K. ªERIFAKI, H. BÖKE: PLASTER CHARACTERISTICS OF BYZANTINE WALL PAINTINGS IN WESTERN ANATOLIA
471–478
PLASTER CHARACTERISTICS OF BYZANTINE WALL
PAINTINGS IN WESTERN ANATOLIA
ZNA^ILNOSTI OMETOV BIZANTINSKIH STENSKIH SLIK V
ZAHODNI ANATOLIJI
Kerem ªerifaki
*
, Hasan Böke
Department of Conservation and Restoration of Cultural Heritage, Ýzmir Institute of Technology, Ýzmir, Turkey
Prejem rokopisa – received: 2021-11-21; sprejem za objavo – accepted for publication: 2022-07-12
doi:10.17222/mit.2022.555
In this study, the execution technique and material characteristics of the plaster layers of Byzantine wall paintings from three ar-
chaeological sites (Anaia, Olympos and Aigai) from western Anatolia were examined for the purpose of their conservation.
Throughout the study, mineralogical and chemical composition of the plaster layers were determined with a scanning electron
microscope coupled with an X-ray energy dispersive system (SEM-EDS) and X-ray diffraction analyses (XRD). Reflected light
microscopy (RLM) and SEM studies conducted on polished cross-sections of samples revealed the stratigraphy and
microstructural properties of the plaster layers of the Anaia Church (4
th
–12
th
centuries), Baºpýnar Church (13
th
century) in
Olympos and a Byzantine settling and the Chapel (13
th
century) in Aigai. Throughout the research, pure lime and aggregates in
the plasters of Aigai, as well as magnesium-rich lime and straw in the plasters of Anaia and Baºpýnar were determined. The
plasters from the Anaia and Baºpýnar Churches exhibit characteristics similar to the samples from the post-iconoclastic period
due to a similar raw-material use. In the samples from Aigai, plastering techniques such as marmorino, intonachino and
cocciopesto, indicating earlier periods, were observed. As a result of this study, it was concluded that the number and character-
istics of the plaster layers from the paintings had been applied intentionally in accordance with the technique of the painting.
Plaster characteristics of the wall paintings in western Anatolia from the Byzantine period were found to be similar to the ones
in Ýstanbul, the Balkans, Crete and Cyprus. The results of this study will guide the conservation efforts of the wall paintings.
Keywords: wall paintings, Byzantine, lime plaster, post-iconoclastic
V tem ~lanku avtorji predstavljajo tehnike izvedbe in materialne lastnosti plasti ometov, ki so nosilci bizantinskih stenskih slik,
iz treh arheolo{kih najdi{~ v zahodni Anatoliji (Anaia, Olympos in Aigai), z namenom njihovega konserviranja. Mineralo{ka in
kemijska sestava plasti ometov je bila dolo~ena z vrsti~nim elektronskim mikroskopom sklopljenim z energijsko disperzijsko
spektroskopijo (SEM-EDS) in rentgensko difrakcijo (XRD). Stratigrafske in mikrostrukturne lastnosti slojev ometa cerkve
Anaia (4. do 12. stol.), cerkve Baºpýnar (13. stol.) v Olymposu ter bizantinske naselbine in kapelice (13. stol.) v Aigai, so bile
dolo~ene z odbojno svetlobno mikroskopijo (RLM) in SEM analizo na spoliranih pre~nih prerezih vzorcev. V ometih iz mesta
Aigai je bila odkrita prisotnosti ~istega apna in agregatov, ometi iz Anaia in Baºpýnar, pa so bili proizvedeni iz apna bogatega z
magnezijem in slame. Ometa v cerkvah Anaia in Baºpýnar imata glede na uporabljene primarne materiale podobne
karakteristike kot vzorci iz post-ikonoklasti~nega obdobja. Pri vzorcih iz Aigai so bile uporabljene tehnike izdelave ometov, kot
so marmorino, intonachino in cocciopesto, kar ka`e na poznej{e obdobje. Rezultati preiskave so pokazali, da je bilo {tevilo in
lastnosti slojev ometa za poslikave izbrano zavedno, glede na tehniko stenske slike. Lastnosti ometov za stenske slike v zahodni
Anatoliji v bizantinskem obdobju so enake s tistimi v Istanbulu, na Balkanu, Kreti in Cipru. Rezultati te {tudije so zato dobro
vodilo pri konserviranju stenskih slik.
Klju~ne besede: stenske slike, Bizanc, apneni omet, post-ikonoklasti~en
1 INTRODUCTION
Wall paintings applied on lime plaster were fre-
quently used in history, going as far back as the Neolithic
settlements.
1
The raw materials of these plasters and the
ways they were applied differed among civilizations.
Lime, which first found use in portable art works with
the development of pyrotechnology,
2
became widespread
over time and one of the most common binders used in
plasters and mortars. Lime has been an indispensable
material in the Roman architecture. Although it was a
continuation of the Roman Empire, the materials used in
the Byzantine architecture show regional differences.
The plasters used in the Byzantine period are especially
different from the ones used in the Roman period and in
Western Europe. Of the early Middle Ages, it is known
that a simplification of the classic Roman technique for
wall-painting plasters occurred.
3,4
This simplification re-
duced the number of plaster layers, while the main com-
ponents stayed the same. However, after Iconoclasm (8
th
and 9
th
centuries AD), straw and animal hairs started to
be used as inert charges with a small quantity of sand in
wall painting plasters.
4
In this period Byzantine render-
ings of wall paintings consisted of two layers, arriccio
(the rough plaster) and intonaco (the fine plaster), which
consisted of lime, straw, chopped hog bristles, and con-
taining only a small quantity of sand. Crushed straw is a
common ingredient of the plasters in the Byzantine pe-
riod. Because of this material, Byzantine plasters are
more related to traditional clay renderings.
4
The purpose
of using straw was to reduce the shrinkage and prevent
the cracking of plaster. The water-retention character of
Materiali in tehnologije / Materials and technology 56 (2022) 5, 471–478 471
UDK 75.052:726(37) ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 56(5)471(2022)
*Corresponding author's e-mail:
keremserifaki@iyte.edu.tr (Kerem ªerifaki)

straw also slows down the drying of plaster, providing a
longer working time, as required for the fresco tech-
nique.
5
During the Byzantine period, bricks were used in
the construction of walls. Brick walls absorb moisture
from the plaster. Because of this, the thickness of plas-
ters was increased to provide the moisture needed for
frescoes.
4
The thickness of the arriccio and intonaco can
be 20–30 mm in total.
5
The Craftsman’s Handbook of Cennino Cennini,
written in the 14
th
century, is a historic source that gives
important information on the material use and method of
painting in the Byzantine era.
6,7
Its author recommends
two units of sand and one unit of lime for plaster while
both ingredients must be well sifted. Before plastering,
the wall has to be swept and well wetted. After that, the
first layer of plaster is applied onto the wall. The first
layer should be fairly rough and quite uneven to provide
an appropriate surface for the adhesion of the second
plaster layer. The picture is drawn on the first plaster
layer when it is dry. After wetting down the first layer, a
thin and quite even layer of plaster is applied over it.
7
Dionysios of Fourna
8
provided detailed information
on the painting technique of the late Byzantine period. In
his Painters’ Guide, he gives instructions for the tech-
nique of Manuel Panselinos. Fourna recommends slaking
lime in a trough and mix the lime, straw and tow with
plenty of water. Straw and tow should be clean and the
mixture should be left to settle for two or three days until
it becomes appropriate for plastering. Before plastering a
wall, he also recommends wetting its surface well.
Fourna recommends polishing the plaster after wetting it
with water. Afterwards the picture composition should
be drawn on the plaster with a compass and ochre brush-
work. The second polishing, the application of the color
black and the third polishing should be carried out.
7,8
Beside the ancient sources, studies concerning the
late Byzantine-period wall paintings support this signifi-
cant change in the material use. Studies conducted on the
wall paintings from the Kariye Museum (from the early
14
th
century) detected two plaster layers composed of
lime and a large amount of coarse straw without any
sand aggregate.
9
Similarly, in the Sancta Sophia Church
in Trabzon, intonacoes were composed of lime, coarse
straw and negligible amounts of sand.
10
Wall paintings of
the Protaton Church (13
th
century) were executed on a
fine plaster layer of pure lime (1 mm) over a plaster layer
(of up to 30 mm) composed of lime, straw and small
amounts of aggregate.
11
A similar implementation was
K. ªERIFAKI, H. BÖKE: PLASTER CHARACTERISTICS OF BYZANTINE WALL PAINTINGS IN WESTERN ANATOLIA
472 Materiali in tehnologije / Materials and technology 56 (2022) 5, 471–478
Figure 1: Studied monuments and samples collected from the archaeological sites

also found on the wall paintings of the Monastery of
@i~a (13–14
th
centuries) in Serbia.
12
Linen fibers in St.
Theodori Church (16
th
century) and straw in St. Trinity
Church (14
th
century) were detected in the fine plasters
of the wall paintings.
13
Hein et al. also found consider-
able amounts of animal and plant fibers in plaster layers
of the wall paintings of the churches at the Mani Penin-
sula dating back to 10–15
th
centuries.
14
Winfield reported
the use of lime plaster composed of lime, sand or lime
dust in the wall paintings from the 1
st
to the 4
th
century
and the use of straw in the plasters of the wall paintings
of Direkli Kilise (10
th
century), Samanlý Kilise in
Cappadocia, St Nicholas Church in Myra (12
th
century),
Byzantine churches of Pontus from the 10
th
to the 15
th
century and the Parecclesion of the Kariye Church (14
th
century).
7
Beside Ýstanbul and Anatolia, straw was de-
tected in the middle and later period Byzantine wall
paintings from Yugoslavia, Russia, Cyprus and Italy.
7,15
In this study, material characteristics of the plasters
of wall paintings from three byzantine settlings from
western Anatolia are determined. Painting samples be-
long to the buildings, constructed in a close time range.
The studied wall paintings from Aigai (Manisa) and
Baºpýnar – Olympos (Ýzmir) were dated to the 13
th
cen-
tury and the ones from Anaia (Aydýn) to the 4–12
th
cen-
tury (Figure 1).
2 EXPERIMENTAL PART
Experimental studies were carried out on small frag-
ments of the wall paintings, provided by the head of ar-
chaeological excavation teams from their archives. Be-
side the substructure of the Anaia Church, two further
structures were widely demolished and only small frag-
ments of wall paintings survived. During sampling, spe-
cial care was taken not to damage the in-situ conserved
paintings. Plaster samples collected from the soil fill and
classified by the excavation team during excavations
were preferred for studying. During the sampling, visual
similarities between the samples from the western fortifi-
cation wall of Aigai (2
nd
century) and samples from the
byzantine settlement (13
th
century) were observed.
Therefore, a sample from the fortification wall was in-
cluded into the study for comparison.
The samples were labelled with letters or abbrevia-
tions showing the archaeological site that they were col-
lected from (Aigai: Aig, Olympos: O, Anaia; An). The
samples that had more than one layer, had the same basic
sample number, but with different decimal numbers (Ta-
ble 1).
The chemical and microstructural properties of the
plaster layers were determined with a scanning electron
microscope (SEM) coupled with an X-ray energy disper-
sive system (EDS) and a polarized light microscope
(PLM). SEM-EDS analyses were performed on polished
sections of the samples using Philips XL 30S-FEG. The
elemental mapping acquired with the SEM was used for
distinguishing plaster layers with aggregates of different
origins. PLM analyses were carried out with a Nikon
Eclipse E400 polarized light microscope.
The mineralogical compositions of fine plaster layers
were determined with X-ray diffraction analyses (XRD)
performed on fine-powdered plaster samples with grain
sizes of less than 53 μm, with a Philips X-Pert Pro X-ray
diffractometer. The instrument operated with a CuK
 ra-
diation and a Ni filter adjusted to 40 kV and 40 mA in a
range of 5–60°, with a scan speed of 1.60 °/min. Mineral
phases of each X-ray diffraction spectrum were identi-
fied with the Philips X-pert Graphics and Identity soft-
ware program.
Table 1: Plaster samples collected from the excavation archieves and
their original locations
Sample name Structure Layers
Anaia site
An1 Substructure of Anaia Church 1
An2 Substructure of Anaia Church 1
An3 Substructure of Anaia Church 1
An4.1 An4.2 Substructure of Anaia Church 2
An5 Substructure of Anaia Church 1
An6.1 An6.2 Substructure of Anaia Church 2
An7.1 An7.2 Substructure of Anaia Church 2
An8 Synthronon of Anaia Church 1
An9 Substructure of Anaia Church 1
An10 Substructure of Anaia Church 1
An11 Substructure of Anaia Church 1
Olympos site
O1.1 O1.2 O1.3 Baºpýnar Church 3
O2.1 O2.2 Baºpýnar Church 2
O4 Baºpýnar Church 1
O5 Baºpýnar Church 1
O6 Baºpýnar Church 1
O7 Baºpýnar Church 1
O8 Baºpýnar Church 1
O9 Baºpýnar Church 1
O10 Baºpýnar Church 1
O11.1 O11.2 Baºpýnar Church 2
Aigai site
Aig1.1 Aig1.2
Iron Gate region – Byzantine
settlement
2
Aig2.1 Aig2.2 Aig2.3
Bottom of the western fortifica-
tion wall
3
Aig3.1 Aig3.2
Iron Gate region – Byzantine
settlement
2
Aig4.1 Aig4.2 Aig4.3 Byzantine Chapel 3
Aig5.1 Aig5.2 Aig5.3 Byzantine Chapel 3
3 RESULTS AND DISCUSSIONS
One of the important factors affecting the characteris-
tics of plaster layers in Byzantine wall paintings is the
painting technique and the requirements for this tech-
nique. The fresco and secco techniques were both widely
used for Byzantine wall paintings. Depending on the
painting technique used, the workmanship and the mate-
rial of the surface to be painted, the number and thick-
K. ªERIFAKI, H. BÖKE: PLASTER CHARACTERISTICS OF BYZANTINE WALL PAINTINGS IN WESTERN ANATOLIA
Materiali in tehnologije / Materials and technology 56 (2022) 5, 471–478 473

ness of the plaster layers vary. For the fresco technique,
it is necessary to prevent rapid drying of the plaster so
that the working time can be extended. Therefore,
multi-layered plasters were used and the thickness of the
plaster layers was increased if the surface was con-
structed of brick or similar materials that absorb water.
In the secco technique, a painting was executed on dry
plaster and it was important to obtain a smooth, white
surface, to which the painting could be applied. In this
case, thinner and fewer layers of plaster were used as
there was no need to prevent rapid drying of the plaster.
In a comprehensive study, which is part of this paper,
the painting technique used for the wall paintings was
determined by evaluating the diffusion of paint layers
into the plaster layer and the presence of a carbonation
line between these layers.
16
It was observed that the sam-
ples taken from the Anaia Church consisted of a single
thin layer, covering a brick vault surface, while the sam-
ples taken from the Baºpýnar Church and Aigai Byzan-
tine settlement were composed of multilayered rough
and fine plasters, covering stone wall surfaces (Fig-
ure 2). It was determined that all the samples, except for
one, taken from the Anaia Church were executed in the
secco technique; the paintings from the Olympos
Baºpýnar Church were painted in the fresco technique,
but these were sometimes completed in the secco tech-
nique; and the samples from the Byzantine settling of
Aigai were painted in the fresco technique.
16
The pig-
ments used for the painting were earth-based inorganic
pigments: ochres for red and yellow, green earth (celado-
nite and glauconite) for green, natural ultramarine (lapis
lazuli) and Egyptian blue for blue, calcite and kaolin for
whites, carbon black for black and grey formed the pal-
ette of these paintings.
16
Although the stratigraphy and microstructural charac-
teristics of the fine plaster layers from the three sites
vary, they exhibit similarities in some parts (Figure 3).
For all the samples, the thickness of the fine plaster lay-
ers decreases from the one on the wall to the surface
layer. The layer on the surface contains the finest aggre-
gates or no aggregates and generally has a less-porous
structure. The interface of the plaster layers has a
rougher surface than the surface of the painting, provid-
ing good adhesion of the layers. In some of the samples,
fine plaster layers could only be distinguished with mi-
croscopic methods.
Table 2: Elemental compositions (w/%) of the binders of fine plaster
layers determined with SEM–EDX
Na
2
O MgO Al
2
O
3
SiO
2
K
2
O CaO FeO
Anaia An3 – 3.4 – 3.7 – 92.9 –
Olymp
os
O1.1 0.1 6.9 1.1 8.6 – 79.9 0.6
O1.2 0.3 8.9 0.8 2.9 – 85.5 0.3
O1.3 – 6.3 0.8 3.8 – 87.0 –
Aigai
Aig2.1 1.4 0.8 – – – 96.9 –
Aig2.2 0.7 0.5 5.3 18.9 2.7 71.1 –
Aig2.3 0.6 0.7 3.7 10.4 0.3 81.9 1.2
K. ªERIFAKI, H. BÖKE: PLASTER CHARACTERISTICS OF BYZANTINE WALL PAINTINGS IN WESTERN ANATOLIA
474 Materiali in tehnologije / Materials and technology 56 (2022) 5, 471–478
Figure 2: Schematic sections of the wall paintings: a) walls of the Baºpýnar Church, b) vault of the substructure of the Anaia Church
Figure 3: SEM images of the cross-sections of plaster samples from:
a) Anaia (An3), b) Olympos (O1), c)Aigai (Aig4)

The chemical compositions of the binder of plasters
determined with SEM-EDS for the Anaia and Olympos
samples are similar (Table 2). They mainly include CaO
and MgO due to the lime binder and SiO
2
due to the
superfine aggregates. On the Aigai samples, the superfi-
cial fine plaster (FP1) is mainly made of pure CaO, while
other plaster layers that lie under this layer are made of
fat lime and aggregates.
In the substructure of the Anaia Church, two different
plaster implementations were detected. In most of the
samples, fine plasters consist of one or two layers with
similar characteristics (Figure 3). The thickness of these
layers is generally between 6–10 mm. In one sample
(An-05), it is 15 mm. Pores, developed due to the decay
of the organic tows (straw or chaff), dominate the micro-
structure of these plasters. Microcracks, small amounts
of aggregates and residues of organic fibers are also visi-
ble. Secondly, in one sample (An-04) a compact and
white fine plaster layer (2.5 mm), mainly composed of
lime and marble dust, over a lime plaster layer, com-
posed of lime, sand, and a small amount of organic fi-
bers, was detected. These two layers adhere well to each
other. Fine plasters of the Baºpýnar Church are composed
of successive layers of roughly 1–3 mm in thickness. In
the microstructure of the plaster layers, microcracks and
pores, developed due to the decay of organic fibers, were
detected. The adhesion of the plaster layers to each other
is sound. The absence of distinct carbonation line in the
interface shows that the plaster layers were executed
while the underlaying layer was still wet.
The samples from Aigai consist of two or three layers
of fine plasters. The thickness of fine plaster 1, located
on the surface, is in a range of 1–3.5 mm. The thickness
of fine plaster 2 is in a range of 4–7.5 mm. The thickness
of fine plaster layer 3 is in a range of 4.5–7 mm even
though it is not certain that it comprises the whole layer.
The particle size of the aggregates used in the plasters
has a wide range between 2.2 mm and 45 μm. Although
semi-rounded aggregates were observed in the plaster
matrix, the majority of the aggregates have sharp edges.
The aggregate types used in the plaster layers of Aigai
were determined with PLM and elemental mapping (Fig-
ure 4). Except for one sample (Aig3), the paint layer was
applied on a fine plaster composed of lime and marble
powder (marmorino). Under the marmorino, one or two
fine plaster layers composed of lime and sand (into-
nachino) are present. On one of the samples, the use of
cocciopesto was observed. Tow (straw or chaff), which is
an important ingredient of Byzantine lime plasters,
4
is
absent in all layers. Both the stratigraphy of fine plasters
and the use of marble powder "in intonaco" are consis-
tent with the suggestions by Vitruvius and Pliny. All of
the samples selected from the Byzantine settlement, to-
gether with the one sample taken from the fortification
K. ªERIFAKI, H. BÖKE: PLASTER CHARACTERISTICS OF BYZANTINE WALL PAINTINGS IN WESTERN ANATOLIA
Materiali in tehnologije / Materials and technology 56 (2022) 5, 471–478 475
Figure 4: a, b, c) SEM images of the interface of the successive fine plaster layers (O1) from Baºpýnar, d) SEM image of the polished section of
marmorino and intonachino layers (Aig2), e) elemental mapping (Si) of marmorino and intonachino layers of Aig2, f) elemental mapping (Ca) of
marmorino and intonachino layers of Aig2

wall (2
nd
century AD), represent similar stratigraphies,
raw material usage, chemical and mineralogical charac-
teristics.
Table 3: Mineralogical compositions of the plasters
Minerals
calcite
aragonite
vaterite
dolomite
magnesite
diopsite
andesine
albite
quartz
Sample Anaia site
An1 +++ +
An2 +++ + +
An3 +++ +
An4.1 +++ ++
An4.2 +++ +
An5 +++ +
An6.1 +++
An6.2 +++ ++
An7.1 +++
An7.2 +++ ++
An8 +++
An9 +++ + +
An10 +++ + + +
An11 +++
Olympos site
O1.1 +++ + + +
O1.2 +++ + + +
O2.1 +++ + +
O2.2 +++ + + +
O4 +++ + + +
O5 +++ + ++
O6 +++ + + ++
O7 +++ + ++
O8 +++ + + + +
O9 +++ + + +
O10 +++ + + + + +
O11.1 +++ + + + +
O11.2 +++ + + ++
Aigai site
Aig1.1 +++
Aig1.2 +++ + + +
Aig2.1 +++ +
Aig2.2 +++ + + +
Aig2.3 +++ + + +
Aig3.1 +++ ++ + +
Aig3.2 +++ + + +
Aig4.1 +++
Aig4.2 +++ + + +
Aig4.3 +++ ++ + +
Aig5.1 +++ +
Aig5.2 +++ + + +
Aig5.3 +++ + + +
The number of (+) represents the abundance of mineral peaks.
Mineralogical characteristics of the fine plaster layers
were determined with an XRD analysis (Table 3). In the
XRD patterns of the samples from the substructure of the
Anaia Church, CaCO
3
in the calcite, aragonite and
vaterite forms and quartz were detected (Figure 5).
Quartz peaks are due to small amounts of aggregates
present in the plaster. Aragonite and vaterite are poly-
morphs of calcite that may be formed as a result of
self-healing of plasters in the presence of magnesium.
17
Elemental composition of the samples determined by
SEM -EDS revealed that the MgO content of the plasters
is in a range of 2.99–9.42 %. The presence of these poly-
morphs may be attributed to the recrystallization of dis-
solved calcite due to the wet-dry cycle occurring in the
substructure of the Anaia Church. In the 12
th
century, the
church was surrounded by the walls of a citadel, enclos-
ing all the openings of the substructure, except for the
entrance. Afterwards the entrance of the substructure
was covered by the earthquakes in the middle of 13
th
and
14
th
century. The rising damp from the ground and the
water drained from the vaults of the substructure created
a humid atmosphere and wet surfaces that may have
caused this phenomenon. In the SEM analyses, a differ-
ent type of crystallization of CaCO
3
was observed in the
voids. But, since the sizes of the cracks and voids are so
wide, due to the loss of tows, it is not possible to attrib-
ute this to self-healing. A thin calcium carbonate layer
that covers the original paint surface and decreases the
legibility of the wall paintings is also linked to the
recrystallization of dissolved calcite.
16
K. ªERIFAKI, H. BÖKE: PLASTER CHARACTERISTICS OF BYZANTINE WALL PAINTINGS IN WESTERN ANATOLIA
476 Materiali in tehnologije / Materials and technology 56 (2022) 5, 471–478
Figure 5: XRD patterns of the Olympos and Anaia samples display
different polymorphs of CaCO
3
(calcite, vaterite, aragonite) formed
with the recrystallization of dissolved calcite due to the wet-dry cycle
in the humid atmosphere
Figure 6: XRD patterns of the Aigai samples show the differences due
to a different aggregate usage

In the XRD patterns of the Nif-Olympos samples,
calcite, quartz, dolomite and vaterite peaks were deter-
mined (Figure 5). The mineralogical compositions of the
plasters of the Aigai samples show the differences de-
pending on the type of the aggregates (Figure 6). In
marmorinos, sharp calcite peaks derived from the car-
bonated lime and marble aggregates were detected. In
the Intonachino and cocciopesto samples, quartz, albite
and andesine due to the aggregates, and calcite due to the
carbonated lime were detected.
The raw-material compositions of the plasters and
pigments can be used for date estimations. Since the
technology has not changed dramatically throughout the
centuries, the information gathered as a result of the ex-
perimental studies intended for the determination of
raw-material compositions could only define approxi-
mate time intervals. In this respect, fine plaster samples
of the three sites were divided into two groups. In the
first group, only lime and aggregates of different origins
(marble, silicates or crushed brick) were determined. The
plasters of the second group were composed of lime,
negligible amounts of aggregate and large amounts of or-
ganic fibers. These differences in the plaster characteris-
tics may indicate that the samples may belong to differ-
ent periods. The samples from Aigai and one sample
from the Anaia Church were dated to the period before
iconoclasm. The samples from the Baºpýnar Church
(Nif-Olympos) and ten samples from the Anaia Church
were dated to the period after iconoclasm (Table 4).
4 CONCLUSIONS
In this study, stratigraphy, chemical composition,
mineralogical characteristics and microstructural charac-
teristics of the Byzantine wall paintings from three sites
(Anaia, Baºpýnar and Aigai) from western Anatolia were
determined. Furthermore, dating of the plasters was sug-
gested by comparing the material characteristics and exe-
cution techniques.
Lime plaster types, composed of pure lime and fine
marble, brick and sand aggregates, known as marmorino,
cocciopesto and intonachino, were used in Aigai.
The samples from the Byzantine settlement of Aigai
show strong similarities with the 2
nd
century wall-paint-
ing sample from Aigai. Considering different raw materi-
als used for the 12–13
th
century samples from Anaia and
Olympos and various post-iconoclastic samples from the
literature survey, we can notice a confusion regarding the
layering of the Aigai settlement. The samples so far
thought to belong to the Late Byzantine period may actu-
ally belong to the Roman period. The most probable ex-
planation for this confusion is the fact that Roman paint-
ing fragments were carried out down the slope and filled
the ruins of the Byzantine period.
The plasters of the Anaia Church and Baºpýnar
Church were mainly composed of magnesium-rich lime
and straw that were common plastering materials of the
post-iconoclastic period of Anatolia. The main conserva-
tion problem with these plasters is the pores that oc-
curred due to the decay of the organic tows. The adhe-
sion of the plaster layers to each other is generally
sound, but the layers are detaching from their supports.
Vaterite and aragonite, the polymorphs of calcite, are
present in the fine plasters of the Anaia Church. Vaterite
is also present in the samples from the Baºpýnar Church.
These polymorphs may be the indicators of the
recrystallization of dissolved lime in the pores of the
plasters due to a water penetration and humid environ-
ment. A thin layer of calcium carbonate deposits present
on the paint surfaces is an indicator of this process.
One sample from the Anaia Church shows similari-
ties with the pre-iconoclastic period wall paintings with
the presence of a superficial marmorino layer and aggre-
gates, without any tow in the subjacent plaster layer. A
single picture layer was observed in the excavated part of
the substructure. However, the presence of a plaster frag-
ment from the early period is important as it shows that
there may be more than one paint layer in the parts of the
building that have not been excavated yet. This informa-
tion may help the future excavations that will be carried
out at the site.
Acknowledgment
The authors extend their thanks to the researchers
from the Centre for Materials Research of the Ýzmir In-
stitute of Technology for the XRD and SEM-EDS analy-
ses carried out during the experimental stage of this
study.
K. ªERIFAKI, H. BÖKE: PLASTER CHARACTERISTICS OF BYZANTINE WALL PAINTINGS IN WESTERN ANATOLIA
Materiali in tehnologije / Materials and technology 56 (2022) 5, 471–478 477
Table 4: Possible dating of the plaster samples based on the raw-material compositions
Site Anaia Baºpýnar-Olympos Aigai
Sample name
An-01
An-02
An-03
An-04
An-05
An-06
An-07.1
An-07.2
An-07.3
An-08
An-09
An-10
An-11
O-01
O-02
O-03
O-04
O-05
O-06
O-07
O-08
O-09
O-10
O-11
Aig-01
Aig-02
Aig-03
Aig-04
Aig-05
Before icono-
clasm
(8–9
th
cen. AD)
 
After icono-
clasm
(8–9
th
cen. AD)
 

5 REFERENCES
1
D. S. Çamurcuoðlu, The wall paintings of Çatalhöyük (Turkey): Ma-
terials, technologies and artists, Unpublished PhD thesis, University
College London / Institute of Archaeology, London 2015
2
W. D. Kingery, P. B. Vandiver, M. Prickett, The beginnings of
pyrotechnology, Part II: Production and use of lime and gypsum
plaster in the pre-pottery Neolithic Near East, J. Field Archaeol., 15
(1988) 2, 219–244, doi:10.2307/530304
3
Z. Murat, Wall paintings through the ages: the medieval period (Italy,
twelfth to fifteenth century), Archaeol. Anthropol. Sci., 13 (2021) 11,
1–27, doi:10.1007/s12520-021-01410-4
4
P. Mora, L. Mora, P. Philippot, Conservation of wall paintings,
Butterworths, London 1984
5
R. Cormack, Wall paintings and mosaics, The Oxford handbook of
Byzantine studies, Oxford University Press, Oxford 2008, 385–395
6
C. Cennini, The Craftsman’s Handbook (Il Libro dell’Arte), Trans-
lated by D. V. Thompson, Dover, New York 1960
7
D. C. Winfield, Middle and later Byzantine wall painting methods. A
comparative study, Dumbarton Oaks Papers, 22 (1968), 61–139,
doi:10.2307/1291276
8
Dionysius of Fourna, The ‘painters manual’ of Dionysius of Fourna,
translated into English by P. Hetherington, The Sagittarius Press,
London 1974
9
R. J. Gettens, G. L. Stout, A monument of Byzantine wall painting:
The method of construction, Stud. Conserv., 3 (1958) 3, 107–119,
doi:10.2307/1504904
10
J. Plesters, Sancta Sophia, Trebizond – A note on the materials and
technique, Stud. Conserv., 8 (1963) 4, 131–135, doi:10.1179/
sic.1963.020
11
S. Daniilia, A. Tsakalof, K. Bairachtari, Y. Chryssoulakis, The
Byzantine wall paintings from the Protaton Church on Mount Athos,
Greece: tradition and science, J. Archaeol. Sci., 34 (2007) 12,
1971–1984, doi:10.1016/j.jas.2007.01.016
12
I. Holclajtner-Antunovi}, M. Stojanovi}-Mari}, D. Bajuk-Bogda-
novi}, R. @iki}, S. Uskokovi}-Markovi}, Multi-analytical study of
techniques and palettes of wall paintings of the monastery of @i~a,
Serbia, Spectrochim. Acta A., 156 (2016), 78–88, doi:10.1016/
j.saa.2015.11.031
13
N. Civici, M. Anastasioui, T. Zorba, K. M. Paraskevopoulos, T. Dilo,
F. Stamati, M. Arapi, Studying wall paintings in Berati Castle (Alba-
nia): Comparative examination of materials and techniques in XIV
th
and XVI
th
century churches, J. Cult. Herit., 9 (2008) 2, 207–213,
doi:10.1016/j.culher.2007.08.004
14
A. Hein, I. Karatasios, D. Mourelatos, Byzantine wall paintings from
Mani (Greece): Microanalytical investigation of pigments and plas-
ters, Anal. Bioanal. Chem., 395 (2009) 7, 2061–2071, doi:10.1007/
s00216-009-2967-6
15
G. Taglieri, D. Rigaglia, L. Arrizza, V. Daniele, L. Macera, G.
Rosatelli, G. Musolino, Microanalytical investigations on a
Byzantine fresco of the Dormitio Virginis from Sicily, J. Cult. Herit.,
40 (2019), 155–162, doi:10.1016/j.culher.2019.05.016
16
K. ªerifaki, Determination of Byzantine wall painting techniques in
western Anatolia, PhD Thesis, Ýzmir Institute of Technology, Ýzmir
2017
17
M. Singh, S. V. Kumar, S. A. Waghmare, P. D. Sabale, Aragon-
ite–vaterite–calcite: Polymorphs of CaCO3 in 7
th
century CE lime
plasters of Alampur group of temples, India, Constr. Build. Mater.,
112 (2016), 386–397, doi:10.1016/j.conbuildmat.2016.02.191
K. ªERIFAKI, H. BÖKE: PLASTER CHARACTERISTICS OF BYZANTINE WALL PAINTINGS IN WESTERN ANATOLIA
478 Materiali in tehnologije / Materials and technology 56 (2022) 5, 471–478