mospheric contamination. The preparation method was ordinary spin-coating in a laboratory atmosphere.A forthcoming publication will elucidate the growth mode of the same system with a different application technique, i.e. dip-coating. In this case, i.e. with no centrifugal acceleration present, there is no layer-by-layer growth.
ExperimentalThe chemical surface constitution of the silicon wafers was investigated by XPS (VG ESCALAB MK2 and 22Oi-XL) and AES (Perkin Elmer PHI 660). The wafer surface and the film thickness of the applied organic films were monitored [8] by means of the ellipsometer Auto EL 11, Rudolph Res. The morphology of the wafers and films was imaged with the Nanoscope 111 Multimode AFM (Digital Instruments) in the Tapping Mode with minimized force in air by means of silicon cantilevers (Nanosensors, L.O.T.Orie1) having a nominal tip radius of less than 10 nm. Each tip was characterized before and after imaging by means of calibration samples to avoid artifacts in imaging.The prepolymer was dissolved in l,l,l-trichloroethane to allow the application out of solution. In order to obtain different mean film thickness, the concentration was varied from 0.002 to 0.2 mg/mL). For spin-coating, the wafers were mounted horizontally on the spin-coater (laboratory atmosphere), fully wetted with the prepolymer solution, and after a pause of 3 s, accelerated to So00 rpm. During the rotation time of 15 s the film suddenly changed color, indicating the evaporation of the solvent. The convolution between AFM tip and sample (W. Hanrieder, R.Mock, H. Meixner, Ultramicroscopy 1992,4244,169) will not hamper the determination of heights as long as access to the height levels to he compared is given. This procedure neglects the convolution between AFM tip and sample (Ref.[lo]), which results in an over-estimation of the area fraction of protrusions and in an under-estimation of the area fraction of depressions.