M. Járó - L. Költő szerk.: Archaeometrical research in Hungary (Budapest, 1988)

Analysis - GEGUS Ernő, BORSZÉKI János: Investigation of archaeological metal findings by a laser-microspectral analysis method and characterization of results using pattern recognition methods

always used a radiation source with cross-spark excitation in solving our tasks. The energy added with a medium-voltage spark discharge enhances the energy of the micro-plasma, thus the analysis lines of trace elements appear in the spectrum, too, but evidently at different detection limits for each element; in general with a content of about one hundredth or one thousandth per cent. In addition to metals, also phosphorus and arsenic, and antimony may be detected but only up to a limit of one tenth- one hundredth per cent. However, the method is not suitable for the detection of carbon (the cross-electrode material is carbon, too!), sulphur, selenium, oxygen, or halides. Laser-microspectral analysis is not limited to the analysis of electrically conducting samples [15], It was also used in our investigations to identify the materials of Roman fresco pigments, ores, slags, ceramics, etc. The control of the absorption of laser radiation in nonconducting materials and their vaporization is much more difficult to regulate than in the case of metal samples, but details of this question are beyond the framework of this publication. The conditions of laser-microspectral analysis suitable for investigating archaeological metal findings are set out in the following table, based on our experince with the instruments LMA—

1 and PGS-2 (Zeiss, Jena).Our experience can also be interpreted for other similar apparatus (e.g. LMA—

10). Conditions of investigations Laser source: type LMA—

1 (Zeiss, Jena), ruby resonator, controlled with a passive Q-switch; in the general case d=0.25 mm (for surface investigation d=2 mm); pumping light: xenon lamp 1000 Ws, resonator circuit U = 2kV, C = 508 /iF, L = 127 /iH. Cross-spark discharge: carbon electrodes 5 mm in diameter, turned at 20° cone, changed for each sample; electrode gap 0.9 mm, electrodes 1 rnrn over the sample; resonator circuit U = 2.3 kV, C = 2.5 uF, L = 125 [Ai. Registragion of spectra : plane-grating spectrograph of Ebert mounting type PGS—

2 (Zeiss, Jena), grating blazed for UV range (280 nm, 1st order), 651 grooves/mm, rec.lin.disp. 0.76 nm/mm; slit width 0.020 mm;quartzcylindrical lense for increasing luminous intensity, focused for the 220—380 mn range; illumination onto the slit using an achromatic lense f=75.8 mm, slit height 11 mm; emulsion: fine-grain, half-tone (type Agfa-Gevaert Scientia 34B50 or ORWO WU-2); determination of gradation using laser spectrum of iron sample (15 laser shots integrated), 3-step slit filter, 3-lense illumination without quartzcylindrical lense; developing of emulsion in Kodak D—

19, 20°C, 4 min. Evaluation of spectra: estimation of blackenings of analysis lines in spectrum projector, or their measurement in micro-densitometer, calculation of logarithmic intensities (Y and AY) using ß-transformation of Török—

Zimmer, and background correction. Analysis lines (nm): Ag 338.2 and 328.0 (in silver samples 255.3 and 272.2) Al 3082 (characteristic for soil) As 234.9 Au 267.6