senso s
A icle
Non-Des uc i e Analy ical In es iga ion o Deco a i e
Wallpape s Samples o he Nine een h Cen u y be o e
Thei Res o a ion
Ila ia Cos an ini 1, Kepa Cas o 1, Ma ia Dolo es Rod iguez-Laso 2, Juan Manuel Mada iaga 1,* and Go ka A ana 1
Ci a ion: Cos an ini, I.; Cas o, K.;
Rod iguez-Laso, M.D.; Mada iaga,
J.M.; A ana, G. Non-Des uc i e
Analy ical In es iga ion o Deco a i e
Wallpape s Samples o he
Nine een h Cen u y be o e Thei
Res o a ion. Senso s 2021,21, 4416.
h ps://doi.o g/10.3390/s21134416
Academic Edi o s: John K. Delaney
and Da id W. Messinge
Recei ed: 31 May 2021
Accep ed: 25 June 2021
Published: 28 June 2021
Publishe ’s No e: MDPI s ays neu al
wi h ega d o ju isdic ional claims in
published maps and ins i u ional a il-
ia ions.
Copy igh : © 2021 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
1Depa men o Analy ical Chemis y, Facul y o Science and Technology, Uni e si y o he Basque Coun y
UPV/EHU, P.O. Box 644, 48080 Bilbao, Basque Coun y, Spain; [email p o ec ed] (I.C.);
[email p o ec ed] (K.C.); [email p o ec ed] (G.A.)
2Depa men o Pain ing, Uni e si y o he Basque Coun y, P.O. Box 644, 48080 Bilbao, Spain;
ma iadolo es. [email p o ec ed]
*Co espondence: [email p o ec ed]
Abs ac :
In his wo k, deco a i e wallpape s (19 h cen u y) om an his o ical palace loca ed in
Oia zun (Basque Coun y, Spain) we e analyzed be o e hei es o a ion. Mic o-ene gy dispe si e
X- ay luo escence spec oscopy, Raman spec oscopy, and a enua ed o al e lec ance in a ed
spec oscopy we e used o in es iga e he elemen al and molecula composi ion o pigmen s, he
p esence o binde s, and he s a e o conse a ion o he pape suppo . The aim o he in es iga ion
was ying o unde s and he possible deg ada ion pa hways and iden i y he aw ma e ials in o de
o choose he bes es o a ion p o ocol acco ding o he o iginal aspec o wallpape s. As s a ed om
bo h he elemen al dis ibu ion and he iden i ica ion o mine al phases by Raman spec oscopy,
he mos used pigmen was lead ch oma e. I was mixed wi h o he pigmen s such as ul ama ine
blue, zinc ch oma e, hema i e, and a acami e among o he s o ob ain di e en shades and hey we e
applied mixed wi h an animal glue. B ass, iden i ied hanks o elemen al mic o-EDXRF maps, was
employed as a shiny deco a i e elemen . In addi ion, a pa ial deg ada ion o cellulose was de ec ed
due o i s na u al ageing, he acidic na u e o lignin, and o a phenomenon o humidi y o he walls.
P obably he deposi ion o black pa icula e ma e was he cause o he da kening o he pain ing
su aces.
Keywo ds:
deco a i e wallpape s; aman spec oscopy; mic o-EDXRF; ATR-FTIR; pigmen s;
cellulose
1. In oduc ion
A wo ks on pape such as manusc ip s, li hog aphs, wa e colo s, e c., a e one o he
mos sensi i e ca ego ies o a wo ks especially when hey a e exposed in open spaces, like
deco a i e wallpape s, and hey need special a en ion o be p ese ed. The impo ance o
he s udy o wallpape s has ecei ed mo e a en ion o e he pas decades as a esul o he
inc easing in e es o conse a ion o p i a e cul u al he i age p ese ed in he in e io o
his o ical houses and palaces. P i a e he i age has been con e ed in o public he i age and
many his o ic houses ha e been adap ed o museums o a galle ies.
The o igins o he manu ac u e o wallpape can be loca ed in he Eas om whe e
hey a i ed in Eu ope h ough he comme cial ou es. A he end o he 17 h cen u y,
Eu opean manu ac u e s we e al eady de eloping pape as a wallco e ing. G adually, i
became an essen ial deco a i e elemen in he homes o he bou geois, o which a g ea
in e es quickly a ose in making hese p oduc s a oiding impo om he Eas [1].
A he end o he 18 h cen u y, he manu ac u e o wallpape s eached a maximum
in p oduc ion. F ance, England, Ge many, Swi ze land, and Belgium we e some o he
leading coun ies in he c ea ion and p oduc ion o deco a i e wallpape s. In his pe iod,
some impo an ac o ies we e ounded in Eu ope, such as he one in Rixheim by Juan
Senso s 2021,21, 4416. h ps://doi.o g/10.3390/s21134416 h ps://www.mdpi.com/jou nal/senso s
Senso s 2021,21, 4416 2 o 12
Zube in 1790, o ha o he Joseph and Pie e Du ou b o he s in he Bu gundy egion
(F ance). Be o e 1830, wallpape s we e made by pain ing, s enciling, eng a ing, o mos
o en by p in ing wi h woodblocks in dis empe colo s on joined shee s o handmade
pape . A e his da e, hanks o he de elopmen o enginee ing echniques, he mode n
con inuous pape p oduc ion o wallpape s s a ed [2].
The echniques o execu ion, he deco a ions, and he ma e ials used in he Wes we e
di e en om he o ien al wallpape s [
3
]. Howe e , he use o oxic pigmen s o hei
ealiza ion had been la gely documen ed in all cases [
4
,
5
]. Fo his eason, he in e es in he
s udy o wallpape s has been also ocused on he impac ha hese ma e ials may ha e had
on he heal h in he pas and especially nowadays. P e ious esea ch unde lined he heal h
impac o conside able amoun o oxic p oduc s con aining lead (Pb), ch omium (C ),
a senic (As), and me cu y (Hg) widely used as pigmen s o in e io wall deco a ion [
6
–
8
].
Qui e amous is he case o he manu ac u e William Mo is, who used g een a senic
pigmen s [
9
]. Howe e , he use o oxic pain ings is no only some hing ha happened
in he pas . In ac , he esea ch o Mielke e al. [
2
] conce ning he con en o Hg and Pb
in coa ings o la ex pain , unde lines hei isk mainly o he child en who may inges
pain chips and consequen ly de elop lead poisoning. Mo eo e , in 1977, he Uni ed S a es
En i onmen al P o ec ion Agency (EPA) banned Pb-based pain s, and in 1991, he same
agency banned Hg in he o m o phenyl-me cu y ace a e, added as a p ese a i e in house
pain s. Thus, echnicians and es o e s should be awa e o he dange o he ma e ials wi h
which hey come in o con ac in o de o adop adequa e p o ec i e measu es.
As men ioned, in compa ison wi h o he ype o a wo ks, wallpape s a e easily
deg adable by in insic ac o s, due o he ma e ials and echniques used, o ex insic
agen s such as ligh , humidi y, mic o-o ganisms, e c. In ac , hei agili y is s ongly
a ec ed by he cha ac e is ics o he a chi ec u al wall and o he en i onmen [
10
]. In
addi ion, he cons an changes in people’s as e o which hey ha e been subjec ed o e
he yea s ha e acili a ed hei disappea ance. We can almos conside hem as epheme al
a wo ks depending on he ends o he momen [1,11].
Rega ding he pigmen s, hey can show deg ada ion phenomena (ch oma ic change
o example) caused by hei na u al aging o by he p esence o con aminan s in he
en i onmen . On he o he hand, he cellulose ma ix is he one ha unde goes he g ea es
deg ada ion due o eac ions ha can induce bo h chemical and physical changes. In
mos cases, he de e io a ion is i e e sible, hus, i is impo an o unde s and he causes
ha could ha e p o oked i . In pa icula , cellulose deg ada ion occu s h ough he acid
hyd olysis, oxida ion, c oss-linking, ing opening, chain scission, and a ia ion o he
amo phous/c ys alline a io [
12
], some imes induced by he p esence o some pigmen s
(coppe pigmen s) [13].
Due o he complexi y in he composi ion o pape a wo ks, some au ho s ha e
poin ed ou he use ulness o using many analy ical echniques o he s udy o aw ma e i-
als and o he in es iga ion o he conse a ion s a e o he suppo [
14
–
16
]. Fo his aim,
non-des uc i e po able sys ems we e de eloped and success ully employed du ing in
si u campaigns o elemen al (ene gy dispe si e X- ay luo escence spec oscopy, scanning
elec on mic oscopy—ene gy dispe si e X- ay spec oscopy, lase -induced b eakdown
spec oscopy) and molecula s udy (FTIR and Raman spec oscopy in di e en modes,
su ace-enhanced Raman spec oscopy) o cul u al he i age ma e ials [
17
]. In mos o he
cases, deco a i e wallpape s a e made by a supe imposi ion o a ious pain ing laye s.
Fo his eason, due o he pene a ion powe o he beam, he in e p e a ion o he esul s
is c i ical since he spec a eco ded, bo h wi h elemen al and wi h molecula echniques,
can gi e in o ma ion co esponding o many laye s. In his ci cums ance, no des uc i e
bench op ins umen allows o selec speci ic poin s and, in some cases, pe o m maps o
he dis ibu ion o elemen and compounds, o a be e cha ac e iza ion o ma e ials and
he s udy o he manu ac u ing echnique [13,18].
In he cu en s udy, as pa o he es o a ion wo ks, a se ies o deco a i e wallpa-
pe s belonging o he nine een h cen u y we e s udied by non-des uc i e elemen al and
Senso s 2021,21, 4416 3 o 12
molecula analyses. A i s , mic o-ene gy dispe si e X- ay luo escence spec oscopy
(mic o-EDXRF) was ca ied ou o assess he elemen al composi ion o wallpape samples.
Then, Raman spec oscopy and a enua ed o al e lec ance Fou ie ans o m in a ed
spec oscopy (ATR-FTIR) measu emen s we e pe o med wi h he aim o de e mine he
ma e ials used o he ealiza ion o wallpape s (pigmen s and binde media) and o assess
he s a e o conse a ion o pape suppo .
2. Ma e ials and Me hods
2.1. His o ical In o ma ion and Samples
The samples unde s udy we e aken du ing he es o a ion wo ks o A izmendi
Enea-U diñola Palace in Oia zun (Basque Coun y, Spain) ca ied ou o con e he
building in o a museum. The building, buil in he 18 h cen u y, has main ained i s ac ual
appea ance o mo e han h ee cen u ies and ep esen s an example o he li es yle o
he mo e powe ul classes o he p o ince o Gipuzkoa in he 19 h and 20 h cen u ies.
Du ing he Spanish ci il wa , he A izmendi Enea palace ell in o he hands o he F ancois
au ho i ies whe e hey es ablished hei headqua e s.
The wallpape s we e made o ellum pape o indus ial manu ac u e. As hey a e
made o con inuous pape , hey do no come be o e 1830. The Zube & Cie ac o y in
Rexheim (Alsace, F ance) specialized in he manu ac u e o wallpape s, and managed
o make he i s con inuous pape in i s pape mill in 1830, ob aining a pa en o i in
1832. Howe e , no ma ks o he p oduc ion company we e ound so hei o igin emains
unknown.
Deco a i e wallpape can ep esen di e en hemes depending o he pe iod and
p o enance [
19
]. In he pas , he design could show epe i i e schemes o lowe s and
medallions in undula ing a angemen s [
20
]. In o he cases, hey imi a ed expensi e
ma e ial such as pain ed wall deco a ion, ma ble, i o y, and mos o en ex iles wi h
imi a ions o ich ab ics such as moi és and el e s [
2
]. The wallpape s s udied in he
p esen wo k belong o he la e design wi h a se ial image. The execu ion echnique was
he glue empe a applied by means o xylog aphic s amping p ocedu es.
The analyzed samples we e ob ained om he deco a i e wallpape s a e being
emo ed om he palace and be o e being es o ed in he conse a ion and es o a ion lab-
o a o y o he Depa men o Pain ing o he Uni e si y o he Basque Coun y (UPV/EHU).
A o al o i e samples (wp1-wp2-wpP3-wp4-wp5) o di e en sizes we e analyzed ( he
la ges had a size o abou 7
×
4.5 cm). They showed he main shades used o deco a ion
including g een, pu ple, black, whi e, g ey, and b own colo s. The s a e o conse a ion
o wallpape was qui e good al hough some a eas, in co espondence wi h he ceiling,
p esen ed a ch oma ic discolo a ion o he pape suppo .
2.2. Ins umen a ion
2.2.1. Mic o-Ene gy Dispe si e X- ay Fluo escence Spec oscopy (mic o-EDXRF)
The elemen al maps we e acqui ed using a M4 TORNADO EDXRF spec ome e
(B uke Nano GmbH, Be lin, Ge many). The analyses we e pe o med unde acuum
(20 mba ) in o de o imp o e he iden i ica ion o he ligh e elemen s. The la e al eso-
lu ion used o spec al acquisi ions was 20 mic ome e s. The maps we e ob ained using
M-QUANT so wa e. To ob ain he quan i a i e maps, he assignmen o he elemen s and
he decon olu ion o he spec al in o ma ion we e ca ied ou . The maps we e ob ained
by conside ing he K-alpha line o each elemen .
2.2.2. Mic o-Raman Spec oscopy
Wallpape samples we e analyzed using a con ocal Renishaw InVia Raman spec om-
e e , coupled o a Leica DMLM mic oscope. The spec a we e acqui ed wi h he Leica
50×N
Plan (0.75 NA) lens wi h a 2
µ
m spa ial esolu ion and he Olympus 100
×
(0.8 NA).
The minimum heo e ical spo diame e using he 532 nm lase was, o he Leica 50
×
and Olympus 100
×
, 0.9
µ
m and 0.8
µ
m, espec i ely, while using he 785 nm lase , i was
Senso s 2021,21, 4416 4 o 12
1.7
µ
m and 1.1
µ
m, espec i ely. Addi ionally, o isualiza ion and ocusing, ano he Leica
5
×
N Plan (0.12 NA) and a 20
×
N Plan EPI (0.40 NA) lens we e used. Fo ocusing on and
sea ching o poin s o in e es , he mic oscope implemen s a mo o ized s age (XYZ). The
powe applied was se a he sou ce a a maximum o 50 mW while on he sample was
always less han 20 mW o a oid possible he modecomposi ion o he samples. No mally,
10–300 scans, each las ing 1–20 s, we e accumula ed o achie e a sui able signal- o-noise
a io a an ope a ing spec al esolu ion o ≤1 cm−1.
2.2.3. A enua ed To al Re lec ion In a ed (ATR-FTIR)
ATR-FTIR spec a we e collec ed by a Jasco 6300 spec opho ome e equipped o
s anda d ansmi ance and a enua ed o al e lec ance (PIKE Mi acleTH) measu emen s.
The equipmen is ea u ed by a Ge/KB beam spli e and a deu e a ed L-alanine doped
iglycene sul a e (DLATGS) de ec o wi h Pel ie empe a u e con ol. All spec a we e
acqui ed in he middle in a ed egion ( om 4000 o 400 cm
−1
), accumula ing 128 scans,
wi h a spec al esolu ion o 4 cm
−1
. Da a acquisi ion was ca ied ou by Jasco Spec a
Manage Sui package, whils da a ea men was pe o med by O igin 2018.
2.2.4. pH-Me e
Fo he analysis o he pH o he cellulose a pH-me e (mic opH 2000 C ison) was
used. The elec ode (52 07, C ison) o pH measu emen s on la su aces was employed.
I s main cha ac e is ic is ha i s diaph agm and he memb ane a e on he same plane.
3. Resul s and Discussion
3.1. Cha ac e iza ion o Ma e ials
The wallpape s samples we e analyzed a i s wi h mic o-EDXRF o ob ain he
elemen al dis ibu ion o he compounds employed o hei ealiza ion. Fo his pu pose,
mic o-EDXRF maps we e eco ded like shown in Figu e 1. Al hough a o al o i e samples
we e analyzed, he selec ed a eas shown in he elemen al dis ibu ion maps o
Figu e 1
we e ep esen a i e o he whole analyzed samples as hey included all colo s o he
wallpape s; g een, b own, pu ple, black, g ey, and whi e. The size o he sample was la ge
han ha shown in he images.
Senso s 2021, 21, x FOR PEER REVIEW 4 o 13
50× N Plan (0.75 NA) lens wi h a 2 μm spa ial esolu ion and he Olympus 100× (0.8 NA).
The minimum heo e ical spo diame e using he 532 nm lase was, o he Leica 50× and
Olympus 100×, 0.9 μm and 0.8 μm, espec i ely, while using he 785 nm lase , i was 1.7
μm and 1.1 μm, espec i ely. Addi ionally, o isualiza ion and ocusing, ano he Leica
5× N Plan (0.12 NA) and a 20× N Plan EPI (0.40 NA) lens we e used. Fo ocusing on and
sea ching o poin s o in e es , he mic oscope implemen s a mo o ized s age (XYZ). The
powe applied was se a he sou ce a a maximum o 50 mW while on he sample was
always less han 20 mW o a oid possible he modecomposi ion o he samples. No mally,
10–300 scans, each las ing 1–20 s, we e accumula ed o achie e a sui able signal- o-noise
a io a an ope a ing spec al esolu ion o ≤1 cm−1.
2.2.3. A enua ed To al Re lec ion In a ed (ATR-FTIR)
ATR-FTIR spec a we e collec ed by a Jasco 6300 spec opho ome e equipped o
s anda d ansmi ance and a enua ed o al e lec ance (PIKE Mi acleTH) measu emen s.
The equipmen is ea u ed by a Ge/KB beam spli e and a deu e a ed L-alanine doped
iglycene sul a e (DLATGS) de ec o wi h Pel ie empe a u e con ol. All spec a we e
acqui ed in he middle in a ed egion ( om 4000 o 400 cm−1), accumula ing 128 scans,
wi h a spec al esolu ion o 4 cm−1. Da a acquisi ion was ca ied ou by Jasco Spec a Man-
age Sui package, whils da a ea men was pe o med by O igin 2018.
2.2.4. pH-Me e
Fo he analysis o he pH o he cellulose a pH-me e (mic opH 2000 C ison) was
used. The elec ode (52 07, C ison) o pH measu emen s on la su aces was employed.
I s main cha ac e is ic is ha i s diaph agm and he memb ane a e on he same plane.
3. Resul s and Discussion
3.1. Cha ac e iza ion o Ma e ials
The wallpape s samples we e analyzed a i s wi h mic o-EDXRF o ob ain he ele-
men al dis ibu ion o he compounds employed o hei ealiza ion. Fo his pu pose, mi-
c o-EDXRF maps we e eco ded like shown in Figu e 1. Al hough a o al o i e samples
we e analyzed, he selec ed a eas shown in he elemen al dis ibu ion maps o Figu e 1 we e
ep esen a i e o he whole analyzed samples as hey included all colo s o he wallpa-
pe s; g een, b own, pu ple, black, g ey, and whi e. The size o he sample was la ge han
ha shown in he images.
Figu e 1. Mic o-EDXRF maps om one o he deco a i e wallpape samples (sample wp2).
Figu e 1. Mic o-EDXRF maps om one o he deco a i e wallpape samples (sample wp2).
The mos abundan elemen s iden i ied in he i s sample we e calcium (Ca) and
sul u (S) (Figu e 1) and hei dis ibu ion was isible e en whe e he pain laye was los .
Senso s 2021,21, 4416 5 o 12
In addi ion, Pb and C we e dis ibu ed ai ly e enly h oughou he sample. Aluminum
(Al) was p esen in well-de ined a eas whe e a pu ple pigmen was applied, whe eas silicon
was in black and iole a eas. Fo he g een-colo ed a eas, mic o-EDXRF analysis showed
he p esence o coppe (Cu), deno ing he use o a g een o blue coppe pigmen . In he
same g een a eas, zinc and chlo ine we e also e iden . On he o he hand, in b own and
black zones, i on (Fe) was ound.
A e mic o-EDXRF, Raman analyses we e ca ied ou o know he molecula composi-
ion o he aw ma e ials and o iden i y any deg ada ion p ocess hey may ha e unde gone
o e he yea s.
The use o calci e as well as gypsum was ecognized in many poin s o analysis e en
whe e he pain ed laye was no p esen . They we e applied as ille s o cellulose like
epo ed in o he in es iga ions [
14
]. In pa icula , calcium ca bona e was employed also
like he alkaline ese e o he pape suppo o neu alize he acidic na u e o some
componen o cellulose (lignin) and a oid i s oxida ion and acidic hyd olysis. These esul s
a e in conco dance wi h he dis ibu ion o Ca in he EDXRF analysis (see abo e).
Raman spec a o he black a eas showed he p esence o ca bon black pigmen (Raman
bands: 1350 and 1605 cm
−1
, Figu e 2a). I on oxides hema i e (Raman bands: 225, 244, 292,
410, 490, and 610 cm
−1
, Figu e 2b) and goe hi e (Raman bands: 167, 206, 244, 300, 398, 480,
and 550 cm
−1
, Figu e 2c) we e also de ec ed in black a eas mixed wi h ca bon black, as
obse ed also wi h he op ical mic oscope ( ed and yellow g ains mixed wi h he black
pigmen ). Thei p esence explains he e idence o i on ound by mic o-EDXRF in ha a ea.
Senso s 2021, 21, x FOR PEER REVIEW 5 o 13
The mos abundan elemen s iden i ied in he i s sample we e calcium (Ca) and
sul u (S) (Figu e 1) and hei dis ibu ion was isible e en whe e he pain laye was los .
In addi ion, Pb and C we e dis ibu ed ai ly e enly h oughou he sample. Aluminum
(Al) was p esen in well-de ined a eas whe e a pu ple pigmen was applied, whe eas sil-
icon was in black and iole a eas. Fo he g een-colo ed a eas, mic o-EDXRF analysis
showed he p esence o coppe (Cu), deno ing he use o a g een o blue coppe pigmen .
In he same g een a eas, zinc and chlo ine we e also e iden . On he o he hand, in b own
and black zones, i on (Fe) was ound.
A e mic o-EDXRF, Raman analyses we e ca ied ou o know he molecula com-
posi ion o he aw ma e ials and o iden i y any deg ada ion p ocess hey may ha e un-
de gone o e he yea s.
The use o calci e as well as gypsum was ecognized in many poin s o analysis e en
whe e he pain ed laye was no p esen . They we e applied as ille s o cellulose like e-
po ed in o he in es iga ions [14]. In pa icula , calcium ca bona e was employed also
like he alkaline ese e o he pape suppo o neu alize he acidic na u e o some com-
ponen o cellulose (lignin) and a oid i s oxida ion and acidic hyd olysis. These esul s
a e in conco dance wi h he dis ibu ion o Ca in he EDXRF analysis (see abo e).
Raman spec a o he black a eas showed he p esence o ca bon black pigmen (Ra-
man bands: 1350 and 1605 cm−1, Figu e 2a). I on oxides hema i e (Raman bands: 225, 244,
292, 410, 490, and 610 cm−1, Figu e 2b) and goe hi e (Raman bands: 167, 206, 244, 300, 398,
480, and 550 cm−1, Figu e 2c) we e also de ec ed in black a eas mixed wi h ca bon black,
as obse ed also wi h he op ical mic oscope ( ed and yellow g ains mixed wi h he black
pigmen ). Thei p esence explains he e idence o i on ound by mic o-EDXRF in ha
a ea.
Figu e 2. Raman spec a o (a) ca bon black (532 nm exci a ion lase ), (b) hema i e oge he wi h
he Raman bands o lead ch oma e (LC), ul ama ine blue (UB), gypsum, and calcium ca bona e,
and (c) goe hi e (785 nm exci a ion lase ) om a black a ea (sample wp2).
Acco ding o Raman spec oscopy, he pu ple shade was ob ained by a mix u e o
ul ama ine blue (Al6Na8O24S3Si6, Raman bands a 258, 545, 584, 806, 1092, 1124, 1290, and
1642 cm−1, (Figu e 3a) and he syn he ic o ganic pigmen aliza in lake (PR 83:1) (Raman
bands a 484, 655, 901, 1190, 1292, 1325, 1354, and 1476 cm−1, (Figu e 3b) [21,22]. The unex-
pec ed p esence o silicon (Si) de ec ed in he black a eas by mic o-EDXRF would be ex-
plained by he use o ul ama ine blue, since a pu ple laye was ound below in ha a ea.
Figu e 2.
Raman spec a o (
a
) ca bon black (532 nm exci a ion lase ), (
b
) hema i e oge he wi h he
Raman bands o lead ch oma e (LC), ul ama ine blue (UB), gypsum, and calcium ca bona e, and (
c
)
goe hi e (785 nm exci a ion lase ) om a black a ea (sample wp2).
Acco ding o Raman spec oscopy, he pu ple shade was ob ained by a mix u e o
ul ama ine blue (Al
6
Na
8
O
24
S
3
Si
6
, Raman bands a 258, 545, 584, 806, 1092, 1124, 1290, and
1642 cm
−1
, (Figu e 3a) and he syn he ic o ganic pigmen aliza in lake (PR 83:1) (Raman
bands a 484, 655, 901, 1190, 1292, 1325, 1354, and 1476 cm
−1
, (Figu e 3b) [
21
,
22
]. The
unexpec ed p esence o silicon (Si) de ec ed in he black a eas by mic o-EDXRF would be
explained by he use o ul ama ine blue, since a pu ple laye was ound below in ha a ea.
He e, we ha e an example o he p oblem explained abo e, how he supe imposi ion o
se e al laye s can make di icul he in e p e a ion o EDXRF spec a in his case. Howe e ,
om he compa ison o EDXRF maps o Al and Si, i was also e iden ha a silicon-based
Senso s 2021,21, 4416 6 o 12
pigmen was used in black a eas in mix u e wi h ca bon black and i on oxides bu i could
no be iden i ied by Raman spec oscopy.
Senso s 2021, 21, x FOR PEER REVIEW 6 o 13
He e, we ha e an example o he p oblem explained abo e, how he supe imposi ion o
se e al laye s can make di icul he in e p e a ion o EDXRF spec a in his case. How-
e e , om he compa ison o EDXRF maps o Al and Si, i was also e iden ha a silicon-
based pigmen was used in black a eas in mix u e wi h ca bon black and i on oxides bu
i could no be iden i ied by Raman spec oscopy.
Figu e 3. Raman spec a o (a) ul ama ine blue, (b) aliza in ed (532 nm exci a ion lase ), and (c)
lead ch oma e (785 nm exci a ion lase ) om a pu ple a ea (sample wp2).
In addi ion, in mos poin s analyzed and acco ding o he dis ibu ion o lead and
ch ome (by EDXRF), he yellow pigmen lead ch oma e (PbC O4, Raman bands a 136,
326, 338, 358, 377, 400, and 840 cm−1, Figu e 3c) was iden i ied by Raman spec oscopy.
This pigmen was employed o ob ain di e en colo s in mix u es wi h o he pigmen s.
Fo example, in g een a eas, i was mixed wi h ul ama ine blue.
Addi ionally, he yellow pigmen zinc ch oma e (ZnC O4 Raman bands: 112, 141, 343,
358, 410, 774, 872, 893, and 941 cm−1, Figu e 4a) was iden i ied in he a eas whe e zinc
appea ed, mixed wi h lead ch oma e and ul ama ine blue in o de o ob ain he g een
colo . On he o he hand, in he same a eas, in mino amoun , wo g een pigmen s, he
coppe sul a e b ochan i e (Raman bands: 156, 178, 196, 242, 319, 390, 421, 446, 482, 506,
596, 610, 620, and 972 cm−1, Figu e 4b) and he basic coppe chlo ide a acami e
(Cu2(OH)3Cl, Raman bands: 122, 150, 509, 816, 910, and 970 cm−1, Figu e 4c) we e iden i-
ied, in conco dance wi h he dis ibu ion o coppe and chlo ine in he EDXRF maps.
P e ious esea ch men ioned a acami e, alone o oge he wi h b ochan i e, as a deg ada-
ion p oduc o coppe pigmen s such as malachi e o azu i e, al hough i s use as pigmen
has been also demons a ed [23,24]. In ou case, no Raman spec a o malachi e was ec-
o ded, hus, he use o a acami e mixed on pu pose wi h b ochan i e was assumed. In he
g een a eas, he p esence o gypsum was also ound in many poin s as well as ca bon
black.
Figu e 3.
Raman spec a o (
a
) ul ama ine blue, (
b
) aliza in ed (532 nm exci a ion lase ), and (
c
) lead
ch oma e (785 nm exci a ion lase ) om a pu ple a ea (sample wp2).
In addi ion, in mos poin s analyzed and acco ding o he dis ibu ion o lead and
ch ome (by EDXRF), he yellow pigmen lead ch oma e (PbC O
4
, Raman bands a 136, 326,
338, 358, 377, 400, and 840 cm
−1
, Figu e 3c) was iden i ied by Raman spec oscopy. This
pigmen was employed o ob ain di e en colo s in mix u es wi h o he pigmen s. Fo
example, in g een a eas, i was mixed wi h ul ama ine blue.
Addi ionally, he yellow pigmen zinc ch oma e (ZnC O
4
Raman bands: 112, 141, 343,
358, 410, 774, 872, 893, and 941 cm
−1
, Figu e 4a) was iden i ied in he a eas whe e zinc
appea ed, mixed wi h lead ch oma e and ul ama ine blue in o de o ob ain he g een colo .
On he o he hand, in he same a eas, in mino amoun , wo g een pigmen s, he coppe
sul a e b ochan i e (Raman bands: 156, 178, 196, 242, 319, 390, 421, 446, 482, 506, 596, 610,
620, and 972 cm
−1
, Figu e 4b) and he basic coppe chlo ide a acami e (Cu
2
(OH)
3
Cl, Raman
bands: 122, 150, 509, 816, 910, and 970 cm
−1
, Figu e 4c) we e iden i ied, in conco dance wi h
he dis ibu ion o coppe and chlo ine in he EDXRF maps. P e ious esea ch men ioned
a acami e, alone o oge he wi h b ochan i e, as a deg ada ion p oduc o coppe pigmen s
such as malachi e o azu i e, al hough i s use as pigmen has been also demons a ed [
23
,
24
].
In ou case, no Raman spec a o malachi e was eco ded, hus, he use o a acami e mixed
on pu pose wi h b ochan i e was assumed. In he g een a eas, he p esence o gypsum was
also ound in many poin s as well as ca bon black.
In he da k b own lines, whe e i on was de ec ed by EDXRF, he p esence o a mix u e
o lead ch oma e and hema i e was ound. In addi ion, gypsum as well as calci e we e used
on pu pose as pigmen s, due o he dis ibu ion o calcium and sul u in well-de ined a eas
in he EDXRF maps.
The second analyzed sample was cha ac e ized by a ci cula b igh deco a ion applied
o e a whi e and g ey backg ound wi h some black a eas ha simula e a ex ile ma e ial,
like shown in Figu e 5. EDXRF analyses showed ha also in his case he mos abundan
elemen in he sample was calcium, dis ibu ed in a homogeneous way h oughou he
a ea.
Senso s 2021,21, 4416 7 o 12
Senso s 2021, 21, x FOR PEER REVIEW 7 o 13
Figu e 4. Raman spec a o (a) lead ch oma e (LC) in mix u e wi h zinc ch oma e (ZC), (b) b o-
chan i e, and (c) a acami e (785 nm exci a ion lase ) om a g een a ea (sample wp2).
In he da k b own lines, whe e i on was de ec ed by EDXRF, he p esence o a mix-
u e o lead ch oma e and hema i e was ound. In addi ion, gypsum as well as calci e we e
used on pu pose as pigmen s, due o he dis ibu ion o calcium and sul u in well-de ined
a eas in he EDXRF maps.
The second analyzed sample was cha ac e ized by a ci cula b igh deco a ion ap-
plied o e a whi e and g ey backg ound wi h some black a eas ha simula e a ex ile ma-
e ial, like shown in Figu e 5. EDXRF analyses showed ha also in his case he mos abun-
dan elemen in he sample was calcium, dis ibu ed in a homogeneous way h oughou
he a ea.
In addi ion, in his sample, he laye ed p ocess o manu ac u ing he wallpape was
no iced hanks o he use o mic o-EDXRF maps. Indeed, he me allic deco a ion, which
was no e iden in he en i e sample by he naked eye, was clea ly isible in he mic o-
EDXRF maps, and i was composed by coppe (Cu) and zinc (Zn). As men ioned, wallpa-
pe manu ac u e consis ed in he supe posi ion o a ious colo ed laye s using blocks o
wood on which he d awing was eng a ed. In ou sample, a i s , a whi e laye was ap-
plied abo e which he Zn and Cu deco a ion was sp ead. Subsequen ly, he g ey laye
and las ly, he black one, we e applied acco ding o he desi ed deco a ion.
E en in hese samples, acco ding o he dis ibu ion o ch ome and C , lead ch oma e
was de ec ed in many poin s ega dless o he colo . In he g ey a eas, i was employed
oge he wi h ul ama ine blue, calci e, and ca bon black. On he o he hand, in whi e
a eas, calcium ca bona e (CaCO3, Raman bands: 155, 282, 712, and 1086 cm−1, Figu e S2b)
was mainly iden i ied in mix u e wi h gypsum (CaSO4·2H2O, Raman bands, 414, 494, 620,
670, 1008, 1135 cm−1, Figu e S2a) and g ains o lead ch oma e. In addi ion, he whi e pig-
men ba ium sul a e (BaSO4, Raman bands: 453, 460, 615, 645, 987, 1138, and 1164 cm−1,
Figu e S2c) was iden i ied. In he black a eas, whe e he p esence o i on and silicon we e
ound by mic o-EDXRF, he ca bon black was mixed wi h ul ama ine blue and he yellow
pigmen s wi h lead ch oma e and goe hi e.
Figu e 4.
Raman spec a o (
a
) lead ch oma e (LC) in mix u e wi h zinc ch oma e (ZC), (
b
) b ochan i e,
and (c) a acami e (785 nm exci a ion lase ) om a g een a ea (sample wp2).
Senso s 2021, 21, x FOR PEER REVIEW 8 o 13
Figu e 5. Mic o-EDXRF maps om a deco a i e wallpape (sample wp4). A e he sys ema ic applica ion o b ass all o e
he pape , o he pigmen ed laye s we e applied.
As al eady men ioned, some analyzed samples we e cha ac e ized by he p esence
o a golden deco a ion in addi ion o black, whi e, and g ey colo s. The elemen al compo-
si ion ob ained in a b igh ci cula a ea and in a g ey a ea is shown in Figu e S3. Analyses
ca ied ou in di e en poin s showed he di e ence in he concen a ion o coppe and
zinc in he me allic ci cles. The high concen a ion o hese elemen s sugges ed he p es-
ence o b ass like deco a i e elemen due o i s b igh ness and used as a subs i u e o
gold. This deco a i e compound has also been iden i ied in o he s udies ca ied ou on
wallpape belonging o he nine een h cen u y [25]. Acco ding o he elemen al maps, a
bina y b ass was used since, in addi ion o Cu and Zn, no o he me al, such as aluminum,
manganese, o silicon was de ec ed. Howe e , in he EDXRF spec a eco ded in he ci -
cula a eas (Figu e S3a), o he elemen s (calcium, i on, lead, silicon, sul u , ch ome) we e
eco ded, due o he pene a ion capabili y o he EDXRF echnique, and coming om he
pape suppo and pain ed laye s. They we e mo e e iden in he spec a collec ed in a
g ey a ea (Figu e S3b). A semi-quan i a i e analysis o he b ass was ca ied ou wi hou
conside ing he elemen s om unde lying laye s. The pe cen age o Zn and Cu was 29.4%
and 70.6%, espec i ely, which was compa ible wi h he composi ional ange o α-b ass
ha has a zinc con en o less han 35%. (α-β b asses ha e be ween 35% and 46.6% o Zn
while β-b asses be ween 46.6% and 50.6% o Zn) [26]. The addi ion o zinc in luences he
co osion esis ance, which dec eases i a majo amoun is added. As a esul , a low b ass
zinc con en will be o en mo e esis an han b ass wi h highe zinc g ade [27]. The alpha
b asses a e duc ile, much mo e when hey a e cold-wo ked and annealed a he han ho -
wo ked because, i ho -wo ked, impu i ies end o seg ega e a he g ain bounda ies and
make he b ass e y weak [26].
In addi ion, o ca y ou a comple e cha ac e iza ion o he ma e ials, ATR-FTIR anal-
yses we e also pe o med. In all, ATR-FTIR spec a eco ded on he back side o he pape
suppo (Figu e S4a) he p esence o he peaks belonging o calcium ca bona e, ib a ion
modes we e isible a 713 (ν4 CO3, symme ic), 873 (ν2 CO3, asymme ic), 1420 (ν3 CO3,
asymme ic), and 1800 cm−1 (ν1 + ν4 CO3, symme ic) [28]. In addi ion, some bands we e
compa ible wi h he undamen al s e ching and bending ib a ions obse ed in he in-
a ed egion o SO42− (ν3, 1115 and 1143 cm−1) and o H2O (ν2 1620 cm−1, ν1 3400, ν3 3540
cm−1) o gypsum [29]. Thus, hese esul s we e in acco dance wi h EDXRF maps con i m-
ing he use o calci e as alkaline cha ge o pape suppo and o he gypsum as ille o he
cellulose.
Figu e 5.
Mic o-EDXRF maps om a deco a i e wallpape (sample wp4). A e he sys ema ic applica ion o b ass all o e
he pape , o he pigmen ed laye s we e applied.
In addi ion, in his sample, he laye ed p ocess o manu ac u ing he wallpape was
no iced hanks o he use o mic o-EDXRF maps. Indeed, he me allic deco a ion, which
was no e iden in he en i e sample by he naked eye, was clea ly isible in he mic o-
EDXRF maps, and i was composed by coppe (Cu) and zinc (Zn). As men ioned, wallpape
manu ac u e consis ed in he supe posi ion o a ious colo ed laye s using blocks o wood
on which he d awing was eng a ed. In ou sample, a i s , a whi e laye was applied
abo e which he Zn and Cu deco a ion was sp ead. Subsequen ly, he g ey laye and las ly,
he black one, we e applied acco ding o he desi ed deco a ion.
E en in hese samples, acco ding o he dis ibu ion o ch ome and C , lead ch oma e
was de ec ed in many poin s ega dless o he colo . In he g ey a eas, i was employed
Senso s 2021,21, 4416 8 o 12
oge he wi h ul ama ine blue, calci e, and ca bon black. On he o he hand, in whi e
a eas, calcium ca bona e (CaCO
3
, Raman bands: 155, 282, 712, and 1086 cm
−1
, Figu e S2b)
was mainly iden i ied in mix u e wi h gypsum (CaSO
4·
2H
2
O, Raman bands, 414, 494, 620,
670, 1008, 1135 cm
−1
, Figu e S2a) and g ains o lead ch oma e. In addi ion, he whi e
pigmen ba ium sul a e (BaSO
4
, Raman bands: 453, 460, 615, 645, 987, 1138, and 1164 cm
−1
,
Figu e S2c) was iden i ied. In he black a eas, whe e he p esence o i on and silicon we e
ound by mic o-EDXRF, he ca bon black was mixed wi h ul ama ine blue and he yellow
pigmen s wi h lead ch oma e and goe hi e.
As al eady men ioned, some analyzed samples we e cha ac e ized by he p esence o
a golden deco a ion in addi ion o black, whi e, and g ey colo s. The elemen al composi ion
ob ained in a b igh ci cula a ea and in a g ey a ea is shown in Figu e S3. Analyses ca ied
ou in di e en poin s showed he di e ence in he concen a ion o coppe and zinc in
he me allic ci cles. The high concen a ion o hese elemen s sugges ed he p esence o
b ass like deco a i e elemen due o i s b igh ness and used as a subs i u e o gold. This
deco a i e compound has also been iden i ied in o he s udies ca ied ou on wallpape
belonging o he nine een h cen u y [
25
]. Acco ding o he elemen al maps, a bina y b ass
was used since, in addi ion o Cu and Zn, no o he me al, such as aluminum, manganese,
o silicon was de ec ed. Howe e , in he EDXRF spec a eco ded in he ci cula a eas
(Figu e S3a), o he elemen s (calcium, i on, lead, silicon, sul u , ch ome) we e eco ded,
due o he pene a ion capabili y o he EDXRF echnique, and coming om he pape
suppo and pain ed laye s. They we e mo e e iden in he spec a collec ed in a g ey a ea
(Figu e S3b). A semi-quan i a i e analysis o he b ass was ca ied ou wi hou conside ing
he elemen s om unde lying laye s. The pe cen age o Zn and Cu was 29.4% and 70.6%,
espec i ely, which was compa ible wi h he composi ional ange o
α
-b ass ha has a
zinc con en o less han 35%. (
α
-
β
b asses ha e be ween 35% and 46.6% o Zn while
β
-
b asses be ween 46.6% and 50.6% o Zn) [
26
]. The addi ion o zinc in luences he co osion
esis ance, which dec eases i a majo amoun is added. As a esul , a low b ass zinc con en
will be o en mo e esis an han b ass wi h highe zinc g ade [
27
]. The alpha b asses a e
duc ile, much mo e when hey a e cold-wo ked and annealed a he han ho -wo ked
because, i ho -wo ked, impu i ies end o seg ega e a he g ain bounda ies and make he
b ass e y weak [26].
In addi ion, o ca y ou a comple e cha ac e iza ion o he ma e ials, ATR-FTIR
analyses we e also pe o med. In all, ATR-FTIR spec a eco ded on he back side o
he pape suppo (Figu e S4a) he p esence o he peaks belonging o calcium ca bona e,
ib a ion modes we e isible a 713 (
ν4
CO
3
, symme ic), 873 (
ν2
CO
3
, asymme ic), 1420
(
ν3
CO
3
, asymme ic), and 1800 cm
−1
(
ν1
+
ν4
CO
3
, symme ic) [
28
]. In addi ion, some
bands we e compa ible wi h he undamen al s e ching and bending ib a ions obse ed
in he in a ed egion o SO
42−
(
ν3
, 1115 and 1143 cm
−1
) and o H
2
O (
ν2
1620 cm
−1
,
ν1
3400,
ν3
3540 cm
−1
) o gypsum [
29
]. Thus, hese esul s we e in acco dance wi h EDXRF
maps con i ming he use o calci e as alkaline cha ge o pape suppo and o he gypsum
as ille o he cellulose.
In addi ion, hanks o he FTIR analyses, he p esence o an o ganic compound
was ound. Indeed, he band a 1642 cm
−1
co esponds o a undamen al s e ching
o he amide ca bonyl g oup (–CO–NH–, 1655 cm
−1
) and he de o ma ion o amide II
(–NH
2
, 1632 cm
−1
) [
30
]. In he FTIR spec um me hylene g oup, abso bencies a 2853
and 2923 cm
−1
we e also e iden , and he b oad band cen e ed a abou 3300 cm
−1
(N-H
s e ching), indica ing hyd oxyl and amide g oups pa ially masked by he b oad abso p-
ion band o he OH g oup [
31
]. The e idence in he ATR-FTIR spec um was compa ible
wi h he p esence o animal glue. In addi ion, he FTIR bands belonging o he animal glue
we e e iden in he spec um eco ded in a pu ple a ea (Figu e S4b) showing he use o an
o ganic binde o he applica ion o pigmen s in he wallpape s.
Senso s 2021,21, 4416 9 o 12
3.2. S a e o Conse a ion o he Wallpape s
The analy ical echniques we e applied o assess he s a e o conse a ion o pain
laye as well as o he pape suppo .
Rega ding he s a e o conse a ion o pigmen s, a ch oma ic al e a ion o he pic o ial
laye , in he o m o da kening, was isible. No deg ada ion p oduc s o he o iginal
pigmen s we e iden i ied as esponsible o he ch oma ic change by Raman o FTIR
spec oscopy. The p esence o ca bon black, ound in many poin s o he sample, seemed
o be he main ac o o discolo a ion o he wallpape and i could be conside ed a
con amina ion due o he p esence o black pa icula e ma e s. Howe e , due o he
endency o lead ch oma e o da ken [
32
,
33
], caused by a educ ion o he ch oma e ions o
C (III)-compounds, i s in ol emen in he deg ada ion p ocess canno be excluded despi e
he deg ada ion p oduc (C
2
O
3·
2H
2
O) was ne e eco ded by Raman spec oscopy in he
s udied wallpape . Indeed, in he a eas o b own colo , a g ea e quan i y o lead ch oma e
was iden i ied and a he mic oscopical le el, hey appea ed da k yellow. The e o e, i is
possible ha hose a eas we e yellow o iginally wi h a di e en shade depending on he
pigmen (hema i e, goe hi e, gypsum, e c.) mixed wi h lead ch oma e.
Mic o-EDXRF analyses we e also pe o med in some a eas, in he back side o he sup-
po , whe e he pape showed an e iden ch oma ic change due p obably o an oxida i e
p ocess. The elemen al maps sugges he p esence o sul a e compounds in co espondence
wi h he da ke a eas as shown by he o e lapping o he sul u , calcium, and ba ium
maps (Figu e S5). Acco ding o he Raman analyses pe o med in hese a eas, he main
compounds iden i ied we e gypsum and ba ium sul a e. The e idence o ba ium sul a e
and gypsum could be ela ed wi h he p esence o a pain laye sp ead on he walls be o e
he applica ion o he wallpape s.
In he Raman spec a eco ded in he same a eas (Figu e 6), he Raman bands o
cellulose, he p incipal componen o pape , we e isible. Cellulose is composed by bo h
c ys alline and amo phous phases. The mos abundan na i e c ys alline o m is cellulose
I ha exis s as a mix u e o polymo phs I
a
and I
b
[
34
]. In addi ion o cellulose, he
pape ma ix is composed in a mino amoun by hemicellulose and lignin. Hemicellulose
and he amo phous componen o he cellulose ac as a linke be ween he c ys alline
cellulose ib es [
12
]. Lignin is a he e ogeneous polyme ha occu s in woody and ascula
issues. I o ms a ma ix su ounding he cellulose in woody cell walls, which p o ec s he
hemicellulose and cellulose ha oge he o m he holocellulose [35].
Senso s 2021, 21, x FOR PEER REVIEW 10 o 13
Figu e 6. Raman spec um o pape suppo whe e bands o cellulose (C) and lignin a e shown (L)
(sample wp5).
In Figu e 6, wo bands belonging o cellulose we e isible loca ed a 1095 cm−1 (C–O–
C o β-glycosidic linkage o cellulose and hemicellulose) and a 1120 cm−1 (C–O–C o he
α-glycosidic linkage o hemicellulose). Gene ally, he aging p ocesses o cellulose a o
he pa ial b eaking o he α-glycosidic and β-glycosidic. In pa icula , as demons a ed
by p e ious esea ch, a dec ease in he in ensi y o he band co esponding o he α-gly-
cosidic bond, compa ed o he β -glycosidic bond can occu , because i is a weake link,
he e o e, he hemicellulose seems o be mo e sensi i e o deg ada ion p ocesses [36].
In addi ion o he cellulose Raman bands, hose belonging o lignin we e e iden ,
al hough e y weak, wi h he mos in ense one a 1597 cm−1 co esponding o he symme -
ic a oma ic ing s e ch. All he Raman bands o lignin (1272, 1334, 1453, 1597, and 1660
cm−1) p esen in he Raman spec um eco ded in he samples ha e been epo ed in he
s udy o Aga wal e al. [37] o he spec um o sp uce milled-wood lignin (MWL). The
p esence o lignin in he pape ma ix could be one o he causes o he cellulose oxida ion,
due o i s acidic na u e, which was isible by he naked eye like a yellowing (Figu e S5).
Fo his eason, nowadays, lignin is emo ed om cellulosic pulp be o e he manu ac u -
ing p ocess o pape . In addi ion, he absence o some Raman bands a 2893 and 2936 cm−1
could be a ibu ed o he aging o cellulose as epo ed in ano he in es iga ion [38]. In
o de o e i y he pH o he pape suppo , some analyses wi h a pH-me e we e ca ied
ou in he yellowed a eas. Ten poin s we e conside ed, and he a e age o he alues ob-
ained was be ween 4.23 and 5.57, in he mos yellowed and in he bes p ese ed a eas,
espec i ely, con i ming an acid pH o he cellulose suppo . This ac sugges ed an im-
po an deg ada ion p ocess o cellulose unde ling he loss o e ec i eness o he alkaline
ese e.
Figu e 6.
Raman spec um o pape suppo whe e bands o cellulose (C) and lignin a e shown (L)
(sample wp5).
