Silicon retains conductivity at ultra-low charge levels 01.03.2020

Researchers from the American National Institute of Standards and Technology (NIST) have come up with a new method for measuring the mobility of charged particles in silicon, which, if not overturned, then significantly expanded the understanding of charge transfer processes in semiconductors.

The method proposed by the scientists made it possible to carry out the most sensitive measurements of the speed of movement of an electric charge in silicon, and this is an indicator of its effectiveness as a semiconductor. As a result, the new method will make it possible to more accurately assess the effect of certain dopants on the silicon conductivity and will form the basis for improving the characteristics of semiconductor devices. This is a chance to improve the performance of the chips almost for nothing only through a better understanding of the processes. Carry out tuning, so to speak.

Traditionally, the mobility of electrons and holes in silicon was measured by the Hall method. This method assumes that contacts are soldered on a sample of silicon (semiconductor) to pass an electric current. The disadvantage of this method is that defects or impurities appear at the soldering points, which introduce distortions into the measurement results.

For the purity of the experiment, scientists from NIST used a non-contact method. The silicon sample was first exposed to light of low intensity in the form of ultrashort pulses of visible light, and then the sample was irradiated with radiation pulses in the far infrared or microwave range. Weak visible light produced a photodoping effect on silicon: charged particles appeared in the silicon layer in the form of electrons and holes.

Visible light, for obvious reasons, could not penetrate into the thickness of silicon. For this purpose, the photodoped sample was irradiated with terahertz radiation (in the far infrared range), for which silicon is transparent. And the more charged particles in the sample, the more light penetrates or is absorbed by the sample. It is important to note here that for a more accurate measurement of the electron mobility in the sample, its thickness should have been quite large, up to 1 mm. This ruled out the influence of defects on the sample surface on measurements.

However, the number of electrons and holes "introduced" by visible light in the sample had to be as small as possible in order to lower the sensitivity threshold during measurements. Usually, for this, the sample was irradiated with one photon, but in the case of a thick sample, one photon knocked out insufficiently charged particles in silicon. A way out was found in irradiating the sample with two photons of visible light. After that, terahertz radiation freely passed through the sample with a minimum number of charged particles in the bulk of the material. According to scientists, the threshold of sensitivity was reduced by a factor of 10 from 100 trillion charge carriers per cm2 to 10 trillion.

As soon as the threshold of sensitivity was lowered, the surprising became clear. The mobility of electrons in silicon turned out to be able to grow even to a very rarefied state of carriers in the material, which no one suspected before. Actually, the mobility itself turned out to be 50% higher than previously thought. For a control check, a similar experiment was carried out with gallium arsenide (GaAs), also a photosensitive semiconductor. It was found that the mobility of charge carriers in this material continues to grow as their density decreases. The carrier density limit measured by the new method turned out to be about 100 times lower than previously thought.

In the far or not so far future, semiconductors will be able to operate at very low charge levels. At least the theoretical limit has been pushed far enough. These are highly sensitive solar panels, and single-photon detectors (hello to quantum computers!), ultra-efficient electronics and much more.