Cultural relics suffer from corrosion and damage caused by the fragile composition of the material and the surface pollutants resulting from environmental changes. Laser cleaning, characterized by high efficiency, environmental friendliness, and strong controllability, is an effective method for removing surface pollutants without harming the relics, rendering it suitable for cleaning various types of cultural relics. Recognizing the advantages of ultrashort pulse lasers in the realm of cultural relic cleaning, along with the critical role of the laser wavelength in this process, two primary research focuses have emerged: picosecond and multi-wavelength laser cleaning of cultural relics. This paper addresses the requirements for applying laser cleaning to stone cultural relics. Comprehensive numerical simulations and experiments were conducted on the removal of surface contaminants using picosecond lasers at wavelengths 1064, 532, and 355 nm. Concurrently, a variety of real-time and offline detection methods were employed in a coordinated manner to analyze the effects of picosecond laser cleaning at these three wavelengths, considering aspects such as sample surface morphology, roughness, and elemental content. The ultimate goal was to determine the most suitable wavelength parameters for cleaning stone cultural relics.

To address the specific demands for laser cleaning of stone cultural relics, in this study, a systematic investigation is conducted through numerical simulations and experimental research on the surface pollutants of stone cultural relics treated with a picosecond laser at three wavelengths: 1064, 532, and 355 nm. A two-temperature model is applied to simulate the interaction between picosecond laser pulses of the three wavelengths and surface contaminants, considering differences in the ablation process and depth at the same laser energy density. Building on this foundation, the damage threshold of the picosecond laser at three wavelengths and the cleaning thresholds for surface pollutants on stone cultural relics under specific conditions are measured. In the experiment, a coaxial charge-coupled device (CCD) is applied for real-time monitoring and control of the laser-cleaning process. The surface roughness and elemental content before and after laser cleaning are measured using a confocal microscope and a fluorescence spectrometer, respectively.

The damage thresholds were 0.26±0.06, 1.45±0.06, and (2.1±0.08)J/cm² at the wavelengths of 1064, 532, and 355 nm, respectively (Fig. 5), and the cleaning thresholds were 0.198±0.033, 0.573±0.114, and (0.739±0.249)J/cm² respectively under specific conditions (Fig. 7). For an energy density of 2 J/cm² pulse, the laser ablation depths were respectively 171, 106, and 62 μm for 1064, 532, and 355 nm laser pulses, and in numerical simulations, the ablation depths were 164, 104, and 58 μm respectively for the same energy density of 2 J/cm² pulse (Fig. 4). The experimental results are similar to those obtained by numerical simulations. Thermal ablation plays a major role in the surface ablation of contaminant materials by picosecond pulsed laser at this energy density, whereby the sample surface undergoes continuous cleaning utilizing picosecond lasers at these three wavelengths. Real-time monitoring via a CCD system was utilized for online observation until no alterations in surface contaminants were detected. Subsequently, a comparative analysis was performed on various factors, including surface morphology, roughness, and elemental content after cleaning. The objective was to evaluate the variations in cleaning efficiency among lasers of different wavelengths. The surface morphology (Fig. 8), roughness, and elemental content of the pollutants before and after laser cleaning at different wavelengths were compared and analyzed. The experimental results show that the surface roughness of the sample after laser cleaning at the three wavelengths of 1064, 532, and 355 nm was 13.71, 30.816, and 20.789 μm, respectively (Fig. 9). The content of S, the main component of the surface contaminants after laser cleaning at the three wavelengths, was 0.41%, 4.09%, and 1.15%, respectively (Table 4). The Fe content on the surface of the sample after laser cleaning at the three wavelengths was 0.42%, 2.13%, and 4.16%, respectively (Table 4).

The results obtained from experiments and numerical simulations are approximate. When irradiating the surface with a picosecond pulse laser at the energy density of 2 J/cm², thermal ablation plays a primary role in ablating contaminants from the surface of the material. The cleaning thresholds, efficiency, surface roughness, and elemental content were used for a comparative analysis of laser cleaning at the three wavelengths of 1064, 532, and 355 nm, thereby demonstrating that the 1064 nm laser had the smallest damage and cleaning thresholds under specific conditions, as well as the highest cleaning efficiency. The surface roughness of the sample was the smallest after 1064 nm wavelength laser cleaning, and the contents of S and Fe, the main components of the surface contaminants, were the least. Considering various factors, including the laser cleaning threshold, cleaning efficiency, post-cleaning surface roughness, and contaminant element content, the 1064 nm wavelength laser demonstrated the lowest cleaning threshold, highest efficiency, and most thorough cleaning. Overall, its cleaning performance surpassed that of lasers with 532 nm and 355 nm wavelengths under the same parameters.