• No results found

Material Analysis of the Compo Ornamentation in Rooms 129 and 132 at Kasteel de Haar

N/A
N/A
Protected

Academic year: 2021

Share "Material Analysis of the Compo Ornamentation in Rooms 129 and 132 at Kasteel de Haar"

Copied!
132
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Material Analysis of the Compo Ornamentation in Rooms 129 and 132 at Kasteel

de Haar

Research into the cause and effect of the degradation process of compo ornaments.

Master thesis

Conservation and Restoration Master’s programme

Historic Interiors Specialisation 2018

Ms. Jasmine Guest, 11353791

University of Amsterdam

Thesis supervisor: mw. M.N. van Schrojenstein Lantman MA

2nd reader: dhr. dr. H. Herman den Otter

(2)

Figure on cover page: Ceiling Ornamentation Room 129. 1892-1912. Compo ornamentation. Kasteel de Haar. Photo by author, 2017.

(3)

Table of Contents

Table of Contents...3 Summary ...6 Samenvatting...7 Acknowledgments ...8 1 Introduction ...9

1.1 Research question, sub questions and scope of research ...9

1.2 Method ...10

1.3 Relevance of research ...10

2 History and current condition ...11

2.1 Object history and description of Kasteel de Haar and rooms 129 and 132 ...11

2.1.1 Kasteel de Haar ...11

2.1.2 Rooms 129 & 132 ...11

2.1.3 History of use ...13

2.1.4 Maple and Co. LTD and George Jackson and Sons ...14

2.2 Material buildup and history ...15

2.2.1 Ceiling support ...15 2.2.2 Paper ...15 2.2.3 Compo ...16 2.2.4 Paint ...16 2.3 Past Conservation ...16 2.4 Current condition ...18 2.4.1 Room 129 ...20 2.4.2 Room 132 ...21 2.5 Conclusion ...22

3 Current Scientific knowledge of compo ornamentation ...23

3.1 History ...23

3.2 History of compo at Kasteel de Haar ...26

3.3 Characteristic properties ...27

3.4 Conclusion ...30

(4)

4.4.1 Microscopy and SEM-EDX of paint samples ...39 4.4.2 Results ...39 4.5 Discussion ...40 4.5.1 Fibrous plaster ...40 4.5.2 Paper ...40 4.5.3 Compo ...41 4.5.4 Paint ...41 4.5.5 Conclusion ...41

5 Physical characteristic testing ...45

5.1 Thermo-setting heat testing ...45

5.1.1 Procedure ...45 5.1.2 Results ...47 5.1.3 Discussion ...47 5.2 Hygroscopicity test ...47 5.2.1 Procedure ...48 5.2.2 Results ...49 5.2.3 Discussion ...51 5.3 Conclusion ...52

6 Discussion and conclusion ...53

6.1 Further research ...54

6.2 Advice for restoration and preventive conservation ...54

Works Cited ...56

List of figures ...61

Appendix I – Term index ...63

Appendix II – Fibrous plaster production method ...65

Appendix III – George Jackson & Sons, Compo Ornament Archive ...66

Appendix IV – Damage Mapping ...70

(5)

Appendix VI – Gas-Chromatography Mass Spectrometry (GC-MS) ...84

Appendix VII - Scanning Electron Microscopy with Energy Dispersive X-ray Analysis (SEM-EDX) .67

Appendix VIII- Hygroscopicity test ...95

Appendix IX – Temperature and Relative Humidity (RH) of Room 129 at Kasteel de Haar, 2016-2017 ...98

(6)

de Haar are in a severe state of degradation. Almost all the ornaments have deformed in some way. The aim of this thesis is to gain insight into the material configuration of the compo from Kasteel de Haar and the causes for its degradation. Scientific analysis of the compo from Kasteel de Haar will determine its susceptibility and appropriate treatment recommendation and preventive conservation practices.

Kasteel de Haar, designed by Pierre Cuypers, has a rich collection of archival materials. This provided essential details such as the materials in the ceiling construction as found in work estimates. It also provided confirmation that an English company, Maple & Co., was responsible for the design and installation of rooms 129 and 132. Compo archival research also determined that George Jackson and Sons was likely the company that produced the compo ornamentations. Interviews and archival reviews at the company in London provided information about the production method of the compo, the ratio its ingredients and professional insight into the severity of the degradation of the compo ornaments. Furthermore, scientific research (microscopy of the compo and paint layers by SEM-EDX, fibre analysis, and THM-Py-GC/MS) was performed for further understanding of the material and of the degradation process. A sample of compo from room 129 was tested for its equilibrium moisture content and it was also heat tested to determine its thermos-setting ability. This scientific analysis determined the exact composition of the compo at Kasteel de Haar in order to draw conclusions about how the materials of the rooms interacted and how the degradation of the compo ornaments manifested itself.

This research, scientific analysis and physical testing, concluded that the main factors in the degradation process of the compo ornaments are the environmental factors of temperature, and humidity. Hypothetically, the degradation phenomena that occurred primarily in the deformation of the ornaments where shrinking, cracking, curling, and crazing. These damages lead to further damage such as delamination of the compo from the paper support and delamination of the paper itself. Secondly, the lack of structural stability of the castle and vibrations could have also been a major factor in the degradation and the large cracking of the thicker compo ornaments, however this factor was not explored within the scope of this research but is recommended to fully understand it’s relevance in relation to the hypothesis presented here.

(7)

Samenvatting

Materiaal analyse van compo versieringen in kamers 129 en 132 bij Kasteel de Haar Onderzoek naar de oorzaak en gevolg van het afbraakproces van compo ornamenten.

Jasmine Guest, Universiteit van Amsterdam, juni 2018

Het volgende scriptieonderzoek betreft een analyse van de compo versieringen op de plafonds van de ruimten 129 en 132 (1903-1905) in Kasteel de Haar, Utrecht, Nederland. Compo is een thermoplastisch en hygroscopisch materiaal dat uit dierlijke lijm, krijt, hars en olie bestaat. Beide kamers van het Kasteel de Haar zijn in een ernstige staat van degradatie. Bijna alle ornamenten zijn in zekere mate vervormd. Het doel van het onderzoek is om inzicht te krijgen in de materiaal configuratie van compo en de mogelijke oorzaken voor de afbraak te bepalen. Aan de hand van deze informatie zijn aanbevelingen voor behandeling en preventieve conservering bepaald.

Literatuur, analysitsch onderzoek en archiefonderzoek verstrekten informatie over Kasteel de Haar en compo. Kasteel de Haar, ontworpen door Pierre Cuypers, heeft een rijke collectie aan archiefmateriaal. Dit leverde essentiële details, zoals de materialen van de plafondconstructie die konden worden afgeleid uit werkaantekeningen. Daarnaast werd er bevestigd dat een Engels bedrijf, Maple & Co., verantwoordelijk was voor het ontwerp en de installatie van de twee kamers. Uit archief onderzoek naar compo werd vastgesteld dat George Jackson and Sons het bedrijf was dat de compo versieringen vermoedelijk produceerde. Een bezoek aan het bedrijf in Londen verstrekte informatie over de productiemethode van compo, de verhouding van de ingrediënten en een professioneel inzicht in de ernst van de afbraak van de compo ornamenten. Tot slot werd er analytisch onderzoek

(microscopie, XRF van de compo en verflagen met SEM-EDX, vezelanalyse en THM-Py-GC / MS) uitgevoerd om een uitgebreid inzicht te verkrijgen in het materiaal en afbraakproces. Een monster uit kamer 129 werd getest op zijn evenwichtsvochtgehalte. Thermoplastische warmte testen werden uitgevoerd om te bepalen of het materiaal nog buigzaam was bij verwarming. Deze wetenschappelijke analyse bepaalde de exacte configuratie van de compo bij Kasteel de Haar. Hierdoor kon

geconcludeerd worden welke interactie er tussen de materialen plaatsvindt en wat de oorzaak van de degradatie is.

Uit analytisch en fysisch onderzoek is geconcludeerd dat de omgevingsfactoren, temperatuur en vochtigheid, de belangrijkste factoren zijn bij het afbraakproces van de compo ornamenten. Er was hypothetisch vastgesteld dat de degradatie van de ornamenten voornamelijk plaatsvond daar waar de ornamenten vervormden. Deze belangrijkste schades leiden tot verdere schade zoals het loskomen van de papierdrager van het plafond en van het ornament. Het gebrek aan structurele stabiliteit van het kasteel en trillingen kan ook een belangrijke factor zijn voor de delaminatie en de scheuren in de grotere compo ornamenten. Verder onderzoek naar de structuur van het plafond zal hiervoor moeten worden uitgevoerd. Tot slot is er verder onderzoek nodig naar de fysische kenmerken van de compo, voor een volledig begrip van het degradatiemechanisme van het compo materiaal.

(8)

Gratitude is also due to Julia van Velzen, for her initial interest in the compo ornaments at Kasteel de Haar, which first introduced me to the project. To Judith Bohan, who introduced the degradation of the compo ornaments as a thesis subject. Gratitude is also due to those that have supported my research at the Kasteel de Haar foundation, Lucas van Haastert from Bergy Bouwbedrijf BV and Katrien Timmers, curator of the museum.

David Serra, director of George Jackson and Sons, who was very accommodating and knowledgeable regarding the history and fabrication procedure of the company for compo

ornamentation. Compo experts and fabricators at George Jackson and Sons, for the expertise in the production of the ornaments and knowledge of materials new and old.

I would like to thank all those at the RCE (Dutch Cultural Heritage Agency) for their help and support. Bart Ankersmit, my thesis point of contact for guidance and correspondence with the RCE, his guidance and input were essential for this research thesis. Luc Megens, RCE analysist, for his help with SEM-EDX analysis and interpretation. Henk van Keulen for his THM-Py-GC/MS analysis of the binding media in the compo.

Finally, the Dutch translation of my summary is credited to Santje Pander. Finally, I would like to thank my mother, Andrée Cadieux, for her continuous support and thorough editing.

(9)

1 Introduction

This case study is within the context of the master thesis project of the master’s program of conservation and restoration of cultural heritage at the University of Amsterdam, Historic Interiors specialisation. The thesis project is a scientific research paper based on an analytical case study.

The case study for this thesis will be the ceiling composition ornamentation at Kasteel de Haar, (Haarzuilens) Utrecht in rooms 129 and 132. These rooms were brought to my attention by Judith Bohan, who had earlier that year been commissioned to restore wallpaper in other rooms at the castle.

1 Common adhesives and filling materials in the conservation field were tested on the composition

ornaments, but these were found to be unsatisfactory because of the lack of adhesion and incompatibility. Therefore, it was determined that a more in depth analysis of the material was required.

This thesis project endeavors to conclude the issue of the unknown material buildup of the composition ornaments in the rooms and its substrates.2 Currently, the compo ornaments and paint

have begun to delaminate and flake off. The future conservation goal would be to restore the ceiling by consolidating the ornaments. This compo material, dating to the late 19th century, has degraded

more severely and unlike other compo ornaments.(Wetherall 26–29) The support material which makes up the ceiling is also unknown, the ceiling support appears to be a smooth pink plaster like material. It is hypothesized that the pink ceiling material is a form of asbestos cement sheets. To properly restore the decorative elements of the room, a thorough analysis of the unknown material is essential to determine its susceptibility and to determine the causes of the degradation. The hypothesis is that the compo material is composed of a mixture of animal glue, chalk, resin and oil which has degraded due to the temperature and relative humidity fluctuations.

The desire of the Kasteel de Haar foundation is to restore the ceilings and to implement preservation practices for the ceilings. After two years of conservation treatments throughout the castle, the hope is to enhance the museum experience by allowing visitors full access to almost all the rooms at Kasteel de Haar including rooms 129 and 132. Understanding the current condition and susceptible degradation elements of the ceilings will contribute to the conservation of the rooms and their preservation.

1.1 Research question, sub questions and scope of research

The present state of the rooms is that they are in dire need of conservation. The castle has hired an outside company to begin a conservation treatment of the ceilings, therefore this thesis and scope of research will be to determine configuration of compo and the cause of its degradation and the appropriate treatment for future preservation of the rooms and future conservation treatments of compo with a similar configuration. In order to properly structure the research and for an efficient method, the following questions have been derived from research methodology:

How does the material composition of the compo décor elements of the rooms 129 and 132 at Kasteel de Haar impact the degradation and based on these findings, what is the susceptibility?

(10)

tested, will indicate that have been major contributors to the compo ceiling degradation.

1.2 Method

To answer the research question of this thesis, research, analytical and physical testing will be executed. The research of the material composition and degradation of the decorated ceiling in rooms 129 and 132 at Kasteel de Haar was sub-divided into the following sub-questions.

1. What is the current state of degradation in bedrooms 129 & 132 at Kasteel de Haar?

In chapter 2, History and current condition, a discussion of the historical use of the rooms and of the current state of the compo will be determine the cause of the present damages. The most

threatening agents of deterioration based on literary research are water and temperature. The causation of the climate in the rooms, argued to be the most influential factor, will be related to the history and past use of the rooms.

2. What is the current scientific knowledge of compo and its surrounding materials?

Chapter 3, Current scientific knowledge, will investigate literary information regarding the layers that make up the ceiling work of both rooms and of the material configuration of the layers. This

information will guide the investigative scientific analysis of the material to further research the degradation phenomena.

3. What is the material composition of the compo ceiling ornamentations?

Chapter 4, scientific analysis and results, will outline the procedures for the scientific analysis of each material found in the ceiling construction and the results of the analyses. THM-Py-GC/MS (Gas chromatography–mass spectrometry) will help determine the resin, oil and binding media. Microscopy will determine the material composition of the pink ceiling support, compo and the paint layers. SEM-EDX (Image Scanning electron microscopy with energy dispersive X-ray spectroscopy) will further determine the elemental composition of the compo that was not identified in the THM-Py-GC/MS testing and the type of paint used on the ceilings. Finally, fiber analysis will be used with polarized light to identify the fibers in the paper that determines the type of paper used between the compo and the pink ceiling support. The results of the scientific analysis should answer the following question:

4. What is the susceptibility of the materials in the rooms?3

This questions will be answered in chapter 5, physical characteristic testing. This chapter will analyse the characteristics of this aged compo with regards to water sorption and temperature. Chapter 6,

discussion and conclusion, concludes the research question of determining the composition of the ceilings

of rooms 129 and 132 at Kasteel de Haar in order to determine the main factors of its degradation. The research will also provide recommendations for the preservation and restoration of the compo decorated ceiling.

1.3 Relevance of research

The rooms 129 and 132 at Kasteel de Haar have previously been examined by Tineke Slot Jorritsma during the 2007 conservation project and very briefly by Judith Bohan in 2017. Both reports concluded that a more thorough investigation into the material buildup of the ceilings is required in

(11)

order to properly recommend a conservation treatment. This research will not only help with the treatment proposal of the two rooms at Kasteel de Haar, but also serve as technical research into the compo ornamentation fabrication procedure during the turn of the 20th century. Little historic

literature can be found on the recipes and production techniques because they were undisclosed company secrets (Serra). Furthermore, research of the chemical and physical degradation phenomena of compo will contribute to the conservation and restoration of the material which have been used on a variety of surfaces in historic interiors, furniture, and frames.

2 History and current condition

This chapter will outline the relevant history of Kasteel de Haar and rooms 129 and 132. In order to properly asses a material buildup of the compo ornaments and the surrounding materials an analysis of the current condition of the ornaments and the extent of the degradation is necessary. Additionally, the history of the castle and its use will aid in determining how changes in climate affects the degradation of the ceilings.The object history will include past conservation treatments of the room with a short summary of reports by Tineke Slot Joritsmas and Judith Bohan. These past conservation reports and conclusions contribute to the understanding of the current condition and damage mapping of the ceilings. Finally, the materials that are found in the ceiling are described, based on archival estimates addressed to Cuypers by Maple and Co. LTD, London. This information is essential for the following chapter of current scientific knowledge to establish the material and degradation present for further research.

2.1 Object history and description of Kasteel de Haar and rooms 129 and 132

2.1.1 Kasteel de Haar

Kasteel de Haar is currently owned by the original founding family, the Van Zuylen van Nijevelt van de Haar family. The castle, although still occupied by the family for one month a year, has adapted its function from a domestic dwelling to one of the most visited Dutch museums. Visitors can now have guided tours of the castle, and it is used as a space for exhibition, theatre, and for other public events.

The castle is built on the ruins of a medieval castle, dating back to the 13th century. From 1892

to 1912, the castle was rebuilt and redesigned by the famous Dutch architect, Pierre Cuypers. The rebuild was commissioned by the baron Etienne van Zuylen van Nijevelt van de Haar. The new design was all inclusive extending from the architecture, to the interior design, to the gardens. (About De Haar

Castle) The design of the entire castle is in the neo-gothic and neo-renaissance style.(RCE)

2.1.2 Rooms 129 & 132

The rooms that this thesis focuses on, 129 and 132, are designed in the neo-Classical Adam style. They significantly differ from the rest of the Neo-Gothic castle and were not designed by Cuypers. Instead, the rooms were made and installed by the London based company, Maple & Co. LTD. In room 129, the ceiling is decorated in an oval form with urns, palmettos and leaf tendrils at the corners. Within the oval are eight medallions with scenes of putti with attributes that supposedly symbolize the four elements.(Bohan 34) (Figure 2-1) In Room 132, the Adam style is continued, but in a simpler design. The oval shape design on the ceiling, is subtler with no roundels or extra embellishments. (Figure 2-2)

(12)

Figure 2-1 KDH.129, Ceiling, made by author, 2018.

(13)

2.1.3 History of use

In order to most accurately hypothesise the cause of the degradation of the compo ornaments in rooms 129 and 132, the history of the usage of the rooms must be studied. As previously

mentioned, the castle is used by the van Zuylen family for the month of September every year, and in the other months, the castle entertains events and tours. The two rooms investigated in this thesis are not on the main tour route and were only visited in the past on guided tours.(Kimberly Letter and Katriene Timmers) They may be used by the family when they are at the castle, but there is no

supporting documentation to confirm this. The future use of the rooms, after the two-year renovation project, will most likely be a part of the main touring route of the castle. (Kimberly Letter and Katriene Timmers) Understanding the historic use of the rooms will enable to model the historic climate. This will subsequently help understand the possible impact of the climate on compo in the past and in the future.

Since the beginning, the castle was equipped with the most modern technologies, electricity, radiators, hot and cold water, etc. Radiant heat was therefore always use even though the rooms are fitted with fireplaces. The castle was originally built as a summer house and therefore, was never heated or used all year long. (Kimberly Letter and Katriene Timmers) Since the 1950s, the family, when visiting Kasteel de Haar, stayed in the adjacent châtelet which is less grand and more

comfortable. (Heijenbrok et al. 424) In 2000, the castle and grounds were bought by the municipality and the Kasteel de Haar foundation was established. This allowed for government funding for desperately needed restorations. The grounds were bought by Vereniging Natuurmonumenten and the castle became open to the public. (561) This change in stewardship of the castle allowed for the following restorations; foundation repair, reinstallation of the heating system, and other restoration projects in the castle, châtelet and grounds.

During the last restoration project, in 2007, water leaked through the roof into rooms 129 and 132, along with other rooms on the first floor. This leakage was minimally documented, however it could be a factor in the degradation of the rooms based on its probable impact on climate in the rooms. A summary of the damages and possible causes for the degradation can be found in Appendix X. Further research into the compo material dating to the time of its installment at Kasteel de Haar

(14)

changes of room 129, recorded over a two year period, are provided in appendix IX. This data will be discussed in relation to the degradation phenomena of the compo in chapter 5.

2.1.4 Maple and Co. LTD and George Jackson and Sons

Maple and Co. LTD, London, is the company that was hired to design and install the Adam style rooms at Kasteel de Haar. Estimates and correspondence between the company and Pierre Cuypers for the main rooms can be found at the Het Nieuwe Instituut (state archive for Dutch

Architecture and Urban Planning) where descriptions of the work and the materials to be installed are outlined. Maple and Co. was founded in 1841 by John Maple and James Cook, and officially became the company Maple and Co. in 1851, after the partnership dissolved. The Company remained a family business until the Second World War, after which, it was taken over by a larger furnishing company. (Barty-King 131) Maple and Co. officially ended in 1997, when the Allied Maples Group went into bankruptcy. (V&A Collections)

Figure 2-3 Antique print, Advert Maple & Co. furniture, 1893, www.pemgshop.top.

Even though Maple & Co. were commissioned to work on these rooms, the ceiling and architectural ornaments where not fabricated by them. At the time, George Jackson & Sons was a well-established compo and fibrous plaster producer, also located in London. George Jackson, from whom the company acquired its namesake, was a supplier and contemporary of Robert Adam. (Simon; Cooper Hewit Museum) The company, founded in 1780, is still in business and continues to make

(15)

ceiling and architectural ornaments. (V&A Archives) In fact, examination of their personal archives and those of the Victoria and Albert Museum of London, indicates that molds and images of their studio match the ceiling ornaments in rooms 129 and 132 at Kasteel de Haar.4 It is assumed that

George Jackson & Sons was most likely the manufacturer of the compo ornaments. (Serra)

2.2 Material buildup and history

The following quotes from the estimate by the Maple & Co. gives a description of the work commissioned on the rooms at Kasteel de Haar. Although rooms of 129 and 132 are not specifically mentioned, descriptions of the materials used and work estimates provide a good sense of the materials used in those rooms. The following rendering of the material buildup of the layers of the ceiling are based on archival research and on-location observation.

Figure 2-4 Ceiling and ornamentation material buildup, created by author, 2018.

2.2.1 Ceiling support

What was previously referred as the pink support, Maple and Co. quotes in the archive work estimate as being ‘fibrous plaster slabs’ made of ‘Keen’s cement’ and secured to battens.

The work estimate found in the archives of Kasteel de Haar at Het Nieuwe Institute sates the following: “[…] form the ceiling in line with underside of beams by fixing battens suspended by irons if necessary, supply and fix fibrous plaster slabs to same and finish in Keen’s cement.”(Correspondentie,

Bestellingen En Financiële Stukken van ‘Maple & Co.’’’) The fibrous plaster board was first patented by C.F.

Bielefeld. The fibrous ‘slabs’ are made of sulphate of lime, gypsum, or plaster of Paris, with a fiber composed of an open weave jute, which is strengthened with wood.(Millar and Bankart 133) The founder’s son of George Jackson and Sons, introduced fibrous plaster to the London market. This new material completely reinvented the production of interiors with a faster and more economical production and installation method. Fibrous plaster was used by George Jackson and Sons in theaters, stately homes, museums, etc. (George Jackson Limited, sec.History)

Based on the similarity of the ornaments in the rooms and the documented compo ornaments and molds from the George Jackson and Sons and the Victoria and Albert museum archives, and because George Jackson and Sons commonly used fibrous plaster in their interiors. It is therefore assumed that the company installed the two rooms at Kasteel de Haar.

2.2.2 Paper

Very little can be found about the paper support to which the compo ornaments are adhered. In the Het Nieuwe Instituut archived correspondence between Pierre Cuypers and Maple and Co. LTD, the lining of the ceiling is described as ‘white lining paper’.(Correspondentie, Bestellingen En Financiële

(16)

2.2.4 Paint

Further analysis of the paint will be conducted to determine the exact type of paint used in the rooms, but the following excerpt from the Maple & Co. estimates from the state archives confirms the following; “[…] peindre a trois couches et a vernis d’huile sans poli.” Decoration des chambers a

choucher au second. (Correspondentie, Bestellingen En Financiële Stukken van ‘Maple & Co.’’’) This 1900 French

quote is about the application of paint in the rooms at Kasteel de Haar. It states that the paint should be applied in three layers and varnished in oil without shine. Therefore, the paint applied in the bedrooms at the castle, are expected to have three layers of oil paint, applied one after another with a matt finish. The matt finish was already observed from visual observation of the paint in the rooms. If no modern paint has been applied since the 1950s, it should be all oil paint. This can be officially determined by microscopic analysis of a cross-section of a sample taken from room 129 which will be analysed and concluded in Chapter 4 Scientific research and analysis .

2.3 Past Conservation

Two conservators have made preliminary reports in order to make estimates of the work required for the conservation of the compo ceiling ornaments. Another brief report is also accessible about the work completed on the ornamentation in the Baroness’ room. This room, although not examined in this thesis, was completely restored and offers some guidance for the research of rooms 129 and 132. The first of the complete estimate assessments of the rooms 129 and 132 was made by Tineke Slot Jorritsma during the previous restoration project at the castle in 2007, which included the restoration in the Baroness’ room. The second was most recently by Judith Bohan in 2017.

The last restoration project at Kasteel de Haar was completed by the SRAL5 in 2009, in which

many rooms were restored. In the Pilot Kasteel de Haar Executed Restoration Work report, a summary of the completed project, by Tineke Slot Jorritsma, of the restoration of the ‘carton pierre’ or compo ornamentation in the rooms of the Baroness is given. Unlike the ceiling compo ornaments researched in this thesis, the ornaments featured in this room have a support wire integrated in the ornaments for additional structural support.(SRAL) This difference in production of the compo ornament may have been necessary for these ornaments, since they are not adhered to a support, but draped in the corner of a frame, over a mirror.

The first assessment by Tineke Slot Jorritsma, reported the possibility of treating the ornaments. She concluded that four types of compo material were used in the rooms. (Slot Jorritsma 1) The carton-pierre material is a term used interchangeably in her report with compo. She concluded not only the material being carton-pierre, but also noted the panel ceiling structure, coloured pink, and the a paper surface, to which the carton pierre is adhered to. The degradation of the rooms in 2007 described by her is similar to what can be found today. The compo exhibits a lot of shrinking and warping at many and frequent intervals, with fractures of about 2-7mm. The warping and distortion

(17)

was apparent to all sides of the rooms, where in some locations the warping had ripped the paper support from the pink ceiling sheets. She also noted cracks in room 129 which had accumulated a black substance; which may be dirt accumulation or mould.(Slot Jorritsma 8) Jorritsma concluded that the ornaments were made of carton-pierre, but this is later discredited by Judith Bohan in her report in 2017. Carton-pierre, by definition, has a paper element in their composition, but the ornaments in rooms 129 and 132 do not. Therefore, the ceiling ornaments are not carton-pierre.

In Judith Bohan’s report from December 2017 on the inventory and testing of ornament ceilings in rooms 129 and 132, Kasteel de Haar, Haarzuilens,(Bohan 39) she reported some consolidation testing using common conservation adhesives which proved to be inconclusive. The consolidation adhesives that she used were Evacon, Mowilith DMC 2, Mowilith 20 with 20% ethyl acetate, and Plextol (type not provided). The adhesives had adverse effects on the paper support due to its sensitivity to moisture and did not adhere to the deformed compo ornaments due to their brittle nature.It is therefore imperative that research is done on all of the materials of the ceilings to ensure proper consolidation and further restoration of the ceiling.

Kasteel de Haar has employed Verlaan & Bouwstra Architects, to lead the work on the compo ceiling in rooms 129 and 132. The company has also been involved in a past pilot restoration project at the castle. (Verlaan & Bouwstra) Research on the ornaments and the material structure by the

(18)

gathered, but (presumably) not all.

(19)

Figure 2-6 Shrinking, curling and delamination of compo in room 129, photo by author, 2017.

These damages are caused by internal and external factors. The internal factors are the material buildup of the compo and its susceptibility. This is hypothesized as the cause for the shrinkage and cracking observed. The external factors cause the damages on the compo and surrounding materials due to temperature and humidity in the rooms. These factors cause the most prominent form of the deformation and the delamination of the compo from the paper substrate and the pink support.(Bohan 38)

The degradation of the ceiling compo ornaments has been documented and is included in a damage mapping and assessment of the rooms. (For damage mapping of both rooms please refer to appendix IV) The complete loss of material, seems to be more frequent in room 129, but both rooms have lacunae, where the compo material has completely fallen off removing the paper support along with it in certain sections. All of the compo material, except for the putti roundels in room 129, have small cracks that run along the thinnest and protruding parts of the ornaments. A microscopic image of a crack in the ornament is found in Figure 2-7. The location of this sample featured in the

(20)

Figure 2-7 KDH.129.DL.01, Microscopic image of ceiling ornament sample, x50 magnification, LED daylight, photo by author, 2018.

Figure 2-8 Location of sample in figure 2-5, photo by author, 2018.

2.4.1 Room 129

Room 129 shows a lot of nuances in the damages according to the thickness of the ornament. The larger compo borders have a large surface area adhered to the support and demonstrate very little shrinkage. Crazing has occurred on the ‘peaks’ of the almost all the ornaments. (figure 2-7) The larger ornamentation includes the putti roundels, all the boarders, and the chalices in each of the corners.

(21)

Larger cracks have appeared on the outer oval boarder. (figure 2-9) Delamination, shrinking and material loss is almost consistent across the entirety of the ceiling in the smaller ornaments. It should be noted that the degradation appears to be more severe in the corners that are closer to the external walls of the castle. The corner with the most damage is in the North corner. Causes for this increase in severity will be further researched and explained in chapter 4, results and interpretations. About 80% of the smaller ornamentation has some sort of damage. The larger ornamentation has significantly less with only about 10%.

Figure 2-9 Dark cracks on the 'peaks' of the compo, large ornament molding room 129, photo by author, 2018.

2.4.2 Room 132

In the larger room, 40% of the ornamentation is severely affected. The small ornamentation in the corners have the same degradation of shrinkage, cracking and delamination as in room 129. (Figure 20-10) The large oval boarders also have the perpendicular cracks lengthwise. Different to the previous room, more damage has occurred to the paint layer on the flat surface of the ceiling. Long vertical cracks flow along almost from one side to another. Cracks in the paint layers are also present along the edge of the ceiling, where it meets the cornice between the wall and the ceiling. (Figure 2-11) All of the ornamentation has the small cracks or crazing in the ‘peaks’ of the design, where the paint is cracked through to the compo layer and mould has grown.

(22)

Figure 2-10 Curling and delamination of compo ornaments in room 132, photo by author, 2018.

Figure 2-11 Cracking and delamination of the paint and paper layer, photo by author, 2018.

2.5 Conclusion

Archival research and past conservator reports of rooms 129 and 132 provide some confirmations but further research is required to obtain a better understanding of the material

configuration found in the layers of the ceilings. The ceilings and ornamental structure is composed of fibrous plaster ‘slabs’, lining paper, compo ornamentation, and paint. The degradation of the

ornaments consists of delamination, shrinkage, cracking and deformation. The compo ornaments have shrunk and cracked to create gaps of about 2-7mm, causing curling of the individual ornament pieces. The cracks in the ornaments have penetrated the top paint layers to the compo material, where mould has grown. Based on the determined timeline and use of the rooms certain climate deductions can be made. In the following chapter current scientific knowledge of compo ornament the possible causing factors for the degradation of the ceiling ornaments will be explored.

(23)

3 Current Scientific knowledge of compo ornamentation

This chapter will define the configuration of the compo ornamentation in rooms 129 and 132 and its degradation throughout time based on current scientific knowledge.

3.1 History

History of compo and its reversed molds production method is rich and dates back to ancient Egypt. The Adam style compo recipe will be the focus of this chapter. The ‘thermo-setting6 material’

has had many different names since its origins such as “Italian ‘pastiglia’, gesso, carton-pierre, papier-mâché, carta pesta, carta gesso, fibrous slab, bulk gold, pressed on decoration, gilder’s compo, paste compo, London composition, French compo, pate coulante, plaster compo, plaster, stucco” etc…(Millar and Bankart 283–96; Grimm) Compo ornamentation requires negative molds for fabrication, same as ornamentation made of plaster and stucco. What differentiates the compo fabrication from that of plaster and stucco is the requirement for glue to harden. Plaster and stucco only require water to properly set. For this reason, they are not considered to be part of the compo family.(Wetherall 26) Carton-pierre and papier-mâché, are also materials that require molds and glue for their fabrication. Both materials are of French origin. Papier-mâché was first patented by E F Bielefeld in 1847. Carton-pierre was accredited to M Hire around the same time as paper- maché but differed by reducing the paper to a pulp and adding plaster to the glue and resin. Carton-pierre has also included kaolin, flour and chalk whiting over the years in its many recipe variations. Carton-pierre and papier-mâché are often used interchangeably with compo, but are not the same material because of the use of paper. Compo material contains no paper material.

The origins of compo ornamentation can be traced back to variations of the material that was decorative plaster used in ancient Egyptian ornaments. The Egyptians used a variation of compo to decorate their furniture and mummies.Romans decorated their tombs with a similar material. Theophilus, in the 12th century mentions the use of adding whiting to an oil resin.(Wetherall 26) The

writings of Cennino Cennini in The Craftsman’s Handbook from the 15th century describes ‘how you may

model on a wall’ with either varnish or wax mixed with flour (whiting). (Cennini, chaps.CXXVIIII, CXXX)

Jonathan Thornton correlates the direct origins of compo material, or press-mold decorations, to Italian Renaissance which used a mixture of organic binders. These pressed-mold ornaments where not used for architectural design, like at Kasteel de Haar, but for decorating painted sculptures, wooden boxes, and picture frames. (Figure 3-1) The use of compo ornamentations was described by many Renaissance writers with varying recipes including ingredients of gypsum, lead carbonate, wood, marble dust, eggs, pigments, wool, oils and resins. The material is also related to the Louis XIV period in France where is was used as a substitute for wood carvings in picture frames, mirrors, and on some buildings. Because the mixtures were based in organic binders, the ornaments were prone to

(24)

Figure 3-1 Casket with grotesque ornament and Medusa heads, 1500-10, Italy, wood with glass and molded pastiglia decoration, 120 x 200 x 12 cm, Unknown, Victoria & Albert Museum, London, England. Gilt pastiglia

boxes were mostly made in Venice and Ferrara from about 1480 until 1550. Pastiglia or pasta is the name given to lead white paste , bound with egg white.(“Casket | V&A Search the Collections”)

Compo, as referred to in this thesis, is believed to have been first made by a Swiss clergyman, Mr. Liadort. (Millar and Bankart 397) In England, compo was a closely guarded secret of Italian workmen, but was revealed to John Jackson, the grandfather of the firm George Jackson & Sons. Robert Adam, architect, is accredited with introducing compo material to England and it can be found in almost all of his houses across, England, Edinburgh, Glasgow, Dublin and others in the United Kingdom. Robert Adam used the compo material to adorn ceilings, walls, chimney-pieces, doors, etc. in his interiors. (397)

(25)

Figure 3-2 Compo molds, 'compo' inside negative, 19th century, hand carved fruitwood, London, England, W.28-1989, from George Jackson and Sons company donated to the museum in 1988. © Victoria and Albert Museum, London

Figure 3-3- Compo molds, 'compo' negative and positive production, 19th century, hand carved fruitwood, London, England, W.28-1989, from George Jackson and Sons company donated to the museum in 1988. © Victoria and Albert Museum, London

Research on site at George Jackson and Sons confirmed that the recipe for the compo at Kasteel de Haar has remained the same since the inception of the company. It was also confirmed that the recipe is very similar to the one given in the 1899 publication by William Millar, Plastering, Plain and

Decorative. This recipe will be explored and provided in the following section. Unfortunately, the

company no longer has access to archival information about passed projects from the beginning of the 20th century, consequently confirmation of the company having produced the ornaments at Kasteel de

Haar could not be found. It is still hypothesized however, as described earlier, that the compo at Kasteel de Haar was produced and installed by George Jackson and Sons. Based on the interview at George Jackson and Sons studio, the archival information from Het Nieuwe Instituut (the state archive for Dutch architecture and urban planning) and the recipe provided in the book by W. Millar, a fair assumption can be made of the materials used in the compo at Kasteel de Haar. The following two sections will discuss the material configuration of the compo material and the physical characteristics that were expected for this material based on its material configuration.

(26)

Figure 3-4 Estimate from Maple & CO. LTD, London, Het Nieuwe Instituut, HAAR.110366093,details of work commissioned in bedrooms, including materials and costs, 1903, photo by author, 2018.

3.2 History of compo at Kasteel de Haar

The recipe given in the technical book by William Millar, Plain Plastering and Decorative, for London compo was confirmed to be very similar to that used by George Jackson and Sons during the studio visit with the director David Serra and an experienced compo maker. The recipe is as follows:

The following quantities will make a 16lb. batch. This is an old recipe, and is most likely used in London shops at the time. “Dissolve 16lbs. of town glue and 5 pints of water in one pot, and dissolve 9lbs. of ground resin and 31/2 pints of linseed oil in another pot. When both are dissolved, pour the glue into the resin-pot, and stir well with a stout stock. Sprinkle in sifted whiting until the consistency of thin dough, and then turn the mass out on a slab, and knead it well, adding whiting as

(27)

The recipe provided in the article The History and Techniques of Composition by Judith Wetherall for Adam compo suggests the use of colophony as the resin, which is boiled in the oil. The article also provides a very detailed procedure of how the compo was made.

Although compo material will be identified by material analysis, the exact ratio and types of protein can be difficult to establish and production procedure can therefore be ambiguous. Therefore, thorough archival research is necessary to determine the most accurate assumption of what material are present and to what proportion.

3.3 Characteristic properties

Compo is a thermo-setting material, pressed into molds to create a sculptural relief. Thermo-setting materials are characteristically soft or ‘plastic’ with heat and harden when cooled. (Thornton et al. 405) The molds used to make the ornaments were traditionally made of either wood, metal or plaster.(Wetherall 26) Compo is mostly known in woodworking and furniture making, but was used as architectural detailing, predominately in England, during the 18th century. It was first made popular by

Robert Adam in the 1760s as the main exponent of his neoclassical style.

The compo ornaments at Kasteel de Haar are made of two thermo-setting components, resin and animal glue (Serra). Understanding the physical and chemical characteristics of the material can significantly helped to understand its degradation process and consequently, the environmental factors in rooms 129 and 132 that catalysed the process. To an amount, the degradation of compo ornaments is directly correlated to the ingredients used and their respective ratios.(Thornton et al. 8) Further research with GC-MS (Gas chromatography–mass spectrometry) and SEM-EDX (Image Scanning electron microscopy with energy dispersive X-ray spectroscopy) of cross-sections will also help understand the exact ingredients in the compo at Kasteel de Haar and at what ratios they were used. The most influential factor is dependent on the interior climate condition which cause the most degradation in long-term exposure to heat and dampness of both the compo and the substrate.(8)

One of the major characteristics of the decomposition of a mixture composed of a rein and an animal glue, is its tendency to shrink. The resin and glue component oxidize and cross-link over time, causing the material to become more brittle. (Koob, sec.Animal Hide or Skin Glues) This type of material, aka compo, is also water sensitive.

The Article by David W. von Endt and Mary T. Baker, The Chemistry of Filled Animal Glue

Systems, provides an in-depth summary of the chemical structure of a filled animal glue and the physical

characteristics of animal glue that changes with the addition of a filler (whiting). Both authors are chemist for the Conservation Analytical Laboratory at the Smithsonian Institute in Washington, D.C.. This article is sued to compare the animal based material of gesso to compo. Gesso is composed of two main ingredients, an animal glue binder and calcium sulfate or carbonate as a filler. Comparatively, compo is also mainly composed of an animal glue binder and a calcium filler, but has the addition of oil and a natural resin.(Thornton, “Early Practice and Materials of Gap-Filling in the West,” sec. Gesso, Composition)

Analysing the chemical and physical properties of protein glues and the effects of fillers on these properties, helps to understand the degradation process of these materials. This understanding provides recommendations for its treatment and conservation. Hide glues are composed of proteins which contain polar and ionisable chemical groups. These chemical groups interconnect to form networks which have similar properties to cross-linked polymers as shown in figures 3-5 and 3-6

(28)

Figure 3-5 The hydrogen bonding between protein chains, Hydrogen in a bonded state, rendering by Jasmine Guest, 2018.

Figure 3-6 The hydrogen bonding between protein chains, Hydrogen bonds broken by either water or heat, rendering by Jasmine Guest, 2018.

Figure 3-7- The hydrogen bonding between protein chains, Hydrogen bonds reformed with two less hydrogen bonds, rendering by Jasmine Guest, 2018.

It is the various combinations of amino acids that distinguish the properties of the variety of proteins. The hydrogen bonding, although weaker than the covalent bonds, allows the glues to interact with certain substrates. The ability to interact with substrates determines whether or not it is a

7The bonding found in collagen is technically not cross-linking because it is a hydrogen bond, but the properties are similar to those of a cross-linked polymer.

(29)

powerful adhesive or not. As seen in Figure 3-8, when the collagen is heated in water, the hydrogen bonds break, creating single coils to make a solution. When the solution begins to cool, the collagen chains reform to a 3 coil structure again, but can line up only in sections, changing the bonding structure. This causes a pattern or network of the coils, held together by hydrogen bonds, also known as gelatin.

Figure 3-8- Gelatin formation by the heating and cooling of collagen, rendering by Jasmine Guest, 2018. The coiled structure created by the hydrogen bonds are the cause for the ‘spring’ or flexible nature of the animal glues. The hydrogen bonding, as stated before, can be altered with the addition of water molecules. The presence water molecules will weaken the network because it will bond where the protein chains are linked together. This network weakened by water molecule can then more easily be pulled apart causing an expansion. When the water evaporates, and the structure begins to form again, there is shrinkage in the bulk material. (von Endt and Baker, sec.collagen structure)

The moisture or heat that breaks the cross-linking of the hydrogen bonds are the predominant characteristics of the material, but this is changed with the introduction of a filler. The addition of a filler changes the reformation of the hydrogen bonding altering the strength and elasticity of the animal glues.(sec. filled animal glues) The addition of a filler will decrease the mobility of the protein chains, therefore increasing the tensile modulus of the material.(sec. Tensile Modulus)8 The addition of

filler however will decrease the tensile strength of the material because calcium is in the filler which is non-reinforcing.9 This is why the protein network fails more easily when under stress and the adhesion

properties decreases when a filler is introduced.

In a cross-linked polymer, when the filler has been filled beyond the maximum of the distance between the hydrogen bonds, there is an extreme decrease in both the tensile modulus and tensile strength of the material. Although the tensile strength is decrease, the addition of filler will increase the impact strength of the filled animal glue.(sec.Tensile Strength)

The addition of a filler also gives more water resistance to the animal glue. More specifically, the spherical filler particles of the calcium in the network of the collagen chains of the animal glue

decrease the permeability of water into the material. This concept in demonstrated in Figure 3-9, where the reaction of a filled material has been schematically illustrated.

(30)

Figure 3-9- Spherical filler molecules in the collagen network and the change in reaction to stress,rendering by Jasmine Guest, 2018.

The failure in gesso is subject to ‘stress softening’ which is caused by fluctuations in humidity. Stress softening occurs when the chains in the glue network have been irreversibly bonded out of position. This will cause the animal glue filled material to deform by curling and shrinking. Therefore, when a fluctuation in humidity is high, creates a softening in the material. With each softening, the material becomes easier to deform, causing degradation of the material. (von Endt and Baker, sec.Moisture Effects) This theory applied to gesso, is assumed to be similar to that of compo, since they are both composed of mostly animal glue and filler. The role of the oil and resin ingredients also have an effect on the reaction to stress. Further research and testing is required to understand the effects of oil and resin in the compo.

3.4 Conclusion

To further determine which of these factors is most relevant in the case study of the two rooms at Kasteel de Haar, analysis will be done on the exact material configuration of the compo, including a hygroscopicity test and a temperature stress test. Determining the material configuration will render a more scientific hypothesis on the causing factors of its degradation. As stated by many sources, compo material is hygroscopic and thermos-setting, making it sensible to fluctuation in temperature and relative humidity. (Rivers and Umney 776; Wetherall 27; Thornton et al. 9; Thornton, “Early Practice and Materials of Gap-Filling in the West,” sec. Gesso) To expand on this knowledge, particular attention to climate conditions and its correlation to causes in degradation will be presented. The hygroscopicity test will grant an understanding of the hysteresis of the material as a response to humidity. The temperature stress test will present a rough estimate of the Tg temperature10 of the

compo material.

10 “The glass transition temperature (Tg) of a polymer is important […] as an indication of the softness of the polymer. […] As the temperature is raised, the polymer changes from glassy state to a liquid one. This change is

Unstressed Stressed

(31)

4 Scientific research and analysis

This chapter will focus on the scientific analysis conducted on each material which compose the ceilings in rooms 129 and 132 at Kasteel de Haar. Analysis of the results will enable a more comprehensive understanding of the sensitivity of the material and consequently determine

considerations for its restoration and conversation. The scientific analysis is organized by each of the layers comprising the decorated ceilings. Each section will answer the question of what exact material is in each layer.

Current literature provides little information regarding scientific research conducted on compo and on its sensitivity to environmental conditions. The methods for the optical analysis identification of the materials included; visual inspection, on-location microscopy with a Dinolite and microscopy of a cross-section. The scientific analysis consisted of THM-Py-GC/MS (Gas chromatography–mass spectrometry), and SEM-EDX (Image Scanning electron microscopy with energy dispersive X-ray spectroscopy). The question of environmental factors contributing to the degradation of the ceiling can be answered by the results from these analytical tests and study of the composition of the material buildup.

4.1 Fibrous plaster

Fibrous plaster is referred to in the archive estimate as Keen’s cement. The patent dated 1838 of Keen’s cement by John Danforth and Richard Wynn Keen describes how to make and colour it. The patent states that Keen’s cement is made of gypsum steeped in alum, heated to very intense

temperatures, ground to a powder and sifted.(Keen’s Cement 230) The ground gypsum is again heated to a point where nearly all the crystallized water is gone. (Buza, chap.7) After the water of crystallisation has been removed, the gypsum is dissolved in water. For coloured cement, as observed at Kasteel de Haar, the patent instructs “[…] dissolve half a pound of copperas or sulphate of iron, in each galon of water […] the result will be a pale red cement.” The material has a very hard and smooth finish which can be shined. The material found on the ceiling support in both rooms 129 and 132 is also very hard and smooth, with a pale red colour and a shiny appearance. (Keen’s Cement 230, 231) (Figure 4-1)

Further research should be conducted to determine if the production method of Keen’s cement from the 1838 patent is comparable in composition to that of the Keen’s cement found at Kasteel de Haar.

(32)

In Figure 4-1, we see a lacuna in the ceiling, where the paint and paper layers have peeled away from the fibrous boards, which reveals the pink material. Figure 4-2 and Figure 4-3, show a magnified image of a small damage in the wall in room 132 where the layer buildup to the pink support can be seen under reflected light and UV (ultraviolet light). Under reflected light the fibrous plaster has a spotted pattern, with large flecks of concentrated material. Under UV the large white spots fluoresce a light green-yellow colour. Other large flakes fluoresce as a light violet. The light violet fluorescing could be indicative of calcium carbonate. (Danielle Measday, sec.Pigments, Inks and Dyes) The remainder of the pink support does not fluoresce.

SEM-EDX analysis was completed to determine the elemental configuration of the plaster layer.11 A sample (KDH.132.02) of the fibrous plaster material from room 132 was embedded in

acrylic resin (Polypol) to achieve a cross-section of the layer buildup. This sample was taken from the lacuna in the wall support included in the pink support material and paint layers. It was not possible to get a sample on the ceiling closer to the compo ornaments, because of the hardness of the pink support and the brittleness of the paint. It is assumed that the pink support is the same on the walls as is on the ceiling, because they exhibit the same colour, surface texture, and hardness. Figure 4-3 depicts large flecks of material which were later examined using SEM-EDX. The back scattering image produced by SEM-EDX (BSE image) (Figure 4-5) renders different shades based on the heaviness of the element. Heavier elements appear whiter and lighter elements appear darker. (Stuart 52) In the

11 The full SEM-EDX along with conclusions by the RCE can be found in Appendix VII.

Figure 4-2 Small damage in wall, room 132, Kasteel de Haar, Dinolite image, reflected light, 50x, photo by author.

Figure 4-3 Small damage in wall, room 132, Kasteel de Haar, Dinolite image, 50x, UV, photo by author.

(33)

lighter layers 2, 3, 4 and 5, it was concluded that the layers were mainly composed of a heavier element of lead, therefore indicating the presence of a lead based paint. In the dark layer 1, it was concluded that the material is mostly calcium, the main consistent of Keen’s cement. The 6th layer will be

described in section 4.4.

Figure 4-4 KDH.132.02, cross-section, licra microscope, 50x magnification, refracted light, photo by author.

Figure 4-5 BSE image, KDH.132.02, Kasteel de Haar, 2018, BSE by Luc Megens (RCE).

6 5 4 3 2 1 6 5 4 3 2 1

(34)

Figure 4-6 SEM-EDX area spectrum of the support layer in sample KDH.132.02. SEM-EDX report done by Luc Megens (RCE), 2018.

Figure 4-7 SEM-EDX spectrum of bright red material in support in sample KDH.132.02. SEM-EDX report done by Luc Megens (RCE), 2018.

The elemental composition of the layer 1 in figure 4-4 and 4-5 can be found in the spectra in figures 4-6 and 4-7. The spectra indicates the presence of iron, magnesium, aluminum, silicone, sulfur and potassium. (Megens)12 As a result of analysis combined with the archival information, we can

conclude that the fibrous pink support is most likely Keen’s cement.

4.2 Paper

Based on archival data mentioned in Chapter 3 Current Scientific Knowledge, the paper layer is described as lining paper, which is known to be a cheaper quality paper. Based on the book

Papermaking: the history and technique of an ancient craft by Dard Hunter, the paper used to line the ceiling

for these compo ornaments is most likely machine made wood pulp paper. This paper might also be sensitive the acidic hydrolysis making it a risk factor to the degradation of the compo. (Magdalena Marosz 105) The paper was produced in infinite lengths and was used for newspapers, magazines, etc.

12 Aluminum and sulfur are consistent with the archival description of aluminum sulphate as an accelerator in the production process of Keen’s cement.

(35)

(Hunter 361) which suggests it is of an inferior quality. The use of this poor quality paper most likely is a factor in the degradation of the ceiling ornamentation in rooms 129 and 132 at Kasteel de Haar. The type of pulp and production method will be further determined with fiber analysis of the paper.

4.2.1 Fiber Analysis of the paper component

Fiber analysis of the paper lining was completed with the help of Bas van Velzen, paper conservator and professor of book and paper conservation at the University of Amsterdam. Mr. van Velzen is currently in the final steps of his PhD research on fibre analysis using microscopy of paper fibres under polarized light. His research methodology on the determination of the physical

characteristics of paper with microscopy allows for the determination of the paper's main fiber source and production procedure. With the knowledge of the type and production procedure of the paper, more specific characteristics can be drawn with regards to its degradation and influences by the environment.

The fiber analysis was executed by the removal of a sample of paper, about 3mm2 in size from

an ornament from room 129. It is assumed that the paper is the same in both rooms. One sample was analysed from room 129 assuming the compo material and the other materials of the ceilings are the same in both rooms. The paper sample was positioned on a clean slide in a 1:1 ratio in a water/ethanol solution. The sully saturated sample was then slowly pulled apart dispersing the fibers for proper analysis. A slide cover was then applied and the sample was analysed under polarized light. The fibers were identified and a z-stack image was made of a selected area, which housed three identifying fiber types. Another two z-stack images were taken as other reference images to further analyzed the paper. All images are identified with a title, magnification, polarization angle, production procedure and fiber type as depicted in figures 4-8, 4-9 and 4-10.

(36)

Figure 4-9 Paper support, from sample KDH.129.07, 200x xpol, image by Bas van Velzen, 2018.

Figure 4-10 Fungal hyphen, from sample KDH.129.07, 200x xpol, image by Bas van Velzen, 2018.

4.2.2 Results

The fiber analysis of visual examination at a high magnitude resulted in the confirmation of the type of fiber and other contaminates in the paper by surrounding materials. Three types of cell

(37)

structures are present in the paper fiber confirming that is a machine pulp esparto paper. The three different fibers characteristic to esparto are identified in Figure 4-8 as follows: 1-long thin blue bent cell, 2- saw edge and 3-comma blue and orange shaped. (Bas van Velzen) Only the distinctive fibers esparto paper fibers were detected. The black scattered material, seen in Figure 4-9, are minerals from contamination of the compo ornaments and the fibrous plaster. The dark scattered spots are typically minerals in the sample as contaminants from external sources. In Figure 4-10, there is a fungal hyphen which is a pollution contamination. The presence of a fungus confirms there is mould in the room which in turn is related to elevated humidity. Therefore, a high humidity level must have been present in the room for a fungus to have to have grown. (National Trust 82)

The fiber analysis clearly identified esparto as the fiber in the ceiling lining paper based on the distinctive cells found in the samples from rooms 129 and 132 of Kasteel de Haar.

4.3 Compo ornaments

The following analysis is based on the research and historical context outlined in Chapter 3

Current Scientific knowledge of compo ornamentation.

Based on the London compo recipe in William Millar’s Plain Plastering and Decorative instructive book the compo in rooms 129 and 132 are hypothesized to be composed of a lower grade animal glue, chalk, colophony resin and linseed oil.

4.3.1 Microscopy, SEM-EDX and THM-Py-GC/MS of compo

Optical analysis is simply the visual observation of the material and is done to obtain a

preliminary state of the degradation process. As previously stated, the first analysis without a visual aid allowed for an initial assessment of the state of degradation of the rooms and of the material buildup of the ceilings. Section 1.3 Current State of Degradation described cracking, shrinking and delamination of the compo ornaments, paper and paint layers. The description of the extent of degradation also included the presence of mould across th entire ceiling within the small cracks in the ornaments. Optical analysis of the compo material consists of documentation of the rooms using photography damage mapping and proper assessment of the current condition. The microscopic analysis was done on location using a Dino Lite and a microscope. A visual understanding of the cracks and mould was assessed with the Dinolite. The compo material and paint layers of cross-section from both rooms were subsequently analyzed with a microscope under reflected light and UV. Further analysis of the compo material was done using SEM-EDX to determine the elemental compo of the material, and to verify its consistency in both rooms and in different ornaments. THM-Py-GC/MS was also used to deduce information of the compo’s binding media, resin and oil, and to compare the results to historical recipes.

4.3.2 Results

The Dinolite ultraviolet light setting shows a blue-green fluorescence in the paint layer and minimal grey fluorescence in the compo material.

(38)

Figure 4-13 and

Figure 4-14 depict a magnified image of a lacunae in the ‘compo’ ornament in room 129. The lacunae in the ornament reveals the multiple layers of the paint and compo. The top paint layer fluoresces a muted green-yellow colour, while the under-colour fluoresced a non-muted colour. The top layer of the compo is fluorescing yellow while the lower compo layer is fluorescing purple. Figure 4-14 has a muted fluorescence, however it should be considered as having the same florescence as that depicted in Figure 4-12 as both are the same top paint layers. This difference in florescence is due to the amount of surface dirt. Based on the data presented in A Summary of Ultra-Violet fluorescent

Materials relevant to Conservation by Danielle Measday from the Australian Institute for the Conservation

of Cultural Material, he top white paint layer, which fluoresced as green-yellow, is composed of lead. If the white paint had been a modern acrylic, it would not have fluoresced. If it was a zinc white, it would have fluoresced bright blue with star shaped reflective particles. In

Figure 4-14, the compo material, fluorescing purple is concluded to be composed of calcium carbonate13 as the main component. (Danielle Measday) The top layer of the compo, fluorescing

yellow-green, is most likely an oil based. (Johnson, pt.II) Further analysis with incident and ultra-violet light will conclude the microscopic analysis.

(39)

Figure 4-11 KDH.129.DL.01, Dino Lite Image, 50x, photo

taken by author. Figure 4-12 KDH.129.DL.01.UV, Dino Lite Image, 50x, 356 nm, photo taken by author.

Figure 4-13 KDH.129.DL.03, Dino Lite Image, 50x, photo

taken by author. Figure 4-14 KDH.129.DL.03.UV, Dino Lite Image, 50x, 365 nm, photo taken by author. According to the work of Mary Lou Florian, an expert on biomaterial degradation of cultural material and past chief conservator at the Royal British Columbia Museum, the fluorescence of mould increases and decreases according to the life cycle of fungi. Therefore the non-fluorescing mould should be further analyzed to determine if it is active or not. (Florian et al. 185)

The compo area spectra from the SEM-EDX analysis showed the presence of carbon, oxygen and calcium. This is what was expected based on the historical recipes of the compo as outlined in chapter 2. The calcium was also, as predicted, present in large quantities. The dense material and apparent texture of the compo suggest a very homogeneous mixture. (Megens)

The THM-Py-GC/MS testing on the compo ornamentation was to determine what animal glue, oil and resin was used in its production. It is expected that the results of that test combined with recipes found in historical literature will determine the ratio of the components in the compo. During the on-site research of the studio of George Jackson and Sons, it was noted that their ‘secret’ recipe was similar to that of W. Millar, but with a couple of minor changes. THM-Py-GC/MS testing is the method chosen to determine these changes. THM-Py-GC/MS testing was conducted on two compo samples, KDH.129.04 (04) and KDH.129.05 (05). Sample 04 was from a larger thicker ornament and sample 05 was from a smaller, thinner ornament. These two samples were chosen to see if there is a difference in recipe between the two types of compo. Both samples were chosen from room 129,

(40)

4.4 Paint

The paint layers in rooms 129 and 132 at Kasteel de Haar could be a relevant factor in the degradation process of the compo ornaments. The multiple paint layers could have affected the hysteresis of the compo material and could have attributed to other degradation phenomena.

Based on the archival research outlined in Chapter 2 History and current condition, the paint should consist of three consecutive layers of white oil paint with a matte finish. Visual analysis on-location, revealed that some discrepancies were present in the ceiling paint finishes. In room 129, large

ornaments of the urns and the putti roundels were a lighter white colour than the rest of the ceiling. It is therefore hypothesized that the larger white ornaments in room 129 have had a more recent layer of paint added whereas the rest of the ceiling in 129 and the ceiling in 132 should have the expected three consecutive layers of white oil paint because there is no evidence if new paint.

4.4.1 Microscopy and SEM-EDX of paint samples

Five paint layers of paint layers were analysed from both rooms. These include one thinner more delicate compo ornamentation, one thicker and larger ornamentation, and one from the walls. The paint was first analysed with microscopy using dark field and UV lighting. The samples were then analysed using SEM-EDX to further confirm the presence of pigments and bulking materials in the paint.

4.4.2 Results

Layers 2, 3 and 4 of sample KDH.132.02, Figure 4-5, are significantly lighter, and therefore have heavier elements, while the paint layers 5 and 6 are much darker indicating light elements. The SEM-EDX results show a presence of lead in layers 2, 3 and 4. The darker paint layer (layer 3) contained a much higher calcium concentration and minimal lead, suggesting that it was a ground layer for the subsequent lead white paint layer 4. In the large uneven paint layer 5, the SEM-EDX showed the presence of calcium, lead, silicon, aluminum, barium, sulfur and zinc. This was hypothesized as a ground layer for the new modern paint, layer 6.(Megens) The top layer, due to its visual appearance and lack of fluorescence under UV is a more modern paint. The SEM-EDX analysis showed that the layer contains calcium and titanium. Titanium white was not widely commercially available until the 1920s (Feller 298) and is therefore a modern addition to the room.

4.5 Discussion

The following section will conclude all of the findings of each material composing the structure of the decorated ceiling of rooms 129 and 132 at Kasteel de Haar. Conclusions have been drawn with regards to the material configuration of each layer and the environmental influences on those layers.

14 See appendix VI for the complete THM-Py-GC/MS report.

Referenties

GERELATEERDE DOCUMENTEN

Also, all base colors are representative of the palettes’ primary color and are available at the 500 variation or using their color names, e.g., MaterialBlue or MaterialBlue500 are

To conclude, this research study examines process, IT, a forecasting model, and organization and people perspectives of forecasting to determine the factors of influence on

Staff turnover rate at Gracesemi in 2007 is 66.7%; totally 6 employees left SA unit and the average total number employed during 2007 is 9. Literature searching shows that there is

The comparison of four different reference houses, consisting of a row house built before 1946, between 1946-1965, 1976-1979 and 1989-2000, shows the influence of the dependent

In order to answer hypothesis 2, a higher number of choice sets leads to differences in attribute importance (price becomes more important than brand), and

Comparing percentages of cases where duration of the interval is above the median compared to cases with duration below the median, one can assume contact with supervisor

Therefore this thesis aims to answer the following research question: “To what extent is automation of the material handling process at Company X feasible?” The goal

Patients of multiple surgery specialties, with different urgency classes should be booked onto sessions of their perform- ing surgeon, taking into account a combination of