The uniqueness of the Linevo-1 monument in terms of studying pottery lies in the fact that clay and molding masses for making vessels were found here. This made it possible to trace the changes that occur in them during thermal exposure. The results of petrographic, X-ray phase and thermal studies of vessels are published for the first time. The algorithm of derivatogravimetric measurements, data processing and interpretation is described in sufficient detail. The possibilities of applying physico-chemical methods to the study of ceramics in general are discussed. As an alternative to the generally accepted approach to using the results of thermal analysis to detect the firing temperature, another one is proposed, in which a comparative analysis of the preservation of clay components in the molding masses of ceramics is important to determine the quality of firing vessels from different monuments and even different parts of the same product.

Keywords: ceramics, methods of natural sciences, thermal analysis, quality of firing.

Introduction

This work continues a series of publications on the physical and chemical study of ancient ceramics [Drebushchak V. A., Mylnikova, Drebushchak T. N. et al., 2003a, b; Drebushchak V. A., Mylnikova, Drebushchak T. V., Boldyrev, 2003; Drebushchak V. A., Mylnikova, Drebushchak T. N., 2006; Drebushchak V. A., Mylnikova, Drebushchak T. N. et al., 2006; Molodin, Mylnikova, 2003, 2004; Molodin, Mylnikova, Parzinger, Shneevais, 2003; Drebushchak V. A., Mylnikova, Drebushchak T. N., Boldyrev, 2005; Drebushchak, Mylnikova, Molodin, 2007; et al.]. The results of such studies are usually presented in the following sections: They are accompanied by conclusions about the firing temperature, sometimes about the oxidative or reductive nature of the firing atmosphere, and subsequent arguments about the possible interpretation of the information obtained [Grebenshchikov, Derevyanko,

The work was carried out within the framework of the Integration program, project No. 25 "Cultural variability on the monuments of the Urals and Western Siberia in the Bronze Age - Early Iron Age", and the RGNF project No. 09-01-00398a "Specialization of settlements in the forest-steppe zone of Western Siberia at the turn of the Bronze and Early Iron Ages: a comparative analysis experience".

page 60
2001; Lamina, Lotova, Dobretsov, 1995; Mylnikova. 1999]. During the implementation of the integration project in 2003-2005, we came to the conclusion that there are no sufficient physical and chemical justifications for such conclusions [Drebushchak V. N., Mylnikova. Drebushchak T. N. et al., 2006]. Only the temperature cannot be determined unambiguously from the products of thermal transformations, since all reactions of this kind are kinetic in nature, and so far no one has been able to determine the temperature and firing time simultaneously (and it is unlikely that it will be possible in the near future). This conclusion is related not to the imperfection of existing methods of physical and chemical research, but to the fundamental laws of substance change during thermal transformations [Drebushchak V. A., Mylnikova, Drebushchak T. N., Boldyrev, 2005]. However, it was found that accurate quantitative determinations of the mass loss of the sample during calcination (derivatogravimetric analysis) and the results of X-ray phase analysis allow us to reliably establish differences in the amount of clay minerals and thinners in the molding masses. This opens up prospects for reconstructing the features of ceramic manufacturing technology [Drebushchak, Mylnikova, and Molodin, 2007]. We abandoned the concept of" firing temperature", the algorithm for determining which differs among authors who use different methods of thermal analysis (thermogravimetric, differential thermal, thermomechanical, differential scanning calorimetry), which makes it difficult to compare the results with each other, and proposed the term "degree of thermal transformations". We measure the mass loss of a ceramic sample when calcined to 850°C. The values of mass loss in the temperature ranges 22-350 and 350-600°C allow us to determine the position of the test sample on the diagram of the degree of preservation of the clay component and evaluate the quality of its firing relative to other samples [Drebushchak V. N., Mylnikova, Drebushchak T. N. et al., 2006].

In this paper, the results of studying ceramics from the Linevo-1 monument by thermogravimetry and X-ray powder diffraction methods are published. Samples for research were taken mainly from full vessels, from each of them three or four from different parts (bottom, body, shoulder, corolla - part of the neck with the edge of the vessel), which made it possible to conduct a comparative analysis of molding masses and the degree of thermal transformations of the clay part in different parts of the product. In addition to information about the results of the study, the algorithm of derivatogravimetric measurements (DTG), data processing and interpretation is described in sufficient detail.

Monument and samples

The settlement of Linevo-1 has been known in the archaeological literature since the mid-1980s. The monument is located 2 km northeast of the village of Zarechnoye, Toguchinsky district, Novosibirsk region, on the southern bank of Lake Linevo, staritsa of the Pni River. The terrace at this point forms a promontory, housing depressions were 1.5-3.0 m above the current level of the lake and 5-6 m above the level of the Pni river. The monument was opened by V. A. Zach. He also uncovered 500 m2 of the settlement area, where two dwellings were excavated [Zakh, 1986, 1997], one of which is attributed to the Irmen culture, the other to the Zavyalovo culture, to its early Linev stage (the transition time from the Bronze to the Iron Age) [Zakh, 1997, p. 93; Troitskaya,Zakh, 1997, p. 93]. Sidorov, 1989, p. 104]. For the second period, a date was named-the end of the IX-beginning of the VIII century BC [Zakh, 1997, p. 92].V. V. Bobrov did not agree with this attribution of the complex [1995, 1999]. Some researchers considered it possible to attribute monuments of this type to one of the variants of the Bolynerechenskaya culture of the transition period of the Upper Ob region [Kosarev, 1981, p. 202; 1987, p. 302; Chlenova, 1994, p. 84; Mogilnikov, 1986, p.30-31]. Large-scale works in 2003-2005 (2,454 m2 of continuous area; fig. 1, 2) allowed us to look at Linevo-1 in a new way. Now it is obvious that there is no reason to assign housing complexes to different chronological periods, all the buildings are simultaneous and date from the transition from Bronze to iron time [Mylnikova, Durakov, Mzhelskaya et al., 2003, 2005; Mylnikova, Durakov, Mzhelskaya, Kobeleva, 2004]. Analysis of the inventory, stratigraphy and planigraphy of the monument allows us to date it to the VIII-VI centuries BC.

Ceramic materials obtained in the course of 2003-2005 can be divided into groups according to the manufacturing technology and ornamentation (Molodin and Mylnikova, 2005) (Fig. 3), among which the late Irmennian occupies a dominant position. Binocular microscopy and petrographic analysis revealed several recipes for molding masses: clay + rock fragments + chamotte; clay + rock fragments; clay + chamotte with signs of organic matter; clay.

The unique feature of the Linevo-1 monument in terms of studying pottery is that clay and molding masses for making vessels were found here (in the work they are called "clays"). Portion-a red clay blank (pile size 0. 2x0. 2 m) found in sq. U-F / 10-11 of dwelling 17. In the filling of the pit of dwelling 16, on the border of the hearth, in sq. R/26, the lower part of the vessel filled with clay was found. Nearby was a pile of similar clay. Another one is fixed in the interhospital space (between the-

page 61

Figure 1. Excavation plan for 2003-2005. a -clay; b-object; c-sample numbers.

between pits 16a and 15), in sq. M / 29, on the edge of the pit (see Fig. 1). This made it possible to study the molding masses of ceramics, as well as clays in the pre-firing state, and to trace the changes that occur in them under thermal influence.

Methodology

Measurements of mass loss during calcination from room temperature to 850°C with a heating rate of 20°C / min were carried out in an argon current (20 ml / min) on TG-209 Netzsch thermal scales. Samples weighing 47.00 ± 0.15 mg were placed in gold crucibles. Several measurements of the empty crucible were made. The reproducibility of mass values at heating 10 micrograms to 500°C was no worse than at calcination 20 micrograms to 850°C.

The ceramic phase composition was measured using a Bruker D8 GADDS diffractometer with a two-coordinate Hi-Star detector. CuKa radiation, graphite monochromator, 0.5 mm collimator. 40 kV and 40 mA high-voltage generator mode.

page 62
Reflection angle range 5-65° (three frames). The distance from the sample to the detector is 25 cm. A range of angles slightly more than 20° was captured per frame (for crosslinking diffractograms).

Results

Samples of clays for making ceramics. The results of thermogravimetric measurements (TG) show significant differences between the samples in weight loss during calcination to 850°C (Fig. 4, A): N 1 (dwelling 17) - 14.1 %, N 2 (inter - core space) - 7.5, N 3 (dwelling 16) - less than 2 %. Calcite is absent in the clays. 4 (B) clearly shows the peaks of dehydration (near 100°C) and hydroxyl decomposition (near 500°C). The main part of water is removed from the sample at the first stage, and at its completion, a portion is released, which is observed on the derivative of mass loss as a "shoulder" (clearly visible on the curve of sample N 2). This two-step nature is observed in some varieties of clays, and the ratio of mass loss at two stages depends on the cationic composition.

The physical and chemical principles of the location of points on the diagram of the degree of preservation of the clay component are described in detail [Drebushchak V. N., Mylnikova, Drebushchak T. N. et al., 2006]. Briefly recall that the horizontal axis corresponds to the mass loss of the sample in the temperature range 22-350°C (dehydration), the vertical axis - in the range 350-600°C (decomposition of hydroxyls). The lower the amount of mineral admixture in the molding mass and the higher the content of pure clay (including chamotte), the more the points on the diagram move up to the right. When lightly fired, some of the hydroxyl decomposes, increasing the porosity of the clay. This increases the water content and reduces the amount of hydroxyls in the ceramic. The more intense the low-temperature firing, the more the points on the diagram shift to the right and down. When firing at high temperature, when hydroxyls are already completely removed, thermal transformations in clay are reduced to its sintering, closing of pores. The final product of such firing is a glassy mass with a low moisture capacity. The higher the temperature and the more intense the firing, the more the points on the diagram move to the left and down.

4, the points of clay samples from the Linevo-1 settlement are located far from each other and significantly below the line obtained from measurements of clay samples from different places on the Chicha-1 monument (Drebushchak V. A., Drebushchak T. N., Molodin et al., 2005). The reason for this discrepancy is the composition of the starting materials. Without going into the details of the classification of clay minerals and their properties.

2. Stratigraphic section along the F / 52 - F / 4 line.

a-turf; b-black humus layer; c-black layer with loam; d - brown golden; e-gray golden; e-the same with coals; g-yellow loam mixed with black soil; h - soot layer with coals; i-black ashy soil.

page 63

3. Ceramics from the Linevo-1 settlement.

1-5 - group I-Irmen culture; 6, 9, 10-group II - with cross-comb-stream ornamentation; 7,8 , 11-13-group III - with ornamental elements of Molchanov culture; 14-group IV - with features of Early Iron Age ceramics; 15,16 -group V-with features of samodelkin-type tableware.

page 64

4. Results of analysis of clay samples.

A - TG curves: mass loss; B-DTG curves: mass loss rate (the arrow indicates the "shoulder" at the peak of dehydration); C-diagram of the degree of preservation of clay minerals (contour circles represent all the samples under study); D - DTG curves of samples N 1 and 2 and calculation of the correction of the composition of sample N 1 by diagram of the degree of safety (for explanations, see the text).

to discuss the question of the ratio of water and hydroxyls, you can greatly simplify the description of the chemical composition and use an approximate formula

∑Mi+∑Mj ++[AlxSi5,5-xO10(OH)2]*nH 2O,

where M I and M j are monovalent and divalent cations, respectively. The number of water molecules in clay minerals depends on temperature, humidity, and cationic composition. If the samples are kept under the same conditions (temperature and humidity), the amount of water in them will depend only on the latter. The introduction of thinners into clays with different cation compositions will lead to the fact that on the diagram of the degree of preservation of the clay component, the points of the samples will be located along lines with different slopes.

Sample N1 is represented by two dots on the diagram (Fig. 4, C). The lower one corresponds to the values obtained strictly in accordance with the algorithm described above (m1- mass loss in the range of 22-350°C, m2- in the range of 350-600°C). It definitely falls out of the "otoshchiteli" line. 4, D: the peaks of mass loss of samples N 2 and 3 in the range of 350-600°C are combined, and the peak of dehydroxylation for N 1 is very large, the mass loss is significant, and since clay dehydroxylation is a kinetic process [Rouxhet, 1970; Nahdi, Perrin, Pijolat et al., 2002], then part of the peak goes beyond the temperature range conventionally assumed to be the temperature limits of hydroxyl decomposition. For most clay and ceramic samples, the peak of dehydroxylation is usually located slightly lower, and the peak of mass loss at this stage completely falls within the selected range of 350-600°C. We encountered an unexpected case of distortion of the results due to the conditional division of mass loss into two temperature ranges. Therefore, for the peak of dehydroxylation above 600°C (shaded part in Fig.D) the mass loss was calculated (it turned out to be 1.07%) and added to the value obtained for the range 350-600°C. The amount of mass loss due to hydroxyls increased, and the point on the diagram (Fig. 4, C) shifted up (shown by the arrow) and almost coincided with the "otoshchiteli"line.

X-ray studies have shown that the phase composition of samples N 1 - 3 is identical (Fig. 5, A). All reflexes can be attributed to quartz and clay minerals (muscovite, illite, etc.). They are reproduced on all three diffractograms, no traces of the presence of sample N 3, the point of which in Fig. 4, B It is located at the origin, and no additional crystal phase is found that is absent in the other two. That is, we can conclude that sample N 1 is pure clay, N 2 and 3 are molding mass. The additive in sample N 3 is burnt clay-

page 65

on-chamotte, which was confirmed by petrographic analysis; sample N 2 also contains chamotte, but it is 2 times less than in the previous one.

Samples of ceramics from dwelling No. 15. One vessel (N 4) was examined. Three fragments from different parts (bottom, trunk and corolla) were taken from it. Thermogravimetric measurements showed that the corolla and body samples have very close mass loss values, both total-5.59 and 5.43%, and in the temperature range from 600 to 850°C - 0.85 and 0.76 % (Fig. This corresponds to the reproducibility of the measurement results with small variations in the composition due to imperfection of promes and sampling. However, in the temperature ranges of 22-350 and 350-600°C, significant differences in weight loss are observed: the corolla sample - 2.98 and 1.76 %, respectively, and the body - 3.29 and 1.38 %. The former loses less water than the latter, but more hydroxyl. If you combine these two intervals into one, the total weight loss is 4.74 % for the corolla and 4.67 % for the trunk. The values are again very close. Thus, we have a redistribution of mass loss between dehydration and dehydroxylation. This variation of changes in the mass loss of samples from different parts of the vessel corresponds to differences in the quality of firing [Drebushchak, Mylnikova, and Molodin, 2007]. The total mass loss of the bottom sample was 9.0 %, which is almost 2 times more than that of the trunk and corolla samples. The biggest difference occurs in the range of 22-350°C, it is less at a temperature of 350-600°C. In the range of 600 - 850°C, the differences in weight loss between the bottom, body and corolla are insignificant.

page 66

5. Diffraction patterns of clay and ceramic samples.

A - samples of clays N 1 - 3; B-d - samples vessel N 4: B - beater (1), the body (2), bottom (3), In - the same on a larger scale in the range of angles 17 - 30° (arrow denotes the peak at 28°), G - Corolla and sample clay N 1 (arrows point to the reflexes, the clay, but missing ceramics); D - samples halo (1) and body (2) vessel N 5 (arrows point to the reflexes related to mica and feldspars); E - samples of the Corolla (1), the body (2) and bottom (3) of the vessel N 6 (arrows point to the reflexes related to feldspars); W - samples Corolla (1), shoulder (2), of the body (3) vessel N 8 and clay N 1; Z - samples Corolla (1), shoulder (2), bottom (3) of the vessel N 9 (arrows point to the peaks of atomically) and the body of the vessel N 8 (without ataxites) (4); And samples of the Corolla (1), shoulder (2), the body (3) vessel N 10 (arrows point to the peaks of atomically), tulov vessels N 9 (high content of ataxites) (4) and 8 (without ataxites) (5); K - sample a whisk (7), the body (2) and bottom (3) of the vessel N 11; L - samples of the whisk (7), shoulder (2), the body (5) and bottom (4) vessel N 12; M-samples of haloes vessels N 7 (1), 13 (2) and clay N 2 (3).

The points of samples of different parts of vessel No. 4 on the diagram of the degree of preservation of the clay component lay down so that the difference in the composition of the molding masses used to make the bottom, on the one hand, and the body and corolla, on the other, becomes obvious (Fig. 6, B): in the molding mass of the bottom, there are almost 2 times The points of the body and corolla samples are located along the "firing" line and indicate large thermal changes in the clay of the molding mass of the body compared to the corolla.

page 67

6. Results of analysis of vessel samples from the dwelling 15.

A - curves of TG (a) and DTG (b); B-diagram of the degree of preservation of clay minerals (contour circles represent all samples under study).

1-corolla, 2-body, 3-bottom.

The results of thermogravimetric analysis allow us to draw the following conclusions: 1) the bottom of the vessel was made of a molding mass with a large (approximately twice) clay content compared to the body and whisk. In this case, this can not be an accident, for example, the result of insufficient promesa; 2) the body was made of molding mass of the same composition as the corolla, but burned better (the result of the position of the vessel in the firing device).

The X-ray diffraction patterns of the three N4 vessel samples are similar (see Fig. 5, B). All peaks are located at the same reflection angles and have almost the same intensity. The phase composition of clay and thinners in all three samples is identical. The greatest difference in intensity is observed at an angle of 28°: for the corolla sample, the peak is significantly higher than for the trunk and bottom (see Fig. In this connection, it is interesting to compare the diffractograms of samples of this vessel and clays (see Fig. 5, A). The intensity of the reflex at an angle of 28° for clay sample N 3 is significantly lower than for the other two, as is the degree of preservation of the clay component. Thus, there is a correlation between the degree of clay preservation and the intensity of the reflex at an angle of 28°. We are not talking about the amount of clay component, but about the degree of preservation of its sorption properties. This observation does not yet allow us to draw any reasonable conclusions, but the issue can be put up for discussion.

When comparing the diffraction patterns of clay samples N 1 and the corolla of the vessel N 4 (see Fig. 5, D), it is clear that the latter does not have any additional peaks. On the contrary, at least three broad reflexes (at angles of 12.5, 35, and 62°) demonstrated by the clay sample are absent from the ceramic diffraction pattern. These reflexes belong to clay minerals, and it is quite natural that they disappear after firing. Since, as already explained above, the clay samples contain chamotte as a thinning agent, it can be concluded that it is also present in the molding mass of vessel No. 4. This explains why the trunk and corolla samples with approximately twice the amount of sand or rock debris as the bottom sample have the same diffraction patterns. The diffractograms of chamotte and clay differ precisely in the very reflexes that disappear after firing the latter.

Ceramic samples from dwelling 16. Fragments of two vessels (N 5 and 6) were examined. The results of thermogravimetric measurements show that the samples of the corolla and body of vessel No. 5 differ insignificantly in total weight loss, at the level of 1 % (Fig. 7, A). Such differences may be due to the heterogeneity of the molding masses, variations in the ratio of thinners and clay minerals. In the diagram of the degree of preservation of the clay component (Fig. 7, B), the points corresponding to the corolla and body lie quite far from the "otoshchiteli" line, indicating significant thermal transformations. The vessel was qualitatively burned, especially the body.

Diffractograms of the corolla and trunk samples of vessel No. 5 (see Fig. 5, E) are similar to each other, contain the same number of reflections at the same reflection angles, i.e. the molding masses are identical. However, if we compare it with the diffractograms of clay samples and vessel N 4 (see Figs. 5, A, B), we will see additional reflections at angles of 8,9; 17,8; 22,1; 23,6; 24,3° and much more intense-at angles of 27.5, 28 and 30.5°. The first two reflexes (at 8.9 and 17.8° C) probably belong to mica (mica in these samples is visible to the naked eye), the rest - to feldspar (plagioclase), which are part of the otoschitel.

Samples of vessel No. 6 do not differ significantly in total weight loss, at the level of 1 % (see Fig. 7, C). The peak of hydroxyl loss is clearly visible on all three gra-

page 68

7. Results of analysis of ceramic samples from the dwelling 16.

A - curves of TG (a) and DTG (b) of vessel samples N 5; B-diagram of the degree of preservation of clay minerals; C - curves of TG (a) and DTG (b) of vessel samples N 6. 1 - corolla, 2 - body, 3 - bottom.

figs, but on the curve of the Tulov sample it is very large, which is typical for ceramics with weak heat treatment. In the diagram of the degree of preservation of the clay component (see Fig. 7, B), the points corresponding to the corolla and body of vessel No. 6 lie very close to the "otoshchiteli" line, confirming their weak heat treatment. The bottom was exposed to a stronger thermal effect. Whether this is the result of firing the vessel or its operation, it is impossible to determine from the available data. The TG curve of the corolla sample shows a peak at 750°C, characteristic of calcite. The mass loss here is about 0.4 %, which corresponds to a mineral impurity of less than 1 % (mass fraction).

The diffractograms of the N6 vessel samples are similar (see Fig. 5, D) are distinguished by the absence of mica reflexes at small angles (8.9 and 17.8°). Obviously, in this case, only plagioclase was used as a thinning agent for the molding mass.

Samples of ceramics from the dwelling 17. Seven vessels were examined, from which different numbers of samples were taken (N 7-corolla; N 8-corolla, shoulder, trunk; N 9 - corolla, shoulder, bottom; N 10-corolla, shoulder, trunk; N 11-corolla, trunk, bottom; N 12 - corolla, shoulder, trunk, bottom; N 13-corolla).

The difference in mass loss between the body and shoulder samples of vessel No. 8 is very small (Fig. 8, A), within the measurement error range (Drebushchak, Mylnikova, and Molodin, 2007). Their curves actually merge into one. Differences in the corolla sample measurement results against the background of data for other vessels should also be considered very small. Total mass loss for veins-

8. Results of analysis of samples of vessel No. 8 from dwelling 17.

A - curves of TG (a) and DTG (b); B-diagram of the degree of preservation of clay minerals.

1-corolla, 2-shoulder, 3-torso.

page 69
chik, tuloyu and shoulder are respectively 14.66; 15.00 and 15.03 %. Recall that the difference of less than 0.1 % at any of the temperature ranges (22-350, 350 - 600 and 600 - 850°C) between the samples of ancient ceramics is considered unreliable. The peak of hydroxyl loss is clearly visible in all three graphs, but it is slightly larger on the corolla sample curve, and the peak of mass loss is smaller. This ratio between dehydration and dehydroxylation indicates that the corolla was burned slightly weaker than the body and shoulder. This is confirmed by the diagram of the degree of preservation of the clay component (Fig. 8, B): the point corresponding to the corolla lies closer to the "otoshchiteli" line than the points of the body and shoulder samples.

There are no significant differences between the diffractograms of the N8 vessel samples (see Fig. When compared with the diffractogram of a clay sample N 1 from the same dwelling, differences are found in three peaks (in Fig. 5, F are marked with arrows): reflexes at angles of 12.6 and 28° for the clay sample and at 33.3° for ceramic samples. In discussing diffractograms of N1 - 3 clay samples, we have already drawn attention to the reflex near the angle of 28°, which correlates with the content of unburned clay minerals in a mixture of clay and chamotte. In this case, when comparing the unburned molding mass and the vessel N 8 from dwelling 17, we again encounter the disappearance of the reflex at an angle of 28°. Based on the results of X-ray and thermogravimetric measurements, it can be concluded that vessel No. 8 was made of pure clay, without the use of any otoshchiteli, even chamotte, which is confirmed by binocular and petrographic studies.

Between the samples of vessel No. 9, the differences in mass loss when heated to 600°C are very small (Fig. 9, A). The differences increase at high temperatures. The total weight loss for the corolla, shoulder, and bottom is 9.18, 9.47, and 10.24%, respectively. This is about a third less than in the samples of vessel N 8 and clay N 1.It can be assumed that the difference is due to the presence in the ceramics of thinners that do not change the mass when heated, which make up about a third of its total mass. The diffractograms of the N9 vessel samples are also similar to each other. The molding masses of the corolla, shoulder and bottom have the same composition of thinners (Fig. 9, B). Their presence is especially clearly visible when compared with the diffractogram of the sample of the body of vessel No. 8, in the manufacture of which otoschiteli were not used (see Fig. 5,3). The degree of heat treatment and the number of thinners in the molding masses of vessel No. 9 can be judged from Fig. The points of the three samples are located approximately along the "firing" line, parallel to that for the samples of vessel N 8, but shifted to the center of coordinates by approximately one third. For the manufacture of vessel No. 9, otoschiteli were used in the amount of about a third of the total mass. The points are located quite far from the "otoshchiteli" line. This indicates a fairly long-term heat treatment of ceramics. The best quality of firing is observed at the corolla of the vessel, the worst - at the bottom.

Samples of vessel N10 lose weight almost equally up to a temperature of 400°C (Fig. With further heating, the torso sample loses about 1% less mass than the corolla and shoulder samples. It is clear that this difference is mainly due to the content of hydroxyl in clay minerals. And since the differences in the amount of hydroxyl remaining in ancient ceramics, all other things being equal, arise due to the different quality of firing, it can be concluded that the corolla and shoulder experienced a less intense thermal effect than the body. The total weight loss for the corolla, shoulder, and trunk is 12.12, 11.74, and 10.82%, respectively. This is less than in the samples of vessel N 8 and clay N 1, but more than in the samples of vessel N 9. Apparently, the molding masses of vessel No. 10 contain thinners, but

9. Results of analysis of samples of vessel No. 9 from dwelling 17.

A-curves of TG (a) and DTG (b); B-diagram of the degree of preservation of clay minerals (the arrow indicates the displacement of the "roasting" line relative to that for vessel No. 8).

1-corolla, 2-shoulder, 3-bottom.

page 70

10. Results of analysis of samples of vessel No. 10 from dwelling 17.

A-curves of TG (a) and DTG (b); B-diagram of the degree of preservation of clay minerals (the arrow indicates the displacement of the "firing" line relative to that for vessel No. 8).

1-corolla, 2-shoulder, 3-torso.

in a smaller amount compared to vessel No. 9. This is also confirmed by the location of the points in the diagram (Fig. 10, B): they lie closer to the points of the samples of vessel N 8 (without thinners) than the points corresponding to vessel N 9. The corolla and shoulder are slightly less burned than the torso (in Fig. 10, B, their points are located close to each other and not far from the "otoshchiteli" line).

X-ray studies have shown that samples from different parts of the N10 vessel have a similar phase composition (see Figs. When compared with the diffractogram of a sample of the body of vessel N 9, in which there are approximately twice as many otoshchiteley (estimated by the results of thermo-gravimetry), it is clear that the intensity of reflexes related to otoshchiteley (mainly field chemicals) in samples of vessel N 10 is less, which is very noticeable both in the range of angles 22-26° and and at 30.6°. This confirms the conclusion about a smaller amount of otoschitel compared to vessel N 9. The presence of a reflex at an angle of 10.6°, which is absent on the diffractogram of the sample of vessel N 9, indicates that the phase compositions of otoschitel in the molding masses of the compared vessels differ. This phase with a large unit cell probably refers to clay minerals, possibly one of the varieties of mica. Interestingly, the interplanar distance in this phase significantly differs from that in mica found in the samples of vessel No. 5, where the reflex is observed at an angle of 8.9°.

The TG curves of the bottom, trunk, and corolla samples of vessel No. 11 (Fig. II, A) diverge at the dehydration stage and then run almost parallel. The mass loss rate lines clearly show that the dehydration peaks at 145°C have different amplitudes. The maximum speed is at the bottom sample, and the minimum speed is at the corolla. After the 300°C mark, the lines almost merge. The peaks of hydroxyl loss in the range of 460-500°C are also approximately equal. This indicates the same quality of firing of different parts of the vessel. The total weight loss is 10.62 % for the bottom, 8.98 % for the trunk and 8.54 % for the corolla. Since these values are significantly lower than for the clay sample N 1, it can be said that the molding mass of the vessel N 11 contains approximately 30 (bottom) to 40 % (corolla) of thinners. 11, B), the points of the corolla, trunk, and bottom samples lie approximately along the line,

11. Results of analysis of samples of vessel No. 11 from dwelling 17.

A - curves of TG (a) and DTG (b); B-diagram of the degree of preservation of clay minerals.

1-corolla, 2-body, 3-bottom.

page 71

12. Results of analysis of samples of vessel No. 12 from dwelling 17.

A - TG curves; B - DTG curves; C-diagram of the degree of preservation of clay minerals.

1-corolla, 2-shoulder, 3-trunk, 4-bottom.

a line drawn from the origin (this line is called the auxiliary otoshchitelei line in the diagram). Thus, if the points of vessels N 8 and 9 were located mainly along the "roasting" line (the same number of otoshchiteli, but different quality of roasting), then the points of vessel N 11 were located along the "otoshchiteli" line (the quality of roasting is almost the same, and the number of otoshchiteli in the samples is different). Interestingly, the firing quality of vessel N 11 is approximately the same as that of vessels N 7 and 8. The diffractograms of three samples of vessel N 11 are similar to each other (see Fig.
12, A) diverge at the stage of dehydration, then the lines of the bottom and corolla run almost parallel to each other, and the trunk and shoulder lines are close and intersect. Differences in total weight loss for the corolla, trunk, and shoulder samples are within 1.3 %, while the bottom sample differs from them by more than 3 %. 12, And it is difficult to see any regularity. Figure 12, B. The rate of mass loss clearly shows that the body, shoulder, and corolla samples behave completely differently from the bottom sample. The first three have a double dehydration peak, with maxima near 100 and 240°C. The reason for this two-step approach is not entirely clear, and various guesses can be made, but we will not focus on the explanation. The DND sample has a single intense dehydration peak and a reliably recorded hydroxyl loss peak, which is absent in the other three samples. 12, C), the points corresponding to the corolla, shoulder, and trunk are located at the very origin of the coordinate axes. The molding masses of these samples are characterized by a small amount of clay minerals (initial or remaining after firing and operation of the vessel). The point corresponding to the bottom is far away from the first three. There is no doubt that the bottom is made of a different molding compound. This conclusion is confirmed by the results of the analysis of the composition of otoschiteles.

The diffractograms of the N12 vessel samples (see Fig. 5, L) differ markedly. For example, the molding mass of the shoulder contains a minimum amount of otoshchiteli. The intensity of plagioclase reflexes in the range of angles 21.5-26° is very low compared to the other samples, and reflexes at angles 8.9, 17.8, and 19.9° are generally absent. This sample should contain the maximum amount of clay minerals and lose the greatest amount of water. However, its weight loss during dehydration is the smallest among all the samples studied. The molding mass of the bottom contains a significant amount of plagioclases, and in addition, some mineral with a large interplane distance (very intense reflex at an angle of 8.9°). It would seem that the mass loss of this sample should be small. However, when dehydrated, it loses the maximum amount of water. At the same time, judging by the diagram of the degree of preservation of the clay component, the bottom was subjected to a rather strong thermal effect. It is impossible to explain such discrepancies only by roasting or the amount of otoschitel. Both factors are clearly at work here at the same time. Samples of different parts of vessel No. 12 differ both in composition and in the degree of preservation of clay minerals. This is a very heterogeneous vessel. At the same time, there is no doubt that the fragments of the corolla, torso and shoulder really belong to the same product. They have a characteristic double

page 72

13. Results of analysis of vessel samples N 7 and 13 from dwelling 17.

A - curves of TG (a) and DTG (b); B-diagram of the degree of preservation of clay minerals.

1 - corolla of vessel N 7; 2-corolla of vessel N 13.

dehydration peak, which has not yet been observed in other ceramic samples from the Linevo-1 settlement.

Two vessels (N 7 and 13) from dwelling 17 (see Figs. 3. 1, 16) are represented only by corolla samples. 13, A), both total (N7 - 11.55 %, N13 - 5.28%) and at the stages of dehydration and dehydroxylation. The molding masses are clearly of different compositions. The degree of firing of the corollas also varies. The peak of hydroxyl decomposition is approximately the same, but the peak of dehydration in the sample of vessel N 7 is much larger, and since the degree of preservation of clay minerals is related to the ratio of mass loss at these two stages, it will be different (Fig. 13, B). The point of the sample of vessel N 7 is located in the region of high values of hydroxyl, and quite far from the line "otoshchiteli", and the point of the sample vessel N 13 - in the region of average values and quite close to the line "otoshchiteli". The corolla of the first vessel was significantly more calcined than the second.

Diffractograms of vessel samples N 7 and 13 (see Fig. 5, M) are different: the former has more reflexes than the latter. The molding mass of the whisk of the vessel N 13 does not contain otoshchiteli, except chamotte. The diffraction pattern of this ceramic sample is similar to that of clay sample N2.

Conclusions

Since our tasks included not only the publication of new data on the ceramics of a particular monument, but also a detailed description of the methodology for interpreting the results of analytical work on thermogravimetry and radiography, the conclusions should be divided into two parts: the possibilities of the methods used and information about ceramics from the Linevo-1 settlement obtained using analytical techniques.

I. The results of thermogravimetric measurements allow us to determine the differences in the quality of firing of different samples of ancient ceramics. We are not talking about "measuring" or "calculating" the temperature, as is often found in the literature. Roasting quality is a comparative characteristic. It can be judged by the ratio of mass loss of the sample due to dehydroxylation and dehydration. Mass loss by these two mechanisms occurs in different temperature ranges. Using the diagram of the degree of preservation of the clay component, you can determine which samples were subjected to strong firing, and which were subjected to weak firing. The accuracy and reproducibility of thermogravimetric measurements allowed us to conclude that different parts of the same vessel experienced different thermal effects. This may be due to the use of the simplest roasting devices such as a bonfire.

Thermogravimetric and radiographic studies reveal differences in the composition of molding masses: different ratios of clay minerals and thinners; the presence of various minerals. The first is determined by the total mass loss: the larger it is, the less mineral thickeners there are. The second is determined by the diffractogram of the sample, by the position of the reflexes.

Based on the results of thermogravimetric and radiographic measurements, it is possible to reconstruct the manufacturing technology of ancient ceramic dishes. A comparative analysis of samples of different parts of the same vessel allows us to draw conclusions about which of them are made from the same molding masses, and which are made from different ones; for which parts of the vessel a molding mass with a small number of otoschitel was used, and for which - with a large one.

According to thermogravimetric measurements, it is possible to determine the functional purpose of the ancient vessel: whether it served as a storage container

page 73
some materials or used for cooking or for storing fire. During primary firing, the degree of thermal transformation in the thickness of the ceramic wall is different: the maximum load is experienced by the outer surface, the minimum - by the middle. If the product served as a container, this picture is saved. When using a cooking vessel, its outer surface is repeatedly exposed to fire, while the inner surface is not heated to more than 100°C (with water). Accordingly, the first one is characterized by the maximum degree of thermal transformations, while the second one is characterized by the minimum degree. If the vessel was used for storing fire, then its inner surface was exposed to the greatest thermal impact. This type of research is most convenient to conduct on samples of thick-walled dishes [Chicha..., 2009, pp. 150-176].

II. At the settlement of Linevo-1, molding masses with different contents of chamotte were used as an otoshchitel for the manufacture of dishes. This recipe is most typical for the ceramics of the Irmen culture (although samples with an admixture of chamotte and crushed rock, but with a predominance of the former). Such rock fragments, which included plagioclases and micas, were used as thinners, which was recorded in samples of ceramics of groups II and III. Composition of molding masses in the manufacture of different parts of the vessel (bottom, body, corolla) it could change significantly. The same feature is noted for the ceramics of the Late Irma and Sargat cultures at the Chicha-1 settlement (Drebushchak V. A., Mylnikova, Drebushchak T. N. et al., 2006, pp. 74-75). Obviously, the use of molding masses with different amounts and quality of ingredients for the production of one product is an innovative feature of the culture of the transition period from the Bronze to the Iron Age. At the same time, the settlement has vessels made entirely from homogeneous molding masses, as well as from clay without additives. These are mainly dishes with features of Early Iron Age ceramics.

The findings on vessels N 7 and 12 are particularly interesting. According to the manufacturing technology, ornamentation technique, and physical and chemical characteristics of the molding masses, they are not locally produced. In most cases, the presence of chamotte was detected by the three methods used for ceramics that were considered to be classical Irma ceramics by all indicators. The molding masses of dishes that have analogues on the monuments of the Tom region contain an admixture of crushed rock and chamotte in various percentages.

The degree of roasting of different parts of the vessels is different, apparently due to the heterogeneity of the flame, i.e. the simplest roasting devices of the bonfire type, most often open, were used.

Thus, derivatogravimetric analysis in combination with X-ray phase analysis and petrographic analysis provide quite large opportunities for characterizing pottery technology.

List of literature

Bobrov V. V. Perekhodnoe vremya ot bronzy k zhelezu v lesostepnom Priobye [Transition time from bronze to iron in the forest-steppe Ob region]. November 1995: tez. dokl. Novosibirsk: Izd-vo IAET SB RAS, 1995, pp. 24-26.

Bobrov V. V. Complexes of ceramics with figurative-stamp ornamentation in the South-east of Western Siberia / / Archeology, Ethnography, museum business. Kemerovo: Kem. state University, 1999, pp. 32-43.

Grebenshchikov A.V., Derevyanko E. I. Pottery of ancient tribes of the Amur region (beginning of the Early Iron age). Novosibirsk: IAET SB RAS Publ., 2001, 120 p. (in Russian)

Drebushchak V. A., Drebushchak T. N., Molodin V. I., Mylnikova L. N., Boldyrev V. V., Derevyanko E. I. Rezul'taty i vozmozhnosti thermogravimetrii drevnoi keramiki (po materialam gorodishche Chicha-1) [Results and possibilities of thermogravimetry of ancient ceramics (based on the materials of the Chicha-1 ancient settlement)] Novosibirsk State Pedagogical University un-ta, 2005, vol. 1, pp. 101-111.

Drebushchak V. A., Mylnikova L. N., Drebushchak T. N. Comprehensive study of ancient ceramics: some questions of the results interpretation methodology / / Annual Review in Cultural Heritage Studies. - 2006. - Vol. 39. - P. 316-350 (in cor. and rus. yaz.).

Drebushchak V. A., Mylnikova L. N., Drebushchak T. N., Boldyrev V. V. Roentgenograficheskoe issledovanie drevnoi keramiki [Radiographic study of ancient ceramics]. IV National Conference on the use of X-ray, Synchrotron radiation, Neutrons and Electrons for materials Research (RSNE-2003): tez. dokl. - Moscow, 2003, p. 104.

Drebushchak V. A., Mylnikova L. N., Drebushchak T. N., Boldyrev V. V., Derevyanko E. I. K voprosu o thermogravimetricheskom analiz drevnoi keramiki [On the issue of thermogravimetric analysis of ancient ceramics]. Novosibirsk: Publishing House of IAET SB RAS, 2003a, vol. 9, part 2, pp. 183-186.

Drebushchak V. A., Mylnikova L. N., Drebushchak T. N., Boldyrev V. V., Derevyanko E. I. Metody estestvennykh nauk v izuchenii keramiki gorodishche Chicha-1 [Methods of natural sciences in the study of ceramics of the Chicha-1 settlement]. Sotsial'no-demograficheskie protsessy na territorii Zapadnoy Sibiri (drevnost ' i srednevekovye). - Kemerovo: Kem. State University, 20036 , pp. 154-156.

Drebushchak V. A., Mylnikova L. N., Drebushchak T. N., Boldyrev V. V., Molodin V. I., Derevyanko E. I., Mylnikov V. P., Nartova A.V. Fiziko-khimicheskoe issledovanie keramiki (na primere izdelii perekhodnogo vremeni ot bronzovogo k zheleznom veku) [Physical and chemical research of ceramics (on the example of products of transition time from the Bronze to the Iron Age). Novosibirsk: Publishing House of the Siberian Branch of the Russian Academy of Sciences, 2006, 98 p. (Integration Projects of the Siberian Branch of the Russian Academy of Sciences; issue 6).

Zakh V. A. Linevo-1-monument of transition time from bronze to iron / / Scythian epoch of Altai. - Barnaul: Alt Publishing House. State University, 1986, pp. 94-95.

Zakh V. A. The Bronze Age of the Prisalairya region (based on the materials of the Izylinsky archaeological microdistrict). Novosibirsk: Nauka Publ., 1997, 132 p. (in Russian)

page 74
Kosarev M. F. Bronzovyi vek Zapadnoy Sibiri [The Bronze Age of Western Siberia]. Moscow: Nauka Publ., 1981, 280 p.

Kosarev M. F. Late Bronze Age and Transition time from the Bronze Age to the Iron Age. Transition time from the Bronze Age to the Iron Age // The Bronze Age of the forest belt of the USSR, Moscow: Nauka Publ., 1987, pp. 296-1304.

Lamina E. V., Lotova E. V., Dobretsov N. N. Mineralogy of ancient Baraba ceramics. Novosibirsk, IAET SB RAS Publ., 1995, 126 p. (in Russian)

Mogilnikov V. A. K etnokul'turnoi situatsii na Altae v skifskoe vremya [On the ethno-cultural situation in the Altai during the Scythian period]. Barnaul: Alt State University Publ., 1986, pp. 29-32.

Molodin V. I., Mylnikova L. N. Binocular microscopy of ceramics of the Chicha-1 settlement / / Socio-demographic processes on the territory of Siberia (antiquity and Middle Ages). - Kemerovo: Kem. State University, 2003, pp. 147-151.

Molodin V. I., Mylnikova L. N. Ceramics of the Chicha-1 settlement as a source on the history of transition time from the Bronze Age to the Iron Age // History and culture of Siberia in the research and educational space (dedicated to the anniversary of Professor E. I. Solovyova): mat-ly region, nauch. - Practical conference-Novosibirsk: Novosibirsk Publishing House. state ped. un-ta, 2004, pp. 101-106.

Molodin V. I., Mylnikova L. N. Keramika poseleniya Linevo-1 perekhodnogo vremeni ot bronzovogo k zheleznom veku [Ceramics of the Linevo-1 settlement of the transition period from the Bronze Age to the Iron Age]. Problemy arkheologii, etnografii i antropologii Sibiri i sopredel'nykh territorii: (Materials of the Annual Session of the IAET SB RAS 2005), Novosibirsk: Izd. 11, part 1, pp. 400-405.

Molodin V. I., Mylnikova L. N., Parzinger G., Shneevais Y. Keramika gorodishcha Chicha-1 (tekhnologicheskie aspekty) [Ceramics of the Chicha-1 settlement (technological aspects)]. Istoricheskiy opyt khozyaistvennogo i kul'turnogo osvoeniya Zapadnoy Sibiri: sb. nauch. trudy-Barnaul: Izd. Alt. gosudarstvennogo un-ta, 2003, pp. 299-311.

Mylnikova L. N. Pottery of Neolithic tribes of the Lower Amur (based on the materials of the Neolithic settlement of Condon-Pochta). Novosibirsk: Publishing House of IAET SB RAS, 1999, 160 p. (in Russian)

Mylnikova L. N., Durakov I. A., Mzhelskaya T. V., Kobeleva L. S. Arkheologicheskoe izuchenie poseleniya Linevo-1 (Novosibirsk region) [Archaeological study of the Linevo-1 settlement (Novosibirsk region)]. Problemy arkheologii, etnografii i antropologii Sibiri i sopredel'nykh territorii: (Materials of the Annual Session of the IAET SB RAS 2004). RAS, 2004, vol. 10, part G, pp. 390-393.

Mylnikova L. N., Durakov I. A., Mzhelskaya T. V., Mylnikov V. P., Nevzorova I. V., Savin A. N., Parinov R. O. Studies of the Linevo-1 settlement during the transition period from the Bronze Age to the Iron Age // Problemy arkheologii, etnografii i antropologii Sibiri i sopredel'nykh territorii [Problems of Archeology, Ethnography and Anthropology of Siberia and Adjacent Territories: Materials of the Annual Session of the IAET SB RAS 2003]. Novosibirsk: Publishing House of the IAET SB RAS, 2003, vol. 9, part 1, pp. 459-463.

Mylnikova L. N., Durakov I. A., Mzhelskaya T. V., Savin A. N., Kobeleva L. S., Syatkin V. P., Parinov R. O. Raboty na poselenii Linevo-1 v 2005 g. [Works on the settlement of Linevo-1 in 2005]. Problemy arkheologii, etnografii i antropologii Sibiri i sopredel'nykh territorii: (mat-ly Godovoy sessii IAET SB RAS 2005). Novosibirsk: Izd-vo IAET SB RAS, 2005. - Vol. 11, part 1. - pp. 431-436.

Troitskaya T. N., Zakh V. A., Sidorov E. A. Novoe o zavyalovskaya kul'tury [New information about zavyalovskaya culture]. Zapadno-sibirskaya lesostep ' na rubezhe bronzovogo i zheleznogo veka [West Siberian forest-steppe at the turn of the Bronze and Iron Ages]. Tyumen: Publishing House of the Tyumen State University, 1989, pp. 103-116.

Molodin V. I., Parzinger G., Krivonogov S. K., Kazansky A. Yu., Chemyakina M. A., Matasova G. G., Vasilevsky A. N., Ovcharenko A. S., Grishin A. E., Ermakova N. N., Dergacheva M. I., Chicha - gorodishche perekhodnogo ot bronzy k zhelezu vremeni v Barabinskaya lesostepi [The ancient settlement of the transition from Bronze to Iron time in the Barabinsk forest-steppe]. N. Fedeneva, O. A. Nekrasova, L. N. Mylnikova, I. A. Durakov, L. S. Kobeleva, A.V. Zubova, T. A. Chikisheva, D. V. Pozdnyakov, A. S. Pilipenko, A. G. Romashchenko, I. V. Kulikov, V. F. Kobzev, O. I. Novikova, S. K. Vasiliev, I. Shneevais, K. Privat, V. V. Boldyrev, V. A. Drebushchak, T. N. Drebushchak, E. I. Derevyanko, A. P. Borodovsky, D. Borgarit, I. Reiche, S. V. Kuzminykh, Zh. V. Marchenko. Novosibirsk: Publishing House of IAET SB RAS, 2009, vol. 3, 248 p.

Chlenova N. L. Monuments of the Late Bronze Age in Western Siberia, Moscow: Pushchinsky Scientific Center of the Russian Academy of Sciences, 1994, 170 p.

Drebushchak V.A., Mylnikova L.N., Drebushchak T.N., Boldyrev V.V. The investigation of ancient pottery: Application of thermal analysis // J. Therm. Anal. Cal. - 2005. - Vol. 82(3). - P. 617 - 626.

Drebushchak V.A., Mylnikova L.N., Molodin V.I. Thermogravimetric investigation of ancient ceramics: metro-logical analysis of sampling // J. Therm. Anal. Cal. - 2007. - Vol. 90(1). - P. 73 - 79.

Nahdi K., Perrin S., Pijolat M., Rouquerol F., Ariguib N., Ayadi M. Nucleation and anisotropic growth model for isothermal kaolinite dehydroxylation under controlled water vapour pressure // Phys. Chem. Chem. Phys. - 2002. - Vol. 4. - P. 1972 - 1977.

Rouxhet P.G. Kinetics of dehydroxylation and of OH-OD exchange in vacrocrystalline micas //Amer. Mineral. - 1970. - Vol. 55. - P. 841 - 853.

The article was submitted to the Editorial Board on 03.08.09.

page 75

Permanent link to this publication: