Phenylene-ethynylene oligomers are molecules that have interested researchers in the field because they not only conduct electric current, but also appear to turn their conductivity on and off in observations with a scanning tunneling microscope probe. Scientists have thought that these molecules might be used as molecular switches, if their intermittent conductivity could be attributed to some internal property of the molecules that could be controlled.
Instead, physicists Ganesh K. Ramachandran, and Stuart M. Lindsay and chemist Alex Primak from Arizona State University and Theresa J. Hopson, Adam M. Rawlett and Larry A. Nagahara from Motorola Labs report that the molecules' apparent "switching" behavior is more likely the result of an unexpectedly weak bond with the layer of current-conducting gold molecules on which they are arranged. When the bond is broken, the molecules lose contact with the gold surface, and the electrical connection is turned off.
"There were a number of possible explanations for their conductance turning on and off," said Lindsay. "We eliminated that it was caused by the loss of the contact between the top probe and the molecule, or by a change in the molecule itself, so, ergo, it had to be in the contact between the molecule and the gold surface."
To test whether the intermittent on-off behavior of the molecules could be due to changes in their complex structure, the team first tested much simpler molecules known as n-alkanedithiols - chains of carbon and hydrogen atoms with hydrogen-sulfur (thiol) end segments that bond with gold contacts. The simple composition of the alkanedithiols made it extremely unlikely that any significant change in the molecules' structure could occur that would affect conductivity.
"Alkanedithiols have absolutely no internal mechanism for turning themselves on and off. We understand those molecules like our own families - they are very simple things," said Lindsay. "But we noticed that even these molecules occasionally switch themselves off."
This led to the second question of whether the random on-off behavior (known as "stochastic switching") could be due to breaking the connection through a loss of contact between the microscope probe and the molecule. The conducting molecules are scattered in a single-molecule layer of similar but non-connecting molecules, all bonded to an underlying layer of gold. If the conducting molecules bent or slumped, they might shrink below the layer of the surrounding non-conducting molecules and lose contact with the probe, like a bent yarn in a shag carpet disappearing into the carpet pile.
To test for the loss of the top contact, the team attached gold nano-particles to the top of the molecules. Larger than the molecules themselves, the gold particles could not disappear in the non-conducting layer, and would drift off if the bond with the alkanedithiol molecules was broken. The team found that the attached particles stayed in place, and, though the frequency of switching was reduced, the on-off switching behavior continued.
"Since switching occurred even with simple molecules and with a solid top contact, our conclusion was that it had to be caused by the contact between the molecule and the gold surface," said Lindsay. "In a sense, this was surprising because the assumption has been that the gold-thiol bond underneath was too strong to break.
"People said, 'if that's true, molecules would be breaking loose from the surface.' But, as a matter of fact they absolutely do do that - that's how the molecule layers are prepared. At a relatively mild temperature, some molecules break off and others come in, so in many ways it's not surprising at all," he said.
To further confirm that a break in the bottom bond was causing the switching, the team measured the rate of rate of switching at increasingly higher temperatures (which should cause increasingly less-stable gold-thiol bonds) and found that the switching behavior increased with higher temperature.
"If you believe the strength of the chemical bond that's in the literature," said Lindsay, "then nothing should happen if you gently heat the substrate - the molecules should just sit there, because the heat should not be enough to break the bond and for the molecules to move. But they do move - we know that - and they move more at higher temperatures."
The problem, according to Lindsay, is probably in the bonding behavior of gold, which is always used for the electrode material because it is easy to form films with.
"We simply need to find other materials with better bonding properties," he said. "Then we can move on and continue looking for molecules with exciting electrical properties."
The research was supported by a grant from the National Science Foundation. Stuart Lindsay is the director of the Center for Single Molecule Biophysics, a component of ASU's new Arizona Biodesign Institute. The current report is the first published research product of the institute.