THIS PAGE IS UNDER CONSTRUCTION (March 6, 2007)

Main-Group and Main-Group Metal Chemistry 

Our work centers around main-group chemistry, the interaction of main-group elements with transition metals, and the control of such interactions.

Variable Coordination Around a Central Main-Group Element
Calixarenes are a unique class of compounds that posssess an unusual combination of features:  constraint and flexibility.  The size of the central cavity supplies the constraining feature, while the ability of the phenolic groups to rotate provides flexibility.  This is best illustrated in Scheme 1 summarizing our work on variable coordination of a single phosphorus atom within the framework of a calix[4]arene.  Initially, insertion of the phosphorus into para-Rcalix[4]arene (R = tert-butyl, H) yields the six-coordinate zwitterionic compound 1 via loss of two moles of dimethylamine.  Removal of the third mole of dimethylamine, either by heat or treatment with acid, yields the three-coordinate phosphite  2.  Methylation of 2 gives the four-coordinate 3.  Finally, deprotonation of 3 yields the five-coordinate phosphorane 4.

Scheme 1 (R = tert-butyl, H)

  



 

The size of the calix[4]arene cavity is ideal to support a single main-group atom, while the flexibility of the backbone adopts to the coordination requirements of the central atom.  This is best illustrated by examining the structures of the six-coordinate and five-coordinate derivatives, 1 and 4, respectively. 

 

In 1 (R = H), the calix[4]arene backbone adopts the cone conformation, ideal for the six-coordinate phosphorus atom.                                                                                          In 4 (R = tert-butyl), one of the  phenolic units "flips" to accommodate the five-coordinate phosphorus atom in an almost perfect trigonal bipyramidal geometry, with O(1) and O(3) at the axial positions and O(2), O(4), and C(1) at the equatorial sites.

 

Control of Ligand/Metal Interaction

The calix[4]arene appears best to support a single main-group atom.  In efforts to incorporate two atoms, i.e., a main-group (ligand) atom and a transition metal, we chose the larger calix[5]arene.  The small increase in cavity size should provide room for both atoms, while still providing some constraint to control the ligand/metal interaction.  Treatment of para-tert-butylcalix[5]arene with one equivalent of tris(dimethylamino)phosphine yields the mono-phosphorus compound 5 (Scheme 2).  Insertion of tungsten proceeds smoothly to yield complex 6.  The structure of 6 reveals that the phosphorus lone pair is pointing directly at the vacant coordination site on tungsten.  However, the P---W distance is 3.15 Å, well outside the longest bond lengths reported for P-W bonds (although a small phosphorus-tungsten coupling constant is observed in the NMR spectrum, indicative of a weak interaction).  Several factors may account for the long distance.  The calix[5]arene backbone might prevent approach of the phosphorus to the tungsten.  In addition, amido and imido ligands are known to be excellent π-donors to high oxidation-state-metals; thus, the tungsten may not want any further electron density.  If the latter is the determining factor, then replacing one of the nitrogen ligands with a ligand that places less density on the metal might allow the phosphorus to bind to the tungsten.  In fact, this is exactly what occurs.  Treatment of 6 with trifluoromethanesulfonic acid yields 7 via replacement of tert-butylamido with triflate.  The PW distance decreases to 2.74 Å (with a corresponding increase in the  phosphorus-tungsten coupling constant), indicating bond formation.   In addition, the calixarene backbone reorients to adopt the cone conformation.  We see below that sometimes a conformational change alone can lead to a drastic alteration in the ligand/metal interaction.

 

Scheme 2

 

 


 



In 6, although the phosphorus lone pair is pointing directly at the vacant coordination site on tungsten, the distance is too long for a "bond." In 7, the excellent π-donor  amido ligand is replaced with the much weaker binder triflate resulting in formation of the P-W bond.  In addition, the calixarene backbone reorients to adopt the cone conformation.  In some cases, a conformational change alone can lead to a drastic alteration in the ligand/metal interaction (see below).  

  

The importance of the calixarene geometry in controlling the phosphorus/metal interaction is illustrated when 5 is treated with tetrakis(dimethylamino)titanium (Scheme 3).  In this reaction, two isomers are formed, 8-cone and 8-1,2-alt, that differ only in the calix[5]arene conformation (with the calix[5]arene adopting an approximate cone conformation in one and an approximate 1,2-alternate conformation in the other).  The P---Ti distance in 8-cone is 3.69 Å, well outside the range of any conceivable interaction.  However, the 1.2-alternate conformation allows a significantly closer approach of the phosphorus (P---Ti distance = 2.90 Å).  Part of the reason for this decreased distance in 8-1,2-alt may be the reduced steric repulsion of the dimethylamino groups in this conformation; however, this is not the only reason, since the tungsten complex 7 exhibits the cone conformation for the calix[5]arene, and it has the shorter P---W distance compared to 6 in which the calix[5]arene adopts an approximate 1,2-alternate conformation.

 

Scheme 3  

 



In 8-cone, the P---Ti distance is 3.69 Å, well outside the range of any conceivable interaction. In 8-1,2-alt, the 1.2-alternate conformation allows a significantly closer approach of the phosphorus (P---Ti distance = 2.90 Å). 

  
Efforts are currently underway to further understand the constraints of calix[5]arenes with a single phosphorus ligand (5) as well as derivatives containing two phosphorus ligands.