Mechanochemical and solution synthesis, and crystal structures and IR and solid-state (CPMAS) NMR spectroscopy of some bis(triphenylphosphine)silver(i) mono- and di-hydrogencitrate systems

Article abstract of DOI:10.1039/c2dt12452a

The complex [(Ph₃P)₂Ag(H₂cit)]·EtOH (1; H₂cit^- = dihydrogencitrate = C₆H ₇O₇^-) contains [(Ph₃P) ₂Ag(H₂cit)] molecules in which the silver atom is coordinated to two PPh₃ molecules and the two oxygen atoms of one of the 'terminal'/1-carboxylate groups of the dihydrogencitrate group. The molecules form centrosymmetric hydrogen-bonded dimers in the solid. In [{(Ph₃P)₂Ag}₂(Hcit)], (2), unsymmetrical deprotonation of the citrate grouping is found, from the 1- and 3- (i.e. terminal and central) carboxylates: [(Ph₃P)₂Ag(O ₂CCH₂C(OH) (CH₂COOH)CO₂)Ag(PPh ₃)₂]. The above complexes, as well as [(Ph ₃P)₃Ag(H₂cit)] (3) were prepared via conventional solution methods, involving the reaction of trisilver(i) citrate, citric acid and triphenylphosphine, and by a mechanochemical method involving the reaction of silver(i) oxide, citric acid and triphenylphosphine. IR studies of 1-3 show the presence of coordinated carboxylate and free carboxylic acid groups in the mono- and di-hydrogencitrate ligands, and the formation of 2 from 1 shows that dihydrogencitrate deprotonation can occur upon dissolution of 1 in protic solvents. High-field (9.40 T) ³¹P CPMAS NMR spectra were recorded and analysed, yielding heteronuclear ¹J( ^107/109Ag,³¹P) and homonuclear ²J( ³¹P,³¹P) spin-spin coupling constants.

Full text of DOI:10.1039/c2dt12452a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 5409
www.rsc.org/dalton
PAPER
Mechanochemical and solution synthesis, and crystal structures and IR and
solid-state (CPMAS) NMR spectroscopy of some bis(triphenylphosphine)
silver(I) mono- and di-hydrogencitrate systems†
Graham A. Bowmaker,*a John V. Hanna,*b Brian W. Skeltonc and Allan H. Whitec
Received 20th December 2011, Accepted 8th March 2012
DOI: 10.1039/c2dt12452a
The complex [(Ph3P)2Ag(H2cit)]·EtOH (1; H2cit− = dihydrogencitrate = C6H7O7−) contains
[(Ph3P)2Ag(H2cit)] molecules in which the silver atom is coordinated to two PPh3 molecules and the two
oxygen atoms of one of the ‘terminal’/1-carboxylate groups of the dihydrogencitrate group. The
molecules form centrosymmetric hydrogen-bonded dimers in the solid. In [{(Ph3P)2Ag}2(Hcit)], (2),
unsymmetrical deprotonation of the citrate grouping is found, from the 1- and 3- (i.e. terminal and
central) carboxylates: [(Ph3P)2Ag(O2CCH2C(OH) (CH2COOH)CO2)Ag(PPh3)2]. The above complexes,
as well as [(Ph3P)3Ag(H2cit)] (3) were prepared via conventional solution methods, involving the reaction
of trisilver(I) citrate, citric acid and triphenylphosphine, and by a mechanochemical method involving the
reaction of silver(I) oxide, citric acid and triphenylphosphine. IR studies of 1–3 show the presence of
coordinated carboxylate and free carboxylic acid groups in the mono- and di-hydrogencitrate ligands, and
the formation of 2 from 1 shows that dihydrogencitrate deprotonation can occur upon dissolution of 1 in
protic solvents. High-field (9.40 T) 31P CPMAS NMR spectra were recorded and analysed, yielding
heteronuclear 1J(107/109Ag,31P) and homonuclear 2J(31P, 31P) spin–spin coupling constants.
tribasic acid citric acid (C6H8O7 = ‘H3cit’) and its various depro-
Introduction
tonated forms, referred to collectively as ‘citrate’,10–16,18 with
The many crystallographic studies of complexes of silver(I) that
have been reported demonstrate its structural versatility in its
coordination chemistry.1,2 Many of these complexes involve
anionic ligands such as halides, pseudohalides and oxyanions.
Some types of oxyanion complex, such as those involving car-
boxylate ligands, have long been known,3,4 and we have pre-
viously characterized a number of examples involving formate,
the simplest possible carboxylate ligand, by crystal structure and
spectroscopic methods.5,6 Silver–carboxylate systems are of con-
siderable interest in a number of different areas of chemistry,
such as the synthesis and mechanism of action of silver metal
catalysts,7–9 the preparation of silver colloids, coatings, nanopar-
ticles and nanowires,9–17 and as antibacterial agents.18,19 Silver(I)
is known to deactivate cellular enzymes and DNA by coordinat-
ing to electron-donating groups such as carboxylates.20 While
many structural studies have been reported on silver(I) mono-car-
boxylate complexes,1,21–26 there is a paucity of data on those of
familiar oligo-/poly-carboxylates. For example, a number of the
interesting applications of silver–carboxylate systems involve the
structures available in ref. 27 and 28. However, there has been
no systematic study of families of silver(I) complexes involving
citrate ligands, complexes of which are sometimes difficult to
crystallize and, while there are references in the patent literature
to ‘silver dihydrogen citrate’ and its antimicrobial action,18 there
have been no reports of a proper characterization of this com-
pound or any complexes thereof. The compound ‘silver citrate
hydrate’ (C6H5O7Ag3 = Ag3cit)·(xH2O) ‘trisilver citrate’ is com-
mercially available (e.g. Aldrich 361259), but its structure is not
known and results of a recent solid-state 109Ag CPMAS NMR
study suggest that the silver coordination in this compound is
quite different from that in silver(I) carboxylates involving
monobasic carboxylic acids.29
In the present work we undertook a study of complexes of
silver(I) mono- and di-hydrogencitrate with triphenylphosphine.
Use of the triphenylphosphine ligand PPh3 allows the isolation
of silver carboxylate units, which might otherwise form poly-
meric structures,28 and provides an NMR probe by way of the
31P nucleus. 31P CPMAS NMR spectra were acquired with the
1
aim of measuring the heteronuclear J(107/109Ag,31P) and homo-
2
nuclear J(31P, 31P) spin–spin coupling constants, which provide
aSchool of Chemical Sciences, University of Auckland, Private Bag
92019, Auckland, New Zealand. E-mail: ga.bowmaker@auckland.ac.nz
bDepartment of Physics, University of Warwick, Coventry, CV4 7AL, UK
cSchool of Chemistry and Biochemistry M313, The University of Western
Australia, Crawley 6009, Western Australia
information about the metal–ligand bonding in the complexes.
Infrared spectroscopy was also applied to provide evidence con-
cerning the nature of the citrate ligand present and its mode of
bonding to the silver atom. X-ray crystal structures were deter-
mined where possible to provide more detailed structural
†CCDC 763388 and 763389. For crystallographic data in CIF or other
electronic format see DOI: 10.1039/c2dt12452a.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 5409–5417 | 5409

View Article Online
information and geometric parameters for correlation with the
NMR data.
was filtered and washed with a little ice-cold EtOH. Yield
0.715
g (83.5%). Found: C 63.3, H 4.9%; calcd for
The ready accessibility of a number of novel species and/or
forms by way of mechanochemical synthesis30–50 has encour-
aged us to explore some triphenylphosphine–silver(I) mono- and
di-hydrogencitrate systems in this manner. We chose to use silver(I)
oxide Ag2O as the source of silver(I) in these reactions. Most of
the mechanochemical syntheses of metal complexes reported to
date use metal salts or other metal complexes as the source of
the metal ion,30–37,46–50 but some recent reports demonstrate the
possibility of using metal oxides, particularly those of groups 2
and 12,42–45 and the present work shows that this approach may
be extended to the Group 11 metal oxide Ag2O.
C78H66Ag2O7P4: C 64.4, H 4.6%.
Method 2. Silver(I) oxide (0.070 g; 0.3 mmol) and citric acid
monohydrate (0.063 g, 0.3 mmol) were ground together in a
mortar and pestle to form a homogeneous mixture and then
water (0.08 g) was added and grinding was continued to produce
an off-white paste. This was allowed to partially dry by evapor-
ation, then triphenylphosphine (0.315 g, 1.2 mmol) was added
with grinding followed by addition of ethanol (0.07 g). Upon
further grinding the mixture changed to a viscous oil, which sud-
denly stiffened to produce an off-white paste. This was ground
further, following addition of further ethanol (0.06 g) and the
mixture was allowed to dry by standing in a fume cupboard; the
latter procedure was repeated following the addition of further
ethanol (0.4 g). The IR spectrum of this product was identical to
that of the product 2 prepared by Method 1 above.
Experimental
Chemicals
[(Ph3P)3Ag(H2cit)], 3. Method 1. A mixture of trisilver(I)
citrate (0.51 g, 1.0 mmol), citric acid (0.4 g, 2.1 mmol) and
PPh3 (2.5 g, 9.5 mmol) and ethanol (20 ml) was heated with stir-
ring. The resulting solution was filtered while hot and the filter
was washed several times with hot EtOH to dissolve solid
product which had separated in the filter. The filtrate deposited a
colourless microcrystalline product that was collected and
washed with ice-cold ethanol. Yield 1.4 g (42%). Found: C 67.2,
H 5.2%; calcd for C60H52AgO7P3: C 66.4, H 4.8%.
Method 2. Silver(I) oxide (0.056 g; 0.23 mmol) and citric acid
monohydrate (0.097 g, 0.46 mmol) were ground together in a
mortar and pestle to form a homogeneous mixture and water
(0.04 g) was then added and grinding was continued to produce
an off-white powder. To this was added triphenylphosphine
(0.362 g, 1.38 mmol) with grinding followed by addition of
ethanol (0.09 g). Upon further grinding the mixture changed to a
viscous oil, which suddenly stiffened to produce an off-white
paste. This was ground following addition of further ethanol
(0.3 g) and the mixture was allowed to dry by standing in a fume
cupboard. The latter process was repeated twice but the IR spec-
trum of the product was not the same as that of the product 3
above. The dry product was placed in a glass tube, ethanol
(0.4 g) was added, and the tube was sealed and placed in a 60 °C
oven for 2 h. The resulting mixture was dried in an open tube at
60 °C to yield a white product. The IR spectrum of the product
was identical to that of the product 3 prepared by Method 1
above.
Silver(I) oxide (BDH), triphenylphosphine (RdH), citric acid
monohydrate (Merk) and sodium dihydrogencitrate (Sigma)
were used as received. Trisilver(I) citrate was prepared by
mechanochemical reaction of silver(I) oxide (0.9 mmol) and
citric acid (0.6 mmol) as described below.
Preparation of compounds
[(Ph3P)2Ag(H2cit)]·EtOH, 1. Method 1. A mixture of trisilver(I)
citrate (Ag3cit ≡ Ag3C6H5O7) (0.51 g, 1.0 mmol), citric acid
monohydrate (0.84 g, 4.0 mmol) and PPh3 (1.6 g, 6.0 mmol)
and ethanol (20 ml) was heated with stirring. The resulting sol-
ution was filtered while hot and allowed to cool slowly in a hot
water bath. The colourless crystalline product was collected and
washed with ice-cold ethanol. Yield 2.2 g (84%). Found: C 61.5,
H 5.0%; calcd for C44H43AgO8P2: C 60.8, H 5.0%.
Method 2. Silver(I) oxide (0.070 g; 0.3 mmol) and citric acid
monohydrate (0.126 g, 0.6 mmol) were ground together in a
mortar and pestle to form a homogeneous mixture and then
water (0.04 g) was added and grinding was continued to produce
an off-white powder. To this was added triphenylphosphine
(0.315 g, 1.2 mmol) with grinding followed by addition of
ethanol (0.06 g). Upon further grinding the mixture changed to a
viscous oil, which suddenly stiffened to produce an off-white
paste. This was ground further, following addition of further
ethanol (0.06 g) and the mixture was allowed to dry by standing
in a fume cupboard. The resulting solid mass was broken up by
grinding and placed in a 50 °C oven for 30 m. The IR spectrum
of this product was identical to that of the product 1 prepared by
Method 1 above. Heating for a longer period (50 °C oven for
2 d) resulted in loss of EtOH, as indicated by disappearance of
the EtOH ν(OH) band at 3497 cm−1) and formation of the desol-
vated complex [(Ph3P)2Ag(H2cit)], 1b. The EtOH solvate 1 was
regenerated by grinding the desolvated product with EtOH
(0.1 g) followed by air-drying under ambient conditions.
Mechanochemical reactions between Ag2O and citric acid.
These were carried out in order to compare the products of such
reactions with those carried out in the presence of triphenylphos-
phine: silver(I) oxide (0.3, 0.6, or 0.9 mmol) and citric acid
monohydrate (0.6 mmol) were ground together dry in a mortar
and pestle and then water (0.1 g) was added and the grinding
was continued to produce an off-white paste. The products were
dried in the air under ambient conditions.
[{(Ph3P)2Ag}2(Hcit)], 2. Method 1. [(Ph3P)2Ag(H2cit)]·EtOH,
1 (0.9695 g) was dissolved in boiling EtOH (12 ml) and the
resulting solution decanted from a small amount of undissolved
white solid and allowed to cool slowly in a warm water bath.
The colourless crystalline product that deposited upon cooling
Structure determinations
Full spheres of CCD area-detector diffractometer data were
measured at ca. 100 K yielding Nt(otal) reflections, these merging
5410 | Dalton Trans., 2012, 41, 5409–5417
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 1 A selection of crystallographically established P2AgO2C discrete silver(I) atom environments
Anion/silver
Ag–P/Å
Ag–O/Å
P–Ag–P/°
O–Ag–O/°
P–Ag–O/°
H2citrate (1)/1
2.3904(5)
2.4417(5)
2.4355(13)
2.4615(12)
129.203(16)
53.73(4)
98.43(4)
−125.82(4)
Hcitrate (2)/1
Hcitrate (2)/2
2.403(3)
2.421(3)
2.409(11)
2.455(11)
131.95(8)
136.78(17)
129.62(3)
124.13(3)
127.72(3)
127.96(3)
142.18(4)
126.15(3)
54.9(4)
51.8(4)
53.33(7)
52.3(1)
52.8(1)
53.5(1)
51.6(1)
51.3(1)
101.3(3)
−115.6(3)
2.413(4)
2.424(5)
2.501(15)
2.528(11)
104.3(4)
−111.8(3)
O2CCH3/1a
(anhydrate)
2.4332(8)
2.4483(8)
2.420(2)
2.438(2)
106.03(7)
−114.83(7)
O2CCH3/2
2.4264(8)
2.4608(8)
2.379(3)
2.510(3)
102.0(1)
−125.61(8)
(12H2O·34EtOH)/1b
(12H2O·34EtOH)/2
O2CCF3/1c
2.4461(8)
2.4490(8)
2.438(2)
2.451(3)
110.75(8)
−115.28(7)
2.4063(8)
2.4582(8)
2.431(2)
2.449(3)
103.10(9)
−122.50(7)
2.423(1)
2.445(1)
2.526(3)
2.542(4)
97.13(9)
−112.41(9)
O2CC(OH)CH3/1d
2.4345(9)
2.4671(9)
2.425(3)
2.508(4)
104.7(1)
−126.5(1)
a Ref. 21(redetermination, cf. ref. 25). b Ref. 23 (redetermination of material recorded as a ‘sesquihydrate’ in ref. 21). c Ref. 22. d Ref. 24.
to N unique after ‘empirical’/multiscan absorption correction
(Rint cited) which were used in the full matrix least squares
refinements on F2, refining anisotropic displacement parameter
forms for the non-hydrogen atoms, hydrogen atom treatment fol-
0.75. 2θmax = 125°; Nt = 25 678, N = 10 450 (Rint = 0.061), No =
5353. R1 = 0.13, wR2 = 0.35 (a = 0.20); S = 1.11.
Variata. Data were weak and limited and measured using a
monochromatic Cu Kα radiation source. In the refinement, one
of the phenyl rings in triphenylphosphine ligand 3 and all in
ligand 4 were modelled as disordered over pairs of sites of equal
occupancy with isotropic displacement parameter forms and con-
strained geometries.
2
lowing a riding model (reflection weights: (σ2(Fo ) + (aP)2)−1 (P
2
2
= (Fo + Fc )/3)); No with I > 2σ(I) were considered ‘observed’.
Neutral atom complex scattering factors were employed within
the SHELXL 97 program.51 Pertinent results are given below
and in Table 1 and the figures, the latter depicting the non-hydro-
gen atoms with 50% probability amplitude displacement envel-
opes, hydrogen atoms, where shown, having arbitrary radii of
0.1 Å.
Spectroscopy
Infrared spectra were recorded at 4 cm−1 resolution on samples
suspended in KBr using a Nicolet 8700 FTIR spectrometer, or
on dry powders using a Perkin Elmer Spectrum 400 FT-IR spec-
trometer equipped with a Universal ATR sampling accessory.
The ATR spectra were converted to conventional absorbance
spectra using OMNIC software.
Crystal/refinement data
[(Ph3P)2Ag(O2CCH2C(OH)(CO2H)CH2COOH)]·C2H5OH,
[(Ph3P)2Ag(H2cit)]·EtOH 1. C44H43AgO8P2, M = 869.6. Tricli-
Solid-state 31P CPMAS NMR data were obtained at ambient
temperature on Bruker DSX-400 (9.40 T) spectrometer operating
at a 31P Larmor frequency of 161.92 MHz. Conventional cross-
polarization52 and magic-angle-spinning53 techniques, coupled
with spin temperature alternation54 to eliminate spectral artefacts,
were implemented using a Bruker 3.2 mm double-air-bearing
probe in which MAS frequencies of ∼10–12 kHz were achieved.
ˉ
nic, space group P1, a = 12.2314(5), b = 12.8542(5), c =
14.4510(4) Å, α = 82.679(2), β = 87.727(3), γ = 64.362(4)°, V =
2031.4(2) Å3. Dc (Z = 2) = 1.422 g cm−3. Monochromatic Mo
Kα radiation, λ = 0.71073 Å. μMo = 0.63 mm−1; specimen: =
0.41 × 0.28 × 0.25 mm; Tmin/max = 0.99. 2θmax = 65°; Nt =
37387, N = 14 800 (Rint = 0.039), No = 9552. R1 = 0.036, wR2 =
0.073 (a = 0.030); S = 0.90.
1
An initial H π/2 pulse width of 3 μs, recycle delay of 15–30 s,
1
[(Ph3P)2Ag(O2CCH2C(OH)(CH2COOH)CO2)Ag(PPh3)2],
[{(Ph3P)2Ag}2(Hcit)] 2. C78H66Ag2O7P4, M = 1454.9. Monocli-
nic, space group P21/c, a = 24.568(2), b = 12.4749(7), c =
23.1076(10) Å, β = 106.704(6)°, V = 6783.2(8) Å3. Dc (Z = 4) =
1.425 g cm−3. Monochromatic Cu Kα radiation, λ = 1.54183 Å.
Hartmann–Hahn contact period of 2–5 ms and a H decoupling
field of 80–85 kHz (during acquisition) were used in these 31P
experiments. The reported 31P chemical shifts were externally
referenced to the IUPAC standard of 85% H3PO4 via a secondary
solid reference of ammonium dihydrogen phosphate ((NH4)
(H2PO4)) which exhibits a shift at δ 1.0 ppm; this secondary
μCu = 6.0 mm−1; specimen: = 0.13 × 0.12 × 0.04 mm; Tmin/max
=
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 5409–5417 | 5411

View Article Online
reference material was also used to set the Hartmann–Hahn con-
dition. The 2D 31P CPCOSY experiment was implemented with
the TPPI (time proportional phase incrementation) method55–57
for acquisition of phase-sensitive data in both the F1 and F2
dimensions. The application of this technique has been discussed
in detail elsewhere.58,59 The recycle delay, contact period, 1H π/2
pulse width and MAS rate were the same as those implemented
in the above 1D 31P CPMAS experiments. A total of 256 F1
increments were acquired into 256 word blocks, with both
dimensions zero-filled to 1 K words and weighted with Gaussian
multiplication prior to Fourier transformation.
synthesis involving more than two reactants, but these syntheses
have been carried out using a single solvent.37,45 The present
study addresses the situation of a three-component reaction A +
B + C → product, where a solvent that is appropriate for assist-
ing the reaction A + B is not appropriate for C. In the present
case the reaction between Ag2O and citric acid is best assisted
by the solvent water, in which the citric acid is readily soluble,
but water is not an appropriate solvent for PPh3. We therefore
carried out the reactions in two steps, using water for the
solvent-assisted reaction between Ag2O and citric acid, and
ethanol for the solvent-assisted reaction of PPh3 with the product
of the first reaction. Using this approach, the number of cycles of
mechanochemical treatment required to form pure product was
reduced to a minimum. An additional feature of the mechano-
chemical synthesis as applied to the present system is that the
reaction to produce 1 produced a pure product using the stoichio-
metric amounts of the three reactants as shown in eqn (1),
whereas the solution reaction required an excess of citric acid to
suppress the dissociation reaction (2). The reason for this differ-
ence is that in the mechanochemical reaction there is insufficient
solvent present to accommodate the citric acid produced in the
dissociation, so the dissociation is suppressed in this case by the
very nature of the solvent-assisted mechanochemical reaction!
Results and discussion
In principle the dihydrogencitrate complex 1 can be prepared by
addition of the stoichiometrically required amount of citric acid
to a 1 : 6 mixture of trisilver(I) citrate, Ag3cit and PPh3:
Ag3cit þ 2H3cit þ 6PPh3 ! 3½ðPh3PÞ2AgðH2citÞꢀ
ð1Þ
However, the product of this reaction when carried out in
EtOH solution consisted of a mixture of 1 and the monohydro-
gen citrate complex 2, indicating that deprotonation of the dihy-
drogencitrate to monohydrogencitrate occurs under these
conditions with the formation of the binuclear complex 2:
2½ðPh3PÞ2AgðH2citÞꢀ ! ½ðPh3PÞ2Ag2ðHcitÞꢀ þ H3cit
ð2Þ
Crystal structures
Indeed, it was subsequently found that attempted recrystalliza-
tion of the pure dihydrogencitrate complex 1 from EtOH pro-
duced the pure monohydrogencitrate complex 2. In order to
prepare pure 1 from EtOH solution it was necessary to add
excess citric acid to the reaction mixture in order to suppress the
dissociation reaction (2) (see Experimental above and Crystal
structures below). In the presence of excess PPh3 the reaction
analogous to (1) yielded the 1 : 3 complex 3:
In many carboxylate structures, the potentially oligodentate car-
boxylate ligand functionalities serve to bridge metal atoms to
yield oligo- or polymeric forms (the complexes of silver being
no exception),1 a common motif being the Ag(O·CX·O)2Ag
eight-membered ring, as found in the structure of the parent
silver acetate,60 possibly augmented by bridges further afield
from the oxygen atoms. However, in the present structurally
characterised species, it is evident that the stratagem of occupy-
ing coordination sites about the silver atom with multiple PPh3
donors has been effective in assisting to thwart such multiple
interactions, and all carboxylate–silver interactions are cleanly
bidentate chelate in CO2AgP2 environments in arrays which are
discrete neutral mono- and bi-nuclear.
Ag3cit þ 2H3cit þ 9PPh3 ! 3½ðPh3PÞ3AgðH2citÞꢀ
ð3Þ
The above complexes were more cleanly prepared by solvent-
assisted mechanochemical synthesis in a two-step process using
Ag2O, citric acid and PPh3 as starting materials, e.g. For
complex 1:
In the solvated complex 1, the 1-carboxylate group of citric
acid is deprotonated (as in the structure of sodium dihydrogenci-
trate),61 and supplanted by O,O′-complexation with a silver
atom, the complex being [(Ph3P)2Ag(O2CCH2C(OH)(COOH)
CH2COOH)], one molecule, together with a molecule of
ethanol, comprising the asymmetric unit of the structure. The
silver atom environment (Table 1) is similar to those found in
other silver(I) monocarboxylate phosphine complexes;21–25 there
is a pronounced asymmetry (0.03 Å) in the Ag–O distances, the
oxygen atom associated with the shorter being hydrogen-bonded
intramolecularly by the 3-hydroxyl group (Fig. 1(a)) (O,H⋯O
2.660(2), 1.95 Å). The hydrogen atom of the associated 3-car-
boxylate group hydrogen-bonds to the other coordinated oxygen
atom of an inversion-related molecule so that a centrosymmetric
dimer is formed (O,H⋯O 2.638(2), 1.82 Å). Hydrogen-bonds
between the uncoordinated terminal 5-carboxylate group and the
ethanol solvent molecule knit the dimers into a one-dimensional
chain (Fig. 1(b)) (O,H(51)⋯O(10) 2.584(2), 1.76; O,H(10)⋯O
3Ag2O þ 6H3cit þ 12PPh3 ! 6½ðPh3PÞ2AgðH2citÞꢀ þ 3H2O ð4Þ
In the first step the appropriate amounts of Ag2O and citric
acid were reacted together in a single water-assisted mechano-
chemical treatment. The product of this reaction was then reacted
with PPh3 in an ethanol-assisted mechanochemical reaction. In
the case of compounds 1 and 2, the second reaction was com-
plete after only one or two mechanochemical treatments. For
compound 3, monitoring of the product by IR spectroscopy
showed that the reaction was not complete after three mechano-
chemical treatments, so the method of solvent-assisted solid-state
synthesis was applied,36 treating the reaction mixture with a
small amount of ethanol and heating in a closed tube at 60 °C.
Recent reports have demonstrated the use of Group 2 and Group
12 metal oxides as reactants in mechanochemical synthesis,42–45
and the present examples extend this to the case of the Group
11 metal oxide Ag2O. Also, previous studies have demonstrated
the possibility of carrying out solvent-assisted mechanochemical
(52) (2 − x, y, 1 − z) 2.821(2), 2.09 Å).
ˉ
5412 | Dalton Trans., 2012, 41, 5409–5417
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 2 Molecular projection of [{(Ph3P)2Ag}2(Hcit)], 2.
Scheme 1
Infrared spectroscopy
Fig. 1 (a) Molecular projection of [(Ph3P)2Ag(H2cit)]·(EtOH), 1. (b)
The extended one-dimensional hydrogen-bonded array.
The mid-range IR spectra (4000–1500 cm−1) of the citrate com-
plexes 1, 2, 3 are shown in Fig. 3. These contain bands due to
the coordinated citrate anion, together with some bands (not
labelled) due to the triphenylphosphine ligand. A broad ν(OH)
band derivative of the carboxylic acid groups occurs in the range
2500–3500 cm−1, similar to the range observed in citric acid
itself (Fig. 4(a)). The large width and complex structure of the
ν(OH) band is probably the result of extensive hydrogen-
bonding, which is evident in the structures of compounds 1 and
2 (see Crystal structures, above).
The asymmetric CO stretches νa(CO) of the carboxylic acid
and carboxylate groups occur at ca. 1720 and 1560 cm−1
respectively in complexes 1, 2 and 3 (Fig. 3(a), (c), (d)). These
values are in good agreement with those found in related
systems.6,62,63 In the spectrum of complex 1 the νa(CO) bands
occur at 1746, 1719 and 1561 cm−1, consistent with the presence
of two carboxylic acid groups and one carboxylate group in this
dihydrogencitrate complex. Upon conversion to the monohydro-
gencitrate complex 2, the absorption intensity in the 1560 cm−1
region increases at the expense of that in the 1720 cm−1 region,
consistent with the deprotonation of the dihydrogencitrate to
monohydrogencitrate upon formation of 2 from 1. Although no
The other structurally characterized silver(I)/citrate compound
is a complex of the Hcit2− anion, [{(Ph3P)2Ag}2(Hcit)] ≡
[(Ph3P)2Ag(O2CCH2C(OH)(CH2COOH)CO2)Ag(PPh3)2],
2,
unsolvated, and with a single molecule comprising the asym-
metric unit of the structure (Fig. 2). Although the determination
is of inferior precision, it establishes clearly that, rather than both
of the (‘terminal’) 1,5-carboxylates being symmetrically deproto-
nated (or, for that matter, the alcohol OH), one terminal and the
central 3-carboxylates are those which chelate the two silver
atoms; the data of Table 1 suggest that the interaction of the 3-
carboxylate may be the weaker of the two, indicative of signifi-
cantly different basicity. An internal intraligand hydrogen-bond
is found between the carboxylic hydrogen and the more strongly
coordinated oxygen atom of the more distant chelating deproto-
nated group (O,H(51)⋯O(12) 2.54(2), 1.70 Å).
Based on the above structures, an overall scheme for the for-
mation of 1 and 2 is shown in Scheme 1, where the mechano-
chemical reaction is represented in the formation of 1 and the
solution reaction is represented in the formation of 2.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 5409–5417 | 5413

View Article Online
Fig. 3 Mid-range IR spectra of (a) [(Ph3P)2Ag(H2cit)]·EtOH, 1. (b)
[(Ph3P)2Ag(H2cit)], 1b. (c) [{(Ph3P)2Ag}2(Hcit)], 2. (d) [(Ph3P)3Ag-
(H2cit)], 3.
crystal structure has been accessible for complex 3, the relative
intensities of the νa(CO) bands at 1718, 1561 cm−1 are consistent
with the proposed formulation [(Ph3P)3Ag(H2cit)], these bands
being assigned to two carboxylic acid and one coordinated car-
boxylate group respectively.
The IR spectrum of complex 1 also contains a ν(OH) band at
3497 cm−1 ascribed to the lattice EtOH present in the solvated
structure (Fig. 3(a)). This band disappears when the complex is
desolvated by heating it to 50 °C (Fig. 3(b); see Experimental).
The desolvation has an unexpectedly large effect on the νa(CO)
bands of the carboxylic acid and carboxylate groups, which
appear at 1703, 1582 cm−1 respectively. The shifts in these
bands upon desolvation are greater than those caused by a
change in the state of deprotonation of the citric acid (cf. 1 and
2; Fig. 3(a) and (c)) and by a change in the number of coordi-
nated PPh3 ligands (cf. 1 and 3; Fig. 3(a) and (d)), which is con-
sistent with loss of the lattice EtOH causing a change in the
structure of the complex. This is consistent with the 31P CP
MAS NMR spectra of the solvated form 1 and the desolvated
form 1b (see below). Nevertheless, the process is completely
reversible; grinding the desolvated 1 with EtOH under ambient
Fig. 4 Mid-range IR spectra of (a) citric acid monohydrate, H3cit·H2O,
and of the products of H2O assisted mechanochemical reactions of
nAg2O: H3cit·H2O for (b) n = 0.5, (c) n = 1, (d) n = 1.5; the spectrum of
NaH2cit (e) is included for comparison (see text).
conditions results in the reformation of the EtOH solvate, as evi-
denced by its IR spectrum, which is identical to that of the initial
solvated complex (Fig. 3(a)). Reversible structural changes
associated with solvation–desolvation of a complex solvate have
been reported previously.35
The mechanochemical syntheses of complexes 1, 2, and 3
were carried out in two steps, the first being a water-assisted
reaction between Ag2O and citric acid, and the second an EtOH-
assisted reaction between the product of the first reaction and an
appropriate amount of PPh3. It was therefore of interest to
5414 | Dalton Trans., 2012, 41, 5409–5417
This journal is © The Royal Society of Chemistry 2012

View Article Online
investigate the nature of the first reaction, and this was achieved
by carrying out an IR study of the products of the following
water-assisted mechanochemical reactions
nAg2O þ H3cit ! product
ð5Þ
for n = 0.5, 1, 1.5 (see Experimental). Possible products of these
reactions are AgH2cit, Ag2Hcit and Ag3cit respectively. The IR
spectra of citric acid and the products of the above reactions are
shown in Fig. 4. From these it is clear that the reactions for n =
0.5, 1 produce a mixture of citric acid (νa(CO) = 1753, 1721,
1685 cm−1); cf. Fig. 4(a)) and a product with νa(CO) = 1563,
1535, 1517 cm−1. This latter product is the only one produced
for the n = 1.5 case, indicating that the product is Ag3cit. Thus
the products of the n = 0.5, 1 reactions are mixtures of Ag3cit
and citric acid. Therefore, although the PPh3 complexes 1 and 2
formally contain AgH2cit and Ag2Hcit respectively, the initial
mechanochemical reactions using the appropriate proportions of
Ag2O and citric acid do not produce these latter two compounds.
Rather, the PPh3 complexes of AgH2cit and Ag2Hcit form only
after reaction of the initially formed mixture of Ag3cit and citric
acid with PPh3. The above results cast some doubt on the exist-
ence of compounds such as AgH2cit and Ag2Hcit, despite refer-
ences in the patent literature to ‘silver dihydrogen citrate’ and its
antimicrobial action.18 This is rather surprising, as the monoso-
dium salt of citric acid, NaH2cit exists,61 and its IR spectrum is
shown in Fig. 4(e). From this it is clear that mono-deprotonation
of citric acid results in substantial changes in the IR spectrum, so
that if the product of the reaction whose spectrum is shown in
Fig. 4(b) had been AgH2cit, then it would not contain bands
with exactly the same wavenumbers and relative intensities as
citric acid.
Solid-state NMR spectroscopy
Fig. 5 Solid-state 31P CPMAS NMR spectra acquired at 9.40 T of (a)
[(Ph3P)2Ag(H2cit)]·EtOH, 1. (b) [(Ph3P)2Ag(H2cit)], 1b. (c)
[{(Ph3P)2Ag}2(Hcit)], 2. (d) [(Ph3P)3Ag(H2cit)], 3.
The solid-state 31P CPMAS NMR spectra of complexes 1, 2,
and 3 are shown in Fig. 5 and an expansion and simulation of
the spectrum of complex 1 is shown in Fig. 6. The derived NMR
parameters are listed in Table 2. In general, the spectra of the
silver complexes are expected to consist of partially resolved
multiplets due to the individual 31P chemical shifts resulting
from the chemically inequivalent P nuclei bound to each Ag
centre. Each resonance displays 1J(107,109Ag,31P) scalar coupling
between the 31P (I = 1/2) and 107,109Ag nuclei (I = 1/2), and
further splitting of these doublets can occur via 2J(31P, 31P) coup-
ling.64 For the present compounds, this is well illustrated by
compound 1, which contains two inequivalent P atoms bound to
the same Ag atom, resulting in an ABX spin system (A, B = 31P;
X = 107,109Ag) for each Ag isotope. The 31P spectrum shows the
AB part of the ABX spectrum, which in general consists of two
overlapping sets of AB quartets,65 the separate patterns for coup-
ling to the 107,109Ag isotopes sometimes remaining unresolved.
In the case of 1 the spectrum consists of two overlapping, nearly
identical AB quartets, in which the 107,109Ag isotopic splitting is
partially resolved (Fig. 5(a), Fig. 6). The relative intensities of
the peaks (as well as correlation peaks in the 2D 31P CPCOSY
spectrum, not shown) allow unambiguous assignment of the two
AB sub-spectra, from which the parameters in Table 2 are
obtained. The spectrum of the desolvated complex 1b (Fig. 5(b))
Fig. 6 Expansion of the 31P CP MAS NMR spectrum of [(Ph3P)2Ag
(H2cit)]·EtOH, 1 and stick diagram of the spectrum simulated as the AB
107
part of an ABX system with δA = 7.39, δB = 12.71, JAX = 384 Hz,
JAX109 = 442 Hz, JBX107 = 475 Hz, JBX109 = 546 Hz, JAB = 132 Hz.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 5409–5417 | 5415

View Article Online
Table 2 31P CPMAS NMR parameters
environments. The mechanochemical synthetic method provides
another example of the recently demonstrated use of metal
oxides in such syntheses and, further, it demonstrates the appli-
cation of two-step solvent-assisted mechanochemical synthesis
in which different solvents are used for the two reaction steps,
the solvents being chosen on the basis of the solubility of a com-
ponent or components of the reaction in the solvent concerned.
The above complexes may be regarded as adducts of Ag(H2cit)
or Ag2(Hcit) with two or three molecules of PPh3 per Ag atom,
but mechanochemical studies showed that neither Ag(H2cit) nor
Ag2(Hcit) can be prepared as distinct compounds, despite the
existence in the literature of references to compounds such as
‘silver dihydrogen citrate’. The present work shows that the only
stable product of the reaction between silver oxide and citric acid
is trisilver citrate, Ag3cit. IR studies of 1–3 show the presence of
coordinated carboxylate and free carboxylic acid groups in the
mono- and di-hydrogencitrate ligands, and the formation of 2
from 1 shows that dihydrogencitrate deprotonation can occur
upon dissolution of 1 in protic solvents. In such cases, the
present work shows that mechanochemical synthesis can proceed
to a unique product using stoichiometric quantities of reactants,
whereas solution synthesis requires an excess of the acid to
prevent unwanted deprotonation reactions.
Compound
δ(31P)/ppm 1J(107,109Ag,31P)/Hz
[(Ph3P)2Ag(H2cit)]·EtOH, 1 7.6, 12.9
412, 501 (2J(PP) = 134 Hz)
450
ca. 465
160
[(Ph3P)2Ag(H2cit)], 1b
[{(Ph3P)2Ag}2(Hcit)], 2
[(Ph3P)3Ag(H2cit)], 3
8.8
ca. 7.9
5.25
is much simpler, being effectively a simple doublet, indicating
that loss of the lattice EtOH causes a significant change in the
structure of the complex, such that the two Ag–P bonds become
nearly equivalent.
The spectrum of 2 consists of an unresolved multiplet that has
not been analysed in detail, although peak separations compar-
able in magnitude to 1J(107,109Ag,31P) in 1 are evident. The
complex spectrum is not surprising because the compound con-
tains two inequivalent Ag(PPh3)2 groups that would give rise to
two ABX spin systems, with a total of sixteen lines in the 31P
spectrum. These are clearly not all resolved, possibly because of
broadening of the lines that might be the result of the disorder
observed in the crystal structure of this complex (see Experimen-
tal). There is also a very weak shoulder in the spectrum at about
15 ppm that is probably due to the presence of a small amount
of 1 present as an impurity.
Acknowledgements
The spectrum of 3 consists of a dominant peak at ca. 5 ppm
that is assigned to the 1 : 3 complex [(Ph3P)3Ag(H2cit)], with a
number of weaker peaks at lower field that can readily be
assigned to the EtOH-solvated 1 : 2 complex 1. The dominant
peak is a strongly overlapped doublet with a splitting that is
assigned to 1J(107,109Ag,31P) coupling (Table 2).
We would like to thank EPSRC and the University of Warwick
for partial funding of the solid state NMR infrastructure at
Warwick, and acknowledge additional support for this infrastruc-
ture obtained through Birmingham Science City: Innovative
Uses for Advanced Materials in the Modern World (West Mid-
lands Centre for Advanced Materials Project 2), with support
from Advantage West Midlands (AWM) and partial funding by
the European Regional Development Fund (ERDF).
The 1J(Ag,P) coupling constants in Table 2 are consistent with
the structures of 1 and 2 and with the proposed structure of 3. It
has previously been shown that one of the factors that determines
the magnitude of 1J(M,P) coupling constants in metal-phosphine
complexes is the number of phosphine ligands coordinated to the
metal atom M, a reduction in magnitude occurring with an
increase in the number of phosphine ligands.66–68 In particular,
the values are similar to those reported previously for other
silver complexes containing an oxyanion and two or three co-
ordinated PPh3 ligands.5,6,66 In the case of the formate com-
References
1 (a) R. J. Lancashire, in Comprehensive Coordination Chemistry, ed.
G. Wilkinson, R. D. Gillard, J. A. McCleverty, Pergamon, Oxford, 1987,
vol. 5, p. 775; (b) M. C. Gimeno and A. Laguna, in Comprehensive
Coordination Chemistry II, eds-in-chief, J. A. McCleverty and
T. J. Meyer, Elsevier Pergamon, 2004, vol. 6, ed. D. E. Fenton, p. 911.
2 C. E. Holloway, M. Melnik, W. A. Nevin and W. Liu, J. Coord. Chem.,
1995, 35, 85.
1
plexes [(Ph3P)nAg (O2CH)], J(Ag,P) reduces from ca. 430 (n =
2) to ca. 230 Hz (n = 3) with increasing phosphine ligand
coordination, which is similar to the change observed for the
corresponding dihydrogencitrate complexes 1 and 3 (Table 2).
3 A. Angel and A. V. Harcourt, J. Chem. Soc. Trans., 1902, 81, 1385.
4 A. Angel, J. Chem. Soc. Trans., 1906, 89, 345.
5 G. A. Bowmaker, Effendy, J. V. Hanna, P. C. Healy, G. J. Millar, B.
W. Skelton and A. H. White, J. Phys. Chem., 1995, 99, 3909.
6 G. A. Bowmaker, Effendy, J. V. Hanna, P. C. Healy, J. C. Reid,
C. E. F. Rickard and A. H. White, J. Chem. Soc., Dalton Trans., 2000,
753.
7 J. Shibata, K.-I. Shimizu, S. Satokawa, A. Satsuma and T. Hattori, Phys.
Chem. Chem. Phys., 2003, 5, 2154.
8 K. Vehlow, K. Köhler, S. Blechert, S. Dechert and F. Meyer,
Eur. J. Inorg. Chem., 2005, 2727.
9 L. Jiang, Z. Wu, D. Wu, W. Yang and R. Jin, Nanotechnology, 2007, 18,
185603.
10 Z. S. Pillai and P. V. Kamat, J. Phys. Chem. B, 2004, 108, 945.
11 O. Siiman, L. A. Bumm, R. Callaghan, C. G. Blatchford and M. Kerker,
J. Phys. Chem., 1983, 87, 1014.
12 J. Clarkson, C. Campbell, B. N. Rospendowski and W. E. Smith, J.
Raman Spectrosc., 1991, 22, 771.
13 C. H. Munro, W. E. Smith, M. Garner, J. Clarkson and P. C. White, Lang-
muir, 1995, 11, 3712.
Conclusion
The complexes [(Ph3P)2Ag(H2cit)]·EtOH (1), [{(Ph3P)2Ag}2-
(Hcit)], (2) and [(Ph3P)3Ag(H2cit)] (3) were prepared by conven-
tional solution methods, involving the reaction of trisilver(I)
citrate, citric acid and triphenylphosphine, and also by a mechan-
ochemical method involving the reaction of silver(I) oxide, citric
acid and triphenylphosphine. Complexes 1 and 2 have been
structurally defined by single crystal X-ray studies as the neutral
mononuclear [(Ph3P)2Ag(O2CCH2C(OH)(COOH)CH2COOH)]
and binuclear [(Ph3P)2Ag(O2CCH2C(OH)(CH2COOH)CO2)Ag-
(PPh3)2] molecular species respectively, the carboxylate groups
being cleanly bidentate chelate in CO2AgP2 coordination
5416 | Dalton Trans., 2012, 41, 5409–5417
This journal is © The Royal Society of Chemistry 2012

View Article Online
14 C. Rodger, W. E. Smith, G. Dent and M. Edmondson, J. Chem. Soc.,
Dalton Trans., 1996, 791.
15 K. K. Caswell, C. M. Bender and C. J. Murphy, Nano Lett., 2003, 3, 667.
16 F.-K. Liu, P.-W. Huang, Y.-C. Chang, F.-H. Ko and T.-C. Chu, J. Mater.
Res., 2011, 19, 469.
17 A. Jakob, T. Rüffer, H. Schmidt, P. Djiele, K. Körbitz, P. Ecorchard,
T. Haase, K. Kohse-Höinghaus, S. Frühauf, T. Wächtler, S. Schulz,
T. Gessner and H. Lang, Eur. J. Inorg. Chem., 2010, 2975.
18 A. B. Arata, US Patent Pat, 20060100273, 2006.
19 J. Adaskaveg, H. Foerster, B. A. Holtz, E. Hoffman, D. Gubler and
E. Erickson, Acta Hortic. (ISHS), 2006, 704, 277.
20 V. Sambhy, M. M. MacBride, B. R. Peterson and A. Sen, J. Am. Chem.
Soc., 2006, 128, 9798.
21 S. W. Ng and A. H. Othman, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun., 1997, 53, 1396.
22 S. W. Ng, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1998, 54,
743.
23 J. V. Hanna and S. W. Ng, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun., 1999, 55, IUC9900031.
39 T. Friscic and W. Jones, Cryst. Growth Des., 2009, 9, 1621.
40 T. Friscic, A. V. Trask, W. Jones and W. D. S. Motherwell, Angew. Chem.,
Int. Ed., 2006, 45, 7546.
41 A. N. Swinburne and J. W. Steed, CrystEngComm, 2009, 11, 433.
42 T. Friscic and L. Fabian, CrystEngComm, 2009, 11, 743.
43 E. H. H. Chow, F. C. Strobridge and T. Friscic, Chem. Commun., 2010,
46, 6368.
44 F. C. Strobridge, N. Judas and T. Friscic, CrystEngComm, 2010, 12,
2409.
45 T. Friscic, J. Mater. Chem., 2010, 20, 7599.
46 C. J. Adams, H. M. Colquhoun, P. C. Crawford, M. Lusi and A.
G. Orpen, Angew. Chem., Int. Ed., 2007, 46, 1124.
47 D. Braga, S. L. Giaffreda, F. Grepioni, A. Pettersen, L. Maini, M. Curzi
and M. Polito, Dalton Trans., 2006, 1249.
48 D. Braga, M. Curzi, A. Johansson, M. Polito, K. Rubini and F. Grepioni,
Angew. Chem., Int. Ed., 2006, 45, 142.
49 L.-F. Yang, M.-L. Cao, Y. Cui, J.-J. Wu and B.-H. Ye, Cryst. Growth
Des., 2010, 10, 1263.
50 V. Strukil, L. Fabian, D. G. Reid, M. J. Duer, G. J. Jackson, M. Eckert-
Maksic and T. Friscic, Chem. Commun., 2010, 46, 9191.
51 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2007,
64, 112.
52 A. Pines, M. G. Gibby and J. S. Waugh, J. Chem. Phys., 1973, 59, 569.
53 E. R. Andrew, A. Bradbury and R. Eades, Nature, 1958, 182, 1659.
54 E. O. Stejskal and J. Schaefer, J. Magn. Reson., 1975, 18, 560.
55 G. Bodenhausen, R. L. Vold and R. R. Vold, J. Magn. Reson., 1980, 37,
93.
56 D. Marion and K. Wuethrich, Biochem. Biophys. Res. Commun., 1983,
113, 967.
57 T. Allman, J. Magn. Reson., 1989, 83, 637.
24 J. V. Hanna and S. W. Ng, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun., 2000, 56, 24.
25 B. Femi-Onadenko, Z. Kristallogr., 1980, 152, 159.
26 Y. Zhao, P. Zhang, B. Li, X. Meng and T. Zhang, Inorg. Chem., 2011, 50,
9097.
27 D. W. Hartley, G. Smith, D. S. Sagatys and C. H. L. Kennard, J. Chem.
Soc., Dalton Trans., 1991, 2735.
28 D. S. Sagatys, G. Smith, R. C. Bott, D. E. Lynch and C. H. L. Kennard,
Polyhedron, 1993, 12, 709. (ref. 27 and 28 record two of the three (only)
structural characterisations available for silver(I)/citrate complexes. The
third is a more complex form entailing borate as well as citrate as the
complexing anionic species (I. I. Zviedre, V. K. Bel’sky and
E. M. Shvarts, Latv. Khim. Z., 1998, 99 (CCDC: 159197 (LUQHIO)),
which is less helpful).
29 G. H. Penner and W. Li, Inorg. Chem., 2004, 43, 5588.
30 G. A. Bowmaker, N. Chaichit, C. Pakawatchai, B. W. Skelton and A.
H. White, Dalton Trans., 2008, 2926.
58 J. V. Hanna, M. E. Smith, S. N. Stuart and P. C. Healy, J. Phys. Chem.,
1992, 96, 7560.
59 J. V. Hanna, R. D. Hart, P. C. Healy, B. W. Skelton and A. H. White, J.
Chem. Soc., Dalton Trans., 1998, 2321.
60 L. P. Olson, D. R. Whitcomb, M. Rajeswaran, T. N. Blanton and B.
J. Swertka, Chem. Mater., 2006, 18, 1667.
31 G. A. Bowmaker, N. Chaichit, C. Pakawatchai, B. W. Skelton and A.
H. White, Can. J. Chem., 2009, 87, 161.
32 G. A. Bowmaker, B. W. Skelton and A. H. White, Inorg. Chem., 2009,
48, 3185.
33 G. A. Bowmaker, C. Pakawatchai, S. Saithong, B. W. Skelton and A.
H. White, Dalton Trans., 2009, 2588.
34 G. A. Bowmaker, C. Pakawatchai, S. Saithong, B. W. Skelton and A.
H. White, Dalton Trans., 2010, 39, 4391.
35 G. A. Bowmaker, J. V. Hanna, B. W. Skelton and A. H. White, Dalton
Trans., 2009, 5447.
36 G. A. Bowmaker, J. V. Hanna, B. W. Skelton and A. H. White, Chem.
Commun., 2009, 2168.
37 G. A. Bowmaker, C. Di Nicola, C. Pettinari, B. W. Skelton, N. Somers
and A. H. White, Dalton Trans., 2011, 40, 5102.
61 J. P. Glusker, D. van der Helm, W. E. Love, M. L. Dornberg, J.
A. Minkin, C. K. Johnson and A. L. Patterson, Acta Crystallogr., 1965,
19, 561.
62 D. L. Pavia, G. M. Lampman and G. S. Kriz, Introduction to Spec-
troscopy, Brooks/Cole, 2001, pp. 60–62.
63 K. Nakamoto, John Wiley & Sons, Inc., New York, 5th edn, 1997, pp.
59–62.
64 G. A. Bowmaker, J. V. Hanna, C. E. F. Rickard and A. S. Lipton, J.
Chem. Soc., Dalton Trans., 2001, 20.
65 R. K. Harris, Nuclear Magnetic Resonance Spectroscopy, Longman,
London, 1986, pp. 54–57.
66 L. J. Baker, G. A. Bowmaker, D. Camp, Effendy, P. C. Healy,
H. Schmidbaur, O. Steigelmann and A. H. White, Inorg. Chem., 1992,
31, 3656.
38 G. A. Bowmaker, Effendy, J. V. Hanna, P. C. Healy, S. P. King,
C. Pettinari, B. W. Skelton and A. H. White, Dalton Trans., 2011, 40,
7210.
67 E. C. Alyea, J. Malito and J. H. Nelson, Inorg. Chem., 1987, 26, 4294.
68 E. L. Muetterties and C. W. Alegranti, J. Am. Chem. Soc., 1972, 94,
6386.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 5409–5417 | 5417