Investigating the binding properties of porous drug delivery systems using nuclear sensors (radiotracers) and positron annihilation lifetime spectroscopy - Predicting conditions for optimum performance

Article abstract of DOI:10.1039/c0dt01499k

Understanding how the size, charge and number of available pores in porous material influences the uptake and release properties is important for optimising their design and ultimately their application. Unfortunately there are no standard methods for screening porous materials in solution and therefore formulations must be developed for each encapsulated agent. This study investigates the potential of a library of radiotracers (nuclear sensors) for assessing the binding properties of hollow silica shell materials. Uptake and release of Cu²⁺ and Co²⁺ and their respective complexes with polyazacarboxylate macrocycles (dota and teta) and a series of hexa aza cages (diamsar, sarar and bis-(p-aminobenzyl)-diamsar) from the hollow silica shells was monitored using their radioisotopic analogues. Coordination chemistry of the metal (M) species, subtle alterations in the molecular architecture of ligands (Ligand) and their resultant complexes (M-Ligand) were found to significantly influence their uptake over pH 3 to 9 at room temperature. Positively charged species were selectively and rapidly (within 10 min) absorbed at pH 7 to 9. Negatively charged species were preferentially absorbed at low pH (3 to 5). Rates of release varied for each nuclear sensor, and time to establish equilibrium varied from minutes to days. The subtle changes in design of the nuclear sensors proved to be a valuable tool for determining the binding properties of porous materials. The data support the development of a library of nuclear sensors for screening porous materials for use in optimising the design of porous materials and the potential of nuclear sensors for high through-put screening of materials.

Full text of DOI:10.1039/c0dt01499k

View Article Online / Journal Homepage / Table of Contents for this issue
This article is published as part of the Dalton Transactions themed issue entitled:
Radiopharmaceuticals for Imaging and Therapy
Guest Editors Stephen Faulkner (Oxford University) and Nick Long (Imperial College)
Published in issue 23, 2011 of Dalton Transactions
Image reproduced with permission of Martin W. Brechbiel
Articles in the issue include:
PERSPECTIVES:
Multimodal radio- (PET/SPECT) and fluorescence imaging agents based on metallo-radioisotopes:
current applications and prospects for development of new agents
Flora L. Thorp-Greenwood and Michael P. Coogan
Dalton Trans., 2011, DOI: 10.1039/C0DT01398F
Radiometallated peptides for molecular imaging and targeted therapy
João D. G. Correia, António Paulo, Paula D. Raposinho and Isabel Santos
Dalton Trans., 2011, DOI: 10.1039/C0DT01599G
Towards translation of 212Pb as a clinical therapeutic; getting the lead in!
Kwon Yong and Martin W. Brechbiel
Dalton Trans., 2011, DOI: 10.1039/C0DT01387K
ARTICLES:
First dinuclear Re/Tc complex as a potential bimodal Optical/SPECT molecular imaging agent
Alexandre Boulay, Marine Artigau, Yvon Coulais, Claude Picard, Béatrice Mestre-Voegtlé and Eric
Benoist
Dalton Trans., 2011, DOI: 10.1039/C0DT01397H
Synthesis, cytotoxicity and cellular uptake studies of N3 functionalized Re(CO)3 thymidine
complexes
Mark D. Bartholomä, Anthony R. Vortherms, Shawn Hillier, John Joyal, John Babich, Robert P.
Doyle and Jon Zubieta
Dalton Trans., 2011, DOI: 10.1039/C0DT01452D
Visit the Dalton Transactions website for more cutting-edge inorganic, organometallic and
bioinorganic research
www.rsc.org/dalton

Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 6278
www.rsc.org/dalton
PAPER
Investigating the binding properties of porous drug delivery systems using
nuclear sensors (radiotracers) and positron annihilation lifetime
spectroscopy—Predicting conditions for optimum performance
Eskender Mume,a,b Daniel E. Lynch,c Akira Uedonod and Suzanne V. Smith*a
Received 1st November 2010, Accepted 25th January 2011
DOI: 10.1039/c0dt01499k
Understanding how the size, charge and number of available pores in porous material influences the
uptake and release properties is important for optimising their design and ultimately their application.
Unfortunately there are no standard methods for screening porous materials in solution and therefore
formulations must be developed for each encapsulated agent. This study investigates the potential of a
library of radiotracers (nuclear sensors) for assessing the binding properties of hollow silica shell
materials. Uptake and release of Cu2+ and Co2+ and their respective complexes with polyazacarboxylate
macrocycles (dota and teta) and a series of hexa aza cages (diamsar, sarar and
bis-(p-aminobenzyl)-diamsar) from the hollow silica shells was monitored using their radioisotopic
analogues. Coordination chemistry of the metal (M) species, subtle alterations in the molecular
architecture of ligands (Ligand) and their resultant complexes (M-Ligand) were found to significantly
influence their uptake over pH 3 to 9 at room temperature. Positively charged species were selectively
and rapidly (within 10 min) absorbed at pH 7 to 9. Negatively charged species were preferentially
absorbed at low pH (3 to 5). Rates of release varied for each nuclear sensor, and time to establish
equilibrium varied from minutes to days. The subtle changes in design of the nuclear sensors proved to
be a valuable tool for determining the binding properties of porous materials. The data support the
development of a library of nuclear sensors for screening porous materials for use in optimising the
design of porous materials and the potential of nuclear sensors for high through-put screening of
materials.
is functionalised, mesoporous silica can be readily internalised by
plant and animal cells and is reported to have shown no significant
cytotoxicity.7
Introduction
Significant progress in the fabrication of supramolecular objects
with novel phase morphologies, architectures and functions has
been achieved in recent times.1–6 Of particular importance are
the hollow spherical aggregates, which have potential for en-
capsulating cargo molecules within the “empty” core domains.7,8
Breakthrough in the synthesis of mesoporous silica materials with
controlled particle size, morphology and porosity has enabled new
applications in medicine, environment and industry.1–6
The silica matrices are particularly attractive because of their
intrinsic chemical stability and biocompatibility. Silica is an
essential component of cells in the human body and amorphous
silica is known to be non-toxic and biodegradable.9 Amorphous
silica is ultimately excreted via the urine. Further, when the surface
A particularly significant advancement in the field has been
the ability to engineer controlled porosity in these materials for
applications such as drug delivery. The silica particle acts as a
shield for its cargo molecule (i.e. drugs or peptides or proteins)
and is expected to transport its cargo to the desired site without
any significant loss prior to localisation. The release of the cargo
molecules can then be triggered by change in pH at the target site
or degradation of the particle in vivo. Control of such processes
is only possible if the silica materials can be produced with well-
defined surface areas and tuneable pores sizes and volumes.
The release mechanism of biodegradable polymer-based drug
delivery systems relies on hydrolysis of the carrier, often triggered
by suspension in water. For these materials the cargo must be
loaded using organic solvents. Organic solvents are not always
ideal for biological product as they can trigger denaturation or
aggregation of the proteins.
aCentre for Antimatter-Matter Studies (CAMS) at the Australian Nuclear
Science and Technology Organisation (ANSTO), Menai, NSW, 2234,
Australia. E-mail: suzanne.smith@ansto.gov.au
bCentre for Antimatter-Matter Studies (CAMS) at the Australian National
University, Canberra, ACT, 0200, Australia
For many mesoporous silica nanoparticles, absorption to the
surface of the pores is reported to be controlled by size or
morphology of the pores.10 However it is difficult to predict under
cExilica Ltd, The Technocentre, Puma way, Coventry, CV1 2TT, UK
dInstitute of Applied Physics, University of Tsukuba, Tsukuba, Japan
6278 | Dalton Trans., 2011, 40, 6278–6288
This journal is
The Royal Society of Chemistry 2011
©

what conditions the cargo molecules can be loaded and released.
Therefore formulation for uptake and release must be determined
for each cargo molecule and application. This is largely because it
is difficult to determine the actual charge within the pores of the
material. Some researchers have reported that the pore size does
not influence the release of the cargo.11 Other researchers have
reported that the morphology of mesoporous silica nanomaterials
controls the release behaviour.12 This tends to suggest that the
pores are of one type and therefore the affinity between the pore
surface and the cargo molecule governs the release rate and not
the size of the pores.
Understanding how the synthetic process changes the overall
size and porosity of the particle is crucial for optimising the
design of such systems. Numerous tools are used to provide
information on the porosity in these types of materials, they
include Small Angle Neutron Scattering and X-ray spectroscopy
(SANS, SAXS), nitrogen gas adsorption-desorption, Brunauer
Emmett Teller (BET), Scanning Electron Microscopy (SEM),
Transmission Electron Microscopy (TEM) and Positron Anni-
hilation Lifetime Spectroscopy (PALS).13 Collectively they can
provide information about the surface area and potentially the size
and concentration of pores in the material. Positron annihilation
technique is the powerful characterization tool for studying the
size and concentration of free volume and defects in materials.
The material is exposed to a 22Na source and the lifetime of the
positron in the material is used to determine the micropore size.
The positron lifetime (t) is the reciprocal of the integral of the
positron r+(r) and the electron r-(r) densities at the site where the
annihilation takes place.
longer positronium lifetime and is generally attributed to ortho-
positronium formation.
However, these tools provide little information about the
chemistry within the nanopores of the silica shells for loading the
desired cargo molecules. Hence the pharmaceutical industry has
to investigate a wide range of conditions (through trial and error)
in order to determine the best condition for loading and release
of molecules. The study of absorption of cargo agents (metals
and/or drugs) to solid materials is generally performed using
techniques such as atomic absorption spectroscopy, X-ray fluo-
rescence spectroscopy, inductive coupled plasma-atomic emission
spectrometry (ICP-AES), chelate titration, polarographic analysis,
X-ray photoelectron spectroscopy, ATR/IR spectroscopy, elec-
tron spin resonance (ESR) spectroscopy and UV/Vis spec-
troscopy. Such techniques require either large test samples or
solutions, or have comparatively complex testing procedures to
determine the amount of cargo molecule absorbed. These methods
can be extremely costly in time and use of expensive cargo
molecules (i.e. drugs, peptides and proteins).
An alternative approach is the use of radiotracers, either the ra-
dioisotope of the metal ion of interest or a radioisotopic analogue
to quantitatively track the movement of common cargo molecules
within and from the materials. Gamma emitting radioisotopes are
particularly useful as the gamma signal can be easily detected in
various phases (i.e solid and liquid) using a gamma counter and
they are highly sensitive. Test samples can be kept to a minimum
(milligrams and microlitres) and analysis is rapid (secs) without
a significant amount of processing. Furthermore, studies can be
conducted under a range of solvent conditions including high
electrolyte concentrations.14,15,16
ꢀ
t = constant x ( r+(r) r-(r) dr)-1
(1)
This study reports the development and application of a library
of radiotracers (nuclear sensors) of varying size, charge and
hydrophobicity. Radiometal complexes of a series of ligands (see
Fig. 1) were prepared to probe the chemistry and the availability
of the pores within a hollow silica shell material.17 The effect of
pH on the binding and release from the hollow silica particles is
presented. The intent of this work is to develop a library of nuclear
A PAL spectrum of a polymer is usually found to consist of three
different types of mean lifetimes (free positron, para-positronium
and ortho-positronium) which correspond to different annihilation
processes. For example, eqn (1) shows that a larger pore, which
has a lower average electron density, is expected to have a
Fig. 1 Selected ligands radiolabelled for investigating binding properties of hollow silica shells. Ligand abbreviations: 1,4,7,10-tetraazacyclododecane-
1,4,7,10-tetraacetic acid (dota); 1,4,8,11-tetraazacyclo-dodecane-1,4,8,11-tetraacetic acid (teta); 3,6,10,13,16,19-hexaazabicyclo[6.6.6]eicosane-1,8-
diamine (diamsar); 1-N-(4-aminobenzyl)-3,6,10,13,16,19-hexaazabicyclo[6.6.6]eicosane-1,8-diamine (sarar); bis-(1-N-(4-aminobenzyl)-3,6,10,13,16,19-
hexaazabicyclo[6.6.6]eicosane-1,8-diamine (bis-(p-aminobenzyl)-diamsar). Metal complexes of these ligands will be denoted as (M-ligand)n+, e.g.
(Co-diamsar)2+
.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6278–6288 | 6279
©

Table 1 Retention factor (Rf) and mobile phases used to separate the 57/natCo2+ and each 57/natCo-Ligand system
57/natCo2+
57/natCo-Ligand
Rf
Mobile phase
Rf
0.1 M sodium acetate pH 4.5: H2O: MeOH: ammonium hydroxide 20 : 18 : 2 : 1 v/v);
used to separate (57/natCo-dota)2- from free 57/natCo2+
<0.2
>0.8
>0.8
<0.2
0.15 M ammonium acetate pH 6.5; used to separate (57/natCo-diamsar)2+
,
(
57/natCo-sarar)2+ or (57/natCo-bis-(p-aminobenzyl)-diamsar)2+ from free 57/natCo2+
Table 2 Lifetime components for hollow silica shells and Co-doped
hollow silica shells
sensor to determine the chemistry within porous materials in order
to optimize their design and application as delivery vehicles.
Sample t2(ns)
t3(ns)
t4(ns)
R2(nm) R3(nm) R4(nm)
Experimental
No.1
No.2
1.017 0.03 2.57 0.05 17.8 0.2 0.34
0.64 0.01 2.33 0.02 0.18
0.67
0.62
1.68
—
—
Materials and equipments
All reagents and solvents used were of analytical grade and
obtained from commercial sources. Hollow silica shells were
supplied by Exilica Ltd, The Technocentre, Puma Way, Coventry,
CV1 2TT, UK. Water used for experimental purposes was Milli-Q
grade (18 X cm). Buffers and mobile phases in this study were
prepared as required.
the overall time in the furnace was 3 h. After such time the sample
was removed to cool at ambient temperature.
Co-doped hollow silica shells
High specific activity (SA) radioisotopes 57Co2+ (t
=
To 300 mg of hollow silica shells was added 0.1 M potassium
dihydrogen phosphate (PBS) solution (5 mL), vortexed and the
beads left to swell for 5 min. To the suspension was added cobalt(II)
chloride (0.82 g) in PBS. The mixture was then left to agitate for
24 h and then centrifuged (1000 rpm for 1 min) and the supernatant
decanted off. The Co-doped hollow silica shells were washed with
Milli-Q water (3 ¥ 1 mL), dried in air and analysed by PALS.
1/2
271.8 days; SA = 2.59 ¥ 105 GBq g-1) and 64Cu2+ (t1/2 = 12.7 h;
SA = 1.28 ¥ 106 GBq g-1) were obtained from MDS Nordion
and ANSTO Radiopharmaceuticals and Industrials (ARI) respec-
tively. The gamma emissions used to monitor each radioisotope
were 122.1 and 136.5 keV for 57Co2+ and 511 keV for 64Cu2+.
Analytical standard solutions of CoCl2 (0.01 M) and CuCl2
(0.01 M) were supplied by Sigma Aldrich and Riedel de Haen
respectively. CuCl2, CoCl2 and 1,4,7,10-tetraazacyclododecane-
1,4,7,10-tetraacetic acid (dota), 1,4,8,11-tetraazacyclododecane-
1,4,8,11-tetraacetic acid (teta) were purchased from Sigma Aldrich
and Macrocyclics (USA) respectively.
PALS (Positron Annihilation Life-time Spectroscopy) analysis
The lifetime spectra of the positrons were measured by using
a digital oscilloscope (LeCroy Wavepro) and two scintillation
detectors based on Hamamatsu H3378 photomultiplier tubes and
BaF2 scintillators.18 About 4 ¥ 106 counts were accumulated for
each lifetime spectrum. The lifetime spectrum of positrons, SLT (t),
is given by
The equipment used for the binding and loading capacity studies
include Eppendorf 5804R centrifuge, Wallac Wizard 1480 gamma
counter and MyLab SLRM-2 M Intelli Mixer rotomix. Mass
spectroscopy was conducted using AB Applied Biosystems MDS
SCIEX API 2000 LC/MS/MS System mass spectrometer with
an electrospray positive ionization method. Bruker Avance DPX
ꢁ
SLT (t) =
li Ii exp(-li t),
(2)
1
400-1H NMR (400 MHz), H recorded in D2O; chemical shifts
1
where li and Ii are respectively the i-th components of the
annihilation rate of positrons and its intensity.
for H signals for each compound were measured relative to the
internal standard of dioxane. 1H NMR (D2O) d: 3.75 (s).
The lifetime of the positrons, ti, is given by 1/li. The observed
spectra were analysed with a time resolution of about 170 ps (full-
width-at-half-maximum, FWHM) by using the RESOLUTION
computer program.19 The data are presented in Fig. 2, 3 and
Table 2.
Preparation of hollow silica shells
The procedure for fabricating silica shells used is summarised be-
low and detailed elsewhere.17 Poly(1-methylpyrrol-2-ylsquaraine)
(PMPS) were prepared by refluxing equimolar amounts of 1-
methylpyrrole and squaric acid in butan-1-ol for 18 h. The particles
formed were filtered and washed (using a Soxhlet) with hot
ethyl acetate. To the PMPS beads (1 g) was added 30 mL of
9 : 1 TEOS:ethanol and the mixture was sonicated for 10 s and
magnetically stirred for a further 30 min. The resultant product
was vacuum filtered and then suspended in 0.1 M HCl and stirred
for 30 min. The PMPS beads were filtered off and ◦then transf◦erred
to a crucible. The crucible was heated from 150 C ◦to 660 C at
Synthesis of hexa-aza cages
(Cu-diamsar)2+. Diamsar was prepared in a similar manner to
1
that previously described and characterised by H NMR.20 The
CuCl2 (1.28 g; 11.5 mmol) was added to diamsar in ethanol (3.0 g;
9.55 mmol; 100 mL). The blue solution formed was evaporated to
dryness under reduced pressure and the residue (Cu-diamsar)Cl2
(4.8 g) used without further purification for the synthesis of (Cu-(p-
nitrobenzyl)-diamsar)2+ and (Cu-bis-(p-nitrobenzyl)-diamsar)2+.
◦
10 C min-1. The temperature was then held at 660 C such that
6280 | Dalton Trans., 2011, 40, 6278–6288
This journal is
The Royal Society of Chemistry 2011
©

with 6 M HCl. Each fraction was evaporated to dryness under
vacuum and the resultant residue dissolved in water (2 ¥ 40 mL)
and rotary evaporated to dryness. The addition of absolute ethanol
(20 mL) to the residue and sonication for 3 min resulted in
the precipitation of blue crystals. The crystalline product was
filtered and washed with 10 mL of cold ethanol and dried under
vacuum. Yield for (Cu-(p-nitrobenzyl)-diamsar)2+ and (Cu-bis-
(p-nitrobenzyl)-diamsar)2+ were 0.81 g (29%) and 0.38 g (11%)
respectively. An electrospray mass spectrometry (ESI/MS) in an
aqueous solution (10-6 M) at a cone voltage of 25 V was conducted.
(Cu-(p-nitrobenzyl)-diamsar)2+; ESI/MS: displayed a major signal
at m/z = 547.2 [M+ HCl]+ and m/z = 256.2 [M+ 2H]2+ with a minor
signal of m/z 511.4 [M + H]+. (Cu-bis-(p-nitrobenzyl)-diamsar)2+;
ESI/MS: displayed a major signal at m/z = 682.4 [M+ HCl]+ and
m/z = 324.0 [M+ 2H]2+ with a minor signal of m/z 646.5 [M +
H]+.
Fig. 2 Positron Annihilation Life-time Spectroscopy (PALS) spectrum of
hollow silica shells.
(bis-(p-aminobenzyl)-diamsar)2+
(Cu-(p-nitrobenzyl)-diamsar)2+ and
(Cu-bis-(p-nitrobenzyl)-diamsar)2+
Sodium borohydride (600 mg; 1.59 mmol) in water (0.5 mL) was
slowly loaded on to the catalyst [10% palladium on activated
charcoal] (110 mg) under nitrogen. The (Cu-(bis-p-nitrobenzyl)-
diamsar)2+ (384 mg; 0.06 mmol) dissolved in NaOH solution
(0.5 mL; 0.04% NaOH) was then added slowly to the suspension
under nitrogen. The reaction was allowed to proceed until the
colour of the solution turned from blue to clear (30 min at 23 ◦C).
The Pd/C catalyst was filtered off (0.22 mm) and hydrochloric acid
(1 M) was added dropwise until gas evolution ceased (~0.5 mL,
1 M HCl). The clear solution was lyophilised to produce a white
powder. 1H NMR: 3.00 (s, 12H, NCH2CH2N); 3.31 (s, 12H,
NCCH2N); 3.96 (s, 4H, ArCH2); 6.94 (d, 4H, Ar–H); 7.34 (d,
4H, Ar–H); ESI/MS, m/z = 525.6 [M + H]+.
The complexes were prepared using a modified literature
procedure.21 The (Cu-diamsar)Cl2 (2.5 g; 5.58 mmol) was dissolved
in ethanol water solution (w/w 1 : 1; 50 mL). The pH of the
solution was adjusted to 4.5 by the addition of 4% NaOH solution.
Then the 4-nitrobenzaldehyde (1.18 g; 7.81 mmol) and sodium
cyanoborohydride (0.42 g; 6.68 mmol) were added and the reaction
◦
mixture was left to stir under nitrogen gas for 5 h at 23 C. The
ethanol was then evaporated off and the blue solution diluted
in water (2 L) and loaded onto a Sephadex SP C-25 column
(55 ¥ 7.5 cm swelled in water). The column was washed with
various sodium citrate solutions (1 L of 0.1 M followed by 4 L
of 0.3 M). Three fractions representing three blue bands on the
column were collected. They represent (Cu-diamsar)2+, (Cu-(p-
nitrobenzyl)-diamsar)2+ and (Cu-bis-(p-nitrobenzyl)-diamsar)2+ in
order of elution. Each fraction was diluted in water (3 fold) and
then absorbed on to three different Dowex 50 W ¥ 2 columns
(10 ¥ 1.2 cm swelled in water) and washed with water at a flow
rate of 14 mL min-1. The respective columns were then washed
with 1 M HCl (1 L) and respective copper complexes were eluted
Sarar
Sarar was synthesized using a similar procedure as previously
described.21
Sodium borohydride (380 mg; 10.0 mmol) in water
(0.6 mL) was slowly loaded on to the catalyst [10% pal-
ladium on activated charcoal] (65 mg) under nitrogen. The
Fig. 3 The lifetime and relative intensities of micropores in undoped and Co-doped hollow silica shells.
The Royal Society of Chemistry 2011 Dalton Trans., 2011, 40, 6278–6288 | 6281
This journal is
©

(Cu-(p-nitrobenzyl)-diamsar)2+ (500 mg; 0.76 mmol) dissolved in
NaOH solution (0.4 mL; 0.04% NaOH) was then added slowly
to the suspension under nitrogen. The reaction was allowed to
proceed until the colour of the solution turned from blue to clear
(20 min at 23 ◦C). The Pd/C catalyst was filtered off (0.22 mm) and
hydrochloric acid (1 M) was added dropwise until gas evolution
ceased (~0.7 mL, 1 M HCl). The clear solution was lyophilised to
Where C = M ion or M-ligand bound to hollow silica shells
D = M ion or M-ligand in solution
m = amount of hollow silica shells (mg)
V = total volume of the solution (mL)
Release of 64/natCu2+, 57/natCo2+, and 57/nat Co-Ligand from hollow
silica shells
1
produce a white powder. H NMR: 2.92 (s, 12H, NCH2CH2N);
The hollow silica shells were loaded with each nuclear sensor [i.e.
64/natCu2+, 57/natCo2+, (57/natCo-dota)2-, (57/natCo-diamsar)2+, (57/natCo-
sarar)2+ or (57/natCo-bis-(p-aminobenzyl)-diamsar)2+ in a similar
manner to that described above and once equilibrium was reached,
centrifuged (7.5 rpm for 2 min) and the supernatant removed
carefully. The loaded hollow silica shells were then exposed to a
fresh solution of buffer (1.5 mL) and vortexed for 5 s. The mixtures
were left to agitate at 23 ◦C up to 24 h. Then solutions were sampled
for the release of nuclear sensor over varying time intervals (10,
30, 60, 90, 120 min and 24 h) in a similar to that described above.
The radioactivity associated with solutions and hollow silica shells
were measured in a gamma counter. As described above the moles
of M ion and M-Ligand associated with the hollow silica shells
were calculated at each time point.
3.00 (s, 6H, NCCH2N); 3.10 (s, 6H, NCCH2NCCH2); 3.58 (s, 2H,
ArCH2); 6.81 (d, 2H, Ar–H); 7.13 (d, 2H, Ar–H); ESI/MS, m/z =
420.4 [M + H]+.
Complexation of 57/natCo2+ by dota, diamsar, sarar and
bis-(p-aminobenzyl)-diamsar - (57/natCo-Ligand)
A typical procedure involved exposing equimolar amounts
(1.5 ¥ 10-2 M) of the Ligand (dota, diamsar, sarar or bis-(p-
aminobenzyl)-diamsar) to a Co2+ solution doped with 57Co2+ in
sodium phosphate buffer (PBS) pH 7. The mixture was then
vortexed for 5 s and incubated at 23 ◦C for 1 h. The total volume
of the final solutions was 480 mL. The complexation reaction was
monitored at 23 ◦C by instant thin layer chromatography (ITLC-
SG) until > 95% of the 57/natCo-Ligand was formed. The mobile
phase and retention factor (Rf ) for 57/natCo2+ and 57/natCo-Ligand
are summarised in Table 1. The percentage of 57/natCo2+ complexed
by the Ligand was calculated for each reaction.
Results and discussion
Preparation of hollow silica shells and measurement of porosity
Uptake of 64/natCu2+, 57/natCo2+, (57/natCo-dota)2-,
The hollow silica shells were prepared by the over-coating of
sacrificial polymeric template particles with a silicon precursor
followed by thermal calcination in a furnace at 660 ◦C. The
hollow silica shells were specifically designed to act as molecular
containers. They have a thin permeable shell to promote increased
flow of target molecules and solution into and from the shells. The
surface area (SSA) of the hollow silica shells was determined using
the BET isotherm to be > 370 m2 g-1.22
In the present study, positron annihilation lifetime spectroscopy
(PALS) was used to determine the microporosity in the hollow
silica shells. The probability or intensity of ortho-positronium
formation is proportional to the number of free pores in hollow
silica shells. A comparison of spectra for undoped and Co-
doped hollow silica shells is illustrated in Fig. 2. Visual analysis
of the spectra shows there is a sharp decrease in the intensity
of the Co-doped silica shell spectrum compared to that of the
metal free sample. This clearly reflects a change in the sample.
Fig. 3 and Table 2 summarises the analysis of these spectra and
the determined life-time and intensity of the free positrons and
positronium formation (pick-off annihilations) on exposure to the
22Na source.
Analysis of the results using eqn (1) show the spectrum (No.1)
for the hollow silica shells decomposed into four components [ti
where i = 1 to 4] (see Fig. 3a). The pore diameters within the
hollow silica shells were estimated from the lifetimes of each ortho-
positronium component using the Tao–Eldrup model.23 From
Fig. 3(a) we can see that the hollow silica shells have four lifetime
components (t1 to t4) while Co-doped silica shells have only three
lifetime components (t1 to t3). The second and third lifetimes
present in both spectra appear to be due to pick-off annihilation
of Ps and relate to two pore sizes of 0.34 nm and 0.67 nm radii.
The longest lifetime, t4 found present only in the hollow silica
(
(
57/natCo-diamsar)2+, (57/natCo-sarar)2+ and
57/natCo-bis-(p-aminobenzyl)-diamsar)2+ by hollow silica shells
A typical procedure involved mixing 10 mg (accurately weighed in
triplicate) of ho◦llow silica shells in 1480 mL of appropriate buffer
(pH 3–9) at 23 C. Buffer solution ranged from 0.1 M glycine in
0.1 M sodium chloride for pH 3.0; 0.1 M sodium acetate for pH 4.0
and 5.0; 0.1 M potassium dihydrogen phosphate for pH 6.0, 7.0
and 8.0; 0.1 M glycine in 0.1 M sodium chloride, pH adjusted
with sodium hydroxide to pH 9.0. The resulting mixture was then
vortexed for 5 s and left to agitate for 5 min. The hollow silica
shells were then exposed to either free 57/natCo2+ or 64/natCu2+ (1 ¥
10-3M) or 57/natCo-Ligand (1 ¥ 10-4M) in 20 mL. Uptake of the
metal ions and 57/natCo-Ligand were monitored until equilibrium
was reached. At set time intervals (0, 5, 10, 30, 60, 90, 120 min and
24 h) the reaction mixture was centrifuged (7.5 rpm for 2 min). The
supernatant was sampled (20 mL in triplicate) and the radioactivity
associated with solutions and hollow silica shells were measured
using the gamma counter. The moles of metal (M) ions and metal–
ligand complex (M-Ligand) associated with the hollow silica shells
were calculated for each time point using eqn (3). In addition, the
distribution coefficients (Kd) at the various time intervals were
calculated as outlined in eqn (4).
B×% activityassociated withhollowsilicashells
A =
(3)
weight(mg)of hollowsilicashells
Where A = moles of M ion or M-Ligand bound to milligrams of
hollow silica shells and B = initial moles of M ion or M-Ligand
(%C / m)
Kd =
(4)
(%D /V )
6282 | Dalton Trans., 2011, 40, 6278–6288
This journal is
The Royal Society of Chemistry 2011
©

shell relates to the presence of a larger pore of 1.68 nm radius.
Fig. 3(b) show the relative intensity of each t value. It show the
disappearance of the t4 lifetime (intensity 0%) in the Co-doped
samples. This suggests that the major trapping of the Co(II) ion
occurs in the larger pore of 1.68 nm.
Preparation of hexa aza cage derivatives
The hexa aza cage systems were synthesised via a metal template
synthesis of the Co(II)-diamsar species as previously described.20
The cobalt ion was removed using sodium cyanide followed
by extraction into acetonitrile. The recrystallised diamsar is
then complexed with a slight excess of copper chloride and
reductive amination of (Cu-diamsar)2+ with p-nitrobenzaldehyde
afforded both the copper complex of the bis and mono alkylated
species of diamsar (Cu-(p-nitrobenzyl)-diamsar)2+ and (Cu-bis-(p-
nitrobenzyl)-diamsar)2+ respectively, as the hydrochloride salt. The
final products were purified on a Sephadex SP C-25 and Dowex
50 W ¥ 2 columns. The mass spectra analysis gave for (Cu-(p-
nitrobenzyl)-diamsar)2+ a major signal at m/z = 547.2 [M+ HCl]+
and m/z = 256.2 [M+ 2H]2+ with a minor signal of m/z 511.4
[M + H]+. and (Cu-bis-(p-nitrobenzyl)-diamsar)2+ a major signal
at m/z = 682.4 [M+ HCl]+ and m/z = 324.0 [M+ 2H]2+ with a
minor signal of m/z 646.5 [M + H]+, indicating pure products (in
supplementary material). The de-metallation from the cage was
performed by using the reducing agent sodium borohydride with
palladium on carbon in basic media. This afforded sarar and bis-
(p-aminobenzyl)-diamsar (Fig. 1) as a white powder. All other
Ligands were obtained from commercial sources.
2+
Fig. 4 Percent of 57/natCo
complexed by dota, teta, diamsar, sarar
and bis-(p-aminobenzyl)-diamsar after 1 h. [M2+] = 10-3 mol L-1; Buffer:
0.1 mol L-1 PBS pH 7; T = 23 ◦C; Time = 1 h.
complexed by selected ligands at millimolar concentrations after
1 h at pH 7.
Metal (M) binding properties of hollow silica shells
An accurately weighed amount of hollow silica shells was incu-
bated in buffer solutions (i.e. pH 3–9) at room temperature for at
least 5 min. The suspensions were then exposed to each nuclear
sensor and the resultant reaction mixture was then sampled at
varying time intervals (5 min to 24 h) after centrifugation. The
amount of nuclear sensor bound to the hollow silica shells was
determined by counting in a gamma counter. Fig. 5 to 8 illustrate
selected data.
Preparation of nuclear sensors
A range of nuclear sensors (radiotracers) was prepared for
investigating the chemical properties of porous materials. All
ligands (see Fig. 1) were chosen based on their ability to form
thermodynamically stable complexes with Cu2+ or Co2+ metal
ion.24,25
Stability constants (log K) reported for dota complexes of Hg2+,
Cu2+, Ni2+ and Co2+ are 26.4, 22.2, 20.0 and 20.2 at 25 ◦C,
[0.1 M (CH3)4N(NO3)] respectively and 25.7, 21.6, 19.9 and 16.6
respectively for teta.26–28 The Hg2+ complex with sar, the parent
hexa aza cage ◦of sarar, log K is 28.1 ([OH-] = 0.1 M, I = 0.5 M
NaClO4, at 25 C), which is significantly larger than the Hg2+ dota
or teta complexes.29 The [Cu(sar)]2+ complex is expected to be
even more stable but the very slow dissociation of the Cu2+ cation
precludes the determination of its log K value. The exchange of
radiolabelled Cu2+ ion with the copper complex (< 6% in 18 h at
~ 25 ◦C) is consistent with this analysis.
Fig. 5 The effect of pH on metal binding properties of hollow silica shells
at ambient temperature. [M2+] = 10-3 mol L-1; silica shell = 10 mg; V =
1.5 mL; Time = 10 min; T = 23 ◦C; number of samples (n) = 3.
Each ligand was radiolabelled with either nat/57Co2+ or nat/64Cu2+
respectively.
The radiolabelling of each ligand was performed by mixing
the metal ion (M) with the Ligand (L) at a molar ratio of 1 : 1
at ~10-2 M and room temperature. ITLC was used to monitor
the complexation of the M ion. Once the complexation reaction
was complete the final products were diluted in the various
buffer solutions and monitored for breakdown over time. Those
complexes that demonstrated stability over a 24 h period were
used for the binding studies. Fig. 4 shows the percentage of M ion
Fig. 5 illustrates the binding profile for Cu2+ and Co2+ over a
range of pH. The data clearly indicate the absorption of these metal
ions is dependent on pH, with maximum absorption occurring
over pH 6–8. While the absorption curves for Cu2+ and Co2+ are
similar, the total amount of M ion bound and the rate at which
equilibrium is reached were significantly different for the two metal
ions. The total amount of Cu2+ bound was 1.5 fold higher than Co2+
in pH 8 buffer. The rate at which the Cu2+ ion was absorbed was
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6278–6288 | 6283
©

Fig. 6 Metal ion bound to hollow silica shells verses time at pH 8. [M2+] = 10-3 mol L-1; silica shell = 10 mg; V = 1.5 mL; T = 23 ◦C; number of samples
(n) = 3.
considerably faster than Co2+ ion, reaching maximum absorption
within 10 min compared to 90 min for the latter species (see Fig. 6).
Low uptake of both metal ions at pH 3 to 5 indicates the
pores of the hollow silica shells are most likely negatively charged,
attracting the H+ ion more readily. In contrast, the poor uptake
at pH 9 is mostly likely due to either the Cu2+ and Co2+ ion
preferentially binding to buffer (i.e. glycine) or the formation of
soluble hydroxide species. ITLC was used to monitor the formation
of metal hydroxide species (Rf 0) in the absence of hollow silica
shells and was found to be absent. Only at a Co2+ concentration
of 0.75 x10-2 M in glycine buffer (pH 9) is there any evidence of
precipitation, hence it is more likely for the present studies the Co2+
is forming complexes with glycine buffer. Thus the Co2+ complex
is not as readily available for the uptake by hollow silica shells.
Overall the metal ion binding properties of the hollow silica shells
at pH 6–8 correlates well with the desolvation energies for inner
sphere H2O from the respective metal ion (i.e. Cu2+ = ~ 5 ¥ 108 s-1
and Co2+ = ~4 ¥ 105 s-1).30 The affinity and total amount of metal
ion bound correlates with the Irving William series Cu2+ > Co2+.
affinity of the (Co-teta)2- was also tested at pH 3 to 9 and found to
behave in a similar manner to the (Co-dota)2- over the same pH.
Of particular significance is the dramatic enhancement in
the binding of the (Co-bis-(p-aminobenzyl)-diamsar)2+ species
at pH 6, 7 and 8 compared to both (Co-diamsar)2+ and (Co-
sarar)2+. The binding affinity for all three of the positively charged
M-Ligand species is equivalent at pH 9. In view of this, the
binding of the positively charged M-Ligand was fast and in the
following sequence Co-(bis-(p-aminobenzyl)-diamsar)2+ > (Co-
diamsar)2+ > (Co-sarar)2+ at pH 9 (see Fig. 8). Comparison of the
rate of uptake for these hexa aza cage complexes by hollow silica
shells over time show that with the more hydrophobic species,
[(Co-sarar)2+ and (Co-bis-(p-aminobenzyl)-diamsar)2+], binding
was complete in only 10 min, while equilibrium for (Co(diamsar)2+
took 90 min. Data also show that the total amount absorbed for
each nuclear sensor is equivalent at 1440 min.
Collectively it could be speculated that the pores within the
hollow silica shells are most likely negatively charged and that
affinity for the positively charged molecules will be preferred.
In addition the more hydrophobic molecules will be absorbed
preferentially at the high pH (neutral through to basic). This is
consistent with the behaviour observed for the higher absorption
of negatively charged species to the hollow silica shells at pH 3
to 4. At the lower pH the pores within the hollow silica shells are
most likely protonated and the positively charged nuclear sensors
can not compete for the binding sites as effectively as H+ ion.
The PALS study described earlier support the hypothesis that the
nuclear sensors are most likely to be absorbed into the larger pore
(radius = 1.68 nm) of the hollow silica shell.
Metal complex (M-Ligand) binding properties of hollow silica
shells
The uptake of each Co-Ligand species by the hollow silica shells
was also monitored over time (up to 24 h) or until equilibrium
was reached at each pH. The moles of nuclear sensor associated
with the hollow silica shells (10 mg) were calculated at each time
point. In general, the M-Ligand absorption by the hollow silica
shells was selective and rapid at room temperature. The reactions
were often complete within 10 min with mild shaking or 1 min
with vortexing. Fig. 7 and 8 illustrate selected results.
Fig. 7 compares the binding of a series of nuclear sensors to the
hollow silica shells over a range of pH. The data show how the
effect of charge and hydrophobicity of M-Ligand can significantly
influence its binding to the hollow silica shells. The positively
charged nuclear sensors; (Co-diamsar)2+, (Co-sarar)2+ and (Co-
bis-(p-aminobenzyl)-diamsar)2+ show increased binding affinity as
the pH is increased (from pH 5–9). Notably, negatively charged
species (Co-dota)2- were preferentially weakly absorbed at pH 3
and 4 compared to positively charged nuclear sensors. The binding
Release of metal (M) and metal complexes (M-Ligand) from
hollow silica shells
The binding studies were conducted over a range of pH in order
to identify the conditions under which the nuclear sensor is stably
bound and most likely to be released. Together with the PALS data
they can assist to gain an understanding of whether the size or
charge of the pores govern the binding properties, and ultimately
how this information can be used to guide reengineering of these
types of materials.
6284 | Dalton Trans., 2011, 40, 6278–6288
This journal is
The Royal Society of Chemistry 2011
©

Fig. 7 Effect of pH on the binding of metal complexes (Co-Ligand) to hollow silica shells. [Co-Ligand] = 10-4 mol L-1; silica shell = 10 mg; V = 1.5 mL;
Time = 10 min; T = 23 ◦C; number of samples (n) = 3.
Fig. 8 Moles of (Co-Ligand)2+ absorbed into hollow silica shells at pH 9 over time. (Co-Ligand)2+ = 10-4 mol L-1; silica shell = 10 mg; V = 1.5 mL; T =
23 ◦C; number of samples (n) = 3.
Depending on the results obtained in Fig. 7 hollow silica shells
were loaded with the nuclear sensors at their optimum pH. The
mixtures were centrifuged and the hollow silica shells loaded with
the nuclear sensor were separated and re-exposed to fresh buffer
solution of the same pH. Initial binding conditions for the M
species, negatively and positively charged M-Ligand species were
conducted at pH 6 to 8, pH 3 to 5 and pH 6 to 9, respectively. The
mixtures were then vortexed and left to agitate at 23 ◦C for 24 h.
The release of nuclear sensor was monitored over time. Fig. 9 to
11 show percentage of nuclear sensor released after 24 h.
Fig. 9 shows that twice as much Cu2+ ion (30.0 0.4%) is released
at pH 7 compared to Co2+ (16.0 0.1%) after 24 h. For pH 6 and 8
the amount released for both metal ions is similar, (17.0 0.4%). In
contrast the negatively charged species (Co-dota)2- and (Co-teta)2-
are released more rapidly, with ~50–60% and ~80% between pH 3–
5, respectively after 24 h (see Fig. 10).
The positively charged hexaza cage complexes absorbed into
hollow silica shells were exposed to solutions of pH 6 to 9 (see
Fig. 9 Percentage of M released from hollow silica shells after re-exposure
to fresh buffer solution. [M] = 10-6 mol L-1; silica shell = 10 mg; V = 1.5 mL;
T = 23 ◦C; number of samples (n) = 3.
Fig. 11). Up to 20% [19.0 0.2% to 20.0 2.5%] (Co-diamsar)2+
was released from hollow silica shells within minutes, with no
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6278–6288 | 6285
©

◦
and incubated with gentle agitation at 23 C for 24 h in various
buffer solutions. Selected data are given in Fig. 12 and 13. Kd
values greater than 100 have been classified as moderate binders
while those in the region of 1000 or more are considered to be
high binding. Values in the region of 50 to 10 are weak binders
and therefore would be released rapidly when exposed to the same
buffer conditions. The data presented in this study (Fig. 13) show
that the Kd value for (Co-dota)2- and (Co-teta)2- were lower than
that for the positively charged M-Ligand. This may be attributed to
the higher affinity of (M-Ligand)2+ for exchange sites on the hollow
silica shells. More importantly the high Kd indicates the optimum
solvent conditions under which an agent should be bound and
the large volumes of solvent that would be required to release the
agent from the hollow silica shells.
Fig. 10 Percent of (Co-Ligand)2- released from hollow silica shells after
re-exposure to fresh buffer solution. [Co-Ligand] = 10-6 mol L-1; silica
shell = 10 mg; V = 1.5 mL; T = 23 ◦C; number of samples (n) = 3.
Conclusion
further loss over the study period of 24 h. For (Co-sarar)2+ loss is
highly dependent on pH with 50.0 0.7%, 30.0 0.7%, 10.0 0.1%
and 0% released after 24 h at pH 6, 7, 8 and 9, respectively. For
the more hydrophobic nuclear sensor, (Co-bis-(p-aminobenzyl)-
diamsar)2+ the behaviour is quite different. There is 30.0 0.7%
and 35.0 0.3% released at pH 7 and 9, respectively, and in
contrast, less than 26.0 0.2% is released at pH 6 and 8. As
demonstrated with the Co-doped silica shells using PALS, the
nuclear sensors are most likely absorbed into the larger pore
(1.68 nm radius).
The work presented here represents the first report of the
application of radiometal complexes for the characterization of
porous materials. The work clearly shows the absorption to the
hollow silica shells is governed by size, charge and hydrophobicity
of its micropores. The micropores in the hollow silica shells are
sensitive to subtle changes in the molecular architecture of the
ligands and their resultant complexes demonstrated by distinct
changes in rate of absorption and release. The binding of the
nuclear sensors is most likely due to hydrogen, electrostatic or
ionic bonding. Distribution coefficients proved to be valuable
and sensitive indicators of release rates, and potentially could
be useful for defining formulation and application of the porous
material.
Distribution coefficients for nuclear sensors with hollow silica shells
Distribution coefficients (Kd) are often calculated in an effort
to obtain an indication of strength of binding between a solute
and resin in the presence of a specified solution. Distribution
coefficients are generally used for metal purification where it is
important to determine conditions of highest and weakest binding
to the exchange resins. This information is then used to determine
the optimum conditions for separation.
For the current study the Kd was determined by exposing a
known quantity (10 mg) of hollow silica shells in a fixed volume
(1.5 mL) and known concentration of nuclear sensors (1 ¥ 10-4 M)
The use of nuclear sensors was found to be a very sensitive and
fast method for assessing the performance of porous materials.
An advantage of this approach is the minimal amount of
product required for analysis. The process is highly adaptable
for high-throughput screening of the next generation of porous
or controlled release materials. It has the potential to predict the
type of molecules and the conditions under which they might
be encapsulated by porous material. Using this strategy for the
screening of porous materials may provide an opportunity for real
cost savings in development of product formulations.
Fig. 11 Percent of (Co-Ligand)2+ released from hollow silica shells after re-exposure to fresh buffer solution. [Co-Ligand] = 10-6 mol L-1; silica shell =
10 mg; V = 1.5 mL; T = 23 ◦C; number of samples (n) = 3.
6286 | Dalton Trans., 2011, 40, 6278–6288
This journal is
The Royal Society of Chemistry 2011
©

Fig. 12 Distribution coefficient of M after re-exposure to fresh buffer solution. [M] = 10-3 mol L-1; silica shell = 10 mg; V = 1.5 mL; T = 23 ◦C; number
of samples (n) = 9.
Fig. 13 Distribution coefficient of M-Ligand after re-exposure to fresh buffer solution. [M-Ligand] = 10-4 mol L-1; silica shell = 10 mg; V = 1.5 mL; T =
23 ◦C; number of samples (n) = 9.
9 Chemwatch 10451 NC317TCP Version No: 3 CD 2010/4 Page 1 of 15.
Acknowledgements
10 M. Vallet-Regi, A. Ra´mila, R. P. del Real and J. Pe´rez-Pariente, Chem.
Mater., 2001, 13, 308.
We thank Australian Research Council funding for the ARC
Centre of Excellence for Antimatter-Matter Studies.
11 I. I. Slowing, B. G. Trewyn, S. Giri and V. S. Y. Lin, Adv. Funct. Mater.,
2007, 17, 1225.
12 B. G. Trewyn, C. M. Whitman and V. S. Y. Lin, Nano Lett., 2004, 4,
2139.
13 S. V. Smith, J. Phys.: Conf. Ser., 2009, 185, 012044.
14 R. Naik, G. Wen, M. S. Dharmaprakash, S. Hureau, A. Uedono, X.
References
Wang, X. Liu, P. G. Cookson and S. V. Smith, J. Appl. Polym. Sci.,
1 S. Huh, J. W. Wiench, J. C. Yoo, M. Pruski and V. S. Y. Lin, Chem.
Mater., 2003, 15, 4247.
2010, 115, 1642–1650.
15 R. Rajkhowa, R. Naik, L. Wang, S. V. Smith and X. Wang, J. Appl.
Polym. Sci., 2011, 119, 3630–3639.
2 B. G. Trewyn, C. M. Whitman and V. S. Y. Lin, Nano Lett., 2004, 4,
2139.
16 G. Wen, R. Naik, P. G. Cookson, S. V. Smith, X. Liu and X. G. Wang,
Powder Technol., 2010, 197, 235.
17 D. E. Lynch, Y. Nawaz and T. Bostrom, Langmuir, 2005, 21,
6572.
3 K. Suzuki, K. Ikari and H. Imai, J. Am. Chem. Soc., 2004, 126, 462.
4 J. Y. Ying, Chem. Eng. Sci., 2006, 61, 1540.
5 J. Y. Ying, C. P. Mehnert and M. S. Wong, Angew. Chem., Int. Ed.,
1999, 38, 56.
6 C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli and J. S. Beck,
Nature, 1992, 359, 710.
18 H. Saito, Y. Nagashima, T. Kurihara and T. Hyodo, Nucl. Instrum.
Methods Phys. Res., Sect. A, 2002, 487, 612.
19 P. Kirkegaard, M. Eldrup, O. E. Mogensen and N. J. Pedersen, Comput.
Phys. Commun., 1981, 23, 307–335.
20 G. A. Bottomley, I. J. Clark, I. I. Creaser, L. M. Engelhardt, R. J.
Geue, K. S. Hagen, J. M. Harowfield, G. A. Lawrance, P. A. Lay, A. M.
7 D. R. Radu, C. Y. Lai, K. Jeftinija, E. W. Rowe, S. Jeftinija and V. S. Y.
Lin, J. Am. Chem. Soc., 2004, 126, 13 216.
8 I. Slowing, B. G. Trewyn and V. S. Y. Lin, J. Am. Chem. Soc., 2006, 128,
14792.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6278–6288 | 6287
©

Sargeson, A. J. See, B. W. Skelton, A. H. White and F. R. Wilner,
Aust. J. Chem., 1994, 47, 143.
21 N. M. Di Bartolo, A. M. Sargeson, T. M. Donlevy and S. V. Smith,
J. Chem. Soc., Dalton Trans., 2001, 2303.
22 Unpublished data.
26 S. Chaves, R. Delgado, J. J. R. Frausto and Da Silva, Talanta, 1992, 39,
249.
27 E. T. Clarke and A. E. Martell, Inorg. Chim. Acta, 1991, 190, 27.
28 A. E. Martell, R. M. Smith, Critical Stability Constants, vol. 62nd
Supplement, Plenum Press, New York, 1989.
23 D. W. Gidley, W. E. Frieze, T. L. Dull, J. Sun, A. F. Yee, C. V. Nguyen
and D. Y. Yoon, Appl. Phys. Lett., 2000, 76, 1284.
24 S. V. Smith, Q. J. Nucl. Med. & Mol. Imag., 2008, 52(2), 193.
25 S. V. Smith, J. Inorg. Biochem., 2004, 98, 1874and reference within.
29 L. Grøndahl, A. Hammershøi, A. M. Sargeson and V. Thom, Inorg.
Chem., 1997, 36, 5396.
30 M. Pourbaix, in Atlas of Electrochemical Equilibria in Aqueous Solu-
tions, NACE, Texas, 1974.
6288 | Dalton Trans., 2011, 40, 6278–6288
This journal is
The Royal Society of Chemistry 2011
©