Synthesis and crystallographic characterisation of a homologous series of bis-tridentate phosphine oxide NP3O3 Fe(II), Co(II), Ni(II) and Cu(II) complexes

Article abstract of DOI:10.1016/j.ica.2020.119870

Despite the large numbers of mono- and bidentate phosphine oxide complexes, examples of well characterised discrete multidentate phosphine oxide transition metal complexes are relatively rare. Herein, we report the coordination chemistry of the triphosphine oxide ligand N{CH₂P(O)Ph₂}₃ (NP₃O₃, 1) with a series of first row transition metals. Ligand 1 was found to form a series of bis-tridentate homoleptic complexes with Fe(II), Co(II), Ni(II) and Cu(II) salts of the type M(BF₄)₂·6H₂O. The anion exchange reaction of BF₄^— for BPh₄^— aided in the crystallisation of three of the four complexes. X-Ray crystallographic analysis of this series: [Fe(N{CH₂P(O)Ph₂}₃)₂](BPh₄)₂ (2), [Co(N{CH₂P(O)Ph₂}₃)₂](BPh₄)₂ (3), [Ni(N{CH₂P(O)Ph₂}₃)₂](BPh₄)₂ (4) and [Cu(N{CH₂P(O)Ph₂}₃)₂](BF₄)₂ (5) confirmed either almost idealised octahedral or distorted octahedral structures and tridentate coordination of 1. Complex 5, Cu(II) d⁹, displayed a classic Jahn-Teller distortion showing a prominent elongation of the axial M–O bonds. All complexes were found to be paramagnetic which precluded NMR spectroscopy analysis. Magnetic susceptibility measurements on solid samples of the complexes confirmed their number of unpaired electrons and high spin states.

Full text of DOI:10.1016/j.ica.2020.119870

Journal Pre-proofs
Research paper
Synthesis and crystallographic characterisation of a homologous series of bis-
tridentate phosphine oxide NP3O3 Fe(II), Co(II), Ni(II) and Cu(II) complexes
Samuel Page, Daisy Rogers-Simmonds, Andrew White, Philip W. Miller
PII:
DOI:
Reference:
S0020-1693(20)31069-0
https://doi.org/10.1016/j.ica.2020.119870
ICA 119870
To appear in:
Inorganica Chimica Acta
Received Date:
Revised Date:
Accepted Date:
6 May 2020
4 July 2020
4 July 2020
Please cite this article as: S. Page, D. Rogers-Simmonds, A. White, P.W. Miller, Synthesis and crystallographic
characterisation of a homologous series of bis-tridentate phosphine oxide NP3O3 Fe(II), Co(II), Ni(II) and Cu(II)
complexes, Inorganica Chimica Acta (2020), doi: https://doi.org/10.1016/j.ica.2020.119870
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier B.V.

Synthesis and crystallographic characterisation of a homologous series of bis-tridentate phosphine
oxide NP3O3 Fe(II), Co(II), Ni(II) and Cu(II) complexes
Samuel Page, Daisy Rogers-Simmonds, Andrew White and Philip W. Miller*
[*] Dr Philip W. Miller, Department of Chemistry, Imperial College London, South Kensington,
London, UK, SW7 2AZ
E-mail: philip.miller@imperial.ac.uk
Abstract
Despite the large numbers of mono- and bidentate phosphine oxide complexes, examples of well
characterised discrete multidentate phosphine oxide transition metal complexes are relatively rare.
Herein, we report the coordination chemistry of the triphosphine oxide ligand N{CH2P(O)Ph2}3
(NP3O3, 1) with a series of first row transition metals. Ligand 1 was found to form a series of bis-
tridentate homoleptic complexes with Fe(II), Co(II), Ni(II) and Cu(II) salts of the type M(BF4)2.6H2O.
—
—
The anion exchange reaction of BF4 for BPh4 aided in the crystallisation of three of the four
complexes. X-Ray crystallographic analysis of this series: [Fe(N{CH2P(O)Ph2}3)2](BPh4)2 (2),
[Co(N{CH2P(O)Ph2}3)2](BPh4)2 (3), [Ni(N{CH2P(O)Ph2}3)2](BPh4)2 (4) and [Cu(N{CH2P(O)Ph2}3)2](BF4)2 (5)
confirmed either almost idealised octahedral or distorted octahedral structures and tridentate
coordination of 1. Complex 5, Cu(II) d9, displayed a classic Jahn-Teller distortion showing a
prominent elongation of the axial M-O bonds. All complexes were found to be paramagnetic which
precluded NMR spectroscopy analysis. Magnetic susceptibility measurements on solid samples of
the complexes confirmed their number of unpaired electrons and high spin states.

Introduction
Phosphine oxides are a broad class of ligand that have been used to prepare numerous well-defined
transition metal, lanthanide and actinide complexes. They have found wide ranging applications for
the preparation of luminescent and near-IR emitting materials,[1,2] as extractants for rare earth and
actinide elements,[3-5] capping agents for quantum dots[6] and as ligands in homogeneous
catalysis.[7-9] Common phosphine oxides of the type: R3PO (R = alkyl or aryl) typically coordinate in
a monodentate fashion via the oxygen atom and can be considered qualitatively to be hard bases
owing to the significant ionic nature of the P-O bond. They have therefore been more widely
exploited for the coordination of harder and less polarisable metal ions such as the lanthanides and
higher oxidation state transition metals. Although a wide range of diphosphine oxide and mixed
phosphine oxide-P, N, O and carbene donor ligands have been used to tune the stability, solubility,
catalytic and luminescent properties of complexes[10] the number of triphosphine oxide ligands is
much lower despite their potential for tuning the physical properties of resulting complexes.
Recent examples of such polyphosphine oxides include a series of P3O3 ligands, reported by
Morris et al.,[11] (Fig. 1) that have the potential to act as an extraction agent for nuclear fuel
treatment and were demonstrated to effectively coordinate to yttrium(III). Y(III) is often used as
model diamagnetic ion for investigating behaviour of related paramagnetic lanthanides owing to its
ionic radius and similar chemical behaviour. The triphosphine oxide ligand (RP*,SP*-P3O3, Fig. 1),
which has two chiral diphenylphosphinic amide groups connected through a phenylphosphine oxide
moiety, was also found to effectively coordination Y(III)[12] and Eu(III)[13] in a tridentate fashion
generating complexes with either one or two of the ligands occupying metal coordination sites.
Further examples include the structurally similar DPPPO and DPPYPO ligands,[14] and the more rigid
TPPM ligand[15] that have been used to prepare a range of nine-coordinate and emissive Eu(III)
complexes. In contrast to discrete tridentate complexes above, the P3O3 ligand (Fig. 1) was found to

form coordination polymers with La(III) and Pr(III) ions.[16] The P3O3 ligand chelates in a bidentate
manner, forming an eight membered ring with the lanthanide ion, while the third -CH2P(O)Ph2 arm
bridges to a second metal ion chelating generating a coordination polymer chain.
Our group has recently explored a range triphosphine ligands for small molecule activation
and catalytic applications.[17-21] The synthesis of these ligands is relatively straight forward and
involves a phosphorous-based Mannich reaction of a phosphonium chloride salt with an amine to
generate N-centred triphos ligands. We have prepared a series of symmetrical N-triphos ligands
with different -PR2 groups and conducted a number of coordination studies with a range of
transition metals.[22] The coordination chemistry of the equivalent oxide N-triphos ligands has not
been widely investigated and could find potential applications in the areas of metal sequestration
and for preparation of electroluminescent and magnetic materials. To the best of our knowledge,
only one complex with the N-triphos oxide ligand has been reported, the tetrahedral beryllium
complex [(NP3O3)Be(OH2)]2+.[23] Herein, we report the synthesis and structural characterisation of
a homologous series of bis-tridentate first row transition metal: Fe(II), Co(II), Ni(II) and Cu(II)
complexes with the N-triphos oxide ligand NP3O3.
Results and Discussion
The N-triphos oxide ligand (NP3O3, 1) was prepared in high yield via the known phosphorous-based
Mannich reaction of bis(hydroxymethyl)diphenyl phosphonium chloride with formaldehyde in
methanol followed by oxidation of the isolated N-triphos ligand with hydrogen peroxide solution
(Scheme 1). 1H and 13C NMR spectra conformed to the expected ligand structure and the 31P{1H}
NMR spectrum displayed a characteristic higher frequency shift in resonance to +30.4 ppm upon
oxidation compared to the parent N-triphos ligand. The ES-MS mass spectrum showed the expected
molecular ion peak and a characteristic P(O)Ph2 fragment at 201 m/z. Single crystals suitable for X-
ray diffraction were obtained on recrystallisation of NP3O3 via vapour diffusion from

dichloromethane and diethyl ether, crystallising as a solvate of water and hydrogen peroxide. The
molecular structure of NP3O3 (Fig. 2) shows pseudo three-fold symmetry around the slightly
puckered central nitrogen atom of the ligand. The crystal system and space group were found to
differ from a previously reported unsolvated structure of this ligand.[24] No significant difference
between the P=O bond lengths of the solvate and unsolvated structure was observed.
Reaction of two equivalents of NP3O3 with Fe(BF4)2.6H2O and NaBPh4 in a mixture of
THF:butan-1-ol at RT for one hour yielded a pale orange/brown precipitate that was isolated by
filtration and washed with diethyl ether. The 31P{1H} NMR spectra of the product in both deuterated
1
chloroform and acetonitrile displayed no signals across the full spectrum range while the H NMR
spectra showed broad and unresolved signals that are characteristic of a paramagnetic complex. This
indicated a likely paramagnetic high-spin d6 Fe(II) metal centre in a probable octahedral coordination
environment. The absence of a free ligand signal suggested complete reaction of the ligand and
coordination of all the phosphine oxide arms of the ligand to the metal centre. When the complex
was dissolved in deuterated DMSO the 31P{1H} and 1H NMR spectra showed signals indicative of the
free NP3O3 ligand, which therefore demonstrates the lability of this ligand system when there is a
competitive ligand present. IR spectroscopy of the product displayed a negligible variation in the
P=O stretching frequencies compared to the free NP3O3 ligand and indicates that the phosphine
oxide group largely retains its double bond character upon metal coordination. The ES-MS mass
spectrum displayed peak at 687 m/z which can be assigned to the molecular ion
[Fe(N{CH2P(O)Ph2}3)2]+2.
Isolation of the [Fe(N{CH2P(O)Ph2}3)2](BPh4)2 complex and recrystallisation via vapour
diffusion from dichloromethane and diethyl ether, in air, yielded single crystals suitable for X-ray
diffraction. The molecular structure of Fe(N{CH2P(O)Ph2}3)2](BPh4)2 (Fig. 3) revealed a bis-tridentate
homoleptic complex with an almost idealised octahedral arrangement, solvated with
dichloromethane. The three phosphine oxide arms of each ligand coordinate to the central Fe(II)

-
centre via the oxygen atoms to form the octahedral coordination environment while the two BPh4
-
anions remain outside the coordination sphere. Attempts to obtain crystals of the BF4 adduct
-
proved unsuccessful and demonstrated that the large BPh4 counter ions were important in
obtaining good quality crystals. There is no significant lengthening of the P=O double bonds on
coordination to the metal centre (Table 1), as was suggested by the minimal changes in the IR
stretching frequency compared to free ligand. The O-Fe bond distances are within the expected
ranges for metal phosphine oxide complexes. The central apical nitrogen of the ligand remained
uncoordinated, resulting in tridentate coordination on the ligand, a feature that we have commonly
observed with many of our N-triphosPh complexes. The central nitrogen also adopts a greater degree
of planarity compared to the nitrogen atom in the free ligand. Interestingly, two ligands coordinate
to the metal centre to form a bis-tridentate homoleptic complex; we have not previously observed
two N-triphosPh ligands coordinating to a metal centre owing to the steric crowding that would result
at the metal centre from a direct M-PPh2 bond. The extra length of the Fe-O=PPh2 bond in the NP3O3
complex reduces such steric crowding around the metal centre and enables coordination of two
NP3O3 ligands in a tridentate fashion.
Reaction of Co(BF4)2.6H2O with two equivalents of NP3O3, with an in-situ anion substitution
of [Co(N{CH2P(O)Ph2}3)2](BF4)2 with NaBPh4 in an analogous manner to the Fe complex above,
1
resulted in the formation of a pale-pink precipitate [Co(N{CH2P(O)Ph2}3)2](BPh4)2 (3). 31P{1H} and H
NMR spectra displayed similar features to that of the Fe complex showing either no signal or broad
unresolved signals respectively, again indicative of an expected d7 Co(II) paramagnetic metal centre.
The ES-MS mass spectrum displayed peak at 688 m/z indicating the molecular ion
[Co(N{CH2P(O)Ph2}3)2]+2 with two coordinating NP3O3 ligands. Single crystals suitable for X-ray
diffraction were grown via vapour diffusion of diethyl ether into acetonitrile. The molecular
structure of 3 (Fig. 4) was similar to that of 2, and showed the chelation of two NP3O3 ligands to the
octahedral Co(II) centre.

The reactions of Ni(BF4)2.6H2O and Cu(BF4)2.6H2O with two equivalents of NP3O3 (alongside
NaBPh4 for M = Ni) were performed in THF:propan-2-ol mixtures and yielded off-white and pale-
1
yellow solids respectively. 31P{1H} and H NMR spectra were similar to those observed for the
paramagnetic Fe(II) and Co(II) complexes and showed broad unresolved signals. Mass spectrometry
again gave molecular ion peaks that were consistent with [M(N{CH2P(O)Ph2}3)2]+2 complexes. Single
crystals of both were grown via vapour diffusion of diethyl ether into acetonitrile solutions. Single
crystal X-ray analysis confirmed two analogous bis-NP3O3 octahedral structures,
[Co(N{CH2P(O)Ph2}3)2](BPh4)2 (4) and [Ni(N{CH2P(O)Ph2}3)2](BF4)2 (5) (Fig. 5 and Fig. 6). The crystal
structures of 3-5 were solvated by acetonitrile. The Ni(II) structure shows no significant deviation in
Ni-O bond lengths, however, the Cu(II) d9 complex displays a classic Jahn-Teller distortion with
significant axial elongation of Cu-O(5) 2.318(3) Å compared to the equatorial bonds Cu-O(3) 1.958(3)
Å and Cu-O(7) 2.043(3) Å (Fig. 6, Table 2). The P-O bond lengths, and corresponding infra-red
stretches, for the series 2-5 all negligible differences from the free NP3O3 ligand (Table 1).
Magnetic susceptibility measurements were attempted using the Evans method however
low solubilities of complexes in common solvents CDCl3 and CD3CN precluded accurate
measurements and resulted in values much smaller than predicted. Changing the solvent to d6-
DMSO improved solubility but resulted in significant complex decomposition as evidenced by the
appearance of a free ligand peak in the 31P NMR spectrum. A magnetic susceptibility balance was
therefore used to take measurements on solid samples, Table 3 provides a summary of the magnetic
susceptibility measurements for complexes 2-5. Complex 2 was found to be high spin Fe(II) d6 with
four unpaired electrons with a magnetic moment of µeff = 4.81 µB as might be expected for a first-
row transition metal in an octahedral field in +2 oxidation state. Complex 3, Co(II) d7 was also found
to be high spin with a magnetic moment of µeff = 4.12 µB which corresponds to three unpaired
electrons. The Ni(II) d8 and Cu(II) d9 complexes 4 and 5 have only one possible electronic
configuration in an octahedral field resulting in two and one unpaired electrons respectively as
confirmed by the magnetic susceptibility measurements (Table 3).

In conclusion, the coordination chemistry of the tridentate phosphine oxide ligand N,N,N-
tris(diphenylphosphinomethylene)amine (NP3O3) has been investigated with a series of first-row
transition metals (Fe, Co, Ni, Cu) in the +2 oxidation state. It is a rare example of a tridentate
phosphine oxide ligand system that has been demonstrated to form a series of discrete transition
metal coordination complexes. All complexes formed octahedral geometries with two ligands
chelating in a tridentate manner via the oxygen atoms of the P(O)PPh2 moieties. The structures of
complexes 3-5 formed near idealised octahedral geometries while 5, Cu(II) d9, displayed a significant
axial elongation owing to a Jahn-Teller distortion. Magnetic susceptibility measurements confirmed
the paramagnetic nature of the complexes and established their expected number of unpaired
electrons. Elemental analysis of complexes 2-3 was found to give lower than expected percentage of
carbon. This is likely due to slight contamination from NaBF4 salts that form from the anion
substitution reaction. The ease of synthesis of NP3O3 via oxidation of N-triphos ligands opens up the
possibilities of a much wider array of complexes with a variety of P(O)R2 groups that could be used
to alter the solubility, magnetic and luminescent properties of such complexes and potentially find
applications in the fields of metal sequestration, electroluminescence and magnetic materials as has
been seen for mono and bidentate phosphine oxide ligands. We therefore anticipate that such
tridentate phosphine oxide ligands will attract further interest for the preparation of novel transition
metal and lanthanide complexes.
Experimental
General considerations
All chemicals were purchased from commercial sources and used as received without further
purification. Unless otherwise stated, all of the reactions have been performed using standard
1
Schlenk line techniques. All solvents were degassed prior to use. H, 13C and 31P{1H} NMR spectra
were recorded on a Bruker AV 400 MHz NMR spectrometer at 400 MHz, 100 MHz and 162 MHz
respectively. IR spectra were recorded on a Perkin Elmer Spectrum 100 FT-IR spectrometer. Melting

point measurements were recorded on an OptiMelt Automated Melting Point System. Mass
spectrometry analyses were performed by the Mass Spectrometry Service, Imperial College London.
Elemental analyses were performed by Mr. Stephen Boyer of the School of Human Sciences, London
Metropolitan University. Crystallographic analyses were performed by Dr. Andrew White of the
Department of Chemistry at Imperial College London. Magnetic susceptibility measurements were
recorded using a Sherwood Scientific Magnetic Susceptibility Balance Mark 1, samples were ground
into fine powder and loaded into a sample tube and the data were corrected for diamagnetism. N-
TriphosPh was synthesised according to the literature procedure.[25,26]
Synthesis of N{CH2P(O)Ph2}3 (NP3O3) (1)
To a solution of N-TriphosPh (0.5 g, 0.817 mmol) in THF (10 mL), was added hydrogen peroxide (30 %
solution in water) (0.25 mL, 2.45 mmol) and the resulting colourless solution was stirred at RT for 2
hrs. The mixture was then dried over MgSO4, filtered and the solvent removed under reduced
pressure to yield a white solid. The solid was washed with diethyl ether and dried in vacuo. Single
crystals suitable for X-ray diffraction were grown by vapour diffusion of diethyl ether into a
1
dichloromethane solution (0.44 g, 0.67 mmol, 82 %). H NMR (CDCl3, 400.2 MHz): δ 4.23 (d, 6H, CH2,
2JPH = 6.4 Hz), 7.2-7.9 ppm (m, 30H, C6H5). 31P{1H} NMR (CDCl3, 162 MHz): δ 30.4 ppm (s). 13C{1H}
1
3
NMR (CDCl3, 100.6 MHz): δ 128.4-131.6 (m, C6H5), 58.1 (dt, CH2, JP,C = 84.2 Hz, JP,C = 7.4 Hz). IR
(solid, cm-1): 693 (s), 716 (s), 857 (m), 1118 (w), 1167 (s, νP=O), 1435 (m). ES-MS (positive ion): m/z
calcd. for C39H37NO3P3 ([M+H]+) 660.1981. Found: 660.1998 (100 %). Melting point 129.2-129.4 °C.
Anal. Calcd. for C39H36NO3P3.2H2O: C, 67.33; H, 5.80; N, 2.01. Found: C, 67.08; H, 5.82; N, 2.27.
Synthesis of [Fe{N(CH2P(O)Ph2)3}2](BPh4)2 (2)
To a solution of 1 (0.15 g, 0.227 mmol) and NaBPh4 (0.078 g, 0.227 mmol) in THF (5 mL) was added
Fe(BF4)2.6H2O (0.038 g, 0.114 mmol) in butan-1-ol (5 mL) and stirred at room temperature for one
hour. The resulting pale-orange precipitate was isolated by cannula filtration, washed with diethyl
ether and dried in vacuo (0.12 g, 0.06 mmol, 52 %). Single crystals suitable for X-ray diffraction

were grown by vapour diffusion of diethyl ether into a dichloromethane solution. IR (solid, cm-1):
693 (s), 857 (s), 1129 (m), 1164 (s, νP=O), 1435 (m), 2982 (vw), 3049 (vw). ES-MS (positive ion): m/z
calcd. for [C78H72FeN2O6P6]2+ ([M]2+) 687.1583. Found: [M]2+ 687.1609 (1 %), [1+H]+ 660.1850 (100 %).
The analogous compound [Fe{N(CH2P(O)Ph2)3}2](BF4)2 (2a) could be prepared and isolated in a
similar way by omitting the use of NaBPh4 (0.05 g, 0.032 mmol, 28 %). Anal. Calcd. for
C78H72B2F8FeN2O6P6: C, 60.49; H, 4.69; N, 1.81. Found: C, 59.14; H, 4.64; N 1.88.
Synthesis of [Co{N(CH2P(O)Ph2)3}2](BPh4)2 (3)
To a solution of 1 (0.150 g, 0.227 mmol) and NaBPh4 (0.078 g, 0.227 mmol) in THF (5 mL) was added
Co(BF4)2.6H2O (0.038 g, 0.114 mmol) in 1-butanol (8 mL) and stirred at room temperature for one
hour. The resulting pale-pink precipitate was isolated by cannula filtration, washed with hexane and
dried in vacuo (0.172 g, 0.085 mmol, 75 %). Single crystals suitable for X-ray diffraction were grown
by vapour diffusion of diethyl ether into an acetonitrile solution. IR (solid, cm-1): 693 (s), 857 (s), 1118
(m), 1165 (w, νP=O), 1435 (m), 2986 (vw), 3042 (vw). ES-MS (positive ion): m/z calcd. for
[C78H72CoN2O6P6]2+ ([M]2+) 688.6574. Found: [M]2+ 688.6606 (3 %), [1+H]+ 660.2171 (100 %). The
analogous compound [Co{N(CH2P(O)Ph2)3}2](BF4)2 (3a) could be prepared and isolated in a similar
way by omitting the use of NaBPh4 (0.064 g, 0.041 mmol, 36 %).
C78H72B2F8CoN2O6P6: C, 60.37; H, 4.68; N, 1.81. Found: C, 60.28; H, 4.51; N 1.91.
Anal. Calcd. for
Synthesis of [Ni(N(CH2P(O)Ph2)3)2](BPh4)2 (4)
To a solution of 1 (0.150 g, 0.227 mmol) and NaBPh4 (0.078 g, 0.227 mmol) in THF (10 mL) was added
Ni(BF4)2.6H2O (0.039 g, 0.114 mmol) in propan-2-ol (12 mL) and stirred at room temperature for one
hour. The resulting pale-yellow precipitate was isolated by cannula filtration, washed with hexane
and dried in vacuo (0.16 g, 0.08 mmol, 69 %). Single crystals suitable for X-ray diffraction were grown
by vapour diffusion of diethyl ether into an acetonitrile solution. IR (solid, cm-1): 697 (s), 857 (s), 1166
(s, νP=O), 1435 (m), 3042 (vw). ES-MS (positive ion): m/z calcd. for [C78H72NiN2O6P6]2+ ([M]2+)
688.1585. Found: [M]2+ 688.1607 (3 %), [1+H]+ 660.1376 (100 %)]. The analogous compound

[Ni{N(CH2P(O)Ph2)3}2](BF4)2 (4a) could be prepared and isolated in a similar way by omitting the use
of NaBPh4 [Ni(NP3O3)2](BF4)2 (0.042 g, 0.027 mmol, 24 %).
Synthesis of [Cu(N(CH2P(O)Ph2)3)2](BF4)2 (5)
To a solution of 1 (0.150 g, 0.227 mmol) in THF (10 mL) was added Cu(BF4)2.6H2O (0.039 g, 0.114
mmol) dissolved in propan-2-ol (12 mL) and stirred at room temperature for one hour. The resulting
orange/yellow precipitate was isolated by cannula filtration, washed with hexane and dried in vacuo
(0.022 g, 0.014 mmol, 12 %). Single crystals suitable for X-ray diffraction were grown by vapour
diffusion of diethyl ether into an acetonitrile solution. IR (solid, cm-1): 541 (s), 695 (s), 712 (s), 861 (s),
1024 (s), 1058 (s, br), 1117 (m), 1158 (s, νP=O), 1436 (m). ES-MS (positive ion): m/z calcd. for
[C78H72CuN2O6P6]2+ ([M]2+) 690.6556. Found: [M]2+ 690.6541 (13 %), [1+H]+ 660.1808 (100 %).
Acknowledgements
SP is grateful to Imperial College London for a President’s Scholarship award and DRS is grateful to
Imperial College London for a UROP busrary.
Tables, Schemes and figures
Table 1. P-O bond lengths (ꢀ) for NP3O3 and complexes 2-5.
NP3O3
(1)
[Fe(NP3O3)2] [Co(NP3O3)2] [Ni(NP3O3)2] [Cu(NP3O3)2]
(BPh4)2 (2)
(BPh4)2 (3)
1.491(3)
1.492(3)
1.492(3)
(BPh4)2 (4)
(BF4)2 (5)
P(3)-O(3) 1.485(2) 1.4918(11)
P(5)-O(5) 1.491(2) 1.4935(11)
P(7)-O(7) 1.477(2) 1.4901(11)
1.4952(12) 1.504(3)
1.4957(12) 1.494(4)
1.4959(12) 1.497(3)
Table 2. M-O bond lengths (ꢀ) for complexes 2-5.
[Fe(NP3O3)2] [Co(NP3O3)2] [Ni(NP3O3)2] [Cu(NP3O3)2]
(BPh4)2 (2)
M(1)-O(3) 2.135(3)
M(1)-O(3)’ 2.135(3)
M(1)-O(5) 2.128(3)
M(1)-O(5)’ 2.128(3)
M(1)-O(7) 2.117(2)
M(1)-O(7)’ 2.117(2)
(BPh4)2 (3)
2.1140(11)
2.1140(11)
2.0655(10)
2.0656(10)
2.1166(11)
2.1167(11)
(BPh4)2 (4)
(BF4)2 (5)
2.0811(12) 1.958(3)
2.0811(12) 1.958(3)
2.0451(11) 2.318(3)
2.0451(11) 2.318(3)
2.0846(11) 2.043(3)
2.0847(11) 2.044(3)
Table 3. Magnetic susceptibility data for complexes 2-5.

[Fe(NP3O3)2] [Co(NP3O3)2] [Ni(NP3O3)2] [Cu(NP3O3)2]
(BPh4)2 (2)
(BPh4)2 (3)
(BPh4)2 (4)
(BF4)2 (5)
d-electron count
d6
4
d7
d8
d9
unpaired electrons (high-spin)
effective magnetic moment (µeff) 4.81 µB
Spin only magnetic moment (µso) 3.91
3
2
1
4.12 µB
3.24
2.65 µB
1.83
2.09 µB
1.32
Figure 1. Examples of tridentate phosphine oxide ligands.
Scheme 1. Synthesis of NP3O3 via the phosphorus-based Mannich reaction and subsequent
oxidation, followed by coordination to M2+ (M= Fe, Co, Ni or Cu) to form the homologous series of
bis-chelate complexes 2-5.

Fig. 2
The crystal structure of NP3O3, 1. Solvent and hydrogen atoms omitted for clarity.
Fig. 3
The structure of the Ci-symmetric complex cation present in the crystal of 2. Solvent
atoms, counter ions and hydrogen atoms omitted for clarity.

Fig. 4
The structure of the Ci-symmetric complex cation present in the crystal of 3. Solvent
atoms, counter ions and hydrogen atoms omitted for clarity.
Fig. 5
The structure of the Ci-symmetric complex cation present in the crystal of 4. Solvent
atoms, counter ions and hydrogen atoms omitted for clarity.

Fig. 6 The structure of the Ci-symmetric complex cation present in the crystal of 5. Solvent atoms,
counter ions and hydrogen atoms omitted for clarity.
References
[1]
[2]
[3]
[4]
M. Hatanaka, Y. Hirai, Y. Kitagawa, T. Nakanishi, Y. Hasegawa, K. Morokuma, Chem. Sci. 8
(2016) 423–429. https://doi.org/10.1039/c6sc03006h.
Y. Hirai, T. Nakanishi, Y. Kitagawa, K. Fushimi, T. Seki, H. Ito, Y. Hasegawa, Angew. Chemie -
Int. Ed. 55 (2016) 12059–12062. https://doi.org/10.1002/anie.201606371.
M.Y. Alyapyshev, V.A. Babain, Y.A. Ustynyuk, Russ. Chem. Rev. 85 (2016) 943–961.
https://doi.org/10.1070/rcr4589.
M. Gilmore, É.N. McCourt, F. Connolly, P. Nockemann, M. Swadźba-Kwaśny, J.D. Holbrey, ACS
Sustain. Chem. Eng. 6 (2018) 17323–17332.
https://doi.org/10.1021/acssuschemeng.8b04843.
[5]
F. Arnaud-Neu, J.K. Browne, D. Byrne, D.J. Marrs, M.A. McKervey, P. O’Hagan, M.J. Schwing-
Weill, A. Walker, Chem. - A Eur. J. 5 (1999) 175–186. https://doi.org/10.1002/(SICI)1521-

3765(19990104)5:1<175::AID-CHEM175>3.0.CO;2-P.
[6]
[7]
M. Green, J. Mater. Chem. 20 (2010) 5797–5809. https://doi.org/10.1039/c0jm00007h.
N. Almora-Barrios, I. Cano, P.W.N.M. Van Leeuwen, N. López, ACS Catal. 7 (2017) 3949–3954.
https://doi.org/10.1021/acscatal.7b00355.
[8]
[9]
J. Gao, B. Yang, C. Chen, J. Catal. 369 (2019) 233–238.
https://doi.org/10.1016/j.jcat.2018.11.007.
T.M. Shaikh, C.M. Weng, F.E. Hong, Coord. Chem. Rev. 256 (2012) 771–803.
https://doi.org/10.1016/j.ccr.2011.11.007.
[10] A.W.G. Platt, Coord. Chem. Rev. 340 (2017) 62–78.
https://doi.org/10.1016/j.ccr.2016.09.012.
[11] P.E. Sues, A.J. Lough, R.H. Morris, Inorg. Chem. 51 (2012) 9322–9332.
https://doi.org/10.1021/ic3010147.
[12] C. Popovici, I. Fernández, P. Oña-Burgos, L. Roces, S. García-Granda, F.L. Ortiz, Dalt. Trans. 40
(2011) 6691–6703. https://doi.org/10.1039/c1dt10194c.
[13] F.J. Sainz-Gonzalo, M. Casimiro, C. Popovici, A. Rodríguez-Diéguez, J.F. Fernández-Sánchez, I.
Fernández, F. López-Ortiz, A. Fernández-Gutiérrez, Dalt. Trans. 41 (2012) 6735–6748.
https://doi.org/10.1039/c2dt30257h.
[14] K. Miyata, Y. Hasegawa, Y. Kuramochi, T. Nakagawa, T. Yokoo, T. Kawai, Eur. J. Inorg. Chem.
(2009) 4777–4785. https://doi.org/10.1002/ejic.200900598.
[15] Y. Kuramochi, T. Nakagawa, T. Yokoo, J. Yuasa, T. Kawai, Y. Hasegawa, Dalt. Trans. 41 (2012)
6634–6640. https://doi.org/10.1039/c2dt30538k.
[16] A.M.J. Lees, A.W.G. Platt, Polyhedron. 24 (2005) 427–433.
https://doi.org/10.1016/j.poly.2004.12.003.
[17] A. Phanopoulos, N.J. Brown, A.J.P. White, N.J. Long, P.W. Miler, Inorg. Chem. 53 (2014) 3742–
3752. https://doi.org/10.1021/ic500030k.
[18] A. Phanopoulos, A.J.P. White, N.J. Long, P.W. Miller, ACS Catal. 5 (2015) 2500–2512.

https://doi.org/10.1021/cs502025t.
[19] S.L. Apps, P.W. Miller, N.J. Long, Chem. Commun. 55 (2019) 6579–6582.
https://doi.org/10.1039/c9cc01496a.
[20] S.L. Apps, A.J.P. White, P.W. Miller, N.J. Long, Dalt. Trans. 47 (2018) 11386–11396.
https://doi.org/10.1039/c8dt02471e.
[21] M.J. Hanton, S. Tin, B.J. Boardman, P. Miller, J. Mol. Catal. A Chem. 346 (2011) 70–78.
https://doi.org/10.1016/j.molcata.2011.06.010.
[22] A. Phanopoulos, A.J.P. White, N.J. Long, P.W. Miller, Dalt. Trans. 45 (2016) 5536–5548.
https://doi.org/10.1039/c6dt00170j.
[23] F. Cecconi, C. A. Ghilardi, S. Midollini, A. Orlandini, Inorg. Chem. Commun. 3 (2000) 350–353.
https://doi.org/10.1016/S1387-7003(00)00092-7.
[24] O. Walter, IUCrData. 2 (2017). https://doi.org/10.1107/s2414314617014791.
[25] G. Markl, G.Y. Jin, Tetrahedron Lett. 22 (1981) 1105–1108.
[26] A. Phanopoulos, N. Long, P. Miller, JOVE-JOURNAL Vis. Exp. (2015).
https://doi.org/10.3791/52689.
Author statement
Samuel Page: Investigation, Writing-Original draft preparation.
Daisy Rogers-Simmonds: Investigation, Validation.
Andrew White: Investigation, Visualization.
Philip Miller: Conceptualization, Supervision, Methodology, Writing- Reviewing and Editing.
Synthesis and crystallographic characterisation of a homologous series of bis-tridentate phosphine
oxide NP3O3 Fe(II), Co(II), Ni(II) and Cu(II) complexes

Samuel Page, Daisy Rogers-Simmonds, Andrew J. P. White and Philip W. Miller*
[*] Dr Philip W. Miller, Department of Chemistry, Imperial College London, South Kensington,
London, UK, SW7 2AZ
E-mail: philip.miller@imperial.ac.uk
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
Highlights


A new series tridentate phosphine oxide Fe(II), Co(II), Ni(II) and Cu(II) complexes with the
NP3O3 ligand
X-ray crystallographic characterisation reveals bis-chelates with idealised octahedral
coordination geometries


Cu(II) complex displays a classic Jahn-Teller distortion
Complexes were found to be paramagnetic