The plasmachemical removal process of corrosion layers is based on a reduction effect of RF hydrogen low-pressure plasma, and it is used for archaeological objects. Incrustation layers on artifact surface become brittle and porous due to plasma processing. The structure and composition of corrosion layers is changed. Therefore, it is much easier to recover the original surface of the plasma treated artifacts in contrary to those treated by conventional ways. Moreover, we can save time on invasive and thus dangerous mechanical removal of corrosion layers as for example sanding is. After plasma treatment, we can observe fine details of the original surface and memory of tools used during its manufacturing. These details are important information on the origin and manufacturing methods of the artifacts. The plasma reduction process leads to the removal of impurities from cavities as well, and a function of mechanical components of archaeological object can be restored. Moreover, chlorides can be easily removed from the corrosion layers and thus any significant post-corrosion is protected. Each archaeological object is original and it has its own “corrosion history”. First, the object had been exposed to the atmosphere for a long time. Then, it had been often placed in a tomb or grave or it otherwise got into the soil or sea. Thus, each archaeological object was exposed to different corrosion stress (humidity, composition of corrosive environment, etc.). Due to these facts, any universal way of a corroded object treatment is very difficult or even impossible to propose. In this work, the problem was solved using model samples of common metals which were treated at various plasma treatment conditions. Archaeological objects made of iron are the most common artifacts, and the typical corrosion products on iron are akaganeite, rokuhnite, and szomolnokite. These three corrosion products were created on the model samples in laboratory and then, the plasmachemical reduction was applied for their removal. The experiment was done in a Quartz cylindrical reactor with capacitive coupled RF plasma created using outer electrodes. We used discharge power from 100 W to 400 W in a continuous or pulsed regime (duty cycle of 75 %, 50 % and 25 %). Flowing plasma was created in pure hydrogen at pressure of 150200 Pa. Sample temperature was monitored by a thermocouple, and it did not exceed 200C during all these experiments. This temperature is regarded as a limit temperature for metallographic changes of archaeological iron. Higher temperature can cause destruction of archaeological iron objects. The optical emission spectroscopy of OH radical was used for the process monitoring. We focused on the monitoring of OH-radicals generated in the plasma, which are characteristic species formed by this process. Each corrosion product has a different time evaluation of generated OH-radicals, which is closely related to the degradation of a given corrosion product. Corrosion layers were analyzed before and after the plasmachemical reduction by SEM-EDX. We have found that the plasmachemical reduction is not very suitable for the szomolnokite corrosion product, which is degraded with difficulty and at high applied powers, only. However, very good removal efficiency was obtained for the rokuhnite and akaganeite corrosion.

This thesis presents results of the corrosion layers removal which could be found on the archaeological artefact surfaces. The low pressure low temperature plasma reduction was used for this purpose. Brass samples were chosen for this study. Two different ways have been used to form model corrosion layers. Several sets of corrosion layers were prepared in laboratory in two different corrosion atmospheres, namely ammonia atmosphere and atmosphere of hydrochloric acid. These samples were placed into desiccator. Small quantities of sand were added to some sets of samples so samples with sandy incrustation were prepared. The corrosion layers had been usually formed during four weeks. The second way, which was used to prepare model corrosion layer, was the natural corrosion in soil or compost. In this case, the corrosion layers had been formed approximately 2 years. The samples were treated in the low pressure (150 Pa) cylindrical Quartz reactor (90 cm long and 9.5 cm in diameter) with a pair of external copper electrodes connected via the matching network to a radiofrequency generator (13.56 MHz). The flows of working gases were set by independent mass flow controllers. Whole system was continuously pumped by the rotary oil pump which was separated from the discharge reactor by liquid nitrogen trap with aluminium chips eliminating dust and reactive species from the gas flow. Each sample was placed on a glass holder at the reactor center. Plasma was generated in pure hydrogen or in mixture of hydrogen and argon. Total flow of working gas was 50 sccm. Different ratios of gas mixture were tested, the ratio 30 sccm hydrogen and 20 sccm argon flows was the best. RF discharge was used in a continuous and pulsed regime. Pulsed mode was carried out with various duty cycle at the frequency of 1000 Hz. There were two ways of temperature monitoring. The sample temperature during the treatment was monitored by a K-type thermocouple installed inside the sample in the first case. Thermometer optical probe was connected to the sample surface by a small stainless plate and allowed continuous sample temperature monitoring in the second way. Safe object temperature for copper and copper alloys is 100–120 °C. To avoid exceeding this temperature, power control or the duty cycle in pulse mode were automatically controlled if thermometer optical probe was used. Plasma chemical treatment is based on generation of reactive atomic hydrogen in plasma discharge. The main reactions during reduction were reactions between oxygen and chloride contained in the corrosion layer and the hydrogen ions and neutral atoms generated in the plasma. These reactions create an unstable OH radical, which emits light in the region of 306–312 nm. This radiation was detected by the optical emission spectroscopy using Ocean Optics HR4000 spectrometer with 2400 gr/mm grating. Data obtained from this method were used to calculate rotational temperatures and integral intensity of OH radicals that were used for the process monitoring. Corrosion layer was not completely removed during the reduction, but due to the reactions which occur in the plasma corrosion layer became brittle and after plasma chemical treatment can be removed easily. The SEM-EDS material analyses were carried out before and after treatment of some samples. Some samples were analysed by XRD analysis. EDS analysis showed that amount of oxygen and chloride was decreased, mainly at 400 W pulse mode.

No description available.