Formation of 1 ∶ 1 and 2 ∶ 2 complexes of Ce(III) with the heteropolytungstate anion α₂-[P₂W₁₇O₆₁]¹⁰⁻, and their interaction with proline. The structure of [Ce₂(P₂W₁₇O₆₁)₂(H₂O)₈]¹⁴⁻

Abstract

The “1 ∶ 1” complex of Ce³⁺ and α₂-[P₂W₁₇O₆₁]¹⁰⁻ is formed in an equimolar solution of Ce(NO₃)₃ and K₁₀[P₂W₁₇O₆₁], but attempts to isolate a solid yield salts of the 1 ∶ 2 complexes first described by Peacock and Weakley in 1971. With a 3-fold excess of Ce³⁺ however, a crystalline ammonium salt (anion 1) could be isolated in 90% yield based on tungstate. Structural analysis of these crystals (monoclinic, C2/c; a = 46.1124(6); b = 12.7451(1); c = 26.4624(2) Å, β = 93.570(1)°, Z = 4; 18856 independent reflections, R1 = 0.0799; wR2 = 0.1594) revealed a head-to-head dimeric anion with each 9-coordinate Ce(III) cation attached to four oxygens that surround the lacuna of one tungstophosphate anion, a terminal O(W) of the second tungstophosphate, and four water molecules. Phosphorus-31 NMR spectra of solutions of 1 revealed two sets of resonances, A and B, the relative intensities of which could be fitted to an equilibrium 2A ≡ B, with K = 1.46 ± 0.04 M⁻¹ at 23 °C. Further NMR analysis suggests a scheme in which 1, which is in rapid exchange with the “monomer” (2), is also in slow exchange with a labile associate of Ce³⁺ and the 1 ∶ 2 complex. The exchange process between 1 and 2 is responsible for the non-observance of one of the expected 9 lines in the tungsten-183 NMR spectrum and an exchange-broadened P-NMR line. Addition of proline to form a ternary complex [Ce(P₂W₁₇O₆₁)(proline)(aq)]⁷⁻(K_form = 4.5 ± 0.1 M⁻¹ from ¹H-NMR) restores the missing W-NMR resonance.

---

Introduction

The chemistry of polyoxometalates continues to attract interest as a result of their realized and potential applications in fields as disparate as catalysis, materials science, structural biology, and medicine.¹ In recent years attention has been focussed on very large polyoxoanions which span the interface between discrete molecular species and extended lattices, and which can demonstrate novel host–guest chemistry, and unusual optical, magnetic and other properties.² The structures of such large polyoxoanions can be mentally decomposed into smaller discrete “building blocks” with various linking atoms or groups. Simple, early examples of such composite polyoxometalates are complexes such as [Mn(Nb₆O₁₉)₂]¹²⁻³ and [Eu(PW₁₁O₃₉)₂]¹¹⁻,⁴ in which the linkers are transition metal and lanthanide cations. The latter type of complexes, first reported by Peacock and Weakley, are easily synthesized from the lacunary polyoxoanion “ligand” and the lanthanide cation which adopts a square prismatic 8-coordinate coordination geometry. Large numbers of such complexes with lanthanide and actinide(4+) cations and a variety of polyoxometalate ligands have been characterized since the first reports.^2,5 Much research has been directed at the use of the tungstophosphate anion [P₂W₁₇O₆₁]¹⁰⁻ (α₂ isomer) for radioactive waste treatment via sequestration of transuranium elements.⁶

Although Peacock and Weakley had noted the existence of 1 ∶ 1 complexes such as [Ce(PW₁₁O₃₉)(H₂O)_x]⁷⁻, and these were later characterized by spectroscopic and electrochemical methods,^7,8 the first structural studies revealed that the anions assembled into linear polymeric arrays upon crystallization.⁹ We have recently examined the 1 ∶ 1 complex of the metastable α₁ isomer of [P₂W₁₇O₆₁]¹⁰⁻ with Ce³⁺ and showed that it crystallizes in the form of a head-to-head dimer, [{Ce(H₂O)₄(P₂W₁₇O₆₁)}₂]¹⁴⁻.¹⁰ Very recently, Francesconi et al. have described the structure of [{Eu(H₂O)₃(P₂W₁₇O₆₁)}₂]¹⁴⁻.¹¹

This paper describes the synthesis, and solution and solid state characterization of the 1 ∶ 1 complex of the α₂ isomer of [P₂W₁₇O₆₁]¹⁰⁻ with Ce^III. Owing to the existence of the stable 1 ∶ 2 complex, the solution equilibria are more complicated than for the α₁ system (for which no 1 ∶ 2 complex has yet been detected) and evidence for two dimers is presented. The formation of labile complexes with proline has been detected by NMR.

Experimental

Materials

All chemicals were reagent grade and used as supplied. Water was purified by passage through a purification train. The following lacunary anion salts were prepared according to published methods and were identified by infrared spectroscopy and ³¹P NMR: α₂-K₁₀P₂W₁₇O₆₁,¹² α₂-Li₁₀P₂W₁₇O₆₁,¹³ and α-Na₁₂P₂W₁₅O₅₆.¹²

Synthesis of (NH₄)₇[Ce(P₂W₁₇O₆₁)]·nH₂O

This complex was first isolated as a minor by-product from the synthesis of [Ce₄(OH₂)₉(OH)₂(P₂W₁₆O₅₉)₂]¹⁴⁻.¹⁴ More efficient syntheses were subsequently developed.

Method 1. A hot (80 °C) solution of α₂-K₁₀P₂W₁₇O₆₁·xH₂O (3.5 g, ca. 0.72 mmol) in water (40 mL) was added dropwise over a period of 45 min to a hot (80 °C) solution of Ce(NO₃)₃·6H₂O (1.25 g, 2.8 mmol) in 10 mL water. Formation of an oily phase (dark orange–red) at the bottom of the beaker was observed upon complete addition of the polyoxotungstate solution. After stirring the solution was stirred at 80 °C for an additional 5 min, 30 mL of 4 M NH₄Cl solution was added. The reaction mixture turned into a clear orange solution which was stirred at the same temperature for an additional 5 min before the solution was cooled to room temperature (pH 2.8). The formation of crystalline material occurred after a few hours and the final product (3.12 g, 90 mol%) was collected the next day.

Method 2. To a solution of α-Na₁₂P₂W₁₅O₅₆ (4.23 g, ca. 1.0 mmol) dissolved in 100 mL 1 M LiCl solution (pH 7.7), a solution of Na₂WO₄·2H₂O (0.66 g, 2.0 mmol) in 5 mL water was added (pH 8.3). After stirring the solution vigorously for 5 min (pH 8.3), a solution of Ce(NO₃)₃·6H₂O (0.86 g, 2.0 mmol) dissolved in 1 M HOAc (6 mL) was added dropwise to the reaction mixture to form a light yellow–orange suspension. The suspension was heated in a water bath from room temperature to 85 °C over the course of 10 min, and kept at this temperature for an additional 25 min. While stirring, 2.0 ml 3 M KCl were added to the suspension. The mixture was then filtered with suction once and gravimetrically twice to obtain a clear orange solution (pH 5.0). The solution was subsequently heated in a hot water bath for 1 additional min, and 35 mL of a 4 M NH₄Cl solution were added while stirring (pH 5.0). The solution was allowed to stand at room temperature for one week to obtain a orange crystalline product (2.50 g, 52 mol%).

The products of the three preparations gave identical P-NMR spectra. Found: N, 1.94; H, 1.44; Ce, 3.04; P, 1.17; W, 63.11; Cl, 0.10; K, 0.20%. Calc. for (NH₄)_6.75K_0.25[Ce(P₂W₁₇O₆₁)]·0.27H₂O·0.15NH₄Cl: N, 1.96; H, 1.66; Ce, 2.84; P, 1.26; W, 63.40; Cl, 0.10; K, 0.20%. IR (ν/cm⁻¹): 1085 (s), 1055 (w), 1024 (w), 1016 (w) 943 (s), 923 (s), 918 (s), 812 (vs), 783 (s). Ce^IV/III redox potential (literature⁸): 0.654 V (0.651 V) vs. Ag/AgCl.

Crystal structure determination

Single crystal X-ray analyses were performed on a Bruker-Siemens SMART CCD single crystal diffractometer equipped with an Mo-Kα anode and graphite monochromator (λ = 0.71073 Å). The orange crystal (0.17 × 0.12 × 0.04 mm) was mounted on a glass fiber under mineral oil to prevent water loss, and placed in a nitrogen stream at 173(2) K. The final unit cell was calculated using a least squares refinement of reflections culled from the entire data set. The data was corrected for Lorentz and polarization effects. An empirical absorption correction based on equivalent reflections¹⁵ was applied as incorporated in the SADABS program.¹⁶ All structures were solved using direct methods and refined against F² using the routines included in the SHELXTL-PC software suite.¹⁷ Nitrogen atoms of ammonium cations were modeled as oxygen atoms because the former could not be distinguished from the latter. K, P, W and Ce atoms were refined anisotropically; oxygen atoms were refined isotropically. Hydrogen atoms were not included in the refinement model. (NH₄)₁₃K[{Ce(P₂W₁₇O₆₁)}₂](H₂O)₁₈, M = 9204.95, monoclinic, a = 46.1124(6), b = 12.7451(1), c = 26.4624(2) Å, β = 93.570(1), U = 15522.0(3), T = 173(2) K, space group C2/c, Z = 4, μ(Mo-Kα) = 25.825 mm⁻¹, 84148 reflections measured, 18856 unique (R_int = 0.1610) which were used in all calculations. The final wR(F²) was 0.1594.

CCDC reference number 169242.

See http://www.rsc.org/suppdata/dt/b1/b105967j/ for crystallographic data in CIF or other electronic format.

Physical measurements

Phosphorus-31 NMR was measured on a Varian Mercury 300 MHz spectrometer (P resonance frequency 121.472 MHz) interfaced to a Sun Microsystems workstation with the following parameters: spectral width, 10000 Hz; acquisition time, 6.4 s; pulse delay, 5 s; pulse width, 34.1°. All spectra were referenced to external 85% H₃PO₄, and the temperature was controlled by a VT temperature control unit. Tungsten-183 NMR spectra were recorded on a Bruker AM 300WB spectrometer (W resonance frequency 12.504 MHz), equipped with an Aspect 3000 computer with the following parameters: spectral width, 8064.516 Hz; acquisition time, 1.0158 s; pulse delay, 0.5 s; pulse width 45° tip angle. All spectra were referenced to external 2 M Na₂WO₄. Infrared spectra were recorded with a Nicolet FT-7000 spectrometer as KBr pellets. Cyclic voltammetry was measured at ambient temperature on a BAS100A system. A glassy carbon working electrode (diameter 3mm), a platinum wire counter electrode and an Ag/AgCl reference electrode (3M NaCl) were used. Approximate formal potential values (E_1/2) were calculated from the cyclic voltammograms as the average of the cathodic and anodic peak potentials for each corresponding oxidation and reduction wave.

P-NMR Titration

Titration of α₂-[P₂W₁₇O₆₁]¹⁰⁻ with Ce(NO₃)₃ was carried out as follows: A concentrated solution of α₂-[P₂W₁₇O₆₁]¹⁰⁻ was prepared by dissolving ca. 7.20 g of α₂-Li₁₀P₂W₁₇O₆₁·22H₂O in 10.0 mL D₂O, and 0.5 mL of the solution was placed in an NMR tube. To this solution, Ce(NO₃)₃ (1.50 M) in D₂O was added in 5 or 10 μL increments, and the ³¹P-NMR spectrum was measured at 23 °C. On adding Ce(NO₃)₃, new peaks belonging to the [Ce(α₂-P₂W₁₇O₆₁)₂]¹⁷⁻ complex (4) appeared, until the ratio of α₂-[P₂W₁₇O₆₁]¹⁰⁻ and Ce(NO₃)₃ reached 2 ∶ 1. Plotting the integration ratio change between the peaks due to α₂-[P₂W₁₇O₆₁]¹⁰⁻ and [Ce(α₂-P₂W₁₇O₆₁)₂]¹⁷⁻ against the amount of Ce(NO₃)₃ added gave the exact concentration (1.50 M) of the α₂-[P₂W₁₇O₆₁]¹⁰⁻ solution.

Results and discussion

Synthesis

The complex was isolated as a mixed potassium–ammonium salt from the reactions involving the precursor anions, α₂-[P₂W₁₇O₆₁]¹⁰⁻ or α-[P₂W₁₅O₅₆]¹²⁻ (methods 1 and 2, respectively).

α₂-[P₂W₁₇O₆₁]¹⁰⁻ + Ce(NO₃)₃ (3 eq.) → [Ce(α₂-P₂W₁₇O₆₁)]⁷⁻ (Method 1)

α-[P₂W₁₅O₅₆]¹²⁻ + WO₄²⁻ (2 eq.) + Ce(NO₃)₃ (2 eq.) → [Ce(α₂-P₂W₁₇O₆₁)]⁷⁻ (Method 2)

The direct method (method 1) between the monovacant lacunary anion α-[P₂W₁₇O₆₁]¹⁰⁻ and Ce(NO₃)₃ gave excellent isolated yields (90%). The trivacant lacunary derivative α-[P₂W₁₅O₅₆]¹²⁻ also gave the desired product under appropriate conditions in the presence of additional tungstate, although excess amounts of Ce(NO₃)₃ were necessary. The formation of an oily phase during the synthesis and its disappearance upon addition of NH₄Cl (method 1) suggested the importance of the counter ions for this synthesis. Attempts to optimize the synthesis for the correct (1 ∶ 1) stoichiometry were unsuccessful, and considerable amounts of the 1 ∶ 2 complex (4) were formed.

X-Ray structural analysis

As illustrated in Fig. 1, the complex exists in dimeric form (1) in the solid state. The asymmetric unit contains a 1 ∶ 1 [(H₂O)₄Ce(α₂-P₂W₁₇O₆₁)]⁷⁻ unit, and the complete anion is generated by a center of symmetry. The linkage between the two 1 ∶ 1 units occurs in a head to head fashion, via two Ce–O–W bridges formed between each cerium atom and one terminal oxygen atom belonging to the W₂O₁₀ duplet of the adjacent α₂-P₂W₁₇O₆₁ anion. Each cerium atom is nine-coordinate (monocapped square antiprism), with four sites occupied by one anion unit (acting as a tetradentate ligand), four by water molecules, and the ninth site by a terminal oxygen of the other anion. The average Ce–O bond length for the bonds formed with the ligand‘s oxygens in the lacuna (2.48 Å) is essentially equal to the sum of the Ce–O ionic radii for an 8-coordinate Ce³⁺ and 2-coordinate O²⁻ (2.49 Å)¹⁸ and to the average Ce–O bond length in the [Ce(α₂-P₂W₁₇O₆₁)₂]¹⁷⁻ complex (2.46 Å).¹⁹ The bond length for the Ce–O bond joining the two halves of the anion (Ce–O2A, 2.57 Å) is comparable to the overall average Ce–OH₂ bond distance (2.56 Å) and to the sum of the ionic radii for a 9-coordinate Ce³⁺ and 2-coordinate O²⁻ (2.55 Å).¹⁸ The W–O framework of the anion does not show unusual features and the W–O and P–O distances are consistent with previous crystallographic reports of complexes formed with the α₂-P₂W₁₇O₆₁ anion.^19,20

Fig. 1 (left) Polyhedral representation of [Ce(α₂-P₂W₁₇O₆₁)(H₂O)₄]₂¹⁴⁻ in the ammonium salt (white circles: cerium, grey circles: water oxygens) and (right) ball-and-stick representation of selected atoms (phosphorus atom is black).

NMR analysis

The P-NMR spectrum of a solution of the isolated ammonium salt (20 mM in D₂O) exhibits two pairs of resonances, A₁ (−17.61 ppm), A₂ (−14.57 ppm), and much weaker B peaks at −14.81 and −14.06 ppm, see Fig. 2(a). In order to understand this spectrum a series of measurements was made on solutions of Li₁₀[α₂-P₂W₁₇O₆₁] with increasing amounts of Ce(NO₃)₃, see Fig. 3. Four resonances are observed when 0 < Ce/P₂W₁₇ < 0.5, two for unreacted [P₂W₁₇O₆₁]¹⁰⁻ at −6.70 (P₁) and −13.39 ppm (P₂) and two for the 1 ∶ 2 complex [Ce(P₂W₁₇O₆₁)₂]¹⁷⁻ at −13.90 and −13.93 ppm. The 1 ∶ 2 complex is the only detectable species present when Ce/P₂W₁₇ = 0.5. When 0.5 < Ce/P₂W₁₇ < 1.0 the two NMR peaks gradually transform into the B resonances and the A resonances appear and increase in intensity. These results demonstrate that the species responsible for the B resonances is in fast exchange with the 1 ∶ 2 complex, but not with the species responsible for the A resonances. That “A” and “B” are connected by a slower exchange process was shown by 2D (EXSY) spectroscopy (see ESI). The relative intensities of the A and B peaks at Ce/P₂W₁₇ = 1.0 are concentration dependent, see Fig. 2(b) and (c), and are consistent with a “monomer(A)–dimer(B)” equilibrium, characterized by K = 1.46 ± 0.04 M⁻¹ and ΔH = −18.2 ± 0.8 kJ mol⁻¹, (Fig. 4 and ESI). Although it is tempting to identify B with the dimer revealed by the crystal structure, this does not explain the fast exchange with the 1 ∶ 2 complex and slow exchange with the monomer, [Ce(P₂W₁₇O₆₁)(H₂O)_x]⁷⁻.²¹ We therefore propose equilibria involving four species, as shown in Scheme 1—the crystal dimer (1) and the “monomer” (2) in fast exchange, the 1 ∶ 2 complex (4) and an “ion-pair” of 4 with Ce³⁺ (3) also in fast exchange, and a “slow” exchange equilibrium between 3 and 1, 2. This scheme is further supported by the following W-NMR data.

Fig. 2 P-NMR spectra of [Ce(α₂-P₂W₁₇O₆₁)]⁷⁻: (a) 20 mM of isolated complex at 23.0 °C (40 scans), (b) 150 mM at 23.0 °C (8 scans), (c) 15 mM at 23.0 °C (128 scans), and (d) 150 mM at 66.8 °C (8 scans) of in situ-prepared complex {[α₂-P₂W₁₇O₆₁]¹⁰⁻ + Ce(NO₃)₃}.

Fig. 3 Chemical shift variation during the titration of α₂-[P₂W₁₇O₆₁]¹⁰⁻ with Ce(NO₃)₃ (open circles and squares: B₂ and B₁ resonances, closed circles and squares: A₂ and A₁ resonances). Broken lines denote stoichiometry of the 1 ∶ 2 and 1 ∶ 1 complexes.

Fig. 4 Plot of the integrated intensity of the B₁ resonance versus the square of the intensity of the A₁ resonance in D₂O at 23.0 °C. The sample was prepared by mixing Li₁₀(α₂-P₂W₁₇O₆₁) and Ce(NO₃)₃ in a 1 ∶ 1 ratio (initial concentration 150 mM). The slope (monomer–dimer equilibrium constant) is 1.46 ± 0.04 M⁻¹ (R² = 0.998).

Scheme 1 Suggested equilibria in D₂O. Those represented by horizontal arrows are labile on the NMR timescale.

The W-NMR spectrum of a saturated solution of the ammonium salt (∼20 mM 1) shows eight resonances (2 ∶ 2 ∶ 2 ∶ 1 ∶ 2 ∶ 2 ∶ 2 ∶ 2), Fig. 5(a). Attempts to generate a more soluble lithium salt by ion-exchange led to the formation of large amounts of the 1 ∶ 2 complex [spectrum shown in Fig. 5(e)], but the spectrum of the ammonium salt could be reproduced from an equimolar mixture of Li₁₀[α₂-P₂W₁₇O₆₁] and Ce(NO₃)₃, Fig. 5(b). The “missing” line in this spectrum is assigned to W1,W2 [Fig. 1(b)], which are anticipated to have very different chemical shifts in a static dimer structure, but which become equivalent under monomer–dimer exchange at a rate, resulting in an undetectable coalesced line. Evidence for the intermediate exchange rate, is also provided by the broad A₁ resonance (from the phosphorus adjacent to the cerium atoms) in the P-NMR spectrum, which narrows as the temperature is increased, Fig. 2(b) and (d).

Fig. 5 W-NMR spectra (see Table 1 for chemical shift values) of (a) [Ce(α₂-P₂W₁₇O₆₁)]⁷⁻ (isolated complex ca. 20 mM), 230000 scans, (b) in situ-prepared [Ce(α₂-P₂W₁₇O₆₁)]⁷⁻ (50 mM Li₁₀(α₂-P₂W₁₇O₆₁) + {50 mM Ce(NO₃)₃}, 40000 scans, (c) (b) + DL-proline (20 eq.), 44000 scans, (d) (b) + L-proline (20 eq.), 52000 scans, (e) lithium salts of [Ce(α₂-P₂W₁₇O₆₁)₂]¹⁷⁻ (1 g in 3 mL D₂O), 46000 scans. The broadness of the lowest fieldline is attributed to slow rotational interconversion of syn and anti isomers of the complex [see ref. 5(a)].

Table 1 W-NMR chemical shifts for spectra illustrated in Fig. 5 (the integration ratios are shown in parentheses)

a Splitting of virtual “doublet” resulting from loss of mirror symmetry in the AA complex.
a		121	−135	−174	−176	−209	−210	−215	−226
		(2W)	(2W)	(2W)	(1W)	(2W)	(2W)	(2W)	(2W)
b		120	−136	−176	−177	−210	−211	−216	−227
		2W	2W	2W	1W	2W	2W	2W	(2W)
c	304	118	−135	−172	−175	−209		−215	−224
	(2W)	(2W)	(2W)	(2W)	(1W)	(4W)		(2W)	(2W)
d	304	118	−135	−172	−175	−208	−210	−215	−223
	(2W, 33 Hz^a)	(2W, 29 Hz^a)	(2W, 31 Hz^a)	(2W, 19 Hz^a)	(1W)	(2W, 6 Hz^a)	(2W)	(2W)	(2W)
e	323	145	−136	−171	−179	−211	−213	−217	−226
	(2W, br.)	(2W)	(2W)	(2W)	(1W)	(2W)	(2W)	(2W)	(2W)

---

We have shown elsewhere that amino acids (AA) form weak complexes with [Ce(α₁-P₂W₁₇O₆₁)(H₂O)_x]⁷⁻ and that these may be used to discriminate between the enantiomers of the heteropolyanions.¹⁰ When D,L-proline is added to a solution containing [Ce(α₂-P₂W₁₇O₆₁)(H₂O)_x]⁷⁻ both P- and W-NMR spectra are modified, the A₁ resonance is narrowed²¹ and the “missing” tungsten resonance is observed [Fig. 5(c)]. If L-proline is added, several of the W resonances are split as a result of the loss of mirror symmetry in the AA complex.²² These results imply that attachment of the amino acid to the monomer impedes the monomer–dimer exchange. The formation constant of the proline–polyanion complex was estimated to be 4.5 ± 0.1 M⁻¹ from ¹H NMR spectra (ESI). This value is comparable to those found for the corresponding complexes of the α₁ isomer¹⁰ and for AA complexes of Eu^III chelates.²³

Conclusions

Isolation of the “1 ∶ 1” complex (1, in dimeric form) of Ce³⁺ and the α₂ isomer of [P₂W₁₇O₆₁]¹⁰⁻ was achieved by the use of a 3-fold excess of Ce³⁺. Solutions of 1, or an equimolar mixture of Ce³⁺ and [P₂W₁₇O₆₁]¹⁰⁻, are involved in equilibria linked to the stable 1 ∶ 2 complex. This behavior stands in contrast to that of mixtures of Ce³⁺ and the α₁ isomer of [P₂W₁₇O₆₁]¹⁰⁻ which yield only a 1 ∶ 1 complex in rapid exchange with the dimeric form.¹⁰ There is no convincing evidence to date to suggest that 1 ∶ 2 complexes of any trivalent lanthanides with α₁-[P₂W₁₇O₆₁]¹⁰⁻ can be formed, and this has been attributed to steric factors.^24,25 Addition of proline to solutions of 1 prevents exchange between monomer and dimer forms by binding to the Ce³⁺ centre. The formation constant of the proline complex is comparable to that found for other amino acid complexes of lanthanides.

Acknowledgements

This work was supported in part by the US Department of Energy (grant no.: DE-EF07-96ER14695), National Science Foundation (CHE9727417), Mitsubishi Chemical Co., and the Department of Chemistry at Georgetown University. The authors would like to thank Dr Knut Wassermann and Dr Michael H. Dickman for helpful discussions, and expert technical suggestions. We are grateful to Prof. Robert E. Bachman for training in and assistance with X-ray crystallography, and Prof. Angel C. de Dios and Dr Veeradej Chynwat for assistance with NMR spectroscopy.

References and notes

1. (a) M. T. Pope, Heteropoly and Isopoly Oxometalates, Springer-Verlag, Berlin, 1983; (b) M. T. Pope and A. Müller, Angew. Chem., Int. Ed. Engl., 1991, 30, 34; (c) Y. Izumi, K. Urabe and M. Onaka, Zeolite, Clay, and Heteropoly Acid in Organic Reaction, Kodansha, Tokyo, 1992, 99; (d) C. L. Hill and C. M. Prosser-McCartha, Coord. Chem. Rev., 1995, 143, 407; (e) T. Okumura, N. Mizuno and M. Misono, Adv. Catal., 1996, 41, 113; (f) Chem. Rev., special edition on polyoxometalates, ed. C. L. Hill; 1998, 98; (g) R. Neumann, Prog. Inorg. Chem., 1998, 47, 317.
2. A. Müller, F. Peters, M. T. Pope and D. Gatteschi, Chem. Rev., 1998, 98, 239.
3. C. M. Flynn, Jr. and G. D. Stucky, Inorg. Chem., 1969, 8, 333; C. M. Flynn, Jr. and G. D. Stucky, Inorg. Chem., 1969, 8, 335.
4. R. D. Peacock and T. J. R. Weakley, J. Chem. Soc. A, 1971, 1836.
5. (a) J. Bartis, S. Sukal, M. Dankova, E. Kraft, R. Kronzon, M. Blumenstein and L. C. Francesconi, J. Chem. Soc., Dalton Trans., 1997, 1937; (b) Q.-H. Luo, R. C. Howell, M. Dankova, J. Bartis, C. W. Williams, W. DeW. Horrocks, Jr., V. G. Young, Jr., A. L. Rheingold, L. C. Francesconi and M. R. Antonio, Inorg. Chem., 2001, 40, 1894.
6. (a) D. A. Malikov, M. S. Milyukova and E. V. Kuzovkina, Sourem. Metody Razd. I Opred. Radioaktiv. Elem., M., 1989, 99 from Zh. Khim., 1990, abstr. no. 1V233; D. A. Malikov, M. S. Milyukova and E. V. Kuzovkina, Chem. Abstr., 1990, 112, 241827m; (b) N. Y. Kremlyakova, A. P. Novikov, B. F. Myasoedov and N. V. Katargin, J. Radioanal. Nucl. Chem., 1990, 145, 183; (c) M. S. Milyukova, D. A. Malikov, E. V. Kuzovkina and B. F. Myasoedov, J. Radioanal. Nucl. Chem., 1986, 104, 81; (d) Y. M. Kulyakov, T. I. Trofimov, V. Y. Frenkel, I. A. Lebedev and B. F. Myasoedov, J. Anal. Chem. (Transl. of Zh. Anal. Khim.), 1981, 36, 1691; (e) Y. M. Kulyako, I. A. Lebedev, V. Y. Frenkel, T. I. Trofimov and B. F. Myasoedov, Sov. Radiochem. (Engl. Transl.), 1981, 23, 671; (f) N. P. Molochnikova, V. Y. Frenkel, B. F. Myasoedov and I. A. Lebedev, Sov. Radiochem. (Engl. Transl.), 1982, 24, 250; (g) N. P. Molochnikova, V. Y. Frenkel and B. F. Myasoedov, J. Radioanal. Nucl. Chem., 1988, 121, 409; (h) M. S. Milyukova, N. S. Varezhkina and B. F. Myasoedov, Sov. Radiochem. (Engl. Transl.), 1990, 32, 361; (i) N. P. Molochnikova, V. Y. Frenkel and B. F. Myasoedov, Radiokhimiya, 1989, 31, 65; (j) J. M. Adnet, P. Brossard and J. Bourges, Proc. Int. Conf. Technol. Expo. Future Nucl. Syst.: Emerging Fuel Cycles Waste Disposal Options 1993, 1993, 2, 1008; J. M. Adnet, P. Brossard and J. Bourges, Chem. Abstr., 1994, 120, 146793d; (k) A. M. Fedoseev and I. G. Tananaeuv, Radiokhimiya, 1994, 36, 422.
7. J. P. Ciabrini and R. Contant, J. Chem. Res.(M), 1993, (S)391, (M)2720.
8. N. Haraguchi, Y. Okaue, T. Isobe and Y. Matsuda, Inorg. Chem., 1994, 33, 1015.
9. M. Sadakane, M. H. Dickman and M. T. Pope, Angew. Chem., Int. Ed., 2000, 36, 2914.
10. M. Sadakane, M. H. Dickman and M. T. Pope, Inorg. Chem., 2001, 40, 2715.
11. Q. H. Luo, R. Howell, J. Bartis, M. Dankova, W. DeW. Horrocks and L. Francesconi, Pacifichem 2000, Honolulu, HI, Dec. 14–19, 2000 abstr. INOR 1078.
12. R. Contant, Inorg. Synth., 1990, 27, 104.
13. J. Bartis, Y. Kunina, M. Blumenstein and L. C. Francesconi, Inorg. Chem., 1996, 35, 1497.
14. A. Ostuni and M. T. Pope, C. R. Acad. Sci., Ser. IIc: Chim., 2000, 3, 199.
15. R. Blessing, Acta Crystallogr., Sect. A, 1995, 51, 33.
16. G. M. Sheldrick, SADABS, Absorption Correction Method, University of Göttingen, Germany, 1996.
17. G. M. Sheldrick, SHELXTL 97, Program for the Refinement of Crystal Structures, release 97–2, Bruker Analytical Instruments, Madison, WI, USA, 1997.
18. R. D. Shannon, Acta Crystallogr., Sect. A, 1976, 32, 751.
19. A. Ostuni, M. Sc. Thesis, Georgetown University, Washington D. C., USA, 1998.
20. V. N. Molchanov, L. P. Kazanskii, E. A. Torchenkova and V. I. Simonov, Sov. Phys. Crystallogr. (Engl. Transl.), 1979, 24, 96.
21. The peak-widths at half-height of [Ce(α₂-P₂W₁₇O₆₁)]⁷⁻ (ca. 10 mM in D₂O at 22 °C) were 6.29 Hz and 3.06 Hz for the peaks at −17.40 and −14.44 ppm, respectively. In the presence of 16 eq. of L-proline, we observed two peaks at −17.56 and −14.59 ppm with peak-widths at half-height of 3.78 and 2.80 Hz, respectively.
22. No line-splitting is observed in the presence of D,L-proline, since the (enantiomeric) adducts interconvert rapidly on the NMR timescale.
23. (a) J. Kido, Y. Okamoto and H. G. Brittain, J. Org. Chem., 1991, 56, 1412; (b) R. Hulst, N. K. De Vries and B. L. Feringa, J. Org. Chem., 1994, 59, 7453.
24. J. Bartis, M. Dankova, J. J. Lessmann, Q.-H. Luo, W. DeW. Horrocks, Jr. and L. C. Francesconi, Inorg. Chem., 1999, 38, 1042.
25. However, 1 ∶ 2 complexes with U⁴⁺ and Th⁴⁺, and the mixed-ligand species [Ce(α₁-P₂W₁₇O₆₁)(α₂-P₂W₁₇O₆₁)]¹⁷⁻ have been isolated and structurally characterized, see ref. 19 and A. Ostuni, R. E. Bachman, A. K. Jameson and M. T. Pope, manuscipt in preparation.

---

Footnote

† Electronic supplementary information (ESI) available: representative P-31 EXSY spectrum for a solution of “1 ∶ 1” complex. Plot of lnKvs. 1/T for the equilibrium species A (1, 2) and B (3, 4). Determination of formation constant of proline complex; plot of inverse cerium-induced chemical shift of Hα versus [proline]/[1]. See http://www.rsc.org/suppdata/dt/b1/b105967j/

---