Multi-methodical study of Early Modern Age archaeological glazed ceramics from Prague

A series of scientific methods (X-ray analyses, optical and electron microscopy, Raman spectroscopy and thermal analyses) was used to research the ceramic bodies and glazes of forty vessels from Renaissance Prague. The set of Early Modern Age archaeological glazed ceramics contained tableware, kitchenware and technical ceramics. The main aim was to characterise ceramic materials and glazes used over two centuries. Attention was also paid to the identification of defects and corrosion products of the glazes. The research was conducted with an emphasis on the context of the original use of the ceramic artefacts and the environment of the waste pits from which they were excavated. The archaeological finds include three technical ceramic vessels (a rectifier, a bowl and a jar), which together could have formed a distillation apparatus. Since these vessels represent a rare finding, they were subjected to a detailed investigation. Based on the similarity of ceramic bodies and glazes, it was proved that the three technical ceramic vessels were made in the same workshop and were parts of one distillation apparatus. The results of the performed analyses determined that the studied vessels were manufactured from raw materials with a high content of a plastic component and that the raw materials did not change significantly during the period from the 15th to 18th centuries. The firing temperature of most of the vessels did not exceed 1000 °C. It was proved that all the glazes were medium- or high-lead content glazes and were coloured with ionic pigments.


Introduction
Glazed ceramics are often studied from the perspective of decoration techniques, the macroscopic description of materials or the identification of pigments. Present-day research is also focused on identifying fabrication centres of workshops and how they developed; moreover, emphasis is placed on the manufacturing process or firing conditions. It is of course problematic to create an overall view or conclusion about glazed production in Central Europe because these sets are randomly examined. The study of Early Modern ceramic production in the Czech Republic, represented mainly by archaeological finds, is one of the rapidly developing topics. This study is one of the first using archaeometric data to evaluate Early Modern glazed ceramics from Prague. Archaeological finds from the waste pits of Prague's palaces were used for this purpose.
Early Modern ceramics production in Prague shows that Bohemian pottery production fully reflected the era's broader Central European trends in terms of shape and is comparable with assemblages of artefacts found in Germany [1][2][3][4][5], Austria [6][7][8], Switzerland [9][10][11], Hungary [12] and Poland [13,14]. Any comparison with production from Slovakia is to some degree limited by the state of published assemblages. Generally speaking, in terms of ceramic production it can be said that during the Early Modern Age the Czech lands were part of developed Europe. The period from the 16th century to the first half of the 17th century is typified by the emergence and chiefly the increasing number of inner-glazed kitchenware and tableware. The main reason for the use Open Access *Correspondence: martina.kohoutkova@vscht.cz 2 Central Laboratories, University of Chemistry and Technology Prague, Technická 5, 166 28 Prague, Czech Republic Full list of author information is available at the end of the article of glazes in kitchenware was to prevent non-sintered ceramic bodies from absorbing water; the second reason was aesthetic. Starting in the last third of the 16th century, there was a gradual increase in pottery glazed on both sides-particularly among ceramic tableware. As in the rest of Europe, Bohemia witnessed an optimisation of ceramic forms. An individualisation of tableware is seen in the form of plates and other dishes intended for single servings. The artefact assemblages also feature numerous individual finds of ceramic items-primarily objects of a decorative character-that are not directly related to the consumption of food. A different example of the diversity of finds are technical ceramic vessels. The main aim of this work was to characterise ceramic bodies and glazes over two centuries and to find differences in the composition of both components (ceramic bodies and glazes). An important part of the study was focused on identification of the composition of glazes including colourants, their defects and corrosion products formed on the surface of glazes. Special attention was paid to investigation of technical ceramics with the aim, above all, to characterise their ceramic bodies and glazes and to ascertain whether these "chemical vessels" could be part of a single distillation apparatus (Fig. 1). An equally important part was the evaluation of historical glazes and their possible degradation. Scientific methods commonly used in archaeometry were used to meet the objectives-X-Ray analyses [15][16][17], optical microscopy and SEM technique in combination with X-Ray techniques [15,[18][19][20][21][22][23][24], thermal analyses [21,22] and Raman spectroscopy [25][26][27]. The main advantage of used techniques when they are used for the study of rare and precious finds is that the same samples can be investigated by a combination of these techniques.

Archaeological background
This study is based on finds from waste pits from the Prague Castle complex and the Salm Palace in Prague's Hradčany (Fig. 2a) district. These are predominantly finds discovered during the initial period of archaeological research in the 1920s and early 1930s. The archaeological excavations in Salm Palace were undertaken in 2009 and 2010. Finds from all waste pits cover time periods of varying length from the second half of the 15th to the first half of the 18th century. Each of these finds was subjected to a detailed analysis [28], on the basis of which it was dated. Concerning dating, the main emphasis was placed on the wider spatial context of the waste pits, as the dating disputableness of such findings cannot be overlooked. From historical reports we know that waste pits were often, and repeatedly, jumbled up or partially removed [29][30][31][32][33]. It is possible that chronologically older objects could become a part of the younger finding context. Furthermore, in general, "luxury" items have been part of living culture much longer than ordinary consumer ceramics.
Of the original set of 783 complete or substantially reconstructed objects, 40 vessels were selected for the  [34]; b Liber de arte distillandi; Das buch der rechten kunst zu distilieren. Hieronymus Brunschwygk (1450-1512) [35] present study based on their typological context, shape and colour of glaze. The analysed set of samples comprises pots, pipkins, jugs, bowls, cups, a mug, a distillation lid and a flowerpot. The finds studied here were evaluated as a single set with emphasis on dating the individual vessels. In order to follow technological changes in ceramics production in time, namely at the turn of the Early Modern and Modern Ages in Prague, three time groups were defined. The division of the three-hundredyear-long section into these three time groups is based on the historical development of Prague and the Czech lands. In the first time period (1450-1550), the Late Medieval traditions of ceramic production clearly disappeared. Glazed ceramics appear in the assemblages of found artefacts, but even with regard to written evidence regarding the restriction of trade with glazes, glazed ceramics were probably not a common, widespread article [36]. The text of the fourth article of the statutes of the Potters' Guild of Prague Old Town from 1535 states that there is a ban on "trafficking" with what were most likely scarce glaze materials without the knowledge of both of Prague's older guilds (Old Town and New Town). Specifically, this applied to lead and iron ore. Another document providing evidence of problems with the supply of the raw materials required to produce glazes is a letter from 1551, in which the Bohemian chamber of commerce approached the governor of the Křivoklat region with a request to consider the possibility of mining lead ore near Častonice. There is a reference to potters' clay in the seventh article of the statues from 1535, in which the Prague pottery masters call on the city councillors to intervene on their behalf with the abbot of Strahov, who is preventing them from mining clay in the vicinity of the monastery. When this is taken in combination with instructions from 1590 associated with collecting customs duties for goods brought into Old Town over the Prague Bridge, in which there is mention of "clay, when the potters' black is being brought from the tenancy", it is highly likely that, in the 16th century, Prague potters were using, among other things, the grey to dark grey, layers of carbonised vegetation remnants and even the coal that came with the Cenomanian clay and claystone mined in the Petřín area. In the middle of the 16th century (1550-1650), there was an increase in ceramic production in Prague and glazed ceramics started to become widely available. Vessels fired in a reducing atmosphere make up only a single-digit percentage of finds. In this period, pottery from finely levigated clay with a slightly grainy fabric was made in Prague. The thin-walled vessels were turned on a fast-rotating potter's wheel. The ceramic assemblages are again dominated by barrel-shaped pots whose maximum width is located in the middle of the body. The majority of pots are glazed on the inside. Flat forms are represented by pans on a tripod. Another form is deep bowls. The European-wide phenomenon of this period is slip ware. This pottery was brought to Prague from nearby Beroun. The use of painted decoration appears most frequently on shallow bowls and jugs of all sizes. From the perspective of the pottery assortment, it can be stated that there was an expansion in the formal spectrum of production. These are mainly shallow bowls with a flanged rim; plates and cups newly appear. Undoubtedly, new Prague inhabitants coming from Italy, Spain and Germany participated in the development of crafts. In the second half of the 16th century, these national minorities formed a significant and relatively large population of Prague [37]. The beginning of the third period (1650-1750) is marked by the end of the Thirty Years' War (1648). In the second half of the 17th century there were significant changes in the power organization of Europe. These changes were then reflected in all areas of life.
So the oldest group of findings is dated to the second half of the 15th and the first half of the 16th century. The second time group is defined by the second half of the 16th century to the first half of the 17th century. The youngest group of findings then comes from the second half of the 17th to the first half of the 18th century.
In addition to the usual kitchenware and tableware, three vessels which can be described as technical ceramics were analysed-a distillation lid (no. 2, Fig. 1), a jar (no. 7, Fig. 1) and a rectification bowl (no. 3, Fig. 1). In the case of technical ceramics, glazing helped to increase the resistance of the surface (the rectification lid and the bowl were glazed on the inside only, while the jar was glazed on both sides). Ceramic vessels that may have formed a part of chemical equipment are rarely found in archaeological collections [2,[38][39][40][41][42]. Most often they come in the form of shards and only rarely can researchers analyse entire vessels. The above-mentioned findings of technical ceramics were discovered in Waste Pit B near the Old Provost House, the oldest residential building at Prague Castle located in the central part of the castle complex, adjoining St. Vitus Cathedral on the southwest. Historians identify the Old Provost House with the so-called Episcopal Court which was first mentioned in archival documents as early as the 11th century. The list of capitular houses from 1486 [43] states that the house was divided into two parts during this period, with the archbishop using the western part and the provost inhabiting the eastern part. The building was not reunified until 1660 [44].
During the rescue archaeological research in July 1925, archaeologists explored Waste Pit B, which contained a rich collection of finds (ceramics, glass, iron) dating back to between 1450 and 1550 [45]. The assemblage was recorded [46]. Of the ceramic material found in the pit, 187 vessels were reconstructed, including three technical ceramic vessels which may have formed a distillation apparatus ( Fig. 2b; Liber de arte distillandi). The fact that these objects were located (and most likely used) in an ecclesiastical building makes the whole assemblage particularly interesting. The vessels were reconstructed immediately after the completion of the archaeological research in the area. The method and process used for the reconstruction were thoroughly recorded. Along with keeping a detailed survey log, the researchers also recorded the laboratory examination of the vessels [47][48][49]. The laboratory logs contain a detailed description of the daily process of the vessels' reconstruction and conservation, also specifying the binders used. In addition to these routine work records, the laboratory logs contain information about extraordinary events such as a doctor's visits, a funeral, the president's birthday [48] and the visit of an American archaeological expedition from Harvard University [49].
The set of artefacts from Waste Pit B is only one of nine assemblages from Prague Castle and Hradčany which were analysed in this study. Of 783 vessels, we focused on 40 (Table 1), which were chosen to help answer questions related to the way ceramics were produced during the Early Modern Age in Prague [50].

Experimental procedure
All analysed vessels come from waste pits from the Prague Castle complex and the Salm Palace. The preservation of all analysed objects in a similar environment is essential for their comparability. Samples for analysis were preferably selected from entire vessels of a similar type, in particular pipkins. The present investigation aims above all to characterise the ceramic bodies and glazes of 40 objects: 13 pipkins, 6 bowls, 10 pots, 3 cups, 2 plates and a jar, 2 jugs, a distillation lid, a flowerpot and a beer stein.
The studied finds were divided into three groups according to the probable time periods of their manufacture (Table 1). Detailed research was concentrated on three technical ceramic vessels, possibly parts of a unique distillation apparatus that was excavated from the different layers of Waste Pit B, and the same sequence of analysis was performed for all studied finds.
The average chemical composition of both the ceramic bodies and the glazes of all 40 vessels was determined by X-ray fluorescence analysis (XRF). The advantage of this method is that the measurement is taken from a relatively large sample area (in this case an area 2.7 cm in diameter), which leads to the suppression of errors due to possible inhomogeneities. The chemical composition of the ceramic bodies was determined from powdered samples.
The samples were ground in an agate mortar and pressed into boric acid pots.
An ARL 9400 XP + sequential WD-XRF spectrometer was used for the analysis and the acquired data were evaluated using the UniQuant 4 standardless software; the measured data were rounded based on the measurement error. The identified chemical compositions of all 40 vessels are listed in Table 2. The data for the rectification lid, the jar and the bowl were processed into PC diagrams (Fig. 3). Principal Component Analysis was performed using XLStat statistical software for Excel, which effectively summarized the XRF data according to quantitative variables (identified oxides). Subsequently, the XRF data were used to calculate normative minerals using the MINLITH program [51,52]. This program was originally created for sedimentary rocks but it is also used to suggest the possible raw materials used for the production of historical ceramics. In this work, the program was used to obtain an idea of the possible presence of clay minerals in raw materials because meta-clays, which are usually present in ceramic bodies, cannot be detected by XRD [53,54]. The calculated normative minerals and the input data for the evaluated "chemical" vessels are listed in Table 6. Chemical compositions of glazes were determined from the flat surfaces of ceramic shards using an Axios PANanalytical sequential XRF spectrometer; the data were evaluated using the Omnian standardless software. The glazes were analysed before and after chemical cleaning (using a 5% acid solution) of corrosion layers (Tables 4, 5). Mineralogical composition was identified by X-ray diffraction analysis (XRD) using an X'Pert PRO θ-θ powder diffractometer and CuKα radiation over the angular range of 5-70° (2θ). Data evaluation was performed using the Panalytical High Score Plus 4.0 software package. Semi-quantitative analysis was performed using RIR (Reference Intensity Ratio) values from the PDF4 + database of the HighScore Plus software. Mineralogical compositions of the glazes were determined from the surfaces of ceramic shards. In the case of ceramic body samples, powder samples were used for the analysis. Hydrothermally treated samples (after 100 h at 230 °C (2.77 MPa) in a Teflon lined autoclave) were measured over the angular range of 2-15° 2θ to examine for the presence of clay minerals.
Glaze colour was first evaluated visually and subsequently a spectrophotometric colouring measurement was used to interpret the effect of colouring oxides. The same method was employed to evaluate ceramic bodies in which the colour is influenced particularly by the presence of colouring compounds in raw materials (especially Fe and Ti compounds) and by firing conditions. The results of the colour measurement (using the CIE L*a*b* colour space) are presented in Table 7. A    Datacolor Mercury 2000 reflection spectrophotometer, working in the visible spectrum range, was used for the measurement.
Parameters of the ceramic body materials (bulk density, weight absorption, apparent porosity and apparent density) were determined by weighing in water (before and after the sample was boiled for 2 h, followed by storage in water for 24 h). The data for the rectification lid, the jar and the bowl are presented in Table 6 below.
Thin sections of selected archaeological samples were examined using an optical microscope [55] in polarized light, with or without crossed polarizers. Mineralogical composition and the size and shape of grains in the samples were identified by optical microscopy (OM) using Olympus BX60 and BX51 optical microscopes and NIS-Elements 2.03 software. Defects in the glazes were studied using an Olympus SZX9 stereomicroscope and an Olympus BX60 optical microscope.
Corrosion layers and pigments were evaluated using electron microscopy (TESCAN VEGA 3 LMU scanning electron microscope equipped with EDS analyser, OXFORD Instruments INCA 350 and JEOL JSM-6460 LA equipped with EDS analyser). The samples for SEM-EDS analyses were not polished due to the aim to analyse corrosion products on the surface of the glazes that could be destroyed by polishing.
The materials used for the glaze were identified by micro-Raman spectroscopy using a Nicolet DXR (Thermo Scientific) dispersive Raman microscope. The measurement principle used spot measurements taken directly on the particles in the glazes on the cross-section. Analyses were performed using a 780 nm diode laser, with maximum power of 5-7 mW and acquisition time of 5 min. Measurements were taken in the range between 3300 and 50 cm −1 . The spectral resolution was 4 cm −1 . The laser was focused on the sample through 50 × and 100 × objective lenses. Raman spectra were processed with the Omnic 9.0 program.
In order to identify clay minerals, the samples of ceramic bodies of the three technical ceramic vessels were exposed to hydrothermal treatment in Teflon lined autoclaves at 230 °C (2.77 MPa) for 100 h.  Table 2 presents the chemical composition of the ceramic bodies (determined by XRF) in the form of oxides. Figure 3 shows the principal components of the ceramic bodies of all the analysed archaeological finds and their distribution into groups by date of origin. The mineralogical compositions obtained by an evaluation of diffraction patterns of all samples of the analysed collection are shown in Table 3.

Evaluation of the ceramic bodies
The mineralogical compositions were also studied by optical microscopy. Thin section preparation is a destructive process and therefore could not be applied to all ceramic findings. 15 thin sections were prepared from the studied vessels, with each group represented by 5 thin sections. All evaluated samples from the oldest group of findings (nos. 2, 3, 5, 7 and 8) and one (no. 19) from the second time group-the second half of the 16th century to the first half of the 17th centurycontained significantly bimodal temper in the form of quartz and solitary feldspars. The relative abundance of inclusions in the samples, estimated by using comparative charts, is c. 10-20%. Inclusions in fine fractions (30-150 µm) are sub-angular to sub-round. Big inclusions of feldspar temper (Fig. 4a) are rare (under 1%). The matrix is non-calcareous with fine inclusions (about 10 µm in diameter) of quartz and muscovite and contains elongate voids. Some samples from the second time group (nos. 20, 21 and 22) and some from the youngest group of findings (nos. 29 and 35) show an optically active non-calcareous matrix rich in mica grains, and a sporadic mixing of two types of clays was identified (Fig. 4b). Most samples from the youngest group (nos. 27, 31 and 36) and one from the second time group (no. 18) show a relative abundance of inclusions (20%) and no big inclusions of temper (Fig. 4c) and iron pigmentation (nos. 18,27,36) in the non-calcareous matrix.

Evaluation of the glazes
The chemical compositions of the glazed surfaces were measured before and after the removal of corrosion crusts (Tables 4, 5, Figs. 5, 6). The analysed finds had been cleaned and washed with water during the first restoration in the 1920s, but the corrosion crusts were left.
Most of the degraded vessels from the set of studied samples were parts of a distillation apparatus (rectification bowl, distillation lid and collection jar), which is related to their use (Fig. 5, the red squares on the right of the chart). XRD analyses determined that the majority of the glazes (77%) did not contain any crystalline phase and that the other glazes contained mainly quartz and rarely cassiterite.

Distillation apparatus
The chemical composition of the three technical ceramic vessels (rectification lid no. 2, jar no. 7 and bowl no. 3) is shown in Fig. 2 (black circle) and Table 3 (in green). Figure 7 shows XRD patterns of the ceramic body of the bowl and the detail of the XRD pattern of the same sample after hydrothermal loading. XRD analyses of the hydrothermally treated sample demonstrates the presence of montmorillonite and kaolinite in the sample.
The normative minerals (Table 6) characterizing the probable raw material mixtures used for the vessels (with the simplifications mentioned in the Experimental Procedure) were calculated using the chemical composition determined by XRF. The table is supplemented with Optical microscopy was used to study the microstructure of selected vessels by evaluating thin sections (Fig. 8). Figure 8a shows the glaze layer on bowl no. 3 with a thickness ranging between 240 and 0 µm and signs of corrosion. Figure 8c, d show very slightly an "optically active" matrix with mica plates. Figure 8b represents the isometric grain of plagioclase feldspar, also detected by XRD analysis (Table 3). The meta-clays contained in the ceramic bodies became rehydrated/rehydroxylated as a result of their use and subsequent preservation in a waste pit. In the course of thermal analysis, their dehydration/dehydroxylation was monitored during the heat load [56,57]. Figure 9 shows the TG-DTA curves illustrating the dehydration (50-250 °C) and dehydroxylation (350-520 °C) processes in the samples.
The chemical composition of the glazes of the three technical ceramic vessels before and after the surfaces were cleaned was determined by XRF analysis (Figs. 5,  6). The chemical composition converted to molar percentage (main components) and the colouring of glazes, including CIE L* a* and b* values, are listed in Table 7.
The presence of corrosion products in the form of inorganic acid salts was confirmed by XRF analysis. The chemical composition of the surface of the glaze was measured before and after cleaning. It is evident that corrosion products were partially removed during purification as the content of P and Ca decreased significantly and conversely the content of Si increased. The molar percentages were calculated from values measured by XRF ( Table 7). The conversion to molar percentages is performed mainly to identify the function of lead oxide in the structure-whether it is involved in the formation of a glass mesh or not. Changes of conditions (waste pits, deposits) had a significant effect on the process of degradation and the kinetics of the corrosion process. Crystallization of corrosion products on the surface of the glazes enhanced the further leaching of cations from the glassy network [58]. Figure 10a, b show SEM images of green glazes with a thickness ranging between 40 and 130 μm and signs of degradation (especially cracking). In the case of jar no. 7, the inside glaze layer is about 100 μm thick (Fig. 10a) whereas the thickness of the outside glaze layer fluctuates around 60 μm. The corrosion products were identified using XRD as lead and calcium phosphates. Scanning electron microscopy also revealed the presence of quartz grains directly in the glaze layers (Fig. 10c).
The chemical composition of the glazes of the technical ceramic vessels was also determined using EDS analysis (Table 8), which demonstrated that the glazes with high lead content were coloured with ionic colourants (largely Cu). Figure 10e shows the corrosion crust in the form of calcium and lead phosphates, also detected using XRD. Chloride corrosion products were identified in the sample from the jar (Fig. 10f ). Sulphide corrosion products were detected in the crack above the quartz grain within the glaze of the bowl (Fig. 10g).
The green colouring in the lead glazes resulted from the combination of ionic colours-particularly Cu and Fe oxides. The glazes also contain a very small amount of Sb oxide (Tables 5, 7). SEM analysis also identified 0.5 wt% of Sn in the glaze of the bowl. Micro-Raman spectroscopy played an important part in studying the material composition of the glazes. The signal at 468-465 cm −1 , associated with peaks at 201-204, 264-267, 356-358 and 396-398 cm −1 , is due to the presence of quartz, SiO 2 (Fig. 11), which was detected in all the investigated samples from the lid, the jar and the bowls [59]. All samples show a strong peak at 146 cm −1 and weak peaks at 205, 394, 514 and 637 cm −1 , attributed to the presence of titanium oxide-anatase. Characteristic bands of rutile were not found. Raman signals of yellow grains in green glazes at 130-132 cm −1 may be attributed to Pb-Sn-based compounds known as leadtin yellow (Pb 2 SnO 4 or PbSn 1-x SixO 3 ) or to Pb-Sb based compounds known as Naples yellow (Pb(SbO 3 ) 2 or Pb(SbO 4 ) 2 ) [27,60]. The strongest peak characteristic of lead-tin yellow is found at 130-132 cm −1 . The identification of Naples yellow is complicated due to the presence of anatase, as their principal peaks (approximately 145 cm −1 ) overlap. In all the samples, the spectra of the green glaze grains show a broad band around 900-970 cm −1 . The presence of copper in the green lead glazes was not confirmed by Raman spectroscopy, as copper can be dispersed in ionic form with no specific vibration in Raman spectra [61].

Evaluation of the ceramic bodies
The binary diagram SiO 2 -Al 2 O 3 of ceramic bodies, using the XRF data listed in Table 2 grouped by dating (Fig. 12), shows that the first group of the earliest vessels (red) has the smallest differences in composition.
The majority of the ceramic materials from the group of the earliest vessels were tempered using bimodal temper with large inclusions of quartz and feldspars. The ceramic bodies of these vessels probably contain a higher amount of a non-plastic component than most of the ceramic bodies from the other two groups.  The presence of a high-temperature phase mullite shows that the firing temperature of some vessels reached, or slightly exceeded, 1000 °C. Such vessels fired at a higher temperature were identified in all of the studied chronological groups (Fig. 13).

Evaluation of the glazes and their degradation
A chemical analysis of the glazes proved that the studied samples were medium-or high-lead content glazes with a PbO content of 35-67 wt%. The lowest PbO content was identified in the samples taken from the distillation lid (no. 2) and the flowerpot (no. 21). Their lower lead content is caused by the leaching of Pb during their use. Corrosion products, especially in a form of lead compounds, were identified on the surfaces of the samples. Figure 14 shows the colours of the ceramic bodies and the glazes, including colouring agents (determined using the XRF results listed in Table 5), corrosion products (determined using XRD) and identified defects. The majority of the glazes (77%) did not contain any crystalline phase (according to XRD). The other glazes contained mainly quartz. Cassiterite-one of the oldest known opacifiers-was identified (by XRD) in the glaze of four vessels with black glazes (pot no. 12, pipkins nos. 30 and 35 and small bowl no. 34) [62]. The presence of kaolinite and gypsum on the glaze surfaces was caused

Distillation apparatus
Special attention was paid to the analyses of the technical ceramic vessels (rectification lid no. 2, jar no. 7 and bowl no. 3). The aim was to prove that these vessels were products of the same workshop and could have originally formed one distillation apparatus. The chemical composition of the three vessels is very similar (Fig. 4, black circle, and Table 4, in green). As far as the composition of the ceramic bodies is concerned, only small differences occur in the proportion of CaO, MgO, and K 2 O contents, demonstrating an overall homogeneity of the raw material. The results of the evaluation of thin sections (Fig. 6) are consistent with XRD (Table 4) and XRF (Tables 2, 3) results. All three vessels exhibit a similar microstructure, matrix and size, as well as a similar type of non-plastic material (particularly silica grains). Occasionally, mica (muscovite), K-feldspars and Fe-pigmentation were also observed. The greatest differences between the XRD results and the calculated mineralogical composition are in the content of individual clay minerals (Table 6). Clay minerals convert to amorphous meta-clays at temperatures between 500 and 1000 °C and so they cannot be detected by XRD. At temperatures above 900 °C, they convert to defective spinel and then to mullite. Traces of mullite were identified only in the jar sample. A very slightly "optically active" matrix with lamellar mica grains was observed in the bowl sample. We may assume that the vessels were fired at temperatures between 850 and 980 °C as evidenced by the very slightly "optically active" matrix and the absence of rutile (anatase transforms into rutile at temperatures over 920 °C). Similarities in ceramic body parameters (Table 6, right hand side) of all three vessels indicate that the vessels were fired within approximately the same temperature range. The results of the thermal analyses are consistent with the vessels' mineralogical composition determined by XRD and optical microscopy. The largest amount of water was released from bowl no. 3 (Fig. 7a, red line). We may assume that this is related to the higher content of kaolinite clay (the characteristic dehydration/ dehydroxylation and formation of spinel when kaolinitic clay is fired, Fig. 7a, red line) and the highest temperature load (approx. 100 °C) during its usage in comparison to the other two vessels [63]. The lowest degree of rehydroxylation was observed in the jar (Fig. 7a, brown line). In this case, this is probably due to the intensive thermal load during firing, as XRD proved traces of mullite in its ceramic body. DTA results (characteristic dehydration/ dehydroxylation and the formation of spinel during the clay calcination, see Fig. 7a, brown line) confirmed this assumption.
The green glazes, present on all of the examined vessels, are uneven with thicknesses ranging between 40 and 200 μm (Figs. 8a, 10a, b) and show significant signs of degradation that was probably caused not only by their preservation in a waste pit but also by their usage. The corrosion products on the green glazes were identified by XRD as lead and calcium phosphates, both usually originating in waste pits from kitchen waste, such as bones. The occurrence of kaolinite and gypsum on the glaze surfaces was caused by secondary contamination from the previous restoration work, or from the surrounding environment. We assume that in a composition range of up to ca 30 mol%, lead cations tend to act as network modifiers. Lead ions behave similarly as alkali metal cations and form an ionic bond with non-bridging oxygens at up to about 30 mol% of PbO. The chemical characteristics of the Pb-O bond change from an ionic character to be more covalent when the PbO content increases above 30 mol% to better suit the needs of the glass formation of an oxide glass network. When the PbO content is increased to over 60 mol%, lead forms its own glassy structure, due to the properties of lead ions to produce a glassy network over a wide range of concentrations of up to more than 80 mol% PbO, if the melt is cooled rapidly [64][65][66]. Pb ions leached from the glazed surfaces are therefore involved in the formation of the corrosion products detected by SEM. X-ray fluorescence analysis of the cleaned glazes confirmed that the surface of the glazes was covered by corrosion products containing lead. Other observed defects were of a different type. Large grains of quartz in the raw material caused cracking in the glaze resulting from volume changes accompanying the transformation of α → β quartz during the process of cooling after a vessel was fired. The varying PbO content indicates the degree of corrosion in the lead component, depending on how long the vessel was used and exposed to the corrosive environment: vapour from distilled solutions in the case of the lids, boiling liquid in the case of the bowl and hot distillate in the case of the jar.  Table 8 Conclusion Forty ceramic bodies and glazes from three hundred years of ceramic production were examined. This study analysed items of ordinary ceramic kitchenware, comprising chiefly cooking pots and tripod cooking vessels. However, the analysed set also included several items of tableware in the form of shallow painted bowls, one flowerpot, and one three-legged bowl glazed on both sides. As far as the representation of individual shapes is concerned, the assemblage is standard and corresponds to the ceramic finds dating to the Modern Age urban environment in the Czech lands. In general, it applies that most ceramic assemblages are dominated by barrel-shaped pots glazed only on the inner side. Deep bowls and cooking vessels make up only a single-digit percentage of discoveries. Tableware is represented by jugs and shallow bowls. Other forms of ceramics are usually represented by random solitary pieces. On the case of Prague, it is possible to compare  finds from Republic Square (Náměstí Republiky) in Prague's Old Town [67], from the Theatine Palace in Lesser Town [68] and the waste pit in Prague's New Town [29]. Similar conclusions have also been reached during a comparison of the ceramic assortment found in South Bohemia [69]. In this particular assemblage of found artefacts, a unique position is held by three vessels associated with technical/chemical ceramics. They are exceptional particularly with regard to the condition in which they have been preserved. These types of items appear rarely in archaeological find assemblages and, as a rule, are usually in fragments. It was proved that the studied vessels were manufactured from similar raw materials free of calcite admixtures. The ceramic materials were tempered mostly by quartz. The results of chemical and mineralogical analyses of the ceramic bodies showed that the raw materials that were used did not significantly change from the 15th to the 18th centuries. Differences were observed only in the group of the oldest vessels, with a clay matrix containing very poorly sorted inclusions. The grain size distribution was significantly bimodal. Most of the samples from the second and third group did not contain large inclusions, and some contained more mica in the matrix. The orientation of all the samples showed the vessels were produced by rotation. The firing temperatures were in the range of about 850-1050 °C. The range is determined by the degree of the dehydroxylation of mica, the "optical activity" or "inactivity" of the matrix, and the absence or presence of mullite. In the case of the glazes, it was proved that the studied samples were medium-or high-lead content glazes without any significant differences over the centuries. The distinction between the oldest vessels (dating to 1450-1550) and the younger ones is probably due to an incident in 1541, when a large fire occurred at the site of the findings (Prague's Hradčany district). Because of the fire, it is possible that a large quantity of utility ceramics were replaced in Hradčany households In the second period (1550-1650), the ceramic craft significantly developed and a substantial number of kitchen vessels were replaced by newly manufactured items. During the Thirty Years' War and the subsequent period (1650-1750), it can be assumed that the production processes from the time of the development of the craft were preserved. This is probably the reason for the similarity of the vessels from the second and third groups and their difference from the vessels from the first-oldestgroup. In the case of the glazes, it was proved that the studied samples were medium-or high-lead content glazes, again without any significant differences over the centuries. Three technical ceramic vessels showed a similar type of ceramic body and glaze. Based on the results of the performed analyses, we may assume that these vessels were made from similar raw materials in the same workshop and that they were part of one distillation apparatus. All three vessels show pronounced degradation caused by their usage and their subsequent preservation in a waste pit. The vessels had to sustain contact with hot and rather acidic liquids, which is a