Neil J. Williams

Possible Steric Control of the Relative Strength of Chelation Enhanced Fluorescence for Zinc(II) Compared to Cadmium(II): Metal Ion Complexing Properties of Tris(2-quinolylmethyl)amine, a Crystallographic, UV−Visible, and Fluorometric Study

The idea is examined that steric crowding in ligands can lead to diminution of the chelation enhanced fluorescence (CHEF) effect in complexes of the small Zn(II) ion as compared to the larger Cd(II) ion. Steric crowding is less severe for the larger ion and for the smaller Zn(II) ion leads to Zn-N bond length distortion, which allows some quenching of fluorescence by the photoinduced electron transfer (PET) mechanism. Some metal ion complexing properties of the ligand tris(2-quinolylmethyl)amine (TQA) are presented in support of the idea that more sterically efficient ligands, which lead to less M-N bond length distortion with the small Zn(II) ion, will lead to a greater CHEF effect with Zn(II) than Cd(II). The structures of [Zn(TQA)H(2)O](ClO(4))(2).1.5 H(2)O (1), ([Pb(TQA)(NO(3))(2)].C(2)H(5)OH) (2), ([Ag(TQA)(ClO(4))]) (3), and (TQA).C(2)H(5)OH (4) are reported. In 1, the Zn(II) is 5-coordinate, with four N-donors from the ligand and a water molecule making up the coordination sphere. The Zn-N bonds are all of normal length, showing that the level of steric crowding in 1 is not sufficient to cause significant Zn-N bond length distortion. This leads to the observation that, as expected, the CHEF effect in the Zn(II)/TQA complex is much stronger than that in the Cd(II)/TQA complex, in contrast to similar but more sterically crowded ligands, where the CHEF effect is stronger in the Cd(II) complex. The CHEF effect for TQA with the metal ions examined varies as Zn(II) >> Cd(II) >> Ni(II) > Pb(II) > Hg(II) > Cu(II). The structure of 2 shows an 8-coordinate Pb(II), with evidence of a stereochemically active lone pair, and normal Pb-N bond lengths. In 3, the Ag(I) is 5-coordinate, with four N-donors from the TQA and an oxygen from the perchlorate. The Ag(I) shows no distortion toward linear 2-coordinate geometry, and the Ag-N bonds fall slightly into the upper range for Ag-N bonds in 5-coordinate complexes. The structure of 4 shows the TQA ligand to be involved in pi-stacking between quinolyl groups from adjacent TQA molecules. Formation constants determined by UV-visible spectroscopy are reported in 0.1 M NaClO(4) at 25 degrees C for TQA with Zn(II), Cd(II), and Pb(II). When compared with other similar ligands, one sees that, as the level of steric crowding increases, the stability decreases most with the small Zn(II) ion and least with the large Pb(II) ion. This is in accordance with the idea that TQA has a moderate level of steric crowding and that steric crowding increases for TQA analogs tris(2-pyridylmethyl)amine (TPyA) < TQA < tris(6-methyl-2-pyridyl)amine (TMPyA).

Strong Metal Ion Size Based Selectivity of the Highly Preorganized Ligand PDA (1,10-Phenanthroline-2,9-dicarboxylic Acid) with Trivalent Metal Ions. A Crystallographic, Fluorometric, and Thermodynamic Study

The selectivity of the rigid ligand PDA (1,10-phenanthroline-2,9-dicarboxylic acid) for some M(III) (M = metal) ions is presented. The structure of [Fe(PDA(H)(1/2))(H(2)O)(3)] (ClO(4))(2).3H(2)O.(1)/(2)H(5)O(2) (1) is reported: triclinic, P1, a = 7.9022(16) A, b = 12.389(3) A, c = 13.031(3) A, alpha = 82.55(3) degrees , beta = 88.41(3) degrees , gamma = 78.27(3) degrees , V = 1238.6(4) A(3), Z = 2, R = 0.0489. The coordination geometry around the Fe(III) is close to a regular pentagonal bipyramid, with Fe-N lengths averaging 2.20 A, which is normal for a 1,10-phenanthroline type of ligand coordinated to seven-coordinate Fe(III). The Fe-O bonds to the carboxylate oxygens average 2.157 A, which is rather long compared to the average Fe-O length of 2.035 A to carboxylates in seven-coordinate Fe(III) complexes. The structure of 1 supports the idea that the Fe(III) is too small for ideal coordination in the cleft of PDA, and the structure shows that the Fe(III) adapts to this by inducing numerous small distortions in the structure of the PDA ligand. The log K(1) values for PDA at 25 degrees C in 0.1 M NaClO(4) were determined by UV spectroscopy with Al(III) (log K(1) = 6.9), Ga(III) (log K(1) = 9.7), In(III) (log K(1) = 19.7), Fe(III) (log K(1) = 20.0), and Bi(III) (log K(1) = 26.2). The low values of log K(1) for PDA with Al(III) and Ga(III) are because these ions are too small for the cleft in PDA, which requires a large metal ion with an ionic radius (r(+)) of 1.0 A. In(III) and Fe(III) (r(+) = 0.86 and 0.72 A for a coordination number (CN) of 7) are somewhat too small for the cleft in PDA but may adapt by increasing the coordination number, which increases the metal ion size, and have high log K(1) values. Very large log K(1) values are found, as expected, for Bi(III) (r(+) = 1.17 A, CN = 8), which fits the cleft quite well. Fluorescence studies show that Y(III) produces the largest CHEF (chelation enhanced fluorescence) effects, followed by La(III) and Lu(III), in the PDA complexes. Metal ions with nonfilled d or f subshells produce very large quenching of the fluorescence, as do heavy-metal ions such as In(III) and Bi(III), which have large spin-orbit coupling effects. The Al(III)/PDA complex produced an intense broad band at longer wavelength than the pi*-pi emissions of the PDA ligand, which is at a maximum at pH 6, and the possibility that this might reflect an exciplex, where one PDA ligand in the Al(III) complex pi-stacks with the excited state of a second PDA ligand, is discussed.

Metal Ion Complexing Properties of the Highly Preorganized Ligand 2,9-bis(Hydroxymethyl)-1,10-phenanthroline: A Crystallographic and Thermodynamic Study

Metal ion complexing properties of the ligand 2,9-bis(hydroxymethyl)-1,10-phenanthroline (PDALC) are reported. For PDALC, the rigid 1,10-phenanthroline backbone leads to high levels of preorganization and enhanced selectivity for larger metal ions with an ionic radius of about 1.0 A that can fit well into the cleft of the ligand. Structures of PDALC complexes with two larger metal ions, Ca(II) and Pb(II), are reported. [Ca(PDALC) 2](ClO 4) 2 ( 1) is triclinic, Pi, a = 7.646(3), b = 13.927(4), c = 14.859(5) (A), alpha = 72.976(6), beta = 89.731(6), mu = 78.895(6) degrees , V = 1482.5(8) A (3), Z = 2, R = 0.0818. [Pb(PDALC)(ClO 4) 2] ( 2) is triclinic, Pi, a = 8.84380(10), b = 9.0751(15), c = 12.178(2) (A), alpha = 74.427(3), beta = 78.403(13), mu = 80.053(11) degrees , V = 915.0(2) A (3), Z = 2, R = 0.0665. In 1, the Ca(II) is eight-coordinate, with an average Ca-N of 2.501 A and Ca-O of 2.422 A. The structure of 1 suggests that Ca(II) is coordinated in a very low-strain manner in the two PDALC ligands. In 2, Pb(II) appears to be eight-coordinate, with coordination of PDALC and four O donors from perchlorates bridging between neighboring Pb atoms. The Pb has very short Pb-N bonds averaging 2.486 A and Pb-O bonds to the alcoholic groups of PDALC of 2.617 A. It is suggested that the Pb(II) has a stereochemically active lone pair situated on the Pb(II) opposite the two N donors of the PDALC, and in line with this, the Pb-L bonds become longer as one moves around the Pb from the sites of the two N donors to the proposed position of the lone pair. There are two oxygen donors from two perchlorates, nearer the N donors, with shorter Pb-O lengths averaging 2.623 A. Two oxygens from perchlorates nearer the proposed site of the lone pair form very long Pb-O bond lengths averaging 3.01 A. The Pb(II) also appears to coordinate in the cleft of PDALC in a low-strain manner. Formation constants are reported for PDALC in 0.1 M NaClO 4 at 25.0 degrees C. These show that, relative to 1,10-phenanthroline, the hydroxymethyl groups of PDALC produce a significant stabilization for large metal ions such as Cd(II) or Pb(II) that are able to fit in the cleft of PDALC but destabilize the complexes of metal ions such as Ni(II) or Cu(II) that are too small for the cleft.

Complexation of Metal Ions of Higher Charge by the Highly Preorganized Tetradentate Ligand 2,9-Bis(hydroxymethyl)-1,10-Phenanthroline. A Crystallographic and Thermodynamic Study

The metal ion selectivity for M(III) (M = metal) ions exhibited by the highly preorganized ligand PDALC is investigated (PDALC = 2,9-bis(hydroxymethyl)-1,10-phenanthroline). The structures are reported of [Bi(PDALC)(H(2)O)(2)(ClO(4))(3)] x H(2)O (1), monoclinic, P2(1)/c, a = 12.8140(17), b = 19.242(3), c = 9.2917(12) A, beta = 91.763(2) degrees, V = 2289.9(5) A(3), Z = 4, R = 0.0428; [Th(PDALC)(NO(3))(4)] x 3 H(2)O (2), monoclinic, P2(1)/n, a = 7.876(3), b = 22.827(9), c = 12.324(5) A, beta = 94.651(6) degrees, V = 2208.4(15) A(3), Z = 4, R = 0.0669; [Cd(PDALC)(2)](ClO(4))(2) (3)), triclinic, P1, a = 7.5871(16), b = 13.884(3), c = 14.618(3) A, alpha = 74.081(2) degrees, beta = 88.422(2) degrees, gamma = 78.454(2) degrees, V = 1450.2(5) A(3), Z = 2, R = 0.0267. The Bi in 1 is best regarded as 9-coordinate, with four short bonds to the PDALC, and two short bonds to the coordinated water molecules, with three long bonds to perchlorate oxygens. The Bi-N bonds at 2.35 A are by a considerable margin the shortest Bi-N bonds to 1,10-phenanthroline (phen) type ligands, which is suggested to be due to the Bi adapting to the metal ion size requirements of PDALC. The Th(IV) in 2 is 12-coordinate, with four bonds to PDALC, and the four chelated nitrates, with close to normal bond lengths to the PDALC ligand. The Cd(II) in 3 is 8-coordinate, with Cd-N and Cd-O bonds that are similar to those found in other 8-coordinate Cd(II) complexes. The five known structures of PDALC complexes, including the three reported here, suggest that the M-N bonds to PDALC are quite easily varied in length in response to differing metal ion sizes, but that the M-O bonds are more constrained by the rigid ligand to be close to the ideal value of 2.50 A. The formation constants (log K(1)) for M(III) ions with PDALC show that for small metal ions such as Ga(III) and Fe(III), log K(1) is only slightly higher than for phen, suggesting that these metal ions are too small to coordinate to the alcoholic oxygen donors of PDALC. For larger metal ions such as Bi(III), Gd(III), Th(IV), and UO(2)(2+), log K(1) for PDALC is higher than log K(1) for phen by more than 5 log units, which stabilization is attributed to the fact that PDALC is preorganized for complexation with large metal ions with an ionic radius of about 1.0 A. The fluorescence of M(III) complexes of PDALC is discussed. PDALC free ligand gives fluorescence typical of phen ligands, with the protonated form giving a broad less intense band, and the non-protonated form of the ligand giving an intense structured set of bands. Large lanthanide ions without partially filled f-subshells, such as La(III), Lu(III), and also Y(III), give a fairly strong CHEF (chelation-enhanced fluorescence) effect, while those with partially filled f-subshells, such as Gd(III), Yb(III), and Tb(III), strongly quench the fluorescence of PDALC. A heavy element such as Bi(III) has strong spin-orbit coupling effects that act to quench the fluorescence of PDALC almost completely, which effect is enhanced by the covalence of the Bi-N bonds.

Selectivity of the Highly Preorganized Tetradentate Ligand 2,9-Di(pyrid-2-yl)-1,10-phenanthroline for Metal Ions in Aqueous Solution, Including Lanthanide(III) Ions and the Uranyl(VI) Cation

Some metal ion complexing properties of DPP (2,9-Di(pyrid-2-yl)-1,10-phenanthroline) are reported with a variety of Ln(III) (Lanthanide(III)) ions and alkali earth metal ions, as well as the uranyl(VI) cation. The intense π-π* transitions in the absorption spectra of aqueous solutions of 10(-5) M DPP were monitored as a function of pH and metal ion concentration to determine formation constants of the alkali-earth metal ions and Ln(III) (Ln = lanthanide) ions. It was found that log K(1)(DPP) for the Ln(III) ions has a peak at Ln(III) = Sm(III) in a plot of log K(1) versus 1/r(+) (r(+) = ionic radius for 8-coordination). For Ln(III) ions larger than Sm(III), there is a steady rise in log K(1) from La(III) to Sm(III), while for Ln(III) ions smaller than Sm(III), log K(1) decreases slightly to the smallest Ln(III) ion, Lu(III). This pattern of variation of log K(1) with varying size of Ln(III) ion was analyzed using MM (molecular mechanics) and DFT (density functional theory) calculations. Values of strain energy (∑U) were calculated for the [Ln(DPP)(H(2)O)(5)](3+) and [Ln(qpy)(H(2)O)(5)](3+) (qpy = quaterpyrdine) complexes of all the Ln(III) ions. The ideal M-N bond lengths used for the Ln(III) ions were the average of those found in the CSD (Cambridge Structural Database) for the complexes of each of the Ln(III) ions with polypyridyl ligands. Similarly, the ideal M-O bond lengths were those for complexes of the Ln(III) ions with coordinated aqua ligands in the CSD. The MM calculations suggested that in a plot of ∑U versus ideal M-N length, a minimum in ∑U occurred at Pm(III), adjacent in the series to Sm(III). The significance of this result is that (1) MM calculations suggest that a similar metal ion size preference will occur for all polypyridyl-type ligands, including those containing triazine groups, that are being developed as solvent extractants in the separation of Am(III) and Ln(III) ions in the treatment of nuclear waste, and (2) Am(III) is very close in M-N bond lengths to Pm(III), so that an important aspect of the selectivity of polypyridyl type ligands for Am(III) will depend on the above metal ion size-based selectivity. The selectivity patterns of DPP with the alkali-earth metal ions shows a similar preference for Ca(II), which has the most appropriate M-N lengths. The structures of DPP complexes of Zn(II) and Bi(III), as representative of a small and of a large metal ion respectively, are reported. [Zn(DPP)(2)](ClO(4))(2) (triclinic, P1, R = 0.0507) has a six-coordinate Zn(II), with each of the two DPP ligands having one noncoordinated pyridyl group appearing to be π-stacked on the central aromatic ring of the other DPP ligand. [Bi(DPP)(H(2)O)(2)(ClO(4))(2)](ClO(4)) (triclinic, P1, R = 0.0709) has an eight-coordinate Bi, with the coordination sphere composed of the four N donors of the DPP ligand, two coordinated water molecules, and the O donors of two unidentate perchlorates. As is usually the case with Bi(III), there is a gap in the coordination sphere that appears to be the position of a lone pair of electrons on the other side of the Bi from the DPP ligand. The Bi-L bonds become relatively longer as one moves from the side of the Bi containg the DPP to the side where the lone pair is thought to be situated. A DFT analysis of [Ln(tpy)(H(2)O)(n)](3+) and [Ln(DPP)(H(2)O)(5)](3+) complexes is reported. The structures predicted by DFT are shown to match very well with the literature crystal structures for the [Ln(tpy)(H(2)O)(n)](3+) with Ln = La and n = 6, and Ln = Lu with n = 5. This then gives one confidence that the structures for the DPP complexes generated by DFT are accurate. The structures generated by DFT for the [Ln(DPP)(H(2)O)(5)](3+) complexes are shown to agree very well with those generated by MM, giving one confidence in the accuracy of the latter. An analysis of the DFT and MM structures shows the decreasing O--O nonbonded distances as one progresses from La to Lu, with these distances being much less than the sum of the van der Waals radii for the smaller Ln(III) ions. The effect that such short O--O nonbonded distances has on thermodynamic complex stability and coordination number is then discussed.

Metal Ion Complexing Properties of Dipyridoacridine, a Highly Preorganized Tridentate Homologue of 1,10-Phenanthroline

DPA (dipyrido[4,3-b;5,6-b]acridine) may be considered as a tridentate homologue of phen (1,10-phenanthroline). In this paper some of the metal ion complexing properties of DPA in aqueous solution are reported. Using UV-visible spectroscopy to follow the intense π-π* transitions of DPA as a function of pH gave protonation constants at ionic strength (μ) = 0 and 25 °C of pK(1) = 4.57(3) and pK(2) = 2.90(3). Titration of 10(-5) M solutions of DPA with a variety of metal ions gave log K(1) values as follows: Zn(II), 7.9(1); Cd(II), 8.1(1); Pb(II), 8.3(1); La(III), 5.23(7); Gd(III), 5.7(1); Ca(II), 3.68; all at 25 °C and μ = 0. Log K(1) values at μ = 0.1 were obtained for Mg(II), 0.7(1); Sr(II), 2.20(1); Ba(II), 1.5(1). The log K(1) values show that the high level of preorganization of DPA leads to complexes 3 log units more stable than the corresponding terpyridyl complexes for large metal ions such as La(III) or Ca(II), but that for small metal ions such as Mg(II) and Zn(II) such stabilization is minimal. Molecular mechanics calculations (MM) are used to show that the best-fit M-N length for coordination with DPA is 2.60 Å, accounting for the high stability of Ca(II) or La(III) complexes of DPA, which are found to have close to this M-N bond length in their phen complexes.

“Straining” to Separate the Rare Earths: How the Lanthanide Contraction Impacts Chelation by Diglycolamide Ligands

The subtle energetic differences underpinning adjacent lanthanide discrimination are explored with diglycolamide ligands. Our approach converges liquid-liquid extraction experiments with solution-phase X-ray absorption spectroscopy (XAS) and density functional theory (DFT) simulations, spanning the lanthanide series. The homoleptic [(DGA)3Ln]3+ complex was confirmed in the organic extractive solution by XAS, and this was modeled using DFT. An interplay between steric strain and coordination energies apparently gives rise to a nonlinear trend in discriminatory lanthanide ion complexation across the series. Our results highlight the importance of optimizing chelate molecular geometry to account for both coordination interactions and strain energies when designing new ligands for efficient adjacent lanthanide separation for rare-earth refining.

Recommended Guanidine Suppressor for the Next-Generation Caustic-Side Solvent Extraction Process

The guanidine recommended for the Next-Generation Caustic-Side is N,N ,N -tris(3,7-dimethyloctyl)guanidine (TiDG). Systematic testing has shown that it is significantly more lipophilic than the previously recommended guanidine DCiTG, the active extractant in the commercial guanidine product LIX -79, while not otherwise changing the solvent performance. Previous testing indicated that the extent of partitioning of the DCiTG suppressor to the aqueous strip solution is significantly greater than expected, potentially leading to rapid depletion of the suppressor from the solvent and unwanted organic concentrations in process effluents. Five candidate guanidines were tested as potential replacements for DCiTG. The tests included batch extraction with simulated waste and flowsheet solutions, third-phase formation, emulsion formation, and partition ratios of the guanidine between the solvent and aqueous strip solution. Preliminary results of a thermal stability test of the TiDG solvent at one month duration indicated performance approximately equivalent to DCiTG. Two of the guanidines proved adequate in all respects, and the choice of TiDG was deemed slightly preferable vs the next best guanidine BiTABG.

Synthesis and reactivity of (benzoxazol-2-ylmethyl)phosphonic acid.

An efficient three step synthesis of (benzoxazol-2-ylmethyl)phosphonic acid (6-H(2)) is described along with IR, mass spectrometry (MS), and (1)H, (13)C, and (31)P NMR spectroscopic characterization data, and a single crystal X-ray diffraction structure determination. 6-H(2) is unstable in acidic aqueous solutions (pH < 4) undergoing ring-opening to give [(2-hydroxyphenylcarbamoyl)methyl] phosphonic acid (7-H(2)) that is characterized by IR, MS, and NMR methods. The protonation constants (pK(a)) for 7-H(2) have been measured, and crystal structure determinations for (NH(4))(7-H) and K(7-H)·DMF are described. Reactions of NaOH and KOH with 6-H(2) in MeOH/H(2)O solutions led to isolation and crystal structure determinations of the salts [Na(6-H)·H(2)O](2), K(6-H), Na(3)(6)(6-H)·H(2)O, and [K(2)(6)](2)·3H(2)O. The complexation reactions of 7-H(2) with La(III), Nd(III), and Gd(III), as a function of pH, were also examined by titrametric methods, and a model for the 1:1 anion binding with Ln(III) cations is proposed.