ChapterPDF Available

A new imaging method of fluorescence induced by multispectral UV for studying historical musical instruments coatings


Abstract and Figures

In art conservation, UV luminescence is a widespread technique: traditionally, longwave UV-A lamps (365 nm-emission peaks) are used to highlight the distribution of varnishes or retouches, and some pigments such as the red lakes. Recently, a new method of luminescence has been tested, induced by a multispectral UV-Vis radiation source (11 narrow bands in the 295-405 nm range), proving that luminescence images obtained using multiple excitation bands provide a good characterization of the luminescent properties of the material. The application to historic musical instruments, where non-invasive approaches are strongly encouraged, can be surely promising: for example, for bowed instruments the detection of specific coloring materials such as madder lake can be decisive for identifying specific finishing methods adopted in violin making by the great Masters of the past. In this work, we applied the multispectral UV luminescence method to mock-ups mimicking historical coating materials for bowed instruments, focusing on madder lake identification before and after artificial aging.
Content may be subject to copyright.
UV-Vis Luminescence imaging techniques
Técnicas de imagen de luminiscencia UV-Vis
Marcello Picollo
Maartje Stols-Witlox
Laura Fuster-López
Técnicas de imagen de
Imaging techniques
No. 1 | 2019
Editor in Chief/Editor jefe
Laura FUSTER-LÓPEZ, Universitat Politècnica de València (Spain)
Assistant Editor/Asistente editorial
Ana Mª GARCÍA-CASTILLO Universitat Politècnica de València (Spain)
Scientic Editors/Editores cientícos
Marcello PICOLLO “Nello Carrara” Institute of Applied Physics of the National Research Council (Italy)
Maartje STOLS-WITLOX University of Amsterdam (The Netherlands)
Laura FUSTER-LÓPEZ Universitat Politècnica de València (Spain)
Editorial Board/Equipo editorial
Vasilike ARGYROPOULOS University of West Attica (Greece)
Néstor BARRIO Tarea. Instituto de Investigaciones sobre el Patrimonio Cultural. Universidad Nacional de San Martín. (Argentina)
Anne Laurence DUPONT Centre de Recherche sur la Conservation. Centre National de la Recherche Scientique. Muséum National
d’Histoire Naturelle, Ministère de la Culture (France)
Davide GASPAROTTO Paintings Department, J. Paul Getty Museum (USA)
Marcello PICOLLO “Nello Carrara” Institute of Applied Physics of the National Research Council (Italy)
Benjamin ROUS Netherlands Institute for Conservation+Art+Science+ (The Netherlands)
Maartje STOLS-WITLOX University of Amsterdam (The Netherlands)
Anna VILA Fundación La Caixa (Spain)
Andrea CASINI “Nello Carrara” Institute of Applied Physics of the National Research Council (Italy)
Julio M. DEL HOYO-MELÉNDEZ The National Museum Krakow (Poland)
Joanne DYER The British Museum (UK)
Marina GINANNI Opicio delle Pietre Dure (Italy)
Miquel HERRERO-CORTELL Universitat de Lleida (Spain)
Tuulikki KILPINEN Conservator in private practice (Finland)
Yosi POZEILOV Los Angeles County Museum of Art (USA)
Roxanne RADPOUR University of California Los Angeles (USA)
Alessandro SIDOTI Biblioteca Nazionale Centrale di Firenze (Italy)
Marie SVOBODA J. Paul Getty Museum (USA) ·
Masahiko TSUKADA Tokyo University of the Arts (Japan)
John TWILLEY Consulting scientist (USA)
Rita L. AMOR-GARCÍA Independent conservator-researcher. Cons-Graf (UK)
Ruth DEL FRESNO-GUILLEM Independent contemporary art consultant, researcher and conservator (Canada)
Annette S. ORTIZ-MIRANDA Postdoctoral Researcher. Northwestern University / Art Institute of Chicago.
Center for Scientic Studies in the Arts (USA)
No. 1 | 2019
(Eds. Marcello Picollo, Maartje Stols-Witlox and Laura Fuster-López)
No. 1
UV-Vis Luminescence Imaging Techniques
Técnicas de imagen de luminiscencia UV-Vis
Marcelo Picollo
Maartje Stols-Witlox
Laura Fuster-López
Editorial Universitat Politècnica de València, 2019
Ref.: 6583_02_01_01
ISBN: 978-84-9048-828-7 (Print on demand)
Legal Deposit: V-3649-2019
Layout design
Enrique Mateo, Triskelion Diseño Editorial
Byprint Percom SL
Printed in Spain
UV-Vis Luminescence Imaging Techniques / Técnicas de imagen de luminiscencia UV-Vis
This book is licensed under a Creative Commons Atribution-NonCommercial-NonDetivates-4.0
International Licensed
Image cover: Fragment of a multispectral UV luminescence image of the
mock-ups series P6 exposed to 11 excitation bands at different levels of aging.
Source: Cavalerietal.
Imagen de portada: Fragmento una imagen de luminiscencia multiespectral UV de la
serie de probetas P6 expuestas a 11bandas de excitación a diferentes niveles de enve-
jecimiento. Fuente: Cavaleri et al.
7 Introduction to the Series
Presentación de la colección
Laura Fuster-López
11 Introduction to the Volume
Introducción al volumen
Marcello Picollo, Maartje Stols-Witlox, Laura Fuster-López
19 Foreword. Some Observations about Images
Prólogo. Algunas observaciones acerca de las imágenes
Néstor Barrio
29 Hints on the Luminescence Phenomena Theory
Aspectos clave en la teoría de los fenómenos luminiscentes
Mauro Bacci
35 UV-Induced Visible Luminescence for ConservationDocumentation
Luminiscencia visible inducida por UV para ladocumentación en conservación
E. Keats Webb
61 Broad-Band, Photo-Induced, Steady-State Luminescence Imaging in Practice
Imágenes de luminiscencia de banda ancha, foto-inducidas y estado estable en la práctica
Giovanni Verri
103 Photo-Luminescence Imaging Spectroscopy forPolychrome Objects
Espectroscopía de imagen fotoluminiscente para objetos policromados
John K. Delaney
119 Examination of Luminescence of Cross Sections (Fluorescence Microscopy)
Examen de la luminiscencia de secciones transversales (Microscopía de fluorescencia)
Barbara H. Berrie, Mathieu Thoury
UV-Vis Luminescence Imaging Techniques |
Técnicas de imagen de luminiscencia UV-Vis
141 The Application of Multi-band Imaging Integrated with Non-Invasive Spot Analyses for the
Examination of Archaeological Stone Artefacts
Aplicación de técnicas de imagen multi-banda integradas con el análisis no invasivo de puntos para el exa-
men de objetos arqueológicos de piedra
Susanna Bracci, Roberta Iannaccone, Donata Magrini
161 A New Imaging Method of Luminescence Induced by Multispectral Uv for the Study of Coatings on
Historical Musical Instruments
Un nuevo método de imagen de luminiscencia inducida por UV multiespectral para el estudio de recubri-
mientos en instrumentos musicaleshistóricos
Tiziana Cavaleri, Giacomo Fiocco, Tommaso Rovetta, Piercarlo Dondi, Marco Malagod, MonicaGulmini,
Anna Piccirillo, Marco Pisani, Massimo Zucco, Marco Gargano
181 Vis-NIR Reflection and Luminescence Hyperspectral Imaging for the Determination of CdS-based
Pigment Stoichiometry
Imágenes hiperespectrales de reflexión y luminiscencia Vis-NIR para la determinación de la estequiometría
de pigmento a base de CdS
Chiara Grazia, Claudia Sapienza, Costanza Miliani, Aldo Romani
201 Exploring the Ultraviolet Induced Infrared Luminescence of Titanium White Pigments
Explorando la luminiscencia infrarroja inducida por ultravioleta de pigmentos blancos de titanio
Annette T. Keller, Roland Lenz, Alessia Artesani, Sara Mosca, Daniela Comelli, Austin Nevin
233 Rediscovering Lost Decorations of 15th Century Wall Paintings in the Chapel of Our Lady of the
Fountains (La Brigue, France) by Means of UV-Vis Luminescence Imaging
Redescubrimiento de decoraciones perdidas en pinturas murales del siglo XV en la Capilla de Nuestra
Señora de las Fuentes (La Brigue, Francia) mediante luminiscencia UV-Vis
Sarah Boularand, Odile Guillon, Fanny Bauchau, Jean-Marc Vallet
261 Luminescent Staining with Rhodamine B to Study the Penetration of Calcium Hydroxide-based
(Ca (OH)2) Nanoparticulated Consolidants to Preserve Rock Art
Tinción luminiscente con Rodamina B para estudiar la penetración de consolidantes nanoparticulados a
base de hidróxido de calcio (Ca (OH)2) en la conservación de arte rupestre
Gemma Barreda-Usó, Mª Antonia Zalbidea-Muñoz
279 A survey on the Use of Ultraviolet Induced Visible Luminescence in Paper Conservation
Informe sobre el uso de la luminiscencia visible inducida por ultravioleta en la conservación depapel
Aafke Weller, Bas Van Velzen, Ewout Koek
A New Imaging Method of Luminescence Induced by Multispectral UV for the Study of Coatings on Historical Musical Instruments
Un nuevo método de imagen de luminiscencia inducida por UV multiespectral para el estudio de recubrimientos en instrumentos musicaleshistóricos
Cavaleri et al.
Abstract: in art conservation, UV luminescence is
a widespread technique: traditionally, long wave UV-A
lamps (365 nm-emission peak) are used to highlight
the distribution of varnishes or retouches, and of some
pigments such as the red lakes. Recently, a new meth-
od of luminescence has been tested, induced by a mul-
tispectral UV-Vis radiation source (11narrow bands in
the 295-405nm range), proving that luminescence im-
ages obtained using multiple excitation bands provide
Resumen: la luminiscencia UV es una técnica generaliza-
da en la conservación de obras de arte: tradicionalmente, las
lámparas UV-A de onda larga (pico de emisión de 365 nm) se
utilizan para resaltar la distribución de barnices o repintes,
y de algunos pigmentos como las lacas rojas. Recientemente,
se ha probado un nuevo método de luminiscencia inducido
por una fuente de radiación UV-Vis multiespectral (11 bandas
estrechas en el rango de 295-405 nm), lo que demuestra que
las imágenes de luminiscencia obtenidas utilizando múltiples
Tiziana Cavaleria*, Giacomo Fioccob,c, Tommaso Rovettab, Piercarlo Dondib,d, Marco Malagodib,e,
MonicaGulminic, Anna Piccirilloa, Marco Pisanif, Massimo Zuccof, Marco Garganog
aFondazione Centro per la Conservazione e il Restauro La Venaria Reale, Italy.
bArvedi Laboratory of Non-Invasive Diagnostics, CISRiC, University of Pavia, Italy.
cChemistry department, University of Turin, Italy.
d Electrical, Computer and Biomedical Engineering department, University of Pavia, Italy..
eMusicology and Cultural Heritage department, University of Pavia, Italy.
fIstituto Nazionale di Ricerca Metrologica (INRIM), Italy.
gPhysics department, University of Milan, Italy.
UV-Vis Luminescence Imaging Techniques |
Técnicas de imagen de luminescencia UV-Vis
bandas de excitación proporcionan una buena caracterización
de las propiedades luminiscentes del material. La aplicación a
instrumentos musicales históricos, donde el enfoque no invasi-
vo es fundamental, puede ser prometedora: por ejemplo, para
los instrumentos de cuerda frotada, la detección de materiales
colorantes como laca de granza puede ser decisiva para identi-
car métodos de acabado especícos adoptados en la fabrica-
ción de violines por los grandes maestros del pasado. En este
trabajo, se aplica el método de luminiscencia UV multiespectral
a reproducciones que imitan a los materiales empleados en re-
cubrimientos históricos en instrumentos de cuerda frotada, y se
centra en la identicación de laca de granza antes y después del
envejecimiento articial.
Palabras clave: luminiscencia UV; imágenes multiespectra-
les; instrumentos musicales de cuerda frotada; materiales de
recubrimiento; laca de granza; barnices.
Dentro de la preservación de los objetos del patrimo-
nio cultural, los instrumentos musicales históricos, y en
particular los de cuerda frotada tales como violines, vio-
las, violonchelos y contrabajos, requieren una atención
particular, en especial porque todavía se mantienen en
condiciones de funcionar, aunque tocarlos podría con-
ducir a deformaciones estructurales (Goli, Fioravanti,
Busoni, Carlson y Mazzanti, 2012), desgaste, pérdida
de barniz (Rovetta, Invernizzi, Licchelli, Cacciatori y
Malagodi, 2018) y desvanecimiento del color (Fiocco
etal., 2018).
Según las recetas históricas (Tirat et al., 2016;
Weththimuni etal., 2016) y los últimos resultados cien-
tícos en este tema, se sabe que los materiales utiliza-
dos en el pasado por los fabricantes de violines en los
“tratamientos de acabado” eran en su mayoría sensibles
a la luz, entre ellos, mezclas de resinas naturales y acei-
tes secantes en barnices y lacas como material colorante
a good characterization of the luminescent properties
of the material. The application to historic musical in-
struments, where non-invasive approaches are strongly
encouraged, can be surely promising: for example, for
bowed instruments the detection of specific color-
ing materials such as madder lake can be decisive for
identifying specific finishing methods adopted in violin
making by the great Masters of the past. In this work,
we applied the multispectral UV luminescence method
to mock-ups mimicking historical coating materials for
bowed instruments, focusing on madder lake identifica-
tion before and after artificial aging.
Key words: UV-luminescence; multispectral im-
aging; bowed musical instruments; coating materials;
madder lake; varnishes.
Among the preservation of cultural heritage
artefacts, historical musical instruments - and in
particular the bowed stringed ones such as vio-
lins, violas, cellos and double basses - need special
attention, especially because they are still kept in
working condition, even though playing them could
lead to structural deformations (Goli, Fioravanti,
Busoni, Carlson and Mazzanti, 2012), wear, loss
of varnish (Rovetta, Invernizzi, Licchelli, Cacciatori
and Malagodi, 2018) and color fading (Fiocco etal.,
According to historical recipes (Tirat etal., 2016;
Weththimuni et al., 2016) and the latest scientif-
ic results in this topic, we know that the materials
used in the past by the violin makers in the “finish-
ing treatments” were mostly light-sensitive, among
them, mixtures of natural resins and siccative oils
in varnishes, and lakes as coloring material (Echard
A New Imaging Method of Luminescence Induced by Multispectral UV for the Study of Coatings on Historical Musical Instruments
Un nuevo método de imagen de luminiscencia inducida por UV multiespectral para el estudio de recubrimientos en instrumentos musicaleshistóricos
Cavaleri et al.
(Echard y Lavadrine, 2008). La fotodegradación de los
barnices y la decoloración de los tintes en los pigmentos
tipo laca provocan alteraciones graves que complican o
incluso impiden la identicación de los materiales origi-
nales: por esta razón, se recomiendan encarecidamente
un enfoque multianalítico (Gulmini etal., 2013). Estos
materiales se consideran entre los más relevantes en
la caracterización y, por lo tanto, en la preservación de
instrumentos musicales históricos. Esto también signi-
ca que, cuando se diseñan medidas de conservación
preventivas, se deben establecer estrategias metodoló-
gicas adecuadas para monitorizar su estado de conser-
vación (Fichera etal., 2018).
Entre las técnicas no invasivas, la imagen de luminis-
cencia inducida por una lámpara ultravioleta de 365nm
(Dondi etal., 2017) generalmente se aplica como método
preliminar para investigar la distribución de diferentes
materiales de recubrimiento en instrumentos musicales
históricos, aunque la técnica tiene muchas limitaciones
(Invernizzi, Fechera, Licchelli y Malagodi, 2018).
Por ejemplo, estudios recientes sobre reproduccio-
nes de sistemas de recubrimientos antiguos coloreados
han demostrado que después de largos ciclos de enve-
jecimiento articial, la luminiscencia rojiza de la laca de
granza ya no es detectable con lámparas ultravioletas
de 365nm ni con otras técnicas no invasivas comunes
(Fiocco etal., 2018).
Este estudio piloto probó un nuevo método de lumi-
niscencia, inducido por una fuente UV-Vis multiespectral
(11bandas en el rango de 295-405nm) en reproduccio-
nes que simulan sistemas de recubrimientos históricos
en instrumentos de cuerda frotada. Este método ya se
había aplicado en el estudio no invasivo de las propieda-
des de luminiscencia de pigmentos, colorantes (Cavaleri,
Gargano, Pisani, Zucco y Buscaglia, 2017), aglutinantes
y barnices (Cavaleri, Gargano, Pisani y Zucco, 2018), de-
mostrando que es una herramienta muy útil en el proce-
so de identicación de materiales.
and Lavadrine, 2008). Photo-degradation of var-
nishes and the fading of the dyes in lake pigments
lead to severe alterations that complicate or even
impede identification of the original materials: for
this reason, multi-analytical approaches are strong-
ly suggested (Gulmini etal., 2013). These materi-
als are considered among the most relevant in the
characterization and therefore in the preservation
of historical musical instruments. This means also
that, when planning preventive conservation mea-
sures, proper methodological strategies should be
set up for monitoring their condition (Fichera etal.,
Among the non-invasive techniques, imaging of
luminescence induced by a 365nm ultraviolet lamp
(Dondi et al., 2017) is usually applied as a prelimi-
nary method to investigate the distribution of dif-
ferent coating materials on historic musical instru-
ments, although the technique has many limitations
(Invernizzi, Fechera, Licchelli and Malagodi, 2018).
For example, recent studies on reproductions of
colored ancient coating systems have shown that
after long cycles of artificial aging, the reddish lumi-
nescence of the madder lake is no longer detectable
with 365nm ultraviolet lamps nor with other com-
mon non-invasive techniques (Fiocco etal., 2018).
This pilot study tested a new method of lu-
minescence, induced by a multispectral UV-Vis
source (11 bands in the 295-405 nm range) on
mock-ups simulating historical coating systems on
bowed instruments. This method had been already
applied elsewhere for the non-invasive study of
the luminescence properties of pigments, dyes
(Cavaleri, Gargano, Pisani, Zucco and Buscaglia,
2017), binders and varnishes (Cavaleri, Gargano,
Pisani and Zucco, 2018), demonstrating that it is
a powerful tool to support the material identifica-
tion process.
UV-Vis Luminescence Imaging Techniques |
Técnicas de imagen de luminescencia UV-Vis
Se sabe que la laca de granza emite una luminiscencia
rojiza con un amplio rango de excitación (250-550nm)
(Miliani, Romani y Favaro, 1998). Sin embargo, hasta
donde se sabe, aún no se ha estudiado su comporta-
miento de emisión para bandas de excitación estrechas.
Además, no se han estudiado los efectos del envejeci-
miento sobre esta luminiscencia. Por lo tanto, el estudio
actual incluye probetas envejecidas articialmente para
reproducir diferentes estados de conservación.
El objetivo principal de este trabajo era evaluar la de-
tectabilidad de la laca de granza en diferentes recubri-
mientos y después de haber sido sometidos a diferentes
ciclos de envejecimiento articial. Un paso adicional de
la investigación se centró en vericar la no invasividad
de este método, un requisito para su aplicación futura en
instrumentos de cuerda frotada históricos reales.
Materiales y métodos
Reproducciones. Se prepararon tres juegos de
muestras con varios estratos cada una según las recetas
históricas para barnizar instrumentos de cuerda frotada
de los siglos XVI al XVIII (Tirat etal., 2016; Weththimuni
etal., 2016).
Se pulieron fragmentos de madera de arce de
5×1×1 cm (alto×ancho×profundidad) y se aplicó un
estrato de imprimación de caseinato de amonio: 15%
de caseína (Kremer Pigmente, cod. 63200) disuelto en
agua desionizada mezclada con amoníaco (solución al
30%, 1 mL). El barniz era una mezcla de 75/25p/p de
aceite de linaza prensado en frío (Kremer Pigmente,
cod. 73054) y colofonia (Kremer Pigmente, cod.60300),
preparado siguiendo una antigua receta: el aceite de li-
naza se calentó hasta 270°C y se mantuvo a esta tem-
peratura durante 3horas; el aceite de linaza cocido se
enfrió a 150°C y se añadió colofonia molida; por lo tan-
to, la mezcla resultante se calentó nuevamente a 250°C
durante otros 30minutos (Spinella et al, 2017). La laca
de granza se adquirió como un pigmento comercial, que
It is known that madder lake emits a reddish
luminescence with a broad excitation range (250-
550 nm) (Miliani, Romani and Favaro, 1998), nev-
ertheless, at the best of our knowledge, its emission
behavior for narrow excitation bands has not been
reported. Moreover, the effects of aging on this lu-
minescence have not been studied. Therefore, the
current study includes artificially aged mock-ups in
order to reproduce different states of preservation.
The main aim of this work was to evaluate mad-
der lake detectability in different coating systems,
and after being subjected to different artificial ag-
ing cycles. An additional step of the research fo-
cused on verifying the complete non-invasiveness
of this method, a requirement for future application
to real historic bowed instruments.
Materials and Methods
Mock-ups. Three sets of multi-layered mock-ups
were prepared according to historical recipes for
varnishing bowed string instruments from 16th to
18th centuries (Tirat etal., 2016; Weththimuni etal.,
Maple wood slabs of 5×1×1 cm (h×w×d) were
polished and a ground layer of ammonium casein-
ate was applied: 15% of casein (Kremer Pigmente,
cod. 63200) dissolved in deionized water mixed
with ammonia (30% solution, 1 mL). The varnish
was a mixture of 75/25 w/w cold pressed linseed
oil (Kremer Pigmente, cod. 73054) and colophony
(Kremer Pigmente, cod. 60300), prepared follow-
ing an ancient recipe: linseed oil was heated up to
270°C and it was maintained at this temperature
for 3 hours; the cooked linseed oil was cooled to
150°C and milled colophony was added; therefore
the resulting mixture was heated again at 250°C for
other 30min (Spinella et al., 2017). Madder lake was
A New Imaging Method of Luminescence Induced by Multispectral UV for the Study of Coatings on Historical Musical Instruments
Un nuevo método de imagen de luminiscencia inducida por UV multiespectral para el estudio de recubrimientos en instrumentos musicaleshistóricos
Cavaleri et al.
se preparó tratando un extracto acuoso de granza con
alumbre de potasio e hidróxido de sodio (laca de gran-
za elaborada a mano de Fantuzzi Colori Vegetali). Cada
material se analizó previamente mediante diferentes
técnicas espectroscópicas (espectroscopía FT-IR, es-
pectroscopía XRF, microanálisis EDX), conrmando la
composición química descrita en las hojas de datos del
Como se muestra en la Tabla 1, para crear diferentes
sistemas de recubrimiento, la laca de granza (ML, de sus
siglas en inglés) se dispersó en diferentes niveles de la
estratigrafía y en diferentes concentraciones:
· ML en el barniz (probetas P3y P4, 7% y 20% en
peso de ML respectivamente en la capa de colo-
fonia de aceite de linaza), esparcidas sobre la im-
primación sin color a base de caseína;
· ML en la imprimación (probetas P5y P6, 1% y 4%
en peso de ML respectivamente en la capa a base
de caseína), sin barniz;
· ML en la imprimación (probetas P7 y P8, respecti-
vamente 1% y 4% en peso de ML respectivamente
en la capa a base de caseína), a continuación barni-
zado con barniz de aceite-colofonia no pigmentado.
Las concentraciones de ML se basaron en la posi-
bilidad de obtener dispersiones que podrían actuar
como una imprimación coloreada o como un barniz
coloreado. Las diferentes capas se aplicaron a mano
con un pincel en una capa (parte media de la probe-
ta) o dos capas superpuestas (parte superior de la
probeta) para crear diferentes intensidades de color.
El primer conjunto de probetas fue analizado an-
tes del envejecimiento (t0), el segundo y el tercero
se analizaron después de 240h (t1) y 520h (t2) de
envejecimiento artificial respectivamente. Se eligie-
ron tiempos de envejecimiento de hasta 520 h para
observar las características de diagnóstico de los
purchased as a commercial pigment, which was pre-
pared by treating an aqueous madder extract with
potash alum and sodium hydroxide (handmade mad-
der lake Fantuzzi Colori Vegetali). Each material
was previously analyzed by different spectroscopic
techniques (FT-IR spectroscopy, XRF spectroscopy,
EDX microanalysis), confirming the chemical com-
position described in the product datasheets.
As shown in Table 1, in order to create different
coating systems, madder lake (ML) was dispersed at
different levels of the stratigraphy and in different
· ML in the varnish (mock-ups P3and P4, respec-
tively 7% and 20% in weight of ML in the linseed
oil-colophony layer), spread on the uncolored
casein-based ground;
· ML in the ground layer (mock-ups P5 and P6,
respectively 1% and 4% in weight of ML in the
casein-based layer), without varnish;
· ML in the ground layer (mock-ups P7 and P8,
respectively 1% and 4% in weight of ML in the
casein-based layer), then varnished unpigment-
ed oil-colophony varnish.
The ML concentrations were based on the pos-
sibility of obtaining dispersions that could act as a
colored ground layer or as a colored varnish. The
different layers were applied manually with a brush
in one coat (middle part of the mock-up) or two su-
perimposed coats (upper part of the mock-up) in or-
der to create different color intensities.
The first set of mock-ups was analyzed before
aging (t0), the second and the third ones were
analyzed after, respectively, 240 h (t1) and 520h
(t2) of artificial aging. Aging times up to 520h were
chosen in order to observe the diagnostic features
UV-Vis Luminescence Imaging Techniques |
Técnicas de imagen de luminescencia UV-Vis
materiales involucrados a través del proceso de de-
gradación progresiva. Dado que el tiempo de enveje-
cimiento de la luz artificial no puede correlacionarse
directamente con el envejecimiento natural, se tomó
la decisión de extender el período hasta 520h.
Envejecimiento articial. Las probetas se expusie-
ron a luz día articial utilizando una cámara de enveje-
cimiento Heraeus Suntest CPS (Hanau, Alemania) equi-
pada con una lámpara de xenón ltrada (vidrio de cuarzo
revestido que simula un vidrio de ventana de 3mm de
espesor, con longitudes de onda <300nm) con una irra-
diación promedio de 750W/m2. El sistema de ventila-
ción de aire de la cámara mantuvo la temperatura inter-
na a 50°C aproximadamente.
Instrumentación y elaboración de imágenes. El nue-
vo método de imágenes de luminiscencia multiespectral
se llevó a cabo utilizando once ltros de paso de banda de
interferencia (Edmund Optics) con un ancho de banda de
of the involved materials through the progressive
degradation process. Since the time of artificial
light aging cannot be directly correlated to natural
aging, the choice was made to extend the period up
to 520h.
Artificial aging. Mock-ups were exposed to
artificial daylight using a Heraeus Suntest CPS
(Hanau, Germany) aging chamber equipped with a
filtered xenon lamp (Coated quartz glass simulating
a 3 mm thick window glass, cutting wavelengths
<300nm) with an average irradiation of 750W/m2.
The air-venting system of the chamber kept the
internal temperature at about 50°C.
Instrumentation and image elaboration.
The new method of multispectral luminescence
imaging was performed using eleven interference
bandpass filters (Edmund Optics) with a 10 nm
bandwidth (FWHM, full width half maximum) and
Table 1 | Outline of the mock-ups composition. When no coating
is present, the wooden support treated with ammonium caseinate
remains visible.
Tabla 1 | Esquema de la composición de las probetas. Cuando no hay
recubrimiento, el soporte de madera tratado con caseinato de amo-
nio permanece visible.
P3 P4 P5
2 coats of madder lake (7%) in oil-colophony
2 coats of madder lake (20%) in oil-colophony
2 coat of madder lake (1%) in ammonium
1 coat 1 coat 1 coat
no coating no coating no coating
P6 P7 P8
2 coats of madder lake (4%) in ammonium
oil-colophony varnish spread on 2 coats of
madder lake (1%) in ammonium caseinate
oil-colophony varnish spread on 2 coats of
madderlake (4%) in ammonium caseinate
1 coat oil-colophony varnish spread on 1 coat oil-colophony varnish spread on 1 coat
no coating no coating no coating
A New Imaging Method of Luminescence Induced by Multispectral UV for the Study of Coatings on Historical Musical Instruments
Un nuevo método de imagen de luminiscencia inducida por UV multiespectral para el estudio de recubrimientos en instrumentos musicaleshistóricos
Cavaleri et al.
10nm (FWHM, ancho máximo medio completo) y con pi-
cos de transmitancia que iban de 295nm a 405nm en pa-
sos de 10nm. Los rasgos espectrales de los ltros (Fig.1)
se caracterizaron con un espectrofotómetro (HR4000,
Ocean Optics) y una fuente de luz de banda ancha (EQ-
400LDLS Energetiq). Los ltros se montaron en una rue-
da delante de una lámpara de arco de xenón (Arc Lamp,
Cairn Research) y la radiación ltrada se centró en las
supercies de la probeta con un reector parabólico. Con
este sistema, se crearon 11 bandas de excitación en el
rango de 295-405nm, con las cuales se estudió la luminis-
cencia inducida dependiente de las diferentes longitudes
de onda de excitación.
Las imágenes de luminiscencia multiespectral se ad-
quirieron con dos dispositivos diferentes. En primer lu-
gar, se realizó una cámara detectora monócroma (Ascent
A4000, detectores de Si enfriados, 2048×2048píxeles)
con un sensor de alta eciencia cuántica y lente de cuar-
zo, normalmente utilizada en aplicaciones fotónicas: esto
permitió medir incluso señales de baja luminiscencia y
vericar la aplicabilidad fundamental del método. En se-
gundo lugar, se utilizó una cámara de fotograma completo
Nikon 810D DSLR (7360x 4912píxeles) con una lente
Nikkor AF de 50mm: esta segunda cámara se seleccionó
para probar la posibilidad de utilizar una cámara comer-
cial de alta gama para la adquisición de información útil
y able acerca de las propiedades luminiscentes de los
with transmittance peaks ranging from 295 nm
to 405 nm in 10nm steps. The spectral features
of the filters (Fig. 1) were characterized with a
spectrophotometer (HR4000, Ocean Optics) and a
broadband light source (EQ-400LDLS Energetiq).
Filters were mounted on a wheel in front of a
xenon arc lamp (Arc Lamp, Cairn Research) and
the filtered radiation was focused on the mock-
up surfaces with a parabolic reflector. With this
system, 11 excitation bands were created in
the 295-405 nm range, with which the induced
luminescence dependent of the different excitation
wavelengths were studied.
Multispectral luminescence images were
acquired with two different devices. Firstly, a
monochrome detector camera (Ascent A4000,
cooled Si detectors, 2048×2048 pixels) was used
with a high quantum efficiency sensor and quartz
lens, normally used in photonic applications: this
allowed measuring even low luminescence signals
and verifying the fundamental applicability of the
method. Secondly, a Nikon 810D DSLR full-frame
camera (7360×4912pixel) was used with a 50mm
AF Nikkor lens: this second camera was selected to
test the possibility of using a high-level commercial
camera for the acquisition of reliable and useful
information about the luminescent properties of
Figure 1 | Transmittance of the 300-400 nm band-pass
Interference Filters, Edmund Optics.
Figura 1 | Transmitancia de los ltros de interferencia de paso de
banda de 300-400nm, Edmund Optics.
UV-Vis Luminescence Imaging Techniques |
Técnicas de imagen de luminescencia UV-Vis
materiales. La conguración experimental se muestra en
la Figura 2: el ejemplo muestra la imagen adquirida con la
cámara DSLR a λexc 320nm, la posición de las probetas y
de tres objetivos de referencia, así como su luminiscencia.
Para ambos dispositivos, se unieron diferentes l-
tros de paso largo a las lentes (como se especica en la
tabla2) con el n de excluir la parte reejada de la radia-
ción de excitación. Tal y como se muestra en esta tabla,
fue posible examinar cuatro rangos espectrales diferen-
tes con la cámara cientíca. Con la cámara comercial
solo pudieron examinarse dos combinaciones de ltros,
debido a la transmisión espectral especíca de la lente
comercial, que absorbe la radiación UV hasta 350nm,
evitando la detección de cualquier señal por debajo de
ese valor.
the materials. The experimental setup is shown in
Figure 2: the example shows the image acquired
with the DSLR camera at λexc 320nm, the position
of the mock-ups and of three reference targets as
well as their luminescence.
For both devices, different long-pass filters (as
specified in Table 2) were attached to the lens in
order to exclude the reflected part of the excitation
radiation. As shown in this table, with the scientif-
ic camera it was possible to examine four different
spectral ranges. With the commercial camera just
two combinations of filters were possible, due to
the specific spectral transmission of the commercial
lens, which absorbs UV radiation up to 350nm, pre-
venting the detection of any signal below that value.
Figure 2| Experimental setup (left) and luminescence image ac-
quired with the Nikon camera at λexc 320 nm showing mock-ups
and reference target positions (right). References have been used
to correct the images since using long-pass colored filters a chro-
matic alteration is inevitably added to the acquired image.
Figura 2 | Conguración experimental (izquierda) e imagen de lumi-
niscencia adquirida con la cámara Nikon a λexc 320 nm que muestra
las probetas y las posiciones de referencia del objetivo (derecha). Se
usaron referencias para corregir las imágenes, ya que al usar ltros de
color de paso largo, inevitablemente se agrega una alteración cromá-
tica a la imagen adquirida.
A New Imaging Method of Luminescence Induced by Multispectral UV for the Study of Coatings on Historical Musical Instruments
Un nuevo método de imagen de luminiscencia inducida por UV multiespectral para el estudio de recubrimientos en instrumentos musicaleshistóricos
Cavaleri et al.
Con respecto a las imágenes de luminiscencia ad-
quiridas con la cámara Ascent, para cada longitud de
onda de excitación se adquirieron dos imágenes mono-
cromáticas: la primera es la imagen de la luminiscencia
que proviene de la supercie de la probeta, la segunda
es la imagen de una hoja de referencia no luminiscente
bajo las mismas condiciones de iluminación. Al dividir
la primera imagen por la segunda, se obtuvo la imagen
normalizada corregida por falta de uniformidad debido
a variaciones en la iluminación. Dado que las imágenes
resultantes no se corrigen para los diferentes factores
dependientes de la longitud de onda (emisividad de la
fuente, transmitancia del ltro, luminiscencia espectral
de referencia blanca), solo fue posible comparar áreas
(probetas) dentro de la misma imagen. La comparación
entre áreas en diferentes imágenes (adquiridas a dife-
rentes longitudes de onda de excitación) fue imposible.
Para las imágenes de luminiscencia adquiridas con la
cámara Nikon, fue necesario establecer un proceso de
postproducción razonable. Usando ltros de color de
Regarding the luminescence images acquired with
the Ascent camera, for each excitation wavelength
two monochromatic images were acquired: the rst
one is the image of the luminescence coming from
the mock-up surface, the second one is the image of
a non-luminescent reference sheet under the same
illuminating conditions. By dividing the rst image
by the second, the normalized image corrected for
non-uniformity due to variations in the illumination
was obtained. Since the resulting images are not
corrected for the different wavelength-dependent
factors (source emissivity, lter transmittance, white
reference spectral luminescence) only comparing
areas (mock-ups) within the same image was possi-
ble. Comparison between areas in different images
(acquired at different excitation wavelengths) was
For the luminescence images acquired with the
Nikon camera, it was necessary to set up a reasonable
post-production process. Using long-pass colored
Table 2| Instrumental set-up and conditions. Tabla 2 | Conguración instrumental y condiciones.
(xenon + UV lters)
(xenon + ltros UV)
Instrumentation and acquisition range / Instrumentación y rango de adquisición
Camera / Filter combination
Combinación Cámara / Filtro
Sensitivity range
Rango de sensibilidad
Camera/Filter combination
Combinación Cámara / Filtro
Sensitivity range
Rango de sensibilidad
300 nm
Ascent A4000 +
Edge BasicTM Long Wave pass 325 nm 325-980 nm
Nikon D810 +
IR/UV Cut lter + Cyan lter 380-780 nm
310 nm
320 nm
330 nm
Ascent A4000 +
Edge BasicTM Long Wave pass 355 nm 355-980 nm340 nm
350 nm
360 nm Ascent A4000 + BrightLine®
long-pass lter 380 nm 380-980 nm
370 nm
380 nm
Ascent A4000 +KV418 420-980 nm
Nikon D810 +
IR/UV Cut lter +
Cyan lter + KV418
420-720 nm390 nm
400 nm
UV-Vis Luminescence Imaging Techniques |
Técnicas de imagen de luminescencia UV-Vis
paso largo, inevitablemente se agrega una alteración
cromática a la imagen adquirida. Además, no hay están-
dares luminiscentes correctos colorimétricamente que
sean comparables a los utilizados comúnmente para Vis,
por lo que el procedimiento de corrección de color su-
pone un desafío. Se intentó realizar una corrección de
color usando tres referencias luminiscentes ad-hoc, in-
cluidas en la conguración, cada una de las cuales exhibía
un luminiscente promedio característico en el azul, en el
verde y en el rango rojo del espectro Vis (Fig.2-dcha).
Desafortunadamente, dos de estas referencias mostra-
ron diferentes rangos de luminiscencia en comparación
con los de las muestras, produciendo una saturación
distinta para algunas bandas de excitación. Solo el están-
dar rojo luminiscente, producido/introducido original-
mente para la orientación del láser UV/Vis/NIR, podría
usarse en los pasos de normalización, con la conciencia
de que, al usar un solo canal, el cambio colorimétrico
sería inevitable en la secuencia de imágenes. Este es-
tándar de referencia se midió y caracterizó usando un
espectrómetro de luminiscencia Perkin Elmer LS-50B
(rango de excitación 200-800nm, precisión de longitud
de onda 1.0 nm, reproducibilidad de longitud de onda
0.5 nm). Los espectros resultantes (Fig. 3) mostraron
que la luminiscencia ocurre en un rango limitado para las
11bandas utilizadas. Esto llevó a la conclusión de que el
objetivo rojo es un buen estándar luminiscente para las
aplicaciones comentadas.
Las imágenes en bruto se corrigieron para los pa-
rámetros de disparo (f/número, tiempo de exposición,
ISO), posteriormente se multiplicaron por el factor de
corrección obtenido midiendo la irradiancia efectiva re-
cibida por la supercie de la muestra como se observa
en la Figura 4y se presentan en la siguiente sección.
Medición de la irradiación de la fuente de luz. Para
evaluar si la técnica de luminiscencia multiespectral es
realmente no invasiva (es decir, no daña los recubri-
mientos de la supercie, paso necesario de cara al uso
de la técnica con instrumentos históricos reales), se
filters, inevitably a chromatic alteration is added
to the acquired image. Moreover, luminescent
colorimetrically-correct standards comparable to
those commonly used for Vis are not available, thus
the procedure of color correction is challenging. An
attempt was made to perform a color correction
using three ad-hoc luminescent references, included
in the set-up, each exhibiting a characteristic
average luminescence in the blue, in the green and
in the red range of the Vis spectrum (Fig. 2-right).
Unfortunately, two of these references showed
different ranges of luminescence compared to those
of the samples, producing a distinct saturation for
some excitation bands. Only the red luminescent
standard, originally produced/introduced for
UV/Vis/NIR laser targeting, could be used in the
normalization steps, with the awareness that
using a single channel, a colorimetric shift may be
unavoidable in the images sequence. This reference
standard was measured and characterized using a
Perkin Elmer LS-50B luminescence spectrometer
(excitation range 200-800nm, wavelength accuracy
1.0 nm, wavelength reproducibility 0.5 nm). The
resulting spectra (Fig. 3) showed that luminescence
occurs in a limited range for the 11bands used. This
led to the conclusion that the red target is a good
luminescent standard for the reported applications.
Raw images were corrected for the shooting
parameters (f/number, exposition time, ISO),
subsequently multiplied with the correction factor
obtained by measuring the effective irradiance
received by the sample surface as shown in
Fig.4and are presented in the next section.
Light source irradiance measurement. In order
to evaluate whether the multispectral luminescence
technique is indeed non-invasive, (i.e. does not
damage the surface coatings, a necessary step
towards future use with real historical instruments),
UV irradiance measurements were carried out
A New Imaging Method of Luminescence Induced by Multispectral UV for the Study of Coatings on Historical Musical Instruments
Un nuevo método de imagen de luminiscencia inducida por UV multiespectral para el estudio de recubrimientos en instrumentos musicaleshistóricos
Cavaleri et al.
llevaron a cabo mediciones de irradiancia UV para cada
banda de excitación de la fuente de luz. Se colocó un fo-
todiodo de referencia con un área nominal de 1cm2a la
misma distancia de la lámpara donde se colocó la mues-
tra (aproximadamente 1,2m). Para establecer la energía
máxima en vatios/cm2 para cada longitud de onda, se
usó la forma consistente en que un fotodiodo convierte
la potencia óptica en corriente eléctrica: el fotodiodo se
colocó en la posición de máxima exposición a la lámpara.
La corriente generada por el fotodiodo se registró con
un multímetro calibrado para cada ltro aplicado delan-
te de la lámpara y se reguló a 80W. Esta es la potencia
máxima utilizada en el experimento. Las corrientes re-
gistradas para cada ltro se multiplicaron por la curva
de eciencia nominal del propio fotodiodo (en vatios/
amperios). De esta manera, se estableció la energía
máxima en vatios/cm para cada longitud de onda. La in-
certidumbre de la medición, principalmente debido a la
curva de respuesta, fue inferior al 10%. Los valores de
irradiación UV utilizados varían entre 1,37mW/cm2 (a
λexc340nm) a 15,49mW/cm2(a λexc370nm), lo que pue-
de considerarse potencialmente perjudicial en términos
for each excitation band of the light source. A
reference photodiode with a nominal area of 1 cm2
was positioned at the same distance from the lamp
where the samples were placed (about 1.2 m) in
the position of maximum irradiance with respect to
illuminated area. To measure and set the maximum
energy in W/cm2 for each wavelength, the current
generated by the photodiode was measured with a
calibrated multimeter and converted in optical power
multiplying the value corresponding to each lter by
the nominal efciency curve of the photodiode itself.
A value of 80 W was then selected as the maximum
allowed power for the light source and used for the
experiment. The uncertainty of the measurement,
mainly due to the responsivity curve, was below
10%. Values of UV irradiance used range between
1.37mW/cm2(at λexc 340nm) to 15.49mW/cm2 (at
λexc 370 nm), which can be considered potentially
harmful in terms of conservation. Although there are
no precise regulations for UV radiation in cultural
heritage conservation, some guidelines (MIBACT,
2000) recommend not exposing the cultural object to
Figure 3| Luminescence spectra of the red reference target used
for the color correction of the raw images. The eleven spectra
show similar behavior when varying the excitation wavelength.
Figura 3| Espectros de luminiscencia del objetivo de referencia rojo
utilizado para la corrección de color de las imágenes en bruto. Los
once espectros muestran un comportamiento similar al variar la lon-
gitud de onda de excitación.
Figure 4 | Irradiance of the excitation bands used in the multi-
spectral UV luminescence technique.
Figura 4 | Irradiación de las bandas de excitación utilizadas en la
técnica de luminiscencia UV multiespectral.
UV-Vis Luminescence Imaging Techniques |
Técnicas de imagen de luminescencia UV-Vis
de conservación. Aunque no existen regulaciones pre-
cisas para la radiación UV en la conservación del patri-
monio cultural, algunas pautas (MIBACT, 2000) reco-
miendan no exponer el objeto cultural a más de 1,2mW/
cm2para el promedio de objetos fotosensibles y no más
de 0,4mW/cm2 para objetos fotosensibles (Fig.4). Estos
niveles deben considerarse como valores máximos ab-
solutos para la exposición continua, mientras que en los
análisis realizados se superó la cantidad instantánea de
radiación UV, aunque se puede considerar que la dosis
total cumple con los niveles de seguridad aceptables
para la mayoría de los objetos y solo si están raramen-
te expuestos a condiciones extremas y por períodos de
tiempo muy cortos.
Tal y como se mostró en un estudio anterior, después
de ciclos de 240 h (t1) y 520h (t2) de envejecimiento
articial, el pigmento laca de granza disperso en siste-
mas de recubrimiento que simulan instrumentos histó-
ricos de cuerda frotada, apenas es detectable con otras
técnicas no invasivas. La observación microscópica bajo
luz Vis permite la identicación de un número muy bajo
de partículas de laca que retienen su color rojo disperso
en el aglutinantes aceitoso o proteico, pero la mayoría
ha cambiado de rojo a blanco (Fig.5, ver por ejemplo las
manchas blancas en la imagen P6 en t2) (Fiocco et al.,
Mientras que la laca ya no es detectable con otras
técnicas, las imágenes adquiridas con la cámara cientíca
(Ascent A4000) muestran y permiten mediciones de la in-
tensidad de la luminiscencia emitida por la laca en las pro-
betas para cada banda de excitación. Además, para cada
banda de excitación es posible documentar y calcular las
diferencias entre las tres partes de la probeta, correspon-
dientes a 0, 1y 2capas de material de recubrimiento (ver
Tabla 1), utilizando los valores de nivel de gris extraídos
de imagen. Incluso se pueden detectar ligeras diferencias
en términos de luminiscencia, correspondientes a las di-
ferentes capas de los materiales.
more than 1.2mW/cm2for averagely photosensitive
objects and not to more than 0.4mW/cm2for highly
photosensitive objects (Fig. 4). These levels have
to be considered as absolute maximum values for
continuous exposition, whereas in the reported
analyses the instantaneous amount of UV radiance
was exceeded, but the total dose may be considered
to meet acceptable safety levels for most objects and
only if they are exposed extremely rarely and for very
short periods of time.
As shown in a previous study, after cycles of
240 h (t1) and 520 h (t2) of artificial aging mad-
der lake pigment dispersed in coating systems
simulating historical bowed instruments is barely
or no longer detectable with other non-invasive
techniques. Observation under optical microsco-
py allows for the identification of a very low num-
ber of madder particles that retain their red color
dispersed in the oily or proteinaceous binders, but
most have changed from red to white (Fig. 5, see
for example the white spots on the image P6at t2)
(Fiocco etal., 2018).
While the madder is no longer detectable with
other techniques, the images acquired with the
scientific camera (Ascent A4000) show and allow
measurements of the intensity of luminescence
emitted by the madder in the mock-ups for each
excitation band. Moreover, for each excitation
band it is possible to document and calculate the
differences between the three parts of the mock-
up, corresponding to 0, 1and 2 coats of coating
material (see Table1), using the grey level values
extracted from the image. Even slight differences
in terms of luminescence, corresponding to the
different layering of the materials, can still be
A New Imaging Method of Luminescence Induced by Multispectral UV for the Study of Coatings on Historical Musical Instruments
Un nuevo método de imagen de luminiscencia inducida por UV multiespectral para el estudio de recubrimientos en instrumentos musicaleshistóricos
Cavaleri et al.
Figure 5 | Optical microscopy, Vis light images: mock-ups
P4(2 coats of 20% ML in oil-colophony varnish) and P6(2coats
of 4% ML in ammonium caseinate) at t0, t1(240h) and t2(520h)
artificial aging.
Figura 5 | Microscopía óptica, imágenes de luz Vis: probetas P4(2ca-
pas de 20% ML en barniz de aceite de colofonia) y P6 (2capas de
4% ML en caseinato de amonio) en t0, t1(240h) y t2(520h) enveje-
cimiento articial.
UV-Vis Luminescence Imaging Techniques |
Técnicas de imagen de luminescencia UV-Vis
Considerando, por ejemplo, las probetas P4 para to-
das las bandas de excitación, la luminiscencia total (de-
bido tanto al aglutinante aceitoso como a la laca) es en
general más intensa después del envejecimiento y se no-
tan ligeras diferencias en la intensidad de luminiscencia
entre los tiempos de envejecimiento t1y t2. El gráco de
la Figura 6se reere en particular a la porción de la pro-
beta preparada con dos capas de 20% de ML en barniz
de colofonia de aceite. El mayor cambio en términos de
luminiscencia se puede ver en particular a λexc 370 nm,
donde el valor de intensidad aumenta de 44,18en t0a
111,54 en t1 y disminuye ligeramente en t2 (99,81).
Una segunda diferencia notable aparece a λexc 310nm,
con un gran aumento de 13,58 en t0a 42.37en t1y una
ligera disminución a 36.58en t2.
En contraste, en ausencia de la laca de granza y su
aglutinante (Fig. 7, probetas preparadas solo con ca-
seinato de amonio en la madera) la intensidad de la lu-
miniscencia no cambia sustancialmente entre t0 y t1,
mientras que disminuye considerablemente después de
520horas de envejecimiento articial (t2) para las ban-
das de excitación 300-320nm, 370nm y 400nm.
Esta diferencia conrma que las propiedades de
luminiscencia especícas del sistema laca-agluti-
nante son realmente detectables con esta técnica.
Además, en las imágenes de luminiscencia adquiridas
con la cámara Nikon antes del envejecimiento, la emisión
rojiza/rosada de la laca es claramente evidente tanto en
aglutinantes proteicos como aceitosos antes del enve-
jecimiento, y se puede evaluar por separado para cada
banda de excitación (Figs. 8, 9, 10, ver probetas en t0).
Cuando la laca de granza se dispersa en un aglutinan-
te oleoso, en particular cuando se aplica en dos capas, la
luminiscencia rojiza se sigue distinguiendo claramente
incluso después de 240horas de envejecimiento arti-
cial (Fig. 8, ver la parte superior de las probetas en t1)
durante varias bandas de excitación y es particularmen-
te evidente para λexc 300nm, 310nm, 340y 400nm.
Considering for example the mock-ups P4for all the
excitation bands, the total luminescence (due to both
the oily binder and the lake) is in general more intense
after aging and slight differences in luminescence in-
tensity are noticeable between aging times t1and t2.
The graph at gure 6refers in particular to that portion
of the mock-up prepared with two coats of 20% ML in
oil-colophony varnish. The greatest change in terms of
luminescence can be seen in particular at λexc 370 nm,
where the intensity value increases from 44.18at t0to
111.54at t1and slightly decreases at t2(99.81). A sec-
ond notable difference appears at λexc 310nm, with a
great increase from 13.58at t0to 42.37at t1and a
slight decrease to 36.58at t2.
In contrast, in the absence of the madder lake
and its binder (Fig. 7, mock-ups prepared only with
ammonium caseinate on wood) the luminescence
intensity does not change substantially between
t0and t1, whereas it decreases considerably after
520hours of artificial aging (t2) for the excitation
bands 300-320nm, 370nm and 400nm.
This difference confirms that the specific lumi-
nescence properties of the lake-binder system are
indeed detectable with this technique.
In the luminescence images acquired with the
Nikon camera before aging, moreover, the reddish/
pinkish emission of the madder is clearly evident
both in proteinaceous and oily binders before ag-
ing, and it can be evaluated separately for each exci-
tation band (Figs. 8, 9, 10, see mock-ups at t0).
When madder lake is dispersed in an oily binder,
in particular when applied in two coats, the reddish
luminescence is still clearly distinguishable after
240hours of artificial aging (Fig. 8, see the upper
part of the mock-ups at t1) for several excitation
bands and it is particularly evident for λexc 300nm,
310nm, 340and 400nm.
A New Imaging Method of Luminescence Induced by Multispectral UV for the Study of Coatings on Historical Musical Instruments
Un nuevo método de imagen de luminiscencia inducida por UV multiespectral para el estudio de recubrimientos en instrumentos musicaleshistóricos
Cavaleri et al.
Por el contrario, cuando más laca se dispersa en un
aglutinante proteico, la emisión rojiza/rosada es eviden-
te después de 240 horas de envejecimiento solo cuando
se usan bandas de excitación entre 300nm y 330nm o
400nm (Figs. 9, 10, ver maquetas en t1).
Después de 520 horas de envejecimiento, el tono
rosado/rojizo de la luminiscencia que indica la presen-
cia de una laca ya no es visible en ningún aglutinante
(Figuras 8, 9, 10, ver probetas en t2).
El método de luminiscencia inducida por una fuente
multiespectral de UV-Vis propuesto (11 bandas estre-
chas en el rango de 295-405 nm) ha demostrado ser
muy eciente para detectar la luminiscencia rojiza / ro-
sada de la laca de granza dispersa en diferentes sistemas
In contrast, when madder is dispersed in a pro-
teinaceous binder, the reddish/pinkish emission is
evident after 240hours of aging only when using
excitation bands between 300nm and 330 nm or
400nm (Figs. 9, 10, see mock-ups at t1).
After 520hours of aging, the pinkish/reddish hue
of luminescence indicating the presence of madder
lake is no longer visible in either binder (Figs. 8, 9,
10, see mock-ups at t2).
The proposed method of luminescence induced
by multispectral UV-Vis source (11narrow bands in
the 295-405nm range) has shown to be very efficient
for detecting the reddish/pinkish luminescence
of the madder lake dispersed in different coating
Figure 7 | Luminescence intensity values obtained measuring
grey level intensities (mean and standard deviation) of the imag-
es recorded with the Ascent camera, for each λexc on the mock-up
of wood treated only with the casein-based ground, before aging
(t0) and after two different periods of aging (t1and t2).
Figura 7 | Valores de intensidad de luminiscencia obtenidos midiendo
las intensidades de nivel de gris (media y desviación estándar) de las
imágenes grabadas con la cámara Ascent, para cada λexc en la maqueta
de madera tratada solo con tierra a base de caseína, antes del enveje-
cimiento (t0) y después de dos períodos diferentes de envejecimiento
(t1y t2).
Figure 6 | Luminescence intensity values obtained measuring
grey level intensities (mean and standard deviation) of the images
recorded with the Ascent camera for each λexc on mock-up P4with
2coats of 20% ML in oil-colophony varnish, before aging (t0) and
after two different periods of aging (t1and t2).
Figura 6 | Valores de intensidad de luminiscencia obtenidos midiendo
las intensidades del nivel de gris (media y desviación estándar) de las
imágenes grabadas con la cámara Ascent para cada λexc en la maqueta
P4 con 2 capas de 20% ML en barniz de aceite-colofonia antes del
envejecimiento (t0) y después de dos períodos diferentes de enveje-
cimiento (t1y t2).
UV-Vis Luminescence Imaging Techniques |
Técnicas de imagen de luminescencia UV-Vis
de recubrimiento, tanto en aglutinantes oleosos como
proteicos. De esta forma, la técnica puede ser una herra-
mienta no invasiva útil en la identicación de la compo-
sición de los recubrimientos (revelando la presencia de
laca de granza y, por tanto, proporcionando indicaciones
del color original) en instrumentos históricos de cuerda
Sin embargo, otros colorantes rojos pueden tener
propiedades de luminiscencia similares y la literatura so-
bre las recetas históricas utilizadas en el pasado por los
systems, both in oily and proteinaceous binders.
This way, the technique can be a helpful non-
invasive tool for supporting the identification of
the composition of the coatings (revealing the
presence of madder lake and therefore providing
indications of the original color) on historic bowed
Nevertheless, other red colourants may have
similar luminescence properties and the litera-
ture on the historical recipes used in the past by
Figure 8 | Multispectral UV luminescence of mock-ups series
P4 (mock-ups prepared with 20% madder lake in oil-colophony
varnish: 2 coats; 1 coat; no coating at the bottom) exposed to
11excitation bands at different levels of aging (t0, t1, t2).
Figura 8 | Luminiscencia multiespectral UV de probetas serie P4(pro-
betas preparadas con un 20% de laca en barniz de aceite-colofonia:
2capas; 1capa; sin capa en el fondo) expuestas a 11bandas de excita-
ción a diferentes niveles de envejecimiento. (t0, t1, t2).
A New Imaging Method of Luminescence Induced by Multispectral UV for the Study of Coatings on Historical Musical Instruments
Un nuevo método de imagen de luminiscencia inducida por UV multiespectral para el estudio de recubrimientos en instrumentos musicaleshistóricos
Cavaleri et al.
fabricantes de violines no es tan exhaustiva como para
excluir otros tipos de lacas. Por tanto, la investigación
adicional debe centrarse en la comparación y el análisis
de otros materiales con esta técnica con el objetivo de
mapear las características de luminiscencia de cada ma-
terial; cuando se haya dado ese paso, dependiendo de
los resultados, la técnica podría implementarse en pro-
tocolos de diagnóstico multianalíticos.
Curiosamente, este trabajo ha demostrado que con
la radiación UV multiespectral, la emisión rojiza/rosada
violinmakers is not so exhaustive that other types
of lakes can be excluded. Therefore, further re-
search should focus on comparison and analysis of
other materials with this technique with the aim of
mapping luminescence features of each material;
when that step has been taken, depending on the
outcomes, the technique could be implemented into
multi-analytical diagnostic protocols.
Interestingly, this work has demonstrated
that with multispectral UV, the reddish/pinkish
Figure 9 | Multispectral UV luminescence of the mock-ups
series P6(mock-ups colored with 4% madder lake in ammonium
caseinate: 2coats; 1coat; no coating at the bottom) exposed to
11excitation bands at different levels of aging (t0, t1, t2).
Figura 9 | Luminiscencia multiespectral UV de la serie de probetas
P6 (maquetas coloreadas con 4% de laca en caseinato de amonio:
2capas; 1capa; sin capa en el fondo) expuestas a 11bandas de excita-
ción a diferentes niveles de envejecimiento ( t0, t1, t2).
UV-Vis Luminescence Imaging Techniques |
Técnicas de imagen de luminescencia UV-Vis
de la laca de granza todavía se puede detectar después
de 240horas de envejecimiento articial. Cuando la laca
se dispersa en un aglutinante oleoso, permanece detec-
table con la excitación probada. Dispersada en un aglu-
tinante proteico, la laca permanece detectable usando
bandas de excitación entre 300nm y 330nm o 400nm.
Este hecho es un punto de partida importante para la in-
vestigación, ya que anteriormente se había demostrado
que la laca de granza era poco o nada detectable con téc-
nicas tradicionales no invasivas después de largos perío-
dos de envejecimiento articial.
emission of madder lake can still be detected af-
ter 240hours of artificial aging. When madder is
dispersed in an oily binder it remains detectable
with excitation tested. Dispersed in a protein-
aceous binder, the madder remains detectable us-
ing excitation bands between 300nm and 330nm
or 400nm. This fact is an important starting point
for the research, as madder lake was shown earli-
er to be barely or not detectable after long peri-
ods of artificial aging with traditional non-invasive
Figure 10 | Multispectral UV luminescence of the mock-ups
series P8(mock-ups colored with 4% madder lake in ammonium
caseinate in 2coats, 1coat and with no coating at the bottom, then
varnished with oil-colophony varnish) exposed to 11 excitation
bands at different levels of aging (t0, t1, t2).
Figura 10 | Luminiscencia multiespectral UV de la serie de probetas
P8(maquetas coloreadas con 4% de laca en caseinato de amonio en
2capas, 1capa y sin capa en la parte inferior, barnizadas con barniz
de aceite-colofonia) expuestas a 11 bandas de excitación a diferentes
niveles de envejecimiento (t0, t1, t2)..
A New Imaging Method of Luminescence Induced by Multispectral UV for the Study of Coatings on Historical Musical Instruments
Un nuevo método de imagen de luminiscencia inducida por UV multiespectral para el estudio de recubrimientos en instrumentos musicaleshistóricos
Cavaleri et al.
Las pruebas realizadas para esta investigación han
demostrado que los niveles de exposición son algo su-
periores a los niveles conocidos recomendados para la
exposición continua a los rayos UV de los objetos del
patrimonio. Sin embargo, como solo se usan tiempos
de exposición cortos para UV multibanda, la dosis total
recibida por los objetos es baja. Por lo tanto, el método
puede considerarse seguro para instrumentos de cuer-
da frotada. El tiempo de exposición corto, que varía de
1sa 30s, dependiendo de la emisión de intensidad de la
fuente UV y la transmitancia especíca de los ltros de
paso de banda, asegura que la cantidad total de radia-
ción UV recibida por el objeto esté muy por debajo de
los niveles de seguridad. Además, estos tiempos de ex-
posición pueden reducirse aún más al aumentar el valor
ISO o la apertura de la lente de la cámara utilizada para
la adquisición de imágenes.
Aplicaciones futuras
El método propuesto permite la identicación de
materiales luminiscentes UV como los que se pueden
encontrar en instrumentos musicales históricos. El pro-
cedimiento puede ajustarse de acuerdo con diferentes
respuestas de material a fuentes de excitación UV selec-
cionadas. El método proporcionará datos útiles para los
conservadores y cientícos de la conservación que ayu-
darán a la detección preliminar de barnices y retoques
en instrumentos musicales históricos, así como en otros
objetos de arte barnizados. Después de experimentos
adicionales con el método, el procedimiento se aplicará
en futuras investigaciones a estudios de casos reales, en
particular a violines históricos de colecciones de museos
como el del Museo del Violino en Cremona.
The tests carried out for this research have
shown that the exposure levels are somewhat
above the known levels advised for continuous
UV exposure of heritage objects. However, as only
short exposure times are used for multiband UV, the
total dose received by the objects is low. Therefore,
the method can be considered to be safe for bowed
instruments. The short exposure time, varying from
1 s to 30s depending on the UV source intensity
emission and the specific transmittance of the band
pass filters, ensures that the total amount of UV ra-
diation received by the object is far below the safe-
ty levels. Additionally, these exposure times, can be
even more reduced by increasing the ISO value or
the lens’s aperture of the camera used for the image
Future applications
The proposed method allows the identification
of UV-luminescent materials such as those that can
be encountered on historic musical instruments.
The procedure can be finetuned according to dif-
ferent material responses to selected UV excitation
sources. It will provide useful data for conservation
scientists and conservators that will help the pre-
liminary detection of varnishes and retouches on
historical musical instruments, as well as on other
varnished artworks. After further experiments with
the method, in future research the procedure will
be applied to real case studies, in particular to his-
toric violins of museum collections such as that of
the Museo del Violino in Cremona.
References | Bibliografía
Cavaleri, T., Gargano, M., Pisani, M., Zucco, M., and Buscaglia, P. (2017). 300-400nm UV uorescence multispectral imaging for
organic pictorial materials identication. In Carrero, J. A., Fernández, L. A., Gómez, O., and Gómez, L (Ed), Book of Abstracts.
Technart 2017. Non-destructive and microanalytical techniques in art and cultural heritage, pp.110. Bilbao, España: Universidad
del País Vasco.
UV-Vis Luminescence Imaging Techniques |
Técnicas de imagen de luminescencia UV-Vis
Cavaleri, T., Gargano, M., Pisani, M., and Zucco, M. (2018). Imaging of visible uorescence induced by multispectral UV excitation for
organic materials identication in works of art. In, 3rd International Conference on Innovation in Art Research and Technology - INART
2018, Parma, Italy.
Dondi, P., Lombardi, L., Invernizzi, C., Rovetta, T., Malagodi, M., and Licchelli, M. (2017). Automatic analysis of UV-induced uorescence
imagery of historical violins. Journal on Computing and Cultural Heritage, 10(2), pp.1-13.
Echard, J.P., and Lavedrine, B. (2008). Review on the characterisation of ancient stringed musical instruments varnishes
and implementation of an analytical strategy. Journal of Cultural Heritage, 9(4), pp. 420-429.
Fichera, G.V., Albano, M., Fiocco, G., Invernizzi, C., Licchelli, M., Malagodi, M., and Rovetta, T. (2018). Innovative monitoring plan
for the preventive conservation of historical musical instruments. Journal of Studies in Conservation, 63(1), pp. 351-354.
Fiocco, G., Rovetta, T., Gulmini, M., Piccirillo, A., Canevari, C., Licchelli, M., and Malagodi, M. (2018). Approaches for detecting
madder lake in multi-layered coating systems of historical bowed string instruments. Journal of Coatings, 8(5), pp. 171-187.
Goli, G., Fioravanti, M., Busoni, S., Carlson, B., and Mazzanti, P. (2012). Measurement and modelling of mass and dimensional
variations of historic violins subjected to thermo-hygrometric variations: The case study of the Guarneri “del Gesù” violin (1743)
known as the “Cannone”. Journal of Cultural Heritage, 13(3), pp.145-153.
Gulmini, M., Idone, A., Diana, E., Gastaldi, D., Vaudan, D., and Aceto, M. (2013). Identication of dyestuffs in historical textiles: strong
and weak points of a non-invasive approach. Dyes and Pigments, 98(1), pp.136-145.
Invernizzi, C., Fichera, G.V., Licchelli, M., and Malagodi, M. (2018). A non-invasive stratigraphic study by reection FT-IR
spectroscopy and UV-induced uorescence technique: The case of historical violins. Microchemical Journal, 138, pp. 273-281.
MIBACT (2000). Atto di indirizzo sui criteri tecnico-scientici e sugli standard di funzionamento e sviluppo dei musei. Elaborati del Gruppo
di lavoro. (D.M. 25.7.2000)
Miliani, C., Romani, A., and Favaro, G. (1998). A spectrophotometric and uorimetric study of some anthraquinoid and indigoid
colorants used in artistic paintings. Journal of Spectrochimica Acta Part A, 54, pp. 581–588.
Rovetta, T., Invernizzi, C., Licchelli, M., Cacciatori, F., and Malagodi, M. (2018), The elemental composition of Stradivari’s
musical instruments: new results through non-invasive EDXRF analysis. Journal of X-Ray Spectrometry, 47(2), pp. 159-170.
Spinella, A., Malagodi, M., Saladino, M.L., Weththimuni, M.L., Caponetti, E., and Licchelli, M. (2017). A step forward in disclosing the
secret of Stradivari’s varnish by NMR spectroscopy, Journal of Polymer Science, Part A: Polymer Chemistry, 55(23), pp.3949–3954.
Tirat, S., Degano, I., Echard, J.P., Lattuati-Derieux, A., Lluveras-Tenorio, A., Marie, A., Serfaty, S., and Le Huerou, J.Y. (2016). Historical
linseed oil/colophony varnishes formulations: Study of their molecular composition with micro-chemical chromatographic
techniques. Microchemical Journal, 126, pp.200–213.
Weththimuni, M.L., Canevari, C., Legnani, A., Licchelli, M., Malagodi, M., Ricca, M., and Zefro, A. (2016). Experimental characterization
of oil-colophony varnishes: A preliminary study. International Journal of Conservation Science, 7(2), pp.813–826.
... A soiling mixture was prepared and then applied following the method proposed by Wolbers. Subsequent cleaning tests were performed following-as well as reviewing-the methodologies commonly used in other art conservation fields, in particular that of easel paintings [8][9][10]. The aim of the work was testing some already-known cleaning methods on bowed string musical instrument surface and developing an analytical method to monitor the cleaning results. ...
Full-text available
Restoration and conservation procedures for historical musical instruments involve several issues, also connected with their frequent being played. One of the most delicate procedures for their preservation is the cleaning of surfaces from soil and dirt which have accumulated over the years. In fact, when external particles reach the surface, a fraction of them can deposit on it. Moreover, the contact with the player can generate chemical-physical changes, rapidly warming and wetting the surfaces through sweat deposition. This work focused on the cleaning methods of surfaces of bowed string musical instruments by a systematic and analytical approach. The selective cleaning procedure of varnished surfaces from grime and soil needs to be performed without compromising the original matter. Therefore, a dirty surface was reproduced on a set of varnished mock-ups and different water-based cleaning systems—generally used by restorers—were tested. The procedures were monitored in each step with several analytical methods: multispectral imaging (near-infrared (NIR), IRFC, visible imaging (VIS), UV-induced visible fluorescence), stereomicroscopy, XRF and FTIR spectroscopies allowed us to non-invasively outline the cleaning system efficacy. The results highlighted different levels of cleaning and, in some cases, it was possible to identify the best selectivity for the different procedures.
Defendente Ferrari, one of the most important painters of the Piedmontese Renaissance, was extremely active in the early 16th century. His Madonna with Child and two Saints (1505–1510 ca.), which was originally located in the Collegiate Church of Carmagnola (Turin province, Italy) now in the collections of Palazzo Madama - Museo Civico d’Arte Antica in Turin, was recently analyzed and restored at the Centro Conservazione e Restauro La Venaria Reale, as part of a study and conservation project focused on the artist and his circle. This study presents an interdisciplinary and multi-analytical archaeometric approach resulting from the collaboration of several universities and research centers. The adopted protocol allowed for clarifying some peculiarities of the artist's painting technique, such as the use of the dotted - a technique applied to render the nuances, volume, and flesh tones of the figures - and that of thin black cross-hatched strokes, drawn to shape the clothes and to define the related volumes and folds. Moreover, specific degradation processes were studied, which apparently occur also in other contemporary artworks with similar features. Of particular interest is the alteration of the blue color of the Virgin's garment, made of azurite and presently strongly blackened. SWIR reflectography performed with a high-resolution InGaAs scanner prototype allowed for imaging both the detailed underdrawing and the careful hatching used to highlight the shadows. The combined application of optical microscopy, SEM-EDS, µ-Raman, FT-IR spectroscopy, and GC–MS has been fundamental for analyzing some specific pigments, binders, and varnishes, as well as for inferring the cause for Virgin's garment blackening. A complete mapping of the pigments was provided through the combination of FORS (350–2000 nm), portable µ-Raman, MA-XRF, and Hypercolorimetric Multispectral Imaging. The latter also allowed for observing the distribution of what is supposed to be either the original or the ancient organic finishing layer, which was maintained during the conservation treatment. This research has brought to an in-depth knowledge of the artist's painting technique and helped the adoption of the best choices for an effective and minimal-invasive restoration intervention.
Full-text available
Musical instrument coatings are generally made by multi-layered systems of organic and inorganic materials, applied on the wood substrate by the violin makers during the finishing process. This coating has paramount relevance for several aspects: protection from sweat and dirt, increase of specific acoustic features, and especially aesthetic effects. In fact, the colour of historical bowed string instruments represents a very peculiar characteristic of each workshop. Among the various colourants, lakes are the most challenging to detect because of their sensibility to the alteration processes. In this work, non-invasive and micro-invasive procedures were applied to a set of mock-ups mimicking historical coatings systems prior and after artificial ageing, in order to highlight the overall information that can be recovered for the detection of madder lake in historical bowed instruments. A set of techniques, including colourimetry, visible and UV-light imaging, stereomicroscopy, Fibre Optics Diffuse Reflectance spectroscopy (FORS), X-ray Fluorescence spectroscopy (XRF), Scanning Electron Microscopy coupled with Energy-Dispersive X-ray microprobe (SEM-EDX), and Fourier-Transform Infrared spectroscopy (FTIR) were used in order to evaluate the pros and cons in the detection of organic and inorganic component of madder lake at low concentration levels.
Full-text available
It is commonly thought that the varnishes used by the great violin-maker Antonio Stradivari may have a role in determining not only the esthetical features but also the acoustic properties of his instruments, and the idea of a “lost secret” is still widespread among musicians and violin-makers. Previous scientific researches on varnish samples of Stradivari's instruments revealed that they were generally made by a mix of linseed oil with and colophony or metal rosinates in different ratios ranging between 75/25 (oil/resin) and 60/40 (oil/rosinate). However, it is still not clear whether the mixture composition can be related to any structural and/or functional feature of the resulting varnish. To investigate this aspect, we prepared varnishes with different linseed oil/colophony (w/w) ratios and applied NMR techniques to achieve information about their chemical–physical characteristics. Here, we show that the two components strongly interact in the solid state and that only the varnish prepared from 75/25 (w/w) linseed oil/colophony mixture displays unique properties in terms of dynamic homogeneity unlike coatings with other compositions. Our results suggest that the so-called “secret” of Stradivari's finish could not be related to unknown ingredient(s) but to a specific oil/resin composition that provides the best performance. © 2017 Wiley Periodicals, Inc. J. Polym. Sci., Part A: Polym. Chem. 2017
Full-text available
Historically, the varnishes had the aim to protect the bowed musical instruments by the external agents and to confer them an aesthetic value. During the 17th and 18th century, in Italy, the bowed instruments, especially violins, were generally covered by a layer of varnish made with several natural materials such resins, oil or hide glue: i.e., instruments by the great violin maker Antonio Stradivari were covered often with a layer of varnish made of linseed oil and colophony in the ratio 3:1, respectively. The main aim of this work was to study the modifications that occur in those kinds of varnishes, after exposing them to some factors of degradation. In order to study the different properties of organic coatings and their suitable compositions, different mixtures of linseed oil and colophony were recreated in the laboratory following an ancient recipe: linseed oil and colophony were mixed together with different ratios (50/50 and 75/25, respectively) and then, they were applied on Maple wood samples and on glass slides for experimental purposes. In order to investigate the different external factors which cause the varnish layer degradation, samples were analyzed by different techniques before and after different ageing processes (thermo-hygrometric cycles, exposition to UV lamp and to acid vapors). Out of strong experimental evaluation, all the results suggested that the composition of 75/25 (oil: colophony) is much better as a varnish for musical instruments.
The Cultural Heritage preservation, by means of preventive actions aimed to reduce the risk of deterioration and to protect the constituent materials, implicates several difficulties for the conservators. The musical instruments can represent a particular class of artworks, because their conservation is not limited to the material preservation, but also to the original function of the object (produce sounds).Some historic instruments, held in the collection of Museo del Violino in Cremona, were selected in order to evaluate the suitability of the non-invasive analytical techniques to monitor the ongoing material transformation. The historic violins ‘Vesuvio’ 1727c. by Antonio Stradivari and ‘Carlo IX’ 1566c. by Andrea Amati were selected for their higher and lower use in performances over the years 2016/2017. Moreover, the cello ‘ex-Cristiani’ 1700 by Antonio Stradivari was analysed after a concert with the aim to evaluate any modification of the varnishes
In the Cultural Heritage field, stratigraphic analysis is commonly carried out by applying microscopic and micro-analytical techniques to cross-sections of micro-samples taken from artwork. This methodology provides the most comprehensive information on the object's physical structure, but necessarily implies the loss of precious or unique fragments. For this reason, sampling is rarely feasible on musical instruments having such a historical and commercial value as Stradivari or Guarneri violins. Having been played over the centuries, these masterpieces have undergone different kinds of degradation and wear, and subsequent invasive restorations. Varnish layers have therefore been thinned or even removed from the surface and other materials could have been added overlaying the original ones, with consequent variations in coating thickness and composition over time. In the present work, a novel non-destructive and non-invasive methodology combined UV-Induced Fluorescence (UVIFL) photography and reflection FT-IR spectroscopy to investigate how and to what extent varnish thicknesses vary on the surface of historical violins as well as to reconstruct the stratigraphic layer order and chemical composition. Firstly, UV-induced variations and IR diagnostic bands were identified on numerous model samples reproducing musical instrument stratigraphy. Then the methodology was applied to the study of several Stradivari and Guarneri violins and its limitations and strengths were showed in three representative case studies.
During recent decades, many researchers have tried to understand the main influences on the extraordinary sound and beauty of the masterpieces made by the ancient violin makers. This is still a challenge for many others today. Mainly because of a lack of written historical documents, the rediscovery of some of the ancient violin-making processes was made possible thanks to scientific analyses performed on their materials by means of diagnostic techniques. However, understanding which substances were adopted is a very hard task, because the analyses are influenced by many factors: for example, alterations, wear, retouches, and the heterogeneity of materials. This paper presents some new EDXRF results collected on eight-stringed musical instruments made by Antonio Stradivari between 1669 and 1734 (“Clisbee” 1669, “Hellier” 1679, “Ford-Rougemont” 1703, “Joachim-Ma” 1714, “Russian Federation” viola 1715, “Cremonese” 1715, “Vesuvius” 1727, and “Scotland” 1734) and now preserved at the Museo del Violino in Cremona. A brief comparison with a modern violin made by Simone Ferdinando Sacconi (“Hellier copy” 1941), one of the most eminent violin makers of the 20th century and one of the greatest experts on Stradivari's work, is also provided. This represents the first comparative analysis of a wide number of ancient musical instruments made by the same violin maker over an extended period. A non-destructive and non-invasive approach was followed to (a) understand the elemental composition differences between the best conserved and most worn-out surfaces; (b) check if there are elemental similarities among the finishing materials of violins made in different years by the same violin maker; (c) give new suggestions about the materials used. To distinguish the best conserved areas from the worn-out ones, a preliminary investigation by UV-induced fluorescence photography was performed. In addition, stereomicroscopic observations and Fourier transform infrared spectroscopy (FTIR) analyses were performed on selected areas to validate the hypotheses. The results, in some cases comparable with previous research on Stradivari instruments, have increased the pool of information about materials and treatments adopted in the Stradivari workshop.
In recent years, UV-induced fluorescence (UVIFL) photography has proven to be very effective when studying the surface of historical musical instruments, such as violins. This technique makes it possible to highlight superficial details not clearly perceptible with visible light (e.g., retouchings, superficial distribution of varnishes, or wear). The data retrieved are also an important guide for further noninvasive spectroscopic analyses used when the chemical composition of the surface needs to be investigated. However, UVIFL imagery interpretation of a historical violin is no trivial task. In fact, constant playing and the multiple restorations over the centuries have produced very complex surfaces. This work presents an automatic tool designed to facilitate this kind of analysis. Using a quantized histogram in HSV color space, the distribution of the main fluorescence colors on an instrument’s surface can be highlighted, recurrence of the same color in different areas of the same violin can be detected, or different violins can be compared. UVIFL images of seven Stradivarius violins kept in the Museo del Violino in Cremona, Italy, were used as a test set. The results achieved endorse the validity of the proposed approach.
This paper presents a study regarding the hygro-thermal conditions to which the violin Guarneri “del Gesù” (1743), known as the “Cannone”, is subjected during its conservation and occasional use in concerts with special attention on its mass and dimensional variations. Several environmental measurement campaigns were planned and carried out using relative humidity and temperature probes. The violin mass variation was measured continuously inside the display case where it is conserved, and before and after concerts by means of a special exhibition frame integrating a precision balance. These measurements enabled reproducing the thermal and hygrometric variations to which the violin is normally subjected using a purposely-developed portable climatic chamber, and also enabled measuring the consequent hygroscopic and thermal deformations in selected points by means of a purposely-developed measuring frame. An empirical model for computing the mass variations according to the variation of environmental conditions was implemented and verified and the typical mass variation consequent to the use of the violin during concerts was also determined. The violin's thermal and hygroscopic deformations were measured in selected points for given temperature and relative humidity steps. The paper includes a discussion about the possible impact of hygro-thermal variations on violin conservation.