Addition of a single electron to a neutral (uncharged) molecule generates a unique chemical species, called an anion radical (or radical anion by some) that simultaneously has a unit of negative charge and an unpaired electron. This can be accomplished, in many cases, by simply treatment with an alkali metal (which makes the alkali metal cation the counterion to the anion radical). In general, the electron will prefer (Aufbau principle) to enter the lowest energy antibonding molecular orbital (LUMO). In the context of the resulting anion radical, this same MO then becomes the SOMO (singly occupied MO of a radical species). The LUMO of the neutral molecule is therefore uniquely important among the ABMO’s because the added electron will prefer to occupy that MO and also because it becomes the SOMO of the anion radical, which (more or less) exclusively controls the distribution of the odd electron, and therefore the position of likely reactivity of the radical. Further, assuming that we are dealing with a hydrocarbon system as the neutral precursor, and that this hydrocarbon itself is nonpolar (has zero charge accumulations at all carbons), the SOMO also uniquely controls the distribution of negative charge. In these cases, the spin and charge are said to be tightly coupled, and in essence the spin and charge “travel together”, i.e., the spin and charge at each carbon are equal in absolute magnitude. This interesting result obtains because, in the assumed kind of nonpolar system, the filling of the BMO’s evidently results in Qi = 0 at each position. Therefore any charge and any spin density must arise from the population of the SOMO. Needless to say, if a polar functionality such as a ketone function is present, the filling of the BMO’s does result in non-zero charge accumulations at most atoms, so that both the SOMO and the BMO’s contribute to the overall charge densities. Nevertheless, the SOMO continues to control the spin distribution. In such systems, the spin and charge are said to be loosely coupled. They are not uncoupled, because the SOMO continues to contribute most of the charge density and all of the spin density. You have already calculated, as part of the very first problem set in Unit 1, the charge and spin densities in the butadiene anion and cation radicals, so that you can actually verify these things. Please note also that the spin and charge densities in the anion and cation radicals are identical (except for sign, in the case of charge).
Scheme 1. Formation of an anion radical from a neutral molecule by single electron acceptance.
It should be noted that any organic molecule has antibonding MO’s, so that essentially any molecule is capable of forming a corresponding anion radical. When the LUMO is of very high energy, it may be difficult to transfer an electron to the molecule by chemical means (such as an alkali metal) or even by electrochemical means (cathodic reduction). Recall that pi bonds are weaker than sigma bonds, in general. Thus, the energy of the pi BMO of ethene is not as low as the sigma BMO of a C-C sigma bond. Since there is a pairing relationship (with respect to the nonbonding level) between the BMO and the corresponding ABMO, the pi ABMO of the ethene pi bond is not as high in energy as the sigma ABMO of a C-C sigma bond (e.g. in ethane). Consequently, the addition of an extra electron to a pi bond is much easier than to a sigma bond. Nevertheless, the ABMO corresponding to the pi bond of ethene is still relatively high in energy, so that experimentally, it is difficult to generate the ethene anion radical. However, conjugation tends to decrease the energy of the LUMO of a pi system (consider butadiene vs. ethene), and correspondingly the addition of an electron becomes much easier to such conjugated systems (Scheme 2). Finally, the insertion of an electronegative atom such as oxygen has a strong energy lowering effect upon the LUMO, so that carbonyl pi bonds are especially easy to reduce to anion radicals. In general, there is a very nice linear relationship between the LUMO energy and the reduction potential of the substrate.
Scheme 2. Lowering of LUMO energies and raising of HOMO energies via conjugation and incorporation of a heteroatom.
Examples of Stable (Persistent) Anion Radicals.
If sufficient delocalization and or heteroatom presence is provided, anion radicals are not only relatively easily formed, but they can be stable enough to persist in solution in the absence of proton donors or oxygen. The blue benzophenone anion radical is very commonly seen in the organic laboratory as a drying agent for ethereal solvents such as tetrahydrofuran (THF). The idea is that small amounts of water which may be present in the solvent would quickly protonate the basic anion radical. Atmospheric oxygen would also be removed by reaction with the anion radical. Therefore if the blue color of the benzophenone anion radical persists, the solutions can be considered to be very dry and oxygen-free. The procedure for drying such a solvent is more or less as follows. Starting with a flask of already rather pure and anhydrous solvent (obtained commercially), a small amount of benzophenone is added followed by the addition of potassium metal. The initial amounts of the benzophenone anion radical will form and then disappear as a result of reaction with water and/or oxygen. After a while, the blue color will persist, showing that the solvent is now rigorously anhydrous and anaerobic. Distillation of the solvent then yields a very pure sample. Note that the benzophenone anion radical (ketyl) benefits not only from extended conjugated (over both phenyl rings) but also from the electronegative heteroatom effect.
Scheme 3. The benzophenone anion radical (ketyl): A persistent anion radical.
Another very common anion radical seen in the organic lab is the biphenyl anion radical (green). Usually this is prepared by reaction of biphenyl with lithium metal, often in THF solvent. This anion radical, and other anion radicals of aromatic systems, is often used as an electron transfer reductant. It is interesting to note the relative stability of the anion radicals of aromatic system, even though the addition of an electron to an ABMO diminishes the net bonding of the system. Evidently, some of the special stability associated with the aromaticity of the neutral substrate still remains in the anion radical.
Scheme 4. The biphenyl anion radical: A one electron reducing agent.
Stable Anion Radicals. The tetracyanoethylene anion radical has already been mentioned as an example of an isolable anion radical salt. Another example is the tetracyanoquinodimethane anion radical (Scheme 5).
Scheme 5. The TCNQ anion radical.
The Cyclooctatetraene Anion Radical. The cyclooctatetraene anion radical is a persistent anion radical in solution in the absence of air and moisture (Scheme 5). It is especially interesting for a number of reasons. First of all, its neutral precursor, cyclooctatetraene has a non-planar, tub shape. This is not especially surprising in view of the fact that the planar form would be a cyclic, conjugated system consisting of 8 pi electrons. It would, accordingly, be classified as antiaromatic, according to the 4n+2 Rule. The anion radical, however, is a 9 electron system and although not aromatic, it is not antiaromatic either. As it turns out the anion radical of this system is actually planar and a true cyclic, conjugated system.
Scheme 6. The planar cyclooctatetraene anion radical.
Another aspect of the COT anion radical is that it disproportionates to a somewhat surprising extent to the corresponding dianion and a molecule of neutral cyclooctatetraene (Scheme 6). The driving force for this can be readily seen, since the dianion, as a 10 electron system, can be considered to be aromatic.
Scheme 7. Disproportionation of an anion radical.
The reduction of 1,3-butadiene by alkali metals dissolved in liquid ammonia results in conjugate addition of hydrogen to the diene system, i.e., formation of 2-butene (cis and trans isomers). As background, we should note that alkali metals dissolved in liquid ammonia to give deep blue solutions of solvated electrons in the very polar solvent, ammonia. These electrons are available to transfer to an ABMO of an organic molecule. Although it is difficult to reduce ethene to its anion radical, the lower energy LUMO of butadiene easily accepts an electron to form the corresponding anion radical. Hopefully you have already calculated the charge densities in the 1,3-butadiene anion radical (first unit) and found that most of the negative charge resides on C1 and C4 of the diene. It would not be a surprise, therefore, to find that if protonation of this anion radical occurs, it will occur on one of these terminal atoms. The result is the formation of an allylic type radical. Since the SOMO in the allyl radical is an NBMO, this radical is easily reduced (by more solvated electrons) to the corresponding allylic anion. This anion is then protonated (by ammonia or an alcohol which has been added to the solvent) to yield 2-butene. This protonation occurs rather selectively on the terminal atom to yield 2- rather than 1-butene. The mechanism of this reduction, which is typical of the reduction of conjugated dienes (or enones) by metals in liquid ammonia, is shown in Scheme 7.
Scheme 8. Reduction of a conjugated diene with alkali metal in liquid ammonia.
In a similar way, naphthalene is reduced to the specific dihydroisomer shown in Scheme 9. Incidentally, the position of protonation of an anion radical of an alternant hydrocarbon system can be confidently predicted by a knowledge of the SOMO distribution, since this MO exclusively controls both the spin and the charge in these systems. In the naphthalene anion radical, the SOMO has the largest coefficients on the alpha positions, so that initial protonation occurs there. Reduction of the resulting radical to an anion is followed by protonation, also at the alpha position.
Scheme 9. Birch reduction of naphthalene by sodium in liquid ammonia, containing tert-butyl alcohol.
Incidentally, even benzene can be reduced to 1,4-dihydrobenzene in this way. These reactions represent examples of Birch reduction.
The reductive coupling of two ketone molecules by magnesium metal to give a 1,2-diol, often termed a pinacol coupling, is considered to take place via the anion radical of the ketone (recall that the anion radical of a ketone is called a “ketyl”). We should pause here to consider a few special aspects of ketyl anion radicals. First of all, the ketone carbonyl pi bond is highly polarized, with a large fractional negative charge on oxygen and a corresponding positive charge on carbon. This corresponds to a BMO which is more heavily weighted on oxygen than on carbon (Scheme 10). Perhaps not too surprisingly then, the ABMO shifts over to having a heavier weight on carbon. Since the ABMO will be the SOMO of the anion radical, and since the SOMO controls the spin density distribution, the radical character will be greater on carbon than on oxygen. Since, however, the neutral ketone already had much negative charge on oxygen before being converted to the anion radical, and the carbon had positive charge, the SOMO does not exclusively control the charge distribution, i.e., the spin and charge are uncoupled. The electron density which goes onto carbon from the SOMO is used in part to neutralize the positive charge on carbon, and that which goes onto oxygen further increases the net negative charge on oxygen. In effect then, this ketyl has the radical character mostly on carbon and the charge density mostly on oxygen. Radical coupling, then, occurs at the site of highest spin density, i.e., carbon.
Scheme 10. Spin and Charge Uncoupling in a Ketyl
Since magnesium has two electrons outside of a closed shell, it is a good electron transfer donor (reducing agent). In a heterogeneous reaction, it transfers an electron to the ketone, giving a ketyl. Two of these ketyl anion radicals than couple to give the magnesium salt of the double conjugate acid of the pinacol (Scheme 11).
Scheme 11. Coupling of two ketyl anion radicals in the formation of pinacols.
Anion Radical Cycloadditions
Here at UT, we have been developing some very novel anion radical cycloaddition chemistry. Recall that the formation of cyclobutanes by cycloaddition of two alkene molecules is extremely difficult and, for practical purposes, quite useless. However, it turns out that placing electrons in an antibonding MO raises the energy of the species sufficiently that such reactions are quite feasible in some cases. Intramolecular reactions such as that shown in Scheme 12 have been found to proceed quite efficiently under electrochemical reducing conditions or using persistent anion radicals of aromatic systems to generate the required anion radicals.
Scheme 12. Intramoolecular Anion Radical Chain Cyclobutanation
The mechanism proposed for these reactions is shown in Scheme 13.
Scheme 13. Mechanism of Anion Radical Chain Cyclobutanation
Examples of intermolecular anion radical cyclobutanation have also been developed in this department and are currently under exploration.