**6. Conclusions**

Two techniques related to ultrasonic cleaning of Si wafers and sonochemical modification of Si, SiGe and a-Si/SiGe surfaces in hydrocarbon solutions of chloroform (CHCl3) and dichloromethane (CH2Cl2) are outlined.

In spite of our lack of knowledge of the exact sonication mechanisms even in distilled water, this research field can be considered to be among potential candidates to develop a new class of environmental friendly cleaning steps in siliconbased technologies. Some progress has recently been made in understanding a unique potential capability of sonicated water in Si wafer cleaning processes. The underlying mechanisms related to the fundamental properties of cavitation and bubble implosion events, the role of a thin interphase layer between the bubble and the surface placed in the sonicated liquid can offer new far-reaching implications and importance for heterogeneous liquid–solid systems.

It is demonstrated that organic particle contaminants are effectively removed during the kHz-frequency sonication of crystalline Si wafers in distilled water over the first 40–60 min. When ultrasonically processing the wafers for treatment times less than ≈60 min at the peak acoustic intensity of about 400 W/cm<sup>2</sup> , the dangling bonds at the air/oxide and oxide/wafer interface can be activated. That affects barriers of the free carrier migration at the interfaces, as revealed by the current– voltage curves, and acts as recombination centers, accelerating the surface photovoltage decays. A healing of the bonds may occur at longer cleaning times (from 60 to 120 min) with a partial recovery of the interfaces and a consequent reversing of the observed changes. The potential of using distilled water in environmental friendly and non-toxic ultrasonic cleaning step in crystalline Si wafer preparation is addressed.

In fact, the studies described above do not reveal information about the full complexity of subsurface defect distribution effects. Therefore, there remains a wide number of uncertainties, e.g., the fundamental problem of whether or not the ultrasonic processing exploiting high acoustic powers is capable of promoting effective cleaning without surface deterioration effects.

To improve the photovoltaic response of Si wafers, SiGe and amorphous silicon (a-Si)/SiGe surfaces, a sonochemical treatment in hydrocarbon solutions of chloroform (CHCl3) and dichloromethane (CH2Cl2) can be employed. The use of the sonochemical reaction slows down the observed surface photovoltage decay and enhances its magnitude in SiGe and a-Si/SiGe thin layers grown on Si. The average surface-integrated photovoltage and decay time can increase up to 50%. This effect is not observed in distilled water, indicative of the fact that CH-containing radicals can lead to the observed improvements. It is suggested that the effect can be explained as follows. The hydrocarbon solution is decomposed and produces hydrocarbon chains, which are then decomposed further away into hydrogen and carbon. The reactive Si dangling bonds revealed on the surface of Si, a-Si or SiGe alloy layers are saturated by the hydrocarbon species to passivate the surface.

More work needs still to be done beyond the description of a very few links that have been highlighted above. In particular, the following experiments could pave the way for new mechanisms of surface passivation, activation of the interphase regions dangling bonds as well as cleaning of surfaces due to the ultrasonic processing.
