J Culher 2016 07 001

G Model
CULHER-3106; No. of Pages 10 ARTICLE IN PRESS

Journal of Cultural Heritage xxx (2016) xxx–xxx
Available online at
ScienceDirect
www.sciencedirect.com
Original article
Preliminary investigation on the use of the Q-switched Nd:YAG laser

to clean corrosion products on museum embroidered textiles with
metallic yarns
Bojana Radojković a,∗ , Slɑvicɑ Ristić a , Suzɑnɑ Polić b , Radmila Jančić-Heinemann c ,
Dɑnijelkɑ Rɑdovɑnović d
a
Institute Gosa, Milana Rakića 35, Belgrade, Serbia
b
Centrɑl institute for conservation, Terazije 26, Belgrade, Serbia
c
Faculty of Technology and Metallurgy, Karnegijeva 4, Belgrade, Serbia
d
Ethnographic Museum in Belgrade, Studentski trg 13, Belgrade, Serbia
a r t i c l e i n f o a b s t r a c t
Article history: The classic methods of cleaning old-fashioned textile museum objects with metallic yarns do not often
Received 26 February 2016 yield expected results. The use of laser in the conservation of these objects plays an important role as a
Accepted 3 July 2016 very efficient and ecologically friendly technique. Conservation and restoration of historical exhibits with
Available online xxx
metal-textile combinations is becoming ever more complex due to a huge variety of damaging factors.
The paper presents the results of the corrosion laser cleaning effects on ethnographic textile with silver
Keywords: coated copper yarns from the holdings of the Ethnographic Museum in Belgrade, using Nd:YAG laser. The
Laser cleaning
testing of the cleaning effects was performed by optical and scanning electron microscopy. The chemical
Corrosion
Metal yarns
analysis of yarns was done by the EDX and XRD analyses. The parameters for successful and safe cleaning
Phosphine of corrosion products on metallic yarns were determined.
SEM © 2016 Elsevier Masson SAS. All rights reserved.
XRD
1. Research aims cleaning method than the classic ones. The evaluation of the tested
areas can lead to important conclusions and a further laser cleaning
The paper presents the results of the research on laser appli- of the shirt will be based on the obtained results.
cation for cleaning corroded metal yarns on embroidered textile
artworks. The aim of this work is to determine optimum laser
parameters for a successful cleaning of corroded metal yarns with-
2. Introduction
out any damage to metal strips and underlying cotton, on a sample
of a female shirt from the museum collection. Different laser wave-
The deterioration of the museum objects depends on environ-
lengths and fluencies were used in dry and wet experimental
mental factors such as humidity, temperature, light, pollutants,
conditions. Morphological and chemical modifications in the laser-
microorganisms and insects. It induces structural damage, modifi-
cleaned zones were investigated by optical and scanning electron
cation of materials, corrosion and agglomeration of pollutants and
microscopes, as well as by energy-dispersive (EDX) and X-ray Pow-
microorganisms on artworks stored in museum depots or exhibited
der Diffraction (XRD) spectroscopy. The obtained results confirmed
in galleries.
that for these samples, which are a combination of two differ-
Numerous research works describe the classic methods imple-
ent materials, silvered copper strip and cotton, the application of
mented in cleaning of metallic yarns in textile museum objects
Nd:YAG lasers with appropriated parameters is a more effective
[1,2]. Traditional cleaning methodologies usually apply mechanical
removal or chemical reactions, or a combination of these two meth-
ods [3]. An ideal cleaning treatment should remove a tarnished
layer without affecting either the underlying metal surface or the
∗ Corresponding author. Tel.: +381 64 83 89 828.
fabric.
E-mail addresses: bojana.radojkovic@institutgosa.rs (B. Radojković),
slavica.ristic@institutgosa.rs (S. Ristić), suzanapolicradovanovic@gmail.com
Lasers can controllably and selectively remove undesired layers
(S. Polić), radica@tmf.bg.ac.rs (R. Jančić-Heinemann), rada.n@sezampro.rs from surfaces, with high accuracy. Results of laser cleaning depend
(D. Rɑdovɑnović). on a number of parameters related to laser and object surface
http://dx.doi.org/10.1016/j.culher.2016.07.001
1296-2074/© 2016 Elsevier Masson SAS. All rights reserved.
Please cite this article in press as: B. Radojković, et al., Preliminary investigation on the use of the Q-switched Nd:YAG
laser to clean corrosion products on museum embroidered textiles with metallic yarns, Journal of Cultural Heritage (2016),
G Model
2 B. Radojković et al. / Journal of Cultural Heritage xxx (2016) xxx–xxx
characteristics. The most complicated case is the use of lasers in to artworks stabilization, and possible further conservation proce-
cleaning items with multi-component structures [4–7]. dures.
Laser cleaning is an effective cleaning technique of metal cor- Silver objects can form a thin patina, depending on the aggres-
rosion layers since it provides a high degree of control that allows siveness of the environment [24]. Silver is particularly attacked by
objects with a considerable amount of surface detail and differ- sulphide gases. Silver sulphide usually appears as a black layer of
ent materials to be successfully and safely cleaned. There are many corrosion on silver museum objects and this kind of corrosion is
investigations of lasers applications in cleaning glass [8], ceramics known as tarnish.
[9], stone [10,11], easel and wall painting [11,12], metal objects
[11,13–15], textile [5–7,16–18], etc. Determination of the most
appropriate methodology for successful laser cleaning contains an 4. Sample description
optimization of laser parameters in accordance with material prop-
erties. Also, a good knowledge about the ablation mechanism that The Ethnographic Museum in Belgrade stores a great number
occurs is required [1,16,17,19]. of textile objects which are part of the world cultural heritage and
Nd:YAG lasers are highly effective in conservation of artworks, it puts significant efforts to provide the necessary conditions for
as it has been confirmed in numerous tests. Siano et al. [11] present preventive protection [4]. In spite of all efforts, in the early nineties
a state-of-the-art, innovation and a new perspective application of of the past century, changes as dark and brittle yarns were observed
Nd:YAG lasers in cleaning different artworks. The cleaning of dirty, on several objects with metallic yarn embroidery. With time, this
corroded surfaces of two copper alloys archaeological objects with phenomenon became visible on a growing number of exhibits.
Nd:YAG laser, with a wavelength of = 1064 nm and a laser pulse of The studied shirt, dated from 1914, was made of cotton embroi-
120 ns, is presented in [13]. The most complicated implementation dered with cotton and silver-coated copper yarns. The photos of
of laser in cleaning objects with multi-component compositions the shirt are shown in Fig. 1 (a and b were recorded in 1964 and
such as a metal and textile combination is given in [1,5–7,16,19,20]. c–e in 2013). Silver-coated copper sequins were fixed with glass
Abdel-Kareem et al. [16] used the Nd:YAG laser radiation of a beads using a cotton thread. It is also decorated with glass beads,
wavelength of 532 nm for cleaning some corroded copper thread strung on a cotton thread (Fig. 1b). The parts with the metallic yarn
samples with cotton fibre core. Nd:YAG lasers influence on sil- embroidery are sewn between the pieces of tulle [4].
ver and copper plates and silk bands as on real artefacts (satin Donja Brnjica is a village five kilometers north of Priština. In the
with tarnished silver yarns and corroded silvered copper yarns) whole Serbian cultural heritage, so consequently on Kosovo and
were investigated in the study of Degrigny et al. [1]. The Nd:YAG Metohija, embroidery is a very important method of decorating
laser used in this study had high fluence but a lower puls fre- textile items. With industrialization which at that time took place
quency in comparison with the tests presented here. Also, their in Serbia, embroidery techniques were improved. Wedding shirts
experiments were conducted in the helium atmosphere. The laser were specially decorated with gilded and silver yarns.
cleaning of yarns in museum textiles, consisting of a silver ribbon The shirt embroidery was in an excellent condition in 1964
wrapped around a group of silk fibres, was reported in [5–7,19]. (Fig. 1a and b). The shirt was kept in a storeroom treated with dif-
The investigation of Elnaggar et al. [6] used picosecond laser scan- ferent pesticides. Mothballs based on paradihlorbenzol were placed
ning pulses (1064 nm) for the removal of tarnish from the surface in textiles depots from 1968 to 1984, after the regular, annual air
of silvered metal yarns. Unlike previous research, in the study the ventilation. From 1984 to 1999, the pest control of the depots was
results of which are analysed here, the sample was made of cotton done with fumigants such as Nuvan 7 and phosphine. Afterwards
embroidered with silver-coated copper ribbon wrapped around a protection was carried out by the herbal strewing solution (tansy).
group of cotton fibres. Also, the nanosecond pulse regime was used During the years, defects caused by corrosion have appeared and
for irradiation. Although there are certain published results, this have slowly led to the destruction of the embroidered parts.
domain has not been sufficiently researched because every object is The tests of corrosion causes began in 2000. The measurement
a problem that requires a study and comprehensive research before on the metal yarns showed that the pH value was 3.5, which means
cleaning is attempted. that the reaction was substantially acidic. The damaged silver sam-
This paper shows the results of the research conducted with the ples, tested by emission spectroscopy, showed significant presence
aim of determining the optimum parameters in the cleaning pro- of phosphorus (9%).
cess of the corrosion product on a female shirt from Donja Brnjica, Metal phosphide tablets, used in depot fumigation, develop
Kosovo and Metohija, Serbia. The shirt was handmade, in the early phosphine. In the atmosphere, which has a higher oxygen con-
twentieth century and it represents a part of a Serbian national tent, phosphine tends to transform into a stable form – phosphoric
wedding costume. A very small number of similar specimens are acid. Phosphine reacts with precious metals and causes corrosion.
preserved. With the increase of temperature and humidity, the reaction rate
The results obtained by using Nd:YAG laser, were analysed by increases. Also, condensation of water vapor over abrasive and
an optical microscope and a scanning electron microscope (SEM) hygroscopic dust particles on the surface of museum objects can
with the EDX and XRD analyses. stimulate the growth of microbes and accelerate corrosion.
The cleaning of the metallic yarn on the shirt was performed
by applying a classic method: it was treated with a 75% aqueous
3. Degradation of embroidered textile artworks solution of phosphoric acid (H3 PO4 ). After washing with distilled
water, neutralization was carried out in a sodium bicarbonate, 1%
In embroidered textile objects, various metal yarns can be found. aqueous solution, followed by another rinsing with distilled water.
The thickness of precious metal coating is measured in microns Some parts of the embroidery were protected with paraloid B82.
[21]. The corrosion processes can occur under external, mechanical, During the cleaning process, the embroidery began to fall off from
chemical and biological influences [22,23]. the fabric (Fig. 1e). Detail tests were performed in order to deter-
Common corrosion products, generated in a reaction of copper mine the effect of chemicals on the textile. These tests included:
and its alloys with the environment, are copper oxides, basic cop- visual analysis of the samples, determination of the raw material
per sulphates, copper chloride and basic copper carbonates. Some composition, effects of the chemical treatment of the samples, cal-
corrosion products protect the underlying metal from further cor- culation of the concentration of the acid and neutralizer regarding
rosion. However, copper chloride cleaning is necessary with a view to pH values, as well as testing shrinkage and strength [4].
G Model
B. Radojković et al. / Journal of Cultural Heritage xxx (2016) xxx–xxx 3
Fig. 1. The shirt, Donja Brnjica, Kosovo and Metohija, Serbia photographed in 1964 (a and b) and 2013 (c–e).
The preliminary protection measures, mentioned above, did not Table 1

Experimental parameters.
give good results. Corrosion process had not been stopped. The pho-
tographs taken in 2013 (Fig. 1c–e) show the parts of the embroidery Zone Fluence Wavelength Irradiation time (s) Comment
with loosen yarns that are in a very poor state (Fig. 1d). (mJ/cm2 ) (nm) (and number of pulses)
1 100 532 5 (100) Dry

2 200 532 5 (100) Dry
5. Experiment
3 250 532 5 (100) Dry
4 200 1064 5 (100) Dry
The experiment was performed using a commercial Nd:YAG 5 200 1064 15 (300) Dry
laser, Thunder Art Laser, a product of the Quanta System. The laser 6 200 1064 15 (300) Wet
can operate with three wavelengths, 1064 nm, 532 nm and 355 nm. 7 290 1064 15 (300) Wet
8 380 1064 15 (300) Wet
It operates in the Q-switch mode. The pulse duration is 8 ns. The
repetition rate is 20 Hz and the unfocused laser beam diameter
is 10 mm, with the Gaussian distribution of energy. The energy
of the laser beam can be changed depending on the wavelength experimental parameters for the analysed zones are provided in
as follows: for 1064 nm maximum energy to 1000 mJ, for 532 nm Table 1. Dry cleaning was performed on zones 1–5, while the parts
to 550 mJ, and for wavelength 355 nm to 200 mJ. The laser has of the sample from zones 6–8 were wetted before irradiation. The
a mobile, articulated arm through which the laser beam is con- surface was sprayed with water before cleaning. Wet laser cleaning
ducted with the aid of seven mirrors. The laser arm can approach was performed with an aim to get better cleaning efficiency by the
the sample under different angles and from a different distance. 1064 nm wavelength. The diameter of the focused laser beam on
The effect of the laser on the sample was performed under nor- the sample surface was around 8.0 mm.
mal atmospheric conditions. The preliminary tests, performed on The morphology of the sample surface was recorded with a USB
silvered copper plates (with different number of pulses and fluen- optical microscope and a scanning electronic microscope (SEM),
cies), had shown that the effects of corrosion cleaning were visible JEOL JSM–6610LB. The SEM is linked to the energy dispersion X
for the number of pulses above 100 and the laser fluence above spectrometer (EDX), INSA350, for the preliminary analysis of the
100 mJ/cm2 . A smaller number of pulses, 30, required high flu- sample chemical composition. Before the SEM analysis, the sam-
ence above 800 mJ/cm2 , which caused damage on cotton fibers. The ple was steamed with a 20 nm thick layer of gold, with the aim of
G Model
Fig. 2. a: part of the shirt embroidery with laser cleaning zones; b and c: zoomed parts with corrosion layers.
increasing the electrical conductivity, namely, obtaining a better phenomena associated with the interaction of laser radiation with
quality of the picture of the sample area structure. the encrustation, the corrosion layers and the base substrate.
The samples of the embroidery fallen off the shirt were stud- The corrosion layer cleaning from metallic yarn embroidery
ied on a powder diffractometer PHILIPS PW 1710 with a purpose by laser is a complex phenomenon, its explanation requires the
of obtaining as much information as possible about the corrosion knowledge of the characteristics of laser lights and materials. It
products. is based on several processes: absorption of laser energy within a
very short period of time (several ns), melting of material in the
6. Results and discussion heating layer depending on the applied energy, and ablation or
evaporation of material. Depending on the fluence and the quan-
Museums around the world, have been using phosphine-based tity of the absorbed energy, mechanical expansion waves can be
products for protection against insects and rodents for years. Phos- formed, consisting of evaporated material and ambient gas, which
phine is extremely volatile and reacts with metals, particularly with also remove parts of the surface layers and discards them from the
copper, zinc, silver and gold [25–27]. The tests conducted at the irradiated zone. The strength of the expansion wave increases if a
Vinča Institute in Belgrade [25] showed that phosphine gas, used sample is damp (very fast water evaporation), and thereby creates
as a fumigant in this museum, most probably had accelerated the the conditions for more efficient cleaning of corrosion products.
corrosion process on the shirt. High temperature, and relative air This paper analyses the results of the irradiation effect of the
humidity accelerates the reaction, as well as a possible presence of Nd:YAG lɑser beams with two wavelengths, 1064 nm and 532 nm,
ammonia. The adsorbed layers of phosphine can react under certain on the shirt parts with metallic embroidery. Fig. 2a shows a
conditions with metals, creating phosphides that, further reacting photograph of a sample, with the dimensions of approximately
with the air, turn into metal and phosphorus oxides. Phosphorus 50 × 50 mm, with the zones exposed to the laser light with different
oxides react with humidity from the air, resulting in phosphorus fluences, wavelengths, number of pulses and quantity of moisture.
acid, which, with other acids possibly existing in the atmosphere, Fig. 2b and c show the zoomed parts with the corrosion layers. On
can create insoluble or poorly soluble salts with metals. A particular the basis of the SEM images, it was estimated that the thickness of
problem is a possibility of forming a gold (or silver)/electrolyte- the embroidery yarns was 0.6 mm, and that the metal strip width
phosphorous acid/copper (or zinc) compound. Then, in the anodic was 0.7 mm, with the metal strip thickness of around 10 ␮m. The
corrosion process, copper or zinc, which are electrically positive, thickness of the silver layer could not be determined because cor-
dissipate and cause crumbling of the yarn. rosion had deteriorated it. The thickness of the corrosion products
Preservation of the artworks surface during laser cleaning is was in the range from 10 ␮m to 50 ␮m. All corrosion layers were
related to the selection of the optimal process parameters, the laser to be removed.
beam wavelength, the energy density, the laser pulse length, and Before cleaning the embroidery sample, the preliminary tests
the environmental conditions in which the cleaning is performed. were done on a 0.1 mm thick copper plate covered with a sil-
These factors directly affect the thermal and photomechanical ver layer of 5 ␮m, in order to determine the laser fluence that
G Model
Fig. 3. SEM of irradiated zones on trial sample, silvered copper plate.
does not damage the silver layer. The trial tests were carried out This phenomenon is difficult to avoid during laser corrosion clean-
with two wavelengths (1064 and 532 nm) and with the number ing because a reduced input fluence reaches the copper surface
of pulses from 20 to 200. The fluence was from 100–800 mJ/cm2 . which is below the corrosion product [28].
Fig. 3 shows the SEM images of the irradiated surface zones with: Zones 6–8 were irradiated with the same = 1064 nm, the same
(a) = 1064 nm, = 100 mJ/cm2 , (b) = 532 nm, = 200 mJ/cm2 , (c) number of pulses and with fluence values increasing from 200
= 1064 nm, = 300 mJ/cm2 and (d) = 532 nm, = 375 mJ/cm2 . (Fig. 5) to 380 mJ/cm2 . The difference in the experimental condi-
The damage, such as a detached silver layer are visible in the tions between zones 5 and 6 is the moisture percentage. The results
zones irradiated with the fluences above 300 mJ/cm2 . Therefore, in show that, there was no significant improvement in cleaning, in the
this embroidery sample treatment, the fluences below 300 mJ/cm2 presence of water (zone 6), when lower energies were used. This
were used, except for one zone (zone 8) (Fig. 2). is in accordance with the results that Degrigny et al. [1] obtained
in their study. Only, the increase of fluence, in combination with
6.1. Optical microscopy the increased moisture, contributes to better cleaning effects, con-
firmed by the pictures of zone 8 (Fig. 5g–i). On the other hand, it is
The results of the laser beam-sample interaction were recorded necessary to have an insight about how much moisture can damage
with an optical USB Digital Microscope, Edmund Optics, magni- the textile parts of the sample (i.e. what is the stability of the fabric
fication 10×–200×, progressive scan CMOS sensor, 6 LEDs and a colour).
working distance of 8.5–112.0 mm. The chosen parameters of corrosion product cleaning are below
The value = 532 nm was used to expose zones 1, 2 and 3, and the damage threshold of metallic yarns and textiles [16,19], because
the fluence increased from 100 to 250 mJ/cm2 . The pulse number the aim was to determine the efficient cleaning threshold. The
was the same for all three zones. It is obvious that the cleaning optimization of the procedure may be achieved by the change of
efficiency, increased with the increase of fluence (Fig. 4a–e). The the laser beam incidence angle in order to irradiate the partially
cleaning effects are visible in the central part of the irradiated zones shielded parts as well.
with an approximate diameter of 5 mm, due to laser beam with the As an illustration, zone 5 (Fig. 4) is presented with two pho-
Gaussian energy distribution. The photo of zone 4 recorded with tographs (200×), taken at two different planes of the metallic
200× magnification is presented in Fig. 4f. embroidery on the shirt. It is obvious that the internal layers were
The comparison of zones 2 and 4 (Fig. 4) shows that cleaning not cleaned enough and that it is necessary to increase the num-
is more efficient with the wavelength in the near infra-red area ber of pulses and change the beam incidence angle. A difference
compared to the cleaning with the wavelength in the visible part is visible between the plane where the laser beam was focused
of the spectrum (523 nm). and the plane behind, where the fluence had a lower value. Due
The laser beam with the 1064 nm wavelength was used for zones to the absorption and dissipation the fluence was significantly
4 to 8 (Figs. 4d, f and 5). The fluence for zones 4 (Fig. 4d and f) and reduced and the cleaning effects were poorer in the inner layers
5 (Fig. 4g–i) was the same. The number of pulses was increased (Fig. 5).
for zone 5, which resulted in more efficient removal of corrosion Thermal effects on laser-cleaned textile fibers such as fiber soft-
deposits on the metallic yarns (Fig. 4h). On the cleaned surface, dark ening (melting), shape changing and a sort of “welding” between
areas caused by further oxidation of Cu2 O to CuO can be noticed. fibers are not visible in the used range of energy and the high
G Model
Fig. 4. The 1–5 zones of cleaning shirt embroidery.
number of pulses. A lot of crystals of corrosion products are spread products. Also, spectra are determined on the cotton fibre in the
on the cotton fibers (Fig. 5). core of the metal yarn. In zone 2, cleaned with = 532 nm, spec-
trum 4 shows the presence of phosphorus. Spectrum 2 in zone 5,
6.2. Image analysis cleaned with the same fluence as zone 2 but with = 1064 nm and a
higher number of pulses, shows that there is not more phosphorus
For the cleaning degree determination, the images obtained by on the cotton fibre. Also, phosphorus is not found in spectrum 5
optical microscopy were analysed and processed by the Image-Pro of zone 6, which confirms that phosphorus is eliminated not only
Plus programme. The laser-cleaned zones were marked while the from the metal part, but also from the organic fibre of the metal
areas of copper and corrosion products are separated by differ- yarn, when = 1064 nm is used.
ent colour and their percentage was automatically calculated. The In these study, laser cleaning was applied just on the part of
spaces between the yarns were not considered. The obtained results the embroidery without the cotton-based material. The part of the
presented on the diagram with image-Pro Plus photos (Fig. 6) show shirt embroidery was protected with paraloid B82. In this study an
that the percentage of the cleaned areas increases with the fluence unprotected part of the embroidery was used. The EDX spectra of
and wavelength. There is no significant difference between the per- the untreated yarns areas are consistent with spectrum 3 of zone
centage of the cleaned areas when either wet or dry conditions are 2.
used in a combination with laser beams with the same fluence and The EDX analysis (Table 2) shows that there is silver in the cor-
= 1064 nm. rosion products but not on all metal surfaces (silver which was
damaged to a great extent by corrosion). Phosphorus was registered
6.3. SEM and EDX analysis in the spectra taken from the parts covered with residue corro-
sion products or textile yarns around which metal (copper) yarns
The metallic yarns surface structures and morphology were were wrapped. This means that the corrosion had progressed to
monitored by an optical microscope and a scanning electron micro- such an extent that the silver layer had completely disappeared in
scope. Microscopic tests, conducted by SEM, allow the study of the some places and only the copper base of the yarn remained. The
fluence threshold that is safe and efficient for corrosion product presence of phosphorus was also evident in the corrosion products
cleaning. The chemical analyses of the cleaned zones were done which also confirm the results of the previous tests and the con-
with the aim of determining the degree of cleaning, the composi- clusion that phosphine gas, used as a fumigant, had most probably
tion of the corrosion products and the metallic yarns. Fig. 7 shows accelerated the corrosion process on the shirt.
the SEM photographs of zones 2 to 7, as well as the spectra in these The microscopic analyses performed on the corroded samples
zones on the metal strips and the cotton fibers by energy-dispersive before laser cleaning, show the presence of salt crystals in the most
spectroscopy (EDX). The spectra are determined both on the yarn damaged parts. The greyish-black colour at some places can origi-
areas with the corrosion products, and on the areas without these nate from phosphide that did not react further. It is also necessary
G Model
Fig. 5. Laser cleaning zones 6–8.
Table 2
EDX results, processing option: all elements analysed (normalized).
Spectrum C O P Cu Ag
Zone 2
Spectrum 1 56.62 17.78 25.60
Spectrum 2 40.96 9.68 49.35
Spectrum 3 30.64 56.13 6.14 6.99 0.10
Spectrum 4 55.07 41.53 1.24 2.15
Zone 5
Spectrum 1 47.23 15.60 37.16
Spectrum 2 56.64 42.36 1.00
Spectrum 3 55.12 15.21 29.00 0.68
Zone 6
Spectrum 1 48.22 14.54 36.81 0.43
Spectrum 2 45.90 11.35 42.38 0.36
Spectrum 3 38.91 43.27 4.28 12.41 1.14
Spectrum 4 53.85 20.73 25.42
Spectrum 5 57.13 41.78 1.09
Zone 7
Spectrum 1 46.93 14.82 37.58 0.67
Spectrum 2 55.22 5.24 39.55
Fig. 6. Analysis of laser cleaned zones with Image Pro Plus programme.
to mention, that the decomposition of copper phosphide is much

slower than the decomposition of gold and silver phosphide. The 6.4. X-ray diffraction analysis
information about the percentage of the content of chemical ele-
ments, can lead to a conclusion about the occurring reactions type The samples of the copper embroidery on the shirt were tested
and its extent. This information certainly helps in determining fur- on a powder diffractometer PHILIPS PW 1710 under the follow-
ther testing and rehabilitation steps. ing conditions: operating voltage: U = 40 kV, current intensity:
The results of the microscopic analyses of the metal yarns before I = 30 mA, X-ray radiation from anticathode copper (Cu), wave-
and after laser cleaning confirm that there is a noticeable decrease length: CuK˛ = 0.154178 nm, graphite monochromator, test range:
of salt crystals and increase in the percentage of copper in the metal 10–110◦ 2, step: 0.02◦ 2 time constant: 0.8 s per step.
yarn samples after laser cleaning. The laser cleaned surface retained The obtained data on the position of the diffraction maximums
more of the original topographical structure. 2 (◦ ), the value of the interplanar distances dhkl (nm)
G Model
Fig. 7. SEM photo and EDX spectra for zones 2, 5, 6 and 7.
for all hkl reflections, as well as the respective relative Due to the nature of the samples it was not possible to make a
intensity I/Imax are presented both graphically and in tables. perfectly flat and smooth surface on which x-rays diffract, hence
Based on the obtained values of the intensity I/Imax and the somewhat higher errors in the 2 measurement.
interplanar distances d and by comparing them with the ref- Only a single crystal phase was identified in the sample. This
erence data and the Joint Committee on Powder Diffraction was copper that has a structure of the most dense tesseral packing.
Standards (JCPDS standards), the present crystal phases were The Miller indices of individual reflections of this crystal phase are
identified. designated in Fig. 8.
G Model
chemical methods, because the only waste generated in this

process consists of impurities ejected from the artwork surface.
The obtained results can also be useful in conservation of some
metals in the industry. In the forthcoming period, it is necessary to
study the aging process, the change of colour and pigment on textile
objects, the appearance of re-corrosion processes and other modi-
fications that may occur after the implementation of laser cleaning
in the process of conservation of textile objects with metallic
yarns.
Acknowledgements
This research was financially supported by the Ministry of Edu-

cation, Science and Technological Development of Serbia under
Projects TR-35046 and TR-34028. The authors especially thank the
Fig. 8. XRD diagrams of the corrosion products, the Miller indices. Ethnographic Museum in Belgrade, Serbia.
The experimentally determined parameters of a unit cell References

a = 0.36202(6) nm show a slight increase (a = 0.0053 nm) com-
pared to aCu of pure copper (aCu = 0.36149 nm), which indicates that [1] C. Degrigny, E. Tanguy, R. Le Gall, V. Zafiropulos, G. Marakis, Laser cleaning of
this is probably a copper alloy with a very small percentage of some tarnished silver and copper yarns in museum textiles, J. Cult. Herit. 4 (2003)
152s–156s.
other metal of greater radius. [2] C. Frade, P. Cruz, E. Lopes, M.M. Sousa, J. Hall, R. Santos, A. Aguiar- Ricardo, T.
From the very weak and wide diffraction maximums on the Casimiro, Cleaning classical Persian carpets with silk and precious metal yarn:
small 2 angles, it can be concluded that the sample contains a conservation and ethical considerations, in: J. Bridgland (Ed.), Proceedings
of ICOM-CC 16th Triennial Conference, Lisbon, September 19–23, 2011
small quantity of at least one other crystal phase or, which is (CD ROM).
more probable, several crystal phases. The value of the interplanar [3] A. Timar-Balazsy, D. Eastop, Chemical principles of textile conservation,
distances d indicates that these are probably hydrated chlorides, Butterworth-Heinemann, Boston, Oxford, 1998.
[4] D. Rɑdovɑnović, Degradation of metallic yarn on textile objects at the Ethno-
chlorates (atacamite, Cu2 Cl(OH)3 , JCPDS-78-372, ICSD-61252; Cu-
graphic Museum in Belgrade, Ethnographic Museum, Belgrade, 2008 (In
chloride dehydrate, CuCl2 (H2 O)2 , JCPDS-88-1697, ICSD-40290; Serbian).
Cu-perchlorate hexahydrate, (Cu(H2 O)6 )·(ClO4 )2 , JCPDS-79-728, [5] M. Sokhan, F. Hartog, D. McPhail, Surface analysis of the laser cleaned metal
ICSD-65683) and sulphates (brochantite, CuSO4 ·3Cu(OH)6 , JCPDS- yarns, in: Lasers in the conservation of artworks LACONA V proceedings,
Osnabrück, Germany, Sept. 15–18, 2003, pp. 237–244, http://dx.doi.org/
87-454, ICSD-59288; chalcanthite, CuSO4 ·5(H2 O), JCPDS-11-646, 10.1007/3-540-27176-7 29.
ICSD-4305) copper, which cannot be determined with certainty due [6] A. Elnaggar, P. Fitzsimons, A. Nevin, I. Osticioli, M. Ali, K. Watkins, Investigation
to their very minimal presence and a low degree of crystallinity. of ultrafast picosecond laser system for cleaning of metal decorations of 17th
C. gloves of King Charles I. Investigation of silk by microscopy, e-Preservation
Sci. (e-PS) 12 (2015) 14–19 (ISSN: 1854-3928 print edition).
7. Conclusion [7] B. Taarnskov, P. Pouli, J. Bredal-Jørgensen, in: Radvan<ET-Al> (Ed.), Lasers in
the conservation of artworks VIII, Taylor and Francis Group, London, 2011, pp.
67–73, http://dx.doi.org/10.1201/b10567-12 (Print ISBN: 978-0-415-58073-
The analysis of the Nd:YAG laser cleaning of the corrosion prod- 1).
ucts on the metallic yarns of female shirt has shown that this [8] B. Radojković, S. Ristić, S. Polić-Radovanović, Study of ruby laser beam interac-
method can help in resolving a long-lasting problem of appearance tion with glass, FME Transactions 41 (2013) 109–113.
[9] S. Ristic, S. Polic-Radovanovic, B. Katavic, M. Kutin, Z. Nikolic, M. Puharic, Ruby
of corrosion on metallic embroidery on textiles. laser beam interaction with ceramic and copper artifacts, J. Russ. Laser Res. 31
The basic goal was to remove corrosion products and other (2010) 401–412.
deposits on the yarns without applying classic methods. Some [10] M. Speranza, M. Sanz, M. Oujja, A. De los Rios, J. Wierzchos, S. Pérez-Ortega,
M. Castillejo, C. Ascaso, Nd-YAG laser irradiation damages to Verrucaria
parameters for successful and safe cleaning of corrosion products nigrescens, Int. Biodeter. Biodegr. 84 (2013) 281–290.
on metallic yarns were varied and the results show that the safe [11] S. Siano, J. Agresti, I. Cacciari, D. Ciofini, M. Mascalchi, I. Osticioli, A.A. Mencaglia,
laser fluences should not be higher than 250 mJ/cm2 for all wave- Laser cleaning in conservation of stone, metal, and painted artifacts: state of
the art and new insights on the use of the Nd:YAG lasers, Appl. Phys. A-Mater.
lengths and the number of pulses between 100 and 300 depending 106 (2012) 419–446.
on wavelengths (100 for = 1064 nm and 300 for = 532 nm). The [12] R. Bordalo, P.J. Morais, H. Gouveia, C. Young, Laser cleaning of easel paintings:
increase of the number of pulses, and the decrease of fluences an overview, Laser Chem. ArticleID90279 (2006), http://dx.doi.org/10.1155/
2006/90279 (9 pages).
resulted in a more efficient removal of corrosion deposits on metal-
[13] H. Garbacz, E. Fortuna-Zalesna, J. Marczak, A. Koss, A. Zatorska, G.Z. Zukowska, T.
lic yarns. Onyszczuk, K.J. Kurzydlowski, Effect of laser treatment on the surface of copper
The comparison of different zones (Figs. 4 and 5) shows that alloys, Appl. Surf. Sci. 257 (2011) 7369–7374.
[14] B. Katavić, S. Ristic, S. Polic-Radovanovic, Z. Nikolic, M. Puharic, M. Kutin, Anal-
cleaning is more efficient with the wavelength in the near infra-
ysis of the threshold of damage of copper and aluminium in interaction with
red, compared to the cleaning with the wavelength in the visible Ruby laser, Chem. Ind. 64 (2010) 447–452.
part of the spectrum (523 nm). [15] R. Pini, S. Siano, R. Salimbeni, M. Pasquinucci, M. Miccio, Tests of laser cleaning
The chosen parameters of corrosion products cleaning are below on archaeological metal artefacts, J. Cult. Herit. 1 (2000) S129–S137.
[16] O. Abdel-Kareem, M. Harith, Evaluating the use of laser radiation in cleaning
the limit of damage to metallic yarns and textile. The optimiza- of copper embroidery yarns on archaeological Egyptian textiles, Appl. Surf. Sci.
tion of the procedure may be achieved by changing the laser beam 254 (2008) 5854–5860.
incidence angle in order to irradiate the partially shielded parts as [17] F. Ferrero, F. Testore, Surface degradation of linen textiles induced by laser
treatment: comparison with electron beam and heat source, AUTEX Res. J. 2
well. (2002) 109–114.
The laser cleaning of museum objects consisting of combination [18] S. Ristić, S. Polić-Radovanović, B. Radojković, Laser cleaning of textile artefacts
of different materials has a lot of advantages, both in a qualitative with corroded metal yarns, in: Z. Anastasijevic (Ed.), Proceedings of the 6th
international scientific conference, OTEH 2014, Belgrade, October 09–10, The
and economic sense, especially when cleaning corroded metal Military Technical Institute, Belgrade, 2014, pp. 649–655.
yarns in textile fabrics. The tests analysed in this paper confirmed [19] J. Lee, J. Yu, Y. Koh, Experimental study on the effect of wavelength in the laser
that laser technology is environmentally more suitable than classic cleaning of silver yarns, J. Cult. Herit. 4 (2003) 157s–161s.
G Model
[20] I. Rezic, L. Ćurković, M. Ujevic, Simple methods for characterization of metals [25] L.J. Petkovska, B. Radak, Results of emergency tests–analysis of the state of the
in historical textile yarns, Talanta 82 (2010) 237–244. process and causes of degradation of metal yarns on textile objects at the Ethno-
[21] http://www.conferences.saxo.ku.dk/traditionaltextilecraft/keynote speakers/ graphic Museum in Belgrade, Institute for nuclear sciences Vinča, Belgrade,
presentations/Anna Karatzani.pdf (May 26th 2015). 2000 (In Serbian).
[22] E.J. Bond, T. Dumas, S. Hobbs, Corrosion of metals by the fumigant phosphine, [26] R.J. Brigham, Corrosive effects of phosphine, carbon dioxide, heat and humidity
J. Stored. Prod. Res. 20 (1984) 57–63. on electronic equipment: phase II, Bonanza Printing & Copying Centre Inc.,
[23] B. Gutarowska, A. Michalski, Microbial degradation of woven fabrics and pro- Ottawa, ON, 1999.
tection against biodegradation, in: H.-Y. Jeon (Ed.), Woven fabrics, InTech, [27] K. Gherdán, T.G. Weiszburg, Zs. Bendő, F. Kristály, T. Váczi, N. Zajzon, 2014
Rijeka, Croatia, 2012, pp. 267–293, http://dx.doi.org/10.5772/2594. Phosphine fumigation damage: corrosion of metal and metal-textile compos-
[24] S. Polić-Radovanović, S. Ristić, B. Jegdić, Z. Nikolić, Methodological and techni- ite museum objects, in: Proceedings of the 11th international conference on
cal aspects application of new technology to protect cultural heritage, Institut non-destructive investigation and microanalysis for the diagnostics and con-
Gosa, Beograd, 2010 (In Serbian; https://www.books.google.rs/books?id= servation of cultural and environmental heritage, Museo Arquelogico Nacional,
vkGFAwAAQBAJ&pg=PP2&lpg=PP2&dq=Slavica+Risti%C4%87,+metodolo%C5 Madrid, June 11–13, 2014, http://dx.doi.org/10.13140/2.1.2948.5766.
%A1ki&source=bl&ots=2Vm9eOOnU3&sig=wMJF470LfSNjhUPEm7m812uznsA [28] J.M. Lee, K.G. Watkins, Laser removal of oxides and particles from copper sur-
&hl=sr-Latn&sa=X&ved=0ahUKEwifirHG1NbKAhUnlHIKHd3kDpUQ6AEIGDAA faces for microelectronic fabrication, Opt. Express Special Focus Issue 7 (2000)
#v=onepage&q=Slavica%20Risti%C4%87%2C%20metodolo%C5%A1ki&f=false). 69–76.

J Culher 2016 07 001

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

J Culher 2016 07 001

Uploaded by

Copyright:

Available Formats

G Model

CULHER-3106; No. of Pages 10 ARTICLE IN PRESS

Preliminary investigation on the use of the Q-switched Nd:YAG laser

The preliminary protection measures, mentioned above, did not Table 1

1 100 532 5 (100) Dry

Fig. 3. SEM of irradiated zones on trial sample, silvered copper plate.

Fig. 4. The 1–5 zones of cleaning shirt embroidery.

Fig. 5. Laser cleaning zones 6–8.

to mention, that the decomposition of copper phosphide is much

Fig. 7. SEM photo and EDX spectra for zones 2, 5, 6 and 7.

chemical methods, because the only waste generated in this

This research was ﬁnancially supported by the Ministry of Edu-

The experimentally determined parameters of a unit cell References

You might also like