Dusts on Busts

Dusts on Busts

Dust on Plaster Surfaces:

Focussing on the Portrait Busts in the Rococo Hall of the

Duchess Anna Amalia Library in Weimar

An M.A. Preventive Conservation Dissertation presented to the

Department of Arts & Social Sciences, Northumbria University at Newcastle

Aurelia Badde

Student Number: 08019847

CV0704

September 2009

Tutor: A. Jean E. Brown Programme Leader

Preventive Conservation Burt Hall

Northumbria University Newcastle upon Tyne

NE1 8ST Tel: 0191 227 3331 Mobile: 0777 597 4708

3

Contents

Abstract ............................................................................................................................................... 5

Kurzbeschreibung ................................................................................................................................ 5

Introduction ......................................................................................................................................... 6

Methods ............................................................................................................................................... 7

Chapter 1 ................................................................................................................................................. 9

Dust and Plaster .................................................................................................................................. 9

Plaster as a Statuary Material .......................................................................................................... 14

Plaster’s Properties ........................................................................................................................... 18

Dust on Plaster Surfaces ................................................................................................................... 21

Cleaning Plaster Surfaces ................................................................................................................. 28

Chapter 2 ............................................................................................................................................... 32

Case Study – The Plaster Portrait Busts in the Duchess Anna Amalia Library ................................ 32

A Brief History of the Duchess Anna Amalia Library and Its Art Collection ................................... 34

The Portrait Busts in the Duchess Anna Amalia Library .................................................................. 38

On the Sculptors, Materials and Provenances of the Portrait Busts ................................................. 44

On the History of Maintenance of the Plaster Busts ......................................................................... 47

Experimental 1: Analysis of Cross-Sections of Paint-layers taken from Five Busts Selected by Criteria of their Accession Date to the Collection ....................................................................... 50

Excursion:.............................................................................................................................................. 63

Coating Plaster Surfaces – a German Tradition in Preventive Practice? ........................................ 63

Survey: ................................................................................................................................................... 71

Questionnaire on Maintenance Concepts and Dry Cleaning Methods for Plaster Surfaces on Display .............................................................................................................................................. 71

4

Chapter 3 ............................................................................................................................................... 83

Planning Preventive Care and Maintenance for the Plaster Busts in the Rococo Hall .................... 83

Climatic Conditions in the Rococo Hall ........................................................................................... 84

Experimental 2: Evaluation of Clima as Recorded by Six Loggers for T and RH ...................... 85

Dust Impact on the Plaster Busts of the Rococo Hall ....................................................................... 88

Experimental 3: Dust Monitoring – the Distribution Pattern within the Rococo Hall .............. 98

Experimental 4: Microscopic and Image Analysis of Sampled Dust ....................................... 100

Experimental 5: Dust Analysis with XRF and FTIR and Chromatographic Methods .............. 107

Evaluating a Maintenance Plan for the Busts in the Rococo Hall .................................................. 109

Experimental 6: Vacuum Cleaning the Plaster Surfaces ........................................................... 110

Conclusions .......................................................................................................................................... 113

An explanation of why the research has made a valuable contribution to Preventive Conservation ... 115

Thanks .................................................................................................................................................. 116

References ............................................................................................................................................. 117

Online Resources ................................................................................................................................. 128

List of Illustrations .............................................................................................................................. 130

Picture Credits ..................................................................................................................................... 132

List of Appendices ................................................................................................................................ 132

Appendix I, Über die Behandlung von Gipsabgüssen behufs deren Erhaltung ................................... 133

Appendix II, Lab. Report Erfurt........................................................................................................... 139

Appendix III, Lab. Report Hildesheim ................................................................................................. 176

5

Abstract The conservation and reinstallation of the plaster portrait busts in the Rococo Hall of the Duchess Anna Amalia Library in Weimar calls for a preventive care and maintenance concept. Former tradition in their maintenance was evident by various superimposed layers of protective light-coloured coatings that were applied to the plaster surfaces. The removal of these paint layers during restoration and conservation of the busts, was largely prompted by the damage they had experienced by fire extinguishing water in 2004. Comparative material analysis identified the paint layers as secondary, non-original measures for “whitening” soiled or otherwise damaged surfaces. The tradition of applying layers of paint or other coating materials onto plaster surfaces is one focus of investigation. Research into this tradition, as well as a survey into other plaster collections’ maintenance practices, places these surfaces in Weimar into a broader context.

Investigation and analysis of the dust impact on the plaster surfaces in the Rococo Hall is carried out with consideration to the environmental parameters. Research into different preventive approaches for maintenance of plaster statuary, as well as a survey on dry cleaning methods for plaster, contributes to the evaluation of an appropriate methodology in context to the plaster busts’ installation in the Rococo Hall. A practical dry cleaning trial assists in establishing a cleaning methodology as well as a routine for the care of these now uncoated plaster surfaces.

Kurzbeschreibung Mit der Wiedereinrichtung der restaurierten Gipsbüsten im Rokokosaal der Herzogin Anna Amalia Bibliothek, stellt sich die Frage nach dem Umgang mit unbeschichteten Gipsoberflächen in Bezug auf ihre Pflege und nachhaltige Erhaltung. Die Tradition in der vormaligen präventiven Praxis war an Abfolgen von hellen Farbanstrichen auf den Gipsoberflächen der Büsten im Rokokosaal nachzuweisen. Diese Farbbeschichtungen hatten im Jahr 2004 im Zuge der Löscharbeiten Schaden genommen und wurden während der Restaurierungen der Büsten entfernt. Ein analytisch unternommener „Querschnitt“ durch die Übermalungspraxis an den im Rokokosaal ausgestellten Gipsbüsten erfasst eine über zweihundertjährige Sammlungs-, Aufstellungs- und Behandlungsgeschichte und weist die ursprünglich materialsichtige Aufstellung der Gipsbüsten nach. Das Phänomen Farbanstriche auf Gips stellt einen Focus dieser Arbeit dar. Mittels einer Umfrage wird es in historischen, auch in internationalen Kontext gesetzt und kann für die Zeit ab dem späteren 19. Jh. als durchaus üblich betrachtet werden – auch als ästhetische Maßnahme, an diesem schmutzanfälligen und allgemein als minderwertig erachtetem Material. Die Umfrage evaluiert ferner den heutigen konservatorischen Umgang mit ungefassten Gipsoberflächen; verschiedene Reinigungsverfahren für Gipsoberflächen werden mit direktem Bezug zu der Aufstellungssituation der Büsten im Rokokosaal bewertet. Die Konzepterstellung für die Pflege und nachhaltige Erhaltung der Gipsoberflächen in ihrer jetzigen Aufstellung im sanierten Rokokosaal, wird anhand eines Klima-Monitorings und einiger Feldversuche zur Einschätzung der Staubbelastung auf den Gipsoberflächen, zur Staubverteilung im Saal, wie auch durch verschiedene an Staubproben unternommenen Analysen, erarbeitet. Die Interaktion von Staub auf ungefasstem Gips wird in dieser Arbeit besprochen. Die Untersuchung stellt die aufgrund ihrer Positionierung im Saal besonders belasteten Gipsoberflächen heraus. Praktische Trockenreinigungs-proben tragen mit bei zur Etablierung einer anzuwendenden Reinigungsmethodik für die Gipsoberflächen im Rokokosaal.

6

Introduction

The Duchess Anna Amalia Library suffered a fire disaster on September 2nd, 2004.The conservation

and restoration of a total of 65 plaster portrait busts that were installed in the Rococo Hall at the time,

was carried through in 2005 to 2007, after thorough drying of the plaster busts. The questions that

arose during damage assessment and while conceptualizing the conservation plan, instigated this

study, which deals with the surfaces of these plaster busts and assesses methods for their preventive

care.

Most of the surfaces of the plaster busts situated in the Rococo Hall of the Duchess Anna Amalia

Library, were covered with various layers of paint. This, in Germany, is a fairly common phenomenon

in the treatment of historical plaster surfaces in reaction to dust accumulation. Fire extinguishing

water however, caused a large extent of damage to the plaster busts. The paint layers, although they

themselves disintegrated to some extent, acted in many areas as protective shields against water and

soot and other particles that became encrusted to the surfaces.

The reinstallation of the plaster portrait busts in the Rococo Hall of the Duchess Anna Amalia Library

after their restoration calls for a preventive care and maintenance concept. The now uncoated plaster

surfaces have experienced 22 months of dust-fall since reinstallation. The environment and its impact

on the plaster surfaces therefore require investigation. Plasters materiality and its vulnerabilities will

be characterized and plaster cleaning will be discussed in this study, with the aim of evaluating a dry

cleaning methodology for the plaster surfaces in the Rococo Hall.

Continuing discussion over re-application of a surface protection film stands to reason and would

maintain the tradition in surface treatments to these busts. The motivation for applying coatings of

paint onto plaster, can to some extent be interpreted as preventive, in that the hydrophilic and highly

porous and soft plaster is thus shielded from its immediate environment. Yet the sheer amount of

superimposed layers, suggests that these coated surfaces in the Library were not necessarily more

manageable in terms of cleaning. Their removal disclosed soiling on the plaster surfaces. The off-white

shades of colouring demonstrate the historical appreciation of light coloured portrait busts in the

Rococo Hall. Analysis of the paint layers ascertained that their application was also corrective

inasmuch as they hid surface soiling and damage histories. This particular tradition in treatment

practice to plaster surfaces will be explored and compared with other viable concepts of preventive

care and surface maintenance for plaster statuary.

7

Methods

Published literature on plaster cleaning is not comprehensive; the literature research on plaster

cleaning methods therefore also focuses on unpublished studies and dissertation theses and aims at

giving an overview of methods and approaches.

Surveying coating practices, maintenance concepts and dry cleaning methods, aims at gaining a

broader, international view into maintenance concepts for plaster statuary. A wide participation in the

survey is sought by posting a questionnaire on conservation mailing lists and also by approaching cast

collections and specialized conservators individually. The questionnaire seeks parallels to the practice

of coating plaster statuary outside of Germany. It also seeks information, which is relevant especially

for the appraisal of this genre of sculpture – plaster portrait busts – which was typical and very widely

spread in the neo-classical era and its wake.

Analysis is carried out on cross-sections of paint-layers taken from five busts, selected by criteria of

their accession date to the collection. The number of coatings is compared and the respective binding

media determined by gas-pyrolysis chromatographic and mass-spectrometric methods. Pigment

analysis is achieved with X-ray and infrared methods of spectometry. Analysis of five cross-sections

gives material evidence, which will represent a probe into more than 200 years of one collections

history and a maintenance practice that is documented for hundreds of plaster busts in this collection.

The understanding of this treatment history is realized by direct comparison to records and to

experiences of maintenance to plaster statuary and is achieved by way of literature research and the

survey.

The general distribution of dust settlement on the plaster busts, since the re-opening of the Duchess

Anna Amalia Library to the public in October 2007, is established by naked eye inspection, as well as

photographic recording and subsequent comparative sampling. A review of literature on methods of

dust sampling and dust analysis in the cultural heritage sector assists toward determining appropriate

analytical methods.

The pattern of distribution is influenced by the air movements within the two floors of the hall, which

in turn are dictated by the air ventilation system. Microscopic inspection of the samples and image

analysis with subsequent statistical evaluation give information on the particle-size distribution

curves. Comparison of distribution curves of different samples from different locations within the

library hall, gives some indications on the relation between airflow and particle size distribution. The

material composition of the sampled dust particles is analyzed with methods of X-ray-fluorescence and

infrared spectroscopy as well as gas-pyrolysis chromatography. Climate measurement and statistical

evaluation of their recordings defines the environmental parameters under which dust particles collect

onto the plaster surfaces.

Finding a suitable method and a routine for cleaning the plaster busts displayed in the Rococo Hall of

8

the Duchess Anna Amalia Library is the practical aim of this study. A trial is carried through on some

busts, selected by the rate of their respective soiling, aiming at evaluating the feasibility as well as

operating expense. The efficacy of this cleaning methodology gives some indication toward the

respective adherence of particle matter to the plaster surfaces. This in turn indicates whether the

climate values in the Rococo Hall, specifically the RH, are suitable for bare plaster surfaces in the

presence of particle matter. A cleaning routine can be established in reference to the respective

locations of the plaster busts within the hall.

9

Chapter 1

Dust and Plaster

“Think of me not in grief, but cheerfully. – Keep the plasters clean, as that is the main thing.”1

Wilhelm von Humboldt (1767-1835), diplomat, philosopher, linguist, educationalist and university

founder, was probably referring to the plaster casts of famous ancient sculpture he had collected for his

family home, Tegel Palace,2 when on his deathbed he spoke these words to his family, on April 2nd

1835. It is perhaps the most poignant reference to be found on the notorious problem of plaster

soiling and it represents a general concern for the effects of the environment on bare plaster surfaces.

Humboldt’s remark is interpreted as the appeal of an entire generation, directed at it’s posterity, to

safeguard and cherish the material and cultural legacy of German neoclassicism.3

Figure 1: Adolph Menzel’s pastel drawing ‘Plaster Model Storeroom in the Altes Museum’ of 1848 depicts

the storage situation of the cast collection during the construction of the Neues Museum.

1 “Gedenket meiner nicht in Trauer, sondern in Heiterkeit. – Haltet mir die Gipse rein, denn das ist die

Hauptsache.” quoted from Gabriele von Bülow, in: Tochter Wilhelm von Humboldts. Ein Lebensbild. Aus den Familienpapieren Wilhelm von Humboldts und seiner Kinder 1791-1887, 4th edition, Berlin 1894, p. 348. All following English translations of German text passages are the authors, unless otherwise cited.

2 See: http://www.monumente-online.de/07/05/streiflichter/08_Humboldtschloesschen.php; see also Paul Ortwin Rave.

3 Bernhard Maaz 1993, p. 14; Frank Matthias Kammel 2001, p. 66.

10

Both Wilhelm von Humboldt’s own collection of casts as well as his prominent role in the organisation

and increase of the Berlin academy’s and the museums’ plaster casts collections,4 can serve as

examples for a very pronounced trend involving an intense propagation of plaster statuary.5 The

history of an inflationary production and distribution of pre-modern and modern era reproductions of

Greek and Roman sculpture, throughout the late 17th, the 18th and most especially the 19th century,

fills numerous volumes.6 This was a period in time, in which every discipline of the arts studied,

sketched and modelled after life-casts in plaster7 and after plaster casts of ancient sculpture. Style and

imagery of ancient sculpture dominated the contemporary sculpture and a vivid interest and

admiration for classical styles and forms became a solid educational ideal.8 Cast-galleries were installed

in museums and within private collections throughout Europe.9 Schools, universities and academies

collected and literally amassed plaster casts.10 These casts – primarily replicas of ancient statuary and

then later on increasingly also of medieval, modern era and contemporary sculpture – either filled

entire halls or were placed in the spaces in between the original sculpture, with encyclopaedic yet cost-

effective zeal.11

4 Gertrud Platz-Horster, pp. 276-277. For the history of the Berlin cast collections see also: Bernhard Maaz

1993, pp. 30-41; Sibylle Einholz 1991, 1992 and 1996; Veit Stürmer, Henning Wrede. See Frank Matthias Kammel 1992, for the history of the Berlin cast collection of medieval sculpture.

5 Siegrid Düll.

6 Francis Haskell and Nicholas Penny’s Taste and the Antique still being one of the most acknowledged publications on the subject, see especially pp. 31-42, pp. 79-91 and pp. 117-124.

7 Life-casting as well as casting death masks has a tradition in our cultural history which reaches into Egyptian and Ethiopian antiquity, Nicholas Penny, pp. 194-196; see also Bettina Uppenkamp, pp. 138-142 or Frank Matthias Kammel, 2001, pp. 55-56.

8 Bernhard Maaz 1993, pp. 13-16; Heinz Ladendorf; Peter Bloch.

9 See Thomas Lochmann and Klaus Stemmer 2000, pp. 228-229, for a brief overview of the early development of some cast collections out of the art cabinets into academies: i.e. the French Academy in Rome in 1666, the cast collection of the Academy of Arts in Berlin in 1695, the collection of casts at the court of Düsseldorf around 1700, later forming the famous Mannheimer Antikensaal, in 1803 moving into the Munich Academy. The collection in the Academia Real in Madrid was formed in 1761 and in Dresden in 1782, both from Anton Raphael Mengs vast collection of roman casts, see Thomas Kidderlen. Further see for example Donna Kurtz or John Kenworthy-Browne for portrayals of English collection histories of original sculpture and casts; also Frank Matthias Kammel 2001, pp. 52-55 for a broader view on the European development of plaster cast collections.

10 See Adolf H. Borbein, pp. 32-33: Earliest cast collections of German universities are in Göttingen, 1767 and Bonn, 1819. The website of the International Association for the Conservation and the Promotion of Plaster Cast Collections, http://www.plastercastcollection.org/en/database.php, lists all institutions by country that are relevant as plaster cast collections. Each collection is briefly described and the respective founding year is often indicated.

11 Bettina Uppenkamp pp. 146-147. The tremendous mercantile aspect of this extremely busy trade of casts is described by Hans-Ullrich Cain, Adolf H. Borbein and Gerhard Rupp, pp.337-341.

11

Plaster casts were also very present in other fields of education, in anatomy and anthropology,

palaeontology, phrenology and ethnology, and largely also in architecture.12 In an age where travelling

and tourism was reserved for the very privileged, and photographic reproduction could not yet

communicate an impression of an object, plaster casts served as – in Johann Wolfgang von Goethe’s

words – the “truest facsimiles”.13 And even though Goethe’s and Humboldt’s collections of casts were

both famously outstanding, the 19th century saw plaster busts and statues settling onto pediments,

mantelpieces, chests and bookshelves in many private homes.14 Any household priding itself on

education would acquire cast copies of famous ancient and Renaissance sculpture or of contemporary

busts of distinguished scholars.15

Sculptors would manufacture, show and sell their works in plaster editions, as this provided a constant

flow of income, whereas the much more laborious and cost-intensive bronze-casting or marble

working of a sculpture mostly occurred upon order.16 The intrinsic and sustainable value of an original

work of art in marble or in bronze could however never be replaced by a plaster cast. While some casts

would keep or later take on the quality and aura of an original and valuable work – for example original

sculptor models or when the original sculpture it represents suffered deterioration or destruction17 or

by its collection context, as in the case of the plaster busts in the Rococo Hall – the great majority of

plaster casts were serial mass-reproductions in a comparatively very cheap and easily transportable

and replaceable material.18

While 18th century form-idealism had followed Johann Joachim Winckelmann (1717 -1767) in

12 Frank Mathias Kammel 2001, pp. 55-56.

13 “gute Gypsabgüsse, als die eigentlichsten Facsimiles”, see Johann Wolfgang von Goethe, Weimar Edition, Sophien-Ausgabe, Weimar 1887-1912, Department I, Vol. 32, p. 322 (WA, I, 32, p. 322). See Klaus Stemmer 1999, p. 9 and 2000, p. 230 on the lasting impact Goethe played on the reception of antiquity in Germany. See Gabriele Oswald 2007, on Goethe’s art and cast collection.

14 Sibylle Einholz 1993, p. 16, quotes from Adolf Glaßbrenner’s Schilderungen aus dem Berliner Volksleben of 1841, p. 48: “The Italian plaster-figure tradesman carries on his head a long wooden board, on which stand the busts of Schiller and Goethe, of Napoleon, Blücher, crowned plaster heads, praying children, a club-swinging Hercules, a midicaean Venus, a big dog, a famous statesman and various cats with moving heads, and calls out: figurica, beautiful figurica buy!” For German text see: http://opus.kobv.de/zlb/volltexte/2007/1252/pdf/GlaAbrenner_Volksleben_1.pdf

15 Hans-Ulrich Cain, pp. 207-208, Bettina Uppenkamp, p. 146.

16 Bernhard Maaz, loc. cit., Frank Matthias Kammel 2001, p. 51.

17 Sibylle Einholz, 1993, pp. 19-20, recounts some examples of plaster casts serving for reconstruction of 19th century marble sculpture, which was damaged in World War II. Cf. also Malcolm Baker http://www.vam.ac.uk/collections/sculpture/sculpture_features/cast_collection/cast_courts_masterpiece/index.html

18 Bernhard Maaz, loc. cit., Bettina Uppenkamp, loc. cit.

12

accepting a plaster cast as a precise replacement of an original – in fact, even appreciating plaster’s

unpretentious material qualities and its abstracting white colour as a distinct improvement over the

original - this had gradually started to change in the course of the later 19th century.19 The rejection of

plaster’s materiality followed in conjunction with a progressively obsolete academic tradition in

education, and also with the desire for original artwork and precious materials in exhibitions.20 Apart

from the dull-white colour of plaster, it was foremost its susceptibility towards dust which largely and

sustainably degraded the appreciation of this material.21 The so-called reverse shadow effect, which

plaster surfaces develop when soiling accumulates particularly on the protruding and horizontal areas

of a sculpture (see Fig. 8), was most heavily criticized.22 While some engaged in finding suitable

methods for (mass) cleaning and also for (mass) treating plaster statuary by impregnation, others

simply coated soiled plaster surfaces with paint.23 A controversial debate ensued over the question of

these over-paints amongst museum professionals and archaeologists: Some argued in favour of the

material imitating quality of an over-paint , which could produce the semblance of bronze, marble,

ivory, sandstone or other material.24 They also stressed the additional effect the over-paints had of

protecting the plaster surfaces25 and of creating surfaces resistant to cleaning.26 Others saw a moral

danger in such proceedings and begged the ethical question of an “honest plaster cast” with a surface

as “delicate as a raw egg”27 versus the “deceptive baublery” of an over-painted cast.28 In 1877 the Royal

Ministry for Education in Berlin then called together an expert group of 23 museum-directors,

chemists, sculptors and other authorities, to assay the methods for the preservation and treatment of

plaster casts employed in the Berlin museums as well as in other German collections. The Commission

19 Hans-Ulrich Cain, pp. 206-207. Bettina Uppenkamp, p. 147-148 and Frank Matthias Kammel 2001, p. 49.

20 Bettina Uppenkamp, loc. cit.

21 Frank Matthias Kammel 2001, p. 58.

22 Robert Daun, p. 194; Bettina Uppenkamp loc. cit.; see also Louise Cone p. 12.

23 Cf. note 29 and pp. 61-68. 24 Hans-Ulrich Cain, pp. 209-210. Frank Matthias Kammel 2001, pp. 58-62.

25 Robert Stiassny, p. 13.

26 Robert Daun, p. 194; Frank Matthias Kammel 2001, loc. cit.; Bettina Uppenkamp, p. 148.

27 “Ein Gipsabguß sei so zart wie ein rohes Ei”: archaeologist Kekulé von Stradonitz, quoted by Berthold Daun, p. 194.

28 Hans-Ulrich Cain, p. 210, Bettina Uppenkamp, p. 148 and Frank Mathias Kammel 2001, p. 59, all quote Botho Gräf, archaeologist in Jena, in: Museumskunde, Vol. 4, 1908, p. 64: “ein weißer Gipsabguß ist ehrlicher, ein bemalter eine täuschende Spielerei, sie kann vielleicht erfreulich in der Ausstattung von Wohngemächern wirken, in ein Museum gehört sie nicht.”

13

for Consultation on the Treatment and Conservation of Plaster Casts came together in a number of

sessions, with the aim of finding a suitable and widely acceptable aesthetic and maintenance approach

for plaster and its surface qualities.29

A marked and general depreciation, which plaster statuary effectively started to experience at the

beginning of the 20th century, is in fact attributed to plasters own unique material and reproducible

qualities.30 The hybrid quality of mixed material collections with plaster replicas next to original art

was now increasingly criticized and plaster casts stood for bad taste and were polemically villified.

Reproducibility was suddenly perceived as a defect, and the plaster casts began to move out of the

galleries.31 “Living art instead of plaster!”32 was the new credo of the new century, when the “mass

scrap of a mass production”33 progressively moved into storage and subsequently experienced great

neglect and damage, as well as vast losses.34 Twentieth century political history in Germany is then the

background for the removal of already largely depreciated plaster casts into the furthest corners, attics

and cellars - called storage. Or, even more simply, plaster casts were broken into pieces and disposed

of.35

The gradual but noticeable re-appraisal of plaster casts, which recommenced noticeably in the 1970s

of the 20th century, is often attributed to contemporary art’s exploration of this material since the

1960s.36 Technology, material history and diversity of plaster casts have just recently again been in the

29 Kommission zur Beratung über die Behandlung und Konservierung von Gipsabgüssen, Sonderabzug aus den

Verhandlungen des Vereins zur Beförderung des Gewerbefleisses, Berlin 1877, see below pp. 61-68 and cf. Apendix I for excerpt. Louise Cone 1996, pp. 15-19 and 2005, pp. 120-124, describes a commission, which established itself in Copenhagen in 1893/94 to define methods for impregnating and cleaning plaster. In 1888 the Thorvaldsens Museum had sent letters to various sculpture collections in Europe, asking them about their methods for keeping their plasters clean.

30 Sibylle Einholz 1993, p. 17, Hans-Ulrich Cain, p. 210.

31 Frank Matthias Kammel 2001, p. 49-58.

32 “Lebendige Kunst statt Gips” B. E. Werner in the Deutsche Allgemeine Zeitung, November 9th 1931, quoted by Frank Matthias Kammel, 2001, p. 48.

33 Sibylle Einholz 1996, p. 11: “’Massenschrott‘ vermeintlicher Vielfachproduktion in minderem und verachtetem Material”.

34 Hans-Ulrich Cain, p. 211; Bernhard Maaz 1993, p. 35.

35 Frank Matthias Kammel 2001, pp. 62-65, records the practice of destroying plaster casts in German Museums still in the 50s and 60s and even up to the 70s of the 20th century.

36 Frank Matthias Kammel 2001, p. 65; Bettina Uppenkamp, pp. 154-159; Hans-Ulrich Cain, p. 20; and see Malcolm Baker, http://amethyst.vam.ac.uk/collections/sculpture/sculpture_features/cast_collection/ cast_courts_masterpiece/index.html

14

focus of various symposia.37 Conditions and methods that can provide sustainable conservation for

plaster statuary need to be considered an integral issue within this ‘revival’ and are hence in the focus

of this study.

Plaster as a Statuary Material

“One does not consider enough what an invaluable material plaster is, that one can send sculpture’s best,

to some extent identically into the distance and, without affecting the first owner, can provide a second

ownership and a doubled pleasure.”38

It is possible to cast a block of plaster and to sculpt, carve and rasp into it perhaps much more easily

than into stone or wood,39 but this method is not a very common practice in sculpting. Modelling in

plaster occurs frequently, yet plaster’s most significant use is very clearly as a casting material within

the statuary processes of creation and reproduction.

A traditional method for creating statuary is by first modelling the original sculpture in a flexible

material, like clay, and then casting this model by means of a so-called waste mould40 or lost mould, -

which, as the name suggests, is destroyed in the process.41 When the plaster mould has set, the clay

model is scraped out of the mould and replaced by liquid plaster, which is filled into the mould while

this is slewed, so that the fluid plaster reaches all areas of the negative form.42 The mould is then

removed from the cast by chipping it away and is thus destroyed. The plaster model43 that is produced

from a lost cast might then be reworked, by cutting away seams, which occur when the mould is cast

37 Plaster Casts: Making, Collecting and Displaying from Classical Antiquity to Present, International Conference at

Oxford University (24-26 September 2007); Das Original der Kopie. The Originality of Copies, Colloquium (2nd - 3rd November 2007), Humboldt University Berlin; and planned Plaster and Plaster Casts: materiality and practice, 12-13 March 2010, conference at the Victoria and Albert Museum, London.

38 “Man bedenkt nicht genug, was für ein unschätzbares Mittel der Gips ist, daß man durch ihn das plastische Beste gewissermaßen identisch in die Ferne senden und, ohne den ersten Besitzer zu beeinträchtigen, einen zweyten Besitz und einen verdoppelten Genuß verschaffen kann.” Johann Wolfgang von Goethe in a letter to Ludwig Achim von Armin, dated Jena, 26. June 1806, cf. note 13: WA, IV, 50, p. 140.

39 See for example Edward Phelps.

40 Karlheinz Volkart, Gypsum and Plaster Dictionary, entry 469, spells mould and mold. 41 Plaster can also be cast into moulds made of wood or of clay and nowadays commonly in latex rubber or

silicone. Archaeological evidence suggests that plaster moulds began to replace earthenware ones in the Hellenistic eastern Mediterranean in about 200 BC for earthenware and metal as well as for plaster casting, see Nicholas Penny, p. 195.

42 See for instance the diagram in Henry Lie, p. 43.

43 The first plaster cast of a sculpture is often referred to as the artists’ model, when for instance the original model in clay or wax is lost or the sculptor has reworked it.

15

in sections. Superficial flaws that might have been formed in the casting process (i.e. air bubbles and

the like) are then evened out.

Plaster properties allow an easy surface manipulation as well as the reworking and re-piecing of

entire sculptural volumes. Plaster is usually cast in most parts as a hollow,44 and complicated

structures of a cast sculpture will be supported by the insertion of metal armatures,45 during or after

casting, and also by building the mould in sections (see Fig. 2). The cast sections of the statue are

assembled using armatures and plaster as joining materials.

Figure 2: Francesco Carradori ‘Making a Plaster Mould’, etching, 1802.

A durable mould is then usually formed onto this first plaster model, the surface of which is sealed,

traditionally with oil or shellac and also lubricated, traditionally with soap, wax or grease. The durable

mould serves for the production of a series of cast plaster copies. Serial production is a most common

practice in sculpture workshops and plaster has always served in various stages of the reproduction

44 As this decreases the costs, the setting time and the pressure that setting plaster exudes due to its expansion,

as well as the weight that amasses during casting procedure and of the finished product – thus being in total much more manageable.

45 In former times also often wood, see Moritz Kiderlen, p. 102 for a large variety of traditionally used arming materials in plaster casts, for instance bone and cane.

16

processes.46 Plaster reproductions can serve as a pointing model for copies in marble,47 or as models

for sand or wax casting bronze copies.48

Durable moulds are made in many small wedge-shaped plaster sections, each worked and fitted onto

the volumes of the casting model. Usually clay borders create each section, which is then filled with

liquid plaster. Very complex sculptural volumes will require hundreds of these small sections, each of

them having been cast and moulded into the undercuts and onto protrusions of the sculptural volumes,

removed after hardening, neatly conically shaped, coated with shellac, lubricated with soap or grease

and keyed into the next section. They are often given a little handle and numbered (see Fig. 3). The

whole assembly of small pieces is then bound together by an outer ‘mother-mould’, which in turn will

be formed onto the small-piece as well as the larger wedged surfaces of the piece-mould (again in

plaster), and will be made out of various sections, and bound together. One opening, usually the base,

is left open. This opening is then used for casting. When revealed, the cast is covered with a network of

thin raised lines, which correspond to the individual sections of the mould.49

Small piece moulds can produce many series of casts. If conserved properly, they last much longer than

any flexible mould and hence produce many more replicas. Before silicone was introduced into the

casting procedure as a flexible mould, gelatine was applied onto a model pre-treated with shellac or

wax.50 The gelatine was treated with alum to make it more durable and a two or more pieced ‘mother-

mould’ or case was fitted onto it, made of plaster, which in turn was treated in shellac. The gelatine can

be finely cut open to pull out the model and a small series of casts (four to six) can thus be produced.51

Modern day casting procedure has to some extent replaced heavy plaster ‘mother-moulds’ by epoxy or

polyester resin cases onto latex rubber or silicone, which are highly flexible and accurate casting

materials and thus largely used for reproducing sculpture.52

46 Cf. Christa Landwehr on ancient casting methods in plaster and in wax, excavated in Baiae, near present day’s

Naples.

47 Plaster pointing models are for example beautifully preserved in Antonio Canova’s Gipsoteca in Possagno.

48 See Tony Birks, pp. 23-87.

49 Nicholas Penny, p. 196. See for example Moritz Kiderlen, pp. 97-103, for a detailed description of casting procedures in small-piece moulds.

50 Having replaced the semi-flexible casting materials wax and resin at about 1850, Moritz Kiderlen, p. 98.

51 This procedure, among others, is practised to this day at the Replica Workshop of the National Museums in Berlin. For a simple gelatine mould, see diagram in Henry Lie, p. 43.

52 Tony Birks, pp. 36-39, illustrates this method in detail. To a much higher degree than gelatine, latex and especially silicone require the thorough application of a barrier material. See Annette Schulz, p. 58 on possible damages caused to original surfaces by casting procedures. See Jeffrey P. Maish et al. on the use of Cyclododecane as a barrier in casting marble with latex. Frédérique Chantepie has developed an interesting approach: She successfully tested polyacrylic acid, Carbopol 940, as a barrier material for silicone casting a

17

Figure 3: A small-piece mould.

plaster original, which detaches itself easily.

18

Plaster’s Properties

Gypsum plaster is also referred to as plaster of Paris or simply as gypsum or simply as plaster,

sometimes also by its Italian name, Gesso.53 Plaster is prepared by heating di-hydrated calcium

sulphate, which is naturally deposited in gypsum minerals or selenite rocks, at 100-150 °C, to drive off

some of the crystallization water and thus attain hemi-hydrated calcium sulphate.54

Figure 4: ‘The Plaster Kiln’ by Jean Louis Théodore Géricault, 1822-1823, oil on canvas.

Rehydration of the then finely ground hemi-hydrated calcium sulphate - by adding water - will allow it

to set and harden into calcium sulphate di-hydrate55 as a solid in a felt of needle-like fine crystals. The

setting procedure produces heat56 and expansion57, the latter being very useful for casting procedures.

Liquid plaster, with a consistency comparable to milk, is used for casting. It readily fills into all surface

53 Nicholas Penny, pp. 194-199.

54 Hemi refers to 0.46 to 0.68 parts water in crystalline form, depending on the source of the raw gypsum powder. On different types of hemi-hydrate and an-hydrate plasters and methods of their processing see Giorgio Torraca, pp. 65-67 and extensively Annette Schulz, pp. 29-36; also Gipsdatenbuch pp. 13-19 and 54-58. Mainly type α-hemihydrate-gypsum is used for casting and other statuary purposes. It is produced under elevated pressure, see: http://www.lafargeprestia.com/caso4___h2o.html.

55 Two parts water in crystalline form.

56 About 3.900 Cal/mole, Arthur Beale et al., p. 18.

57 By 1%, according to Bettina Uppenkamp, p. 137.

19

volumes of the mould and will trace all different surface textures precisely and smoothly. The

appearance of set and dry cast plaster is white and dense. Yet gypsum plaster is most often not pure,

in that it can contain other minerals. Sulphur, iron, silicates and other clay impurities can have an

effect on the colour of set and especially on aged plaster. Deposits of calcium carbonate or dolomite

will affect the porosity of set plaster.58

The porosity of plaster is largely dependent on the solids-to-water-ratio in the original mix. Only 17 %

of the original water content however is effectively used during the crystallization process.59

Generalizing, one can assume the closest crystal structure and thus higher density in a plaster with the

least amount of water used for its preparation.60 The surplus water creates the pores within the

crystal structure of plaster, from which it then slowly evaporates. The strength of set plaster is directly

related to its density;61 respectively porosity. Persistently moist conditions will induce further

crystalline and textural changes and re-crystallisation.62 Therefore aged plaster is noticeably softer

and much more open-pored than a newly cast surface.

The strength of cast plaster is reduced as a result of its solubility in water.63 Plaster is hydrophilic

because its structure is characterized by water constituting an integrally larger part of the crystal

structure, containing two water molecules for every molecule of calcium sulphate. This quite naturally

attracts water constituent molecules.64 The solubility rate of plaster is high: at 20 °C, slightly over 2 g of

calcium sulphate di-hydrate go into a solution per 1 litre of water.65 Repeated drops of water onto a

plaster surface will rapidly produce erosion – leaving wetted paper pulp on a probe plaster surface to

58 Louise Cone 1996, p. 7.

59 Violine Pillard, p. 280.

60 This fact stands in odd contrast to the phenomenon of the so-called casting skin, the plaster cast’s surface layer, which, although it is produced with plaster of a very runny, liquid consistency, forms a harder protective layer, like a shell to the more porous sub-superficial plaster, discussed for example by Carin Petterson and Louise Cone and verifiable with X-ray CT-examination: Liquid plaster of a first layer is filled into a lubricated casting mould, and once set, acquires a surface boundary of higher density. Lower densities of second and third layers of casting layers are visible with X-ray CT. Lubrication of the cast could possibly enhance the density factor of the casting skin, another probability is moisture extraction to the first layers by impact of the subsequent layers during casting.

61 Gipsdatenbuch, p. 55, correlates solid to water ratio of set plaster with bending tensile strength, ball impression hardness and water absorption. The average values for set plaster: bending tensile strength ≥ 2.5 N/mm2; for hardness ≥ 10 N/mm2.

62 Louise Cone 1996, p. 9.

63 Louise Cone, loc. cit.

64 Cf. Torraca, p.1, Violine Pillard, p. 280.

65 Arthur Beale et al., p. 19, Annette Schulz, p. 73, Louise Cone 1996, p. 8 Ute Griesser, p. 49 and Violine Pillard, p. 279, cite marginally different values for plasters’ solubility.

20

dry, will easily prove this.66 Erosion of the casting skin and of surface volumes indicate a history of

abrasive washing treatments. Plaster’s porosity makes for a specifically large surface area, and the

many cavities allow the infiltration of water molecules, which leads to erosion and hence to gradual

superficial degradation.

Plaster’s porous hydrophilic nature is also the cause for the strong adhesion and the infiltration of dirt into

its surface structure. Because of its material nature, plaster’s use and reception has largely been restricted to

the interior. Additionally plaster has traditionally also been modified during or after casting, in order to

cause hardening and water repellence.67 Setting for instance, can be accelerated by adding small amounts

of inorganic acids and their salts or retarded by the addition of organic materials, usually organic acids and

their salts, or of sugar, pectin (ground mallow root) or organic colloids,68 but also by the addition of sodium

phosphate or sand. All additional substances will in fact retard the setting rate to a varying degrees.69

Additives, such as marble dust, alum, borax, potassium, magnesia, animal glue or dextrin, have

traditionally been used to achieve the increased hardening of set plaster. Immersion of a cast plaster object

in a saturated solution of alum at 90 °C is one traditional method used for hardening plaster, yet is reported

to cause brittleness and to greatly increase plasters solubility in water.70 The addition of organic media,

such as sour milk or oils, traditionally rendered this hygroscopic material more resistant to water and thus

more easily washable.71 Coatings of oil, shellac72 or milk on cast plaster surfaces serving this purpose, are

accounted for just as various recipes and patents using combinations of some of the above-mentioned

together with barium sulphate, calcium silicate, phosphoric acid, paraffin or fatty emulsions or soap

treatments and many others.73

66 Erosion is produced at an observably more pronounced rate in accordance to high (basic) PH values of the

liquid. Di-hydrated Calcium Sulphate (CaSO4 .2H2O) in solution has a pH value of about 7. Plaster reacts very sensitively to acid factors lower than PH 5 and to alkalinity higher than pH factor 9. Cf. Course Documents of Gilbert Delcroix 2004, at ESBAT (Ècole Supérieure des Beaux-arts de Tours), recorded for example by Amélie Chédeville, in a table in Annex VI.

67 Annette Schulz, pp. 39-49; Ute Griesser, pp. 47-49.

68 L. Hüttmann / R. Tormin, p. 62; Annette Schulz, loc. cit.

69 Alfred Bohnhagen, pp. 188-192; Annette Schulz, pp. 36-37.

70 Martin Weber, p. 18; Alfred Bohnhagen, p. 189.

71 On washing plaster surfaces, see below, p. 26.

72 Oil or shellac coatings will most often indicate a plaster object’s former use in casting procedures.

73 Alfred Bohnhagen, pp. 190-192; L. Hüttmann / R. Tormin, p. 63; Annette Schulz, pp. 43-46.

21

Dust on Plaster Surfaces

Dust and dirt are sometimes used as synonyms and like dirt, the term dust is a hypernym. Technically,

dust contains dirt74 and dirt, by definition, is a material in a wrong place, “a foreign matter”.75 Dust on

the other hand, is ubiquitously present in all indoor and outdoor environments,76 and its accumulation

on a surface is termed as dirt. While dust is generally categorized as a pollutant, it is also categorized

as having the specific quality of being able to adsorb and to absorb pollutants.77 The term pollutant is

used for an extremely complex mixture of airborne atmospheric matters, which occur as dispersed

solid particles, liquid droplets or gas.78 Both the solid and liquid matters are called particulates, and an

aerosol is the definition of an assembly of solid or liquid particulates suspended in gases and

vapours.79 Particulates are also termed as small discrete objects.80 They are categorized by their sizes,

by their chemical compositions and by their sources of origin. The sizes of particulates range between

that of a molecule (approximately 0.0002 μm in diameter) to 500 μm.81 Particles that are large enough

to settle in still air are generally referred to as dust; they are removed from the air by gravity and other

inertial forces.82

The size of dust is defined as larger than 0.5 μm (see Fig. 5).83 Particles larger than 2.5 μm are defined

as coarse particles,84 and particles above 20 μm in size have been evaluated to have significant settling

74 http://www.thefreedictionary.com/dirt names among other meanings of the word dirt: earth or soil; a filthy

or soiling substance, such as mud or dust; excrement.

75 Science for Conservators, Vol. 2, Cleaning, pp. 13-14.

76 Sally Woodcock, p. 125.

77 See Lidia Morawska and Tunga Salthammer 2003a, Barbara Appelbaum, pp. 97-98 and 112-113, Dario Camuffo et al., p. 128 or reCollections, Caring for Collections across Australia, Dust and Pollutants, see: http://archive.amol.org.au/recollections/3/pdf/dust.pdf.

78 Stephen Hackney 2008/2009.

79 Stephen Hackney, loc. cit.; Gary Thomson, p. 150, points out that aerosols unfortunately do not remain suspended in air for ever, but at some point settle and adhere to surfaces, forming solid layers of black dirt. Lidia Morawska and Tunga Salthammer, 2003a, explain aerosols as an assembly of liquid and solid particles suspended in a gaseous medium long enough to enable observation and measurement.

80 Lidia Morawska and Tunga Salthammer 2003a, p. 8.

81 Kenneth Wark and Cecil F. Warner, pp. 9-10. See Garry Thomson, pp.147-148, on units of measurement for pollutants.

82 Gary Thomson 1986, p. 131. See Kenneth Wark and Cecil F. Warner, loc. cit. and pp. 165-169 for terminal or settling velocities for particles of varying sizes.

83 Lidia Morawska and Tunga Salthammer 2003a, p. 8-9. For a bibliographic review on sizes of dust particles see Diameter of a Speck of Dust, The Physics Factbook, http://hypertextbook.com/facts/2003/MarinaBolotovsky.shtml.

22

velocities.85 Dust is formed by crushing or other mechanical breakage of a parent material.86 It is

further defined into the categories ‘soil dust’, ‘road dust’ and ‘house dust’.87 The German Guideline VDI

4300-8 (VDI, 2001)88 states: “There is currently no generally binding definition of the term settled house

dust. To delimit the term from suspended particulate matter, it is meant to mean all types of particles,

which are encountered indoors in deposited form. The dust may be solids of the most varied inorganic or

organic materials, which can be found of natural or synthetic origin. The term includes not only fractions

which originate indoors themselves, but which also are introduced from outside.”

Figure 5: These dust particles were collected off pedestal surfaces in the Rococo Hall (in July 2009) and photographed at 25-fold microscopic magnification after separating them from the larger amount of fibrous material present in the sample. 1,342282 µm equals 1 pixel in this image’s resolution, hence particle sizes of under 1,3 µm are not visualized.

84 Lidia Morawska and Tunga Salthammer, loc. cit. In comparison, the diameters of human hairs range from 17

to 181 µm, according to http://hypertextbook.com/facts/1999/BrianLey.shtml.

85 See above, note 82, for settling velocities of particulate matter.

86 Lidia Morawska and Tunga Salthammer, loc. cit.

87 For instance Gary Thomson, 1965, p. 151, Helen Lloyd et al. 2007a, p. 138; Nazaroff, pp. 17 and 113. See also Lidia Morawska’s Motor Vehicle Emissions as a Source of Indoor Particles on shapes and sizes of combustion and abrasion particles in road dust, in: Lidia Morawska and Tunga Salthammer, pp. 297-318.

88 Translated in Lidia Morawska and Tunga Salthammer 2003a, p. 9.

23

Outdoor air pollution is generated both anthropogenically – by various combustion processes of the

industry, energy generation, transportation89 and other human activities – and also naturally. Some

natural sources of pollutants are the sea (salts),90 soil and rock debris, forest fires (a combination of

smoke91, soot and gas), volcanic debris and particles formed by natural gaseous emissions.92 Both the

rate and the quality of pollution is largely dependent on the respective region.93 Industrially generated

gaseous pollutants are dominantly acidic. Especially reactive gases are sulphur dioxide, nitrogen oxide

and hydrogen sulphide. Their transport is conducted by water vapour, which catalyses their acidic

reactivity.94 Their salts, sulphates, nitrates and sulphides are produced in the presence of oxygen, they

are thus termed secondary pollutants.95 Ozone96 found on ground level is also termed as a secondary

pollutant,97 because it is produced when nitrogen dioxide and carbon monoxide (chiefly produced by

combustion of fossil gas) and volatile organic compounds (VOC)98 react with the atmosphere in the

presence of sunlight. Alkaline pollutants have also been investigated on their damaging effects to

collections materials, and fresh concrete used in building materials is named as a chief source for

alkaline particles found in collections.99

89 The prime source of 20th century air pollution being the motor vehicle, significantly the diesel motor, see

Anne Lisbeth Schmidt, et al. and Peter Brimblecome 2004, p. 87.

90 See for example David Thickett on corrosion processes to metals in the presence of chlorides, dust and RH in collections situated in coastal and inland regions.

91 Smoke contains, apart from carbon oxides, sulphur dioxide, ammonium sulphate and iron and traces of other elements, see Gary Thomson, p. 150.

92 Hydrogen sulphide, ammonia, nitrogen oxide, and hydrocarbon, see Lidia Morawska and Tunga Salthammer 2003 a, p. 4.

93 Stephen Hackney 2008/2009.

94 Gary Thomson, p. 152-154 describes some reactions and induced effects of sulphur dioxide, hydrogen sulphide and nitrogen oxides on materials in museums. For a summary, see Table 1 in Norbert S. Baer and Paul N. Banks, respectively Table 1 in Peter Brimblecombe, 1990 or see Barbara Appelbaum, pp. 97-98.

95 Jean Brown 2008/2009; Stephen Hackney 2008/2009.

96 Garry Thomson1965, pp. 157-159, already discusses the attack of cellulose and natural resin varnishes in museums by ozone.

97 For example http://en.wikipedia.org/wiki/Air_pollution.

98 See for instance Michail Kokonoglou’s and Lorraine Gibson’s poster on analysis of volatile organic compounds in indoor air, IAP Copenhagen 2001, http://www.natmus.dk/cons/reports/2002/iap_meeting/iap2001.pdf; Peter Brimblecombe, 1990 and René Van Grieken et al. discuss some effects of formaldehyde on museum objects.

99 Peter Brimblecombe, 1990, p. 2; Nazaroff et al., p. 17 as well as Norman S. Baer and Paul N. Banks, p. 10, refer to a series of investigations by K. Toishi and T. Kenjo in the late 1960s and early 70s on the effects of alkaline aerosols emitted by setting concrete. Barry Thomson, p. 156, defines the effects of ammonia for the

24

Sand, clay, and ash particles, silica crystals, salts and mineral fibres present in house dust, might have

originated naturally outdoors or might be composite flakes of building materials of the respective

indoor situation. Soot and tar as well as fat and grease products that are present in house dust, might

equally have been produced indoors (by heating or cooking for example) or else imported from

outdoors by air exchange100 or the presence of human beings.101

Textile fibres – stemming from clothes, carpets, furniture etc. – are stated to constitute the larger

amount of particle matter of sampled dust in historic and museum interiors (see Fig. 7).102 Human hair

and skin fragments form another large proportion of particulate matter to be found within sampled

house (i.e. museum) dust.103

Figure 6: These fibres were separated from dust samples collected off pedestal surfaces in the Rococo Hall (in

July 2009, cf. Fig. 5) and photographed at 6-fold microscopic magnification.

deterioration of art.

100 On correlation of outdoor and indoor pollution and see for instance August Krogh, Stephen Hackney 1984, Norbert S. Baer and Paul N. Banks, Table 2, Peter Brimblecombe 1990 Table 4. Nazaroff, pp. 21-54, compares air pollution in three Californian museums with particle filters built into sophisticated heating, ventilation and air conditioning systems (HVAC) to two Californian museums with no pre-filtering.

101 Soot consists of elemental carbon particles and is the main concern of the studies described in Nazaroff et al. and by James R. Druzik and Glenn A. Cass.

102 See for example Young Hun Yoon and Peter Brimblecombe 2000a.

103 Nazaroff et al. p. 113; Marie-Louise Jacobsen; Young Hun Yoon and Peter Brimblecombe 2000b. Human beings are stated to lose up to 100 hairs a day, they are also cited to lose 1 to 2 grams of skin flakes a day, see: http://www.wdr.de/tv/quarks/sendungsbeitraege/2004/0413/006_unsichtbar.jsp.

25

Apart from shedding skin and losing hair, human beings also release a range of organic compounds as

bio-effluents – ketones, aldehydes, alcohols, esters, phenol and toluene – and also inorganic gases such

as carbon dioxide, carbon oxide, hydrogen sulphide and ammonia.104 Plant parts and plant pollen,

insect parts and insect excrement are also naturally generated components within house dust, just as

are spores of moulds and other microorganisms.105 Dust mites for instance, “cosmopolitan guests in

human habitation,”106 feed on the human skin flakes present in dust.

The term soiling refers to the accumulation of particles on surfaces.107 So, coming back to the presence

of dust on bare plaster surfaces, and with the knowledge of plaster’s surface reactivity: The fact that

dust is, on top of all described above – hygroscopic,108 poses a particular concern. Since “dust can

interact with an object’s surface, causing physical damage and chemical alteration”,109 the potent

sources for plasters deterioration, prevalent in dust, require detailed and contextual analyses.

Resolution 6 of the Commission for Consultation on the Treatment and Conservation of Plaster Casts of

1877 reads:110 “The protection of (…) casts will be essentially achieved by a most possibly dust-free

installation of collection rooms and the prevention of humidity in these, which should be duly considered

when constructing new or remodelling existing collection rooms.” Air filtration and humidity control are

– for obvious reasons – the pre-condition for sustainable conservation of plaster statuary. Showcase

installation is advised, equally the covering of plaster statuary within storage with cotton cloth or with

polyethylene sheets.111

104 Peter Brimblecombe 1990, pp. 2-3.

105 See for instance Marie-Louise Jacobsen. The relation between poor house-keeping and biological infestation of art and artefacts is a fundamental parameter in preventive conservation, see Jean Brown, Bob Child, Helen Lloyd, Hannah J. Bainbridge, Barbara Appelbaum or for example http://www.nps.gov/history/museum/publications/MHI/CHAP5.pdf [last accessed August 12th, 2009]

106 http://en.wikipedia.org/wiki/House_dust_mite [last accessed August 12th, 2009]

107 The rate of soiling is expressed with R for reflectance; see Carlotta Grossi et al., http://www.iaq.dk/iap/iaq2003/posters/soiling.pdf, or Peter Brimblecombe and Carlotta Grossi 2004.

108 Particularly salts and fibres within house dust have hygroscopic qualities, see for instance Barry Knight, p. 20 and Helen Lloyd et al. 2007a, p. 138.

109 See: http://www.nationaltrust.org.uk/main/w-collections-dust.

110 “Der Schutz der Abgüsse wird wesentlich durch eine möglichst staubfreie Einrichtung der Sammlungsräume und durch Bewahrung derselben vor Feuchtigkeit gefördert, worauf bei der Errichtung neuer oder dem Umbau alter Sammlungsräume gebührend Rücksicht zu nehmen ist.” Kommission zur Beratung über die Behandlung und Konservierung von Gipsabgüssen, see below, pp. 63-66.

111 Annette Schulz, p. 76; Michalis Doulgeridis and Maria Kliafa. Any immediate contact of any material with a plaster surface should however be considered in respect to possible hazards they might represent, like for example fire load, humidity accumulation, condensation due to micro environment or volatile compounds in their material composition, and is definitely not advisable for pre-soiled surfaces.

26

Figure 7: This cast plaster bust of Julius Caesar shows heavy soiling and the so-called

‘reverse shadow effect’.

The catalyst for all deterioration processes to plaster surfaces seem to be high humidity levels: Apart

from forming a “hard-to-remove film”112 of dark grey-brown to black hue, pitting as a result of plaster

surfaces’ interactions with accumulated “old dust”113 in the presence of humidity is a known

phenomenon in the field. Cementation processes of dust to some material surfaces in the presence of

humidity and their effects have been investigated.114 Humidity seems to be the most important factor

112 Barry Knight, loc. cit. Louise Cone, p. 13, refers to the result of many years of collected dirt to a plaster surface

as ‘natural patina’.

113 German Guideline VDI 4300-8 makes a distinction between ‘old dust’ and ‘fresh dust’. Old dust is settled dust of an unknown age, fresh dust is defined as dust, of which the age is determined by the planning of time of measurement, Lidia Morawska and Tunga Salthammer 2003a, p. 9.

114 Barry Knight, p. 20, reports on investigations by Young Hun Yoon that prove the cementing material within dust to be chiefly of organic nature. Both hygroscopic salts and sugar (probably of fungal or microbial origin) present in dust, react strongly to high humidity levels and are assumed responsible for the cementing reaction of dust to surfaces. See also Holly Dawes, who states that cementation starts taking place at humidity levels above 35%.

27

in the cementation of dust to surfaces.115 Although research yields no comparable analytical study on

the effects of pollutants on plaster surfaces, visual evidence proves that various corrosive processes to

plaster surfaces do take place due to dust deposition and its adherence.

Dust is also a welcome host for fungal growth, which is evidenced on plaster as darkened areas with a

pitting phenomena.116 Chemical as well as physical alterations to plaster are commonly evidenced in

colour changes. Visible plaster soiling can appear as yellow and brown staining, which might be due to

impurities present in the plaster itself.

Soot and grease - due to heating117 and handling - are a big topic for all surfaces. Staining due to

handling, results from a combination of perspiration products of the hands and of other materials that

the hands will have touched, like the remains of food or creams. Areas of handling acquire a gloss as

well as heightened tackiness and thus allow for increased adherence of particulate matters. The

perspiratory glands of the hands produce a constant flow of a clear, odourless acidic liquid, which is

slightly salty to taste. Water is the main constituent of human sweat (99 %), but organic compounds

like urea, glucose, lactic and amino acids and others are just as present as are an-organic compounds,

mainly sodium chloride, also ammonia, phosphates, sulphates, magnesium and calcium nitrates.118

Areas of handling on plaster surfaces are often shaded a dark yellow to brown (see Figure 9). In

addition, inherent impurities in plaster will be visually more pronounced in areas of handling due to

plaster’s porosity and the capillary action, which is induced by introducing hand-sweat. Very often

dark fingerprints can be found on plaster surfaces, which are result of infiltration of hand-sweat into

pre-soiled surfaces.119 Equally, past cleaning interventions with brushes, sponges or feathers are

visible both in the erosion of surface volumes as well as by deep ingression of soiling, sometimes

combined with a pronounced gloss.120 Constructive and ancillary materials, like metal and wood,

cause discolorations but can also induce cracking of entire sculptural volumes in the presence of high

humidity levels.121

115 See for example Peter Brimblecombe et al. 2004, or Helen Lloyd et al. 2007a.

116 See Lorenza D’Alessandro and Francesca Persegati, pp. 99-102. Helgard Weber as well as Arthur Beale et al., p. 20, discuss the action of micro-organisms specifically to coating media on plaster surfaces. Arthur Beale, p. 19, and Jorun Rebekka Ruppel, p. 177, remark on fungal growth being attracted to plaster surfaces by the presence of remnants of cleaning materials, like corn starch or agar-agar poultices.

117 Louise Cone 1996, p. 12, describes the heating system’s impact on the plaster collection of the Thorvaldsens Museum in Copenhagen.

118 Ute Griesser, pp. 53-54.

119 Peter Brimblecombe 1990, p. 5, Ute Griesser, p. 55.

120 Louise Cone, p. 14.

121 Violine Pillard.

28

Figure 8: The rate of soiling on plaster surfaces is more evident in the presence of areas of recent losses.

Cleaning Plaster Surfaces

Total immersion of the plaster object in clear water for a few hours, until the plaster is fully soaked, is

a method that fortunately is only rarely found advocated.122 As is the proposed subsequent washing

off of still adhering dirt with clear running water and soft rubbing with brushes or sponges.123 Most

older and contemporary published recipes and recommendations for cleaning plaster however, do take

plaster’s delicate and brittle, porous and soluble properties into account, by involving low to no

122 For example Philippe Clérin, p. 84 or Victor H. Wager, p. 86: “Mere dust can be nicked off, but when of old

standing, the cast will have to be washed. Do not under any circumstances wash it as you would wash a dish. This will send the dirt in. The only way is to totally immerse the cast, leaving it untouched until it can absorb no further water. When the bubbles cease to rise, leave for another ten minutes or so and then wipe it, UNDER THE WATER, with a sponge. Do not lift the model above the surface, as it will collect again the dirt that is floating there. When thoroughly sponged, collect all the dirt from the water with blotting paper and a saucer, or allow the tap to run until all the dirt has flowed away over the sides. Only when the surface of the water is clean, can the cast be removed. If the sponge is used before the cast is sufficiently saturated, the wiping will send the dirt in, so try wiping on an unimportant part first.” Friedrich Rathgen, p. 34, states that the general appearance of a cast thus treated seems only little changed, especially if viewed from some distance. See Louise Cone, p. 17, who describes this method to be historically documented in Florence in 1888.

123 Philippe Clérin, loc. cit.; Friedrich Rathgen, loc. cit. recommends washing plaster casts (if not too old) with soap and water and brushes. Cf. Barbara Appelbaum, p. 173, who writes: “Do not, ever, use water to wash down the surface of a plaster, as the water combined with abrasion will cut into the surface.”

29

humidity transport and action. Because water is so easily absorbed by plaster, soiling is irreversibly

carried into the surface by wet methods.124

Poultice-cleaning with methylcellulose gels or starch pastes for example, which are removed while still

moist and flexible, are widely propagated.125 The use of agar-agar pastes,126 employed in the same

manner as starch pastes, is also frequently discussed in the relevant literature. It is probably

comparable to the suggestion of alginic acid127 in its application and action, or to latex milk.128 The

combination of methylcellulose and fine clay powder (like attapulgite), as a paste, produces a poultice

with drier properties than a methylcellulose gel on its own. It is highly adsorbing and is equally

removed from the surface while still moist (see Fig. 9).129 Laponite® gel, thickened with fumed silica,

is left on the surface to dry into soft crumbling flakes.130 Similarly, a paste consisting of magnesium

oxide and water is applied to plaster surfaces and left there to dry. Soiling is removed from the plaster

surface in the process of the dried magnesium oxide falling off or being brushed off.131

All of the described poulticing methods act by binding the layer of soiling to the poulticing film and

thus, in the best case, gently separating it from the plaster substrate on removal. Staining and tide-lines

can however be produced when the poulticing method contains too much water, and also when

poulticing is applied only partially. In addition, too high surface tensions of the poulticing materials

can cause structural damage to the plaster surface when removed.132

124 CCI Notes 12/2, p. 2, include the valuable warning that uncoated plaster objects might be sculptor models, i.e.

maquettes, and thus might carry artist’s markings which would be obliterated by moisture intensive cleaning methods.

125 For example Eduard Uhlenhuth, p. 59; Friedrich Rathgen, loc. cit.; Anette Schulz and Ute Griesser each list an entire range of gel and paste systems, which are accounted for in plaster cleaning.

126 Mentioned already in 1915 by Friedrich Rathgen. See Anette Schulz, p. 82, who describes this practice with the modification of allowing the agar-agar film to dry on the plaster surface and remoistening it before removal.

127 Didier Besnainou.

128 Anette Schulz, pp. 82-83. Judit Gasca Miramón et al.

129 Astrid Lorenzen and Bertrand Porte; Johanna Kapp.

130 Ursula Haller and Ulrich Schießl. Jorun Rebekka Ruppel points out the yellowing that can be produced with this comparatively moist method, p. 180.

131 Carin Pettersson, p. 38; Louise Cone, p. 22, describes this method of being especially effective on stains caused by impurities such as sulphure and iron.

132 Carin Pettersson for example tested polyvinyl-acetate and rice-starch cleaning applications to plaster surfaces and by microscopic evaluation found they produced erosion.

30

Figure 9: Poultice cleaning plaster surface. Surface soiling is separated from the plaster surface on removal of the

poultice.

The dry removal of dirt films has been suggested by way of sanding, most often however in the

context of a subsequent application of a whitening coating.133 Laser cleaning is a dry cleaning method

that is currently under investigation for heavily soiled plaster.134 However, yellow and brown

discolorations are effected very notably in areas of increased soiling, which might relate to prior

handling.135 Simple dusting, such as wiping with soft cloths or with dusting feathers is documented to

cause gloss to plaster surfaces.136 A variety of erasers are recommended in plaster cleaning,137 and

have largely replaced breadcrumbs in this use.138 The abrasive effect erasers can have on plaster is

noted, as is the fact that soiling is thereby rubbed into deeper surface zones.139

133 For example Friedrich Rathgen, p. 35, who remarks that only schooled workers should be allowed to

undertake this measure. See also Anette Schulz, p. 78.

134 For example Søren Rasmussen and Birgitte Lydik Clausen.

135 Jorun Rebekka Ruppel, p. 180; Eric Tanguy et al. have particularly focussed on the problem of yellowing in their trials with laser cleaning.

136 Louise Cone, see note 123.

137 See Ute Griesser, pp. 63-67.

138 Anette Schulz, p. 78.

139 Ute Griesser, p. 67.

31

Non-contact removal of loose dust from plaster surfaces with compressed air was formerly achieved

with bellows,140 and is generally advisable for the removal of loose particles, if the pressure is set well

below two bar and the plaster surface volumes are intact. Vacuuming is also an advisable method for

removing dust, however the conjunct use of even soft brushes on plaster surfaces as an aid in

vacuuming has been noted to be abrasive.141

140 Anette Schulz, loc. cit.

141 Ute Griesser, p. 63, note 50. And see below, pp. 69-80, in the survey, for a variety of recommendations on dry cleaning methods.

32

Chapter 2

Case Study – The Plaster Portrait Busts in the Duchess Anna Amalia Library

The Duchess Anna Amalia Library is singular both in terms of its history and also its impact as the

most important library for German literature between 1750 and 1850. Although the era of

Enlightenment to later Romanticism was the time during which the profile of the library was most

prominently formed, its holdings today contain a total of one million literary sources, manuscripts and

books, ancestral registers, maps as well as musical scripts, dating from the 9th to the 21st century142.

The Duchess Anna Amalia Library is, since 1998, part of the UNESCO World Heritage site “Classical

Weimar”.143

The Library gained sad world wide fame when, on September 2nd, 2004, a corroded clamp joint of the

copper aluminium wiring of an electric cable started smouldering behind wooden panelling in the attic,

gradually causing the most devastating library fire in Germany since World War II. The flames

destroyed about two thirds of the roof.144 50,000 books were entirely lost, 62,000 volumes were

partially damaged by the flames and by extinguishing water. While the two upper floors of the library

building stood in flames, about 28,000 volumes could be evacuated from the lower floors of the library

in unharmed conditions. 850,000 volumes were luckily stored in alternate magazines at the time and

were thus spared from the fire.

Central pieces of the inventory of art works survived the fire, yet the irreplaceable losses suffered by

the Duchess Anna Amalia Library include a large oil on canvas painting that had been attached to the

ceiling above the second Gallery as well as 35 paintings, which had adorned the stairway leading to the

third floor and on the third floor in the second gallery. The 90 marble, limestone, terracotta and mainly

plaster sculptures that had been on display in the library could be evacuated in a comparatively less

damaged state. A total of 65 portrait busts in plaster were plucked from their pedestals from the

burning building, while fire-extinguishing work was underway (see Fig. 11). These busts are the

subjects of this research.

The restoration of the building and the artwork, which was to be re-integrated into the ensemble of

the Rococo Hall was accomplished by late Summer of 2007. The re-opening took place on October 24th,

2007 – only 1147 days after the fire-disaster – on the 268th birthday of Duchess Anna Amalia (1739-

1807) and the German ‘Day of Libraries’.145

142 See http://www.klassik-stiftung.de/einrichtungen/herzogin-anna-amalia-bibliothek.html.

143 See http://www.klassik-stiftung.de/stiftung/unesco-weltkulturerbe.html.

144 See http://www.dw-world.de/dw/article/0,,1317151,00.html.

145 Walther Grunwald, Michael Knoche and Hellmut Seemann; J. Flade; http://www.anna-amalia-library.com/en/

33

For more than a decade prior to the fire, the library director, Michael Knoche, had put the restoration

of the historical library rooms and the entire building on the urgency-agenda. In 1992, risk

assessment had determined the urgent necessity for a thorough modernization of the building and the

need for a relocation of the vast dimensions of the library holdings from different storage commodities

into a new library extension. The Study Centre, which lies just opposite the historical main building, is

composed of two refurbished historical palaces and a recently constructed adjacent building.

Underground multi-storey storage spaces connect these two library ensembles. The scheduled date for

the evacuation of all books from the historical library rooms into underground storage was scheduled

for only a few weeks after the night of the fire. This is the particularly bitter background story to the

disaster as well as to the overwhelming demonstration of support, which the Duchess Anna Amalia

Library experienced thereafter.146 The nation wide and even worldwide attention and sympathy,

which the disaster in Weimar provoked, was and still is outstanding. To date, about 35 million € have

been received from public funds and private donors for the restoration and replacement of the book

collection. This represents one fourth of the total amount necessary, which is estimated at 67 million

€. Comparable copies are replacing primarily those works, which were completely burned or for

which restoration costs are comparatively too high. A total of 22,000 works from the 16th to the 20th

centuries have since been re-integrated into the collection. 10,000 of these books were gifts from

other libraries and private collectors after the fire, and 12,000 were second-hand purchases. The

restoration of the 20,000 water-damaged and 8,000 fire-damaged volumes is still underway and is still

dependent on the generosity of donors.147

Figure 10: September 2004, the plaster statuary was salvaged from the burning Rococo Hall into the under-

ground storage area.

146 Michael Knoche 1999a, 2005, 2006a and 2006b, 2007a; Hellmut Seemann; Annette Seemann, pp. 108-118;

Walther Grunwald.

147 Mathias Hageböck 2005 and 2006; Corinna Herrmann and Cornelia Ripplinger; Rembert Unterstell.

34

A Brief History of the Duchess Anna Amalia Library and Its Art Collection

The Duchess Anna Amalia Library is named after Anna Amalia, Duchess of Saxe-Weimar-Eisenach

(1739-1807), who arranged for the courtly book collection to be moved from the Ducal Palace into a

separate building, thereby providing for a much more effective public accessibility than was formerly

possible. The Green Palace was chosen as the location of the new library building, a small princely

Renaissance residence constructed in 1563 by Nicolaus Gromann some 200 meters away from the

ducal residence, the Wilhelmsburg. Johann Georg Schmid and August Friedrich Straßburger remodelled

the Green Palace into a modern library building between 1761 and 1766 under the directive and close

supervision of the Duchess.148

Figure 11: View of the Ducal Library from North-West (the building behind the trees, seen from direction of

the palace), Aquarelle painting by Georg Melchior Kraus, ca. 1800.

The central and representative room of the Duchess Anna Amalia Library, the so-called Rococo Hall is

one of the most beautiful library halls in Germany.149 The interior of the Renaissance palace was cored

to accommodate a wooden construction, which spreads over three storeys. An oval interior with two

galleries was assembled on a rectangular floor space of 21 by 11 meters. The decoration of the buttress

capital, the gallery balustrades and of the ceiling above the second gallery was designed in the style of

the late Rococo. The original material chosen for the interior decoration – a woad-tinted chalk paint

on all the wood work and composition gold accentuating all decorative elements, like the batten,

cartouches and capitals – was comparatively simple, yet achieved an undoubtedly splendid ensemble

for the 30,000 volumes, which moved in in 1766.150

148 Jürgen Beyer.

149 See Laura Padgett’s photographical documentation of 1999 and Candida Hofer’s famous photos of the Rococo Hall in her book on Weimar.

150 Ulrike Steierwald; Michael Knoche 2007b.

35

Anna Amalia’s special interest and initiative in all matters of construction, design, equipment, staffing

and accessions of the Ducal Library, as it was then called, is remarkable for her time and is explained

by her ancestry. Born Anna Amalia of Brunswick-Wolfenbüttel as fifth child to Princess Philippine

Charlotte of Prussia and Karl I, Duke of Brunswick-Wolfenbüttel, she grew up in a court, which

situated the renowned Bibliotheca Augusta, esteemed throughout the 17th century as the eighth

wonder of the world, forming the largest library holdings north of the Alps.151 The Library-Rotunda in

Wolfenbüttel was erected in 1710 as the first freestanding profane library building in Europe and is

regarded as paragon for the Ducal Library in Weimar - both for its interior design and in terms of its

collection of art.152

Princess Anna Amalia became Duchess to Saxe-Weimar-Eisenach in 1756 by marriage to Ernst

August II Konstantin of Saxe-Weimar-Eisenach (1737-1758). Ernst Augusts untimely death two years

into their marriage left her as regent for their infant son Karl August until his maturity in 1775. Anna

Amalia’s regency is acclaimed because of her particular patronage of the arts and of music.153 Her

appointment of the poet, writer, literary critic and translator of William Shakespeare’s works into

German, Christoph Martin Wieland (1733-1813), for the education of her two sons, marks the

beginning of a period, which was later named Weimar Classicism. This term describes a cultural and

literary movement with a widespread, profound and lasting impact on the arts and on aesthetics, on

science, philosophy, psychology and literature. Weimar Classicism was chiefly distinguished by

Christoph Martin Wieland, by Johann Wolfgang von Goethe (1749-1832), by Johann Christoph

Friedrich von Schiller (1759-1805) and by Johann Gottfried Herder (1744-1803). These names are

closely linked to the Duchess Anna Amalia Library, their portraits are manifold in these historical

rooms, which is as much home to their works as it is to the literature they read and cherished.154

Like other libraries of the early modern period, the Grand Ducal Library in Weimar was not solely a

collection of books but a collection of books in context with the ducal collections of art, nature and

history. Not too much is known about the early art collection and its display in the Rococo Hall, yet it is

assumed that it mainly consisted of the paintings from the former library rooms in the Wilhelmsburg

– 90 oil on canvas portraits depicting ancestors and other great people connected to the progeny –

which first decorated the library hall.155 This changed in 1774, when a fire disaster struck and

151 See http://www.hab.de/museum/besuch/index-e.htm.

152 Michael Knoche 2007b, p. 234; Annette Seemann, p. 48.

153 Michael Knoche 1999c; Joachim Berger.

154 Michael Knoche, loc. cit. and see http://en.wikipedia.org/wiki/Weimar_Classicism

155 Michael Knoche 2007b, p. 234; Bettina Werche 2007a, p. 163.

36

destroyed the Wilhelmsburg and all the objects that could be salvaged from the ducal art collections

were brought into the Ducal Library. A transfer of further art pieces and cultural treasures from other

residences of the ducal family followed.156 An annex built onto the south of the Green Palace in 1804 by

Heinrich Gentz (1766-1811) allowed not only for the establishment of an own Art Cabinet in context

with the Rococo Hall, it gave further space for a seemingly never-ending accumulation of art and other

objects of artistic, cultural and scholarly interest. Any new acquisitions seemed to have gained

immediate access to the Green Palace and Anna Amalia’s own considerable legacy of books, musical

scripts, etchings, drawings, minerals and other types of objects were added to the library collection

after her death in 1807.157 In fact, the Green Palace transmuted to a ‘Library Museum’ in the period

during which Johann Wolfgang von Goethe was superintendent of the library (from 1797 to 1832).

The library staff came to act as custodians, conservators and as tour guides for all parts of the Ducal

art collections.158

After the fire in the palace in 1774, the library had assumed the function of a collecting receiver for

anything that seemed worthy. This even developed into the absurd159 when any new acquisition, like

for example, some mysterious Thuringian stone tablets, made their way into the Green Palace, as well

as what was thought to be Schiller’s exhumed skull, which for a while was kept inside an inbuilt locker

of one of the wooden pediments in the Rococo Hall in 1826, later to be replaced by his death-mask.160

The Rococo Hall and the annexed building to the south, as well as all the secondary rooms and

stairwells and interconnecting corridors of the Green Palace remained considerably crammed with

objects and paintings, as well as with glass-cabinets and tables filled with precious and interesting

items well into the early 20th century. Although the Drawing School accepted a large part of the art

collection in 1824, when it moved out of the Grand-Ducal Palace161 into its own building and the Grand

Ducal Museum opened in 1868 with a large portion of the collection of paintings and graphic art,

sculptures and casts of ancient sculptures. The systematic removal of the artwork from the library

continued during the second half of the 19th century: The Wittumspalais, the residence of the Dowager

Duchess after the fire in the Wilhelmsburg, was made into a museum for the families’ history in 1871.

156 Gert-Dieter Ulferts.

157 See Bärbel Raschke for a description of the dowager duchess’ private library.

158 Annette Seemann, pp. 48-49; See Ulrike Steierwald, pp. 94-95, for the musealogical aspect that entered the concept of librarianship in Weimar, citing a letter from Goethe to his colleague Christian Gottlieb Voigt in 1811.

159 Michael Knoche 2007b, p 236.

160 See Klaus Fahrner on Schillers portraits and the skull-intermezzo in the library hall.

161 The title of Grand Duke was awarded to Carl August at the Congress of Vienna in 1815.

37

The prehistoric items from the libraries Art Cabinet were merged into the collection of Germanic

Museum Jena in 1876. One further site of memorial culture was created for Weimar with the opening

of the Goethe National Museum in 1885. It was finally the abdication of the last sovereign in Weimar,

Grand Duke Wilhelm Ernst (1876-1923), in 1918 and the accommodation of large parts of this

dynasty’s art, treasure, curiosa and memorabilia collections in the Grand Ducal Palace, then called the

Palace Museum, which ended the library’s role as the central site for all art collections. The art, which

remained on display in the library hereafter, can be regarded as little more than citations of the

different collection concepts and happenstances of the 17th to 19th centuries.162

Figure 12: View into the interior of the Rococo Hall in 1978 with Carl-August’s oil on canvas portait placed

in the central position, crowned by a plaster cast of a monumental bust of Johann-Wolfgang von Goethe on the balustrade of the first gallery.

The period of intense collection and acquisition of art work and other items of scientific or cultural

value described here, is naturally also reflected in the number of accessions of books and other literary

and musical sources to the library: While the library holdings in 1691 – the year in which public

access was first granted to the Ducal Library in the palace itself – records 1300 volumes, in 1766 about

30,000 volumes first move into the Green Palace. The year of Johann Wolfgang von Goethe’s death,

162 Michael Knoche 2007b; Ulrike Steierwald, pp. 94-97.

38

1832, records 80,000 volumes, the year 1875 records 170,000 volumes and the year 1913 records

300,000 volumes in the library holdings. The literary volumes augmented over the following 100

years - until the fire in 2004 - close to one million volumes, while the art collection however further

dispersed due to different political and administrative re-organisations.163

After the dissolution of the Grand-Dukedom, the Grand-Ducal Library was re-named into Thuringian

Regional Library in 1920. In 1969 the Thuringian Regional Library was incorporated into the National

Research and Commemorative Site for Classical German Literature in Weimar (NFG)164 – the

predecessor of today’s Foundation of Weimar Classics. The library was once more re-named, into

Central Library of German Classicism, when the library of the former Goethe Society had been merged

into its holdings. It was during this period, throughout the 1970s, that most of the still remaining art-

collection was relocated out of the library and into various museum holdings within the NFG. The

Duchess Anna Amalia Library finally received its name in 1991 on the occasion of its 300th birthday, in

a newly re-unified Germany and upon a vote of the library staff. In hindsight this name seems to be

the most appropriate identification for the unique atmosphere and the history of this library.165

The Portrait Busts in the Duchess Anna Amalia Library

The portrait busts – of which close to 223 are accounted for in the library’s collection of a total of 313

sculptures, medallions and reliefs – are an outstanding feature of the art collection, which is

prominently displayed in the Rococo Hall.166 An early reference to the portrait busts in the Rococo Hall

dating from 1788 is noteworthy because it mentions the very unusual character of this collection of

busts: Next to the portrait paintings and busts of Saxon dukes and princes, “busts of now living scholars,

worked into stone and in plaster by the adept hands of the local court sculptor Mr. Klauer” are noted.167

The “now living scholars” this travelogue alludes to, were amongst others Johann Wolfgang von Goethe,

the philosopher and poet Johann Gottfried Herder, Christoph Martin Wieland and the translator,

163 Michael Knoche 1999b.

164 Nationale Forschungs- und Gedenkstätten der klassischen deutschen Literatur in Weimar.

165 Barbara Koch.

166 Gabriele Oswald, 1995, was the first to attempt a systematic overview and catalogue of the library’s dispersed sculpture collection, based on a two volume hand-written register of the collection, in order of materials, that had been established between 1848 and 1853, with numerous later addenda: “Verzeichnis der im Kunst=Cabinet auf Großherzogl. Bibliothek befindlichen Gegenstände”.

167 Friedrich Carl Gottlob Hirsching, Versuch einer Beschreibung sehenswürdiger Bibliotheken Teutschlands nach alphabetischer Ordnung der Oerter, 4 Vols, Erlangen 1786-1791, here 3rd Vol., 1788, p. 168, cited after Michael Knoche 2007b, p. 236: “Büsten jetzlebender Gelehrten, die von der geschickten Hand des dasigen Hofbildhauers Hrn. Klauer, in Stein und in Gyps gearbeitet sind.”

39

journalist and music teacher Johann Christoph Bode (1730-1793). These men were of common

descent, yet distinguished inhabitants of Weimar with close connections to the court – who regularly

visited the library to borrow books and to find themselves there, eternalized in paintings on the wall

and also as portrait busts on the high pedestals upon which they are partly still placed to this day

amongst the literary treasures.

The message of this self-confident display clearly indicated that there was no reservation in Weimar

between aristocracy and common people of special merit, the merit being scholarly or artistic

excellence. A wholly new library concept and self-manifestation is seen in this trend, one that

dissociates itself from the ceremonial formalisms of the past, which mainly promoted its own history

and lineage.168 The baroque portraits of the court’s forefathers were in fact gradually removed to the

upper galleries and replaced in the inner cell of the main room in the Rococo Hall by an influx of neo-

classical busts as well as by paintings portraying contemporary personages. This notable modernity in

the library concept of display is also reflected by an innovative directive issued on February 26th 1798

for the terms of library use, in which the opening hours (two mornings a week), the loan period (12

weeks) and also the admission of “young people” to the library is regulated.169

Although this new trend can be traced back to Anna Amalia, it was during the reign of her son, Duke

and later Grand Duke Carl August (1757-1828, sovereign since 1775), in which the exceptional

iconographic program was firmly established. Johann Wolfgang von Goethe is also a key figure for the

library’s inventory. He was called to Weimar in 1775, at the age of 26 and remained here for the rest

of his life, holding a succession of state offices and becoming Duke Carl August’s chief advisor. Goethe

was superintendent for all institutions of science and the arts after 1788. He shared the

superintendence of the Ducal Library with Privy Councillor Christian Gottlob von Voigt (1743-1819).

Carl August’s son, Grand Duke Carl Friedrich (1783-1853, sovereign since 1828) together with his

Russian born wife Maria Pawlowna (1786-1859), continued this process. So it was three generations

of the Weimar dynasty who, as contemporaries of the era around 1800, actively participated in this

exceptional installation of a memorial site for great men (and very few women) still within the

lifetime of those commemorated here and also in close proximity to these people, whom they all knew

personally.170

168 Michael Knoche 2007b, pp. 236-237.

169 Of the 475 registered library users in the period between 1798 and 1809, three dozen were grammar school pupils. Their parents or teachers signed their library tickets. For the directive of 1798 "Vorschrifft, nach welcher man sich bey hießiger Fürstl. Bibliothek, wenn Bücher ausgeliehen werden, zu richten hat", see Annette Seemann, pp. 38-44.

170 Bettina Werche, 2007a, pp. 163-164.

40

The singularity of the library collection and display cannot be stressed strongly enough: there was no

comparable library collection, let alone library presentation within Germany at the time. The usual

sculptural and pictorial programme then found in German libraries is that of casts of ancient

sculptures and portraits. Carl August and Goethe equally acquired a collection of casts of ancient Greek

and Roman sculptures and busts for the library. These were installed in different and varying locations

within the library and sometimes carried over to the Ducal Palace to serve for tuition in the Drawing

School. An inventory, which was established between 1848 and 1853 names 59 different plaster casts

of Greek and Roman original sculpture in the Art Cabinet.171 Yet this number is comparatively small in

relation to the 208 portrait busts of modern era and contemporary people.

Figure 13: View into the interior oval of the lower level of the Rococo Hall (from the opposite angle as in

Fig. 12) in 2009.

The Bibliotheca Augusta in Wolfenbüttel most probably acted toward the development of the specific

Weimar program, in that it displayed busts of deceased scholars. The library of Wörlitz Palace in

Anhalt-Dessau is another exception to the rule, in that contemporary philosophers, statesmen, poets

and writers were depicted in grisaille fresco-technique on the interior walls in amongst deceased and

ancient scholars. Anna Amalia, Carl August and Goethe knew the library in Wörlitz. Its programme of

display had been greatly inspired by Duke Leopold III Friedrich Franz of Anhalt-Dessau’s visits to

England and Italy in the 1760s.172

The marked transformation within late 18th century portraiture is closely linked to enlightened

171 Katharina Krügel.

172 Gabriele Oswald 2003, p. 99. Malcolm Baker 1995, describes a comparable display of busts in the Wren Library at Trinity College.

41

thought: the replacement of representative portraits of noble status by individualizing and

physiognomically oriented portraiture is set within a greater inflationary movement of very specific

and extensive interest in the genre of portrait busts, lasting well into the 20th century.173 So a

canonizing and social function is evident in these painted, sculpted and cast portraits filling the Rococo

Hall, representing a social trend. They attest to a culture of highly developed and busy literacy, in which

the production and an ever-wider distribution of portraits was also a form of communication.174

Neither those who wrote, nor those who read were willing any longer to do without the portraits of

the authors of the books they cherished - as tangible evidence of the portrayed character, spirit and

virtue.175

Goethe readily presented his own bust, modelled and then cast in 1780 by the above mentioned

sculptor, Martin Gottlieb Klauer – who from 1781 to 1801 was sculptor to the court and during that

time created 40 busts for the library collection – to a great many of his friends and acquaintances.176

Likewise he was in the habit of collecting not only casts of ancient sculptures and busts for his own

home, but also of contemporaries, to whom he wished to remain close. A letter to the writer, lawyer

and philosopher Friedrich Heinrich Jacobi (1743-1819) of whom Goethe received a bust in 1784,

demonstrates this: “Your memory is also vivid amongst us and we conversed with your bust, which is a

good likeness and of which I now own a cast, just recently. You will make your friends a good gift with it.

Even if the presence and the very delicate artistry might be to an extent missing in such a work of art, it

is worth a lot to the absentee. Now I have thought that plaster is very perishable, in a few years the first

and the best casts are more or less spoiled, therefore Klauer shall now carve a head out of Saxon marble

and if it turns out well, I shall be pleased. (…) This evening the Herders will come to visit and also Frau

von Stein. We will be thinking of you.”177 Herder, in a letter to the same Jacobi, dated on December 20th

1784, writes that he, together with his wife Karoline and Goethe, had thought of Jacobi with “devoted

173 Bernhard Maaz 2001; Ursula Merkel; Ursel Berger.

174 See Malcolm Baker 2001, pp. 28-30: Fine finishing of surface details encourages the viewers engagement into equal dialogue with the portrayed, just as much as the typical treatment and working of the socle and the placement of a busts at eye height. The artificial nature of a bust as a truncated fragment of a body is the enhancing factor for the semblance of reality.

175 Ulrike Steierwald, pp. 96-97.

176 On substantial serial manufacture of a bust in context to the celebrity of the subject, see John Cage, p. 36.

177 “Dein Andencken ist unter uns auch lebendig und wir haben uns neulich mit deiner Büste unterhalten die recht gut gerathen ist, und wovon ich nun einen Abguß besitze. Du wirst deinen Freunden ein angenehm Geschenck damit machen. So ein Kunstwerck wenn auch die Gegenwart und der ganz delikate Kunstsinn manches daran vermisst, bleibt doch für den Abwesenden sehr viel werth. Nun habe ich gedacht der Gyps ist sehr vergänglich in einigen Jahren sind die ersten und besten Ausgüsse mehr oder weniger verdorben, deswegen soll Klauer nun einen Kopf aus sächsischem Marmor hauen und wenn er geräth werde ich mich sehr freuen. (…) Heute abend kommen Herders zu mir und Frau v. Stein. Wir werden dein gedencken.” Letter from Goethe to Friedrich Jacobi, dated 12th November 1784, Johann Wolfgang von Goethe, cf. Note 13, WA IV, 6, p. 386-388,

42

love” and had laired around his likeness “as to a monument of (his) presence”.178

Of the 223 busts, which form part of the art collection of the library, 11 portray ancient Greek and

Roman heads, four allegorically portray the elements and 62 portray aristocracy, who are passed and

contemporary members of the house of Saxe-Weimar-Eisenach and of other related feudal houses,

like Saxon-Coburg, Saxon-Gotha and the Prussian monarchy. Busts for example of King Maximilian I. of

Bavaria (1756-1825) and of Tsar Alexander I. of Russia (1777-1825) or of Otto Duke of Bismarck

(1815-1898), can also be classified as representations of dynastic and political interests.

Quite a number of foreigners are celebrated here in the form of portrait busts. For instance the

collection holds three busts of William Shakespeare (ca. 1564-1616) and one of the French writer

Guillaume Thomas Raynal (1711-1796). The French writer Voltaire (1694-1778) is portrayed here, as

is the Bavarian Christoph Willibald Gluck (1717-1787), by the famous Jean-Antoine Houdon (1741-

1828), who was much admired by the Duchess and her son Carl August. Johann Kaspar Lavaters

(1741-1801) close connection to Goethe and the impact of his “Physiognomische Fragmente zur

Beförderung der Menschenkenntnis und Menschenliebe” on late 18th and 19th century portraiture explain

the relatively early entry of the bust of this Swiss poet and physiognomist into the library collection.179

Busts of Napoleon Bonaparte (1769-1821), of Arthur Wellesley, Duke of Wellington (1769-1852) and

of Madame Dubarry (1743-1793) also form part of the collection of plaster busts, but seem, in

comparison to the colossal busts of the Italian poets Torquato Tasso (1544-1595) and Dante Alighieri

(ca. 1265-1321), to have been collected in another context.

The core group of contemporaries portrayed in Weimar’s Ducal Library as busts though – Goethe

(fourteen portrait busts, two of which marble), Herder (four busts, one of which marble), Schiller (four

busts, one of which marble, and his death-mask) and Wieland (three busts, two of which marble) – was

consistently expanded between 1782 and the 1850s and beyond, to include other poets and writers,

philosophers, musicians, actors, painters, sculptors, clergymen and librarians. They all stood in some

relationship with Weimar, the library and the court or can at least be associated by the appreciation

they received here.180 To name only a few of the better known of these personages, are the

philosopher Gotthold Ephraim Lessing (1729-1781), the art-historian and archaeologist Johann

178 Herder, Briefe 5, Nr. 73, p. 89, cited from Gabriele Oswald 2003, p. 104: “voll treuer Liebe” and “als an einem

Denkmal Deiner Gegenwart gelagert”.

179 Lavaters “Physiognomical Fragments for the Advancement of Knowledge on Human Nature and Philanthropy”, published in four volumes between 1775 and 1778, is described as the cause for an (economic) boom in sculpted portraiture, Petra Rau, p. 61. John Cage describes the impact of Lavater’s Essays on Physiognomy and Franz Joseph Gall’s phrenological studies on British portraiture as well as of the use of sculpted busts in support of phrenological theories, pp. 41-46.

180 Cf. Malcolm Baker 2001, on a group identity of portrait busts through their shared convention, p. 22: “Entering a space where many busts are displayed challenges us to register them as part of a larger communal whole – members, as it were, of a club – while coming face to face with them as individuals.”

43

Joachim Winckelmann (1717-1768), the dramatist August von Kotzebue (1761-1819), the author and

philosopher Novalis (1772-1801), the composer Georg Friedrich Händel (1685-1759), the composer

Ludwig van Beethoven (1770-1827), the philosopher Immanuel Kant (1724-1804), the fairy-tale

author, theologian and teacher Johann Carl August Musäus (1735-1787), the actor and dramatic

author August Wilhelm Iffland (1759-1814), the painter Friedrich Preller the elder (1804-1877) and

the sculptors Bertel Thorvaldsen (1770-1844) and Christian Daniel Rauch (1777-1857)181

When the “now living scholars” and artists died however, their busts remained in their position within

the hall. Instead of replacing those who had deceased gradually with busts of contemporary and

topical artists or scholars, who were equally connected to Weimar – for example perhaps Franz Liszt

(1811-1886), Richard Wagner (1813-1883) or Friedrich Nietzsche (1844-1900) – the neo-classical

period remained fixed in the Rococo Hall and was evermore enhanced. Everyone who had some

context to this famous period of Weimar Classicism stood a good chance of being integrated into the

library hall in the form of a painting, an etching or a cast bust. Reverend Röhr (1777-1848) for

instance, would have remained completely unknown, had he not held the funerary elegy on Goethe in

1832, and hence his plaster portrait is amongst these 223 busts. And in 1848 – the year in which

Weimar largely celebrated Goethe’s 100th birthday – the Rococo Hall was chosen as the location for

the festivities, as the most manifest site of and for Weimar Classicism.182

Figure 14: The first gallery of the Rococo Hall, 2008.

181 See Gabriele Oswald, 1995, pp. 237-271, for an alphabetical list of all 300 sculptures by name of those

portrayed.

182 Michael Knoche 2007b, pp. 238-239.

44

This staging and memorialisation of a special epoch has remained the inherent programme for the interior

of the Rococo Hall, also long after the ever-expanding library holdings far outgrew their own dimensions and

tapped into 20th century sources with all the implications and upheavals the various extreme political

settings had on literature in Germany in that century. This special period, called Weimar Classicism, has

remained untouched and adulated throughout and the Rococo Hall deeply committed to the memory of a

time past.183

On the Sculptors, Materials and Provenances of the Portrait Busts

Only approximately 20 of the portrait busts were worked in marble and a few in a local limestone, the

rest, apart from very few terracotta works, are all plaster casts. The difference in price between a

marble bust and a plaster was considerable and easily explains this ratio of sculptural materials in the

library collection. Naturally the prices paid for plaster casts also depended on the originality of the

artwork and on its serial distribution.

The first sculptor to be employed by the court was Martin Gottlieb Klauer in 1781, who had been

commissioned for portrait busts by the Duchess since 1773. The busts he created of Duchess Anna

Amalia, of her husband Ernst August II Konstantin and of Christoph Martin Wieland in 1780 and 1781

for example, cost 26 Taler184 a piece.185 This is a price for a commissioned and original artwork, that is

a portray, first modelled in clay then cast from a mould and re-cast from the first plaster model. This

price should not be confused with those of casts sold by art dealers, for example Carl Christian

Heinrich Rost (1741-1798) in Leipzig, for which a relative mass serial production can be presumed.

Rost was one of Goethe’s main suppliers for copper etchings and books, he had also distinguished

himself as an important contractor and dealer for casts after ancient sculptures and also for casts of

portraits of famous people. His third sales catalogue of 1786 informed those interested, that he had

acquired the best moulds of the Ferrari brothers,186 which made him one of the most sought after

dealers for plaster casts in the region, since he had the sole rights also for the models of the Dresden

183 Ingrid Arnhold.

184 The Reichstaler, was a standard silver coin of the Holy Roman Empire, it was introduced in 1754 and weighed about 23,93 grams of silver, one tenth of the Cologne Mark, which was a unit of weight equivalent to 233.856 grams.

185 From Klauer’s contract book 1779-1797 GSA: 96/1576. I owe this information to Ulrike Müller-Harang of the Foundation of Weimar Classics, whom I wish to thank very much.

186 Oswald 2007a, p. 284, states evidence of Wieland’s communications about the Italian brothers Giugio and Giacomo Ferrari, from whom the Duke and Goethe bought various casts, which were accounted for in the ducal casket accounting book (Schatullrechnungen) of 1776.

45

collection of Greek and Roman originals and over many models from the Vatican museums.187 He

commissioned etchings to illustrate the statuary he had on offer in a second volume to his sales

catalogue of 1794, which he sold separately from the text volume for the comparatively steep price of

2 Taler – which was approximately his price for a plaster cast of a bust.188 In comparison, casts of

Ludwig Klauer’s (the younger Klauer) busts of Schiller and Wieland were advertised as casts for 4 Taler

each in a Weimar Journal in 1805.189 The difference in price to the Rost casts might be accounted for

by the fact that these casts were original in the sense that their author was selling them from his own

workshop – even though their manufacture can already be termed as serial reproductions. The fact

that these two bust were brand new productions, both completed in 1805 – Schiller’s bust after his

death mask – might well also have had an impact on the price.

In his position as court-sculptor and teacher at the drawing school (1781-1801), Martin Gottlieb

Klauer received a yearly salary of 200 Taler, for which he had to produce, amongst other duties, one

bust a year for the library. The Berlin born sculptor Christian Friedrich Tieck (1776-1851) succeeded

Klauer in this position in 1801, but only remained in Weimar for a few years due to the insufficient

contract situation.190 He in turn was succeeded in 1805, by Carl Gottlob Weißer (1780-1815), who had

trained with Tieck. Weißer profited from the completion of construction works on the new Ducal

Palace, which had demanded a large amount of funding,191 and was thus commissioned to copy Tieck’s

plaster bust of the librarian Voigt into white Carrara marble at some point after 1805.192

Carrara marble is a material that is comparatively much more expensive to acquire and it demands a

much longer time in working. Weißer worked a total of 7 marble busts in the time between 1805 and

1812.193 Amongst those he portrayed in marble was Anna Amalia’s brother, Friedrich August Herzog

von Braunschweig (1740-1805), who had died while visiting his sister’s court. This marble bust was

contracted by Duke Carl August. The accounting book for the art-works for the library notes 44 Taler

in payment to Weißer for this bust in 1806.194 By comparison, the price that was paid to the sculptor

187 Anzeige aller Kunstwerke der Rostischen Kunsthandlung zu Leipzig, Zweyte Abteilung, Leipzig 1786.

188 Petra Rau, 2003, pp. 62-65 and catalogue entry 84, in: Antlitz des Schoenen, p. 268.

189 Journal des Luxus und der Moden (Weimar) Vol. 20, 1805 Tome 11, pp. 736-737, again thanking Ulrike Müller-Harang.

190 Gabriele Oswald, 1999, pp. 98-99.

191 Duke Carl August and his family moved into the completed eastern wing of the new palace building in 1803.

192 Gabriele Oswald 1995, pp. 37-38.

193 Ibid.

194 See Gabriele Oswald, 2007b, p. 225, ThHStA Weimar, Sign. A 1037. It is also not clear whether Weißer received this sum on top of a yearly salary as court sculptor.

46

Johann Heinrich Dannecker for his marble bust of Schiller in 1806 by Schiller’s friend and brother-in-

law Wilhelm von Wolzogen, was 500 florins – approximately 7.5 times the amount paid to Weißer for

a marble bust.195 In 1826 Carl August bought this same marble bust of Schiller for the library out of

Charlotte Schiller’s legacy for the sum of 200 ducats, which was more than 12 times the amount

Weißer had received for his work.196

A total of 73 sculptors works form the library’s sculpture collection. Dannecker was one of the leading

German sculptors of his time and ranks with Alexander Trippel, Johann Gottfried Schadow, Christian

Daniel Rauch, Ernst Rietschel, Christian Friedrich Tieck, Ludwig Wilhelm Wichmann, Adolf von

Donndorf, Ludwig Schwanthaler, and on an international level with Bertel Thorvaldsen, Antonio

Canova, Jean-Antoine Houdon, Jean Pierre Antoine Tassaert and David d'Angers. These sculptors are

all represented in the library’s collection. Their international acclaim mingles in with works of the

leading sculptresses of their time, Angelica Bellonata Facius and Elizabet Ney – as well as with those of

the regionally famous Klauer, Weißer and Johann Peter Kaufmann in Weimar and with those of

Friedrich Wilhelm Doell and Friedemann Hunold, court sculptors in nearby Gotha, respectively in

Dessau.197

Peter Kaufmann (1764-1829) had trained with Antonio Canova in Rome and succeeded Weißer in the

position as court sculptor in Weimar in 1817, against Goethe’s initial objections. Goethe’s concerns

toward fixing a new position as court sculptor pertained not only to Weißer’s sad and solitary

existence, which had ended in suicide, but to the comparatively bad contract situation Weimar offered

to sculptors, which did not allow for establishing of a solid sculpting tradition.198 Kaufmann’s

additional adeptness as a model-maker and caster finally convinced and so he became the author of

four of the marble busts in the library’s collection.199

195 Gabriele Oswald, 1995, p. 135. Florin was the older name for Gulden or Guilder, both these names originated

from the gold coins they described, but became common names for silver or base metal coins, one Gulden or Florin was worth approximately 2/3 of a Taler at this time. According to http://de.wikipedia.org/wiki/Gulden#Deutscher_Bund, 500 Florin at this time would have been worth 333 Taler.

196 Gabriele Oswald, loc. cit. The Ducat is a gold coin with a fixed AGW of 0.1107, which is 3.4909 grams of .986 gold. One Ducat was worth 2 2/3 Taler and 4 Florin. So, the price Carl August paid in 1826 would have been 800 Florin or Guilder, or 532.8 Taler. (This hypothetical calculation was made on the basis of the relative monetary values given on the mentioned Wikipedia-site, note 195) and does not take into account any differences in values between those of the Holy Roman Empire or those of the States of the German Confederation.)

197 Bettina Werche 2007a, pp.166-167. Gabriele Oswald 1995, pp. 232-239 lists the 73 sculptors in alphabetical order.

198 Described in three of Goethe’s letters to Carl August of 26th and 27th May, and 14th of June, 1816, Johann Wolfgang von Goethe, cf. note 13, WA, IV, 27, pp. 26-32 and 56-58.

199 Gabriele Oswald 1999, pp. 99-100.

47

The portrait busts provenances are sometimes recorded in the two volume hand-written “Register of

Objects in the Art-Cabinet of the Grand-Ducal Library”, which was established between 1848 and 1853,

with numerous later addenda.200 28 of these records indicate that the respective busts were presented

to the library, either by the portrayed himself, by the sculptor or by a third party. Two thirds of those

objects, which have no known provenance, are estimated to have most likely originated from the

ducal family’s property.201 The reasons for their respective order and the date of their accession to the

library require further detailed archival research.

On the History of Maintenance of the Plaster Busts

As outlined above, the primary focus of research on the portrait busts of the Duchess Anna Amalia

Library was the identification of the items of the libraries collection, which in the 19th and 20th

centuries had been dispersed throughout the various different institutions and holdings that today

form the Foundation of Weimar Classics.202 Accession date and respective positioning of the different

portrait busts within the Rococo Hall have since been the focus of more detailed investigations.203 So

far however, no research at all has concentrated on maintenance and treatment of the plaster busts of

the library collection. It was only the restoration campaign in reaction to the fire disaster of 2004 that

prompted this query into the history of the plaster busts surface treatments and it is hoped that future

archival research might follow up on some of the observations and questions raised within this

investigation.

With three exceptions, the appearance of the 65 plaster busts positioned in the Rococo Hall prior to

the fire of September 2004 was one of slightly to heavily soiled over-painted plaster surfaces.204 These

white or off-white coatings of paint of varying material thicknesses greatly suffered under the impact

of fire extinguishing water. A thorough wetting of the busts led to an elevation of the paint layers,

which in many areas ripped open. In most cases however, the coatings protected the underlying

plaster surfaces from the force of water. A mixture of soot, smoke, gas and chemical additives to the

fire fighting foam – which was also implemented in some areas of the Rococo Hall – settled as a sticky

200 “Verzeichnis der im Kunst=Cabinet auf Großherzogl. Bibliothek befindlichen Gegenstände”, transcribed for

the collection of sculpture by Gabriele Oswald 1995, pp. 129-229.

201 Gabriele Oswald 1995, p. 128.

202 Gabriele Oswald 1995.

203 Bettina Werche 2001, Bettina Werche 2007b; Katharina Krügel.

204 Three plaster busts were not treated with a coating i.e. left bare. Six further plaster busts, which were installed in the hall at the time, had bronze or terracotta material imitating coatings. They do not form part of this study.

48

film onto the surfaces, into which particle matter adhered. The immediate emergency response action

on the plaster busts in the night of September 2nd, 2004 was their salvaging into an underground

storage area. Their thorough drying was subsequently closely monitored under controlled

environmental conditions and lasted over a period of up

Figure 15: The bust of Johann Wolfgang von Goethe was severely damaged at the base.

Yet the over-paints had to a large extent protected the plaster surfaces from the direct impact of fire-extinguishing water.

to one and a half years. Damage assessment and conceptualization of the conservation plan then

called for the characterization of the nature and contextual value of the over-paints – to weigh the

necessity for their conservation against their removal. Their partly heavy deterioration was finally as

much reason behind the decision for their removal as was the fact that the multiple superimposed

paint layers had in some areas accumulated so much material thickness as to hide surface detail to an

unacceptable extent. The contextual value of the over-paints was questioned under the aspect of their

originality, their function and with the broader perspective of comparison to similar surface

treatments in other collections of plaster casts.

The originality of the paint layers was questioned with the presumption that an originally intended

marble imitation would have been achieved on a relatively new and unsoiled plaster surface. Similarly,

the application of a coating of paint for the prevention of dust-settlement can also be presumed to have

been applied to a bare and clean plaster surface before it was put on display. Stratigraphical analysis

however, not only revealed dirt layers in between various different paint layers of different shades of

off-white on these busts surfaces, but also dirt layers covering the underlying plaster surface. The

initial presentation of the portrait busts in the Rococo Hall as well as in the Art Cabinet in the annex

49

building of 1804, was therefore in most cases identified as that of bare plaster surfaces which were

then subject to dust-fall over a period of unknown time, before a first application of a surface coating.

In most cases, more than one coating was uncovered, in one instance up to seven different layers were

superimposed on top of the plaster surface, always revealing upon their removal a stratigraphy of

soiling in between the layers of paint. The uncovering of the paint layers during the restoration and

conservation campaign further revealed that very often, apart from soiling, damage histories had lain

hidden underneath the coatings of paint.

Figure 16: During stripping, the reconstruction of the rear of the head of Herder’s bust showed underneath

three coatings of paint on top of a washed down surface.

In fact, a surprisingly large amount of damage histories and repair interventions to the busts were

uncovered during the restoration campaign. The fragility of plaster accounts for the high quota of

damages to these busts’ installation in a publicly accessible space, which was always effective as a

library. In addition, a relatively busy movement history, which these busts must have experienced,

suggests itself on the basis of the above outlined collection accumulation and subsequent

distributions and repositioning of the busts over various parts of the historical library as well as

during the various stages of the collection’s dispersal. Temporary transferrals of plasters from the

library to the drawing-school for educational purposes is documented for the collection of casts of

ancient sculptures and might well also apply to the collection of portrait busts. Goethe’s refusal to a

loan request for Dannecker’s plaster Schiller-bust of 1794 from the library for the adornment of a

festive ceremony in 1802, on “the very natural grounds that a plaster bust has yet not ever been

returned from a festivity in an undamaged state”205, is further support for the assumption of an

205 “Schillers einzige Original-Büste, auf der Weimarischen Bibliothek befindlich, eine frühere herzliche Gabe

Danneckers, wurde zu jenem Zwecke verlangt und aus dem ganz natürlichen Grunde abgeschlagen, weil man noch nie eine Gypsbüste unbeschädigt von einem Feste zurückerhalten habe.” Johann Wolfgang von Goethe, see note 13, WA I, 35, p. 124.

50

extensive movement history for this collection of plasters.

Some detailed material and stratigraphical analyses of five busts surface coatings form part of this

investigation and will be described below. The results prove that the various coatings of paint were

applied on the plaster busts in the Duchess Anna Amalia Library as corrective measures. But while one

can assume that the intent behind the application of white to off-white shades of paint onto pre-soiled

or pre-painted surfaces was not one of exact marble imitation inasmuch that no marbling texture was

evident in any of these coatings,206 it is apparent that the general appearance these busts were to

maintain (by these over-paints) was white to off-white. This is not surprising, considering white-

coloured statuary was the categorical imperative to neoclassical aesthetics as first formulated by

Winckelmann in 1764207 and arrogated amongst others by Kant, Herder and Hegel.208 The arguments

brought forward – that sculpture needed to eschew any colour or colouristic effect because its essence

lay in the tactile exploration of shape, and that only in sculpture could the perfect fusion of the mental

and spiritual with the physical and sensual and take place and thus needed to be partly abstracted and

idealised by the colour white – were not exclusively German idealist thinking.209 The quintessential

understanding that only a white sculpture is “true” and not only the form and contour, but the spirit,

mind and character of a person can only, justly and respectfully be portrayed in a white material thus

forms the intrinsic context to these busts appearances. A question that poses itself within this

understanding is the point of time within this collection’s history at which these surfaces received

their layers of “whitening”.

Experimental 1: Analysis of Cross-Sections of Paint-layers taken from Five Busts Selected by Criteria of their Accession Date to the Collection

The aim of this experimental part of research is to analyze the paint layers on the plaster surfaces in

the Rococo Hall. As already stated, experience during restoration was that dirt and repair histories had

206 White coloured coatings on plaster are quite commonly associated with marble imitations, see for example

Barbara Appelbaum, p. 173 or Moritz Kidderlen, p. 109.

207 White having the quality of rendering the objects appearance larger than life and reflecting light from it, thus being both aesthetically and symbolically significant for sculpture: Johann Joachim Winckelmann, Geschichte der Kunst des Altertums, Dresden 1764, in: Studien zu Deutschen Kunstgeschichte Vol. 343, Baden-Baden 1966, VIII, 4,2 §19.

208 Cf. Hanna Philipp, Hans Ulrich Cain and Ingeborg Kader. See especially Karina Türr, pp. 95-124, on a heated debate about polychromy in sculpture, which lasted well into the 20th century.

209 On the colour white in neoclassical statuary in a European context cf. Alex Potts.

51

lain covered underneath these coatings. Seeking to affirm these experiences analytically, five busts

surface treatments were singled out for analysis. The selection of five samples of over-paint210 was

conducted solely by criteria of the respective busts accession date into the Rococo Hall. The intent

behind this method of selection was to “cross-section” into the most possibly large time-frame of

surface exposure within the library hall and into the treatment history of these plaster surfaces.

Approximately 120 years history of surface exposure and maintenance practice are represented here

as spot samples:

Figure 17: Raynal’s bust was seriously damaged in the night of the fire.

The plaster bust of the French writer Guillaume Thomas Raynal of 1782 by Jean Pierre Antoine

Tassaert is confirmed as having been placed in the Rococo Hall since 1783.211 It is listed in an

inventory of 1818-1851212 and also in the inventory of 1848-1853213. Fig. 17 visualizes the severe

210 Samples of over-paint had been collected during restoration, cf. note 217.

211 See Bettina Werche, 2001 and 2007b: Jean-Baptiste Gaspar d’Ansse de Villoison’s Epistolae Vimarienses, Zürich 1783.

212 Repertorium Großherzogliches Kunstkabinett 1818-1851, p. 13.

52

damage this bust experienced in 2004. As in all five cases of this investigation, a small sample of the

coating was embedded in casting resin and prepared for microscopic analysis as well as photography

in reflected light and with UV-fluorescence excitation.214 Two layers are clearly discernible in Fig. 18,

which shows the sample with a magnification factor of 100. The first layer is white and comparatively

thicker and denser with a few larger inclusions.

Figure 18: Microscopic inspection of the cross-section of Raynal´s coatings reveals two layers

of paint, between which there is soiling.

The top layer has a conglomerate consistency; many smaller and larger inclusions make up a rough

matrix. The hue is less white and seems comparatively slightly yellowish. Fig. 19 shows a small trace of

dirt below layer 1 and very clearly, a layer of dirt between layers 1 and 2. At double magnification the

dirt layering in between the paint-stratigraphy is more evident and the image produced with UV-

animation (Fig. 19) highlights the interlaying as well as the surface soiling.

213 Verzeichnis Kunstbesitz Bibliothek 1848-1853 XXIIIb, No. 154.

214 The analyses were carried through at the Technical College Erfurt by Frank Mucha. Casting resin: Technovit, 2000 LC, polished with Micromesh, grain sizes 1500-12000. Microscope: Axioplan, Zeiss. Camera Axiocam MRc 5, Zeiss. Software: Carl Zeiss Axiovision Rel. 4.7. Microscopic Photography: reflected light, dark field, zoom factors 50-200 and UV-fluorescence excitation with filters FT (510 nm) and LP (520 nm). See Apendix III.

53

Figure 19: UV-animation of the same cross-section, cf. Figure 18.

The plaster bust of Johann Wolfgang von Goethe of 1790 by Alexander Trippel is listed in the

inventory of 1818-1851 and again in the inventory of 1848-1853.215 Its exact date of accession to the

library collection however is not known, but one safely can assume it to have been before 1818.

Goethe’s surface stratigraphy proves very similar, if not identical to that of Raynal. Figure 15 shows the

conditions prior to restoration. The cross-section (Fig. 20) gives evidence of two layers of paint. The

first thin layer is again more dense and white and lies on soiled plaster substrate. The second layer is

thicker, its composition is coarser and it has a yellowish tint.

215 Repertorium Großherzogliches Kunstkabinett 1818-1851, p. 12; Verzeichnis Kunstbesitz Bibliothek 1848-

1853 XXIIIe, No. 229.

54

Figure 20: 200-fold magnification of the cross-section into Goethe’s surface.

Figure 21: UV-investigation highlights the soiling.

55

Figure 22: The bust of Torquato Tasso prior to restoration.

The exact date of accession of the plaster bust of Torquato Tasso of 1816 by a C. Petscherbur216 of

Russia to the Rococo Hall is also not clear. It is listed in the inventory of 1848-1853 and might well

have been donated to the library by the Russian Princess Maria Pawlowna, who married Carl Friedrich

in 1804.217 Its installation in the library therefore occurred at some point during the first half of the

19th century, between 1804 and 1849. Fig. 22 shows the condition of damage prior to restoration. 50-

fold magnification of the cross-section, reveals staining as well inclusions in the plaster substrate, (see

Fig. 23). Again, two layers of paint are discernible and again, soiling is evident to the plaster surface

underneath the first coating. At 200-fold magnification (Fig. 24), these two layers of paint seem

identical to those on Raynal’s and Goethe’s surfaces, with a coarse-grained yellowish top coating on a

denser and whiter first coat of paint. Here, as in the case of Goethe (Figs. 20 and 21), there is no

apparent soiling between layer 1 and layer 2.

216 Neither is it clear if C. Petscherbur in fact stands for an artists signature or for the city of St. Petersburg, cf.

catalogue entry 17.11-17.15, in: »Ihre Kaiserliche Hoheit« Maria Pawlowna – Zarentochter am Weimarer Hof.

217 Verzeichnis Kunstbesitz Bibliothek 1848-1853 XXIIIb, No. 114.

56

Figure 23: A cross-section of Tasso’s surface with magnification factor 50.

Figure 24: Magnification factor 200 of the same cross-section shows dirt deposition

on the plaster surface.

57

Figure 25: Rauch’s bust’s surface coating was heavily damaged by fire-extinguishing water..

The plaster bust of Christian Daniel Rauch of 1857 by Ernst Friedrich August Rietschel (Fig. 25), is

clearly indicated as having been donated to the library by Grand Duchess Maria Pawlowna in 1857.218

A cross-section of the surface at 200-fold magnification (Fig. 26) unfortunately proves no adhering

plaster substrate to this sample. The surface condition was however soiled, as is also evident in Fig. 25.

Dirt depositions are evident in the cross-section, in between a first and a second layer of paint. This

second layer seems very similar to the respective second layers on Raynal, Goethe and Tasso. It has an

irregular texture and is slightly yellowish. A third and coarser layer is applied on top of layer 2.219

218 Verzeichnis Vermehrung Kunst- und Münz-Kabinett 1818f. p. 247: April 1857 “1857 Present of Grand Duchess

Maria Paulowna”. “1857 Geschenk der Großherzogin Maria Paulowna”.

219 This sample unfortunately was too fragmented to perform any material analysis.

58

Figure 26: A cross-section of Rauch’s surface-coatings in 200-fold magnification.

Figure 27: Schiller’s bust prior to restoration shows water

damage and soiling.

59

The plaster bust of Friedrich von Schiller of 1794 by Johann Heinrich Dannecker is clearly indicated

to have gained access to the collection in 1896, the year it was cast after the original of 1794, which it

replaced in the library in 1896.220 Figure 27 shows the state of damage to the bust before restoration.

One layer of paint is evident in the cross-section (Fig. 28). The image produced with UV-light allows

clear discerning of soiling on the plaster substrate as well as on top of the coating of paint (Fig. 29).

Figure 28: A cross-section into the bust of Schiller’s surface reveals one coating of paint.

Figure 29: UV-light highlights soiling on the plaster and also on top of the coating of paint. 220 Verzeichnis Kunstbesitz Bibliothek 1848-1853 XXIIIb, Nr. 109.

60

Bust Raynal Goethe Tasso Rauch Schiller

Accession to the Rococo Hall

1782 or 1783 before 1818 before 1848 1857 1896

Number of paint layers

2 2 2 3 1

Soiling on plaster substrate

yes yes yes n.a. yes

Soiling in between layers of paint

yes not detected not detected yes n.a.

Analysis of binding media,

layer 1:

fatty tempera

- - - alkyd resin

Pigment-analysis,

layer 1

zinc white - - - lithopone

Analysis of binding media,

layer 2

alkyd resin

traces of paraffin

- alkyd resin - n.a.

Pigment-analysis,

layer 2

- - lithopone - n.a.

Presumed date of application

layer 1

after 1857,

perhaps before 1896?

after 1857,

perhaps before 1896?

after 1857,

perhaps before 1896?

after 1857,

perhaps before 1896?

after 1896,

presumably after 1930s,

perhaps 1941?

Presumed date of application

layer 2

after 1896,

probably after 1930s,

perhaps 1941?

after 1896,

probably after 1930s,

perhaps 1941?

after 1896,

probably after 1930s,

perhaps 1941?

after 1896,

probably after 1930s,

perhaps 1941?

n.a.

Table 1: Summary of Experimental 1

Summarizing, these cross-sections represent a time-span from 1783 through to 2004, in which the

respective plaster surfaces were kept subject to the library environment and received various

maintenance treatments. These treatments can be characterized as over-paints onto pre-soiled

surfaces. Analyses of binding media to layers 1 and 2 on Raynal resulted in the identification of a

mixture of protein colloid (animal glue), fat and oils for paint layer 1, which can be defined as a fatty

tempera. The binding media used in paint layer 2 on Raynal was identified as an alkyd resin with

linseed oil as an additive. 221 Comparative analysis of the binding media used in layer 2 on Tasso and

221 Both analyses were conducted with THM- (Thermally assisted hydrolysis and methylation) Pyrolyse-

Gaschromatography/Mass-spectrometry (THM-Py-GC/MS), cf. Apendix III, p. 19-22.

61

respectively in layer 1 on Schiller revealed an alkyd resin with probably linseed oil as an additive.222

Pigment analysis on paint layer 1 on Raynal revealed zinc white.223 The pigment used in paint layer 2

on Tasso, respectively in paint layer 1 on Schiller was identified as lithopone, a mixture of barium

sulphate and zinc sulphide.224 The yellowish hue to this paint layer was not identifiable as a pigment. It

is of organic nature and might well stem from the combined impact of soiling and fire extinguishing

water.225

The results of these analyses can naturally only to be regarded as comprehensively valuable for these

selected five surface histories and can not offer more than a rough idea of the entire collections

surface histories. Visual and material experiences during restoration had in fact often ascertained

more than two different layers of paint. However these findings by microscopic and material analyses

into five probes have brought some valuable insights into material qualities of the over-paints on the

plaster surfaces in the Rococo Hall and prove that a comparison of five different stratigraphies can

undoubtedly promote some understanding for the respective timing in maintenance history as

identified by these over-paints:

One first layer of tempera and zinc white could be identified for Raynal’s surface and is visually

identical to the first layer of paint on Goethe, Tasso and Rauch. This first layer was applied onto pre-

soiled surfaces. Since Rauch’s bust was installed in the Rococo Hall in 1857 and the stratigraphy of its

surface coatings gives visual indication of the (same?) first white paint layer, one might thus assume

the year 1857 as the terminus post quem for the application of a first white layer of paint. One can also

go on to assume that it might have been applied to these surfaces before 1896 – the year of accession

of the bust of Schiller to the library hall – because the bust of Schiller did not receive this type of

coating. It is of course equally possible that Schiller’s bust was at this unknown point in time (when a

first white coating was applied to the four other surfaces discussed here) already installed in the

library hall, but did not receive this same treatment, because its surface was not yet soiled.

Comparative analysis of paint layer 2 on Raynal and Tasso and of layer 1 on Schiller revealed it to be

composed of alkyd resin with linseed oil additive and lithopone. This coating must have been applied

to these surfaces at some point after 1896, which is the year in which Schiller’s bust was installed in

the library. Research into the use of alkyd resins does suggest that this specific mix of binding media

222 With THM-Py-GC/MS, cf. Apendix III, p. 23-26.

223 With x-ray fluorescence analysis and additional micro-chemical spot-testing, cf. Apendix III, p. 27.

224 With x-ray fluorescence analysis and Fourier-Transform Infrared-Spectroscopy (FT-IR-Spectroscopy), cf. Apendix III, 28-31. Lithopone is also known as Orr's Zinc White, after John B. Orr. It was patented and first manufactured in 1874 in Glasgow, see Rutherford John Gettens, George Leslie Stout, p. 125. In Germany lithopone was manufactured since 1876, see Bernd Hering, pp. 135-145.

225 As assumed by Frank Mucha of the laboratory of the Technical College in Erfurt.

62

was not known before the early 1930s.226 A systematic and ‘radical’ over-painting of the entire

interior of the Rococo Hall occurred in 1941.227 This might have been the point in time for the

application of the second layer of paint on the bust’s surfaces (respectively layer 1 on Schiller).

Thorough archival research into order and contracts of restorative works within the library hall might

probably reveal more definite dates for these two surface applications. The similarity between the five

surface stratigraphies does lead to the assumption that these two different coating interventions could

well have been carried through on larger scale bases. Some commands for a “whiting” of the plaster

surfaces in the Rococo Hall or a record of payment for such a presumably large order might be

traceable. The time frames these findings suggest for an archival research are the later part of the 19th

century, respectively the 1930s and later, perhaps in 1941.

226 Tom Learner, pp. 77-78.

227 Matthias Grunewald, p. 88, citing: Die Allgemeine Thüringer Landeszeitung, Issue No. 220, August 12th, 1941.

63

Excursion:

Coating Plaster Surfaces – a German Tradition in Preventive Practice?

There are in fact two questions indicated by the question mark in the title to this section. They formed

the initial motivation for this investigation, because a re-coating of the plaster surfaces in the Rococo

Hall is in continued discussion for reasons of plaster’s susceptibilities and in keeping with the

tradition in the maintenance of these surfaces. While always referring to the relatively thick white

over-paints – as illustrated with the analyses on properties of the paint layers in the Rococo Hall and

described above – and not to sealing films on plaster which were applied during casting procedures –

the first of these two questions therefore reads: is it a purely German phenomenon to apply paint to

plaster statuary? And the second question reads: can one at all categorize these over-paints as

preventive measures against dust-fall? The first question will be largely answered in the next section,

which is formed by the evaluation of a survey: A questionnaire forms part of this research and looks

into current practices of applying or re-applying a coating to plaster surfaces. It surveys into

experiences with existing over-paints, as gathered by colleagues who care for plaster statuary in other

collections, also outside of Germany. In fact, the evidence for layers of paint as surface treatments to

plaster statuary is indeed huge.

Perhaps because of plaster’s manifold use in casting procedures, which always entails the application

of agents for release and lubrication to plaster surfaces, surface sealing or “finishing” is a very

common phenomenon on plaster and was traditionally realized with varnishes, wax, oil, milk or other

(see above, p. 20).228. One can safely assume the level of inhibition toward applying sealing or coating

agents to plaster to be quite low – even lower, if there is already a prior coating agent on a plaster

surface. But while sealing agents, which are applied for casting procedure will consist of a very fine

layer of shellac, oil, wax or soap, the paint layers encountered on the surfaces of the busts in the library

have a different quality, in that they are thick and non-transparent and were obviously aesthetically

motivated when creating uniform white or off-white surfaces and purposefully hiding heavy soiling

and damage histories.

The surface treatment of over-painting is evident on the largest contingent of the plaster sculpture in

the various collections, which form the Foundation of Weimar Classics. Multiple superimposed layers

228 Cf. Catherine Chevillot, CCI Notes 12/2, p. 2 or Barbara Applebaum, p. 173 or

http://www.vam.ac.uk/collections/sculpture/sculpture_features/caring_marble_plaster/caring_plaster/index.html. Recorded this period in Weimar in a letter from Dannecker to Schiller in 1794: “Listen, something comes to my mind in regard of your portrait, I soaked it in oil, so that when vermin should soil it one can wash it off easily with a wet finger. It will probably arrive blotched, but just put it in the sun, it will surely get an even tone.” cited after Christian von Holst, Johann Heinrich Dannecker: Der Bildhauer. Exhibition catalogue, Stuttgart 1987, p. 210: “Höre es fallt mit noch etwaß wegen deinem Brust Bild ein, ich habe es mit einem Öhl getränkt, damit wann ungeziefer es verschmuzen sollte man es leicht mit einem nassen Finger abwaschen kan. Es wird vermutlich flekend ankommen, stelle es aber nur in die Sonne es wird gewieß einen gleichen Ton bekommen.”

64

Figure 30: Plaster casts of Christian Daniel Rauch’s bust of Johann Wolfgang von Goethe in storage.

hide information on an objects history and cause difficulties in identifying objects’ provenances.

Confusion on issues of provenance is caused especially by the serial nature of plaster casts. Over-

paints render individual cast editions, which might have been produced at different points in time and

might have wholly different provenances, onto one surface quality. Apart from the loss of plaster’s

material quality, the object loses its historical allocation. An illustration of this phenomenon is given by

this photograph of some of the casts of Rauch’s Goethe bust of 1820 in storage of the Schiller Museum

in Weimar (see Fig. 30) The respective top layer of paint on this series of casts seems identical and

therefore can be presumed to have been applied at one given point in time, over which however there

is no secondary documentation. Upon naked eye inspection, these busts all seem identical. Yet the

different inventory numbers attached to the bases prove various accessions and therefore different

provenances and collection contexts.229 Only detailed investigations and materials analysis will give

information on possible sub-coatings and other indications for each bust’s history of acquisition,

display and maintenance. 230

229 Likewise, the only solid evidence for the collection context of the library busts is a number, in red, which was

sometimes revealed while removing paint layers. Red numbering had evidently been applied to the plaster busts in connection with the inventory of 1848-1851. Many other different numbers however were also applied to different areas of these busts’ surfaces and onto different layers of paint. Marking and writing onto plaster casts' surfaces is curiously wide-spread and fits into with the general depreciation this sculptural material has received.

230 The phenomenon of thick over-paints is not known exclusively to the Foundation of Weimar Classics. My own work experience at the National Gallery, the Sculpture Collection and Museum for Byzantine Art and the Collection of Classical Antiquities of the National Museums in Berlin stands for some pre-occupation

65

To clarify, the term over-paint is not used here when classifying a material imitation, as is often found

on casts of bronze or terracotta, marble or sandstone originals. For instance, the Replica Workshop of

the National Museums in Berlin produces marble imitations on their plaster casts to this day, when so

ordered. When no specific material quality is ordered, the current surface treatment to fresh casts –

that is to “naked” plaster surfaces – is a coating with an off-white toned emulsion paint.231 This

treatment can safely be classified as a preventive surface treatment against soiling.

The expert group that formed the Commission for Consultation on the Treatment and Conservation of

Plaster Casts in Berlin issued the following resolutions in 1877:232

“1. The conservation of plaster casts which are assembled in art collections, does not allow the coating of

a cast with paint, the application of a transparent colouration or of a veneer is also not recommended: a

coating with paint will be allowed only to undoubtedly spoiled casts, by way of exception.

2. It seems admissible in contrast, to protect the surface of a plaster cast in need of conservation against

dirt and to facilitate its necessary cleaning by impregnation.

3. The attempted aim is the discovery of a method, which in virtue of choice and treatment of plaster

before casting or its substitution by another casting material, will produce casts, which will for their

conservation not require further treatment of the finished cast.

4. In addition, it is advisable to search for new and better methods to give the surface of the plaster cast a

consistency, which protects it as much as possible against damage and continual soiling and which

facilitates its cleaning as much as possible.

5. Badly spoiled plaster casts should best be replaced by new casts. It is therefore desirable to conduct the

outward acquisition of casts wherever possible, in a manner in which, instead of the finished casts real

moulds are acquired, optionally at least always acquiring two casts, of which one is designed for

reproduction.

6. The protection of the casts will be essentially achieved by an as much as possible dust-free installation

of collection rooms and the prevention of humidity in these, which should be considered when

constructing new or remodelling existing collection rooms.

Comments on the preceding resolutions

To 1: With their vote on 1, the commission wishes to express the opinion “that clean new plaster casts, as

long as the question of polychromy is not an issue, do not receive a coating of paint. It avows itself in this

with this topic.

231 Thanking Bertold Just and Thomas Schelper of the Replica Workshop for this information.

232 See Appendix I.

66

respect to be against the so-called veneer, as it has for example been applied in the local museum.233

Because although with this latter method the plaster surface is covered only with an extremely thin layer,

the commission assured itself at close inspection, that the appearance, which the veneer gives to the

casts, is in the predominant majority of cases similar to which a thick coating of paint would produce.

Older casts, which have been soiled to such an extent by some outside influences, that they cannot be

cleaned, and which cannot be replaced by new casts, can in the opinion of the commission, appear at least

somewhat in accordance with their purpose, uniformly tinted by a careful coating with paint. Plaster

casts, which are intended for common school education, may for reasons of their preservation, be covered

with a proper paint coating.

To 2.: In smaller plaster collections, as they exist in some universities, polytechnical and art schools, the

smaller number of objects and the comparatively weak public attendance allow to keep the plaster casts

adequately clean, without necessitating the application of washing media. In such cases the casts should

be kept in the state in which they came out of the moulds. In comparison, casts in large and very

frequented collections cannot be kept clean without regular periodical washing. They must therefore be

prepared on their surfaces in such a manner, which allows the treatment with soapy water. This can be

achieved by the application of a liquid, which will penetrate the plaster surface, close its pores and thus

harden it: yet it should be chosen so that it changes the appearance of the casting skin as little as possible.

A part of the commission wishes that such impregnations should also change the colour of the plaster as

little as possible, while notably the sculptors within the commission would not mind seeing a warmer

shade emerge from the impregnation.234

To 3.: Customary types of plaster have very different properties, which to an extent do not satisfy its

function. Moreover the usual method of applying an impregnation generates an uneven colouration of

the cast. These grievances would not exist, if the casts were produced from a mass, which had a uniform

colour and which, without impregnation would allow its washing. To initiate such a desirable aim, the

commission advises the high ministries to offer a considerable prize for the production of such a casting

material. This material should, if it is not plaster, be similar to plaster in colour and should not be inferior

in respect to its ease of handling as well as to its fine grain size.235

233 Cf. Gertrud Platz-Forster, pp. 282-288. Frank Matthias Kammel, 2001, p. 58: The archaeologist Carl Bötticher

had ordered a coating for the Berlin plaster cast collection of ancient sculpture of yellowing wax, which prompted a vehement discussion about the question of treating plaster surfaces, splitting an entire generation of museum scholars.

234 Bernhard Maaz 1993, p. 39, notes an increase in toning plaster with warmer shades of off-white to yellow being contextual to late 19th centuries' interest in polychrome sculpture. Many sculptors commenced treating their Carrara marble statues with different agents (for example tea or chestnut brew) in this period, to break the glaring white of freshly hewn marble into warmer tones or to highlight areas.

235 Cf. Petra Rau, pp. 71-72, Friedrich Wilhelm Doell, court-sculptor in Gotha, and his brother Carl Friedrich had commenced developing a mass of plaster with clay additives, which was subsequently fired, already in 1790.

67

To 4.: Even if the casting material named under 3. can be achieved, it would hardly be launched

everywhere at once. The collections would in fact for a longer period of time be in the situation to have to

preserve the casts made in the old manner from the outside and would need to impregnate the casts for

the purpose of their cleaning. That is why it remains important to find a new manner of impregnation,

which achieves the conditions, specified under 2., to a higher degree than the known methods of

impregnation. Therefore the commission advises making this question the purpose of a second

competition.

To 5.: The realisation of the desire expressed under 5., which does not seem to demand further

motivation, would allow the distribution of the best casts of excellent sculptures at cheap prices and thus

the propagation of purer outlooks on art as well contributing to the advancement of artistic practice.

The commission deems it appropriate, that major museum moulding shops are expanded accordingly

and agree on reproducing a certain amount of especially precious objects in the recommended fashion.

For the purpose of easier replacement of damaged casts, it is advisable that moulding shops deliver casts

to museums and to educational institutions at the price of manufacture.

To 6.: The conditioning of display rooms for plaster casts is in various respects important for their

conservation. The commission does not however see itself in the position to advise specific directives,

restricting itself rather to a few suggestions, which refer to the prevention of dust and of humid and

spoiled air.

The advice concerning heating is to maintain a possibly constant and not too high temperature

throughout the year. Radiators with very high temperatures should be avoided, because they act

detrimentally by singeing dust particles. Low-pressure water radiators, implemented in vertical pipe-

systems with low area sizes, are deemed especially advisable.

Ventilation should primarily seek to keep humidity at bay, which is especially deleterious in winter. The

occasional occurrence of a draught is partially deemed as downright beneficial, generally not

detrimental. The simple method that is common in Italy, to keep the rooms cool and at the same time airy

by duly opening windows and closing shutters in summertime is deemed recommendable.

For those cases in which artificial lighting should become necessary, the employment of sun-burners is

recommended, as have been introduced in the South-Kensington-Museum with great success, and which

quickly abduct combustion products.236

Floors give least cause for the production of dust if they are manufactured out of hard and fairly smooth

material. Paving with tiles from Mettlach237 is deemed recommendable, one with rough bricks

236 Cf. Geoffrey N. Swinney.

237 See http://www.villeroy-boch.com/en/us/home/the-company/the-company/history.html

68

objectionable, common wooden flooring is not recommended.

For the cleaning of plaster casts, repeated dusting within short periods is of first and foremost service. As

a tool for this, the use of feather dusters is generally to be used after the use of the bellows, while the

feathers can easily be carriers of fattiness or adherent dirt particles. The bellows should be fitted with a

padded mouthpiece. Heavily soiled casts should be cleaned with soapy water within larger intervals.”

This report238 deserves full translation here, because it delivers insight into some highly interesting

views on concepts for display and maintenance of plaster statuary dating from the late 1870s. These

are to an extent remarkably referable to current thinking and approaches, on for example air-

exchange, humidity values or temperature constancy for the preservation of materials. It also gives an

idea of the pure market value of the propagation and trade of casts and moulds had in this period of

time, which presumably was also the incentive behind prompting a contest to find an adequate

material in replacement of plaster as a casting material.

And while the commissions’ intent on finding a suitable casting material to replace plaster seems not

to have been realized in the desired fashion,239 the development of coating materials evidently

remained in focus for some time in Berlin. It culminated perhaps between 1916 and 1921, when the

Neues Museum’s cast collection of ancient sculpture had been assigned into the holdings of the

university's archaeological seminar. Friedrich Rathgen, Director at the Chemical Laboratory of the

Royal Museums in Berlin, developed the use of Zapon-lack240 as an airbrush method of mass-coating

plaster casts, after various failed attempts at finding suitable methods for mass cleaning. A total of 2500

casts of ancient sculpture were thus treated with a mixture of solved Zapon-lack, lithopone and ochre,

in a hall in today’s Humboldt-University, in which its canteen is situated, see Figure 31.

238 Cf. note 29. A photocopy of the original text is in Appendix I, however only these relevant pages. The

subsequent text deals with the contests of 1875 for the invention of a casting material that could replace plaster, the contest was repeated in 1877.

239 Gertrud Platz-Forster, loc. cit.; Frank Matthias Kammel, 2001, loc. cit.

240 Solved cellulose nitrate which had been developed in Dresden in 1899, see Friedrich Rathgen, p. 36. This so-called “Noack-Coating”, is today still evident on a large number of plaster surfaces in Berlin and was named after the archaeologist Ferdinand Noack, Veit Stürmer and Henning Wrede, p. 18. On Zapon-lack see http://cameo.mfa.org/materials/record.asp?key=2170&subkey=10272&Search=Search&MaterialName=zapon&submit.x=0&submit.y=0 and also Marc Gilberg, http://206.180.235.133/jaic/articles/jaic26-02-004_4.html.

69

Figure 31: The University hall in which 2500 casts were sprayed with Zapon-lack in

1916-1921.

The valuable information this report contains on the question of over-paints, can be summarized into

the following points:

• Over-paints on plaster surfaces are not regarded as acceptable from an aesthetic point of view

• Other finishing veneers are likewise considered unacceptable for plaster surfaces

• Over-paints should be applied to plaster surfaces as a last resort, only in the case of irreparable

soiling,

• or for purpose of preservation, when casts are intended for heavily frequented public

buildings, like for example schools

• rather than paint or finish plaster surfaces with some veneer, material impregnation would be

preferable for the “purpose of their cleaning”

• small collections with “comparatively weak public attendance” do not require the application of

a material that facilitates washing down plaster surfaces

• “larger and more frequented collections of plaster casts do necessitate the application of a surface

protecting agent that allows cleaning treatment with soapy water”

• the material qualities, fineness of grain and texture and usefulness of plaster as a casting

material is fully appreciated, any material, which might be found in substitution should have all

of these qualities as well as the same colouring

Reading this report, the answer to the second question posed above, whether one can categorize the

purpose of the thick and white over-paints as being preventive measures against dust-fall, is clearly

yes. While the commission argued that over-paints were only permissible for heavily soiled surfaces,

their additional feature of creating manageable surfaces in terms of regular and necessary cleaning

with soapy water is clear. Hence the content of this report is easily referable to the practice on the

70

plaster surfaces in the Rococo Hall, as it is also to a large proportion of plaster surfaces in other

collections of plaster casts.

While realizing that the paint layers were applied to the plaster surfaces of the busts in the Rococo Hall

for aesthetic reasons and also to facilitate future cleaning, the multiplicity of layering does in fact

suggest that this was not fully achieved – to the extent that the first layer of paint was at a later point

in time re-covered with further coatings.

71

Survey:

Questionnaire on Maintenance Concepts and Dry Cleaning Methods for Plaster Surfaces

on Display

A questionnaire was sent directly to specialized sculpture conservators, to several German and

international museum conservation studios and to plaster cast collections that are linked to

www.plastercastcollection.org. It was also hosted on the SPAD241 working group webpage of ICOM-CC

and on the stone and wall-painting group of ICON, UK.

A total of 32 specialized conservators participated in the survey.242 The responses came from within

Germany, France, Belgium, Austria, the Netherlands, Spain, Italy, Sweden, Norway, Denmark, Canada

and the United States of America. The institutions and collections in which the participants experience

with plaster statuary has been gathered are: Universität Göttingen, Archäologisches Institut Göttingen,

Antikenmuseum der Universität Leipzig, Universität Halle, Museum für Abgüsse Klassischer Bildwerke

München, Akademisches Kunstmuseum der Universität Bonn, Winckelmann Institut Humboldt

Universität, Gipsformerei und Skulpturensammlung der Staatlichen Museen zu Berlin,

Skulpturensammlung und Abgusssammlung Dresden, Deutsches Historisches Museum, Musée des

Monuments Français, Musée Rodin, Collection de la Ville de Paris, Collection de l’Institut d’Art de

l’Université de la Sorbonne, Château and Musée du Blois, Musée Girodet, Musée de Beaux-Arts in Tours,

Château-Musée Dieppe, Musée de Beaux-Arts de Marseille, Musées d’Avignon, Institut für Klassische

Archäologie Universität Wien, Stichting Beeldende Kunst Amsterdam, Real Academia de Bellas Artes de

San Fernando, Musei Vaticani, National Museum in Stockholm, Vigeland-museet Oslo, Thorvaldsens

Museum, Statens Museum for Kunst, Ny Carlsberg Glyptothek, Art Galley of Ontario, The Metropolitan

Museum of Art, The Peabody Museum, Harvard University, The Henry Ford Museum and the Minnesota

Historical Society.

The questionnaire aims at identifying maintenance concepts for plaster casts and their surfaces. Since

the practical focus of this thesis is to establish a dry cleaning methodology for the uncoated plaster

surfaces that are now again on open display in the Rococo Hall, different methods for dry (and in-situ)

cleaning are put to question. One other focus is on plaster maintenance in regard to environmental

parameters of display and storage and the reasons for potential showcase display of plaster statuary

are queried. And, since research into literature has not been very forthcoming for conservation

manuals on questions of maintenance and cleaning of plaster statuary, information on any such a

manual was requested.

241 Sculpture, Polychromy, and Architectural Decoration is abbreviated SPAD.

242 I’d like to express my very great thanks and appreciation to all of those who participated and also to all who assisted in distributing the questionnaire. Since there was a lot of positive feedback as well as interest in the evaluation of this survey, this evaluation will be distributed amongst the participants. The responses were received within the time frame June 24th to August 10th, (extended to September 10th) 2009.

72

One other line of investigation pursues the practice of coating plaster surfaces and aims at possibly

categorizing different types or purposes of coatings applying to different types of plaster statuary. The

question is how specialized conservators perceive these coatings today, in view of their function and

their original context and in terms of their aesthetic qualities. Also, a question toward current

practices of coating plaster surfaces seeks to assess whether such an approach i.e. coating plaster for

reasons of preservation, might at all still be topical.

The first question was general, in that it aimed at identifying the different types of plaster statuary,

which would be related to within the survey.

1. Are the plaster sculptures you work with A. casts of ancient Greek and Roman original sculpture 16 B. casts of medieval original sculpture 13 C. casts of modern era statuary 20 D. original sculptor models 23 E. architectural models 13 F. others were specified as casts of meso-american sculpture, of architectural elements, of holy figures and of christmas groups, life and and death masks as well as landscape models

4

If more than one of these categories please mark several. If C. please specify whether portrait busts of neo-classical era (ca. mid18th to mid 19th century):

Diagram 1: Evaluation of question 1 on the types of sculpture that this survey represents

The largest contingent of surveyed plaster sculpture is clearly D, original sculptor models, followed by

casts of modern era statuary and cast of ancient Greek and Roman original sculpture.243

243 Only very few of the respondents answered to the additional question as to the casts of modern era statuary

(C) being portrait busts of neo-classical era, which was put with the aim of evaluating whether it is foremost this genre, the 19th century plaster portrait bust, that is most heavily or widely to be found coated with paint or other media. Cf. Catherine Chevillot, p. 174, who on reviewing plaster surface treatments of the late 19th century, makes this observation. I can now only assume that this question should have been put more precisely as a direct question (i.e.: Have you in your experience found especially the portrait bust to be

0

5

10

15

20

25

A. casts of ancient Greek and Roman original sculptureB. casts of medieval original sculptureC. casts of modern era statuaryD. original sculptor models E. architectural modelsF. other (please specify)

73

The second question related to coatings on plaster surfaces.

2. In your experience, are plaster surfaces coated or sealed with shellac, wax, oil, paint or other

media?

30 of 32 participants answered Yes. 19 of these answered Yes and No. 2 of 32 answered No.

If yes, which type of plaster sculpture (Question 1) is coated? A = 12, B = 9, C = 14, D = 15, E = 8, F

= 4

Diagram 2: Evaluation of question 2

The answers to question 2 indicate a large proportion of plaster statuary to be coated. The diagram

below illustrates the type of plaster statuary that the respondents have encountered as coated.

Diagram 3: Statistical evaluation 2 to question 2

almost always coated or over-painted?) in order to receive an evaluable feedback.

0

5

10

15

20

25

30

35

Are plaster surfaces coated?

YesNoYes and No

0

2

4

6

8

10

12

14

16

Which type of plaster sculpture is surface-coated?

A. casts of ancient Greek and Roman original sculptureB. casts of medieval original sculptureC. casts of modern era statuaryD. original sculptor models E. architectural modelsF. other

74

The answers to the second part of question 2, which pertained to the types of sculpture (as indicated

in question 1) and their surface treatments, are shown in diagram 3 as proportionally relatively

identical to the types of plaster statuary that are indicated in the diagram to question 1. These checks

lead to the assumption, that all types of plaster casts and statuary need not be but can be similarly

treated with a surface coating of some type.

Question 2.2. then queried into the technical purpose of the coatings, as perceived by specialist

conservators:

2.2. In your assessment, were the coatings applied for A. technical purposes (casting procedures) 17 B. imitation of original material (bronze, terracotta or other) 26 C. prevention of dust settlement and permeation 14 D. aesthetic purposes (hiding disfiguring dirt layers or damage and repair histories) 25

E. other purposes (specified as original artist’s coating and past restoration, the latter might well be attributed to D, making that number total in 26)

2

The diagram evaluation below illustrates the purpose of hiding past restoration, i.e. damages, repairs

and dirt, category D, to be perceived as proportionally nearly equally common as category B, material

imitation. 13 participants concluded the prevention of dust settlement to be a purpose for coating.

However, multiple checks were possible, so, the prevention of dust settlement might possibly also be

termed an additional purpose.

Diagram 4: Evaluation of question 2.2

The third question was posed with the intent of finding similar surface histories to those in the

Rococo Hall. The categories A and B of the previous question (2.2.) were to be excluded, because the

0

5

10

15

20

25

30

In your assessment, were the coating applied for

A. technical purposes (casting procedures)B. imitation of original material (bronze, terracotta or other)C. prevention of dust settlement and permeationD. aesthetic purposes (hiding disfiguring dirt layers or damage and repair histories)E. other purposes (please specify)

75

paint layers referred to in this question are those that are neither material imitations nor coatings

stemming from casting procedure but layers of paint that were brought onto pre-soiled or otherwise

unsightly or damaged plaster surfaces.244

3. Are multiple superimposed layers of paint evident on some surfaces? 21 replied Yes, 6 replied

No, 1 replied Yes and No, 1 replied sometimes, 1 replied seldom.

If yes, please describe the shade(s) of colour you have encountered in paint layers on plaster

surfaces:

The colours of the over-paints were in the most cases (by 12 respondents) described as white, ivory or

off-white tones. Grey and all ochre tones (red, yellow, grey, green) were named, but also brown, tan

tones and blue. Some respondents gave additional information on types of binding media they

encounter, which however will not be discussed here, as the materials for coating plaster do not form

part of this study.

Diagram 5: Evaluation of answers to question 3

The next question (3.1.) was posed in order to assess the current practice of removing coatings of

paint from plaster surfaces.245

244 It was not entirely clear on reading the participants checks and answers whether this distinction was

understood. I first considered deleting those answers from this evaluation, where I assume an original paint layer (i.e. material imitation) to be meant, however can on reading these contributions conclude that over-paints are just as likely to occur in various colours and that some material imitations have been realized in these over-paints, which seem to have been to an extent applied to cover damage and/or soiling histories.

245 The one reference I have so far found on stripping plaster is on Victoria & Albert Museum’s site Caring for Plaster: “For the inexperienced, stripping plaster is difficult to do well. Abrasives are likely to damage the surface whilst paint strippers may penetrate the plaster, causing damage or stains. It is best to consult a specialist, who can offer advice on the best course of action. However, before consulting a conservator give some thought to the issues involved. For example, why was the bust painted in the first place? A common reason for overpainting is to conceal old stains or disguise old restoration fills. If this is the case, no matter how well the paint is removed, the

0

5

10

15

20

25

Are multiple superimposed layers of paint evident on some plaster surfaces?

YesNoyes and nosometimesseldom

76

3.1. Have you in your practice removed coatings from plaster surfaces? 18 answered Yes and 12

No

If yes, please name the reason for their removal: The reasons given for the respective removal

were all very similar: to uncover the original plaster or painted surface, to uncover sharpness of

surface detail, because the coating had deteriorated, because the coating disfigured the cast, for

aesthetic reasons, to clean the underlying surface and to repair damages and faulty older restorations

as well as for casting purposes.

Diagram 6: Evaluation of responses to question 3.1.

And the following question assesses whether it is common practice to coat or re-coat plaster surfaces:

3.2. Have you in your practice applied coatings onto plaster surfaces or onto previously coated

plaster surfaces? 15 replied with Yes and 14 replied with No (some did not indicate).

If yes, please name the reason for application and please specify whether previously coated:

The reasons named for an application were: for aesthetic reasons, to appease blotchy appearances, to

cover staining or un-cleanable plaster surfaces, for material imitation, as material imitation when a

(new) cast was too large to fit into a show case, as surface consolidant and protectant, prevention of

dust settlement, to ensure longevity to a new cast. Many of those who checked No added that they

retouched pre-existing layers of coatings.

piece may look no better afterwards. If peeling paint is the problem, this can be treated by a conservator – stripping is not your only option. Find out all you can about your object. If the paint is an original finish, or if any underlying paint is original, this may add to the historical information and authenticity of the piece. Removing original finishes often reduces both historical and monetary value.” [http://www.vam.ac.uk/collections/sculpture/sculpture_features/caring_marble_plaster/caring_plaster/index.html last accessed August 31st, 2009]

0

2

4

6

8

10

12

14

16

18

20Have you removed coatings from plaster surfaces?

YesNo

77

Diagram 7: Evaluation of question 3.2.

Questions concerning environment and installation followed and the respondents recommended

climate values proved to be very varied in that the highest values for relative humidity (RH) ranged

between 70 % and 35 %, while the lowest recommended values for RH ranged between no value at all

or between 20 % and 55 %. Some respondents left this field blank altogether (and one answer is my

personal favourite, because it states that absolute numerical values are not as important as keeping a

constant RH). However, one other participant to this survey made the very valid specification that the

internal armature, which might be present within plaster statuary, necessitates closer climate

control.246 In one case additional recommendations on temperature values (between 20 and 22 °C)

were given, which most probably apply to the environment of display.247

4. Which RH values do you recommend for display and storage of plaster sculpture?

MAX RH ranging between 35 and 70 % MIN RH ranging between 20 and 55 %

246 On this topic, cf. Violaine Pillard, p. 279, who is most probably referring to internally armed (with iron)

highly fragile plaster casts of leaves by Geoffroy-Dechaume, when she states plaster conservation can be guaranteed in an environment of 24 °C to 38 °C to attain relative humidity levels of between 35% and 45%.

247 Cf. Standards in the Museum Care of Archaeological Collections, Museum & Galleries Commission, 1992, note 2, p. 57: on the recommended average 18°C for display of collection materials (for which 10°C are specified as min., and 25°C specified as max.) “meets thermal comfort requirements for people. It should not be exceeded and whenever possible, temperature should be kept below this level. The lower the temperature, the slower is the rate of deterioration of materials”, www.collectionslink.org.uk/assets/userfiles/index.php?file=000072.pdf [last accessed August 31st, 2009].

0

2

4

6

8

10

12

14

16

Have you applied coatings onto plaster surface or onto previously painted plaster surfaces?

YesNo

78

Diagram 8: Statistical evaluation of question 4

Perhaps in accordance, climate control was the category least chosen as a reason behind experiences

of showcase installation of plaster statuary, the most chosen category being protection against theft

and vandalism, A, followed closely by category D, protection against soiling:

5. Is plaster sculpture displayed in showcases? 22 answered with Yes and 7 answered with No If yes, which criteria apply for show case installation? A. Protection against theft and vandalism 21 B. Climate control 5 C. Protection against soiling 19 D. Mounting difficulties 11

Diagram 9: Statistical evaluation of question 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 240.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

80.00%

Which RH Values do you recommend for Plaster Statuary

MAX RHMIN RH

0

5

10

15

20

25

Reasons for Showcase Installation

A. Protection against theft and vandalismB. Climate controlC. Protection against soilingD. Mounting difficulties

79

Only two respondents indicated the knowledge and use of guidelines, which was asked in question 6,

fixed to a manual for preventive care and maintenance. One of these responses came from Belgium

and the restorer stated it to be a vademecum to protect and maintain artefacts, issued by the IRPA in

Brussels in 1986. The other respondent is Dutch, but did not specify on the manual he has conferred.

The only two useful guidelines for preventive care and maintenance I could trace in my research

contain information and advice looking after domestic plaster objects with practical instructions on

how to clean and preserve plaster: CCI Notes 12/2, Care of Objects Made of Plaster of Paris, issued by

the Canadian Conservation Institute, and the site ‘Caring for Plaster’ of the Victoria and Albert

Museum.248

6. In your practice, do you follow the guidelines of a manual for preventive care and cleaning

for uncoated plaster surface? Yes / No.

If yes, please name / describe the manual:

Dry cleaning methods for uncoated plaster formed the following part of this survey, question 7, in

which various methods (known from own work experience and encountered in the literature or in

conversations with colleagues), were named for review by practitioner specialists. Two types of

erasers were named for checking, Wallmaster® and Wishab®, one being a vulcanized natural foamed

erasing sponge and the other a synthetic rubber i.e. factice.249 Special attention was given to the

question of the use of the brush in combination with either a vacuum cleaner or compressed air:

For 7. Please indicate the dry cleaning method(s) you apply to plaster surfaces A. compressed air 11 B. compressed air and brush 13 C. vacuum cleaner 25 D. vacuum cleaner and brush 31 E. foamed natural rubber sponge, e.g. Wallmaster® 14 F. synthetic rubber (factice) eraser, e.g. Wishab® 20 G. others methods / materials specified were the erasers Absorene®, Gomma Blanda Milan® and Staedtler Mars Plastic, soft and Groom/Stick® and Anjusil® latex milk as well as agar and other non-residue gels of different composition, depending on frailty of the surface and soiling.

The results showed that dry cleaning plaster with a vacuum cleaner and brush is clearly the method

employed most often, as assessed by this survey, and the combined use of a brush (presumably soft

brush) is in fact more pronounced than either nominally contact-free method (vacuum or compressed

air) on its own. Some classified that low pressure and precaution should be implemented with

compressed air, others classified the use of a brush as being abrasive. Likewise some classified the

248http://www.vam.ac.uk/collections/sculpture/sculpture_features/caring_marble_plaster/caring_plaster/index.

html

249 Cf. Martina Noehles’ review and bibliographical article on a variety of latex sponges, who notes reports of staining to paper due to use of the yellow Wishab® sponge.

80

action of various erasers to be too destructive and caution was pronounced for the use of anything

with too much tack, like for instance Groom/Stick®.

Diagram 10: Statistical evaluation of question 6

The last questions sought general advice for or against methods for dry cleaning plaster and while

some respondents omitted filling out this part, some very valuable comments were made and are

repeated below in full, because they assist in illustrating the relative diversity and also some

contradiction in approaches to dry cleaning plaster:

8. Would you care to offer recommendations for dry-cleaning uncoated plaster surfaces?

• To empirically trial test the appropriate cleaning method

• Laponite® RD gel and laser (both have drawbacks though)

• Compressed air

• Factice erasers, Staedtler Mars Plastic eraser soft, vacuum cleaner and brush

• Vacuum cleaner

• Vacuum cleaner

• Compressed air and vacuum cleaner and brush

• If I don’t know the nature of the residual not to proceed with cleaning

• Gel (Laponite® RD) and pastes of attapulgite and sepiotlithe

• Pre-testing for the determination of the appropriate method, avoiding too much humidity

and abrasion

• Laponite® RD gel with fumed silica

• Peeling gel, after pre-testing

• Paste of attapulgite and carboxymethyl cellulose (CMC)

• The determining criterion is the minimum possible abrasion

0

5

10

15

20

25

30

35

Dry cleaning methods used on plaster surfaces

A. compressed airB. compressed air and brushC. vacuum cleanerD. vacuum cleaner and brushE. foamed natural rubber sponge, e.g. Wallmaster®F. synthetic rubber (factice) eraser, e.g. Wishab®

81

• Methods with very low abrasion because of the feeble strength of plaster

• Regular, possibly contact-free vacuuming

• In the case of heavy soiling, agar-agar

• Stratigraphical analysis of the surface and of the superimposed elements

• Combination of vacuuming and rubber sponges worked well on the 800 casts we recently

treated

• Laser cleaning can work very well, however, if there is any coating on the plaster, below the

grime and perhaps unknown at the time of cleaning, the laser cleaning will cause the coating

to discolour to blotchy yellow areas

8.1. Which dry cleaning methods would you dis-advise to apply to uncoated plaster surfaces?

• It depends on the object

• All abrasive methods

• C, D, E of question 7 (vacuuming with and without the aid of a brush, and foamed natural

rubber sponge)

• E and F of question 7 (foamed natural rubber sponge and factice rubber eraser)

• A, B, C and F of question 7 (compressed air, compressed air and brush, vacuum cleaner,

Wishab®

• Brush

• Erasing methods can lead to blotchy surfaces

• Any kind of eraser and sponge

• Any abrasive method or rubbing of the surface

• Natural latex, because it leaves residues

• Some erasers cause gloss to plaster surfaces and abrade, micro-sandblasting is too aggressive

• Rubbing and brushing

• It depends on the fragility and the porosity of the surface, compressed air can be damaging to

fragile epidermis

• The brush has to be very very gentle! If not I can not recommend it.

• It all depends on the fragility of the plaster surface

• Abrasive erasers or sponges

• All types of scratching

• All rough tools and instruments

• Nothing with too much tack, such as `groom stick’- can risk removing loose sitting plaster, no

stiff bristle brushes, leaves scratches.

• Whichever method causes friction to the surface

• Using materials like sponges with too much abrasion

82

Summarizing, this survey has established that the type and purpose of the over-paints in the Rococo

Hall are not at all a singular case. Coating soiled plaster (still) seems to be a habitual practice. The

removal of over-paints is a relatively common practice, as is the (re-) application of coating material to

plaster.

The recommendations on dry cleaning plaster surfaces the survey received, though varied, give a good

idea of the problematic nature plaster surfaces have, as well as of the high amount of respect and

caution, with which conservators approach this material.

A general deficit is notable in the field of guidelines and published literature on issues of preventive

conservation for plaster statuary – not least by the very varied recommendations, which this survey

received on the question of environmental parameters (in this case RH).

This study only marginally deals with wet and poultice cleaning methods for plaster, because it focuses

on establishing a preventive maintenance plan for bare plaster surfaces which are on open display in a

controlled environment and which have all recently been poultice-cleaned and stripped of various

super-imposed layers of white to off-white layers of paint. Studying the relevant literature on the topic

of plaster cleaning and conservation however, which is as yet still largely un-comprehensively

published,250as well the respondents contributions, has allowed reflection on the fact that language

barriers exist, also in conservation. One example for instance, the Laponite® RD gel251, which some of

the German contributors recommend, seems to be known only regionally and the carboxy-

methylcellulose and attapulgite poultice mix seems to be widely appreciated for cleaning plaster mainly

in France.252 I would therefore like to conclude this survey with the observation, that more

communication and networking would benefit this very specialized field of work (and devotion). The

overall very positive feedback and interest the questionnaire received, reflects a renewed interest in

plaster statuary, perceivable in various international symposia, which focus also on materiality and

technology of plaster casts.253

250 Mainly being formed by yet unpublished semester reports or diploma theses, for example: Astrid Lorenzen,

Anette Schulz, Ute Griesser, Louise Cone, Johanna Kapp.

251 See Ursula Haller and discussed by Jorun Rebekka Ruppel.

252 See Astrid Lorenzen and more recently Johanna Kapp.

253 Les Moulages de Sculptures antique et l’Histoire de l’Archaéologie, Actes du Colloque International, (October 24th, 1997), and yet unpublished proceedings of: Plaster Casts: Making, Collecting and Displaying from Classical antiquity to Present, International Conference at Oxford University (September, 24th-26th 2007); Das Original der Kopie. The originality of copies, Colloquium (November, 2nd--3rd 2007), Humboldt University Berlin; planned: Plaster and Plaster Casts: materiality and practice, 12th-13th March 2010, conference at the Victoria and Albert Museum, London.

83

Chapter 3

Planning Preventive Care and Maintenance for the Plaster Busts in the Rococo Hall

Planning a strategy for the preventive care and maintenance for the plaster busts in the Rococo Hall

involves the identification of the agents, which might cause deterioration. In consideration to plaster’s

material properties and to the situation of installation of those plaster busts, which have received no

surface sealing, these can be identified as particulate matter settling on their surfaces for prolonged

periods of time in the presence of high levels of humidity.254 These relevant factors of environmental

impact, which are relative humidity levels and dust impact, have therefore been investigated and will

be presented below.

The placement of the busts on wooden pedestals is determined by the collection context and its history

of display within the library hall. Vibration proofing was achieved at the time of their installation by

buffering the base areas of the pedestals with cork. The bust’s base surfaces are fixed onto acid

cardboard with mounting wax255 and the board itself is mounted onto the pedestal surface with

mounting wax.

Because the re-application of a sealing agent to the plaster surfaces in the Rococo Hall continues to be

a point of discussion, and stands to reason on the basis of the tradition in maintenance to these

surfaces, the last section has dealt with the topic of surface coatings on plaster extensively. And while it

is valid to consider the application of a surface sealer in keeping with this tradition, it is equally valid

to investigate into methods of preventive care and maintenance for plaster surfaces, which do not

encompass sealing agents. This in view of the evidence, which suggests that the original display of the

plaster surfaces in the Rococo Hall was unsealed, but more especially because this study’s findings

suggest that the treatment method of over-painting soiled plaster surfaces evidently stems from a

period in time, in which the vacuum cleaner had not yet reached broad conservation practice.256

The situation of installation of the plaster busts in the Rococo Hall is such that their placement into

showcases is not feasible. Two collections of plaster statuary in France, Galerie David d´Angers, in

Angers and Musée Rodin in Meudon, Paris, are examples for bare plaster surfaces on open display and

they both show proof of good maintenance and preventive care in the form of solid cleaning routines,

254 As already stated, this research yielded no study or assessment of factors that influence soiling cementation to

plaster surfaces. The actual values that influence cementation processes are therefore unknown. Cf. also Holly Dawes, who describes that cementation processes of dust to objects’ surfaces commences at values as low as 35% RH.

255 Micro-crystalline wax, see http://www.museumwax.co.uk/.

256 See for example http://inventors.about.com/library/inventors/blvacuum.htm on the history of the development of the vacuum cleaner.

84

which are carried through by specialized conservators. In both cases, cleaning of heavily pre-soiled

plaster had been conducted with poulticing methods since the early 1980s and cleaned plaster

surfaces are since vacuumed in the location of their display at regular intervals, to remove recurring

dust settlement.257

A risk factor for the plaster busts, which should not be underestimated, is that of handling. The

maintenance plan, which this research proposes, therefore takes this risk into account by the choice

of the most contact-free method for a routinely effective dust removal from the plaster surfaces.

Climatic Conditions in the Rococo Hall

There is no heating system to the library hall. The displacement ventilation system which was installed

during renovation of the building, pumps air into both levels of the library from beneath each window

(8 windows on each level) regularly at 0,3 m/s. This air is heated and its RH is regulated in accordance

with outside conditions and the computer-controlled sensor watch. The climate values are regulated

for the interior and for the materials that form the collection of the library.258 The interior fitting of the

hall consists mainly of painted wood. The main mass of the collection in the Rococo Hall is formed by

the historical manuscripts and leather-bound books. The art collection in the Rococo Hall consists of

paintings on canvas and on wood and of some marble and lime-stone sculptures and busts as well as

of 46 bare plaster bust and 9 coated259 plaster busts. A period of monitoring was carried out (see

below) and established that the recorded highest peaks of relative humidity were reached at 62.5 %

and highest temperature values at 21 °C. The average constant RH in the Rococo Hall however lies

within the range of 54 to 57%, the average temperature at 17 to 18 °C.

257 Personal communication by conservator Agnès Cascio, whom I wish to thank. The conjunct use of a soft

brush while vacuuming is a point of difference, see also above section, survey on dry cleaning methods.

258 Jochen Flade, 2009-b, portrays the extensive drying program this sophisticated ventilation system performed during renovation and drying of the fire and water damaged interior of the Rococo Hall between 2004 and 2007.

259 These coatings either represent material imitations (terracotta and bronze) or else the respective plaster busts were not installed in the library hall prior to the fire of 2004, therefore did not undergo water-damage and subsequent restoration.

85

Experimental 2: Evaluation of Clima as Recorded by Six Loggers for T and RH

Six loggers for temperature and relative humidity were placed at different locations (see floor-plan

Figs. 33 and 34) of the library within the time period on February 2nd to August 3rd 2009.260 Their

recordings match, no large deviations are discernible for different locations on the two levels of the

Rococo Hall.

Figures 32 and 33: Floor plan of the two levels of the Rococo Hall, showing the placements of the six loggers.

The recorded temperature values in the time period February to August 2009 range from 14 °C to 21

°C. The recorded humidity values range between the low peak of 48 % (very seldom) and the highest

peak of 62.5 % (very seldom). The average RH lies in the range of 54 - 57 %.

Comparing 2 Loggers recordings in, for instance a week in May 2nd to 9th (see diagrams 11 and 12),

will show relatively constant curves: Logger 2 is placed in level 1 of the Rococo Hall, close by a window

/ air inlet. Logger 6 is placed in the 1st Gallery, on the balustrade. The comparison shows that while

temperature values are slightly lower on the lower level of the library hall, the humidity values are

260 Testo Loggers, 175-H2.

86

slightly higher to those recorded by logger 6 on the upper level.

Diagram 11: Recorded values of a logger (No 2, cf. floorplan, Fig. 33), placed on the first level of the Rococo Hall in one week in May 2009.

Diagram 12: Recorded values of a logger (No 6, cf. floorplan, Fig. 34), placed on the second level of the Rococo Hall in one week in May 2009.

Comparing these recorded values of loggers 2 and 6 to logger 3 (that is positioned at approximately

the same location as logger 2, only one level higher, in the 1st Gallery, cf. floorplan Fig. 34), for the time

period April, 20th to 26th, shows the same relative stability in the curves and confirms the area of the

windows to be those, which show slightly higher levels of humidity and slightly lower levels of

temperature (see diagrams 13 - 15). This is in accordance with the arrangement of the ventilation

system within the hall and the airflow it induces.

00:00:0007:00:00

14:00:0021:00:00

04:00:0011:00:00

18:00:0001:00:00

08:00:0015:00:00

22:00:0005:00:00

12:00:0019:00:00

02:00:0009:00:00

16:00:0023:00:00

06:00:0013:00:00

20:00:0003:00:00

10:00:0017:00:00

00:00:0007:00:00

14:00:0021:00:00

0

10

20

30

40

50

60

70

Logger 202.05. - 09.05.2009

[%rF] Channel 1[°C] Channel 2

00:00:0007:00:00

14:00:0021:00:00

04:00:0011:00:00

18:00:0001:00:00

08:00:0015:00:00

22:00:0005:00:00

12:00:0019:00:00

02:00:0009:00:00

16:00:0023:00:00

06:00:0013:00:00

20:00:0003:00:00

10:00:0017:00:00

00:00:0007:00:00

14:00:0021:00:00

010203040506070

Logger 602.05. - 09.05.2009

[%rF] Channel 1[°C] Channel 2

87

Diagram 13: Recorded values of a logger (No 2, cf. floor plan, Fig. 33), placed on the first level of the Rococo Hall in one week in April 2009.

Diagram 14: Recorded values of a logger (No 3, cf. floor plan, Fig. 34), placed on the second level of the Rococo Hall in one week in May 2009.

Diagram 15: Recorded values of a logger (No 6, cf. floorplan, Fig. 34), placed on the second level of the Rococo Hall in one week in May 2009.

00:00:0007:00:00

14:00:0021:00:00

04:00:0011:00:00

18:00:0001:00:00

08:00:0015:00:00

22:00:0005:00:00

12:00:0019:00:00

02:00:0009:00:00

16:00:0023:00:00

06:00:0013:00:00

20:00:0003:00:00

10:00:0017:00:00

00:00:00

0

10

20

30

40

50

60

Logger 220.04. - 26.04.2009

[%rF] Channel 1[°C] Channel 2

00:00:0006:00:00

12:00:0018:00:00

00:00:0006:00:00

12:00:0018:00:00

00:00:0006:00:00

12:00:0018:00:00

00:00:0006:00:00

12:00:0018:00:00

00:00:0006:00:00

12:00:0018:00:00

00:00:0006:00:00

12:00:0018:00:00

00:00:0006:00:00

12:00:0018:00:00

0

10

20

30

40

50

60

Logger 320.04. - 26.04.2009

[%rF] Channel 1[°C] Channel 2

00:00:0007:00:00

14:00:0021:00:00

04:00:0011:00:00

18:00:0001:00:00

08:00:0015:00:00

22:00:0005:00:00

12:00:0019:00:00

02:00:0009:00:00

16:00:0023:00:00

06:00:0013:00:00

20:00:0003:00:00

10:00:0017:00:00

0

10

20

30

40

50

60

Logger 620.04. - 26.04.2009

[%rF] Channel 1[°C] Channel 2

88

Dust Impact on the Plaster Busts of the Rococo Hall

This investigation focuses on the total of 46 restored plaster busts which were re-installed on their

original pedestals over two floors of the hall in October 2007 and did not received a surface coating.

Their porous surfaces are a particular challenge for cleaning. A comprehensive overview of the pattern

in dust settlement that has taken place over the 22 months, which have passed since the re-opening of

the library to the public, is the very first step in establishing a cleaning methodology and routine. In

order to understand the distribution and impact of dust fall on of the plaster statuary installed within

the Rococo Hall, the busts have not been dusted or otherwise cleaned since Autumn 2007.

Figure 34: The plaster bust of Johann Wolfgang von Goethe by Christian Daniel Rauch is placed

on the first level of the Rococo Hall. Behind it stands the colossal marble bust of Goethe by David D´Angers, the pedestal of which is large enough to block visitor’s access to this area.

89

Most published research on dust monitoring in historical houses states the visitor to be the number

one importer of particle substances.261 The air ventilation system in the library hall is fitted with a

HEPA262-prefilter, which standardly retains particulate matter as small as 0.3 μm with an efficiency of

99.97%. The 16 windows were all sealed in the course of the renovation of the building. Comparing

visitor routes to quantity of dust settlement on the plaster busts was therefore the first undertaking of

this study: Up to 300 visitors are permitted entrance into the library on six days of every week. No

more than 25 visitors are allowed in simultaneously at any time. The visitors are required to wear felt

overshoes in order to decrease the amount of coarse-grained soiling particles’ inflow into the room.263

Figure 35: 25 visitors are allowed into the lower level of the Rococo Hall simultaneously in the company of wardens.

The visitors’ route is restricted to the lower of the two floors (see Fig. 35 and floor plan, Fig. 36), and a

large area of this floor is cordoned off to the public. The 1st Gallery floor (see floor plan, Fig. 37), on

which the larger amount of busts is situated – 32 versus 14 busts on the lower floor – is frequented by

staff members only. The volumes, which are kept in these rooms, can be used by order to the

261 For example Gary Thomson; Holly Lawes et al.; Helen Lloyd et al. 2002. 262 Thanks to Thilo Wendt of Ingenieurbüro Six, Rudolfstadt /Weimar for this information. HEPA stands for High

Efficiency Particulate Air. 263 See for instance Barry Knight, p. 19. An obvious defect the wearing of overshoes does have, as commented on

by Konrad Katzer, Chief Conservator at the Foundation of Weimar Classics, is that these relatively bulky overshoes tend to make people shuffle or to drag their feet. This causes scratching to the historical oak floor-boards, when grains of sand or other coarse particles get adhere to the soles of the overshoes.

90

Figures 36 and 37: Floor plan of the lower level, respectively of the 1st Gallery (level two of the Rococo Hall),

indicating the positions of the plaster busts as red circles.

Study Centre, which is situated on the 2nd Gallery floor, one level above the level shown in Figure 14.

Figures 38 and 39, below, show what was, at the outset of this investigation, assumed to be the

distribution pattern of dust settlement in accordance to the presence of human beings in the two levels

of the hall. The areas that are not accessible to the public on the lower level of the hall are cordoned

off. The floor plan of the lower level of the Rococo Hall thus indicates the visitor route (Fig. 38) by the

black dots, which in turn indicate the pattern of dust fall, as the initial assumption for dust distribution

suggested. Likewise, Figure 39 correlates dust fall settlement (indicated by black dots) on the basis of

the knowledge of the transit route for members of staff. This “route” is used some two to seven times a

day, as it is a quick path between two office areas of the library. The 1st Gallery floor is not accessible to

visitors, yet there is some staff activity here and likewise also in the lower hall, in context to book

requests.

A first assessment of the impact level of dust on the plaster surfaces was carried out by naked eye

inspection, followed by photographic recording. The actual picture that presented itself on close and

comparative inspection of the plaster surfaces in respect to the dust that has accumulated on them,

differs slightly to the initial assumption, as is illustrated in Figures 38 and 39 above. On the lower level,

those busts situated within the interior oval, show more dust accumulation than the busts situated in

91

areas of the room further away from the visitor route. Within the oval, the busts that are placed in

closer proximity to the entrance / exit path, show comparatively higher accumulation than those

placed in the rear of the oval. And the individual dust pattern on each of the busts varies. One

outstanding notable and common feature on the surfaces of the busts on the lower level of the hall, is

that dust settles on the top of the bust (crown of the head, in the hair) and on the shoulders. Inspecting

the pedestals’ top surfaces shows that the dust settles largely to the rear of these surfaces. Common

sense suggests that dust will settle in areas that are level and in areas where there is less movement of

air i.e. caused by peoples´ movements and by the air conditioning system.

Figures 38 and 39: Floor plan of the lower level, respectively of the 1st Gallery, indicating the presumed dust

distribution in the two levels of the hall in relation to human activity.

Detailed photos of three busts that are placed on the lower level of the Rococo Hall (cf. floor plan, Figure

41, below) illustrate these findings:

The first bust, a portrait of Johann Wolfgang von Goethe (No. 1 in floor-plan, Figure 41, below), is

situated in an apparently sheltered area, away from the visitor flow. A very large pedestal placed in the

niche in front of the exit door, on which an over-sized marble bust of Goethe is placed, keeps the

visitors out of this area, as do various cordons (cf. Figure 34). The crown of Goethe´s head is evenly

92

covered with smaller to large particles. They seem, with naked eye inspection, to consist mainly of

textile fibres (see Figure 40). Apart from Goethe’s hair and forehead, the shoulders are covered in dust

at the rear of the bust as is also the top surface of the plinth. This same situation is evidenced on the

bust of Johann Karl August Musäus (No 2. on floor plan, Figure 41, see Figures 43 and 44).

Figure 40: The crown of Goethe´s plaster bust, cf. above Fig. 34, cf. No. 1 in floor plan, Figure 41.

The bust of Guillaume Thomas François Raynal, placed within the interior oval of the lower level of the

hall, to which the visitors have access (No. 3 on floor plan, Fig. 41), is shown here to represent those

busts which show a larger amount of dust accumulation. Although the front areas of the bust seem to

be less affected by dust-fall, a close inspection shows a comparatively larger amount of dust evenly

covering all planes that are relatively level.

93

Figures 41 and 42: Floor plan of the lower level, respectively of the 1st Gallery, indicating the

dust distribution in the two levels of the hall, as perceived with naked eye inspection.

Figure 43: Particle matter on the crown of Musäus’ bust’s head, (cf. No. 2 in floor plan, Fig. 41).

94

Figure 44: Dust deposition is very evident on the rear of the pedestal surface and on the rear of Musäus plinth.

Figure 45: Dust accumulation on the bust of Raynal, which is situated in the interior oval of the lower level of the

Rococo Hall (cf. No. 3 in floor plan, Figure 41).

95

The situation on the first gallery regarding dust fall was at first slightly surprising. There is no notable

higher accumulation of dust on the plaster surfaces that are close to the area of transit, as was

assumed and illustrated in Fig. 39. However, a rather large accumulation of particles is distinct on all of

the front surfaces of the busts that are placed along the balustrade and also on the top surfaces of the

respective pedestals. The plaster surfaces of those busts that are not near the void behind the

balustrade, have seemingly accumulated only very little dust. An interesting feature in the settlement of

dust on the busts placed along the “route of staff transit”, is the fact that dust seems to settle in the

respective area away from the route. One bust, indicated in floor plan Figure 42 as No. 4, of Carl August

Böttiger, is situated along the “route”: Its surfaces show no visible dust accumulation. However the

busts situated on the balustrade show a considerable amount of particle matter, which collects foremost

and distinctly on the surfaces facing the void into the lower Gallery.

As stated above, the displacement ventilation system which was installed during renovation of the

building, pumps air into both levels of the library from beneath each window (8 windows on each

level) regularly at 0,3 m/s. The area size of the balustrade of the second Gallery (see Fig. 47) defines

the air space, which is housed in by a glass cube. The space on the second Gallery, which is outside this

glass cube, is used as reading and conference room, on the third level of the Rococo Hall. Inside the

glass cube, alongside its upper outer edges, fitted into the halls’ ceiling, is the outlet system for the

controlled air exchange.

Figure 46: Carl August Böttiger’s bust is situated on the first Gallery (cf. Figure 42, no. 4 in floor plan);

its surfaces show no visible dust accumulation.

96

Figure 47: Ludwig Tieck’s bust is positioned on the gallery balustrade on the second

level of the library hall, No. 5 in floor plan Figure 42.

There is a distinct difficulty in assessing the respective and comparable amounts of dust that

accumulate on surfaces by naked eye inspection only. Firstly, an unknown amount of particle matter is

not so dark as to be distinguishable on a white or light surface. And a large amount of particle matter is

so small that our eyes do not perceive it. Photographing dust as a tool in visual assessment of dust

impact also poses certain problems. Especially those busts, which are situated along the balustrade on

the first gallery and show more dust accumulation on their fronts than on their rears, bear difficulties

in the recording of dust settlement on their surfaces. The front areas of the busts are covered with fine

particles and fibres. However, it is not possible to photograph the front of these busts in detail, because

they are fixed onto the pedestal, which is formed by the balustrade with mounting wax. This problem

applies to all of the busts that are placed along the balustrade.

97

Figure 48: Visualising dust on the left shoulder of the bust of Ludwig Tieck.

Figure 49: Removing particles from a pedestal surface with sticky tape eases visual

assessment of the degree of particle depositions within the bust’s vicinity.

The pattern of distribution of settled particle matter in the Rococo Hall, as is evident with naked eye

inspection, is apparently greatly influenced by the air movements within the two floors of the hall,

which in turn is dictated by the air ventilation system. Although photographic documentation is a

useful aid, this method of evaluation cannot be considered comprehensive. Sampling was thus carried

through on different surfaces relevant to the placement of the busts, at different locations on the two

floor-levels; to verify the distribution pattern, which was previously established with naked eye

98

inspection.264 Qualitative analyses of dust matters, collected off pedestal surfaces on each level of the

library hall, were achieved at the laboratory of the technical college in Erfurt and will be discussed

below. An evaluation of the degree of microbiological infestation present in the collected dust was

realized at the technical college in Hildesheim and will also be discussed below.

Experimental 3: Dust Monitoring – the Distribution Pattern within the Rococo Hall

A period of monitoring by collecting dust on sampling substrates and their direct comparison, was

carried through to attain a quantifiable verification of the distribution pattern as established by visual

inspection, illustrated in floor-plan Figures 41 and 42. Finding an easily implementable method of

assaying dust distribution with subsequent particle evaluation was the imperative aim. Comparative

quantitative analysis was realized using microscopic photography of the samples, followed by image

analysis. Six glass trays were distributed each on the lower level and on the second level of the hall

(Figures 56 and 57, below) for a time period of 4 weeks, from June 22nd to July 20th, 2009.

Figure 50: Positioning the glass trays for dust sampling.

264 The focus of investigation was dust settlement relevant to the positions of the busts. See Barry Knight on the

vertical profiles of dust deposition in historical houses with visitor flow: “Many large dust particles were found up to 30 cm from the floor, either from visitors’ shoes or kicked up from the carpets. These particles were too heavy to rise much higher, and the amount decreased with increasing height. The rate increased again between about 80 to 150 cm, roughly hip-to-head height, and consisted mostly of clothing fibres dislodged by friction. Above 150 cm, the rate decreased rapidly and was very low above 200 cm.”

99

The 12 glass trays were positioned according to visually perceived higher rates of soiling, respectively

behind the bust, at the rear of the pedestal (see Figure 50). The respective position was chosen so that the

placement guaranteed free air space above it, yet at the same time considering a relative rate of sheltering

against visitor or staff movement. At the end of the term of four weeks monitoring, a second glass tray was

placed onto each dust-soiled surface of the sampling trays, fastened with tape and numbered and packed for

safe transferral to the laboratory for microscopic inspection (see Figure 51, below).

Figure 51: Comparing deposition rates by visual Figure 52: Comparing sampling with Gel-lifter

analysis of glass tray sample collectors. and with glass trays.

An additional test series aimed at comparing two different types of sampling material, encountered in

the relevant literature – glass trays and gelatine film.265 Figure 52 illustrates comparative sampling

with a glass tray and Gel-lifter.266 The sampling location for this comparative sampling was 1.3, see.

floor plan Figure 56, the period of time covered was two weeks, from July 20th to August, 3rdThe

question behind this comparative sampling test was whether, as might be supposed, dust might

265 For example Anne Lisbeth Schmidt et al.; David Thickett et al.; Helen Lloyd et al., 2007 a and b; Morten

Ryhl-Svendsen, Maria-Louise Jacobsen; Barry Knight. Holly Lawes et al., Lidia Morawska, Tunga Salthammer (2003b) and Thomas Schneider. There exist many other and more sophisticated methods for analysing particle distribution in interiors. The aim behind this choice of sampling method was to evaluate recordedly easily implementable and easily evaluable methods available to a conservator.

266 The Gel-lifter is produced by P-B Miljø A/S, Enebærvej 7, 8850 Bjerringbro, Midtjylland, DK and is discussed by Morten Ryhl-Svendsen as well as Marie-Louise Jacobsen. My thanks go to Sakarias Petersen-Bach at P-B Miljø A/S for sending me a free sample.

100

adhere more strongly to gelatine and thus be less prone to be removed from the sampling collector by

air movements. A glass tray and a piece of gelatine film were thus positioned next to each other in a

sheltered location as described above and left there for a period of two weeks (Figure 52), after which

they were covered with a glass tray, secured and transported to the lab for microscopic evaluation and

photographic recording.

Figure 53: Comparative microscopic investigation for the assessment of dust distribution.

Experimental 4: Microscopic and Image Analysis of Sampled Dust

Each tray was photographed with magnification factor 6 of the microscope and with a digital camera

with pixel resolution of 2584 x 1936.267 A glass tray measures 76 x 26 mm. At least two photographs

were made of each tray. Those trays that visually showed higher rates of dust accumulation (sample

1.3 and sample 2.4) were photographed in eight sections, each from left to right.

A total of 36 jpg-images, which each covered a photographed area of ca. 15.14 mm x 11.35 mm, which

was visualized at image size 2584 x 1936 pixel, were then the basis for an evaluation of accumulated

distribution with image analysis and subsequent statistical evaluation of particle counts.

267 Microscope: Axioplan, Zeiss. Camera Axiocam MRc 5, Zeiss. Software: Carl Zeiss Axiovision Rel. 4.7.

Microscopic Photography: reflected light, dark field, zoom factor 6 and 25.

101

Figure 54: Sampling location 1.3, image a. Image size 2584 x 1936 pixel, in mm 15.14 x 11.35.

Total particle count 1994. Average particle size 52.618 pixel. Pixel size of image 5.86592179 µm. Average particle size ca. 294 µm (0.294 mm). Note the high amount of scratches, identifiable as thin almost vertical lines.

Figure 55: Detail of image of particle counting, generated with ImageJ.

102

ImageJ,268 is an open source image processing and analysis program written in Java, which allows it to

run on Linux, Mac OS X and Windows. The respective particle counts this study established were

achieved with the following method: The image was loaded in ImageJ (File > Open), then the resolution

was reduced to 8-bit grey scale (> Image > Type > 8 bit). In the next step, a threshold of the image was

set, in order to distinguish the background from the particle matter to be counted. (> Process > Binary

> Make Binary). Then the particles can be counted (> Analyze > Analyze particles: determing options

Size = 0 – infinity; Circularity 0.00 – 1.00; Show = nothing; Display results = on; Summarize = on). This

issues an image, in which each particle is numbered (see Figure 55). A spreadsheet that lists the

requested data, which in this case was particle count and area coverage (cf. table on this page) is also

issued, as well as a summary of all particle counts of one counting session as well as an evaluation of

average, minimum and maximum area coverage (=size) of each counted particle. The respectively highest

particle count of each glass tray was then compared to the respectively lowes. These two respective particle

counts were each chosen for statistical comparison, using excel, see Table 2 and Diagram 16. The area size for each of the counts in Table 2 is determined by the size of the image, which served for

calculation and measures ca. 15.14 mm x 11. 35 mm. The location in the hall, which amassed,

respectively the highest minimum and maximum values is location 1.1 (see floor plan Figure 56, below

and Figure 34, above), Johann Wolfgang von Goethe. The exit situation is just behind the shelf against

which Goethe is placed.

Table 2: Highest and lowest

respective values of total particle

counts on images sized 15.14 x

11.35 of glass tray samplers.

Transferring these particle counts onto the floor-plan illustrations gives a good idea about the

distribution pattern of particle settlement within the hall in a summer month. 269 Comparing the actual

268 http://rsbweb.nih.gov/ij/features.html, discussed by Morten Ryhl-Svendsen as well as Marie-Louise

Jacobsen.

269 Holly Lawes et al., p. 28, have observed that dust deposition is not consistent throughout the year.

Sampling Location In the Rococo Hall MAX Particle

Count

MIN Particle Count

1.1 3545 2622 1.2 1740 1684 1.3 1994 1577 1.4 1602 1465 1.5 2174 1343 1.6 660 527

2.1 1312 716 2.2 1179 823 2.3 1469 1349 2.4 1841 1135 2.5 1201 1090 2.6 1103 741

103

counts with the visual assessment (as shown in floor plan Figures 41 and 42), in fact, show that the

sampling locations within the interior oval on the first level are impacted less strongly than assumed.

Yet in comparison to an area where there is no visitor activity (compare position 1.6), they show a

higher quota of particle deposition. Sampling locations 1.5, 1.3 and 1.4 are similarly affected; they all

represent the interior oval of the lower level of the hall.

Figures 56 and 57: Distribution of 12 glass trays for dust sampling on the two levels of the Rococo Hall. The entrance and exit situations for the visitors are situated at the bottom of the floor-plan, Fig. 56.

Sampling location 1.2. is affected comparatively higher than visual evidence of particle matter on the

plaster surfaces suggests. Here the particle count compares with the values, which particle counting

established for the sampling locations within the interior oval.

The values, which represent particle distribution rates on the upper level of the hall are all within a

similar range. One sampling position is affected slightly higher, position 2.4, on the balustrade. Position

2.6 however, shows a proportionally similar rate of deposition, although this sampling location is

comparatively sheltered. Interestingly enough, the lowest impact values on the second level of the hall,

where there is no visitor activity, are all proportionally higher than in sampling position 1.6 on the

lower level of the hall. The evaluated distribution rate on the 1st Gallery is thus comparable to the dust

impact evaluated for relevant locations on the first level, yet is slightly less pronounced.

104

Diagram 16: Excel-evaluation of two images produced from each of the 12 glass trays.

Diagram 17: Evaluation of 5 particle counts with respect to the particle size distribution.

Image resolution puts a lower limit to the sizes of particle matter that is possible to determine with this

method of evaluation. In the case of these images, 5.586592179 µm equals the size of 1 Pixel. This

means that nothing smaller than 0,0056 mm could be visualized and counted. The particle counts,

which were realized, show that the number of particles, which have area coverage sizes of 1 pixel, are

1.1 1.2 1.3 1.4 1.5 1.6 2.1 2.2 2.3 2.4 2.5 2.60

500

1000

1500

2000

2500

3000

3500

4000

Particle counts from 12 locations within both levels of the hall

MAX Particle CountMIN Partic le Count

Sampling location

Par

ticle

cou

nt

105

particularly high; see graph evaluation above (Diagram 17), which correlates area coverage with

particle count. There is an even scale in the distribution of sizes up to ca. 1000 pixel, which is 5,8 mm

(thus presumably representing fibrous structures), which increases proportionally to the decrease in

size into the range of smaller matter. One can hence assume, that the proportional linearity in particle

size distribution is also present below particles with area coverage in the size of less than 1 pixel (= 5.8

µm). That particulate matter however could not be visualised and could therefore not be assessed

with the method of quantitative evaluation applied here.

Diagram 18: Comparison of particle size distributions.

Diagram 19: Comparison of particle size distributiuons in the range below 50 pixel (=293 µm, = 0.29 mm)

These curves (Diagram 18 and 19) represent a pronouncedly similar proportional distribution of

particle sizes, of two particle counts with very differing absolute values (3545 vs. 660, see Table 2, p.

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 500

500

1000

1500

2000

2500

3000

3500

Paricle Size Distribution, Sample 1.1Pixel Sizes 1 - 50

Particle Sizes in Pixel

Parti

cle

Cou

nt

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 500

100

200

300

400

500

600

700

Particle Size Distribution, Sample 1.6Pixel Sizes 1 - 50

Particle Size Distribution

Par

ticle

Cou

nt

106

100). This distribution rate matches to that, which was evaluated for the first gallery, see Diagrams 20

and 21.

Diagram 20: Comparison of particle size distributiuon in the range below 50 pixel on the first Gallery.

Diagram 21: Comparison of particle size distributiuon in the range of below 50 pixel on the first Gallery, cumulatively sampled in a sheltered area, shows a very similar distribution curve to the curve representing the situation on the balustrade, see Diagram 20 above.

Errors in particle counting can be introduced by smudging and scratches on the glass (see Figure 54).270

All images show a large amount of scratches. Increasing contrast values in areas of smudging, diminishes

count information, which impacts the value. This method however does prove to be a relatively quick one

to gain an overall idea of various sized particles present, while realizing that a large contingent of particle

matter is not comprehensive part of deposited matter.

270 Scratches were apparent on all images. Only some images had smudges, these were corrected by

manipulating contrast, which automatically limits information on smaller particle matter.

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 500

200400600800

10001200140016001800

Particle Size Distribution, Sample 2.4Pixel Sizes 1 - 50

Particle Sizes in Pixel

Part

icle

Co

un

t

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 500

200400600800

10001200140016001800

Particle Size Distribution, Sample 2.6Pixel Sizes 1 - 50

Particle Size in Pixel

Par

ticle

Cou

nt

107

Summarizing, it can be stated that dust impact is more pronounced on the lower level of the hall, the

most heavily impacted area being the entrance/exit situation. The interior oval is relatively

homogenously impacted with a comparatively high rate of depositions. The airflow is induced by the

air exchange system and is upward bound within the library hall. Thus a relatively even distribution of

amount of particle matter, which is proportionally also evenly scaled in its size distribution, can settle

over both levels of the hall equally.

Evaluating the use of Gel-lifter versus glass trays met with a simple mistake: the refraction index of

gelatine271 is different to that of glass (refractive index = 1.5). Thus the particles counted on the gel

samples (photographed in reflected light and evaluated with the method described above) were

approximately one third of the particle number counted on the glass tray. A comparative evaluation of

these two types of samplers could therefore not be achieved on the basis of the images produced in

reflected light.

Experimental 5: Dust Analysis with XRF and FTIR and Chromatographic Methods

Two samples of loose particle dust, which had been swept together with a soft brush from various

pedestal surfaces in direct proximity to the busts on the two levels of the Rococo Hall seemed to consist

mainly of fibres fluffed together densely.

Non-fibrous materials were shaken out of the bunch of intertwining dust and were analyzed with X-

ray fluorescence.272 The elements Si, S, Ca, Fe, were detected most strongly, as well as traces of Ti and

Zn. Fourier-Transform-Spectroscopy identified plaster, calcium, siliciumdioxide (sand) as well proteins

in the sample, which was collected from the lower level of the library hall. Exactly the same material

components were identified on the sample of dust collected from the second level. Thermally assisted

hydrolysis and methylation Pyrolysis-Gaschromatography/Mass-spectrometry detected a large

amount of crushed fibrous material, within the non-fibrous mix of sampled dust.273 It identified:

drying oils, palm-/coconut oils and fats, animal oils and collagenous proteins, carbohydrates, cellulose,

271 Method of identification of gelatines refractive index is described in:

http://webphysics.davidson.edu/faculty/dmb/edibleopticalmaterials/find_n.htm

272 Cf. Appendix III, pp. 2-15.

273 CF. Appendix III, pp. 16-18.

108

di-terpenodide resins and the synthetic resins: acrylic resin, isophthalic acid resin, polystyrol and

polystyrol-copolymers, orthophthalic acid alkyd resin, polystyrene and polystyrene-copolymere as

well as phenolic resin. The results conclude the organic matter present in the sampled dust to be

composed of mainly fibrous material, skin flakes and particles of paint.

Figure 58: Dust sampled by dusting with a soft paintbrush, cf. Figures 5 and 6 for a 25-fold magnification.

The pH value of the collected dust was determined from the particles, which had been sampled on

the glass-trays after the microscopic analysis was accomplished, and was established at 7.766.274

One sample with loose dust particles, sampled with the same method as described above (Fig. 58), and

also with sticky tape275, respectively with strips of Gel-lifter samplers that had both contact-sampled

dust from book-shelves, were sent in to a specialized laboratory for identification of microbiological

infestation.276 There the samples were dyed with Cotton Blue (to be able to detect chitin and cellulose),

and cultivated in two different growth mediums (malt extract medium and di-chloran-glycerin-agar).

Microscopical investigation showed no marked infestation. The amount of bacteria and fungi (mainly

formed by types of penicillium and aspergillium), which was thus cultivated, was not conspicuously

274 Carried out by Frank Mucha: The particles were set into 5 ml of Rectapur water, VWR (pH value of 6.280).

The pH meter used was Lab 850, Scott instruments. The pH was read as 7.766 when a constancy of the electrode transmitted value was achieved (electrode: Blue Line 27, Schott instruments). See Anne Lisbeth Schmidt et al. who note a pH value of 7.6 as being non-aggressive.

275 Tesa klar.

276 Cf. Appendix IV.

109

different to the amount to be expected in “usual house dust.” The analysis report closes with the

remark, that especially in view of the climate values recorded within the library hall, neither the types

nor the extent of infestation were at all conspicuous.

The comparative evaluation of sticky tape versus Gel-lifter as sampling aids proved microscopically

identical, both in the amount and type of adhering dust particles.

Interestingly enough however, a relatively large amount of skin particles were analysed on the sticky

tape. While sampling, the Gel-lifter had in fact been much easier to use than the sticky tape, because it

is fitted with a non-sticky end piece and has a nice thick consistency, which facilitates its handling.

Sticky tape has the tendency to stick to ones fingers whilst one tears it off the role, which proved

slightly bothersome during sampling. This then became very evident with microscopic analysis in

transmitted light, when fingerprints, which were on the sticky tape, were visualized in blue (Cotton

Blue).277

Evaluating a Maintenance Plan for the Busts in the Rococo Hall

The settlement rates of particle matter vary within the two levels of the library hall. Yet the distribution

pattern as analyzed, indicates that even those areas, which visually do not account for an increased

deposition of particle matter, show a respectable amount of dust accumulation.

Initially, the planning for the busts’ maintenance revolved around establishing a cleaning routine in the

order of an established rate of particle depositions, in relation to different relevant positions in the

Rococo Hall. This plan was however adapted, after microscopic and image analysis assisted toward

determining factual evidence of a relatively even particle distribution.

Routine cleaning of all plaster surfaces with a vacuum cleaner forms the basis of the proposed

maintenance plan. A trial cleaning was conducted on all the surfaces, which allowed for a calculation of

the time, in which this measure can prudently be carried out. Visual examination determined the

point of time in which sufficient surface cleaning could be accomplished by vacuuming.

There is obviously a limit to visual estimation of remaining particle matter on surfaces, 278 just as

there is also a limit to a vacuum cleaner’s ability to remove all adhering particles from a porous

surface, while its nozzle is kept in a safe enough distance in order to diminish possible impact. This

limit is taken into account on the basis of proposing a relatively tight cleaning routine and with the

perspective of future poultice-cleaning in those cases, in which the plaster surfaces commence to

277 Appendix IV, p. 3.

278 On this largely Nazaroff et al., pp. 91-99, who developed a method for predicting visual appearance based upon surface coverage and compared the predictions with the results of optical analysis.

110

‘grey’ due to gradual accumulation of very fine particle matter. Only time and close observation will tell

whether this proposed plan will work in respect to the bust’s longer-term appearances.

Sampling location 1.3 shows a fair average high impact level and can serve as a good example when

assessing the routine, in which regular cleaning should best be carried through. The total number of

particles, which settled on the glass tray over four weeks, and could be visualized, was established with

the method described above, at 3545 counts over an area of 15.14 mm x 11. 35 mm. Those particle

matters that were too small to be visualized remain an unknown factor; yet can assumed to be

considerable.

An acceptable absolute rate of particle deposition that should prompt its removal, is however difficult

to asses, as it is purely subjective. My personal recommendation would be to vacuum-clean the plaster

surfaces every eight weeks. Three hours of specialist cleaning effectively cost ca. 120,- €.279 The

established cost budget for housekeeping in the Rococo Hall will ultimately accommodate these man-

hours. The maximum rate of soiling, which the decision-makers consider to be acceptable, will

ultimately determine the intervals between cleaning and thus the yearly expenditure for the plaster

busts’ maintenance.

Experimental 6: Vacuum Cleaning the Plaster Surfaces

A backpack vacuum cleaner280 and a small ladder were necessary to follow through the planned

cleaning trial. The time it effectively took to vacuum all bare plaster surfaces in the Rococo Hall was a

little under three hours, on August 3rd. 2009. This includes the time spent on the ladder, rewiring and

re-plugging as well as examining surface qualities in regard to their degree of soiling versus their

degree of cleaning. The vacuum cleaner’s nozzle was purposefully not fitted with any ancillary

material, such as a brush or a cloth. The logic behind this, claims that a method for non-contact

removal of loose particle matter need not necessarily fit its instruments for the contingent case of

279 This is the gross amount, which is basically equivalent for free-lance contraction as well as the gross cost per

hour for employed conservation staff which is ca. 35,- to 40,- € per hour.

280 In this case type 25.8230 KÄRCHER was used. It has a permeability rate of 0,1% and is equipped with paper filter bags type 25.8374 KÄRCHER and with fleece filters type 25.8375 KÄRCHER, which is a dust-filter type M (specified as MAK > 0,1 mg/m³ and for wood-dust). The products differ on the different websites of KÄRCHER (.com; .co.uk; .ch; .at). This product information was researched at www.KÄRCHER.at.

111

Figure 59: Dust on Dantes’ shoulder.

Figure 60: Dust removed off Dantes’ shoulder.

impact, because impact is just not going to occur. The fact that there is no padding material, will most

probably exact a higher degree of caution. The nozzle, which was used, has a diameter of 3,5 cm; it

was kept in a safe enough distance from the surfaces during vacuuming, while at the same time

carefully trying to reach all surface depositions for suction. The minimum timed effort for cleaning a

bust was 3 minutes, and the maximum 10 minutes, depending on rate of soiling, on size and

sculptural volumes of each respective bust.

112

Perception of levels of dustiness versus levels of cleanliness is subjective,281 photographic

evaluation also has its limits (see Figs. 59 and 60 above). My own subjective perception was the basis

for establishing the achieved condition as being sufficiently clean.

Figure 61: Vacuum cleaner and ladder.

281 See for instance Lloyd et al. 2007, p. 140.

113

Conclusions

As far as the five surface topographies of paint versus soiling that were analyzed within this research,

can represent the entirety of maintenance history of the plaster surfaces in the Rococo Hall, the paint

coatings on the plaster busts have been identified as aesthetically motivated measures of treatment.

Evaluation of the comparative analyses suggests that one first coating was applied at some point in

the later 19th century, definitely after 1857, perhaps before 1896. A second coating was applied most

presumably in the first half of the 20th century, perhaps in 1941. Both of these applications can be

categorized as corrective, because they covered soiled surfaces and also past restorations. Their

preventive action can also be categorized, in that they protected the surfaces from further soiling

while facilitating future cleaning.

Cleaning plaster by washing down their surfaces with (soapy) water is documented for Germany in

the late 19th century. Coating heavily soiled surfaces with white-coloured paint is also documented in

Germany for this period of time and later. The survey this study undertook, assessed that this

treatment method is not particular for Germany. Plaster has apparently very often been coated after

its surfaces accumulated soiling. Subsequent re-coating is an equally common phenomenon.

In view of the question that prompted this research, whether a new treatment of surface sealing

should be applied to these busts, the point of conclusion of this study is:

While plasters´ material properties do call for heightened attention toward appropriate preventive

thought and action, the conservation measure of applying paint onto (soiled) plaster surfaces was in

fact largely habitual in a period of time, when appropriate methods of dry cleaning plaster – such as a

vacuum cleaner – were not yet developed.

The original display of the plaster busts’ surfaces at their accession into the library’s collection was an

uncoated one. This should also be taken into account, while appreciating their restored and conserved

appearances and while reflecting on methods for their preservation.

The plan for the preventive care and maintenance, which this study established, bases on the given

factors of a stable, controlled and calculable environment, in which these restored plaster surfaces are

now displayed, as well as on the relatively low (and controlled) number of visitors to the library hall.

The pattern of dust distribution has been determined in relation to the plaster busts’ positions of

installation, and the quality of particulate matters, which are present in the library’s environment,

have been analyzed. There is no conspicuous feature in their material qualities.

The vacuum cleaner (when fitted with appropriate filters to reduce recirculation of sucked up

particles back into the environment) can be regarded as the most appropriate instrument for

regularly removing dust from these plaster surfaces, in situ. A routine for vacuuming the plaster

114

surfaces is proposed within this study. It aims at allowing for as little accumulation of particulate

matter onto the surfaces as is practicable.

Time as well as close observation will tell whether this proposed methodology of maintenance is

sustainably effective, taking into account that smallest fractions of particle matter are prospectively

not going to be effectually removed with the proposed cleaning method. A gradual greying of the

plaster surfaces might at a later point become evident. In that case, poulticing methods for cleaning

plaster can become effective to reduce the adhering soiling. All prior abrasive methods however, to

which one can also to an extent count the use of a soft brush on pre-soiled plaster, are not

recommended. Apart from their possibly abrasive action in the presence of particle matter on the

plaster surface, experience shows that soiling can thereby be introduced into deeper surface zones.

Cementation processes of dust to plaster surfaces require future study and analysis, with particular

focus on the humidity values. But meanwhile, this study on the surfaces of the plaster busts in the

Rococo Hall and on the methods for their sustained conservation, ends with Wilhelm von Humboldt´s

recommendation: “Keep the plasters clean, as that is the main thing.”

115

An explanation of why the research has made a valuable contribution to Preventive

Conservation

This research on the plaster busts of the Rococo Hall followed multiple foci: characterization of their

materiality, analysis of their environment and reflection on the history of their display and treatment.

Its intent was to raise awareness for plaster’s material properties and its susceptibilities – in reference

to establishing an appropriate approach for these busts’ future sustainable conservation in their site of

installation.

Preventive conservation involves an overall understanding and working knowledge of an object of art

and of its composition. Determining agents of decay in the environment that can impact its

composition, provides the objective evidence necessary for risk assessment. While investigating the

environmental parameters of the situation of the plaster busts’ installation in the Rococo Hall, this

research hence also aimed at understanding the busts’ history of maintenance and display by

analyzing some surface profiles of soiling versus coating.

Knowledge of an objects’ history contributes to the understanding of its cultural and material values.

This understanding was further broadened by comparing one particular tradition in maintenance and

treatment of plaster surfaces to past and current practices in other collections of plaster statuary.

These different lines of research have provided the ground for an assessment of a viable concept and

methodology for these busts’ future maintenance and which was ultimately achieved by the practical

cleaning trials that were carried through on the plaster surfaces. It is hoped that the collection of

plaster casts in the Rococo Hall of the Duchess Anna Amalia Library in Weimar will benefit by this

research, in that the future approach to their maintenance has received an appropriate framework of

contextual and material understanding. It is likewise hoped that the valuable contribution this

research makes to preventive conservation is perceived to apply not only to these particular objects,

but also to an entire genre within sculpture – to the plaster cast and its surface qualities.

116

Thanks

I wish to thank Konrad Katzer and Michael Knoche for their very kind support and interest in my

study as well as for their trust in its realisation. The Herzogin Anna Amalia Biblitohek (HAAB) and the

Klassik Stiftung Weimar (Foundation of Weimar Classics) very generously supported this study with a

scholarship as well as with funding the analysis, for which I am very grateful. My thanks also go to

Bettina Werche, Gabriele Oswald, Ulrike Müller-Harang, Katharina Krügel and Michael Oertl for

valuable information on the Art Collection of the Duchess Anna Amalia Library. I would like to thank

Jean Brown for her steady and patient support throughout this last year as well as for her trust in my

ability to conclude this thesis. And I wish very much to thank my parents, and especially Mum, for

scrutinizing my English.

Aurelia Badde, Sept. 13th, 2009

117

References

Ingrid Arnhold, Ein Zeitalter wird besichtigt. Die Großherzogliche Bibliothek 1832-1871, in: Herzogin Anna Amalia Bibliothek – Kulturgeschichte einer Sammlung, Munich 1999, pp. 108-117.

Lorenza D’Alessandro, Francesca Persegati, Scultura e Calchi in Gesso, Storia, tecnica e conservazione, Rome 1987.

Barbara Appelbaum, Guide to Environmental Protection of Collections, Madison, Connecticut, 1991.

Aurelia Badde, Bernhard Illerhaus, 3D-μCT reveals the work processes of cast plaster statues, their internal structures, damages and restorations, in: Optical Methods for Art and Archaeology, Renzo Salimbeni, Lucca Pezzati (ed.), Proceedings of SPIE Vol. 5857, 2005, 58570U-1.

Norbert S. Baer, Paul N. Banks, Indoor Air Pollution: Effects on Cultural and Historic Materials, in: The International Journal of Museum Management and Curatorship, 1985, Vol. 4, pp. 9-20, reprinted in: Simon J. Knell (ed.), The Care of Collections, London 1994, pp. 150-163.

Hannah J. Bainbridge, An introduction to Preservation Housekeeping at the Bowes Museum, Preventive Conservation, Course Documents, CV0700BNN, Preventive Conservation, School of Applied Arts and Sciences, Northumbria University, 2008/2009.

Malcolm Baker (1995), The Portrait Sculpture, in: David McKitterick (ed.), The Making of the Wren Library, Trinity College Cambridge, Cambridge 1995, pp. 110-137.

Malcolm Baker (2001), A Sort of Corporate Company, Approaching the Portrait Bust in it’s Setting, in: Exhibition catalogue ’Return to Life: a new look at the portrait bust‘, Henry Moore Institute, National Portrait Gallery, Scottish National Portrait Gallery, Penelope Curtis et al. (ed.), Leeds 2001, pp. 20-35.

Marcello Barbanera, Les Collections de Moulages au XIXe Siècle: Étapes d’un Parcours entre Idéalisme, Positivisme et Esthétisme, in: Les Moulages de Sculptures antique et l’Histoire de l’Archéologie, Actes du Colloque International, 24e octobre 1997, Geneva 2000, pp. 57-73.

Mary Beard, Cast: Between Art and Science, in: Les Collections de Moulages au XIXe Siècle: Étapes d’un Parcours entre Idéalisme, Positivisme et Esthétisme, in: Les Moulages de Sculptures antique et l’Histoire de l’Archéologie, Actes du Colloque International, 24e octobre 1997, Geneva 2000, pp. 157-166.

Massimo Becattini, Il “gesso” in età neoclassica, in: Il mondo antico nei calchi della Gipsoteca, Exhibition catalogue, Cassa di Risparmio di Firenze, Florence 1991, p. 17.

Maike Berchthold, Gipsabguss und Original, ein Beitrag zur Geschichte von Werturteilen, dargelegt am Beispiel des Bayerischen Nationalmuseums München und anderen Sammlungen des 19. Jahrhunderts, Dissertation Thesis, Universität Stuttgart, 1987.

Joachim Berger, Anna Amalia von Sachsen-Weimar-Eisenach (1739-1807), Denk- und Handlungsräume einer »aufgeklärten« Herzogin, in: Ereignis Weimar-Jena. Kultur um 1800. Ästhetische Forschung 4, Heidelberg 2003.

118

Ursel Berger, Kopf ohne Büste, Zum Plastischen Porträt im 20. Jahrhundert, in: Exhibition catalogue ’Hauptsache Köpfe: plastische Porträts von der Renaissance bis zur Gegenwart aus der Skulpturensammlung‘, Staatliche Kunstsammlungen Dresden 2001, pp. 40-46.

Didier Besnainou, Un nouveau gel de nettoyage des plâtres, in: Le Plâtre, l’art et la matière, Paris 2001, pp. 375-378.

Jürgen Beyer, Die Baugeschichte, Vom Herzogsschloss des 16. Jahrhunderts zur Großherzoglichen Bibliothek von 1849, in: Die Herzogin Anna Amalia Bibliothek, Nach dem Brand in neuem Glanz, Berlin 2007, pp. 125-144.

Tony Birks, The Alchemy of Sculpture, Abingdon, Somerset, Gloucestershire, 1998.

Peter Bloch, Schadow und die Europäische Bildhauerkunst seiner Zeit, in: Johann Gottfried Schadow und die Kunst seiner Zeit, Exhibition catalogue Düsseldorf, Nürnberg, Berlin, Cologne 1994, pp. 33-40.

Alfred Bohnhagen, Der Stukkateur und Gipser, Leipzig 1914, Reprint Zentralantiquariat der DDR, Leipzig 1987.

Adolf H. Borbein, Zur Geschichte der Wertschätzung und Verwendung von Gipsabgüssen antiker Skulpturen (inbesondere in Deutschland und Berlin), in: Les Moulages de Sculptures antique et l’Histoire de l’Archéologie, Actes du Colloque International, 24e octobre 1997, Geneva 2000, pp. 29-43.

Peter Brimblecombe, The Composition of Museum Atmospheres, in: Atmospheric Environment, Vol. 24B, 1990, No. 1, pp. 1-8.

Peter Brimblecombe, Damage to Cultural Heritage, in: Air Pollution and Cultural Heritage, Proceedings of the International Workshop on Air Pollution and Cultural Heritage, Sevilla, Spain, December, 1st-3rd 2003, Leiden 2004, pp. 87-90.

Peter Brimblecombe, Carlotta Grossi, The rate of darkening of material surfaces in C. Saiz-Jimenez, in: Air Pollution and Cultural Heritage, Proceedings of the International Workshop on Air Pollution and Cultural Heritage, Sevilla, Spain, December, 1st-3rd 2003, Leiden 2004, pp. 193-198.

Peter Brimblecombe, Kate Frame, David Howell, Barry Knight, Helen Lloyd, David Thickett, Young Hun Yoon, Accumulation and Cementation of Dust Indoors, 6th Indoor Air Quality 2004 Meeting (IAQ2004). Padova, Italy, November, 10th-12th 2004 [http://www.isac.cnr.it/iaq2004/ last accessed on September 3rd 2009]

Jean Brown, Introduction to Environmental Management, Environmental Management, Course Documents, CV0701BNN, Preventive Conservation, School of Applied Arts and Sciences, Northumbria University, 2008/2009.

John Cage, Busts and Identity, in: Exhibition catalogue ’Return to Life: a new look at the portrait bust‘, Henry Moore Institute, National Portrait Gallery, Scottish National Portrait Gallery, Penelope Curtis et al. (ed.), Leeds 2001, pp. 36-48.

Hans-Ulrich Cain, Gipsabgüsse. Zur Geschichte ihrer Wertschätzung, in: Anzeiger des Germanischen Nationalmuseums, Nürnberg 1995, pp. 200-215.

119

Dario Camuffo, René van Grieken, Hans-Jürgen Busse, Giovanni Sturaro, Antonio Valentino, Adriana Bernardi, Nigel Blades, David Shooter, Kristin Gysels, Felix Deutsch, Monika Wieser, Oliver Kim, Ursula Ulrych, Environmental monitoring in four European museums, Atmospheric Environment 35, Supplement 1, Elsevier Science Ltd., pp. 127-140.

CCI Notes 12/2, Care of Objects Made of Plaster of Paris, Canadian Conservation Institute, Ottawa 2002.

Frédérique Chantepie, Un agent de démoulage réversible faisant écran à la migration des huiles de silicone dans le matériau plâtre: Etude et application, in: Techne, Nº25, 2007, pp. 51-57.

Amélie Chedeville, Etude et restauration de Sabazios: sculpture gallo-romaine en tuffeau, II°-III° siécle, Tours, musée de L’Hôtel Goüin, Inv. 2005.029.0001, diploma thesis, Ecole supérieure des Beaux-Arts Tours, cycle conservation restauration des œuvres sculptées, 2008 (unpublished manuscript).

Catherine Chevillot, La question des revêtements de surface des plâtres du XIXe siècle, in: Le Plâtre, l’art et la matière, Paris 2001, pp. 375-378.

Bob Child, Insects, Environmental Management, Course Documents, CV0701BNN, Preventive Conservation, School of Applied Arts and Sciences, Northumbria University, 2008/2009.

Philippe Clérin, Das große Buch des Modellierens und Bildhauens, Stuttgart, Wien 1993 (2nd edition)

Jeremy Colls, Air Pollution. An introduction, London 1997, pp. 341.

Louise Cone (1996), The Cleaning of Plaster of Paris Sculptures, Semester Report, Konservatorskolen, Monumentalkunst, Copenhagen 1996 (unpublished manuscript).

Louise Cone (2005), Fra snavs til patina, Afrensning af gipsskulpturer i Danmark – historisk set, in: Afstøbningssamlingen – død eller levende?, Copenhagen 2005, pp. 119-125.

Berthold Daun, Die Bemalung antiker Gipsabgüsse, in: Museumskunde, Vol 11, pp. 193-197.

James R. Druzik, Glenn A. Cass, A New Look at Soiling of Contemporary Paintings by Soot in Art Museums, in: Third Indoor Air Quality Meeting (IAQ), Oxford 2000 [see: http://iaq.dk/iap/iaq2000/2000_contents.htm last accessed on September 3rd 2009]

Michalis Doulgeridis, Maria Kliafa, Plaster Sculptures on Exhibition, in: The Object in Context: Crossing Conservation Boundaries, Contributions to the 21st Munich Congress (August, 28th-September, 1st 2006), David Saunders, Joyce H. Townsend, Sally Woodcock (ed.), London 2006, p. 303.

Siegrid Düll, Bemerkungen zur Kulturgeschichte des Gipabgusses. Zum Wert der Gipsabgußsammlungen bei Wilhelm von Humboldt und Johann Wolfgang von Goethe, in: Das Spiel mit der Antike: zwischen Antikensehnsucht und Alltagsrealität; Festschrift zum 85. Geburtstag von Rupprecht Düll. Sigrid Düll, Otto Neumaier, Gerhard Zecha (ed.), Arianna Vol. I, Möhnesee 2000, pp. 213-222.

Sibylle Einholz (1991), „Feiner, weißer… Gips!“ Zur Bedeutung eines umstrittenen Materials. 100 Jahre Gipsformerei der Staatlichen Museen in Charlottenburg, exhibition edition, Heimatmuseum Charlottenburg, Berlin 1991.

120

Sibylle Einholz (1992), Enzyklopädie in Gips. Zur Sammlungsgeschichte der Berliner Museen, in: Bär von Berlin, Vol. 41, 1992, pp. 75-96.

Sibylle Einholz (1996), Orte der Kontemplation und Erziehung. Zur Geschichte der Gipsabgußsammlungen in Berlin, in: Zu der Meisterwerke mittelalterlicher Skulptur. Die Berliner Gipsabgußsammlung, Hartmut Krohm (ed.), Berlin 1996, pp. 11-39.

Klaus Fahrner, Zwei Schillerporträts aus der Kunstsammlung der Weimarer Bibliothek, in: Herzogin Anna Amalia Bibliothek – Kulturgeschichte einer Sammlung, Munich 1999, pp. 104-106.

Jochen Flade (2009a), Die Generalsanierung der Herzogin Anna Amalia Bibliothek in Weimar nach der Brandkatastrophe 2004. Darstellung der holzrestauratorischen Maßnahmen im Rokokosaal, in: VDR Beiträge zur Erhaltung von Kunst und Kulturgut, Vol. 1, 2009, pp. 29-51.

Jochen Flade (2009b), Die bauliche und denkmalpflegerische Sanierung – Erläuterungen zu den Holzrestaurierungen im Rokokosaal, in: Instandsetzungspraxis an der Herzogin-Anna-Amalia-Bibliothek in Weimar, Gerd Geburtig (ed.), Stuttgart 2009, pp. 90-122.

Rutherford John Gettens, George Leslie Stout, Painting materials: a short encyclopaedia, New York, 1966.

Adolf Glaßbrenner, Berliner Volksleben – Bd. 1, Leipzig 1847 [http://opus.kobv.de/zlb/volltexte/2007/1252/ POI, last accessed on September 3rd 2009]

Gispdatenbuch, Bundesverband der Gips- und Gipsbauplattenindustrie e.V., Darmstadt 1995.

Johann Wolfgang von Goethe, Weimar Edition, (Sophien-Ausgabe), Weimar 1887-1912, Online-Resource, Cambridge, ProQuest Information and Learning Co, 1997.

René van Grieken, Filip Delalieux, Kristin Gysels, Cultural heritage and the environment, in: Pure and applied Chemistry, Vol. 70, No. 12, 1998, pp. 2237-2331.

Ute Griesser, Problematik der Reinigung empfindlicher Oberflächen; Dargestellt an Gipsskulpturen aus den Environments von George Segal aus dem Museum Ludwig, Köln, Diploma Thesis, Fachbereich Konservierung und Restaurierung von Kunst- und Kulturgut der Fachhochschule Köln, Cologne 1994 (unpublished manuscript).

Carlota Grossi, Young Hun Yoon and Peter Brimblecombe, Indoor Air Quality in Museums and Historic Properties, IAQ 2003 Norwich Poster Abstracts [http://www.iaq.dk/iap/iaq2003/posters/soiling.pdf last accessed on September 3rd 2009]

Walther Grunwald, Die Gebäudesanierung, in: Walther Grunwald, Michael Knoche, Hellmut Seemann (ed.), Die Herzogin Anna Amalia Bibliothek, Nach dem Brand in neuem Glanz, Berlin 2007, pp. 17-121.

Walther Grunwald, Michael Knoche, Hellmut Seemann (ed.), Die Herzogin Anna Amalia Bibliothek, Nach dem Brand in neuem Glanz, Berlin 2007.

Stephen Hackney (1984), The Distribution of Gaseous Air Pollution within Museums, in: Studies in Conservation, Vol. 29, Nº3 (Aug. 1984), pp. 105-116

121

Stephen Hackney (2008/2009), Atmospheric Pollutants, Lecture in Environmental Management, Course Documents, CV0701BNN, Preventive Conservation, School of Applied Arts and Sciences, Northumbria University, 2008/2009.

Matthias Hageböck (2005), Der Brand in der Herzogin Anna Amalia Bibliothek und seine Folgen aus der Sicht des Bucheinbands, Einbandforschung Vol. 16 (April 2005), pp. 8-13 [http://www.uni-muenster.de/Forum-Bestandserhaltung/forum/hageboeck.html last accessed on July 3rd 2009]

Mathias Hageböck (2007): Vorbereitung und Durchführung der Buch- und Papierrestaurierung nach dem Brand in der Herzogin Anna Amalia Bibliothek, in: Thüringer Museumshefte, 2007, Vol.16/2, pp. 19-25.

Ursula Haller, Ulrich Schießl, Reinigung ungefasster Gipsoberflächen – eine neue Methode, in: Zeitschrift für Kunsttechnologie und Konservierung, 12, Vol. 2, Worms 1998, pp. 274-282.

Pamela B. Hatchfield, Pollutants in the Museum Environment: Practical Strategies for Problem Solving in Design, Exhibition and Storage, London, 2002, pp. 203.

Bernd Hering, Weiße Farbmittel, Fürth 2000.

Corinna Herrmann, Cornelia Ripplinger: Herzogin Anna Amalia Bibliothek II. „Mengenrestaurierungen“: Konzept und Umsetzung am Beispiel eines Vergabepakets wasser- und hitzegeschädigter Papiereinbände, in: Papierrestaurierung 8, 2007, Vol. 4, pp. 14-16.

Günter S. Hilbert, Sammlungsgut in Sicherheit, Beleuchtung und Lichtschutz, Klimatisierung, Sicherungstechnik, Brandschutz, Berlin, 2002.

Candida Hofer, Weimar, Munich 2007.

»Ihre Kaiserliche Hoheit« Maria Pawlowna – Zarentochter am Weimarer Hof, Exhibition catalogue, Foundation of Weimar Classics and Art collections, Weimar 2004.

L. Hüttmann, Der Gipser als Zementierer, Tüncher und Stukkateur, Weimar 1842, revised and reprinted by. R. Tormin, Weimar 1883, reprinted in Fulda 1996.

Maria-Louise Jacobsen, Dust – developing a method for sampling and analyzing dust in museums, in: Museum Microclimate, Copenhagen 2007, p. 27 http://www.natmus.dk/sw53828.asp last accessed on September 3rd 2009]

Ingeborg Kader, Gipsabgüsse und die Farbe “Weiss” in: Les Moulages de Sculptures et l’histoire de l’Archéaologie, Actes du colloque international, Paris 24e octobre 1997, Geneva 2000, pp. 121-155.

Frank Matthias Kammel, “(…) eine ebenso belehrende wie unterhaltende Anschauung (…)” Zur Geschichte der Abgußsammlung nachantiker Skulptur an den Berliner Museen, in: Jahrbuch Preußischer Kulturbesitz, Vol. 28, 1991, pp. 159-193.

Johanna Kapp, Etude et restauration de trois sculptures du Château Musée de Dieppe · Saint Jacques en plâtre d’Eugène Paul Bénet (vers 1898) · Tête de saint Jacques et Livre, fragments en pierre de Caen (fin XVe/début XVIe siècle). Diploma Thesis, Ecole supérieure des Beaux-Arts Tours, cycle conservation restauration des œuvres sculptées, 2007 (unpublished manuscript).

122

John Kenworthy-Browne, Private Skulpturen-Galerien in England 1730-1830, in: Klaus Vierneisel, Gottlieb Leinz (ed.), Glyptothek München 1830-1980. Jubiläumsausstellung zur Entstehungs- und Baugeschichte, (Exhibition Catalogue, Glyptothek München), Munich 1980, pp. 334-353.

Moritz Kiderlen, Die Sammlung der Gipsabgüsse von Anton Raphael Mengs in Dresden, Dresden 2006.

Barry Knight, Dust in Historic Houses, Conservation and Management, in: Conservation Bulletin, Issue 45, 2004, pp. 18-24.

Michael Knoche (1999a), Prekäre bauliche Verhältnisse, in: Herzogin Anna Amalia Bibliothek – Kulturgeschichte einer Sammlung, Munich 1999, pp. 216-217.

Michael Knoche (1999b), Bestandentwicklung in Bänden, in: Herzogin Anna Amalia Bibliothek – Kulturgeschichte einer Sammlung, Munich 1999, p. 230.

Michael Knoche (1999c), “Ein anderer Geist war über Hof und Stadt gekommen”, in: Gerhard Schuster, Caroline Gille (ed.), Wiederholte Spiegelungen, Weimarer Klassik, 1759-1832, catalogue for the permanent exhibition in the Goethe-Nationalmuseum, Munich 1999, pp. 35-44.

Michael Knoche (2005), The Herzogin Anna Amalia Library after the Fire, in: IFLA Journal 31, 2005, pp. 90-92, [http://archive.ifla.org/V/iflaj/IFLA-Journal-1-2005.pdf last accessed on September 3rd 2009]

Michael Knoche (2006a), 500 Tage nach dem Brand der Herzogin Anna Amalia Bibliothek, Talk held in Dresden on the 95. German Day of Librarians on March 22nd 2006 [http://www.opus-bayern.de/bib-info/volltexte/2006/260/pdf/Dresden500Tage.pdf last accessed on September 3rd 2009]

Michael Knoche (2006b), Die Bibliothek brennt: ein Bericht aus Weimar, Göttingen 2006.

Michael Knoche (2007a), Die Herzogin Anna Amalia Bibliothek, Zurück im historischen Gebäude, in: Walther Grunwald, Michael Knoche, Hellmut Seemann (ed.), Die Herzogin Anna Amalia Bibliothek, Nach dem Brand in neuem Glanz, Berlin 2007, pp. 13-16.

Michael Knoche (2007b), Die Weimarer Bibliothek als Büchersammlung, Museum und Erinnerungsort, in: Anna Amalia, Carl August und das Ereignis Weimar, (Klassik Stiftung Weimar) Hellmut Th. Seemann (ed.), Göttingen, 2007 pp. 231-243.

Barbara Koch, Mein Weg durch drei Bibliotheken, in: Herzogin Anna Amalia Bibliothek – Kulturgeschichte einer Sammlung, Munich 1999, pp. 221-222.

Michail Kokonoglou and Lorraine Gibson, Analysis of Volatile Organic Compounds in Indoor Air, IAP Copenhagen 2001 [http://www.natmus.dk/cons/reports/2002/iap_meeting/iap2001.pdf last accessed on September 3rd 2009]

Konrad Kratzsch, Kostbarkeiten der Herzogin Anna Amalia Bibliothek, Leipzig 1993.

August Krogh, The Dust Problem in Museums and How to Solve it, in: The Museums Journal, January 1948, Volume 47, Number 10, pp. 183-188.

Heinz Ladendorf, Antikenstudium und Antikenkopie (Abhandlungen der sächsischen Akademie der Wissenschaften zu Leipzig, Philosophisch-historische Klasse 46,2), Berlin 1953, pp. 69-74.

123

Christa von Hees Landwehr, Die antiken Gipsabgüsse aus Baiae: griechische Bronzestatuen in Abgüssen römischer Zeit, (Archäologische Forschungen Vol.14), Berlin 1985.

Holly Lawes, To dust or not to dust, in: ICON news, Nº5, 2006, pp. 37-38.

Holly Lawes, Caroline Bendix, Peter Brimblecombe, Chris Calnan, Three Years of Library Dust, in: The National Trust, Views 41 [www.nationaltrust.org.uk/main/w-conservation-dust-library_dust.pdf last accessed on September 3rd 2009]

Tom Learner: The Analysis of synthetic resins found in twentieth century paint media in: Resins - Ancient and Modern, Preprints of the SSCR 2nd Resins Conference held in Aberdeen, September, 13th-14th 1995, M. M. Wright, J. H. Townsend (ed.), Aberdeen 1995, pp. 76-84.

Henry Lie, Technical Features in Rosso’s Work, in: Medardo Rosso, Second Impressions, New Haven, London 2003, pp. 69-93.

Helen Lloyd, Housekeeping, Preventive Conservation, Course Documents, CV0700BNN, Preventive Conservation, School of Applied Arts and Sciences, Northumbria University, 2008/2009.

Helen Lloyd, Kathy Lithgow, Peter Brimblecombe, Youn Hun Yoon, Kate Frame, Barry Knight, The effects of visitor activity on dust in historic collections, in: The Conservator, Vol. 26, 2002, pp.72-84.

Helen Lloyd et al. (2007a): Helen Lloyd, Caroline Bendix, Peter Brimblecombe, David Thickett, Dust in Historic Libraries, in: Museum Microclimates, Tim Padfield & Karen Borchersen (ed.), National Museum of Denmark, Copenhagen 2007, pp. 135-144. [http://www.frilandsmuseet.dk/graphics/konferencer_mm/microclimates/pdf/lloyd.pdf last accessed on September 3rd 2009]

Helen Lloyd et al. (2007b): Helen Lloyd, Peter Brimblecombe, Katy Lithgow, Economics of Dust, Studies in Conservation 52, 2007, pp. 135-146.

Thomas Lochman, Gipsabguss-Sammlungen zur Goethezeit, in: Das Haus zum Kirschgarten und die Anfänge des Klassizismus in Basel, Historisches Museum Basel 1996, pp. 185-196.

Astrid Lorenzen, Bertrand Porte, Restauration de deux Modèles Monumentaux en Plâtre de François Sicard, Diploma Thesis, École Régionale des Beaux-Arts de Tours, 1989 (unpublished manuscript).

Bernhard Maaz (1993), Klaus Stemmer, Bernhard Maaz, Berliner Gypse des 19. Jahrhunderts, Exhibition catalogue Abgusssammlung antiker Plastik Berlin, 1993.

Bernhard Maaz, Die Crux der Subjektivität, Die Bildnisbüste im Zeitalter ihrer Popularisierung, in: Exhibition catalogue ’Hauptsache Köpfe: plastische Porträts von der Renaissance bis zur Gegenwart aus der Skulpturensammlung‘, Staatliche Kunstsammlungen Dresden 2001, pp. 26-32.

Ursula Merkel, Aufbruchsjahre der Moderne, Aspekte der Porträtplastik vom Historismus bis zum Expressionismus, in: Exhibition catalogue ’Hauptsache Köpfe: plastische Porträts von der Renaissance bis zur Gegenwart aus der Skulpturensammlung‘, Staatliche Kunstsammlungen Dresden 2001, pp. 33-39.

124

Lidia Morawska, Motor Vehicle Emissions as a Source of Indoor Particles on shapes and sizes of combustion and abrasion particles in road dust, in: Indoor Environment, Airborne Particles and Settled Dust, Lidia Morawska, Tunga Salthammer (ed.), Weinheim 2003, pp. 297-318.

Lidia Morawska, Tunga Salthammer (2003a), Fundamentals of Indoor Particles and Settled Dust, in: Indoor Environment, Airborne Particles and Settled Dust, Lidia Morawska, Tunga Salthammer (ed.), Weinheim 2003, pp. 3-46.

Lidia Morawska, Tunga Salthammer (2003b), Introduction to Sampling and Measurements Techniques, in: Indoor Environment, Airborne Particles and Settled Dust, Lidia Morawska, Tunga Salthammer (ed.), Weinheim 2003, pp. 49-55.

Elisabet Tebellius-Murén, Gips som form och gjutmassa, in: Gips –Tradition I Konstens Form, Stockholm, 1999.

The National Trust manual of Housekeeping, the care of collections in historic house open to the public, The National Trust (ed.), Amsterdam 2006.

William W. Nazaroff, Mary P. Ligocki, Lynn G. Salmon, Glen R. Cass, Theresa Fall, Michael C. Jones, Harvey I. H. Liu, Timothy Ma, Airborne Particles in Museums, Los Angeles 1993. [http://www.getty.edu/conservation/publications/pdf_publications/airborne.pdf last accessed on September 3rd 2009]

Martina Noehles, Die Kunst des Radierens - Radiermittel im Überblick, in: PapierRestaurierung Vol. 3 (2002) - No.1, pp. 22-28, http://206.180.235.135/iada/pr01jb_g.pdf [last accessed August 21st, 2009]

Peter D. Osborn, The Engineer’s Clean Air Handbook, Butterworths, London 1989.

Gabriele Oswald (1995), Die Plastiksammlung der Herzogin Anna Amalia Bibliothek, 2 Vols., Diploma Thesis, Institute for Art History at the Martin-Luther-University in Halle-Wittenberg, 1995.

Gabriele Oswald (1999), Die Porträtplastik, in: Herzogin Anna Amalia Bibliothek – Kulturgeschichte einer Sammlung, Munich 1999, pp. 98-100.

Gabriel Oswald (2003), “Wir haben uns neulich mit Deiner Büste unterhalten”, Sehen und gesehen werden – die Porträtplastik, in: Antlitz des Schönen, Klassizistische Bildhauerkunst im Umkreis Goethes, Rudolstadt: Thüringer Landesmuseum Heidecksburg, 2003, pp. 91-115.

Gabriele Oswald (2007a), Die Anfänge klassizistischer Bildhauerkunst in Deutschland am Beispiel Weimar, in: Anna Amalia, Carl August und das Ereignis Weimar, Hellmut Th. Seemann (ed.), Göttingen, 2007, pp. 282-291.

Gabriele Oswald (2007b), Goethes Plastiksammlung im Spiegel seines Kunstverständnisses, PhD Dissertation-Thesis, Martin-Luther-Universität Halle-Wittenberg, 2007 (unpublished manuscript).

Laura Padgett, Der Rokokosaal, in: Herzogin Anna Amalia Bibliothek – Kulturgeschichte einer Sammlung, Munich 1999, pp. 11-26.

Nicholas Penny, The Materials of Sculpture, New Haven, London 1993, pp. 194-199.

125

Carin Pettersson, Plaster of Paris, Cleaning of Plaster Casts, Graduating Thesis, BA/Sc, Program in Conservation of Cultural Property, Gothenburg University, Institute of Conservation, Gothenburg 2002 (unpublished manuscript).

Edward Phelps, Carving in Plaster of Paris, Leicester 1963.

Hanna Philipp, Winckelmann und das Weiß des Rokoko, in: Antike Kunst, Vol. 39, 1996, pp. 88-99.

Christiane Pinatel, Les envois de moulages d’antiques à l’École des Beaux-Arts de Paris par l’Académie de France à Rome, in: Les Moulages de Sculptures antique et l’Histoire de l’Archéologie, Actes du Colloque International, 24e octobre 1997, Geneva 2000, pp. 75-120.

Violine Pillard, La corrosion des armatures de fer dans le plâtre: l´exemple d´un moulage du sculpteur Geoffroy-Dechaume, in: Le Plâtre, l’art et la matière, Paris 2001, pp. 279-291.

Gertrud Platz-Horster, Zur Geschichte der Berliner Gipssammlung, in: Berlin und die Antike, Exhibition catalogue, Berlin 1979, pp. 273-292.

Hans Portsteffen, Oberflächenreinigung – eine Einführung, in: Oberflächenreinigung – Material und Methoden, C. Weyer (ed.), Stuttgart 2006, pp. 27-30.

Alex Potts, Neoclassicism and the Ambiguity of Sculptural Polychromy, in: The Color of Life, Polychromy in Sculpture from Antiquity to the Present, Roberta Panzinelli et al. (ed.), J. Paul Getty Museum and the Getty Research Institute, Los Angeles 2008, pp. 84-88.

Giuseppe Pucci, Les Moulages de Sculpture Ancienne et l’Esthétique du XVIIIe Siècle, in: Les Moulages de Sculptures antique et l’Histoire de l’Archéologie, Actes du Colloque International, 24e octobre 1997, Geneva 2000, pp. 45-55.

Bärbel Raschke, Die Bibliothek der Herzogin Anna Amalia, in: Herzogin Anna Amalia Bibliothek – Kulturgeschichte einer Sammlung, Munich 1999, pp. 83-86.

Søren Rasmussen, Birgitte Lydik Clausen, Rensning af gips og andre materialer med laserlys, Danske Museer 41, 2000, Nº2, pp. 41-44.

Friedrich Rathgen, Über die Auffrischung verschmutzter Gipsabgüsse, in: Museumskunde, Vol. 8, 1/2, 1917, pp.33-40.

Petra Rau, “Unter diesen Göttern zu wandeln”, Kunsthandel, Kunstjournale und Kunstmanufakturen im 18. Jahrhundert, in: Antlitz des Schönen, Klassizistische Bildhauerkunst im Umkreis Goethes, Rudolstadt: Thüringer Landesmuseum Heidecksburg, 2003, pp. 59-89.

Paul Ortwin Rave, Wilhelm von Humboldt und das Schloss zu Tegel, Leipzig 1950.

Gerhard Rupp, Gips, Zink und Bronze – Berliner Vervielfältigungsfirmen im 19. Jahrhundert, in: Ethos und Pathos, Die Berliner Bildhauerschule 1786-1914, Beiträge, Peter Bloch, Sibylle Einholz, Jutta von Simson (ed.), Berlin 1999, pp. 337-351.

Jorun Rebekka Ruppel, Waschen, Peelen, Lasern – zum Problem der Reinigung von Gipsabgüssen, in: Oberflächenreinigung – Material und Methoden, C. Weyer (ed.), Stuttgart 2006, pp. 176-182.

126

Morten Ryhl-Svendsen, The Viking Ship Museum in Roskilde. An Investigation of Dust in the Exhibition, IAQ 2003: 5th meeting of the Indoor Air Pollution Working Group, Norwich. [http://iaq.dk/iap/iaq2003/2003_18.htm last accessed on September 3rd 2009]

Anne Lisbeth Schmidt, Airborne Dust in a Museum Environment, IAP Copenhagen, 4th Meeting of the Indoor Air Pollution Working Group, 2001. [http://iaq.dk/iap/iap2001/2001_20.htm last accessed on August 12th, 2009]

Anne Lisbeth Schmidt, Pernille Bronée, Kåre Kemp, Jes Fenger, Airborne Dust in a Museum Environment, IAP Copenhagen, 4th Meeting of the Indoor Air Pollution Working Group, 2001, presentation 20. [http://iaq.dk/iap/iap2001/2001_20.htm last accessed August 13th, 2009]

Thomas Schneider, Sampling of Surface Dust in Buildings, in: Indoor Environment, Airborne Particles and Settled Dust, Lidia Morawska, Tunga Salthammer (ed.), Weinheim 2003, pp. 82-104.

Annette Schulz, Gipsplastiken, Geschichte – Materialkunde – Reinigung von ungefassten Oberflächen, Diploma Thesis, Institut für Technologie der Malerei an der Staatlichen Akademie der Bildenden Künste in Stuttgart, 1992 (unpublished manuscript).

Science for Conservators, Vol. 2: Cleaning (Conservation Science Teaching Series), The Conservation Unit of the Museums and Galleries Commission, London, New York, 4th edition 1996.

Annette Seeman, Die Geschichte der Herzogin Anna Amalia Bibliothek, Frankfurt/Main, Leipzig 2007.

Hellmut Seeman, Nach dem Brand in neuem Glanz, in: Walther Grunwald, Michael Knoche, Hellmut Seemann (ed.), Die Herzogin Anna Amalia Bibliothek, Nach dem Brand in neuem Glanz, Berlin 2007, pp. 10-12.

Ulrike Steierwald, Zentrum des Weimarer Musenhofes, Die Herzogliche Bibliothek 1758-1832, in: Herzogin Anna Amalia Bibliothek – Kulturgeschichte einer Sammlung, Munich 1999, pp. 62-103.

Klaus Stemmer, Bernhard Maaz, Dietmar Vogel, Berliner Gypse des 19. Jahrhunderts: Von der Idee zum Abguss, Exhibition catalogue Abguss-Sammlung antiker Plastik, Berlin 1993.

Klaus Stemmer, “Seid heiter und haltet mir die Gipse rein…” Vorbemerkungen zur Kulturgeschichte des Gipsabgusses im 19. Jahrhundert, in: Das Spiel mit der Antike: zwischen Antikensehnsucht und Alltagsrealität; Festschrift zum 85. Geburtstag von Rupprecht Düll. Sigrid Düll, Otto Neumaier, Gerhard Zecha (ed.), Arianna Vol. I, Möhnesee 2000, pp. 227-232.

Robert Stiassny, Vom Gipsmuseum der Wiener Kunstakademie, in: Museumskunde, Vol. 6, 1910, p. 1-17.

Veit Stürmer, Henning Wrede, Ein Museum im Wartestand. Die Abgußsammlung antiker Bildwerke, in: Winckelmann-Institut der Humboldt Universität zu Berlin, Vol. 3, Berlin 1998.

Geoffrey N. Swinney, The evil of vitiating and heating the air. Artificial lighting and public access to the National Gallery, London, with particular reference to the Turner and Vernon collections, in: Journal of the History of Collections 2003, Vol. 15, 1, pp. 83-112

Eric Tanguy, Nathalie Huet, Armand Vinçotte, Laser Cleaning of Patrimonial Plasters, in: Lasers in the Conservation of Artworks LACONA V Proceedings, Osnabrück, Germany, (Sept. 15th-18th 2003), Berlin, Heidelberg 2005.

127

David Thickett, Chloride, Dust and RH, Risks in Coastal and Inland Areas, Presentation Abstracts and Additional Notes from the 6th meeting of the Indoor Air Pollution Working Group, Padova, September 2004. [http://www.isac.cnr.it/iaq2004/abstract/thickett.pdf last accessed on August 12th, 2009]

David Thickett, Soek-Joo Rhee, Sarah Lambarth, Libraries and archives in historic buildings, in: Museum Microclimates, Tim Padfield & K. Borchersen (ed.) National Museum of Denmark 2007, pp. 145-155 [http://www.frilandsmuseet.dk/graphics/konferencer_mm/microclimates/pdf/ thickett.pdf last accessed on September 3rd 2009]

Garry Thomson, Air Pollution: A Review for Conservation Chemists, in: Studies in Conservation, Vol. 10, Nº4 (Nov. 1965), pp. 147-167.

Garry Thomson, The Museum Environment, London 1986 (2nd edition).

Giorgio Torraca, Porous Building materials, Material Science for Architectural Conservation, Rome 1988 (3rd edition).

Karina Türr, Farbe und Naturalismus in der Skulptur des 19. Jahrhunderts, Mainz 1994.

Eduard Uhlenhuth, Vollständige Anleitung zum Formen und Gießen oder genaue Beschreibung aller in den Künsten und Gewerben dafür angewandten Materialien als: Gyps, Wachs, Schwefel, Leim, Wien, Pest, Leipzig 1892 (3rd edition).

Gert-Dieter Ulferts, Bibliothek als Museum, Reminiszenzen des vorklassischen Weimar, in: Walther Grunwald, Michael Knoche, Hellmut Seemann (ed.), Die Herzogin Anna Amalia Bibliothek, Nach dem Brand in neuem Glanz, Berlin 2007, pp. 157-162.

Rembert Unterstell, The Fate of Weimar’s Renowned Anna Amalia, in: German Research, Volume 27, 2005, Issue 2, pp. 6-9 [http://www.dfg.de/en/german_research/download/german_research_2_05.pdf last accessed on September 3rd 2009]

Bettina Uppenkamp, Potentiale der Bescheidenheit. Über kunstvolle und kunstlose Möglichkeiten in Gips, in: Material in Kunst und Alltag, Monika Wagner, Dietmar Rübel (ed.), Hamburger Forschungen zur Kunstgeschichte I, Berlin 2002, pp. 137-161.

Karlheinz Volkart, Gypsum and Plaster Dictionary, English-French-German, Wiesbaden, Berlin, 1971.

Victor H. Wager, Plaster Casting for the Student Sculptor, London 1952.

Helgard Weber, Die Gipsabgusssammlung des Bode-Museums – Geschichte, Bedeutung und Probleme, in: Proceedings to the Berlin colloquium, November 19th-20th, 2007, ’Das Bode-Museum auf dem Weg zur Wiedereröffnung, Projekte und Restaurierungen’ (in print).

Martin Weber, Die Kunst des Bildformers und Gypsgießers: oder gründlicher Unterricht wie Büsten, Statuen, Vasen, Urnen, Ampeln, Konsolen oder Kragsteine, Rosetten, Laub- und Simswerk, Reliefbilder und andere dergleichen plastische Gegenstände auf dem Wege des Abformens und Abgießens nachzubilden, wie sie zu schleifen, zu polieren, zu firnissen, zu bronzieren und zu restaurieren sind. Für Künstler und Techniker, vornehmlich Architekten, Stukkateure, Bildhauer und Gipsgießer vom Fach, Weimar, 1861.

128

Bettina Werche (2001), Überlegungen zu Wiedereinrichtung des Rokokosaales mit Kunstobjekten, in: Kolloquium im Rückblick 2001 [http://www.klassik-stiftung.de/fileadmin/downloads/haab/werche_wiedereinrichtung_rokokosaal.pdf, last accessed on September 3rd 2009]

Bettina Werche (2007a), Rokokosaal, Die Kunstausstattung des zentralen Bibliotheksraumes, in: Die Herzogin Anna Amalia Bibliothek, Nach dem Brand in neuem Glanz, Berlin 2007, pp. 13-16.

Bettina Werche (2007b), Käuters Skizze des Rokokosaales der Großherzoglichen Bibliothek, in: Anna Amalia, Carl August und das Ereignis Weimar, (Klassik Stiftung Weimar) Hellmut Th. Seemann (ed.), Göttingen, 2007, pp. 244-271.

Johann Joachim Winckelmann, Geschichte der Kunst des Altertums, Dresden 1764, in: Studien zu Deutschen Kunstgeschichte Vol. 343, Baden-Baden 1966.

Sally Woodcock, Review of the papers presented at UKIC’s and the Tate Gallery’s joint conference proceedings, January 1990, Dirt and pictures separated, in: Studies in Conservation, Vol. 36, No. 2 (May, 1991), pp. 124-125.

Young Hun Yoon, Peter Brimblecombe (2000a), Clothing as a Source of Fibres within Museums, Journal of Cultural Heritage 1, 2000, pp. 445-454.

Young Hun Yoon, Peter Brimblecombe (2000b), Dust at Felbrigg Hall, The National Trust, Views Issue 32, 2000, pp. 31-32 [http://www.nationaltrust.org.uk/main/w-collections-dust-felbrigg.pdf last accessed on September 3rd 2009]

Online Resources

[last accessed on September 3rd 2009]

AATA Online, database of the Getty Conservation Institute, Los Angeles, in association with the International Institute for Conservation of Historic and Artistic Work, London, http://aata.getty.edu/nps/

Australian Museums & Galleries Online, Haymarket NSW: http://archive.amol.org.au/recollections/3/pdf/dust.pdf

The Conservation and Art Materials Encyclopedia Online (CAMEO), Museum of Fine Arts, Boston, MA: http://cameo.mfa.org/materials/record.asp?key=2170&subkey=10272&Search=Search&MaterialName=zapon&submit.x=0&submit.y=0

Conservation Online, Preservation Department of Stanford University Libraries, Palo Alto, CA: http://206.180.235.135/iada/pr01jb_g.pdf

Deutsche Welle (Radio), Bonn: http://www.dw-world.de/dw/article/0,,1317151,00.html

The Free Dictionary, Farlex, Huntingdon Valley, PA: http://www.thefreedictionary.com/

Herzogin Anna Amalia Bibliothek, Weimar: http://www.anna-amalia-library.com/en/

Herzog August Bibliothek, Wolfenbüttel: http://www.hab.de/museum/besuch/index-e.htm

129

ICCROM, International Centre for the Study of the Preservation and Restoration of Cultural Property, Rome, http://www.iccrom.org/eng/02info_en/02_01library_en.shtml

The Indoor Air Pollution Workgroup, Conservation Department of the National Museum of Denmark, Copenhagen: http://www.iaq.dk/

ImageJ, Image processing and analysis in Java: http://rsbweb.nih.gov/ij/index.html

International Association for the Conservation and the Promotion of Plaster Cast Collections (A.I.C.P.M.): http://www.plastercastcollection.org/en/database.php

INP – Institut nationale du patrimoine: http://exlibris.inp.fr

About.com: Inventors: New York Times Company, New York City: http://inventors.about.com/library/inventors/blvacuum.htm

Journal of the American Institute for Conservation, American Institute for Conservation of Historic and Artistic Works (AIC), Washington, D.C.: http://206.180.235.133/jaic/articles/jaic26-02-004_4.html

Klassik Stiftung Weimar, Weimar: http://www.klassik-stiftung.de/

Lafarge Prestia, Cholburi, Thailand: http://www.lafargeprestia.com/caso4___h2o.html

Monumente Magazin, Deutsche Stiftung Denkmalschutz, Bonn: http://www.monumente-online.de/07/05/streiflichter/08_Humboldtschloesschen.php

Museums and Galleries Commission (MGC), Westminster: http://www.collectionslink.org.uk/assets/userfiles/index.php?file=000072.pdf

Museum Wax, Richmond: http://www.museumwax.co.uk/

The National Trust, Swindon: http://www.nationaltrust.org.uk/main/w-index.htm

Tim Padfield, Conservation Physics, Copenhagen: http://www.padfield.org/tim/cfys/index.php

The Physics Factbook™, Glenn Elert (ed.): http://hypertextbook.com/

Testo: Alton, Hampshire: http://www.testo.co.uk/

Victoria and Albert Museum, South Kensington: http://www.vam.ac.uk/

Villeroy & Boch AG, Mettlach: http://www.villeroy-boch.com/en/us/home/the-company/the-company/history.html

WebPhysics, Department of Physics, Davidson College, Davidson, NC http://webphysics.davidson.edu/

Westdeutscher Rundfunk (TV), Köln: http://www.wdr.de/tv/quarks/sendungsbeitraege/2004/0413/006_unsichtbar.jsp

Wikipedia Encyclopedia, San Francisco, CA: http://en.wikipedia.org/wiki/Main_Page

130

List of Illustrations

Figure 1: Adolph Menzel’s pastel drawing ‘Plaster Model Storeroom in the Altes Museum’ of 1848 depicts the storage situation of the cast collection during the construction of the Neues Museum. ........................................................................................................................................................9

Figure 2: Francesco Carradori ‘Making a Plaster Mould’, etching, 1802. ................................................ 15 Figure 3: A small-piece mould. ................................................................................................................................. 17 Figure 4: ‘The Plaster Kiln’ by Jean Louis Théodore Géricault, 1822-1823, oil on canvas. ............. 18 Figure 5: These dust particles were collected off pedestal surfaces in the Rococo Hall (in July

2009) and photographed at 25-fold microscopic magnification after separating them from the larger amount of fibrous material present in the sample. 1,342282 µm equals 1 pixel in this image’s resolution, hence particle sizes of under 1,3 µm are not visualized. ................................................................................................................................................... 22

Figure 6: These fibres were separated from the sampled dust, which was collected off pedestal surfaces in the Rococo Hall (in July 2009, cf. Fig. 5) and photographed at 6-fold microscopic magnification. ................................................................................................................. 24

Figure 7: This cast plaster bust of Julius Caesar shows heavy soiling and the so-called ‘reverse shadow effect’. .......................................................................................................................................... 26

Figure 8: The rate of soiling on plaster surfaces is more evident in the presence of areas of loss.28 Figure 9: Poultice cleaning plaster surface. Surface soiling is separated from the plaster surface on

removal of the poultice. ........................................................................................................................ 30 Figure 10: September 2004, the plaster statuary was salvaged from the burning Rococo Hall into

the under-ground storage area.......................................................................................................... 33 Figure 11: View of the Ducal Library from North-West (the building behind the trees, seen from

direction of the Palace), Aquarelle painting by Georg Melchior Kraus, ca. 1800. .......... 34 Figure 12: View into the interior of the Rococo Hall in 1978 with Carl-August’s oil on canvas

portrait placed in the central position, crowned by a plaster cast of a monumental bust of Johann-Wolfgang von Goethe on the ballustrade of the first gallery.............................. 37

Figure 13: View into the interior oval of the lower level of the Rococo Hall from the opposite angle as in Fig. 12, in 2009. ............................................................................................................................ 40

Figure 14: The first gallery of the Rococo Hall, 2008. .................................................................................... 43 Figure 15: The bust of Johann Wolfgang von Goethe was severely damaged at the base. Yet the

over-paints had to a large extent protected the plaster surfaces from the direct impact of fire-extinguishing water. ................................................................................................................. 48

Figure 16: During stripping, the reconstruction of the rear of the head of Herder’s bust showed underneath three coatings of paint on top of a washed down surface .............................. 49

Figure 17: Raynal’s bust was seriously damaged in the night of the fire. .............................................. 51 Figure 18: Microscopic inspection of the cross-section of Raynal´s coatings reveals two layers of

paint, between which there is soiling. ............................................................................................ 52 Figure 19: UV-animation of the same cross-section, see Figure 18. ........................................................ 53 Figure 20: 200-fold magnification of the cross-section into Goethe’s surface (cf. Fig. 15). ........... 54 Figure 21: UV-investigation highlights the soiling. .......................................................................................... 54 Figure 22: The bust of Torquato Tasso prior to restoration. ...................................................................... 55 Figure 23: A cross-section of Tasso’s surface with magnification factor 50. ........................................ 56 Figure 24: Magnification factor 200 of the same cross-section shows dirt deposition on the

plaster surface. ......................................................................................................................................... 56 Figure 25: Rauch’s bust’s surface coating was heavily damaged by fire- extinguishing water.. ... 57

131

Figure 26: A cross-section of Rauch’s surface-coatings in 200-fold magnification. .......................... 58 Figure 27: Schiller’s bust prior to restoration shows water damage and soiling. .............................. 58 Figure 28: A cross-section into the bust of Schiller’s surface reveals one coating of paint. ........... 59 Figure 29: UV-light highlights soiling on the plaster and also on top of the coating of paint. ........ 59 Figure 30: Plaster casts of Christian Daniel Rauch’s bust of Johann Wolfgang von Goethe in

storage. ........................................................................................................................................................ 64 Figure 31: The University hall in which 2500 casts were sprayed with Zapon-lack in 1916-1921.

................................................................................................................................................................ ........ 69 Figures 32 and 33: Floor plan of the two levels of the Rococo Hall, showing the placements of the

six loggers. .................................................................................................................................................. 85 Figure 34: The plaster bust of Johann Wolfgang von Goethe by Christian Daniel Rauch is placed

on the first level of the Rococo Hall. Behind it stands the colossal marble bust of Goethe by David D´Angers, the pedestal of which is large enough to block visitor’s access to this area. ..................................................................................................................................................... 88

Figure 35: 25 visitors are allowed into the lower level of the Rococo Hall simultaneously in the company of wardens. ............................................................................................................................. 89

Figures 36 and 37: Floor plan of the lower level, respectively of the 1st Gallery (level two of the Hall), indicating the positions of the plaster busts as red circles. ....................................... 90

Figures 38 and 39: Floor plan of the lower level, respectively of the 1st Gallery, indicating the presumed dust distribution in the two levels of the hall in relation to human activity.91

Figure 40: The crown of Goethe´s plaster bust, cf. above Fig. 35, cf. No. 1 in floor plan, Fig. 41………………………………………………………………………………………………………………….…..92

Figures 41 and 42: Floor plan of the lower level, respectively of the 1st Gallery, indicating the dust distribution in the two levels of the hall, as perceived with naked eye inspection. ....... 93

Figure 43: Particle matter on the crown of Musäus’ bust’s head, (cf. No. 2 in floor plan, Fig. 42). ................................................................................................................................................................ 93

Figure 44: Dust deposition is very evident at the rear of the pedestal surface and on the rear of Musäus' plinth. ......................................................................................................................................... 94

Figure 45: Dust accumulation on the bust of Raynal, which is situated in the interior oval of the lower level of the Rococo Hall (cf. No. 3 in floor plan, Figure 42). ...................................... 94

Figure 46: Carl August Böttiger’s bust is situated on the first Gallery (cf. Figure 43, no. 4 in floor plan) its surfaces show no visible dust accumulation. ............................................................. 95

Figure 47: Ludwig Tieck’s bust is positioned on the gallery balustrade on the second level of the library hall, No. 5 in floor plan Figure 43........................................................................................ 96

Figure 48: Visualising dust on the left shoulder of the bust of Ludwig Tieck, Fig. 47. ...................... 97 Figure 49: Removing particles from a pedestal surface with sticky tape eases visual assessment of

degree of particle depositioning within the bust’s vicinity...................................................... 97 Figure 50: Positioning the glass trays for dust sampling............................................................................... 98 Figure 51: Comparing deposition rates by visual analysis of glass tray sample collectors.. ............ 99 Figure 52: Comparing sampling with Gel-lifter and with glass trays. ...................................................... 99 Figure 53: Microscopic investigation for the assessment of dust distribution. ................................ 100 Figure 54: Sampling location 1.3, image a. Image size 2584 x 1936 pixel, in mm 15.14 x 11.35.

Total particle count 1994. Average particle size 52.618 pixel. Pixel size of image 5.86592179 µm. Average particle size ca. 294 µm (0.294 mm). Note the high amount of scratches, identifiable as thin, almost vertical lines......................................................... 101

Figure 55: Detail of image of particle counting, generated with ImageJ. ............................................. 101

132

Figures 56 and 57: Distribution of 12 glass trays for dust sampling on the two levels of the Rococo Hall The entrance and exit situations are situated at the bottom of the floor-plan. .. 103

Figure 58: Dust sampled by dusting with a soft paintbrush, cf. Figures 5 and 6 for a 25-fold magnification.. ....................................................................................................................................... 108

Figure 59: Dust on Dantes’ shoulder. .................................................................................................................. 111 Figure 60: Dust removed off Dantes’ shoulder. .............................................................................................. 111 Figure 61: Vacuum cleaner and ladder. .............................................................................................................. 112

Picture Credits

Open source: Figures 1, 2, 4, 11, 12, 31.

Lorenza D’ Alessandro, Francesca Persegati: Figure 3.

Frank Mucha: Figures 5, 6, 18, 19, 20, 21, 23, 24, 26, 28, 29, 54.

Aurelia Badde: Title image and figures 7, 8, 9, 10, 13, 14, 16, 17, 30, 32, 33, 34, 35, 36, 37, 38, 39, 40,

41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 55, 58, 59, 60, 61.

Restaurierung am Oberbaum: Figures 15, 22, 27.

Rütt und Schulz: Figure 25.

List of Appendices

Appendix I: Über die Behandlung von Gipsabgüssen behufs deren Erhaltung (Excerpt)

Appendix II: Lab. report FH Erfurt

Appendix III: Lab. report FH Hildesheim

133

Appendix I, Über die Behandlung von Gipsabgüssen behufs deren Erhaltung

134

135

136

137

138

139

Appendix II, Lab. Report Erfurt

An Frau Aurelia Badde Friedrichstr. 12 10969 Berlin

Frank Mucha Freiberuflicher Chemiker

Naturwissenschaftliches Labor Fachbereich Konservierung und Restaurierung Fachhochschule Erfurt Altonaer Str. 25 99085 Erfurt Tel. 0361/6700 788 Fax. 0361/6700 766 E-Mail: f.mucha@fh-erfurt.de Web: http://res1.fh-erfurt.de/labor/labor.htm

Auftrag vom 23.06.2009 Mein Zeichen: Erfurt, den 11/07/2011 Objekt: Anna Amalia Bibliothek Analysenergebnisse Sehr geehrte Frau Badde, gemäß Ihrem Auftrag vom 23.06.2009 wurden von mir naturwissenschaftliche Untersuchungen an Staub- und Fassungsproben durchgeführt. Anbei sende ich Ihnen die Untersuchungsergebnisse und die Kostenaufstellung. Mit freundlichen Grüßen Frank Mucha Anlage: Untersuchungsergebnisse, Kostenaufstellung

140

Staubprobe 1, untere Etage Röntgenfluoreszenzanalyse (RFA) Mehrere Teilproben des nichtfasrigen Staubanteils der wurde mit einem Spotdurchmesser von 0,3mm angeregt. Das Rhodiumsignal ist messtechnisch bedingt (Röntgenröhre) und ohne Relevanz. In der Probe sind Verbindungen folgender Elemente enthalten: 1a Elemente: Si, S, Ca, Fe

Si S

Rh Ca

Fe

141

1b Elemente: Si, S, Ca

Si

S Rh

Ca

142

1c Elemente: Si, S, Ca, Ti, Fe, Zn

Si

S

Rh Ca

Ti Fe

Zn

Zn

143

1d Elemente: Si, S, Ca, Fe

Si

S

Rh

Ca

Fe

144

1e Elemente: Si, Ca

Si

Rh Ca

145

1f Elemente: Si, S, Ca, Fe

Si S

Rh

Ca Fe

146

Fourier-Transform Infrarot-Spektroskopie (FT-IR-Spektroskopie): Erläuterung zur IR-Spektroskopie: Die Peaks, in der IR „Banden“ genannt, entsprechen Schwingungen der miteinander verbundenen Atome. Da diese unterschiedlich schwer sind und unterschiedlich stark miteinander verbunden sind, ergeben verschiedene Substanzen auch verschiedene und damit charakteristische Schwingungsbanden, die zur Identifizierung genutzt werden. Bei Gemischen addieren sich die Einzelspektren einfach. Die als Referenz genutzten Spektren wurden zum Vergleich mit in die Diagramme gelegt. Ergebnis: Es sind als nicht fasrige Hauptbestandteile Gips, Kalk, Siliciumdioxid (Quarzstaub, Sandstaub) und proteinhaltige sowie weitere organische Substanzen enthalten.

4400 ,0 4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 450 ,0cm -1

% T

G ip s

C a lc ium carb ona t

S ilic ium d iox id(Q uarzs taub )

P ro te in

S taub p rob e 1n ich tfas r ig e B es tand te ile

147

Staubprobe 2, Erste Galerie Röntgenfluoreszenzanalyse (RFA) Mehrere Teilproben des nichtfasrigen Staubanteils der wurde mit einem Spotdurchmesser von 0,3mm angeregt. Das Rhodiumsignal ist messtechnisch bedingt (Röntgenröhre) und ohne Relevanz. In der Probe sind Verbindungen folgender Elemente enthalten: 2a Elemente: Si, Ca

Si

Rh

Ca

148

2b Elemente: Ca

Rh

Ca

149

2c Elemente: Si, S, Ca

Si S

Rh

Ca

150

2d Elemente: Si, S, Ca

Si

S

Rh

Ca

151

2e Elemente: Si, S, Ca, Fe

Si

S

Rh

Ca Fe

152

2f Elemente: Si, S, Ca, Fe

Si S

Rh Ca

Fe

153

Fourier-Transform Infrarot-Spektroskopie (FT-IR-Spektroskopie): Ergebnis: Die Zusammensetzung ähnelt der Staubprobe 1: Es sind als nicht fasrige Hauptbestandteile Gips, Kalk, Siliciumdioxid (Quarzstaub, Sandstaub) und proteinhaltige sowie weitere organische Substanzen enthalten. Der Anteil nicht fasriger Bestandteile ist in der Probe 2 dem rein optischen Eindruck nach geringer.

4400 ,0 4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 450 ,0cm -1

% T

S taub p rob e 2n ich tfas r ig e B es tand te ile

G ip s

S ilic ium d iox id(Q uarzs taub )

C a lc ium carb ona t

P ro te in

154

Organische Analyse nicht fasriger Hauptbestandteile: Staubprobe 1, untere Etage und Staubprobe 2, Erste Galerie THM- (Thermally assisted hydrolysis and methylation) Pyrolyse-Gaschromatographie/Massenspektrometrie (THM-Py-GC/MS) Ergebnis: Es war ein erheblicher Anteil nicht abtrennbarer Feinstfasern in den beiden Proben der nicht fasrigen Bestandteile enthalten. Beide Proben stimmen in Ihrer Zusammensetzung qualitativ überein. Bestandteile: aus eingetragenen Partikeln und Fasern: Öle, tier. Öle / Fette (Hautpartikeln), Cellulose (Fasern), diterpenodide(s) Harz(e), Kunstharze (Acrylharz, Isophthalsäurealkydharz, Polystyrol / Polystyrol-Copolymere, Phenolharze), kollagene (tier.) Proteine (Hautpartikeln) und aus abgetragenen Fassungspartikeln: trocknendes Öl, Orthophtalsäure-Alkydharz Die wichtigsten Pyrolyseprodukte wurden (massenspektrometrisch und durch Vergleich der Retentionszeiten identifiziert) im Pyrogramm indiziert. Diese diagnostischen Marker sind in der Tabelle detailliert aufgeschlüsselt.

155

Tabelle der indizierten Markerverbindungen Peak- Indiz

Name der Markerverbindung Pyrogramm

Name nichtmethylierten Verbindung

Herkunft aus Malschicht-Komponente:

1 Methylmethacrylat - Acrylharz 2 N-Methylpyrrol - Kollagene Proteine 3 1,4-Dimethyl-1H-pyrazole - Proteine 4 Styrol - Polystyrol / Styrol-

copolymere 5 Glycerintrimethylether Glycerin Öle / Alkydharze 6 Anisol Phenol u.a. Phenolharze 7 p-Methylanisole p-Cresol u.a. Phenolharze 8 Pentaerythritoltrimethylethe

r Pentaerythritol Alkydharz

9 5,6-Dimethyluracil - Kollagene Proteine 10 2,4-Dimethoxyphenol - Kohlenhydrate 11 1,3,5-Trimethoxybenzene - Kohlenhydrate 12 1,3-Dimethyltetrahydro-

2(1H)-pyrimidinone - Proteine

13 L-Proline, 1-methyl-5-oxo-, methyl ester

- Kollagene Proteine

14 1,2,4-Trimethoxybenzene - Kohlenhydrate 15 Tri-O-Methyl-3-deoxy-2-

methoxymethyl-D-erythro-pentonsäure

1,4-glycosidisch verknüpfte Polysaccharide

Kohlenhydrate Stärke / Cellulose

16 Suberinsäuredimethylester (Octandisäuredimethylester)

Suberinsäure Trocknende Öle

17 Dimethylorthophthalat Orthophthalsäure Alkydharze, Weichmacher

18 Dimethylisophthalat Isophthalsäure Alkydharze 19 Azelainsäuredimethylester

(Nonandisäuredimethylester)

Azelainsäure

20 Myristinsäuremethylester Myristinsäure Palm- / Cocosöl, tier. Fette + Öle

21 Palmitinsäuremethylester Palmitinsäure Öle 22 Ölsäuremethylester Ölsäure Öle 23 Stearinsäuremethylester Stearinsäure Öle 24 Methyldehydroabietat Dehydroabietinsäure diterpenoide Harze 25 Cholesterin methyl ether Cholesterin Tier. Lipide

156

Pyrogramme der Proben mit den indizierten diagnostischen Markern

Chromatogram PlotsPlot 1: ...ateien\gcms\thm\a proben\baddestaub1a thm.sms RIC MergedPlot 2: ...ateien\gcms\thm\a proben\baddestaub2a thm.sms RIC Merged

10 20 30 40 50 minutes

0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75MCounts

0

1

2

3

MCounts

RIC Merged baddestaub1a thm.sms 2000 CENTROID RAW

RIC Merged baddestaub2a thm.sms 2000 CENTROID RAW

2

1

1

2

3

3

4

4

5

5

6

6

7

7

8

8

9

9

10

10

11

11

12

1213

1314

14

15

15

16

16

17

17

18

18

19

19

20

20

21

21

22

22

23

23

24

24

25

25

Staubprobe 1untere Etage

Staubprobe 2Erste Galerie

157

Bindemittelanalysen der Fasssungsproben Probe1, Raynal Schicht 1 THM- (Thermally assisted hydrolysis and methylation) Pyrolyse-Gaschromatographie/Massenspektrometrie (THM-Py-GC/MS) Ergebnis: Es ist ein Gemisch von Collagen (Glutinleim) mit Ölen / tierischem Öl bzw. Fett als Bindemittel enthalten. Es sind nur geringe Anteile an Suberin- und Azelainsäure enthalten, was auf langsam bzw. nicht trocknende Öle hinweist. Das Fettsäureprofil keinem der üblichen Öle / Fette. Der Art des Öls / Fettes (Gemisches?) kann nicht identifiziert werden. Insgesamt aber liegt eine fette Tempera vor. Die wichtigsten Pyrolyseprodukte wurden (massenspektrometrisch und durch Vergleich der Retentionszeiten identifiziert) im Pyrogramm indiziert. Diese diagnostischen Marker sind in der Tabelle detailliert aufgeschlüsselt. Tabelle der indizierten Markerverbindungen Peak- Indiz

Name der Markerverbindung Pyrogramm

Name nichtmethylierten Verbindung

Herkunft aus Malschicht-Komponente:

1 N-Methylpyrrol Hydroxyprolin Kollagene Proteine 2 Glycerintrimethylether Glycerin Öle Fette 3 5,6-Dimethyluracil - Kollagene Proteine 4 1,4-Dimethylpiperazin-2,5-

dion Glycin Kollagene Protein

5 Dodecansäuremethylester Dodecansäure Öl insbes. Palm- / Cocosöl, Fette

6 Azelainsäuredimethylester (Nonandisäuredimethylester)

Azelainsäure Trocknende Öle

7 Myristinsäuremethylester Myristinsäure Palm- / Cocosöl, tier. Fette + Öle

8 Palmitinsäuremethylester Palmitinsäure Öle Fette 9 Stearinsäuremethylester Stearinsäure Öle Fette

158

Pyrogramm der Probe mit den indizierten diagnostischen Markern

Chromatogram PlotFile: ...tellungen\mucha\eigene dateien\gcms\thm\a proben\badde1a s1 thm.smsSample: Badde1a S1 THM Operator: Scan Range: 1 - 3826 Time Range: 0.00 - 59.97 min. Date: 21.07.2009 18:12

10 20 30 40 50 minutes

0

250

500

750

kCounts RIC Merged badde1a s1 thm.sms 2000 CENTROID RAW

1

2

34

5

6

7

8

9

Seg 1 Seg 2, Messphase, Time: 2.00-60.00, EI-Auto-SIS, 47-650 m/z

630 1276 1914 2549 3179 Scans

159

Probe1, Raynal Schicht 2 THM- (Thermally assisted hydrolysis and methylation) Pyrolyse-Gaschromatographie/Massenspektrometrie (THM-Py-GC/MS) Ergebnis: Es wurde Alkydharz als Bindemittel verwendet. Details: polyfunktioneller Alkohol: Glycerin, mehrbasige Säure: Orthophthalsäure Additive: trocknendes Öl, wahrscheinlich Leinöl (für eine exakte Bestimmung der Ölsorte lag der Ölanteil zu niedrig) Es handelt sich um ein kurzöliges Alkydharz, der Ölzusatz stellt im Vergleich zum Alkydanteil den kleineren Anteil im Bindemittelsystem. Zudem ist in geringen Mengen Paraffin nachweisbar. Die wichtigsten Pyrolyseprodukte wurden (massenspektrometrisch und durch Vergleich der Retentionszeiten identifiziert) im Pyrogramm indiziert. Diese diagnostischen Marker sind in der Tabelle detailliert aufgeschlüsselt. Tabelle der indizierten Markerverbindungen Peak- Indiz

Name der Markerverbindung Pyrogramm

Name nichtmethylierten Verbindung

Herkunft aus Malschicht-Komponente:

1 Glycerintrimethylether Glycerin Öl, Alkydharz 2 Dimethylorthophthalat Orthophthalsäure Alkydharz 3 Azelainsäuredimethylester

(Nonandisäuredimethylester)

Azelainsäure Öl

4 Palmitinsäuremethylester Palmitinsäure Öl 5 Stearinsäuremethylester Stearinsäure Öl

160

Pyrogramm der Probe mit den indizierten diagnostischen Markern

Chromatogram PlotFile: ...n\administrator\eigene dateien\gcms\thm\a proben\badde1a s2 thm.smsSample: Badde1a S2 THM Operator: Scan Range: 1 - 3861 Time Range: 0.00 - 59.97 min. Date: 22.07.2009 17:07

10 20 30 40 50 minutes

0

100

200

300

400

500

kCounts RIC Merged badde1a s2 thm.sms 2000 CENTROID RAW

1

2

34

5 Paraffin

Seg 1 Seg 2, Messphase, Time: 2.00-60.00, EI-Auto-SIS, 47-650 m/z

632 1276 1914 2569 3211 Scans

161

Probe2, Tasso oberste Schicht THM- (Thermally assisted hydrolysis and methylation) Pyrolyse-Gaschromatographie/Massenspektrometrie (THM-Py-GC/MS) Ergebnis: Als Bindemittel wurde Alkydharz verwendet. Details: polyfunktioneller Alkohol: Pentaerythritol, mehrbasige Säure: Orthophthalsäure Additive: Leinöl Es handelt sich um ein kurzöliges Alkydharz, der Ölzusatz stellt im Vergleich zum Alkydanteil einen kleineren Anteil im Bindemittelsystem. Die Peakintegration ergab ein Flächenverhältnis von Palmitinsäuremethylester/ Stearinsäuremethylester von 1,9, was in Übereinstimmung mit aktuellen Publikationen die Ölsorte innerhalb der Gruppe trocknender Öle auf Leinöl festlegt (Leinöl < 2, Walnussöl ~ 3,5, Mohnöl ~ 5,1). Die Zusammensetzung entspricht einem typischen handelsüblichen Alkydharz. Die wichtigsten Pyrolyseprodukte wurden (massenspektrometrisch und durch Vergleich der Retentionszeiten identifiziert) im Pyrogramm indiziert. Diese diagnostischen Marker sind in der Tabelle detailliert aufgeschlüsselt.

162

Tabelle der indizierten Markerverbindungen Peak- Indiz

Name der Markerverbindung Pyrogramm

Name nichtmethylierten Verbindung

Herkunft aus Malschicht-Komponente:

1 Glycerintrimethylether Glycerin Öl 2 Pentaerythritoltetramethylet

her Pentaerythritol Alkydharz

3 Pentaerythritoltrimethylether Pentaerythritol Alkydharz 4 Suberinsäuredimethylester Suberinsäure Öl 5 Dimethylorthophthalat Orthophthalsäure Alkydharz 6 Azelainsäuredimethylester

(Nonandisäuredimethylester)

Azelainsäure Öl

7 Palmitinsäuremethylester Palmitinsäure Öl 8 Stearinsäuremethylester Stearinsäure Öl Pyrogramm der Probe mit den indizierten diagnostischen Markern

Chromatogram PlotFile: ...n\administrator\eigene dateien\gcms\thm\a proben\badde2a s1 thm.smsSample: Badde2a S1 THM Operator: Scan Range: 1 - 3820 Time Range: 0.00 - 59.97 min. Date: 22.07.2009 10:48

10 20 30 40 50 minutes

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

MCounts RIC Merged badde2a s1 thm.sms 2000 CENTROID RAW

12

3

4

5

67

8

Seg 1 Seg 2, Messphase, Time: 2.00-60.00, EI-Auto-SIS, 47-650 m/z

627 1256 1892 2533 3175 Scans

163

Probe4, Schiller, oberste Schicht THM- (Thermally assisted hydrolysis and methylation) Pyrolyse-Gaschromatographie/Massenspektrometrie (THM-Py-GC/MS) Ergebnis: Analog zur Probe 2 (Tasso oberste Schicht) Als Bindemittel wurde Alkydharz verwendet. Details: polyfunktioneller Alkohol: Pentaerythritol, mehrbasige Säure: Orthophthalsäure Additive: trocknendes Öl, wahrscheinlich Leinöl (für eine exakte Bestimmung der Ölsorte lag der Ölanteil zu niedrig) Es handelt sich um ein kurzöliges Alkydharz, der Ölzusatz stellt im Vergleich zum Alkydanteil einen kleineren Anteil im Bindemittelsystem. Die Zusammensetzung entspricht wiederum einem typischen handelsüblichen Alkydharz. Die wichtigsten Pyrolyseprodukte wurden (massenspektrometrisch und durch Vergleich der Retentionszeiten identifiziert) im Pyrogramm indiziert. Diese diagnostischen Marker sind in der Tabelle detailliert aufgeschlüsselt.

164

Tabelle der indizierten Markerverbindungen Peak- Indiz

Name der Markerverbindung Pyrogramm

Name nichtmethylierten Verbindung

Herkunft aus Malschicht-Komponente:

1 Glycerintrimethylether Glycerin Öl 2 Pentaerythritoltetramethylet

her Pentaerythritol Alkydharz

3 Pentaerythritoltrimethylether Pentaerythritol Alkydharz 4 Suberinsäuredimethylester Suberinsäure Öl 5 Dimethylorthophthalat Orthophthalsäure Alkydharz 6 Azelainsäuredimethylester

(Nonandisäuredimethylester)

Azelainsäure Öl

7 Palmitinsäuremethylester Palmitinsäure Öl 8 Stearinsäuremethylester Stearinsäure Öl Pyrogramm der Probe mit den indizierten diagnostischen Markern

Chromatogram PlotFile: ...n\administrator\eigene dateien\gcms\thm\a proben\badde4a s1 thm.smsSample: Badde4a S1 THM Operator: Scan Range: 1 - 3910 Time Range: 0.00 - 59.97 min. Date: 22.07.2009 13:53

10 20 30 40 50 minutes

0

100

200

300

400

500

600

700

800

kCounts RIC Merged badde4a s1 thm.sms 2000 CENTROID RAW

2 3

14

5

6 7 8

Seg 1 Seg 2, Messphase, Time: 2.00-60.00, EI-Auto-SIS, 47-650 m/z

636 1275 1931 2606 3275 Scans

165

Pigmentuntersuchungen Probe Raynal, Schicht 1, weiß Eine Probe der weißen Schicht 1 wurde mit einem Spotdurchmesser von 0,3mm angeregt. Das Rhodiumsignal ist messtechnisch bedingt (Röntgenröhre) und ohne Relevanz. In der Probe sind Verbindungen folgender Elemente enthalten: Hauptanteil: Zink Nebenanteil /Spuren: Schwefel, Calcium

Mikrochemische Untersuchung: Ergebnis: Es handelt sich bei dem Weißpigment um Zinkweiß. Schwefel und Calcium stammen von anhaftenden Verunreinigungen durch das Trägermaterial (Gips). Das Zinkweiß wurde mikrochemisch nachgewiesen, da es im mittleren Infrarot keine charakteristischen Banden zeigt. [A spot test for zinc white: Antoni Palet Casas and Jaime de Andres Llopis; Studies in Conservation 47 (2002) 273-276]

S Rh

Ca

Zn

Zn

166

Probe Tasso, Gelbpigment der obersten Schicht Eine Probe der gelblichen obersten Schicht wurde mit einem Spotdurchmesser von 0,3mm angeregt. Das Rhodiumsignal ist messtechnisch bedingt (Röntgenröhre) und ohne Relevanz. In der Probe sind Verbindungen folgender Elemente enthalten: Hauptanteil: Schwefel, Barium, Zink Nebenanteil /Spuren: -

S

Ba

Ba

Zn

Zn

167

Fourier-Transform Infrarot-Spektroskopie (FT-IR-Spektroskopie): Ergebnis: Es handelt sich bei dem Weißpigment um Lithopone. Gelbliche Verfärbungen: Es handelt sich offenbar um Verfärbungen organischen Ursprunges, da keine farbgebenden Metallionen nachweisbar sind. Ein eventuelles Eisensignal wäre an der eingezeichneten roten Linie (RFA-Spektrum) erschienen. Dies stimmt auch mit der Beobachtung überein, dass es sich mehrheitlich um gelbliche Verfärbungen der weißen Matrix handelt und kaum Pigmentpartikeln erkennbar sind.

4000 ,0 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 526 ,0cm -1

% T

Tassoob ers te S ch ich t

L ithop one

168

Probe Schiller, Pigment der obersten Schicht Eine Probe der gelblichen obersten Schicht wurde mit einem Spotdurchmesser von 0,3mm angeregt. Das Rhodiumsignal ist messtechnisch bedingt (Röntgenröhre) und ohne Relevanz. In der Probe sind Verbindungen folgender Elemente enthalten: Hauptanteil: Schwefel, Barium, Zink Nebenanteil /Spuren: Ca Das Rhodiumsignal der Röntgenröhre tritt aufgrund der geringen Größe des Probenpartikels in Erscheinung, ist aber ohne Bedeutung.

Fourier-Transform Infrarot-Spektroskopie (FT-IR-Spektroskopie): Ergebnis: Es handelt sich bei dem Weißpigment ebenfalls um Lithopone.

S Ba

Ba

Zn

Zn Rh

Ca

169

4400 ,0 4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 450 ,0cm -1

% T

S ch ille rob e rs te S ch ich t

L ithop one

170

Methoden / Messbedingungen: THM- (Thermally assisted hydrolysis and methylation) Pyrolyse-Gaschromatographie/Massenspektrometrie (THM-Py-GC/MS) Es erfolgte zur Erzeugung flüchtigerer, weniger reaktiver und damit besser indizierbarer Methyl-Derivate des Pyrolysats, eine Hydrolyse/Methylierung in situ während der Pyrolyse durch Zugabe von 1µl TMAH (25% Tetramethylammoniumhydroxid in Wasser) Pyrolyse: 20s bei 700°C im Quarzröhrchen mit CDS Pyroprobe 5000; GC/MS-System: Varian Saturn GC 3900 und MS 2100 System; Säule: Varian FactorFour CP8944, 30m x 0,25mm ID DF=0,25); GC: Start 50°C, Hold 2min, Heizrate: 5K/min bis 300°C, Hold 8min; MS: Ionization Mode: EI Auto, Ion Preparation SIS Die Identifizierung der diagnostischen Marker erfolgte durch Vergleich der Retentionszeiten und der Massenspektren mit eigenen Referenzmessergebnissen, der NIST02 Spektrendatenbank und Ergebnissen aus wissenschaftlichen Publikationen. Die diagnostischen Marker sind in den Tabellen detailliert aufgeschlüsselt und im Pyrogramm indiziert.

171

Röntgenfluoreszenzanalyse (RFA) System: Röntgenfluoreszenzspektrometer System: Nanomaster Meßdauer 300s Hochspannung 40kV Strom 800µA Kollimator 300µm Kanalanzahl 1024 Energieeichung E(keV) = 0,020 + 0,032 x K Röhre OXFORD µFocus Be Anode Rh Applikation QUALITATIV System: Röntgenfluoreszenzspektrometer DX-95 mit Analysator DX-4 der Firma EDAX I = 250 µA; U = 25,00 kV X-ray Path: Vacuum Detector Parameters: Resolution 148,9 eV Be Thickness 10,00 µm Für die die Röntgenfluoreszenzuntersuchung wurde ein geeignetes Probenpartikel an der interessierenden Stelle mit einem Spotdurchmesser von 300µm angeregt. Es sind jeweils mehrere Elektronenübergänge möglich, weshalb die Elemente meist mehrere Signale erzeugen. Es werden Elemente ab der Ordnungszahl 23 (Natrium) detektiert. Organische Verbindungen (C, H, O, N) werden nicht detektiert.

172

Fourier-Transform Infrarot-Spektroskopie (FT-IR-Spektroskopie): System: Perkin Elmer FT-IR-Spektrometer Paragon 1000 PC mit i-Series FT-IR-Mikroskop Spectrum BX FT-IR Instrument Setup Setup Software Revision: Version 4,21 15.06.2000 Scan Mode: Resolution : 4,00 cm-1 Apodization : Strong Static Gain : 1 OPD Velocity : 1,5 cm/s Interferogram Direction : Bi-Directional Instrument Details: Source : MIR Beamsplitter : KBr Detector : MCT Detector Location : External Serial Number : 39843 Firmware Revision : 1,03 Scan Parameters: Beam Type : Background Data Type : Single Number of Scans : 60 Scan Start : 4000 cm-1 Scan End : 526 cm-1 Shuttle : Not Fitted APV Filter : Not Fitted

173

Verwendete Literatur Öle [Van den Berg et al.; Chemical changes in curing and ageing oil paints, Triennial meeting (12th), Lyon 1999, 248-253] [Dominique Scalarone et al.; Thermally assisted hydrolysis and methylation-pyrolysis-gas chromatography/mass spectrometry of light-aged linseed oil; Journal of Analytical and Applied Pyrolysis 58 59 (2001) 503-512] [Francesca Cappitelli et al.; An initial assessment of thermally assisted hydrolysis and methylation-gas chromatography/mass spectrometry for the identification of oils from dried paint films; Journal of Analytical and Applied Pyrolysis 63 (2002) 339-348] Challinor: A rapid simple pyrolysis derivatisation gas chromatography – mass spectrometry method for profiling of fatty acids in trace quantities of lipids Alkydharze [Challinor; Structure determination of alkyd resins by simultaneous Pyrolysis methylation; Journal of analytical and applied pyrolysis; 18 (1991) 233-244] [Cappitelli; The chemical characterisation of alkyd resins by THM-GC/MS and FT-IR-spectroscopy; Conservation Science 2002, 243-248] [Tom Learner: The Analysis of synthetic paints by pyrolysis-gas-chromatography-mass spectrometry; Studies in Conservation 46 (2001) 225-241] Kohlenhydrate [D. Fabbri, R. Helleur: Characterization of the tetramethylammonium hydroxide thermochemolysis products of carbohydrates; Journal of Analytical and Applied Pyrolysis 49 (1999) 277-293] [C. Schwarzinger et al.: Pyrolysis-gas chromatography / mass spectrometry and thermally assisted hydrolysis and methylation (THM) analysis of varios cellulose esters; Journal of Analytical and Applied Pyrolysis 58 59 (2001) 513-523] [C. Schwarzinger et al.: Levoglucosan, cellobiose and their acetates as model compound for the thermally assisted hydrolysis and methylation of cellulose and cellulose acetate; Journal of Analytical and Applied Pyrolysis 62 (2002) 179-196] Synthetische Bindemittel, Kunstharze [Tom Learner: The Analysis of synthetic paints by pyrolysis-gas-chromatography-mass spectrometry; Studies in Conservation 46 (2001) 225-241]

174

[Tom Learner: The Analysis of synthetic resins found in the twentieth century paint media; Resins ancient and modern: pre-prints of the SSCR s 2nd resins conference held in Aberdeen 1995 76-84] Diterpenoide Harze [Dominique Scalarone, Massimo Lazzari, Oscar Chiantore; Ageing behaviour and pyrolytic characterisation of diterpenic resins used as art materials: colophony and Venice turpentine; Journal of Analytical and Applied Pyrolysis 64 (2002) 345-361] [Berg, Klaas van den, Pastorova, Ivana, Spetter, Leo, Boon, J. Jaap: State of oxidation of diterpenoid Pinaceae Resins in varnish, Wax lining material, 18th century resin oil paint, and a recent copper resinate glaze; Triennial Meeting(11th), Edinbourg 1996 930-938] [Pastorova et al.: Analysis of oxidised diterpenoid acids using thermally assisted methylation with TMAH; Journal of analytical and applied pyrolysis, 43 1997 41-57] [van den Berg et al.: Mass spectrometric methodology for the analysis of highly oxidized diterpenoid acids in old master paintings; Journal of mass spectrometry : JMS 35 2000 512-533] Proteine [John M. Challinor, Thermally Assisted Hydrolysis and Derivatisation Techniques for the Characterisation of Organic Materials, 1998, 77 ff., Dissertation] [G. Chiavari, N. Gandini, P. Russo, D. Fabbri; Characterisation of Standard Tempera Painting Layers Containing Proteinaceous Binders by Pyrolysis (/Methylation)-Gas Chromatography-Mass Spectrometry; Chromatographia Vol. 47, No. 7/8, April 1998] [Xu zang, Johnie C. Brown, Jasper D.H. van Heemst, Amanda Palumo, Patrick G. Hatcher: Characterization of amino acids and proteinaceous materials using online tetramethylammonium hydroxide (TMAH) thermochemolysis and gas chromatography-mass spectrometry technique; Journal of Analytical and Applied Pyrolysis 61 (2001) 181-193] [Ilaria Bondaduce and Maria Perla Columbini: Gas chromatography / mass spectrometrie for the characterization of organic materials in frescoes of the Monumental Cemetery of Pisa (Italy); Rapid Communications in Mass Spectrometry 17 (2003) 2523-2527]

175

Leistungs- und Preisverzeichnis für naturwissenschaftliche Untersuchungen Substanzklasse / Methode Kosten je Probe in

EURO Bindemittelklassen- und Pigmentanalyse mittels FT-IR-Spektroskopie (ATR oder Mikroskop) Eingrenzung der Bindemittelklasse: Öl, Harz, Wachs, tierischer Leim, pflanzlicher Leim Pigmentanalyse durch kombinierte Röntgenfluoreszenz- und Infrarotspektroskopie

40,00 80,00

Pigmentbestimmung mit Röntgenfluoreszenzanalyse (RFA) optional Elementaranalyse von Legierungen, Auflagerungen, Baustoffen etc. Aufnahme eines RFA-Spektrums und Auswertung

meist in Kombination mit der FT-IR-Spektroskopie 40,00

Detaillierte Bindemittelanalyse mittels Pyrolysegaschromatographie-Massenspektrometrie Tierische Leime: Kasein, Hühnerei, Glutinleime (Hautleim, Knochenleim etc.) Trocknende Öle: Leinöl, Mohnöl, Walnussöl Wachse: Bienenwachs, Carnaubawachs, Candellilawachs, Montanwachs, Paraffin Harze: Dammar, Kopale, Mastix, Kolophonium, Schellack, Terpentine, Elemi etc. Pflanzliche Leime: Stärken, Cellulosederivate, Gummen Kunstharze: Alkydharze, Acrylharze, Polyurethane, Copolymere, Polyvinylacetate, Vinyl-Acryl-Copolymere, Epoxidharze, Siliconharze, Additive, Weichmacher Komplett:

120,00

Bestimmung der Anionen bauschädlicher Salze (Chlorid, Nitrat, Sulfat, Phosphat und Fluorid), quantitativ, ionenchromatographisch,

55,00

Bestimmung der Zusammensetzung von Baustoffen (Zuschläge, Bindemittel, Korngrößenverteilung, Sieblinie)

nach Aufwand

Querschliff (Anfertigung) Mikroskopische Aufnahmen (Auflicht oder UV-Fluoreszenz) Untersuchung /Analyse der Schichtfolge

40,00 30,00 nach Aufwand

Dünnschliff Anfertigung Aufnahme (Durchlicht , UV-Fluoreszenz oder Polarisation) Untersuchung und Auswertung

70,00 30,00 nach Aufwand

Faser-Identifizierung Spektroskopisch, mikroskopisch (mit Aufnahme)

65,00

Bestimmung chemisch/physikalischer Größen Schmelzpunkt pH-Wert

20,00 20,00

Holz Untersuchung auf Holzschutzmittel

120,00

Weitere Untersuchungen sowie Erstellung von Gutachten Nach Vereinbarung Es gelten die allgemeinen Vertragsbedingungen der FH Erfurt sowie Haftungsausschluss.

176

Appendix III, Lab. Report Hildesheim

177

178

179

180

181