This is a report on a Early-Concept Exploratory Research (EAGER) project on whether a quenched cooled Na-NH₃ solution can be a high temperature superconductor. 

A superconductor conducts electricity without any resistance and consumes no energy when it is used in electrical transmission lines or in electronic devices. In a normal (metallic or metallic compound) conductor, electrons are responsible crrent flow that entails resistance. When some of these normal conductors are cooled below a critical or transition temeprature, Tc, superconductivity sets in. The mechanism responsible for the transition, according to the Bardeen-Cooper-Schrieffer (BCS) theory, publsihed in 1957, is that individual electrons, mediated by quantized lattice vibrations or phonons, form Cooper pairs. The Cooper pairs behaves as bosons and obey Bose-Einstein statistics, which according to quantum mechanics can flow without any resistance. Superconductivity was found in dozens of metals and metallic compounds prior to 1986.  The highest Tc of these conventional superconductors was found to be ~ 30K or -243 C. The most common application of these conventional supercondcutor is the fabrication of superconducting magnets that are central for MRI machines and particle accelerators. Owing to the low Tc, these magnets are usually submerged in liquid helium.   

In 1986, another class of superconductors, materials containing copper oxides and known as cuprates high Tc superconductos was found. To date, cuprates with Tc as high as 140K has been found. This is important since a more economincal and renewable cryogen, liquid nitrogen with a boiling temeprature of 77K can be used to cool this class of superconductors.

Interestingly, Richard Ogg of Stanford reported in 1946 the observation of superconductivity in quenched cooled NaNH₃ solutions with a Tc of 190K! When a Na atom is dissolved in ammonia ( NH₃), a Na ion with a positive charge, Na⁺ ,and a solvated electron e^-  are found. Both Na⁺ and e^- are self-trapped in cavities ( or bubbles) inside the ammonia liquid. At low Na concentrations, the solution is light blue and non-conducting, at high Na concentrations the solution is bronze and conducting. At intermediate concentrations and below 230K, the solution phase separates into coexisting conducting and no-conducting liquids. What Ogg reported is that if the solution at above 230K is quickly quenched into the solid without undergoing liquid-liquid phase separation, the resultant solid show evidence of trapped magntetic flux, a signature of superconductivity. He saw evidence of trapped flux in 7 out of 200 quenched cooled samples. Ogg interpreted this purported superconductivity as a consequence of Bose-Einstein condensation of a pair of (solvated) electrons. It is remarkable that Ogg proposed this model of electron pairing 11 years before the BCS theory. 

Between 1946 and 1978 there were many attempts to search for superconductivity in the solid NaNH₃ solution. Most of these experiments reached negative conclusions but at least two found possible evidence of superconductivity in electrical resistance measurements. 

In order to reach a firm conclusion on the veracity of Ogg's claim, we carried out three different experiments.

In the first experiment, we improved upon the quench cooling protocol.  We introduced the liquid solution into glass capillaries with an inner diameter of only 0.18 mm and we made standard four electrode resistance and contact-free resistivity measurements while plunging into liquid nitrogen. Owing to the very small diameter, we were able to cool the capillary with solution down to 77K much more quickly, which we estimate to be ~20 msm a factor of 10 faster than prior studies. We found very small but non-zero resistance  in the quenched cooled samples. The very small ( less than 10^-4 ohm) resistance is probably due to that of a filmentary sodium metal precipitated from the solution upon cooling into the solid phase.

In the second experiment, we splat-cooled droplets of liquid NaNH₃ onto a sapphire disk that was kept at 25K. In this experiment presuurized liquid NaNH₃ was pushed  through a 40 μm aperture of a pulse valve that was opened for 5 ms. This micro-jet liquid shot through a screen with 20 μm and landed onto the sapphire disk that was anchored at the cold stage of closed cycle refrigerator. Platinum electrical leads were deposited on the disk for resistance measurements. We estimate the droplets were quenched from room temperature ( liquid ) to 25K ( solid) in ~ 10^-5 seconds, the solid droplets show high rather than zero resistance. 

In the third experiment, we infused the liquid solution into a porous Vycor glass cylinder with pores of only 7 nm in diameter. Inside the porous glass, liquid-liquid phase separation was found to be suppressed down to 77K, however contactless resistivity measurements of the sample also show high instead of zero resistance. 

To summarize, in spite our extensive effort, we failed unforutnaltey, to find any evidence of superconductivity in the NaNH₃ system.

Last Modified: 08/16/2018
Modified by: Moses H Chan