Acta Chim. Slov. 2002, 49, 677-720. 677 CHARACTERIZATION OF BROWN LAYERS ON FAÇADES OF ARCHAEOLOGICAL BUILDINGS IN SLOVENIA Polonca Ropret1, Peter Bukovec2 1Institute for the Protection of Cultural Heritage of Slovenia, Restoration Center, Poljanska 40, 1000 Ljubljana, Slovenia 2Faculty of Chemistry and Chemical Technology, University of Ljubljana, Aškerčeva 5, 1000 Ljubljana, Slovenia Darko Hanžel J.Stefan Institute, Jamova 39, 1111 Ljubljana, Slovenia Received 22-02-2000 Abstract Brown layers have occurred just below the façade surface of some archaeological buildings. They are a consequence of chemical changes, which can give rise to further destruction. Brown layers and undamaged plaster underneath the brown layers consist of CaCO3, SiO2, Ca2SiO4, CaMg(CO3)2, silicate minerals containing Ca, Al, Mg, K, Fe, metal hydroxides and sulphates, with some surface bound water. In both layers coordination of Fe3+ ions is tetrahedral and octahedral. In brown layers there are more Fe3+ ions at octahedral sites and there is an evidence of a change in chemical environment of ferric ions. The difference between brown layers and undamaged plaster underneath the brown layers is in goethite (?-FeOOH) formation. It is microcrystalline and some of the Fe3+ ions are substituted by a non magnetic ion, probably Al3+. Iron in oxidation state 2+ is present only in the brown layer of sample ZMV 7. One of the possible ways of its genesis is reduction of Fe3+ due to SO2 present in the air. Introduction A change of colour on archaeological objects often agitates restorers and architects because it is aesthetically unattractive and it can also give rise to further destruction of those objects. On some constructions of historical value in Slovenia, built or renewed around year 1910, brown layers just below the facade surface have occurred. Those layers are hard and thin, usually 1-2mm. There are parts where brown colour has entered even further into the interior of the plaster. In the Centre for Restoration of the Republic of Slovenia they have prepared colour studies1 of samples taken from building facades. Those studies were done by microscopic observations. Since the borderline between brown layers and undamaged plaster beneath is not sharp, they assumed that brown P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 678 Acta Chim. Slov. 2002, 49, 677-720. layers were not created by adding a pigment into the external layer of building facades, but are a consequence of chemical changes. Some of the former investigations2,3,4 of cements and iron containing material, have shown, that iron could be included into chemical changes of building plaster. D. Hanžel, D. Dimic and D. Lasič studied hydration behaviour of two Yugoslav cements by Mössbauer spectroscopy.2 According to their investigation cement contains minerals with ferric ions at octahedral and tetrahedral sites in ferrites of brownmillerite-like structure having the generic formula 2CaO?(Fe2O3)1-x?(Al2O3)x. Spectra of hydrated samples differ from those of nonhydrated samples. This indicates a change in chemical environment of ferric ions during the process of hydration. There is also an evidence that the ferrite phase might take part in the complex cementing action during ageing. Analysis of thin black layers on building stone3,4 showed that the layers mainly consist of various iron oxides and iron oxide hydroxides, sulphates, soot and silicate minerals, with smaller amounts of metal, rubber and asphalt particles. In low concentration are present also the organic constituents. The aim of this investigation was to characterise brown layers formed on building facades. Various techniques including electronic microanalysis, thermogravimetric and differential thermal analysis, X-ray powder analysis, IR and Mössbauer spectroscopy were used to analyse these layers. The samples Samples were taken from 4 archaeological buildings in Slovenia. Three of them are placed in Ljubljana (Zmaj bridge-sample ZMV 7, Barrier on river Ljubljanica-sample ZAP 10, Auersperger palace-sample AUE 37), and one in Domžale (Social home Domžale-sample DDD 36). From each of the four buildings one sample was analysed. In each case, a sample of corresponding undamaged plaster underneath the brown layer was collected for comparison. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 679 Brown layers and undamaged plaster were separated by a diamond knife. The knife was used because the material is very hard, and to avoid contamination. For electronic microanalysis samples were prepared in a shape of a flat tile. For all other analyses samples were milled into powder and homogenised. Instrumental analysis Electronic microanalyses were done on an electronic microscope JEOL JSM 35, equipped with energy disperse spectrometer Tractor. Thermo gravimetric and differential thermal analyses were taken on a Perkin-Elmer 7 Series Thermal Analysis system in air flow. The heating rate was 5 °C/min. X-ray powder patterns were taken on a Huber Guinier camera 620 (Cu K„). The patterns were compared with standard patterns in the PDF5 using uPDSM computer program.6 Infrared spectra were taken on a Perkin-Elmer FT-IR 1720 X spectrometer in nujol matrix in CsI single crystal windows. Mössbauer spectra were measured by using a Wissel constant acceleration spectrometer. The spectra were analysed by a non-linear least squares fitting procedure. Metallic iron was used as a reference for isomer shift parameter as well as for calibration of the velocity scale. The y-ray source was 57Co in rhodium matrix. Results Elemental composition was determined by electronic microanalysis. The surface of the samples is inhomogeneous. Therefore electronic microanalysis was used only as a qualitative analysis. As it is shown in Table 1 the main component in both layers is calcium. Other elements are present in smaller concentration. Sample AUE 37 -undamaged plaster is unsuitable for this investigation because it is in a form of fine dry particles and it was impossible to get it in a shape of a flat tile. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 680 Acta Chim. Slov. 2002, 49, 677-720. TABLE 1. The results of the electronic microanalysis given in weight percent considering oxides. a.) brown layer b.) undamaged plaster underneath the brown layer 1., 2., 3., 4. – different locations of the same sample Remark: sample AUE 37-b.) is unsuitable for this investigation Ca Si K Mg Al S Fe Ti P Cl DDD 36 a.) 1. 2. 3. b.) 1. 2. 52.26 4.77 19.19 0.09 3.89 12.91 5.68 0.09 77.61 10.04 4.34 1.09 1.86 1.53 1.89 0.48 50.80 2.85 0.77 0.53 12.05 1.03 67.64 19.47 0.79 4.76 0.29 5.76 0.21 87.39 6.58 0.46 1.93 1.37 0.87 0.16 ZMV 7 a.) 1. 2. 3. 4. b.) 1. 2. 3. 4. 5.03 86.47 1.94 2.05 3.30 80.80 8.09 1.98 1.54 2.48 1.03 1.65 80.25 6.77 1.45 2.66 1.54 1.23 2.87 0.65 1.28 73.99 2.25 17.29 0.38 0.85 1.65 0.86 2.60 96.47 70.65 10.57 8.04 4.49 2.15 2.68 75.39 1.21 21.99 69.23 18.13 0.98 5.29 2.64 0.72 1.73 ZAP 10 a.) 1. 2. 3. b.) 1. 2. 3. 80.93 10.82 0.63 1.04 2.92 1.24 0.59 23.76 39.94 4.71 5.22 18.98 5.85 0.47 44.42 9.98 2.87 39.34 0.31 96.52 1.50 0.64 41.57 40.62 1.28 0.95 5.68 2.10 5.77 0.98 93.53 2.63 0.70 0.24 0.39 0.85 0.17 0.10 AUE 37 a.) 1. 2. 3. 74.53 10.51 2.51 1.03 3.07 2.94 4.13 68.77 12.49 4.28 1.61 3.29 3.54 5.05 64.34 13.08 6.92 0.90 3.29 3.23 6.99 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 681 Thermo gravimetric and differential thermal analyses gave similar results for all samples as well as for corresponding undamaged plaster underneath. In both cases TG curves show two significant mass losses, one up to 100 °C and another one more intense in the temperature range of 600-900 °C (Figure 1). At these temperatures DTA curves show two corresponding endothermal peaks. In TG curves a slow, continuous weight loss is noticeable in the temperature range of 100-500 °C. TG analysis of samples DDD 36-undamaged plaster (Figure 2) and ZMV 7-undamaged plaster shows another mass loss at temperature 400-450 oC. FIGURE 1. TG and DTA analysis of the sample DDD 36 – brown layer 100 95 90 -85 -80 75 70 65 60 0 150 300 450 600 T [oC] 750 900 t- 0.04 0.02 0 -0.02 -0.04 -- -0.06 -- -0.08 -0.1 -0.12 -0.14 TG DTA 1050 FIGURE 2. TG and DTA analysis of the sample DDD 36 – undamaged plaster underneath the brown 100 95 90 85 80 75 70 -65 60 0 \/ 150 r 0.06 TG y - - 0.04 - 0.02 DTA ^y _ - 0 - -0.02 - -0.04 - -0.06 - -0.08 - -0.1 - -0.12 300 450 600 T [oC] 750 900 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 682 Acta Chim. Slov. 2002, 49, 677-720. Infrared spectra were taken on a sample DDD 36 – brown layer and on a residual after TG analysis of the same sample. Characteristics: 1. DDD 36 – brown layer a.) Broad band in the area of 3700 – 3000 cm-1. b.) Strong, broad band ranging from 1500 to 1400 cm-1. c.) Broad band in the area of 1200 – 800 cm-1 with one sharp peak at 873 cm-1. d.) Strong, sharp peak at 713 cm-1. 2. The rest after TG analysis of sample DDD 36 – brown layer a.) Sharp band at 3640 cm-1. b.) Broad band in the area of 1200 – 800 cm-1. X-ray powder analyses were done on all samples and on TG residuals. Powder patterns show many broadened or overlapping diffraction lines. It is therefore difficult to determine the exact composition of samples. Components which show the best matching of lines comparing to lines of components from PDF computer database are: CaCO3, SiO2, Ca2SiO4 and CaMg(CO3)2. It was often easier to determine composition of the rest after TG analysis because the lines are sharper and less overlapping. Tables from 2 to 17 show the results of x-ray powder analysis. Mössbauer spectra have been recorded on brown layers of samples DDD – 36 and ZMV – 7 and on the undamaged plaster underneath brown layers of these samples. Spectra of undamaged plaster of both samples consist of two overlapping doublet patterns (Figure 3). Similar patterns are present also in spectra of brown layers (Figure 4) except there is a change in quadrupole splitting parameter of approximately 0.2 mm/s and the ratio of relative intensities also becomes different. In the spectrum of sample ZMV – 7-brown layer (Figure 5) another quadrupole splitting doublet is present, with higher values for isomer shift and quadrupole splitting parameter but a very low intensity. Hyperfine parameters of Mössbauer spectra recorded at room temperature are given in Table 18 and 19. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 683 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 684 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 685 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 686 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 687 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 688 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 689 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 690 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 691 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 692 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 693 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 694 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 695 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 696 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 697 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 698 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 699 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 700 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 701 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 702 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 703 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 704 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 705 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 706 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 707 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 708 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 709 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 710 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 711 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 712 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 713 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 714 Acta Chim. Slov. 2002, 49, 677-720. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 715 FIGURE 3. The Mössbauer spectrum of 57Fe in sample DDD 36-undamaged plaster underneath the brown layer at room temperature I – Fe3+(1) - tetrahedral sites, II – Fe3+(2) - octahedral sites FIGURE 4. The Mössbauer spectrum of 57Fe in sample DDD 36-brown layer at room temperature I – Fe3+(1) - tetrahedral sites, III – Fe3+(3) - octahedral sites P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 716 Acta Chim. Slov. 2002, 49, 677-720. TABLE 18. Hyperfine parameters of Mössbauer spectrum of the sample DDD 36 at room temperature a.) undamaged plaster b.) brown layer oxidation state surface area isomer shift S [mm/s] quadrupole splitting AEQ[mm/s] line width r[mm/s] relative intensity [%] a.) Fe3+(1) 789 + 226 0.29+0.07 1.39 + 0.18 0.94+0.32 90 Fe3+(2) 85+96 0.35+0.07 0.40 + 0.11 0.26+0.18 10 b.) Fe3+(1) 722+217 0.28+0.04 1.62 + 0.11 0.50+0.16 51 Fe3+(3) 611 + 154 0.33+0.04 0.59 + 0.07 0.48+0.08 49 FIGURE 5. The Mössbauer spectrum of 57Fe in the sample ZMV 7 – brown layer at room temperature I – Fe3+(1) - tetrahedral sites, II – Fe3+(3) - octahedral sites, IV – Fe2+ TABLE 19. Hyperfine parameters of Mössbauer spectrum of the sample ZMV 7 at room temperature a.) undamaged plaster c.) brown layer oxidation state surface area isomer shift 5 [mm/s] quadrupole splitting AEQ[mm/s] line width r[mm/s] relative intensity [%] a.) Fe3+(1) 4595 ± 1036 0.28 + 0.05 1.44+0.15 0.70+0.18 68 Fe3+(2) 2152+ 803 0.34 + 0.05 0.43+0.08 0.42+0.10 32 b.) Fe3+(1) 2762 ± 1037 0.40 + 0.04 1.61+0.12 0.40+0.16 31 Fe3+(3) 4625+ 961 0.36 + 0.03 0.69+0.07 0.50+0.08 53 Fe2+ 1420+ 588 0.85 + 0.06 3.18+0.12 0.36+0.16 16 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 717 From the room temperature Mössbauer spectra it was not possible to distinguish between goethite (a-FeOOH) and ferrihydrite (Fe5HO84H2O). It is only obvious that there is a difference in chemical environment of ferric ions in brown layers and in the corresponding undamaged plaster underneath the brown layers. Therefore a low temperature spectrum (Figure 6) at 73 K was recorded for brown layer of sample DDD 36. The spectrum shows a sextet pattern as well as two overlapping doublet patterns. A hyperfine magnetic field appears because some of the particles have reached magnetic order. The linewidths are quite large. The results are shown in Table 20. FIGURE 6. The Mössbauer spectrum of 57Fe in the sample DDD 36-brown layer at temperature 73 K I - Fe3+(1) - tetrahedral sites, III - Fe3+(3) - octahedral sites TABLE 20.Hyperfine parameters of Mössbauer spectrum of the sample DDD 36-brown layer recorded at 73 K . a.) doublet b.) sextet a.) b.) oxidation state surface area isomer shift 5 [mm/s] quadrupole splitting AEQ [mm/s] line width r[mm/s] hyperfine magnetic field H [T] Fe3+(1) 3402 +1280 0.39 + 0.11 1.30+0.61 1.18 + 0.38 / Fe3+(3) 958 + 1392 0.46 + 0.06 0.72+0.16 0.44 + 0.26 / Fe3+(3) 2653 +1035 0.33 - 0.63 1.12 48.6 P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 718 Acta Chim. Slov. 2002, 49, 677-720. Discussion Samples of brown layers and samples of corresponding undamaged plaster underneath consist of elements: Ca, Si, Al, Mg, K, S, Fe, Ti, O, C, H. The DTA endothermic effect at 100 °C (Figure 1) can be ascribed to the desorption of surface bound water. A slow, continuous mass loss in the temperature range of 100-500 °C is a characteristic of OH group polycondensation. A significant mass loss in the temperature range of 600-900 °C belongs to the degradation of carbonates.7 In samples DDD 36-undamaged plaster (Figure 2) and ZMV 7-undamaged plaster are present both CaCO3 and CaMg(CO3)2. On account of that the degradation of carbonates occurs in two steps. CaMg(CO3)2 decomposes in the temperature range of 400-700 °C.7 The IR spectra confirm conclusions from the TG analysis. A broad band (sample DDD 36-brown layer) in the range of 3700-3000 cm-1 can be assigned to vibrations of water.8 Since the band is broad it can be assumed that molecules of water are differently bound to metals present in the sample. In the residual after TG analysis broad band in this area disappears and the sharp band at 3640 cm-1 can be ascribed to stretching mode of O-H group.8 It appears due to partial hydration of metal oxides on air after TG analysis. Strong, broad band in the range of 1500-1400 cm-1 and two sharp peaks at 873 cm-1 and 713 cm-1 belong to -CO3 modes.8 The first one is significant for ?3(C–O) stretching modes, the second one can be ascribed to ? modes of the whole carbonate group and the last one to ?4(O–C–O) bending modes. These bands are missing in the spectrum taken on the rest after TG analysis. The broad band in the range of 1200 - 800 cm-1 appears in both spectra and can be ascribed to M - O modes8 of various metal oxides present in samples. Because IR spectra didn’t give the information which metal oxides are present, we didn’t record spectra of all samples. Components which show the best matching of lines in X-ray powder patterns are CaCO3, SiO2, Ca2SiO4, CaMg(CO3)2, silicate minerals containing Ca, Al, Mg, K, Fe, metal hydroxides and sulphates. It was possible to determine more components in P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… Acta Chim. Slov. 2002, 49, 677-720. 719 residuals after TG analysis. In addition to better crystallised phases, e.g. SiO2, a number of new phases appeared in residuals due to chemical reactions. From the above analyses it could not be concluded that iron is involved in chemical changes of external layers of building facades. Mössbauer analysis turned out to be a convenient method for investigation of iron in poorly crystalline material. Two overlapping doublet patterns (Figure 3) in the room temperature spectra of undamaged plaster of the sample DDD 36 can be ascribed to ferric ions on tetrahedral (Fe3+(1)) and octahedral (Fe3+(2)) sites in ferrites of brownmillerite-like structure 2. In brown layers coordination remains tetrahedral and octahedral (Fe3+(3)), but there are more ferric ions at octahedral sites what can be concluded by comparing relative intensities (Table 18 and 19). The change in quadrupole splitting parameter of about 0.2 mm/s indicates a change in chemical environment of the octahedrally coordinated ferric ions (Fe3+(3)). Parameters of Fe3+(3) agree well with parameters of microcrystalline goethite (?-FeOOH)9 and ferrihydrite (Fe5HO8?4H2O).3,9 To determine which one of these two minerals is present in brown layers the low temperature spectrum at 73 K was recorded (Figure 6). Parameters of this spectrum (Table 20) match with parameters of microcrystalline goethite.9 The hyperfine magnetic field is 1 T lower than that found in literature, which can be explained by aluminium substitution, sample crystallinity and particle size effects.10,11 The difference between brown layers and undamaged plaster underneath the brown layers is in formation of goethite (?-FeOOH). It is microcrystalline and some of the Fe3+ ions are substituted by a non magnetic ion, probably Al3+. Iron in oxidation state 2+ is present only in the brown layer of sample ZMV 7 (Figure 5). One of the possible ways of its genesis is reduction of Fe3+ due to SO2 present in the air. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological… 720 Acta Chim. Slov. 2002, 49, 677-720. Acknowledgements The authors wish to thank to the Ministry of Education, Science and Sport of the Republic of Slovenia for their financial support (Grant PO – 0508-0103). References and Notes 1. I. Nemec, Društveni dom Domžale – barvna študija, Konzervatorski nadzor: Zavod za varstvo naravne in kulturne dediščine, Kranj 1995 2. D. Hanžel, G. Lahajnar, A Study of Two Yugoslav Cements by Mössbauer Spectroscopy of 57Fe, Vestnik slovenskega kemijskega društva, Ljubljana, 1986, 33, 147-150 3. A. J. Nord and T. Ericson, Chemical Analysis of Thin Black Layers on Building Stone, Studies in Conservation, 1993, 38, 25-35 4. W. E. Steger and H. Mehner, The Iron in Black Weathering Crusts on Saxonian Sandstones Investigated by Mössbauer Spectroscopy, Studies in Conservation, 1998, 43, 49-58 5. PDF, Sets 1-49 and 70-86, International Centre for Diffraction Data, Newtown Square, Pennsylvania, USA 1999 6. µPDSM, ”Micro Powder Diffraction Search Match”, Fein-Marquart Associates. Release 4.30 7. Atlas of Thermoanalytical Curves, ed. G. L. Pyay, Akadémiai Kiadó, Budapest 1977 8. K. Nakamoto, Infrared and Raman Spectra, Wiley & Sons, 4th Edition, 1986 9. E. Murad and J. H. Johnston, Iron Oxides and Oxyhydroxydes, Mössbauer Spectroscopy Applied to Inorganic Chemistry, Vol. 2, ed. G. J. Long, Plenum Press, New York 1987 10. S. Morup, Mössbauer Studies of Microcrystalline Materials, Mössbauer Spectroscopy Applied to Inorganic Chemistry, Vol. 2, ed. G. J. Long, Plenum Press, New York 1987 11. D. C. Golden, L. H. Bowen, S. B. Weed and J. M. Bingham, Mössbauer Studies of Synthetic and Soil – occurring Aluminium – Substituted Goethites, Soil Sci. Am. J., 1979, 43, 802-808 Povzetek Na nekaterih arheoloških stavbah je fasada tik pod površino rjavo obarvana. Ta rjava obarvanost je posledica kemijskih sprememb, ki lahko pozročijo tudi nadaljno destrukcijo fasade. Rjavo plast in nepoškodovano plast fasade tik pod rjavo plastjo sestavljajo CaCO3, SiO2, Ca2SiO4, CaMg(CO3)2, silikati, ki vsebujejo Ca, Al, Mg, K, Fe, kovinski hidroksidi in sulfati. V obeh plasteh je tudi nekaj površinsko vezane vode. Fe3+ ioni so tetraedrično in oktaedrično koordinirani. V rjavi plasti je več Fe3+ ionov na oktaedričnih mestih in eksperimentalni podatki kažejo tudi na spremembo kemijskega okolja Fe3+ ionov. Razlika med rjavo plastjo in nepoškodovano plastjo fasade je v tem, da je v rjavi plasti nastal goethit (? - FeOOH), ki je mikrokristaliničen in delež Fe3+ ionov je substituiran z ne – magnetnim ionom, verjetno Al3+ ionom. Železo v oksidacijskem stanju 2+ je prisotno samo v rjavi plasti vzorca ZMV 7. Ena od možnih poti nastanka je z redukcijo Fe3+, zaradi prisotnosti SO2 v zraku. P. Ropret, P. Bukovec, D. Hanžel: Characterization of brown layers on façades of archeological…