Catalytic bond forming reactions promoted by amidinate, guanidinate and phosphaguanidinate compounds of magnesium

Article abstract of DOI:10.1039/c4dt01097c

The synthesis and catalytic properties of a series of magnesium compounds consisting of monoanionic, N,N'-chelating ligands (N∩N = amidinates, guanidinates, phosphaguanidinates) is reported. The compounds were synthesized by (i) insertion of a carbodiimide into an existing Mg-C or Mg-N bond, or (ii) protonolysis of an organomagnesium compound by a neutral pre-ligand. Structural analyses of mono- or bis-(chelate) compounds with general formula Mg(N∩N)X(L)_nand Mg(N∩N)₂(L)_n(X = halide, aryloxide, amide; L = Et₂O, THF; n = 0, 1 or 2) have been performed and the influence that the ligand substituent patterns have on the solid-state structures has been probed. Selected examples of the compounds were tested as (pre)catalysts for the polymerization of lactide, the dimerization of aldehydes and the hydroacetylenation of carbodiimides.

Full text of DOI:10.1039/c4dt01097c

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Catalytic bond forming reactions promoted by
amidinate, guanidinate and phosphaguanidinate
compounds of magnesium†
Cite this: Dalton Trans., 2014, 43,
14302
a
b
b
b
Ryan J. Schwamm, Benjamin M. Day, Natalie E. Mansfield, William Knowelden,
b
a
Peter B. Hitchcock and Martyn P. Coles*
The synthesis and catalytic properties of a series of magnesium compounds consisting of monoanionic,
N,N’-chelating ligands (N∩N = amidinates, guanidinates, phosphaguanidinates) is reported. The com-
pounds were synthesized by (i) insertion of a carbodiimide into an existing Mg–C or Mg–N bond, or (ii)
protonolysis of an organomagnesium compound by a neutral pre-ligand. Structural analyses of mono- or
bis-(chelate) compounds with general formula Mg(N∩N)X(L)
n
2 n
and Mg(N∩N) (L) (X = halide, aryloxide,
Received 14th April 2014,
Accepted 7th May 2014
amide; L = Et O, THF; n = 0, 1 or 2) have been performed and the influence that the ligand substituent
2
patterns have on the solid-state structures has been probed. Selected examples of the compounds were
DOI: 10.1039/c4dt01097c
tested as (pre)catalysts for the polymerization of lactide, the dimerization of aldehydes and the hydro-
acetylenation of carbodiimides.
www.rsc.org/dalton
Introduction
The application of group 2 compounds containing Mg, Ca, Sr,
Ba as catalysts is an attractive area of research due to the rela-
tively low toxicity of the metal and abundance of cheap/easily
1
–4
synthesised starting materials.
Magnesium is particularly
favoured in this respect due to the ubiquitous presence of
organomagnesium compounds in synthetic chemistry labora-
tories, typically in the form of Grignard reagents. Accordingly,
in the last 5 years magnesium complexes have been reported
Fig. 1 Generic amidinate, guanidinate and phosphaguanidinate ligands.
for their activity as catalysts in a range of chemical transform- metallacycle is bonded to a –PR substituent (Fig. 1). The
2
5
–9
10–13
ations including hydroamination, hydroboration,
cross- structural consequences of the different substitution patterns
1
4
15–17
dehydrocoupling,
ring-opening polymerization (ROP),
and the homocoupling of aldehydes
are investigated for a range of four-, five- and six-coordinate
1
8,19
(the Tishchenko reac- compounds.
20
tion). We have recently explored a series of magnesium
amidinate and guanidinate compounds as (pre)-catalysts in
2
1–23
selected bond forming reactions.
In this contribution we
report additional results towards understanding the complex Experimental
equilibrium chemistry involved in these catalytic cycles and
the application of selected examples to ROP catalysis. To sup-
plement the results from existing amidinate and guanidinate
compounds, we have extended this area to include phospha-
guanidinate compounds in which the carbon atom of the
General information
All manipulations were carried out under dry nitrogen using
standard Schlenk-line and cannula techniques, or in a conven-
tional nitrogen-filled glovebox. Solvents were dried over appro-
priate drying agents and degassed prior to use. All compounds
were purchased from Sigma-Aldrich and used as received
unless otherwise stated. MgMeCl (3 M solution in THF) and
a
School of Chemical and Physical Sciences, Victoria University of Wellington,
Wellington, 6012, New Zealand. E-mail: martyn.coles@vuw.ac.nz
MgMeBr (3 M solution in Et O) were received as solutions. LiN-
2
b
Department of Chemistry, University of Sussex, Falmer, Brighton BN1 9QJ, UK
{
3 2 3 2
SiMe } and KN{SiMe } were crystallized from toluene; benz-
†
Electronic supplementary information (ESI) available: Thermal ellipsoid plots
of [7] , [8] and 14. CCDC 996734–996740. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c4dt01097c
aldehyde was freshly distilled and stored under nitrogen; rac-
Lactide was sublimed before use. Ph PC{NCy}{NHCy}, Cy PC
2
2
24
2
2
14302 | Dalton Trans., 2014, 43, 14302–14314
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
i
i
25
25
{
N Pr}{NH Pr}, and Cy
2
PC{NCy}{NHCy} were synthesised as solution in THF, 0.48 mmol) was added drop wise to a clear
reported.
colourless solution of Ph PC{NCy}{NHCy} (0.200 g,
2
NMR spectra were recorded using a Bruker Avance DPX 0.50 mmol) in THF (∼10 mL). The resultant clear, colourless
1
13
1
3
00 MHz spectrometer at 300.1 ( H), 75.4 ( C{ H}) and 121.5 solution was stirred at ambient temperature for 1 h before
3
1
1
1
(
(
P{ H}) MHz, a Varian INOVA system at 300.1 ( H) or 75.3 being added to a clear colourless solution of LiOAr′ (0.113 g,
C{ H}) MHz, or a Varian VNMR 400 MHz spectrometer at 0.50 mmol) in THF (∼10 mL). Stirring for 16 h at ambient
00.1 ( H), 100.3 ( C{ H}) and 161.9 ( P{ H}) MHz. Proton temperature followed by removal of volatiles in vacuo afforded
1
3
1
1
13
1
31
1
4
and carbon chemical shifts were referenced internally to a crude white solid. Extraction from LiCl with hot hexane and
residual solvent resonances; phosphorus chemical shifts were slow cooling to ambient temperature yielded colourless crys-
referenced to external H PO (85% aq.). Coupling constants tals of 12. Yield 0.170 g (48%). Anal. Calcd for C44H N O MgP
3 4 63 2 2
are quoted in Hz. Elemental analyses were performed by (707.26): C, 74.72; H, 8.98; N, 3.96%. Found: C, 74.66; H, 8.88;
1
3
S. Boyer at London Metropolitan University.
Synthesis of Mg(Ph PC{NCy} )Cl(THF) (6). A solution of
N, 3.99%. H NMR (300 MHz, C
6
D
6
): δ 7.66 (d t, JHH = 7.1,
), 7.12–7.02 (m, 6H,
4
2
2
J
PH = 1.5, 4H, m-C
6
H
5
), 7.32 (s, 2H, m-C
6
H
2
MgMeCl (0.08 mL of a 3 M solution in THF, 0.25 mmol) was o,p-C H ), 3.82 (m, 2H, Cy–H ), 3.60 (m, 4H, THF), 2.46 (s, 3H,
6
5
α
added drop wise to a solution of Ph
2
PC(NCy)(NHCy) (0.100 g, Ar′–CH
3
), 1.83 (s, 18H, Ar′–C(CH
3
)
3
, 1.68–1.00 (m, 20H,
), 1.17 (m, 4H, THF). C{ H} NMR (75 MHz, C ):
was stirred at ambient temperature for 16 hours prior to δ 172.5 (d, JPC = 49, PCN ), 160.9 (s, i-C H ), 137.2 (C H ),
1
3
1
0
.25 mmol) in toluene. The resultant clear, colourless solution Cy–CH
2
6 6
D
1
2
6
2
6
2
removal of volatiles under reduced pressure. Crystallization of 135.5 (d, JPC = 17, C
6
H
5
), 134.3 (d, JPC = 19, C
6
H
5
), 132.3 (d,
), 69.6
the crude white solid by slow cooling a hot hexane solution to PC = 19, C ), 129.1 (C
J
6
H
5
6
H
2
), 125.7 (C ), 121.3 (C H
6
H
2
6 2
3
ambient temperature afforded 6 as colourless crystals. Yield (THF), 56.0 (d, J = 18, Cy–C ), 37.2, 35.6 (Cy–CH ), 31.8 (Ar′–
PC
α
2
0
.11 g (84%). Anal. Calcd for C29
H
40
N
2
OMgPCl (523.37): C, C(CH
3
)
3
), 26.2 (THF), 24.9 (Cy–CH
2
), 21.7 (Ar′–CH
3
). The reson-
6
6.55; H, 7.70; N, 5.35%. Found: C, 66.49; H, 7.65; N, 5.33%. ance for C(CH ) could not be confidently assigned due to it’s
3 3
1
3
4
31
1
H NMR (300 MHz, C D ): δ 7.74 (d t, J = 6.4, J = 1.5, 4H, low intensity and possible overlap with other signals. P{ H}
6
6
HH
PH
3
m-C
p-C
0H, Cy–CH ), 1.47 (m, 4H, THF). C{ H} NMR (75 MHz, tion of MgMeCl (0.16 mL of a 3 M solution in THF,
6
H
5
), 7.54 (t,
J
HH = 6.1, 4H, o-C
6
H
5
), 7.12–7.02 (m, 2H, NMR (121 MHz, C
6
D
6
): δ −20.5.
6
H
5
), 4.23 (m, 2H, Cy–H
α
), 4.12 (m, 4H, THF), 1.90–0.82 (m,
Synthesis of Mg(Ph
2
PC{NCy}
2 3 2
)(N{SiMe } )(THF) (14). A solu-
1
3
1
2
C
1
6
2
1
6
D
6
): δ 170.8 (d, JPC = 53, PCN
2
), 134.3 (d, JPC = 20, C
), 129.0 (d, JPC = 7, C
6
H
H
5
5
), 0.48 mmol) was added drop wise to a clear colourless solution
), of Ph PC{NCy}{NHCy} (0.200 g, 0.50 mmol) in THF (∼10 mL).
32.4 (d, JPC = 19, C ), 129.3 (C
9.4 (THF), 56.4 (d, J = 18, Cy–C ), 37.0, 35.7 (Cy–CH ), 26.3 The resultant clear, colourless solution was stirred at ambient
6
H
5
6
H
5
6
2
3
PC
3
α
2
1
1
(
THF), 25.2 (Cy–CH
Synthesis of Mg(Cy
was made in an analogous manner to that described for 6, 16 h at ambient temperature followed by removal of volatiles
using 0.22 mL of a 3 M solution of MgMeBr in Et in vacuo afforded a crude white solid. Extraction from the KCl
0.66 mmol) and 0.250 g Cy PC{NCy}{NHCy} (0.64 mmol), and with hot hexane with slow cooling to ambient temperature
Et O as the solvent. Prolonged storage at 5 °C afforded a small yielded colourless crystals of 14. Yield 0.240 g (74%). Anal.
2
). P{ H} NMR (121 MHz, C
6
D
6
): δ −23.6.
temperature for 1 h before being added to a solution of KN
2
PC{NCy} )Br(OEt
2
) (7). The compound {SiMe } (0.100 g, 0.50 mmol) in THF (∼10 mL). Stirring for
2 3 2
2
O
(
2
2
number of colourless crystals (0.037 g, 10%) that were isolated Calcd for C35
and analysed by X-ray diffraction and elemental analysis. Anal. 6.48%. Found: C, 64.88; H, 9.02; N, 6.55%. H NMR (300 MHz,
58 3 2
H N OMgSi P (648.30): C, 64.84; H, 9.02; N,
1
3
4
Calcd for C H N OMgPBr (581.93): C, 59.85; H, 9.35; N, C D ): δ 7.67 (d t, J
= 8.4, J = 1.3, 4H, m-C H ), 7.10 (t,
2
9
54
2
6
6
HH
PH
6
5
3
4
.81%. Found: C, 59.71; H, 9.57; N, 4.72%.
J
HH = 6.2, 4H, o-C
)Cl(THF) (8). The compound 2H, Cy–H
was made in an analogous manner to that described for 6, 1.26 (m, 4H, THF), 0.48 (s, 18H, SiMe ). C{ H} NMR (75 MHz,
6
H
5
), 7.02 (t, J = 5.1, 6H, p-C
6
H
5
), 3.79 (m,
i
Synthesis of Mg(Cy
2
PC{N Pr}
2
α
), 3.68 (m, 4H, THF), 1.60–1.00 (m, 20H, Cy–CH
2
),
),
1
3
1
3
1
using 0.11 mL of a 3 M solution of MgMeCl in THF
C
6
D
5
): δ 171.9 (d, JPC = 48, PCN
2
), 135.7 (d, JPC = 17, C
6 5
H
i
i
(
0.33 mmol) and 0.100 g Cy
2
PC{N Pr}{NH Pr} (0.31 mmol) in 132.3 (d, JPC = 18, C ), 128.7 (C H
6
H
5
6 5
), 68.8 (THF), 56.2 (d,
3
toluene. Crystallization from hot hexane afforded 8 as colour-
JPC = 18, Cy–C ), 37.5 (Cy–CH ), 26.2 (THF), 26.1, 25.1
α
2
3
1
1
less crystals. Yield 0.09 g (60%). Despite repeated attempts, (Cy–CH
2
), 6.3 (SiMe
3
). P{ H} NMR (121 MHz, C
6 6
D ): δ −20.7.
1
accurate elemental could not be obtained. H NMR (300 MHz,
Synthesis of Mg(Cy
2
PC{NCy} )(N{SiMe )(THF) (15). The
2
3 2
}
C D ): δ 4.86 (m, 2H, CHMe ), 4.80 (m, 2H, Cy–H ), 3.90 (m, compound was made in an analogous manner to that
6
6
2
α
4
J
H, THF), 2.33–0.82 (m, Cy–CH
2
*), 1.43 (m, 4H, THF), 1.20 (d, described for 14, using 0.60 mL of a 3 M solution of MgMeBr
3
HH = 6.3, 12H, CHMe
2
) [*total integral for cyclohexyl methyl- in Et
2
O (1.80 mmol), 0.730 g Cy
2
PC{NCy}{NHCy} (1.80 mmol)
1
3
1
ene groups = 20H]. C{ H} NMR (75 MHz, C D ): δ 170.8 (d, and 0.300 g LiN{SiMe3}2 (1.80 mmol). Stirring for 18 h at
6
6
J = 53, PCN
2 α
), 68.6 (THF), 50.6 (d, J = 18, Cy–C ), 46.9 (d, J = 14, ambient temperature followed by removal of volatiles in vacuo
CHMe ), 34.2 (d, J = 17, Cy–CH
2
2
), 32.6 (d, J = 23, Cy–CH
2
), 30.6 afforded a crude white solid. Extraction from LiCl with hot
(
2
d, J = 12, Cy–CH ), 26.4 (d, J = 21, Cy–CH ), 25.8 (s, CHMe ), hexane and slow cooling to ambient temperature yielded
2
2
2
3
1
1
5.6 (THF). P{ H} NMR (121 MHz, C
Synthesis of Mg(Ph PC{NCy} )(OAr′)(THF) (12) (Ar′
,6- Bu -4-MeC H ). A solution of MgMeCl (0.16 mL of a 3 M Found: C, 63.51; H, 10.53; N, 6.26%. H NMR (400 MHz,
6
D
6
): δ −9.9.
colourless crystals of 15. Yield 1.00 g (84%). Anal. Calcd for
2
2
=
35 70 3 2
C H N OMgSi P (660.40): C, 63.65; H, 10.68; N, 6.36%.
t
1
2
2
6
2
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 14302–14314 | 14303

View Article Online
Paper
Dalton Transactions
C
6
D
6
): δ 3.71 (m, 4H, THF), 2.30 (m, 2H, Cy(P)–H
α
), 2.11–1.11 and all atom sites were set at 80% carbon, 20% oxygen; in
(
m, 42H, Cy( –H + Cy
–CH ), 1.27 (m, 4H, THF), 0.46 (s, addition there was conformational disorder of the solvate. All
2
N)
α
1
(N/P)
1
3
1
1
4
3
8H, SiMe
3
). C{ H} NMR (100 MHz, C
6
D
6
2
): δ 174.8 (d, JPC
=
),
disordered atoms were left isotropic.
9.7, PCN
2
), 68.7 (THF), 38.0 (br), 36.0 (d, JPC = 18, Cy(P)–C
α
3.5 (d, JPC = 23, Cy(P)–CH ), 32.0 (d, J = 12, Cy(P)–CH ), 27.4
2
PC
2
(
6
d, J
.6 (SiMe
P
= 21, Cy(P)–CH
2
), 27.4, 26.6, 26.5 (Cy(N)–CH
2
), 25.2 (THF),
Results and discussion
Synthesis
3
1
1
3
). P{ H} NMR (161 MHz, C ): δ −8.3.
6 6
D
Catalysis (typical procedures)
The synthesis of magnesium compounds incorporating the
6 6
Lactide polymerization. A C D solution of the pre-catalyst N,N′-chelating ligands described in this study was achieved
(0.01 mmol) and 20 equiv. rac-Lactide were combined in an using two approaches. The insertion of carbodiimides into
NMR tube fitted with a J. Young’s tap. NMR spectra were Mg–X bonds (Scheme 1) represents an atom efficient method
2
7–29
acquired after 10 minutes and then at 30 minute intervals.
for the preparation of amidinates (X = alkyl, aryl)
and gua-
2
7
Dimerization of aldehydes. An NMR tube fitted with a nidinates (X = amide). We have used this route to access mag-
J. Young’s tap was charged with 1 mol% of pre-catalyst fol- nesium compounds with the [mesC{NCy} ] amidinate ligand
lowed by the addition of 0.5 mL of a C D solution of 1,4- (mes = 2,4,6-Me C H ), isolating bromide 1 as a pre-cursor to
6 6
3 6 2
−
2
22
dimethoxybenzene (0.11 M). The NMR instrument was locked further derivation. We have also demonstrated that insertion
and shimmed to this sample and an experiment set up to into the Mg–N bonds of Mg(N{SiMe } ) affords the bis(guanidi-
3
2 2
1
i
23
record H NMR spectra (1 scan every 30 seconds) for the dur- nate) compound Mg[(Me Si) NC{N Pr} ] (THF) (2).
3
2
−
2 2
ation of the experiment. 0.1 mL of a C D solution of benz-
Alternatively, the [N∩N] ligand may be introduced to mag-
6
6
aldehyde (4.77 M) was added to the sample in the glovebox. nesium using a protonolysis methodology, employing the
The experiment was started as soon as the sample was in the neutral amidine/guanidine and a suitable organo- or amido-
2
8,30–36
probe (maximum time elapsed after mixing = 2 minutes).
Hydroacetylenation. A solution of the magnesium com- route to prepare compounds incorporating the bicyclic guani-
pound in C (0.5 mL of a 0.0034 M standard solution, dinates derived from 1,3,4,6,7,8-hexahydro-2H-pyrimido[1,2-a]-
.7 μmol) was added to an NMR tube fitted with a J. Young’s pyrimidine (hppH) and 1,4,6-triazabicyclo[3.3.0]oct-4-ene
magnesium reagent (Scheme 2).
We have used this
6 6
D
1
2
1
tap that had been previously charged with a mixture of phenyl (Htbo). However rather than give the expected Mg(N∩N)X(L)n
acetylene (0.022 mL, 0.17 mmol) and N,N′-diisopropylcarbodi- compounds, the solid-state structures showed a preference for
imide (0.028 mL, 0.17 mmol). The NMR tube was heated to these ligands to adopt a bridging coordination mode (3 and 4,
0 °C for 24 h with regular monitoring of the progress of the Fig. 2); a similar coordination of the [hpp]− anion was
37
8
1
catalysis using H NMR spectroscopy. Yields of the propargyla- observed in the tetrameric base-free compound [Mg(hpp)Br]
midine were determined using peaks corresponding to the isolated in low yield by Himmel and co-workers. We have
4
3
0
THF from 1 as an internal standard.
used this approach to furnish the mono- and bis-(chelate) com-
pounds 5 and 9 that incorporate the ethynylamidinate ligand
Crystallography
i
− 23
[
PhCuCC{N Pr}
2
] .
Crystals were covered in inert oil and suitable single crystals
We note that when attempting to isolate heteroleptic com-
were selected under a microscope and mounted on a Bruker pounds of the type Mg(N∩N)X(L) (X = halide), ligand redistri-
n
AXS diffractometer (15) or an Enraf Nonius Kappa CCD diffr- bution can occur affording the bis(chelate) species, Mg-
3
1
actometer ([6]
2 2 2
, [7] , [8] , 12 and 14). Data were collected at (N∩N) (L) with concomitant formation of MgX (L) . Very
2
n
2
n
1
73(2) K using Mo Kα radiation at 0.71073 Å. The structures minor changes to the ligand substituents can influence this
2
6
were refined with SHELXL-97. Additional points are summar- process, illustrated by the contrasting results obtained from
ized below:
Mg(Cy PC{NCy}
the insertion reaction of dicyclohexylcarbodiimide and THF
[
2
2
)Br(Et
2
O)]
2
2
[7] . The coordinated ether
molecule was disordered and was modelled over two positions
with C–C distances restrained to be equal.
Mg(Ph PC{NCy} )(OAr′)(THF) 12. The carbon atoms of the
2 2
THF molecule were disordered over two positions and were
modelled isotropically.
2 2 3 2
Mg(Cy PC{NCy} )(N{SiMe } )(THF) 15. There are two essen-
tially identical molecules in the unit cell. One of the SiMe3
groups is disordered over two positions; the carbon atoms of
the lower occupancy position are modeled isotropically.
Mg(p-tolC{NCy} ) (THF) . One cyclohexyl group and the co-
2
2
2
ordinated THF group are disordered with the two components
required to have the SAME geometry. For the THF solvate it
was impossible to distinguish the position of the oxygen atom
Scheme 1
14304 | Dalton Trans., 2014, 43, 14302–14314
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Scheme 3
Scheme 2
not obtained; subsequent reactions generated this compound
1
in situ. The H NMR of 6 and 8 were consistent with a symme-
trically bound phosphaguanidinate ligand with a single bound
3
1
1
THF molecule. The P{ H} NMR of diphenylphosphaguanidi-
nate derivative 6 shows a singlet at δ −23.6, consistent with
p
this ligand in the N,N’-chelated bonding mode (δ
p
range:
Similarly, the chemical shift for the
Cy P-derivative, 8 (δp −9.9) is consistent with other N,N’-che-
3
8–40,42–47
−17.3 to −23.8).
2
lated ligands incorporating this substitution at phosphorus (δ
Fig. 2 Solid-state structure of 3 and 4, showing bridging coordination range: −7.4 to −11.0).
p
4
4
of the guanidinate ligands.
Metathesis reactions of the Mg(N∩N)X(L) compounds has
n
been successfully performed using MOAr (M = Li, K; Ar =
t
t
2
,6- Bu C H , 2,6- Bu -4-MeC H ) and MN{SiMe } (M = Li, K)
2 6 3 2 6 2 3 2
solutions of MgRBr (R = mes, p-tolyl). In the former case the
to access aryloxide (11 and 12) and bis(trimethylsilyl)amide
derivatives 13–15 (Scheme 4). The H NMR spectrum of each
desired compound Mg(mesC{NCy} )Br(THF) (1) crystallized in
2
1
good yield as the μ,μ-dibromo-bridged dimer. However, a
reduction in bulk at the 2,6-positions of the arylmagnesium
reagent afforded the solvated homoleptic products upon
work-up, with fractional crystallization of Mg(p-tolylC-
compound was consistent with a symmetric N,N’-chelated
3
1
1
ligand, with a single THF at the metal. The P{ H} NMR
chemical shifts for the phosphaguanidinate derivatives vary
according to the phosphorus substituents, as reported above,
with no significant difference observed between the aryloxide
{
NCy} ) (THF) ‡ and MgBr (THF) . A similar ligand redistribu-
2 2 2 2 4
tion was exploited in this work to enable isolation of the
bis(amidinate) species Mg(mesC{NCy} (THF) (10), from the
12 (δ
p
−20.5) and the amide 14 (δ
p
2
−20.7) of the ‘Mg(Ph PC
2 2
)
{
NCy} )(THF)’ fragment. The presence of THF was not indi-
2
mixed amidinate/acetylide species A (Scheme 3).
We have used the protonolysis route to synthesize a series
of phosphaguanidinate magnesium compounds from the reac-
cated in the NMR spectra of the hpp-amide compound Mg
(
hpp)(N{SiMe
3
}
2
) 16, and the corresponding [tbo]− compound
2
4
tion of the neutral species Ph PC{NCy}{NHCy}, and Cy PC
{
(
2
2
i
25
NR′}{NHR′} (R′ = Pr, Cy), with MgMeX (X = Cl, Br) solutions
6–8, Scheme 2). This approach was used rather than the car-
bodiimide insertion route that has successfully been employed
3
8,39
40
by Hill
and Westerhausen to synthesize phosphaguanidi-
nate compounds of Ca, Sr and Ba, primarily due to the lack of
4
1
availability of suitable magnesium phosphanides.
Compounds 6 and 8 were isolated in good yield and charac-
1
13
31
terized by H, C and P NMR spectroscopy. The yield for
compound 7 was not optimized and spectroscopic data were
‡
2 2 2
The composition of the amidinate species Mg(p-tolylC{NCy} ) (THF) was con-
firmed by X-ray diffraction (Fig. S4†) and the cif-file has been uploaded as a part
of this contribution; no additional analytical data were obtained on this
compound.
Scheme 4
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 14302–14314 | 14305

View Article Online
Paper
Dalton Transactions
τ
5
-value of 0.70 and may therefore be considered as a distorted
trigonal bipyramid, with N1 and Br′ defining the apical posi-
tions. This change in geometry arises mainly from an
increased twisting of the plane of the metallacycle relative to
the plane defined by N1–Mg–O, such that the N2–Mg–X angle
is considerably less in the amidinate than in the phosphagua-
nidinate (123.66(8)° and 144.36(6)°, respectively, Fig. 5). The
τ5-values of 0.56 and 0.48 for the two independent magnesium
Fig. 3 Solid-state structure of 16 and 17.
2
centres in propargylamidinate derivative [5] illustrate that this
is a relatively ‘soft’ parameter that can be influenced strongly
by subtle changes in the ligand. Overall it appears that there is
no clear preference for either trigonal bipyramidal or square-
based pyramidal metal geometry in compounds that adopt
this general structure type.
2
1
1
7 showed a guanidinate : amide : THF ratio of 1 : 1 : 1. The
solid-state structures of these compounds showed bridging
ligands, as observed in 3 and 4 (Fig. 3).
Despite a tendency towards asymmetry in the Mg–N bonds
range ΔMgN: 0.025 Å to 0.078 Å), consistent with inequivalent
Structural study
(
X-ray diffraction analysis has been performed on phosphagua- axial and equatorial (or basal and apical) positions for each of
nidinate compounds 6–8, 12, 14 and 15 (Tables 1–5). The the idealized five-coordinate geometries, the C–N bond lengths
results are presented alongside selected examples from our within the metallacycle are indistinguishable (within 3σ). This
previous work.
is consistent with symmetric delocalization of π-electron density
As observed in related amidinate (including 1), and gua- in the heteroallylic unit. The bite angle of the ligand is largely
4
8
22
3
2,49
nidinate
compounds, the Mg(R
2
PC{NR′}
2
)X(L) complexes invariant of the substitution pattern (range 63.93(12)° to 64.76
C metallacycle. The sum of
N∩N] ligands (Fig. 4 for [6] ; Fig. S1 and S2† for [7] and [8] : angles at phosphorus (range ∑angles(P): 309.5° to 311.3°) varies
crystallize as the μ,μ-dihalobridged dimers with chelating (8)°) suggesting a robust MgN
2
−
[
2
2
2
2 2
see ESI†). The magnesium centres are five-coordinate with the little between the Ph P- and Cy P-ligands, and indicates the
remaining site occupied by an O-bound ether molecule (THF availability of P-lone-pair for interaction with additional
5
1
or Et O). The resultant geometries are best described as dis- metals or, in the application of these compounds in catalysis,
2
torted square-based pyramidal in which the apical position is possible interaction with substrate molecules.
5
0
occupied by the ether oxygen atom, with τ -values between
Compound Mg(Ph PC{NCy} )(OAr′)(THF) (12) crystallizes as
2 2
5
0
.30 and 0.40. In contrast, the amidinate structure [1]
2
has a the monomeric compound with a terminal aryloxide ligand
i
Table 1 Crystal structure and refinement data for [Mg(Ph
8]
2
PC{NCy}
2
)Cl(THF)]
2
[6]
2
, [Mg(Cy
2
PC{NCy}
2
)Br(Et
2
O)]
2
[7]
2
and [Mg(Cy
2
PC{N Pr}
2
)Cl(THF)]
2
[
2
[6]
2
[7]
2
[8]
2
Empirical formula
CCDC Number
C
58
H
80
4 2 2 2
N O Mg P Cl
2
C
58
H
108
4
N O
2 2
Mg P
2
Br
2
46 88 4 2 2 2 2
C H N O Mg P Cl
996734
1046.72
173(2)
996735
1163.86
173(2)
996736
910.66
173(2)
M
r
T [K]
Crystal size [mm]
0.25 × 0.20 × 0.15
Triclinic
P ˉ1 (no. 2)
0.20 × 0.20 × 0.20
Triclinic
P ˉ1 (no. 2)
0.15 × 0.10 × 0.05
Monoclinic
P2 /c (no. 14)
1
14.3215(9)
11.5567(7)
15.9545(8)
90
92.786(3)
90
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
β [°]
8.9301(2)
11.0289(3)
16.2637(5)
76.778(2)
84.155(2)
68.586(2)
1451.39(7)
1
9.6349(14)
11.1469(9)
16.5610(19)
88.234(3)
79.663(4)
67.527(11)
1615.5(3)
1
γ [°]
V [Å ]
3
2637.5(3)
2
1.15
Z
D
calc. [mg m−
3
]
1.20
0.23
1.20
1.36
−1
Absorption coefficient [mm
]
0.25
θ Range for data collection [°]
Reflections collected
3.59 to 26.02
21 660
5662 [Rint 0.043]
4710
3.59 to 26.73
19 523
6791 [Rint 0.051]
4905
3.47 to 26.02
17 706
5119 [Rint 0.106]
3115
Independent reflections
Reflections with I > 2σ(I)
Data/restraints/parameters
Final R indices [I > 2σ(I)]
Final R indices (all data)
5662/0/316
6791/112/353
5119/0/262
R
R
1
= 0.042, wR
= 0.054, wR
2
= 0.102
= 0.108
R
R
1
= 0.048, wR
= 0.079, wR
2
2
= 0.101
= 0.112
R
R
1
= 0.072, wR
= 0.134, wR
2
= 0.139
= 0.167
1
2
1
1
2
2
GOOF on F
1.039
0.49 and −0.31
0.998
0.41 and −0.43
1.015
0.42 and −0.28
Largest diff. peak/hole [e Å−
3
]
14306 | Dalton Trans., 2014, 43, 14302–14314
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
23
22
Table 2 Selected bond lengths (Å) and angles (°) for [6]
2
, [7]
2
and [8]
2
presented with the corresponding data from [5]
2
and [1]
2
.
All molecules
2
are located on an inversion centre rendering both magnesium atoms equivalent, except for [5] ; the corresponding parameters for the second mag-
5
nesium centre are presence in italics. The τ -values were calculated according to reference 50
a
a
b
a
b
[1]
2
[5]
2
[6]
2
[7]
2
[8]
2
Mg–N1
Mg–N2
Mg–X
Mg–X′
C1–N1
C1–N2
C1–P
2.123(2)
2.050(2)
2.5532(9)
2.6789(9)
1.331(3)
1.338(3)
—
2.138(2)/2.119(2)
2.060(2)/2.060(2)
2.7202(9)/2.7181(9)
2.5364(9)/2.5406(9)
1.324(4)/1.331(3)
1.333(3)/1.340(3)
—
2.1104(16)
2.0874(16)
2.4112(8)
2.4787(8)
1.330(2)
1.331(2)
1.8905(17)
2.071(2)
2.108(2)
2.5818(10)
2.6882(10)
1.341(3)
1.330(3)
1.898(3)
2.079(3)
2.104(3)
2.4848(17)
2.4078(16)
1.339(4)
1.325(4)
1.889(4)
N1–Mg–N2
N1–Mg–X
N1–Mg–X′
N2–Mg–X
N2–Mg–X′
N1–Mg–O
N2–Mg–O
O–Mg–X
64.76(8)
98.32(7)
165.47(7)
123.66(8)
101.13(7)
98.28(9)
117.09(10)
118.43(7)
91.30(7)
86.55(3)
93.45(3)
64.51(9)/64.74(9)
161.45(7)/161.33(7)
103.20(8)/102.08(7)
96.94(7)/96.78(7)
127.94(8)/132.43(8)
98.84(9)/99.84(0)
121.64(10)/118.01(10)
90.50(7)/91.11(7)
110.00(8)/109.09(7)
88.40(3)/88.36(3)
91.55(3)/91.69(3)
64.16(6)
102.24(5)
162.56(5)
144.36(6)
99.78(5)
100.37(7)
112.86(7)
101.71(6)
92.07(6)
86.97(3)
93.03(3)
64.27(9)
137.56(8)
99.03(7)
100.53(7)
161.48(7)
113.43(10)
101.19(10)
108.25(8)
92.71(7)
63.93(12)
98.59(10)
140.03(11)
161.20(11)
101.55(10)
117.09(13)
102.62(13)
91.36(10)
102.10(10)
87.49(5)
O–Mg–X′
X–Mg–X′
Mg–X–Mg′
86.46(3)
93.54(3)
92.51(5)
τ
5
0.70
0.56/0.48
0.30
0.40
0.35
∑angles(P)
—
—
311.3
309.5
310.9
a
b
X = Br. X = Cl.
Table 3 Crystal structure and refinement data for Mg(Ph
N{SiMe )(THF) 15
2
PC{NCy}
2
)(OAr’)(THF) 12, Mg(Ph
2
PC{NCy}
2
)(N{SiMe
3
}
2
2 2
)(THF) 14 and Mg(Ph PC{NCy} )-
(
3 2
}
12
14
15
Empirical formula
CCDC Number
C
44
H
63
2 2
N O MgP
C
35
H
58
3 2
N OMgSi P
35 70 3 2
C H N OMgSi P
996737
707.24
173(2)
996738
648.30
173(2)
996739
660.40
173(2)
M
r
T [K]
Crystal size [mm]
0.30 × 0.30 × 0.25
Monoclinic
0.25 × 0.25 × 0.20
Triclinic
P ˉ1 (no. 2)
0.28 × 0.22 × 0.16
Triclinic
P ˉ1 (no. 2)
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
β [°]
P2
1
/c (no. 14)
9.5663(2)
23.8131(5)
18.3412(3)
90
95.552(1)
90
4158.58(14)
4
1.13
10.1097(2)
10.4287(2)
19.5149(4)
89.591(1)
79.403(1)
72.968(1)
1931.30(7)
2
14.2018(3)
16.0675(3)
18.2651(3)
87.827(1)
80.418(1)
85.172(1)
4093.92(13)
4
γ [°]
V [Å ]
3
Z
D
calc. [mg m−
3
]
1.12
0.18
1.07
0.17
−1
Absorption coefficient [mm
]
0.12
θ Range for data collection [°]
Reflections collected
3.46 to 26.22
27 354
8147 [Rint 0.059]
5995
3.47 to 25.95
31 193
7494 [Rint 0.040]
6212
2.92 to 27.48
79 546
18 752 [Rint 0.061]
13 481
Independent reflections
Reflections with I > 2σ(I)
Data/restraints/parameters
Final R indices [I > 2σ(I)]
Final R indices (all data)
8147/0/446
7494/0/388
18 752/3/788
R
R
1
= 0.061, wR
= 0.087, wR
2
= 0.145
= 0.162
R
R
1
= 0.038, wR
= 0.051, wR
2
2
= 0.089
= 0.095
R
R
1
= 0.060, wR
= 0.091, wR
2
= 0.143
= 0.157
1
2
1
1
2
2
GOOF on F
1.020
0.44 and −0.44
1.016
0.29 and −0.23
1.059
0.84 and −0.94
Largest diff. peak/hole [e Å−
3
]
(
Tables 3 and 4, Fig. 6); a similar structure was recorded for more symmetrically than in the [Mg(N∩N)X(L)]
2
compounds
2
2
52
amidinate 11. The τ -values of 0.67 and 0.63 for 11 and 12, (max. ΔMgN 0.023 Å) with equal distribution of π-density across
4
respectively indicate a geometry closer to tetrahedral than the –NCN– fragment. The terminal aryloxide ligand
square planar, despite the small bite angle of the ligands (11 approaches a linear geometry at oxygen in compound 11
6
5.89(9)°, 12 65.12(8)°). The N,N’-chelating ligands bond to Mg (163.38(18)°) but is considerably more acute in 12 (155.95(16)°)
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 14302–14314 | 14307

View Article Online
Paper
Dalton Transactions
Table 4 Selected bond lengths (Å) and angles (°) for 12 presented with
22
the corresponding data from 11.
The
τ
4
-values were calculated
according to ref. 52
11
12
Mg–N1
Mg–N2
Mg–O1
Mg–O2
C1–N1
C1–N2
C1–P
2.067(2)
2.054(2)
1.8431(19)
2.005(2)
1.328(3)
1.329(3)
—
2.062(2)
2.085(2)
1.8463(17)
2.004(2)
1.328(3)
1.331(3)
1.892(2)
N1–Mg–N2
N1–Mg–O1
N1–Mg–O2
N2–Mg–O1
N2–Mg–O2
O1–Mg–O2
65.89(9)
65.12(8)
134.82(10)
106.29(9)
131.25(9)
111.31(9)
103.07(9)
163.38(18)
131.44(9)
106.76(9)
139.14(8)
103.39(9)
104.58(8)
155.95(16)
Fig. 4 Thermal ellipsoid plot (30% probability) of [Mg(Ph PC{NCy} )Cl-
2
2
2 2
(THF)] [6] (hydrogen atoms omitted).
a
Mg–O1–Cx
τ
4
0.67
0.63
∑angles(P)
—
312.0
a
11, x = 23; 12, x = 26.
Table 5 Selected bond lengths (Å) and angles (°) for 14 and 15 pre-
22
4
sented with the corresponding data from 13. The τ -values were cal-
culated according to ref. 52
a
1
3
14
15
Mg–N1
Mg–N2
Mg–N3
Mg–O
C1–N1
C1–N2
C1–P
2.077(2)
2.081(2)
1.997(2)
2.0205(18)
1.336(3)
1.333(3)
—
2.0844(14)
2.0668(14)
1.9891(14)
2.0113(12)
1.331(2)
1.330(2)
1.8910(16)
2.084(2)/2.0681(19)
2.0611(19)/2.059(2)
1.992(2)/1.992(2)
2.0295(17)/2.0365(17)
1.337(3)/1.334(3)
1.335(3)/1.340(3)
1.895(2)/1.891(2)
Fig. 5 Schematic projections of the magnesium environments of [6]
left) and [1] (right).
2
(
2
N1–Mg–N2
N1–Mg–N3
N1–Mg–O
N2–Mg–N3
N2–Mg–O
N3–Mg–O
Mg–N3–Si1
Mg–N3–Si2
Si1–N3–Si2
65.54(7)
65.35(5)
65.20(7)/65.43(7)
134.26(8)
108.30(8)
130.83(8)
106.34(8)
105.60(8)
115.41(10)
117.05(10)
127.24(12)
132.06(6)
103.33(6)
130.06(6)
112.33(5)
107.24(6)
117.71(7)
115.89(8)
126.27(8)
134.88(8)/135.98(8)
101.08(8)/99.93(7)
127.07(8)/129.91(9)
108.61(8)/110.08(8)
111.38(8)/108.16(8)
118.07(11)/116.95(11)
116.25(10)/117.35(11)
125.41(12)/125.45(12)
τ
4
0.67
0.69
0.70/0.67
∑angles(P)
—
313.6
311.7/311.5
a
Corresponding values for the second molecule in the unit cell
presented in italics.
Fig. 6 Thermal ellipsoid plot (30% probability) of Mg(Ph
2
PC{NCy}
2
)-
(OAr’)(THF) 12 (hydrogen atoms omitted).
despite no apparent adverse steric contacts. Both values fall
within the rather large range for aryloxide ligands of this type
5
3,54
52
(
146.0(4)°–170.99(9)°).
The bis(trimethylsilyl)amide compounds Mg(R
N{SiMe
)(THF) (14 R = Ph, 15 R = Cy) both crystallize in the suggesting the bulk of the aryl- and trimethylsilyl-groups is
for 14: see ESI†). The τ -values (range 0.67 to 0.70) are
4
PC{NCy}
2
)- similar to those calculated for the aryloxide compounds above,
2
(
3 2
}
P ˉ1 space group (Tables 3 and 5) but are not isostructural; there sufficiently removed from the metal environment that little
are two independent molecules present in the unit cell of 15. influence is exerted on the metal geometry. The chelating
The gross structural features are the same as noted for the ligands are essentially symmetrically bound (max. ΔMgN
2
2
−
amidinate compound 13, with a chelating [N∩N] ligand, 0.023 Å), with equal delocalization in the amidine component.
terminal amide and coordinated THF (Fig. 7 for 15; Fig. S3† The Mg–Namide distance (range 1.9891(14) Å to 1.992(2) Å) are
14308 | Dalton Trans., 2014, 43, 14302–14314
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 7 Thermal ellipsoid plot (30% probability) of one of the indepen-
dent molecules of Mg(Cy
atoms omitted).
2 2 3 2
PC{NCy} )(N{SiMe } )(THF) 15 (hydrogen
typical for terminal amides of this type, and the planar geome-
try at the nitrogen atoms is as expected.
2 n
Fig. 8 Different structural types reported for Mg(N∩N) (L) .
Although the geometric variations are generally small once
a specific structural type has been adopted, the ligand substi-
tuents play an important role in determining the gross struc-
important role, with significant differences between ‘flat’ aro-
matic substituents and more ‘spherical’ alkyl/silyl groups.
The bis(guanidinate) compounds VIII and IX are five-
coordinate (type 8d). Ignoring electronic contributions to the
bonding, this increase in coordination number correlates with
3
1,55,56
ture of the compound.
structural types observed for the bis(chelate) compounds, Mg-
This is illustrated for five
3
1,33,57
(
N∩N) (L) (Fig. 8). Amidinate compounds I–III (Table 6)
2
n
represent examples of non-solvated, four coordinate species
Fig. 8a). Each of these compounds was prepared in the pres-
ence of THF or Et O, whereas similar compounds IV and
the nitrogen of the –NR″ group removing the R″ substituents
2
(
from the metal environment. This is supported by the similar
α values, and is accentuated by the planar geometry of this
nitrogen, which permits rotation such that the substituents
are perpendicular to the metallacycle (VIII 77.2(5)°; IX 79.7(1)°
2
2
8,34
V,
were not exposed to donor solvents during their prepa-
ration and the structures may therefore not represent the pre-
ferred coordination number. Variations of the base-free
60
and 80.7(1)°). The six-coordinate compounds XII to XVII
‘
Mg(N∩N) ’ unit are seen for bimetallic VI and VII (Fig. 8b
2
(type 8e) and distorted octahedral with trans donor solvent
molecules. This enables the metallacycles to be co-planar,
maximizing the distance between R′-substituents. The angle θ
5
7
and c). Five-coordinate magnesium centres (Fig. 8d) are
adopted for guanidinate (VIII and IX)
(
(
formamidinate compounds XVI and XVII, where R = H.
Examination of the data in Table 6 shows the C–N–R′ angle
2
3,27
and amidinate
2
3,58
X and XI)
Fig. 8e) are restricted to amidinates XII–XV,
species. In contrast, bis(solvated) compounds
is generally small, although the presence of SiMe groups in
3
2
3,27,59
including
XIII causes the ligands to twist relative to each other (θ =
3
5
15.7(2)°).
We conclude, that there is no clear correlation between sub-
(
α) is largely invariant of structure type and ligand substitution
stituent pattern and structure type for this series of com-
pattern, with most ligands giving values in the range 120°–
pounds, or for the [N∩N]− ligands in general. This
t
1
(
25°. Exceptions occur when R or R′ = Bu (III and IV) or SiMe
X and XIII) for which values tend towards 130°, and for the
formamidinate ligands in XVI and XVII (values generally
120°). The latter correlates well with the reduced size of the
3
unpredictability can be summarized by comparing compounds
t
i
III and XII where changing from R′ = Bu to R′ = Pr causes an
increase in the coordination number from four (type 8a) to six
<
(
type 8e).
R-group and the adoption of a six-coordinate structure.
Despite the similar α angles, large variations in the angle
between the two metallacycles (θ) exist for compounds with
Catalysis
the same structure type. For example, there is only a slight Selected examples of compounds described in this work have
increase in α when the methyl substituent in I is replaced with been examined as (pre)-catalysts for a variety of bond-forming
p-tolyl in II, but a large decrease in θ, and a τ
sistent with an approximately square-planar metal. Further,
4
-value for II con- reactions.
Lactide polymerization. Magnesium compounds have an
the interplay between these parameters is not a direct corre- established history in promoting the ring-opening polymeriz-
lation, as larger α values in compounds III and IV give greater ation (ROP) of lactide to give poly-lactic acid (PLA). A series of
θ angles and τ
4
-values more fitting for a distorted tetrahedron. Mg(N∩N)X(L) compounds have been developed based on the
n
−
The ability of the substituents to interdigitate may play an β-diketiminate ligand (BDI) system, [HC{C(Me)NAr} ] , with
2
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 14302–14314 | 14309

View Article Online
Paper
Dalton Transactions
Table 6 Collection of structural parameters for bis(chelate) compounds, Mg(N∩N)
2
(L)
n
(see text for references). Structure refers to Fig. 8, α is
defined as the C–N–R’ angle, θ is the inter-planar angle between the two metallacycles (normalized to 0° < θ < 90°), τ-values are calculated accord-
4
5
ing to ref. 52 (τ -) and 50 (τ -)
Coordn no.
Structure
R
R′/R″
L
α/°
θ/°
τ-value
a
b
4
4
4
8a
8a
8a
8a
8a
8b
I
II
III
IV
V
Me c
p-tol
Ph
Bu
2,6(mes)
Me
Ar
—
—
—
—
—
—
121.2(4) to 122.5(4)
123.8(2), 123.9(2)
129.5(2), 130.0(2)
129.7(3), 129.9(3)
128.6(2), 131.2(2)g
121.4(2), 122.2(2)
54.9(2), 54.1(2)
13.3(2)
89.8(2)
80.3(2)
88.2(1)
—
τ
τ
τ
τ
τ
4
4
4
4
4
= 0.40, 0.41
= 0.10
= 0.58
= 0.57
= 0.60
a
Ar
t
Bu
mes
Pr
d
t
e
d
d
(
(
4)
4)
d
i
C H
2 6 3
f
4
VI
Et
—
t
h
Bu
Pr
124.9(2), 125.5(2)
121.8(2)
f
i
4
8c
VII
Me
—
—
—
i
i
5
5
5
8d
8d
8d
VIII
IX
X
N Pr
2
Pr
Pr
THF
THF
121.8(7), 122.1(7)
47.1(5)
44.7(1)
71.3(2)
τ
τ
τ
5
5
5
= 0.57
= 0.50
= 0.74
i
b
N{SiMe
Ph
3
}
2
122.4(2) to 124.5(2)
i
Cy
SiMe
Cy
2
Et O
123.6(3)
j
3
131.0(3)
5
8d
XI
Mes
THF
120.2(1), 121.8(1)
38.9(1)
τ
5
= 0.41
i
6
6
6
6
6
6
8e
8e
8e
8e
8e
8e
XII
Ph
Ph
CuCPh
p-tol
H
H
Pr
THF
THF
THF
THF
THF
THF
120.6(7), 120.7(6)
127.9(3), 130.6(2)
121.0(1), 121.1(1)
121.7(3), 121.8(2)
119.0(2) to 122.9(2)
117.4(2), 118.5(2)
7.4(5)
15.7(2)
0
—
—
—
—
—
—
XIII
XIV
XV
XVI
XVII
SiMe
3
i
Pr
c
Cy
p-tol
o-tol
0
c
b
2.1(1)
0.2(3)
k
a
i
b
c
d
e
f
Ar = 2,6- Pr
2 6 3
C H .
Values given as a range. 4-MeC
6 4
H .
Prepared in the absence of coordinating solvent. 2,4,6-Me
3 6
C
H
2
. α values reported
g
h
t
i
j
k
3 6 4
for chelating ligands only. Angle to Et-substituent. Angle to Bu-substituent. Angle to Cy-substituent. Angle to SiMe -substituent. 2-MeC H .
6
1–63
63–66
64,67–69
X = alkyl,
examined selected compounds to assess their behaviour as polymerization of lactide, their development as useful reagents
ROP catalysts. is limited, with any control of polymerization imparted by the
amide,
or alkoxide.
Accordingly we
Whilst these compounds demonstrate good activity for the
Phosphaguanidinate compounds 6, 8, 12 and 14 were N,N-chelate lost on conversion to the active species. In
initially screened for activity on an NMR scale with 20 equi- addition, the neutral compound hppH is an excellent organo-
valents of rac-lactide. The chlorides 6 and 8 were inactive at catalyst for lactide polymerization (1 mol% hppH gives 99%
7
1
room temperature and 70 °C, consistent with the requirement conversion of L-lactide in 20 seconds), and the observed
of a reactive Mg–X bond (where X = alkyl, amide, alkoxide) to polymerization activity of 16 may therefore be due to the pres-
initiate polymerization. The aryloxide 12 did not show appreci- ence of liberated guanidine.
able activity at room temperature, but was active at 70 °C,
Dimerization of aldehydes (the Tishchenko reaction). In
giving a conversion of <95% after 6 h. This is consistent with a 2007 Hill and co-workers rekindled research in the homo-
slow initiation due to the large aromatic substituent of the coupling of aldehydes promoted by main group elements
7
2
aryloxide.
other than aluminium. They examined the catalytic activity
The amide 14 was considerably more active, with <95% con- of amide compounds of Ca, Sr and Ba, initially supported by
1
version after 30 min at room temperature. The H NMR the BDI ligand. Subsequent work with main group elements
7
3,74
73,75,76
73
spectra of this reaction showed non-coordinated THF, and a includes compounds of Li,
Na,
K,
and most
7
7
shift of SiMe resonances to −0.03 ppm, which is inconsistent recently Zn. Magnesium reagents have been used in the pres-
and selenide anions to promote this
the P{ H} NMR spectra showed that polymerization initiated coupling reaction. We have previously communicated our work
3
1
8,19,78
79
with the liberation of HN{SiMe } (δ 0.09 in C D ). However, ence of thiolate
3
2
H
6 6
3
1
1
2
2
by 12 and 14 proceed with loss of the phosphaguanidine, in this area with amidinate compounds 11 and 13, and gua-
2
1
Ph PC{NCy}{NHCy}, identified by a characteristic singlet at δ
nidinates 16 and 17. We report here the catalytic perform-
ance of the phosphaguanidinate amides 14 and 15.
2
P
2
4
−16.9.
Amidinate- and guanidinate-amides (13 and 16) were also
The esterification of benzaldehyde was examined on an
tested. Both compounds promoted the NMR scale polymeriz- NMR scale in C D using 1 mol% of 14 and 15 with 1,4-
6
6
ation at room temperature, with <95% conversion after 2 h dimethoxybenzene as an internal standard. During the initial
13) and 5 minutes (16). The rapid polymerization promoted 10 minutes, compounds 14 and 15 produce 60% and 46%
by the hpp-derivative 16 is consistent with the lack of a coordi- yields of benzylbenzoate, respectively (Fig. 9), corresponding to
(
nating solvent that would compete with the monomer for turnover frequencies (TOFs) of 360 h− (14) and 276 h (15).
1
−1
7
0
access to the metal. However, in both cases the non-co- This represents a marked increase in the rate of reaction over
1
ordinated ligand was identified in solution by H NMR spec- the same time period when compared with 13 (yield 37%, TOF
troscopy, consistent with rapid loss of the N,N’-chelate.
∼ 220 h− ) and 16/17 (yields ∼20%, TOFs ∼ 120 h ). The rate
1
−1
14310 | Dalton Trans., 2014, 43, 14302–14314
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
2
Scheme 5 [Mg] = ‘Mg(mesC{NCy} )’, solvent omitted.
Fig. 9 Graphical representation of the activity of magnesium amide and
aryloxide compounds for the catalytic dimerization of benzaldehyde.
We reported the hydroacetylenation of carbodiimides using
2
3
of ester production decreases after this time [30 minutes: 70% compound 13, with an NMR yield of 72% after 24 h at 80 °C.
(
14), 60% (15); 60 minutes: 74% (14), 65% (15)], likely due to Subsequent investigations using bis(chelate) compounds 2,
the reduced concentration of substrate in the closed system.
9 and 10 showed similar activities (range 72%–74%) demon-
3
1
1
In contrast to the lactide polymerization catalysis, P{ H} strating that the amide was not a requirement for catalysis.
NMR analysis of the reaction mixtures showed no liberated Indeed, catalytic turnover was also observed for organometallic
phosphaguanidine, suggesting that the ligand remains associ- and organoamide compounds of magnesium, including
ated with the magnesium during the catalysis. The GCMS trace Grignard reagents RMgX. This demonstrated multiple entry
of a hydrolysed reaction between 15 with two equivalents of points into the accepted catalytic cycle (Scheme 5), prompting
benzaldehyde shows the expected organic species PhC(O)H, further investigation through stoichiometric reactions.
PhCH OH, PhC(O)N{SiMe } and PhC(O)OCH Ph, in addition
The procedure for NMR scale catalysis involves the addition
2
3 2
2
to the silyl exchange products PhCH OSiMe and PhC(O) of the magnesium reagent to a pre-mixed, equimolar solution
2
3
2
1
NHSiMe
3
.
2
Peaks for PCy H and CyNvCvNCy are also of the alkyne and the carbodiimide. For 13, it is assumed the
observed, but believed to be due to the facile cleavage of the reaction proceeds via protonolysis of the amide and formation
8
0
P–C bond during sample preparation.
of the terminal acetylide (Step 1, Scheme 5). However, inser-
Hydroacetylenation of carbodiimides. Ethynylamidines, tion of carbodiimide into the Mg–N bond to form a guanidi-
RCuCC{NR′}{NHR′}, are potentially valuable building blocks nate ligand is a viable alternative pathway (see Scheme 1).
to organic compounds containing more complex heterocycles A competition reaction performed using a 1 : 1 : 1 ratio of [13]:
8
1–83
i
i
including isoxazoles and pyrazoles.
Although the prepa- PrNvCvN Pr : PhCuCH in C D , showed that both pathways
6 6
1
ration of the metal-bound amidinates via the insertion of a car- operate under these conditions. The H NMR spectrum
bodiimide into a terminal metal–acetylide bond is facile, showed HN{SiMe } (from Step 1) and {Me Si} NC{N Pr}-
hydrolysis to liberate the free amidine is not viable due to the {NH Pr} (I, Scheme 6), which originates from carbodiimide
sensitivity of the products to these conditions. This has led insertion to form amidinate/guanidinate species B prior to
i
3
2
3
2
i
8
4
to research into the catalytic production of these unsaturated protonation by phenylacetylene. The relative ratio of
8
5
1
molecules, with most work to date focussed on the lantha- HN{SiMe } : I, calculated from H NMR integrals of the SiMe
3
3
2
8
4,86,87
nide elements.
The use of main group compounds as catalysts was first
reported in 2006 using commercially available LiN{SiMe } .
3 2
Subsequently, Hill and co-workers studied the stoichiometric To examine this, 1 equiv. of PrNvCvN Pr was added to A
insertion of carbodiimides into Ca–C bonds of BDI-supported (generated in situ), to form the heteroleptic bis(amidinate)
acetylide compounds. Despite observing catalytic turnover at species C containing the [PhCuCC{N Pr} ] ligand. Sub-
peaks, was 9 : 1 showing that Step 1 dominates the reactivity.
The insertion of the carbodiimide into the Mg–C bond
(Step 2, Scheme 5) is generally considered an irreversible step.
8
8
i
i
8
9
i
−
2
8
0 °C in toluene, loss of the BDI-ligand was detected by NMR sequent addition of 1 equiv. CyNvCvNCy did not show
spectroscopy. Building on these results, Hill’s team has formation the corresponding N,N-dicyclohexyl-2-phenylethyny-
reported catalytic hydroacetylenation of carbodiimides using lamidinate compound. Given the similar reactivity of the two
2
3
the solvated homoleptic amides, M(N{SiMe } ) (THF) (M = carbodiimides during catalysis, this suggests that ‘de-inser-
3
2 2
2
9
0
Mg, Ca, Sr). These simple catalysts show high activities and tion’ of N,N’-diisopropylcarbodiimide does not occur to an
demonstrate improved substrate scope by extending the coup- appreciable extent. Although previous studies showed that C
ling to include bulky carbodiimide derivatives (e.g. R′ = undergoes ligand re-distribution to form the homoleptic
i
2
,6- Pr
2
C
6
H
3
).
bis(amidinate)s 9 and 10, the lack of CyNvCvNCy incorpor-
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 14302–14314 | 14311

View Article Online
Paper
Dalton Transactions
Hill reports that catalysis initiated by the group 2 bis
amide) compounds is subject to a significant product inhi-
(
bition effect. Kinetic analysis of the Sr case suggested that this
proceeded by competitive protonolysis of the metal amidinate
9
0
by the amidine product rather than the phenylacetylene. To
investigate whether this process was operating in our system,
the stoichiometric reaction between Mg(CuCPh)
2
(THF)
ethynylamidine PhCuCC{N Pr}{NH Pr}, was performed
Scheme 7). Quantitative formation of the bis(amidinate) 9
4
and
i
i
(
with formation of phenylacetylene indicates that the terminal
magnesium acetylide ligand is susceptible to protonolysis by
the amidine product, and suggests that Step 3 may be
reversible.
Conclusions
We have demonstrated that amidinate, guanidinate and phos-
phaguanidinate compounds of magnesium are accessible
through standard synthetic procedures. The insertion of carbo-
diimides into Mg–C and Mg–N bonds is an efficient route to
amidinate and guanidinate compounds, respectively, whereas
protonation of organomagnesium groups was the preferred
route to phosphaguanidinate compounds. The substitution
pattern of the ligand may influence the constitution of the
species isolated from the reaction, with ligand redistribution a
possibility in solution. Surveying a range of compounds from
the literature failed to identify a single structural parameter
that correlated with the type of solid-state structure adopted,
although the ability of the ligand substituents to pack efficien-
tly around the small magnesium centre is believed to play an
important role.
Scheme 6
ated into the system is good evidence that formation of the
amidinate ligands is essentially an irreversible process.
The relative protolytic stability of the amidinate and guani-
dinate ligands generated during the catalysis initiated by 13 is
important in determining the mechanism. Experiments show
that the amidinate ligand, [mesC{NCy}
to protonolysis when compared with [{Me Si} NC{N Pr} ] and
2
PhCuCC{N Pr} ] . For example, no reaction was observed
upon heating NMR samples bis(amidinate)s 9 or 10 to 80 °C in
the presence of 5 equivalents of phenylacetylene, whereas the
bis(guanidinate) 2 quantitatively liberates guanidine I under
these conditions. This contradicts the observation that bis
−
2
] , is relatively resistant
i
−
3
2
2
i
−
[
Selected compounds show good activity in a range of bond-
forming catalytic reactions. However, the extent to which the
integrity of the metal reagent is maintained during catalysis
has been questioned. In the case of the ring-opening polymer-
ization of lactide, the loss of the ligand from magnesium was
confirmed and so it was concluded that these species were of
limited use.
Although loss of the ligand from the metal was not
observed for the dimerization of aldehydes, the factors that
control the activity remain unclear. Thus the reason for the
increased activity observed for the phosphaguanidinate com-
pounds is not well understood, but must result from the pres-
(
amidinate) 10 is active in hydroacetylenation and highlights
the limitations of comparing the results of stoichiometric reac-
tions with the complex equilibria present during a catalytic
9
0
experiment. As noted for Hill’s strontium system, there was a
negligible kinetic isotope effect when the catalysis was per-
formed with deuterated phenylacetylene (Fig. S5†) suggesting
that the protonation of the amidinate by phenylacetylene is
not involved in the rate determining step.
2
ence of the terminal –PR groups. Finally, stoichiometric
studies have shown that the hydroacetylenation catalysis is
dominated by the presence of multiple equilibria in solution
and that, provided a polar organomagnesium or magnesium-
amide bond is present at some point during the catalytic cycle,
access to active species is achieved with little loss to activity.
So, although the ligands presented in this study have enabled
some interesting chemistry to be explored, we regretfully con-
clude that their future development in the areas of catalysis
presented in this contribution is, at best questionable.
Pathway 1 in Scheme 8 may be playing a more dominant role
in magnesium-based catalysis than we first assumed.
Scheme 7
14312 | Dalton Trans., 2014, 43, 14302–14314
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
1
6 H.-J. Chuang, H.-L. Chen, J.-L. Ye, Z.-Y. Chen, P.-L. Huang,
T.-T. Liao, T.-E. Tsai and C.-C. Lin, J. Polym. Sci., Part A:
Polym. Chem., 2013, 51, 696–707.
1
1
1
7 Y. Wang, W. Zhao, X. Liu, D. Cui and E. Y.-X. Chen, Macro-
molecules, 2012, 45, 6957–6965.
8 S. P. Curran and S. J. Connon, Angew. Chem., Int. Ed., 2012,
51, 10866–10870.
9 L. Cronin, F. Manoni, C. J. O’Connor and S. J. Connon,
Angew. Chem., Int. Ed., 2010, 49, 3045–3048.
2
2
0 W. Tischtschenko, Chem. Zentralbl., 1906, 77, 1309.
1 B. M. Day, N. E. Mansfield, M. P. Coles and P. B. Hitchcock,
Chem. Commun., 2011, 47, 4995–4997.
2
2
2
2 B. M. Day, W. Knowelden and M. P. Coles, Dalton Trans.,
2
012, 41, 10930–10933.
3 R. J. Schwamm and M. P. Coles, Organometallics, 2013, 32,
277–5280.
4 J. Grundy, M. P. Coles and P. B. Hitchcock, Dalton Trans.,
003, 2573–2577.
Scheme 8
5
Acknowledgements
2
Victoria Master’s by thesis Scholarship (RJS).
25 N. E. Mansfield, J. Grundy, M. P. Coles, A. G. Avent and
P. B. Hitchcock, J. Am. Chem. Soc., 2006, 128, 13879–13893.
2
2
6 G. M. Sheldrick, SHELXL-97, Program for the Refinement of
Crystal Structures, Göttingen, 1997.
7 B. Srinivas, C.-C. Chang, C.-H. Chen, M. Y. Chiang,
I.-T. Chen, Y. Wang and G.-H. Lee, J. Chem. Soc., Dalton
Trans., 1997, 957–963.
Notes and references
1
M. Arrowsmith and M. S. Hill, in Comprehensive Inorganic
Chemistry II, ed. T. Chivers, Elsevier, 2013, vol. 1, pp. 28 J. A. R. Schmidt and J. Arnold, J. Chem. Soc., Dalton Trans.,
189–1216. 2002, 2890–2899.
M. R. Crimmin and M. S. Hill, in Topics in Organometallic 29 M.-D. Li, C.-C. Chang, Y. Wang and G.-H. Lee, Organometal-
Chemistry, ed. S. Harder, 2013, pp. 191–241. lics, 1996, 15, 2571–2574.
A. G. M. Barrett, M. R. Crimmin, M. S. Hill and 30 O. Ciobanu, A. Fuchs, M. Reinmuth, A. Lebkücher,
1
2
3
P. A. Procopiou, Proc. R. Soc. London, Ser. A, 2010, 466, 927–
63.
E. Kaifer, H. Wadepohl and H.-J. Himmel, Z. Anorg. Allg.
Chem., 2010, 636, 543–550.
9
4
5
S. Harder, Chem. Rev., 2010, 110, 3852–3876.
X. Zhang, T. J. Emge and K. C. Hultzsch, Angew. Chem., Int.
Ed., 2012, 51, 394–398.
31 N. Nimitsiriwat, V. C. Gibson, E. L. Marshall,
P. Takolpuckdee, A. K. Tomov, A. J. P. White,
D. J. Williams, M. R. J. Elsegood and S. H. Dale, Inorg.
Chem., 2007, 46, 9988–9997.
6
M. Arrowsmith, M. R. Crimmin, A. G. M. Barrett, M. S. Hill,
G. Kociok-Köhn and P. A. Procopiou, Organometallics, 2011, 32 S. P. Green, C. Jones and A. Stasch, Science, 2007, 318,
0, 1493–1506. 1754–1757.
S. R. Neal, A. Ellern and A. D. Sadow, J. Organomet. Chem., 33 R. T. Boeré, M. L. Cole and P. C. Junk, New J. Chem., 2005,
011, 696, 228–234. 29, 128–134.
J. F. Dunne, D. B. Fulton, A. Ellern and A. D. Sadow, J. Am. 34 A. Xia, H. M. El-Kaderi, M. J. Heeg and C. H. Winter,
Chem. Soc., 2010, 132, 17680–17683. J. Organomet. Chem., 2003, 682, 224–232.
M. R. Crimmin, M. Arrowsmith, A. G. M. Barrett, 35 M. L. Cole, D. J. Evans, P. C. Junk and L. M. Louis, New
3
7
8
9
2
I. J. Casely, M. S. Hill and P. A. Procopiou, J. Am. Chem.
Soc., 2009, 131, 9670–9685.
0 D. Mukherjee, A. Ellern and A. D. Sadow, Chem. Sci., 2014,
J. Chem., 2002, 26, 1015–1024.
36 F. A. Cotton, S. C. Haefner, J. H. Matonic, X. Wang and
C. A. Murillo, Polyhedron, 1997, 16, 541–550.
37 M. P. Coles, Chem. Commun., 2009, 3659–3676.
1
1
1
1
1
1
5, 959–964.
1 M. Arrowsmith, M. S. Hill and G. Kociok-Köhn, Chem. – 38 M. R. Crimmin, A. G. M. Barrett, M. S. Hill,
Eur. J., 2013, 19, 2776–2783.
P. B. Hitchcock and P. A. Procopiou, Organometallics,
2008, 27, 497–499.
39 A. G. M. Barrett, M. R. Crimmin, M. S. Hill,
P. B. Hitchcock, S. L. Lomas, M. F. Mahon and
P. A. Procopiou, Dalton Trans., 2010, 39, 7393–7400.
2 M. Arrowsmith, T. J. Hadlington, M. S. Hill and G. Kociok-
Köhn, Chem. Commun., 2012, 48, 4567–4569.
3 M. Arrowsmith, M. S. Hill, T. J. Hadlington and G. Kociok-
Köhn, Organometallics, 2011, 30, 5556–5559.
4 J. F. Dunne, S. R. Neal, J. Engelkemier, A. Ellern and 40 T. M. A. Al-Shboul, G. Volland, H. Görls and
A. D. Sadow, J. Am. Chem. Soc., 2011, 133, 16782–16785.
5 W. Yi and H. Ma, Dalton Trans., 2014, 43, 5200–5210.
M. Westerhausen, Z. Anorg. Allg. Chem., 2009, 635, 1568–
1572.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 14302–14314 | 14313

View Article Online
Paper
Dalton Transactions
4
4
4
1 B. M. Day and M. P. Coles, Eur. J. Inorg. Chem., 2010, 5471– 66 A. P. Dove, V. C. Gibson, E. L. Marshall, A. J. P. White and
477. D. J. Williams, Dalton Trans., 2004, 570–578.
2 A. C. Behrle and J. A. R. Schmidt, Organometallics, 2013, 32, 67 M. H. Chisholm, J. C. Huffman and K. Phomphrai,
141–1149. J. Chem. Soc., Dalton Trans., 2001, 222–224.
3 A. J. Roering, S. E. Leshinski, S. M. Chan, T. Shalumova, 68 B. M. Chamberlain, M. Cheng, D. R. Moore, T. M. Ovitt,
5
1
S. N. MacMillan, J. M. Tanski and R. Waterman, Organo-
metallics, 2010, 29, 2557–2565.
E. B. Lobovsky and G. W. Coates, J. Am. Chem. Soc., 2001,
123, 3229–3238.
4
4
4
4
4
4
5
4 N. E. Mansfield, J. Grundy, M. P. Coles and P. B. Hitchcock, 69 F. Drouin, T. J. J. Whitehorne and F. Schaper, Dalton Trans.,
Polyhedron, 2010, 29, 2481–2488. 2011, 40, 1396–1400.
5 W.-X. Zhang, M. Nishiura, T. Mashikp and Z. Hou, Chem. – 70 M. P. Coles and P. B. Hitchcock, Eur. J. Inorg. Chem., 2004,
Eur. J., 2008, 14, 2167–2179. 2662–2672.
6 N. E. Mansfield, M. P. Coles and P. B. Hitchcock, Dalton 71 R. C. Pratt, B. G. G. Lohmeijer, D. A. Long,
Trans., 2005, 2833–2841.
7 M. P. Coles and P. B. Hitchcock, Chem. Commun., 2002,
R. M. Waymouth and J. L. Hedrick, J. Am. Chem. Soc., 2006,
128, 4556–4557.
2
794–2795.
72 M. R. Crimmin, A. G. M. Barrett, M. S. Hill and
P. A. Procopiou, Org. Lett., 2007, 9, 331–333.
73 K. Rajesh and H. Berke, Adv. Synth. Catal., 2013, 355, 901–
906.
8 P. C. Andrews, M. Brym, C. Jones, P. C. Junk and M. Kloth,
Inorg. Chim. Acta, 2006, 359, 355–363.
9 J. Cheng, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2011,
6
7, m987.
74 M. M. Mojtahedi, E. Akbarzadeh, R. Sharifi and
M. S. Abaee, Org. Lett., 2007, 9, 2791–2793.
0 A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn and
G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349– 75 T. Werner and J. Koch, Eur. J. Org. Chem., 2010, 6904–6907.
356. 76 D. C. Waddell and J. Mack, Green Chem., 2009, 11, 79–82.
1 J. Grundy, N. E. Mansfield, M. P. Coles and P. B. Hitchcock, 77 J. Li, J. Shi, H. Han, G. Z. H. Tong, X. Wei, D. Liu and
Inorg. Chem., 2008, 47, 2258–2260. M. F. Lappert, Organometallics, 2013, 32, 3721–3727.
2 L. Yang, D. R. Powell and R. P. Houser, Dalton Trans., 2007, 78 C. J. O’Connor, F. Manoni, S. P. Curran and S. J. Connon,
55–964. New J. Chem., 2011, 35, 551–553.
3 A. D. Pajerski, E. P. Squiller, M. Parvez, R. R. Whittle and 79 S. P. Curran and S. J. Connon, Org. Lett., 2012, 14, 1074–
H. G. Richey Jr., Organometallics, 2005, 24, 809–814. 1077.
4 K. W. Henderson, G. W. Honeyman, A. R. Kennedy, 80 N. E. Mansfield, M. P. Coles and P. B. Hitchcock, Polyhe-
1
5
5
5
5
9
R. E. Mulvey, J. A. Parkinson and D. C. Sherrington, Dalton
Trans., 2003, 1365–1372.
5 M. P. Coles, D. C. Swenson, R. F. Jordan and V. G. Young
dron, 2012, 37, 9–13.
81 G. Himbert, W. Schwickerath and G. Maas, Liebigs Ann.
Chem., 1985, 1389–1397.
5
5
5
5
5
6
6
6
6
6
6
Jr., Organometallics, 1997, 16, 5183–5194.
6 M. P. Coles, D. C. Swenson, R. F. Jordan and V. G. Young
Jr., Organometallics, 1998, 17, 4042–4048.
82 G. Himbert and W. Schwickerath, Liebigs Ann. Chem., 1984,
85–97.
83 H. Fujita, R. Endo, A. Aoyama and T. Ichii, Bull. Chem. Soc.
Jpn., 1972, 45, 1846.
84 W.-X. Zhang, M. Nishiura and Z. Hou, J. Am. Chem. Soc.,
2005, 127, 16788–16789.
85 W.-X. Zhang and Z. Hou, Org. Biomol. Chem., 2008, 6, 1720–
1730.
86 W. Yi, J. Zhang, M.-D. Li, Z. Chen and X. Zhou, Inorg.
Chem., 2011, 50, 11813–11824.
87 L. Xu, Y.-C. Wang, W.-X. Zhang and Z. Xi, Dalton Trans.,
2013, 42, 16466–16469.
88 T.-G. Ong, J. S. O’Brien, I. Korobkov and D. S. Richeson,
Organometallics, 2006, 25, 4728–4730.
89 A. G. M. Barrett, M. R. Crimmin, M. S. Hill,
P. B. Hitchcock, S. L. Lomas, M. F. Mahon, P. A. Procopiou
and K. Suntharalingam, Organometallics, 2008, 27, 6300–
6306.
7 A. R. Sadique, M. J. Heeg and C. H. Winter, Inorg. Chem.,
2
001, 40, 6349–6355.
8 M. Fan, Q. Yang, H. Tong, S. Yuan, B. Jia, D. Guo, M. Zhou
and D. Liu, RSC Adv., 2012, 2, 6599–6605.
9 D. Walther, P. Gebhardt, R. Fischer, U. Kreher and
H. Görls, Inorg. Chim. Acta, 1998, 281, 181–189.
0 S. L. Aeilts, M. P. Coles, D. C. Swenson, R. F. Jordan and
V. G. Young Jr., Organometallics, 1998, 17, 3265–3270.
1 M. H. Chisholm, K. Choojun, J. C. Gallucci and
P. M. Wambua, Chem. Sci., 2012, 3, 3445–3457.
2 C. N. Ayala, M. H. Chisholm, J. C. Gallucci and
C. Krempner, Dalton Trans., 2009, 9237–9245.
3 X. Xu, Y. Chen, G. Zou, Z. Ma and G. Li, J. Organomet.
Chem., 2010, 695, 1155–1162.
4 M. H. Chisholm, J. C. Gallucci and K. Phomphrai, Inorg.
Chem., 2002, 41, 2785–2794.
90 M. Arrowsmith, M. R. Crimmin, M. S. Hill, S. L. Lomas,
M. S. Heng, P. B. Hitchcock and G. Kociok-Köhn, Dalton
Trans., 2014, 43, DOI: 10.1039/c3dt53542h.
5 M. H. Chisholm and K. Phomphrai, Inorg. Chim. Acta,
2
003, 350, 121–125.
14314 | Dalton Trans., 2014, 43, 14302–14314
This journal is © The Royal Society of Chemistry 2014