Synthesis and coordination chemistry of phosphine oxide decorated dibenzofuran platforms

Article abstract of DOI:10.1021/ic300301d

A four-step synthesis for 4,6-bis(diphenylphosphinoylmethyl)dibenzofuran (4) from dibenzofuran and a two-step synthesis for 4,6-bis(diphenylphosphinoyl) dibenzofuran (5) are reported along with coordination chemistry of 4 with In(III), La(III), Pr(III), Nd(III), Er(III), and Pu(IV) and of 5 with Er(III). Crystal structure determinations for the ligands, 4·CH₃OH and 5, the 1:1 complexes [In(4)(NO₃)₃], [Pr(4)(NO ₃)₃(CH₃CN)] ·0.5CH₃CN, [Er(4)(NO₃)₃(CH₃CN)]·CH₃CN, [Pu(4)Cl₄]·THF and the 2:1 complex [Nd(4)₂(NO ₃)₂]₂(NO₃)₂· (H₂O)·4(CH₃OH) are described. In these complexes, ligand 4 coordinates in a bidentate POP′O′ mode via the two phosphine oxide O-atoms. The dibenzofuran ring O-atom points toward the central metal cations, but in every case it is more than 4 A from the metal. A similar bidentate POP′O′ chelate structure is formed between 5 and Er(III) in the complex, {[Er(5)₂(NO₃)₂](NO ₃)·4(CH₃OH)}_0.5, although the nonbonded Er...O_furan distance is reduced to ～3.6 A. The observed bidentate chelation modes for 4 and 5 are consistent with results from molecular mechanics computations. The solvent extraction performance of 4 and 5 in 1,2-dichloroethane for Eu(III) and Am(III) in nitric acid solutions is described and compared against the extraction behavior of n-octyl(phenyl)-N,N- diisobutylcarbamoylmethyl phosphine oxide (OφDiBCMPO) measured under identical conditions.

Full text of DOI:10.1021/ic300301d

Article
pubs.acs.org/IC
Synthesis and Coordination Chemistry of Phosphine Oxide
Decorated Dibenzofuran Platforms
Daniel Rosario-Amorin,† Eileen N. Duesler,† Robert T. Paine,*,† Benjamin P. Hay,‡ Lætitia H. Delmau,‡
Sean D. Reilly,§ Andrew J. Gaunt,*,§ and Brian L. Scott∥
†Department of Chemistry and Chemical Biology, University of New Mexico, Albuquerque, New Mexico 87131, United States
‡Chemical Sciences Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, Tennessee 37831, United States
∥
§Chemistry Division and Materials Physics and Applications Division, Los Alamos National Laboratory, Los Alamos,
New Mexico 87545, United States
S
* Supporting Information
ABSTRACT: A four-step synthesis for 4,6-bis(diphenylphosphinoylmethyl)-
dibenzofuran (4) from dibenzofuran and a two-step synthesis for 4,6-bis-
(diphenylphosphinoyl)dibenzofuran (5) are reported along with coordination
chemistry of 4 with In(III), La(III), Pr(III), Nd(III), Er(III), and Pu(IV) and
of 5 with Er(III). Crystal structure determinations for the ligands, 4·CH3OH
and 5, the 1:1 complexes [In(4)(NO3)3], [Pr(4)(NO3)3(CH3CN)]·0.5CH3CN,
[Er(4)(NO3)3(CH3CN)]·CH3CN, [Pu(4)Cl4]·THF and the 2:1 complex
[Nd(4)2(NO3)2]2(NO3)2·(H2O)·4(CH3OH) are described. In these com-
plexes, ligand 4 coordinates in a bidentate POP′O′ mode via the two phosphine
oxide O-atoms. The dibenzofuran ring O-atom points toward the central metal
cations, but in every case it is more than 4 Å from the metal. A similar bidentate
POP′O′ chelate structure is formed between 5 and Er(III) in the complex,
{[Er(5)2(NO3)2](NO3)·4(CH3OH)}0.5, although the nonbonded Er···Ofuran
distance is reduced to ∼3.6 Å. The observed bidentate chelation modes for 4 and 5 are consistent with results from molecular
mechanics computations. The solvent extraction performance of 4 and 5 in 1,2-dichloroethane for Eu(III) and Am(III) in nitric
acid solutions is described and compared against the extraction behavior of n-octyl(phenyl)-N,N-diisobutylcarbamoylmethyl
phosphine oxide (OΦDiBCMPO) measured under identical conditions.
chelates. Furthermore, examples of the NOPOP′O′ class display
useful solvent extraction properties for trivalent f-block element
cations.38−41 Our continuing interest in developing new chelating
conditions, especially for f-block metal cations, has led us to explore
the possibility of using phosphine oxide and methyl phosphine
oxide donor centers to decorate the 4,6-DFB fragment, and
determine if the resulting ligands utilize the DBF ring O-atom in
coordination interactions with hard cations. The syntheses for these
ligands, molecular modeling of potential coordination interactions,
selected coordination chemistry toward hard acceptor ions, In(III),
Ln(III) (Ln = La, Pr, Nd, and Er) and Pu(IV), and a survey of the
liquid−liquid solvent extraction properties with Eu(III) and
Am(III) are described in this report.
INTRODUCTION
■
A variety of organic platforms have been utilized to append
donor groups in geometrically favorable arrangements so that
the resulting molecules generate either strongly chelating or
hemilabile coordination fields for metal cations. Although not
largely popular, the rigid 4,6-dibenzofuran (4,6-DBF) fragment
is one such platform that has been used to create interesting
acyclic1−15 and macrocyclic16−20 ligands. Reports on the
coordination chemistry of these ligands are few, but where
structural data are available, it has been observed that the
DBF O-atom may participate in chelate formation2,3,6,12,14 or
not.4,7−9 In previous ligand construction activities we have used
2-pyridine N-oxide (2-XC5H4NO) and/or 2,6-pyridine
N-oxide (2,6-X2C5H3NO) ring platforms to attach donor
functionalized substituent arms, X = phosphine oxide, -[P(O)-
R2],21,22 phosphonate, -[P(O)(OR)2],23 methylphosphine oxide,
-[CH2P(O)R2],24−33 methylphosphonate, -[CH2P(O)(OR)2],34
methylphosphinic acid, -CH2P(O)(OH)2],34 methylphosphine
sulfide, -[CH2P(S)R2]35 and methylamido, -[CH2C(O)N(H)R]36
and −[CH2CH2C(O)NR2].37 Generally, the 2-XC5H4NO com-
pounds behave as bidentate, chelating NOPO24,33 or NOCO37
ligands toward hard metal cations, while the 2,6-(X)2C5H3NO
compounds act as tridentate NOPOP′O′21−27 or NOCOC′O′
EXPERIMENTAL SECTION
■
General Information. Organic reagents (Aldrich Chemical Co)
and metal nitrates (Ventron) were used as received, and organic
solvents (VWR) were dried and distilled by standard methods.
Caution! All plutonium chemistry was conducted at Los Alamos National
Laboratory inside specialist radiological facilities designed for the safe
handling and manipulation of high specific-activity α-particle emitting
Received: February 10, 2012
Published: June 7, 2012
© 2012 American Chemical Society
6667
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
7.35 (t, JHH = 7.5 Hz, 2H, H4), 5.00 (d, JHH = 6.3 Hz, 4H, H1). 13C{1H}
NMR (125.8 MHz, CDCl3): δ 153.88 (C7), 128.00 (C3), 124.49 (C2),
123.40 (C4), 121.81 (C6), 121.22 (C5), 40.27 (C1). FTIR (KBr, cm−1):
1488, 1433, 1420, 1352, 1326, 1263, 1224, 1188, 1118, 1057, 864, 844,
803, 781, 745, 667, 612, 578, 542. HRMS(ESI): m/z (%): 229.0412
[M-Cl+] (100). C14H10ClO requires 229.0420. Anal. Calcd for C14H10Cl2O:
C 63.42, H 3.80. Found: C 63.74, H 4.21.
4,6-Bis(diphenylphosphinoylmethyl)-dibenzofuran (4). A solution
of 3 (630 mg, 2.38 mmol) in ethyl diphenylphosphinite (3 mL, 13.9
mmol) was refluxed under nitrogen until the solution solidified. The
solid was heated (120 °C, 12 h), and the resulting white residue was
collected and washed with diethyl ether (3 × 10 mL) giving 4 as a
white powder: yield 1.37 g, 96%; mp 238−240 °C. 31P{1H} NMR
radionuclides. The plutonium(IV) solution (0.46 M in aqueous HCl,
isotopic composition: 94% 239Pu, 6% 240Pu, trace 238Pu, 240Pu, and
242Pu) was purified by using anion-exchange chromatography. The
plutonium solution concentration and oxidation state were determined
by measuring the vis−NIR spectrum of a solution sample diluted in
1.0 M HClO4. Reactions were performed under dry nitrogen unless
specified otherwise. Infrared spectra were recorded on a Bruker Tensor
27 benchtop spectrometer. A Varian Cary 6000i spectrophotometer
with a fixed spectral bandwidth of 0.2 nm was used to record solution
electronic absorption spectra at ambient temperature. Solid diffuse
reflectance spectra were collected using a Varian Cary 500 with installed
Internal Diffuse Reflectance Accessory. Solution NMR spectra were
measured with Bruker FX-250 and Avance-300 and -500 spectrometers
using Me4Si (1H, 13C) and 85% H3PO4 (31P) as external standards.
Downfield shifts from the reference resonances were given +δ values.
The atom numbering systems used for the NMR resonance assign-
ments are provided on spectral traces included in the Supporting
Information. The Pu(IV) solutions were contained inside 4 mm
diameter PTFE NMR tube liners to provide multiple containment of
the radioactive isotopes. The high resolution mass spectra were
obtained at the UNM Mass Spectrometry Center by using electrospray
ionization (ESI) with a Waters/Micromass mass spectrometer.
Elemental analyses (CHN) were performed by Galbraith Laboratories.
Ligand Syntheses. 4,6-Diformyl-dibenzofuran (1). An aliquot of
n-butyllithium (1.6 M in hexane, 46.4 mL, 74.3 mmol) was added to
dibenzofuran (5 g, 30 mmol) and N,N,N′,N′-tetramethylethylenedi-
amine (11.1 mL, 74.3 mmol) in dry hexane (30 mL). The resulting
mixture was refluxed (10 min), cooled (0 °C), and DMF (10 mL) was
added. This mixture was allowed to slowly warm (23 °C, 1 h) while
stirring, and then quenched with water (20 mL). The resulting solid
was collected by filtration, washed with water (3 × 20 mL), and
crystallized from hot methanol leaving a bone-white powder, 1: yield
1.8 g, 27%; mp 218−220 °C. 1H NMR (500 MHz, CDCl3): δ 10.71 (s,
2H, H1), 8.25 (d, JHH = 8.0 Hz, 2H, H5), 8.05 (d, JHH = 8.0 Hz,
2H, H3), 7.56 (t, JHH = 7.5 Hz, 2H, H4). 13C{1H} NMR (125.8 MHz,
CDCl3): δ 187.62 (C1), 156.51 (C7), 127.86 (C5), 126.86 (C3), 124.74
(C6), 123.97 (C4), 121.65 (C2). HRMS(ESI): m/z (%): 247.0371
[M+Na+] (100). C14H8NaO3 requires 247.0367.
1
(202.1 MHz, CDCl3): δ 29.8; (MeOH-d4): δ 34.5. H NMR (500
MHz, CDCl3): δ 7.8−7.7 (m, 10H, H5,9), 7.45−7.35 (m, 4H, H11)
7.45−7.35 (m, 10H, H3,10), 7.19 (t, JHH = 8.0 Hz, 2H, H4), 3.91 (d,
JPH = 13.5 Hz, 4H, H1). 13C{1H} NMR (125.8 MHz, CDCl3): δ
154.28 (d, JCP = 5.2 Hz, C7), 132.49 (d, JCP = 98.7 Hz, C8), 131.91
(C11), 131.16 (d, JCP = 9.1 Hz, C9), 128.90 (d, JCP = 5.2 Hz, C3),
128.47 (d, JCP = 11.5 Hz, C10), 124.11 (C6), 123.02 (C4), 119.44 (C5),
115.57 (d, JCP = 7.2 Hz, C2), 31.81 (d, JCP = 66.9 Hz, C1). FTIR (KBr,
cm−1): 3431, 3056, 2944, 2904, 1632, 1591, 1487, 1436, 1408, 1328,
1262, 1191(νPO), 1174, 1120, 1071, 1051, 1029, 998, 885, 851, 832,
806, 782, 743, 725, 694, 569, 536, 512, 462, 407. HRMS(ESI): m/z (%):
597.1742 [M+H+] (35). C38H31O3P2 requires 597.1748; 619.1559
[M+Na+] (100). C38H30O3NaP2 requires 619.1568. Anal. Calcd
for C38H30O3P2: C 76.50, H 5.07. Found: C 75.68, H 5.03.
4,6-Bis(diphenylphosphinoyl)-dibenzofuran (5). Method A. Under
dry nitrogen, a solution of n-BuLi (1.6 M in hexane, 9.3 mL, 15 mmol)
was combined with dibenzofuran (1 g, 6 mmol) and N,N,N′,N′-
tetramethylethylenediamine (2.2 mL, 15 mmol) in dry hexane (20 mL),
and the combination refluxed (10 min). The mixture was cooled (0 °C),
and diphenylphosphinic chloride (2.32 mL, 11.9 mmol) was added. The
resulting mixture was slowly warmed (23 °C) over 2 h, stirred, and then
quenched with water (20 mL). The resulting solid was collected by
filtration and washed with water (3 × 20 mL). The residue was dissolved
in a minimum amount of MeOH, precipitated by addition of Et2O and
washed with Et2O (3 × 10 mL). The bone-white powder obtained was
purified by column chromatography (silica gel eluted with CH2Cl2/
MeOH 98/2) leaving 5 as a white powder: yield 620 mg, 18%;
mp >380 °C (dec). Method B. Under dry nitrogen, a solution of n-BuLi
(1.6 M in hexane, 40.5 mL, 64.8 mmol) in hexane (80 mL) was added
dropwise (40 m, 23 °C) to a solution of dibenzofuran (3.36 g,
20.0 mmol) and N,N,N′,N′-tetramethylethylenediamine (9.78 mL,
64.8 mmol) in dry hexane (80 mL). After addition, the resulting
mixture was refluxed (1 h), then cooled (0 °C), and a solution of Ph2PCl
(7.20 mL, 40.0 mmol) in hexane (50 mL) was added (1 h). The mixture
was stirred (23 °C, 12 h), then quenched with water (10 mL), and the
combination concentrated in vacuo. The residue was extracted with
CH2Cl2 (3 × 30 mL), the organic layer recovered, dried (MgSO4),
hydrogen peroxide (30%, 10 mL) added, and the mixture stirred (23 °C,
3 h). The resulting mixture was treated with water (20 mL), and the
phases separated. The organic fraction was dried (MgSO4), filtered and
the solvent removed in vacuo leaving a brown residue that was treated
with Et2O (50 mL). The resulting pale yellow solid was collected by
filtration and washed with Et2O. This solid was dissolved in CH2Cl2,
filtered through a silica pad, and the volatiles evaporated in vacuo
providing 5 as a white powder (3.6 g). Additional product was recovered
by vacuum evaporation of the combined ether filtrate and wash phases,
and the residue purified by column chromatography (silica gel, elution
with CH2Cl2/MeOH 98/2) leaving a white powder, 5 (2.1 g): combined
yield 5.7 g, 50%; mp >380 °C (dec). 31P{1H} NMR (121.49 MHz,
CDCl3): δ 25.2. 1H NMR (300 MHz, CDCl3): δ 8.15 (d, JHH = 7.5 Hz,
2H, H4), 7.82−7.76 (m, 2H, H2) 7.67−7.60 (m, 8H, H8), 7.45−7.36 (m,
6H, H3,10), 7.36−7.28 (m, 8H, H9). 13C{1H} NMR (75.4 MHz, CDCl3)
δ 156.23 (C6), 133.01 (d, JCP = 6.1 Hz, C2), 132.13 (s, C10), 131.90 (d,
JCP = 10.6 Hz, C8), 131.79 (d, JCP = 107.6 Hz, C7), 128.58 (d, JCP = 12.6
4,6-Bis(hydroxymethyl)-dibenzofuran (2). To a suspension of 1
(1.78 g, 7.94 mmol) in MeOH/CHCl3 (1/1, 40 mL), was added
NaBH4 (587 mg, 15.5 mmol), and the mixture was stirred (23 °C,
12 h) under nitrogen. The mixture was concentrated to 10 mL, and
water (20 mL) was added. The resulting pale yellow solid, 2, was
1
collected by filtration: yield 1.6 g, 88%; mp 188−190 °C. H NMR
(300 MHz, CDCl3): δ 7.92 (d, JHH = 7.2 Hz, 2H, H5), 8.05 (d, JHH
=
7.5 Hz, 2H, H3), 7.38 (t, JHH = 7.5 Hz, 2H, H4) 5.13 (d, JHH = 6.3 Hz,
4H, H1). 13C{1H} NMR (125.8 MHz, d6-DMSO): δ 151.26 (C7),
124.89 (C3), 124.50 (C2), 121.60 (C4), 121.59 (C6), 118.10 (C5),
56.14 (C1). FTIR (KBr, cm−1): 1489, 1432, 1366, 1190, 1074, 1050,
1028, 991, 869, 842, 806, 766, 741, 669, 610. HRMS(ESI): m/z (%):
251.0684 [M+Na+] (100). C14H12NaO3 requires 251.0677.
4,6-Bis(chloromethyl)-dibenzofuran (3). Under dry nitrogen, a
suspension of 2 (780 mg, 3.41 mmol) and triphenylphosphine (2.68 g,
10.3 mmol) in dry CH2Cl2 (50 mL) was cooled (0 °C) and N-chloro-
succinimide (1.36 g, 10.3 mmol) was added. Stirring was continued
(0 °C) until the mixture became homogeneous, and then the solution
was warmed (23 °C) and stirred (12 h). The solvent was vacuum
evaporated, and the residue was purified by column chromatography
(silica gel, eluted with hexane/Et2O, 95/5). Following vacuum
evaporation of the eluant, compound 3 was obtained as a yellow
Hz, C9), 125.03 (s, C3), 124.15 (d, JCP = 6.2 Hz, C5), 123.42 (d, JCP
=
1
powder: yield 830 mg, 92%; mp 136−138 °C. H NMR (500 MHz,
10.6 Hz, C4), 117.00 (d, JCP = 100.2 Hz, C1). FTIR (KBr, cm−1): 3430,
1573, 1470, 1437, 1413, 1393, 1181 (νPO), 1122, 899, 837, 811, 787, 742,
CDCl3): δ 7.88 (d, JHH = 7.5 Hz, 2H, H5), 7.52 (d, JHH = 7.0 Hz, 2H, H3),
6668
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
724, 694, 588, 574, 559, 547, 533, 497. HRMS(ESI): m/z (%): 569.1428
[M+H+] (40). C36H27O3P2 requires 569.1435; 591.1232 [M+Na+]
(100). C36H26O3NaP2 requires 591.1255.
after each addition. As 4 was added, the solution turned a paler yellow
color. NMR of Pu(IV)-4 solution complex: A sample of 4 (4.2 mg,
7.0 μmol) was dissolved in THF-d8 (1 mL), and the 31P and 1H NMR
spectra were recorded. Aqueous Pu(IV)/HCl solution (17.3 μL,
7.96 μmol Pu) was then added to the ligand solution, and the 31P
Syntheses of Metal Complexes. Indium and Lanthanide
Complexes of 4. Solutions of In(NO3)3 and Ln(NO3)3·xH2O
(84 μmol, 1 equiv) in MeOH/EtOAc (1/1, 2 mL) were added
dropwise to solutions of ligand 4 (50−100 mg, 84−168 μmol, 1−2
equiv) in MeOH/EtOAc (1/1, 3 mL). The mixtures were stirred
(23 °C, 1 h), and the resulting clear solutions were evaporated to
dryness leaving solid residues. [In(4)(NO3)3]. 31P{1H} NMR (121.5
1
NMR (referenced to external 85% H3PO4) and H NMR (referenced
to residual internal solvent protio resonances) spectra were recorded.
1H NMR (THF-d8, 25 °C, 300 MHz): δ 8.10 (br, 2H, benzo-H), 7.59
and 7.40 (br, 20H, phenyl-H), 7.00 (t, 2H, benzo-H), 6.32 (br, 2H,
benzo-H), 4.79 (excess protons from Pu(IV)/HCl stock solution
aliquot addition), 3.58 and 1.73 (residual protio resonances from
THF-d8). 31P{1H} NMR (THF-d8, 25 °C, 300 MHz): δ 0.58.
1
MHz, CDCl3): δ 29.7; (MeOH-d4): δ 35.1. H NMR (300 MHz,
CDCl3): δ 7.74−7.69 (m, 10H, H5,9), 7.61−7.28 (m, 14H, H3,10,11),
1
7.25 (m, 2H, H4), 3.88 (d, JHH = 15.0 Hz, 4H, H1). H NMR (300
X-ray Diffraction Analyses. Crystals of the ligands and lanthanide
and indium complexes were placed in glass capillaries and mounted on
a Bruker X8 APEX II CCD-based X-ray diffractometer located at
UNM that is equipped with an Oxford Cryostream 700 low temperature
device and normal focus Mo-target X-ray tube (λ = 0.71073 Å). Crystals
of the plutonium complex were contained in paratone and a wax-sealed
quartz capillary coated in acrylic. The diffraction data were collected with
a Bruker D8 diffractometer outfitted with an APEX II CCD detector
and Bruker Kryoflex low temperature device located at LANL. The
instrument was equipped with a graphite monochromatized Mo Kα
X-ray source (λ = 0.71073 Å) and a 0.5 mm monocapillary. The data
frames were integrated with the Bruker SAINT software package42 and
absorption corrections were applied with SADABS.43 In no case was
crystal decay observed. The structures were solved and refined by direct
methods and difference Fourier techniques using the Bruker SHELXTL
software package.44 Lattice and data collection parameters for the ligands
and the metal complexes are presented in Tables 1 and 2, respectively.
MHz, MeOH-d4): 7.78−7.72 (m, 10H, H5,9), 7.56−7.47 (m, 12H,
H10,11), 7.16−7.15 (m, 4H, H3,4), 4.12 (d, JHH = 12.9 Hz, 4H, H1).
FTIR (KBr, cm−1): 3057, 2912, 1591, 1527, 1495, 1439, 1397, 1294,
1273, 1250, 1200, 1163, 1140(νPO), 1124, 1095, 1011, 887, 849, 827,
802, 787, 764, 741, 727, 689, 623, 532, 513, 501. Anal. Calcd for
[In(4)(NO3)3], C38H30InN3O12P2: C, 50.86; H, 3.37; for [In-
(4)2(NO3)3], C76H60InN3O15P4: C, 61.10; H, 4.05. Found: C, 58.73;
H, 4.10. [La(4)(NO3)3]·H2O. 31P{1H} NMR (121.5 MHz, MeOH-d4):
δ 35.6. 1H NMR (300 MHz, MeOH-d4): δ 7.72−7.68 (m, 10H, H5,9),
7.53−7.39 (m, 12H, H10,11), 7.14−7.09 (m, 4H, H3,4), 4.07 (d, JHH
=
13.8 Hz, 4H, H1). FTIR (KBr, cm−1): 3059, 2916, 1628, 1591, 1469,
1437, 1296, 1190, 1159, 1142 (νPO), 1094, 1028, 997, 883, 847, 818,
791, 733, 695, 625, 559, 532, 513. Anal. Calcd for [La(4)(NO3)3]·
H2O, C38H32LaN3O13P2: C, 48.58; H, 3.43. Found: C, 48.55; H 3.55.
[Pr(4)(NO3)3]·H2O. 31P{1H} NMR (121.5 MHz, DMF-d7): δ 34.6
1
(br). H (300 MHz, DMF-d7): δ 8.12−8.03 (m, 8H, H9), 7.82 (d,
JHH = 7.2 Hz, 2H, H5), 7.58−7.56 (m, 12H, H10,11), 7.35 (d, JHH
=
6.9 Hz, 2H, H3), 7.14 (d, JHH = 6.9 Hz, 2H, H4), 4.5 (br, 4H, H1).
FTIR (KBr, cm−1): 3059, 2918, 1730, 1684, 1591, 1491, 1437, 1298,
1190, 1167, 1142 (νPO), 1122, 1095, 1028, 999, 885, 847, 814, 791, 733,
692, 579, 561, 532, 513. Anal. Calcd for [Pr(4)(NO3)3]·H2O,
C38H32N3O13P2Pr: C 48.48, H 3.43. Found: C, 48.36; H, 3.52.
[Er(4)(NO3)3]·H2O. FTIR (KBr, cm−1): 3059, 2914, 1732, 1684,
1591, 1473, 1437, 1296, 1190, 1150, 1140 (νPO), 1122, 1094, 1028, 997,
885, 847, 818, 791, 734, 692, 579, 561, 532, 513. [Nd(4)2(NO3)3]·H2O.
FTIR (KBr, cm−1): 3057, 2916, 1591, 1484, 1437, 1304, 1194, 1143
(νPO), 1122, 1095, 1052, 1027, 998, 887, 847, 819, 789, 738, 694, 626,
578, 534, 512. Anal. Calcd for [Nd(4)2(NO3)2](NO3)·H2O,
C76H62NdN3O16P4: C, 59.46; H, 4.15. Found: C, 59.40; H, 4.18.
Erbium Complex of 5. A solution of Er(NO3)3·5H2O (88 μmol,
1equiv) in MeCN/MeOH (4/1, 3 mL) was added dropwise to a
solution of ligand 5 (50 mg, 88 μmol, 1 equiv) in MeOH/EtOAc (1/1,
3 mL). The resulting mixture was stirred (23 °C, 1 h), and the clear
solution was evaporated to dryness leaving a solid residue. [Er-
(5)2(NO3)3]·4H2O. FTIR (KBr, cm−1): 2924, 2854, 1591, 1472, 1439,
1418, 1300, 1203, 1150 (νPO), 1122, 1097, 1030, 903, 839, 816, 783,
744, 731, 690, 592, 559, 546, 528, 507. Anal. Calcd for [Er(5)2-
(NO3)2]·4H2O, C72H60ErN3O19P4: C, 55.35; H, 3.87. Found: C,
55.33; H, 3.65.
Table 1. Crystallographic Data for Ligands 4·CH3OH and 5
4·CH3OH
C39H34O4P2
5
empirical formula
crystal size (mm)
formula weight
crystal system
space group
unit cell dimen.
a (Å)
C36H26O3P2
0.21 × 0.22 × 0.39 0.08 × 0.18 × 0.32
628.60
triclinic
P1
568.51
monoclinic
P21/n
11.1295(3)
12.1172(3)
13.6427(3)
88.335(1)
69.154(1)
76.794(1)
1671.17(7)
2
13.3352(2)
11.0946(2)
18.8531(3)
90
b (Å)
c (Å)
α (deg)
β (deg)
94.015(1)
90
γ (deg)
V (Å3)
2782.45(8)
4
Z
T, (K)
150(2)
150(2)
Dcald (g cm−3
)
1.249
1.357
μ(mm−1
)
0.170
0.194
Plutonium Complex of 4. Method 1. Aqueous Pu(IV)/HCl
solution (9.1 μL, 4.2 μmol Pu) was added to tetrahydrofuran (THF,
700 μL), and this solution was combined with a sample of 4 (2.9 mg,
4.9 μmol) in MeOH (750 μL). The vial was sealed and left to stand.
After 6 days, the vial cap was loosened to allow solvent to slowly
evaporate. Crystals formed in the vial by the following day, and these
were used in the X-ray diffraction and the vis−NIR spectroscopic
analyses. Method 2. A sample of 4 (4.6 mg, 7.7 μmol) was dissolved
in THF (1 mL), and combined with aqueous Pu(IV)/HCl solution
(18.2 μL, 8.37 μmol Pu). Vapor diffusion of Et2O into the THF
solution produced a yellow powder that was air-dried: yield 5 mg, 66%.
This solid was found to be insoluble or sparingly soluble in THF,
CH2Cl2, EtOH, EtOAc, MeCN, and acetone, and it was used for the
diffuse reflectance vis−NIR analysis. Pu(IV)-4 spectrophotometric
titration: Aqueous Pu(IV)/HCl solution (9.1 μL, 4.2 μmol Pu) was
added to THF (700 μL), and a vis−NIR spectrum of the clear yellow
solution was recorded. A sample of 4 (4.6 mg, 7.7 μmol) in THF
(1.0 mL) was added in aliquots, and the vis−NIR spectrum was recorded
min/max transmission
reflection collected
0.9365/0.9654
40022
0.9402/0.9845
61938
independent reflections [Rint]
11076[0.0150]
0.0393(0.1058)
9199[0.0296]
0.0407(0.1038)
final R indices [I > 2σ(I)]
R1 (wR2)
final R indices (all data) R1(wR2) 0.0489(0.1145)
0.0573(0.1140)
All heavy atoms were refined anisotropically, and all carbon hydrogen
atoms were included in ideal positions and refined isotropically (riding
model) with Uiso = 1.2Ueq of the parent atom unless noted otherwise.
The structure refinements were well behaved except as indicated in the
following notes. 4·CH3OH: colorless prisms were obtained by dissolving
100 mg of 4 in MeOH/CH2Cl2 (1/1, 10 mL), concentrating this
solution until cloudy (∼5 mL), adding a few drops of CH2Cl2, and
holding the resulting solution at −20 °C (12 h). The MeOH lattice
solvent molecule is disordered over two positions with occupancies set at
51.73% (O1s,C1s) and 48.27% (O2s, C2s). The H-atoms on the O- and
6669
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
6670
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
Scheme 1
C-atoms were placed in ideal positions with Uiso = 1.5Ueq of the parent
atom. 5: colorless prisms were grown from MeOH solution of 5 held at
−20 °C. [In(4)(NO3)3]: thin prisms were isolated by slow evaporation
of a solution of the complex dissolved in a minimum volume of MeCN/
MeOH (20/80) at 23 °C. [Pr(4)(NO3)3(CH3CN)]·0.5CH3CN:
colorless prisms were obtained by slow evaporation (23 °C) of a
MeCN/MeOH (90/10) solution of the complex. The H-atoms of the
CH3CN molecules were included in ideal positions and refined with
fixed Uiso = 1.5Ueq of the parent C-atom. [Er(4)(NO3)3(CH3CN)]·
CH3CN: thin colorless blades were grown by cooling (−20 °C) a
MeCN/MeOH (90/10) solution of the complex. The outer sphere
CH3CN is disordered over two positions with occupancies 56.8%
(N1s,C1s,C2s) and 43.2% (N2s,C3s,C4s). The H-atoms were included
in ideal positions and refined with fixed Uiso = 1.5Ueq of the parent
carbon atom. [Pu(4)Cl4]·(THF): Single crystals were obtained from
Method 1. The presence of twin components resulted in relatively large
esd values associated with the bond lengths and angles. Repeated
attempts at structure determinations with different crystals revealed
similar or worse twinning, and the low solubility of the compound
prevented growth of crystals from other solvent systems suitable for
single crystal X-ray diffraction. [Nd(4)2(NO3)2]2(NO3)2·(H2O)·
4(CH3OH): colorless needles were obtained by slow evaporation
(23 °C) of a CH3OH solution of the complex. The crystals were of marginal
quality as a result of the presence of a large number of disordered
solvent molecules in the crystal lattice. There were no signs of crystal
decay during data collection. In an initial refinement model, one water
molecule, disordered over two positions with occupancies of 65% (O1w)
and 35% (O2w), and four CH3OH molecules, one disordered over two
positions with occupancies of 58.7% (O5s,C5s) and 41.3% (O4s,C4s),
were included; however, the R-factors were high. Significant improvement
of the R-factors was achieved through use of the SQUEEZE procedure.45
PLATON estimates 171 electrons unaccounted for in the solvent
accessible void volume which leads to an estimate of 9.5 CH3OH
molecules or 16 water molecules. Except for the molecules mentioned
above, the remaining lattice solvent structure was not satisfactorily
modeled. All ordered non-hydrogen atoms were refined anisotropically.
The H-atoms on the ligand 4 were included in fixed ideal positions with
Uiso = 1.2Ueq of the parent C-atoms. The ordered CH3OH H-atoms were
fixed in ideal positions and refined with Uiso = 1.5Ueq. {[Er(5)2(NO3)2]-
(NO3)·4CH3OH}0.5: colorless prisms were obtained by slow evaporation
(−20 °C) of a MeCN/MeOH (90/10) solution of the complex. The Er
resides on a crystallographic 2-fold axis so that there are two ligands and
two nitrate ions coordinated to Er1. The third nitrate is not coordinated,
and it is equally disordered over two positions (50%) about an inversion
center. There are two outer sphere CH3OH molecules one of which is
ordered [O1s, C1s], and it was refined anisotropically. The second is
disordered over two positions with occupancies of 65% and 35%.
Distribution Studies. Materials. All salts and solvents were
reagent grade and were used as received. Extraction experiments were
carried out using 1,2-dichloroethane (OmniSolv, EM Science) as the
diluent. The aqueous phases were prepared using nitric acid (J. T. Baker,
Ultrex II) and europium nitrate (Aldrich, 99.9%). Distilled, deionized
water was obtained from a Barnstead Nanopure filter system (resistivity
at least 18.2 MΩ-cm) and used to prepare all the aqueous solutions. The
radioisotope 241Am was provided by the Radiochemical and Engineering
Research Center (REDC) of ORNL. The radiotracer 152/154Eu was
obtained from Isotope Products, Burbank, CA. Both were added as
spikes to the aqueous phases in the sample equilibration vials in the
extraction experiments.
Methods. Phases at an 1:1 organic to aqueous (O:A) phase ratio
were added to 2 mL polypropylene microtubes, which were then
capped and mounted by clips on a disk that was rotated in a constant-
temperature air box at 25.0
0.5 °C for 1 h. After the contacting
period, the tubes were centrifuged for 5 min at 3000 rpm and 25 °C in
a Beckman Coulter Allegra 6R temperature-controlled centrifuge. A
250 μL aliquot of each phase was subsampled and counted using a
Canberra Analyst pure Ge Gamma counter. Counting times were
sufficient to ensure that counting error was a small fraction of the
precision of the obtained distribution ratios, considered from a
combination of volumetric, replicate, and counting errors to be 5%.
Americium and europium distribution ratios were calculated as the
ratio of the volumetric count rates of the 241Am and 152/154Eu isotopes
in each phase at equilibrium.
Molecular Mechanics Calculations. Method. Geometry opti-
mizations of the free and metal-bound forms of NOPOPO, 4 and 5
were carried out with the MM3 force field46 using a points-on-a-sphere
metal ion47 as implemented in PCModel software.48 Conformational
searches to locate the most stable form for each structure were
performed using the GMMX algorithm provided with this software.
Input files required to repeat these calculations including additional
parameters for treating the metal-dependent interactions are available
as Supporting Information.
RESULTS AND DISCUSSION
■
Ligand Syntheses and Characterization. The syntheses
of the target ligands 4 and 5, developed in the present study,
are summarized in Scheme 1. The 4,6-diformyl-dibenzofuran, 1,
was prepared essentially as described in the literature8 although
the realized yield of purified compound was typically about half
of that previously reported. The syntheses of 4,6-bis-
(hydroxymethyl)-dibenzofuran, 2, and 4,6-bis(chloromethyl)-
dibenzofuran, 3, were accomplished in a manner similar49 to pro-
cedures briefly outlined by Hanton11 and Cram16 for preparation
6671
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
of 4,6-bis(bromomethyl)-dibenzofuran. Both steps are efficient,
and 3 was obtained as a yellow powder. Subsequent reflux of 3
with excess Ph2POEt under nitrogen, without additional solvent,
gave a solid mass that, following removal of the excess Ph2POEt,
left 4,6-bis(diphenylphosphinoylmethyl)-dibenzofuran, 4,6-[Ph2P-
(O)CH2]2DBF, 4, as a white powder.50 The compound is soluble
in CHCl3, CH2Cl2, THF, and hot CH3CN, less soluble in MeOH
and insoluble in cold CH3CN. The HRESI-MS displays a parent
ion, [M+H+], and an infrared spectrum contains a strong
absorption at 1191 cm−1 that is tentatively assigned to the νPO
stretching frequency based upon assignments made in related
phosphine oxide compounds.24,28 The 31P NMR spectrum shows
a single resonance, whose chemical shift, δ 29.8 in CDCl3 and
34.5 in MeOH-d4, compares favorably with shifts observed for
diphenylphosphinoylmethyl-decorated pyridine and pyridine
N-oxide compounds,24,28 and the 1H and 13C{1H} NMR spectra
are consistent with the proposed structure. Most notably, the 1H
and 13C{1H} resonances for the methylene group in the
−CH2P(O)Ph2 arms appear at δ 3.91 (JPH = 13.5 Hz) and δ
31.8 (JCP = 66.9 Hz), respectively.
O1(−x+1, −y+1, −z+2)···H1S, 1.73(9) Å, and the angle,
O1S−H1S−O1(−x+1, −y+1, −z+2), 166(5)°.
Ligand 5 was prepared in a one-pot synthesis (Method A), as
summarized in Scheme 1, by reaction of dibenzofuran with
n-BuLi/TMEDA solution (1:2) followed by treatment of the
putative 4,6-Li2-DFB with 2 equiv of Ph2P(O)Cl. This approach
was initially chosen over an alternative route that would utilize
oxidation of the known precursor diphosphine,1,2 4,6-[Ph2P]2-
DFB, 6, since mixed substituent derivatives, 4,6-[RR′P(O)CH2]2-
DFB, of potential interest for extraction measurements might be
reached efficiently by using more easily prepared alkyl and aryl
phosphinic chlorides than with the mixed substituent phosphines.
Surprisingly, this direct approach, proved to be somewhat
inefficient as indicated by the low, unoptimized yield (18%) of 5.
Synthetic variations that might provide improved yields have not
yet been explored. The modest yield encountered with Method
A led us to prepare 5 via oxidation of the known phosphine, 6,
(Method B), and the yield of 5 improved (50%). Compound 5
displays a parent ion, [M+H+] in the HRESI-MS, and the IR
spectrum contains a strong absorption at 1181 cm−1 tentatively
assigned to a PO stretching mode. The 31P NMR spectrum
The molecular structure of 4 was confirmed by single crystal
X-ray diffraction analysis. A view of the molecule is shown in
Figure 1 and selected bond lengths are listed in Table 3. The
1
has a single resonance, δ 25.2, and the H and 13C{1H} NMR
spectra are consistent with the proposed structure. The
molecular structure of 5 was further confirmed by single crystal
X-ray diffraction analysis, and a view of the molecule and selected
bond lengths are presented in Figure 2 and Table 3, respectively.
Figure 1. Molecular structure of 4,6-[Ph2(O)CH2]2DFB·CH3OH,
4·CH3OH, (thermal ellipsoids, 50%) with lattice CH3OH omitted for
clarity.
Figure 2. Molecular structure of 4,6-[Ph2(O)]2DFB, 5, (thermal
ellipsoids, 50%).
Table 3. Selected Bond Lengths for Ligands 4·MeOH and
5 (Å)
The dibenzofuran backbone is planar with the two PO bond
vectors twisted out of the plane by different amounts as indicated
by the dihedral angles, O2−P1−C2−C1 −54.9(1)° and O3−
P2−C11−C10 4.4(1)°. The PO bond lengths, P1−O2,
1.4849(10) Å and, P2−O3, 1.4857(9) Å are identical and
comparable with the bond lengths in 4·MeOH. It is noted that
during the course of this study a report of the synthesis of 5 by
hydrogen peroxide oxidation of 6 appeared along with a crystal
structure determination,15 but parametric data are absent in the
report, and apparently the X-ray diffraction data have not yet
been deposited.
As noted in the Introduction, a synthesis for the diphosphine,
4,6-(Ph2P)2DBF, 6, has been previously reported,1,2 and
options for bidentate PP′ vs tridentate POP′ chelation on
transition metal centers have been evaluated. The rigid character
of the backbone, the large separation of the P-atom donor centers
(5.741 Å), and the calculated bite angle range (117−147°)
suggested that this diphosphine would prefer to form bimetallic
complexes. Indeed, coordination of 6 to a Co2(CO)6 fragment
bond type
4·MeOH
1.4934(8)
5
P−O
P1−O1
P2−O2
P1−C1
P1−O2
P2−O2
P1−C2
P2−C11
C1−O1
C12−O1
1.485(1)
1.4857(9)
1.812(1)
1.804(1)
1.385(1)
1.385(1)
1.4879(8)
1.810(1)
1.807(1)
1.381(1)
1.383(1)
P−C
P2−C12
C14−O3
C13−O3
C−O
molecular unit contains a MeOH solvent molecule that is
disordered nearly equally over two sites, and it is omitted from
the view in Figure 1. The dibenzofuran backbone is planar, and
the PO bond vectors are rotated out of the backbone plane
in opposite directions. The PO bond lengths, P1−O1
1.4934(8) Å and P2−O2 1.4879(8) Å are comparable to the
PO bond length in [Ph2P(O)CH2]2C5H3NO, 1.480(3) Å.25
The MeOH solvent molecule is hydrogen bonded with one
phosphoryl oxygen atom as indicated by the nonbonded
distances, O1S···O1(−x+1, −y+1, −z+2), 2.667(11) Å, and
6672
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
gave a bimetallic complex, Co2(CO)6(6), in which the ligand
bridges the two Co(0) atoms, and the furan O-atom is not
involved in the ligand/metal interaction.3 However, subsequent
coordination chemistry with Ru(III) revealed complexes
containing tridentate POP′ chelate interactions.10 To adopt this
structure the nonbonded P···P separation decreased by ∼1 Å, and
the bite angle increased by ∼25° relative to the free ligand.
Molecular modeling suggested that this deformation results in
steric strain in excess of 3 kcal/mol.4 The reaction of 6 with
Re2Cl82− also produced a complex with a tridentate POP′ chelate
interaction on one of the Re centers and a monodentate P
interaction on the other.10 Given these observations, it was of
interest to explore the coordination preferences for both the
bis(methylphosphine oxide), 4, and the more rigid bis(phosphine
oxide), 5.
Ligand Computational Modeling. Qualitative CPK
model building suggests that both of the bis(phosphine
oxide)-type ligands, 4 and 5, should, at the least, be able to
act as bidentate, POP′O′ chelating ligands on trivalent or
tetravalent f-block element cations. However, it is not clear if
the furan O-atom can join with these donor centers to form
tridentate chelate structures without incurring serious ligand
strain. Therefore, computational molecular mechanics (MM)
assessments of the ligand strain energies developed during the
transformation of the ligands from their free, unbound resting
states to metal bound bidentate (PO donors only) and
tridentate (both PO and Ofuran donors) chelate structures
were undertaken to address this issue.51 Such strain energies,
which can be evaluated in the absence of other factors such as
auxiliary inner-sphere ligation or outer-sphere solvation,
provide a quantitative measure of the intrinsic ability of the
ligand to achieve the bound conformation. When evaluated in
the tridentate bonding mode with Pr(III), the computed strains
developed in 4 and 5, relative to their free ligand confor-
mations, are 13.8 and 11.9 kcal/mol, respectively. These results
indicate that both 4 and 5 are structurally less organized for
tridentate metal coordination than NOPOPO, which exhibits a
corresponding ligand strain of only 8.8 kcal/mol. Bidentate
chelation interactions by 4 and 5 result in less ligand strain than
tridentate chelation. However, with strain energies of 9.3 kcal/mol,
4, and 9.0 kcal/mol, 5, these ligands are still more strained than
tridentate NOPOPO. Views of the computed energy-minimized
tridentate metal complexes are shown in Figure 3.
Figure 3. Geometry optimized structures for tridentate coordination
in 1:1 ligand Pr(III) complexes: (a) NOPOPO, (b) ligand 4, and (c)
ligand 5.
A search of the Cambridge Structural Database53 for crystal
structures involving f-element cations coordinated to mono-
dentate triaryl phosphine oxides, pyridine-N-oxides, and ethers
was conducted. As shown in Figure 4a, phosphine oxide
interactions exhibit an average PO−M bond angle of 163
10° with no clear preference for C−PO−M dihedral angles.
As illustrated in Figure 4b, pyridine-N-oxide interactions
display N−O−M angles of 132
8° and dihedral angles
covering a wide range, 90 33°. Finally, as indicated in Figure 4c,
ether interactions exhibit C−O−M angles in a more narrow range,
125
4°, with an out-of-plane angle, 0
8°, consistent with
prior observations that the ether oxygen donor atom prefers sp2
hybridization.52
As shown in Figure 5, these data can be used to visualize the
donor orientation provided by the binding conformers of
NOPOPO, 4, and 5. The resulting picture is consistent with the
ligand strain analysis results. NOPOPO, which develops the
least amount of strain upon metal complexation, exhibits the
most convergent set of vectors, that is, this scaffold allows all
three donor-metal interactions to achieve geometries closest to
those observed for monodentate analogues. In both 4 and 5,
the architectures position the two PO donors to simulta-
neously coordinate the metal, but fail to correctly orient the
ether oxygen donor group. The degree of vector convergence is
better in 5 than in 4, which exhibits the largest strain on metal
coordination. Thus, both ligand strain analysis and evaluation of
donor orientation predict that, of these three ligands, 4 is least
likely to engage in tridentate coordination.
The computed ligand strain energies reflect the degree of
binding site organization provided by these scaffolds.51 An
important component to this organization is how well the spatial
arrangement of donor groups in the binding conformation is able
to complement the metal ion. As illustrated in an earlier review
on the role of donor atom orientation in metal ion binding,52
a
convenient device for visualizing the degree of complementarity
offered by a ligand binding conformer is to add vectors to each
donor atom that represent the direction for optimal metal
interaction with the donor atom. When all vectors converge at the
correct M−L distance, the ligand is complementary. The
orientation of these vectors, which reflect optimal M−L−X
angles and M−L−X−X dihedral angles, can be deduced from
crystal structure data for complexes involving simple mono-
dentate ligands.
Coordination Chemistry. With these results in mind, the
coordination chemistry of 4 was initially explored with several
small, trivalent p-block metal cations, Al(III), Ga(III), and
In(III), as their nitrate salts, with ligand:metal combining ratios
6673
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
There is also a new absorption at ∼1200 cm−1(Δν ∼+10 cm−1)
that may arise from an unbound PO group in a monodentate 4.
The 31P NMR spectrum for a 1:1 mixture of 4 and In(NO3)3
dissolved in MeOH-d4 shows a single resonance, δ 35.1, that is
only slightly shifted from the free ligand in MeOH-d4, δ 34.5. A
2:1 mixture and a 4:1 mixture show a single resonance at δ 35.4
and 34.7, respectively, indicating that the ligand is undergoing
rapid exchange on the In(III) in methanol solution. Crystalli-
zation of a crude sample of the indium(III) complex, obtained
from the 1:1 combination of 4 and In(NO3)3, from a mixed
MeCN/MeOH solvent mixture, provided single crystals for
which a single crystal X-ray structure determination reveals the
composition [In(4)(NO3)3]. A view of the complex is shown in
Figure 6, and selected bond lengths are listed in Table 4. The
Figure 4. Depiction of the observed location of f-block metal ions
coordinated with monodentate (a) triaryl phosphine oxides, (b)
pyridine N-oxides, and (c) ethers in the Cambridge Structural
Database.51 Observed metal locations are illustrated by green volumes,
and the black vector emanating from each oxygen atom represents an
average direction of approach for the metal ion.
Figure 6. Molecular structure and atom labeling scheme for
[In(4)(NO3)3] (thermal ellipsoids, 50%) with H-atoms omitted for
clarity.
In(III) ion displays a seven coordinate pentagonal bipyramidal
inner coordination sphere generated by two oxygen atoms from
the PO groups of a bidentate ligand 4, four oxygen atoms
from two bidentate nitrate ions, and one O-atom from a
monodentate nitrate ion. A similar bidentate/monodentate
nitrate coordination condition on In(III) was observed in the
complex [In(bipy)2(NO3)3] in which the bipyridyl ligands are
both bidentate, one nitrate ion is bidentate and two nitrate ions
are monodentate.54 The furan O-atom is 4.209 Å removed from
the In(III), and the nonbonded In1···O3 vector is 22.3° out of
the DBFring plane. These features are consistent with the
absence of a coordinate bond interaction between the indium
ion and the DBF O-atom. The average In−O(P) bond length,
2.0967 0.0002 Å, is relatively short compared to bond lengths
in several phosphine oxide complexes of indium halides:
[In(Ph3P(O))2(Br)2(CH2Br)], 2.281(6) Å (avg)55 and [In-
(Me2PhP(O))2Cl3], 2.196(7) Å (avg).56 The average PO
bond length in the coordinated ligand, 1.5108 0.0004 Å, is
elongated relative to the average PO bond length in the free
ligand, 1.4907 0.0028 Å. Attempts to isolate and structurally
characterize 2:1 complexes of these Group 13 cations that
might contain 4 binding with both bidentate and monodentate
coordination modes have so far been unsuccessful. However,
the fact that the nitrate ions show mixed coordination modes in
the [In(4)(NO3)3] solid state structure support the suggestion
that 2:1 complexes may contain mixed ligand binding
Figure 5. Donor group orientation in the binding conformation of (a)
NOPOPO, (b) ligand 4, and (c) ligand 5 illustrated with 2.5 Å vectors
emanating from each donor oxygen atom in the optimal direction for
metal binding (see Figure 4). P-phenyl groups are replaced with
carbon atoms for clarity.
of 1:1 and 2:1. Under the conditions explored, the CH
elemental analyses suggest that in each case these p-block metal
cations form mixtures that contain both 1:1 and 2:1
complexes.50 The IR spectra for the complexes display a strong
band tentatively assigned to the PO stretching mode with a
coordination shift, ΔνPO ∼50 cm−1, relative to the free ligand.
This coordination shift is consistent with metal−ligand binding
at least through one or both phosphine oxide donor groups.
6674
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
6675
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
conditions with one or more nitrate counter-ions displaced to
the outer coordination sphere.
The coordination chemistry of 4 with the lanthanide cations
La(III), Pr(III), Nd(III), Eu(III), and Er(III), all as their nitrate
salts, was examined by using 1:1 and 2:1 ligand:metal
combining ratios in MeOH/EtOAc (1:1) solvent. Under the
conditions explored, the 1:1 complexes, like the free ligand, are
very soluble in CH2Cl2 and CHCl3, while in MeOH, MeCN,
and DMF solubility is reduced. The complexes obtained from
2:1 combinations are also soluble in CH2Cl2 and CHCl3, but
the solubility in the other solvents is much lower than for the
complexes isolated from 1:1 combinations. Each complex was
obtained as a solid powder following evaporation of solvent.
Attempts were made to deduce the precise compositions of the
complexes by elemental analyses prior to mixed-solvent recrys-
tallizations that produced the single crystal samples used for
X-ray diffraction analyses. Satisfactory agreements were
obtained for complexes formulated as [La(4)(NO3)3]·H2O,
[Pr(4)(NO3)3]·H2O, and [Nd(4)2(NO3)3]·MeOH. The anal-
yses for the remaining samples suggested formation of mixtures
of 1:1 and 2:1 complexes. For single crystal samples, the
agreements between measured and calculated CH composi-
tions, based upon the crystallographically determined compo-
sitions, were not within 0.4%. This is due, in part, to the
presence of loosely held lattice solvent molecules. High resolution
mass spectra for selected samples were also collected; however,
parent ions were not observed. This suggests that the complexes
do not survive the ionization conditions of the ESI-MS spec-
trometer employed. The IR spectra for the samples with 1:1 and
2:1 compositions are essentially identical with all showing
significant ΔνPO shifts in the range 49−52 cm−1. The 31P NMR
spectrum for the 1:1 complex [La(4)(NO3)3] in MeOH-d4 shows
a single resonance, δ 35.6, shifted ∼1 ppm from the free ligand.
Further additions of 4 to the solution resulting in 2:1 and 4:1
ligand:La(III) reactant ratios gave spectra with a single resonance,
δ 35.2, consistent with rapid ligand exchange. Attempts were also
made to study the complexation of 4 with La(III) in aqueous
solution by using spectrophotometric-based metal titrations, but
low binding energy (log K1 ∼ 0.2) hindered this analysis. As a
result, single crystal X-ray diffraction analyses provide the most
definitive composition and structural characterization for the
crystalline complexes.
The inner sphere structures for the 1:1 complexes [Pr(4)-
(NO3)3(CH3CN)]·0.5CH3CN and [Er(4)(NO3)3(CH3CN)]·
CH3CN are shown in Figures 7 and 8, respectively, and selected
bond lengths are summarized in Table 4. The inner sphere
structures are identical except for small differences in the Ln-O
and Ln-N distances because of variations in lanthanide ionic radii
and the greater polarizing strength of Er(III). The structures
contain a bidentate POP′O′-bonded ligand 4, three bidentate
nitrate groups and one N-bonded acetonitrile molecule. The eight
oxygen atoms and one nitrogen atom, in both cases, produce a
tricapped trigonal prismatic coordination polyhedron. The
bidentate chelate interaction in the Pr(III) complex is slightly
asymmetric, Pr1−O1 2.333(1) Å and 2.367(1) Å, while the inter-
action is more symmetric in the Er(III) complex, Er1−O2
2.240(2) Å and 2.248(2) Å. The DBF ring, including the O-atom,
is planar but not O-bonded to the Ln ions; the nonbonded
Ln···ODBF separations are 4.280 (Pr) and 4.265 Å (Er),
respectively, and the nonbonded M···Ofuran vectors make an
angle of 18° with the DBF plane. The Ln-O(P) bond lengths are
similar to those found in Pr(III) and Er(III) complexes containing
alkyl decorated POP′O′NO″ ligands31 and slightly shorter than
Figure 7. Molecular structure and atom labeling scheme for
[Pr(4)(NO3)3(CH3CN)]·0.5CH3CN (thermal ellipsoids, 50%) with
lattice CH3CN and H-atoms omitted for clarity.
Figure 8. Molecular structure and atom labeling scheme for
[Er(4)(NO3)3(CH3CN)]·CH3CN (thermal ellipsoids, 50%) with
lattice CH3CN and H-atoms omitted for clarity.
found with the phenyl substituted POP′O′NO″.24,28 The PO
bond lengths are slightly elongated over the values observed in
the free ligands.
Considerable effort was devoted to isolation of X-ray quality
single crystals of 2:1 complexes containing 4. Unfortunately, in
all cases, crystal quality was poor as a result of the presence of
disordered solvent molecules that could not be adequately
modeled. The refinement for one complex, {[Nd(4)2(NO3)2]-
(NO3)·(H2O)·4(CH3OH)}0.5, following application of
SQUEEZE to obtain a solvent free structure, is presented
since the qualitative structural features are worth noting. A view
of the inner sphere structure is shown in Figure 9. The
structure contains two ligands, 4, and both are bonded in a
POP′O′ bidentate mode. The DBF O-atoms are well removed
from the Nd(III) at 4.22 and 4.20 Å. This result indicates that
there is sufficient space to accommodate two bidentate POP′O′
coordinated molecules of 4 in the Nd(III) inner coordination
sphere, but additional structural conclusions should not be
drawn given the crystal quality.
The coordination chemistry of 4 toward Pu(IV) was also
explored. Initially it was noted that addition of aqueous
Pu(IV)/HCl solution to a MeOH solution of 4 produced a
yellow precipitate, as did addition of Pu(IV)/HCl solution to 4
in hot CH3CN. These precipitates, assumed to be a Pu(IV)
complex with 4, did not redissolve in CH2Cl2 or hot MeOH.
6676
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
Figure 9. Molecular structure and atom labeling scheme for
+
−
[Nd(4)2(NO3)2 ](NO3 )·(H2O)·4(CH3OH) with outer sphere ni-
trate ion, lattice solvent molecules, and H-atom omitted for clarity
(thermal ellipsoids, 30%).
Figure 11. Solution electronic absorption spectrum of 2−6 mM Pu(IV)
in THF as a function of added 4. Equivalents of added 4: 0 (1), 0.44 (2),
0.88 (3), 1.77 (4). For legibility, spectra 2, 3, and 4 have been offset on
the y-axis by 15, 34, and 55 M−1 cm−1, respectively.
Therefore, the complexation of Pu(IV) with 4 was followed by
visible/near-IR spectrometry in THF solution. As shown in
Figure 10, the spectra of aliquots of the initial yellow Pu(IV)
evident by blue shifts in the transitions at 637, 735, and 903 nm
to 630, 727, and 892 nm, respectively. In addition, there are
marked profile changes in the electronic transitions in the regions
of 650−720 and 780−870 nm, with new maxima appearing at
661 and 786 nm. From the spectra in Figure 11 it appears that
only one Pu(IV) complex is formed with 4. The spectra of 0.88,
1.77, and 3.6 equiv of ligand added are essentially identical,
consistent with the assignment of a 1:1 Pu(IV):4 complex
stoichiometry as the only product. This complex is readily
precipitated from THF solution as a powder in 66% yield.
However, once isolated, the complex exhibits poor solubility in a
wide range of organic solvents tested. To obtain NMR spectra,
samples of the complex were prepared in situ by mixing Pu(IV)/
HCl with 1.1 equiv of 4 in THF-d8 solution. The 31P NMR
spectrum exhibits a paramagnetically broadened resonance at
0.58 ppm, which is assigned to the 1:1 Pu(IV):4 complex. This
resonance is significantly shifted from the sharp “free” ligand
resonance at 24.2 ppm (THF-d8) or 29.8 ppm (CDCl3). The 1H
NMR spectrum is more complicated because of paramagnetic
line broadening induced by the proximity of the Pu(IV) ion (5f4
electronic configuration) and the presence of a large resonance at
4.79 ppm due to the protons from the Pu(IV)/HCl solution
used in the sample preparation. However, by using the free ligand
1H NMR spectrum in THF-d8 as a guide, along with integral
values from the in situ spectrum of the [Pu(4)Cl4] complex,
tentative assignments can be made. The peaks at 8.10, 7.00, and
6.32 ppm are in the expected aryl region. All three have equivalent
integration values, and can be assigned to the ortho-, meta-, and
para-protons on the benzo rings of the DBF backbone, but we
cannot say which peaks correspond to Ho, Hm, or Hp positions.
The combined integrations of the peaks at 7.59 and 7.40 ppm sum
to a value nine times as large as each of the 8.10, 7.00, and 6.32
ppm peaks. Therefore these two peaks likely account for
overlapping ortho, meta, and para protons on the phenyl rings
(although the integration of these two peaks should be ten times
as great instead of nine to accurately account for all of the protons
on the phenyl rings). Finally, the alkyl CH2 protons, which appear
as a doublet (due to 31P coupling) at about 4.1 ppm in the THF-
d8 free ligand spectrum, are either too proximate to the
paramagnetic Pu(IV) ion to be observed because of line
Figure 10. Solution electronic absorption spectra of various
concentrations of aqueous Pu(IV)/HCl in THF. The Pu(IV)
concentrations are 2.0 mM (blue), 5.9 mM (green), and 22.7 mM
(red).
hydrochloric acid stock solution, spiked into THF, differ
depending on the concentration of the Pu(IV)/HCl in THF,
but they resemble the spectra of [PuCl6]2− in CH3CN57 (green
trace). This suggests that the [PuCl6]2− anion may not be the
only species present in THF solution prior to addition of 4 (e.g.,
other plutonium chloride or solvation species).58 However, the
differences in the initial spectra do not affect the form of the
spectra once 4 is added. Figure 11 shows the vis−NIR spectra as
increasing equivalents of 4 are added to the Pu(IV)/HCl solution
in THF solvent. Visually, the original distinct yellow color
becomes less intense with each addition. The electronic
transitions for actinide ions are generally broader and more
sensitive to the local ligand environment than the more “core-
like” lanthanide ions, allowing changes in the profiles of the 5f-5f
and 5f-6d transitions to be a useful tool in following speciation
changes of actinide ions. The formation of a Pu(IV) complex is
6677
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
broadening or they are masked by the large resonance of excess
protons arising from the Pu(IV)/HCl solution aliquot. The solid-
state vis−NIR diffuse reflectance spectrum of the isolated Pu(IV)
complex with 4 is shown in Figure 12 (blue trace). It shows
Figure 14. Molecular structure and atom labeling scheme for [Er(5)2-
(NO3)2](NO3)·2CH3OH (50% thermal ellipsoids) with H atoms,
outer sphere nitrate and lattice solvent omitted for clarity.
sits at over 4.2 Å away from the Pu(IV) ion. The Pu−O bond
distances to the phosphine oxide O-atoms are 2.207(11) and
2.227(16) Å, while the Pu−Cl distances range from 2.552(5) to
2.577(7) Å. The closest geometrical and chemical comparison
in the literature is the cis-[PuCl4(Ph3PO)2] complex,59 in which
the triphenylphosphine oxide Pu−O distance is 2.221(4) Å, and
the Pu−Cl distances are 2.570(2) and 2.572(2) Å. Both are
statistically identical to the corresponding bond lengths in
[Pu(4)Cl4]. The bite angle, O−Pu−O, for 4 in [Pu(4)Cl4] is
94.2(4)°,which is larger than the O−Pu−O angle of 89.2(2)° in
[PuCl4(Ph3PO)2]. This difference probably results from the
geometric constraints imposed by the DBF backbone compared
to the relative freedom of the two unattached Ph3PO ligands
in [PuCl4(Ph3PO)2]. Additional related examples of Pu(IV)/
phosphine oxide complexes with aromatic ligand backbones
have been described.26,29,60 For example, a ten-coordinate
Pu(IV) complex, [Pu(NO3)2(NOPOP′O′)2](NO3)2·1.5H2O·
0.5 MeOH (NOPOP′O′ = 2,6-[(C6H5)2P(O)CH2]2C5H3NO),
contains two NOPOP′O′ ligands bonded in a tridentate mode
through the two PO and one N−O functionalities.26 The
average Pu−O(P) distance, 2.347 Å, is over 0.1 Å longer than
the bond lengths in [Pu(4)Cl4]. This is likely a reflection of the
larger coordination number and increased steric congestion in
the 1:2 complex. In another ten coordinate 1:2 complex,
[Pu(NOPO)2(NO3)3][Pu(NO3)6]0.5, the related bifunctional
NOPO ligand, (2-[(C6H5)2P(O)CH2]C5H4NO), adopts a
bidentate chelate bonding mode in which the Pu−O(P)
distances are also over 0.1 Å longer than in [Pu(4)Cl4]. Finally,
structures for two 1:1 Pu(IV):POPO complexes (POPO = 2,6-
[(C6H5)2P(O)CH2]2C6H4) have been reported.29 In an eight
coordinate [Pu(POPO)(NO3)2Cl2] complex, the Pu−O(P) bond
distances are similar to those in [Pu(4)Cl4], while in a nine
coordinate complex, [Pu(POPO)(NO3)3(OMe)], the Pu−O(P)
bond lengths are slightly longer at 2.301(2) and 2.276(2) Å.
Clearly, the solid-state structural results, regarding the
denticity of 4 when bound with hard trivalent and tetravalent
metal ions, are consistent with the molecular modeling analysis.
It is worth noting that the nonbonded M···Ofuran distances in
the structurally similar 1:1 complexes of 4 are comparable, M =
Pr3+ (4.280 Å), Er3+ (4.263 Å), Pu4+ (4.292 Å), and In3+ (4.209 Å),
but the differences do not directly parallel the differences in metal
crystal radii:61 Pr3+ (1.28 Å, CN = 8), Er3+ (1.14 Å, CN = 8), Pu4+
(1.00 Å, CN = 6), and In3+ (1.063 Å, CN = 8). The differences
Figure 12. vis−NIR diffuse reflectance spectrum of solid [Pu(4)Cl4]
(blue spectrum, right axis). The solution electronic absorption
spectrum of 3.5 mM Pu(IV) in THF with 0.88 equiv added 4 (red
spectrum, left axis).
excellent correlation of the electronic transitions compared to the
solution spectra (Figure 12, red trace), supporting a conclusion
of identical speciation in both solid and solution phases.
Confirmation of the 1:1 Pu(IV):4 stoichiometry is provided
by a structure determination on an X-ray diffraction quality
single-crystal obtained by slow evaporation of a THF/MeOH
solution of the complex. A view of the molecular structure is
shown in Figure 13, and pertinent bond distances are provided
Figure 13. Molecular structure and atom labeling scheme for
[Pu(4)Cl4]·THF (50% thermal ellipsoids) with lattice THF and all
H-atoms omitted for clarity.
in Table 4. The Pu(IV) ion is six coordinate, bound to four
chloride anions and two oxygen atoms from the bidentate
ligand, 4. The geometry about the Pu(IV) ion is best described
as distorted octahedral with the O atoms bound in a cis
arrangement enforced by the requirements of the DBF
backbone. The DBF oxygen atom is not coordinated, and it
6678
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
may more reflect the residual effective polarizing strengths of the
respective metal cation−anion units and their respective impacts on
the framework structure of 4. Some support for this conclusions is
found in the two nonbonded Nd···Ofuran distances in the 2:1
complex [Nd(4)2(NO3)2]2(NO3)2 (4.221 and 4.204 Å). Here,
despite expected steric crowding because of the proximity of two
large bidentate ligands in the inner coordination sphere, both Ofuran
atoms are drawn closer to the Nd(NO3)2+ unit than is the case
with the 1:1 complexes containing Ln(NO3)3 units.
In any case, it is apparent that the anticipated weak M-ODBF
bond strength and the stabilization that might be gained
through tridentate POP′O′O″-metal interaction are not
sufficient to offset the added strain energy incurred in forming
the 1:1 or 2:1 complexes with tridentate 4.
Recalling that the computational modeling analyses suggested
that the ligand strain energy for tridentate binding of 5 was less
than with 4, the coordination chemistry of 5 with Pr(III), Eu(III)
and Er(III) nitrates, using 1:1 and 2:1 combining ratios, was also
explored. In each case, complexes form; however, diffraction
quality single crystals of only one complex, {[Er(5)2(NO3)2+]-
(NO3−)·4(CH3OH)}0.5, have so far been obtained. A view of the
molecule is shown in Figure 14 and selected bond lengths are
presented in Table 4. The structure is complicated by disorder
shown by one outer sphere nitrate ion and a MeOH solvent
molecule; however, the inner sphere structure about the Er(III) is
well-defined. The Er(III) ion is coordinated with two ligands, 5,
and two nitrate ions. The Er ion resides on a crystallographic
2-fold axis so that one DBF ligand and one nitrate ion in the inner
sphere are related to the second ligand and nitrate by the 2-fold
axis. Both molecules of 5 are chelated in a bidentate POP′O′
mode with Er−O bond lengths identical with those found in
[Er(4)(NO3)3(CH3CN)], and the nitrate ions are bidentate. The
eight inner sphere O-atoms generate a bidisphenoid coordination
polyhedron. The DBF O-atoms are closer to the Er(III) ion but
still well removed, 3.639 Å, and nonbonding. Here too, the base
strength of the DBF O-atom is apparently so weak that it is not
able to offset the strain energy developed in forming a tridentate
chelate interaction.
Figure 15. Americium and europium distribution ratios as a function
of the initial nitric acid concentration. Organic phase: 4, 5, or CMPO
at 10 mM in 1,2-DCE. Aqueous phase: trace 241Am and 0.1 mM of
europium nitrate in nitric acid. O/A = 1, T = 25 °C.
Solvent Extraction. The solvent extraction properties of 4
and 5 toward Am(III) and Eu(III) in aqueous nitric acid
solutions were investigated and compared to the well studied
TRUEX extractant, n-octyl(phenyl)-N,N-diisobutylcarbamoyl-
methylphosphine oxide (OΦDiBCMPO = CMPO).62 Both 4
and 5 are insoluble in the preferred organic phase extraction
solvent, dodecane. Therefore, the distribution ratios, D =
[Morg/Maq], for all three compounds were measured under the
same conditions, as a function of the concentration of nitric
acid, by using 0.01 M solutions of each in 1,2-dichloroethane
(DCE). The variations of D with increasing [HNO3] are
summarized in Figure 15. All three molecules show the unique
property wherein the respective D's increase with increasing
nitric acid concentration. Both 4 and CMPO reach their
maximum D value at ∼1 M [HNO3] and then begin to decrease
with further increases in [HNO3]. This probably results from
competing protonation of the phosphine oxide donor sites at high
acid concentrations.63 It is also apparent that the D's for 4 at all
acid concentrations are significantly smaller than those for the
CMPO ligand, and this is consistent with the trial spectrophoto-
metric titration analysis data for 4 with La(III) in aqueous
solution. In contrast, the maximum D values for extraction of
both Am(III) and Eu(III) by 5 are reached at significantly higher
nitric acid concentration, ∼5 M compared with ∼1 M with 4. It is
also clear from Figure 15 that 5 is a stronger extractant for both
Figure 16. Americium and europium distribution ratios as a function
of the initial ligand concentration. Organic phase: variable
concentrations of 4 or CMPO in 1,2-DCE. Aqueous phase: trace
241Am and 0.1 mM of europium nitrate in 1 M nitric acid. O/A = 1,
T = 25 °C.
Eu(III) and Am(III) than 4 at all acid concentrations. This
observation is consistent with the MM computed strain energy
analysis that indicated lower ligand strain and better donor vector
convergence for 5 compared to 4. It is interesting that 5, under
identical conditions, appears to be a slightly better extractant than
CMPO for Am(III) and Eu(III) in HNO3 solutions above 0.5 M.
This suggests that further comparative studies with more
hydrocarbon soluble derivatives of 5, CMPO, and NOPOPO
are warranted. Lastly, it is noted that, under the conditions used
for this study, 5 has a small preference to bind Eu(III) over
Am(III) (α = 2.3 at 3 M HNO3). The opposite is typically true
with CMPO extractions of Am(III) and Eu(III).63
The ligand dependencies on D's for 4 and CMPO in DCE at
constant nitric acid concentration (1 M) were also examined
for both Am(III) and Eu(III) extractions, and the data are
summarized in Figure 16. The ligand dependency for 5 was not
measured in this study because of its more limited solubility
range in DCE. From slope analyses, it is deduced that two
molecules (slope = 2) of both 4 and CMPO are involved in
6679
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
their extraction complexes. It can also be seen from Figure 16
that CMPO is the stronger extractant of this pair at all ligand
concentrations studied. These results also encourage develop-
ment of hydrocarbon solvent soluble analogues of 5 for
additional detailed nitric acid extraction and ligand dependency
extraction analyses, and those activities are in progress.
ACKNOWLEDGMENTS
■
The authors thank Professor R.D. Hancock and his group for
their efforts to obtain aqueous solution formation constant data
for 4 in combination with La(III). Financial support for this
study at the University of New Mexico was provided by the
Division of Chemical Sciences, Geosciences and Biosciences,
Office of Basic Energy Sciences, U.S. Department of Energy
(Grant DE-FG02-03ER15419 (R.T.P)). In addition, funds
from the National Science Foundation assisted with the
purchases of the X-ray diffractometer (CHE-0443580) and
NMR spectrometers (CHE-0840523 and −0946690). B.P.H.
and L.H.D. acknowledge support from the Division of
Chemical Sciences, Geosciences and Biosciences, Office of
Basic Energy Sciences, U.S. Department of Energy. A.J.G. and
S.D.R. thank the U.S. Department of Energy, Office of Science,
Early Career Research Program (contract DE-AC52-
06NA25396). R.T.P. wishes to thank Prof. R. G. Bergman
and Dr. V. Chan for unpublished details regarding the synthesis
of 3.
CONCLUSIONS
■
A high yield synthesis for the trifunctional ligand 4,6-
bis(diphenylphosphinoylmethyl)dibenzofuran, 4, has been
developed, and ligand strain energies for potential bidentate
POP′O′ and tridentate POP′O′O″ chelate interactions have
been assessed with molecular mechanics calculations. In addi-
tion, a synthesis for 4,6-bis(diphenylphosphinoyl)dibenzofuran,
5, has also been devised, and the strain energies for bidentate
and tridentate chelate structures computed and compared with
values obtained for 4 and NOPOPO. For both 4 and 5, the
strain energy is greater for the tridentate structure relative to
the bidentate structure. Subsequent combinations of the ligands
with several trivalent metal nitrates and Pu(IV) chloride
produced coordination complexes for which single crystal X-ray
diffraction structure determinations reveal formation of 1:1 and
2:1 complexes. In all cases 4 and 5 are found to act as bidentate
POP′O′ chelates. Consistent with the computational analyses,
no examples were isolated and structurally characterized where
4 or 5 adopt tridentate or metal bridged structures. This
suggests that the DBF fragment O-atom is not a sufficiently
strong donor site to offset the strain energy encountered in
forming tridentate structures such as suggested for the acyclic
analogue of 5, 1,5-bis(diphenylphosphinoyl)-3-oxapentane.64 A
survey of the extraction performance of 4 and 5 in 1,2-DCE
indicates that 5 is a stronger extractant for Am(III) and Eu(III)
than 4 at all nitric acid concentrations studied (0.01−6 M) and
slightly stronger than CMPO at nitric acid concentrations >1M.
Further development of dialkyl and alkyl(aryl) phosphine oxide
analogues of 5 are underway that will provide additional insight
into the complexation and extraction performances of this
ligand.
REFERENCES
■
(1) Haenel, M. W.; Jakubik, D.; Rothenberger, E.; Schroth, G. Chem.
Ber. 1991, 124, 1705−1710.
(2) Kranenburg, M.; van der Burgt, Y. E. M.; Kamer, P. C. J.; van
Leeumen, P. W. N. M.; Govbitz, K.; Fraamje, J. Organometallics 1995,
14, 3081−3089.
(3) Vogl, E. M.; Bruckmann, J.; Kruger, C.; Haenel, M. W.
J. Organomet. Chem. 1996, 520, 249−252.
(4) Vogl, E. M.; Bruckmann, J.; Kessler, M.; Kruger, C.; Haenel, M.
W. Chem. Ber. 1997, 130, 1315−1319.
(5) Hlavinka, M. L.; Hagadorn, J. R. Chem. Commun. 2003, 2686−
2687.
̈
̈
(6) Skar, M. L.; Svendsen, J. S. Tetrahedron 1997, 53, 17425−17437.
(7) (a) Kanemasa, S.; Oderaotoshi, Y.; Yamamoto, H.; Tanaka, J.;
Wada, E. J. Org. Chem. 1997, 62, 6454−6455. (b) Kanemasa, S.;
Oderaotoshi, Y.; Sakaguchi, S.; Yamamoto, H.; Tanaka, J.; Wada, E.;
Curran, D. P. J. Am. Chem. Soc. 1998, 120, 3074−3088.
(8) Deng, Y.; Chang, C. J.; Nocera, D. G. J. Am. Chem. Soc. 2000,
122, 410−411.
(9) Kaul, R.; Deechongkit, S.; Kelly, J. W. J. Am. Chem. Soc. 2002,
124, 11900−11907.
(10) Kuang, S.-M.; Fanwick, P. E.; Walton, R. A. Inorg. Chem. 2002,
41, 405−412.
ASSOCIATED CONTENT
■
(11) Caradoc-Davies, P. L.; Hanton, L. R. Dalton Trans. 2003, 1754−
1758.
S
* Supporting Information
Selected IR, vis−UV, and NMR spectra, CIF files for the crystal
structures, PCModel input files for NOPOPO, 4 and 5. This
material is available free of charge via the Internet at http://pubs.
acs.org. X-ray data have also been deposited with the Cambridge
Crystallographic Data Centre with the deposition numbers
833444 (4·MeOH), 833445 (4·2H2O), 833446 (4·H2O),
833447 [Pr(4)(NO3)3(CH3CN)]·0.5(CH3CN), 833448 [Er(4)-
(NO3)3(CH3CN)]·(CH3CN), 833449 [In(4)(NO3)3], 833450
[Nd(4)2(NO3)2]2(NO3)2·(H2O)·4(CH3OH), 833451 {[Er-
(5)2(NO3)2](NO3)·4(CH3OH)}0.5, and 833452 (5). These files
may be accessed free of charge at http://www.ccdc.cam.ac.uk/
conts/retrieving.html.
(12) Pintado-Alba, A.; de la Riva, H.; Nieuwhuyzen, M.; Bautista, D.;
Raithby, P. R.; Sparkes, H. A.; Teat, S. J.; Lopez-de-Luzuriaga, J. M.;
́
Lagunas, C. M. Dalton Trans. 2004, 3459−3467.
(13) Porter, R. M.; Danopoulos, A. A. Polyhedron 2006, 25, 859−863.
(14) De la Riva, H.; Nieuwhuyzen, M.; Fierro, C. M.; Raithby, P. R.;
Male, L.; Lagunas, M. C. Inorg. Chem. 2006, 45, 1418−1420.
(15) Han, C.; Xie, G.; Li, J.; Zhang, Z.; Xu, H.; Dend, Z.; Zhao, Y.;
Yan, P.; Liu, S. Chem.Eur. J. 2011, 17, 8947−8956.
(16) Robbins, T. A.; Cram, D. J. J. Chem. Soc., Chem. Commun. 1995,
1515−1516.
(17) Agbaria, K.; Biali, S. E. J. Org. Chem. 2001, 66, 5482−5489.
(18) Li, F.; Delgado, A.; Coelho, A.; Drew, M. G. B.; Fel
́
ix, V.
Tetrahedron 2006, 62, 8550−8558.
(19) Li, F.; Delgado, R.; Drew, M. G. B.; Fel
́
ix, V. Dalton Trans. 2006,
5396−5403.
AUTHOR INFORMATION
■
́
(20) Li, F.; Li, L.; Delgado, R.; Drew, M. G. B.; Felix, V. Dalton Trans.
Corresponding Author
2007, 1316−1324.
(21) McCabe, D. J.; Russell, A. A.; Karthikeyan, S.; Paine, R. T.;
Ryan, R. R.; Smith, B. Inorg. Chem. 1987, 26, 1230−1235.
(22) Conary, G. S.; Russell, A. A.; Paine, R. T.; Hall, J. H.; Ryan, R. R.
Inorg. Chem. 1988, 27, 3242−3245.
*E-mail: rtpaine@unm.edu (R.T.P.); gaunt@lanl.gov (A.J.G.).
Notes
The authors declare no competing financial interest.
6680
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681

Inorganic Chemistry
Article
(23) Russell, A. A.; Meline, R. L.; Duesler, E. N.; Paine, R. T. Inorg.
Chim. Acta 1995, 231, 1−5.
used to correlate ligand structure with metal ion binding affinity in a
number of systems. (a) Hay, B. P.; Rustad, J. R.; Hostetler, C. J. J. Am.
Chem. Soc. 1993, 115, 11158−11164. (b) Hay, B. P.; Zhang, D.;
Rustad, J. R. Inorg. Chem. 1996, 35, 2650−2658. (c) Dietz, M. L.;
Bond, A. H.; Hay, B. P.; Chiarizia, R.; Huber, V. J.; Herlinger, A. W.
Chem. Commun. 1999, 1177−1178. (d) Hay, B. P.; Dixon, D. A.;
Vargas, R.; Garza, J.; Raymond, K. N. Inorg. Chem. 2001, 40, 3922−
3935. (e) Lumetta, G. J.; Rapko, B. M.; Garza, P. A.; Hay, B. P.;
Gilbertson, R. D.; Weakley, T. J. R.; Hutchison, J. E. J. Am. Chem. Soc.
2002, 124, 5644−5645. (f) Hay, B. P.; Oliferenko, A. A.; Uddin, J.;
Zhang, C.; Firman, T. K. J. Am. Chem. Soc. 2005, 127, 17043−17053.
(52) Hay, B. P.; Hancock, R. D. Coord. Chem. Rev. 2001, 212, 61−78.
(53) (a) Cambridge Structural Database, November, 2012. (b) Allen,
F. H. Acta Crystallogr. 2002, B58, 380−388. (c) Bruno, I. J.; Cole, J. C.;
Edgington, P. R.; Kessler, M.; Macrae, C. F.; McCabe, P.; Pearson, J.;
Taylor, R. Acta Crystallogr. 2002, B58, 389−397.
(24) Rapko, B. M.; Duesler, E. N.; Smith, P. H.; Paine, R. T.; Ryan,
R. R. Inorg. Chem. 1993, 32, 2164−2174.
(25) Engelhardt, U.; Rapko, B. M.; Duesler, E. N.; Frutos, D.; Paine,
R. T. Polyhedron 1995, 14, 2361−2369.
(26) Bond, E. M.; Duesler, E. N.; Paine, R. T.; Neu, M. P.; Matonic,
J. H.; Scott, B. L. Inorg. Chem. 2000, 39, 4152−4155.
(27) Bond, E. M.; Duesler, E. N.; Paine, R. T.; Noth, H. Polyhedron
̈
2000, 19, 2135−2140.
(28) Gan, X.-M.; Duesler, E. N.; Paine, R. T. Inorg. Chem. 2001, 40,
4420−4427.
(29) Matonic, J. H.; Enriquez, A. E.; Scott, B. L.; Paine, R. T.; Neu,
M. P. Nucl. Sci. Tech. 2002, 3, 100−105.
(30) Gan, X.-M.; Paine, R. T.; Duesler, E. N.; Noth, H. Dalton Trans.
̈
2003, 153−159.
(54) Junk, P. C.; Skelton, B. W.; White, A. H. Aust. J. Chem. 2006, 59,
147−154.
(31) Gan, X.-M.; Rapko, B. M.; Duesler, E. N.; Binyamin, I.; Paine, R.
T.; Hay, B. P. Polyhedron 2005, 24, 469−474.
(32) Pailloux, S.; Shirima, C. E.; Ray, A. D.; Duesler, E. N.; Paine, R.
T.; Klaehn, J. R.; McIlwain, M. E.; Hay, B. P. Inorg. Chem. 2009, 48,
3104−3113.
(55) Peppe, C.; Nobrega, J. A.; Hernandes, M. Z.; Longo, R. L.;
Tuck, D. G. J. Organomet. Chem. 2001, 626, 68−75.
(56) Robinson, W. T.; Wilkins, C. J.; Zeying, Z. J. Chem. Soc., Dalton
Trans. 1990, 219−227.
(57) Ryan, J. L.; Jorgensen, C. K. Mol. Phys. 1963, 7, 17−29.
(58) Nikitenko, S. I.; Moisy, P. Inorg. Chem. 2006, 45, 1235−1242.
(59) Berthon, C.; Boubals, N.; Charusnikova, I. A.; Collison, D.;
Cornet, S. M.; Den Auwer, C.; Gaunt, A. J.; Kaltsoyannis, N.; May, I.;
Petit, S.; Redmond, M. P.; Reilly, S. D.; Scott, B. L. Inorg. Chem. 2010,
49, 9554−9562.
(60) Matonic, J. H.; Neu, M. P.; Enriquez, A. E.; Paine, R. T.; Scott,
B. L. J. Chem. Soc., Dalton Trans. 2002, 400−405.
(61) Shannon, R. D.; Prewitt, C. T. Acta Crystallogr. 1969, B25, 925−
930.
(62) Horwitz, E. P.; Kalina, D. G.; Diamond, H.; Vandegrift, G. F.;
Schulz, W. W. Solvent Extr. Ion. Exch. 1985, 3, 75−109.
(63) Horwitz, E. P.; Martin, K. A.; Diamond, H.; Kaplan, L. Solvent
Extr. Ion Exch. 1986, 4, 449−494.
(64) (a) Turanov, A. N.; Karandashev, V. K.; Baulin, V. E. Solvent
Extr. Ion Exch. 1999, 17, 1423−1444. (b) Turanov, A. N.;
Karandashev, V. K.; Baulin, V. E.; Yarkevich, A. N.; Safronova, Z. V.
Solvent Extr. Ion Exch. 2009, 27, 551−578.
(33) Pailloux, S.; Shirima, C. E.; Ray, A. D.; Duesler, E. N.; Smith, K.
A.; Paine, R. T.; Klaehn, J. R.; McIlwain, M. E.; Hay, B. P. Dalton
Trans. 2009, 7486−7493.
(34) Gan, X.-M.; Binyamin, I.; Rapko, B. M.; Fox, J.; Duesler, E. N.;
Paine, R. T. Polyhedron 2006, 25, 3387−3392.
(35) Gan, X.-M.; Duesler, E. N.; Parveen, S.; Paine, R. T. Dalton
Trans. 2003, 4704−4708.
(36) Binyamin, I.; Pailloux, S.; Duesler, E. N.; Rapko, B. M.; Paine, R.
T. Inorg. Chem. 2006, 45, 5886−5892.
(37) Binyamin, I.; Pailloux, S.; Hay, B. P.; Rapko, B. M.; Duesler, E.
N.; Paine, R. T. J. Heterocycl. Chem. 2007, 44, 99−103.
(38) Bond, E. M.; Engelhardt, U.; Deere, T. P.; Rapko, B. M.; Paine,
R. T. Solvent Extr. Ion Exch. 1997, 15, 381−400.
(39) Bond, E. M.; Engelhardt, U.; Deere, T. P.; Rapko, B. M.; Paine,
R. T. Solvent Extr. Ion Exch. 1998, 16, 967−983.
(40) Nash, K. L.; Lavallette, C.; Borkowski, M.; Paine, R. T.; Gan,
X.-M. Inorg. Chem. 2002, 41, 5849−5858.
(41) Sulakova, J.; Paine, R. T.; Chakravarty, M.; Nash, K. L. Sep. Sci.
Technol., submitted for publication.
(42) SAINT + 7.01, 2003 (UNM), SAINT+ 7.06, 2003 (LANL);
Bruker AXS, Inc.: Madison, WI.
(43) Sheldrick, G. M. SADABS 2.10, 2003 (UNM), SADABS 2.03,
2001 (LANL); University of Gottingen: Gottingen, Germany.
̈
̈
(44) SHELXTL, 6.14; Bruker AXS, Inc.: Madison, WI, 2001.
(45) van der Sluis, P.; Spek, A. L. Acta Crystallogr. 1990, A46, 194−
201.
(46) (a) Allinger, N. L.; Yuh, Y.-H.; Lii, J.-H. J. Am. Chem. Soc. 1989,
111, 8551−8566. (b) Lii, J.-H.; Allinger, N. L. J. Am. Chem. Soc. 1989,
111, 8566−8575. (c) Lii, J.-H.; Allinger, N. L. J. Am. Chem. Soc. 1989,
111, 8576−8582.
(47) Hay, B. P. Coord. Chem. Rev. 1993, 126, 177−236.
(48) PCModel, version 9.3; Serena Software: Bloomington, IN.
(49) The syntheses for 2 and 3 are based upon unpublished
procedures: Chan, V. S.; Bergman, R. G., private communication. Full
details of the syntheses as employed in the present study are
summarized in the Experimental Section.
(50) Elemental analyses (CHN) for the ligand 4 and its metal
complexes were performed on multiple samples. Unfortunately, in
some cases agreement with proposed compositions is not within
0.4%. Scatter in replicate determinations suggests that some of the
problem resides with difficulty in the analysis methodology. In
addition, single crystal samples of complexes readily lose some lattice
solvent that results in variable and low CH determinations. Analyses
for noncrystalline samples of [In(4)(NO3)3] and [Er(4)(NO3)3]
suggest formation of mixtures of 1:1 and 2:1 complexes.
(51) Even when computed in the absence of other factors, such as
the presence of auxiliary inner-sphere ligation and counter-ions, outer-
sphere solvation, etc., ligand strain energies have been successfully
6681
dx.doi.org/10.1021/ic300301d | Inorg. Chem. 2012, 51, 6667−6681