) of the Griffin Alkyd FDOC series, using a multi-analytical

approach based on the integration of gas chromatography–mass

spectrometry (GC/MS), high performance liquid chromatography

coupled with electrospray ionization mass spectrometry with a

tandem quadrupole-time of flight mass spectrometer (HPLC–ESI-

Q-ToF), and flow injection analysis (FIA) in the ESI-Q-ToF mass

spectrometer [23].

In this work a red sample from Zara’s ‘‘Racconta storie’’ (2003)

painted with alkyd FDOC was also studied by ATR-FTIR. In this

case, its FTIR spectrum was similar to the FTIR spectra of the

Cadmium Red paint replica investigated, and a comparison was

made in terms of ageing and paint stability.

Finally, the same set of colours was also investigated after

artificial ageing under an acetic acid atmosphere. Acetic acid is

one of the significant atmospheric pollutants found inside muse-

ums. The wood of furniture, especially non-aged wood, wood

panels and synthetic polymeric materials used in museum

Table 1

Colour name, code and formulation of the FDOC studied in this work.

Colour name Index colour Pigment name Pigment chemical structures Filler

Titanium White PW6 Titanium white TiO2 Dolomite

Ivory Black PBK9 Ivory black Bone char (C, CaCO3, Ca3(PO4)2) Absent

French Ultramarine PB29 Sodium alumino-silicate complex containing

sulphur

Na6(AlSiO4)6S4 Dolomite

Phthalo Blue PB15 Copper phthalocyanine Dolomite

Viridian Green PG18 Hydrated chromium (III) oxide Cr2O3�2H2O Absent

Phthalo Green PG7 Chlorinated copper phthalocyanine Dolomite

Cadmium Yellow PY35 Cadmium/zinc sulphide (Cd,Zn)S2 Dolomite

Winsor Lemon PY3 Arylamide yellow Dolomite

Cadmium Red PR108 Cadmium (II) sulphoselenide CdSSe Dolomite

Winsor Red PR170 + PR188 Naphtol carbamides Dolomite

216 C. Duce et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 130 (2014) 214–221


equipment, namely adhesives, varnishes or other plastics can

release large amounts of volatile organic compounds such as acetic

acid. Although the amounts of acetic acid released by these mate-

rials are low, when they are used in closed spaces (e.g. showcases),

relatively high concentrations (> higher 3000 g m�3 inside the

showcases, 0.1 and 100 g m�3 outside) can be reached. These

concentrations can have a significant effect on the objects exhibited

[24,25]. Acetic acid can also attack other materials, such as copper

alloys and copper materials [26], paper [27], shell [28] or other

calcareous materials [29]. Acetic acid can also attack polymeric

materials, causing depolymerization by acid hydrolysis [30,31].

Experimental

Equipment and ATR-FT-IR spectra measurements

Infrared spectra were recorded using a Perkin Elmer Spectrum

100 FTIR spectrophotometer, equipped with a universal attenuated

total reflectance accessory (ATRU) and a TGS detector. A few

micrograms of powder scratched from each paint reconstruction

were used. The following spectrometer parameters were used;

resolution: 4 cm�1, spectra range: 450–4000 cm�1, number of

scans: 32. Spectrum software was used to process FTIR spectra.

Materials

Ten commercial alkyd resin samples were purchased from

Winsor & Newton (Griffin series, fast drying oil colours) (Table 1).

The pure pigments were produced by Abralux colori Berghè

(Cadmium Yellow-PY35, Viridian green-PG18, Cadmium red-

PR108, Winsor lemon-PY3, French ultramarine-PB29 and Titanium

white-PW6) and Zecchi colori Firenze (Ivory black-PBK9).

Two phthalocyanine pigments were synthesized in our labora-

tory: Phthalo blue-PB15 and Phthalo green-PG7. The synthesis pro-

cedures are shown in the Supplementary Data (SD).

The pure pigment Winsor red 170–188 is a mixture of naphtol

carbamides. However, this pigment is not available because the

percentage composition is unknown as the manufacturers does

not provide the exact composition of any individual colour. Glacial

acetic acid was purchased from Sigma-Aldrich-Fluka Chemical Co.

1005706 (Milan, Italy).

Sample preparation

Natural ageing of FDOC

The FDOCs were applied on glass microscope slides in order to

obtain approximately 0.5–1.0 mm thick layers. The paint replicas

were naturally aged in our laboratory (21 ± 2 �C; 65% relative

humidity). Naturally aged samples were tested 48 h, 120 h,

192 h, 264 h, 384 h, 720 h, 1050 h, 1770 h, 3300 h, 5600 h

(8 months) and 12800 h (18 months) after their preparation to

evaluate all changes due to natural ageing.

Ageing with CH3COOH of Winsor & Newton alkyd paint colours

For the artificial ageing with acetic acid, the FDOC paint replicas

were placed in closed glass cases at 25 �C for six months in an

atmosphere saturated with glacial acetic acid.

Patrizia Zara’s paints

Small portions of samples (3–4 mg) were taken from the lateral

part of the work (scratched with a stainless steel knife). We

selected a point where the artist had used the pure red colour (to

avoid analysing colour mixtures) in the painting ‘‘Racconta storie’’

(2003).

Results and discussion

Natural ageing of FDOCs

Based on TG data [21] after 48 h, the physical drying of the

FDOC paint replicas was complete as the paints had lost 18–19%

of solvent and were dry to the touch. The composition of the dry

paints from TG data was 10–40% filler (when present), 20–50%

pigment, and 40–70% w/w alkyd resin.

Ageing of the paint replicas was investigated by ATR-FTIR by

comparing FTIR spectra obtained 48 h after preparation up to

12,800 h. Fig. 3 shows a comparison of FTIR spectra of PW6 FDOC

(a) and PBK9 FDOC paint replicas (b) after 48 h curing at room

temperature and after 10 and 18 months ageing, along with their

corresponding pigment. Analogous FTIR spectra of the other paint

cm-1

A

470

3319

2921

2850

1819

1729

1429

1258

1167

1121

1071

877

726639

1635

p

a

b

c

4000 3200 2400 1800 1400 1000 600

4000 3200 2400 1800 1400 1000 600

cm-1

A

1409

1015

960872

777

697

599 558

463

3314

2921

2850 1730

1455

1413

1258

1169

1024

961

872

741

631

1635

599 558

p

a

b

c

A (PW6)

Fig. 3. FTIR spectra of alkyd FDOC paint replicas. (A) Titanium White pigment

(PW6, p curve); Titanium White FDOC paint replica after 48 h drying (a), after

10 months of natural ageing (b) and after 18 months of natural ageing (c). (B) Ivory

Black pigment (PBK9, p curve); Ivory Black FDOC paint replica after 48 h drying (a),

after 10 months of natural ageing (b), and after 18 months of natural ageing (c).



replicas prepared with the FDOC investigated are reported in the

Supplementary Data (Fig. SD-1: a–h).

The spectra show the typical peaks of alkyd resins: the broad

band at 3314 cm�1 due to OH groups, the peaks at 2921 and

2850 cm�1 to CAH asymmetric and symmetric stretching vibra-

tions of methylene groups, and at 1729 cm�1 to C@O stretching

of carboxylic acids and esters [32–35]. In the fingerprint region

we observed a peak at 1429 cm�1, due to the CH2 bending which

overlaps with the OACAO bending of the CO32� of dolomite

(CaMg(CO3)2) used as a filler. This overlap was observed in all paint

samples, apart from Viridian green and Ivory black, which do not

contain any filler. However, in PBK9, the stretching vibrations char-

acteristic of calcium carbonate (1455, 1413 and 872 cm�1) and cal-

cium phosphate (1015 and 558 cm�1) are present, as expected for

pigments from bone samples [21,36]. The peak at 1258 cm�1

(strong and slightly rounded) is likely due to the esters, and at

1120 cm�1 is due to CAH, CAC and CAO stretching. The band at

728–741 cm�1 is due to CAH bending of aromatic ring by phthalate

units, and the band at 875 cm�1 is due to stretching CAO of CO32�

[37]. In PW6 the large peak below 700 cm�1 is due to the pigment

titanium white.

After 48 h drying, none of paint replicas showed absorptions at

3011, 1654 and 723 cm�1 typical of isolated double bonds, indicat-

ing that unsaturated fatty acids were autooxidated [9,13]. With

ageing, we observed an increase and broadening of the OH group

peak around 3300 cm�1 and an increase in the bands in the region

1730–1680 cm�1 due to carbonyl stretching and the broad shoul-

der at 1635 cm�1. The increase in these bands may indicate the

oxidation reactions that take place during natural ageing and that

lead to the formation of alcohols and carbonyl species [1,19,38].

The broadening and increase in the band at 1635 cm�1 may be

due to the stretching vibration of C@O groups in diketones and

their enol form [19,38].

A decrease in methylene CAH absorptions at 2930 and

2855 cm�1 was also observed, in agreement with Muizebelt et al.

[19] and Perrin et al. [39], due to the oxidation of double bonds.

A low fraction of methylene groups are also susceptible to disap-

pearance through Norrish type I and Norrish type II reactions [39].

The spectral changes observed in this work on FDOC are similar

to those observed by Chiantore et al. [1] in the ATR-FTIR study of

Winsor & Newton alkyd colours (without drier) after 10 years of

ageing.

DSC analysis of these samples also showed that the glass tran-

sition temperature Tg was far above the room temperature (around

60 �C) after only 48 h, implying that the paint films are brittle at

room temperature [21]. With ageing the Tg values of all the sam-

ples slightly increased, but after 250 days, once the autooxidation

and cross-linking associated with chemical drying had ended, the

alkyd colours were very stable.

The absorbance value of the band at 1635 cm�1, described

above, which increases with ageing, is a representative parameter

of oxidation and can be used to follow the ageing process. Fig. 4

shows the trends in FTIR absorbance values at 1635 cm�1 as a func-

tion of the ageing time of all the FDOC paint replicas investigated.

FTIR spectra were normalized in the 1300–1200 cm�1 region for

PG18; in the 1200–800 cm�1 region for PBK9 and PB29; in the

1600–1300 region for Titanium White PW6 Titanium dioxide,

Phthalo Blue PB15 Copper phthalocyanine, Phthalo Green PG7

Chlorinated copper phthalocyanine, Cadmium Red PR108

Cadmium sulphoselenide, Winsor Red PR170-PR188 Naphtol

carbamides, Winsor Lemon PY3 Arylamide yellow, and Cadmium

Yellow PY35 Cadmium zinc sulphide.

The normalization regions were selected on the basis that they

do not change during ageing. The trends in these data show a

plateau around 100–200 days. After 200–300 days of natural

ageing the absorbance value of the band at 1635 cm�1 increased

again. This behaviour, which is more evident for PB15, PB29,

PG18 and PY3, implies the completion of an initial process, proba-

bly curing, i.e. a state in which most of the cross-linking related to

the coating film formation is completed. This is inagreement with

the results reported by Bartolozzi et al. based on the analysis of the

CHCl3 extracts of PB29 [22], and with our results obtained by DSC

data [21]. After almost 250 days, the exothermal DSC peak, associ-

ated with the peroxide decomposition and radical recombination,

disappears suggesting that the autooxidation and cross-linking of

the resin, via the double bonds consumption, have terminated.

After curing, the increase in the absorbance of the band at

1635 cm�1 could be related to a further oxidation due to ageing.

For other FDOCs, these two steps are not clearly visible. For PY35

and PR170-188, this parameter reached a plateau within our obser-

vation time. The different behaviours may be due to a different

interaction between the alkyd resin and the pigment, which may

affect the curing and the stability of resin, as evidenced also by

TG results [21].

On the basis of these data, the 1635 cm�1 absorption value

investigated over time in selected paint replicas prepared with

FDOCs, chosen as a model, may be used to setup a non-invasive/

microinvasive ATR-FTIR method to follow the ageing process of

paintings, without requiring any extractive procedure.

In this work we investigated a red sample taken from ‘‘Racconta

storie’’ by Patrizia Zara, a young Italian painter who uses Griffin

series FDOCs in her hyper-realistic paints. Fig. 5 shows the

ATR-FTIR spectrum of the red sample. In this case we know that

the filler (dolomite) and the resin formulation is the same as those

used for the paint replicas, although the pigment used in the origi-

nal colour is unknown.

The open symbol in the Cadmium Red plot refers to the value

found from processing the 10 year old red FDOC sample from ‘‘Rac-

conta storie’’. This value, which has been arbitrarily indicated at

600 days instead of 3650 days to avoid to change the x scale, is

approximately the same of the value obtained after 18 month age-

ing. Thus it results on a plateau.

By comparing the spectrum in Fig. 5 with our paint replicas pre-

pared with the red FDOCs we found a good match with Cadmium

Red (PR108) (Fig. SD-1). The absorbance value at 1635 cm�1 of the

FTIR spectrum of the Patrizia Zara red sample, which is 10 years

old, was on the plateau of the related trend reported in Fig. 4. This

is in agreement with DSC results obtained from the analysis of

samples taken from other Patrizia Zara paints and confirmed that,

after autooxidation had terminated, the alkyd colours were very

stable [21].

Study of ageing of FDOC in CH3COOH vapours

This aim of this part of the study was to highlight the main

chemical groups of FDOCs involved in the acid degradation using

ATR-FTIR spectroscopy. Despite the treatment of FDOC paint repli-

cas under an atmosphere saturated with acetic acid being far from

the actual conditions, the results obtained suggest which region of

the FTIR spectrum needs to be explored in order to highlight the

damage to real samples.

Macroscopically, all the FDOC paint replicas after six month’s of

ageing under acetic acid atmosphere showed a white efflorescence

on the surface, which was particularly visible in the PBK9 sample.

The two phthalocyanine alkyd colors (PB15 and PG7) completely

lost their pigmentation producing a white residue. This decolor-

ation could be explained by the reaction of acetate with Cu(II) of

phthalocyanine pigments which leads to the formation of stable

copper (II) acetate salts and free phthalocyanine, which are light

yellow colour species [26].

Fig. 6 shows representative FTIR spectra of Titanium White

(PW6), Phthalo Blue (PB15), Cadmium Yellow (PY35) and



Cadmium Red (PR108) FDOCs (dried for 48 h) before and after

6 months under an acetic acid atmosphere. FTIR spectra of the

other FDOCs investigated are reported in the Supplementary Data

(Fig. SD-2: a–f).

The spectra of acid aged samples containing dolomite as a filler

show the decrease in the absorption of peaks at 877 and 1432 cm�1

due to carbonate. This decrease corresponds to an increase in the

absorption of peaks assigned to HCO3� (671 and 1545–1580 cm�1)

0,02

0,04

0,06

0,08

PBK9

Ab

s

Days Days

Days Days

Days Days

Days Days

DaysDays

0,02

0,04

0,06

0,08

0,10

0,12

0,14

0,16 PW6

Ab

s

0,00

0,02

0,04

0,06

0,08

0,10

0,12

0,14PB15

Ab

s

0,02

0,04

0,06PB29

Ab

s

0,1

0,2

0,3PG18

Ab

s

0,00

0,05

0,10 PG7

Ab

s

0,00

0,02

0,04

0,06

0,08

0,10

0,12 PR108

Ab

s

P.Zara red FDOC (2004)

0,00

0,02

0,04

0,06

0,08

0,10

0,12

0,14

0,16

0,18 PR170- PR 188

Ab

s

0,03

0,04

0,05

PY3

Ab

s

0 100 200 300 400 500 600 0 100 200 300 400 500 600

0 100 200 300 400 500 600 0 100 200 300 400 500 600

0 100 200 300 400 500 600 0 100 200 300 400 500 600

0 100 200 300 400 500 600 0 100 200 300 400 500 600

0 100 200 300 400 500 600 0 100 200 300 400 500 600

0,00

0,05

0,10PY 35

Ab

s

Fig. 4. Trend of FTIR normalized absorbance value at 1635 cm�1 as function of ageing.



which forms from the reaction of dolomite with H+ ions of acetic

acid. In all the spectra there was also a decrease in the aliphatic

CAH group absorptions at 2920 and 2850 cm�1 and in the C@O

stretching absorption at 1730 assigned to the esters present in

the resin. These changes may be due to a partial depolymerization

and hydrolysis of the resin with the loss of organic volatile species

[40]. The increase in the absorption in the 3600–3300 cm�1 region

assigned to OAH stretching vibrations and in the 1720–1580 cm�1

region assigned to the C@O stretching vibrations, could be due to

the formation of mono- or di-carboxylic acids [40].

We observed several particular spectral features in samples

PY35, PR108 and PB29 FDOCs (see Fig. SD-2 a), due to the different

chemical composition of the pigments. The aged spectra of samples

PR108 and PY35, which both contain Cd and S complexes, show the

increase in the absorption at about 1115 cm�1 likely due to the

formation of S@O species.

Conclusions

FTIR-ATR spectroscopy was used to characterize the spectral

changes occurring during the natural ageing of paint replicas

prepared with alkyd FDOCs and after their exposure to an acetic

acid atmosphere for six months. We found that the absorbance

in the OH stretching region (3100–3600 cm�1) and the absorbance

value at 1635 cm�1 increased during curing and natural ageing.

The increase in absorbance of the band at 1635 cm�1 highlighted

two different processes during the ageing of the resin. The first pro-

cess terminated between approximately 200–300 days and can be

assigned to the curing of the resin, as confirmed by DSC data. After

300 days, the ageing of the film formed by the resin occured. On

this basis the absorbance value at 1635 cm�1 could be selected

as a parameter to follow the ageing process of paintings prepared

with FDOC using a non-invasive/microinvasive ATR-FTIR method,

without any extractive procedure.

4000 3200 2400 1800 1400 1000 600

cm-1

A

3289

2957 2925

2874

2851

2521 2161

1509

1432

1239

1145

1064

995

875

746

728

711

633607

514

1726

1647

Fig. 5. FTIR spectra of a red sample from ‘‘Racconta storie’’. (For interpretation of

the references to colour in this figure legend, the reader is referred to the web

version of this article.)

3319

29212850

1729

1429

1258

1167

1121

1071

1040

877

742

726

639

cm-1

A

3361 29151723

1574 1408

1344

12571119

1024979

880

656 615

A (PW6)

a

b

cm-1

A

3423

2922

2851

1731

15991580

1432

13341258

11191070

972

877

726721

651

3449

2921

2850 1977

1728

1546

1409

13671261

11191023877

721661

B (PB15)

a

b

cm-1

A

3314

29222851

1730

1599

1431

1257

1165

1119

10701040

980

876

728

633608

35303403

29192850

1729161914191256

1114

877727

671600

C (PY35)

a

b

4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600

4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600

cm-1

A

3312

2921

2850

1820

1729

1599

1429

1257

11651119

1070

981

876

728

634608

35323403

29172850

1727

16181571

1407

13191260

1115

875

798

727

670

601

D (PR108)

a

b

Fig. 6. FTIR spectra of alkyd resin paint replicas prepared with FDOCs (dried for 48 h) before (curve a) and after (curve b) 6 months treatment under an acetic acid

atmosphere. (A) Titanium White (PW6), (B) Phthalo Blue (PB15), (C) Cadmium Yellow (PY35), (D) Cadmium Red (PR108).



The absorbance value at 1635 cm�1 found in the red sample col-

lected from ‘‘Racconta storie’’ [[[[painted by Patrizia Zara about

10 years ago]]] was very similar to the value found in the FTIR

spectrum of Cadmium red FDOC after 550 days of natural ageing,

thus suggesting that after the end of chemical drying (autooxida-

tion), alkyd colours are very stable.

After six months of artificial ageing in an acetic acid atmo-

sphere, all the paint replicas looked very damaged. Although this

treatment was extreme with respect to the actual conditions, these

results suggest that the carbonate (dolomite) of the filler and the

carbonyl group of the esters present in the resin were involved

in the reaction with the acid vapours. Their absorptions could thus

be investigated in order to follow any degradation due to acid pol-

lution in real paintings.

Due to the recent use of FDOCs in contemporary art, the devel-

opment of any technique able to investigate their chemical charac-

teristics and ageing process is of interest in order to establish the

appropriate conservation and restoration strategies of paintings.

Acknowledgments

This work was carried out as a part of the ‘‘Preventative Conser-

vation of Contemporary Art’’ Project (COPAC 2011–2013), funded

by PAR-FAS Regione Toscana (Tuscany, Italy). The authors

gratefully acknowledge Prof. A. Bianchi and Prof. A. Bencini

(Dipartimento di chimica, Università di Firenze) for their help in

the synthesis of copper phthalocyanine (PB15) and chlorinated

copper phthalocyanine (PG7).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in

the online version, at http://dx.doi.org/10.1016/j.saa.2014.03.123.

References

[1] R. Ploeger, D. Scalarone, O. Chiantore, J. Cult. Heritage 9 (2008) 412.[2] M.D. Soucek, T. Khattab, J. Wu, Prog. Org. Coat. 73 (2012) 435.

[3] R.A. Sailer, M.D. Soucek, J. Appl. Polym. Sci. 73 (1999) 2017.[4] Bieleman, Additives for Coatings, Weinheim, 2000.[5] S.T. Warzeska, M. Zonneveld, R. van Gorkum, W.J. Muizebelt, E. Bouwman, J.

Reedijk, Prog. Org. Coat. 44 (2002) 243.[6] A. Spyros, D. Anglos, Anal. Chem. 76 (2004) 4929.[7] R. Ploeger, D. Scalarone, O. Chiantore, Polym. Degrad. Stab. 94 (2009) 2036.[8] S. Wei, V. Pintus, M. Schreiner, J. Anal. Appl. Pyrol. 104 (2013) 441.[9] M. Lazzari, O. Chiantore, Polym. Degrad. Stab. 65 (1999) 303.[10] D. Scalarone, M. Lazzari, O. Chiantore, J. Anal. Appl. Pyrol. 58 (2001) 503.[11] V. Stava, D. Vesely, P. Kalenda, Pigm. Resin Technol. 37 (2008) 67.[12] R. van Gorkum, E. Bouwman, Coord. Chem. Rev. 249 (2005) 1709.[13] S.M. Cakic, I.S. Ristic, J.M. Vladislav, J.V. Stamenkovic, D.T. Stojiljkovic, Prog.

Org. Coat. 73 (2012) 401.[14] G. Stamatakis, U. Knuutinen, K. Laitinen, A. Spyros, Anal. Bioanal. Chem. 398

(2010) 3203.[15] A. Phenix, J. Am. Inst. Conserv. 41 (2002) 61.[16] J. Mallegol, J.L. Gardette, J. Lemaire, J. Am. Oil Chem. Soc. 77 (2000) 257.[17] J. Mallegol, J. Lemaire, J.L. Gardette, Prog. Org. Coat. 39 (2000) 107.[18] Y. Ouldmetidji, L. Gonon, S. Commereuc, V. Verney, Polym. Test. 20 (2001) 765.[19] W.J. Muizebelt, J.C. Hubert, R.A.M. Venderbosch, A.J.H. Lansbergen, R.P.

Klaasen, K.H. Zabel, J. Coatings Technol. 70 (1998) 53.[20] M. Soucek, M.K.G. Johansson, Prog. Org. Coat. 73 (2012) 273.[21] C. Duce, L. Bernazzani, E. Bramanti, A. Spepi, M.P. Colombini, M.R. Tiné, Polym.

Degrad. Stab. (2014), http://dx.doi.org/10.1016/j.polymdegradstab.2014.03.035.

[22] G. Bartolozzi, V. Marchiafava, V. Mirabello, M. Peruzzini, M. Picollo,Spectrochim. Acta Part A: Mol. Biomol. Spectrosc. 118 (2014) 520.

[23] J. La Nasa, I. Degano, F. Modugno, M.P. Colombini, Anal. Chim. Acta 797 (2013)64.

[24] A.J. Cruz, J. Pires, A.P. Carvalho, M.B. de Carvalho, J. Cult. Heritage 9 (2008) 244.[25] V.C. Sharma, U.S. Lal, T. Singh, Stud. Conserv. 48 (2003) 203.[26] J. Tetreault, E. Cano, M. van Bommel, D. Scott, M. Dennis, M.G. Barthes-

Labrousse, L. Minel, L. Robbiola, Stud. Conserv. 48 (2003) 237.[27] A.L. Dupont, J. Tetreault, Stud. Conserv. 45 (2000) 201.[28] T. Baird, N.H. Tennent, Stud. Conserv. 30 (1985) 13.[29] A.W.V.B. Brokerhof, Maarten James & James, Science Publishers Ltd., 1996, p. 7.[30] J.R. Campanelli, M.R. Kamal, D.G. Cooper, J. Appl. Polym. Sci. 48 (1993) 443.[31] E.S.-R.a.I. Vancsó-Szmercsányi, Chem. Zvesti 26 (1972) 390.[32] Learner, Analysis of Modern Paints, Getty Publication, Canada, 2004.[33] N. Sonoda, Techne No. 8, 1998, p. 33.[34] S.N. Gan, B.Y. Tan, J. Appl. Polym. Sci. 80 (2001) 2309.[35] S.M. Cakic, I.S. Ristic, V.M. Jašo, R.Z. Radicevic, O.Z. Ilic, J.K.B. Simendic, Prog.

Org. Coat. 73 (2012) 415.[36] E. Tomasini, G. Siracusano, M.S. Maier, Microchem. J. 102 (2012) 28.[37] U. Knuutinen, P. Kyllonen, e-Preserv. Sci. 3 (2006) 9.[38] R. Ploeger, S. Musso, O. Chiantore, Prog. Org. Coat. 65 (2009) 77.[39] F.X. Perrin, M. Irigoyen, E. Aragon, J.L. Vernet, Polym. Degrad. Stab. 70 (2000)

469.[40] E.S. Rethy, I.V. Szmercsanyi, Chem. Zvesti 26 (1971) 7.


ftir study of ageing of fast drying oil colour (fdoc) alkyd paint replicas

Documents