Weizmann Institute scientists have provided one of the answers to this question. Making simple and elegant use of a chemical theory of liquids, they developed a way to predict the minimal possible size of bipolar transistors, one of the major types of transistors commonly used in microelectronics. They then managed to manufacture such a tiny structure using the experimental semiconductor copper indium diselenide. With an inner core of just 20 nanometers (billionths of a meter) and total width of 50 nanometers - less than one-thousandth the width of a human hair - the device is five times smaller than today's smallest standard transistors of this type.

This research, reported recently in Applied Physics Letters, was performed by doctoral student Shachar Richter, working with Prof. David Cahen of the Materials and Interfaces Department, Dr. Yishay Manassen, formerly of Weizmann's Chemical Physics Department and now a professor of physics at Ben-Gurion University of the Negev, and Dr. Sidney Cohen, head of Weizmann's Surface Analysis Unit.

In his research, Richter used atomic force microscopy - a technique in which a phonograph-like stylus probes the surface of a material - to manipulate atoms in a semiconductor. Normally, such microscopes can only shift atoms on the surface of a material, but Richter, building on earlier research by Prof. Cahen, managed to move these atoms around inside the semiconductor.

Richter achieved his results by applying a voltage to the semiconductor and passing a current through the material. Aided by the slight heating produced by the current, the voltage caused atoms called dopants, which determine the material's conductivity, to be propelled in a particular direction. Even though only 100 to 200 dopants were moved in this manner, this sufficed to produce a tiny transistor. It consisted of a hemispherical layer of relatively high conductivity containing the redistributed dopants, flanked on both sides by material with different conductivity.

Next, Richter used the same microscope stylus - at low voltage - to map the conductivity of this miniature structure. Richter's new mapping method, called scanning spreading resistance, reveals the precise path that would be taken by an electric current flowing through a transistor of this type. This new type of measurement, developed independently by Belgian researchers around the time of Richter's study, promises to become an important tool for evaluating miniature electronic devices.

These findings don't necessarily mean that microelectronic devices will eventually get as small as Richter's transistor. His device, however, can serve as a valuable research tool for studying the limits of miniaturization.

Different polymers are often mixed together in liquid form and then allowed to harden in order to impart the best properties of each to the final product. During the mixing, certain types of polymers are more likely than others to move to the surface, forming a coating that thickens with time.

It is important to know which constituent is more likely to move to the outside: for example, a polymer that excels in dirt-resistance or lubrication will be of greatest benefit if it settles closer to the surface. It is also important to know whether the polymer layer at the surface will have time to thicken sufficiently before the molten mixture hardens.

A series of Weizmann Institute studies -- the latest of which appears in a recent issue of Physical Review Letters -- helps to answer these questions and may lead to better polymer materials.

Prof. Jacob Klein and his student Ullrich Steiner of the Institute's Materials and Interfaces Department have determined that the more flexible polymer chains in a mixture -- the ones easier to bend -- will be found at the surface.

They also discovered that the rate of thickening is primarily controlled by the weak forces of attraction between neutral molecules, known as van der Waals forces. Due to the weakness of these forces, the thickening is extremely slow -- it proceeds in accordance with a mathematical formula involving approximately the power of 4. If, for example, a wetting layer grows to a certain thickness after one hour, then it will double its thickness after 16 hours (2 to the 4th power), and double once again after 256 hours (4 to the 4th power), and so on -- until the mixture hardens and this process stops.

These studies were carried out at the Weizmann Institute's 3 MV Van de Graaff particle accelerator, where the nuclei of polymers were bombarded by atoms propelled to high speeds. An analysis of the reactions occurring within these nuclei provides detailed data about the formation of the polymer surface coating and its properties.

These findings advance the basic understanding of processes taking place in polymer mixtures. They may also help optimize the surface properties of materials made up of different polymer components.