Full text of "Plan for the stabilization and removal of wall paintings at Çatalhöyük"

ii»

mximmmf

HHKF

■■-■■■■■■■''

HHHElHiHi

hrh

liiii
1

■I

BUHHIftM

EBBS

wMmKa

,-.■:.■■■■'■?:

H8(
H

ililil

liHlIll

Iwrafl

■ ■• ■-■■■■ ■

UNIVERSITY^

PENNSYLVANIA.

LIBRARIES

PLAN FOR THE STABILIZATION AND REMOVAL OF WALL
PAINTINGS AT CATALHOYUK

Catherine E. Turton

A THESIS

Historic Preservation

Presented to the Faculties of the University of Pennsylvania in
Partial Fulfillment of the Requirements for the Degree of

MASTER OF SCIENCE
1998

Mkuu

Supervisor

Frank\}. Matero

Associate Professor of Architecture

Reader

Catherine S. Myers

Principal, Myers Conservation

Graduate Group Chair

Frank G.NyI aIero

Associate Professor of Architecture

. ...

Acknowledgements

I would like to thank my advisor, Professor Frank G. Matero, whose scholarship
and dedication to the project were a source of inspiration. I also wish to thank Catherine
S. Myers. Her generous contributions of time, support and direction were invaluable to
me in this research.

Thanks to Dr. Rollin Lakis and Xue-Qin Wang for the time and energy they
invested training us in the use of the scanning electron microscope. I am especially
grateful to Michael Dorsch of Colloid Environmental Technologies and the Black Hills
Bentonite Company for their timely responses in supplying the clays for this project. I
would also like to thank ProSoCo, Inc. for their generosity in supplying the consolidant,
T-1919 Conservare® Consolidation Treatment

I want very much to thank my parents for their love and encouragement, and
especially their patience.

Finally, a very special thanks to Felix, whose unfailing support, kindness, serenity,
and understanding have given me a tremendous sense of peace. His lighthearted spirit
and incredible sense of merriment saw me through even the most trying phases of
research.

CONTENTS

1.0 Chapter 1: Introduction 1

1 . 1 Statement of purpose 1

1 .2 Current research 3

1 .3 Catalhoytik 5
2.0 Chapter 2: Sample Development 30

2.1 Existing Conditions 31

2.2 Characterization of Original Material 33

2.2.1 Analysis of Stratigraphies 36

2.2.2 Media Characterization 42

2.2.3 Pigment Identification 46

2.2.3.1 Microchemical spot tests 46

2.2.3.2 Polarized Light Microscopy 49

2.2.3.3 Scanning Electron Microscopy With

Energy Dispersive Spectroscopy (SEM/EDS) 54

2.3 Creation of sample prototype 58
2.3. 1 Materials for Laboratory Facsimiles 58

2.3.1.1 Formation and Characteristics of Clays 61

2.4 Production of Samples 66
3.0 Chapter 3: Testing Program 74

3.1 Introduction 75

3.1.1 Detachment techniques 77

3.1.2 Requirements of Research 78

3.1.3 Materials: Adhesives and Consolidants 80

3.2 Preliminary Testing A: Individual tests 85

3.2. 1 Test for Visible Alteration of the Plaster Surface 86

3.2.2 Test for Surface Consolidation 88

3.2.3 Surface Consolidation and Preconsolidation with Readhesion,

Part I 90

3.2.4 Surface Consolidation and Preconsolidation with Readhesion,

Part n 93

3.2.5 Test for Readhesion / Facing Adhesives 96

3.2.6 Surface Consolidation of Powdering Paint with Consolidation

and Readhesion 99

3.2.7 Facing Adhesives: Methods and Materials:

Preparation for Stacco and Strappo 102

3.2.8 Reattachment to a New Support / Backing 110

3.2.9 Consolidation of Disaggregating Mudbrick with Ethyl Silicates 1 17

3.3 Evaluation of Preliminary Testing A 122

3.3.1 ASTM D4214: Evaluating the Degree of Chalking of Exterior
Paint Films 123

3.3.2 ASTM D3359-90: Measuring Adhesion by Tape Test 127

3.3.3 Determining Depth of Penetration of Consolidants: Iodine

Vapor Test 132

3.3.4 Water Drop Test: CRATerre 135

3.4 Preliminary Testing B: Compatibility of Treatments 141

3.5 Final Testing Program 152
3.5.1 Methodology 153

3.6 Evaluation of Final Testing Program 160

3.6. 1 ASTM D66 1 -86: Evaluating Degree of Cracking of

Exterior Paints 162

3.6.2 ASTM D772-86: Evaluating Degree of Flaking (Scaling)

of Exterior Paints 1 63

3.6.3 ASTM D660: Evaluating the Degree of Checking of

Exterior Paints 164

3.6.4 ASTM D 1535-80: Specifying Color by the Munsell System 166
4.0 Chapter 4: Conclusions 182

4.1 Final Results 182

4.1.1 Surface consolidation 183

4.1.2 Readhesion 184

4.1.3 Consolidation 184

4. 1 .4 Facing adhesives for mural detachment 1 85

4.1.5 Detachment methods 185

4. 1 .6 Reattachment to a new support 1 86

4.1.7 Compatibility of treatments 186

4.1.8 Final Testing Program 187

4. 1 .9 Detachment of consolidated samples 1 88

4.2 Final Results: Data 188

4.3 Conclusions 219

4.4 Recommendations for Further Research 220
Appendices

Appendix A: Catalhoyuk Treatment History 222

Appendix B: Overview of Detachment Techniques 229

Bibliography 258

Index 275

List of Figures

Figure 1: General Map of Turkey. From Mellaart, James, Udo Hirsch,

and Belkis Balpinar. The Goddess from Anatolia. West Germany:

Udo Hirsch, 1989. 26

Figure 2: James Mellaart working on wall paintings. From

Mellaart, James, Udo Hirsch, and Belkis Balpinar. The Goddess

from Anatolia. West Germany: Udo Hirsch, 1989. 27

Figure 3: Bull painting (Shrine F.V/1). From Mellaart, James, Udo Hirsch,

and Belkis Balpinar. The Goddess from Anatolia. West Germany: Udo

Hirsch, 1989. 27

Figure 4: Wall painting with hand prints (Shrine E. VII/8). From Mellaart,

James, Udo Hirsch, and Belkis Balpinar. The Goddess from Anatolia.

West Germany: Udo Hirsch, 1989. 28

Figure 5: Scale copy of wall painting (Shrine E. VIA/50). From Mellaart,

James, Udo Hirsch, and Belkis Balpinar. The Goddess from Anatolia.

West Germany: Udo Hirsch, 1989. 28

Figure 6: Wall painting (Shrine E. VII/21). From Mellaart, James, Udo Hirsch,
and Belkis Balpinar. The Goddess from Anatolia. West Germany: Udo
Hirsch, 1989. 28

Figure 7: Painted relief of human figure (Shrine E. VII/21). From Mellaart,

James, Udo Hirsch, and Belkis Balpinar. The Goddess from Anatolia.

West Germany: Udo Hirsch, 1989. 29

Vll

Figure 8: Wall painting with hand prints (Shrine E. VIB/15). From Mellaart,

James, Udo Hirsch, and Belkis Balpinar. The Goddess from Anatolia.

West Germany: Udo Hirsch, 1989. 29

Figure 9: Wall painting with human figures (Shrine F. V/l). From Mellaart,

James, Udo Hirsch, and Belkis Balpinar. The Goddess from Anatolia.

West Germany: Udo Hirsch, 1989. 29

Figure 10: Photomicrograph of cross-section "Sheena Wall Black." (BL1) 41

Figure 11: Photomicrograph of cross-section "CH 95 Mell 1014 5/9 8F" (RD1) 41
Figure 12: Photomicrograph of cross-section "CH 95 Mell 1014 5/9 8F" (RL1) 41
Figure 13: Photomicrograph "CH 95 Mell 1014 5/9 8F" (RD1 ) pigment particles 53
Figure 14: Photomicrograph "CH 95 Mell 1014 5/9 8F" (RL1) pigment particles 53
Figure 15: Photomicrograph "Sheena Wall Black" (BL1) pigment particles 53

Figure 1 6: X-ray dot map of red paint layer (Sample RD 1 ) 56

Figure 17: Photomicrograph of Sample RD1 56

Figure 18: EDS spectrum of red paint layer (RD 1 ) 57

Figure 19: Acid soluble content: wall and relief plasters. From Kopelson,
Evan. "Analysis and Consolidation of Architectural Plasters from
Catalhoyiik, Turkey." Master's thesis, University of Pennsylvania,
1996. 70

Vlll

Figure 20: Particle size distribution : wall plaster. From Kopelson,

Evan. "Analysis and Consolidation of Architectural Plasters from
Catalhoyiik, Turkey." Master's thesis, University of Pennsylvania,

1996 71
Figure 21: Results of dry sieving. From Kopelson, Evan. "Analysis
and Consolidation of Architectural Plasters from Catalhoyiik,

Turkey." Master's thesis, University of Pennsylvania, 1996. 71.

Figure 22: Mural painting replica: finish layer 1 72

Figure 23: Mural painting replica: finish layer 2 72

Figure 24: Mural painting replica: finish layer 3 72

Figure 25: Mural painting replica: finish layer 4 73

Figure 26: Mural painting replica: finish layer 5 73

Figure 27: Mural painting replica: finish layer 6 73

Figure 28: Mural painting replica: finish layer 7 73

Figure 29: Surface consolidation test 89

Figure 30: Surface consolidation, preconsolidation and readhesion. Part I 92

Figure 31: Surface consolidation, preconsolidation and readhesion, Part II 95

Figure 32: Readhesion / Facing adhesives test 98

Figure 33: Surface consolidation, consolidation, and readhesion test 101

Figure 34: Surface consolidation, consolidation, and readhesion test 101

Figure 35: Surface consolidation, consolidation, and readhesion test 101

Figure 36: Facing adhesive tests 107

Figure 37: Drying colletta 107

Figure 38: Facing adhesive tests 107

Figure 39: Heating colletta 107

Figure 40: Sample with colletta facing 108

Figure 41: Sample with polyvinyl alcohol facing 108

Figure 42: Sample faced with Acryloid B-67 108

Figure 43: Sample faced with Acryloid B-72 108

Figure 44: Appearance after removal of colletta facing 109

Figure 45: Appearance after removal of polyvinyl alcohol facing 109

Figure 46: Appearance after removal of Acryloid B-67 facing 109

Figure 47: Appearance after removal of Acryloid B-72 facing 109

Figure 48: Sample E4-PVA foam core support 1 14

Figure 49: Sample E5-Plaster of Paris, PVA and foam core support 1 14

Figure 50: Sample E6-PVA and foam core support 1 14

Figure 5 1 : Sample E7-Traditional plaster support 1 14

Figure 52: Sample E8-Modified plaster support 1 15

Figure 53: Sample E9-Plasti-tak support 1 15

Figure 54: Sample El 1- Traditional plaster support 1 15

Figure 55: Sample ElO-Modified plaster support 1 15

Figure 56: Sample E12-Foam tape support 1 16

Figure 57: Sample E13-Traditional plaster support 1 16

Figure 58: Sample E15-Traditional plaster support 1 16

Figure 59: Sample E14- Failed plaster support 1 16

Figure 60: Chalking test (Method A): Untreated 125

Figure 61: Chalking test (Method A): Surface consolidated 125

Figure 62: Chalking test (Method A): Consolidated 125

Figure 63: Chalking test (Method A): Preconsolidated/Consolidated 125

Figure 64: Chalking test (Method B): Untreated 126

Figure 65: Chalking test (Method B): Surface consolidated 126

Figure 66: Chalking test (Method B): Consolidated 126

Figure 67: Chalking test (Method B): Preconsolidated/Consolidated 126

Figure 68: Measuring Adhesion by Tape Test (Method A): Untreated 130
Figure 69: Measuring Adhesion by Tape Test (Method A): Surface consolidated 130

Figure 70: Measuring Adhesion by Tape Test (Method A): Consolidated 130
Figure 71: Measuring Adhesion by Tape Test (Method A):

Preconsolidated/Consolidated 130

Figure 72: Measuring Adhesion by Tape Test (Method B): Untreated 131
Figure 73: Measuring Adhesion by Tape Test (Method B): Surface consolidated 131

Figure 74: Measuring Adhesion by Tape Test (Method B): Consolidated 131
Figure 75: Measuring Adhesion by Tape Test (Method A):

Preconsolidated/Consolidated 131

Figure 76: Iodine vapor test 134

Figure 77: Preparation for water drop test 137

Figure 78: Water drop test: 10 seconds 138

Figure 79: Water drop test: 2 hours 138

Figure 80: Water drop test: 10 seconds 139

Figure 8 1 : Water drop test: 2 hours 1 39

Figure 82: Water drop test final result: UT1 140

Figure 83: Water drop test final result: UT2 140

Figure 84: Water drop test final result: ES 1 140

Figure 85: Water drop test final result: ES2 140

Figure 86: Water drop test final result: AQES 1 140

Figure 87: Water drop test final result: AQES2 140

Figure 88: Preliminary test samples 146

Figure 89: Test materials 146

Figure 90: Secondary facing removal 146

Figure 91: Compatibility of treatments: D-l 147

Figure 92: Compatibility of treatments: D-2 147

Figure 93: Compatibility of treatments: D-3 147

Figure 94: Compatibility of treatments: D-4 148

Figure 95: Compatibility of treatments: D-5 148

Figure 96: Compatibility of treatments: D-6 148

Figure 97: Compatibility of treatments: D-7 149

Figure 98: Compatibility of treatments: D-8 149

Figure 99: Compatibility of treatments: D-9 149

Figure 100: Compatibility of treatments: D-10 150

Xll

Figure 101: Compatibility of treatments: D-12 150

Figure 102: Compatibility of treatments: D-17 150

Figure 103: Compatibility of treatments: D-18 151

Figure 104: Compatibility of treatments: D- 19 151

Figure 105: Compatibility of treatments: D-21 151

Figure 106: Prototype A: Untreated 168

Figure 107: Prototype A: After detachment 168

Figure 108: Prototype B: Untreated 168

Figure 109: Prototype B: After detachment 168

Figure 1 10: Prototype C: Untreated 169

Figure 111: Prototype C: After detachment 169

Figure 1 12: Prototype D: Untreated 169

Figure 113: Prototype D: After detachment 169

Figure 1 14: Prototype G: Untreated 170

Figure 115: Prototype Gj: After detachment 170

Figure 1 16: Prototype G 2 : After detachment 170

Figure 1 17: Prototype H: Untreated 170

Figure 118: Prototype H: After detachment 170

Figure 1 19: Prototype I: Untreated 171

Figure 120: Prototype I: After detachment 171

Figure 121: Prototype J: Untreated 1 7 1

Figure 122: Prototype J\\ After detachment 171

Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure

Figure

23: Prototype J2: After detachment

24: Substrate Sample A

25: Diagram A: after detachment

26: Substrate Sample B

27: Diagram B: After detachment

28: Substrate Sample C

29: Diagram C: after detachment

30: Substrate Sample D

31: Diagram D: After detachment

32: Substrate Sample Gi

33: Diagram G\: After detachment

34: Substrate Sample G2

35: Diagram Gt: After detachment

36: Substrate Sample H

37: Diagram H: After detachment

38: Substrate Sample I

39: Diagram I: After detachment

40: Substrate Sample Ji

41: Diagram J\\ After detachment

42: Substrate sample J 2

43: Diagram J2: After detachment

44: ASTM Standard: Checking

171
172
172
173
173
174
174
175
175
176
176
176
176
177
177
178
178
179
179
179
179
180

XIV

Figure 145: ASTM Standard: Cracking 180

Figure 146: ASTM Standard: Flaking 180

Figure 147: Sample overview 181

List of Tables

Table 1 : Description of Original Plaster Samples 35

Table 2: Cross Section Examination: "Ch 95 Mell 1014 5/9 8F" (RD1 ) 37

Table 3: Cross Section Examination: "Ch 95 Mell 1014 5/9 8F" (RL1 ) 38

Table 4: Cross Section Examination: "Sheena Wall Black" (BL1 ) 40

Table 5: Examination with Polarized Light Microscopy 50

Table 6: Visible Alteration of the Plaster Surface 189

Table 7: Surface Consolidation 190

Table 8: Surface Consolidation and Preconsolidation with Readhesion, Part I 191

Table 9: Surface Consolidation and Preconsolidation with Readhesion, Part H 192

Table 10: Readhesion / Facing Adhesives 193

Table 1 1: Readhesion / Facing Adhesives 194
Table 12: Surface Consolidation of Powdering Paint with Consolidation

and Readhesion 195
Table 13: Surface Consolidation of Powdering Paint with Consolidation

and Readhesion 196
Table 14: Surface Consolidation of Powdering Paint with Consolidation

and Readhesion 197
Table 15: Facing Adhesives: Methods and Materials, Preparation for

Stucco and Strappo 198
Table 16: Facing Adhesives: Methods and Materials, Preparation for

XVI

Stucco and Strappo 1 99

Table 17: New Support / Backing: Traditional Materials 200

Table 18: New Support / Backing: Non-Traditional Materials 201

Table 19: ASTM D42 14-89: Chalking Test Results 202

Table 20: ASTM D3359-90: Measuring Adhesion by Tape Test (Method A) 203

Table 2 1 : ASTM D3359-90: Measuring Adhesion by Tape Test (Method B) 204

Table 22: Water Resistance Data 205

Table 23: Water Resistance Data 205

Table 24 Water Resistance Data 206

Table 25: Water Resistance Data 206

Table 26: Water Resistance Data 207

Table 27: Water Resistance Data 207

Table 28: Compatibility of Treatments 208

Table 29: Compatibility of Treatments 209

Table 30: Compatibility of Treatments 210

Table 3 1 : Compatibility of Treatments 21 1

Table 32: Result of Final Detachment Tests 212

Table 33: Result of Final Detachment Tests 213
Table 34: % Distribution of Each Layer Remaining on Substrate

After Detachment 2 1 4

Table 35: Results of ASTM Cracking, Flaking and Checking Tests 215

XVII

Table 36: Results of ASTM Cracking, Flaking and Checking Tests 216

Table 37: Munsell Color Notations: Before and After Treatment 217

Table 38: Munsell Color Notations: Interpretation 218

XVlll

Chapter 1 : Introduction

1.1 Statement of Purpose

This research aims to develop methods for the emergency stabilization and
removal of the wall paintings on earthen plaster at Catalhoyuk, a Neolithic settlement
located on the Konya Plain in south central Turkey. The site, first excavated in the 1960s
by James Mellaart, represents one of the earliest known examples of a society in
transition from an economy based on hunting and gathering to one based on the
domestication of plants and animals. Neolithic mudbrick walls coated with multiple
layers of earthen plaster, wall paintings, and elaborate plaster reliefs are part of the
extensive physical evidence revealed at Catalhoyuk that has served to dramatically alter
traditional views of prehistoric Anatolia and the Near East in general.

Mural paintings discovered at Catalhoyuk are some of the earliest yet found on
man-made walls. Continued excavation of the site, combined with its increased exposure
to the elements will inevitably result in the destruction of building fabric. The impending
loss of any remaining paintings required an evaluation of methods for the transfer and
reattachment of the painted plasters. Due to its invasive nature, the detachment of wall
paintings is a controversial conservation intervention. However, our ability to
comprehend the historical and aesthetic information presented in these superimposed

Chapter 1: Introduction

paintings depends both upon their state of conservation, and our understanding of how
they may have been altered due to the effects of both time and man. 1 An extensive
research program for the conservation and separation of the mural paintings was carried
out at the Architectural Conservation Laboratory of the University of Pennsylvania. The
project began with an overview of conservation literature focusing on detachment
techniques (See Appendix B). A facsimile type was developed for laboratory testing
prior to on-site treatment testing." Executed on 6-by-6-inch and 12-by- 12-inch gypsum
board and terra cotta tiles, the samples consisted of 14 layers of plaster and paintings
using materials similar in character to the original painted plasters. Treatments and
materials were selected based on the following criteria: characterization of the plasters
including optical and physical properties; identification of on-site deterioration
mechanisms; performance of various treatments under controlled conditions; and the
compatibility of selected materials and treatments as tested on the facsimile paintings.

For further information on conservation and detachment techniques, see Paolo Mora and Laura Mora,
Conservation of Wall Paintings (London: Butterworth's, 1984).

" Plaster characterization consisting of a series of tests including bulk sample and microscopic examination,
cross and thin section examination, scanning electron microscopy with energy dispersive spectroscopy,
chemical analysis, and granulometry was carried out by Evan Kopelson and described in, "Analysis and
Consolidation of Architectural Plasters from Catalhoyiik, Turkey" (master's thesis. University of
Pennsylvania, 1996), 95-97.

Chapter 1: Introduction

Based on knowledge of the site, testing considered the following conditions and treatment
requirements:

Surface consolidation of powdering paint and plasters

Interlayerdetachment/preconsolidation

Consolidation of the earthen plasters

Evaluation/selection of facing adhesives for mural detachment

Evaluation/selection of detachment methods

Compatibility of treatments
Visual assessment and standardized tests, developed by the American Society for Testing
and Materials (ASTM), CRATerre, and the Federation of Societies for Coatings
Technology were used to evaluate methods and materials.

Following a comprehensive study of both traditional and modified methods and
materials, two levels of detachment were selected, researched, and ultimately tested on
the laboratory facsimiles. These techniques, known as strappo and stacco, were
originally developed in Italy for the detachment of paintings on lime plaster. Strappo
refers to the detachment of the paint layer alone; stacco, to the detachment of the painted
surface including the underlying plaster layer. A third technique, stacco a massello,
refers to the removal of entire walls. 3

This technique was considered in a separate phase of research.

Chapter 1: Introduction

1.2 Current Research

In 1993, Catalhoyiik was reopened as a full-scale excavation by the Catalhoyuk
Research Trust, under the auspices of the British Institute of Archaeology at Ankara.
Current excavation, headed by Ian Hodder of Cambridge University, aims to continue
archaeological research and to develop a program for tourism and heritage management.
Excavation will be carried out over the next twenty years.

For the past three years, the Architectural Conservation Laboratory of the
Graduate Program in Historic Preservation of the University of Pennsylvania has
undertaken a research program, currently under the direction of Frank G. Matero, Lindsay
Falck, and Catherine Myers, to develop an integrated architectural conservation program
at Catalhoyiik. The research has focused on a broad range of issues: the in situ
stabilization of wall paintings, plaster reliefs, and whole buildings; the development of
non-destructive transfer methods for the wall paintings, reliefs, and architectural
elements', and the development of techniques for the separation of multiple layers of wall
paintings. Additional research carried out at the Architectural Conservation Laboratory
includes the development of apparatus for the detachment, removal, and transport of the
architectural elements, by Professor Lindsay Falck and Caitlin Moore. Also contributing
to this research is a thesis by Elizabeth Moss which explores the environmental
alterations to plastered mudbrick walls during excavation and methods for mitigating
their effects. The relevance of this type of comprehensive research program is global.

Chapter 1: Introduction

The remainder of this introduction will establish a context for the chapters that
follow, beginning with a brief description of the location of the site and prevalent
environmental conditions. A summary of architectural features and a cursory description
of the population are included to further understanding of the context within which the
wall paintings were created. The second chapter, "Sample Development," describes the
development of prototype and preliminary test samples. Chapter 3, "Testing Program"
describes the preliminary testing program, the assessment of results, and the procedure
and results for the detachment of the prototype samples. Chapter Four, "Final Results,"
offers an assessment of data, draws conclusions, and presents recommendations for
further study. Appendix A offers a summary of the treatment history of the painted
plasters at Catalhoyuk. Appendix B provides a brief historical background of detachment
techniques and a compilation of case studies reflecting traditional and modified methods
for the detachment of mural paintings from a number of substrate materials.
1.3 Catalhoyuk

The site consists of a double mound formation known as Catalhoyuk East and
Catalhoyuk West. It is situated nearly 3000 feet above sea level, on the Konya plain, the
largest alluvial plain in Turkey, approximately 14 km north of Cumra, in south central
Turkey.

Considerable seasonal temperature changes are common. The semi-arid climate
may range in temperature from freezing in the winter to greater than 20°c in the summer.

Chapter 1: Introduction

Rainfall is normally under 300 mm per year. 4 The landscape of the Konya Plain is
dominated by the volcano, Kara Dag, to the south and by the Taurus Mountains to the
west. The Carsamba Cay, its main water source, presently splits beyond Cumra into three
branches. Geoarchaeological investigations indicate that a distributary of the Carsamba
River ran between the eastern and western mounds at Catalhoyiik, and that the site itself
lies in the former bed of a Pleistocene lake.

Regional soil studies conducted by Driessen and de Meester in the 1960s indicated
this area as a former backswamp. 5 The soils were classified as Class n, "good soils with
moderate limitations or risks of damage," in other words, adequate for agricultural uses. 6
Location of the settlement at the junction of two different types of Carsamba fan soils, one
suitable for irrigation, the other not, has been compared to the relationship between
settlement location and soil distribution found at other ancient sites. 7 Game may have been
attracted to the moist soil, while the adjacent soil, moistened by capillary action, could have
supported the production of grains. Research by Ellison and Harris conducted in southern
England sustains the notion of a link between settlement site and the juncture of different

Neil Roberts, Peter Boyer and Romola Parish. "Preliminary Results of Geoarchaeological Investigations at
Catalhoyiik" in On the Surface: Catalhoyiik 1993-95, Ian Hodder, ed. (Cambridge: McDonald Institute for
Archaeological Research; London: British Institute of Archaeology at Ankara, 1996), 19.
' Ian Todd, Ctf/fl/ Hiiyiik in Perspective (California: Cummings Publishing, Inc., 1976), 1 13.

P.M. Driessen and T. de Meester, Soils of the Qmira Area, Turkey (Holland: Centre for Agricultural
Publishing and Documentation, 1969; quoted in Ian Todd, Catalhoyiik in Perspective (California: Cummings
Publishing, Inc. 1976), 114.

Todd found a correlation between this relationship and one discussed by Webley regarding certain
Palestinian sites, found in areas containing two different soil types. D. Webley, " Soils and Site Location in
Prehistoric Palestine" in E.S. Higgs, Ed. Papers in Economic Prehistory (Cambridge: Cambridge
University Press, 1972) 169-180, referred to in Ian Todd, £atal Hiiytik in Perspective (California: Cummings
Publishing, Inc., 1976). References: 114, 163.

Chapter 1: Introduction

soil types. 8 Examination of excavated material-faunal remains, paleoecological data, and
soil studies-corroborates the concept of a settlement with an economy based on
domesticated cattle, hunting, agriculture, and trade. 9
The Architecture

Discovery and initial excavation of this site, begun in the 1960's under the direction
of James Mellaart, covered a thirtieth of the 16-hectare east mound, and revealed a
sophisticated Neolithic settlement dating to the seventh and sixth millennia B.C. 1 This is
the largest-known early prehistoric settlement in the Near East with the most extensive area
of architecture as yet uncovered for the period. Its location and the abundance of natural
resources favored the existence of a seemingly continuously stable settlement. Evolution of
this culture is believed by Mellaart to have continued, undisturbed, for a period of at least
800 years.

The most studied mound, the East Mound, is approximately twenty-three meters
high and measures approximately 600-by-350-meters in size. It is believed that
Mellaart's excavation revealed fourteen building levels dating from the eighth and

A. Ellison and J. Harris "Settlement and Land Use in the Prehistory and Early History of Southern
England: A Study based on Locational Models" in D.L. Clarke, Ed. Models in Archaeology (London:
Methuen, 1972) 91 1-962, referred to in Ian Todd, Catal Hiiyitk in Perspective (California: Cummings
Publishing, Inc. 1976), References: 163.

Todd, Catal Hiiyiik in Perspective, 119.
10 The smaller western mound appears to be predominantly Chalcolithic in date. The discovery of Neolithic
remains continuing under this mound indicates that the two may have formed one big settlement ranging over
50 to 60 acres. If this is true, it is the largest Neolithic site known in the Near East. From James Mellaart, The
Goddess from Anatolia (West Germany: Udo Hirsch, 1989) Vol.2, 6. The disturbances of a Hellenistic
occupation and later manipulation of the mound for agricultural use have hindered detection of subsurface
architecture. Roger Matthews, "Surface Scraping and Planning" in On the Surface: Catalhoyiik 1993-95 Ian
Hodder, ed. (Cambridge: McDonald Institute for Archaeological Research: London: British Institute of
Archaeology at Ankara, 1996), 99.

Chapter 1: Introduction

seventh millennia B.C. ' The East Mound is almost completely Neolithic, exhibiting
evidence of Hellenistic, Roman, and Byzantine disturbances in discrete locations. 12 The
site has suffered extensive deterioration since its closing in 1965. Neolithic walls have
collapsed, large sections of plaster have been exposed, and the tops of walls have been
worn away by the foot traffic of tourists. A massive irrigation program has lowered the
water table causing the loss of organic materials previously preserved in waterlogged
conditions.'"

Excavation of the eastern mound at Catalhdyiik revealed a large village, the
architecture largely intact. The settlement appears to have evolved over time, exhibiting
an organic, cellular development nearly devoid of right angles, rather than planned
growth. 14 Most of the buildings were plastered and painted numerous times. Wall
paintings, some of the earliest yet found on man-made walls, are super-imposed on, and
often separated by many campaigns of undecorated plaster. The sense of order and plan
evident in the layout at Catalhoyiik suggests the existence of established customs and
standards.

Buildings were constructed of unbaked mud brick, usually up to or over one meter
in length, made from local alluvium and occupation sediments mixed with vegetal

The exact number is currently in dispute.
" For further information, see Roger Matthews, "Surface Scraping and Planning" in On the Surface:
£atalhoytik 1993-95 Ian Hodder, ed. (Cambridge: McDonald Institute for Archaeological Research;
London: British Institute of Archaeology at Ankara, 1996), 81.

' Ian Hodder. ed.. On the Surface: Catalhoyiik 1993-95 (Cambridge: McDonald Institute for
Archaeological Research; London: British Institute of Archaeology at Ankara, 1996), 2.

Chapter 1: Introduction

stabilizers. In some cases, the mortar layer is as thick as the mud bricks. The houses are
timber-framed and have squared oak post and beam construction with mud bricks filling
the openings between the posts. Timber was also used in the construction of roofing,
verandas, and paneling.

Generally, changes in mud bricks and mortar correspond to changes in floor and
building levels. Although evidence suggests that different types of bricks were used
concurrently during the construction of a wall, these variations, coupled with a slight
overhang between one level and the next often indicate the construction of new walls
directly atop earlier ones following the orientation of previous buildings. In such cases,
both sections were usually rendered with a continuous layer of plaster. Examination of
these plasters revealed two different types. 15 One was an earthen plaster, and the other, a
finer marly type, was probably obtained from the Pleistocene lake bed.

Multiple applications of white wall plaster, as well as elaborately prepared plaster
features and niches, are common. Plastered posts and the lower panels of houses were
either painted red all over or just red for the posts, or with a pattern on the lower panels. 1
In many instances, multiple layers of red painted plaster were found within the white
plaster sequences on walls, floors, platforms and benches. Paintings were generally

14 Roger Matthews. "Surface Scraping and Planning" in On the Surface: Qatalhoyuk 1993-95, Ian Hodder.

ed. (Cambridge: McDonald Institute for Archaeological Research; London: British Institute of Archaeology

at Ankara, 1996), 85.

1:1 Wendy Matthews and Shahina Farid, "Exploring the 1960s' Surface: the Stratigraphy of Catalhbyiik" in

On the Surface: Qatalhbyuk 1993-95, Ian Hodder, ed. (Cambridge: McDonald Institute for Archaeological

Research; London: British Institute of Archaeology at Ankara, 1996), 275-276.

16 James Mellaart, £atalhdyiik: A Neolithic Town in Anatolia (New York: Mc-Graw Hill Book Company)

149.

Chapter 1: Introduction l_0_

executed above the main platforms of the north and east walls, although they occasionally
stretched around corners to adjacent walls.

Plaster floors were made from alluvial and lake-derived deposits with particle
sizes ranging from silty clays to medium coarse sandy silt loam. 1 Similar materials were
used to plaster the walls at £atalhoyuk. 18 Thin section examination of the white,
calcareous silty clay revealed sediments composed of up to 95% pure calcium and
magnesium carbonates, similar to soft lime deposits from the region. To this day,
villagers mix these deposits with water to plaster floors and walls.

Houses were constructed on a rectangular plan, each consisting of approximately
twenty-five square meters. Interior spaces generally consisted of one main large room
and, in some cases, additional smaller rooms with an entrance or opening raised off the
floor plane. Entrance to each house was gained through a hole in the roof accessed by a
wooden ladder placed against the south wall of the main room. This opening also
provided ventilation for the hearth and oven below. Roofs consisting of a layer of reeds
covered with a thick layer of mud were staggered to allow each building access to light.
Individual residential units, as opposed to communal living spaces, are as yet difficult to

17 Results of thin section examination described in Wendy Matthews, Charles French, Timothy Lawrence
and David Cutler, " Multiple Surfaces: the Micromorphology," in On the Surface: ^atalhoyiik 1993-95, Ian
Hodder, ed. (Cambridge: McDonald Institute for Archaeological Research; London: British Institute of
Archaeology at Ankara, 1996), 304.

Wendy Matthews, Charles French, Timothy Lawrence and David Cutler, " Multiple Surfaces: the
Micromorphology," in On the Surface: ^atalhoyiik 1993-95, Ian Hodder, ed. (Cambridge: McDonald
Institute for Archaeological Research; London: British Institute of Archaeology at Ankara, 1996), 304.

Chapter 1: Introduction 11

discern due to the lack of doorways that would suggest adjoining rooms. 19 Party walls
were rare, internal spaces physically defined by double or triple walls, which may have
served to support the roofing, to act as insulation, or perhaps to emphasize boundaries of
personal space.

The proximity of buildings, ranging from 2-35 cm apart, did not allow room for
streets or alleys. Therefore, communication between neighbors was carried out at roof
level. Furniture consisted mainly of built in platforms, benches, hearths, and ovens.
Most rooms contained bins for grains or storage. The outer wall of the settlement, as
uncovered by Mellaart's team, was void of any opening."

Analysis of occupation debris indicates a settlement consisting of residences and
open spaces, the latter used primarily for rubbish disposal and sanitation. Mellaart made
a distinction in building use between dwellings and shrines based on the richness of
decoration and distribution of the remains found in each. However, sufficient evidence
has not been found to differentiate the sacred from the secular within the settlement."'
Activities associated with both domestic and burial practices are often evidenced within

1 The discovery of what may be a door between houses indicates that more than one form of access may
have existed between some houses. Ian Hodder, "Conclusions," in On the Surface: Qatalhoyuk 1993-95,
Ian Hodder, ed. (Cambridge: McDonald Institute for Archaeological Research; London: British Institute of
Archaeology at Ankara, 1996), 363.

20 James Mellaart, "Catal Hiiyuk in Anatolia: Excavations which Revolutionize the History of the Earliest
Civilizations. Part II-Shrines and Buildings," in The Illustrated London News: Archaeological Section No.
2122 (February 2, 1963) 160.

"' The buildings classified as shrines do not differ in structure, size, or internal organization from the houses
and, for the most part, are intermingled with them throughout the settlement. Also no distinction was found
to exist in the construction materials of either building type. Although both contained below-platform
burials, Mellaart believed that "shrines" housed the monumental reliefs and wall paintings. All buildings
contain hearths, ovens, and storerooms and were kept exceptionally clean. Ian Hodder. "Contextual

Chapter 1: Introduction 12

the same building. The presence of red ochre paint on platforms above human skeletal
remains as well as in occupation and fire installation deposits, indicate ritual/burial
activity throughout the settlement." Because many buildings display evidence of
multiple use and/or changes in use all of the buildings are now considered dwellings.

Often what were once buildings became open spaces and vice versa. Variations
within the continuity of building layout have been detected between buildings.
Occupation levels in which evidence of multi-use and transitions exist, show that changes
have occurred over the long term. Elaborate buildings appear to have been reconstructed
on the site of other elaborate buildings. And, artistic motifs have been observed to
survive from one level to the next. Surface examinations indicate the controlled use of
fire within architectural spaces, perhaps to signify the end of use of a space.

At present, the settlement is understood as an agglomeration of architectural
spaces ranging in elaboration from simple or decorative to symbolically complex." No
evidence has yet indicated the existence of large-scale public buildings. The distribution
of evidence for crop processing, animal tending, and obsidian manufacture amongst
domestic spaces combined with the simplicity of the two part plan, indicate that

Archaeology: An Interpretation of Catal Hiiyiik and a Discussion of the Origins of Agriculture" in Institute
of Archaeology Golden Jubilee Bulletin Number 24 (London: Institute of Archaeology, 1987) 44.
"" Wendy Matthews, Charles French, Timothy Lawrence and David Cutler, " Multiple Surfaces: the
Micromorphology," in On the Surface: Catalhoyiik 1993-95, Ian Hodder, ed. (Cambridge: McDonald
Institute for Archaeological Research; London: British Institute of Archaeology at Ankara, 1996), 317.
" Jonathon Last, "Surface Pottery at Catalhoyiik," in On the Suiface: Catalhoyiik 1993-95, Ian Hodder, ed.
(Cambridge: McDonald Institute for Archaeological Research: London: British Institute of Archaeology at
Ankara, 1996), 164.

Chapter 1: Introduction 13

£atalhbyuk functioned as a village rather than as an urban center. 24 Nonetheless, the
elaborate wall paintings, painted plaster reliefs, and burial customs occurring at
£atalhoyuk suggest that this population lived in a very complex symbolic culture that
appears to have revolved around continuity and ancestor worship." 5
The Population

The extensive physical evidence discovered at (Tatalhoyuk has dramatically
altered traditional views of prehistoric Anatolia and the Near East in general. Its
significance is twofold: the physical evidence provides information on the culture and
social systems of a Neolithic society; and in turn, we have gained a heightened
understanding of human resourcefulness and adaptability. Here it was discovered, a
civilization with sophisticated artistic ability and complex religious beliefs had existed.
The wall paintings, plaster reliefs and sculpture, as well as evidence of weaving,
woodwork, metallurgy and obsidian working demonstrate an advanced level of
achievement.

The population of CJatalhoyiik probably survived on an economy of irrigation
agriculture, animal husbandry, hunting, and domestic industry. 26 Evidence of the
domestication of dogs and cattle was produced during excavation of Mellaarfs Level XII,

24 Roger Matthews in On the Surface: Catalhoyiik 1993-95, 86-87.

' Ian Hodder, "Conclusions," in On the Surface: Catalhoyiik 1993-95, Ian Hodder, ed. (Cambridge:
McDonald Institute for Archaeological Research; London: British Institute of Archaeology at Ankara,
1996), 366.

Documentary evidence of a simple irrigation system was noted in one of the fragmentary paintings
excavated by Mellaart. James Mellaart, The Goddess from Anatolia (West Germany: Udo Hirsch, 1989)
Vol. 2, 38.

Chapter 1: Introduction 14

representing a pre-seventh millennia B.C. settlement." The importance of cattle to the
population of the settlement both as a source of food and transport, is seen in its
continuous pictorial representation, in both naturalistic and abstract forms, in material
culture remains. The aurochs was represented in wall paintings found throughout the
settlement levels, as were bucrania, which were often plastered and incorporated into the
architecture. Despite the domestication of plants and animals, hunting was popular for
sustenance and sport, and was represented in wall paintings and in the ornament of
weapons and arrowheads in the burials of men.

The population had a wide range of locally available foodstuffs including game,
fish, fruit, nuts, vegetables and dairy products. Although domesticated cereals have been
discovered, dental microwear studies conducted on human skeletal remains and a lack of
grinding artifacts indicate that cereal was probably not a staple of the diet. Evidence

indicates that the people ate mainly tubers and pulses." Although occupation deposits
rarely reveal fish bones, wall paintings discovered by Mellaart depict nude women with
nets and fish. Further, the practice of food and beverage storage, which was new to the
sedentary culture, probably led to the production of beer and wine through the natural
fermentation process.

Archaeological studies have shown that the population at Catalhoyiik was a mix
of Eurafricans, 59%, descended from an Upper Paleolithic type, Proto-Mediterraneans,

27 James Mellaart, The Neolithic of the Near East (London: Thames and Hudson, Ltd., 1975) 98.
' Theya Molleson and Peter Andrews, "Trace Element Analyses of Bones and Teeth from Catalhoyiik," in
On the Surface: Catalhoyiik 1993-95, Ian Hodder, ed. (Cambridge: McDonald Institute for Archaeological
Research; London: British Institute of Archaeology at Ankara, 1996), 265.

Chapter 1: Introduction I5_

17%, and brachycephalic Alpines, 24%. The average life expectancy for both men and
women was approximately thirty years and the maximum population at any given time
would probably have been approximately 5-6,000. 29

The site is rich with material evidence of the culture. The level of skill of a
portion of the population is attested to by the quality and complexity of the wall
paintings. The stationary culture, made possible by an economy based on agriculture and
the domestication of animals, allowed for the development of specialized craftsmanship.
Production areas or workshops have not been identified. Although only finished products
have been found in very clean buildings, it is unlikely that so wide an array of finished
products were imported." It is believed that production areas within the settlement have
yet to be excavated." '
Wall Paintings

The excavated wall paintings at £atalhoyiik are some of the earliest yet found on
man-made walls. As many as 80 two-part sequences of ground and finish plaster layers,
each measuring 0.5mm or less, were revealed in the examination of representative earthen
plaster and mudbrick samples sent from ^atalhoyiik. Paintings, most often executed on
the dense white finish plasters, were microscopically observed in cross-section
examination between many of the super-imposed sequences. Several unpainted plaster
layers often separate them. Laboratory and in situ examination of the plasters reveal

2q Ibid, 99.

10 Well-made objects in obsidian, imported flint, stone, metal, beads, textiles, and pottery, wooden statues

and vessels have all been found on site.

31 James Mellaart, Goddess from Anatolia, Vol. 2 (West Germany: Udo Hirsch. 1989) 7.

Chapter 1: Introduction 16

several types of deterioration including: the loss of cohesive strength within discreet
layers and adhesive strength between individual layers of the plaster and mural paintings,
macro-biological growth, and salt migration. Causes of deterioration may be linked to
drastic environmental changes brought on by excavation.

The degree of preservation of newly excavated wall paintings suggests that they
were usually painted over while still in good condition. " During the 1960s, layers of plain
plaster were often manually removed in order to reveal paintings on preceding layers.
Removal at this time was a slow process, conducted with small dental knives and scalpels to
remove the undecorated plaster layer by layer from the painting. This type of technique
requires enormous care to prevent damage to the painted surface.

Many of the wall paintings uncovered in the 1960s were removed to a conservation
laboratory. Detachments were carried out in two ways, the block method, in which the
painting was removed along with a portion of the mud brick wall (stacco a massello); and
the peeling method, in which the surface of the painting was coated with an adhesive and a
linen facing, permitted to dry, and peeled along with a small amount of plaster away from
the mudbrick wall (strappo).

Both the dry climate, and the retention of substantial portions of the architecture
may explain the outstanding state of preservation of many of the wall paintings directly
after excavation. Newly exposed paintings however, begin to degrade after just a short

" Paintings are rarely discovered directly on the surface of a wall during excavation. Discovery usually
occurs due to losses within overlying plaster layers.

Chapter 1: Introduction 1_7_

time in the open air, and are often found detached from the walls, fragmented and buried
in rubble.

Pigments were derived mainly from minerals, including among others, azurite,
cinnabar, malachite, and galena, all of which occur naturally on the Anatolian Plateau. 33
Most often, the paintings were executed on a cream or white ground, the bristle marks of a
fine brush sometimes visible. Generally, paintings were confined to a single wall panel,
although exceptions occur in which a scene continues from one wall around the corner onto
the next.

The preliterate society at Catalhoyuk probably used imagery and symbolism as a
means of communication to formulate and illustrate belief systems, history, religion and
ritual. Although some of the paintings may be purely decorative, the subject matter of others
strongly indicates religious and philosophical beliefs. Repetition of designs and their
continuous use throughout the duration of the settlement indicate that these were established
forms of expression. If so, functional representations would not necessarily have been
realistic. Images may not represent daily situations at all, but may instead refer to myth,
legend, or history. Wall paintings must be observed in a context encompassing all material
culture remains, if they are to aid in the interpretive process.
A Society in Transition

As stated above, the settlement at Catalhbyiik represents the transition of a society
from an economy based on hunting and gathering, the Paleolithic, to one based on the

33 James Mellaart, Catal Hiixiik: a Neolithic Town in Anatolia, (London: Thames and Hudson, 1967) 131

Chapter 1: Introduction

domestication of plants and animals, the Neolithic. It would not be surprising then that as
elements of the lifestyles carry over, so too would the focus of imagery and symbolism.

The oldest mural paintings discovered date to the Paleolithic period; the most
famous of which are located in France and Spain. The material culture of the Paleolithic
people reflected their nomadic lifestyle and most often took the form of movable objects.
Monumental cave paintings were the exception. Placement and style of these Paleolithic
paintings, located sometimes deep within caves' interiors, indicate that they may have
served ritual purposes. Their imagery consisted mainly of wild animals and game, which
are believed to represent the culture's main source of food; female statuettes; and
geometric motifs. 34

Paintings at (Tatalhoyiik share a similar subject matter. Animal scenes are
common and may illustrate hunting scenes, although none depict an actual kill. Human
figures appear to dance around animals, such as the deer and the aurochs.

The two periods also share a color palette. Artists of both the Paleolithic and the
Neolithic made great use of red and black paints. So often in fact, that it is difficult not to
imbue them with symbolic meaning, usually interpreted as life and death. This
interpretation is supported by the red ochre burials found in the Paleolithic, later at
£atalhoyuk and at other settlements throughout the world. Traces of red paint were
discovered on many of the architectural elements at (Tatalhoyuk.

34 Early examples also exist at Beldibi. Okiizlii in, Kara In, Kiirtun and the Palanli caves as cited in James
Mellaart, The Goddess from Anatolia (West Germany: Udo Hirsch. 1989) Vol. 2, 3.

Chapter 1: Introduction 19_

Symbols or motifs, such as the repetition of triangles, circles, "swastikas," and
hands and opposing triangles, alternating in color between red and black with white dots,
forming positive and negative bands or combined in groups of four creating a rhomb, can
be seen in both Paleolithic cave paintings and in the mural paintings of Catalhoyuk..

Another motif common to both periods may be found in the depiction of women.
Sexual characteristics are emphasized in the Paleolithic art of Western Europe and a
number of Near Eastern settlements of a much later date. The Venus figures of the
Paleolithic are believed by Mellaart to have evolved at Catalhoyuk into a figure he called
the "Goddess." He wrote repeatedly of a "goddess" image in painting, relief, and figurine
form, using the term as a blanket category for the majority of the discovered female and
questionably female images. This figure is shown in many variations and compositions:
in reciprocal or cross forms, connected or in mirror image, natural, and abstract.
Sometimes naturalistic representations are combined with the abstract and include
animals. Current scholarship rejects many of these theories due to the lack of supporting
evidence. 35

The figures have been perceived in a number of ways. At times, they appear to be
pregnant, or to be giving birth to humans, wild animals or birds. Sometimes they seem to
be twinned or joined to a mate. Some of the excavated murals very clearly depict female
figures with nets, fishing.

35 For further discussion, see Ian Hodder, "Contextual Archaeology: An Interpretation of Catal Huyuk and a
Discussion of the Origins of the Origins of Agriculture" in Institute of Archaeology Golden Jubilee Bulletin
Number 24 (London: Institute of Archaeology, 1987) 45. The author finds no indication of clearly female

Chapter 1: Introduction 20

Women were also shown with wild animals. The depiction of a female figure
with two vultures is common in Anatolia. The union of a symbol of fertility and one so
closely associated with death and rebirth indicates a powerful symbolism. It has been
suggested that the relationships between men and women, domestic and wild, and life and
death have been worked out in the imagery, the architecture, and even the city plan of
£atalhoyiik. 36 The connection between women, as the givers of life, and predators, or
dangerous animals, implies that women may have had a dual symbolic role relating to
both death and renewal. 37 They appear at least symbolically, to have had a role in all

aspects of the cycles of human and animal life. The relatively brief life expectancy of
the population of £atalhoyiik, approximately thirty years, gives credence to the belief that
the symbolism refers predominantly to fertility and reproduction.

The subject of death appears to be a recurring theme amongst the wall paintings and
reliefs. Birds were often depicted in association with headless human beings. These images
are believed to represent the excarnation process. If this is true, the vultures perform part of
a burial rite by stripping the flesh from the bones of the dead before their final burial and
ultimate rebirth. In some cases, vultures have human feet, indicating human participation in
the process.

characteristics on the relief figures described by Mellaart as pregnant goddesses giving birth, sometimes to
humans, others to animals.

36 Ian Hodder, "Contextual Archaeology: An Interpretation of Catalhoyuk and a Discussion of the Origins
of the Origins of Agriculture" in Institute of Archaeology Golden Jubilee Bulletin Number 24 (London:
Institute of Archaeology, 1987).

37 Ian Hodder, "Conclusions," in On the Surface: Catalhoyiik 1993-95, Ian Hodder, ed. (Cambridge:
McDonald Institute for Archaeological Research; London: British Institute of Archaeology at Ankara.
1996), 365.

Chapter 1: Introduction 21

Imagery at Catalhoyiik is sometimes expressed in positive and negative images that
emphasize the relationship between the figure and the background. This interplay is
deliberate and what at first appears to be an abstract pattern may in fact be recognizable
figures. 9 The relationship between positive and negative space undoubtedly had symbolic
meaning, each side of equal importance.

Thus, the concerns of these people, as expressed through their illustrations, appear
to be fertility, death, and rebirth. These are universal concepts, represented materially in
cultures all over the world.
Continuity

Striking examples of wild animal imagery support the idea of a continuous belief
system. A pair of modeled leopards coated with numerous campaigns of painted plaster
exhibit only slightly evolving designs. The aurochs is found continuously in the imagery of
Catalhoyiik in the form of bucrania, plastered and painted and set into benches or hung on
walls and in wall paintings. Portrayed most often as a sexless bull, the symbol has been
equated with male fertility as well as domestication of the wild. Whatever the

18 Walter A. Fairservis, Jr., The Threshold of Civilization, An Experiment in Prehistory (New York: Charles
Scribner's Sons, 1975), 155.

39 James Mellaart The Goddess of Anatolia (West Germany: Udo Hirsch, 1989) Vol. 2, 28.

40 The equation of bulls with male virility is however, a western perception not necessarily borne out in this
context. Naomi Hamilton, "Figurines, Clay Balls, Small Finds and Burials," in On the Surface: Catalhoyiik
1993-95. Ian Hodder, ed. (Cambridge: McDonald Institute for Archaeological Research; London: British
Institute of Archaeology at Ankara, 1996), 252.

Chapter 1: Introduction 22

interpretation, the symbolic significance of this image is witnessed by its use in other
cultures. 41

Mellaart noted the similarity between excavated wall paintings and much later
kilim patterns. In his article, "The Leopard Shrines of Chatal Huyuk," he described a
building "decorated with several super-imposed layers of textile pattern, imitating
kilims." 4 " It is now believed that the striking similarities between Neolithic wall
paintings and designs found centuries, even millennia later in Anatolian kilim design, are
inherited from Neolithic symbolism. 4 "

Catalhoyuk is believed to be the only ancient site as yet to exhibit such a wide
variety of both naturalistic and abstract imagery. Mellaart divided the wall paintings
and reliefs into the following categories:

Monochromatic plaster panels usually painted with some form of red,

• Geometric patterns in both monochrome and polychrome, sometimes repetitive
or mirror image, either rectilinear or curvilinear,

• Symbolic images including circles, quatrefoils, crenellations, stars, swastikas,
and triangles,

Bucrania was used as an architectural element at an earlier site, Mureybet la in northern Syria, dating
from between 10,000-7500 BC.

" James Mellaart, "The Leopard Shrines of Chatal Huyuk." in The Illustrated London News:
Archaeological Section No. 2246, June 4, 1966) 25.

" James Mellaart, Udo Hirsch. and Belkis Balpinar conducted an extensive study of the relationship
between kilim designs and the imagery found in the wall paintings at Catalhoyuk. Although specific belief
systems cannot be deciphered, the discovery of similarities in the design or symbolism of people separated
by time or space may prompt continued study of the paths of communication, knowledge, and trade which
may not otherwise have been considered.
44 James Mellaart, The Goddess from Anatolia (West Germany: Udo Hirsch, 1989) Vol.3, 38.

Chapter 1: Introduction 23

• Panels covered with handprints or silhouettes, usually with five fingers, or hands
framing panels with geometric or naturalistic designs,

• Naturalistic images depicting human figures, bulls, birds, vultures, leopards,
deer, ibexes, bees, found either by themselves or grouped into elaborate scenes,
such as hunts, fishing, or funerary rites; also scenes depicting bears, boars, and
bulls surrounded by humans, some of whom wear leopard skins, and

• Images depicting landscapes such as a tree ripe with fruit surrounded by ibexes;
the settlement with a yurt-like structure and the volcano Hasan Dag in the
background. 43

Through artistic imagery, viewed in a social and cultural context across time and
space, we have the potential to increase our general understanding of a past civilization.
Preservation and presentation of these paintings may serve to enhance our understanding of
the lifestyle of £atalhoyuk's inhabitants. With this in mind, the conservation, analysis, and
interpretation of wall paintings at £atalhoyuk should include the context of the painting
within the room, the building, the site, the region, and finally, within the Near East. 46

45 James Mellaart, Cata/ Hiiyiik: A Neolithic Town in Anatolia (London: Thames and Hudson, 1967) 132,
149.

46 Dangers arise when Western perceptions influence the interpretation process. An example of this is
described in the report, "Figurines, Clay Balls, Small Finds and Burials" by Naomi Hamilton in On the
Surface: Qatalhoytik 1993-95. Ian Hodder, ed. (Cambridge: McDonald Institute for Archaeological
Research; London: British Institute of Archaeology at Ankara, 1996), 226. She notes the risks involved in
the interpretation of the large relief figures positioned atop animal heads as goddesses giving birth. This
interpretation "fits beautifully with Western attitudes to virility and its appropriate symbols, but less well
with evidence."

Chapter 1: Introduction 24

Conclusion

Comprehension of Near Eastern cultures may be improved through careful
observation of the excavation at Catalhoyuk. Documentation of the wall paintings,
architecture, pottery, figurines, and burials has been carried out and published by James
Mellaart, Ian Todd, Ian Hodder, et al. Interpretations too, have been written and rewritten
as methodologies and theories have changed.

Although the imagery found at Catalhoyuk can be tied to Paleolithic cave paintings,
contemporary Near Eastern sites, and the Anatolian kilims design of the eighteenth through
twentieth centuries, discussions about universal meanings must be conducted very carefully
and critically. In a preliterate society, images and meaning may evolve or change without
record. Wall paintings, relief plasters, and, to a lesser extent, figurines with dramatic
imagery, indicate that a powerful symbolism was at work. References to the most basic of
human concerns: fertility, death, and rebirth abound. An unbiased study and comparison of
these paintings with those of other cultures and later periods may indicate a continuum of
belief, transfer of knowledge, or the use of like symbolism to connote totally different ideas.

Accurate interpretation of archaeological sites requires that a sufficient amount of
the settlement tract be excavated. Research from a wide range of disciplines may
increase our potential to grasp the basic concepts of the lifestyle of an ancient people.
Only with these concepts, and the contextual relationships between excavated remains,

Mellaart reconsidered previous conclusions in a series, The Goddess from Anatolia, (James Mellaart, Udo
Hirsch and Belkis Balpinar, 1989.) They explored continuity of material culture of the area from the wall
paintings at Catalhoyuk to present day Anatolian kilim design.

Chapter 1: Introduction 25_

can we begin to identify possible interpretations of any art. Even with such information,
images representing religious or mythological beliefs may be interpreted as literal
depictions of everyday life and vice versa. Images of animals and humans may represent
themselves or deities. Basic knowledge of day to day life is helpful but is in no way a
substitution for knowing the minds of the subjects under study.

Reopening the excavation at £atalhoylik has underscored the need for
collaboration between independent disciplines. Excavations may be better understood
and interpreted by integrating information from many disciplines including, but not
limited to, archaeology, conservation, art and architectural history, ethnoarchaeology,
anthropology, botany, and folkloristics.

Significance of this excavation lies not simply in the light it has shed on the
cultural aspects of a Neolithic society, but of equal importance, its potential to broaden
our understanding of the adaptability of humans to their environments. At £atalhoyiik a
society was, for the first time, dependent on the domestication of plants and animals
rather than hunting and gathering.

Chapter 1 : Introduction

[X .

Fig. 1. General map of Turkey. From Mellaart, James. Udo Hirsch. and Bclkis Balpinar. The Goddess from Anatolia.

West Germanv: Udo Hirsch, 1989.

Chapter 1: Introduction

Fig. 2. James Mellaart working on excavated wall painting From Mellaart. James. Udo Hirsch. and Belkis Balpinar The
Goddess from Anatolia West Germany: Udo Hirsch, 1989.

Fig. 3 Painting from north wall of Shrine F.V/1 showing bull surrounded by humans and other animals From Mellaart.
James. Udo Hirsch. and Belkis Balpinar The Goddess from Anatolia West Germany: Udo Hirsch. 1989

Chapter 1 : Introduction

Fig

4 Plaster relief and hand prints on red painted wall in Shrine E. VI1/8. From Mellaart. James. Udo Hirsch. and
Belkis Balpinar The Goddess from Anatolia West Germany: Udo Hirsch. 1989.

^nkJS BUs

si ,v

i,lg€ ft ,,

4 *■*

Fig. 5. Scale copy of painting from Shnne E. VIA/50 FromMellaart. James. Udo Hirsch. and Belkis Balpinar. The
Goddess from Anatolia. West Germany: Udo Hirsch. 1989.

Fig. 6. Wall painting with alternating red and black triangles from Shnne E VII/21 From Mellaart, James. Udo Hirsch.
and Belkis Balpinar The Goddess from Anatolia West Germany Udo Hirsch. 1989

Chapter 1: Introduction

Fig. 7. Plaster relief of painted human figure from Shrine E. VI1/21 From Mellaart. James. Udo Hirsch. and Belkis
Balpinar The Goddess from Anatolia. West Germany: Udo Hirsch. 1989

Fig. 8. Wall painting with hand prints from Shnne E. VIB/15 FromMellaarl. James. Udo Hirsch. and Belkis Balpinar
lite Goddess from Anatolia. West Germany: Udo Hirsch, 1989.

Fig. 9. Wall painting showing human figures in leopard skins from Shnne F V/l. From Mellaart. James. Udo Hirsch.
and Belkis Balpinar The Goddess from Anatolia West Germany: Udo Hirsch. 1989.

Chapter 2: Sample Development

The plasters and mural paintings at £atalhoyiik are irreplaceable records of the art
and culture of a Neolithic settlement. As with any similar research program, it was
critical to prevent the permanent alteration or destruction of original material. For this
research, facsimiles provided the most practical means by which to test treatments. 48 The
creation of laboratory facsimiles required the characterization and replication in the
laboratory, of the materials and conditions at the site.

To accurately produce a laboratory facsimile, it was necessary to understand the
materials and technology used to construct Neolithic wall paintings. Several methods of
examination informed sample formulation. Microscopic examination of representative
earthen plaster and mudbrick samples reveals as many as eighty two-part sequences of
ground and finish plaster layers, each measuring 0.5 millimeters or less. Paintings, most
often executed on the dense white finish plasters, were observed in cross-section between
many of the superimposed sequences. To replicate this stratigraphy, samples were made
of multiple plaster and paint layers, similar in substance and character to those found at
^atalhoyiik.

8 Since the testing program required a large amount of sample, it was not possible to use original material.
49 Variables such as paint composition, thickness of layers, presence of dirt or soot, and adhesion to the
substrate were not strictly reproduced.

Chapter 2: Sample Development 31_

In order to reduce variables, the sample prototype was based exclusively on the
characterization of the plaster and paint constituents and did not replicate conditions
requiring accelerated aging, such as deterioration. 50 Other conditions occurring at the
site, such as interlayer detachment and friability of the plaster surface, developed without
inducement.
2.1 Existing Conditions:

Previous research indicates that the plaster was composed of materials locally
available to the native population at Catalhoyuk. 51 The mudbrick walls were often coated
with up to 80 superimposed phases of two-part ground/finish plaster layers made from
local marly soils. 32 Microscopic examination revealed the presence of entire shells within
the plaster layers suggesting that, although the technology to burn and slake lime existed,
local soils were probably used without burning."

A preparatory layer ranging from 1-3 mm thick often coated the mudbrick
surfaces as revealed by microscopic examination of Catalhoyuk plaster samples." 4 A
light brown ground layer and a finer white finish coat were usually applied over the
preparatory layer. The most common stratigraphy showed superimposed sequences

5 Aging and deterioration may cause the disintegration of organic binding media resulting in the loss of

cohesion of plaster and paint particles. Exposure to the elements, ultraviolet light, salt migration, and

macro-biological growth serve to intensify molecular degradation.

M In his earlier thesis, Evan Kopelson characterized the plaster using the following tests: bulk sample

examination, cross section examination, thin section examination, scanning electron microscopy with x-ray

spectroscopy, and wet chemical analysis of organic material content, soluble salt content, acid-soluble

content, granulometry, and pH determination. Evan Kopelson. "Analysis and Consolidation of

Architectural Plasters from Catalhoyuk, Turkey" (masters thesis. University of Pennsylvania, 1996).

"" See Introduction for detailed description of the mudbrick walls.

"Kopelson. 130-131.

34 Microscopic examination was conducted on a Nikon SMZ-U stereoscope at 4x to 75x magnification.

Chapter 2: Sample Development 32_

consisting of ground layers, 0.5 mm or less in thickness, followed by finish layers as thin
as 0.25 mm. Ground layers made from pale brown calcareous silty clay were usually
thicker than the finish layers and often contained white inclusions. The finish layers may
have been burnished, as they were often smooth and compacted and composed of finer
whitish calcareous silty clay with very few medium or coarse particles. The extraordinary
thinness of the strata suggests that the soils were probably sieved and mixed with enough
water to form a substance similar in consistency to a limewash or slurry.

Previous analysis of these plasters indicate that they generally consist of fine
calcite particles, quartz, feldspars, and unburned shell fragments bound in a clay matrix. 5
Although it was not possible to identify the clay precisely, geological studies corroborate
other evidence that it is a local clay, montmorillonite, smectite. 56 It is remarkable that the
same technology and materials were used for the entire sequence of wall plasters.
Compositional analyses, which focused on ground/finish sequences rather than individual
layers, indicated the consistent use for all layers of a plaster mix of approximately 50%
calcium carbonate, 40-47% clay and 3-10% sand. Thin section examination of the white
plasters revealed a material made of up to 95% pure calcium and magnesium carbonates,
a composition similar to the soft lime deposits currently found in the region/

55 Kopelson, 130-131.

5 Results of the x-ray diffraction analysis of the clay component of the plaster had not yet been obtained at

the time this research was conducted.

>7 Wendy Matthews, Charles French. Timothy Lawrence and David Cutler, "Multiple Surfaces: the

Micromorphology," in On the surface: £atalhoyuk 1993-95, Ian Hodder, ed. (Cambridge: McDonald

Institute for Archaeological Research; London: British Institute of Archaeology at Ankara, 1996), 304-306.

Chapter 2: Sample Development 33_

Paints appear to have been composed of a limited number of pigments bound with
an unknown medium. Red and black, two of the most commonly used colors in
prehistoric painting are often interpreted as having symbolic powers, usually pertaining to
life and death. Blacks were carbon-based. The red pigments were iron oxides and have
been used since the Paleolithic Era. Of these, red ochre is known to have symbolic
meaning in primitive societies to this day. Red painted imprints of hands discovered on
walls at Catalhoyiik have also been observed at Upper Paleolithic and Native American
archaeological sites.
2.2 Characterization of Original Materials

Analysis of pigments and organic binding media was based on three small plaster
samples from the site. These multi-layered samples, obtained during the 1995 season,
were inconsistent and showed three to fourteen layers, all without substrate. They did
however, retain traces of painted plaster on their surfaces. One small sample labeled
"sheena wall black" showed traces of black paint. Two other samples labeled "Ch 95
Mell 1014 5/9 8F" showed traces of light red and dark red paint, respectively. The
samples were microscopically examined in cross section for stratigraphy. Pigments were
characterized using microchemical spot tests and scanning electron microscopy with
energy dispersive spectroscopy (SEM/EDS). Although the possibility of finding organic
materials was improbable, samples were studied for the presence of organic binding

Chapter 2: Sample Development 34_

medium. Staining techniques, both simple direct reactive staining; and fluorescence

staining, showed no indication of organic materials.'

Based on these results, three original sample types were selected as models for
laboratory facsimiles. These samples, ranging from (1 inch) 3 to (2 inches) 3 were chosen
because they retained painted surfaces, possibly mural painting surfaces; and because they
consisted of multiple super-imposed ground/finish layers, most representative of the site.

This aspect of research warrants additional study. It is recommended that a wide sampling of materials
from various locations be analyzed with these and other methods for identification of organic materials.

Chapter 2: Sample Development

The following table describes the samples.

Table 1

Sample

Description of Original Plaster Samples

RL
(RL.RL1)

••Ch95Mell 1014 5/9
8F'

Sample consists of approximately fourteen layers. Traces of light red paint appear on
the surface.

RD
(RD, RD1)

"Ch95MeU 1014 5/9
8F'

Sample consists of approximately three plaster layers. Traces of dark red paint appear
on the surface.

BL
(BL, BL1)

"Sheena wall black"

Sample consists of approximately seven plaster layers. Traces of black paint appear on
surface

Chapter 2: Sample Development 36_

2.2.1 Analysis of Stratigraphies
Summary:

Microscope examination of cross sections provides a method for studying the
general structure of mural paintings, including stratigraphies. This type of analysis,
combined with surface examination and the identification of organic and inorganic
components, facilitates an understanding of the techniques and materials used at different
stages of execution. In addition, it may allow the investigator to observe the
superimposition of earlier paintings, deterioration mechanisms, paint penetration, and
discontinuities or transformations within the sequence. 59
Objective:

The aim was to microscopically characterize stratigraphies and pigment particles
where apparent.
Methodology:

Unpolished cross sections were examined with a compound microscope at 100 x
magnification in reflected visible light. 60 They were then photographed.
Stratigraphic descriptions and accompanying photomicrographs of each sample follow.

59 Paolo and Laura Mora and Paul Phillipot, The Conservation of Wall Paintings (London: Butterworths,
1984), 21-23.

60 Embedded samples were not polished because the plaster was fragile, even after consolidation with
Acryloid B67. The extensive loss of material caused by cross-sectioning and polishing dissolved surface
material and muddled definition of stratigraphies.

Samples were photographed using the Nikon Optiphot polarized light microscope illuminated by fiber
optics at 50X using Kodak Royal Gold 200 ASA film.

Chapter 2: Sample Development

Observations:
Table 2

Cross section examination: Sample RD1

Layer

Type

Description

Finish plaster

(?)

Densely compacted thick off-white homogeneous layer with black and
brown inclusions; uneven distribution; voids; vertical cracking

Ground
plaster

Densely compacted thick light brown heterogeneous layer with black,
brown, and yellow inclusions; uneven distribution; voids; irregular
cracking

Finish plaster

Loosely compacted thick white homogeneous layer; quartz inclusions;
uneven distribution

Dark red
paint layer

Bright orange red particles; heterogeneous in size and shape

Stratigraphic analysis of cross section of sample "Ch 95 Mell 1014 5/9 8F" (RD1) exhibiting traces of dark
red paint on its outermost layer.

Sample RD1

Sample RD1 consists of approximately three layers of plaster and one layer of
paint. The edges of Layers 2 and 3 are not well defined. 62 Layers follow the typical light
brown ground/fine white finish sequence although Layer 1 is darker than the finish coat
directly underlying the paint layer. The ground layer is light brown with black, brown
and yellow inclusions. The thick white finish layer is unevenly distributed. The paint
layer on the surface of the sample is made up of bright red-orange pigment particles.
There is no evidence of dirt or soot accumulation between the plaster layers. This may
indicate frequent replastering or cleaning of the wall surface prior to new applications of
plaster.

This may be due to the penetration of the BioPlastic embedding material into the sample.

Chapter 2: Sample Development

Table 3

Cross section examination: Sample RL1

Layer

Type

Description

Ground

Loosely compacted thick coarse light brown layer; brown and black
inclusions

Finish

Compacted fine white homogeneous layer; discontinuous; barely visible

Ground

Loosely compacted thick coarse light brown layer; larger brown and
black inclusions

Finish

Compacted fine white homogeneous layer; discontinuous; barely visible

Ground

Compacted light beige heterogeneous layer; brown, yellow, black
inclusions

Finish

Compacted fine white homogeneous layer; very few black inclusions;
discontinuous

Ground

Compacted thin light beige homogeneous layer; black inclusions;
discontinuous

Finish

Compacted fine white homogeneous layer; very few black inclusions;

discontinuous

Ground

Compacted thin light beige homogeneous layer; black inclusions;
discontinuous

Finish

Compacted fine white homogeneous layer; very few black inclusions;
discontinuous

Ground

Compacted thick light beige heterogeneous layer; brown, yellow, black
inclusions; significant losses due to cross sectioning

Void

Discontinuous separation between layers

Ground

Loosely compacted thick coarse light brown layer; brown and black
inclusions

Finish

Dense, discontinuous off-white finish layer; very few inclusions

Paint Layer

Deep red-orange and black particles; heterogeneous in size and shape

Stratigraphic analysis of cross section of sample "Ch 95 Mell 1014 5/9 8F" exhibiting traces of light red
paint on its outermost layer.

Sample RL1

Sample RL1 consists of approximately thirteen plaster layers, one horizontal void
indicating detachment, and one dark red paint layer. Micro cracking occurs vertically
throughout the sample. Significant losses occurred during cross-sectioning preparation.
The upper-most layers, 1 1-14, just below the paint layer are difficult to distinguish due to

Chapter 2: Sample Development 39

damage and losses incurred during the sample cutting process. Layer 13, a loosely
compacted light brown ground layer, is the most heterogeneous, with black, yellow,
brown, and quartz inclusions of varying sizes. The finish layers 6, 8, and 10, are very
white with very few inclusions. They are clearly delineated from the ground plasters.
Layers 2 and 4 are extremely thin discontinuous finish plasters, barely visible even under
magnification. The surface paint layer is made up of very fine deep red-orange and tiny
black particles covered with a discontinuous layer of dirt.

Chapter 2: Sample Development

Table 4

Cross section examination: Sample BL1

Layer

Type

Description

Ground

Loosely compacted, thick light brown heterogeneous layer with black and
brown inclusions; discontinuous; very large voids

Ground

Loosely compacted, thick light brown heterogeneous layer with black and
brown inclusions; discontinuous; large voids

Ground

Loosely compacted thick light brown heterogeneous layer with black,
brown, and yellow inclusions; discontinuous; large voids

Ground

Loosely compacted off-white heterogeneous layer with black, brown, and
yellow inclusions; coarse particles; discontinuous; large voids

Ground

Compacted, thick light beige homogeneous plaster layer, finer grained;
yellow and brown inclusions; black inclusions appear to be charcoal;
discontinuous; large voids

Finish

Compacted, off-white homogeneous plaster layer, brown and black
inclusions

Finish (?)

Compacted, off-white, homogeneous fine-particled plaster; very few
black inclusions

Paint Layer

Very fine opaque black particles; heterogeneous in size and shape

Stratigraphic analysis of a cross section of the sample called "sheena wall black" exhibiting traces of red
paint on its outermost layer.

Sample BL1:

Very little black pigment remained on Sample BL1. Cross section examination
revealed seven plaster layers and one fine fragile black paint layer consisting of very fine
opaque particles. Layers 6 and 7 are densely compacted homogeneous finish layers.
Cracking occurs vertically throughout the sample. Large voids and damage incurred in the
sample cutting process, are visible mainly in layers 1-5, heterogeneous discontinuous
ground layers that are almost indistinguishable from one another.

Chapter 2: Sample Development

Fig 10 Photomicrograph of cross section of "Sheena Wall Black" (BL1 )
sample showing traces of black paint on surface

','■> t

VfljB

Fig II Photomicrograph of a
cross section "Ch 95 Mell 1014
5/9 8F" (RD1) sample showing
traces of dark red paint on surface

Fig. 12 Photomicograph of a cross
section "Ch 95 Mell 1014 5/9 8F"
sample showing traces of light red
paint on surface (RL1 )

Chapter 2: Sample Development 42

2.2.2 Media Characterization
Summary:

The paints at Catalhoyiik appear to have been film-forming materials composed of
pigment and binder. Pigmented layers applied over a white plaster finish coat exist as
separate layers, apparently applied on dry plaster. Based on microscopical examination of
samples in cross section; the layers range in thickness from 0.25 mm to 0.5 mm, and have
weak cohesion and adhesion. Considering their appearance in cross-section and
archaeological precedents, it is possible that these paint layers were bound with an organic
binding medium.

Organic compounds have been used to strengthen colorants and adhere them to
surfaces for thousands of years. These compounds, usually of human, animal, and vegetal
origin, were used as vehicles or binders in paints, mortars and plasters to increase cohesive
and adhesive properties. Mediums and binders believed to have been used prehistorically

for painting include water, urine, blood, eggs, and hot animal fat

1 The function of binding media is to bind the pigment particles into a workable mass and adhere them to
the substrate. Historically, organic binders may be grouped into four main categories: lipids, oils and fats;
proteins, such as, egg, casein and animal glues; glycerides or sugars; and empyreumatic materials, such as
tars and bitumens.

M For further discussion, see R. J. Forbes, Studies in Ancient Technology, Vol. Ill (Netherlands: Brill.
1965) and L.J. Majewski, "The Conservation of Wall Paintings in Archaeological Excavations"
Preservation and Reproduction of Clay Tablets and The Conservation of Wall Paintings. Colt
Archaeological Institute (London: Bernard Quaritch Ltd.) 24-36.

Chapter 2: Sample Development 43

Detection of organic materials of this age is unlikely due to the natural
decomposition of organic binders. 65 Nonetheless, preliminary investigation for the presence
of organic binding media was conducted using simple microscopical techniques.

Direct reactive staining is a simple laboratory method for characterizing the most
commonly occurring binding media, notably glues, oils, egg, and gums. The presence of
proteinaceous binding media such as gelatin, casein, and egg, may be detected by staining
cross-sections with amido black, an acid stain that reacts with basic proteins. Staining with
a solution of oil red may identify oils.

Ultra-violet fluorescence microscopy was selected as a preliminary method for
examining cross-sectional samples, both to enhance the sample's stratigraphy and to
indicate the presence of organic binding materials. 66 It was understood that the age of the
sample and the limitations of the technique made it unlikely that this method would yield
results.

' Disintegration or decomposition of a binding medium due to age or exposure to the elements occurs when
the medium becomes powdery and falls away from the paint film. The resulting loss of cohesion and
increased exposure of the pigment particles often results in the formation of a friable powdery surface. In
addition to the decomposition processes that commonly occur with age, molecular degradation may be
intensified by salt migration and exposure to ultraviolet light. For further information on the composition
and properties of binding media, see Liliane Masschelein-Kleiner, Ancient Binding Media, Varnishes and
Adhesives (Rome: ICCROM, 1985), and Adam Karpowicz , "Aging and Deterioration of Proteinaceous
Media," Studies in Conservation, Volume 26, Number 4, November 1981, 153-160.

Fluorescent materials emit visible light of longer wavelength when exposed to shorter wavelengths of
ultra-violet and blue radiation. Dyes originally selected for biological applications may tag broad classes of
organic compounds to produce secondary fluorescence. Based on research in the past two decades, a group
of fluorescent dyes have been selected for characterizing organic binding materials for application in the
fields of fine art and architectural conservation. Although useful for preliminary study, confirmation is often
required. Additional research on this subject is recommended. More appropriate tests include: FTIR, GC-
NIS, and HPLC. For further information, see Mortimer Abramowitz, Fluorescence Microscopy. The
Essentials, Volume 4 (United States: Olympus America, Inc., 1993); and Richard Wolbers and G. Landrey,
"The use of direct reactive fluorescent dyes for the characterization of binding media in cross-sectional

Chapter 2: Sample Development 44

Objective:

The aim of this study was to indicate the presence of surviving organic media in
paint layers and plasters through direct reactive staining and fluorescent staining of cross-
sectional samples.
Methodology:
Direct reactive staining:

Two unpolished cross-section samples of each painted plaster, RL1, RL2, RD1,
RD2, Bl 1 , B12, were analyzed for the possible presence of organic binding media. Based on
the knowledge of ancient binding media, samples were tested for the presence of fatty
materials, sugars, and proteins. The cross sections were first examined under low
magnification in reflected light. 67
Fluorescence microscopy:

Two cross sections of each of the three painted plaster samples were studied using
two different fluorescent dyes, triphenyl tetrazolium chloride (TTC), for the presence of

examinations," in AIC Preprints. American Institute for Conservation 15 lh Annual Meeting, Vancouver,
British Columbia, (Washington, D.C.:AIC, 1987) 168-202.

' 7 The samples were not polished. A solution of amido black, AB1, composed of lg of dye, 450ml glacial
acetic acid, 450ml 0.1M aqueous sodium acetate, and 100ml glycerin, will stain the sample blue in the
presence of egg yolk protein. The sample was immersed in the solution for five minutes and washed with
5% acetic acid. Another solution of amido black, AB3, composed of lg dye, 900ml H 2 0, and 100ml
glycerin, will stain the sample blue in the presence of gelatin. The sample was immersed in the solution for
five minutes and rinsed with \% acetic acid. Oil red, composed of 6 ml of a stock solution: 0.5 g dye in
100ml isopropanol, and 4 ml H 2 0, turns a sample pink in the presence of triglycerides and cholesterides.
The sample was soaked for six minutes in pure isopropanol. and then immersed in the stain for an additional
ten minutes. The sample was then washed three times, first with H 2 0, then with 607c isopropanol, and
finally with distilled H : again.

Chapter 2: Sample Development 45_

gums; and fluorescein isothiocynate (FITC), for proteins such as animal glues. They were
observed using a Nikon Alphaphot 2 microscope with Episcopic Fluorescent Attachment
EF-D.
Observations:

No appreciable results were obtained from either method. It is likely that any
organic binding media still present within the samples has deteriorated.

68 A 4% solution of triphenyl tetrazolium chloride was used to test for the presence of reducing sugars. This
stain is sensitive to the action of reducing agents and in ultra-violet illumination will produce a dark reddish
brown stain if carbohydrates are present. A 0.1% solution of fluorescein isothiocynate was applied to test
for the presence of proteins. A positive bright yellow stain will occur in the presence of egg, animal glues,
or casein.

69 The limitations of this technique are well noted. For example, it has been observed that porous samples
with a low binder content can be penetrated by polyester resin embedding materials. Coated particles
within the sample can inhibit the reaction of stains with binding media. For further information see,
Michele Derrick, Luiz Souza, Tanya Kieslich, Henry Florsheim and Dusan Stulik, "Embedding Paint Cross
Section Samples in Polyester Resins: Problems and Solutions." Journal of the American Institute for
Conservation, Volume 33, No. 3 (Fall/Winter 1994): 227-228.

Chapter 2: Sample Development 46_

2.2.3 Pigment Identification
2.2.3.1 Microchemical spot testing
Summary:

Pigments may be inorganic (mineral) or organic (animal or vegetable) in origin.
Mineral pigments are often derived from oxides, sulphides, carbonates, and sulphates found
in the earth; organic pigments from animals, wood, fruits and plants. Prehistoric pigments
were derived from clays and calcium carbonates for white; and earths, wood, and charcoal
for black, brown, yellow, red and green.

To create laboratory facsimiles, it was critical to characterize the materials,
technology, and structure of the original plasters and paints. For film forming paints, such
as those believed to have been used at £atalhoyuk, both pigments and media must be
considered. The presence of inorganic pigments was investigated because they would be
more likely to survive and because they have been identified in paints in other prehistoric
applications.

Inorganic pigments may be identified through a variety of methods. The most
simple and accessible of these is microchemical spot testing.

70 For further information, see Paolo and Laura Mora and Paul Philippot, The Conservation of Wall
Paintings (London: Butterworths, 1984) 56-73; Rutherford J. Gettens and George L. Stout, Painting
Materials (New York: Dover Publications, Inc., 1966); R. J. Forbes, Studies in Ancient Technology Vol. Ill
(Netherlands: Brill, 1965)210-264.

Chapter 2: Sample Development 47

Objective:

The aim was to identify the pigments found on the wall paintings at Catalhoyuk
using simple laboratory methods, notably microchemical spot testing and polarized light
microscopy. 71
Methodology:

Utilizing simple microchemical spot tests, pigments found on the surface layers of
a selection of samples were analyzed. 72
Observations:

Two samples from the site were found to have traces of red pigment on their
uppermost layers (Ch 95 Mell 1014 5/9 8F: RL2, RD2). One sample showed traces of

71 Previous paint analysis was reported on by Pamela French in a conservation report dating from 1968.
Plaster and pigment samples were preliminarily analyzed using chemical spot tests. Potassium ferrocyanide
was used to detect the presence of iron in a sample showing traces of red-brown paint on its surface. The
same red-brown pigment was tested for the presence of blood. Positive results were obtained in both tests
but further analysis was recommended. Pamela French, "Clay and Paint Samples: Preliminary Testing,"
(n.p, n.d.)

" A tungsten needle was used to loosen and collect tiny fragments of the pigments at the edges of the paint
layers. The resulting powdery samples were transferred to microscope slides. All the pigments were
observed under the stereo microscope. A drop of concentrated HN0 3 was added to each red sample and
then heated to evaporation on a hot plate to bring the particles into the ferric state. Confirmatory tests
proceeded as follows. Additional samples were obtained and applied to slides as before. The pigment
particles were again treated with single drops of nitric acid and heated to evaporation. A drop of HC1 was
then added to each sample and they were heated to evaporation to remove the HNO, The resulting residues
were mixed with a drop of water acidified with HC1. A fragment of potassium thiocyanate (KSCN) was then
added to each sample. The presence of iron is indicated if a pink or red coloration is produced, its quantity,
by the intensity of the color. The black pigment was tested for the presence of calcium and for phosphate.
After heating the sample, the incombustible residue was dissolved in dilute HC1. (NH 4 ) 2 C0 3 was added to
that to form a precipitate. The formation of a white precipitate is a positive indication of calcium carbonate.
The residue was then tested for the presence of phosphate. Pigment particles were placed on filter paper,
moistened with one drop of ammonium molybdate solution and warmed. A drop of benzidene solution was
added and it was held over NH 3 . The formation of a brilliant blue color indicates the presence of
phosphate. Dispersed samples were also prepared and analyzed by polarized light microscopy.

Chapter 2: Sample Development 48

black paint on its surface (Sheena wall black). Microchemical spot testing indicated that
both red pigments are iron oxides. The black pigment appears to be bone black.

Chapter 2: Sample Development 49

2.2.3.2 Polarized Light Microscopy
Summary:

Pigments may be identified by determining their specific optical characteristics
with the polarizing light microscope. Using a combination of plane and cross- polarized
light, one may identify the following characteristics:

■ Relative value of the index of refraction (n > 1 .66; n < 1 .66)

■ Particle size, shape and color

■ Isotropism / anisotropism

■ Birefringence
Objective:

This study aims to confirm the results of microchemical spot testing for pigment
identification.
Methodology:

Pigment samples were removed from the surface layers of a selection of samples
with a tungsten needle. The pigment particles were microscopically separated from other
particles and mounted on glass slides with Cargille Meltmount™ (n 1 .66). Optical
characteristics were observed at 100-400x with a Nikon Optiphot 2-Pol and recorded.
Samples were compared with pigment slide and photomicrographic references.
Representative samples were photographed. (See Figs. 13, 14, 15)

Chapter 2: Sample Development

Table 5

Examination with Polarized Light Microscopy: Pigment particles

Sample

Description

Ch 95 Mell 1014 5/9 8F

Plane polarized light:

■ Minute crystals

■ Heterogeneous in size and shape

■ Average size was 1 pj in diameter

■ n>1.66

■ Deep orange-red with black edges.
Crossed polars:

" Birefringent spherulites

■ Isotropic particles
Result:

The pigment is an iron oxide, most likely red ochre.

Ch 95 Mell 1014 5/9 8F

Plane polarized light:

■ Minute crystals

■ Heterogeneous in size and shape

■ Average size was lu in diameter

■ n>1.66

■ Deep orange-red
Crossed polars:

■ Birefringent spherulites

■ Isotropic particles
Result:

The pigment is an iron oxide, most likely red ochre.

Sheena wall black

Plane polarized light:

■ Coarse, irregular particles

■ Heterogeneous in size and shape

■ Average size was 5u. in diameter

■ n> 1.66

■ Translucent gray to opaque black
Crossed polars:

■ Some gray particles, characteristic of uncharred calcium phosphate
Result:

The pigment is bone black.

Chapter 2: Sample Development 51

Observations:
RL1:

Microchemical testing confirmed that the pigment in the red paint layer was an
iron oxide. Microscopic examination using polarized light and comparison with known
samples and photomicrographic references indicate that the pigment is red ochre, an
earthy form of iron oxide composed mainly of clay and silica. Particles were
heterogeneous in shape and composition and ranged from opaque to orange-red to
transparent in transmitted light. The average size was lu in diameter. In plane polarized
light, pigment particles were orange to deep orange-red with slightly black edges.
Crossed polar examination showed isotropic particles and birefringent spherulites with
refractive indices above 1.66. Examination using polarized light microscopy indicates
that the pigment is probably a hydrated form of iron oxide, such as red ochre.

RD1:

Microchemical testing confirmed that this pigment is an iron oxide. Microscopic
examination and comparison with known samples and photomicrographic references
indicate that the pigment is red ochre. Particles were heterogeneous in shape with an
average size of 1 u in diameter and ranged from opaque to orange-red to transparent in
transmitted light. As in Sample RL1, particles ranged from orange to deep orange-red
with slightly black edges in plane polarized light. Crossed polarization showed isotropic
particles and birefringent spherulites with refractive indices above 1.66. Examination

Chapter 2: Sample Development 52

using polarized light microscopy indicates a hydrated type of iron oxide, such as red

ochre.

BL1:

Microscopic examination and comparison with known samples suggests that the
pigment is bone black, a pigment made by charring animal bones. Particles were
heterogeneous and coarse in shape and were translucent blackish-brown in transmitted
light. The average size was 5u in diameter In plane polarized light, particles ranged from
translucent gray to opaque black. In crossed polarized light, visible particles were gray, a
characteristic of uncharred calcium phosphate. The refractive index varied from n = 1.66
ton > 1.66.

Chapter 2: Sample Development

Fig 1 3 "Ch 95 Mell 1014 5/9 8F" (RD 1 ) iron oxide pigment particles

«. . >.~ •&* fiL I** ++**wL

&VX

™* ■ 3*

Fig. 1 4 "Ch 95 mell 1 1 4 5/9 8F" (RL 1 ) iron oxide pigment particles

Fig 1 5 "Sheena Wall Black" bone black pigment particles

Chapter 2: Sample Development 54_

2.2.3.3 Scanning Electron Microscopy with Energy Dispersive Spectroscopy
(SEM/EDS)

Summary:

The scanning electron microscope may provide information on the morphology,
particle size, shape, and texture of a specimen. It generates a three-dimensional
micrograph by detecting the interactions of an electron beam on the surface of the sample.
As an electron beam scans the surface of the specimen, an image is formed. Low points
on the specimen emit low electron signals, thereby producing a corresponding dark point
in the image. X-rays generated by this interaction are recorded and analyzed by a
computer to assess elemental makeup. Composition may therefore be related to surface
morphology.

The capabilities of magnification, ranging from 20,000 x to 100,000 x, and a
depth of field of approximately 300u, provide a much greater level of examination than
does optical microscopy. The energy dispersive x-ray analyzer may be used to produce an
elemental spectrum and corresponding dot map illustrating the elemental composition of
the sample.
Objective:

The aim of this examination was to confirm the identification of the red pigment
previously indicated by microchemical spot tests and polarized light microscopy.

" For further information, see Dr. Sheldon Moll, "Scanning Electron Microscopy: A Versatile Analytical
Tool," in Technology and Conservation (Spring 1976):24-28; Walter C. McCrone, John Gustav Delly and
Samuel James Palenik, The Particle Atlas, 6 Volumes, ...

Chapter 2: Sample Development 55

Methodology:

A cross-section of sample RD1 embedded in polyester/methacrylate resin, was
coated with carbon to provide a conductive surface, and examined using the JEOL 6400
scanning electron microscope. The sample was observed at several magnifications.
Energy dispersive x-ray spectra were produced to assess the relative quantity of the
elemental components. X-ray dot maps show the occurrence of specific elements. 74
Observations:

Scanning electron microscopy with energy dispersive spectroscopy revealed a
calcium-rich finish layer directly below the paint layer in sample RD1 . The spectrum and
x-ray dot maps representing the paint layer confirm that the red pigment has a high iron
content. Silica is also present in this layer. 75 The pigment is probably hydrated iron
oxide (Fe 2 3 • »H 2 0). (See Figures 16, 17, 18)

Pigment identification, including microchemical spot testing, polarized light
microscopy, and for one sample, scanning electron microscopy with energy dispersive
spectroscopy, indicated that the pigments are iron oxide and bone black. This preliminary
examination, combined with the results of earlier plaster characterization, served to
inform the development of the laboratory facsimiles. 76

4 SEM/EDS examination was carried out at the Laboratory for Research on the Structure of Matter at the
University of Pennsylvania with the help of Rollin Lakis and Xue-Qin Wang.

? The presence of silica is not uncommon in iron oxide pigments. Some natural red oxide pigments contain
as much as 70% iron oxides and 25% silica.

These findings should be confirmed by additional pigment analysis on a wider selection of samples
during this campaign. However, for the production of mural painting facsimiles, only those pigments
identified in this study were used.

Chapter 2: Sample Development

Image: MDR: SCREENO . IM LUT . : MDR : SCREENO . LX
16 grey levels - fine. TYPE: Monochrome.

Fig. 16 X-ray dot map showing elemental composition of the red paint layer (Sample RD1 )

"^ ;'%*4^JL^

Finish plaster

Paint layer

'IT J** *

Fig 17 Photomicrograph of sample RD1: 600 x

Chapter 2: Sample Development

X-RAY: 0-20 keU Window : UTW

Live: 60s Presets 60s Remaining: Os

Real: 72s \7V. Dead

C°

Sp c c

I'ltteiU'r

< -.0

FS= SK

1EM1 : B PRINT LflVER

K j C T

niimiHMftiiwiintftwff^mfWWrfffiillll

5.033 keU

ch 264=

10.2 >
1H0 cts

Fig. 18 EDS spectrum of the red paint layer (Sample RD1)

Chapter 2: Sample Development 58_

2.3 Creation of Sample Prototype

2.3.1 Materials for Laboratory Facsimiles

Summary:

In order to develop and test methods and materials for the detachment of the wall

paintings at Catalhoyuk, a sample prototype was developed. Extensive empirical

testing of materials and techniques preceded its development.

All preliminary testing of materials and methods was carried out on samples made
with plasters applied to gypsum board substrates. These were replaced by 12-by- 12-inch
terra cotta panels to more closely represent the character of the original mud brick
substrate.

The basic formulation for the facsimile plasters was determined by nine
characterization techniques completed in previous research, also at the Architectural

77 Facsimiles were created at the University of Pennsylvania's Architectural Conservation Laboratory.

78 Artist's whiting, a very finely divided powder derived from high calcium, or dolomitic limestone, marble,
shell or chemically precipitated calcium carbonate was used in the initial testing program. Dry ingredients
were mixed with enough water to form medium thick (yogurt-like) consistency. The plaster was brushed on
in Vi millimeter thick layers to 24" gypsum board panels. The application of at least 50 layers of plaster to
20 of these panels resulted in severe cracking. The application of successive layers caused severe alligator
cracking and extensive detachment of nearly all the plaster from the surface of the substrate. Attempts to
simulate a three-millimeter thick preparatory coat resulted in extensive cracking. Trial plasters made from
artist's whiting did not exhibit sufficient levels of cohesive or adhesive strength. Rabbit skin glue and
gelatin were added to the mix in various concentrations but neither added sufficient strength to the plaster.
Hydrated lime, a dry powder derived from the hydration of quicklime with enough H 2 to form a
hydroxide, was substituted for artist's whiting. It imparted sufficient cohesive and adhesive strength to the
plasters to allow for the required build-up of layers. Pre-testing consisted of numerous applications of the
hydrated lime plaster mix using varying proportions. These were applied to untreated gypsum board panels
and panels coated with Bin Primer, a pigmented shellac which prohibits the absorption of water into the
substrate, theoretically diminishing the extent of cracking and detachment caused by loss of water from the
plasters. The new plaster mix was successful, exhibiting minimal cracking. The difference in the
performance between primed and unprimed panels was negligible, demonstrating that the primer was
unnecessary. Further attempts to build up a thick preparatory layer using the hydrated lime/clay/sand
plaster with additions of both rabbit skin glue and straw were still unsuccessful.

Chapter 2: Sample Development 59_

Conservation Laboratory of the University of Pennsylvania. 79 Sieving and wet
gravimetric analysis determined particle size distribution, one of the major factors
determining the character of plasters. Results of dry sieving are presented in Figure 2 1 .
Analysis of the plasters obtained from the site reveal a material composed of
approximately 50-60% calcium carbonate, 40-47% clays and silts and 3-10% sand, a
composition similar to that of the marly, clayey soils commonly found in the region. (See
figures 19 and 20 for acid -soluble content results and particle size distribution
information)
Objective:

The aim of this phase was to select materials and techniques for the creation of
facsimile samples, similar in physical and chemical properties to those used in the
construction of the Catalhoyiik mural paintings.
Calcium carbonate

Lime is the predominant material found in plasters and mortars of most primitive
civilizations. 80 Hydrated lime served as the calcium carbonate component.
Sand

The function of sand in a plaster or mortar is to strengthen it by eliminating or
decreasing the amount of shrinkage during drying. Finer binding particles fill the voids
between the coarse grains of sand. The mixture becomes bound more tightly following

79 For details, Kopelson, "Analysis and Consolidation of Architectural Plasters from Catalhoyiik. Turkey."

80 Ralph Mayer, The Artist's Handbook of Materials and Techniques (New York: The Viking Press, 1970)
340.

Chapter 2: Sample Development 60

shrinkage. Dry sieving indicated that the plasters at Catalhoyuk contain only a very small
proportion of sand-sized particles, suggesting that the mixture was strengthened by the
addition of organic binding components.

Sand, purchased in bulk from a local building supplier, was sieved until an
adequate amount of required grain sizes was obtained.
Clay

Geological studies indicate that smectite, a highly reactive montmorillonite clay,
is the predominant clay in this region. It was difficult to obtain smectite for this study due
to a lack of availability amongst United States suppliers. Therefore, bentonite, a
montmorillonite clay with very similar characteristics, was substituted for use on
laboratory facsimiles. ~ Two forms of the clay, calcium bentonite and sodium bentonite
were tested, and eventually used for the prototype mixtures. 83

The performance of the montmorillonite clay is critical to determining appropriate
conservation treatments. The following section discusses the formation, characteristics,
and properties of clays in order to explain their importance in the original plasters.

Traces and impressions of vegetal stabilizers in certain plaster components indicate plant matter as at
least one of the organic materials used in the production of the plasters.
" Bentonite clays were obtained from the Black Hills Bentonite Company in Mills, Wyoming.

High sodium content bentonite, unlike the low sodium calcium bentonite, is film forming. It is highly
reactive to water, and swells. The ability of sodium bentonite to adsorb a considerable quantity of water is
due to the substantial internal surface area of clays in the montmorillonite family. Calcium montmorillonite
may swell anywhere from 45%- 145% in water. Although this is much more striking than the reaction
between water and other clays, such as kaolinite, which increases only 20% in volume, it is still far less
reactive than sodium montmorillonite, which swells from 1400%-1600% in water. Plasters made from
sodium bentonite exhibited more shrinkage cracking in both ground and finish layers than those made from
calcium bentonite. Failure to build up a thick preparatory coat with either of these clays resulted in the
decision to omit the layer completely. The addition of organic materials, such as rabbit skin glue to several

Chapter 2: Sample Development 61_

2.3.1.1 Formation and characteristics of clays

Clays are essentially a weathering product of the disintegration and chemical
decomposition of igneous rocks and some types of metamorphic rocks formed when
silicate minerals interact with water by hydrogen exchange, a process based on the
exchange of cations for hydrogen ions. Clays may form at temperatures ranging from
4°C to approximately 400°C in a time span ranging anywhere from hours to millions of
years. Clay formation, the hydration of solids, occurs slowly at the earth's surface
causing some of the material to alter at a different rate. This results in a soil composed of
a combination of clay minerals formed in place, old partially altered minerals from the
parent rock, and original minerals in an unaltered state. Some clays, such as those
obtained from the northern United States bentonite beds, were formed by the

transportation and deposition of volcanic glass in an aqueous environment.

Beginning in the nineteenth century, clay minerals were defined by their crystal
size, as fine-grained minerals whose particle diameters were less than 0.002mm or 2pm
effective spherical diameter, a parameter based on resolution limitations of the
petrographic microscope. The development of X-ray diffraction has allowed for the
identification of different mineral species found in the <2pm-grain size. Characteristics
of chemical reactivity in clays are determined by the internal chemical structure of the

of the samples did not significantly alter their properties. All of the samples exhibited extensive cracking
and a near total lack of adherence to the substrate.

84 Bruce Velde, "Composition and Mineralogy of Clay Minerals" in Origin and Mineralogy of Clays
(Germany: Springer- Verlag Berlin Heidelberg, 1995) 5-7.
^ Dorothy Carroll. Clay Minerals: A Guide to Their X-ray Identification (Colorado: The Geological

Society of America, Inc., 1970) 1.

Chapter 2: Sample Development 62

clay mineral. Physical characteristics may be attributed to size and specific crystal shape.
Clays are both physically and chemically active. They attract water molecules, combining
to form pastes, slurries, and suspensions, thereby altering their effective physical particle
size. Clays may become chemical agents of transfer or transformation by taking ions or
molecules onto their surfaces or into their internal structures.

Because of the small particle size of the material, the properties of clay minerals
must be identified by the combination of analytical techniques such as x-ray diffraction,
thermal stability, and infrared spectroscopy. Results of these tests provide a composite
knowledge of the sample.

The small grain size of clay crystals imparts a large surface area in comparison to
the volume of the particle. Because clay minerals are usually sheet-shaped, they have an
even greater surface area than cubic or spherical minerals of the same grain size. Clay
particles may be divided into three groups based on their shapes, flakes, laths, and
needles.

The small mineral crystals that form clays attract polar water molecules to their
surfaces through weak charge forces that eventually cover the crystals with several layers
of weakly bonded water molecules. The aqueous solution is thickened, changing its
viscosity. Combination of the mineral particles with water results in the formation of a
plastic material. These sheet-like structures, sometimes called phyllosilicates or layer-
lattice silicates, compact more densely than minerals of other shapes because they can be
stacked tightly in parallel layers with large surfaces facing one another. The sheets are

Chapter 2: Sample Development 63

made up of two structural units, a silicon-oxygen tetrahedron and an aluminum-oxygen-
hydroxyl octahedron. These tetrahedral and octahedrally coordinated cations are bound to
oxygens in a multi-layer structure in which the tetrahedra are linked to an octahedral
sheet.

Swelling clays, such as smectites and bentonites, incorporate water molecules into
their structures. The shape of the clay particle changes as water goes into and out of the
structure making it particularly susceptible to changes in atmospheric conditions. A
humid environment will tend to keep expansive clays constantly hydrated, while a dry
atmosphere will keep it hydrated only occasionally. The volume of a clay particle can
vary by as much as 95% depending on hydration.
Montmorillonites and Smectites

Smectites are clays with a 10-A structure of low charge. This type of clay, known
as a swelling or expanding clay, allows hydrated ions to be absorbed between the layers,
thereby increasing interlayer distances. The presence of hydrated cations increases the
normally 10A unit to either 12.5A for a single water layer state or to 15.2A for a double
water layer state. Soils containing large concentrations of smectites shrink extensively
and crack upon drying.

The group name smectite includes both di- and trioctahedral molecular
arrangements. Montmorillonite is a dioctahedral smectite. Montmorillonites are clay
minerals that have water (H2O) molecules rather than potassium cations (K + ) or
magnesium cations (Mg +-) ions tightly bonding their component layers. They have

Chapter 2: Sample Development 64

alternately been called smectites (British usage), montmorillonoids, or montmorins. They
are dioctahedral expanding aluminous minerals in which two tetrahedral layers are almost
exclusively occupied by silicon. The charge imbalance is caused by divalent ion
substitutions, iron (Fe) or magnesium (Mg), for the trivalent aluminum ions in the
octahedral site. Dioctahedral smectite is formed by the reaction of rocks with solutions
from neutral to alkaline pH. Sometimes dioctahedral smectite concentrates in a restricted
space and forms a bentonite deposit. In such deposits, the smectite occurs as a
monomineralic layer.
Bentonite clays

Bentonite clays are highly colloidal plastic montmorillonite clays that come from
the Cretaceous beds near Fort Benton, Wyoming. They swell to several times their
original volumes when placed in water and will form thixotropic gels even when the
amount of bentonite is small. The term bentonite refers to clays formed by the alteration
of volcanic ash in situ. It referred mainly to the Wyoming material until about 1940,
when it came to connote clays that are highly plastic, colloidal, and swelling, without
reference to a specific origin. They are most often composed of di-octahedral smectite
clay minerals. The composition of the smectite varies from bentonite to bentonite.
Variations may occur within the lattice structure of the smectite, in the occupation of the
octahedral and tetrahedral layers, or in the nature of the exchangeable cations. Most
bentonites carry a calcium cation (Ca ++) as the most prevalent ion. Fewer, most often
from the Wyoming beds, carry a sodium cation (Na + ) as the prevalent ion. It appears that

Chapter 2: Sample Development 65

in addition to the presence of water, bentonite formation may depend upon the presence
of a significant amount of magnesium oxide (MgO) in the volcanic ash, since where
magnesia is lacking, it does not alter to smectite. The alteration takes place concurrently
or soon after accumulation of the volcanic ash rather than later or as part of the
weathering process.

Smectite formation is actually the process of devitrification of the natural glass of
the ash and the crystallization of the smectite. The ash will usually consist of an excess
of silica and alkalis. The bentonites that swell the most are those which carry a sodium
cation (Na + ) as the primary exchangeable cation.

Evidence suggests that at relative humidities below 30%, a calcium
montmorillonite will form a skeletal double layer of water, rather than a single layer at a
thickness of about 4.4A. At relative humidities between 30% and 80%, two full layers
develop with layer thickness increasing to 5.9A. Sodium montmorillonite, dried at room
temperature tends to develop a single water layer between the silicate layers with a

spacing of about 12.5A, while calcium montmorillonite under the same conditions
develop two with a spacing of about 14.5-15.5 A. Sodium montmorillonite will absorb a
far greater volume of water however, at high humidities and in the presence of a large
quantity of water than calcium bentonite. 86

For further discussion of clays, see Dorothy Carroll, Clay Minerals: A Guide to Their X-Ray
Identification (Colorado: The Geological Society of America. Inc., 1970); Ralph E. Grim, Clay Mineralogy
(New York: McGraw-Hill Book Company, 1968); C.S. Ross and E. V. Shannon, "Minerals of Bentonite
and Related Clays and Their Physical Properties" J. Am. Ceram. Soc. 9: 77-96 (1926); and Bruce Velde,
"Composition and Mineralogy of Clay Minerals" Origin and Mineralogy of Clays (Germany: Springer-
Verlag Berlin Heidelberg, 1995).

Chapter 2: Sample Development 66

The chemical composition and physical properties of expansive clays, especially
their hygrophilic character, must be considered in all aspects of study and treatment,
including the development of a sample prototype and selection of treatment methods and
materials.
2.4 Production of Samples

Approximately fifty samples of multi-layered mural paintings provided the means
for testing various aspects of treatment as well as combinations of treatment. Of these
nine samples were reserved for final tests.

The plaster composition was the same for all the sample groups. Each two layer
sequence consisted of a ground layer composed of 50% lime, 40% bentonite, and 10%
sieved sand; and a finish layer composed of 50% lime, 47% bentonite, and 3% sand
mixed with enough water to create a material of medium thick, or yogurt-like consistency
(approximately lml/g).

The first group of samples were unpainted and composed of multiple
ground/finish plaster sequences brush applied to 12-by- 12-inch concrete blocks and 6-by-
6-inch gypsum wallboard panels. 87 These samples allowed the researcher to observe the
response of the plasters to solvents, resins, and adhesives and their methods of
application.

The second group of samples was composed of painted multiple ground/finish
plaster sequences brush applied to 6-by-6-inch and 12-by- 12-inch gypsum wallboard

Chapter 2: Sample Development 67_

panels. This group of samples was used to test treatments and materials. Evaluation was
based on standardized tests and visual assessment of the effects of treatment on the
optical properties of the painted plasters. These additional samples provided an
opportunity to evaluate aspects of treatment prior to testing on the final group of
facsimiles.

The third group of nine samples labeled A-J was reserved for the final testing
program. Several combinations of methods and materials including preconsolidation,
consolidation, and detachment, were tested on this group of samples. Each sample
consisted of seven ground/finish plaster sequences brush-applied to 12-by- 12-inch terra
cotta panels. To ensure uniformity, a sufficient amount of plaster was always mixed to
cover all the samples using the same batch. 88

The first six sequences were made from plaster using calcium bentonite. In order
to represent the occurrence of increased deterioration and interlayer delamination at the
surface observed in in situ plasters, the seventh and final sequence was made from the

more expansive sodium bentonite clay.

87 Plasters executed on concrete block exhibited the most severe conditions of powdering paint and
interlayer detachment and were quite useful in the study of surface consolidants and preconsolidants.

88 All the samples however, did not react similarly. Two to three samples out often inconsistently exhibited
cracking in several of the campaigns. Discrepancies may have occurred due to inconsistent misting
technique. Perhaps insufficient mixing of the plaster led to an uneven distribution of its components. The
swelling, shrinking, and subsequent deformation of the plasters that occurred each time they were misted
caused a loss of adhesion between layers and. at times, between the plaster and the substrate. Voids were
detected below the plaster surface by tapping it lightly. The resulting plaster surface was friable and
powdery to the touch.

89 In preliminary laboratory testing, re-wetting of the ground plaster made from sodium bentonite resulted in
significant weakening of its bond to the substrate. The newly applied finish layer cracked extensively,
drawing with it, the ground plaster from the surface of the substrate.

Chapter 2: Sample Development 68_

Prior to application, the substrates were saturated with several spray applications
of water to prevent them from absorbing moisture away from the plasters. In order to
minimize the extensive cracking associated with rapid drying, samples were covered with
sealed frames after each layer was applied. Covers were left in place for one day, but
lifted periodically to allow for the samples to be misted with water. After 24 hours, the
covers were removed. Samples were misted continuously throughout the second day and
were then permitted to dry for an additional three days. After each layer was dry, it had to
be re-saturated before the next layer could be applied. Murals were applied to each dry
finish layer. Paintings required approximately two days to dry before additional layers of
plaster could be applied.
Wall paintings

Mural facsimiles were copied from some of the Catalhoyiik paintings. On both
preliminary test and prototype samples, motifs from the original murals were painted on
the finish layer of each ground/finish sequence. Since a painting medium has not been
identified, a paint type was selected with similar appearance to the original (matte, thin),
compatible physical and chemical properties to the substrate, water solubility, and
nominal alteration to the substrate. 90

1,0 Watercolor was chosen as the paint most like that found at Catalhoyiik. Cotman watercolors, ivory black,
made from gum arabic, dextrin, and nearly pure amorphous carbon, indian red, made from gum arabic,
dextrin, and natural iron oxide; and yellow ochre, made from gum arabic, dextrin, and synthetic iron oxide,
were used. Watercolor paints consist mainly of transparent pigments ground to an extremely fine texture
added to an aqueous solution of gum arabic binder, which does not undergo any chemical change when it
dries or dissolves. This type of paint dries by the evaporation of the solvent and involves no chemical
reaction which would alter its nature. It may re-solubilize by the addition of more water to the dry material.
Variations in the properties of hardening or solubility require that the binding material and the pigment must

Chapter 2: Sample Development 69_

The researcher attempted to create the most fragile sample type possible without
inducing deterioration through accelerated aging. 91 This was accomplished by applying a
weaker pictorial layer lacking a strong organic binder on a plaster substrate made of
expansive clay. It created a fragile system, similar to that found on site.
Division of final test samples

The group of samples reserved for final testing was divided into three parts. Six
of the samples were set aside to be tested using the most successful treatments identified
during preliminary testing. One sample was kept as a control. The last two samples were
consolidated with ethyl silicates prior to detachment in order to represent research
conducted on the consolidation of earthen plasters and mudbrick.

be combined in proportions appropriate to each specific color. The gum binder is thin, so watercolors tend
to merely stain the substrate rather than form tangible distinct films. For further discussion see, Reed Kay,
The Painter's Guide to Studio Methods and Materials, (New Jersey: Prentice-Hall, Inc., 1983) 128-131.
91 Accelerated aging was not carried out due to time constraints.

9 " Loss of the painted surface occurred due to the saturation of the plaster required before application of
each new layer.

Chapter 2: Sample Development

Acid-Soluble Content: Results

Sample

Mass i (grams)

Mass2 (grams)

Percent Carbonates

Wall plaster 1
(Sample A)

25.16

11.80

53.10%

Wall plaster 2
(Sample C)

24.93

9.97

60.05%

Relief plaster 1
(Sample B)

3.69

1.77

52.03%

Relief plaster 2

33.32

16.00

51.98%

Mud brick
(Sample D)

21.57

18.54

14.05%

Fig. 19. Acid-soluble content of wall and relief pasters Note high carbonate content

From Kopelson. Evan. "Analysis and Consolidation of Architectural Plasters from

Catalhoyiik. Turkey" Master's thesis. University of Pennsylvania. 1996

Chapter 2: Sample Development

G'»v«l

Sand

Coarse tc

medium

Fin*

Sill Clay

U.S. standard sieve sizes
| i
5 8 S 1 i

i i i t t

•0

o CO

c
£ -0

III

T"*- '

III

"if
il

III

II
il

i i
i i

III

i i

::i:

i i
i i

k 1

ii
ji

i
i i

II I

! ii

I HIE

! Ii

III Hi! I

il
il

i Ii

HI III

! Ii

o © ©
Grain diameter, mm

Fig. 20. Particle size distribution of wall plaster primarily composed of silts and clays with very few sand-sized
particles From Kopelson. Evan. "Analysis and Consolidation of Architectural Plasters from Catalhoyiik. Turkey."

(master's thesis. University of Pennsylvania. 1996).

Particle Size Distribution: Dry Sieving Results

Wall Plaster

Relief Plaster

Initial Mass:

101.37 grams

Initial Mass:

102.49 grams

Mass retained (g)

Percent passing

Sieve diameter
(mm)

Mass retained (g)

Percent passing

0.14

99.86 %

2.36

100%

0.13

99.73

1.18

0.05

99.95

0.32

99.41

0.600

0.10

99.85

0.83

98.59

0.300

0.80

99.07

2.28

96.34

0. 150

3.52

95.64

4.35

92.05

0.075

4.71

91.04

2.10

pan

0.54

Fig 21 Results of dry sieving. From Kopelson. Evan. "Analysis and Consolidation of Architectural Plasters from
Catalhoyuk. Turkey." (master's thesis. University of Pennsylvania. 1996).

Chapter 2: Sample Development

Fig 22. Mural painting replica: Finish layer 1

Fig. 23. Mural painting replica Finish Layer 2

Fig. 24 Mural painting replica: Finish layer 3

Chapter 2: Sample Development

Fig. 25. Mural painting replica: Finish layer 4

Fig. 26 Mural painting replica: Finish layer 5

Fig 27: Mural painting replica: Finish laver 6

Fig 28 Mural painting replica: Finish layer 7

Chapter 3: Testing Program

This phase of research aimed to inform the on-site treatment of the murals at
^atalhoyiik. Two phases of intervention were determined to be essential and became the
focus of research. The first, preliminary methods for stabilization, aimed to treat surfaces
in need of immediate stabilization and to prepare them for removal. The second type of
intervention focused on the development of methods for detaching the mural paintings.
Methods and materials were tested on simulated samples under laboratory controls.
Summary:

This chapter describes the laboratory research developed to inquire into effective
wall painting detachment methods for £atalhoyuk. In six parts, it handles the
development and execution of treatment systems and methods for their evaluation. It
begins with Section 3.1, a description of the condition of the painted plasters on site and
requirements for their treatment. Section 3.2, "Preliminary Testing A" describes the
testing program in which individual materials were considered to treat several distinct
conditions before detachment. Section 3.3, "Evaluation of Preliminary Testing A"
describes the methods by which these individual phases of treatment were assessed.
Because a combination of treatments is often required to combat the numerous problems
encountered by painted earthen plasters following excavation, Section 3.4, "Preliminary
Testing B," describes the program designed to address the compatibility of the successful
treatments indicated in Sections 3.2 and 3.3. Section 3.5 describes the final testing
program, which focused on the most successful treatments identified during preliminary

Chapter 3: Testing Program 75_

testing. Section 3.6 offers an evaluation of the results of the final testing program.
Supporting data and conclusions are located in Chapter 4.
3.1 Introduction

The condition of wall plasters and murals at Catalhoyiik varies depending on
location within the site and degree of exposure. Many paintings protected by overlying
plaster layers or infill remain in good condition. Field and laboratory examination of
exposed paintings however, revealed problems common to archaeological wall paintings,
such as degradation caused by changes in atmospheric pressure or relative humidity
following excavation. The crystallization of salts due to water infiltration, macro-
biological growth, erosion of the plaster surface, decomposition of the paint layer,
cracking of the paint and plaster layers, interlayer detachment, delamination from the wall
support, and surface accretions are problems often identified at archaeological sites.

The plasters and mural paintings at Catalhoyiik show a loss of cohesive strength
within discreet layers and a loss of adhesive strength between individual layers. Most of

93 For further information see, O.P. Agrawal and Kamal K. Jain, "Problems of Conservation of the Wall
Paintings in India" International Symposium on the Conservation and Restoration of Cultural Property
(Tokyo: Tokyo National Research Institute of Cultural Properties, 1984) 31-39; Wesley Bliss, "Preservation
of the Kuaua Mural Paintings" American Antiquity, No. 3, 1948, p. 218-223; H. Hodges, Neolithic wall
paintings: problems of conservation (IIC United Kingdom Group. 1964); Sergio Arturo Montero, "The
Conservation of Archaeological Painting" //; Situ Archaeological Consen>ation, Proceedings of Meetings:
April 6-13, 1986. Ed. Henry W. M. Hodges (Mexico: Instituto Nacional de Antropologi'a e Historia and
California: The J. Paul Getty Trust, 1987) 98-105; Watson Smith. "Kiva Mural Decorations at Awatovi and
Kawaika-a. Papers of the Peabody Museum of American Archaeology and Ethnology" Vol. 37
( Massachusetts: Peabody Museum, 1952) 34-35; Constance Silver. "Architectural Finishes of the
Prehistoric Southwest: A Study of the Cultural Resource and Prospects for its Conservation" (masters
thesis: Columbia University, ) 115-118; and Qi Yingtao, "Studies on Conservation of the Grotto Temples
and Mural Paintings of Ancient Graves in China" International Symposium on the Conservation and
Restoration of Cultural Property (Tokyo: Tokyo National Research Institute of Cultural Properties, 1984)
19-29.

Chapter 3: Testing Program 76_

the delamination occurs in the outermost plaster layers. As excavation continues, and the
requirement for surface stabilization increases, it will be necessary to implement effective
treatments. One of the first aspects of such treatment is application of the appropriate
preconsolidant and/or consolidant.
Function of Preconsolidant:

Since consolidation of the plaster may not immediately effect the paint layer, and
since consolidation of the plaster may not always be feasible, it is necessary to develop
methods for treating surfaces separately. The function of a preconsolidant is to impart
strength to the surface and to enable additional treatment.
Function of Consolidant:

Application of a consolidant is necessary when the primary or secondary bonds
holding a material together have been broken. Dissolution in a solvent carrier allows the
consolidant to penetrate deeply into a material, in this case earthen plaster, and strengthen it
by binding loose particles together. Its main function is to increase intergranular cohesion.

Chapter 3: Testing Program 77_

Function of an Adhesive:

Adhesives may be made of natural or synthetic materials. They are used to treat
cleavage, flaking paint and plaster by filling gaps between pieces, adhering to both surfaces
and providing an adequately strong and rigid interface between pieces. 94

Wall paintings located in rooms that had been burned at the time of occupation
exhibit more complicated problems. Burned paint cracked and fell off the walls. The
paint layers themselves were very thin and prone to disintegration. The resulting heat
sometimes fused the paintings to underlying layers or to the mudbrick wall supports. In
addition to natural deterioration mechanisms, continued excavation of the site places the
remaining paintings at risk of destruction.
3.1.1 Detachment Techniques

Various methods for removing murals have been developed over the past 100-150
years. These primarily European methods, have usually responded to murals executed on
lime renders as opposed to clay-based renders. Three primary detachment techniques
differ by their level of removal. They are stacco a massello, the removal of a painting
with all or part of its supporting wall; stacco, the removal of a painting with its rendering;
and strappo, the removal of the paint layer alone. 9 They provide the reference for
developing removal and reattachment methods for murals on clay supports, such as those
at £atalhoytik.

4 Andrew Wheatcroft, ed., Science for Conservators Volume 3: Adhesives and Coatings (London: The
Conservation Unit of the Museums and Galleries Commission, 1987) 14.

95 A brief history, detailed review of detachment techniques, and related case studies may be found in
Appendix B.

Chapter 3: Testing Program 78

Two types of detachment techniques were used previously at Catalhoyiik. The
"block method," in which portions of the mudbrick wall were removed with the painted
plaster; and the "peeling method," in which consolidated paintings faced with Japanese
tissue, fine linen, and size, are peeled from the support with a portion of mud plaster still
adhering, were used with varying degrees of success. 96 Conservation reports describing
these methods combined with a literature review of related case studies provided the
foundation for a research design.
3.1.2 Requirements of Research

Based on conditions at the site, several specific problems were identified as
requiring immediate treatment before removal. Treatments for these conditions were
tested in the laboratory.
Conditions:

■ Loss of cohesion in paint (powdering paint)

■ Loss of cohesion in surface lime-based renders (powdering plaster)

■ Loss of adhesion between plaster layers (interlayer detachment)

■ Surface cleaving and cracking

■ Loss of cohesion in earthen plasters (disaggregation)

Previous treatments, as described in reports written by Pamela French, dating from 1968-1974 have been
summarized in Appendix A. Pamela French, "Report on work carried out on the murals from Catal Hiiyiik
in the Archaeological Museum, Ankara in September 1968;" "Report to the British Academy on the work
done on the Catal Huyiik Paintings;" "The Catal Hiiyiik Wall Paintings;" "The Continuation of the
Conservation of the Catal Huyiik Wall Paintings undertaken in the Museum of Anatolian Cultures. Ankara,
Summer. 1973;" and, "Report to the British Academy on the Conservation of the Catal Hiiyiik Wall
Paintings, Summer 1973, Spring 1974." unpublished reports.

Chapter 3: Testing Program

In order to address these conditions, certain preliminary stabilization treatments
had to be tested in addition to various types of wall painting detachment. Specific tests
included:
Treatments:

Consolidation of the painted layer

Consolidation of lime plaster

Reattachment between plaster layers

Flattening of cleavage

Selection/application of facing adhesives for mural detachment

Selection/application of detachment methods

Compatibility of treatments.
Treatments were tested on laboratory facsimiles. Tests were selected according to
the treatment objective, the types of material to be treated, and the nature of deterioration.
Past research focusing on detachment methods was considered.

A selection of materials representing traditional and current practice on similar
materials and conditions in the field of conservation were chosen for preliminary study.
This study aimed to determine which, if any, traditional and/or non-traditional materials
would consolidate the plaster and friable paint without altering their optical properties,
permit the detachment of the plaster or paint layer from the substrate; and be retrievable.

Chapter 3: Testing Program 80_

The effects of adhesives, fixatives, preconsolidants and consolidants, alone and in
combination, were observed on facsimile samples made of three to six layers of painted
plaster on gypsum board substrates.
3.1.3 Materials: Adhesives and Consolidants

A selection of materials representing a range of adhesives and consolidants was
considered for testing. They included natural and synthetic materials chosen from seven
groups of adhesives and consolidants: collagen-based adhesives; acrylic resins; acrylic
emulsions; gums; polyvinyl acetate emulsions; polyvinyl alcohol; and waxes. It was
important that the materials selected were affordable, exhibited low toxicity, and did not
present a fire hazard. The performance of each was observed at various concentrations
according to its function. 97 Adhesive mixtures were also studied. The following is a
brief description of each.
■ Thermoplastic acrylic resins: Acryloid B-67

Acryloid B-67 is an isobutyl methacrylate polymer commonly used in the
conservation of paintings and objects. It is durable, flexible, resistant to discoloration, and
to water, alcohols, alkalis and acids. It does not become insoluble or degrade significantly
under normal conditions of exposure. 98 B-67 may be obtained commercially as solids or
in a 45% solution in mineral spirits. For this research, it was tested as both a
preconsolidant and a facing adhesive.

" The functions of surface consolidation, readhesion, and facing adhesion require materials with different
concentrations. For example, since it is not desirable for a facing adhesive to penetrate the surface of a
painting, a higher concentration would be required than if the same material were used to consolidate a
painted surface. See specific tests for most successful applications.

Chapter 3: Testing Program 81

■ Thermoplastic acrylic resins: Acryloid B-72

Acryloid B-72 is an ethyl methacrylate copolymer used for the consolidation of
paintings and objects. In addition to the properties listed above (see Acryloid B-67), its
low reactivity with sensitive pigments made it especially desirable for this project. Its use
in the field of art conservation is well established and documented." Acryloid B-72 may
be obtained commercially in solids and as a 50% solution in toluene. It was tested as a
surface consolidant and as a facing adhesive.

■ Acrylic emulsion: Rhoplex AC-33

During the first phases of testing, Rhoplex AC-33, a low viscosity acrylic resin
emulsion was studied. Under optimum conditions, it is known to dry to a clear firm film
that does not degrade under normal conditions of exposure. For this research, it was
tested as a preconsolidant.

■ Acrylic emulsion: Plextol B500

Plextol B500 is an aqueous emulsion of a thermoplastic acrylic polymer of ethyl
acrylate and methyl methacrylate. It forms a clear colorless film that remains elastic at
room temperature and is soluble in most organic solvents. It has been used as a binder for

98 C.V. Horie, Materials for Conservation (London: Butterworth & Co., Ltd., 1987). 106-109.

A testing program developed to evaluate the performance of seven different surface consolidants on the
basis of specific ideal properties was carried out at the Instituto Centrale del Restauro. Tests were designed
to measure the adhesive and cohesive strength of the treated surface, resistance to abrasion, solubility,
alterations in optical properties, resistance to biodeterioration, and reaction to the effects of accelerated
aging. The materials studied were Calaton CA, Primal AC33 (also known as Rhoplex AC33), Primal AC55,
Acryloid B-72, Lucite 45, commercial white shellac, and Gelvatol 40-20. Acryloid B-72 in a 5% solution
in toluene demonstrated the best overall performance for the criteria. Cited in Constance Silver,
"Architectural Finishes of the Prehistoric Southwest: A Study of the Cultural Resource and Prospects for its
Conservation" (master's thesis, Columbia University, 1987), 145-146.

Chapter 3: Testing Program 82_

pigmented coating and exhibits the desirable properties of resistance to weathering and
aging. It was selected for testing as a facing adhesive.
■ Collagen-based adhesive: rabbit skin glue

Rabbit skin glue is a natural high polymer organic colloid made from mammalian
collagen, the primary protein of skin, bone, and sinew. These collagen molecules are
connected by a few covalent and many hydrogen bonds. Rabbit skin glue is sold in
granule form and reactivated by soaking in water and heating until dissolution. Due to its
molecular configuration, it has the ability to shift easily and repeatedly between viscous
and firm states by simple heating and cooling. These molecules are long and flexible and
take a helicoidal form in solution. The gelatin swells when set in cold water for at least
three hours. It must be heated prior to use to a temperature not exceeding 60°C. As it
cools, the glue passes through a tacky phase before returning to a gel state. It hardens
quickly into a dried film as water diffuses into the substrate and the gel begins to form.
The gelatin eventually returns to its original dry state, its contraction rate relating directly
to the amount of water used to make the solution. 100 Rabbit skin glue was initially
considered as an adhesive.

100 For further information, see: John R. Hubbard, "Animal Glues," in Handbook of Adhesives (New York,
New York: Van Nostrand Reinhold Company Inc., 1977); C.V. Horie, Materials for Conservation (London,
England: Butterworth & Co. Ltd., 1987) 142-143; Arthur H. Landrock, Adhesives Technology Handbook
(Park Ridge, New Jersey: Noyes Publications, 1985), 160; and Liliane Masschelein-Kleiner, Ancient
Binding Media, Varnishes and Adhesives (Rome, Italy: ICCROM, 1985), 66-68.

Chapter 3: Testing Program 83

■ Collagen-based adhesive: gelatin

Gelatin is a proteinaceous organic compound made up of large molecules of high
molecular weight. Like rabbit skin glue, it is an animal product derived from collagen
which appears slightly yellow in its natural dried form and remains flexible due to its
water content. Dried gelatin will swell to many times its normal volume when immersed
in water, dilute acids or alkalis. This colloid which produces highly viscous solutions, is
widely used as a light adhesive for conditions such as flaking paint. 101 Gelatin was
initially considered as an adhesive.

■ Gums: gum arabic

Gum arabic is derived from the Acacia tree found in Africa, India and Australia. It
is a non-crystalline amorphous substance composed predominantly of carbon, hydrogen
and oxygen. Plant gums in general are "salts of complex organic acids, usually with
calcium, magnesium, and potassium." Sugar units combine with the acidic part of the
molecule to form complex acids. They are crushed or ground to a powdered form and
mixed with boiling water before use as a strong adhesive or binder. 102 Gum arabic was
initially selected for use as an adhesive.

■ Polyvinyl alcohol

Polyvinyl alcohol is produced by the hydrolysis of polyvinyl acetate. It is soluble
in water, resistant to light, and uncomplicated to use. Films derived from this resin

10 ' For further information, see Rutherford J.Gettens and George L. Stout, Painting Materials: A Short
Encyclopedia (New York: Dover Publications, Inc., 1966), 25; H.J. Plenderleith and A.E. Werner, The
Conservation of Antiquities and Works of Art (London: Oxford University Press, 1971), 169-170.

Chapter 3: Testing Program 84_

demonstrate strength, flexibility, and resistance to petroleum solvents, oils, and fats.
Various concentrations of polyvinyl alcohol in water were tested to find the most
appropriate for use as a facing adhesive. A 20% solution proved to be the least
concentrated mixture that could be used without penetration of the plaster surface.
Polyvinyl alcohol was tested as a facing adhesive for mural detachment.

Data from initial testing was utilized and expanded upon in section 3.2.5,
Readhesion/Facing Adhesives. Successful treatments from previous tests were combined
to evaluate their compatibility.
■ Waxes: Microcrystalline wax

Microcrystalline waxes are made up of very thin crystals dispersed in an
amorphous, viscous material. The wax is normally prepared by melting in a double boiler
and mixing it with polyterpene resin. It is known for its properties of adhesion and
flexibility. It has been used successfully in conjunction with the application of gentle
pressure and heat to reattach flaking paint and relax cleavage. 1 3 The proportion of
microcrystalline wax to polyterpene resin used in this program for the purposes of
reattaching flaking plaster and paint was 3: 1 . l04 Microcrystalline waxes were tested for
treatment of surface cleavage and flaking.

02 For further information, see Rutherford J.Gettens and George L. Stout, Painting Materials, A Short
Encyclopedia (New York: Dover Publications, Inc., 1966). 28-29.

103 For further information, see H.J. Plenderleith and A.E. Werner, The Consen>ation of Antiquities and
Works of Art (London: Oxford University Press, 1971), 169-170.

Conservation Materials Ltd. "The Keck Formula "in Conservation Materials Ltd. Catalog (Nevada:

Conservation Materials Ltd. nd.), 35.

Chapter 3: Testing Program 85_

■ Polyvinyl acetate emulsion: Vinamul 6825

Vinamul 6825 is a polyvinyl acetate emulsion miscible in water. It is
recommended as a consolidant for non-metallic objects except when penetration is
required. It has been used as a facing material for wall paintings in climates unsuitable
for glue. 105 Product literature recommends a concentration of one part emulsion to three
parts water. Vinamul 6825 was tested for use as a facing adhesive.

■ Traditional adhesive: "colletta"

Colletta is a collagen-based adhesive developed for the detachment of wall
paintings by the methods of stacco or strappo at the Instituto per il Restauro in Rome. °
A variety of concentrations of both the strappo and stacco formulas were tested to
identify the least concentrated solution capable of effectively detaching the painting.
Preliminary stacco and strappo tests utilized the colletta glue at full strength. Lower
concentrations were evaluated in subsequent tests to minimize the strength of the facing
adhesive during removal. Colletta was tested as a facing adhesive for mural detachment.
3.2 Preliminary Testing A: Individual tests

Preliminary research program "A" is outlined in the following pages. Data and
conclusions are located in Chapter 4.

105 Conservation Resources (U.K.) Ltd. Chemical Data Sheet: Vinamul 6825 (6#/5,)(England: Conservation
Resources (U.K.) Ltd., 1992).

The recipe used for the stacco method is a mixture of bone glue soaked in water for twelve hours until
swelled, then heated until fluid, then mixed with molasses which is used as a plasticizer; vinegar, used as a
fluidizer; oxgall, used as a surface active agent; and a small amount of fungicide. Because the strappo

Chapter 3: Testing Program 86_

3.2.1 Test for Visible Alteration of the Plaster Surface
Summary:

Prior to testing a combination of materials, the components of each material were
individually evaluated based on visual assessment.
Objective:

The aim of this test was to observe possible alteration to plasters caused by
solvents, adhesives, and consolidants.
Methodology:

A simple testing program was devised to observe the visual effects of adhesives,
solvents, and consolidants on the surface of the facsimile plasters. This test was
conducted on one 6- by-6- inch unpainted multi-layer sample on gypsum board. Twine
and tacks created a sixteen square grid. Several drops of the following materials were
applied to separate cells within the grid: water, ethanol, isopropanol, acetone, rabbit skin
glue (57c in water), Acryloid B-72 (7% and 10% in xylene), and Aquazol50® (5% and
10% in water, ethanol, and isopropanol), a 50/50 preconsolidant mixture of Aquazol50®
(5% in ethanol )/T- 1919®, and pure T- 1919. Results, such as darkening, cracking, and
gloss were observed and recorded in Table 6.

method depends upon contraction of the glue to detach the painting, the plasticizer is omitted from the
mix..

Chapter 3: Testing Program 87

Observations:

Undesirable results such as the formation of a glossy film, discoloration of the
surface, and cracking, indicated that rabbit skin glue, Acryloid B-72®, and Aquazol 50®
required further observation before they could be considered for the final testing

107

program.

1 One of the biggest challenges in the consolidation of matte paint is the preservation of the optical
properties of the painted surface. Darkening of the surface due to the formation of a glossy film is a
common problem. One solution is to treat the sample using multiple applications of dilute solutions of the
consolidating agent. For further information, see Eric F. Hansen, Rosa Lowinger, and Eileen Sadoff,
"Consolidation of Porous Paint in a Vapor-Saturated Atmosphere, A Technique for Minimizing Changes in
the Appearance of Powdering, Matte Paint," in Journal of the American Institute for Conservation, Volume
32, No.l (Spring 1993): 1-14.

Chapter 3: Testing Program 88

3.2.2 Test for Surface consolidation
Summary:

The plasters and paint layers at Catalhoyuk are detached and there is loss of
adhesion between paint and plaster layers. The paints have also lost cohesive strength.
These conditions may require preliminary treatment aimed at readhering separated layers,
reattaching paint and flaking plasters, relaxing cleavage, and reestablishing cohesion within
the paint layers.
Objective:

The aim of surface consolidation testing was to identify a material that will
strengthen the surface of the plaster while preserving its original appearance.
Methodology:

Adhesives were brushed onto the surface of a preliminary test sample (an
unpainted multi-layer plaster sample on concrete block) through Japanese tissue. The
condition of the plaster was comparatively stable prior to testing. Changes in surface
appearance are noted in the table below. Results were qualitatively assessed.
Observations:

All of the adhesives, except gelatin in a 5% solution in water were successful at
decreasing the friability of the plaster surface. Undesirable effects however, such as
discoloration and residue sometimes accompanied the strengthening. Aquazol 50® in 5%
solutions in ethanol and water, and Acryloid B-72 in a 5-15% solution in toluene attained
the most favorable results. Data is located in Table 7.

Chapter 3: Testing Program

Fig. 29. Results of test aimed to strengthen the surface of the plaster while preserving its original appearance

Chapter 3: Testing Program 90

3.2.3 Test for Surface Consolidation and Preconsolidation with Readhesion, Part I
Summary:

Detachment of the plaster layers at Catalhoyiik occurs most often in outermost
layers, farthest from the mudbrick substrate. This condition appears to result from salt
crystallization, macro-biological growth, and/or changes in the ambient atmosphere during
excavation. Treatments were needed to preconsolidate the plaster surface and to readhere
delaminating layers.
Objective:

The aim of this test was to identify materials capable of increasing the cohesive
and adhesive strength of the plaster surface without alteration of its optical properties and
to improve interlayer attachment between plaster layers exhibiting cleavage and

108

separation.
Methodology:

Adhesives were brushed through Japanese tissue onto an unpainted 6" gypsum
board panel coated only with a layer of sodium bentonite mixed with water. Prior to
treatment, the clay plaster exhibited severe shrinkage cracking with extensive cleavage
and cupping. Areas that swelled during treatment were covered by silicone release mylar
and manipulated by lightly rolling over the surface with a glass stirring rod. Pressure was
then applied by weighting them with 2-inch marble cubes. Reaction to treatment and
changes in surface appearance and condition are noted in the following table. Weighted

" These tests were conducted in conjunction with one another.

Chapter 3: Testing Program 9\_

samples have been identified as such. Results were qualitatively assessed and are listed

in Table 8.

Observations:

Three of the nine treatments produced acceptable results. Aquazol 50® in a 5%
solution in water strengthened weak friable surfaces. Gelatin in a 5% solution in water
partially relaxed cleavage and promoted readhesion with only a slight darkening of the
surface. Water was applied to two different test cells. In the first case, it demonstrated the
best performance by relaxing cleavage, promoting readhesion and decreasing friability
without visual alteration of the surface. The second application however, began to
dissolve the clay film. Cleavage was partially relaxed but began to cup again while
drying. Test results indicated that further study was required.

Chapter 3: Testing Program

Aquazol 50
(5% ethanol)

RhoplexAC-33
(1%-H,0)

Aquazol 50
(5%-H,0)

's~:

• $ r -^ r ~'i *V-< fM-*

> -:-cw

*&■*

Gum arable
(10%-H,O)

Aquazol 50
( 1 0%-isopropanol )

Gelatin
(5%-H,0)

Gum arabic
(5%-H,0)

H :

Fig 30 Results of test for surface consolidation and
prcconsolidation with readhesion (Part I)

Chapter 3: Testing Program 93

3.2.4 Test for Surface Consolidation and Preconsolidation with Readhesion, Part II
Summary:

This test continued the aims of Preconsolidation and Surface Consolidation with
Readhesion, Part I.
Objective:

The aims of this test were to strengthen the plaster surface while preserving the
original appearance and to improve interlayer attachment between plaster layers exhibiting
cleavage and separation.
Methodology:

Treatments were applied to a 12-by- 12-inch multi-layer painted concrete block.
The condition of the plaster prior to testing was quite poor, exhibiting delamination from
the substrate, interlayer separation and a very friable surface. Adhesives were brushed
onto the surface of the sample through Japanese tissue. Samples that swelled were
covered with silicone release Mylar and manipulated slightly by rolling lightly over the
surface with a glass stirring rod. They were then weighted with 2" marble cubes.
Changes in surface appearance and condition are noted in Table . Results were assessed
in Table 9.

Chapter 3: Testing Program 94

Observations:

Aquazol 50® in 5% solutions in water and ethanol and gelatin in a 5% solution in
water decreased friability and promoted adhesion but caused a slight darkening of the
plasters. The use of water again produced positive results by increasing readhesion to the
substrate.

Chapter 3: Testing Program

(5% in HO)

5% \nW0f.% . ' (0.5% in H 2 0)

'/xquazol 50
(5% in H,0)

Gelatin

<5%i*p 2 0)

Aquazol 50 . Aqu

(5% in ethanol)' ' .J!2M*^

Untreated

Fig. 3 1 . Result of test designed to assess surface consolidation,
preconsolidation, and readhesion treatments (Part II)

Chapter 3: Testing Program 96

3.2.5 Test for Readhesion/Facing Adhesives
Objective:

This test considers the effects and relationship of two treatments. The first
addresses adhesion between detached plaster layers and between the plaster and the
substrate. The second treatment, which relies on the success of the first, concerns attaching
a tissue facing to the surface without penetration of the adhesive into the plaster.

The selection of a facing adhesive depends upon several factors: resistance of the
paint layer to water or to other solvents required for the detachment procedure, the degree of
humidity of the environment in which the work is to take place, the method of detachment,
the reversibility of the adhesive, and the state of preservation of the wall painting. Because
the paint layer was weak and water- soluble, it was necessary to improve its resistance.
Evaluation of facing adhesives was carried out to identify a reversible adhesive capable of
adhering a tissue facing to the plaster surface without penetration.
Methodology:

The following tests were carried out in conjunction with one another on a 24-by-
24-inch painted multi-layer gypsum board panel (D) divided into nine treatment sections.
The condition of the plaster prior to treatment was extremely poor, exhibiting severe
cracking, interlayer separation, and extensive delamination from the substrate. Two
additional conservation materials were studied, microcrystalline wax, and Plextol B500.
The first test area, Dl, was treated with an application of Acryloid B-72 in xylene. After
that, it was divided in half and treated as Dl A and DIB, respectively. Section D2 was

Chapter 3: Testing Program 97

treated with a spray application of water. Following that, it was divided into D2A and
D2B. The remainder of the sample sections was treated as single units. Treatment
details, assessed qualitatively, are described briefly in Tables Dl through D9.
Observations:

The use of Acryloid B-72 as a surface consolidant prevented losses in the paint
layer previously caused by the application of water and other aqueous substances used for
preconsolidation. Darkening of the plaster however, indicated the need to experiment
with more dilute solutions. The combination of B-72 as a surface consolidant with water
or Aquazol 50® in a 5% solution in water produced encouraging results. Readhesion was
improved, cleavage was relaxed, and both treatments allowed for manipulation of the
surface without damage to the plaster. Results of tests using microcrystalline wax and
Plextol B500 were unsatisfactory. Incomplete penetration of the wax resulted in a
darkened film left on the surface of the plaster. Plextol B500 was brushed through
Japanese tissue. Attempts to remove this tissue caused full-scale detachment of the
plaster, indicating that the adhesive, although inappropriate for use as a preconsolidant,
might prove useful as a facing adhesive in later detachment tests. Results are located in
Tables 10 and 11.

Chapter 3: Testing Program

Fig. 32 Results of test considering the treatment of detached plaster layers
and pieces in conjunction with the attachment of facing adhesives.

Chapter 3: Testing Program 99

3.2.6 Test for Surface Consolidation of Powdering Paint with Consolidation and

Readhesion

Summary:

Surface consolidation may be required to arrest cohesion loss, flaking, or other
types of deterioration of a painted surface. Ideal properties of a surface consolidant are:

■ The material must be capable of quickly bonding loose particles or flakes of paint or
rendering to the support.

■ Penetration must be sufficient to establish a bond to the underlying layers

■ Residue must be soluble and removable

■ Flexibility should be sufficient to withstand mechanical shock.

■ The material should be clear and should not alter the optical properties of the plaster
or painting.

■ Any surface treatment should offer resistance to biological and atmospheric
deterioration as well as to static electricity and the accumulation of dust.

■ Ideally, the consolidant should remain soluble in case removal is required in the
future/ 09

Objective:

The purpose of the test was to treat the loss of adhesion and cohesion of the paint
layers without altering the optical properties of the plaster or painting.

Paolo and Laura Mora, Paul Philippot, The Conservation of Wall Paintings (London: Butterworth's,

1984), 217-222.

Chapter 3: Testing Program 100

Methodology:

The following test was carried out on three 12-by- 12-inch painted multi-layer
gypsum board panels. The aim was to determine whether the plaster could be
consolidated after the application of a surface consolidant. The materials were either
brushed on or spray applied as noted in the following tables. After treatment, the surface
was weighted. Results are described in Tables through .
Observations:

The application of Acryloid B-72 as a surface consolidant prevented losses in the
paint layer without disturbing the effectiveness of the preconsolidant treatment.
Applications of both water and Aquazol 50® (5% in water) facilitated flattening of
cleavage and readhesion of small pieces of plaster to the substrate. In addition, water
followed by the application of pressure appeared to improve compaction of the plasters
by decreasing voids between separated layers. Aquazol 50® (5% in water) strengthened
the bond between plaster pieces. The application of Aquazol 50® (5% in isopropanol)
facilitated flattening of cleavage and strengthened the bond between plaster pieces,
however, it caused a yellow discoloration of the surface. Results of the combination of
Acryloid B-67 and Acryloid B-72 treatments were unacceptable because although the
crumbly plaster layer appeared stronger, the surface was significantly darkened and
delaminated pieces were not readhered to the substrate. Data is located in Tables 12, 13,
and 14.

Chapter 3: Testing Program

101

Fig. 33. Combination of treatments. Top row: Acryloid B-72 (8% in xylene) x 3, followed by water.
Bottom row: Acryloid B-72 (3% in xylene) x 2. followed by AquazolSO (5% in water)

Fig. 34. Combination of treatments Acryloid B-72 (3% in xylene) x 2. followed bv water

Fig. 35. Combination of treatments Acryloid B-67 caused discoloration of the lower left hand corner

Chapter 3: Testing Program 102

3.2.7 Facing Adhesives: Methods and Materials - Preparation for Stacco and

Strappo

Summary:

The selection of facing adhesives was based on the conditions previously outlined
in "Readhesion/Facing Adhesives." Requirements for removal include removal with
minimal risk to the painted surface and easy dissolution. The following tests assess the
ease of facing and residue removal of traditional and non-traditional adhesives.
Objective:

The aim of this study was to adhere a tissue facing to the surface of a multi-layer
painted plaster sample, permit it to dry, and then remove it without causing damage or
alteration of optical properties. These tests determined the reversibility and suitability of the
following facing adhesives for stacco and strappo:

Colletta: full strength

Polyvinyl alcohol: 20% in water

Acryloid B-72: 20% in xylene

Acryloid B-67: 20% in mineral spirits

Plextol B500: 50% solids

Vinamul 6825: 50% solids

Chapter 3: Testing Program 103

Methodology:

Six facing adhesives were applied to six 6 x 6 inch painted multi-layer panels. Two
of the adhesives, colletta and polyvinyl alcohol, are water-soluble. ' Two thermoplastic
acrylic resins, Acryloids B-67 and B-72, are soluble in hydrocarbons. A layer of adhesive
was brushed onto the surface of the plaster. The tissue was applied directly to the glue-
coated surface and another layer of adhesive was brushed on top of that. The adhesives
were permitted to cure for approximately seventy-two hours. Then, a poultice was applied
to soften the adhesives in preparation for their removal. Residue was removed using the
appropriate solvent and cotton swabs or pads. Results are outlined in Tables 15 and 16.
Observations:
Colletta:

Applied to the plaster at full strength, colletta formed a strong bond between the
plaster and the tissue facing within minutes. Contraction of the glue caused partial
delamination from the substrate. Poultice applications and blotting of the plaster surface
with pads soaked in hot water to remove the residue of the adhesive caused severe
swelling of the clay plasters. Mechanical removal of the residue was only partially
successful. Colletta was embedded between the brushstrokes of the plaster. Complete
reversibility through poultice application and mechanical removal was impossible. Both
pigment and plaster losses were substantial. The surface was slightly discolored.

Chapter 3: Testing Program 104

Acryloids B-67 and B-72:

20% solutions in mineral spirits and xylene were very viscous." ' Results of the
removal of the facings applied with thermoplastic acrylic resins were poor. The panel faced
with Acryloid B-72 required the application of xylene poultices for six hours before the
facings could be removed. The procedure caused minor swelling of the clays and some
separation between those layers where deterioration already existed. Residue removal was
unsuccessful and a great deal of effort produced little result. Some pigment loss occurred.
A residual glossy film caused a darkening of the plaster surface even after two additional
poultice applications.

The facing adhered with Acryloid B-67 took approximately two hours longer than
the others to soften to a degree which allowed safe removal of the tissue. Even after eight
hours, removal resulted in the loss of fragile layers. The surface remained extremely
tacky, residue removal was difficult and for the most part, unsuccessful. Residue left on
the surface remained unchanged after two additional poultice treatments. Some pigment
was lost due to attempts to mechanically remove the adhesive. The plaster remained
darkened with a uniform glossy surface film.

110 Colletta was made in bulk and stored following recommendations in Paolo and Laura Mora, Paul
Phillipot, The Conservation of Wall Paintings (London: Butterworth's. 1984). 347.
The acrylic resins took nearly twice as long as the colletta to hold the tissue in place.

Chapter 3: Testing Program 105

Plextol B500:

The adhesive swelled but did not resolubilize after the application of solvents. Full-
scale delamination of the plaster layer from the substrate occurred during attempts to
remove the stiffened facings.
Vinamul 6825:

A methanol poultice facilitated removal of the Vinamul 6825 facing but a
discontinuous residue remained, unacceptably darkening the surface of the plaster.
Polyvinyl alcohol (PVOH):

A 20% solution in water was less viscous than the acrylic resins and did not take
as long to bond with the tissue facing. Removal of the polyvinyl alcohol facing adhesive
was easily carried out by dabbing the surface with a cotton pad dipped in water and
wrapped in gauze to prevent fibers from catching on irregularities in the surface texture of
the panel. Clays swelled only minimally. No permanent deformation or delamination
took place. Minor pigment loss occurred. The plaster was darkened very slightly but no
adhesive film remained visible on the surface.

Of these treatments, the most successful was the polyvinyl alcohol adhesive.
Facing and residue removal, easily accomplished using a poultice and water-soaked pads,
resulted in only a slightly darkened surface. Negligible pigment and plaster loss occurred.
Colletta treatment resulted in detachment of plasters from the substrate due to contraction
of the glue. Although this condition made reversal of the facing adhesive more difficult

Chapter 3: Testing Program 1Q6

for these tests, it is the desired outcome when performing an actual detachment. Further
testing using lower concentrations of the adhesive was indicated.

Chapter 3: Testing Program

107

Fig. 36 Facing adhesive tests

Fig. 37. Drying colletta

Fig. 38. Facing adhesive tests

Fig. 39 Heating colletta

Chapter 3: Testing Program

108

Fig 40 Sample faced with colletta and
Japanese tissue

Fig. 4 1 . Sample faced with polyvinyl alcohol and
Japanese tissue

Fig 42. Sample faced with Acryloid B-67 and
Japanese tissue

Fig 43 Sample faced with Acryloid B-72 and
Japanese tissue

Chapter 3: Testing Program

109

Fig 44 Appearance after removal of colletta facing

Fig. 45. Appearance after removal of polyvinyl
alcohol facing

Fig. 46 Appearance after removal of Acryloid B-67
facing

*ig. 47 Appearance after remo\al of Acrvloid
B-72 facing

Chapter 3: Testing Program 1 10

3.2.8 Test for Reattachment to New Support/Backing
Summary:

After a mural painting has been detached, it must be reattached to a stable new
support. '" Requirements of a new support are as follows:

■ Preservation of the integrity of the original texture and form of the painted
surface. The character of a painting is compromised when surface qualities are
not retained.

■ Materials similar enough in character to the painted layer to ensure that the two
won't separate or sustain other types of damage due to changes in ambient
temperature or humidity.

■ Low thermal conductivity and capacity

■ Resistance to moisture and changes in relative humidity. As the leading cause of
deterioration in mural paintings, the introduction of moisture must be avoided
through the application of an impermeable material.

■ Lightweight, uncomplicated, reversible Resistance to solvents

■ Resistance to ultraviolet light.

■ Resistance to biological attack 1 13

"~ For further information including traditional and modern methods for the reattachment of paintings to
new supports, see The Conservation of Wall Paintings, Chapter 1 1, "Application to the New Support."
Because this information refers to rigid supports of both traditional and synthetic materials and textile
supports stretched on frames for the reattachment of fresco paintings on lime plaster, it is necessary to
modify techniques and materials to meet the demands of earthen materials and the condition of the
paintings.

4 Paolo and Laura Mora and Paul Philipot, The Conservation of Wall Paintings (London: Butterworths,
1984)262-265.

Chapter 3: Testing Program 111

Attachment of strappoed murals to a new support required an immediate and simple
solution in order to proceed with other tests, such as the removal of facing adhesives.
Although this aspect of work was not the focus of the research, a limited study was carried
out to determine appropriate temporary supports.
Objective:

The aim was to cheaply and expediently determine methods and materials for
adhering detached paintings to a new support while creating a bond strong enough to
allow for facing and residue removal. It was important that the adhesive was a "contact"
adhesive and did not penetrate into the mural painting. It was also important that the
surface of the new support had a degree of cushion to accept the irregularities in the mural
surface.
Methodology:

Twelve samples on gypsum board were detached with both colletta and polyvinyl
alcohol facings. The backs of all the detached paintings were coated with two layers of
30% polyvinyl acetate emulsion to inhibit penetration of the backing adhesive. Several
traditional reattachment techniques using a new lime plaster support were tried without
success. They led to the development of alternate methods, also initially unsuccessful.

114 Traditional supports made of slightly modified lime plaster have been used for the reattachment of
frescos for centuries. A modified version of this type of support was tested for this research. Panels of Vi"
plywood were cut slightly larger than the samples to be detached. Two coats of 30% polyvinyl acetate
emulsion (PVA) were applied to seal the surface of the panels. Grids of/2" metal lath were then stapled to
the surface of each panel. The surface of each support was sprayed generously with water before the
application of plaster. The lime plaster, consisting of one part slaked lime, two parts sand, and V 8 part PVA
emulsion was applied in a layer thick enough to cover the lath completely. Once the plaster became firm, it
was sprayed again with water. Small slashes were made in the surface with a spatula and a second coat,

Chapter 3: Testing Program 1 12

Time constraints required the researcher to examine temporary forms of support to
advance to the next phase of research, facing removal."' Temporary solutions included:
reattachment to foam core supports using polyvinyl acetate emulsion brushed around the
perimeters of the detached paintings; to plywood supports using Plasti-Tak® a kneadable,
non-staining poster mount material; and finally, to foam core supports using Scotch
Brand 3M Foam Tape.

Although this information only applies to laboratory tests, aspects of it may be
useful for later phases of work. Data is summarized in Tables 17 and 18.
Observations:

Due to the inconsistent and often unacceptable results obtained from the
reattachment of paintings to plaster supports, it was necessary to select a satisfactory
temporary support." 6 Modifications to these methods might have led to successful results,

about Vt inch in depth was applied. At this time, one coat of PVA emulsion was applied to the back of the
detached painting. When the second layer of plaster was firm but still impressionable, another coat of PVA
was brushed on to the back of the paintings and they were then set in the mortar. Small sandbags were used
to weight the surface of the paintings and the new supports were permitted to cure. Five samples, E-5, E-7,
E-13, E-14, and E-15, were applied to this type of plaster. Three more samples, E-8, E-10, and E-l 1 were
applied to plaster which had been modified by the addition of bentonite clay intended to increase the
potential of forming a stable bond between the painting and the substrate by making the two more similar in
character. Results of this program were unacceptable. Modifications to these methods might have led to
successful results, but time constraints limited possibilities for further testing.

115 Data was required of all phases of testing prior to the beginning of the 1997 field season at Catalhoyiik.
Traditional lime plaster supports did not form stable bonds with the detached paintings. In every case,
detached paintings attached only partially to the new supports. This instability caused significant
deformation and loss during facing removal. Application of a water poultice causes swelling of the partially
attached earthen plasters. Unsupported portions of the painting deform more erratically than those still
attached. Facings and/or adhesive residue cannot be completely removed because there is nosupport for the
painting. Samples from which the facings could be removed, exhibited permanent deformation when the
clays swelled, deformed and dried, forming voids between the painting and the new support. Clay modified
supports exhibited deep cracking in addition to the problem of partial bonding with the paintings.

Chapter 3: Testing Program 1 13

but time constraints limited possibility for further testing. Additional research is
recommended.

Attachment of the paintings to foam core supports by brushing PVA around their
perimeters yielded widely inconsistent results, making it an unsatisfactory technique.

Plasti-Tak and Scotch Brand Foam Tape provided both contact adhesive and
cushion. It must be emphasized that neither of these materials is in any way meant to act
as a conservation solution." 8 Rather they were chosen to allow the researcher to proceed
with testing for the compatibility of surface fixatives, preconsolidants, consolidants, and
facing adhesives. 119 Further research to develop an appropriate support for these plasters,
although beyond the scope of this thesis, is recommended.

Although both temporary methods produced acceptable results, foam tape requires
little pressure to adhere the thin painting layer to its substrate, significantly reducing the
potential for damage during reattachment. It provided an economical and expedient
solution to the problem.

117 The back of one of the paintings undergoing this treatment was overlaid with Plaster of Paris coated
gauze strips for support and adhered to a foam core panel with PVA around its perimeter. Attempts to
remove the facings resulted in total destruction of the plaster. Reattachment of a second painting using this
method also failed. Only one painting treated with PVA around its perimeter exhibited favorable results.
Facings and adhesive residue were removed easily and with minimal alteration to the surface appearance.

118 The removal of either of these adhesives would cause damage to the paintings.

119 Manageable pieces of Plasti-Tak were kneaded by hand and flattened onto the reverse of the paintings until
they were covered with a thin uniform layer of adhesive. Paintings were then adhered to PVA emulsion-coated
plywood panels and gently weighted with small sandbags to fully secure them. The Plasti-tak® was unaffected
by both the water poultice and the application of steam used to remove the facings. Paintings showed little to
no damage or loss. Surface appearances were virtually unaltered. Facings were removed easily, due in part to
the secure bond between the paintings and the supports. Strips of foam tape covered the backs of the PVA
emulsion-coated paintings. Paintings were easily transferred to a foam core panel. The water poultice and
steam treatment used to remove facing materials did not affect the bond between the foam tape and the
painting. Samples showed minimal deterioration, loss, or alteration of optical properties.

Chapter 3: Testing Program

114

Fig. 48. Sample E4: PVA and foam core

Fig. 49 Sample E5: Plaster of Pans backing. PVA
and foam core support

Fig 50 Sample E6: PVA and foam core support

Fig. 51. Sample E7: Traditional plaster support

Chapter 3: Testing Program

115

Fig. 52 Sample E8: Modified plaster support

Fig 53 Sample E9: Plasti-tak support

Fig. 54 Sample E10: Modified plaster support

Fig. 55. Sample Ell: Plaster support

Chapter 3: Testing Program

116

Fig 56 Sample E 12: Foam tape

Fig 57. Sample E 13: Traditional
plaster support

Fig. 58 Sample E15: Traditional
plaster support

Fig. 59 Sample El 4: Failed plaster support

Chapter 3: Testing Program 1 17

3.2.9 Consolidation with Ethyl Silicates
Summary:

Field tests conducted in 1996 focused on the consolidation of earthen plaster and
mudbrick with ethyl silicates. Based on laboratory research, the ethyl silicate monomer
T- 1919 was the most extensively tested. These silica esters are effective consolidants
under certain conditions but do not function as adhesives in, for example, adhering
detached layers. Therefore, if detachment has already occurred, pieces of mudbrick will
not be held together. 120

The extreme friability of the plasters requires preconsolidation, a preliminary
treatment that may also handle superficial delamination. To treat these conditions, two
preconsolidants were tested, a 5% solution of Aquazol50 in ethanol in a 50/50 (v/v)
mixture with the T- 1919 monomer, and Acryloid B-72, in a 7% solution in xylene. The
effects of the application of a single preconsolidant followed by the application of a
consolidant as well as the effects of the application of both preconsolidants followed by
the application of a consolidant were observed. Treatments that demonstrated significant
strengthening of the plaster, exhibited unacceptable levels of discoloration. Treatments

that caused only slight discoloration did not substantially increase the strength of the

i pi

plaster. ~

120 For further discussion see, Alejandro Alva Balderrama and Giacomo Chiari, "Protection and
Conservation of Excavated Structures of Mudbrick" in Consen'ation on Archaeological Excavations With
Particular Reference to the Mediterranean Area (Rome: ICCROM, 1984), 1 14.

121 Sections preconsolidated with the Aquazol 50/T-1919 mixture (one spray application) and either a brush
or spray application of Acryloid B-72 (17c in xylene), then consolidated with T-1919 (3cyles/day for 5
days, spray applied) were strengthened but darkened significantly. The section preconsolidated with the

Chapter 3: Testing Program

Although field results were unremarkable, laboratory tests carried out at the
Architectural Conservation Laboratory and at other earthen architecture sites have
demonstrated the effectiveness of ethyl silicates. 1 "" It is believed that a relative humidity
of 50% to 75% promotes curing of ethyl silicates. Extremely low humidity on site may
account for the lack of significant improvement during field tests. Further research in situ
is required to fully test these materials.
Objective:

In order to consider ethyl silicate consolidation of the samples in the research
program, two of the eight prototype samples, G and J, were consolidated with T-1919.
Based on previous fieldwork, one sample was first preconsolidated with a 50/50 mixture

Aquazol50/T-1919 mixture (one spray application) followed by a brush application of Aquazol 50 (5% in
ethanol). then consolidated with T-1919 (3cyles/day for 5 days, spray applied) was strengthened but
darkened significantly. The section preconsolidated with Acryloid B-72 (7% in xylene, brushed on), then
consolidated with T-1919 (3cyles/day for 5 days, spray applied) was strengthened, but darkened
significantly. Sections preconsolidated with the Aquaxol50/T-1919 mixture(one spray application), then
consolidated with T-1919 (3cyles/day for 5 days, spray applied), were only negligibly strengthened and
slightly darkened. The section consolidated with T-1919 (3cyles/day for 5 days, spray applied )only, was
negligibly strengthened, exhibited no color change, and lost more detaching plaster than sections which had
been preconsolidated. A full report of the treatments and results of this program can be found in Evan
Kopelson, "Analysis and Consolidation of Architectural Plasters from Catalhoyiik, Turkey" (masters thesis,
University of Pennsylvania, 1996) 112-129.

122 For additional information, see Giacomo Chiari. "The Treatment of Adobe Friezes in Peru" and Paul M.
Schwartzbaum, Constance S. Silver, and Christopher Wheatley, "The Conservation of a Chalcolithic Mural
Painting on Mud Brick from the Site of Teleilat Ghassul, Jordan," in Third International Symposium on
Mudbrick (Adobe) Preservation, edited by Okan Ustiinkok and Emre Madran (Ankara: ICOM/ICOMOS,
1980); Ricardo Morales Gamarra, "Conservation of Structures and Decorative Adobe Elements in Chan
Chan," Adobe: International Symposium and Training Workshop on the Conservation of Adobe. Final
Report and Major Papers. Lima-Cusco (Peru). 10-22/September 1983, (Rome: ICCROM, 1983)83-89;
and Constance S. Silver, "Analyses and Conservation of Pueblo Architectural Finishes in the American
Southwest," 6 th International Conference on the Conservation of Earthen Architecture: Adobe 90
Preprints, (Los Angeles: The Getty Conservation Institute, 1990) 176-181.

Chapter 3: Testing Program 1 19

of Aquazol 50® (5% in ethanol) / T-1919. Both were then consolidated using the T-1919
monomer at full strength.
Description of Materials

Aquazol 50®, poly (2-ethyl-2-oxazoline), or PEOX, is a non-ionic polymer
adhesive that is highly soluble in water as well as a range of polar organic solvents such
as acetone, dimethyl formamide, ethanol, methanol, methylene chloride, and methyl ethyl
ketone. 124 It is mechanically and thermally stable and it remains stable in weak acids and
bases. It is compatible with other polymeric materials and is heat-sealable. Accelerated
light aging tests did not alter its appearance, ease of handling, or resolubilization.
Films of this polymer remain plastic even in low relative humidity environments such as
that of Catalhbyiik. Because this product is relatively new, and has not been tested in the
field, it will not necessarily be recommended.

T-1919 Conservare OH ® Stone Strengthener, or tetraethylorthosilicate, an ethyl
silicate monomer, demonstrated the most promise in previous laboratory testing. This
system is based on silicic ethyl esters, which have an extremely small molecular structure.

123 These treatments were the most successful of those carried out in the 1996 field season developed by
Constance Silver.

124 Aquazol 50® is an adhesive produced by Polymer Chemistry Innovations by license from the Dow
Chemical Company.

125 Test results indicate that Aquazol 50® drops in molecular weight and decreases in size rather than cross-
linking or increasing in size under aging conditions. After undergoing the equivalent of twenty four years
of natural aging, the polymer will resolubilize in the same solvents it was initially soluble in. Tests of its
adhesive capabilities for conservation applications were conducted at the Analytical Department of the
Winterthur Museum. Results of the tests focused on four working characteristics: "ability to flow and
penetrate, ability to relax flakes, overall security on drying, and visual effects on surrounding design
materials." Data was favorable. The positive results of these tests were based on the treatment of painted
wooden indoor objects. Further study is required to determine the long-term effects of this adhesive on

Chapter 3: Testing Program 120

allowing for deep penetration into the earthen plasters by capillary action while retaining
the original porosity of the surface. As it cures, the silicic ester gel transforms into silica
oxide, a mineral compatible with the earthen plaster. The silica oxide reacts with the clay
particles to form a tri-dimensional network of silica bridges, increasing the water
resistance of the material. 126 The viscosity of this consolidant, which is lower than that of
water, permits the binding of even minute particles. As previously indicated, low relative
humidities may have hindered the reaction of the ethyl silicate monomer at the site. The
performance of consolidation treatments in the field may be improved by covering the
treated sections in a manner similar to the method used to retard drying time of the
plasters during facsimile development.
Methodology

Both the preconsolidant and ethyl silicate monomer systems were applied to
Sample G. The 50/50 (v/v) Aquazol 50(5% in ethanol)/T-1919 mixture was sprayed on
in one application immediately prior to application of the consolidant. Sample J was
treated with only the consolidant. 127 T-1919 was brushed directly on to the surface of
both panels. Three applications at 30 minutes apart were applied daily. Each application
consisted of three saturating applications of consolidant five minutes apart. The process
was repeated at the same time every day for five days. The sample was then permitted to
cure for approximately one month prior to additional treatment.

exposed, expansive earthen plaster walls. Richard C. Wolbers, Mary McGinn, and Deborah Duerbeck,
"Poly (2-ethyl-2-oxazoline): A New Conservation Adhesive" (n.p., n.d.).
Balderrama and Chiari, 1 14.

Chapter 3: Testing Program 121

Observations:

The depth of penetration and effects of the consolidants on the plaster samples
were measured using the Iodine Vapor Test: Determining the Depth of Penetration of
Consolidants; the CRATerre Water Drop Test; and ASTM D3359-90: Measuring
Adhesion by Tape Test (Methods A and B). These tests are described in Section 3.3,
"Evaluation of Preliminary Testing A." Based on simple visual assessment, a minimal
alteration of optical properties occurred after consolidation.

127 Both samples had been sprayed with a 3% solution of B-72 in toluene prior to consolidation in order to

help stabilize the surface.

Chapter 3: Testing Program 122

3.3 Evaluation of Preliminary Testing A

In order to select materials for the final testing program, treatments were
evaluated using the following standardized tests by the American Society for Testing and
Materials (ASTM), CRATerre, and the Federation of Societies for Coatings Technology:
ASTM D 4214-89 " Evaluating the Degree of Chalking of Exterior Paints" and ASTM
D3359-90 "Measuring Adhesion by Tape Test," "The Iodine Vapor Test: Determining the
Depth of Penetration of Consolidants," and the CRATerre Water Drop Test. These tests
facilitated the assessment of materials in treating friable, powdery paint and/or plaster,
reestablishing adhesion between detached pieces or layers, improving adhesive and
cohesive properties of the paint and plaster layers, and penetrating the paint and plaster
(for assessment of the consolidants). The techniques were accessible, affordable and
allowed for visually quantifiable and qualifiable results. The performance of the materials
and methods were visually examined and quantitatively assessed. The performance of
each treatment informed the final testing program.

Chapter 3: Testing Program 123

3.3.1 ASTM D4214-89: Evaluating the Degree of Chalking of Exterior Paint Films
Summary:

ASTM defines chalking as the formation on a pigmented coating of a friable
powder evolved from the film itself at or just beneath the surface. 128 A friable, powdery
paint or plaster layer will transfer this residue to fabric that is swiped across its surface.
The test provides a method for evaluating the degree of chalking transferred to fabric or
to a finger through comparison to photographic reference standards.
Objective:

The aim of this test was to assess the effectiveness of treatments based on the
degree of chalking exhibited by treated and untreated samples.
Methodology:

Tests were carried out on four multi-layer painted samples on terra cotta supports
following the procedures specified in D42 14-89 Section 7 for wood substrates. Methods A
andB.

Sample Treatment

B-72

AQES

Untreated

Surface consolidated with Acryloid B-72 (3% in toluene) x 8
Consolidated with ethyl silicate monomer T-1919
Preconsolidated with a 50/50 (v/v) mixture of Aquazol 50 ( 5% in
ethanol ) / T- 1 9 1 9, then consolidated with T- 1 9 1 9

128 ASTM, "ASTM D42 14-89: Evaluating the Degree of Chalking of Exterior Paint Films," in 1990 Annual
Book of ASTM Standards Section 6: Paints, Related Coatings and Aromatics (Philadelphia: ASTM, 1990),
70-71."

Chapter 3: Testing Program 124

Method A:

A swatch of black velvet wrapped around an index finger is applied to the surface
of the sample with medium pressure. The finger is then rotated at an angle of
approximately 180°. The chalk mark left on the fabric is then compared to Photographic
Reference Standard #1 in the Pictorial Standards of Coatings Defects.
Method B:

A swatch of black velvet wrapped around the index finger is stroked with medium
pressure across the surface of the sample for approximately three inches. The chalk mark
left on the fabric is compared to Photographic Reference Standard #1 in the Pictorial
Standards of Coatings Defects.
Results are shown in Figures 60 and 67 and Table 19.
Observations:

All of the treated samples exhibited a visually appreciable improvement in
chalking resistance when compared to photographic reference standards and with the test
results of an untreated sample. No significant differences in resistance were detected
between the treated samples. All merited a rating of 8 on the ASTM reference scale.

Chapter 3: Testing Program

125

Fig. 60 Chalking Test Method A: Untreated

Fig 6 1 Chalking Test Method A: Surface

consolidated with Acryloid B-72

(3% in toluene) x 8

Fig 62. Chalking Test Method A:

Consolidated with ethyl silicate monomer

T-1919

Fig. 63. Chalking Test Method A:

Preconsohdated with 50/50 mixture of

Aquazol 50(5% in ethanol)/T-1919

Consolidated with T-1919

Chapter 3: Testing Program

126

Fig 64 Chalking Test (Method B):
Untreated

Fig 65. Chalking Test (Method B):Surface

consolidated: Acryloid B-72

(3% in toluene) x 8

Fig. 66. Chalking test (Method B):
Consolidated with T- 19 19

Fig. 67. Chalking test (Method B):

Preconsolidatcd with 50/50 mixture of

Aqua/.ol 50(5% in ethanol)/T- 19 1 9

Consolidated with T- 19 19

Chapter 3: Testing Program 127

3.3.2 ASTM D 3359-90: Measuring Adhesion by Tape Test (Methods A and B)
Summary:

Aims of the preliminary testing program included the reestablishment of adhesion
between detached layers, stabilization of the painted plaster surface, and improvement of
the adhesive and cohesive properties of the paint and plaster layers. Tests designed to
evaluate the adhesion of coating films to metallic substrates were modified to evaluate the
performance of various conservation materials meant to increase the adhesive properties
of the paint and plaster layers. 129
Objective:

This test was intended to supplement treatment assessment by comparing the
adhesive properties of treated and untreated samples.
Methodology:
Tests were carried out on four multi-layer painted samples on terra cotta supports.

Sample Treatment

B-72

AQES

Untreated

ASTM 3359-90 Test Method A: An X-cut was made on the surface of the samples
with a scalpel (each line measuring approximately 40 mm). A three-inch section of one-

129 ASTM, "ASTM D 3359-90: Measuring Adhesion by Tape Test (Methods A and B)" in 1990 Annual
Book of ASTM Standards Section 6: Paints, Related Coatings and Amniotics (Philadelphia: ASTM, 1990),
511-517.

Chapter 3: Testing Program 1_28_

inch wide semi-transparent, pressure sensitive tape was smoothed over the incisions and
pressed with the eraser tip of a pencil. The tape was removed before ninety seconds had
passed at as close to a 180° angle as possible. Adhesion was rated based on a
standardized scale. Two X-cuts were made on each sample, one superficial, (A. 1 ) and
one penetrating (A. 2).

ASTM 3359-90 Test Method B: The procedure described in ASTM D 3359-90,
Test Method B was slightly modified due to the fragility of the plasters. Samples were
placed on a stable horizontal surface. Eight parallel cuts approximately one inch long,
were made with an X-acto knife at V 8 -inch intervals. Each cut was made in one steady
motion using medium pressure. Once all the cuts were made, the surface was brushed
lightly to remove any detached flakes or dust. Eight more cuts were then made the same
distance apart at right angles to the originals, forming a lattice. The surface was again
brushed lightly, taking care not to detach the fragile plaster layers. A three-inch strip of
pressure sensitive tape was applied to the center of the grid and smoothed with a finger.
After approximately forty-five seconds, the tape was removed at as close to a 180° angle
from the surface of the sample as possible. The grid was then inspected for coating loss
and adhesion was rated on the following scale. (Test Method B.2) The entire procedure
was repeated with the same number of cuts made at five-millimeter intervals. (Test
Method B.2)

Chapter 3: Testing Program 129

Ratings were evaluated based on the ASTM scale and combined in Tables 20 and
21 with the test results.
Observations:
Test Method A

Tests performed on the untreated sample caused the most overall damage.
Treated samples showed an appreciable strengthening of the surface as compared to the
untreated sample. All of the samples exhibited significant damage after the deeper
incision. 10 The sample consolidated with T-1919 showed the best results.
Observations:
Test Method B

All of the treated samples showed an appreciable improvement in surface stability
when compared to the untreated samples. The sample consolidated with the ethyl silicate
monomer, T-1919 produced the best results. Those consolidated with eight applications
of Acryloid B-72 (3% in toluene) produced the next best results.

1 ' This ASTM test, designed to assess the adhesion of coating films applied to metallic substrates, requires
that incisions be cut down to the substrate. The amount of pressure required to make that type of incision
caused the destruction of this coating. The process was modified so that incisions penetrated only the
outermost four to five layers.

Chapter 3: Testing Program

130

Fig. 6X Measuring Adhesion by Tape Test
(Method A): Untreated

Fig. 69 Measuring Adhesion by Tape Test

(Method A): Surface consolidated with Acryloid

B-72 (3% in toluene) x 8

Fig. 70 Measuring Adhesion by Tape Test
(Method A): Consolidated with T- 19 19

Fig. 7 1 Measuring Adhesion by Tape Test

(Method A): Preconsolidated with 50/50 mixture

of Aquazol 50 (5% in ethanol)/T-1919

Consolidated with T- 19 19

Chapter 3: Testing Program

131

Fig. 72 Measuring Adhesion by Tape Test
(Method B): Untreated

Fig 73 Measuring Adhesion by Tape Test

(Method B): Surface Consolidated with

Acryloid B-72 (3% in toluene) x 8

Fig. 74 Measuring Adhesion by Tape Test
(Method B): Consolidated with T-1919

IT I i ,

'•SB

T¥ T I r

Fig. 75 Measuring Adhesion by Tape Test

(Method B):Preconsohdated with 50/50

mixture of Aquazol 50

(5%inethanol)/T-1919

Consolidated with T-1919

Chapter 3: Testing Program 1_32_

3.3.3 Determining Depth of Penetration of Consolidants: Iodine Vapor Test
Summary:

It is critical to test the depth of penetration of consolidants into a material prior to
full-scale treatment. Problems often arise when consolidated and unconsolidated zones are
created within a single substrate. These zones contract and expand at different rates and
may undergo considerable stress at their interface, resulting in the loss of surface material.
An accumulation of moisture and salts behind this interface may cause further deterioration
of the mudbrick or plaster. 131 Pre-testing of consolidated sample material accommodates
modification of materials and methods prior to on-site treatment.
Objective:

The purpose of this procedure was to determine the distribution and depth of
penetration of consolidants applied to the prototype samples.
Methodology:

Approximately forty grams of iodine crystals were divided between two glass
containers and placed in a lidded glass chamber. 133 Three treated multi-layer plaster
samples were placed in the chamber and observed at ten-minute intervals for
approximately two hours. The vapor given off by the iodine crystals physically adsorbs

1 This occurrence was also noted during the 1960's mural painting treatments at Catalhoyiik. See Pamela
French, "The Problems of In Situ Conservation of Mudbrick and Mud Plaster" in In Situ Archaeological
Consen'ation, Proceedings of meetings. April 6-13, Mexico (Mexico: Instituto Nacional de Antropologia e
Historia, 1987; California: J. Paul Getty Trust, 1987), 81.

'" Rakesh Kumar and William S. Ginell, "A New Technique for Determining the Depth of Penetration of
Consolidants into Limestone Using Iodine Vapor," in Journal of the American Institute for Consen'ation,
Volume 36, No. 2 (Summer 1997): 143-150.
1 The crystals were left to sit uncovered until the chamber filled with iodine vapor.

Chapter 3: Testing Program 133

onto the surface of organic compounds resulting in a yellow to light brown stain. The first
sample, (ES-1) had been cut from the prototype consolidated with the ethyl silicate
monomer T-1919. The second sample, (AQES-1 ) had been cut from the prototype pre-
consolidated with the 50/50 mixture of Aquazol 50 (5% in ethanol) /T-1919 and
consolidated with T-1919. The third sample, (B72) had been sprayed with eight
applications of Acryloid B-72 (3% in toluene).
Observations:

Minimal changes were observed during the first hour of testing but each of the
samples exhibited maximum staining after approximately two hours. 134 The plaster and
the terra cotta surfaces of the ethyl silicate treated samples exhibited a deep pinkish
yellow color. Although staining is observed throughout the samples, the blotchy
appearance indicates irregular distribution. The surface of the sample treated with
Acryloid B-72 exhibited an orange-yellow stain that did not penetrate the sample.

Sufficient vapor may not yet have accumulated in the chamber before the samples were introduced.

Chapter 3: Testing Program

134

Fig. 76 . Iodine Vapor Test: Determining the Depth of Penetration of Consolidants. (Left) Sample consolidated with
ethyl silicate moonomer T-1919 (Right) Sample preconsohdated with 50/50 mixture of Aquazol50(5% in ethanol)/T-

1919. then consolidated with T-1919.

Chapter 3: Testing Program 135

3.3.4 Water Drop Test: CRATerre
Summary:

In order to evaluate the resistance to water of treated and untreated plasters, a test
was performed in which samples are subjected to the continuous impact of water droplets
for a period of one to two hours or until the sample has been penetrated. Deterioration of
the paint and plaster is monitored and recorded throughout the test, permitting evaluation by
measurement of visually appreciable results.
Objective:

The purpose of this test was to assess the effects of the preconsolidation and
consolidation treatments on the resistance to water of two types of treated samples. Test
results were then compared to those of the untreated samples.
Methodology:

Two samples each of untreated, consolidated, and preconsolidated and
consolidated samples were tested based on a method developed by CRATerre to
determine the effectiveness of impregnating treatments. The surface of each sample was
submitted to the impact of continuous droplets of water to a single spot, at a rate of one
per second from a height of 2.5 meters, for a period of two hours. Samples were
monitored continuously and effects were evaluated every ten minutes. Results are listed
in Tables 22-27.

Chapter 3: Testing Program 136

Observations:

Differences between the untreated and treated samples were significant. Minimal
deterioration occurred in all the samples during the first hour of testing. Shortly into the
second hour, the untreated samples began to exhibit extensive losses of both paint and
plaster. One of the ethyl silicate consolidated samples, (ES- 1 ), exhibited a slight loss of
pigment and plaster due to an anomaly within the plaster layer. The water drop was
centered on a preexisting deformation that masked a void in the plaster. Losses of the
preconsolidated and consolidated samples were minimal compared to those of the
untreated plaster.

Chapter 3: Testing Program

137

Fig. 77. Preparation for water drop test.

Chapter 3: Testing Program

138

Fig. 78. Water drop test after 10 seconds: (Left to right) Untreated (UT1), Untreated (UT2), Ethyl silicate (ESI)

Fig. 79 Water drop test after 2 hours: (Left to right) UT1, UT2, ESI

Chapter 3: Testing Program

139

Fig. 80. Water drop test after 10 seconds: (Left to right) Aquazol50/T-1919+T-1919 (AQES1), Aquazol50/T-I919+T-

1919 (AQES2). Ethyl silicate (ES2)

Fig. 81. Water drop test after 2 hours: (Left to right) Aquazol50/T-1919+T-1919 (AQES1). Aquazol50/T-1919+T-I919

(AQES2), Ethyl silicate (ES2)

Chapter 3: Testing Program

140

, -
1

<*&*4m

k^^Bti

dHH

■BT

■

■rsSflBI^:

*»w _... ^ r ^

Fig. 82. Water drop test final result:
UT1

Fig. 83 Water drop test final result:
UT2

Fig 84 Water drop test final result:
ESI

-r— — — —

Fig 85 Water drop test final result:
ES2

Fig 86. Water drop test final result:
AQES1

Fig 87 Water drop test final result:
AQES2

Chapter 3: Testing Program 141

3.4 Preliminary Testing B: Compatibility of Treatments
Summary:

The purpose of "Preliminary Testing A" was to evaluate materials and methods to
treat the following individual conditions: powdering paint, interlayer detachment, and
cleavage of the surface layer. The program then addressed the detachment of mural
paintings by testing both traditional and non-traditional facing adhesives for strappo and
stucco. Compatible materials were chosen for all aspects of treatment.

For this aspect of research, it was necessary to identify the appropriate facing
adhesive based on the following characteristics:

■ Adequate adhesion over all the painted surface to permit detachment at a
consistent level, whether that be stacco, removal of all the layers; or strappo,
removal of only the top layer.

■ Compatibility with pretreatment, for example, soluble in different solvents
than those required for pretreatment materials.

■ Reversibility.

A combination of conservation treatments is usually required to treat painted
earthen plasters following excavation. The next phase of research, "Preliminary Testing
B," addressed the compatibility of these treatments. Results served to inform the final
testing program.

The problem of consolidation of the paint layer was addressed first. This aspect
of treatment aimed to stabilize the powdering/chalking paint layer so that other aspects of

Chapter 3: Testing Program 142

treatment could be conducted. A surface consolidant was required to accomplish this
without altering the optical properties of the samples and without interfering with other
aspects of treatment. Thus, it needed to affect the surface without penetrating it and still
be thin enough not to prevent the penetration of subsequent treatment materials. Since
aqueous materials were to be used in most other aspects of treatment, it was particularly
important that the surface consolidant was not soluble in water. Several materials were
tested. The most effective were: Acryloid B-72 in a 3% solution in xylene, and, based on
previous tests, Blair Spray Fix, a nitro-cellulose-based surface fixative developed for the
treatment of pastels and temperas. 135

The next phase of testing focused on preconsolidation. The use of aqueous
materials, which swell the clays, may facilitate readhesion when controlled. Results of
preliminary testing indicated that a softened, swelled plaster surface might be flattened by
applying pressure over Mylar. This treatment aids in compacting and readhering cracked
pieces, improving craquelure, and relaxing cleavage. Both water, and water followed by
Aquazol 50 (5% in water) were successful as preconsolidants during preliminary testing.

Finally, four traditional and non-traditional, or synthetic facing adhesives were
tested in conjunction with other aspects of treatment. They were: colletta, polyvinyl
alcohol, Plextol B500, and Vinamul 6825. These adhesives were assessed based on ease

135 Constance S. Silver successfully employed Blair Spray Fix for tests to detach earthen plaster murals at
Mesa Verde. The program is described in "Architectural Finishes of the Prehistoric Southwest: A Study of
the Cultural Resource and Prospects for its Conservation" (Master's Thesis, Columbia University, 1987)
176. Because nitrocellulose darkens considerably with age, the material was not used for the final testing
program here.

Chapter 3: Testing Program 143

of detachment, ease of facing removal, preservation of optical properties, and

compatibility with prior treatment.

Objective:

The purpose of this investigation was to evaluate the compatibility of treatments
based on the following criteria:

■ Preservation of original appearance in terms of: color, gloss, texture, and surface
form

■ Success at achieving the objective: efficient detachment of the plaster stratum as
a whole or of a discreet layer

■ Successful reattachment of the painting to a new support

■ Removal of the facing adhesive with little or no damage to the painted plaster
surface

Materials were selected based on their performance, stability (as observed in
previous testing) and retrievability.
Methodology:

Various combinations of surface fixatives, preconsolidants and facing adhesives
were applied to sixteen 6-by-6-inch multi-layer, painted gypsum wallboard panels in
order to evaluate their compatibility. Detachments were carried out using traditional and
modern materials. The details of each test are listed in Tables 28-31.

Chapter 3: Testing Program 144

Observations:

Both colletta and polyvinyl alcohol successfully met facing adhesive criteria:

■ Does not penetrate the surface of the plaster

■ Facilitates detachment

■ Provides support to detached plaster layer

■ Permits easy removal of the facings and residue

■ Are not adversely affected by other forms of treatment
Colletta:

Results of tests combining colletta facings with other aspects of treatment were
generally positive. Although the strappo technique is designed to detach the paint layer
alone, both the stacco and strappo formulas yielded stacco-\ike results in preliminary
testing. Contraction of the strappo glue caused nearly complete detachment of the plaster
stratum without mechanical intervention. The strappo facing is generally easier to
remove, particularly when the water poultice technique is used in conjunction with
steam. 136 A slight yellow residue often remained on the surface of the samples.
Polyvinyl alcohol:

Polyvinyl alcohol facings were easily removed. The adhesive resolubilized
quickly with water poultices without affecting the plaster and paint layers. Detachment of

136 Differences in the ease of removal of these two facings may be due to a number of circumstances. The
stacco method includes a plasticizer in the colletta recipe, the strappo method does not. The stacco method
requires that the first facing be dry before the application of the second, while in strappo, the two layers are
applied consecutively. The greater tenacity of the stacco facing may be due in part to the additional coat of
glue used to adhere the secondary facing after the first has dried.

137 This may be due in part to the inexperience of the researcher.

Chapter 3: Testing Program 145

the full plaster stratum was achieved easily with few losses and negligible darkening. 138
Based on ease of application and removal, and the preservation of appearance, polyvinyl
alcohol was found to be the most effective facing adhesive.
Plextol B-500 and Vinamul 6825:

Plaster surfaces had to be flooded with solvents before tissue facings attached
with Plextol B-500 and Vinamul 6825 could be removed. Textile facings could be
removed only with significant force, if at all. Dilute solutions of these adhesives were no
less tenacious. Plextol B-500 and Vinamul 6825 were ruled out as facing adhesives
based on the results of the preliminary testing program.

This may have been due in part to a weak bond between the plasters and gypsum board substrate.

Chapter 3: Testing Program

146

Fig 88 Studio overview showing preliminary test samples.

Fig. 89. Test materials

Fig. 90 Secondary facing removal

■•*

Chapter 3: Testing Program

147

Fig 91 Compatibility of Treatments: (Left) Untreated (Right) Sample D-l after detachment with stacco colletta

Fig. 92. Compatibility of Treatments: (Left) Untreated (Right) Sample D-2 after detachment with strappo colletta.

Fig. 93. Compatibility of Treatments: (Left) Untreated (Right) Sample D-3 after detachment with strappo colletta

Chapter 3: Testing Program

148

Fig. 94. Compatibility of Treatments: (Left) Sample D-4 after detachment with siacco colletta (Right) Untreated

Fig. 95. Compatibilih of Treatments (Left) Sample D-5 after detachment with PVOH (Right) Untreated

Fig. 96. Compatibility of Treatments: (Left) Sample D-6 after detachment with PVOH (Right) Untreated

;

Chapter 3: Testing Program

149

Fig. 97. Compatibility of Treatments: (Left) Untreated (Right) Sample D-7 failed detachment withPlextol B500 facing

Fig. 98. Compatibility of Treatments: (Left) Untreated (Right) Sample D-8 after detachment with Vinamul 6825 facing

Fig. 99 Compatibility of Treatments: (Left) Untreated (Right) Sample D-9 failed detachment with Plextol B500 facing

* •. -

Chapter 3; Testing Program

150

Fig. 100. Compatibility of Treatments: (Left) Untreated (Right) Sample D-IO after detachment with Vmamul 6825

Fig. 101. Compatibility of Treatments: (Left) Untreated (Right) Sample D-12 after detachment with PVOH

Fig 102 Compatibility of Treatments: (Left) Untreated (Right) Sample D-17 after detachment with PVOH

Chapter 3: Testing Program

151

Fig. 103. Compatibility of Treatments: (Left) Sample D-18 after detachment with strappo colletia (Right) Untreated

nPBft ^fe « 'V' 1 i

— , — ' — — '—

Fig. 1 04. Compatibility of Treatments: (Left) Sample D- 19 after detachment with stacco colletta (Right) Untreated.

Fig. 105 Compatibility of Treatments: (Left) Sample D-21 after detachment with strappo colletta (Right) Untreated

■.-

Chapter 3: Testing Program 152

3.5 Final Testing Program

This chapter describes the final phase of research, treatment and detachment of the
prototype samples on terra cotta substrates. It describes the detachment procedures and
the final appearance of the paintings. The final conclusions are based on the results of
these tests and are described in Chapter 4.
Summary:

Final conservation treatments were selected based on results of the preliminary
testing program. Two techniques were used to evaluate methods and materials: visual
assessment and observation of performance in standardized tests created by ASTM, the
Federation of Societies for Coatings Technology, and CRATerre. Four specific types of
conditions required treatment before detachment could be carried out:

■ Powdering paint

■ Interlayer detachment

■ Lack of cohesion

■ Lack of adhesion of the plaster and paint layers

Final tests were applied to nine prototype samples composed of seven painted
ground/finish sequences on terra cotta tile substrates. One sample served as a control. Two
samples had been treated, one with ethyl silicates, and one with ethyl silicate and poly (2-
ethyl-2-ozazoline) (see page 117). Thus, the selected conservation materials were applied
to six of the prototype samples, A, B, C, D, H, I. The two most successful of these
treatments were then applied to the consolidated samples, G and J.

Chapter 3: Testing Program 153

Objective:

Tests considered the following:

■ Surface consolidation systems

■ Preconsolidation systems

■ Detachment systems for strappo

■ Detachment systems for stucco

■ Combined treatments
3.5.1 Methodology:

Treatment in Preparation for Detachment
Samples A, B, C, D, H, I
The most effective preliminary treatment involved:

■ Surface consolidation with eight spray applications of Acryloid B-72 (3% in
toluene)

■ Preconsolidation with water and pressure

Based on preliminary tests. Acryloid B-72 was selected as a surface consolidant to
re-establish the cohesive strength of powdering paint. It effectively consolidated the paint
without causing visible alteration or interfering with subsequent treatment. Only minor
changes in porosity allowed penetration of subsequent treatment materials. Moreover, it
made the paint layer to be insoluble in water, thereby permitting subsequent treatments with
aqueous materials.

Chapter 3: Testing Program 154

Therefore samples A, B, C, D, H, and I were prepared for detachment by
consolidation of the surface with Acryloid B-72, and then preconsolidation using water
brushed through Japanese tissue. y Pressure was applied through silicone release Mylar
weighted with sandbags.
Samples G and J

Samples G had been preconsolidated earlier with a 50/50 mixture of Aquazol 50
(5% in ethanol) / T-1919. Both samples G and J were consolidated with the ethyl silicate
monomer T- 1919.

Application of facing adhesives: Samples A, B, C, D, H, I
Summary:

Based on the results of preliminary testing and their ability to meet the established
criteria, colletta and polyvinyl alcohol facings were selected as facing adhesives to detach
the mural paintings from Samples A, B, C, D, H, and I. These adhesives were viscous
enough to prevent penetration of the plaster layers while remaining easily resoluble.
Contraction of both the stacco and strappo forms of colletta alone was enough to detach
large portions of plaster. Polyvinyl alcohol was selected for its easy resolubility and
minimal residue after detachment. Two samples were faced with each adhesive and gauze

1 9 Eight coats of Acryloid B-72 was the maximum that could be applied without causing unacceptable
darkening or surface shine. Two aqueous solutions, Aquazol50® in a 5% solution in water and water alone
were successful at reestablishing the adhesive and cohesive properties of the plasters, readhering
delaminating layers, and facilitating manipulation of cleavage and deformation. Because it has not been
tested in the field, Aquazol50® was not selected. For preconsolidation, water alone was used.

Chapter 3: Testing Program 155

and hemp facings which provide support and protection to the painting during the
detachment procedure.

The two most successful materials used on the first six prototypes were then applied
to the ethyl silicate-consolidated samples, G and J. To permit the testing of two different
facing adhesives, each of the ethyl silicate consolidated samples was divided in half.
Colletta (Samples A, D, H, I )

Stacco colletta was heated in a double boiler and mixed with enough water to make
a 50% solution. The colletta was then brushed on to the surface of the panels and faced
with strips of 100% cotton gauze, overlapping by at least one centimeter, and leaving an
excess of approximately ten centimeters around the perimeters of each painting. 141 The
excess fabric around the border of each painting was folded back upon itself to form a five-
centimeter hem. The panels were then permitted to dry for approximately one to two days.
While they were still soft a second coat of adhesive was applied. A secondary facing
consisting of hemp fabric strips overlapping by approximately one centimeter, was laid on
top of the warm glue. The stacco formula was applied to Samples A and D.

Strappo colletta was heated in a double boiler and mixed with enough water to
make a 75% solution. 14 " Glue was brushed on to the surfaces of the paintings and each was

Cotton gauze and hemp are the traditional materials used for the detachment of mural paintings in the
stacco and strappo methods. For further discussion see Paolo and Laura Mora and Paul Phillipot, The
Conservation of Wall Paintings, (Butterworths: London, 1984).

' ' The cotton gauze must be washed to remove the size, and allowed to dry before application to the
painting. Once applied to the surface of the painting, the gauze was stretched gently to avoid wrinkles or
gaps on the surface.

1 " The concentrations of both colletta glues were chosen empirically based on the most dilute solution
capable of detachment without penetration of the plasters.

Chapter 3: Testing Program 156

faced with strips of 100% cotton gauze in the above-described manner. As required for the
strappo detachment process, a second layer of glue and hemp fabric facings were applied
immediately afterwards. This facing was applied to Samples H and I.
Polyvinyl alcohol (Samples B, C)

Polyvinyl alcohol was dissolved in water at a concentration of 20% and applied to
two panels, Samples B and C. Gauze facings were applied as above and the second layer of
adhesive and hemp facings was applied immediately afterward.
Detachment (Samples A, B, C, D, H, I)

The following section describes the detachment procedure. A very sharp X-acto
knife was used to make cuts around the perimeters of the paintings to be detached.
Plywood panel supports, the same size as the sections to be detached, were placed against
the surfaces of the faced paintings. The hems of the hemp canvases were turned up and
over the rear of the panels and tacked into place. By this point, many of the paintings had
partially detached due to the contraction of the colletta adhesive. If not wholly detached,
one of two methods was used to free the paintings. In some instances, a padded mallet was
tapped lightly over the surface of the support panel. In others, a 12-inch metal spatula was
inserted between the layers of plaster partially freed and those still attached. The support
panel, with the facings still tacked in place, was then pulled from the sample till the painting
was completely freed.

Chapter 3: Testing Program 157

Treatment after detachment:

Following detachment, plaster fragments still attached to the reverse of a painting
were removed prior to reattachment to a new support to allow for a more stable bond
between the painting and the support. 143 A 30% solution of polyvinyl acetate emulsion was
applied to the back of each painting as an isolating layer to further facilitate a strong bond
with the support. Preliminary tests on backing materials and adhesives supported the use of
foam adhesive tape, to both cushion and adhere the fragile painted layers to plywood
supports.

Strips of foam tape were adhered to the back of each painting until completely
covered. Plywood supports coated with a 30% polyvinyl acetate emulsion were pressed
onto the backs of the faced paintings.

Colletta facings were removed with a combination of water poultices and steam. 145

Polyvinyl alcohol facings were treated with water poultices for four to six hours
prior to removing the facing. The procedure proved slightly more difficult than in
preliminary testing.

' Fragments were cleared from the paintings using micro-spatulas and fine scalpels.

144 Scotch Brand Foam Tape functioned as the temporary support for the final testing program.

14 A portable clothing steamer was sufficient for panels of this size. Colletta facings should not be left on

the surface of a painting beyond the point at which they first become firm. As they harden, the tenacity of

the glue makes it very difficult to remove without significant amounts of water and steam, which can be

detrimental to clay plasters. Excessive swelling may result in deformation and detachment. If detachment

occurs, the lack of support behind the plaster layer can make facing and residue removal problematical.

Chapter 3: Testing Program 158

Application of Facing Adhesives II: (Samples Gi, G 2 , Ji, J2)
Colletta and polyvinyl alcohol

The consolidated panels were divided into halves so that each could be tested with
two different facing adhesives. Because the stacco form of colletta yielded the most
favorable results based on visual assessment of samples A, D, H, and I after detachment, it
was applied to one half of each of the consolidated samples. A 20% solution of polyvinyl
alcohol in water was applied to the other half. Primary and secondary facings were applied
using the previously described methodology. Polyvinyl alcohol facings were removed using
water poultices. Colletta facings were removed using a combination of water poultices and
steam.

Results of prototype detachments are described in Tables 32 and 33. Table 34
quantifies the results of each detachment by listing the percentages of discrete plaster layers
remaining on the substrate after detachment. The * symbol in the following tables stands
for the spray-applied preconsolidant made from a 50/50 (v/v) solution of Aquazol 50® (5%
inethanol)/T-1919.
Observations:

Four of the six samples faced with the colletta adhesive detached naturally as a
result of the contraction of the glue. The best result, based on visual assessment of the
reattached paintings, was attained with the stacco formula, which exhibited only minimal
losses and cracking. Despite the more impressive performance of polyvinyl alcohol

Chapter 3: Testing Program 159

facings during preliminary testing, paintings detached using stacco colletta showed the
least deterioration during final testing.

Samples treated with both the Aquazol 50®/T-1919 preconsolidant and the
T-1919 consolidant were detached with both the stacco colletta and polyvinyl alcohol
facings. The stacco sample exhibited significant losses. The sample faced with polyvinyl
alcohol exhibited only slight powdery losses but residue removal caused abrasion of the
paint layer.

Samples consolidated with T-1919 were also detached using stacco colletta and
polyvinyl alcohol facings. The painting detached using the stacco adhesive exhibited
negligible losses at its edges. 146 The polyvinyl alcohol facing resulted in the most
strappo-Mkc detachment. Approximately 95% of the painting detached between the top
two layers.

Each method was successful at removing the surface paint layer and at least one
layer of plaster as a whole. However it was not possible to limit detachment to a specific
consistent layer. Consequently, indiscriminate detachments resulted in extensive

1 Losses of paint and plaster probably occurred due to a lack of support behind the paint layer. After each
painting was detached, fragments of underlying plaster layers were cleared from the back surface until a
consistent layer was reached. Plaster was not removed down to the paint layer. Paintings were reattached
to a new support and facings were removed using water poultices and steam. The introduction of water
caused the clay plasters to swell and deform. The bottom layer often remained adhered to the new support
while the surface layer swelled and partially delaminated. Sections of the paint layer were then left
unsupported, increasing the losses of both paint and plaster during facing and residue removal. Reports on
previous treatment indicate that a lack of support directly behind the paint layer was the cause significant
losses of material during the 1960s program. (See Appendix A) Initially, the removal of supporting plaster
layers from the backs of paintings did not reach down to the paint layer. Consolidation of the painted
surface and the rear surface resulted in a zone of unconsolidated plaster between them. Differential rates of
contraction and expansion between the three zones caused the unconsolidated layer to crumble, leaving the
paint layer unsupported and extremely fragile.

Chapter 3: Testing Program 160

fragmentation of plaster layers remaining on the substrate. Slight but not unacceptable
darkening occurred on the surface of all the samples. Analysis of post-detachment layer
distribution (see Table 34) revealed no consistent pattern of fragmentation. Detailed
detachment results are located in Tables 32 and 33.
3.6 Evaluation of Final Testing Program

This section reports on the effectiveness of treatment measured by various tests
standardized by the American Society for Testing and Materials and the Federation of
Societies for Coatings Technology. Tests were designed to evaluate the performance of
coatings, specifically paint films, on the basis of three conditions: cracking, flaking, and
checking. These tests allowed the researcher to compare the effects of various treatments
on the painted plaster surfaces using standardized rating scales and photographic references.
Samples were classified by color notation using the Munsell System before and after
treatment. Comparison to standardized color chips facilitated the evaluation of treatment-
related color changes.
Summary:

Detachments were evaluated by standardized visual assessment tests. ASTM
standard tests used to assess visual appearance of the samples included: D 661-86:
"Standard Test Method for Evaluating Degree of Cracking of Exterior Paints," D 772-86:
"Standard Test Method for Evaluating Degree of Flaking (Scaling) of Exterior Paints," and
D 1535-80: "Standard Method of Specifying Color by the Munsell System," and D 660:

Chapter 3: Testing Program

"Standard Test Method for Evaluating the Degree of Checking of Exterior Paints." Each of
these tests provides a standard of comparison by which to compare the appearance of the
detached samples to the untreated samples. All of these tests, excluding color specification
are based on pictorial photographic reference standards contained in the Pictorial Standards
of Coatings Defects distributed by the Federation of Societies for Coatings Technology in
Blue Bell, Pennsylvania. 147

Federation of Societies for Coatings Technology, Pictorial Standards of Coatings Defects,
(Pennsylvania: Federation of Societies for Coatings Technology).

Chapter 3: Testing Program 162

3.6.1 ASTM D661-86: Evaluating Degree of Cracking of Exterior Paints
Summary:

ASTM defines three types of cracking: irregular pattern type, in which no definite
pattern is evident; line type, in which the cracks generally occur in parallel lines often
following brush marks; and sigmoid type, in which the cracks form a pattern of intersecting
curves. 148 The Pictorial Standards of Coating Defects consists of silver halide photographs
that provide standards of comparison for each test. 149 Each photograph has a numerical
rating descriptive of the degree of cracking it represents.
Objective:

The purpose of this test was to visually assess and compare the degree of cracking of
treated and untreated samples using pictorial reference standards.
Methodology:

Each sample was compared to the pictorial standards and given a numerical
rating. Comparisons were drawn between the effects of the different treatments and
between treated and untreated samples. Results are listed in Tables 35 and 36. In all of
the following tests, a rating of 2 represents the worst case scenario.

148 ASTM "ASTM D661-86: Evaluating Degree of Cracking of Exterior Paints," in 1990 Annual Book of
ASTM Standards Section 6: Paints, Related Coatings and Aromatic s (Philadelphia: ASTM, 1990), 82-83.
The photographs illustrate line type cracking only but are intended as a reference for all types.

Chapter 3: Testing Program 163

3.6.2 ASTM D772-86: Evaluating Degree of Flaking (Scaling) of Exterior Paints
Summary:

The term flaking refers to the detachment of pieces of a paint layer from its
substrate. 150 The degree of flaking of the painted surface following detachment may be
evaluated by comparison with photographic reference standards.
Objective:

The purpose of this test was to visually evaluate and compare the degree of
flaking on the surface of the treated and untreated samples to pictorial reference standards
and to each other.
Methodology:

Treated and untreated samples were compared to the pictorial standards and given
a numerical rating. The effects of different treatments, and of treated and untreated
samples were compared. Ratings are listed in Tables 35 and 36.

150 ASTM, "ASTM D772-86: Evaluating Degree of Flaking (Scaling) of Exterior Paints," in J 990 Annual
Book of ASTM Standards Section 6: Paints, Related Coatings and Aromatics (Philadelphia: ASTM, 1990),
97-100.

Chapter 3: Testing Program 164

3.6.3 ASTM D660 Evaluating Degree of Checking of Exterior Paints
Summary:

ASTM defines checking as breaks in a film that do not penetrate to earlier layers
or to the substrate. I:i1 "Line checking" is identified by checks on a surface arranged in
horizontal or vertical parallel lines that often follow brushstrokes. "Crowsfoot checking"
describes breaks in a film that form a three way like a crowsfoot. Checks run from the
center and form an angle of approximately 120° between the two ends. "Irregular
checking" does not exhibit a definite pattern but often exhibits a combination of the first
two. The pictorial standards are assigned to a numerical rating system ranging from 2-10.
A rating of 10 refers to films that exhibit no checking.
Objective:

The purpose of this test was to visually evaluate and compare the occurrence of
checking on the surface of both treated and untreated samples to pictorial reference
standards.
Methodology:

Samples were observed and assigned the number best representing the degree of
checking present. Comparisons were then made between the treated and untreated
samples. Ratings are listed in Tables 35 and 36.

151 ASTM, "ASTM D660 Evaluating Degree of Checking of Exterior Paints," in 1990 Annual Book of
ASTM Standards Section 6: Paints, Related Coatings and Amniotics, (Philadelphia: ASTM, 1990), 72-81

Chapter 3: Testing Program 165

Observations:

Both the strappo and the polyvinyl alcohol samples exhibited a significant degree of
flaking, except those that had first been consolidated. The stacco sample preconsolidated
with Aquazol 50/T-1919 exhibited a significant degree of flaking.

One strappo and both consolidated stacco samples exhibited significant cracking,
probably due to the swelling and interlayer detachment caused by the water-based residue
removal method. The remainder of the samples exhibited only slight cracking after
detachment.

One stacco sample exhibited minimal checking after detachment. Neither type, nor
occurrence of deterioration was specific to any one treatment.

Chapter 3: Testing Program 166

3.6.4 ASTM D 1535-80 Specifying Color by the Munsell System
Summary:

The Munsell System of Color Specification assigns color notations based on the
properties of hue, value and chroma, as visually perceived by an observer under normal
daylight conditions. 152 The notations follow the formula: HV/C. A letter expresses hue,
the property that classifies a color by name, such as red. Value, a measure of daylight
reflectance, is expressed on a scale of to 10; 0, representing true black, and 10, true
white. Chroma refers to the degree of saturation of the colored surface, and is expressed
on a scale of to 20. Painted samples may be visually assessed next to standard color
chips provided with the kit. This system provides a method of visually assessing the
effects of treatment on the optical properties of the plasters.
Objective:

The purpose of this test was to evaluate the effects of treatment on the optical
properties of the samples by assigning specific color notations to the painted plaster
before and after treatment.
Methodology:

Prototype panels were observed in the conservation laboratory before and after
treatment under natural daylight illumination from a north window. Painted samples
were compared to chips from the matte edition of the Munsell Book of Color. The colors

/52 ASTM, "ASTM D 1535-80 Specifying Color by the Munsell System," in 1990 Annual Book of ASTM
Standards Section 6: Paints, Related Coatings and Aromatics, (Philadelphia: ASTM, 1990), 182-208.

Chapter 3: Testing Program 167

rated were red, black and the unpainted white plaster. Several color notations were noted
for each color on each sample to account for the mottled surfaces of the prototypes. In
order to better assess the effects of treatment, color notations before and after treatment
are listed together in Table 37. Table 38 interprets the data on the effects of treatment on
painted and unpainted samples.
Observations:

Most samples exhibited a slight yellowing of the surface. The unconsolidated
samples faced with polyvinyl alcohol were the only two to resist yellowing of the
unpainted plaster surface. No other alteration in hue was observed. All the samples
exhibited a slight discontinuous darkening of the surface, due at least in part, to adhesive
residue. In some areas, the paintings were lighter, probably due to abrasion from facing
and residue removal.

Chapter 3: Testing Program

168

Fig. 106 Prototype A: Untreated

Fig. 107 Sample A: After detachment

Fig. 108. Prototype B: Untreated

Fig 109 Sample B: After detachment

Chapter 3: Testing Program

169

Fig. 110. Prototype C: Untreated

Fig 111 Sample C: After detachment

Fig. 112. Prototype D: Untreated

Fig 113. Prototype D: after detachment

Chapter 3: Testing Program

170

[5P^

i*^

1
1

B^N?'" ^ M

l^^l

Fig. 114. Protohpe G: Untreated

Fig. 115. Protohpe G, after
detachment

Fig 1 16 ProtoUpe G,:after
detachment

•

■fcjj —^1 _iA«J

Fig. 117. Prototype H: Untreated

Fig. 118 Prototype H: after detachment

Chapter 3: Testing Program

171

Fig 119 Prototype sample I: Untreated

Fig 120. Prototype sample I: After detachment

Fig. 121 Prototype sample J: Untreated

Fig. 122 Prototype sample J r Fig. 123. Prototype sample J,

After detachment After detachment

Chapter 3: Testing Program

172

Fig. 124 Substrate Sample A: after detachment

Layer 13
Layer 12
Layer 1 1
Layer 10
Layer 9
Layer 8
Layer 7
Layer 6
Layer 5
Layer 4
Layer 3
Laver 2

Fig 125. Sample A: Layers remaining on substrate after detachment

Chapter 3: Testing Program

173

^ r i

igl *.*> M' j v' Sis

"r :

Y' 'J^S^P

Fig. 126. Substrate Sample B: after detachment

Layer 13
Layer 12
Layer 1 1
Layer 10
Layer 9
Layer 8
Layer 7
Layer 6
Layer 5
Layer 4
Layer 3
Layer 2

Fig. 127. Sample B: Layers remaining on substrate after detachment

Chapter 3: Testing Program

174

Fig. 128 Substrate Sample C: after detachment

Layer 13
Layer 12
Layer 1 1
Layer 10
Layer 9
Layer 8
Layer 7
Layer 6
Layer 5
Layer 4
Layer 3
Layer 2

Fig. 129 Sample C: Layers remaining on substrate after detachment

Chapter 3: Testing Program

175

Fig. 1 30 Substrate Sample D after detachment

Layer 13
Layer 12
Layer 1 1
Layer 10
Layer 9
Layer 8
Layer 7
Layer 6
Layer 5
Layer 4
Layer 3
Layer 2

Fig. 131. Sample D: Layers remaining on substrate after detachment

Chapter 3: Testing Program

176

u •

•'4'

,'..•..!( \ ..." N.\ak.iS..,Ai:,^ 1 .,* 'v

<*■?

'•' ixw^

Fig. 132. Substrate Sample G,:
after detachment

Fig. 134 Substrate Sample G,:
after detachment

Fig. 133 Sample G r

Layers remaining on substrate after

detachment

Fig. 135 Sample G,:

Layers remaimng on substrate after

detachment

Chapter 3: Testing Program

177

Fig. 136 Substrate Sample H: after detachment

Layer 13
Layer 12
Layer 1 1
Layer 10
Layer 9
Layer 8
Layer 7
Layer 6
Layer 5
Layer 4
Layer 3
Layer 2

Fig. 137. Sample H: Layers remaining on substrate after detachment

Chapter 3: Testing Program

178

-Jit.j

f*£

Fig. 138 Substrate Sample I: after detachment

Layer 13
Layer 12
Layer 1 1
Layer 10
Layer 9
Layer 8
9 Layer 7
Layer 6
Layer 5
Layer 4
Layer 3
Layer 2

Fig. 139 Sample I: Layers remaining on substrate after detachment

Chapter 3: Testing Program

179

Fig. 140. Substrate Sample J,
after detachment

Fig. 142 Substrate Sample J,
after detachment

Fig 141. Sample J,:
Layers remaining on substrate after detachment

Fig 143. Sample J,:
Layers remaining on substrate after detachment

Chapter 3: Testing Program

180

Fig. 144 Rating Scale for ASTM D660 : Evaluating Degree of Checking of Exterior Paints

Fig. 145 Rating Scale for ASTM D661-86 : Evaluating Degree of Cracking of Exterior Paints

Fig 146 Rating Scale for ASTM D772-86 : Evaluating Degree of Flaking (Scaling) of Exterior Paints

Chapter 3: Testing Program

181

Fig 145. Overview of test samples

Chapter 4: Conclusions

4.1 Final results

This research program addressed two principal requirements for the preservation
of the wall paintings on earthen plaster supports at Catalhbyiik: emergency stabilization
and removal. Results of a series of eighteen tests indicated that two systems of treatment
were successful at strengthening the paint and plaster surfaces and facilitating detachment
of the mural paintings.

This section summarizes the results of the testing program. Test-specific data is
organized into tables beginning on page 189. Each summary of treatment provides a
reference to the corresponding tables. The chapter closes with conclusions and
recommendations for further research.
Summary

Methods for preliminary treatment and emergency stabilization were required to
handle the following conditions:

■ Powdering paint and plaster

■ Interlayer detachment

■ Disaggregation of the earthen plaster

In response to these conditions, the following tests were conducted:

■ Consolidation treatments

■ Facing adhesives for mural detachment

■ Detachment methods

■ Compatibility of treatments

182

Chapter 4: Conclusions 183

Tests were conducted on laboratory facsimiles due to the limited number and size
of painted plaster samples available from the site. Samples were composed of multiple
plaster and paint layers similar in character to those found in Turkey. Fragile conditions,
similar to that found on site, were created by applying a weak pictorial layer to an
expansive clay plaster support.

The loss of cohesive strength within discrete layers and adhesive strength between
individual layers of the Catalhoyiik plasters and mural paintings required two different
types of pretreatment. A preconsolidant was needed to impart cohesive strength to the
surface while enabling the application of additional treatment and preserving the optical
properties of the painted plaster. Another type of adhesive was required to readhere
separated layers, reattach flaking paint and plaster layers, and relax cleavage of the
surface.

The following section summarizes the most successful methods and materials for
each phase of treatment. It is followed by the supporting data, organized into tables for
clarity. Shaded cells indicate a positive result.
4.1.1 Surface Consolidation

Results of preliminary testing indicated that multiple spray applications of an
acrylic resin, Acryloid B-72, in a 3% solution in toluene were the most successful surface
consolidant, making the pigment layer insoluble in water and strengthening the surface of
the painted layer and plaster without altering their optical properties. B-72 was able to
quickly bond loose particles to the support and prevent losses in the paint layer without

Chapter 4: Conclusions 184

interfering with the effectiveness of the preconsolidation treatment. Using multiple
applications of a dilute solution permits sufficient penetration within the plaster to
prevent the formation of a surface film. For more detailed information, please refer to
Tables: 6-9, 12-14, 19-21.

4.1.2 Readhesion

Several tests were used to evaluate materials for the reestablishment of adhesion
between detached layers. Applications of both water and Aquazol 50, a non-ionic
polymer adhesive, in a 5% solution in water, improved adhesion between plaster layers
previously exhibiting cleavage and separation. The effects of water on the behavior of
clay and clay plasters has been explained in section . By using water or another aqueous
material as a preconsolidant, it was hoped that this water-sensitivity could be used to
facilitate manipulation of deformed plaster layers. Aqueous materials cause the clay
plasters to swell and soften. When combined with the application of pressure,
delaminated, deformed and cracked plaster layers show an improvement in readhesion,
relaxation and compaction. For more detailed information, please refer to
Tables: 8-14, 20, 21, 28-31.

4.1.3 Consolidation

Consolidation of disaggregating earthen plaster and mudbrick using ethyl silicates
proved to strengthen the plaster layers and the surface. Consolidated samples exhibited a
visually appreciable strengthening of the surface, as witnessed by their performance in
standardized tests in comparison to untreated samples. In all but the CRATerre Water

Chapter 4: Conclusions 185

Drop Test, the samples consolidated only with T-1919 performed more successfully than
the samples preconsolidated with the Aquazol50/T-1919 mixture, then consolidated with
T-1919. For more detailed information, please refer to
Tables: 12, 13, 14, 19-27,32,33

4.1.4 Facing adhesives for mural detachment

Six types of adhesives were tested as facing materials for mural
detachment. Two of these were effective, colletta, a collagen-based adhesive traditionally
used for the techniques of strappo and stacco, and polyvinyl alcohol, a water-soluble
synthetic adhesive. Both provided adequate adhesion over the entire painted surface to
permit detachment at a consistent level without penetrating the surface of the plaster.
They were not adversely affected by materials used for pretreatment. Both facilitated
detachment. Contraction of the colletta, particularly, causes lifting of the paint layer
without mechanical intervention. Both provided support to the detached plaster layer.
Polyvinyl alcohol facings were easily removed using multiple applications of water
poultices. Colletta facings were most easily removed using a combination of water
poultices and steam. Preliminary tests indicate that either may be used with minimal
alteration of the painted surface. For more detailed information, please refer to
Tables: 10, 11, 15, 16,28-38.

4.1.5 Detachment methods

Two levels of detachment developed for the detachment of paintings on lime
plaster were considered for this research, strappo, removal of the paint layer alone; and

Chapter 4: Conclusions 186

stacco, removal of the paint layer along with its plaster rendering. The removal of walls
in toto was considered in another phase of research. None of the facing adhesives or
techniques provided the control necessary to attain a consistent level of detachment
throughout a sample. Although the recipe and technique for the stacco and strappo forms
of detachment have been designed to accommodate the lifting of either the paint layer
alone or the paint with its rendering, it was not possible to remove discrete layers. For
more detailed information, please refer to
Tables: 28-38.

4.1.6 Reattachment to a new support

Although this aspect of work was not the focus of the research, a simple method
for attaching detached murals to a new backing was required in order to proceed with
testing. In order to continue research, the use of temporary materials was required to
support the paintings. Scotch Brand Foam Tape provided the contact adhesive and
degree of cushion required to support the fragile paintings. It must be emphasized that
this material is in no way meant to act as a conservation alternative. Use of this material
was required only by the need to proceed with testing to obtain results regarding
treatment compatibility and facing adhesives. For more detailed information, please refer
to Tables: 17,18.

4.1.7 Compatibility of treatments

A combination of conservation treatments was required. Once individual
materials were selected for the treatment of powdering paint, interlayer detachment, and

Chapter 4: Conclusions igy

cleavage of the surface layer, it was necessary to address their compatibility with one
another and with methods of mural detachment. Compatibility was assessed using the
following criteria:

■ Preservation of original appearance: color, gloss, texture, surface form

■ Success at achieving the objective: efficient detachment of the plaster stratum
as a whole or of a discreet layer

■ Successful reattachment of the painting to a new support

■ Removal of the facing adhesive with little or no damage to the painted plaster
surface

For more detailed information, please refer to Tables: 28-31.
4.1.8 Final testing program

Results of the preliminary testing program were evaluated to inform the final
testing program. The most effective treatments were used for the pretreatment and
detachment of ten prototype facsimiles. Six samples composed of fourteen layers of
plaster and seven replicated mural paintings were treated with eight spray applications of
Acryloid-B72, 3% in toluene. Water was used in combination with pressure to
preconsolidate the plasters. Paintings were detached using the polyvinyl alcohol and both
the stacco and strappo colletta facings.

The most successful detachments, carried out with the polyvinyl alcohol and
stacco colletta facings were then used to detach the samples consolidated with ethyl
silicates. Results based on visual assessment of the reattached paintings ranged from fair

Chapter 4: Conclusions 188

to very good. Most of the detached murals exhibited a slightly darkened surface and
uneven gloss due to residue from the facing adhesives. Most of the paintings exhibited
minimal losses of the plaster and paint, probably due to a lack of support behind the paint
layer. This lack of support, caused by swelling and delamination of the plaster layers
during facing removal, made it nearly impossible to completely remove adhesive residue.
Contraction of the residue may have been the cause of localized areas of delamination of
the painted surface. For more detailed information, please refer to Tables: 32-38.
4.1.9 Detachment of consolidated samples

The condition of samples consolidated with ethyl silicates did not differ
significantly from other samples after detachment. Consolidated samples detached using
the polyvinyl alcohol facing exhibited less cracking than other samples. Results were
otherwise unremarkable. For more detailed information, please refer to Tables: 32-38.
4.2 Final Results: Data

The following section contains data from all tests conducted during the laboratory
research program.

Chapter 4: Conclusions

189

Table 6

Visible Alteration of the Plaster Surface

H 2

Rabbit skin glue
(5% in H 2 0)

Ethanol

Isopropanol

• Quickly absorbed

• Glossy film

• No visible change

• Slight

formation

softening/swelling

Acetone

B-72

(7% in xylene)

(10% in xylene)

• No visible change

• Slight

discoloration

T-1919®

Aquazol 50®

(5% methanol)

(10% in ethanol)

( 5% methanol)

• No visible change

• Glossy film

• No visible change

formation

• Slight
discoloration

• Negligible
cracking

Aquazol 50®

(10% in isopropanol)

(5%inH 2 0)

(10% inH 2 0)

(5% in ethanol) /T-
1919® 50/50 (v/v)

• Slight

• No visible change

• Slight

• No visible change

discoloration

discoloration
• Cracking

Effects of solvents, adhesives, and consolidants on the optical properties of the plaster surface of a 6-by-6-inch
unpainted multiple-layer sample on gypsum board.

Chapter 4: Conclusions

190

Table 7

Surface Consolidation

Gum arabic (10% in
H,Q)

1 application

• *Darkens, becomes
transparent when
wet

• Less friable

• Little or no visible
alteration

Gum arabic (5% in
HjOj

1 application

• *

• Less friable

• Slight sheen

RhoplexAC-33(5% in

1 application

• *

• Less friable

• Filmy residue

Rhoplex AC-33(10% in

1 application

• *

• Less friable

• Slight filmy residue

B72(5-15% in toluene)

1 application

• *

• Negligible
darkening

• Less friable

B72(5-15% in toluene)
2 applications

• *

• Slight darkening

• Less friable

B72(5-15% in toluene)
3 applications

• *

• Darkened

• Less friable

Aquazol50® (5% in
isopropanol)

1 application

• *

• Yellowish
darkening

Aquazol50®(10% in
isopropanol)

1 application

• *

• Yellowish
darkening

• Less friable

Gelatin (5% in H,0)

1 application

• *

• Somewhat friable

• Little or no visible
alteration

Gelatin (5% in H,0)

2 applications

• *

• Sheen

• Not friable

Aquazol50®(5% in
H,Q)

1 application

• *

• Slight darkening

• Less friable

AquazolSO® (5% in
ethanol)

1 application

• *

• Less friable

• Little or no visible
alteration

AquazolSO® (5% in
ethanol)

2 applications

• *

• Less friable

• Little or no visible
alteration

Test designed to identify a material to strengthen the surface of the plaster while preserving its optical properties
carried out on 1 2-by- 1 2 inch unpainted multi-layer sample on concrete block.

Chapter 4: Conclusions

191

Table 8

Surface Consolidation and Preconsolidation with Readhesion I

Aquazol 50® (5% in ethanol)

1 application

• No swelling

• Remained brittle

• Delaminates from substrate

• Slightly darkened

Rhoplex AC-33 (1% in H.O)

1 application

• Swelling due to H20

• Malleable when rolling

• Can be pressed into plane

• Slightly darkened

• Weighted

• Cleavage partially relaxed

Gum arabic (10% in H,0)

1 application

• Slow absorption rate

• Swelling

• Can be pressed into plane

• Weighted

• Delaminates from substrate

Aquazol 50® (5% in H,0 )

2 applications

• Swelling

• Can be pressed into plane

• Weighted

H,Q

1 application

• Swelling

• Can be pressed into plane

• Remained flattened after
drying

• More compact

• Better adhered to substrate

Aquazol 50® (10% in
isopropanol)

2 applications

• No swelling

• No flattening

• Strengthened

• Well adhered to substrate

Gelatin (5% in H,0 )

1 application

• Swelling

• Partially relaxes cleavage

• Slightly darkened

• Weighted

• Still cleaving but apparently
strengthened

• Well adhered to substrate

Gum arabic (5% in H 2 )

1 application

• Swelling

• Partially relaxes cleavage

• Not quite dry

• Slightly flattened

• Weighted

• Separates from substrate

1 application

• Swelled

• Began to dissolve

• Can be pressed into plane

• Weighted

• Began to cleave after drying

This test, carried out on a 6-by-6-inch unpainted sample of sodium bentonite and water on gypsum board, aimed to
identify materials capable of improving the cohesive and adhesive strength of the plaster surface without altering its
optical properties and to improve interlayer attachment and relax cleavage.

Chapter 4: Conclusions

192

Table 9

Surface Consolidation and Preconsolidation with Readhesion II

Gelatin (5% in H,0)

1 application

• Slight darkening

• Less friable

Rhoplex AC-33 (1% in H,0)

1 application

• Uneven, slight darkening

• Uneven sheen

• Rolled

Rhoplex AC-33 (0.5% in H,0)
1 application

• Darkened

• Slight, filmy sheen

• Less friable

Rhoplex AC-33 (0.5% in H,0)
1 application

• Uneven, slight sheen

• Less friable

• Rolled

B-72(5-15% in toluene)
1 application

• Darkened

• Still slightly friable

Gelatin (5% in H,0)

1 application

• No darkening

• Still friable

Aquazol 50® (5% in ethanol)

1 application

• Slight darkening

• Slightly less friable

Aquazol 50® (5% in H 7 0)

2 applications

• Slightly less friable after 1 coat

• Rolled

• Weighted

• More compacted

Aquazol 50® (5% in H,0)

1 application

• Slight darkening

• Still friable

1 application

• Friable

• More compacted

This test, carried out on one 12-by-12-inch painted multi-layer concrete block, aimed to strengthen the plaster surface
while preserving the original appearance and improve interlayer attachment between plaster layers exhibiting cleavage
and separation.

Chapter 4: Conclusions

193

Table 10

Readhesion / Facing Adhesives

B-72 (8% in xylene): brushed through tissue

• Stabilizes paint layer

• Slightly impedes H : absorption

• Causes brownish discoloration

• Divided in half (D1A and DIB) for additional treatment

D1A

H 2 0: spray applied

• Permits manipulation without smearing

Aquazol 50®(5% in H^O): spray applied

• Improves delamination

• Improves adhesion between pieces

• Relaxes cleavage

• Slightly strengthens

DIB

H 2 0: spray applied

• Improves delamination in isolated areas

• Permits manipulation without smearing

• Improves adhesion between pieces

H 2 0: spray applied

• Improves adhesion to the substrate and between pieces

• Swells the cleaved plaster surface which is then flattened by applying pressure over mylar

D2A

Aquazol 50®(5% in H?0): spray applied

• Swells the plaster surface which is then
flattened by applying pressure over mylar

• Improves craquelure, cleavage

• Allows compaction and readhesion between
cracked pieces

• Dissolves paint layer without preliminary B-72
application

D2B

B-72 (8% in xylene): brushed through tissue
Aquazol 50®[5% in H-.O): spray applied

• Slightly darkens

• Swells the plaster surface which is then
flattened by applying pressure over mylar

• Improves craquelure, cleavage

• Allows compaction and readhesion between
cracked pieces

• Preserves paint layer

Plextol B500(50% solids): brushed through tissue

• Fully adheres plaster to tissue

• Permits full-scale delamination of plaster

Results of tests designed to evaluate the performance of materials as adhesives for separated plaster layers and as facing
adhesives for mural detachment.

Chapter 4: Conclusions 194

Table 11

Readhesion / Facing Adhesives

B-72 (8% in xylene): brushed through tissue
H ^O: spray applied

Readheres plaster to substrate

H : application improves craquelure, cleavage

Allows compaction and readhesion between cracked pieces

B-72 application inhibits loss of paint layer

Causes slight discoloration

Microcrystalline Wax/Mineral Spirits

Incomplete penetration

Significantly discolors plaster and paint layer

B-72 (3% in xylene)- brushed through tissue
Aquazol 50®(5% in H 2 Q): spray applied

• Causes adhesion of the plaster layer to the mylar

• Causes complete delamination of the surface from the substrate

• Permits manipulation to relax cleavage and readhere cracked pieces

B-72 (3% in xylene)- brushed through tissue
Aquazol 50®(5% in H?Q): spray applied

• Permits manipulation to relax cleavage and readhere cracked pieces

• Preserves the paint layer

• Causes slight discoloration

• Provides limited readhesion to the substrate

Microcrystalline wax / Mineral spirits

Exhibits incomplete penetration

Significantly discolors plaster as well as the paint layer

Microcrystalline wax / Mineral spirits

Exhibits incomplete penetration

Significantly discolors plaster as well as the paint layer

Results of tests designed to evaluate the performance of materials as adhesives for separated plaster layers and as facing
adhesives for mural detachment.

Chapter 4: Conclusions

195

Table 12

Surface Consolidation of Powdering Paint with Consolidation and Readhesion

C-4: 2 ground/finish phases, 2 paintings on 12" gypsum board panel

C-4a

C-4b

B-72 (8% in xylene, spray applied): 3 coats

B-72 (8% in xylene, spray applied): 2 coats

H 2 (spray applied)

Weighted

• B-72 fixes paint surface

• Without application of weight, H 2 causes

• Permits manual manipulation of deformation

delamination from substrate

• Does not provide strength

• Permits manipulation of deformation

• HnO causes delamination from substrate

• Slightly darkens

Then:

• Relaxes cleavage

• Aquazol®50 (5% in H 2 0, spray applied)

• Permits readhesion to substrate

• Slightly darkens

• Relaxes cleavage slightly less than C-4a

C-4c

C-4d

B-72 (3% in xylene, spray applied): 2 coats

B-72 (3% in xylene, spray applied): 2coats

Aquazol 50® (5% in H20, brush applied)

Aquazol 50® (5% in isopropanol, brush applied)

• H 2 causes plaster to swell

• Permits manipulation of deformation

• Relaxes cleavage

• Strengthens bond between plaster pieces

• Does not strengthen bond between large

• Causes yellow discoloration

detached pieces and substrate

Result of test carried out on three 12-by- 12-inch painted multi-layer gypsum board panels designed to treat the loss of
adhesion and cohesion of the paint layers without altering the optical properties of the plaster or painting.

Chapter 4: Conclusions

196

Table 13

Surface Consolidation of Powdering Paint with Consolidation and Readhesion

C-5: 2 ground/finish phases, 2 paintings on 12" gypsum board panel

C-5a

C-5b

B-72 (3% in xylene, spray applied): 2 coats

H 2 (spray applied): 2-3 coats

• B-72 consolidates paint surface

• H : causes swelling and delamination of

• HiO causes swelling and delamination of

plaster from substrate

• Permits manipulation of deformation

• Relaxes cleavage

• Does not strengthen bond between large

detached pieces and substrate

• Appears to lessen the severity of voids

C-5c

C-5d

B-72 (3% in xylene, spray applied): 2 coats

H 2 (spray applied): 2-3 coats

• B-72 consolidates paint surface

• HiO causes swelling and delamination of

• H 2 causes swelling and delamination of

plaster from substrate

• Permits manipulation of deformation

• Relaxes cleavage

• Broken portion of outer corner remains

• Appears to lessen the severity of voids

detached

Result of test carried out on three 1 2-by- 12-inch painted multi-layer gypsum board panels designed to treat the loss of
adhesion and cohesion of the paint layers without altering the optical properties of the plaster or painting.

Chapter 4: Conclusions

197

Table 14

Surface Consolidation of Powdering Paint with Consolidation and Readhesion

C-6: 2 ground/finish phases, 2 paintings on 12" gypsum board panel

C-6a

C-6b

B-72 (3% in .xylene, spray applied): 2 coats

• B-72 consolidates paint surface

C-6c

C-6d

B-72 (3% in xylene, spray applied): 2 coats

B-67 (3% in mineral spirits, brushed on): 2 coats

• B-72 consolidates paint surface

• B-67 causes significant surface discoloration

• B-72 consolidates paint surface

• Does not readhere delaminated pieces

• Reserved for later research

• Appears to strengthen plaster (intergranular

cohesion): pieces previously too fragile to

handle without crumbling are now stable

enough to be picked up

Chapter 4: Conclusions

198

Table 15

Facing Adhesives: Methods and Materials
Preparation for Stacco and Strappo

Adhesive

Adhesive performance

Result of Poultice
Application

Final Appearance of Plaster

Colletta
(full strength)

• Very viscous

• Application
repositioned friable
surface material

• Even distribution
required effort

• Bonded quickly

• Contraction of glue
caused delamination
from substrate

• Severe swelling
of the expansive
clays caused by
HiO poultice

• Exacerbation of
interlayer
delamination as
well as full-scale
separation from
the substrate

• Complete reversibility
impossible due to deformation
and delamination from the
substrate

• Significant pigment and plaster
losses caused by deformation
and removal of adhesive
residue

• Surface slightly yellowed

PVOH

(20%)

• Less viscous

• Application did not
damage surface

• Easy to handle

• Distributed evenly

• No adverse
reaction to H 2
poultice

• Adhesive residue easily
removed with H20 soaked
pads

• Negligible pigment loss

• Surface slightly darkened

Acryloid B-72

(20%)

• Viscous

• Application caused
some repositioning
of friable surface
material

• Even distribution
required minimal
effort

• Slight swelling

• Separation
between layers
exhibiting
delamination
prior to treatment
was exacerbated

• Residue removal only partially
successful, even after two
additional poultice treatments

• Pigment loss

• Surface is darkened and
exhibits a filmy residue

Acryloid B-67

(20%)

• Viscous

• Application caused
some repositioning
of friable surface
material

• Even distribution
required minimal
effort

• Significantly

longer dwell time
for poultice to
soften adhesive
to an acceptable
degree

• Residue removal unsuccessful
even after two additional
poultice treatments

• Pigment loss

• Surface is significantly
darkened and exhibits an
unacceptable sheen

Results of test evaluating the performance of traditional and non-traditional materials used as facing adhesives

Chapter 4: Conclusions

199

Table 16

Facing Adhesives: Methods and Materials
Preparation for Stucco and Strappo

Adhesive

Adhesive performance

Result of Poultice
Application

Final Appearance of Plaster

Plextol B500

• Viscous

• Application did not
damage surface

• Easy to handle

• Distributed evenly

• No success with
acetone/toluene
poultice

• Adhesive swells
but does not
resolubilize

• Facings stiffened

• Entire layer adhered to facing

• Delaminated as a whole

Vinamul 6825

• Viscous

• Application did not
damage surface

• Easy to handle

• Distributed evenly

• Limited success
with methanol
poultice

• Darkened

• Uneven filmy residue

Results of test evaluating the performance of traditional and non-traditional materials used as facing adhesives

Chapter 4: Conclusions

200

Table 17

New Support/Backing: Traditional Materials

Sample E-7

Surface consolidant: none
Facing adhesive: Colletta (stacco)
Support: hydraulic lime/sand/PVA
Removal: H 2 poultice

• Clays swelled

• Painting detached from new support

• Residue removal caused abrasion of paint layer

Sample E-13

Surface consolidant: none
Facing adhesive: Colletta (stacco)
Support: hydraulic lime/sand/PVA
Removal: H 2 poultice

• Clays swelled

• Poor bond between painting and new support
caused extensive loss of plaster and pigment

• Residue removal caused abrasion of paint layer

Sample E-14

Surface consolidant: B-72 (3% in toluene) x 1
Facing adhesive: Colletta (strappo)
Support: hydraulic lime/sand/PVA
Removal: H 2 poultice

• Clays swelled

• Poor bond between painting and new support
and residue removal caused losses of both
plaster and pigment

Sample E-15

Surface consolidant: none
Facing adhesive: Colletta: (stacco)
Support: hydraulic lime/sand/PVA
Removal: H 2 poultice

• Clays swelled

• Poor bond between painting and new support
and residue removal caused losses of both
plaster and pigment

Sample E-8

Surface consolidant: B-72 (3% in toluene) x 1
Facing adhesive: Colletta: (strappo)
Support: hydraulic lime/sand/clay/PVA
Removal: H 2 poultice followed by steam

• Unnecessary stress on pictorial layer due to
severe cracking of support plaster

• Inconsistent bond between painting and
substrate

• Residue removal caused due to abrasion of
paint layer

Sample E-10

Surface consolidant: B-72 (3% in toluene) x 1
Facing adhesive: PVOH
Support: hydraulic lime/sand /clay/PVA
Removal: H 2 poultice

• Losses caused by premature removal

• Inconsistent bond between painting and
substrate

Sample E-ll

Surface consolidant: none
Facing adhesive: Colletta: (stacco)
Support: hydraulic lime/sand/clay/PVA
Removal: H 2 poultice followed by steam
• Inconsistent bond between painting and
substrate

Result of tests designed to identify an appropriate new support material

Chapter 4: Conclusions

201

Table 18

New Support/Backing: Non-Traditional Support Materials

Sample E-5

Surface consolidant: none

Facing adhesive: Colletta (stacco-100%)

Support: plaster support failed, detached painting
backed with Plaster of Paris/gauze strips and
adhered to foam core with PVA around
perimeter

Removal: H 2 poultice followed by steam

• Failed

• Destruction of sample

Sample E-4

Surface consolidant: none

Facing adhesive: Colletta (stacco-100%)

Support: PVA applied to perimeter of detached

painting, adhered to foam core
Removal: H : poultice followed by steam

• Facing could not be removed

• Destruction of sample

Sample E-6

Surface consolidant: B-72 (3% in toluene) x 1
Facing adhesive: Colletta (strappo-\00%)
Support: PVA applied at perimeter of detached

painting and adhered to foam core
Removal: H 2 poultice followed by steam

• Inconsistent bond between painting and
substrate

• Residue removal caused abrasion of paint layer

• Significant superficial checking

Sample E-12

Surface consolidant: none
Facing adhesive: Colletta (strappo)
Support: foam tape
Removal: H 2 poultice

• Satisfactory as a temporary support

• Painting well adhered to substrate, facilitated
facing removal

• Slight residue could probably have been
removed with steamer

Sample E-9

Surface consolidant: B-72 (3% in toluene) x 1
Facing adhesive: Colletta (strappo)
Support: Plasti-Tak
Removal: H : poultice followed by steam

• Worked well as a temporary support

• Negligible residue

• No evidence of deterioration

Result of tests designed to identify an appropriate new support material

Chapter 4: Conclusions

202

Table 19

ASTM D4214-89: Chalking Test-Results

Test Method A
Method D-659

Test Method B

Stroke Method

Untreated

The chalk mark compares to
rating No. 6 in Photographic-
Reference Standard #1.

Acryloid B-72

B-72

The chalk mark compares to
rating No. 8 in Photographic-
Reference Standard #1.

The chalk mark compares to

rating No. 8 in Photographic

Reference Standard #1.

T-1919
ES

The chalk mark compares to
rating No. 8 in Photographic-
Reference Standard #1.

The chalk mark compares to
rating No. 8 in Photographic-
Reference Standard #1 .

Aquazol50® /T-1919 +

T-1919

AQES

The chalk mark compares to
rating No. 8 in Photographic-
Reference Standard #1 .

The chalk mark compares to
rating No. 8 in Photographic-
Reference Standard #1.

Results of chalking test designed to assess affects of treatment on multi-layer painted plaster samples

Chapter 4:

Conclusions

203

Table 20

ASTM D3359-90: Measuring Adhesion by Tape Test

Rating Scale: Test Method A

No peeling or removal

Trace peeling or removal along incisions

Jagged removal along incisions up to ' / 16 in. ( 1 .6mm) on either side

Jagged removal along most of incisions up to /a in. (3.2mm) on either side

Removal from most of the area of the X under the tape

Removal beyond the area of the X

Sample

Test Method A.l

Test Method A.2

Rating: 2A

Rating

B72

Rating: 4A

Rating

Rating: 4A

Rating

AQES

Rating: 4A

Rating

Result of test designed to assess affects of treatment on the adhesive properties of multi-layer painted plaster samples

Chapter 4: Conclusions

204

Table 21

ASTM D3359-90: Measuring Adhesion by Tape Test

Rating Scale: Test Method B

The edges of the cuts are completely smooth; none of the squares of the lattice
is detached.

Small flakes of the coating are detached at intersections; less than 5% of the
area is affected.

Small flakes of the coating are detached along edges and at intersections of
cuts. The area affected is 5 to 15% of the lattice.

The coating has flaked along the edges and on parts of the squares. The area
affected is 15 to 35% of the lattice.

The cutting has flaked along the edges of cuts in large ribbons and whole
squares have detached. The area affected is 35 to 65 % of the lattice.

Flaking and detachment worse than Grade 1

Sample

Test Method B.l

Test Method B.2

Rating: IB

B72

Rating: 2B

Rating: IB

ESB72

Rating: 4B

Rating:3B

AQESB72

Rating: 3B

Rating:2B

Result of test designed to assess affects of treatment on the adhesive properties of multi-layer painted plaster samples

Chapter 4: Conclusions

205

Table 22

CRATerre Water Resistance Data: Untreated

Sample

Time Elapsed
(minutes)

Pigment loss
(mm)

Plaster loss
(mm)

Depth of erosion

(no. of layers

affected)

UT-1

6mm

12.5mm

2mm

1 8 mm

3 mm

18mm

5 mm

19 mm

18mm

22mm

20mm

25mm

25 mm

30mm

25 mm

30mm

100

30mm

110

30mm

120

30mm

Table 23

CRATerre Water Resistance Data: Untreated

Sample

Time Elapsed
(minutes)

Pigment loss
(mm)

Plaster loss
(mm)

Depth of erosion

(no. of layers

affected)

UT-2

17.5mm

20mm

22mm

24 mm

3 mm

24mm

3.4mm

25mm

7 mm

25mm

1 1mm

26mm

15.4mm

26mm

19mm

100

26mm

20.6mm

110

26mm

25.1mm

120

26mm

Results of test to assess the water resistance of treated and untreated samples

Chapter 4: Conclusions

206

Table 24

CRATerre Water Resistance Data: Consolidated

Sample

Time Elapsed
(minutes)

Pigment loss
(mm)

Plaster loss
(mm)

Depth of erosion

(no. of layers

affected)

ES-1

9 mm

15mm

16mm

2 mm

16mm

5 . 1 mm

20mm

10mm

20mm

1 0mm

20 mm

10.2mm

20mm

10.3mm

100

20mm

10.3mm

110

20mm

10.4mm

120

20mm

10.5mm

Table 25

CRATerre Water Resistance Data: Consolidated

Sample

Time Elapsed
(minutes)

Pigment loss
(mm)

Plaster loss
(mm)

Depth of erosion

(no. of layers

affected )

ES-2

1mm

2mm

2 mm

100

2 mm

110

2 mm

120

2mm

Results of test to assess the water resistance of treated and untreated samples

Chapter 4: Conclusions

207

Table 26

CRATerre Water Resistance Data: Preconsolidant + Consolidant

Sample

Time Elapsed
(minutes)

Pigment loss
(mm)

Plaster loss
(mm)

Depth of erosion

(no. of layers

affected)

AQES-1

2 mm

3 mm

7.5mm

10mm

12mm

15 mm

16mm

100

1 6mm

110

1 6mm

120

16mm

Table 27

CRATerre Water Resistance Data: Preconsolidant + Consolidant

Sample

Time Elapsed
(minutes)

Pigment loss
(mm)

Plaster loss
(mm)

Depth of erosion

(no. of layers

affected)

AQES-2

0.5mm

0.7mm

100

0.7mm

110

1mm

120

1mm

Results of test to assess the water resistance of treated and untreated samples

Chapter 4: Conclusions

208

Table 28

Compatibility of Treatments

Colletta: Stacco facings (50% in H 2 0)

Sample D-l

Sample D-4

Surface consolidate: B-72 (3% in Toluene) 2 spray

applications

Preconsolidant: H 2

Preconsolidant: Aquazol 50®/H 2 O (5%)

Facing Adhesive: Colletta (stacco-50%)

• Effortless full-scale detachment of plaster

• Slightly yellowed

stratum from substrate

• Minor losses caused by premature facing

• Slightly darkened

removal

• Simple, clean removal of facing with steamer

• Interlayer separation

• Irregular cracking

• Superficial checking

Sample D-l 9

Preconsolidant: Aquazol 50®/H 2 O (5%)

Surface consolidate: Blair Spray Fix

Facing Adhesive: Colletta (stacco-50%)

• Yellowed

• Residue of facing adhesive remains on surface

• Some loss of plaster due to premature facing

removal (± 5%)

• Use of steamer facilitated removal after

preliminary losses

• Interlayer separation

• Irregular cracking

Evaluation of the compatibility of materials used for surface consolidation, preconsolidation and facing adhesion

Chapter 4: Conclusions

209

Table 29

Compatibility

of Treatments

Colletta: Strappo facings (75% in H2O)

Sample D-2

Sample D-3

Surface consolidant: B-72 (3% in Toluene) 2 spray

applications

Preconsolidant: H 2

Preconsolidant: Aquazol 50®/H 2 O (5%)

Facing Adhesive: Collettaf strappo- 75%)

Facing Adhesive: Colletta( strappo- 75%)

• Slightly yellowed

• Most of painting detached prior to manipulation

• Negligible losses

due to contraction of the glue

• Irregular cracking

• Unacceptable powdery losses of both plaster and

pigment (± 20%)

• Significant areas of plaster swelled and

disintegrated after the application of a H :

poultice (damage attributed to HiO rather than

premature facing removal because plaster

disintegrated rather than peeling off with the

gauze)

• Slight residue, damage halted removal

Sample D-18

Sample D-21

Surface consolidant: H 2

Surface consolidant: B-72 (3% in Toluene)] 1 spray

Fixative: Blair Spray Fix

applications

Facing Adhesive: Colletta (strappo-75%)

• Most of painting detached prior to manipulation

due to contraction of the glue

• Slight yellowing

• Full-scale detachment from substrate

• Partial disintegration of plaster due to H :

poultice (losses only reach to next layer)

• Plaster and pigment loss (± 15%)

• Significant interlayer separation

• Slight residue

Evaluation of the compatibility of materials used for surface consolidation, preconsolidation and facing adhesion

Chapter 4: Conclusions

210

Table 30

Compatibility

of Treatments

PVOH facings

(25%inH 2 0)

Sample D-5

Sample D-6

Surface consolidant: B-72 (3% in Toluene) 2 spray

applications

Preconsolidant: H 2

Preconsolidant: Aquazol 50®/H 2 O (5%)

Facing Adhesive: PVOH (25%)

• Effortless detachment

• Negligible loss of plaster and pigment (± 2%)

• Ground layer remained attached to substrate

• Slight darkening

• Slight discoloration

• Irregular cracking

• Negligible pigment loss

• Superficial checking

Sample D-12

Sample D-17

Surface consolidant: B-72 (3% in Toluene) 11 spray

Surface consolidant: Blair Spray Fix

applications

Facing Adhesive: PVOH (25%)

Facing Adhesive: B-72 8%, later removed

Facing Adhesive II: PVOH (25%)

• Removal made somewhat more difficult due to

• Simple removal using H : poultice(quick

prior application of B-72 (8%)

resolubilization of the adhesive allows for

• Duration of poultice treatment was significantly

removal of the facing before significant

increased before the adhesive softened to an

deterioration of the plaster can occur)

acceptable degree

• Negligible discoloration

• Irregular cracking

Evaluation of the compatibility of materials used for surface consolidation, preconsolidation and facing adhesion

Chapter 4: Conclusions

211

Table 31

Compatibility of Treatments

Plextol B500 / Vinamul 6825 facings

Sample D-7

Sample D-9

Surface consolidant: B-72 (39c in Toluene) 2 spray

Surface consolidant: B-72 (3% in Toluene) 2 spray

applications

Preconsolidant: Aquazol 50®/H 2 O (59c)

Preconsolidant: H 2

Facing Adhesive: Plextol B500

• All attempts at removal failed

Sample D-8

Sample D-10

Surface consolidant: B-72 (39c in Toluene) 2 spray

Surface consolidant: B-72 (3% in Toluene) 2 spray

applications

Preconsolidant: Aquazol 50® / H 2 (59c)

Preconsolidant: H 2

Facing Adhesive: Vinamul 6825

Significant darkening

• Partial disintegration of plaster

• Extensive destruction of plaster (application of

• Residue removal abraded paint surface

the alcohol-based solvent dissolved the foam

• Residue of adhesive remains on surface

tape, the adhesive, and the foam core support)

• Application of alcohol-based solvent dissolved

• Unacceptable deformation and cracking of

the adhesive of the foam tape support

plaster

• Deterioration of the support material left little

• Straw component in underlying plaster layers

to stabilize plaster

may have exacerbated some of the difficulties

• Irregular cracking

• Flaking

Evaluation of the compatibility of materials used for surface consolidation, preconsolidation and facing adhesion

Chapter 4: Conclusions

212

Table 32

Results of Final Detachment Tests

Sample

Treatment

Final Result and Appearance

Sample A

Surface consolidant: B-72 (3% in toluene) x 8

Preconsolidant: H ;

Facing Adhesive: Collelta (slacco-50%)

• Indiscriminate detachment

• Remaining layers fragmented

• Slight, irregular surface cracking

• Slightly darkened

• Losses: negligible

• Slight adhesive residue

Sample B

Surface consolidant: B-72 (3% in toluene) x 8

Preconsolidant: H ;

Facing Adhesive: PVOH (20% in H 2 0)

• Indiscriminate detachment

• Remaining layers fragmented

• Slight, irregular surface cracking

• Slightly darkened

• Losses: negligible

• Slight adhesive residue

• Some deformation and interlayer
separation due to swelling of plasters

Sample C

Surface consolidant: B-72 (3% in toluene) x 8

Preconsolidant: H 2

Facing Adhesive: PVOH (20% in H : 0)

• Indiscriminate detachment

• Remaining layers fragmented

• Slight, irregular surface cracking

• Slightly darkened

• Losses: significant, appear as
powdery surface

• Slight adhesive residue

• Some interlayer detachment due to
swelling of plasters

Sample D

Surface consolidant: B-72 (3% in toluene) x 8

Preconsolidant: H ;

Facing Adhesive: Colletta (slacco-50%)

• Most of painting detached prior to
manipulation due to contraction of
glue

• Indiscriminate detachment

• Remaining layers fragmented

• Slight, irregular surface cracking

• Slightly darkened

• Losses: negligible; concentrated
around edges

• Slight adhesive residue

• Some interlayer detachment due to
swelling of plasters

Sample < > ,

Surface consolidant: B-72 (3% in toluene) x 1

Preconsolidant: *

Consolidant: T-191 9

Facing Adhesive: PVOH (20% in H 2 0)

• Indiscriminate detachment

• Remaining layers fragmented

• Slight, irregular surface cracking

• Slight darkening on consolidation

• Losses: negligible(slight powdering)

• Slight adhesive residue, attempts at
removal caused abrasion of paint
layer

Chapter 4: Conclusions

213

Table 33

Results of Final Detachment Tests

Sample

Treatment

Final Result and Appearance

Sample G2

Surface consolidant: B-72 (3% in toluene) x 1

Preconsolidant: *

Consolidant: T-19 19

Facing Adhesive: Colletta (stacco-50%)

• Indiscriminate detachment

• Remaining layers fragmented

• Slight, irregular surface cracking

• Slight darkening on consolidation

• Losses: significant, approx. 15%

Sample H

Surface consolidant: B-72 (3% in toluene) x 8

Preconsolidant: H :

Facing Adhesive: Colletta (strappo-15%)

• Most of painting detached prior to
manipulation, due to contraction of
glue

• Indiscriminate detachment

• Remaining layers fragmented

• Slight irregular surface cracking

• Slightly darkened

• Losses: significant; concentrated
around edges

• Slight adhesive residue

• Some deformation and interlayer
separation due to swelling of plasters

Sample I

Surface consolidant: B-72 (3% in toluene) x 8

Preconsolidant: H :

Facing Adhesive: Colletta {strappo-15%)

• Most of painting detached prior to
manipulation due to contraction of
glue

• Indiscriminate detachment

• Remaining layers fragmented

• Losses: negligible, concentrated
around edges; powdery

• Slight adhesive residue

• Some deformation and interlayer
separation due to swelling ot plasters

Sample Ji

Surface consolidant: B-72 (3% in toluene) x 1

Preconsolidant: FLO

Consolidant T- 191 9

Facing Adhesive: PVOH (20% in H : 0)

• Most of painting (approx. 95%)
detached between 13' h and 14 lh layers
(most strappo-like) behavior

• Slight darkening on consolidation

• Losses: negligible

• Slight adhesive residue; attempts at
removal caused abrasion of paint
surface

Sample J2

Surface consolidant: B-72 (3% in toluene) x 1

Preconsolidant: FLO

Consolidant: T-l 919

Facing Adhesive: Colletta (stacco-50%)

• Most of painting detached prior to
manipulation due to contraction of
glue

• Remaining layers fragmented

• Slight darkening on consolidation

• Losses: negligible (approx. 7%)
concentrated around edges

• Sliaht adhesive residue

Chapter 4: Conclusions

214

Table 34

% Distribution of Each Layer Remaining on Substrate After Detachment:

Sample

Stacco

PVOH

Stacco

AQES/
PVOH

G 2

AQES/
Stacco

Strappo

T-1919/
PVOH

T-1919/
Stacco

Layer 13

2.0%

16.0%

76.0%

10.0%

94.5%

Layer 12

3.0%

28.0%

1.0%

Layer 11

1.0%

45.0%

1.0%

53.0%

19.0%

Layer 10

2.0%

4.0%

1 .0%

10.0%

2.0%

Layer 9

31.0%

15.0%

4.0%

0.5%

1.0%

3.0%

Layer 8

7.0%

35.0%

2.0%

20.0%

1.0%

Layer 7

6.0%

35.0%

2.0%

10.0%

5.5%

Layer 6

0.5%

25.0%

8.0%

20.0%

5.5%

2.0%

4 1 .0%

Layer 5

3.0%

4.0%

2.0%

35.0%

5.0%

Layer 4

0.5%

1 .0%

2.5%

8.0%

0.5%

3.0%

Layer 3

15.0%

21.0%

8.0%

Layer 2

1.0%

14.0%

40.0%

0.5%

5.0%

0.5%

1.0%

26.0%

Layer 1

36.0%

75.0%

7.0%

Terra cotta

15.0%

3.0%

10.0%

1.0%

6.0%

Chapter 4: Conclusions

215

Table 35

Results of ASTM Cracking, Flaking and Checking Tests

Sample

Treatment

Rating

Cracking

Flaking

Checking

Untreated

N/A

Type 1:
Irregular
Rating: 8

Rating 8

Stacco

Surface consolidant: B-72 (3% in toluene) x 8

Preconsolidant: H 2

Facing adhesive: Colletta (stacco-50%)

Type 1 :
Irregular
Rating: 6

Rating: 6

B
PVOH

Surface consolidant : B-72 (3% in toluene) x 8

Preconsolidant: H 2

Facing adhesive: PVOH (20% in H : 0)

Type 1:
Irregular
Rating: 6

Rating: 4

Rating: 8

C
PVOH

Surface consolidant: B-72 (3% in toluene) x 8

Preconsolidant: H 2

Facing adhesive: PVOH (20% in H 2 0)

Type 1:
Irregular
Rating: 6

Rating: 2

Rating: 8

Stacco

Surface consolidant: B-72 (3% in toluene) x 8

Preconsolidant: H 2

Facing adhesive: Colletta (stacco-50%)

Type 1:
Irregular
Rating: 6

Rating: 6

Rating: 8

G,
AQES/
PVOH

Surface consolidant: B-72 (3% in toluene) x 1

Preconsolidant: 50/50(v/v)Aquazol (5% in

EtOH)+T-1919

Consolidant: T-1919

Facing adhesive: PVOH (20% in H 2 0)

Type 1:
Irregular
Rating: 8

Rating: 6

Rating: 8

G 2
AQES/

Stacco

Surface consolidant: B-72 (3% in toluene) x 1

Preconsolidant: 50/50(v/v)Aquazol (5%in

EtOH)+ T-1919

Consolidant: T-1919

Facing adhesive: Colletta (stacco-50%)

Type 1 :
Irregular
Rating: 6

Rating: 4

Rating: 8

Chapter 4: Conclusions

216

Table 36

Results of ASTM Cracking, Flaking and Checking Tests

Sample

Treatment

Rating

Cracking

Flaking

Checking

Untreated

N/A

Type 1:
Irregular
Rating: 8

Rating:

Strappo

Surface consolidant: B-72 (3% in toluene) x 8

Preconsolidant: H 2

Facing adhesive: Colletta (strappo-75%)

Type 1:
Irregular
Rating: 4

Rating:

Strappo

Surface consolidant: B-72 (3% in toluene) x 8

Preconsolidant: H 2

Facing adhesive: Colletta (strappo-75%)

Type 1:
Irregular
Rating: 6

Rating:

Ji
T-1919/
PVOH

Surface consolidant: B-72 (3% in toluene) x 1

Preconsolidant: H 2

Consolidant: T-1919

Facing adhesive: PVOH (20% in H 2 0)

Type 1:
Irregular
Rating: 8

Rating:
6

Rating:

h
T-1919/

Stacco

Surface consolidant: B-72 (3% in toluene) x 1

Preconsolidant: H 2

Consolidant: T-1919

Facing adhesive: Colletta (stacco-50%)

Type 1:
Irregular
Rating: 4

Rating:
6

Rating:

Chapter 4: Conclusions

217

Table 37

Munsell Color Notations: Before and After Treatment

Sample A: Stacco

Sample I

:PVOH

White

Red

Black

White

Red

Black

Untreated

N9.25/
N9.5/

2.5R 6/4
5R 5/4

N2.25/-
N3.25/

Untreated

N9.5/

2.5R 6/4

5R 5/4

N2.0/-

N4.0/

Treated

5Y9/1

7.5R 4/4
7.5R 5/4
7.5R 6/4

N2.25/-
N3.5/

Treated

N9.5/

5R4/6
5R5/6
7.5R 6/6
7.5R7/4

N2.75/-
N3.25/

Sample C: PVOH

Sample D: Stacco

White

Red

Black

White

Red

Black

Untreated

N9.5/

7.5R 5/6
7.5R 6/6

N2.0/-

N4.0/

Untreated

N9.5/

7.5R 5/6
7.5R 6/6

N2.0/-

N4.0/

Treated

N9.5/

7.5R 6/4
7.5R 4/4

N2.75/-
N3.0/

Treated

5Y9/1

7.5R 4/6
7.5R 5/6

N2.75/-

N3.25/

Sample G: AQES/PVOH

Sample G: AQES/Stacco

White

Red

Black

White

Red

Black

Untreated

N9.5/

7.5R 5/6
7.5R 6/6

N2.0/-

N4.0/

Untreated

N9.5/

7.5R 5/6
7.5R 6/6

N2.0/-

N4.0/

Treated

5Y9/1

1 OR 4/6
1 OR 5/6

N2.75/
N3.25/
N4.25/

Treated
G 2

5Y9/1

1 OR 4/6
1 OR 6/6

N2.75/-
N3.25/

Sample H: Strappo

Sample I: Strappo

White

Red

Black

White

Red

Black

Untreated

N9.25/
N9.5/

7.5R 5/6

7.5R 6/6

N2.25/-
N3.5/

Untreated

N9.5/

7.5R 5/6
7.5R 6/6

N2.0/-
N4.0/

Treated

5Y9/1

1 OR 4/6
1 OR 5/6

N2.75/-
N3.25/

Treated

5Y9/1

1 OR 4/6
1 OR 5/6

N3.0/-

3.25/

Sample J: T-1919/PVO]

Sample J: T-1919/Stacco

White

Red

Black

White

Red

Black

Untreated

N9.5/

7.5R 5/6

1 OR 7/4

N2.75/-

N3.5/

Untreated

N9.5/

7.5R 5/6
1 OR 7/4

N2.75/-

N3.5/

Treated
Ji

5Y9/1

1 OR 4/6
1 OR 5/6

N2.75/
N3.25/

Treated

5Y9/1

1 OR 4/6
1 OR 6/6

N2.75/-

N3.25/

Key:

Samples are listed with facing adhesives

AQES refers to samples preconsolidated with the Aquazol50®/T-1919 mixture, then consolidated with

T-1919

T-1919 refers to those samples preconsolidated with T-1919

Chapter 4

Conclusions

218

Table 38

Munsell Color Notations: Interpretation

Sample

Hue

Value

Chroma

Stacco

Unpainted

Red

Black

B
PVOH

Unpainted

lied

Black

C
PVOH

Unpainted

Red

Black

Stacco

Unpainted

Red

Black

AQES/
PVOH

Unpainted

Red

Black

G 2
AQES/

Stacco

Unpainted

Red

Black

Strappo

Unpainted

Red

Black

Strappo

Unpainted

Red

Black

Ji
T-1919/
PVOH

Unpainted

Red

Black

T-1919/
Stacco

Unpainted

Red

Black

Key:

Samples are listed along with facing adhesives

AQES refers to samples preconsolidated with Aquazol50®/T-1919, and then consolidated with T-ll

T-1919 refers to those samples consolidated with T-1919

(+) in hue column refers to yellowing of the surface

= color stayed the same

- color lightened

Chapter 4: Conclusions 219

4.3 Conclusions

Two systems of treatment were successful for strengthening the paint and plaster
surfaces and facilitating detachment of the mural paintings. The first two steps were the
same for each: surface consolidation of the painted plaster with an acrylic resin, Acryloid B-
72, followed by preconsolidation using water combined with the application of pressure.
The use of Acryloid B-72 as a surface consolidant yielded excellent results. Multiple spray
applications of a 3% solution in toluene strengthened the surface without altering the optical
properties of the plaster or paintings, and did not interfere with subsequent treatment.
Furthermore, Acryloid B-72 made the painted surface insoluble in water, which allowed for
preconsolidation of the clay-based renders using aqueous materials. Results of the
experimental program revealed that water facilitated the manipulation of deformation and
cleavage, and increased adhesion between layers.

The third step, application of a facing adhesive for mural detachment, varied
between thee two systems. For one, colletta was used, and for the other, polyvinyl alcohol.
Facings for mural painting detachment must provide consistent coverage to all the painted
surface and adequate support to the detached mural painting without penetration of the
plaster. Colletta, a collagen-based adhesive traditionally used for the detachment of
paintings on lime-based plasters, and polyvinyl alcohol, an adhesive previously tested in the
American Southwest for the detachment of paintings on earthen plaster, were successful at
each of the objectives. They permitted easy removal of the facing materials and residue and
were not adversely affected by other forms of treatment.

Chapter 4: Conclusions 220

4.4 Recommendations for further research

Additional research focusing on materials characterization and analysis is
recommended.

■ Identification of the clay component within the plasters has yet to be carried out.

■ Further analysis for organic materials using more advanced techniques such as FTIR,
GC-MS, and HPLC is recommended. A wide sampling of materials from various
locations should be analyzed with these and other methods for the identification of
organic binding media. A comprehensive understanding of the paint and plaster
technology is critical to interpretation and conservation of the site.

■ Prior to the detachment of paintings on site, continued efforts should be made to
identify an appropriate new support material. Preservation of the detached painting
and the integrity of its original surface are dependent upon the selection of a stable
support material. Additional testing should consider modified lime and clay supports
as well as lightweight support materials such as expanded PVC or epoxy type foam
resins. Because a poor bond to the new support caused some of the deterioration
exhibited by the detached paintings, it may be valuable to reassess the performance of
pretreatment and detachment techniques using a more suitable system of support.

■ Although positive results were obtained using ethyl silicates to consolidate the clay
plasters, they were unremarkable and required a stable environment. Further testing
may be necessary to identify a more suitable consolidant system proven effective in
areas of low relative humidity.

Chapter 4: Conclusions 221

Tests showed excellent results for systems developed to pretreat and detach paint
layers on earthen plasters using the stacco method. At the present time, facsimile testing
indicates that it is not yet possible to perform a controlled detachment of discrete layers or
sequences of painted plaster to a consistent depth. However, since the research was
carried out exclusively on laboratory facsimiles, the evaluation of all treatments,
particularly detachment techniques should be reassessed in situ. The ability to detach
expansive clay-based plasters with weakened pictorial layers without significant losses
within the paint layer is encouraging. Additional research and in situ testing may help to
answer the questions that remain.

Appendix A: (Tatalhoyiik Treatment History

Wall paintings excavated by James Mellaart in the 1960s were conserved,
detached and stored at the Ankara Museum. The following is a summary of treatments as
described by Pamela French in reports dating from 1968-1974. 153
Plaster and pigment characterization

Microscopical examination of one plaster sample was reported. Clay and paint
sample analyses were carried out on a binocular microscope at forty times magnification.
The sample examined was found to have approximately ten layers of plaster with grass
lacunae in the ground layers. The thickness of the preparatory layer was inconsistent,
with variations of up to three millimeters in some sections. Traces of red-brown, red, and
black paint were visible in all the layers.

Chemical spot tests were conducted to help characterize the plaster and pigments,
and to detect the presence of salts. Tests for calcium carbonate content indicated that the
plaster was made from a highly calcareous clay. Iron was detected in analyses of both the

x ' Pamela French, "Clay and Paint Samples: Preliminary Testing," "Report on Work Carried Out on the
Murals from Catal Hiiyiik in the Archaeological Museum, Ankara, in September 1968," "Report to the
British Academy on the Work Done on the Catal Hiiyiik Paintings," "The Catal Hiiyiik Wall-Paintings,"
"The Continuation of the Conservation of the Catal Hiiyiik Wall Paintings Undertaken in the Museum of
Anatolian Cultures, Ankara, Summer, 1973," and "Report to the British Academy on the Conservation of
the Catal Hiiyiik Wall Paintings, Summer 1973, Spring 1974." (n.p., n.d.)

222

Appendix A: Treatment History _ 223

red-brown pigment and a section of unpainted clay plaster. s Tests for salt content
indicated high percentages of nitrates, phosphates, and chlorides.

An area of red-brown painted plaster was tested for the presence of blood.
Positive staining occurred, but results were inconsistent throughout the layer. Although
blood may have been used as a pigment, binder, or wall coating, further testing was
recommended.
Detachment techniques
Three types of detachment were used:

■ Detachment of the paint layer, underlying plaster, and mud brick wall support: the
block method,

■ Detachment of the paint layer and the mud plaster support, and

■ Detachment of the paint layer alone: the peeling method

In each technique, a fine muslin or linen facing was applied to the painted surface with
size or bone glue.
Field Treatment

Following excavation, paintings and reliefs were cleaned in situ and consolidated
with polyvinyl acetate emulsion. A two-layer facing of Japanese tissue and fine linen was
applied to each section with size and permitted to dry for three days. Paint layers were
peeled away with some mud plaster still adhering. The back of each painting was
impregnated with polyvinyl acetate emulsion. Because it did not penetrate to the paint

Both tests were conducted using potassium ferrocyanide.

Appendix A: Treatment History 224

layer, an intermediate layer of unconsolidated plaster remained between the two treated

surfaces.

Evaluation of detached paintings and reliefs

Unmounted paintings detached by the block method in 1962 and 1963 were
evaluated in 1968. They were in poor condition and deteriorating rapidly.
The following conditions were identified:

■ Most of the painted surfaces did not have facings.

■ The synthetic resin used to consolidate the paint, probably polyvinyl acetate,
had contracted, causing severe flaking.

■ Losses within the paint layer were caused by disintegration of the supporting
plaster.

■ Paintings with facings showed slightly less deterioration.
Careful packing preserved paintings detached during the 1965 season.

Overcrowding and repeated fluctuations in temperature and humidity in the storeroom
caused deterioration.

Most of the mounted paintings treated between 1962 and 1965 exhibited the
following conditions:

■ Loss of adhesion between the paint layer and the support plaster

■ Loss of adhesion between the consolidated paint layer and the unconsolidated
support

■ Disintegration of the mud plaster support behind the consolidated paint layer

Appendix A: Treatment History 223

■ Discoloration of the paint surface due to deterioration of the polyvinyl acetate

consolidant

In all cases in which an intermediate layer of unconsolidated plaster remained
between the inner and outer surfaces of a detached painting, differences between the rates
of contraction and expansion of the various layers caused it to crumble. The unsupported
paint layer became severely deteriorated. Attempts to fully consolidate mud plaster
remaining behind the paint layer by inducing the PVA to penetrate completely were
unsuccessful. Partial solubility of the emulsion after the application of solvents indicated
that partial cross-linking of the polymer might have occurred. Because the gel-like
emulsion could not be induced to penetrate, it had to be mechanically removed from the
backs of the paintings. 155 It was decided that the best way to preserve the detached
paintings was to remove most, if not all plaster adhering to the paint layer in order to
prevent the creation of an unconsolidated middle zone between the paint layer and the
rear surface of the plaster.
Retreatment

Retreatment of detached paintings and reliefs was carried out between 1968 and
1973. Sections with unconsolidated plaster still remaining were reduced to a thickness of
approximately one to three millimeters. Multiple applications of polymethyl
methacrylate, at concentrations ranging from 0.5%- 15% were used to consolidate the
backs of the paintings and reliefs.

Appendix A: Treatment History 226

Backing

Following consolidation, it was necessary to identify a strong backing material
that would remain flexible and reversible. A material with the same density as the mud
plaster was desired. A three-part backing system was developed:

■ First layer: 1.5 parts polymethyl methacrylate, 40% in toluene; 1 part marble
powder; 0.5 parts flaked silica;

■ Second layer: 1 part polymethyl methacrylate, 40% in toluene; 1 part marble
powder; 0.5 parts flaked silica; 0.5 parts glass fibers for strength;

■ Third layer: identical to the first, used as an adhesive for strips of pre-washed
cotton muslin applied in a criss-cross pattern.

Facing Removal

Facings were removed easily using very hot water applied with sheets of foam
rubber.
Surface Cleaning

Attempts to remove the polyvinyl acetate consolidant with solvents resulted in a
partially solubilized gel. The greatest degree of solubility was attained with a 50/50
solution of amylacetate and acetone applied with swabs. Removal of the gel caused the
loss of pigment particles. Compresses of blotting paper or cotton and muslin saturated
with solvent were applied repeatedly to remove the gel without sacrificing too much of
the pigment.

x Additional testing showed that even penetration could not be guaranteed beyond a depth of 1.5-2.0

Appendix A: Treatment History 227

Consolidation

Painted surfaces were re-consolidated with multiple applications of polymethyl
methacrylate, in solutions ranging from 3% to 5% in toluene. Heat was used to induce
solvent evaporation, to prevent the back from being affected.
Primary support of relief plasters

Requirements for a primary mounting material were reversibility, simplicity, and
for reliefs, a form of incorporated support. Limited success with expanded polystyrene
led to the temporary use of a plaster cast support for raised areas. Plasticine was used to
form an edging around the painted relief. Vaseline was used as an isolating layer to coat
the back. A plaster made with a solution of 25% polyvinyl acetate emulsion and glass
fibers was poured into the edged relief. Once dry, it was trimmed to the appropriate
thickness. The Vaseline isolating layer allowed for easy separation from the painted
relief. Plaster casts were replaced by expanded polyurethane foam as the more
appropriate support for relief areas prior to application of a rigid support.
Mounting of relief plasters

A material similar in appearance to the wall plaster was needed to fill the area
around mounted sections and link them together. Sheets of expanded polystyrene were
adhered to a wooden frame with polyvinyl acetate emulsion. A space the size of the
detached pieces was hollowed out of the polystyrene and the detached plasters fitted
within them. Holes were drilled in areas of unpainted plaster, and the reliefs with their

millimeters.

Appendix A: Treatment History 228

plaster supports were screwed to wooden mounts. Polystyrene was painted to blend in
with the mud. Losses were inpainted with Rowney Cryla Colors.
Mounting of peeled paintings

Peeled paintings were backed with the same primary material used for the relief
plasters. Plaster supports were not needed in the case of paintings. Expanded polystyrene
sheets, capable of supporting raised and buckled areas, were used with wooden trays to
mount the paintings. Paintings were screwed in as before. The polystyrene and any
pigment losses were inpainted with Rowney Cryla Colors.
Rigid support for exhibition

The most successful rigid support for exhibition was found to be a combination of
polyester resin, glass fibers, and expanded aluminum applied to the existing backing.
1972: Evaluation of sections treated in 1968 and 1969

Results of an evaluation of treatments carried out in 1968 and 1969 were positive.
The synthetic resin remained reversible and showed no apparent alteration. Support
materials were well adhered to paint layers. The aging of the consolidant did not alter the
optical properties of the paint layers.

Appendix B: Overview of Techniques

Introduction

Wall paintings, unlike easel paintings, are integral parts of the architecture to
which they have been applied. The significance of this association is apparent by the
inevitable loss of context that occurs when a painting has been detached. For this reason,
it is imperative that detachment occur only as a last resort when there are no other
treatment options. Archaeological excavation poses just such an example. Mural
detachment is one and sometimes the only means of prolonging the survival of paintings
which would otherwise be destroyed during excavation.

Detachment methods for the transfer of mural paintings have been documented
since antiquity. Three techniques were originally developed for the detachment of
frescoes-paintings executed on fresh moist plaster which become chemically bonded to
the lime ground:

■ Stacco a massello, the oldest, refers to the removal of the painting along with its
rendering and all or part of the wall to which it has been applied.

■ Stacco refers to a technique that removes the paint layer along with its immediately
underlying plaster layers.

■ Strappo, the most delicate of the operations, requires the lifting of the paint layer
alone.

In order to illustrate these techniques, the remainder of this overview will trace

their use through time in a brief historical summary and selection of case studies arranged

by material.

229

Appendix B: Overview of Detachment Techniques 230

Historical Background

Accounts of wall painting detachment go back to at least the first century B.C. As
noted in The Ten Books on Architecture by Vitruvius, "In Sparta, paintings have been
taken out of certain walls by cutting through the bricks, then have been placed in wooden
frames, and so brought to the Comitium to adorn the aedileship of Varro and Murena."
Caligula's attempts to remove wall paintings from Lanuvium were cited by Pliny the
Elder in the first century A.D., in the 35th volume of Historia Naturalis. "Similarly at
Lanuvium, where there are an Atalanta and a Helena close together, nude figures, painted
by the same artist, each of outstanding beauty (the former shown as a virgin), and not
damaged even by the collapse of the temple. The Emperor Caligula from lustful motives
attempted to remove them, but the consistency of the plaster would not allow this to be
done." 157 These accounts illustrate motives for removal as well as the lack of concern
for context that prevailed once the technology was mastered.

The use of detachment techniques is believed to have declined with the Roman
Empire until their rediscovery by the Italians in the fifteenth century. " Although
amateur removals were attempted frequently throughout the interim, they were generally
unsuccessful. By the fifteenth century, frescoes at risk of destruction by demolition,
particularly those with religious subject matter, were moved along with their supporting

156 Marcus Vitruvius Pollio, The Ten Books on Architecture, trans. Morris Hicky Morgan (New York:
Dover Publications, Inc. 1960) 53.

iy7 Pliny the Elder, Gaius Plinius Secondus, Natural History, Book XXXV, trans. H. Rackham
(Massachusetts: Harvard University Press, 1995) 273.

Appendix B: Overview of Detachment Techniques 231

walls. In 1480, the Resurrection by Piero della Francesca was removed along with the
layer of bricks on which it had been painted to the Palazzo Communale of Sansepolcro. 159
In Vita des Spinello Aretino, Giorgio Vasari described the 1501 removal of a Spinello
Aretino fresco. Entire wall sections held together with chains were cut away in order to
detach a painting of the Madonna about to be demolished at the old cathedral and oratory
of Santo Stefano in Arezzo. The mural, fastened with rope, was transported to a new
church in October 1561. 160 Vasari also cited the removal of Ghirlandaio's St. Jerome and
Botticelli's St. Augustine in Ognisanti in Florence. 161 The frescoes were bound with iron
and transported by friars without damage when the old choir of the church of Ognisanti
was destroyed in 1 564. The 1 566 demolition of the old choir of Santa Croce spurred the
removal of the well-known St. John the Baptist and St. Francis by Domenico Veneziano.
The entire supporting wall with the figures was preserved. 162 Examples such as these, of
the preservation of paintings by detachment, set the ethical precedent for modern day
treatments.

Up through the eighteenth century, mural painting transfer relied upon the
removal of entire masses of wall. 163 These large-scale detachments required the skills of

Volker Schaible, " Historisches und Ethisches zur Abnahme von Wanmalerei," Historische Technologie
und Konservierung von Wandmalerei: Vortragstext der dritten Fach-und, Fortbildungstagung (Bern:
Schule fur Gestaltung in Kommission im Verlag Paul Haupt, 1985), 143-150.

Milton Gendel, "Strappato, or the Art of Turning Frescoes into Easel Paintings," in Art News, Volume
67, Issue 6, 1968. 27.

160 Schaible, 143.

161 Schaible, 143.

" For further information, see The Metropolitan Museum of Art, The Great Age of Fresco: Giotto to
Pontormo, (New York: The Metropolitan Museum of Art, 1968).

' The practice of removing entire walls continued sporadically after the eighteenth century and is still used
today in situations requiring emergency stabilization.

Appendix B: Overview of Detachment Techniques 232

architects and engineers. In Florence, architect Nicolo Gasparo Paoletti moved an entire
vault painted by Poccetti. 164 In April 1773 he orchestrated the transfer of a vault by
Ottavio Vannini inside the Villa of Poggio Imperiale, which had been set for
renovation. 163 He was also given credit for the invention of a machine to aid in the
detachment of the paintings of Giovanni da San Giovanni which were moved to the
Academia delle Belle Arte. 166

By the mid-eighteenth century, the techniques of stacco and strappo were being
modified by Antonio Contri, a Ferrarese painter, after learning the technique of
reattaching paintings to hard stone supports. Inspired by the work of Neapolitan Isodoro
Frezza, Contri practiced until he mastered the removal of a very thin surface layer of
intonaco, the perfect strappo, using methods not entirely different from those used
today.' 67 He faced paintings with glue-soaked linen and allowed them to dry for several
days. After making incisions around the paintings, he detached them slowly by hand
from the wall. The backs of the paintings were coated with a more dilute solution of the

154 Gendel, 27. An account of this transfer, by Bottari, may also be found in the preface to the reprint of //
riposo by Borghini.

165 When the renovation was completed in 1813, the vault was transferred for a second time by another
architect, Giuseppe Cacialli. For further information, see The Metropolitan Museum of Art, The Great Age
of Fresco: Giotto to Pontormo (New York: The Metropolitan Museum of Art, 1968).

166 Volker Schaible, " Historisches und Ethisches zur Abnahme von Wanmalerei," Historische Technologie
und Konservierung von Wandmalerei: Vortragstext der dritten Fach-und, Fortbildungstagung (Bern:
Schule fur Gestaltung in Kommission im Verlag Paul Haupt, 1985), 143, citing Abbe de Saint-Non,
"Voyage Pittoresque ou description des royaumes de Naples et de Sicile," 1782. Bd. I, 2, S.8; zit. In J.
Guillerme, L'atelier du temps, Paris 1964, S. 144; hier in der Uebers des Verf.

167 For further information, see The Metropolitan Museum of Art, The Great Age of Fresco: Giotto to
Pontormo (New York: The Metropolitan Museum of Art, 1968).

Appendix B: Overview of Detachment Techniques 233

same glue, backed with linen, covered with hot sand, and weighted. When the backing
was dry, excess sand was removed and the facings were taken off using warm water.

Unfortunately, growing familiarity with removal techniques across Europe
intensified the plundering of antiquities. Well-preserved sites, such as Pompeii in
southern Italy, which was rediscovered in 1748, aroused significant interest among the
rich who had both the desire and the means to obtain the ancient paintings for their own
pleasure. Stone masons, builders, and sculptors worked for hire to carry out detachments.
For example, one of the best known of the sculptors of the late eighteenth century,
Canart, was responsible for detachments at the excavation at Herculaneum. Fascinated
with the process called "lavagna" a traveler, Abbe de Saint-Non described Canart' s
technique in his journal, "Voyage pittoresque ou description des royaumes de Naples et
de Sicile." Cuts were carefully made around the painting, which was supported by four
wooden planks joined by metal latches. The supporting wall was cut from behind the
painting and slate plates placed on either side of it. The whole assembly was eventually
clamped together for transfer using metal straps.

By the nineteenth century, there was a growing concern regarding misuse of the
technology. The nature of the technique led to damage. Traces of paint inevitably
remained on the supports of detached paintings. Shrewd art dealers often profited doubly
from the sale of two parts of the same painting, duping clients into believing each had an

168 Schaible, 144.

169 Schaible, 147.

Appendix B: Overview of Detachment Techniques 234

original fresco. 170 In 1825, Leopoldo Cicognara, writing for the Florence paper
Antologia, blamed overuse of the techniques for the ultimate loss of Italy's national
treasures. Knowledge of the new science had been capitalized upon even by military
figures who symbolically demonstrated political prowess by taking possession of works
of art. Among those who Cicognara criticized was Vivant Denon, General Secretary of
the French Museum, who had been ordered by Napoleon to remove many of Rome's
paintings to the collection of the Louvre. 171 Fortunately, the failure of Napoleon's
campaign in Russia prevented the loss of many of these paintings.

As with any new trend, exploitation and misuse often lead to prohibition. A
growing conservatism resulted. The retention of natural texture had been frowned upon
in the eighteenth and early nineteenth centuries. At that time, paintings were reattached
to linen supports and flattened, losing their character as wall paintings. Not until later in
the century did the loss of context and transformation of wall paintings into decorative
elements become problematic.

Furthermore, the evaluation of treatments over time permitted a new awareness of
technological problems. In order to preserve original surfaces, detached paintings had
been coated with paraffin or waterglass, a sodium silicate liquid with binding
properties. 172 Unfortunately, this coating often turned to a whitish opaque film. Many

170 Giovanni Urbani, "Restoration of Frescoes in Rome and Assisi" The Connoisseur, Vol. CXXXVI No.
549 (November 1955): 158.

171 Schaible, 146.

172 Ralph Mayer, The Artist's Handbook of Materials and Techniques (New York: The Viking Press, 1970),
361.

Appendix B: Overview of Detachment Techniques 235

paintings successfully detached by the strappo method were mined by posl-strappo
treatment:

■ Unsuitable supports slackened over time, causing losses of the paint layer,

■ Inappropriate adhesives applied to the detached paintings caused deterioration
and alteration of optical properties, and

■ In cases of stacco, remaining intonaco often contained moisture, allowing
deterioration to continue even after detachment.

Ethical and technical concerns became grounds for discussion. Toward the beginning of
the twentieth century, the removal of a painting from its context without the threat of
impending destruction became undesirable and highly questionable.

Nineteenth and twentieth century restorations were evaluated by Leonetto Tintori
in, "Methods Used in Italy for Detaching Murals." 174 The Crucifixion by Masaccio was
successfully detached using the stacco method circa 1860. However, paint losses
occurred in the long term due to a residue of glue from the preliminary relining. The
residue was removed and the paintings exhibited no further signs of deterioration. The
fresco, Pippo Spano, by Andrea del Castagno was removed from the Villa Pandolfini in
Florence in 1874. The detached painting was attached to canvas stretched on a wooden

173 By 1866. deterioration by damp and loss by destruction of building fabric are well noted as reasons for
the removal of mural paintings. Cavalcaselle proclaimed the necessity of detaching murals from walls
deteriorating due to the presence of moisture. In the same year, Secco Suardo, in his manual on restoration,
cites the importance of detaching frescoes by great artists in order to preserve them. For further
information, see The Metropolitan Museum of Art, The Great Age of Fresco: Giotto to Pontormo (New
York: The Metropolitan Museum of Art, 1968).

Appendix B: Overview of Detachment Techniques 236

frame using calcium caseinate, a material sensitive to the action of hot water. Use of this
adhesive to adhere the painting to a new support prohibited complete removal of the
strappo glue. Further deterioration occurred due to humidity, causing deformation of the
canvas.

Tintori also assessed the performance of his own treatments. He detached Prato,
a fresco by Niccolo di Pietro Gerini using glue and nitrocellulose as facing adhesives.
The painting was attached to a new support using a mixture of calcium carbonate, vinyl
emulsion, and calcium caseinate. As assessed by Tintori, the optical properties of the
painting were preserved.

Although no longer acceptable as a means by which to accumulate wealth, the
practice of mural detachment was still permissible under the rationale that it was needed
as an educational tool. By the beginning of the twentieth century, the original techniques
of mural detachment, stacco and strappo, were being used to study artist's techniques.
Paintings were removed in order to view the underlying sinopia (preparatory drawings).
Entire exhibits focused on the display of detached paintings along with the drawings on
rough plaster.

More recently, the use of detachment techniques has been reserved for crisis
situations. Methods for applying them have benefited from the need to develop
emergency preservation treatments. For example, improvements to the stacco method

174 Leonetto Tintori, *' Methods Used in Italy for Detaching Murals" Recent Advances in Conservation,
Contributions to the HC Rome Conference, 1961 (London: Butterworth's, 1961 ), 1 18-122.

175 Gendel, 26-29, 64-65.

Appendix B: Overview of Detachment Techniques 237

were spurred by the emergency circumstances of World War II, and again during the
Florence flood in 1966. The introduction of synthetic resins as reattachment adhesives in
the 1960s greatly improved remounting technology. Prior to this, the adhesive used most
often for reattachment was calcium caseinate, a hygroscopic material that caused
efflorescence and deformation of the paint layer in humid conditions.

Steadily throughout this century, technological innovations in visual media have
aided in the preservation of context, the key to a deeper understanding of mural paintings.
Photography has been recognized as a way to record the contextual information of wall
paintings before detachment. 177 Often, the amount of time that passes between the
detachment of a painting and its return to the original context requires detailed visual
documentation to preserve the relationship between the painting and its setting.

As the twentieth century draws to a close, ethical concerns remain. The historical
value of a wall painting depends upon the survival of its context. Original contexts
provide insight for the reconstruction of a period of civilization, and to understanding its
transformation over time.

Stacco a massello, stacco, and strappo must only be used if a painting risks
destruction if left in situ. Extensive examination of a site must be carried out to
determine the potential of less invasive interventions before detachment takes place.
Acceptable interventions may act to reduce the speed of deterioration by eliminating its

176 Eve Borsook, "Effects of Technical Developments on the History of Italian Wall Painting of the
Fourteenth and Fifteenth Centuries" in Conser\>ation ofWallpaintings-the International Scene, ed. Peter
Burman (London: Council for the Care of Churches, 1986), 62.

177 Borsook, 65.

Appendix B: Overview of Detachment Techniques 238

sources and/or by addressing exacerbating structural defects. If circumstances leading to
deterioration cannot be controlled, or if the structure supporting the painting is set for
demolition, detachment is necessary. For example, in an archaeological context, mural
detachment is one and sometimes the only means of prolonging the survival of paintings
which would otherwise be destroyed during excavation.

Generally, the stacco method is preferred for its ability to preserve the original
surface texture and optical properties of a painting. Unfortunately, when structural
deterioration originates in the wall beneath the painting, it often spreads to the plaster
directly underlying the paint layer. These circumstances require employment of the
strappo method. The methodology and materials used for the three types of detachment
have been implemented and modified to treat a wide variety of circumstances throughout
the world. Most of the literature addresses the detachment of paintings on lime plasters,
eilher fresco buono ox fresco secco.

The following summary offers a compilation of the work of both archaeologists
and conservators who have used these techniques on lime plasters, earthen plasters, and
even acrylic paintings on canvas. They reflect diverse conditions and solutions resulting
in individually modified treatments. Although detachment methods have been known for
centuries in Italy, in many cases there was little experience to draw upon on the local
level in both removal process and knowledge of the chemical makeup of the materials at
hand. Consequently, there was rarely a formal basis on which to predict the results of
treatment.

Appendix B: Overview of Detachment Techniques 239

Case Studies

This compilation provides an overview of detachment techniques, their
development over time, and the adaptations developed for specific circumstances. The
majority of the treatments were not monitored over time and therefore have not been
assessed in terms of appropriateness or success.
Earthen plasters
American southwest

The works of Wesley Bliss and Watson Smith in the 1930s are the most well-
known in the southwestern United States. The discovery of two sites in the American
southwest in the 1930s, the Kuaua Pueblo Ruins in New Mexico and the Awatovi and
Kawaika-a Pueblo ruins in Arizona, provided Americans the opportunity to develop
methods of removal for murals on earthen plasters over adobe walls. Both projects dealt
with the problem of detaching paintings from walls with multiple superimposed layers of
painted and unpainted plaster. Because these are earthen plasters, the inherent strength of
lime frescoes, which facilitates the detachment process, could not be relied upon. Each
project required modifications to methods developed in Italy.
Kuaua

Bliss described the preservation treatment of the Kuaua mural paintings in his
1935 thesis for the University of New Mexico and in a 1948 article for American
Antiquity. A workman in a subterranean kiva discovered the mural paintings in 1935.
The walls of the kiva, dried adobe balls set in adobe mortar, were approximately sixteen

Appendix B: Overview of Detachment Techniques 240

inches thick. As many as 85 layers of thin adobe washes, each approximately one-thirtieth
of an inch thick, covered the walls. Many of these had been painted.

Treatment began with the application of thin white shellac to protect the edges of
the painted layers. Superficial layers were scraped away using palette knives. An
accumulation of sand between the plasters and the adobe wall prompted the decision to
remove the paintings before they collapsed. Plain plastered surfaces were coated with
white shellac, painted areas with a dilute solution of celluloid in acetone. Three layers of
tissue were applied as a facing using a wet brush. A layer of molding plaster covered the
wet tissue, followed by the application of three layers of plaster-soaked burlap strips.
Laths were applied to this surface and bound with more plaster-soaked burlap strips.
Wooden frames, attached with more burlap and plaster, were created to fit each wall. The
adobe wall was cut away from the plasters one section at a time to allow for the gradual
shellacking and jacketing of the rear of the plasters. For this step the tissue layer was
omitted because the back jacket was to act as a support for the painting in the laboratory.
Once the base was jacketed, bolts were placed through the walls to hold the front and
back jackets together. The walls were sawn apart and jacketed at the corners. The
packaged walls were then transported to the Anthropology Department at the University
of New Mexico.

The mural paintings exhibited no damage as a result of the jacketing or transport.
Unpainted layers of plaster were scraped away to expose paintings which were then
photographed and hand-copied. The surface of each painting was faced with a layer of

Appendix B: Overview of Detachment Techniques 24J_

washed unbleached muslin attached with a solution of one quart of Eastman stripping
collodian and six ounces of clear Ambroid or Duco cement. After approximately twenty
minutes, the muslin-coated paintings were stripped from the wall. The reverse of each
painting was covered with a mixture of adobe and casein size which, when dry, was
coated with clear Ambroid. The paintings were glued to a wallboard coated with the

178

adobe/casein mixture. The muslin was rolled back upon itself to expose the paintings.
After each layer was detached, its underlying undecorated layers were scraped
down to the next painting. All paintings were displayed in their corresponding
positions. I7y Follow-up condition reports revealed that the paintings were in poor
condition. The instability of the wallboard support as well as the poor aging properties of
the nitrocellulose medium used to remount the paintings exacerbated deterioration of the
paintings. 180

178 No reference was made to the softening of the facing adhesive prior to its removal or to the treatment of
surface residue.

179 Wesley Bliss, "Preservation of the Kuaua Mural Paintings" American Antiquity Vol. XIII, no. 3 (1948):
218-223.

180 Constance S. Silver, "Architectural Finishes of the Prehistoric Southwest: A Study of the Cultural
Resource and Prospects for its Conservation" (master's thesis, Columbia University, 1987). 171.

Appendix B: Overview of Detachment Techniques 242

Awatovi

The Awatovi expedition ran for five field seasons from 1935-1939, under the
direction of Watson Smith in the Jeddito area of Arizona. Mural paintings were
discovered in kivas constructed of stone blocks or sometimes adobe brick, held together
with adobe mortar. A preparatory coat of coarse gray adobe plaster covered the surface of
some of the walls. Successive layers of a fine textured reddish-brown plaster composed
of calcareous material and iron-stained quartz sand mixed with water coated the walls.
Layers ranged from one to six millimeters in thickness. As at Kuaua, super-imposed
painted decorations were found. All pigments but black were inorganic.

Smith and the excavation crew were familiar with the detachment methods
employed at Kuaua but full-scale removal of the walls from Awatovi could not be carried
out due to their sheer number. The walls were treated in situ. The removal of
undecorated plasters was carried out using penknives and scalpels. As at Kuaua,
paintings were photographed and hand-copied. Methods applied at Kuaua provided the

• 181

foundation for the removal of paintings at Awatovi.

Cracks and holes were filled with a patching compound of plasticized calcareous
sand. After extensive testing with a number of different facing adhesives, a mixture of
consisting of: lOOcc Alvar 7-70, lOOcc Acetone, 60cc ethylene dichloride, and 20cc
dibutylphthalate was chosen. One layer of this mixture was applied to the surface of the

181 Watson Smith, "Kiva Mural Decorations at Awatovi and Kawaika-a" in Papers of the Peabody Museum
of American Archaeology and Ethnology, Harvard University (Massachusetts: Peabody Museum, 1952),
13-59.

Appendix B: Overview of Detachment Techniques 243

paintings and permitted to become tacky or in some cases, dry. After the application of a
second layer of adhesive, strips of boiled unbleached linen were pressed against it. When
the facings were completely dried, they were removed by pulling down from the top at a
45° angle from the wall surface. A knife was used to help separate the muslin from the
plaster. Unpainted plaster was scraped away until the next painting was reached. The
detached paintings were applied to glue-sized Untempered Masonite Presdwood,
mounted on a pine frame to reduce shrinkage. The surface of each panel was sanded and
coated with a glue/sand size composed of one part Cologne glue soaked in two parts
water, lcc clove oil, lec carbolic acid, 1 cc thymol in a 50% alcohol solution, and 10 cc
santophen in a 10% water solution. This mixture was applied to the back of each
painting. A synthetic plaster composed of glue, water, and plaster that had been shipped
from the site, was prepared and applied to the back of the painting and permitted to set.
Facings were removed by spraying the surface of the upright panels with acetone until all
the glue had dissolved. The surface was then sprayed with a 4% tincture of formaldehyde
in alcohol to harden the surface. These paintings were found to be in stable condition
even after fifty years. 182
Soviet Middle Asia

Soviet Middle Asian paintings found in the 1940s and 1950s were preserved by
P.I. Kostrov of the Hermitage Museum, who developed techniques based on the results of
conservation treatments carried out on earlier monumental sites. The paintings were

182 Silver, 171.

Appendix B: Overview of Detachment Techniques 244

executed on loess plasters. Deterioration of the paint layer and ground caused the
plasters to break apart when attempts were made to separate them from the wall. They
were consolidated with a synthetic resin dissolved in a xylene base. The paintings were
cut into sections and faced with two layers of gauze adhered with PVA. Grooves were
worked behind the surfaces of the sections to be detached. Excess loess plaster was
disposed of through one of the sides. Wooden frames were propped tightly against the
painted surfaces and the paintings were detached using chisels, hammers, long knives,
and small picks. The back of each painting was coated with polybutylmethacrylate
followed by a wax-colophony mixture and two layers of gauze. Th same wax-colophony
mixture was used to attach the paintings to new supports constructed from galvanized
iron sheets attached to wooden subframes. PVA facings were removed with alcohol and
the paintings were exhibited at the State Hermitage Museum. 183 This research provided
the basis for later conservation methods developed for the conservation of clay plaster,
loess, loess-gypsum, and partially gypsum sculptures excavated in Middle Asia.
Lime plaster on mudbrick
Tumacacori National Monument, Arizona

In "Conservation of Painted Lime Plaster on Mud Brick Walls at Tumacacori
National Monument, U.S.A.," Anthony Crosby briefly describes the 1977 detachment of

183 P.I. Kostrov and E.G. Sheinina, " Restoration of Monumental Painting on Loess Plaster Using Synthetic
Resins" Studies in Consen'ation Vol. 6, Numbers 2 and 3 (August 1961): 90-102.

184 B.J. Stavisky and N.N. Zhukova, "Parallel Progress in Restoration and Archaeology: Discovery and
Restoration of Monumental Painting and Sculpture on a Loess Ground" in ICOM Committee for
Conservation 10 ,h Triennial Meeting Washington, DC, USA Preprints (United States: International Council
of Museums, 1993)395-398.

Appendix B: Overview of Detachment Techniques 245

a severely deteriorated paint film from the earthen plaster walls of a Spanish Colonial
Mission Complex. The friable painted layer was stabilized with a dilute solution of
polyvinyl acetate prior to detachment by the strappo method. Facings of cheesecloth and
muslin were used to detach the paint film. Although some losses occurred during
separation and removal due to the extreme friability of the paint, the treatment was

185

deemed a success, and the painting remounted on a fiberglass support.
China

In China, mural paintings with two types of support have been found in ancient
graves and grotto temples. Their treatment was described in Qi Yingtao's " Studies on
Conservation of the Grotto Temples and the Mural Paintings of Ancient Graves in
China." 186 Rock, brick or stone walls were sometimes coated with a layer of mud and
straw plaster, followed by a lime rendering ranging from three to eight centimeters thick.
Paintings were executed on the lime plaster. Another type of painting was executed
directly on the wall or on top of a fine coat of limewash. Both types exhibited interlayer
detachment, flaking, and cleavage of the painted layer. Transfer methods were developed
for those that could not be preserved in situ.

185 Anthony Crosby, "Conservation of Painted Lime Plaster on Mud Brick Walls at Tumacacori National
Monument, U.S.A.," in Third International Symposium on Mudbrick (Adobe) Preservation (METU: Offset
Printing Workshop, 1980), 59-73.

186 Qi Yingtao, "Studies on Conservation of the Grotto Temples and the Mural Paintings of Ancient Graves
in China" in International Symposium on the Conservation and Restoration of Cultural Property-
Conservation and Restoration of Mural Paintings (1), November 17-21, 1983, Tokyo, Japan (Japan: Tokyo
National Research Institute of Cultural Properties, 1984). 19-29.

Appendix B: Overview of Detachment Techniques 246

Transfer techniques were divided into four categories.

■ Tearing down: removal of the wall along with the painting, used specifically
when individual paintings appear on each brick.

■ Sawing to remove: a saw is used to cut down between the wall and the
rendering

■ Prizing to remove: a lever is inserted to gradually increase the size of a pre-
existing void between the rendering and the wall until the painting is
completely detached

■ Shocking to remove: a chisel is used to slowly jab away the mud layer, carried
out in cases when the rendering and wall are well adhered to one another.

Sawing, supplemented by the other techniques, was the primary method used to
detach mural paintings. Yingtao only briefly mentions "adhering with pasted cloth to
remove," a technique whose name implies a similarity to the stacco or strappo methods.

The mural paintings, often quite large, were first divided into sections. Double
layers of facing, one paper, the other fabric, were attached in early treatments with peach
glue, and subsequently with polyvinyl butyral. The backs of detached paintings were
cleared of fragments and coated with a layer of polyvinyl acetate emulsion. Renderings
were most often reconstructed using a mixture of polyvinyl acetate emulsion, lime, and
hemp fiber. To improve shock resistance, cloth was adhered to the rendering using epoxy
resins. The condition of the paintings after detachment was not discussed.

Appendix B: Overview of Detachment Techniques 247

India

As with the Chinese examples described above, the earliest wall paintings
discovered in India are painted directly on the rock surfaces of caves. Later examples
were executed on earthen plaster bases with white lime, gypsum, or kaolin grounds.
From about the sixth century A.D. onward, lime and sand plasters were used. Both
tempera and true fresco techniques have been discovered.

Transfer methods were developed in India to preserve paintings exhibiting severe
deterioration. 187 Two such methods were used at the Rangmahal Palace at Chamba:

■ Detachment of paintings with renderings in tact: for paintings executed on a
thick lime plaster

■ Detachment of only the paint layer: for paintings executed on clay plaster
Paintings were faced with muslin using PVA emulsion. A second facing of

heavier cloth was adhered with animal glue. Saws, chisels and pointed bars were used to
cut the panels free. Paintings were reattached to an aluminum frame/plaster of Paris
support. Hot water was used to remove the outer facing, ethyl alcohol, the inner.

Paintings were reassembled in their original order and displayed at the National
Museum. Paintings executed on clay plasters were too fragile to withstand the cutting
procedure described above. The paint layer was removed and fixed on a support
consisting of cloth and hard board attached with a casein-lime adhesive.

187 O.P. Agrawal and Kamal K. Jain, "Problems of Conservation of Wall Paintings in India" in
International Symposium on the Conservation and Restoration of Cultural Property-Conservation and
Restoration of Mural Paintings (1), November 17-21, 1983, Tokyo, Japan (Japan: Tokyo National Research
Institute of Cultural Properties, 1984), 31-40.

Appendix B: Overview of Detachment Techniques 248

Another palace complex located in Kulu required the detachment of one of its
murals. Water infiltration had severely deteriorated an otherwise hard lime plaster wall.
Double layer facings of muslin and a heavier cloth were applied with PVA emulsion.
Panels were cut out in 2-by-2-foot sections and reassembled for display at the National
Museum at New Delhi. Facings were removed with ethyl alcohol. The treatments were
deemed successful.
Bulgaria

Archaeological excavations in Bulgaria revealed painted tombs belonging to the
Thracians, an ancient people who settled around the Balkan peninsula until around the
seventh century A.D. One of these, the Maglish Tomb, dates from the beginning of the
second century B.C. A thick mud-lime-sand preparatory plaster coated the uneven
surfaces of its brick walls. A ground layer consisting of sand, marble dust, and lime, and
a finish layer consisting of marble dust and lime followed this. Detachment and
deterioration of the paintings were caused primarily by a lack of adhesion and cohesion of
the preparatory layer. The wall paintings existed mainly as fragments, some still adhering
to the wall, most lying on the ground. In order to facilitate reconstruction, the few
fragments remaining adhered to the walls were detached. The procedure was carried out
without a facing due to the friability of the paint surface. Poor adhesion between the
plaster and the brick wall simplified the process. A net was used to catch falling
fragments during removal. Once the surfaces had been cleared of dirt and salts, the paint
consolidated with multiple coats of Acryloid B-72, 0.5-3%, and the backs cleaned, the

Appendix B: Overview of Detachment Techniques 249

fragments were reassembled for mounting. A metal structure embedded in new plaster
composed of lime, sand, marble dust, asbestos fibers, and 10% PVA, was attached to the
backs of the fragments.
Lime plaster/frescoes
Montenegro

An earthquake in Montenegro in 1979 caused extensive damage to the Complex
of Monastery Podlastva. Mural paintings, in many cases seventeenth century frescoes
covering fifteenth century frescoes, suffered extreme damage, some requiring
detachment. Surfaces were fixed with Acryloid B-72 prior to detachment by the stacco
method. Two linen facings were attached with colletta. The paintings were successfully
separated using vibration. Each painting was reduced to a thickness of 2-to-5-
millimeters and attached to a new support made from a lime-casein adhesive with a 10%
solution of polyvinyl acetate and 3 layers of linen. This was applied to a second support
consisting of layers of Araldite, glass wool, and polyurethane sheets. Warm water was
used to remove the facings. Colletta residue was removed using a warm water pulp
poultice covered by a hot water bottle. More than twenty-five square meters of wall
paintings were detached and remounted beginning in 1983. The procedure was

1 89

considered a success.

188 A detailed report of this project can be found in: Zdravko Barov, "Thracian Painted Tombs: Technical
Notes and Conservation Procedures" in International Symposium on the Conservation and Restoration of
Cultural Property-Conservation and Restoration of Mural Paintings (1), November 17-21, 1983, Tokyo,
Japan (Japan: Tokyo National Research Institute of Cultural Properties, 1984), 197-208.

189 Zdravko Gagovic and Aleksandar Cilikov, "The Report on the Works on the Complex of Monastery
Podlastva from July 5-October 2, 1983" (ICCROM Library, photocopy), 1-3.

Appendix B: Overview of Detachment Techniques 250

Catalonia

Sometimes the removal of a wall painting is dictated by the state of preservation
of its surroundings. In 1982, several Romanesque frescoes were found during restoration
work at the Church of Saint Tomas de Fluvia in Catalonia, beneath several coats of paint
and whitewash and a false vault. 190 More than 60% of the painted surfaces were
separated from the wall. Multiple layers of different periods were super-imposed on the
original. Previous water infiltration caused erosion of the mural paintings and the
crystallization of salts. Although the sources of these problems had been eliminated, the
presence of moisture due to condensation was found to be an inevitable circumstance
within the upper portions of the building that housed the paintings. The paintings, faced
with co/fefta-impregnated cloth strips, were detached from the walls using the strappo
technique, although a stacco-like condition sometimes resulted. Each painting was
reattached to canvas, and the whole, with the addition of an isolating layer, was attached
to polyester and fiberglass soffits. In December 1986, the paintings were returned to their
original locations and reattached using stainless steel tubes. A twelve-centimeter distance
was retained between the mural and the wall to prevent any further deterioration by damp.
The author, although opposed to the detachment of mural paintings, noted that as a last
resort, the treatments were necessary and well worth the effort.

190 Treatment of these paintings was described by Josep Maria Xarrie in "Pintures Murals de Sant Tomas de
Fluvia" in De Mitseus: Quadems de Museologia i Museografia Issue 1 (1988): 1 16-1 19 and English
summary.

Appendix B: Overview of Detachment Techniques 25 1

England

The transfer of the medieval crypt vault paintings at Rochester Cathedral in 1984-
1985, is a good example of both the detachment of an extensive area of painted plaster
and its reattachment in situ. The paintings were partially separated from their grounds,
and actually hung from the vault in some places. Excessive moisture and the previous
application of an organic fixative had severely darkened them. The crypt dates from two
periods ranging from 1080 to 1227. Its use as a coal storage, as well as its proximity to a
gypsum factory resulted in the deposition of soot and gypsum particles on the surface of
the paintings. These were trapped by a hard crystalline layer of salt, which had formed on
the surface due to the movement of moisture through the vault. The friability of the
original rendering would not allow for the consolidation of the painting or its limewash
ground. Transfer of the paintings to a new support was determined to be the only viable
solution.

The vault and soffits were faced in their entirety with a triple layer of silk
crepeline and document tissue strips adhered with viscous water-soluble glue. The
location of cuts was marked in advance to coincide with appropriate elements within the
painting. Ninety sections of the painting were removed using a scalpel to incise the
surface. The badly deteriorated plaster rendering of the vault was removed down to the
rubble structure in order to build up a new sound plaster support for the painting.
Fragments of the lime coat and mortar were removed from the back of each painting
section. An adhesive mortar consisting of a 1:1 mixture of very fine sieved sand and

Appendix B: Overview of Detachment Techniques 252

equally fine sieved lime was used to reattach the paintings to the new coarse support. A
coat of limewash was applied to the backs of the painting to encourage adhesion to the
adhesive mortar. A three-millimeter thick layer of the mortar was then applied and the
paintings were pressed into place using modified printers' rollers. After one day, the
facings were removed with steam. The successful detachment and subsequent
reattachment in situ allowed the conservators to progress to the next stage of treatment,
the removal of the overlying crystalline salt layer.
Rome, Italy

The Crypt of the Basilica of San Lorenzo Fuori le Mura was discovered in Rome
circa 1970. During the following years, the eighth century frescoes suffered severe
deterioration due to the drastic environmental changes brought on by exposure. The
mural paintings were discovered on water-saturated walls composed of brick, stones,
pozzolana, and lime. The plasters were detaching from their supporting walls and
exhibited a complete lack of cohesion. The delaminating paint layer lacked cohesion and
adhesion and was encapsulated by both soluble and insoluble salts. After careful
consideration, a decision was made to detach the paintings. Prior to detachment, the paint
layer was fixed with an acrylic emulsion, Primal AC 33. Heat was applied in an effort to
dry the surface. Two cloth facings were attached using Acryloid B-72 in a 35% solution
in lacquer thinner. Paintings were detached by the stacco method. The back of each

191 David Park and David Perry, "Rochester Cathedral: Conservation of the Crypt Vault Paintings," in Case
Studies in the Consen'ation of Stone and Wall Paintings, Preprints of the Bologna Congress, 21-26
September 1986 (London: The International Institute for Conservation of Historic and Artistic Works,
1987), 182-185.

Appendix B: Overview of Detachment Techniques 253

painting was reinforced with two layers of cheesecloth and one of hemp canvas attached
with lime caseinate. A support consisting of fiberglass, expanded polyurethane, and an
isolating layer of cork was then attached with a mixture of PVA emulsion and calcium
carbonate. Facings were removed with cellulose pulp poultices and lacquer thinner. Salts
were removed, losses were filled and patinated. and a 2.5% solution of Acryloid B-72 in
lacquer thinner was sprayed on in one application. 192 Treatment results were satisfactory.
Florence, Italy

One of the largest detachment programs was initiated when torrential rains caused
the flooding of the Arno River in Florence on November 4, 1966. Many of the city's
most important buildings and collections were affected. The Basilica of Santa Croce, the
Duomo, the Uffizi Gallery, the National Library, the National Museum of History and
Science, the homes of Dante and the Medicis, and the Palazzo Vecchio all suffered
extensive damages, with flood waters reaching as high as ten to twenty feet. The
plaster supports of some paintings suffered damages due to direct contact with the
floodwaters. Others, which had been situated high above the path of the water, began to
deteriorate due to the effects of capillary action. Moisture spread throughout the walls,
encouraging the growth of molds and fungi. Salts, often concentrated in the grounds

'" : A full report of this treatment can be found in: Paul M Schwartzbaum, Ippolito Massari, Giovanna
Pignatelli and Carlo Giantomassi, " Approaches to the Conservation of Mural Paintings in Underground
Structures, Case Studies of Recent Projects by ICCROM Consultants and Staff in International
Symposium on the Conser\<ation and Restoration of Cultural Property-Conservation and Restoration of
Mural Paintings (1), November 17-21, 1983. Tokyo, Japan (Japan: Tokyo National Research Institute of
Cultural Properties, 1984). 41-58.

193 Anon., "Preserving a Priceless Art Treasure" Rohm and Haas Reporter (Philadelphia: Rohm & Haas
Co., Summer 1974), 3-4.

Appendix B: Overview of Detachment Techniques 254

below churches with burial grounds, were carried by the moisture up through the walls,
where they began to crystallize, causing losses within the painted surfaces. In many
cases, detachment of the frescoes was the only option. More than 2,300 square meters of
mural paintings were dried with portable heaters and detached from the walls during an
eighteen month period of continuous work by the Restoration Laboratories of the
Fortezza di Basso in Florence. 194 Their technique, called trasporto, is similar to the
traditional strappo method. The frescoes were coated with animal glue and a layer of
gauze-like canvas. When the glue had dried, the canvas was "pulled off," taking only a
thin layer of the painted wall with it. According to the conservators, none of the pigments
were lost. Upon detachment, paintings were rolled and taken to the Restoration
Laboratories for treatment. New supports consisted most often of a combination of

195

calcium carbonate, Rhoplex acrylic emulsion as a binder, and a glass fiber canvas.
References to this treatment did not offer an assessment of the end result.
India

Two directly superimposed historically significant paintings were discovered at
the Brihadeeswara Temple at Thanjavur, India. The overpainting belonged to the
Nayakka Period dating to the sixteenth century, the underlying layer to the Chola Period
dating to the eleventh century. Rather than sacrifice the later painting, as had been done
in earlier treatments, a decision was made to modify the traditional stacco process. Both

194 Milton Gendel, "Strappato. or the Art of Turning Frescoes into Easel Paintings" An News, Vol. 67, no. 6
(1968): 26.

195 Anon., "Preserving a Priceless Art Treasure" Rohm and Haas Reporter (Philadelphia: Rohm & Haas
Co., Summer 1974), 9-10.

Appendix B: Overview of Detachment Techniques 255

paintings had been executed on lime plaster, each with a two-part ground/finish sequence.
The later painting was faced with layers of gauze and canvas using a 20% solution of
polyvinyl acetate. Detachment, carried out using rubber tipped chisels and a wooden
mallet, was facilitated by the exceptionally smooth surface of the earlier painting. The
detached painting was attached to canvas and a sheet of polyurethane foam using
polyvinyl acetate emulsion. The whole was then attached to a sheet of fiberglass mat
using an epoxy resin adhesive. Facings and residue were removed using a 9: 1 solution of
toluene and acetone. Following the consolidation of the earlier Chola layer, both were

196

displayed. The project, as assessed by the conservator, was a complete success.

Acrylic on canvas

Canada

Two large acrylic murals painted in 1956 by Oscar Cahen on acrylic-grounded
cotton canvas, were removed from the Imperial Oil Building in Toronto, Canada in 1979.
Their supporting structures were set for demolition and renovation. 197 The 9'9" x 22'6"
and 8'6" x 29'8" paintings were adhered with contact adhesive to terra cotta tile brick
walls coated with J4"-l" thick cement (expanded metal lathe embedded in some areas)
and a skim coat of white plaster. Facings of wet strength tissue were applied with
colletta. A modified strappo method was used to roll the paintings on to large diameter

196 The full report on this operation may be found in S. Subbaraman, "Separation of Two Layers of Mural
Painting by Modified Stacco Process" Role of Chemistry in Archaeology (India: The Birla Institute of
Scientific Research, 1991 ), 76-80.

197 Janice Antonacci and Virginia Caswell, " The Emergency Removal of Two Large Murals" in Papers
Presented at the Art Conservation Training Programs Conference, April 28 &29, 1980 (Newark, Delaware:
The Program, 1980) 127-145.

Appendix B: Overview of Detachment Techniques 256

reinforced cardboard tubes covered with cotton fabric. The mural canvases were
carefully detached and the front edges stapled to the cardboard tubes. As each tube was
turned, the bond of the contact adhesive was severed manually with scalpels. Polycotton
fabric was interwoven into the roll to prevent the contact adhesive residue on the reverse
of the painting from sticking to the facings. The rolled paintings were protected with
fabric and polyethylene to prepare for transport and storage. Although additional testing
was prohibited by time constraints, the authors noted that detachment methods would not
have been significantly altered. The paintings were left in storage with their facings in
tact. No mention is made of facing removal or follow-up evaluation.
Conclusion

Comprehensive research programs have been carried out to evaluate the materials
most often used to facilitate the detachment of mural paintings. Both traditional and
synthetic materials have been studied. Giorgio Torraca and Paolo Mora summarized a
comprehensive study of fixatives and consolidants in "Fissativi per Pitture M»ra//." 198
The uses and ideal properties of fixatives are discussed for both dry and humid
environments. Synthetic and traditional fixatives were evaluated for ease of use,
adhesion, resistance to aging, resistance to abrasion, resistance to biological attack,
solubility, and effects on optical properties. The report includes a description of the ideal
properties of a fixative to be used prior to detachment. A great deal of this information

198 Giorgio Torraca and Paolo Mora, "Fissativi per Pitture Murali" in Bollettino dell' Institute Centrale del
Restauro (Italy: ICCROM, 1965), 109-132.

Appendix B: Overview of Detachment Techniques 257

may also be found in The Conservation of Wall Paintings by Paolo and Laura Mora and
Paul Philippot.

"Some Further Testing of Materials Used During the Restoration of Mural
Paintings," compiled by Margaret Hey, includes the research of several notable
investigators. Contributions include the results of microbiological analysis of
experimental supports; evaluations of fixative solutions, adhesives, and support materials;
and the weathering test results of commonly used materials. This report is specifically
geared toward materials used in the detachment process.

"Architectural Finishes of the Prehistoric Southwest: A Study of the Cultural
Resource and Prospects for its Conservation" written by Constance S. Silver in 1987
includes an extensive study of both traditional and synthetic materials used for the
detachment of wall paintings. 201 Painted earthen plaster facsimiles were created in a
conservation laboratory. Methods and materials for detachment were modified and tested
for use on adobe plasters. Detailed assessments of each treatment are included.

' Margaret Hey, Some Further Testing of Materials Used During the Restoration of Mural Paintings
(Amsterdam: ICCROM. 1969), 1-29.

Constance S. Silver, "Architectural Finishes of the Prehistoric Southwest: A Study of the Cultural
Resource and Prospects for its Conservation" (master's thesis, Columbia University, 1987), 172-190.

BIBLIOGRAPHY

Abramowitz, Mortimer. Fluorescence Microscopy. The Essentials, Volume 4. United
States: Olympus America, Inc., 1993.

Agrawal, O.P. "Conservation of Ajanta Wall Paintings." In Case Studies in the

Conservation of Stone and Wall Paintings, Preprints of the Contributions to the
Bologna Congress, September 21-26, 1986, edited by N.S. Bromelle and Perry
Smith. London: International Institute for Conservation, 1986.

Agrawal, O.P. and Jain, K.K. "Problems of Conservation of the Wall Paintings in India." In
International Symposium on the Conservation and Restoration of Cultural
Property: Conservation and Restoration of Mural Paintings (I), November 17-21,
1983, Tokyo, Japan. Tokyo: Tokyo National Research Institute of Cultural
Properties, 1984.

Alvarez, Luciano Cedillo. "Stucco: A Report on the Methodology Developed in Mexico."
In In Situ Archaeological Conservation, Proceedings of Meetings April 6-13, 1986.
Mexico: The Getty Conservation Institute, 1987.

American Society for the Testing of Materials. 7990 Annual Book ofASTM Standards

Section 6: Paints, Related Coatings, and Aromatics. Vol. 6.01. Philadelphia: ASTM,
1990.

Anon. "Preserving a Priceless Art Treasure." Rohm and Haas Reporter 32, no. 2; (Summer,
1974): 1-10.

Antonacci, Janice; Caswell, Virginia. "The Emergency Removal of Two Large Murals." In
Papers Presented at the Art Conservation Training Programs Conference, Newark,
Delaware, April 28 & 29, 1980. Newark: University of Delaware, 1980.

Arrie, Josep Maria. "The Mural Paintings in St. Thomas de Fluvia." In De Museus:
Quaderns de Museologia i Museografia, no. 1, (1988): 1 16-1 19.

Balderrama, Alva and Giacomo Chiari. "Protection and Conservation of Excavated

Structures of Mudbrick." In Conservation on Archaeological Excavations with
Particular Reference to the Mediterranean Area. Rome: ICCROM, 1984.

258

Bibliography 259

Baltoyannis, Stavros. "Conservation and Restoration of the Wall Paintings in the Church of
the Protothronos, Naxos." In Studies in Conservation 21 (1976): 51-62.

Barber, Elizabeth Wayland. Women 's Work: The First 20,000 Years, Women, Cloth, and
Socier\' in Early Times. New York: W.W. Norton & Company, 1994.

Barov, Zdravko. "Thracian Painted Tombs: Technical Notes and Conservation Procedures."
In International Symposium on the Conservation and Restoration of Cultural
Property: Conservation and Restoration of Mural Paintings (I), November 17-21,
1983, Tokyo, Japan. Japan: Tokyo National Research Intitude of Cultural
Properties, 1984.

Beas, Maria Isabel G. "Traditional Architectural Renders on Earthen Surfaces." Master's
thesis, University of Pennsylvania, 1 99 1 .

Billmeyer, Jr., Fred W., Treva Pamer, Max Saltzman. " Pigment Analysis for

Conservation." In Science and Technology in the Sendee of Conservation,
Preprints of the Contributions to the Washington Congress, 3-9 September 1982.
Edited by N.S. Brommelle and Garry Thomson. London: The International Institute
for Conservation of Historic and Artistic Works, 1983.

Birstein, V.J. "On the Technology of the Central Asian Wall Paintings: The Problems of
Binding Media," Studies in Conservation 29, no.l (February 1975): 8-19.

Black, James, comp. Recent Advances in the Conservation and Analysis of Artifacts.
London: Summer's Schools Press/ University of London: 1987.

Bliss, Wesley L. "Preservation of the Kuaua Mural Paintings," American Antiquity 13 no. 3
(1948): 218-223.

Bogovcic, Ivan. "The Museum Presentation of the Peasant Wall Paintings Taken Off After
the Earthquake." In Varstvo Spomenikov 28, 1986: 159-163.

Borrelli Vlad, Licia. "The Restoration of the Isis Hall." In Bollettino Dell ' Instituto
Central e Del Restauro (1967): 6-45.

Borsook, Eve. "Effects of Technical Developments on the History of Italian Wall Painting
of the Fourteenth and Fifteenth Centuries." In Conservation of Wall Paintings - the
International Scene. London: The Council for the Care of Churches, 1986.

Bibliography 260

Bromelle, N.S. and Perry Smith, eds. Case Studies in the Conservation of Stone and Wall
Paintings: Preprints of the Contributions to the Bologna Congress, 21-26
September 1986. London: Internationa] Institute for Conservation of Historic and
Artistic Works. 1986.

Bromelle, N.S.; Elizabeth M. Pye, Perry Smith and Garry Thomson. Preprints of the
Contributions to the Paris Congress, September 2-8, 1984. London: The
International Institute for Conservation of Historic and Artistic Works, 1984.

Brown, Paul Wencil, Carl R. Robbins, and James R. Clifton. "Factors Affecting the
Durability of Adobe Structures." Studies in Conservation 24, no. 1 (February
1978): 23-39.

Carroll, Dorothy. Clay Minerals: A Guide to Their X-Ray Diffraction. Colorado: The
Geological Society of America, Inc., 1970.

Chamot, Emile Monnin and Clyde Walter Mason. Handbook of Chemical Microscopy.
New York: John Wiley & Sons, 1940.

Chiari, Giacomo. "Treatment of Adobe Friezes in Peru." In Third International

Symposiumon Mudbrick (adobe) Preservation, September 29-October 4, 1980.
Edited by Okan Ustunkok Emre Madran. Ankara: ICOM/ICOMOS, 1980.

Conservation materials Ltd. "The Keck Formula" in Conservation Materials Catalog
(Nevada: conservationMaterials Ltd. nd. 35.

Conservation Resources (U.K.) Ltd. Chemical Data Sheet: Vinamul6825 (6815) (England:
Conservation Resources (U.K.) Ltd 1992.

Crosby, Anthony. "Conservation of Painted Lime Plaster on Mud Brick Walls at

Tumacacori National Monument, U.S.A." In Third International Symposium on
Mudbrick (adobe) Preservation, September 29-October 4, 1980. Edited by Okan
Ustunkok Emre Madran. Ankara: ICOM/ICOMOS, 1980.

Derrick, Michele, Luiz Souza, Tanya Kielisch, Henry Florsheim, and Dusan Stuhk.
"Embedding Paint Cross-Section Samples in Polyester Resins: Problems and
Solutions." Journal of the America Institute for Conservation 33, no. 3 (FallAVinter
1994): 227-228.

Douline, A. "Batiments en vouter et compoler en adobe, Niger." Memoir e de CEAATerre.
France: CRATerre, 1990.

Bibliography 261

Driessen, P.M., and T. de Meester. Soils of the Cumra Area, Turkey. Holland: Centre for
Agricultural Publishing and Documentation, 1976.

Ellison, A. and J. Harris. "Settlement and Land Use in the Prehistory and Early History of
Southern England: A Study Based on Locational Models." In Models in
Archaeology. Edited by D.L. Clarke. London: Methuen, 1972.

Espinosa, Agustin. "Conservation and Restoration of the Murals of the Temple of the
Paintings in Bonampak, Chiapas." In In-Situ Archaeological Conservation,
Proceedings of Meetings April 6-13, 1986. Mexico: The Getty Conservation
Institute, 1987.

Fairservis, Jr., Walter. The Threshold of Civilization, An Experiment in Prehistory. New
York: Charles Scribner's Sons, 1975.

Fazio, Paula, et. al. eds. American Society for Testing Materials: Construction, Soil and
Rock, Dimension Stone, Geosynthetics Section 4, Vol. 4.08. Philadelphia: ASTM,
1990.

Federation of Societies for Coatings Technology. Pictorial Standards of Coatings Defects.

Feller, Robert L. Artist's Pigments: A Handbook of Their History and Characteristics, Vol.
I. Washington, D.C.: National Gallery, 1986.

Ferragni, D.; M. Forti, P. Mora, J..M. Teutonico, and G. Torraca. "Injection Grouting of
Mural Paintings and Mosaics." In Preprints of the Contributions to the Paris
Congress, September 2-8 1984. Edited by N.S. Bromelle, Elizabeth M. Pye, Perry
Smith, and Garry Thomson. London: The International Institute for Conservation of
Historic and Artistic Works, 1984.

Fifth International of Meeting of the Experts on the Conservation of Earthen Architecture.
Rome: CRATERRE/ICCROM, 1987.

Forbes, R.J. Studies in Ancient Technology: Volume 111. Netherlands: Brill, 1965.

Franco, Maria Luisa. "Conservation at the Templo Mayor of Tenochtitlan." In In-Situ

Archaeological Conservation, Proceedings of Meetings April 6-13, 1986. Mexico:
The Getty Conservation Institute, 1986.

Bibliography 262

French, Pamela. "Clay and Paint Samples: Preliminary Testing." N.p., n.d.

. "Report on Work Carried Out on the Murals from Catal Hiiyiik in the

Archaeological Museum, Ankara, in September, 1968."

. "Report to the British Academy on the Work Done on the Catal Hiiyuk Paintings."

N.p., n.d.

. "The Catal Hiiyiik Wall-Paintings." N.p., n.d.

"The Continuation of the Conservation of the Catal Hiiyiik Wall Paintings

Undertaken in the Museum of Anatolian Cultures, Ankara, Summer, 1973."

. "Report to the British Academy on the Conservtion of the Catal Hiiyuk Wall

Paintings, Summer 1973, Spring 1974.

. "The Problems of In Situ Conservation of Mud Brick and Mud Plaster." In In Situ

Archaeological Conservation, Proceedings of Meetings, April 6-13, Mexico.
Mexico: Instituto Nacional de Antropologfa e Historia, and California: J. Paul Getty
Trust, 1987.

Gagovic, Zdravko; Aleksandar Cilikov, The Report on the Works on the Complex of
Monastery Podlastva, July 5 - October 2, 1983. Paris: UNESCO, 1984.

Gairola, T.R. "Preservation of the Wall Paintings From the Rang Mahal, Chamba." In

Conservation of Cultural Property in India, Proceedings of the Seminar: February
23-25, 1966. New Delhi: National Museum: Conservation Laboratory, 1966.

Gamarra, Ricardo Morales. "Conservation of Structures and Decorative Adobe Elements in
Chan Chan." In Adobe: International Training and Symposium Workshop on the
Conservation of Adobe. Final Report and Papers. Lima-Cusco(Peru). 10-
22/September 1983. Rome: ICCROM, 1983.

Gedye, lone, Henry Hodges, Andrew Oddy, comp. "The Deterioration and Conservation of
Wall Paintings." Notes for a Short Course in Conservation. London: The British
Museum Research Laboratory, 1973.

Gendel, Milton. "Strappato, or the Art of Turning Frescoes into Easel Paintings." In Art
News 67, no. 6 (1968): 26-29.

Gettens, Rutherford and George L. Stout. Painting Materials: A Short Encyclopedia. New
York: Dover Publications, Inc., 1966.

Bibliography 263

Grim, Ralph E. Clay Mineralogy. New York: McGraw-Hill Book Company, 1968.

Hamilton, Naomi. "Figurines, Clay Balls, Small Finds, and Burials." In On the Surface:
Catalhoyiik 1993-95. Edited by Ian Hodder. Cambridge: McDonald Institute for
Archaeological Research; London: British Institute of Archaeology at Ankara, 1996.

Hansen, Eric F., Rosa Lowinger and Eileen Sadoff." Consolidation of Porous Paint in a
Vapor-Saturated Atmosphere, A Technique for Minimizing Changes in the
Appearance of Powdering, Matte Paint." Journal of the American Institute for
Conservation 32, no. 1 (Spring 1993): 1-14.

Harrison, J.R. "The Mud Wall in England at the Close of the Vernacular Era: A Study in
Building Distribution and the Nature of Materials." In Transactions of the Ancient
Monuments Society. Edited by Michael McGarvie. London: The Ancient
Monuments Society, 1984.

Hey, Margaret. Some Further Testing of Materials Used During the Restoration of Mural
Paintings. Amsterdam: ICOM, 1969.

Hodder, Ian. Symbols in Action: Ethnoarchaeological Studies of Material Culture.
London: Cambridge University Press, 1982.

. Reading the Past: Current Approaches to Interpretation in Archaeology.

Cambridge: Cambridge University Press: 1986.

. "Contextual Archaeology: An Interpretation of Catal Hiiyuk and a Discussion of

the Origins of Agriculture." In Institute of Archaeology Golden Jubilee Bulletin
Number 24. London: Institute of Archaeology, 1987.

. Conclusions to On the Surface: Catalhoyiik 1993-95. Edited by Ian Hodder.

Cambridge: McDonald Institute for Archaeological Research; London: British
Institute of Archaeology at Ankara, 1996.

Hodder, Ian, Michael Shanks, Alexandra Alexandri, Victor Buchli, John Carman, Jonathan
Last and Gavin Lucas. Interpreting Archaeology: Finding Meaning in the Past.
London: Routledge, 1995.

, ed. Symbolic and Structural Archaeology. Cambridge: Cambridge University

Press, 1982.

-, ed. The Archaeology of Contextual Meanings. Cambridge: Cambridge University
Press, 1987.

Bibliography 264

Hodges, Henry, ed. The Meanings of Things: Material Culture and Symbolic Expression.
London: Unwin Hyman, 1989.

, ed. On the Surface: Catalhoyiik 1993-95, 2. Cambridge: McDonald Institute for

Archaeological Research; London: British Institute of Archaeology at Ankara, 1996.

Hodges, Henry W. Neolithic Wall Paintings: Problems of Conservation. England: EC
United Kingdom Group, 1964.

Hodges, Henry W., ed. In Situ Archaeological Conservation, Proceedings of Meetings
April 6-13, 1986. Mexico: Instituto Nacional de Antropologia e Historia de
Mexico/ The Getty Conservation Institute, 1987.

Horie, C.V. Materials for Conservation: Organic Consolidants, Adhesives, and Coatings.
London: Butterworths, 1987.

Horky, Jan. "Restoration of Wall Paintings in a Small Chapel Near Fribourg." In Case
Studies in the Conservation of Stone and Wall Paintings, September 21-26 1986.
Bologna: International Institute for Conservation, 1986.

Hubbard, John R. "Animal Glues." In Handbook of Adhesives. New York: Van Nostrand
Reinhold Company Inc., 1977.

Humane, Kailas; P.C. Gupta. "Strappo Transfer of Murals of Bagh Caves." In
Conservation of Cultural Property in India 22 (1989): 140-141.

ICOM Committee for Conservation. Preprints of the 9th Triennial Meeting Dresden,

German Democratic Republic. Los Angeles: ICOM Committee for Conservation,
1990.

Jedrzejewska, Hanna. "The Warsaw National Museum: Conservation of the Mural
Paintings from Faras." In Museum 19, no. 3, (1966).

Jopling, Carol F., comp. Art and Aesthetics in Primitive Societies, a Critical Anthology.
New York: E. P. Dutton, 1971.

Bibliography 265

Kadokura, Takeo. "Exhibition of Wall-Paintings of the Tumulus Torazuka." In

InternationalSymposium on the Conservation and Restoration of Cultural Property:
Conservation and Restoration of Mural Paintings (I), November 17-21, 1983,
Tokyo, Japan. Japan: Tokyo National Research Institute of Cultural Properties,
1984.

Karpowicz, Adam. " Aging and Deterioration of Proteinaceous Media," Studies in
Conservation 26, no. 4 (November 1981): 153-160.

Kay, Reed. The Painter's Guide to Studio Methods and Materials. New Jersey: Prentice-
Hall, Inc., 1983.

Kopelson, Evan. "Analysis and Consolidation of Architectural Plasters from Catalhbyiik,
Turkey." Master's thesis, University of Pennsylvania, 1996.

Kostrov, P.I., and E.G. Sheinina. "Restoration of Monumental Painting on Loess plaster

Using Synthetic Resins," Studies in Conservation 6, nos.2 and 3 (August 1961): 90-
102.

Kostow-Benczew, Iwan. "The Porous Material Support Used For the Transfer of a Mural
Painting." In Ochrona Zabytkow 25, no 4 ( 1972):291-293.

Kumar, Rakesh and William S. Ginell, "A New Technique for Determining the Depth of
Penetration of Consolidants into Limestone Using Iodine Vapor," Journal of the
American Institute for Conservation 36, no. 2 (Summer 1997): 143-150.

Landrock, Arthur H. Adhesives Technology Handbook. New Jersey: Noyes Publications,
1985.

Larson, Londa J., Kyeong-Sook Kim Shin, and Jeffrey I. Zink. "Photoluminescence

Spectroscopy of Natural Resins and Organic Binding Media of Paintings." Journal
of the American Institute for Conservation 30 ( 1991 ): 89-104.

Last, Jonathon. "Surface Pottery at Catalhbyiik." In On the Surface: Catalhoyiik 1993-95.

Edited by Ian Hodder. Cambridge: McDonald Institute for Archaeological Research;
London: British Institute of Archaeology at Ankara, 1996.

Ling, Roger. Romano-British Wall Painting. United Kingdom: Shire Publications Limited,
1985.

Madjaric, Vlado. "The Exhibition on the Preservation of the Church St. Sophia of Ochrida."
In Museum 9, no. 1, (1956).

Bibliography 266

Majewski. L.J. "On Conservation of Wall Paintings From Archaeological Finds." In
Museum News 5 1 , no 4 ( 1 972): 11-12.

Majewski, L.J. The Techniques and Preservation of the Mosaics and Frescoes in the
Excavations at Sardes, Turkey. Amsterdam: ICCROM, 1969.

Majewski, L.J. "The Conservation ofWall Paintings in Archeological Excavations." In
Preservation and reproduction of Clay Tablets and the Conservation of Wall
Paintings, Colt Archeological Institute. London: Bernard Quaritch Ltd., 1968.

Marinos, Paul G. and George C. Koukis. The Engineering Geology of Ancient Works,

Monuments and Historical Sites: Preservation and Protection Proceedings of an
International Symposium Organized by the Greek National Group oflAEG, Athens,
19-23 September 1988. Rotterdam: Balkema, 1988.

Masschelein-Kleiner, Liliane. Ancient Binding Media, Varnishes and Adhesive s. Rome :
ICCROM, 1985.

Matteini, Mauro. "In Review: An Assessment of Florentine Methods of Wall Painting
Conservation Based on the Use of Mineral Treatments." In The Conservation of
Wall Paintings: Proceedings of a Symposium Organized by the Courtauld Institute
of Art and the Getty Conservation Institute, London, July 13-16, edited by Sharon
Cather. California: Getty Conservation Institute, 1991.

Matthews, Roger. "Surface Scraping and Planning." In On the Surface: Catalhoyuk 1993-
95. Edited by Ian Hodder. Cambridge: McDonald Institute for Archaeological
Research; London: British Institute of Archaeology at Ankara, 1996.

Matthew, Wendy, Charles French, Timothy Lawrence and David Cutler. "Multiple

Surfaces: the Micromorphology." In On the Surface: Catalhoyiik 1993-95. Edited
by Ian Hodder. Cambridge: McDonald Institute for Archaeological Research;
London: British Institute of Archaeology at Ankara, 1996.

Matthews, Wendy and Shahina Farid. "Exploring the 1960s Surface: the Stratigraphy of
Catalhoyuk." In On the Surface: Catalhoyiik 1993-95. Edited by Ian Hodder.
Cambridge: McDonald Institute for Archaeological Research; London: British
Institute of Archaeology at Ankara, 1996.

Mayer, Ralph. The Artist's Handbook of Materials and Techniques. New York: The Viking
Press, 1970.

Bibliography 267

McCrone. Walter C, John Gustav Delly and Samuel James Palenik. The Particle Atlas,
Edition Two. 6 Volumes. Michigan: Ann Arbor Science, 1979.

Mellaart, James. "Excavations at Catal Hiiyiik: First Preliminary Report, 1961." Anatolian
Studies 12 (1962): 41-65.

"Catal Hiiyiik in Anatolia: Excavations Which Revolutionise the History of the

Earliest Civilizations. Part H-Shrines and Buildings," The Illustrated London News:
Archaeological Section No. 2122 (February 2, 1963): 160-164.

. "Deities and Shrines of Neolithic Anatolia: Excavations at Catal Hiiyiik, 1962."

Archaeology 16, no. 1 (March 1963): 29-38.

. "Excavations at Catal Hiiyiik, 1962: Second Preliminary Report." Anatolian

Studies 13 (1963): 43-103.

. :Earliest of Neolithic Cities: The Third Season of Excavations at Anatolian Chatal

Huyuk. Part 1 -Shrines and Images, of 9000 Years Ago." Illustrated London News,
no.2 169 (February 1, 1964): 158-160.

Mellaart, James. "Earliest of Neolithic Cities: Delving Deep into the Neolithic Religion of
Anatolian Catal Huyuk, Part U-Shrines of the Vultures and the Veiled Goddess."
Illustrated London News, no. 2170 (February 8, 1964): 194-197.

. "Excavations at Catal Hiiyiik, 1963: Third Preliminary Report." Anatolian Studies

14 (1964): 39-1 19.

. "A Neolithic City in Turkey." In Scientific American, (April 1964).

. "Neolithic Religion at Catal Huyuk." In Thou Shalt Have No Other Gods Before

Me. New York: The Jewish Museum, 1964.

. "Catal Huyuk West." Anatolian Studies 1 5 ( 1 965): 1 35- 1 56.

. "Excavations at Catal Hiiyiik, 1965: Fourth Preliminary Report." Anatolian

Studies 16(1966): 165-191.

. "The Leopard Shrines of Chatal Huyuk," The Illustrated London News:

Archaeological Section no. 2246 (June 4, 1966): 24-25.

. Cato/ Huyuk, A Neolithic Town in Anatolia. Great Britain: Thames and

Hudson Limited, 1967.

Bibliography 268

Mellaart, James. The Neolithic of the Near East. London: Thames and Hudson, Ltd.,
1975.

Mellaart, James; Udo Hirsch; and Belkis Balpinar. The Goddess from Anatolia. West
Germany: Udo Hirsch, 1989.

The Metropolitan Museum of Art. The Great Age of Fresco: Giotto to Pontormo. New
York: The Metropolitan Museum of Art, 1968.

Miller, Eric G.; Lee, Nicholas J.; and Diane R. Elian. "Remounting and Conservation of
Archaeological Wall Paintings at the British Museum." In Recent Advances in the
Conservation and Analysis of Artifacts, compiled by James Black. London:
Summer's Schools Press/ University of London: 1987.

Mills, John S. and Raymond White. The Organic Chemistry of Museum Objects. London:
Butterworths, 1987.

Mitchell, John. "Early Medieval Wall Paintings Excavated in Germany, Italy, and England:
a Preliminary Survey." In Early Medieval Wall Painting and Painted Sculpture in
England. England: The Individual Authors, 1990.

Molleson, Theya and Peter Andrews. "Trace Element Analyses of Bones and Teeth from
Catalhoyiik. In On the Surface: Catalhoyiik 1993-95, edited by Ian Hodder.
Cambridge: McDonald Institute for Archaeological Research; London: British
Institute of Archaeology at Ankara, 1996.

Montero, Sergio Arturo. "The Conservation of Archaeological Painting." In In-Situ

Archaeological Conservation, Proceedings of Meetings April 6- J 3, 1986. Edited by
Henry W.M. Hodges. Mexico: The Getty Conservation Institute, 1987.

Mora, Laura, Paolo Mora, Giorgio Torraca, and Virginia Anne Bonito. "Treatment of
Peristyle Garden Wall in House of Menander, Pompeii." In Case Studies in the
Conservation of Stone and Wall Paintings: Preprints of the Contributions to the
Bologna Congress, September 21-26, 1986. Edited by N.S. Bromelle and Perry
Smith. London: International Institute for Conservation, 1986.

Mora, Paolo; Laura Mora and Paul Philippot. Conservation of Wall Paintings. London:
Butterworths, 1984.

Mortland, M.M. and V.C. Farmer. International Clay Conference 1978, Proceedings of the
VI International Clay Conference 1978 Held in Oxford, 10-14 July 1978. New
York: Elsevier Scientific Publishing Company, 1979.

Bibliography 269

Nardi, R. "Conservation on the Excavation: The Crypt of Balbus in Rome." In Conference
Proceedings: Preventive Measures During Excavation and Site Protection. Rome:
ICCROM, 1986.

Newey, Hazel; Dove, Simon; and Andrew Calver. "Synthetic Alternatives to Plaster of
Paris on Excavation." In Recent Advances in the Conservation and Analysis of
Artifacts, compiled by James Black. London: Summer's Schools Press/ University
of London: 1987.

Ostaszewska, Maria and Zalewski, Wladyslaw. "Study of Wall Paintings Transfer-History,
Works and Research Trends." In Przenoszenie Malowidel Sciennych Polsce,
Biblioteka Muzealn, I Ochrony Zabytkow Seria b, t.55. Warszawa: Ministerstwo
Kultury i Sztuki, 1979.

Park, David, David Perry "Rochester Cathedral: Conservation of the Crypt Vault

Paintings." In Case Studies in the Conservation of Stone and Wall Paintings:
Preprints to the Bologna Congress, 21-26 September 1986. London: DC, 1986:
182-185.

Pearce, Susan M., ed. Museum Studies in Material Culture. Leicester and London:
Leicester University Press, 1989.

Phillips, M.W. "Experiences in the Use of Acrylic Plaster Adhesives." In Case Studies in
the Conservation of Stone and Wall Paintings, September 21-26, 1986. Bologna:
International Institute for Conservation, 1986.

Phillipot, Paul; Mora, Paolo. "The Conservation of Wall Paintings." In The Conservation
of Wall Paintings with Special Reference to Tropical Conditions, Museums and
Monuments Series, 11. Paris: UNESCO, 1968.

Plenderleith, H.J. and A.E. Werner. The Conservation of Antiquities and Works of Art.
London: Oxford University Press, 1971.

Pliny the Elder, Gaius Plinius Secondus. Natural History, Book 35. Translated by H.
Rackham. Massachusetts: Harvard University Press, 1995.

Pollio, Marcus Vitruvius. The Ten Books On Architecture. Trans. Morris Hicky Morgan.
New York: Dover Publications, Inc., 1960.

Price, Stanley, ed. Conservation of Archaeological Excavations. Rome: ICCROM, 1984.

Bibliography 270

Procacci, Ugo. "Introduction-The Technique of Mural Paintings and Their Detachment."
In The Great Age of Fresco - Giotto to Pontormo, Catalogue of an Exhibition Held
at the Metropolitan Museum of Art, New York, Autumn, 1968. New York:
Metropolitan Museum of Art, 1968.

Pye, Elizabeth M. "The Treatment of Excavated Fragmentary Wall Plaster." In

International Symposium on the Conservation and Restoration of Cultural
Property. Conservation and Restoration of Mural Paintings (1) November 17-21,
1983, Tokyo, Japan. Tokyo: Tokyo National Research Institute of Cultural
Properties, 1984.

Recent Advances in the Conservation and Analysis of Artifacts: Jubilee Conservation
Conference Papers, compiled by James Black. London: University of London
Institute of Archaeology Summer's Schools Press, 1987.

Reddy, V. Rami. Neolithic and Post-Neolithic Cultures. India: Mittal Publications, 1991.

Roberts, Neil, Peter Boyer, and Romola Parish. "Preliminary Results of Geoarchaeological
Investigations at Catalhoyuk." In On the Surface: Catalhoyiik 1993-95. Edited by
Ian Hodder. London: British Institute of Archaeology, 1996.

Ross, C.S. and E.V. Shannon. "Minerals of Bentonite and Related Clays and Their Physical
Properties," J. Am. Ceram. Soc, 9 (1926): 77-96.

Rossi-Manaresi, Rafaella and Antonella Tucci. "Painting Technique of Cima Da

Conegliano and Analytical-Documentary Comments About Pigments. In Preprints
of the 9th Triennial Meeting Dresden, German Democratic Republic, 26-31 August
1990, Volume I. Los Angeles: ICOM Committee for Conservation, 1990.

Samanez Argumedo, Roberto. "Mural Painting on Adobe Wall During Peruvian Colonial
Times: Its Restoration and Conservation." In Case Studies in the Conservation of
Stone and Wall Paintings, Preprints of the Contributors to the Bologna Congress,
September 21-26, 1986, edited by N.S. Bromelle and Perry Smith. London:
International Institute for Conservation, 1986.

Sayre, Edward V.; Majewski, Lawrence J. "Studies for the Preservation of the Frescoes by
Giotto in the Scrovegni Chapel at Padua. II." In Studies in Conservation 8 (1963).

Bibliography 271

Schaible, Volker. "Historisches und Ethisches zur Abnahme von Wanmalerei." In

Historische Technologie und Konservierung von Wandmalerei: Vortragstext der
dritten Fach-und, Fortbildungstagung. Bern: Schule fur Gestaltung in Kommission
im Verlag Paul Haupt, 1985.

Schilling, Michael R. "Color Measurment of the Wall Paintings in the Tomb of Nefertari."
In ICOM Committee for Conservation. Preprints of the 10th Triennial Meeting
Washington, DC, USA; 22-27 August 1993. Paris: ICOM Committee for
Conservation, 1993

Schwartzbaum, P.M.; Ippolito Massari, Giovanna Pignatelli, and Carlo Giantomassi,

"Approaches to the Conservation of Mural Paintings in Underground Structures,
Case Studies of Recent Projects by ICCROM Consultants and Staff." In
International Symposium on the Conservation and Restoration of Cultural
Property: Conservation and Restoration of Mural Paintings (I), November 17-21,
1983, Tokyo, Japan. Tokyo: National Research Institute of Cultural Properties,
1984.

Schwartzbaum, P.M.; Silver, C.S. and C. Wheatley. "The Conservation of a Chalcolithic
Mural Painting on Mudbrick From the Site at Teleilat Ghassul, Jordan." In The
Third International Symposium on Mudbrick (adobe) Preservation. Edited by Okan
Ustunkok and Emre Madran. Ankara: ICCROM, 1982.

Silver, Constance. "Analyses and Conservation of Pueblo Architectural Finishes in the
American Southwest." In International Conference on the Conservation of
Earthen Architecture: Adobe 90 Preprints. Los Angeles: The Getty Conservation
Institute, 1990.

Silver, Constance S. "Architectural Finishes of the Prehistoric Southwest: A Study of the
Cultural Resource and Prospects for its Conservation." Master's thesis,
University of Pennsylvania, 1987.

Silver, Constance S. 1981 Report on the Development of Methods For the Conservation of
Pueblo Indian Mural Paintings in the American Southwest. ICCROM, 1982.

Singh, Purushottam. Neolithic Cultures of Western Asia. London: Seminar Press, 1974.

Smith, Watson. "Kiva Mural Decorations at Awatovi and Kawaika-a." In Papers of the
Peabody Museum of American Archaeology and Ethnology 37. Massachusetts:
Peabody Museum, 1952.

Bibliography 272^

Stavisky, B.J. and Zhukova N.N. "Parallel Progress in Restoration and Archaeology:
Discovery and restoration of monumental Painting and Sculpture on a Loess
Ground." In ICOM Committee for Conservation. Preprints of the 10th Triennial
Meeting Washington, DC. USA, 22-27 August 1993. United States: International
Council of Museums, 1993.

Subbaraman, S. "Separation of Two Layers of Mural Painting By Modified Stacco

Process." In Role of Chemistry in Archaeology: 1st International Colloquium, 15-
18 November 1991. Hyderabad: The Birla Institute of Scientific Research, 1991.

Subbaraman, S.; Jayaramsunderam, C.S. "Separation of the Chola and Nayaka Layers of
Paintings in the Brihadeeswara Temple, Thanjavur, by the Distacco Process." In
Symposium on Mural Paintings Under the Aegis oflndo-U.S. Sub-Commission on
Culture, Held in Ajanta in April 1980. Dehra-Dun: Archaeological Survey of India,
1980.

Tate, J.O., N.H. Tennent and J.H. Townsend, eds. The Proceedings of the Symposium
"Resins in Conservation " Held at the University of Edinburgh, Edinburgh, U.K.
21 s '-22 nd May 1982. Edinburgh: Scottish Society for Conservation and Restoration,
1983.

Tintori, Leonetto. 'A Reassessment of the Character of Materials Employed in Mural
Restoration Based on the Study of Laboratory Samples and Flood Damaged
Frescoes." In ICOM Report 69/42. ICOM, 1969.

Tintori, Leonetto. "Methods Used in Italy for Detaching Murals." In Recent Advances in
Conservation, Contributions to the UC Rome Conference. London: Butterworth's,
1961.

Todd, Ian. Catal Huyuk in Perspective. California: Cummings Publishing Company, 1976.

"The Prehistory of Central Anatolia I: The Neolithic Period." Studies in

Mediterranean Archaeology 60. Goteborg: Paul Astroms Forlag, 1980.

Tomek, Jindrich and Dorothea Pechova. "A Note on the Thin-Layer Chromatography of
Media in Paintings." Studies in Conservation 37 (1992) 39-41 .

Torraca, Giorgio and Paolo Mora, "Fissativi per Pitture Murali." In Bollettino dellTnstituto
Centrale del Restauro. Italy: ICCROM ( 1 965): 1 09- 1 32.

Urbani, Giovanni. "Restoration of Frescoes in Rome and Assisi." Connoisseur 36
(November 1955): 155-160.

Bibliography 273

Velde, Bruce. "Composition and Mineralogy of Clay Minerals." In Origin and Mineralogy
of Clays. Germany: Springer- Verlag Berlin Heidelberg, 1995.

Vunjak, M.; M. Medik. "Removal and Transferal of Wall Paintings From Nubia." In
Zbomik Zastite Spomenika Kultare 16 (1965).

Watson, Smith. "Kiva Mural Decorations at Awatovi and Kawaika-A with a survey of

other wall paintings in the pueblo southwest." In Papers of the Peabody Museum of
American Archeology and Ethnology, Harvard University 37; (1952).

Weaver, M. "The Use of an Inflatable Air-Dome to Produce Controlled Conditions for an
Archaeological Site." In Studies in Conservation 18, no. 2 (1973): 88-93.

Webley, D. "Soils and Site Location in Prehistoric Palestine." In Papers in Economic

Prehistory. Edited by E.S. Higgs. Cambridge: Cambridge University Press, 1972.

Wheatcroft, Andrew, ed. Science for Conservators Volume 3: Adhesives and Coatings.

London: The Conservation Unit of the Museums and Galleries Commission, 1987.

Winter, John. "The Characterization of Pigments Based on Carbon." Studies in
Consen'ation 28 (1983): 49-66.

Wolbers, Richard, Mary McGinn and Deborah Duerbeck. "Poly (2-ethyl-2-oxazoline): A
New Conservation Adhesive." N.p, n.d.

Wolbers, Richard and G. Landrey. "The use of direct reactive fluorescent dyes for the
characterization of binding media in cross-sectional examinations." In AIC
Preprints, American Institute for Conservation 15 th Annual Meeting, Vancouver,
British Columbia. Washington, D.C., 1987.

Xarrie, Josep Maria. "Pintures Murals de Sant Tomas de Fluvia," De Museus: Quadems de
Museologia I Museografia 1 (1988): 1 16-1 19.

Jak Yakar. Prehistoric Anatolia: The Neolithic Transformation and the Early Chalcolithic
Period. Monograph Series of the Institute of Archaeology, Tel Aviv University . Tel
Aviv: Institute of Archaeology of Tel Aviv University, 1991.

Yingtao, Qi. "Studies on Conservation of the Grotto Temples and the Mural Paintings of
Anicient Graves in China." In International Symposium on the Conservation and
Restoration of Cultural Property: Conservation and Restoration of Mural
Paintings (I), November 17-21, 1983, Tokyo, Japan. Japan: Tokyo National
Research Institute of Cultural Properties, 1984.

Bibliography 274

Zielinski, Adam. "In-Situ Conservation of the Temple of Amun Nahkt." In APT Bulletin
21,no.l (1989): 49-57.

Index

acetic acid 44

acrylic 80, 81, 103, 104, 105, 183, 219, 238,

252, 254, 256

Acryloid B-67 80, 81, 100, 102, 104. 198

Acryloid B-72 81, 86, 87, 88, 96. 97, 100, 102,

104, 117, 123, 127. 129, 133, 142, 153, 154,

183, 198,202,219,249,253
adhesive 16, 42, 58, 76, 80, 81, 82, 83, 84, 85,

90,96,97. 103, 104, 105, 111. 112, 113, 119.

122, 127, 141, 143, 144, 154, 155, 156, 157,

158, 159, 167, 183, 184, 185, 186, 187, 188,
191. 198, 200, 201, 203, 204, 208, 210, 211,
212. 213, 215. 216, 219, 226, 236, 237, 241,
243, 247, 249, 252, 255, 256

adhesive strength 16,58,76,90, 183, 191

Adhesives 43, 77, 80, 82, 84, 88, 90. 93, 96,

102, 158, 193. 194, 198, 199, 265, 266, 267.

270, 274
American Society for Testing and Materials3,

122, 160

amido black 43, 44

Anatolia 1

Anatolian Plateau 17

Ankara 4, 5, 7. 8, 9, 10, 12, 13, 14,20,21,23,

32,78, 118,222
Aquazol 50® 86, 87, 88, 91, 94, 97, 100, 1 17.

118, 119, 120, 123, 127, 133, 142, 154, 158,

159, 165, 184, 185, 189, 190, 191, 192, 193,
194, 195, 202, 208, 209, 210, 21 1, 217, 218

aqueous sodium acetate 44

archaeology 5, 6, 7. 8. 9, 10, 11, 12, 13, 14, 19,

20,21,23,32,75,242,244,255
Architectural Conservation Laboratory 2, 4, 58,

59, 118
ASTM 3, 121, 122. 123, 124, 127, 128, 129,

152, 160, 162, 163, 164, 166, 203, 204, 215,

216

aurochs 14, 18, 21

Awatovi 75, 239, 242, 272, 274

azurite 17

bentonite 60, 64, 65, 66, 271

Bliss 75, 239, 241, 260

British Institute of Archaeology 4

Bulgaria 248

calcareous 10

calcium 10, 32, 46, 47, 50. 52, 55, 58, 59, 60, 65,
68, 83, 222, 236. 237, 253, 254

Cambridge University 4, 6

Canada 256

capillary action 6, 120, 254

carbohydrates 45

Carsamba 6

gatalhoyiik 1, 2, 4. 5, 6, 7, 8, 9, 10. 1 1, 12, 13.
14, 15, 17, 18, 19, 20, 21, 22, 23, 24, 25, 30,
31, 32, 33, 42, 46, 47, 58, 59, 60, 68, 69, 74,
75,76,77,78.88.90. 112, 118, 119, 132,

182, 183, 222. 263, 264, 265, 266, 267, 268.
269,271

Catalhoyuk Research Trust 4

Catalonia 250

chalking 123, 124, 125, 126, 141,202

checking 160, 164, 165, 201, 208, 210

China 75, 245, 274

cholesterides 44

chroma 166

cinnabar 17

clays 10,46.59,60,61,62,63.64,66, 104, 112,

142, 198
cleavage 77, 79, 84, 88, 90, 91, 93, 97, 100, 141,

142, 154, 183, 184, 187, 191, 192, 193, 194,

195, 196,219,245

cohesive strength 15, 76, 81, 88, 153, 183

collagen-based adhesive 82, 83

colletta 85, 102, 103, 104, 105, 144, 155, 157,

158, 185. 198, 200, 201, 208, 209, 212, 213.

215,216,219,249
compatibility 3,79, 141, 182, 186, 187,208,

209,210,211

conservation 2

consolidant 67, 76, 80. 81. 85, 86, 97, 99, 100,

113, 117, 120, 121, 122, 132, 142, 153, 159,

183, 189, 200, 201, 207, 208, 209, 210, 211,
212, 213, 215, 216, 219, 220, 225, 226, 228,
257

consolidation iv, v, ix, 2, 3, 31, 36, 59, 67, 69,
76, 79, 80, 8 1 , 87. 88, 89, 90, 93, 99, 1 1 7,
118, 120, 121, 135, 141, 153, 154, 159, 182,
183, 184, 190, 191. 192, 195, 196, 197, 208.
209. 210, 211, 212, 213, 219, 226, 227, 251,
255

Contri 232

275

cracking 37, 38, 58, 60, 67, 68, 75, 78, 86, 87,
90,96, 112, 158, 160, 162, 165, 188, 189,
200,208,209,210,211,212,213

CRATerre 3, 121, 122, 135, 152, 184

Crosby 244,245,261

cross section 37, 38, 40

decomposition 43, 61, 75

delamination 68, 75, 76, 93, 96, 103, 105, 1 17.

188, 193, 194, 195, 196, 198
detachment 1, 2, 3, 4, 5, 31, 38, 58, 67, 74, 75,
77, 78, 79, 84, 85, 96, 97, 105, 1 17, 141, 143,
144, 152, 154, 155, 156, 157, 158, 159, 160,
163, 165, 182, 185, 186, 187, 188, 193, 194,
204, 208, 209, 210, 212, 213, 214, 219, 220,
221, 223, 229, 230, 232, 235, 236, 237, 238,
239, 242, 244, 245, 246, 248, 249, 251, 252,
253, 256, 257, 258
deterioration 2, 8, 15, 31, 36, 68, 69, 77, 79, 99,
104, 110, 113, 132, 136, 159, 165,201,210,
220, 224, 225, 235, 238, 241, 247, 248, 250,
252

disaggregation 78

Driessen and de Meester 6

earthen plaster 1, 239

efflorescence 237

egg yolk protein 44

Ellison and Harris 6

energy dispersive spectroscopy 54

England. .6, 82, 85, 251, 261, 262, 264, 265, 269
ethyl silicate monomer 117, 119, 120, 123, 127,

129, 133, 154
excavation 1, 4, 7, 8, 13, 16, 24, 25, 74, 75, 76,

77, 90, 141, 223, 229, 233, 238, 242

facing adhesives 3, 79, 96, 102, 103, 111, 113,
142, 143, 145, 154, 155, 158, 182, 185, 186,
188, 193, 194, 198, 199, 217, 218, 236, 242

facsimiles 3, 34, 46, 55, 60, 67, 68, 79, 183, 187,
221,258

Federation of Societies for Coatings Technology
3, 122, 152, 160, 161

flaking 77, 83, 84, 88, 99, 160, 163. 165, 183,
224, 245

Florence 231, 232, 234, 235, 237, 253

fluorescein isothiocynate 45

fluorescent dyes 43, 44, 274

French 10, 12, 32, 47, 78, 132, 222, 234, 263,
267

fresco buono 238

fresco secco 238

frescoes.. 229, 230, 235, 239, 249, 250, 252, 254

galena 17

gelatin 43, 44, 58, 82, 83, 88, 94

glacial acetic acid 44

Goddess vii, viii, 7, 13, 16, 19, 21, 22, 24, 268,

269
gum arabic 69, 83

Hodder4, 5,7, 8, 9, 10, 11, 12, 13, 14, 19,20,

21.23.24,32
Hue 166, 218

India... 75, 83, 247, 255, 259, 263, 265, 271, 273
Institute of Archaeology 11, 19, 20, 264, 265,
266,267,269,271,274

iodine crystals 132

Iodine Vapor Test 121, 122, 132

isopropanol 44, 86, 100, 189, 190, 191, 195

Konya 1, 5

Kostrov 243, 244, 266

Kuaua 75, 239, 241, 242, 260

Lime plaster 244, 249

loam 10

magnesium carbonates 10, 32

malachite 17

massello 3, 16, 77, 229, 237

Mellaart 1,7,9, 11, 13, 14, 16, 17, 19,21,22,

23, 24, 268, 269

microchemical spot tests 33, 47, 54

microcrystalline wax 84, 194

Montenegro 249

montmorillonite 32, 60, 63, 64, 65

Mora 2, 257

mudbrick 1,4, 15, 16, 30, 31, 77. 78, 90. 132,

184,244
Munsell 160, 166, 217, 218

276

mural paintings 2, 5, 16, 18, 19, 36, 59, 66, 74,
76, 1 10, 155, 182. 183, 187, 219, 229, 235,
237, 239. 241, 246, 250, 252, 254, 257

murals 20, 68, 74, 75, 77, 78, 1 1 1, 142, 186,
188,235,239,248,256

Neolithic 1

paintings 2, 9, 15, 16, 18, 30, 36, 42, 46, 68, 75,

78, 99, 104, 1 10. 113, 155, 159, 187, 222,

224, 228, 231, 236, 241, 245, 247, 249, 252,

253, 254, 257

paleoecological data 6

Paleolithic 14, 17, 18, 19, 24, 33

penetration 36, 37,84, 85,96,97, 111, 120, 121,

132, 142, 153. 154, 155, 184, 194, 219, 226

photomicrographs 36

Pictorial Standards of Coatings Defects 124, 161

pigments 17, 33, 46, 49, 262, 271, 274

Pleistocene 6, 9

Plextol B500 81. 96. 97, 102, 105, 143, 145,

193, 199,211

Pliny the Elder 230, 270

polarized light microscopy 47, 49, 50, 51, 52, 54,

polyester resins 45, 261

polyvinyl acetate emulsions 80

polyvinyl alcohol 80, 83, 84, 102, 103, 105, 111.

143. 144, 145, 154. 156, 157, 158, 159, 165,

167, 185, 187, 188.219

potassium ferrocyanide 47

powdering paint 3,67,78, 141, 153. 186

preconsolidant 67, 76, 80, 81, 86,97. 100, 113,

117, 120, 142, 143, 158, 159, 183, 184, 207,

208, 209, 210, 211,212, 213, 215, 216
PVA 111, 113,200,201,225,244,247,248,

249, 253

rabbit skin glue 58, 60, 82, 83, 86, 87

reattachment 1.77, 110, 111, 112, 113, 143, 156,

187,237,251,252

RhoplexAC-33 81, 190, 191, 192

Rochester Cathedral 251, 252, 270

Rome 43, 82, 85, 1 17. 1 18, 234, 236, 252, 259,

262,263,267,270,271,273

salts 75, 83, 132, 222, 248, 250, 252

sandbags \\2, 113, 154

scanning electron microscopy 54

smectites 63

Smith 75, 239, 242, 259, 261, 262, 269, 271,

272, 274

spectroscopy 2, 31, 33, 55, 62

stabilization 1,4, 74, 76, 79, 127, 182, 231

stacco 3, 16, 77, 85, 102, 141, 144, 153, 154,

155, 158, 159, 165, 185, 186, 187,200,201,

208, 212, 213. 215, 216, 221, 232, 235, 236,

237. 238, 246, 249, 250, 253, 255
strappo 3, 16, 77, 85, 102, 141, 144, 153. 154,

155, 156, 159, 165, 185, 187, 200, 201, 209,

213, 216, 232, 235, 236, 237, 238, 245, 246,

250, 254, 256
substrates 58, 68, 80, 123, 127, 129. 152

T-1919 86. 117, 118, 119, 120, 123, 127, 129,
133. 154, 158, 159, 165, 185, 189, 202, 212,
213,214,215,216,217,218

terra cotta 2, 58, 67, 123, 127, 133, 152, 256

thermoplastic acrylic resins 80, 81

Tintori 235, 236, 273

Todd 6, 7, 24

transfer 245,246, 247, 251,265,266,270

triglycerides 44

triphenyl tetrazolium chloride 44, 45

Tumacacori 244, 245, 261

ultra-violet fluorescence microscopy 43

ultra-violet illumination 45

University of Pennsylvania 2

value 166,218

Vasari 231

Vinamul 6825 .. 85, 102, 105, 143, 145, 199, 211
Vitruvius 230, 270

Wall paintings 8, 17, 24, 68. 77, 222, 229

Water Drop Test 121, 122, 135, 185

Webley 6

277

3 1198 02418 2839

INI Mil

'^^iiiiliiiiliilll

N/infl/oama/5fl3«?x

i*M&:"<

■

mmm

,m <v I

■ I

■

■ hw

I *'.&*'< *

■ ■

1911

aanHnyi

ffiH
IH

;

WBKmW

■

iS£i

.:•;;•;:■-:,; ''■,,; 93