Our research involves looking at the factors that control the stability of molecules and reactive intermediates. When we understand what causes stability, we can try to engineer into a molecule some of those "stabilizing factors" which will make it more likely that we can synthesize a molecule. Part of understanding stability is understanding instability and that is where we put our efforts.

One of the primary concepts of organic chemistry is that molecules that are aromatic are particularly stable. The flip side of course is that antiaromatic molecules, the antithesis of aromatic molecules, are very unstable. These molecules are very hard to study because of this instability. They don’t stay around long enough to characterize. We have discovered a class of molecules that are both antiaromatic and capable of being studied. They are called fluorenyl cations and consist of a fluorenyl ring with a positive charge which has another positively charged group attached to it (R⁺) as shown below.

The fluorenyl ring was anticipated to be antiaromatic because it contains a 5-membered ring with 4 p-electrons, however the benzene rings, which are fused to the 5-membered ring, might dominate the behavior of the fluorenyl cation, making it aromatic, or at least much less antiaromatic. We have discovered that the fluorenyl cation of 1 is antiaromatic and that we can "titrate" the amount of antiaromaticity by varying the nature of R⁺.

Derivatives of 1 are prepared by oxidation of appropriate neutral precursors as shown below. They have been characterized by the paratropic (upfield) shifts

 of the protons of the ring in the NMR spectrum. We see that if R⁺ is very electron-poor, the fluorenyl cation is antiaromatic. We can calculate other measures of antiaromaticity, such as magnetic susceptibility exaltation or via nucleus independent shift calculations (NICS), and see a linear relationship between the experimental evaluation of antiaromaticity, ¹H NMR shifts, and these calculated properties. The calculations confirm the antiaromaticity of 1 and also suggest that one consequence of having a positively charged substituent (R+) on the fluorenyl cation is to force electron delocalization throughout the fluorenyl ring system, forcing the entire ring system to be antiaromatic. This is in direct contrast to fluorenyl cations without positively charged substituents.

The antiaromaticity of the fluorenyl ring may be affected by its cationic substituent in at least two ways. It appears that the substituent is perpendicular to the plane of the fluorenyl ring, as shown below.

This allows donation of electron density into the "empty" p atomic orbital of the fluorenyl cation, and is consistent with the upfield shift of the carbon bearing the "empty" p atomic orbital. This sigma donation is just another form of hyperconjugation, so a side benefit of our studies is the examination of factors that affect hyperconjugation. With the calculations comes some fairly good structural information, so we are looking at how these effects are transmitted through the dications.

Alternatively, the effect of the cationic substituent might be transmitted simply through the magnitude of the positive charge on the cationic substituent. That is, if R was very electron-poor, the cationic carbon to which it was bonded would be very electron-poor, very positive. The effect of having that positive charge attached to one of the carbons of the fluorenyl system would be to force more complete delocalization of the electrons of the fluorenyl system. The greater the positive charge, the greater the delocalization.

Up until now, our only experimental measure of antiaromaticity has been the paratropic shift of the protons in the NMR spectrum. We are beginning to explore "measuring" the stability of these derivatives through electrochemistry. If it is hard to make the dications electrochemically, that is, a very positive potential is needed to remove the second electron, the magnitude of the potential is a measure of how unstable it is. We can observe these highly reaction dications via cyclic voltametry if we use ultra-microelectrodes and fast scan rates. Our preparation of dications in which R is phenyl substituted with electron donating and withdrawing groups in the meta and para positions has shown a linear relationship between the redox potential for formation of the dication and proton chemical shift and calculated NICS.

Finally, aromaticity/antiaromaticity has been evaluated in terms of the degree of bond length alternation, with aromatic species showing little bond length alternation. We will probably not be able to determine bond lengths through crystal structures, but we can calculate the geometries of the dications with a reasonable degree of assurance. The basis for our confidence in the calculated structures lies in the agreement between experimental chemical shifts and those calculated for the calculated structures. NMR shifts are very sensitive to changes in geometry. This method is called the Harmonic Oscillator Measure of Aromaticity and our initial calculations on the dications for which we have experimental data suggest some interesting results.

We have begun to extend our investigations to indenyl systems (2). These systems

have been predicted to have increased antiaromaticity in comparison to fluorenyl systems. However, when we prepared dications of indenylidene fluorenes, in which R⁺ in 2 was a fluorenyl cation, we found that the fluorenyl system was actually more antiaromatic.

When we attempted to prepare bis-indenylidene dications, which contain two indenyl cations, we found that the dications cyclized. We believed that spacers would prevent this cyclization by moving the ring systems further apart and we recently reported the antiaromaticity of bisfluorenylidene dications with spacers, demonstrating that these dications maintained their antiaromaticity.

 Finally, we have turned our attention to antiaromatic dianions. We have very recently prepared dianions of both bixanthylidene and of tetrabenzon[7.7]fulvalene, with the tetrabenzo[7.7]fulvalene showing the greatest antiaromaticity.

We have also prepared both the dication and dianion of tetrabenzo[5.7]fulvalene, interesting systems in which reduction of the dication to the dianion results in the fluorenyl system changing from antiaromatic to aromatic and vice versa for the tropylium system.

During my sabbatical leave at the University of Oregon in 2008-09, I will also be exploring the potential materials science properties of these species.