Communications film, resulting in terraces of bilayer steps. The regions of arachidic acid in the mixed LB film can be removed by washing in ethanol to leave behind islands of cadmium arachidate molecules. ExperimentalLangmuir-Blodgett films of arachidic acidicadmium arachidate were prepared using a constant-perimeter barrier trough (purpose-built) located in a Class 10000 microelectronics clean room. The subphase was ultrapure water obtained from a commercial reverse osmosisideionizationiUV sterilization system. The arachidic acid (eicosanoic acid) was obtained from Sigma (99 % purity). For salt formation, CdCIp (BDH, Analar Grade) was added to the suhphase to give an overall concentration of 2.5 x lo4 M. The pH was adjusted to 5.7i0.1 by the addition of HCI (BDH, Aristar Grade) or ammonia solution (BDH, Aristar Grade). Transfer of the floating monolayers onto hydrophilic silicon wafers ((100) orientation) was undertaken at a suhphase temperature of 19i1 "C and a deposition pressure of 30 mN m-'. The dipping speed was 2 mm m i d . Transfer ratios for all the monolayers deposited were 1.0oi0.05.A Digital Instruments Nanoscope Ill atomic force microscope was used to examine the topographical nature of the arachidic acidicadmium arachidate LB film surface following dipping and after washing in ethanol (to remove the free acid). All of the high resolution AFM images were acquired in air at room temperature using the contact mode and a 1 pm x 1 pm piezoelectric scan head. A 200 pm narrow-legged silicon nitride cantilever with a small spring-constant ( k = 0.06 Nm-') was used to minimize film damage due to high contact forces. The lower resolution images were acquired in air using the tapping mode in conjunction with a 10 ym x 10 Fm piezoelectric scan head. This technique employs a stiff silicon cantilever oscillating at a large amplitude near its resonance frequency (several hundred kilohertz) which is detected by an optical beam system. AFM images are presented as unfiltered data in gray-scale and were found to be stable and unchanged over long periods of observation.

The sharpness of tips used in scanning tunneling microscopy (STM) is one factor which affects the resolution of the STM image. In this paper, we report on a direct-current (dc) drop-off electrochemical etching procedure used to sharpen tips for STM. The shape of the tip is dependent on the meniscus which surrounds the wire at the air–electrolyte interface. The sharpness of the tip is related to the tensile strength of the wire and how quickly the electrochemical reaction can be stopped once the wire breaks. We have found that the cutoff time of the etch circuit has a significant effect on the radius of curvature and cone angle of the etched tip; i.e., the faster the cutoff time, the sharper the tip. We have constructed an etching circuit with a minimum cut-off time of 500 ns which uses two fast metal–oxide semiconductor field effect transistors (MOSFET) and a high-speed comparator. The radius of curvature of the tips can be varied from approximately 20 to greater than 300 nm by increasing the cutoff time of the circuit.

No abstract

With a scanning tunneling microscope (STM) operating in vacuum, we have studied the lithographic patterning of self-assembling organosilane monolayer resist films. Where the organic group is benzyl chloride, the resist layer can be patterned with electrons down to 4 eV in energy. The patterned films have been used as templates for the electroless plating of thin Ni films. Linewidths down to ∼20 nm have been observed in scanning electron micrographs of the plated films. Still smaller features are observed in STM images of the exposed organosilane films.

The surface morphology of a surface-bound colloidal Pd(II) catalyst and its effect on the particle size of an electroless Ni deposit is examined. The deposited catalyst is found to have a broad distribution of particle sizes with the largest particles reaching approximately 50 nm in diameter. Catalyst surface coverages as low as 20% are found to be sufficient to initiate complete and homogenous metallization. The distribution of particle sizes for the electroless metal deposit, found to be a function of plating time, is broad with the maximum Ni particle size exceeding 120 nm. Results indicate controlling the size of the bound catalyst is the principal determining factor in controlling the particle size of the electroless deposit. Modification of the surface by depleting the concentration of surface functional groups capable of binding catalyst is used to shift the size distribution of bound catalyst to smaller values. A resulting three-to fourfold reduction in the particle size of the electroless deposit is demonstrated.