Extraction and complexation of alkali, alkaline earth, and F-element cations by calixaryl phosphine oxides

Article abstract of DOI:10.1002/(SICI)1521-3765(19990104)5:1<175::AID-CHEM175>3.0.CO;2-P

A series of new calixarene derivatives with phosphine oxide groups on the lower rim and, for comparison, a series of noncalixarene phosphine oxides have been synthesised. Their extraction power for alkali and alkaline earth cations from aqueous metal picr

Full text of DOI:10.1002/(SICI)1521-3765(19990104)5:1<175::AID-CHEM175>3.0.CO;2-P

FULL PAPER
Extraction and Complexation of Alkali, Alkaline Earth, and F-Element
Cations by Calixaryl Phosphine Oxides
F. Arnaud-Neu,[a] J. K. Browne,[b] D. Byrne,[b] D. J. Marrs,[b] M. A. McKervey,[b]
P. OꢀHagan,[b] M. J. Schwing-Weill,*[a] and A. Walker[b]
Abstract: A series of new calixarene
derivatives with phosphine oxide groups
on the lower rim and, for comparison, a
series of noncalixarene phosphine ox-
ides have been synthesised. Their ex-
traction power for alkali and alkaline
earth cations from aqueous metal pic-
rate solution into dichloromethane have
been determined as well as the stability
constants in methanol of the 1:1 com-
plexes of several members of the calix-
arene series. Important selectivity trends
are revealed by both methods. The
extraction power from aqueous nitrate
solutions (1m in HNO3) towards euro-
pium(iii), as a model for trivalent acti-
nides, and thorium(iv), as a model for
tetravalent actinides, has been studied in
detail for eight symmetrical calixarene
derivatives, which differ in either calix-
arene size (4, 5, 6 or 8), the substituent at
the upper rim (tert-butyl or hydrogen) or
the number of methylene groups sepa-
rating the phenolic oxygen atoms from
the phosphorus atoms (1 or 2). The
stoichiometry of the extracted species
was characterised by a classical log ± log
plot analysis. All the calixarenes tested
are far better extractants than typical
noncalixarene extractants, for example,
TOPO and CMPO, currently in use in
the treatment of radioactive waste; they
extract thorium better than europium.
The influence of the nitric acid concen-
tration and of the sodium nitrate con-
centration in the aqueous phase on the
extraction efficiency was also examined
in order to assess the possible applica-
tion of these compounds for the deca-
tegorisation of liquid nuclear waste.
Several thorium complexes have been
characterised by their stability constants
in methanol.
Keywords: calixarenes ´ cations ´
nuclear waste
thermodynamics
´
phosphorus
´
metals.[3] Thus esters and ketones show selectivity within the
alkali cation series, and amides within the alkaline earth
cation series. Thioamides show selectivity for silver and lead.
Other research groups have successfully pursued similar
objectives with lower and upper rim bridged calixarenes,
including calixcrowns and calixspherands.[2, 4]
Introduction
Calixarenes are very amenable to chemical modification at
the upper or lower rim, or both.[1, 2] Lower rim modifications
through the phenolic oxygen atoms have been widely ex-
plored in the design and synthesis of receptors for metal
cations.[1b] Much of our earlier work in this area has
concentrated on calix[4], calix[5]- and calix[6]arenes with
lower rim carbonyl-containing substituents in the form of
esters, ketones, amides, thioamides and carboxylic acids.[3]
Extraction, transport, stability constant and calorimetric
measurements, augmented by NMR, X-ray and computer
simulation studies, provide evidence that many of these lower
rim derivatives have very significant ionophoric properties for
cations, several with good selectivity within groups of
The use of phosphine and phosphine oxide containing
calixarenes as ligands is a relatively recent extension of cation
complexation to the synthesis of transition metal complexes
with catalytic activity.[5] Our work has now established that
phosphine oxides can be employed very successfully for the
separation of selected lanthanides and actinides.[6, 7, 8] Inter-
and intra-group separations of lanthanides and actinides are
important strategies in the management and declassification
of nuclear waste.[9] Of particular importance in separation
processes is liquid ± liquid extraction, in which a charged
metal ion is transferred as a complex from a polar aqueous
phase to another immiscible phase. In the case of nuclear
waste, the aqueous phase is frequently highly acidic and rich
in sodium nitrate, both of which place severe limitations on
the type of extractants that may be employed. Of the various
extractants used in actinide process chemistry, neutral orga-
nophosphorus compounds are among the most useful. For
example, the extracting ability of tributylphosphate forms the
[a] Prof. M. J. Schwing, Dr. F. Arnaud-Neu
Laboratoire de Chimie-Physique, ECPM
1 rue Blaise Pascal, F-67000 Strasbourg (France)
Fax : (‡ 33)3-88-61-89-11
E-mail: schwing@chimie.u-strasbg.fr
[b] J. K. Browne, D. Byrne, Dr. D. J. Marrs, Prof. M. A. McKervey,
P. OꢀHagan, Dr. A. Walker
School of Chemistry, The Queenꢀs University
Belfast BT9 5AG (Northern Ireland)
Chem. Eur. J. 1999, 5, No. 1
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0175 $ 17.50+.50/0
175

M. J. Schwing-Weill et al.
FULL PAPER
basis of the PUREX process for separating plutonium and
uranium.[10] Bifunctional organophosphorus extractants in-
clude carbamoylphosphonates (CMPs) and carbamoylmeth-
ylphosphine oxides (CMPOs).[9]
We have recently reported in a preliminary paper on the
high efficiency in the extraction of EuIII, ThIV, PuIV and AmIII
from inactive and simulated radioactive wastes of tetra-, hexa-
and octaphosphine oxides with or without tert-butyl groups at
the para positions of the phenolic rings[6] (Scheme 1: com-
pounds 1a ± f). Another approach to the use of the calixarenic
platform for the treatment of radioactive waste employs
ˆ
ˆ
calix[4]arenes bearing both P O and C O groups on the
upper rim to simulate the carbamoylmethylphosphine oxides
as the classical extractant CMPO used in the TRUEX
process.[8] Both classes of compounds have been shown to
be excellent extractants, more efficient than TOPO (trioctyl-
phosphine oxide) and the CMPO octylphenyldiisobutylcar-
bamoylmethylphosphine oxide, and to extract ThIV better
than EuIII.
We now present a more complete report on the complexing
properties of calixarenes bearing phosphine oxide groups on
their lower rim, by extending our preliminary report to
include new homo- and hetero-substituted calix[4]- and
calix[5]arene derivatives. We have also extended the metal
ions to include alkali and alkaline earth series.
Scheme 1. The calixarene phosphine oxides studied. Group 1: 1a: R ˆ tBu,
n ˆ 4; 1b: R ˆ tBu, n ˆ 6; 1c: R ˆ tBu, n ˆ 8; 1d: R ˆ H, n ˆ 4; 1e: R ˆ H,
n ˆ 6; 1 f: R ˆ H, n ˆ 8; Group 2: 2a: n ˆ 4; 2b: n ˆ 5; Group 3: 3a: R ˆ
NEt2; 3b: R ˆ O-tBu.
Abstract in French: Une sØrie de nouveaux calixar›nes por-
teurs de groupements oxyde de phosphine sur le bord infØrieur
a ØtØ synthØtisØe ainsi que, à titre de comparaison, une sØrie
dꢀoxydes de phosphine non calixarØniques. Leur pouvoir
extractant vis-à-vis des cations alcalins et alcalino-terreux à
partir dꢀune solution aqueuse de picrate mØtallique vers le
dichloromØthane a ØtØ dØterminØ ainsi que, pour plusieurs
composØs de la sØrie des calixar›nes, les constantes de stabilitØ
dans le mØthanol de leurs complexes 1:1. Des sØlectivitØs
importantes ont ØtØ mises en Øvidence par les deux mØthodes.
Le pouvoir extractant à partir de solutions aqueuses de nitrates
mØtalliques (1m en HNO3) vis-à-vis de lꢀeuropium(iii), pris
comme mod›le dꢀactinides trivalents, et du thorium(iv), pris
comme mod›le dꢀactinides tØtravalents, a ØtØ ØtudiØ en dØtail
pour huit dØrivØs symØtriques des calixar›nes, diffØrant lꢀun de
lꢀautre soit par la taille du calixar›ne (4, 6 ou 8), soit par le
substituant sur le bord supØrieur (tert-butyle ou hydrog›ne) soit
encore par le nombre de groupements mØthylØniques sØparant
les atomes dꢀoxyg›ne phØnolique des atomes de phosphore (1
ou 2). La stoiechiomØtrie des esp›ces extraites a ØtØ caractØrisØe
par la mØthode des pentes. Tous les calixar›nes testØs se sont
The complete series now consists of: group 1, the tetra-,
hexa- and octaphosphine oxide with two methylene groups
between the phenolic oxygen and the phosphorus atom
(Scheme 1); a new group 2 with the tetramer 2a and the
pentamer 2b, in which there is a single methylene group
between the phenolic oxygen atom and the phosphorus atom;
and group 3 with two mixed functionality tetramers 3a and 3b
also with a single methylene spacer, but in which two distal
phosphine oxides have been replaced by two diethyl amide
groups and two tert-butyl ester groups, respectively. For
comparison purposes, several noncalixarene polyphosphine
oxides were tested (Scheme 2) for their extracting power
towards EuIII and ThIV.
All the phosphine oxides with two methylene spacers
between the phenolic oxygen and the phosphorus atom on the
lower rim were synthesised from the parent calixarenes 5 via
the known ethyl acetates 6 (Scheme 3). The sequence
involved reduction of 6 to primary alcohols 7 by use of
DIBAL in toluene, followed by conversion of the alcohols
into tosylates 8 with p-toluenesulfonyl choride in pyridine.
Introduction of diphenylphosphine residues through exposure
of the tosylates to sodium diphenylphosphide in dioxane/
tetrahydrofuran and, finally, oxidation of the resulting phos-
phines 9 to phosphine oxides 1 by use of either dimethyldiox-
irane or hydrogen peroxide in acetone. All the phosphine
oxides were fully characterised spectroscopically and analyti-
Ã
rØvØlØs etre de bien meilleurs extractants que les extractants non
calixarØniques classiques, tels que le TOPO ou le CMPO,
utilisØs couramment dans le traitement des dØchets radioactifs:
ils extraient mieux le thorium que lꢀeuropium. Lꢀinfluence sur
lꢀefficacitØ de lꢀextraction des concentrations initiales en acide
nitrique ou en nitrate de sodium dans la phase aqueuse a
Øgalement ØtØ ØtudiØe, en vue dꢀune possible application de ces
composØs à la dØcatØgorisation des dØchets nuclØaires. Plu-
sieurs complexes du thorium ont ØtØ caractØrisØs par leurs
constantes de stabilitØ dans le mØthanol.
1
cally. H NMR analysis revealed that both tetramers 1a and
1d exist in stable cone conformations in solution. The X-ray
176
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0176 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1

Calixaryl Phosphine Oxides
175±186
The diamide diphosphine oxide 3a has been previously
investigated in extraction studies with alkali and silver
cations.[11, 12] Our studies include both extraction and
complexation studies of the alkali and alkaline earth cations
with all three groups of phosphine oxides. We have extended
our extraction studies of compounds 1a ± f and 2a to euro-
pium(iii) and thorium(iv), for use as models for extraction of
trivalent lanthanides and tetravalent actinides. Some addi-
tional results concerning the stability in methanol of the
complexes of thorium by the various compounds are also
presented.
Experimental Section
General: Melting points are uncorrected. Infrared spectra were obtained
with a Perkin ± Elmer 983G grating spectrophotometer with solid samples
dispersed in KBr clear pressed discs. 1H NMR spectra were recorded at
300 MHz on a General Electric QC300 spectrometer and at 500 MHz on a
General Electric QE500 instrument. 13C and 31P NMR spectra were
recorded at 125 MHz and 202 MHz respectively, with a General Electric
QE500 spectrometer. In all cases tetramethylsilane was used as an internal
Scheme 2. The noncalixarene polyphosphine oxides studied.
standard. Elemental analyses were determined on
a Perkin ± Elmer
2400 CHN microanalyser. Carbon values in microanalysis are frequently
low for calixarenes with cavities capable of retaining solvent molecules. In
some instances acceptable elemental analysis was only observed if one
assumed the inclusion of one or more molecules of solvent. EI mass spectra
were recorded at 70 eV on a VG Autospec instrument with a heated inlet
system. Accurate molecular weights were determined by the peak-
matching method with perfluorokerosene as standard reference. ES mass
spectra were recorded on a Fisons VG-Quatro instrument with electrospray
inlet. Analytical TLC was performed on Merck Kieselgel 60254 plates.
Preparative TLC was carried out with glass plates (20 Â 20 cm) coated with
Merck Kieselgel PF254‡366 (21 g in 58 mL H2O per plate). Flash chromato-
graphy was effected with Merck Kieselgel 60 (230 ± 400 mesh). All
reactions were performed under a nitrogen atmosphere. Commercial
grade solvents were dried and purified by the standard procedures.[13] The
following calixarenes were prepared as previously described in the litera-
ture: 5,11,17,23-tetra-tert-butyl-25,26,27,28-tetrahydroxycalix[4]arene 5a,[14]
5,11,17,23-tetra-tert-butyl-25,26,27,28-tetraethoxycarbonylmethyleneoxyca-
lix[4]arene 6a,[15] 25,26,27,28-tetrahydroxycalix[4]arene 5d,[16] 25,26,27,28-
tetraethoxycarbonylmethyleneoxycalix[4]arene 6d,[15] 5,11,17,23,29-penta-
tert-butyl-31,32,33,34,35-pentahydroxycalix[5]arene,[17]
5,11,17,23,29,35-
hexa-tert-butyl-37,38,39,40,41,42-hexahydroxycalix[6]arene 5b,[18] 5,11,17,
23,29,35-hexa-tert-butyl-37,38,39,40,41,42-hexaethoxycarbonylmethylene-
oxycalix[6]arene 6b,[15] 37,38,39,40,41,42-hexahydroxycalix[6]arene 5e,[16]
37,38,39,40,41,42-hexaethoxycarbonylmethyleneoxycalix[6]arene
6e,[15]
5,11,17,23,29,35,41,47-octa-tert-butyl-49,50,51,52,53,54,55,56-octahydroxy-
calix[8]arene 5c,[19] 5,11,17,23,29,35,41,47-octa-tert-butyl-49,50,51,52,53,54,
55,56-octaethoxycarbonylmethyleneoxycalix[8]arene 6c,[15] 49,50,51,52,53,
54,55,56-octahydroxycalix[8]arene 5 f,[16] 49,50,51,52,53,54,55,56-octa-
ethoxycarbonylmethyleneoxycalix[8]arene 6 f,[15] 5,11,17,23-tetra-tert-
25,26,27,28-tetrahydroxyethyleneoxycalix[4]arene 7a[20] and 25,26,27,28-
tetrahydroxyethyleneoxycalix[4]arene 7d.[20] The calixaryl phosphine
oxides 2a and 3a were prepared by the method of Matt and co-
workers.[5] Phosphine oxides 4a[21a], 4b[21b], 4c[21c] and 4e[21d] were known
compounds.
Scheme 3. Synthesis procedure of the calixarene phosphine oxides. a: n ˆ
4, R ˆ tBu; b: n ˆ 6, R ˆ tBu; c: n ˆ 8, R ˆ tBu; d: n ˆ 4, R ˆ H; e: n ˆ 6,
R ˆ H; f: n ˆ 8, R ˆ H.
General procedure for the formation of calixarene ethyleneoxy alcohols
7a ± f: The following procedure was used to reduce calixarene esters 6a ± f
into the corresponding primary alcohols 7. A 300% molar excess of DIBAL
(1.5m solution in toluene) was added to a solution of the calixarene ester
(ca. 5%) in dry toluene under N2. The reaction mixture was stirred at room
temperature for 24 h. Excess reducing agent was destroyed by dropwise
addition of methanol until hydrogen evolution had ceased. The precipi-
tated inorganic salts were then broken up by the addition of methanol/
water and were subsequently removed by filtration through Celite. The
inorganic residue was washed with hot chloroform, and the organic layer
separated and washed with brine; each brine wash was further back-
crystal structure[6] of the former reveals the disposition of the
putative P O binding sites and their degree of preorganisa-
tion about the lower rim cavity. Compound 2b exist as a stable
cone conformation, whereas the hexamers and octamers are
conformationally mobile at ordinary temperatures, as is the
case with many of the larger calixarene derivatives.
Chem. Eur. J. 1999, 5, No. 1
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0177 $ 17.50+.50/0
177

M. J. Schwing-Weill et al.
FULL PAPER
extracted with chloroform. The combined organic extracts were dried over
MgSO4 and concentrated at reduced pressure to afford the crude calixarene
alcohol. The following ethyleneoxy alcohols were thus prepared.
solid in 74% yield following recrystallisation from ethanol/methylene
1
1
ˆ
chloride. M.p. 170 ± 1728C; IR (KBr) nÄ ˆ 1361, 1178 cm (S O); H NMR
(300 MHz, CDCl3): d 1.05 (s, 54H; C(CH3)3), 1.95 (s, 18H; ArCH3), 3.20 (d,
HB, JBA ˆ 12.08 Hz, 6H; ArCH2Ar), 3.75 (brs, 12H; OCH2CH2), 4.25 (m,
18H; HA, ArCH2Ar, OCH2CH2 overlapping), 6.75 (s, 12H; ArH), 7.31 (d,
J ˆ 7.98 Hz, 12H; ArHCH3), 7.80 (d, J ˆ 8.23 Hz, 12H; SArH); MS (ES):
5,11,17,23,29,35-Hexa-tert-butyl-37,38,39,40,41,42-hexahydroxyethyleneox-
ycalix[6]arene (7b): Calixarene 6b furnished the corresponding alcohol 7b
as a white solid in 77% yield after trituration at 208C in ethanol. M.p. >
‡
m/z: 558.1 [M/4‡H2O] ; C120H144O24S6 (2162.83): calcd C 66.7, H 6.7; found
1
3008C (decomp); IR (KBr) nÄ ˆ 3423 cm (OH); 1H NMR (300 MHz,
C 66.7, H 6.9.
CDCl3): d ˆ 1.16 (s, 54H; C(CH3)3), 3.40 ± 4.00 (brs, 36H; ArCH2Ar,
OCH2CH2 overlapping), 4.45 ± 4.70 (brs, 6H; OH), 7.00 (brs, 12H; ArH);
37,38,39,40,41,42-Hexatosylateethyleneoxycalix[6]arene (8e): Calixarene
7e furnished the dealkylated tosylate 8e as a white solid in 83% yield
following recrystallisation from ethanol/methylene chloride. M.p. 152 ±
‡
MS (ES): m/z: 1237.4 [M‡H] ; C78H108O12 ´ 3H2O (1291.72): calcd C 72.6, H
8.4; found: C 72.5, H 8.1.
1
1
ˆ
1558C; IR (KBr) nÄ ˆ 1358, 1177 cm (S O); H NMR (300 MHz, CDCl3):
d ˆ 2.38 (s, 18H; ArCH3), 3.60 (brs 12H;, OCH2CH2), 3.78 (brs, 12H;
ArCH2Ar), 4.03 (brs, 12H; OCH2CH2), 6.65 (s, 18H; ArH), 7.30 (d, J ˆ
7.88 Hz, 12H; ArHCH3), 7.80 (d, J ˆ 8.18 Hz, 12H; SArH); MS (ES): m/z:
37,38,39,40,41,42-Hexahydroxyethyleneoxycalix[6]arene (7e): Calixarene
6e furnished the alcohol 7e as a white solid in 68% yield after trituration in
1
hot ethanol. M.p. > 3008C (decomp); IR (KBr) nÄ ˆ 3416 cm (OH);
1H NMR (300 MHz, CDCl3): d ˆ 3.55 (brs, 12H; ArCH2Ar), 3.61 (brs,
12H; OCH2CH2), 3.95 (brs, 12H; OCH2CH2), 4.20 (brs, 6H; OH), 6.95
‡
1732.6 [M‡H C7H7] ; C96H96O24S6 (1826.18): calcd C 63.1, H 5.3; found C
62.7, H 5.1.
‡
(brs, 18H; ArH); MS (ES) m/z: 901.4 [M‡H] ; C54H60O12 ´ 2H2O (937.07):
calcd C 69.2, H 6.8; found: C 68.5, H 7.1.
5,11,17,23,29,35,41,47-Octa-tert-butyl-49,50,51,52,53,54,55,56-octatosylate-
ethyleneoxycalix[8]arene (8c): Calixarene 7c furnished the tosylate 8c as a
white solid in 77% yield after recrystallisation from ethanol/methylene
5,11,17,23,29,35,41,47-Octa-tert-butyl-49,50,51,52,53,54,55,56-octahydrox-
yethyleneoxycalix[8]arene (7c): Calixarene 6c furnished the alcohol 7c as
a white solid in 81% yield following recrystallisation from ethanol/water.
1
1
ˆ
chloride. M.p. 128 ± 1308C; IR (KBr): nÄ ˆ 1361, 1176 cm (S O); H NMR
(300 MHz, CDCl3): d ˆ 0.96 (s, 72H; C(CH3)3), 2.17 (s, 24H; ArCH3), 3.78
(brs, 16H; OCH2CH2), 3.87 (brs, 16H; ArCH2Ar), 4.23 (brs, 16H;
OCH2CH2), 6.80 (s, 16H; ArH), 7.12 (d, J ˆ 7.91 Hz, 16H; ArHCH3), 7.66
1
M.p. 255 ± 2578C; IR (KBr): nÄ ˆ 3434 cm (OH); 1H NMR (300 MHz
CDCl3): d ˆ 1.16 (s, 72H; C(CH3)3), 3.58 (brs, 32H; OCH2CH2, ArCH2Ar
overlapping), 3.99 (brs, 16H; OCH2CH2), 4.68 (brs, 8H; OH), 6.97 (16H; s,
‡
(d, J ˆ 8.09 Hz, 16H; SArH); MS (ES): m/z: 1442.3 [M/2‡H] , 1465.9
‡
ArH); MS (ES): m/z: 848.1 [M/2‡Na] ; C104H144O16 ´ 2H2O (1686.3): calcd
‡
[M/2‡2H‡Na] ; C160H192O32S8 (2883.78); calcd C 66.6, H 6.7, S, 8.9; found
C 74.1, H 8.9; found C 73.8, H 8.8.
C 66.7, H 6.7, S 8.5.
49,50,51,52,53,54,55,56-Octahydroxyethyleneoxycalix[8]arene (7 f): Calix-
arene 6 f furnished the dealkylated alcohol 7 f as a white solid in 52% yield
after liberating the crude material from the mixture of DIBAL salts with a
hot solution of chloroform/methanol and subsequent purification by
49,50,51,52,53,54,55,56-Octatosylateethyleneoxycalix[8]arene (8 f): Calix-
arene 7 f furnished the cone isomer of the dealkylated tosylate 8 f as a white
solid in 28% yield after purification by column chromatography (SiO2,
1
CH2Cl2/MeOH 15:1). M.p. 128 ± 1328C; IR (KBr): nÄ ˆ 1358, 1177 cm
1
trituration in acetone. M.p. > 2908C (decomp); IR (KBr) nÄ ˆ 3434 cm
(S O); 1H NMR (300 MHz, CDCl3): d ˆ 2.30 (s, 24H; ArCH3),
ˆ
(OH); 1H NMR (300 MHz [D6]DMSO) d ˆ 3.63 ± 3.66 (t, 16H; OCH2CH2),
3.80 ± 3.95 (brs, 32H; OCH2CH2, ArCH2Ar overlapping), 4.20 (brs,
16H; OCH2CH2), 6.70 (s, 24H; ArH), 7.15 ± 7.23 (d, J ˆ 7.98 Hz, 16H;
ArHCH3), 7.70 (d, J ˆ 8.12 Hz, 16H; SArH); MS (ES): m/z: 1240.2
3.76 ± 3.78 (t, 16H; ArCH2Ar), 4.05 (s, 16H; OCH2CH2), 4.81 ± 4.85 (t, 8H;
‡
OH), 6.81 (s, 24H; ArH); MS (ES): m/z: 1201.6 [M‡H] ; C72H80O16
´
CH3OH (1233.48): calcd C 71.1, H 6.8; found C 70.9, H 6.7.
‡
[M/2‡Na] ; C128H128O32S8 (2434.91): calcd C 62.2, H 5.3; found C 61.8,
General procedure for the formation of calixarene tosylates 8a ± f: The
following procedure was used to transform calixarene ethyleneoxy alcohols
7a ± f into the corresponding tosylates 8a ± f. The calixarene was dissolved
in dry pyridine (ca. 10% solution, with gentle heating as necessary). p-
Toluenesulphonyl chloride (200% molar excess) was added and the
contents shaken to dissolution. The reaction mixture was cooled to 08C in
an ice bath and stirred at that temperature for 2 h, after which time the
solution was subsequently cooled further to 158C for 3 days. The solution
was then filtered to remove the precipitated pyridinium hydrochloride salt
and the filtrate poured onto an ice/water mixture and stirred for 1 h. The
resulting off-white precipitate was filtered, washed well with water and
triturated in hot ethanol to afford the crude calixarene tosylate. The
following calixarene tosylates were thus prepared.
H 5.2.
General procedure for the formation of calixarene phosphines 9a ± f: The
following procedure was used to transform calixarene tosylates 8a ± f into
the corresponding phosphines 9a ± f. Distilled chlorodiphenylphosphine
(150% molar excess to calixarene) was dissolved in dry dioxane (ca. 10%
solution) under N2. Sodium metal (300% molar excess to chlorodi-
phenylphosphine; cut into small slivers) was added slowly. The mixture-
was then heated to reflux for 7 h with strong mechanical stirring. The
intensely yellow reaction mixture was then allowed to cool to room
temperature whilst stirring was continued. The calixarene dissolved in dry
THF (ca. 10% solution) was slowly added dropwise through a pressure-
equalizing dropping funnel and the mixture allowed to stir at room
temperature for 2 days. The red reaction mixture was filtered through
Celite and the inorganic residue washed with dry toluene. The yellow/
orange organic solution was concentrated under reduced pressure to yield a
sticky orange/red semisolid. Methylene chloride was added to the residue
and the organic solution washed with a distilled water/brine (50:50)
mixture. The organic layer was separated and dried over MgSO4 to afford
the crude calixphosphine as a cream solid. The following calixarene
phosphines were thus prepared. Satisfactory analytical data were not
obtained for all of these phosphines due to their ready oxidation in-
air.
5,11,17,23-Tetra-tert-butyl-25,26,27,28-tetratosylateethyleneoxycalix[4]ar-
ene (8a): Calixarene 7a afforded the tetratosylate 8a as a white solid in
96.5% yield an was used without further purification. M.p. 99 ± 1018C; IR
1
1
ˆ
(Kbr): nÄ ˆ 1355, 1170 cm (S O); H NMR (300 Mhz, CDCl3): d ˆ 1.05 (s,
36H, (CH3)3), 2.40 (s,12H, ArCH3), 2.97 (d, 4H, HB, JAB ˆ 12 Hz,
ArCH2Ar), 4.0 ± 4.5 (m, 20H, OCH2CH2 and HA superimposed), 6.68 (s,
8H, ArH), 7.5 (q, 16H, ArH(CH3)); C80H96O16S4 (1441.89): calcd C 66.57, H
6.65, S 8.54; found C 66.79, H 6.88, S 8.88.
25,26,27,28-Tetratosylateethyleneoxycalix[4]arene (8d): Calixarene 7d
furnished the dealkylated tosylate 8d as a white crystalline solid in 91%
yield after recrystallisation from ethanol/methylene chloride. M.p. 180 ±
5,11,17,23-Tetra-tert-butyl-25,26,27,28-tetradiphenylphosphineethylene-
oxycalix[4]arene (9a): Calixarene 8a furnished the phosphine 9a as a white
solid in 61% yield after purification by column chromatography (SiO2,
CH2Cl2/hexane 5:1). M.p. 187 ± 1898C; 1H NMR (500 MHz, CDCl3): d ˆ
1
1
ˆ
1818C; IR (KBr): nÄ ˆ 1355, 1176 cm (S O); H NMR (300 MHz, CDCl3):
d ˆ 2.43 (s, 12H; ArCH3), 3.05 (d, HB, JBA ˆ 13.62 Hz, 4H; ArCH2Ar), 4.13
(t, J ˆ 4.32 Hz, 8H; OCH2CH2), 4.30 (d, HA, JAB ˆ 13.55 Hz, 4H; ArCH2-
Ar), 4.39 (t, J ˆ 4.29 Hz, 8H; OCH2CH2), 6.56 (s, 12H; ArH), 7.32 (d, J ˆ
8.31 Hz, 8H; ArHCH3), 7.76 (d, J ˆ 8.26 Hz, 8H; SArH); MS (ES): m/z:
1.05 (s, 36H; C(CH3)3), 2.65 (m, 8H; OCH2CH2), 2.95 (d, HB, JBA
13.28 Hz, 4H; ArCH2Ar), 3.95 (m, 8H; OCH2CH2), 4.20 (d, HA, JAB
ˆ
ˆ
13.15 Hz, 4H; ArCH2Ar), 6.70 (s, 8H; ArH), 7.10 ± 7.30 (m, 40H; PPh2); 31
P
‡
‡
1217.3 [M‡H] , 1235.5 [M‡H‡H2O] ; C64H64O16S4 (1217.46): calcd C 63.1,
NMR (202 MHz, CDCl3): d ˆ 22.79 (d, PPh2); C100H108P4O4 ´ H2O
H 5.3; found C 63.0, H 5.3.
(1515.86): calcd C 79.3, H 7.3; found C 78.9, H 7.1.
5,11,17,23,29,35-Hexa-tert-butyl-37,38,39,40,41,42-hexatosylateethylene-
oxycalix[6]arene (8b): Calixarene 7b afforded the hexatosylate 8b as a white
25,26,27,28-Tetradiphenylphosphineethyleneoxycalix[4]arene (9d): Calix-
arene 8d furnished the phosphine 9d as a white solid in 68% yield after
178
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0178 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1

Calixaryl Phosphine Oxides
175±186
purification by column chromatography (SiO2, CH2Cl2/hexane 5:1). M.p.
136 ± 1388C; 1H NMR (300 MHz, CDCl3): d ˆ 2.56 (t, J ˆ 7.97 Hz, 8H;
OCH2CH2), 2.97 (d, HB, JBA ˆ 13.45 Hz, 4H; ArCH2Ar), 3.96 (m, 8H;
OCH2CH2), 4.28 (d, HA, JAB ˆ 13.35 Hz, 4H; ArCH2Ar), 6.56 (s, 12H;
ArH), 7.19 ± 7.31 (m, 40H; PPh2); 31P NMR (202 MHz, CDCl3): d ˆ 22.69
(d, PPh2); C84H76P4O4 ´ 2H2O (1309.43): calcd C 77.1, H 6.2; found C 77.2, H
5.8.
13.24 Hz, 4H; ArCH2Ar), 4.15 ± 4.20 (m, 8H; OCH2CH2), 6.68 (s, 8H;
ArH), 7.31 ± 7.45 (m, 24H; POPh2), 7.69 ± 7.80 (m, 16H; POPh2); 31P NMR
‡
(202 MHz, CDCl3): d ˆ ‡30.03 (POPh2); MS (ES): m/z: 1561.6 [M‡H] ;
C
100H108P4O8 ´ CH3OH (1593.90): calcd C 76.2, H 7.1; found C 76.4, H, 6.7.
25,26,27,28-Tetradiphenylphosphorylethyleneoxycalix[4]arene (1d): Calix-
arene 9d was oxidized with DMD as described to afford dealkylated
phosphine oxide 1d as a white crystalline solid in quantitative yield after
recrystallisation from methanol. M.p. 124 ± 1268C; 1H NMR (300 MHz,
CDCl3): d ˆ 2.83 (m, 8H; OCH2CH2), 2.92 (d, HB, JBA ˆ 13.62 Hz, 4H;
ArCH2Ar), 4.02 (d, HA, JAB ˆ 13.41 Hz, 4H; ArCH2Ar), 4.17 (m, 8H;
OCH2CH2), 6.54 (s, 12H; ArH), 7.31 ± 7.46 (m, 24H; POPh2), 7.67 ± 7.74 (m,
16H; POPh2); 31P NMR (202 MHz, CDCl3): d ˆ ‡29.81 (POPh2); MS
5,11,17,23,29,35-Hexa-tert-butyl-37,38,39,40,41,42-hexadiphenylphosphine-
ethyleneoxycalix[6]arene (9b): Calixarene 8b afforded the hexaphosphine
9b as a white solid in 66% yield after purification by column chromato-
graphy (SiO2, CH2Cl2/hexanes 5:1). M.p. 125 ± 1308C; 1H NMR (300 MHz,
CDCl3): d ˆ 1.10 ± 1.20 (m, 54H; C(CH3)3), 2.65 (m, 12H; OCH2CH2),
3.60 ± 4.20 (m, 24H; OCH2CH2, ArCH2Ar overlapping), 6.80 ± 7.60 (m,
72H; ArH, PPh2 overlapping); 31P NMR (202 MHz, CDCl3): d ˆ 22.05 (s,
PPh2); C150H162P6O6 (2246.79): calcd C 80.2, H 7.2; found C 79.8, H 7.5.
‡
(ES): m/z: 1337.6 [M‡H] ; C84H76P4O8 ´ H2O (1355.43): calcd C 74.4, H 5.8;
found C 74.1, H 5.7.
5,11,17,23,29,35-Hexa-tert-butyl-37,38,39,40,41,42-hexadiphenylphospho-
rylethyleneoxycalix[6]arene (1b): Calixarene 9b was oxidized with DMD
as described to furnish the corresponding phosphine oxide 1b as a white
solid in 68% yield after recrystallisation from methanol. M.p. 289 ± 2908C
(decomp); 1H NMR (300 MHz, CDCl3): d ˆ 1.10 (54H; brs, C(CH3)3), 2.95
(12H; brs, OCH2CH2), 3.90 ± 4.25 (24H; brs, OCH2CH2, ArCH2Ar
overlapping), 7.40 (brs, 36H; ArH, POPh2 overlapping), 7.85 (brs, 36H;
POPh2); 31P NMR (202 MHz, CDCl3): d ˆ ‡29.92 (POPh2); MS (ES): m/z:
37,38,39,40,41,42-Hexadiphenylphosphineethyleneoxycalix[6]arene (9e):
Calixarene 8e furnished the hexaphosphine 9e as a white solid in 51%
yield after purification by column chromatography (SiO2, CH2Cl2/hexane
5:1) and subsequent trituration in methanol. 1H NMR (300 MHz, CDCl3):
d ˆ 2.35 (m, 12H; OCH2CH2), 3.70 (m, 12H; ArCH2Ar), 3.80 (m, 12H;
OCH2CH2), 6.40 (t, 6H; ArH), 6.60 (d, 12H; ArH), 7.30 ± 7.40 (m, 60H;
PPh2). Correct elemental analysis could not be obtained for this compound
due to its susceptibility to undergo oxidation in air.
‡
1172.2 [M/2‡H] ; C150H162P6O12 ´ 4H2O (2414.79): calcd C 74.6, H 7.1;
found C 74.9, H 6.7.
5,11,17,23,29,35,41,47-Octa-tert-butyl-49,50,51,52,53,54,55,56-octadiphenyl-
phosphineethyleneoxycalix[8]arene (9c): Calixarene 8c afforded the
octaphosphine 9c as a white powdery solid in 62% yield after purification
by column chromatography (SiO2, CH2Cl2/hexane 5:1). M.p. 229 ± 2318C;
1H NMR (300 MHz, CDCl3): d ˆ 0.94 (s, 72H; C(CH3)3), 2.49 (brs, 16H;
OCH2CH2), 3.85 (brs, 16H; OCH2CH2), 3.93 (brs, 16H; ArCH2Ar), 6.75 (s,
16H; ArH), 7.10 (brs, 48H; PPh2), 7.27 (brs, 32H; PPh2); 31P NMR
(202 MHz, CDCl3): d ˆ 22.90 (PPh2). Correct elemental analysis could
not be obtained for this compound due to its susceptibility to undergo
oxidation in air.
37,38,39,40,41,42-Hexadiphenylphosphorylethyleneoxycalix[6]arene (1e):
Calixarene 9e was oxidized with hydrogen peroxide as described to furnish
the corresponding dealkylated phosphine oxide 1e as a cream solid in 62%
yield after trituration from ether. M.p. 98 ± 1028C; 1H NMR (300 MHz,
CDCl3): d ˆ 2.65 (m, 12H; OCH2CH2), 3.65 (m, 12H; ArCH2Ar), 3.93 (m,
12H; OCH2CH2), 6.43 (m, 6H; ArH), 6.49 (m, 12H; ArH), 7.35 (m, 24H;
POPh2), 7.45 (m, 12H; POPh2), 7.68 (m, 24H; POPh2); MS (ES): m/z:
‡
‡
1003.6 [M/2‡H] , 2006.5 [M‡2H] ; C126H114P6O12 (2006.14): calcd C 75.5,
H 5.7; found C 75.9, H 5.6.
5,11,17,23,29,35,41,47-Octa-tert-butyl-49,50,51,52,53,54,55,56-octadiphenyl-
phosphorylethyleneoxycalix[8]arene (1c): Calixarene 9c was oxidized with
DMD as described to furnish the corresponding phosphine oxide 1c as a
white powdery solid in 89% yield after purification by column chromatog-
raphy (SiO2, CH2Cl2/MeOH 19:1). M.p. 160 ± 1628C; 1H NMR (300 MHz,
CDCl3): d ˆ 0.92 (brs, 72H; C(CH3)3), 2.81 (brs, 16H; OCH2CH2), 3.98
(brs, 32H; OCH2CH2, ArCH2Ar overlapping), 6.75 (brs, 16H; ArH), 7.24
(brs, 48H; POPh2), 7.66 (brs, 32H; POPh2); 31P NMR (202 MHz, CDCl3):
49,50,51,52,53,54,55,56-Octadiphenylphosphineethyleneoxycalix[8]arene
(9 f): Calixarene 8 f furnished the octaphosphine 9 f as a grey solid in 48%
yield upon trituration from ice-cold ethanol. The product was subject to
ready oxidation and was used for subsequent conversion to the corre-
sponding phosphine oxide without purification.
Formation of calixarene phosphine oxides 1a ± f: Oxidation of the
calixarene phosphines 9a ± f to the corresponding phosphine oxides was
carried out using either dimethyldioxirane (DMD)[22] or hydrogen peroxide
in acetone according to the procedures subsequently outlined.
‡
d ˆ ‡29.65 (POPh2); MS (ES): m/z: 1562.0 [M/2‡H] ; C200H216P8O16 ´ H2O
(3141.72); calcd C 76.5, H 7.0; found: C 76.3, H 7.0.
Oxidation with dimethyldioxirane: The calixarene phosphine was suspend-
ed in dry acetone (ca. 5% solution) and the solution cooled to 08C with an
ice bath. Distilled dimethyldioxirane (DMD) (50% molar excess) was
added and the solution stirred at this temperature for 10 min. The yellow
solution was then allowed to warm to room temperature and was stirred at
that temperature for a further 30 min. Removal of the solvent at reduced
pressure furnished the crude calixarene phosphine oxide as a pale yellow
solid.
49,50,51,52,53,54,55,56-Octadiphenylphosphorylethyleneoxycalix[8]arene
(1 f): Calixarene 9 f was oxidized with DMD as described to afford the
corresponding dealkylated phosphine oxide 1 f as a cream solid in 48%
yield after dissolution of the crude material in methanol, filtration to
remove traces of tosylate, and concentration of the filtrate to dryness. M.p.
1
116 ± 1188C; H NMR (300 MHz, CDCl3): d ˆ 2.75 (m, 16H; OCH2CH2),
3.76 (s, 16H; ArCH2Ar), 4.05 (m, 16H; OCH2CH2), 6.58 (s, 24H; ArH),
7.30 (m, 48H; POPh2), 7.65 (m, 32H; POPh2); MS (ES): m/z: 1338.0
‡
‡
[M/2‡H] , 2674.0 [M‡H] ; C168H152P8O16 ´ 3H2O (2728.86): calcd C 74.0,
Oxidation with hydrogen peroxide: Calixarene phosphine was suspended in
dry acetone (ca. 5% solution) and the solution cooled to 08C with an ice
bath. Hydrogen peroxide (30% solution; ca. 5% molar excess) was added
and the solution stirred at this temperature for 10 min at which point it was
allowed to warm to room temperature and was stirred at that temperature
for a further 30 min. Methylene chloride was added and the solution
washed in turn with saturated sodium iodide solution (containing a
drop of acetic acid), 0.1n sodium thiosulphate solution and distilled
water. The organic layer was dried with MgSO4 and concentrated under
reduced pressure to afford the crude calixarene phosphine oxide as an
off-white solid. The following calixarene phosphine oxides were thus
prepared.
H 5.8; found C 73.9, H 5.8.
5,11,17,23,29-penta-tert-butyl-31,32,33,34,35-penta-diphenylphosphoryl-
methyleneoxycalix[5]arene (2b): p-tert-Butylcalix[5]arene[17] was treated
with Ph2P(O)CH2OTs in dry toluene containing sodium hydride exactly as
described by Matt and co-workers[5] for the tetramer analogue 2a to afford
the pentaphosphine oxide 2b in 61% yield. M.p. 220 ± 2218C; 1H NMR
(500 MHz, CDCl3): d ˆ 0.89 (s, 45H; C(CH3)3), 3.05 (d, HB, JBA ˆ 14.73 Hz,
4H; ArCH2Ar), 5.00 (s, 10H; OCH2P(O)Ph2), 5.06 (d, HA, JAB ˆ 14,65 Hz,
4H; ArCH2Ar), 6.51 (s, 10H, ArH), 7.25 (brs, 30H; m- and p-P(O)Ph2),
7.71 ± 7.75 (m, 20H; o-P(O)Ph2); 31P NMR (202 MHz, CDCl3): d ˆ ‡26.32
‡
(P(O)Ph2); MS (ES): m/z: 1880.9 [M] ; C120H125010P5 (1882.19): calcd C
76.6, H 6.7; found C 76.2, H 6.4.
5,11,17,23-Tetra-tert-butyl-25,26,27,28-tetradiphenylphosphorylethylene-
oxycalix[4]arene (1a): Calixarene 9a was oxidized with DMD as described
to furnish the phosphine oxide 1a as a white crystalline solid in 84% yield
after recrystallisation from methanol. M.p. 289 ± 2908C; 1H NMR
(300 MHz, CDCl3): d ˆ 1.05 (s, 36H; C(CH3)3), 2.90 ± 3.10 (m, 12H;
Di-tert-butylesterdiphenylphosphorylmethyleneoxycalix[4]arene
(3b):
The known di-tert-butylacetate of tert-butylcalix[4]arene was treated with
Ph2P(O)CH2OTs in dry THF/DMF containing sodium tert-butoxide exactly
as described by Matt and coworkers[5] for the diamidediphosphine oxide 3a
to afford the diesterdiphosphine oxide 3b in 69% yield. M.p. 319 ± 3208C
OCH2CH2, HB, ArCH2Ar overlapping), 4.00 ± 4.10 (d, HA, JAB
ˆ
Chem. Eur. J. 1999, 5, No. 1
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0179 $ 17.50+.50/0
179

M. J. Schwing-Weill et al.
FULL PAPER
1
1
ˆ
(decomp); IR (Kbr): nÄ ˆ 1743 cm (s, C O); H NMR (500 MHz, CDCl3):
d ˆ 0.88 (s, 18H; C(CH3)3), 1.13 (s, 18H; C(CH3)3), 1.37 (s, 18H; C(CH3)3),
2.88 (d, HB, JBA ˆ13.08 Hz, 4H; ArCH2Ar), 4.47 (s, 4H; OCH2CO2tBu),
4.60 (d, HA, JAB ˆ 12.92 Hz, 4H; ArCH2Ar), 5.78 (s, 4H; OCH2P(O)Ph2),
6.46 (s, 4H; m-ArH), 6.59 (s, 4H; m-ArH), 7.26 ± 7.43 and 7.74-7.78 (m, 20H;
P(O)Ph2); 31P NMR (202 MHz, CDCl3): d ˆ ‡25.31 (P(O)Ph2); MS (ES):
given by Equation (3). If we introduce the distribution coefficient D, as
given in Equation (4), we obtain Equation (5).
Mx‡ ‡xNO3 ‡ nL > MLn(NO3)x
Kex ˆ [MLn(NO3)x]/[Mx‡][NO3 ]x[L]n
(2)
(3)
(4)
(5)
‡
m/z: 1304.2 [M] ; C82H98010P2 (1305.63): calcd C 75.4, H 7.6; found C 75.5, H
7.3.
D ˆ [MLn(NO3)x]/[Mx‡
]
Synthesis of the bicyclic diphosphine oxide 4d: Bis-(diphenylphosphin-
yl)acetylene[21e] (0.5 g, 1.18 mmol) in chloroform (20 mL) was stirred under
nitrogen for 10 min. Cyclopentadiene (0.4 mL, 4.4 mmol) was added
through a syringe and the reaction mixture stirred under reflux for several
hours. The solvent was removed in vacuo and the residue evaporated under
high vacuum overnight to afford 4d as a white crystalline solid (0.56 g,
logD ˆ logKex[NO3 ]x ‡ nlog[L]
With these assumptions, a plot of a logD versus log[L] should be linear and
its slope should be equal to the number of ligand molecules per cation in
the extracted species. To check the validity of the method, a log ± log plot
analysis for the extraction of thorium(iv) and europium(iii) nitrates 10 4 m
by CMPO has been investigated. In agreement with previous studies, a
slope of 3.1 was found for thorium, while a lower value of 2.4 was observed
for europium.[24]
1
95%). H NMR (500 MHz, CDCl3): d ˆ 1.9 (d, J ˆ 7.3 Hz, 1H; CH2), 2.25
(d, J ˆ 6.9 Hz, 1H; CH2), 4.07 (s, 2H; CH), 6.60 (d, J ˆ 2.2 Hz, 2H; C ˆ
CH), 7.29 ± 7.61 (m, 20H; PPh2); 31P NMR (202 MHz, CDCl3): d ˆ ‡24.75
‡
(P(O)Ph2); MS (EI): m/z: 492.14 [M] ; C31H26O2P2 (492.14): calcd C 75.7, H
5.3; found C 76.2, H 5.4.
Extraction measurements
Loading capacity measurements: The loading capacity of the organic phase
Picrate extraction method: The classical picrate extraction method for alkali
and alkaline earth cations with calixarenes has already been described.[22]
In summary, the percentage cation extracted (%E) or the distribution
coefficient D [D ˆ %E/(100 %E)] from water into dichloromethane, at
208C, was determined from the measured absorbance of the picrate anion
remaining in the aqueous phase after extraction at 355 nm. The same
concentration (2.5 Â 10 4 m) was used for the ligand in the organic phase
and the picrate in the aqueous phase.
for thorium has been determined according to the following procedure:
5 mL of the organic phase (2 Â 10 4 m) were vigorously shaken with 5 mL of
3
an aqueous phase containing increasing amounts of cation nitrate (10
±
0.1m) and 1m HNO3 during 15 h. After two hours to allow phase separation,
a given aliquot of the organic phase was mineralized by evaporation at
1008C in the presence of a few drops of 65% HNO3 followed by calcination
at 8008C. The residual thorium was determined by titration according to
the procedure described above.
Stability constant determinations: The stability constant of a complex
formed in a homogeneous medium is strictly speaking a more representa-
tive physicochemical characteristics of the intrinsic affinity of a ligand for a
cation in this given medium than an extraction constant. Stability constants
enable comparison to be made between different ligands, without any
interference from lipophilicity differences which do arise in extraction.
Furthermore, they allow comparison with other more classical macrocyclic
ligands. The stability constants bxy of the complexes, expressed as
concentration ratios according to the general complexation equilibrium
[Eq. (6)] were determined in methanol, at 258C and constant ionic strength
I (Et4NCl), by UV spectrophotometry. Upon addition of the metal
Nitrate extraction method: The extrapolation of the picrate extraction
method to the trivalent lanthanides and tetravalent thorium led to unusual
and so far unexplained selectivities in the lanthanide series, that is, a strong
extraction maximum for Eu3‡
. Since our objective was to develop
extraction processes for lanthanides and actinides that would mimic their
removal from nuclear waste, it was necessary to find an alternative to the
picrate extraction method, which would allow the extraction of metal
nitrates. Furthermore, because liquid nuclear waste has a relatively high
concentration of nitric acid, extractants were needed which would function
in acidic media. For these reasons we have developed a new method for
monitoring metal nitrate extraction in the presence of excess nitric acid,
which is also applicable, with certain approximations, to the determination
of the stoichiometries of the extracted species.
mMx‡ ‡ nL > MmLmn x‡
(6)
Metal nitrate extraction procedure:
A standard experimental set up
chlorides to a solution of the ligand in methanol, the UV-visible spectrum
of the ligand generally undergoes small changes, in the 250 to 300 nm range,
which, in most cases, are sufficient to allow a multiwavelength treatment of
the data by the computer programs Letagrop-Spefo[25] or Sirko.[26] As
evidence was only found for 1:1 complexes, b11 can be abbreviated in the
following text as b. The values given in the data tables correspond to the
arithmetic means of at least three independent determinations. The method
is only appropriate in the stability range 1 ꢀ logb ꢀ 6.
consisted of an aqueous phase containing europium or thorium nitrate
(10 4 m) in 1m HNO3; the organic phase was a solution of calixarene in
dichloromethane, at a concentration adapted to an extraction percentage
ranging from 10 ± 90%. Each phase (1 mL) was thoroughly mixed by
stirring in a stoppered tube at 208C for 12 h. After separation of the two
phases, the concentration of the cation remaining in the aqueous phase was
monitored spectrophotometrically with arsenazo(iii) (2,2''-[1,8-dihydroxy-
3,6-disulpho-2,7-naphtalenebis(azo)]dibenzenearsenic acid) as a coloured
indicator.[23] Arsenazo solution (0.5 mL of 6 Â 10 4 m) was added to a
0.7 mL aliquot of the aqueous phase and the volume was finally adjusted to
50 mL with either a sodium formate/formic acid buffer (pH 2.80) for the
determination of europium or with a 4m nitric acid solution for the
determination of thorium. The absorbances were then determined at
665 nm for thorium and 655 nm for europium. Since the concentration of
arsenazo(iii) is at least 30 times higher than the concentration of the cation,
complete complexation of the cation can be assumed. The extraction
percentage was derived from Equation (1), in which A0 is the absorbance of
the arsenazo solution without the cation and A1 the absorbance of the
blank.
In the case of the short, that is, with one methylene spacer pentamer
phosphine oxide 2b (hereafter, for convenience the phosphine oxides with
two methylene spacers are referred to as long) with the alkaline earth
cations, the stability constants were also investigated by a competitive
‡
potentiometric method with Ag as the auxiliary cation. Nevertheless the
magnitude of the stability constant of the silver complex (3.62 log units)
only allowed a determination of logb for Mg2‡ (4.72 log units), the other
complexes being too strong.
Results and Discussion
%E ˆ 100[(A1 A)/(A1 A0)]
(1)
Log ± log plot analysis: In order to characterize the extraction ability, the
dependence of the distribution coefficient D of a cation between the two
phases upon the calixarene concentration has been examined. If the
Alkali cations: The results obtained with tetramers 1a, 1d and
2a, pentamer 2b and the two mixed tetramers 3a and 3b are
given in Table 1 and are illustrated graphically in Figure 1.
The results clearly reveal an enormous difference in behav-
iour between the tetrameric long phosphine oxide 1a, which
general extraction equilibrium is assumed to be Equation (2) with Mx‡
ˆ
metal ion, L ˆ neutral ligand and with the overlined species referring to
species in the organic phase, the overall extraction equilibrium constant is
180
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0180 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1

Calixaryl Phosphine Oxides
175±186
Table 1. Alkali cations: Percentage extraction %E from water into
extractant, as the stability constants also increase and range
2
dichloromethane and, in brackets, logb in methanol (10 ꢀ ionic strength
‡
‡
from 2.3 for Na to 4.0 for Cs . Clearly the larger cavity of the
I ꢀ 4.5 Â 10 2 m in Et4NCl).
‡
pentamer is better suited for the larger Cs cation than its
‡
‡
‡
‡
‡
Li
Na
K
Rb
Cs
tetra analogue. The characteristics of the two mixed tetramers
3a and 3b are strongly dependent on the nature of the
groups replacing two of the phosphine oxides. Both com-
1a
1d
2a
2b
3a
3b
0.7[a]
±
0.5[a]
±
0.3[a]
±
3.8[a]
0.5[a]
±
0.4[a]
±
3.5[a]
±
3.7[a]
±
3.6[a]
±
3.6[a]
±
‡
pounds show a peak selectivity for Na , as do the corre-
±
sponding tetraamides and tetraesters.[3] The amide groups
7.6[a]
(ꢀ1)
28.5[e]
(ꢀ1)
53.4[a]
(2.6)[b]
8.5[a]
±
6.0[a]
(ꢀ1)
37.9[e]
(2.34)[i]
65.1[a]
(4.5)[b]
17.9[a]
(2.9)[d]
22.3[a]
(1.9)[b]
45.0[f]
(3.28)[j]
42.6[a]
(3.3)[b]
13.7[a]
(2.4)[b]
17.0[a]
(1.4)[b]
50.7[g]
(3.9)[c]
20.0[a]
(2.8)[b]
5.3[a]
±
6.4[a]
±
‡
lead to the highest efficiency for Na , higher than the
56.9[h]
(4.02)[i]
13.0[a]
(ꢀ1)
5.0[a]
±
ester groups, as is predictable from the behaviour of the
‡
homo-substituted analogues. The efficacy order for Na
extraction is 3a > 2b > 3b > 2a > 1d > 1a. This order is
unchanged in terms of stability constants, except the inver-
sion between 2b and 3b and these conclusions are
clearly illustrated in Figures 1a and b. The results are
consistent with the extraction studies made from water to
dichloroethane by other authors for compound 2a, but not for
[a] 0.1 ꢀ sN 1 ꢀ 0.5; [b] sN 1 < 0.1; [c] sN ˆ 0.05; [d] sN ˆ 0.03; [e] sN
ˆ
1
1
1
0.8; [f] sN ˆ 0.7; [g] sN ˆ 0.1; [h] sN ˆ 0.3; [i] sN ˆ 0.02; [j] sN ˆ 0.01
1
1
1
1
1
‡
3a for which these authors observe a decrease from Li to
‡ [11]
Na .
Alkaline earth cations: The results for the same series of
ligands are given in Table 2 and are illustrated in Figure 2.
Clearly the long tetraphosphine oxides 1a and 1d have no
extraction efficacy towards the alkaline earth cations. The p-
dealkylation only slightly increases the extraction power of
1a. No attempt was made to determine the stability constants
of their complexes, as it was assumed, on the basis of the
extraction results, that they would be very small.
The mixed short diphosphine oxide 3b also displayed no
extraction efficacy towards the alkaline earth cations, in
agreement with the generally small affinity of the calixaryl
esters for alkaline earths.[3] The mixed short diamide 3a and
the short phosphine oxide 2a display medium extraction
efficiencies (%E < 25). The stability constants of the com-
plexes fall within the limits of the spectrophotometric method
only for the calcium complex of 2a (2.3), the calcium,
strontium and barium complexes of 3a (4.4, 4.5 and 3.4,
respectively), and the calcium complex of 3b (1.8). The
stability of the calcium complex is higher for 3a than for 2a in
Table 2. Alkaline earth cations: extraction percentages %E from water
into dichloromethane at 208C (0.1 ꢀ sN 1 ꢀ 0.5) and, in brackets, log b
values determined by spectrophotometry, in methanol, at 258C (one
Figure 1. a) Extraction percentages (%E) of alkali picrates from water
into dichoromethane by selected phosphine oxide calixarenes at 208C;
b) Logb values in methanol for the complexation of alkali cations by
selected phosphine oxide calixarenes at 258C, ionic strength between 10
and 4.5 Â 10 2 m in Et4NCl.
2
experiment only). Ionic strength I ranging from 10 to 4.5 Â 10 2 m
(Et4NCl), unless otherwise stated.
2
Mg2‡
Ca2‡
Sr2‡
Ba2‡
1a
1d
2a
2b
3a
3b
0.6
±
3.0
±
5.6
(ꢀ1)
18.2
(4.72)[a]
3.7
±
2.6
±
5.0
±
23.0
(2.35)[b]
62.5
(ꢁ6)
15.8
(4.4)
1.1
0.8
±
3.3
±
9.8
(ꢀ1)
52.2
(ꢁ6)
15.4
(4.5)
1.4
±
0.5
±
3.1
±
8.4
(ꢀ1)
62.3
(ꢁ6)
12.5
(3.4)
1.4
±
does not extract the alkalis (or very little when p-dealkylated
as in 1d), and the tetrameric short phosphine oxide 2a, which
displays an extraction and complexation peak selectivity in
‡
favour of K (%E ˆ 22, logb ˆ 1.9). The extraction ability of a
short phosphine oxide is very much enhanced when the
condensation is increased up to five, along with a change in
the selectivity pattern, which now displays a continuous
increase in extraction efficacy as the cation ionic radius
increases; this increase from tetramer 2a to pentamer 2b
cannot be accounted for by an increasing lipophilicity of the
1.2
±
(1.8)
‡
[a] Competitive potentiometry determination with Ag as auxiliary cation:
‡
I ˆ 10 2 m in Et4NClO4, log b(Ag ) ˆ 3.62 Æ 0.01, sN ˆ 0.09; [b] sN ˆ 0.1
1
1
Chem. Eur. J. 1999, 5, No. 1
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0181 $ 17.50+.50/0
181

M. J. Schwing-Weill et al.
FULL PAPER
the value reached by the tetraamide (98%) and hexaamide
(84%), but is superior to that reached by the p-dealkylated
hexaamide (25%). Moreover, the extraction percentage for
strontium by the pentaphosphine oxide 2b (52% ) is almost as
high as that displayed by the corresponding pentadiethyl-
amide (55%).
Europium(iii) and thorium(iv): It is of major environmental
concern to know how to remove from liquid radioactive
wastes, before their disposal, the harmful long-lived nuclides,
such as caesium and strontium, and the a-emitting actinides.
Of the various extractants used in actinide process chemistry,
neutral organophosphorus compounds are among the most
useful.[9] The PUREX process for plutonium/uranium sepa-
ration, for example, is based on the extracting ability of
tributylphosphate.[10] Efforts to improve the performance of a
group of simple monofuntional organophosphorous extrac-
tants, such as the well-known trioctylphosphine oxide (TOPO,
[(C8H17)3PO]), led to the development of bifunctional
analogues that include carbamoylmethylphosphonates
[(RO)2PO(CO)NR2] (CMPs) and carbamoylmethylphos-
phine oxides [R2POCH2(CO)NR2] (CMPOs), among which
is the classical octylphenyldiisobutylcarbamoylmethylphos-
phine oxide [Ph2POCH2CON(C8H17)2] known as CMPO. In
our study, we decided to extend our extraction and stability
constant measurements to EuIII and ThIV, using the former to
mimic the trivalent actinides and the latter to mimic the
tetravalent actinides present in liquid nuclear waste. The
calixaryl phosphine oxides used for this purpose were 1a ± f, as
well as four other phosphine oxide calixarenes of group 1, with
either dodecyl groups in para position instead of tert-butyl
groups (n ˆ 4) or with n-butyl groups on the phosphorus
instead of phenyl groups (n ˆ 4, 6, 8). Using calixarenes as
building blocks, we hoped to exploit the synergism that might
accrue from combining several phosphine oxide and ether
podands with the receptor potential of macrocycles of the
calixarene family. For comparison purposes, the classical
TOPO and CMPO extractants were also tested, as well as
some bi- and trifunctional noncalixarene phosphine oxides
4a ± e.
Figure 2. a) Extraction percentages (%E) of alkaline earth picrates from
water into dichloromethane by selected phosphine oxide calixarenes at
208C; b) Logb values in methanol for the complexation of alkaline earth
cations by selected phosphine oxide calixarenes at 258C, ionic strength
between 10 2 and 4.5 Â 10 2 m in Et4NCl.
contrast with the trend observed in extraction: this may reflect
the greater influence of lipophilicity expected for 2a, because
of the larger number of aromatic rings. The logb values for 3a
are approximately half the values reached with the corre-
sponding tetradiethylamide.[22]
Evidently, the most efficient compound among the six
studied is the short pentamer 2b, for which %E reaches 62%
for Ca2‡ and Ba2‡, and 18% for Mg2‡. The stability constants
are too high to be evaluated by any of the employed methods
(competitive potentiometry gave a value of 4.7 for magne-
sium). Compounds 3a, 2a and 2b all show an extraction
selectivity for calcium, which can be considered as a plateau
selectivity from calcium to barium for the pentamer and the
mixed diamide, and a peak selectivity for the tetramer 2a.
One may conclude that, in extraction as well as in stability
constants for alkaline earths, the pentameric phosphine oxide
2b is the most efficient of all compounds tested in the present
work. The same conclusion was reached in the alkali series,
but for cesium and rubidium only, the lighter alkalis being
even better extracted or complexed by the mixed diamide
tetramer.
The present results demonstrate i) the very big effect of the
shortening of the carbon chain bearing the phosphine oxide
groups (compare 2a with 1a) and ii) the tremendous effect of
increasing the condensation degree from 4 to 5 (compare 2a
with 2b). The comparison between the calix[5]phosphine
oxide and the corresponding calix[4]-, calix[5]- and calix[6]-
diethylamides shows that the level reached for the calcium
extraction by the pentamer 2b (62%) is certainly inferior to
As the picrate extraction method was not applicable to
experiments intended to mimic the situation with real liquid
wastes, which contain nitrates and nitric acid, we used the new
nitrate extraction technique described above. The results of
this study are presented in Table 3 as percentages of cation
extraction into dichloromethane for selected analytical calix-
arene concentrations, and in Figure 3 as logD versus the log of
free calixarene concentration at equilibrium in the organic
phase for thorium and europium (see Experimental Section
for details).
The main conclusions concerning the calixarene phosphine
oxides are the following:
The long phosphine oxide series 1a ± f:
iii) All calixarenes tested are better extractants for europium
and thorium than TOPO and CMPO. The only exception
concerns the extraction of europium by the octamer 1c.
iii) Thorium is always better extracted than europium.
182
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0182 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1

Calixaryl Phosphine Oxides
175±186
Table 3. Percentage extraction (% E) of thorium(iv) and europium(iii) nitrates
(10 4 m) from 1m HNO3 aqueous solutions into dichloromethane containing the
ligand at various concentrations.
hexamer> octamer> tetramer. In the alkylated series
1a ± c, the hexamer is the best extractant for thorium,
followed by the tetramer and the octamer. With euro-
pium, there is no extraction by the tetramer and the
octamer at concentrations lower than 0.1m. According to
the data in Table 3, the hexamer is better than the
tetramer at a concentration of 2.5 Â 10 2 m, but the log ±
log plot shows an unexpected behaviour of the hexamer,
not represented on Figure 3. The replacement of the p-
tert-butyl groups by dodecyl groups slightly decreases the
extraction of thorium, but increases the extraction of
europium. The replacement of the phenyl groups on the
PO functions by n-butyl groups leads to a complete loss of
extraction efficiency, since there is no extraction at a
ligand concentration 10 3 m, that is, ten times higher than
the cation concentration.
ThIV
EuIII
10 4 m 5 Â 10 4 m 10 3 m 5 Â 10 3 m 10 2 m 2.5 ´ 10 2 m 2.5 ´ 10 2 m
TOPO
CMPO
0
0
0
0
0
0
1.4
1.6
10.2
12.2
64.2
70.4
0[a]
0[b]
1a
1d
1g
0
2.6
5.6
46.9
26.8
79.8
55.0
96.1
100
100
100
100
100
0
24.6
29.6
1b
1e
0
9.4
23.6
79.4
62.5
94.8
98.8
100
30.6
100
100
100
100
100
20.8
94.3
1c
1 f
0
5.1
0
62.6
0
87.9
60.1
100
88.4
100
0
92.0
2a
2b
31.5
55.5
100
86.2
[c]
[c]
[c]
[c]
[c]
[c]
4a
4b
4c
4d
4e
20
100
38
95
100
87
ꢁ 95
±
100
8
ꢁ 95
±
iv) With the classical extractant TOPO with only one
phosphine oxide function per molecule, the slopes of the
log ± log plot analyses are 3 for both thorium and euro-
pium as expected from previous studies.[24] With CMPO,
the slope for thorium has the expected value of 3, but in
the case of europium, the slope is only 2.5. A similar
deviation with respect to the expected slope of 3 was
found by Horwitz and co-workers[27] for the extraction of
americium(iii) into toluene by CMPO, and attributed to a
marked decrease in the measured activity coefficient of
CMPO in toluene as water and nitric acid are also
extracted.
95
85
[a] % E ˆ 18 (CL ˆ 0.25m); [b] % E ˆ 69.5 (CL ˆ 0.25m); [c] No measurement
possible, formation of a film at the interface.
The slopes of the linear plots for the extraction of thorium
by the three dealkylated calixarene phosphine oxides 1d ± f
are close to 2.0, indicating at first sight that the compounds
may extract the cation as 1:2 species, that is, two calixarenes
for one cation. The expected result for a 1:1 complex should
be 1, assuming that one calixarene has enough PO functions to
bind either thorium(iv) or europium(iii). In order to verify the
slope, saturation experiments of the organic phase by
equilibration with increasing cation concentrations in the
aqueous phase (see Experimental Section) were performed
with compound 1d, at a concentration of 2 Â 10 4 m. The
results, shown in Figure 4, indicate that a plateau is reached
for a concentration of thorium in the organic phase equal to
10 4 m, that is, half the concentration of the ligand. This
observation confirms the results of the log ± log plot analysis
for the calixarene 1d, at least in our experimental conditions
in which the ligand was in excess. The deviation from the ideal
expected slope of 1 is even bigger with the p-tert-butyl series,
for which slopes were found to range between 1.8 and 2.6. The
formation of bisligand species of calixarenes with rare earth
cations has already been demonstrated in the case of
carboxylate derivatives of calixarenes:[28±30] it seems therefore
reasonable to conclude that such bisligand species can also be
formed between thorium(iv) and phosphine oxide derivatives
of calixarenes. However, one cannot totally exclude the
posibility of the influence of other factors on the observed
slopes, such as coextraction of nitric acid and activity
coefficients effects.
Figure 3. Extraction of a) thorium(iv) and b) europium(iii) by phosphine
oxide calixarenes and related classical extractants TOPO and CMPO from
metal nitrate solutions (10 4 m) 1m in HNO3 into dichloromethane at 208C.
LogD vs logarithm of the free ligand concentration (numbers in brackets
are the values of the slopes).
The extraction of europium by the dealkylated tetramer 1d,
the octamer 1 f and by the p-tert-butyl tetramer 1a is
characterised by log ± log plot slopes ranging between 2.0
iii) The dealkylated series is better than the alkylated series,
both for thorium and europium. In the dealkylated series
1d ± f, the sequence of efficiency towards both cations is:
Chem. Eur. J. 1999, 5, No. 1
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0183 $ 17.50+.50/0
183

M. J. Schwing-Weill et al.
FULL PAPER
ene platform and the possibility
of preorganization effects on
the extraction ability of the
polyfunctional phosphine ox-
ides, some simple noncalixar-
ene phosphine oxides have
been tested as simple reference
compounds. The results are giv-
en in Table 3 for the most
relevant compounds.
Four compounds, the bifunc-
tional derivatives 4a, 4c and 4d
and the trisphosphine oxide 4e
have an extraction efficiency
for thorium ranging between
87 and 100% at a ligand con-
centration 10 3 m. They can con-
sequently be considered as
good competitors of the calix-
Figure 4. Loading curve for the saturation of a 2 Â 10 4 m dichloromethane solution of calixarene 1d by thorium
nitrate from an aqueous solution 1m in HNO3.
arene phosphine oxides of Table 3. Compound 4a even
reaches, at a ligand concentration of 10 4 m, an extraction
efficiency for thorium(iv) of 20%, which is only slightly lower
than the good extraction performances of the CMPO-like
calixarenes which reach 40 ± 50% extraction under the same
conditions.[8] It may be seen from the comparison between 4a
and 4b that the efficiency decreases with the number of
carbons between the two phosphine oxide functions. Con-
cerning europium, 4a and 4c are better than all calixarenes
tested.
and 2.5, as in the case of thorium. The dealkylated hexamer 1e
gives a slope close to 3, as does TOPO just as if only one
phosphine oxide per calixarene was involved in the complex
formation.
The short tetramer phosphine oxide 2a: The shortening of the
carbon chain between the phenolic oxygen and the phospho-
rus atom by one methylene spacer as in tetraphosphine oxide
2a leads to a big increase of the extraction efficiency with
respect to 1a: at a calixarene concentration of 5 Â 10 4 m, 31%
ThIV is extracted instead of 5.6%, and at concentration of
10 3 m, 55% instead of 27%. However, the changes thus
induced by shortening of the spacer are not as effective as
those resulting from the dealkylation of the para position of
1a. For europium(iii), the shortening of the spacer produces
an even bigger change, as the extraction percentage increases
from 0% with 1a to 86% with 2a, for a ligand concentration
of 2.5 10 2 m. Compound 2a appears to be the best extractant
for europium among all the homo-substituted phosphine
oxides studied in the present work.
Effect of HNO3 and NaNO3 concentrations: All the extraction
work described in the previous sections refer to aqueous
source solutions 1m in nitric acid. A preliminary investigation
of the applicability of these results to the separation of
actinides from the medium level activity liquid wastes, which
are usually of high salinity and acidity (1m in HN03 and 4m in
NaNO3) before embedding of the latter in sub-surface
repositories, led us to study the influence of the concentra-
tions of acid and salt on the extraction ability of the
calixarenes.
The effect of the nitric acid concentration, from 0.1
to 1.3m, is represented in Figure 5, for the p-tert-butyl
tetramer 1a and the dealkylated hexamer 1e, as well
as for the classical extractant TOPO, in the same experi-
mental conditions as those described in the Experimental
Section.
Below a nitric acid concentration of 1m, both calixarenes
show the same decrease in extraction with HNO3 increasing
concentration, as TOPO. However, compound 1e exhibits a
slight increase of the extraction efficiency relative to TOPO
when the nitric acid concentration is higher than 1m. This last
trend is probably interpretable in terms of concomittant
competing effects: coextraction of nitric acid, which tends to
decrease the extent of extraction, and the salt effect of the
anion N03 , which leads to an increase of the extraction. This
salt effect was evidenced by our study of the influence of the
sodium nitrate concentration, from 0 to 4m, on the ThIV
extraction by the hexamer 1e at the concentration 10 3 m, in
presence of HNO3 1m: we found that the higher the
The short pentamer phosphine oxide 2b: The effect of the
increase of the calixarene nucleus to 5, producing the
pentamer 2b, could not be evaluated by the nitrate extraction
method used here: thorium extraction was attempted with
ligand concentrations ranging from 2.5 Â 10 4 m to 4 Â 10 2 m.
The results showed surprisingly the ligand to be an extremely
inefficient extractant, with %E ˆ 6% at 10 2 m and 7.4% at
4 Â 10 2 m. A careful examination of the extraction tubes
showed the presence of a sticky film on the upper surface of
the aqueous layer, even more apparent at higher ligand
concentrations. Europium extraction was also tried with 2b,
but the formation of a film on the upper surface of the
aqueous layer was also observed. This competing phenomen-
on is probably the cause of the apparent poor results of the
extraction power of 2b.
The nonmacrocyclic phosphine oxides 4a ± f: For a better
understanding of the importance of the macrocyclic calixar-
184
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0184 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1

Calixaryl Phosphine Oxides
175±186
[1] a) C. D. Gutsche, Calixarenes, Monographs in Supramolecular Chem-
istry (Ed.: J. F. Stoddart), The Royal Society of Chemistry, 1989;
b) Calixarenes, a Versatile Class of Macrocyclic Compounds (Eds.: J.
Vicens, V. Böhmer), Kluwer, Dordrecht, 1991; c) Calixarenes 50th
Anniversary: Commemorative Issue (Eds.: J. Vicens, Z. Asfari, J. M.
Harrowfield), reprinted from J. Incl. Phenom. Mol. Recogn. Chem.,
Kluwer, Dordrecht, 1994, 19, (1 ± 4).
[2] V. Böhmer, Angew. Chem. 1995, 107, 785; Angew. Chem. Int. Ed. Engl.
1995, 34, 713.
[3] M. A. McKervey, M. J. Schwing-Weill, F. Arnaud-Neu, in Compre-
hensive Supramolecular Chemistry (Ed.: G. W. Gokel), Pergamon,
1996, 1, 537.
[4] a) A. Casnati, A. Pochini, R. Ungaro, F. Ugozzoli, F. Arnaud, S.
Fanni, M. J. Schwing, R. J. M. Egberink, F. de Jong, D. N. Reinhoudt,
J. Am. Chem. Soc. 1995, 117, 2767; b) A. Casnati, A. Pochini, R.
Ungaro, C. Bocchi, F. Ugozzoli, R. J. M. Egberink, H. Strujik,
R. Lugtenberg, F. de Jong, D. N. Reinhoudt, Chem. Eur. J. 1996, 2,
436.
[5] a) C. Loeber, D. Matt, A. De Cian, J. Fischer, J. Organomet. Chem.
1994, 475, 297; b) C. Loeber, C. Wieser, D. Matt, A. De Cian, J.
Fischer, L. Toupet, Bull. Soc. Chim. Fr. 1995, 132, 166; c) C. Dieleman,
C. Loeber, D. Matt, A. De Cian, J. Fischer, J. Chem. Soc. Dalton Trans.
1995, 3097.
[6] J. F. Malone, D. J. Marrs, M. A. McKervey, P. OꢀHagan, N. Thompson,
A. Walker, F. Arnaud-Neu, O. Mauprivez, M. J. Schwing-Weill, J. F.
Dozol, H. Rouquette, N. Simon, J. Chem. Soc. Chem. Commun. 1995,
2151.
Figure 5. Influence of the nitric acid concentration in the aqueous phase
on the extraction (logD) of thorium(iv) nitrate from aqueous solution into
dichloromethane solutions of TOPO and calixarenes 1a and 1e at 208C
([TOPO] ˆ 0.025 m; [1a] ˆ [1c] ˆ 10 3 m).
[7] a) J. F. Dozol, V. Böhmer, M. A. McKervey, F. Lopez-Calahorra, D.
Reinhoudt, M. J. Schwing, R. Ungaro, G. Wipff, European Commis-
sion, Nuclear Science and Technology, 1997, Report EUR 17615 EN;
b) J. F. Dozol, F. Lopez-Calahorra, M. A. McKervey, V. Böhmer, R.
Ungaro, M. J. Schwing, F. Arnaud, D. Reinhoudt, G. Wipff, Proceed-
ings 4th EU Conference on Management and Disposal of Radioactive
Waste, (Ed.: T. Menamin), 1997, 104.
[8] F. Arnaud-Neu, V. Böhmer, J. F. Dozol, C. Grüttner, R. A. Jakobi, O.
Mauprivez, H. Rouquette, M. J. Schwing-Weill, N. Simon, W. Vogt, J.
Chem. Soc. Perkin Trans. 2. 1996, 1175.
[9] K. L. Nash, Solvent Extr. Ion Exch. 1993, 11, 729.
[10] J. Stary, Talanta, 1966, 13, 42.
[11] M. R. Yaftian, M. Burgard, D. Matt, C. Wieser, C. Dieleman, J. Incl.
Phenom. 1997, 27, 127.
[12] M. R. Yaftian, M. Burgard, A. El Bachiri, D. Matt, C. Wieser, C. B.
Dieleman, J. Incl. Phenom. 1997, 29, 137.
[13] D. D. Perrin, W. L. F. Amarego, D. R. Perrin, in Purification of
Laboratory Chemicals, Pergamon, 1966.
[14] C. D. Gutsche, M. Iqbal, Org. Synth. 1989, 68, 234.
[15] F. Arnaud-Neu, E. M. Collins, M. Deasy, G. Ferguson, S. J. Harris,
B. Kaitner, A. J. Lough, M. A. McKervey, E. Marques, B. J. Ruhl,
M. J. Schwing-Weill, E. Seward, J. Am. Chem. Soc. 1989, 111,
8681.
[16] a) C. D. Gutsche, J. A. Levine, J. Am. Chem. Soc. 1982, 104, 2652;
b) C. D. Gutsche, J. A. Levine, P. K. Sujeeth, J. Org. Chem. 1985, 50,
5802.
[17] D. R. Stewart, C. D. Gutsche, Org. Prep. Proced. Int. 1993, 25,
137.
[18] C. D. Gutsche, B. Dhawan, M. Leonis, D. R. Stewart, Org. Synth. 1989,
68, 238.
[19] J. H. Munch, C. D. Gutsche, Org. Synth. 1989, 68, 243.
[20] J. K. Moran, E. M. Georgiev, A. T. Yordanov, J. T. Mague, D. M.
Roundhill, J. Org. Chem. 1994, 59, 5990.
[21] a) T. S. Lobana, R. Singh, Transition Met. Chem. 1995, 20, 501; b) and
c) V. Munyejabo, P. Guillaume and P. Postel, Inorg. Chem. Acta, 1994,
221, 133; d) P. A. W. Dean, M. K. Hughes, Can. J. Chem. 1980, 58, 180;
e) C. Charrier, W. Chodkiewicz, P. Cadiot, Bull. Soc. Chim. Fr. 1966,
1002.
concentration of NaNO3, the better the extraction of ThIV, a
situation that is very favourable to the extension of the
present studies to the further treatment of medium-level
activity liquid wastes.
Stability constants: Preliminary spectrophotometric studies
of the stability constants of the complexes of thorium(iv) with
the calixarenic phosphine oxides 1a, 1b and 2b in methanol
(I ˆ 0.01m Et4NCl) have been completed. The hydroxo species
of the cations have not been taken into account in these
studies. The following values have been obtained for logb of
1:1 complexes: 1a(Th4‡) ˆ 3.5; 1b(Th4‡) ˆ 3.64 Æ 0.08, where-
as with the same method values for Th4‡ of 3.42 Æ 0.06 and
3.64 Æ 0.08 were found for the p-tert-tetradiethyl ester and the
p-tert-hexadiethylamide, respectively.[31] Apparently, from
these results, the nature of the functional group on the lower
rim of these tetramers should play no significant role in the
stability of the thorium complexes in methanol. The situation
seems to be different with the short pentamer: logb for
2b(Th4‡) was found equal to 4.99 Æ 0.08. The value of logb of
the 1:1 YbIII complex of 2b has also been determined and
found to be very high (6.5 Æ 0.1). The value for the 1:1 ThIV
complex of 2a could not be obtained from spectrophotomet-
ric measurements because of insufficient spectral changes
upon complexation in identical conditions. Work is in progress
in order to confirm and extend these preliminary studies by
other methods, in other solvents.
[22] F. Arnaud-Neu, M. J. Schwing-Weill, K. Ziat, S. Cremin, S. J. Harris,
M. A. McKervey, New. J. Chem. 1991, 15, 33.
Acknowledgements
[23] Z. Marczenko, Spectrophotometric Determination of Elements, Wiley,
New York, 1976, 442.
[24] E. P. Horwitz, H. Diamond, K. A. Martin, Solvent Extr. Ion Exch.
1987, 5, 447.
Part of this work was conducted under the European Union RTD
Programme on Nuclear Waste Management and Storage (Contract
F12W-CT90-0062). J.B. thanks the Northern Ireland Department of
Education for a distinction award.
[25] L. G. Sillen, B. Warnquist, Ark. Kemi. 1968, 31, 377.
Chem. Eur. J. 1999, 5, No. 1
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0185 $ 17.50+.50/0
185

M. J. Schwing-Weill et al.
FULL PAPER
[26] V. I. Vetrogon, N. G. Lukyanenko, M. J. Schwing-Weill, F. Arnaud-
Neu, Talanta, 1994, 41, 2105.
[30] T. Kakoi, T. Nishiyori, T. Oshima, F. Kubota, M. Goto, S. Shinkai, F.
Nakashio, J. Membr. Sci. 1997, 136, 261
Â
Á
Â
[27] H. Diamond, E. P. Horwitz, P. R. Danesi, Solvent Extr. Ion Exch. 1986,
4, 1009.
[31] P. Schwinte, These de doctorat de lꢀUniversite Louis Pasteur,
Strasbourg, 1995.
[28] R. Ludwig, K. Inoue, T. Yamato, Solvent Extr. Ion Exch. 1993, 11, 311.
[29] K. Ohto, M. Yano, K. Inoue, T. Yamamoto, M. Goto, F. Nakashio, T.
Nagasaki, Anal. Sci. 1995, 11, 893.
Received: March 5, 1998 [F1039]
186
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0186 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1