Back to Table of Contents

46716-B3
Synthesis, Coordination Chemistry, and Catalytic Studies of a Library of Water-Soluble Polydentate Phosphines Derived from 1,3,5-Triaza-7-phosphaadamantane (PTA)

Donald A. Krogstad, Concordia College

The greening of industry has caused recent research to center around the development of water-soluble phosphines and their transition metal catalysts.  One ligand that has been extensively studied is 1,3,5-triaza-7-phosphaadamantane (PTA) because it is easy to synthesize, highly water-soluble (235 mg/mL), resistant to oxidation, and air-stable. These factors have allowed PTA to be utilized as a ligand in an array of catalytic reactions such as hydroformylations, hydrogenations, and hydroaminations.  Due to the fact that the metal catalysts in many organic transformations are ligated with chelating phosphines, PTA has also recently been used as a synthon for the preparation of potentially bidentate ligands. In all of these studies, however, the chelating compounds utilized only one PTA unit.   To date, no ligands that employ 2 PTA moieties have been reported in the literature.  In an effort to expand upon the existing library of potentially bidentate, water-soluble phosphines, our group attempted to prepare the first ligands that utilize two PTA moieties.

The bis-phosphines, 1,1'-[1,2-phenylenebis(methylene)]bis-3,5-diaza-1-azonia-7-phosphatricyclo[3.3.1.1]decane dibromide (1), 1,1'-[1,3-phenylenebis(methylene)]bis-3,5-diaza-1-azonia-7-phosphatricyclo[3.3.1.1]decane dibromide (2), 1,1'-[1,3-tolylenebis(methylene)]bis-3,5-diaza-1-azonia-7-phosphatricyclo[3.3.1.1]decane dibromide (3), 1,1'-[1,3-anisolylenebis(methylene)]bis-3,5-diaza-1-azonia-7-phosphatricyclo[3.3.1.1]decane dibromide (4) and 1,1'-[1,4-phenylenebis(methylene)]bis-3,5-diaza-1-azonia-7-phosphatricyclo[3.3.1.1]decane dibromide (5)  have been easily prepared in high yields by reacting two equivalents of PTA with 1,2-bis(bromomethyl)benzene, 1,3-bis(bromomethyl)benzene, 1,3-bis(bromomethyl)-5-methyl-benzene, 1,3-bis(bromomethyl)-5-methoxy-benzene,  and 1,4-bis(bromomethyl)benzene in acetone or chloroform (Scheme 1).  The compounds each precipitated from solution as air-stable solids that needed little purification.

Scheme 1.  Synthesis of the Bis-PTA xylenes.

The bis-phosphines were characterized by elemental analysis, ESI-MS, and multinuclear NMR.  Alkylation of the N atoms caused the molecular symmetry of the PTA moieties to be reduced from that of the parent compound.  Consequently, the methylene protons of the PTA units within the phosphines were inequivalent, and in some cases, diastereotopic.  This caused the ¹H NMR of these compounds to be more complicated than that of PTA itself.  The ¹³C NMR spectra of the compounds were well resolved and collectively indicated that the PTA moieties of the bis-phosphines were equivalent on the NMR timescale. This equivalence was also observed in the ³¹P NMR as each bis-phosphine exhibited a single phosphorous resonance.

Compounds 3 and 4 were additionally characterized via single crystal X-ray diffraction.  The geometry about the P atoms of both compounds was compressed pyramidal, as evidenced by their small C-P-C bond angles.  The average tertiary nitrogen, N-C bond distance within 3 (1.460(6) Å) and 4 (1.458(8) Å) was quite comparable, and slightly shorter than those involving the quaternary nitrogen (1.530(5) and 1.533(7) Å).  The PTA moieties of 3 and 4 were orthogonal to the plane of the phenyl rings as evidenced by the average C-N-C-C(ring) torsions angles of 91.5^o (5) and 89.9^o (6) respectively.

Figure 1.  Thermal ellipsoid representations of 3 and 4.  The bromide anions and  solvate molecules have been omitted for clarity.

The primary purpose of this study was to prepare bis-phosphines that could possibly function as water-soluble ligands. Therefore, the solubility of compounds 1-5 was studied in aqueous media.  The substitution pattern of the aromatic ring had a pronounced effect on the water-solubility of the phosphines with the ortho compound 1 being very water soluble (2000 mg/mL) and the para analog (5) far less (12.5 mg/mL). The meta phosphine 2 and its tolyl analog (3) were found to have water-solubilities (810 mg/mL) that were triple that of PTA (235 mg/mL). The difference in water solubility between the ortho (1), meta (2), and para (5) substituted compounds is quite remarkable.  It is hypothesized that this is a consequence of the varied lattice enthalpies of the related systems. The anisolyl compound (4) was determined to be much less soluble (121 mg/mL) than the other meta derivatives, thus indicating that, in addition to the substitution pattern of the ring, the substituents on the ring greatly impact the water-solubility of the compounds. It is assumed that the low solubility of 4 was a consequence of the OCH₃ moiety hydrogen bonding with water to form a polymeric network.

In addition to adding five new phosphines to the library of water-soluble ligands, this study also had great impact on my research program.  First, it allowed four undergraduates to experience the thrill of preparing and analyzing molecules that had never been synthesized.  Two of the students are continuing to work with me this year, and a third went to graduate school to earn her Ph.D. in synthetic chemistry.  Second, this project was important in that it created a new collaboration between my lab and that of Dr. Paola Bergamini of the University of Ferrara in Ferrara, Italy.  The initial preparative results have been quite promising and these synthetic routes will likely lead to new green systems and a long lasting collaboration.

Back to top