·Home ·Table of Contents ·Conservation and Restoration in Art and Architecture

Investigation of Historical Glasses Using Natural Born Radioactivity M. Schiekel, A. Meister, M. Seibitz Technical Univ. of Dresden, Dept. Building Materials, Mommsenstraße 13, 01062 Dresden G. Haase Museum of Decorated Arts, Dresden-Pillnitz, Schloß Pillnitz, Wasserpalais Contact

Introduction

From the 16^th to the 18^th century Saxony had been one of the richest and most developing states in Germany. The Saxon elector Friedrich August II, called August the Strong supported not only the arts but also manufactures and scientists. The Saxon E.W. von Tschirnhaus, promoted by the elector, was one of the last all-round scientists of the 17^th century, dealing with philosophy and also with natural and technical sciences. Tschirnhaus was not only one of the mental originators of European porcelain manufacturing, he also dealt intensively with the improvement of glass production and glass applications, especially in the fields of scientific use. Under his direction products of excellent quality had been manufactured by the Dresden glassworks using special methods. These manufacturing procedures had been applied until the middle of the 18^th century.
The call for glass ware was great in the elector's court and could not been satisfied by only one glasswork. Thus also products made by different works had been used. Recently a considerable number of them have been collected into museums. But nowadays it is sometimes difficult to determine the origin of many pieces.
The glass ware of the 17^th until the early 19^th century was produced using naturally original materials. In that way also small amount of natural born radioactivity such as uranium daughter products and K-40 remained in the final products. Their concentration depends on the location the material had been got from and the glass refining procedures.
Investigating glass ware of known origin it is possible to determine special concentrations of natural born radioactivity as fingerprints of this glassworks.

Experimental

Elemental concentrations of historical glass is usually determined by excitation methods such as x-ray fluorescence [1,2] etc., all requiring an external irradiation with inherent risks of sample damaging due to additional energy deposition. As a different method excluding all these risks the measurement of radiation emitted by natural born radioactive nuclides can be applied. Natural born radioactivity is due to primary uranium or thorium and only a few other nuclides like K-40 with extreme long half lifetimes. In our investigation we used only some selected g-lines emitted from short living daughter nuclides and the 1462 keV-line of K-40 (table 1) as identification lines.

Nuclide	g-Energy in keV	Half life	Characteristic for
Pb-214 Pb-214 Bi-214 Bi-214 Bi-214	295 352 609 1120 1764	26.8 min 26.8 min 19.7 min 19.7 min 19.7 min	uranium
Ra-224	241		thorium
k-40	1462	1.27 · 109 a	potassium
Pb-K-lines	72.8;75.0	-	lead
Table 1: Characteristics of used radiation

Measurements have been done using a well-shielded HPGe-Detector spectrometer with extreme low background radiation in an energy range from 40 to 1800 keV. Detector resolution is 1.3 keV with respect to the 1.33 MeV Co-60 g-line. Due to the low intensity of natural radioactivity a measuring period of 100 h was necessary to obtain not to big statistical errors.
A typical spectrum from an 18^th century glass is shown in fig. 1. Beside the g-lines also a significant peak of Pb-K x-rays has been determined due to excitation of Pb-atoms by the high energy g-lines originating from the sample itself.

Fig 1a: Complete energy range	Fig 1b: Low energy range.
Fig 1: g - Spectrum of a historical glass carafe.

We investigated glass carafes and some covers made in the 18^th century and all used by the Saxon elector's court. The carafes had been classified into four sets based on known or supposed manufacturing works and date. All carafes did not real differ in geometry and mass, so there was no need for expended geometrical corrections.
The characteristics of the sets are listed in table 2. The covers are single pieces.

	Set 1	Set 2	Set 3	Set 4
mean height in mm	145.1	145.0	138.0	133.2
mean max. diameter in mm	98.8	100.0	100.2	98.2
mean wall thickness in mm	3.6	3.8	3.5	3.9
mean mass in g	188	204	183	207
total number of carafes	19	5	5	3
set features
	period: first third of 18th century	period: middle of 18th century	period: 1763 - near 1770	period: 18th century
	engraving: monogram Saxon elector Friedrich August II., King of Poland	engraving: monogram Saxon elector Friedrich August II., King of Poland	engraving: monogram Saxon elector Friedrich August III.	engraving:without
	manufacturer: Dresden glassworks	manufacturer: Dresden glassworks	manufacturer: supposed Glücksburg/Sa glassworks	manufacturer: Dresden or Glücksburg/Sa glassworks
Table 2: Characteristics of investigated glass carafes

Results and Conclusions

Mean values of background free peak intensities normalized to sample mass are shown in table 3. As seen, evident differences were obtained in natural activities comparing the investigated sets. Set 1, for instance, is characterized by small portions of uranium daughter products and a neglectable amount of thorium products.

	Set 1	Set 2	Set 3	Set 4
uranium	5.23	57.54	30.69	54.66
thorium	n.d.	5.69	2.75	5.66
K-40	1111.62	996.17	1169.86	1084.41
Table 3: mean values of g and x-ray yields, normalized to sample mass, in arbitrary units (n.d. = not detectable)

This low concentrations are supposed due to specially refining procedures of the origin material before or throughout the manufacturing process in the Dresden glass-works. Production of different glassworks shows considerably divergent content of radioactive nuclides, so these concentrations can be used to identify the sources of various products. Inside a set of same origin radioactive concentrations only small fluctuate (see fig. 2).

Fig 2: Natural born radioactivity and lead peak intensity for set 2 samples


One carafe was primarily also classified as set 2 product due to same artistic and geometrical characteristics as the other samples of this set. But determination of radioactivity shows that the obtained results prove a better affiliation to the glasses of set 1 (marked as sample No.5 in fig. 2).
The varying concentrations of glass originating from the Dresden glassworks within the same period can be explained by the multiple production of glasses in different pots using also different admixtures. So, for instance, in 1709 "admixture for white melting glasses, . . . colour glasses, . . . crystal mixtures" or simple "a lot of mixtures" had been contained in the Dresden glassworks stocks. Since this time up to 1750 the simple chalk crystal mixture ("poor mixture") and the preciously "English mixture" had been simultaneously produced, giving different glass qualities. The label "poor" or "ordinary" had not only been used for goblets and wine-glasses, but also for unpretentious normal glasses, bottles and small carafes.
In some of the spectra notable peaks at Pb-K-lines exist, due to excitation by g-radiation emitted by the natural born radioactivity from the sample itself. Excitation of element i by photon radiation with energy hu can be described as

Ii ~ ciµi(hv)·I0(hv)Ki·Ai(hv)

(1)

with elemental concentration c_i, mass attenuation coefficient µ_i, primary intensity I₀ and geometrical and absorption factor A_i. The factor K_i includes all atomic constants describing the excited element. Excitation is highly efficient for energies not much about the K-edge. Thus it can be assumed, that only the low energy lines of radioactive sample nuclides have to take into account for Pb-K x-ray excitation. Hence excitation energy range was counted from 90 keV to 360 keV, including the strongest Pb-214 lines. This procedure cannot be used to determine absolute concentrations. Recently produced samples with known lead concentration [2] have been used as calibration standards. Using similar geometrical arrangements Pb-concentration of historical glasses c_Pb,glass can be estimated by

cPb,glass = cPb,calib · (IPb,glass·Iexc,calib)/(IPb,calib·Iexc,glass)

(2)

with I_Pb as background free peak yield of Pb-lines and I_exc as total yield in the limited excitation energy range.

Some selected results of Pb-concentration are shown in table 4 with partly considerable amounts. Although reproducibility of values is poor some trends can be deduced. Unfortunately no details about glass mixtures used in the Dresden glassworks had been published. This know-ledge was handled as a personal secret of the glass producers. However, in 1709, 1725 and 1746 in the glassworks inventory beside Spanish soda, antimony, saltpetre,
chalk, potash, borax,tartaric stone, grey and white sand and glass scrap also always white and red lead are listed, which was intended for the above mentioned "English glass mixture" with higher lead concentration. This attribute has been on record since the death of the glass manufacturer Müller in 1734. He knew the English formula as a result of his family relationships and brought this knowledge along to Dresden in 1713 [4]. Comparable to the desired porcelain production at the beginning of the 18^th century in approximately the same period also manufacturing of "English glass ware" had been invented by order of elector August the Strong ("... Ihro Majt. des Höchst. seel. Königs . . . Befehl") in a quality, that "there are not in least differences in weight and sound compared to real English glasses". In continuation of Tschirnhaus' experiments Müller succeeded in performing this order. The analyses prove that since the first third of the 18^th century (set 2 and set 3) glass production with lead oxide admixtures had been a well mastered procedure. Results also show, that in the Glücksburg glassworks (a part of Dresden glassworks) English glass had been still durably produced up to the end of the 18^th century.

Sample	Sample features	CPbO in %
set 1	Dresden, first third 18th cent.	< 0.15
set 2	Dresden,middle of 18th cent.	1.58 ± 0.53
set 3	Glücksburg, 1763 - near 1770	1.66 ± 0.39
set 4	Dresden or Glücksburg, 18th cent.	0.91 ± 0.18
goblet	Dresden, before 1720	4.32 ± 0.65
cover	Glücksburg, middle of 18th cent.	0.21 ± 0.07
cover	Dresden, near 1730	13.8 ± 2.07
Cover	Dresden, near 1730	4,80 ± 0.95
Table 4: Lead oxide concentration of selected historical glasses

Literature

1. M. Schiekel, J. Heckel, B. Heinrich: Elemental analysis of historical glasses (in German). Wiss. Zt. Techn. Univ. Dresden, 35(1986), p. 41 - 42
2. Ch. Segebade: Principles and some applications of neutron and photon activation analysis (in German). Proc. 4^th International Conference Non-Destructive Testing of Works of Art, Berlin, 1994, p. 326 - 346
3. M. Lätzsch: Private information on elemental concentrations in lead glasses of Reichenbach/OL glassworks
4. G. Haase, D. Arnold, J.H. Pohl: Glass in English mode from the Dresden glassworks - chemical analytical investigation of Dresden glasses (in German). Annales du 10 Congrès du AIHV, Madrid, 1985, p.481 -496

© AIPnD , created by NDT.net

|Home|    |Top|